Statues in Context: Production, Meaning and (Re)Uses (British Museum Publications on Egypt and Sudan) 9789042938076, 9789042938083, 9042938072

Moving beyond typological and stylistic discourses on Egyptian statuary, the papers gathered here seek to explore the ar

332 57 19MB

English Pages 302 [325] Year 2019

Report DMCA / Copyright

DOWNLOAD PDF FILE

Table of contents :
TABLE OF CONTENTS
LIST OF CONTRIBUTORS
2016 COLLOQUIUM PROGRAMME
PREFACE
WHEN STATUES SPEAK ABOUT THEMSELVES1
A PERFECT ‘LIKENESS’? VIEWING LATE PERIOD ARCHAISING SCULPTURE IN CONTEXT
SMALL DIVINE STATUETTES: OUTFITTING RELIGION1
ORIGINS OF METALS FOR COUNTLESS BRONZES
EVIDENCE SUGGESTING ANOTHER BOSTON DYAD OF MENKAURA AND A QUEEN
IN TEMPLE AND HOME: STATUARY IN THE TOWN OF AMARA WEST, UPPER NUBIA
EVIDENCE FROM AMARNA RELIEFS CONCERNING ROYAL STATUES AND THEIR CONTEXTS: HOW ACCURATE WERE THE RELIEFS?
RECONSTRUCTING THE STATUARY OF THE COURTYARD OF THE TEMPLE OF KHNUM ON ELEPHANTINE
EGYPTIAN AND CYPRIOT STONE STATUETTES IN CONTEXT AT LATE PERIOD NAUKRATIS
A CLASSIFICATION OF ‘SACRED’ CACHES IN ANCIENT EGYPT
THOUSANDS OF OSIRIS: THE ARCHAEOLOGICAL CONTEXTS OF THE BRONZES FOUND IN THE TEMPLE OF ‘AYN MANAWÎR AND AT THE SERAPEUM OF MEMPH
LOOKING FOR CONTEXTS: RECENT WORK ON THE KARNAK CACHETTE PROJECT
TAHARQO AND HIS DESCENDANTS: A STATUE CACHE UPSTREAM OF THE FIFTH NILE CATARACT
RITUAL METAL STATUARY IN ANCIENT EGYPT: ‘A LONG LIFE AND A GREAT AND GOOD OLD AGE’
STATUES IN LATE ANTIQUE EGYPT: FROM PRODUCTION AND DISPLAY TO ARCHAEOLOGICAL RECORD
KILLING OR ‘DE-ACTIVATING’ EGYPTIAN STATUES: WHO MUTILATED THEM, WHEN, AND WHY?
Recommend Papers

Statues in Context: Production, Meaning and (Re)Uses (British Museum Publications on Egypt and Sudan)
 9789042938076, 9789042938083, 9042938072

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

BRITISH MUSEUM PUBLICATIONS ON EGYPT AND SUDAN 10

STATUES IN CONTEXT Production, meaning and (re)uses

edited by

Aurélia MASSON-BERGHOFF

PEETERS

STATUES IN CONTEXT

BRITISH

MUSEUM

PUBLICATIONS

ON

EGYPT

STATUES IN CONTEXT

Production, meaning and (re)uses

edited by

Aurélia MASSON-BERGHOFF

PEETERS LEUVEN – PARIS – BRISTOL, CT 2019

AND

SUDAN

10

Cover illustration: Colossal statue of Psamtek I discovered in 2017 in Heliopolis. © Heliopolis Project (Aiman Ashmawy/Dietrich Raue/Kai-Christian Bruhn), photograph by Dietrich Raue. With thanks to the Ministry of Antiquities (Egypt) and the Ägyptisches Museum der Universität Leipzig.

A catalogue record for this book is available from the Library of Congress. ISBN 978-90-429-3807-6 eISBN 978-90-429-3808-3 D/2019/0602/104 © 2019, Peeters, Bondgenotenlaan 153, B-3000 Leuven, Belgium No part of this book may be reproduced in any form or by any electronic or mechanical means, including information storage or retrieval devices or systems, without the prior written permission from the publisher, except the quotation of brief passages for review purposes.

TABLE OF CONTENTS List of contributors ............................................................................................................................................

VII

2016 Colloquium Programme ...........................................................................................................................

IX

Aurélia MASSON-BERGHOFF Preface................................................................................................................................................................

XI

MEANING

I. AND FUNCTION

Elizabeth FROOD When statues speak about themselves...............................................................................................................

3

Campbell PRICE A perfect ‘likeness’? Viewing Late Period archaising sculpture in context....................................................

21

Marsha HILL Small divine statuettes: outfitting religion........................................................................................................

35

II. PRODUCTION: TECHNOLOGY AND WORKSHOPS Aurélia MASSON-BERGHOFF and Ernst PERNICKA Origins of metals for countless bronzes ............................................................................................................

53

Florence Dunn FRIEDMAN with assistance from Michelle PISA Evidence suggesting another Boston dyad of Menkaura and a queen .............................................................

73

III.

VISIBLE STATUES: TEMPLES, PALACES AND HOUSES Neal SPENCER In temple and home: statuary in the town of Amara West, Upper Nubia .......................................................

95

Kristin THOMPSON Evidence from Amarna reliefs concerning royal statues and their contexts: how accurate were the reliefs? ...............................................................................................................................................................

131

Tobias KRAPF Reconstructing the statuary of the courtyard of the temple of Khnum on Elephantine ..................................

147

Ross THOMAS Egyptian and Cypriot stone statuettes in context at Late Period Naukratis .....................................................

159

VI

TABLE OF CONTENTS

IV. BECOMING INVISIBLE:

STATUE CACHES

Guillaume CHARLOUX and Mona Ali Abady MAHMOUD A classification of ‘sacred’ caches in ancient Egypt ........................................................................................

183

Florence GOMBERT-MEURICE Thousands of Osiris: the archaeological contexts of the bronzes found in the temple of ‘Ayn Manawîr and at the Serapeum of Memphis.............................................................................................................................

197

Laurent COULON, Yves EGELS, Emmanuel JAMBON and Emmanuel LAROZE Looking for contexts: recent work on the Karnak Cachette Project................................................................

209

Julie Renee ANDERSON, Salaheldin MOHAMMED AHMED, Mahmoud SULIMAN BASHIR and Rihab KHIDIR

ELRASHEED

Taharqo and his descendants: a statue cache upstream of the Fifth Nile Cataract .........................................

229

V.

AFTERLIVES: REUSE AND DESTRUCTION Deborah SCHORSCH Ritual metal statuary in ancient Egypt: ‘A long life and a great and good old age’ ......................................

249

Troels Myrup KRISTENSEN Statues in late antique Egypt: from production and display to archaeological record ....................................

269

Simon CONNOR Killing or ‘de-activating’ Egyptian statues: who mutilated them, when, and why? .......................................

281

LIST OF CONTRIBUTORS Salaheldin Mohammed Ahmed Qatar-Sudan Archaeological Project (QSAP) [email protected] Julie Renee Anderson Department of Ancient Egypt and Sudan British Museum [email protected] Mahmoud Suliman Bashir National Corporation for Antiquities and Museums, Sudan [email protected] Guillaume Charloux Centre national de la recherche scientifique UMR 8167, Orient et Méditerranée [email protected] Simon Connor Andrew W. Mellon Research Fellow, Metropolitan Museum of Art [email protected] Laurent Coulon École Pratique des Hautes Études, Université Paris Sciences Lettres (EPHE, PSL), UMR 8546, Paris

[email protected] Rihab Khidir National Corporation for Antiquities and Museums, Sudan [email protected] Yves Egels École Nationale des Sciences Géographiques (ENSG) / Institut National de l’Information Géographique et Forestière (IGN), Paris

[email protected] Florence Dunn Friedman Department of Egyptology and Assyriology, Brown University [email protected] Elizabeth Frood University of Oxford [email protected] Florence Gombert-Meurice Louvre Museum [email protected] Marsha Hill The Metropolitan Museum of Art, New York [email protected] Emmanuel Jambon Institut für die Kulturen des Alten Orients (IANES), Tübingen [email protected]

VIII

LIST OF CONTRIBUTORS

Tobias Krapf University of Basel - Swiss School of Archaeology in Greece, Athens [email protected] Emmanuel Laroze Centre National de la Recherche Scientifique (CNRS), UMR 8167, Paris [email protected] Troels Myrup Kristensen Aarhus University [email protected] Mona Ali Abady Mahmoud Egyptian Ministry of Antiquities, French-Egyptian Centre for the Study of the Temples of Karnak, Luxor, Egypt [email protected] Aurélia Masson-Berghoff Department of Greece and Rome British Museum [email protected] Ernst Pernicka Curt-Engelhorn-Zentrum Archäometrie gGmbH an der Universität Heidelberg, Mannheim, Germany [email protected] Michelle Pisa Graphic artist [email protected] Campbell Price Manchester Museum University of Manchester [email protected] Deborah Schorsch The Metropolitan Museum of Art [email protected] Neal Spencer Department of Ancient Egypt and Sudan British Museum [email protected] Ross Thomas Department of Greece and Rome British Museum [email protected] Kristin Thompson University of Wisconsin-Madison [email protected]

2016 COLLOQUIUM PROGRAMME Wednesday 13 July Session: Meaning and function Towards the contexts of statuary from the Mut Temple in Thebes Betsy Bryan New perspectives on the Gem-pa-Aten colossi Dimitri Laboury with Kate Spence and Robert Vergnieux In temple and home: statuary in the Ramesside colonial town of Amara West, Upper Nubia Neal Spencer A perfect ‘likeness’? Viewing Late Period archaising sculpture in context Campbell Price Limestone figurines and their use in the Egyptian and Greek settlement of Naukratis Ross Thomas Session: Reuse and destruction Who destroyed them and why? Killing Egyptian statues Simon Connor Under and next to the Ished-tree: recent finds from the Egyptian-German Mission in the temple of Matariya/Heliopolis Dietrich Raue The statuary from the ‘dump’: the contexts of the assemblage from the western trench in Canopus Damian Robinson The last statues of ancient Egypt Troels Myrup Kristensen Session: The 2016 Raymond and Beverly Sackler Foundation Distinguished Lecture in Egyptology Colossal and processional statuary in ancient Egypt: Where? When? Why? Christian E. Loeben

Thursday 14 July Session: Becoming invisible: statue caches Small divine statuettes: how did they get to the temple and what happened to them here? Marsha Hill

X

2016 COLLOQUIUM PROGRAMME

The Hierakonpolis ivories: human statuettes from the ‘Main Deposit’ Liam McNamara Taharqo and his descendants: a statue cache at the 5th Nile cataract Julie Renee Anderson with Salah eldin Mohamed Ahmed, Mahmoud Suliman Bashir and Rihab Khidir el Rasheed Looking for contexts: recent work of the Karnak Cachette project Laurent Coulon Ritual burial of a god statue at the Temple of Ptah in Karnak Guillaume Charloux Thousands of Osiris: the archaeological contexts of the bronzes found at the Serapeum and in the temple of ‘Ayn Manawîr Florence Gombert Session: Production and workshop (Re)sources project: raw materials’ origins for countless bronzes Aurélia Masson-Berghoff with Ernst Pernicka Large hollow-cast bronze statues of the Third Intermediate Period: recent investigations John Taylor Ascribing statues to workshops and artists Marcel Marée Accessing the (almost) inaccessible: the study of the making of sculpture in situ Giovanni Verri

PREFACE Aurélia MASSON-BERGHOFF

The exhibition Sunken cities: Egypt’s lost worlds, held in 2016 at the British Museum, prompted new ways of thinking about ancient Egyptian statuary. It displayed an exceptional array of statues – including three colossal sculptures of around 5m in height – mainly from Egyptian museums (Fig. 1) (Goddio and Masson-Berghoff 2016). Most of the statues were discovered in the last twenty years during underwater excavations conducted by Franck Goddio and the European Institute for Underwater Archaeology at Canopus and Thonis-Heracleion, off the western Mediterranean coast of Egypt in Abukir Bay. Others came from earlier discoveries across Egypt. Together they shed light on the encounter of ancient Egypt with Greece and the intermingling of these cultures between the Late Period and the Roman era, and on Osirian myth, cult and rituals.

This volume, which draws upon the 25th Annual Egyptological Colloquium organised at the British Museum in July 2016 and additional papers, embraces statues and statuettes of all aspects in all their complexity. It diverges from the strictly chronological or typological approaches usually favoured in such studies, especially in museum contexts. The contributions feature a multiplicity of statues of various types, scales and materials, dating from the Old Kingdom to Late Antiquity, and recovered from sites across Egypt and Sudan, from the western Mediterranean coast to south of the Fifth Nile Cataract (Fig. 2). The life stories of these sculptures unfolded in temples, of course, but also in palaces, houses and tombs, as well as in secondary contexts (notably the ‘caches’). The multifaceted fate(s) experienced by statues since their production is a particular focus.

Fig. 1: Sculptural masterpieces of the Ptolemaic and Roman periods displayed in Sunken cities: Egypt’s lost worlds.

XII

A. MASSON-BERGHOFF

Fig. 2: Map of Egypt and Sudan showing key sites discussed in the volume. Map: Claire Thorne © Trustees of the British Museum.

PREFACE

XIII

Going beyond typological and artistic discourses, the contributors seek to reconstruct the architectural, ritual, political and historical environments in which the statues were produced, initially set up, reused and/or destroyed. Some papers are based on new material, either recent discoveries of statuary in Egypt and Sudan, or barely known examples preserved in museum storerooms. The recontextualisation of previously excavated statues – as well as recent scientific analyses and the use of modern technology – can provide significant new insights into the production, ancient meaning, perception and (re-)uses of statues, and can offer new paths for research. The endurance of statues and the transformations they underwent are emphasised by a number of papers, often reflecting the strong reactions such artefacts evoke in humans. Statues might be empowered through rituals, and then deactivated; used and reused for centuries, or usurped with deference; ritually mutilated and ‘killed’ to remove their agency, or simply destroyed to demonstrate that they held no power. Whether they elicit devotion or hatred, statues did not in Antiquity, and do not now, leave us indifferent. Statues remain relevant to new contexts (though this is outside the scope of the volume); they still inspire strong emotions and acts of creation or destruction. In 2015 the Rhodes Must Fall campaign ended in the removal of a statue of Cecil John Rhodes from the campus of the University of Cape Town and the defacing or vandalising of several colonial-era statues across South Africa. When statues are no longer in their primary or latest contexts of use, it is up to museums – through permanent displays and temporary exhibitions – to invent new ways for attentive visitors or casual passers-by to engage with these most powerful objects left to us by our ancestors. The volume begins by considering the meaning and function of a statue, its purpose and its ‘voice’, and thereby examining the perspectives of the ancient Egyptians themselves. It then follows the life cycle of a statue, from the raw material and the workshop where it was produced, to its deposition in the archaeological record, be that in its original context of use or a location of reuse, disposal or destruction.

these statues fulfilled practical functions. The papers look into the roles of statues, but also into how ancient Egyptians perceived them as effective agents. The inscriptions on the statues themselves help us to consider their primary purposes and meaning. Elisabeth Frood examines statues’ potential autonomous ‘personhood’ and agency as revealed by inscriptions on New Kingdom private statuary. Statues ‘speaking about themselves’ is a phenomenon well-attested in the later Third Intermediate Period and Late Period, but it was already documented in the New Kingdom and is often underestimated (Meskell 2004; Price 2011). The author analyses examples of statue inscriptions that describe not only the environment in which they were placed, but also their expectations and activities, thereby developing each statue’s own persona. Campbell Price focuses on the archaising statue of Tja-iset-imu (British Museum EA 1682), a Dynasty 26 priest of royal statues, and other similar contemporary sculptures, exploring them in the context of a functioning Late Period temple. In their inscriptions, Saite statues reveal that this archaising style is more than simple deference to the past. Within what would have been at the time a ‘competitive monumental environment’, it represented a strategy to attract the attention of passersby and of the gods themselves. Rather than being portraits bearing a resemblance to their owners, they were likened to the statues of ancestors, whose longevity and durability those owners desired to emulate. Drawing upon known religious practices of the 1st millennium BC, Marsha Hill offers some considerations regarding sacred bronze statuettes, which were produced in large quantities during the Late and Ptolemaic periods. She debates the perceptions of these statuettes as mainly votive objects stemming from individual donations and forming testimonies of personal piety, or as material evidence of the elite and sacerdotal classes participating in rituals, notably during festivals. Rather than a popular or elite practice, most of these bronzes could have fulfilled more ritual roles, maintaining the temple cult and functioning within it.

Meaning and function

Complementing typological and stylistic enquiries, scientific studies and new technologies can be relevant to discussions on workshops and the production of statues. They can also help greatly in reconstructing the original appearance of statues. For example, techniques

Moving away from our usual perception of Egyptian statues as pure artworks, the first part of this volume reminds us that in their ancient and original settings

Production: technology and workshops

XIV

A. MASSON-BERGHOFF

investigating surface treatments can identify polychromy, with which statuary was frequently embellished (Fig. 3). The identification of raw materials used in statuary production can inform us about workshops: their location, scale, and procurement of materials. Aurélia Masson-Berghoff and Ernst Pernicka combine the vast corpus of chemical analyses of sacred bronzes of the Late Period with recent lead isotope analyses to discuss the extent to which workshops relied on bronze scrap, and the origins of the raw metals (copper and lead) used in the manufacture of the sacred bronzes. This investigation also has the potential to identify the types of workshops involved in the production of such bronzes, and leads to some broader historical and economic implications. Among available modern technologies, 3D-printing of artefacts has already been used in museums to inform and engage with visitors. For example, the

magical trappings of Tamut’s intact cartonnage case, revealed by CT-scan, were 3D printed for the exhibition Ancient lives, new discoveries: Eight mummies, eight stories (Taylor, Antoine and Vandenbeusch 2014, fig. 69). Such methods can also be of direct help in research, as Florence Dunn Friedman, with assistance from Michelle Pisa, demonstrates. Through the technologies of laser-scanning and 3D printed reproductions, three loose fragments kept in the Museum of Fine Arts can now be securely attributed to at least one dyad similar in size to the one representing king Menkaura and a queen (Museum of Fine Arts, Boston 11.1738). Instead of solid 3D-printed reproductions, only surface printing (in a non-abrasive material) was undertaken, in order to match the fragments safely and accurately. Such research enables a new hypothesis to be developed regarding the wider statue programme in the valley temple of Menkaura, which contained series of similar types.

Fig. 3: Joanne Dyer (Department of Scientific Research at The British Museum) observing painted and gilded details of the limestone statuette EA 68814 from Naukratis, under a digital microscope. The silhouette of a wedjat eye in the small back of the figure is marked in red ochre while the amulet itself was probably originally gilded. Photograph credit: Joanna Fernandes © Trustees of the British Museum.

PREFACE

XV

Fig. 4: Colossal statue of Psamtek I discovered in 2017 in Heliopolis. © Heliopolis Project (Aiman Ashmawy/Dietrich Raue/Kai-Christian Bruhn), photograph by Dietrich Raue. With thanks to the Ministry of Antiquities (Egypt) and the Ägyptisches Museum der Universität Leipzig.

Visible statues: temples, palaces and houses Placing statues in their initial context of use is essential in order to define how they were used during their life and how people interacted with them, and to appreciate the visual (or even emotional) impact they had on the many or the few who had access to them. Primary use-contexts include divine cult statues within their shrines, where they were worshipped, censed, nourished, dressed and addressed. Furthermore, some left the temple to be paraded in processions during festivals, where they might answer oracular questions, or encounter another divine or royal statue in order to partake in a sacred union or mutual regeneration. In other contexts, immovable colossal guardians watched over the entrances of temples, in which accessible statues of gods and deified men acted as benevolent intermediaries between deities and humankind. Statues of the elite

1

I would like to sincerely thank Aiman Ashmawy and Dietrich Raue who lead the Heliopolis Project for allowing us to publish these illustrations.

in their funerary chapels survived on images of banquets; while wishes for fertility and protection for the most vulnerable were placed on statuettes in ordinary homes. An impressive colossal statue of Psamtek I was discovered in 2017 at Heliopolis (Fig. 4; Ashmawy and Raue 2017, 40–1).1 The roughly 8m-tall statue in quartzite is the only known colossal depiction of the Dynasty 26 pharaoh (Fig. 5). It calls to mind the words of Herodotus, who mentioned seeing such colossal statues at Heliopolis (Histories II, 153). Psamtek’s statue once abutted the façade of a Ramesside temple with a courtyard populated with other colossal statues of pharaohs, dyads depicting kings and deities, and other sculptural works (Ashmawy and Raue 2017, 35–6). It is easy to imagine how a cluster of such masterworks must have awed not only the Egyptian population, but also foreigners such as the Greek mercenaries recently enrolled in Psamtek I’s army. The influence of Egyptian statuary on Greek archaic art, notably its colossal sculptures, could have come from such encounters (Ashmawy and Raue 2017, 41). Analysing statues within their original or secondary contexts of use can illuminate the political agenda

XVI

A. MASSON-BERGHOFF

behind a statuary programme. Neal Spencer’s complete survey of all types of statuary uncovered in Amara West – with their related context of discovery and an interpretation of their functions – allows the author to compare that site with other (New Kingdom) Upper Nubian settlements, mainly religious complexes. Attempts to interpret the relative rarity of statuary dated to Dynasty 19 – especially when contrasted to Dynasty 18 temple sites, but also in comparison to contemporary Lower Nubian and Egyptian sites – raise the question of the extent to which the actual provision of statues correlated with the Pharaonic state’s intention, or indeed with modern conceptions of the appearance of ancient towns and temple precincts. Depictions of statues within the ancient buildings they were set are able to shed light on their appearance and function. The exploitation of such representations is not always straightforward, which is in itself revealing. Kristin Thompson addresses information gained from, and challenges posed by, the recontextualisation of statues from Tell el-Amarna based on their depictions. A variety of reliefs – notably from tombs, stelae and temples – helps in better defining the placement and context of use of some statues within buildings, such as the Great Palace, Great Aten Temple and Small Aten Temple, as well as Kom el-Nana. Comparing reliefs depicting buildings with statues to actual finds from those buildings, she explores several avenues of interpretation as to why so few statues are depicted on the otherwise rather detailed reliefs, given that archaeological discoveries indicate that a wealth of statuary existed in reality. Even if statues are no longer visible, it is possible to identify their initial placement and even their types from the shadows or marks they leave on temple walls or pavements, and a number of extrapolations on their role can be suggested from there (as discussed by Christian Loeben in the keynote lecture he delivered at the colloquium on Theban statuary; see also Loeben 2001). Tobias Krapf identified no less than thirty traces of statue placements on the floor of the Ptolemaic to Roman courtyard in the temple of Khnum at Elephantine. Analysis of the traces, combined with the recent study of hundreds of statue fragments and statue bases – dating from Dynasty 18 through the Roman period and including a wide range of divine, royal and private types – conveys the picture of a temple courtyard crowded with an assortment of statues (some reused), amid cultic equipment, architectural elements of all sizes and trees. The central axis, used for

Fig. 5: Reconstruction of the colossal statue of Psamtek I, with human figure for scale. © Heliopolis Project (Aiman Ashmawy/Dietrich Raue/ Kai-Christian Bruhn), illustration by Christopher Breninek, with thanks to the Ministry of Antiquities (Egypt) and the Ägyptisches Museum der Universität Leipzig.

processions and other ritual activities, would have nonetheless imposed some order. A range of religious or ritual practices can be inferred from the recontextualisation of statuary. Reappraising the types and contexts of discovery of about 350 Late Period Egyptian and Cypriot stone statuettes at Naukratis, Ross Thomas explores the religious practices of Egyptian and foreign communities living in the international harbour town. These contemporary stone statuettes emerged from two very different traditions: one Lower Egyptian, and another from a Cypro-Ionian community. According to analyses of distribution patterns across the site, Cypriot figurines appear to be exclusively dedicated within ‘Greek’ sanctuaries, while Egyptian statuettes are mainly found in domestic contexts and sometimes within religious settings.

PREFACE

Becoming invisible: statue caches Countless statues were buried beneath the sacred grounds of temples and other religious complexes. The fourth part of the volume discusses the varied circumstances, occasions, and processes leading to and involved in the creation of statue caches in ancient Egypt and Sudan. One traditional way of seeing these caches is as a collection of votive objects that have reached the end of their life-use. In some cases, this perception is becoming increasingly challenged and nuanced, notably thanks to contextual reassessment. One such assemblage yielded some of the earliest known examples of Egyptian statuary. This is the largely under-studied corpus of ivory statuettes from the ‘Main Deposit’ in Hierakonpolis, excavated by J. E. Quibell and F. W. Green in 1897–8. Liam McNamara reinterpreted the archaeological context of the ‘Main Deposit’ material, not as a cache of discarded votive material placed within a temple complex, but as related to royal ceremonies and the ideology of kingship in the Late Predynastic and Early Dynastic periods (McNamara 2008).2 Most of the objects found within the ‘Main Deposit’ are royal in nature (e.g. Fig. 6) and the original context seems to have been not a temple, but an arena used for royal ceremonies by successive kings. Statue caches are numerous and far from uniform in nature. Guillaume Charloux and Mona Ali Abady Mahmoud offer an archaeological perspective on them, and establish a typological classification of caches based not purely on their content, but rather on their archaeological context. They define ‘sacred’ caches as those in which the contents had received a liturgical burial. Such sacred caches constitute only a fraction of the total number of caches of sacred artefacts discovered in the ancient world. Since their ritual intent is paramount, however, their archaeological examination helps, to a certain extent, in reconstructing some of the gestures and rites carried out in conjunction with burials of statuary. Some of the most common statue caches created in 1st millennium BC Egypt involve dozens, hundreds, or even thousands of bronze statuettes. Instead of the traditional typologies, Florence Gombert-Meurice

2

Study based on unpublished Liam McNamara DPhil thesis: ‘Holy rubbish’? Ivory statuettes from the Hierakonpolis Main Deposit (University of Oxford).

XVII

proposes an alternative presentation and study of such assemblages. She uses the discovery of a deposit of statuettes in their original context, a collapsed chapel in the Temple of Osiris-Iw at ‘Ayn Manawîr, to elucidate the large find of sacred bronzes from the Serapeum made in the 19th century. The analysis of recurring types of Osiris bronze figures made from the same mould, and of the staging of both deposits, reveals a lost liturgy. It seems to relate to a theology operating during specific religious performances, rather than to economic reasons or personal piety. The author lists a variety of criteria for deposits of sacred bronzes which eventually could help to draw up a catalogue of festivals and local religious practices and beliefs, replacing the more traditional classification of bronze types.

Fig. 6: Hippopotamus ivory statuette wearing the cloak of the sed-festival (AN1896-1908.E.326) from the Hierakonpolis ‘Main Deposit’. Photograph by Liam McNamara. © Ashmolean Museum, University of Oxford.

XVIII

A. MASSON-BERGHOFF

The Cachette of Karnak, one of the most famous caches ever uncovered in Egypt, has been recently reassessed (Coulon 2016). L. Coulon, Y. Egels, E. Jambon and E. Laroze summarise the scientific exploitation of several historiographical surveys of the Cachette. This large repository of thousands of statues in stone, bronze and wood alongside other artefacts was created in a single event. The authors briefly discuss the dating and the raison d’être of the cache, as well as the primary context of the statues ritually buried in the Cachette, advocating the use of digital epigraphy to help in such a massive endeavour. The heart of their paper gives the results of experimentation with photogrammetry, combined with archival photographs and plans and Legrain’s notes on the Cachette’s excavation. This process helped create a complete textured 3D model of the context, gaining new spatial information and reconstructing the stratigraphy of the Cachette, which was 15 to 16m deep. The gathering of statues in a cache – alongside temple furniture, or not – is sometimes explained as a consequence of war, carried out either to safeguard such precious and sacred artefacts, or to dispose of disfigured, broken statues following a temple’s destruction. For instance, the creation of caches containing statues of Dynasty 25 rulers at Gebel Barkal and Kerma/Dokki Gel is generally associated with Psamtek II’s military campaign in Nubia (e.g. Bonnet 2011). The recent discovery of a cache of statues depicting the same group of early 7th-century BC Kushite kings much further south, in the Amun temple in Dangeil upstream of the Fifth Nile Cataract, raises some debate. Julie Renee Anderson, Salaheldin Mohammed Ahmed, Mahmoud Suliman Bashir and Rihab Khidir elRasheed consider these groups of statues within the wider historical and political, as well as cultic and cultural, contexts. The disposal of these statues could be linked with religious or ritual changes – such as a temple refurbishment in the Meroitic period and the transfer of the focus of the cult from the divine statues of the ruler to those of gods – rather than attributed to external or even internal political affairs. Afterlives: reuse and destruction The last group of papers explores the manifestations of, and possible reasons behind, the reuse, transformation and destruction of statuary. Alterations to and usurpation of stone statuary, reflecting religious and political shifts or trends, are

well-attested and widely discussed (e.g. Sourouzian 1998; Connor 2015). Metal statues were not spared such practices, although evidence of reuse and transformation seems rather scarce or, at least, not easy to identify without using scientific methods of investigation, such as X-ray radiography and compositional analyses. Deborah Schorsch presents several such examples that include royal and private metal statuary, as well as sacred effigies of gods, that span from the Old Kingdom to the Late Period. In each case, possible reasons behind their modifications and reuse are put forward. The decline of statue production and the complex afterlife of statues in Late Antique Egypt are considered by Troels Myrup Kristensen. Only a handful of new statues are known to have been produced between the late 3rd and early 7th century AD, and those were restricted to specific contexts. This situation dramatically differs from other Mediterranean regions, as indicated by the recent comprehensive survey of Late Antique statues by Smith and Ward-Perkins (2016). In Late Antiquity, the population was still surrounded by ancient Egyptian statuary, and their re-display, transformations, destruction or simple neglect happened at a time of critical religious and social changes. The various modifications and damage that statues often sustained after they were first set up are reviewed by Simon Connor, covering the Pharaonic and subsequent Ptolemaic, Roman, Christian and Islamic periods. When these changes to and/or mutilations of statues were not caused by accident, the causal factors proposed to account for them allow for a wide range of interpretations. Actions may be religiously or ritually motivated, but may also have political (e.g. Arnold 2005) or economic causes. A careful observation of the types and places of these marks of transformation or destruction, associated to the archaeological, historical and cultural contexts, can help us to understand which explanations are the most likely. Acknowledgements It is a pleasure to thank here the participants to the colloquium and the contributors of the additional papers for sharing their latest research, reflections and eagerness for this kind of approach to ancient statuary. Many colleagues in the Department of Egypt and Sudan were instrumental in the preparation and organisation of this event, foremost co-organiser Daniela Rosenow, but also Claire Messenger, Eirini Koutsouroupa, Tania Watkins, Sylwia Janik and Evan York, whose assistance was

PREFACE

greatly appreciated. Thanks are also due to colleagues at the British Museum and other leading academic researchers who chaired the sessions, as well as to Steven Aucott and Alexander Myers of the AudioVisual unit for their technical support on the occasion of the colloquium. I am particularly grateful to Neal Spencer, Marcel Marée and Elisabeth O’Connell for their valuable advice on the content of this volume. My special thanks are extended to Carolyn Jones for her careful copyediting of the papers and Claire Thorne for the map accompanying this introduction. Last, I would like to acknowledge the team at Peeters, Leuven, especially Bert Verrept, for their efficiency and flexibility. Bibliography Arnold, Do. 2005. The destruction of the statues of Hatshepsut from Deir el-Bahri. In C. H. Roehrig, R. Dreyfus and C. A. Keller (eds), Hatshepsut: From queen to pharaoh. New York, 270–6. Ashmawy, A. and Raue, D. 2017. Héliopolis en 2017: Les fouilles égypto-allemandes dans le temple du soleil à Matariya/Le Caire. Bulletin de la Société française d’égyptologie 197, 29–45. Bonnet, C. 2011. Les destructions perpétrées durant la campagne de Psammétique II en Nubie et les dépôts consécutifs. In D. Valbelle and J.-M. Yoyotte (eds), Statues égyptiennes et kouchites démembrées et reconstituées. Paris, 21–32. Connor, S. 2015. Quatre colosses du Moyen Empire ‘ramessisés’ (Paris A 21, Le Caire CG 1197, JE 45975 et 45976). Bulletin de l’Institut français d’archéologie orientale 115, 85–109.

XIX

Coulon, L. (ed.). 2016. La Cachette de Karnak. Nouvelles perspectives sur les découvertes de Georges Legrain. Bibliothèque d’étude 161. Cairo. Goddio, F. and Masson-Berghoff, A. (eds). 2016. Sunken cities: Egypt’s lost worlds. London. Loeben, Ch. E. 2001. Beobachtungen zu Kontext und Funktion königlicher Statuen im Amun-Tempel von Karnak. Leipzig. McNamara, L. 2008. The revetted mound at Hierakonpolis and early kingship: A re-interpretation. In B. Midant-Reynes and Y. Tristant (eds), Egypt at its origins 2. Proceedings of the international conference ‘Origin of the state. Predynastic and Early Dynastic Egypt’, Toulouse (France), 5th–8th September 2005. Leuven, 901–36. Meskell, L. 2004. Object worlds in ancient Egypt: Material biographies past and present. Oxford. Price, C. 2011. Materiality, archaism and reciprocity: The conceptualisation of the non-royal statue at Karnak during the Late Period (c. 752–30 BC). Doctoral thesis, University of Liverpool. Smith, R. R. R. and Ward-Perkins, B. (eds). 2016. The last statues of antiquity. Oxford. Sourouzian, H. 1998. Les statues colossales de Ramsès II à Tanis: Un colosse fragmentaire quartzite, remploi du Moyen Empire (blocs n°1471–1486). In Ph. Brissaud and Ch. Zivie-Coche (eds), Tanis: Travaux récents sur le Tell Sân el-Hagar 1. Paris, 391–419. Taylor, J. H. and Antoine, D. with Vandenbeusch, M. 2014. Ancient lives, new discoveries: Eight mummies, eight stories. London.

I MEANING AND FUNCTION

WHEN STATUES SPEAK ABOUT THEMSELVES1 Elizabeth FROOD

Abstract Three New Kingdom statues are analysed — the Louvre statue of Maanakhtef (E 12926), the Cairo statue of Minmose (CG 1203), and the Cairo statue of Paser (CG 630) — in order to study how object agency and multiple aspects of the person are mobilised through their inscriptions, as well as how those texts exploit the statues’ forms. Selected extracts of the texts are presented in translation and discussed, together with descriptions of the statues. It is argued that the ways in which these texts — as well as others touched on here as comparanda — describe their own environments and their own bodies present complex explorations of concepts relating to image-making, audience, and personhood. * * * When do statues speak about themselves? The title of this essay could seem to be a misstatement, as the answer is arguably ‘all the time’ for Egyptian statues of humans bearing texts. Being inscribed with a name generates an identity, an offering formula generates expectations of interactions with gods and people, including the deified dead, and many statues allude to places of dedication through the gods they address or the spaces that they name. The statues I discuss in this chapter, however, speak about themselves more specifically, describing their surroundings, desires, activities, and their bodies in ways that raise basic issues of object agency and aesthetics. I focus on statues from the New Kingdom, especially the late New Kingdom, when examples become more frequent, explicit, and elaborate.

1

I am grateful to Aurélia Masson-Berghoff and Neal Spencer for their invitation to contribute to this volume, and to Aurélia for her very helpful editing of the text. John Baines read and commented on drafts at different stages and was a wonderful sounding board for different ideas. Thank you also to Campbell Price for sharing his doctoral thesis with me, and to Vincent

One of the functions of representations of elite human or divine forms, in two or three dimensions, is to act as a vessel for non-physical aspects of the person or god, allowing the individual to engage in a range of social transactions and interactions, including participation in rituals, especially the offering cult (e.g. Meskell 2004; Robins 2005; Riggs 2014, 94–8). Partly through their association with the Opening of the Mouth ritual (Fischer-Elfert 1998), human and divine statues are receptive to analyses of their agency, their capacity to ‘exert force on the world’ (Winter 2010, 307). Discussions have been influenced by the anthropologist Alfred Gell’s posthumous 1998 study of forms of agency ascribed to objects in different settings (e.g. Meskell 2004; Price 2011a). An object can become a ‘social being’, so that it ‘does not just reflect the owner’s personhood, it has personhood’, and objects can ‘appear as “agents” in particular social situations’ (Gell 1998, 18). Gell’s theory was developed for less specifically personal objects than statues, such as Kula valuables, those that could potentially have different ‘owners’ or, as with divine images, relate closely to many people. Artefacts that acquire meaning through association with, and representation of, one person extend agency and interaction in other directions. For non-royal statues, this potential is expressed in multiple, intersecting ways: through form and pose (such as open palms signifying address to some entity); attributes held (such as naoi and divine statues); and inscriptions. Annette Kjølby (2007; 2009) has extended Gell’s models productively for non-royal temple statues, examining the social and economic implications of statue-making in the New Kingdom, including form, material, place, and features of inscriptions, on which I focus in this article.

Rondot, Audrey Viger, and Roxane Bicker for figure permissions. I am fortunate to work with a brilliant team of research assistants: Christelle Alvarez, Julia Hamilton, Ellen Jones, and Chiara Salvador all contributed fundamentally to clarifying my ideas and getting this into its final form.

4

E. FROOD

Irene Winter (2010, 317), in her critique of Gell’s model in relation to Mesopotamian statues, argues that we should untangle the hierarchy from the ‘inherently agentive’ (the statue commissioner, referent) to the ‘delegated agent’ (the statue). I attempt here to initiate this untangling by examining how some texts about statues that are inscribed on those same statues express potentials for something like an autonomous ‘personhood’ and agency, making the objects into more than delegated agents. By considering the work’s own conceptualisation of itself, we can start to nuance our ideas of how it performed its creation, presence, and social relations. I build upon my suggestion that some Third Intermediate Period statue texts which describe their own sensory environments could be considered ‘performative statue biographies’, creating identities that are distinct from that of their owner (Frood 2013, 182– 4). Such inscriptions can perhaps be more usefully characterised as ekphrasis. Ekphrasis, literally ‘speaking out’, is a rhetorical description of a work of visual art that enables the object to speak about itself — as with these statue inscriptions — or to be spoken about. It is often considered to have its origins in the earliest Greek literature, Homer’s Iliad and Hesiod (e.g. Francis 2009; also Bartsch and Elsner 2007). The phenomenon, however, is much older and more widespread; only the term can be seen as originating in the Greek world. My use of ekphrasis as a term draws in particular on Zainab Bahrani’s analysis of Mesopotamian contexts. She focuses her discussion around 3rd and 2nd millennium BC temple votive pieces, arguing for ‘its origins in this very ancient votive practice, which originally enabled the image both to reflect on its own status as representation and to participate in the space of life’ (Bahrani 2014, 13 and ch. 7). Her analysis builds on her work on the status of images and the meaning of ṣalmu, ‘image’, as ‘an entity in its own right, a being rather than a copy of a being’, stressing that an image can be ‘an ontological category rather than aesthetic concept’ (Bahrani 2003, 125, 133). Ekphrasis extends this usage, with the artwork articulating its own ontological status, encompassing aesthetic dimensions. Here, the works of Jesper Svenbro (1993) and John Ma (2007; 2013) on dedicatory and honorific captions on Archaic and Hellenistic statue bases (respectively) are important. Whereas Bahrani unpacks philosophical implications of image-making and substitution, Svenbro (esp. chs 1–2) uses a first-person inscription on the

base of a 6th-century BC statue of a young woman as a starting point for a detailed examination of the implications of object voice and its interplay with form, presence, and audience. Ma in turn builds on some of these ideas to examine how syntactical distinctions in captioning present different social, political, and spatial meanings. Egyptian statue texts are much longer and more fully integrated with the body and other iconography than the examples Svenbro and Ma treat, while displaying a comparable potential to communicate their place — their simultaneous ‘hereness’ and ‘elsewhereness’. I examine texts on some New Kingdom statues that speak about the statue. In these texts, a range of voices is found, from third-person descriptions or statements about the statue to the first-person voice of the statue itself, as well as, more rarely, a second-person address to the statue. Such treatments are highly developed in 1st millennium statue inscriptions: temple statues sniff the pungent air (Frood 2013), even complaining when food offerings start to rot and stink (Rizzo 2004); others listen to the singing of priests; and some shabtis yell ‘I am here’ (Perdu 2000–1). So it is not surprising that significant discussions of the implications of statue texts focus on that period. In particular, Campbell Price (2011b) offers a full and detailed study of the conceptualisation of Late Period statues from Karnak, including function and emplacement in his integrated analysis of inscription and monument. I offer preliminary thoughts on some earlier examples, suggesting that such texts not only mark the statue’s agency but also play upon the relationship between the statue and the person it represents. They assert the statue’s separate identity (with Clère 1968, 148), beyond ascriptions of agency or notions of substitution. It is this idea of separateness, of ‘statue-ness’, that may enrich discussions of the materiality and potentials of Egyptian personal monuments. Statue texts The amount of text inscribed on a statue’s surfaces tends to increase and diversify over time. The majority consist of offering formulas concerned with, for example, the maintenance and provision of the offering cult and the owner’s transfiguration in the next world, as well as requests for long life and good old age. Thus, although specifics are often incorporated — lists of temple offerings and festivals on temple statues (e.g. Assmann, Bommas and Kucharek 2005, 326) — many

WHEN STATUES SPEAK ABOUT THEMSELVES

5

texts are little different from those found in other monumental non-royal self-presentations, such as those inscribed in tombs and on stelae. In some sense, they can be ‘read away from statues’ (Ma 2007, 212); they do not engage with the character of a statue as a bearer of text and a manifestation of individual presence, in the process very deliberately eliding distinctions between statue and referent (with Kjølby 2009, 46). The use of the first person in these texts in particular blurs these distinctions. A Dynasty 18 black granite block statue of the steward of Memphis and of the treasuries Amenhotep, from the temple of Osiris at Abydos (Fig. 1), includes the following address to Osiris on the front of the knees: I have come before you, lord of the district of Poqer, that I may make supplications to your ka every day, that I may give you praises, so that you may be exalted [...] so that I may be granted a voice offering of bread and beer, cattle and fowl, alabaster and linen [...] receiving fillets in the wag-festival with green and red linen, that I may process (with) the neshmet-barque as an efficacious ba, so that I may not be separated from the following of Horus. (Hall 1914, pl. 38)

Some of these statements are relevant to both statue and referent, such as the receipt of offerings and being in the following of a god, that is, participating in processions (Schulz 1992, II, 707–8). Others seem to refer more specifically to the desired activities of the transfigured person. The multiple intersections of voice in Egyptian monuments have often been noted (e.g. Assmann 1972), but they may not receive as much attention as they deserve (see also Nyord 2017, esp. 344–6 with n. 42). As John Baines observes (1999, 37) in connection with tomb biographies, the use of the first person is a way of ‘fictionalizing the whole while rendering it immediate and vivid’. We accept that the image we see, the words we read, somehow are those of the person, and yet we know that they are not. In the case of the Amenhotep cited above, the statue’s presence and provision in the temple is linked with the transformation of the individual in the next world, and the first-person voice ensures that any distinction between the two is fuzzy (with Rizzo 2004, 518; cf. Kjølby 2009, 42). In contrast, when statues are described, spoken to, or have their own voice in the texts inscribed on them ‘the fact

Fig. 1: Black granite block statue of Amenhotep from Abydos. Reign of Amenhotep III, Dynasty 18. H. 73.7cm. London, British Museum EA632. © Trustees of the British Museum.

of representation’ is underlined (Ma 2013, 29–30) — their statue-ness is made doubly present. Below I treat some textual and visual puzzles of statue self-reference, puzzles that we generally resolve without thinking about or simply ignore (Ma 2007, 206; 2013, 18). My focus here is the New Kingdom; comparable ideas were expressed, in somewhat different ways, in earlier periods. Shabti figurines and the complex networks of relations and substitutions their texts generate are an obvious example (Nyord 2017). Captions on some Middle Kingdom temple statues recording their status as royal gifts refer to themselves both obliquely

6

E. FROOD

and directly (e.g. Verbovsek 2004, 190–2). An inscription on the papyrus roll in the lap of the Dynasty 13 scribe statue of the vizier Iymeru set up in Karnak follows the individual’s titles with the statement that ‘(this statue) was made at the command of the Dual King…; the gift of a statue [of this companion] (to be) in the following (ḥr šms), in position before (r bw ẖr) His Person l.p.h. in the temple of Amun-Re, Lord of the thrones of the Two Lands, foremost of Karnak’ (reading following Franke, cited in Kubisch 2008, 321; cf. Verbovsek 2004, 380–1, 384–5). This donation record is written as if read by the statue — it reads out the fact of its own existence (compare Frood 2007, 187–8 with n. 100). It also seems to describe its own position seated beneath a statue or image of its king and donor. Ekphrasis and presence I start by returning to the block statue of the royal butler under Amenhotep II, Maanakhtef (Fig. 2: Bisson de la Roque 1927, 108–9; Drioton 1927, 49–51; Frood 2013, 182). The statue, of diorite and about half lifesize (50cm high), was found in a secondary deposition in the temple of Montu at Medamud. It bears traditional offering formulas addressed to many deities in a tabular format on the front of its knees, and a series of vocative addresses concerning Maanakhtef’s provision in the next world on its left side. The text on the right is addressed by Maanakhtef to the temple itself, rather than the god or priests: O great court of Montu [which is in front of?] its lord, may you cause that this twt-statue of the royal butler Maanakhtef be firm (rwḏ) inside the festival court, that it may breathe (ssn.f) myrrh and incense on the flame, that it may partake (pnq.f) of water flooding (from) the altars onto the ground of the court, that it may eat (wnm.f) from what is in the hands of the wab-priests from the remainder of the divine offerings, and see the morning sun in the domain of the one who traverses eternity, that it may follow its god when he circulates through his temple in his festival of that sacred mountain, just as I did when I was on earth. (Urk. IV, 1483, 7–17; Barbotin 2005, 158)

Fig. 2: Diorite block statue of Maanakhtef from Medamud. Reign of Amenhotep II, Dynasty 18. H. 50cm. Paris, Musée du Louvre E 12926. © Musée du Louvre, Dist. RMN-Grand Palais/Christian Décamps.

My translation here of the third person masculine singular pronoun with ‘it’ is an artificial way of suggesting that the text is in large part a description of the statue itself as much as the indwelling, manifest presence of Maanakhtef, a fuzziness of reference comparable to that of the first person noted above. Like the Third Intermediate Period examples, the ‘sensorium’ of the statue is opened up (Winter 2010, 308); the statue, bathed in sunlight, breathes, receives libations, sees and participates in procession. In this it is not unusual: many New Kingdom statues ask that the statue itself do these things (e.g. Urk. IV, 1939, 1–10, and see below). But the descriptions in Maanakhtef’s text are height-

WHEN STATUES SPEAK ABOUT THEMSELVES

ened and extended. The myrrh and incense are thrown on the flame to burn, capturing the moment when scent is most intense. Water gushes across the floor of the court such that you can almost feel its coolness in contrast to the heat of the sun. The nautical associations of pnq ‘bailing’ (e.g. Parkinson 2012, 102–3) may mean this is less about drinking and more about water pouring over your surface. The statue munches offerings. Mention of its being in the following of its own god (nṯr.f: Urk. IV, 1483, 15) can be compared with Ramesside inscriptions in which block statue and cult statue speak to each other (see below). This passage also highlights the parallels between them: the cult statue stands in relation to its god just as the statue of Maanakhtef does to its owner. Later statues are known where one side of the body refers to the owner’s transfiguration and the other to the statue (e.g. Luxor J141: Clère 1995, 87–92, pls 6–7; Copenhagen AEIN 584: KRI, III, 142, 3–15; KoefoedPetersen 1950, pl. 78). This patterning makes play with the distinction between person and statue, countering ephemeral with permanent, the human with the monumental, as is explicit in the final triplet comparing the statue’s experience with the individual’s ‘while on earth’ (reading with Barbotin 2005, 159, contra Helck: Urk. IV, 1483, 17). That Maanakhtef’s speech is an appeal to the court of the temple itself is unusual, reinforcing this sense of monumental presence in monumental space. Although I know no other example from before the Late Period (Price 2011b, 181), an address to high officials on a Dynasty 12 stela from Abydos of the chief priest Wepwawetaa (Fig. 3) asks that ‘you say a thousand of bread and beer, cattle and fowl, alabaster and linen to the temple of Ra’. The phrase r rꜢ-pr n rꜥ is arranged in a small column within line seven and is first of a list of gods in this tabulated format. rꜢ-pr could refer to a physical space or to this community of gods. Wepwawetaa’s inscription also closes with a statement describing the establishment of his statues (twt determined with ) in the temple and their offerings; these latter clauses are again organised in columns. This emphasises that the whole refers to temple environments. Maanakhtef’s appeal is more explicitly about location. The statue had probably been set up in this ‘great court’ or ‘festival court’, an area which Felix Relats Montserrat (pers. comm.) has identified as newly constructed under Thutmose III, perhaps a generation earlier. The statue mobilises the court as a protective place

7

that is positioned to capture the morning sun when it is busy with (hopefully) generous priests. A personal monument speaks to a space and confirms the physicality, and autonomous agency and presence, of both. In this respect, the text is an instance of ekphrasis: ‘even at its most visual … [ekphrasis] finds itself straying to the evocative resonances of the other senses: sound, smell, taste, and touch’ (Bartsch and Elsner 2007, ii). The visualised presence is turned out towards the environment, as in many of the statue texts I discuss. Relatively few statue texts from before the 1st millennium turn inward to describe their own bodies in a developed way. Texts about statues Most non-royal statues from before the Third Intermediate Period that talk about their own actions or presence are more allusive or prosaic than Maanakhtef’s. Short formulas relating to statue presence and function include, for example, a caption that is often prominently sited, especially down the centre of the robe of standing or seated poses, requesting that ‘all that comes forth from the offering table’ of a god be for the individual, or in some cases for his statue (e.g. Weber 2014, fig. 12, 122). These are condensed confirmations of a primary function of statues, to participate in the reversal of offerings (Rizzo 2004), and the locations in ideally more visible areas relates to involvement and interactions with audiences. These often elliptical texts alluding to statue presence can be compared with those that refer to it specifically. The most basic are short requests in addresses or offering formulas, asking that the statue endure (mn) and/or, like Maanakhtef’s, be firm (rwḏ) in the temple or tomb so that it may receive offerings and see the god, or enjoy other comparable benefits. Such references to the statue on its surface are attested from the 3rd millennium onward (e.g. Verbovsek 2004, 315–6, 444–6; for Dynasty 18, Barta 1968, 100, no. 153; 127, no. 153). They are particularly frequent, and occasionally highly elaborate, in the Ramesside period. The back pillar of a kneeling statue from the Karnak Cachette of the high priest of Amun Paser, who was also vizier under Sety I and Ramesses II, bears an offering formula addressed to Amun-Ra-HorakhtyAtum requesting ‘that I may place my twt-statue enduringly and permanently (mn rwḏ), seeing Amun daily’ (KRI, III, 293, 1–2). These requests are closely related to the common one that the owner’s name remain in the

8

E. FROOD

Fig. 3: Limestone stela of Wepwawetaa from Abydos. Red boxes indicate 1) the start of the tabulated list; 2) the determinative for ‘statues’. Reign of Amenemhat II, Dynasty 12. H. 136cm. Munich, Staatliches Museum Ägyptischer Kunst Gl. WAF 35. © Staatliches Museum Ägyptischer Kunst, Photograph: Marianne Franke.

WHEN STATUES SPEAK ABOUT THEMSELVES

temple or tomb, as on another statue of Paser: ‘May you (Amun-Ra) cause that my name be firm (rwḏ) in your temple, that I may receive offerings from your presence and smell the incense of your giving’ (Van Dijk 1993, 122–3, n. 44; KRI, III, 19, 1). Some of the most distinctive examples of such phraseology are on black granite and limestone block statues of the late Dynasty 19 high priest of Amun, Roma, also called Roy, from the Karnak Cachette (Frood 2007, 48–54). The biographical text on the left side of the body of the black granite statue culminates with a request that the king, or possibly Amun, ‘place my ẖnty-statue enduringly (mn) for eternity on the land of his domain perpetually, that Amun may address (wšd) it in his every festival, my name enduring (mn) upon it perpetually’ (KRI, IV, 209, 1–2). An appeal to the living on the back pillar demands that temple staff give it garlands and perform offering rituals for it (KRI, IV, 209, 11–4). Almost every panel of text on the limestone statue refers to the statue. That on the front asks that Amun and his entourage ‘place my ẖnty-statue enduringly and permanently, resting in Karnak forever’ (KRI, IV, 129, 4–5), while that on the right side requests that Amun ‘cause that my name endures upon my ẖntystatue’ (KRI, IV, 130, 9). The offering formula on the back pillar focuses around the life of the statue in the temple, requesting that: my ẖnty-statue endures on earth, my name being visibly carved2 upon it for eternity […] that Amun may address (wšd) it each time he appears, and Mut and Khonsu assent (hn) to it more than the great ones. (KRI, IV, 130, 13–5)

These ‘great ones’ (wrw) probably make a selfdeprecating, yet competitive, comment on the presence of Roma’s statue in a group of other non-royal statues. The language of these verses, especially hn, ‘to assent’, refers to oracular practices (cf. Frood 2007, 184; and Clère 1995, 188–9 where hn also describes statue interactions). The ‘great ones’ could at the same time be living officials participating in oracular consultations

2 3

4

ḫtj has determinatives of an eye and a blade. The Karnak Cachette database includes excellent photographs: [last accessed 6/7/2018]. Clère 1995, 119–23, esp. n. a and pl. 15c; Price 2011b, 160. Such usages may be relatively widespread. An appeal on a naophorous statue of Hormin, overseer of the Memphis jpt under

9

(cf. Clère 1995, 182–3), incorporating the statue as one actor among living human actors. Autonomous statue presence is emphasised at the most detailed levels (with Schulz 1992, II, 709). In a request for its endurance (mn) inscribed on a Dynasty 20 naophorous block statue from the Karnak Cachette, twt is determined with a block statue.3 Regine Schulz (1992, II, 701, n. 3) notes this example as prefiguring the use of the block statue determinative for twt and ḥsy on 1st millennium statues (Price 2011b, 160–1). Although such uses were ancient (e.g. Newberry and Fraser 1894, pls 12, 14), it is the detail of statue on statue that is striking here. Another example is a Ramesside sistrophorous block statue of an Amenemhat, perhaps from Deir el-Medina, bearing an offering formula to Hathor on its left side asking ‘that my name endure in her chapel, (my) memory endure in her shrine’, with sḫꜢ ‘memory’ determined by a block statue.4 These are literal matters of substance and material presence. Minmose: speaking and spoken to The Ramesside high priest’s statue texts about statues are highly participatory. The oracular language of his texts implies two-way interaction and response, in this case with gods in their statue forms. Such an interactive role asserts an agency comparable with Gell’s relational model. Here the most explicit examples are intermediary statues that speak in the first person, which they delineate to some extent from the person of the owner. This strategy can be termed prosopopoeia, a type of ekphrasis (Bahrani 2014, 194; cf. Svenbro 1993, 41–2). Statues that offer to speak with gods on behalf of people are first attested in the New Kingdom and are not common (Kjølby 2009, 41, with n. 59; Otto 1948; cf. Fischer-Elfert and Grimm 2003: a Middle Kingdom statue that mediates with the dead). The best-known examples are on scribe statues of Amenhotep son of Hapu from the reign of Amenhotep III:

Sety I, is addressed to ‘those who will see this ’, as given in Kitchen’s copy text (KRI, I, 314, 15), which must be read as a word for statue.

10

E. FROOD

Perform an offering which the king gives for (me) and libate for me from what you have. I am a herald (wḥmw) whom the king appointed in order to hear the petitions of the humble, in order to bring forward matters of the Two Lands. (Urk. IV, 1833, 17–9)

These texts use the first person to speak as the statue, setting out its role as a distinct agent within a defined social environment and claiming a particular set of social transactions (e.g. Galán 2003; Salvador 2014). Intermediary statues of the Ramesside period are visually and textually more complex, as exemplified by the begging ‘bald ones’ (jꜢs/js) gathered and studied by Jacques Jean Clère (1995), together with their Dynasty 18 precursors (also Franke 1988). Most bald ones speak with the statue’s voice at some point, and a number unusually call attention to their distinctive form, asking for their bald heads to be anointed and their cupped hands to be filled (e.g. Frood 2007, 189–91; 2015). One of the most vivid examples of this body-centred ekphrasis belonged to the high priest of Onuris Minmose, who held office under Ramesses II (Egyptian Museum, Cairo CG 1203: Clère 1995, 73–80, doc. A; surviving height 66cm). Although it is unprovenanced, it may have been set up somewhere at Abydos like so many of Minmose’s monuments (Effland and Effland 2004, 6). This now headless red granite, roughly lifesize statue shows him kneeling with a Hathor-headed sistrum on his knees. His left hand is placed on the side of the emblem’s wig, while his right is above it, palm raised and cupped beneath where his mouth would have been. An inscription in twenty-one columns envelops him, beginning on the right side of the sistrum and finishing on the left with a cryptographic writing of his name. It reads continuously around his body, including two columns on the back pillar, and is indiscriminately carved on the wig of the sistrum, the rear surfaces of Minmose’s shoulders and arms, and his feet. After a list of titles, the text begins, as on many bald ones, with the statue speaking: ‘I am the bald one of Isis the great; I live in her open court’ (cols 3–4). Both here and later, js is written with a determinative of a kneeling man holding a sistrum. This is the only example of this variant determinative in Clère’s corpus (1995, 21, 217). The distinctive orthography mirrors the statue’s form at a detailed level, and a similar mirroring is developed in the text. The following stanzas describe the offerings, mostly drinks, that Minmose desires, commanding an audience to have them ‘placed upon my arms, my mistress kept

fast (smn) in my(?) embrace’ (cols 6–7), thus indicating where the offerings were to be poured (Rizzo 2004, 515–7), while the sistrum is held in its arms. There follows an unmarked shift of voice to a speech placed in the mouth of the viewer/offering bringer and probably addressed at least partly to the statue (with Clère 1995, 76): ‘take for yourself snw-loaves and date-wine to your mouth … given to the mouth of (Min)mose, true of voice, and to the bald one of Isis’ (cols 9–11). The statue then resumes its speech, addressing ‘everyone in the court, noblewomen as much as anyone’ (cols 12–3), drawing attention to its probably bald head as well as other details of physical form: Anoint the servant of Isis; there is oil for it, on its head (tp.f). Look [the image?] of my lady is carved on my head (ḏꜢḏꜢ.j); look, she is at my throat as amulets (wḏꜢw), its [the statue’s] right hand receiving them [offerings?], my left keeping fast the sublime sistrum of its mistress. (cols 13–6)

I follow Ute and Andreas Effland (2004, 12) in reading ‘the image or name of my lady’ in the damaged passage in the first verse here, suggesting that a lacuna after mk gives enough space for another group (with KRI, III, 471, 2a; not noted by Clère). It is difficult to know what this name/image might have been, as is true also of the amulets (wḏꜢw) in the next verse. An amulet around the neck is difficult to imagine because of the position of the right hand in front of the mouth (cf. Clère 1995, 141, fig. 50), although possible, comparing, for example, begging statues that hold menats (e.g. Clère 1995, 114–8, doc G). But no traces are indicated on the photographs published by Clère or in his discussion. Clère (1995, 78, n. p) considered that the carving on the head referred to its baldness. Effland and Effland suggest that both the carving and the amulets are cryptograms of Min and/or Osiris, as attested on other statues of Minmose, reading against the female determinative of nb, thus ‘my lord’. Carving of images and texts on the heads of statues is occasionally attested before the Late Period (e.g. Guichard 2014, 260, cat. 293; Frood in prep. a), so an image of Isis, or her name, could have been present on the head. Although the referents for this ekphrasis are necessarily uncertain, and a match between what is mentioned and what was depicted probably did not have to be exact in any case, its effects are powerful: this is about the body, its needs, its adornments, its musical performances, its visual impact.

WHEN STATUES SPEAK ABOUT THEMSELVES

The shift to the third person with ‘its right hand’, although perhaps an error, is very much in keeping with the shifting perspectives across the statue. The text closes with statements of Isis’ benevolence and mercy, followed by another unmarked shift to address the statue that it should receive oils and libations for the ka of Minmose. Here the change in voice opens up to this indwelling aspect. The statue has an autonomous presence as an js, and through it plays a part in the roles of mediator and musician. ‘Bald one’ was not a title held in life by Minmose or other owners of these statues. The same is true of the designation of Amenhotep son of Hapu as a wḥmw (Price 2011b, 20; Clère 1968, 143–4), which is frequently held by bald one statues, along with the doorkeeper titles (jry-ꜥꜢ) that are known from Ramesside ‘doorstop’ statues (Clère 1968; 1995, 13). These objects, with their projecting doorstop elements, also speak in the first person to declare their place and role as guardians of temple doorways. As Vincent Rondot (2011, 145) observes in his brilliant study of this statue type, ordinary block statues that do not have the projecting feature could have, probably often symbolically, performed the same function. ‘Watchman’ (wršy) seems to be a comparable designation that evokes this role. A text on the left side of a block statue of a charioteer of Ramesses II named Merenptah, from Tell Nabasha, speaks as the statue, describing its position: ‘I am at the side of the august stairway of my lady (Wadjet)’, and requesting offerings and libations ‘for the watchman (pꜢ wršy) (of) Merenptah’.5 Similarly, a column on a lost fragment of a back pillar belonging to the viceroy of Nubia, Setau, under Ramesses II, recorded by Lepsius near the temple of Amenhotep III at Elkab, includes the statement ‘I am your watchman (wrš(y).k) who is in this temple’. This comes after a broken reference to ‘your ẖntystatue’ and a reference to ‘your sublime god’ follows, so the passage is probably part of an address to the deified king.6

5

6

KRI, III, 247, 16–248, 1; Schulz 1992, I, 439–40, II, pls 116a–d, cat. no. 260, with references. The back pillar includes, in a damaged context, phrases that might be part of an address to the statue: ‘May the subjects extend their arms to you, in sight of (ẖr-ḥr: lit. ‘beneath the face of’) Merenptah’, perhaps referring to his own, now erased, image which was carved on the front of the statue’s knees, as well as the statue itself. LD III, 174; LDT IV, 42; KRI, III, 82, 7. My thanks to Vivian Davies, Christine Raedler, and, especially, Susanne Woodhouse

11

Statues can speak and can bear titles of their own. In at least a couple of cases they mention having their own names, as is well-attested for royal statues (e.g. Donohue 1988, 118–23) but rare for non-royal ones (cf. Bahrani 2014, 210). Examples are an appeal on the statue of Didia which asks for recitation of ‘the name of this statue’ (rn n twt pn: KRI VII, 25, 9–10; Frood 2007, 134) and a text in the Saqqara tomb of Mose that requests offerings for his ẖnty-statue, ‘its name enduring forever and ever, carved for perpetuity’ (KRI III, 422, 7–8; Gaballa 1977, 13–4, pl. 28). These statues are at once the person and something more than and different from that. Their owners take on identities that may be unconnected with their biographies, as with two monuments of the mid-Dynasty 18 soldier Amenemheb Mahu, which present him as a ‘singer with the harp of Amun’ (Baines forthcoming) and with the representation of very high-ranking people, such as Iymeru and Amenhotep son of Hapu (above) as well as princes during Dynasty 4, as scribes. In the latter case at least, they are scribes for the king, so the essential element that gives them meaning is implied but not present, and thus they are examples of this sort of distributed role-play from at the least the early 3rd millennium.7 Minmose seems to have liked role-plays: he also dedicated a doorstop statue (Clère 1968) and one of himself grinding grain (Frood in prep. b). Prominent people commissioned numerous statues in a range of iconographies and roles, providing a point of departure for the verbalised realisations found in them. Minmose’s bald statue is distinctive in the way it mixes voices. Statements addressed to the statue on the statue, often in the voice of the owner but here seemingly in the voice of a viewer, are not common. The most discussed are those running around the base of the Dynasty 19 statue of the overseer of the treasury Panehsy.8 That on the right half presents the statue’s role in the social sphere (with Ockinga 1984, 57), addressing ‘(my) snn-image’ and asking that it receives bouquets and offerings, ‘and then my ba will come

7 8

for discussion of this fragment and the fragments of a statue base of Setau also from Elkab and often, perhaps rightly, associated with it (KRI, III, 82, 8–11). My thanks to John Baines for this point. London, British Museum EA 1377. Much cited and discussed, including Otto 1948, 464 n. 2; Ockinga 1984, 57; Schulz 1992, II, 704–5; Van Dijk 1993, 122–3; Assmann 2003; Assmann, Bommas and Kucharek 2005, 330–1; Frood 2007, 166–70.

12

E. FROOD

fluttering, so that he may receive offerings with you’ (Bierbrier 1982, pls 49–51), rather as on Minmose’s statue. The inscription on the left is more transcendent: ‘O my ẖnty-statue, you are here before the lords of the sacred land; place yourself as the memory of my name in the domain of the lords of Tawer, for you are here for me as a refuge (?: jnht); you are my true ḏt-body’. Parallels for the opening verses are known from the back pillar of a naophorous block statue of a treasurer of the Ramesseum, Khay, from Saqqara: ‘O my ẖntystatue, you are in the place of truth beside the lord of the gods; place yourself as the memory of my name in the presence of Wenennefer’ (Schulz 1992, I, 191–2, esp. n. 6 with refs). Metaphors of monuments, including statues, as memorials are well-attested, so that much here is traditional (see Ockinga 1984, 53–6, with n. 9). What is striking, and perhaps relatively new, is the voicing of the idea as an address to the statue. This act asserts a statue’s materiality and its separateness. Another example is an address running in a single column down the lap and between the legs of the colossal statue of the Dynasty 20 third priest of Amun, Amenemope, from the rock-cut group in his Theban tomb: ‘O my twt-statue of stone (jnr) of my city, may I be firm (rwḏ ) … that my name be remembered upon your (sur)face (ḥr ḥr.k)’.9 Here physical impact is heightened by the mentions of fabric and surface, in addition to the scale and ‘dominating presence’ of the statue itself (Ockinga 2009, 33, 56 n. 139). Boyo Ockinga compares the passage with funerary texts which address the Theban mountain and request that it opens to conceal the body, emphasising the living, transcendent quality of the stone (Rummel 2016): stone as body, statue as body and divinised flesh. These texts fit with approaches to statues as ‘personhood distributed or extended in the milieu, beyond the body boundary’ (Kjølby 2009, 35). There is something distinctive here beyond the marking of agency. Rather than deferring reference to the statue and thereby opening up the ‘tranquil paradox’ of statue presence and personal absence (Ma 2007, 205–6; 2013, 28), these texts assert the statue’s presence, its ‘hereness’ (Svenbro 1993, 42–3), which is both closely related to and distinct from the referent. The intermediaries encompass possibilities of representation from role-play to

Statues of the vizier Paser have been mentioned in this discussion. Many of them are inscribed with both traditional and playful examples of statue ekphrasis, and his tomb contains a scene in which he makes a speech about the creation of statues (Assmann 1992). I conclude with the one that is most elaborate in relation to this discussion, a schist standing statue dedicated in the Ptah temple at Memphis (Fig. 4: PM III, 2, 838; KRI, III, 11–3). Like Roma’s statues, it mentions or alludes to itself almost everywhere, and like Panehsy’s, it makes play with material and transcendent aspects of statue self and body. When complete, it was about two-thirds life-size (surviving H. 62cm including a high base). It shows him standing behind a tall plinth which bore an effigy-form figure of Ptah in front of a djed. Only the lower part of the vizier’s body survives, from the top of the thighs down to his sandalled feet, as well as the tips of his fingers held against the djed. The inscriptions were edited by Ramadan el-Sayed (1980), and I largely follow his treatment here (page numbers refer to his article). A dedication formula is inscribed on the top of the base, in front of the right foot: ‘made by the senior scribe of Paser for his lord’ (209). This dedication may have continued on the lost section of the base in front of the left foot, and the fact of dedication may influence other features, in particular the vivid description of the statue body which relates it closely to the transfigured body of Paser. I focus on the inscriptions on two main areas: the plinth of the Ptah statue and the back pillar, although there are allusions to statues elsewhere. For example, the text running down the right side of the djed behind the Ptah statue refers specifically to it — ‘... your tjt-image, South of his Wall, that you may give to (me) all life and dominion …’.10 The column on the left uses a conventional formula to request that ‘you cause my name to be firm

9

10

Ockinga 2009, 55–6 (text 21), pls 14–15, 71. The reading of ḥr.k as ‘(sur)face’ follows Ockinga 2009, 56. It is uncommented there but lovely.

statue as a separate presence, as with ṣalmu in Bahrani’s analysis. The composers and designers of statues knew the myriad complexities of images, as well as the range of qualities and potentials that statues possessed, some of which we now label as agency. Paser’s bodies

El-Sayed notes a small lacuna followed by ‘before me (ḫr.j)’, but this reading is questionable in context and I am unable to verify it.

WHEN STATUES SPEAK ABOUT THEMSELVES

13

(rwḏ) in your temple’ (209), referring to Paser’s own statue, and enforcing the relationship of dependence between the two. The two offering formulas on the plinth are the most visually prominent texts, and they draw attention through their organisation and symmetrical layout (as set out by Borchardt 1925, 177, contra el-Sayed 1980, 210). The formulas radiate out in a line at the top of the plinth from a shared central ḥtp. The texts then continue with a shared central column that starts with Tatenen, directly beneath ḥtp. This shared column includes the only known example of epithets describing Tatenen and the gods associated with him as ‘ones who lengthen lifetimes, and feed those in the underworld’ (LGG VI, 206, 385). Their prominent, repeating position may be a virtuoso compositional display, as are many of the rare and distinctive orthographies in the texts (el-Sayed 1980, 215, 231). The section that fans out from the statue’s right in five columns alludes to the full trajectory of life to afterlife evoked in the epithets: An offering which Ptah, South of his Wall, sublime djed, foremost of the Tjenenet, gives to Tatenen, great one of the council of gods, foremost of the domain of Ptah, those who lengthen lifetimes and feed those in the underworld, that they may grant a long lifetime extolling him (Ptah),11 veneration when (they) receive abundance, contentment through food-offerings, rejoicing in sustenance, until the reckoning of a lifetime to its hour; voice offerings after interment from what has come forth from the presence of the gods, anointing the body (nms ḥꜥw) with the finest sacred oil, as the remainder from the ḏsr-incense of the portico; receiving pure garments provided with bands(?), and clothing imbued with the god’s body; for the ka of the city governor, vizier, Paser, true of voice. (el-Sayed 1980, 210)

I follow el-Sayed (1980, 215) in reading nms and ḏsr as words for anointing and incense, meanings that seem to be attested otherwise only from the Graeco-Roman period. These are appropriate to context, and are sup-

11

El-Sayed reads .sn for sw here: ‘that they grant (Paser) a long lifetime celebrating them (the gods).’ The substitution of sw thus allows for the sportive writing of swꜢš: . I suggest instead that sw remain in the singular as a reference to Ptah.

Fig. 4: Schist standing statue of Paser from Memphis. Reign of Ramesses II, Dynasty 19. H. 62cm. Cairo, Egyptian Museum CG 630, after Borchardt 1925, pl. 116.

ported by the vessel determinative for nms and the substance determinative for ḏsr. They are extensions of and plays upon the more normal meanings of nms ‘to clothe’ and ḏsr as something sacred and set apart. Such poetic phrasing highlights the salience of Paser’s body (ḥꜥw) in the transformation described, all effects intensified by the description of the clothing which follows. The ‘pure garments’ (wꜥbw) are said to be provided or adorned (ḏbꜢ) with wnfw, which el-Sayed renders ‘fringes’, perhaps a mixed orthography of wnḫw, which can refer to bands of linen and types of clothing more generally. The phrasing may refer to patterns of folding and wrapping, in connection with the cord determinative of wnfw (el-Sayed 1980, pl. 48b) and its

14

E. FROOD

association with covering (Wilson 1997, 234). It is perhaps possible that ḏbꜢ ‘to replace (one thing for another)’ (e.g. Parkinson 2012, 65) could be understood here as referring to the tearing up of cloth to create strips for the wrapping of mummies and statues. The following line may also refer to such reuse of temple linens, including cloth that had been used to wrap temple statues (Riggs 2014, 121), the ‘god’s body (ḥꜥwnṯr)’.12 All the different words for textiles here have associations with temples, priestly clothing, and ritual (el-Sayed 1980, 215–6). They can also be compared with how arrangements of cloth on the bodies of Late Period statues are referred to in their texts (e.g. Klotz 2014, 298). The text cascades visual images in a way that may allude to the presence of the god’s body in front of the statue, as well as the future shrouded and decorated corpse of the vizier, while also evoking the body as a site of representation, as in the statue itself. Although the goal of the text is the transfiguration of the body after death, the mentions of reversion of offerings, the sbḫt — a protective gateway structure attested in the underworld and in temples (Spencer 1984, 161– 690) — and of cultic textiles all relate to temple settings. The use of the neutral third person imparts a generalising quality, opening up associations between the wrapped, divinised human body, the bound, effigyform statuette of Ptah above the text, and the statue of Paser in his enveloping vizier’s cloak: ‘the layering of materials freighted with meaning — wood, metal, and cloth for cult statues; bones, flesh (real or otherwise), and wrappings for the dead — made an image that represented a god on earth and that required all the attendant care of ritual and seclusion’ (Riggs 2014, 106). Bones and flesh are particularly relevant for the flanking text on the left side of the plinth, which describes the statue directly after the partly repeated offering formula to Osiris, Nefertem, Horus, and Tatenen: that they place my ẖnty-statue in this shelter, that they transfigure its ba and raise up (its?) sꜥḥ, that they permit it to stand as an akh as before, and place its head (tp.f) upon its bones (qsw.f),

12

13

My thanks to John Baines for this suggestion, and to Christina Riggs for discussion. Wilson 2010, 150–1, no. 48. Captions that describe the divine images held or raised up by statues — Edith Bernhauer’s

that they purify it through their action with pure water which comes from Elephantine, and natron from Elkab, the divine efflux, so that when they integrate its ḥꜥw-body before them, its ḏt-body will always be before the great Ennead in Heliopolis. (el-Sayed 1980, 216)

This is the only certain use of the first person (written with ) in the surviving inscriptions; this may relate, in part, to its dedication for Paser by a third party. So its use here makes it clear that the body treated is that of the statue. The choice of words is more conventional than on the right text, although numerous assonances heighten its effect (e.g. sꜢḫ.sn-bꜢ.f sṯs.sn-sꜥḥ(.f) sꜥḥꜥ.sns(w) m-Ꜣḫ tpt-ꜥ). It begins by stressing physical location — resonating with Maanakhtef’s — this address to the temple court and Panehsy’s characterisation of his statue as a refuge. The description of the statue’s transformation that follows uses ancient formulas concerned with the reconstitution of deceased human and divine bodies (el-Sayed 1980, 218–9). The next example of these formulas that I know in relation to a statue concerns the twt-statue-body on the base of a standing statue of Montuemhat from the Karnak Cachette which asks that Montu place ‘its head upon its bones’ (with el-Sayed 1980, 219 n. k). A number of statue texts use biological metaphors for the statue body, employing fleshy words such as ḥꜥw (Rizzo 2004, 518). Another of Paser’s statues from the Ptah temple asks that Ptah ‘make my body (ḥꜥw) firm when raising up your perfection, my arms shouldering your person every day’ (KRI, VII, 407, 13–5), an ideal description of the statue itself kneeling and holding a figure of the god.13 ‘Perfection (nfrw)’ here also means the body. The early Dynasty 20 stela of the high priest of Amun, Bakenkhons, which describes the devastation wrought on statues at Karnak, refers to them as having fallen ‘on their sides (ḏrww)’, ‘on their backs (psdw)’, and ‘on their noses (fndw)’, all written with bodily determinatives, especially (Boraik 2007, 122, fig. 1, l. 6–7). A century earlier, Paser’s statue too pushes these metaphors very far. Discussing the formula ‘the ba to the sky, the corpse to the underworld, your ẖntyw-statues among the

‘kultbezogener Vermerk’ (2010, 131–2) — are one of the few types that regularly refer to physical appearance, although often elliptically (e.g. Van Dijk 1983; examples cited by Klotz and Leblanc 2012, 675 n. 158).

WHEN STATUES SPEAK ABOUT THEMSELVES

favoured’ on Dynasty 18 and Late Period statues, Assmann, Bommas and Kucharek suggest a tripartite division of the self according to environment, in support of the temple statue as substitute body (2005, 321–31, esp. 329–30; cf. Rizzo 2004, 518–9). Like the human body, Paser’s standing statue has the potential for multiple manifestations that cannot be neatly divided into components. sꜥḥ, a word for statues in later texts (e.g. Price 2011b, 231–8; Frood 2013, 180–1), encompasses its wrapped, divinised nature, and ḏt its capacity for indwelling, as on Panehsy’s statue (Assmann 2003; Riggs 2014, 86, 99–100). But unlike Panehsy’s text, in which the statue is a perch for the fluttering ba, Paser’s statue has its own ba and akh. And its akh is ‘as before’, referring to another phase of existence, maybe Paser’s life. The absence of an n kꜢ n formula at the end of the text shows that it is the statue that stands before the gods in all its transfigured, divinised potential. The arrangement of the inscription in three columns on the back pillar is also complex. The right-hand column is autonomous, referring to being ‘in his presence’, followed by a title string. The central and left column read together and close with a very rare Ramesside example of something close to the full ‘Saite formula’ (el-Sayed 1980, 224; also Raedler 2004, 332–3 with n. 270 for a monumental shabti of Paser bearing the formula). This formula, which is attested from the early New Kingdom onward and common in Dynasty 26, refers to the city god’s protection of the individual/ statue and often refers to the statue as jwny. Karl Jansen-Winkeln (2000) has concluded that the formula was connected with the consecration of statues in temples, whereas David Klotz’s (2016) convincing reassessment of Dynasty 18 examples relates it more to their role in performances. The lost opening of Paser’s text probably included a reference to the city god or Ptah, and was followed by an enumeration of Paser’s epithets and titles. The text continues with a request, probably to that god, to ‘place yourself around him, in front of his ka, in his presence; he is a jwny true of voice [...]’ (el-Sayed 1980, 224). I follow Klotz here (2016, 209), rendering dj as an imperative (rather than Jansen-Winkeln’s passive in Jansen-Winkeln 2000), through which ‘the dedicant essentially demands the City God stop as he passes by’ in procession. Unlike Klotz’s Dynasty 18 examples, in which the ka is the god’s statue (see Jansen-Winkeln 2000, 92–3 for a Ramesside example), the ka in Paser’s text is his own, as on many Late Period statues. Whereas Klotz

15

understands the ka as the dedicant’s statue that is situated in the presence of his own ba who is travelling in the god’s procession, such a complicated explanation is not required for Paser, whose ka is simply his own statue in the presence of the god’s. Klotz (2016, 211) understands jwny as ‘the Heliopolitan’, referring to the statue as a form of Osiris. In Paser’s case the epithet extends the transfigurations undergone by self and statue through the texts ranged over its plinth. On the back pillar, the final groups at the bottom of the central column write jwny pw mꜢꜥ-ḫrw, framed by Paser’s name which closes the columns on either side, especially

(the sr element of his name), which is repeated

immediately below, with closing the offering formulas that wrap around the base (el-Sayed 1980, pl. 49b). So the divinised statue is surrounded and watched over by a component of Paser’s name that approximates the form of the statue. This text and its visual plays perhaps complete the many manifestations of the statue developed throughout the texts. It is tempting to consider this an address to the figure of Ptah that Paser holds. Not only does this formula assert the agency of the city god and the deceased (Klotz 2016), it also proclaims the distinct and separate presence of his own statue. The statue is what matters. These are vivid, complex descriptions of a statue. As on Minmose’s statue, the texts and their orthographies draw attention to the physical, in this case almost flesh and bone, presence of the statue, as well as evoking its transcendent qualities in complex relationship to Paser’s divinised self; in this respect, it can be compared with Panehsy’s text. This treatment opens up a different sort of ‘tranquil paradox’ from that of Ma discussed earlier, here between the obvious stone form and the biological metaphors of humanness and ephemerality that play out across it. Self and statue are not completely integrated; nor is one substituted for the other. Statue selves and statue audiences Statues have distinct and multiple intersecting ontologies. They are manifestations and role-plays of the person they represent, of whom they can describe themselves as substitutes and extensions in a wide variety of ways. As Schulz (1992, II, 705) suggests for Panehsy, they are objects of memory and convergence (‘Annäherung’). But they are also things apart. And as Bahrani argues, awareness of this separateness is at the heart of the human creation of images. For Egyptian

16

E. FROOD

statues from before the 2nd millennium, this separation is manifest mainly in form, and in the position and orientation of their inscriptions, rather than in their content. A vivid example of the former is Idu’s statue pushing itself up through the floor of his Dynasty 6 tomb (Simpson 1976, pl. 29a–c; cf. Franke 1988). I have not focused on the forms of the statues, as against the interplay of form with the texts inscribed on them. Ultimately, forms and poses, from the open mouths and begging postures of the bald-one statues to the cult objects they often hold, convey the fundamental meaning. The texts build upon that meaning and can express nuances that go beyond visual form, in particular to address wider dimensions of time, space, and mode of being. An ‘opposition of person/image’ (Bahrani 2003, 123) is set out in these statue texts. Many of the compositions I discuss are making play with the alternate possibilities that Bahrani indicates with ‘/’ and cross its boundary. Some of what the statue texts assert can be characterised as agency in Gell’s terms: many of the statues are strongly ‘social animals’. But I agree with Bahrani’s criticism of Gell’s model (Bahrani 2014, 201): If we use expressions like the ascription of agency, we are separating these as elements apart from the thing. Agency is, then, an element that can be isolated and analysed, categorised in a different way from the work or the thing, the object or the work of art. … really the point here is that these are things with a different status. They are parts of ontologies and technologies that are different from those we take as being logical and rational, but they are nevertheless real and even logical.

We have thus to be more open to complexity and nuance in analysing what statues do. Egyptian concepts of the person are multiple and shifting; thus they always offer the potential to be understood creatively and anew. Those who created these statues knew this; they were deeply aware of potentials of monuments and reflected upon what to us are philosophical questions of image-making (see Nyord 2017). Such reflections are part of the wider dynamics of monument creation and the perpetual push towards innovation. As Bahrani (2003, 133) observes ‘[i]mage making and portrayal through substitute images were therefore quite literally dependent upon writing presence’. Unlike the Hellenistic world, where a whole genre of epigrams explores the hermeneutic problems in viewing and understanding images (Ma 2013, 16), we have these sorts of things on the statues themselves. From block

statue and musician determinatives to Paser’s texts thematising the myriad ways a statue could be conceptualised, composers were writing presence in new ways. Ekphrasis by its nature draws attention to the viewer or reader, their voice, presence, and subjectivity (cf. Bartsch and Elsner 2007, ii). In the case of the begging statues this addressing of the viewer is particularly punchy. So far in this essay I have only alluded to audience. A significant part of Svenbro’s and Ma’s discussions of statues relates to their encounter with audiences, and how the grammar and voice of the dedicatory formulas relate to the voice of someone who would read them out. For Egyptian material, it is difficult to model how far it was possible to read/declaim inscriptions from the statues themselves, especially since access to the temples in which the material I discuss was set up was probably quite restricted, even if to relatively high concentrations of literate people. Oral performances at creation and/or dedication would have been crucial for the individual and his group, and probably also for the artists and workshops (Baines 2007, 152; Kjølby 2007, I, 96–7). But outside these contexts? Were they ever read again once they were set up? Many statue texts are at least presented as if they might be. The intermediary scribe statues of Amenhotep son of Hapu are somewhat paradoxical in this. The statue reads and writes, activities that seem for and of itself. The papyrus held on its lap is oriented to the statue as if written by it and to be read by it, giving a sense that it is engaged in a self-contained activity for which no other human is required or desired, unless he is reading aloud to us. This is quite in contrast with the outward perspective of the text on the base which identifies and speaks to a diverse human audience. These ‘reading’ effects were understood by Jean Yoyotte (1981) as one of a number of ‘attention grabbing devices’, including the bald ones’ demanding gestures (also Lorand 2016, 237; and see Price, this volume). This may be a productive way of thinking about part of the purpose of these ingenious elaborations in context. The examples I have discussed come from temples, which by the late 2nd millennium were probably becoming cluttered with statues (e.g. Kjølby 2007, I, 40–7). For the southern processional route at Karnak, Chiara Salvador (pers. comm.) imagines that people who walked down the paved, axial pathway through the courts could have been attracted to step off and wander through the displays of statues that were probably set up on either side. Decorum restricted possible forms here — no begging bald ones are known from Karnak before the

WHEN STATUES SPEAK ABOUT THEMSELVES

Late Period — so other strategies were brought into play, including things like the texts I discuss. Despite the superficial frontality of statues, texts wrap around. To read Minmose’s statue it is necessary to walk around it and look very closely to see the musician determinative of js; you have to look at the back of Paser’s statue to see the framing of jwny, the divinised statue, by his name. The idea that most statues were pushed up against walls and inaccessible may be influenced a little too much by the modern experience of the museum (cf. Evans 2012, 77–88 in relation to early Mesopotamian statues). Those who, like Paser and Minmose, had multiple statues might further display their status by commissioning unusual texts, and they or their audiences might also show their learnedness by identifying a distinctive statue and stopping to read its texts. Some 1st millennium statues claim that their owners ‘called out the names of all statues that I passed by’ in the temple (Klotz 2015, 85–6). So some statues probably became well-known. If you were reading, then the effects of ekphrasis could have been quite startling, perhaps forcing you to think differently about the nature of the object in front of you, or perhaps simply forcing you to think, pay attention, react. How could a stone statue be flesh and bone, as a flesh and bone body is transformed into a statue (Riggs 2014, 99–110)? Statues and statue texts, from the very earliest periods, manage their manifestations and interactions — human, divine, and their own distinct status, statue-ness, within that continuum — in complex and diverse ways. The New Kingdom examples treated here demonstrate this complexity, and the fluidity and multiplicity of the selves they represent and present. Bibliography Assmann, J. 1972. Neith spricht als Mutter und Sarg (Interpretation und metrische Analyse der Sargdeckelinschrift des Merenptah). Mitteilungen des Deutschen Archäologischen Instituts, Abteilung Kairo 28, 115–39. ———. 1992. Ein Gespräch im Goldhaus über Kunst und andere Gegenstände. In I. Gamer-Wallert and W. Helck (eds), Gegengabe. Festschrift für Emma Brunner-Traut. Tübingen, 43–60. ———. 2003. Einwohnung. In T. Hofmann and A. Sturm (eds), Menschenbilder – Bildermenschen: Kunst und Kultur im alten Ägypten. Norderstedt, 1–14.

17

Assmann, J., Bommas, M. and Kucharek, A. 2005. Altägyptische Totenliturgien 2: Totenliturgien und Totensprüche in Grabinschriften des Neuen Reiches. Heidelberg. Bahrani, Z. 2003. The graven image. Philadelphia. ———. 2014. The infinite image: Art, time, and the aesthetic dimension in antiquity. London. Baines, J. 1999. Forerunners of narrative biographies. In A. Leahy and J. Tait (eds), Studies on ancient Egypt in honour of H. S. Smith. EES Occasional Publications 13. London, 23–37. ———. 2007. Visual and written culture in ancient Egypt. Oxford. ———. Forthcoming. Ancient Egyptian biographies: From living a life to creating a fit memorial. In E. Frood, J. Stauder-Porchet and A. Stauder (eds), Ancient Egyptian biographies: Forms, contexts, functions. Atlanta. Barbotin, C. 2005. La voix des hiéroglyphes: Promenade au département des antiquités égyptiennes du Musée du Louvre. Paris. Barta, W. 1968. Aufbau und Bedeutung der altägyptischen Opferformel. Ägyptologische Forschungen 24. Glückstadt. Bartsch, S. and Elsner, J. 2007. Introduction: Eight ways of looking at ekphrasis. Classical Philology 102 (1), i–vi. Bernhauer, E. 2010. Innovationen in der Privatplastik: Die 18. Dynastie und ihre Entwicklung. Philippika 27. Wiesbaden. Bierbrier, M. (ed.) 1982. Hieroglyphic texts from Egyptian stelae etc. in the British Museum 10. London. Bisson de la Roque, M. F. 1927. Rapport sur les fouilles de Médamoud (1926) I. Cairo. Boraik, M. 2007. Stela of Bakenkhonsu, high priest of Amun-Re. Memnonia 18, 119–26. Borchardt, L. 1925. Statuen und Statuetten von Königen und Privatleuten im Museum von Kairo. Nr. 1–12942: Text und Tafeln zu Nr. 381–653. Catalogue général des antiquités égyptiennes du Musée du Caire 9. Berlin. Clère, J. J. 1968. Deux statues ‘gardiennes de porte’ d’époque ramesside. Journal of Egyptian Archaeology 54, 135–48. ———. 1995. Les chauves d’Hathor. Orientalia Lovaniensia Analecta 63. Leuven. Dijk, J. van. 1983. A Ramesside naophorous statue from the Teti pyramid cemetery. Oudheidkundige Mededelingen uit het Rijksmuseum van Oudheden te Leiden 64, 49–60.

18

E. FROOD

———. 1993. The New Kingdom necropolis of Memphis: Historical and iconographical studies. Groningen. Donohue, V. A. 1988. The vizier Paser. Journal of Egyptian Archaeology 74, 103–23. Drioton, E. 1927. Rapport sur les fouilles de Médamoud (1926) II: Les inscriptions. Cairo. Effland, U. and Effland, A. 2004. Minmose in Abydos. Göttinger Miszellen 198, 5–17. Evans, J. 2012. The lives of Sumerian sculpture: An archaeology of the early dynastic temple. Cambridge. Fischer-Elfert, H.-W. 1998. Die Vision von der Statue im Stein: Studien zum altägyptischen Mundöffnungsritual. Schriften der Philosophischhistorischen Klasse der Heidelberger Akademie der Wissenschaften 5. Heidelberg. Fischer-Elfert, H.-W., and Grimm, A. 2003. Autobiographie und Apotheose: Die Statue des Zš(š)n ZꜢ-Ḥw.t-Ḥrw im Staatlichen Museum Ägyptischer Kunst München. Zeitschrift für Ägyptische Sprache und Altertumskunde 130 (1), 60–80. Francis, J. A. 2009. Metal maidens, Achilles’ shield, and Pandora: The beginnings of ‘ekphrasis’. American Journal of Philology 130 (1), 1–23. Franke, D. 1988. Die Hockerstatue des Sonbso-mei in Leiden und Statuen mit nach oben gerichteten Handflächen. Oudheidkundige Mededelingen uit het Rijksmuseum van Oudheden te Leiden 68, 59–76. Frood, E. 2007. Biographical texts from Ramessid Egypt. Writings from the Ancient World 26. Michigan. ———. 2013. Sensuous experience, performance, and presence in Third Intermediate Period biography. In R. Enmarch and V. M. Lepper (eds), Ancient Egyptian literature: Theory and practice. Oxford, 153–84. ———. 2015. Temple lives: Devotion, piety and the divine. In P. Giovetti and D. Picchi (eds), Egypt: Millenary splendour. The Leiden collection in Bologna. Milan, 316–23. ———. in prep. a. Imaging the body: The elaboration of statue surfaces in the late New Kingdom. ———. in prep. b. Minmose grinding grain (Berlin ÄM 24179). Gaballa, G. A. 1977. The Memphite tomb-chapel of Mose. Warminster. Galán, J. M. 2003. Amenhotep son of Hapu as intermediary between the people and god. In Z. Hawass and L. Pinch Brock (eds), Egyptology at the dawn

of the twenty-first century: Proceedings of the eighth International Congress of Egyptologists 2. Cairo, 221–9. Gell, A. 1998. Art and agency: An anthropological theory. Oxford. Guichard, H. (ed.) 2014. Des animaux et des pharaons: Le règne animal dans l’Égypte ancienne. Lens. Hall, H. R. 1914. Hieroglyphic texts from Egyptian stelae etc. in the British Museum 5. London. Jansen-Winkeln, K. 2000. Zum Verständnis der ‘Saitischen Formel’. Studien zur altägyptischen Kultur 28, 83–124. Kjølby, A. 2007. New Kingdom private temple statues: A study of agency, decision-making and materiality. Doctoral thesis, Københavns Universitet. ———. 2009. Material agency, attribution and experience of agency in ancient Egypt: The case of New Kingdom private temple statues. In R. Nyord and A. Kjølby (eds), ‘Being in ancient Egypt’: Thoughts on agency, materiality and cognition; Proceedings of the seminar held in Copenhagen, September 29–30, 2006. Oxford, 31–46. Klotz, D. 2014. Replicas of Shu: On the theological significance of naophorous and theophorous statues. Bulletin de l’Institut français d’archéologie orientale 114, 291–337. ———. 2015. The cuboid statue of Ser-Djehuty, master sculptor in Karnak: Los Angeles County Museum of Art 48.24.8 + Cambridge University, Museum of Archaeology and Anthropology 51.533. Revue d’égyptologie 66, 51–109. ———. 2016. Get thee behind me, city god! New Kingdom versions of the so-called ‘Saite formula’. Zeitschrift für Ägyptische Sprache und Altertumskunde 143, 204–13. Klotz, D. and LeBlanc, M. 2012. An Egyptian priest in the Ptolemaic court: Yale Peabody Museum 264191. In C. Zivie-Coche and I. Guermeur (eds), ‘Parcourir l’éternité’: Hommages à Jean Yoyotte 2. Turnhout, 645–98. Koefoed-Petersen, O. 1950. Catalogue des statues et statuettes égyptiennes. Publications de la Glyptothèque Ny Carlsberg 3. Copenhagen. KRI = Kitchen, K. A. 1975–86. Ramesside inscriptions: Historical and biographical. 7 vols. Oxford. Kubisch, S. 2008. Lebensbilder der 2. Zwischenzeit: Biographische Inschriften der 13.–17. Dynastie. Sonderschrift des Deutschen Archäologischen Instituts, Abteilung Kairo 34. Berlin and New York.

WHEN STATUES SPEAK ABOUT THEMSELVES

LD = Lepsius, R. 1897–1913. Denkmäler aus Aegypten und Aethiopien. 12 vols. Leipzig. LDT = Lepsius, R., Sethe, K., and Wreszinski, W. 1897–1913. Denkmäler aus Aegypten und Aethiopien: Text. 5 vols. Leipzig. LGG = Leitz, C. (ed.). 2002–3. Lexikon der ägyptischen Götter und Götterbezeichnungen. 8 vols. Orientalia Lovaniensia Analecta 110–16, 129. Leuven. Lorand, D. 2016. From dedication to favissa: Montuhotep’s journey in Karnak. In L. Coulon (ed.), La Cachette de Karnak: Nouvelles perspectives sur les découvertes des Georges Legrain. Cairo, 231–47. Ma, J. 2007. Hellenistic honorific statues and their inscriptions. In R. Leader-Newby and Z. Newby (eds), Art and inscriptions in the ancient world. Cambridge, 203–20. ———. 2013. Statues and cities: Honorific portraits and civic identity in the Hellenistic world. Oxford; New York. Meskell, L. 2004. Object worlds in ancient Egypt: Material biographies past and present. Oxford. Newberry, P. E. and Fraser, G. W. 1894. El Bersheh 1: The tomb of Tehuti-Hetep. Archaeological Survey of Egypt 3. London. Nyord, R. 2017. ‘An image of the owner as he was on earth’. Representation and ontology in Middle Kingdom funerary images. In G. Miniaci, M. Betrò, and S. Quirke (eds), Company of images: Modelling the imaginary world of Middle Kingdom Egypt (2000–1500 BC). Orientalia Lovaniensia Analecta 262. Leuven; Paris; Bristol, CT, 337–59. Ockinga, B. G. 1984. Die Gottebenbildlichkeit im alten Ägypten und im Alten Testament. Ägypten und Altes Testament 7. Wiesbaden. ———. 2009. The tomb of Amenemope (TT 148) 1: Architecture, texts and decoration. Australian Centre for Egyptology 27. Oxford. Otto, W. 1948. Zur Bedeutung der ägyptischen Tempelstatuen seit dem Neuen Reich. Orientalia 17, 448–66. Parkinson, R. B. 2012. The Tale of the Eloquent Peasant: A reader’s commentary. Lingua Aegyptia, Studia Monographica 10. Hamburg. Perdu, O. 2000–1. Quand des ouchebtis parlent de leur rôle. Bulletin de la Société d’égyptologie de Genève 24, 71–81. PM III2 = Porter, B., and Moss, R. L. B, with J. Malek. 1981. Topographical bibliography of ancient

19

Egyptian hieroglyphic texts, reliefs and paintings III: Memphis. Part 2: Ṣaqqâra to Dahshûr, 2nd revised and augmented edition. Oxford. Price, C. 2011a. Ramesses, ‘King of Kings’: On the context and interpretation of royal colossi. In M. Collier and S. Snape (eds), Ramesside studies in honour of K. A. Kitchen. Bolton, 403–11. ———. 2011b. Materiality, archaism and reciprocity: The conceptualisation of the non-royal statue at Karnak during the Late Period (c. 752–30 BC). Doctoral thesis, University of Liverpool. Raedler, C. 2004. Die Wesire Ramses’ II: Netzwerke der Macht. In R. Gundlach and A. Klug (eds), Das ägyptische Königtum im Spannungsfeld zwischen Innen- und Aussenpolitik im 2. Jahrtausend v. Chr. Königtum, Staat und Gesellschaft früher Hochkulturen 1. Wiesbaden, 277–416. Riggs, C. 2014. Unwrapping ancient Egypt. London. Rizzo, J. 2004. Une mesure d’hygiène relative à quelques statues-cubes déposées dans le temple d’Amon à Karnak. Bulletin de l’Institut français d’archéologie orientale 104 (2), 511–21. Robins, G. 2005. Cult statues in ancient Egypt. In N. H. Walls (ed.), Cult image and divine representation in the ancient Near East. Boston, 1–12. Rondot, V. 2011. De la fonction des statues-cubes comme cale-porte. Revue d’égyptologie 62, 141–57. Rummel, U. 2016. Der Leib der Göttin: Materialität und Semantik ägyptischer Felslandschaft. In S. Beck, B. Backes, I. Liao, H. Simon and A. Verbovsek (eds), Gebauter Raum: Architektur - Landschaft - Mensch. Beiträge des fünften Münchner Arbeitskreises Junge Aegyptologie (MAJA 5), 12.12. bis 14.12.2014. Wiesbaden, 41–74. Salvador, C. 2014. From the realm of the dead to the house of the god: The New Kingdom appeals to the living in context at Thebes. In K. Accetta, R. Fellinger, S. Musselwhite, P. L. Conçalves and W. Paul van Pelt (eds), Current research in Egyptology 2013: Proceedings of the fourteenth annual symposium, University of Cambridge, United Kingdom, March 19–22, 2013. Oxford; Philadelphia, 153–67. el-Sayed, R. 1980. À propos de la statue de Pasar CG. 630 au Musée du Caire. Bulletin de l’Institut français d’archéologie orientale 80, 207–31. Schulz, R. 1992. Die Entwicklung und Bedeutung des kuboiden Statuentypus: Eine Untersuchung zu den sogenannten ‘Würfelhockern’. 2 vols. Hildesheimer ägyptologische Beiträge 33–4. Hildesheim.

20

E. FROOD

Simpson, W. K. 1976. The mastabas of Qar and Idu, G7101 and 7102. Giza Mastabas 2. Boston. Spencer, P. 1984. The Egyptian temple: A lexicographical study. London. Svenbro, J. 1993. Phrasikelia: An anthropology of reading in ancient Greece. New York. Urk. = Sethe, K. et al. 1903–. Urkunden des Aegyptische Altertums. 8 vols. Leipzig; Berlin. Verbovsek, A. 2004. ‘Als Gunsterweis des Königs in den Tempel gegeben…’: Private Tempelstatuen des Alten und Mittleren Reiches. Ägypten und Altes Testament 63. Wiesbaden. Weber, A. 2014. ‘Die beiden Augen des Königs’: Die Statuette des königlichen Schreibers Imen-m-Ipet. In V. M. Lepper (ed.), Persönlichkeiten aus dem alten Ägypten im Neuen Museum: Für das Ägyptische Museum und Papyrussammlung, Staatliche Museen zu Berlin. Petersberg, 105–30.

Wilson, P. 1997. A Ptolemaic lexikon: A lexicographical study of the texts in the temple of Edfu. Orientalia Lovaniensia Analecta 78. Leuven. ———. 2010. Kneeling statue of the vizier Paser. In C. Barclay, R. Grocke and H. Armstrong (eds), Treasures of the Oriental Museum, Durham University. London, 150–1. Winter, I. 2010. Agency marked, agency ascribed: The affective object in ancient Mesopotamia. In I. Winter (ed.), On art in the ancient Near East 2: From the third millennium BCE. Culture and History of the Ancient Near East 34. Leiden; Boston, 307–31. Yoyotte, J. 1981. Le général Djehouty et la perception des tributs syriens: Causerie au sujet d’un objet égaré. Bulletin de la Société française d’égyptologie 92, 33–51.

A PERFECT ‘LIKENESS’? VIEWING LATE PERIOD ARCHAISING SCULPTURE IN CONTEXT Campbell PRICE In memory of my father, James

Abstract The statue of Tja-iset-imu (BM EA 1682) is a striking example of Saite archaism. The appearance of the statue, and the text it carries, make conscious reference to older statue forms, notably of the Old Kingdom. The statue’s base carries an ‘appeal to the living’ text that may refer to ‘seeing/looking at (mꜢꜢ)’ the statue — unusually described as a ‘likeness (mỉtt)’. It is argued that despite a modern expectation that such a comparison ought to be to life — especially when Egyptologists discuss Late Period sculpture — rather, the comparison being made here is to older sculptures, which were used to inspire the creators of Tja-iset-imu’s statue. Acknowledging that the original physical setting of the statue may not have allowed the optimal viewing conditions modern museum-goers might expect, the importance of attracting attention from the living, the dead and the gods is emphasised. Tja-iset-imu’s role in the cult of the statues of King Psamtek I may have provided an additional motivation for the statue’s striking form, and its unusually reflective inscription. * * * Nowadays, pharaonic statues are easy to view in isolation simply as ‘artworks’, susceptible to our own value judgements and aesthetic considerations. This is due in large part to the museum setting of most modern encounters with Egyptian sculpture. Any attempt to reconstruct the ancient context of a piece must acknowledge that statues were intended to function for, and be experienced by, the living, the dead and the gods. This should not, however, be to deny an intended aesthetic impact (e.g. Verbovsek 2015, 141–54), which may have been targeted at any of these groups. Modelling the cognitive impact of sculpture in context is 1

Themes discussed in my doctoral dissertation (Price 2011a), currently being revised for publication. I have addressed aspects of these issues separately (Price 2016a; 2016b; 2017).

hampered by the fact that so few statues — especially those representing non-royal individuals — have survived in what is likely to have been their intended setting. This is particularly true of temple spaces, which were the almost exclusive arena for elite sculpture in the 1st millennium BC. Most moveable (even if not easily portable) temple statues were removed from their original emplacements, either through deliberate caching or violence in ancient times, or having been targeted by antiquities dealers in the more recent past. The original proliferation of statues in temples, especially during the 1st millennium BC, created a competitive monumental environment that obliged artisans and their patrons to seek ways to stand out from the crowd in order to better engage the attention of the passer-by. One means of attention-seeking was the deliberate use of already ancient, highly regarded forms. Even at a rather modest scale, these works had ostentatious aims: to deploy and vaunt knowledge and associations known to artisans and their patrons.1 While inscriptions on non-royal statues have long been recognised as a particularly rich source of religious, genealogical, and occasionally biographical data on the statue owner (e.g. Jansen-Winkeln 2001), they also provide an insight into the conceptualisation of the statue itself, its intended settings and interactions with the living (Price 2011a; see also Frood in this volume). Although we lack much in the way of informative archaeological data for Late Period elite stone statuary and must contend with a partial dataset, some pieces carry retrievable reflections on their own forms and functions in their original contexts.

The statue of Tja-iset-imu (Figs 1, 2) At 125cm tall, the limestone striding statue of Tjaiset-imu is less than life-size in height, yet its impact

22

Fig. 1: Statue of Tja-iset-imu (EA 1682) from the front. London, British Museum. © Trustees of the British Museum.

C. PRICE

Fig. 2: Statue of Tja-iset-imu (EA 1682) from the rear. London, British Museum. © Trustees of the British Museum.

A PERFECT ‘LIKENESS’? VIEWING LATE PERIOD ARCHAISING SCULPTURE IN CONTEXT

23

Fig. 3a: Inscriptions on the belt. London, British Museum (EA 1682). © Trustees of the British Museum.

relies less on its scale than on its appearance. Tja-isetimu strides forward on a thick base, with a short, echeloned wig, back pillar and in each hand a snb-cloth. The statue has been cited several times in passing as a notable example of Saite archaism (e.g. Jones 1990, 36–7; Robins 1997, 227; Russmann 2001, 230–1, cat. 123; Cole 2018, 108, cat. 76), but the statue2 and its inscriptions3 have never received full publication. The statue is typical of the majority of such museum pieces in that it is without firm archaeological context. It is said to be from Giza and was acquired in 1921 through the agency of the Cairo-based dealer, Panayotis

2

3

4

BM EA 1682: Hall 1930, 171–2, fig. 93; Bosse 1936, 17, no. 15. Traces indicate that the statue was at least partially painted. Portions of which are mentioned by De Meulenaere 1965, 21; Perdu 1996, 58, n. M (with partial translation). For the handsome sum of £1,100. Compare this figure with the £1,300 offered for an entire collection of significant objects,

Kyticas.4 Here, I discuss the statue’s form and inscriptions, and consider the resulting implications for meanings of the statue in context.

Inscriptions (Fig. 3) On belt (Fig. 3a) ḥm-nṯr twtw nsw-bỉty P-s-m-ṯ-k ꜥnḫ ḏt ṯꜢỉ-ỉst-n-ỉmw Priest of the statues of the King of Upper and Lower Egypt, Psamtek, living forever (a), Tja-iset-imu (b).

including gilded items, from a Giza merchant in 1909 (Ismail 2011, 385). I am very grateful to Patricia Usick of the Ancient Egypt and Sudan Department of the British Museum for investigating the correspondence relating to the sale of the statue.

24

C. PRICE

Base (Fig. 3b)

m ỉr.ty.fy ḫt n mỉtt=ỉ mr nṯr mꜢ=f as one who will do ritual for my (e) likeness, a god loves the one who looks(?).

(Gourdon 2015, 249–64), but see further discussion below. (b) The owner’s name is written ṯꜢỉ-ỉst-n-ỉmw, with the apparent meaning ‘May Isis seize them’: Ranke 1935, 387, 12; 1952, 398–9. The reed leaf (Gard. M17) is well-attested as a determinative for personal names during early Saite times: e.g. Bothmer 1969, 52. (c) The gs-pr is a common designation in the Old Kingdom for a group of workers: Erman and Grapow 1931, 198.2–16; Strudwick 1985, 222; Jones 2000, no. 969. (d) sỉꜢ-ntt is a rare title attested, for example, on the shabtis of the Dynasty 26–27 priest Djehutyirdis from Tuna el-Gebel (Leer 1957, 20, no. 9). (e) For the reed leaf as the personal pronoun =ỉ, see Leclant 1961, 249–50. (f) Djedu is ancient Busiris in the Delta, the site of Tja-iset-imu’s tomb and the location with which his family was associated (Rodrigo 1999, 240–56). The occurrence of both Osiris of Abydos (Abdju) and of Busiris (Djedu) in the statue’s inscriptions may be a deliberately universalising pairing, and need not give an indication of the origin of the statue itself.

Back pillar (Fig. 3c)

Tja-iset-imu, a priest of royal statues

[1] ḥꜢty-ꜥ ỉmy-r gs-pr ḫtmw-nṯr ỉmy-ỉs ṯꜢỉ-ỉst-n-ỉmw ḏd=f wꜥb nb ꜥḳ.ty.fy [2] r ḥwt-nṯr nt wsỉr nb Ꜣbḏw Noble, overseer of the gs-pr (c), sealer of the god, councillor, Tja-iset-imu, he says: (O) every pure one who will enter into the temple of Osiris Lord of Abydos sḫꜢ=sn rn=ỉ r-gs nṯr nỉs=f kꜢ=ỉ they shall commemorate my name beside the god, he shall invoke my ka m [3] ḏd.ty.fy snw pr m-bꜢḫ ḫft nṯr ḥtp-nṯr ḥr wdḥw=f n ỉmꜢḫ wꜢḥ-ḫt ḫtmw-nṯr ṯꜢỉ-ỉst-n-ỉmw as one who will say ‘snw-offerings gone out in front of the god, offerings upon his offering table’ for the venerated one, god’s offerer, sealer of the god, Tja-iset-imu [4] sꜢ ḫtmw-nṯr sỉꜢ- wꜥb wꜢḥ-ḫt pꜢ-ẖꜥ-r son of the sealer of the god, knower (d), pure priest, offerer, Pa-kha-ru

ḥtp-dỉ-nsw wsỉr nb ḏdw dỉ=f prt-ḫrw t ḥnkt kꜢw Ꜣpdw snw šs snṯr mrḥt An offering which the king gives (to) Osiris, Lord of Djedu (f), that he may give a voice offering of bread and beer, beef and fowl, alabaster, incense, oil m ḫt nb nfr wꜥb rdỉt nb rnpt nb m ḥb nb as everything good and pure (on) every offering, every festival of the year and on every festival n ỉmꜢḫ (ỉmy-)r gs-pr ḫtmw-nṯr ṯꜢỉ-ỉst-n-ỉmw for the venerated one, overseer of the gs-pr, god’s sealer, Tja-iset-imu. Commentary (a) This must refer to King Psamtek I. Following a late Old Kingdom convention, the epithet ꜥnḫ ḏt may be an indication that the named king is still alive

5

Archive P. Lacau, Golenischeff Centre, Paris, photos II, 2 and 4. Compare a fragment in a Japanese private collection, of typically

Tja-iset-imu son of Pa-kha-ru is known from a small number of sources. In addition to the statue in the British Museum, he figures in a genealogy recorded on an offering table seen at auction in Buenos Aires (Rodrigo 1999, 249–50). Fragments of relief from his badly damaged tomb chapel were identified by Pierre Lacau5 at Busiris in the Delta. It is notable that Tja-iset-imu served as an officiant of the ‘statues of the King of Upper and Lower Egypt’, a title only presently attested for him from his British Museum statue. The aforementioned offering table records the same title held by his descendent Wennefer, implying a family concern in the royal statue cult. Several other Saite individuals held titles connected with royal statues (De Meulenaere 2011, 127–32); these relate not only to statues of named rulers, as in the present case, but also to non-specific ‘(royal) statues (twtw)’. Frustratingly for an understanding of the cults,

archaising Saite style, which must come from the same structure: Leclant 1959, 29–30.

A PERFECT ‘LIKENESS’? VIEWING LATE PERIOD ARCHAISING SCULPTURE IN CONTEXT

Fig. 3b: Inscriptions on the upper side of base. London, British Museum (EA 1682). © Trustees of the British Museum.

our chief source for these titles are the terse inscriptions on shabtis of the title-holders (e.g. Janes 2002, 199– 200, cat 101, 204–5, cat. 103). It thus remains unclear whether the statue cult focused on the living king, as may be implied by the use of the term ꜥnḫ ḏt in Tjaiset-imu’s use of the title (see commentary, note [a] above). However, a certain Padineith, vizier under Nectanebo I, was a priest of the statues of King Amasis (ḥm-nṯr n twtw n nỉsw-bỉty wꜢḥ-ỉb-rꜥ) (Bresciani et al. 1983, 44), demonstrating the potential longevity of the statue cult of particular Saite kings.

Fig. 3c: Inscriptions on back pillar. London, British Museum (EA 1682). © Trustees of the British Museum.

25

26

C. PRICE

The recent discovery of portions of a colossal statue of Psamtek I at Heliopolis highlights the existence of well over life-size royal Saite statues (Ashmawy and Raue 2017, 64–5). Although Herodotus (II, 175–6; Rawlinson 1997, 217–8; cf. Lloyd 2007, 372) describes colossal statues at Sais, archaeological evidence from that site is scant and there are no textual attestations for individually named ‘cult colossi’ in the manner of those created for Amenhotep III and Ramesses II (Price 2011b, 403–11). The fact that Tja-iset-imu was involved with the cult of royal images, whether representing a living or deceased king, is of significance for the interpretation of his own statue. As such he is likely to have had more knowledge than most of the proper rituals and interactions expected for statues in context. Situating the statue There is little likelihood that the statue under discussion originated from Tja-iset-imu’s tomb chapel at Busiris; the practice of placing sculpture depicting the deceased in a tomb context seems to have been surprisingly limited in the 1st millennium BC (Bothmer 1969, xxxiii; cf Price 2011a, 173–7). In part this may be because the superstructures of tombs are poorly preserved, especially at Memphis (e.g. Stammers 2009, 26–47), but there is no clear evidence for statue emplacements from what survives. The recent and detailed excavation of the Late Period Assasif necropolis has yielded a similarly striking dearth of statuary, with no evidence of sculptures of the tomb-owners Karabasken, Karakhamun or Nesamenopet discovered in situ (Pischikova 2017, 274). Rather, the almost exclusive context of free-standing non-royal sculpture seems to have been within temple spaces. The purported attribution of Tja-iset-imu’s statue to Giza is plausible, as the site was a well-known centre of activity during Dynasty 26, with contemporary graffiti at the temple of Isis mentioning men with Tja-isetimu’s title wꜢḥ-ḫt (Zivie-Coche 1991, 138, n. e). Specification of ‘Giza’ may, however, belie the statue’s origin from elsewhere in the wider Memphite necropolis, which — despite its common designation as a necropolis — was not only a setting for tombs but also an extensive and active cultic space used by the

6

JE 27972 = CG 1106; Borchardt 1934, 59–60; Jurman 2010, 238.

living during the Late Period (Smith 1974; Nicholson 2016, 19–31). That non-royal statues were set up in this area during Dynasty 26 is implied by an ‘appeal to the living’ on a block statue belonging to a man named Iry, who addresses those ‘who come to the hill to the district of this god, the living Apis … I put this statue in the temple because I did not want to be far from it for the length of eternity’.6 The Memphite necropolis has been particularly prone to disturbance. Intense reuse of spaces, rebuilding and modern explorations with both archaeological and illicit intent have frustrated attempts to trace the precise origins for countless items. Geophysical work carried out by a Scottish mission between 1998 and 2010 revealed evidence of numerous mud-brick structures — representing platform temples and chapels of the Late Period — oriented towards the Serapeum and its approaching processional way (Mathieson and Dittmer 2007, 86–9) (Fig. 4). The functions of these chapels may be compared to cenotaphs at Abydos and Osirian chapels at Karnak (Jurman 2010, 243–4), which both offer possible contexts for non-royal statuary. The well-preserved condition of Tja-iset-imu’s statue may suggest that it was at one point deliberately cached, a situation well-known from both recorded excavations at Saqqara (e.g. Hastings 1997) and also — much more

Fig. 4: Detail of geophysical scan of North Saqqara showing temple platforms of Late Period date.

A PERFECT ‘LIKENESS’? VIEWING LATE PERIOD ARCHAISING SCULPTURE IN CONTEXT

commonly — from an unknowable number of unrecorded contexts.7 As a major elite cemetery for much of pharaonic times, the Memphite necropolis provided a wealth of practical inspiration for later artisans. This phenomenon is well-known, and manifests in the replication of Old Kingdom architectural layouts at Saqqara and Abusir (Stammers 2009, 26–68; Smoláriková 2010, 431–40). The palace gateway of Apries at Mit Rahina (Petrie 1909, pls II–IX; Kaiser 1987, 123–54) plausibly took direct inspiration from the subterranean reliefs of Djoser (Firth and Quibell 1935, pl. 15, 16) and the survival of inked Saite period grids on these Dynasty 3 scenes (Robins 1994, 169–70) suggests mechanisms for transferring two-dimensional motifs. Apries’ royal reliefs also adopted motifs from non-royal Old Kingdom tombs (e.g. Pieke 2017, 267); non-royal Saite tombs — such as that of Tjery at Giza — borrow both textual and figural decoration, with adaptations, from Old and New Kingdom models (el-Sadeek 1984, 217– 63). The motivations for this process of copying and adaptation lay in an interest in — and experience of — individual monuments, and less in emulating the vague ‘spirit’ of particular ages (cf. Price 2017, 395–410), as is often claimed in general surveys of Egyptian art that tend themselves to have a compartmentalised chronological focus. The creator(s) of Tja-iset-imu’s statue could have drawn upon numerous possible referents in accessible Old Kingdom private tombs throughout the Memphite necropolis.8 Familiarity with these structures at Saqqara was no doubt enhanced by investigations for the construction of the vast Sacred Animal Necropolis, which expanded significantly during the Saite period.9 These subterranean tunnels exploited a pre-existing network of Old Kingdom tomb shafts, clearance of which would have exposed decorated surfaces and sculpture partly and fully in the round. While excavating one Old Kingdom tomb chapel at Saqqara, archaeologist Walter Bryan Emery even noted (1965, 6) that a group of engaged statues appeared ‘to have been restored and painted in the Late Period’. A graffito in the name of a Late Period ‘priest of statues’ (ḥm-nṯr n twtw) called

7

8

See, for example the accessible summary in Berman 2014, 150–6, on Auguste Mariette’s excavations of the 1850s, with the mixed nature of material found – including sculpture. In this context, note the intriguing reference by Selim Hassan to a theory held by George Reisner that a group of Old Kingdom statues had been deliberately gathered to ‘decorate’ the Saite

27

Nakhthorheb is known from the Tura-Masara limestone quarries (Usick and Manley 2007, 7, 20 [Sketch 2b]), suggesting the intriguing possibility that someone of Tja-iset-imu’s position may have been involved in acquiring the raw material for his own statue as well as seeking aesthetic inspiration for it. His title ‘overseer of the gs-pr’ may be relevant here (see commentary, note [c]). Viewing the statue in context The British Museum statue of Tja-iset-imu stands out for a number of reasons, and it is worth considering these features of visual salience. His statue is unusual in the context of most surviving Late Period sculpture because it is carved of limestone. While there is doubtless a dearth of preserved limestone sculpture owing to the depredations of the lime kiln and, especially for the Karnak Cachette, the action of subsurface moisture (Price 2016b, 486), limestone may have been employed conspicuously here as the favoured material of Old Kingdom works10 and in contrast to darker stones used more widely in Dynasty 26. The form of the statue has several features that have meaningful referents in much earlier sculpture. Although it would require more extensive survey to quantify, it is worth noting the passing comment of Bothmer (1969, xxxiii) that earlier (i.e. Old–Middle Kingdom) statues do not generally have back pillars that extend above shoulder level; the low back pillar here might have been a deliberate choice to accentuate other old-fashioned features of the piece. The appearance of the shendyt kilt is not in and of itself a striking departure for contemporary elite sculpture, although during the Old Kingdom it is more frequently employed in royal statues. By the Late Period, many non-royal striding statues wear the shendyt and this style of kilt appears to have prevailed during the reign of Psamtek I (Russmann 1973, 38). Most striking about Tja-iset-imu’s appearance is his short, echeloned wig, which had by the Late Period dropped almost completely out of use for sculptures in the round (Perdu 2012, 38), if not for two-dimensional

9

10

temple of Isis at Giza, which Hassan (implausibly) suggests was intended ‘to give it the effect of great age’ (Hassan 1949, 221). For the dating of the creation of the Greater Vaults of Apis in the reign of Psamtek I, see Devauchelle 1994, 100. Perdu 2000, 176–7 suggests that limestone may have been a more easily obtainable substitute for alabaster.

28

C. PRICE

figures — where it featured regularly in archaising reliefs (Bothmer 1985, 101 and pl 4.b).11 This ‘short round tiled’ style had become increasingly popular by the end of the Old Kingdom, with just over half of male statues depicted wearing this style in Dynasty 6 (Tassie 2015, 1884 and fig. 4). The revival of the hairstyle in a small number of sculptures of New Kingdom kings and princes (Bothmer 2004a, 176–7) may have given the wig enhanced significance.12 An instructive parallel for how this style was deployed lies in the self-presentation of Bakenrenef, vizier under Psamtek I. In common with contemporaries at Thebes, the decoration programme of his Saqqara tomb (Bresciani et al. 1988) makes deliberate references to distant pasts — both in terms of date and location. For example, the vaulted burial chamber shows the deceased adoring hours of the day and night and closely replicates scenes in chapels of Hatshepsut and Thutmose III at Deir el-Bahri (el-Naggar 1986, 19–20). A striding statuette of Bakenrenef in dark green schist, less than half the size of Tja-iset-imu’s at around 50cm in height including base, has the same short wig, although without any indication of the echeloned texture.13 A column of inscription on the back pillar includes the title ‘controller of every kilt (ḫrp šnḏyt)’, which does not appear on the more extensive listings of his titles either on the walls of Bakenrenef’s tomb or on his sarcophagus (Pressl 1998, 168–70). ‘Controller of every kilt’ is one of a number of high-ranking titles of the later Old Kingdom associated with attendance on the king (Strudwick 1985, 316), but it is rarely attested afterwards. Its use on Bakenrenef’s statue — which itself wears a shendyt kilt — may therefore be a deliberately archaism, deriving inspiration from accessible Old Kingdom monuments of this time. Rather more explicitly, a kneeling statue of Bakenrenef has a short wig but with a ‘side-lock’ — rare in sculpture and associated with the ritual role of ‘Iunmutef’ (‘pillar of his mother’) — and carries the designation ‘Iun-mutef’14 in the title-string on the back

pillar. As both title and side-lock are extremely unusual in the Late Period, the inscription reflects the iconography of the statue. The same sort of visual punning has been suggested for other Saite sculptures, such as the striding statue of Peftjauemtawyneith, which holds a naos atop the shape of a ḫrp-sign. This has been interpreted as a deliberate reference to the individual’s title ‘controller of the palace’ (ḫrp ꜥḥ) (Bassir 2014, 140 and n. 885), and similar ‘emblematic rebuses’ have been suggested for the iconography of other Late Period statues (e.g. Klotz 2014, 299–300). Such interchange between statue forms and their inscriptions implies a knowledge and intentionality of artisan(s) that surely could only have been fully apprehended by a small number of persons with physical and intellectual access to these objects in their original contexts (cf. Klotz 2012, 144). Such careful, studied responsiveness to earlier works resulted in what in the past has been described, rather dismissively, as the ‘cold, academic school of Saite art’ (Cooney 1950, 203). Doubtless the aim of the Saites themselves was to enhance the efficacy of images through such intericonicity — i.e. the many and varied interrelationships between images.15

11

14

12

13

Compare, for example, the strikingly archaising pair statue Louvre A 89: Russmann 1973, 39, n. 29; Bothmer 2004b, 264, fig. 16.18; Price 2016b, 497, n. 34. Consistent use in both two- and three-dimensional images of the ‘antiquarian’ Prince Khaemwaset may have added special cultural cachet: Charron and Barbotin 2016, cat. nos 3, 5, 7, 51, 60–8, 132 and 135. See also note 14 below. Boston 1970.495: Simpson 1971, 25–30; Bothmer (1985, 100, n. 8) notes that although the head of the statue was broken and reattached, it is original.

‘Likeness’ and presence To my knowledge, Tja-iset-imu’s ‘appeal to the living’ is exceptional in making reference to the statue upon which the text is carved as a mỉtt, conventionally translated as ‘likeness’ in English (Erman and Grapow 1928, 39.12). The use of the term ‘likeness’ in modern visual culture presupposes a mimetic quality: a likeness to life, to a recognisable person or thing. The term is commonly employed in discussions of pharaonic sculpture and is rarely problematised (e.g. Spanel 1988, 1–37, esp. 5). However, a focus on portraiture, especially in statuary of the Late Period, has detracted from understanding the primary function of an Egyptian statue in context: as a perfect, idealised, enduring

15

Brussels MRAH E 7049a + 7049b (formerly Brooklyn no. 82.23): Bothmer 1985, 101, pl. 4c–d on Late Period examples of the side-lock in relief, but without explicitly connecting this to the ‘Iun-mutef’ pose. For the term ‘intericonicity’, and its applicability to New Kingdom tomb chapels see Laboury 2017, 229–58, especially 249; Den Doncker 2017, 333–70.

A PERFECT ‘LIKENESS’? VIEWING LATE PERIOD ARCHAISING SCULPTURE IN CONTEXT

29

body — yet one with the capacity (and intent) to attract attention. Bernard Bothmer was a major proponent of a belief in the deliberate evocation of specific, living people in the treatment of the face in Late Period sculpture.16 A similar approach to Mesopotamian sculpture has been constructively critiqued by Zeinab Bahrani (2014), who highlights the non-mimetic representation of other concerns in sculpture and relief. More Egyptologists ought to heed Bahrani’s (2014, 7) apt quotation of Paul Klee’s dictum, ‘art does not reflect the visible; it makes visible’. Similarly, in pharaonic (especially Late Period) Egypt the intention was not for the statue to look ‘like’ an identifiable living individual. Egyptian visual culture is most often self-referential, more concerned to emulate other ‘art’ than nature. At temples such as Karnak, groups of statues — exemplified by material brought to light in the Cachette — related to and influenced each other in complex ways (e.g. Jansen-Winkeln 2007, 49–79; Price 2011a). It is less straightforward to envision how Tja-iset-imu’s essentially unprovenanced statue fits a certain temple context, but this should not be to deny that it once had one. Neither is the Egyptians’ own terminology entirely distinct. The king is in some contexts said to be the mỉtt ‘likeness’ of certain gods, usually the sun god Ra, with obvious theological implications (Ockinga 1984, 80–100). However, the term is actually only very rarely used in reference to statues.17 One context in which a statue is ‘likened’ (stwt, lit.: ‘cause to be a twt’, ‘cause to be like’ or ‘resemble’; Erman and Grapow 1930, 335.1-11) to the deceased is in the twelfth episode of the Opening of the Mouth ritual, when the son (or sem-priest) requests the sculptors to make a ‘likeness’ of the deceased (Fischer-Elfert 1998, 42–6. n. 109). The emphasis here is on making the deceased more like a durable statue — the central function of mummification, to turn corruptible flesh into a perfect, eternal image.18 Another example where a physical image seems to be referred to occurs in a text of Amenhotep son of Hapu who, in his capacity as Overseer of Works,

describes how ‘I acted with the love of my heart in my controlling of his likeness in this his great temple, in every valuable and hard stone’ (ỉr.n=ỉ m mryt ỉb=ỉ ḥr ḫrp=ỉ mỉty=f m ḥwt=f tn wrt m ꜥꜢt nbt rwdt).19 The many and varied surviving images of Amenhotep III eloquently show how disconnected ancient sculptures are from the modern expectation of a true, consistent ‘likeness’ (Kozloff and Bryan 1992, 125–9). For Late Period private individuals, the word ‘mỉtt’ is most often used in inscriptions to make a comparison with past times or actions. For example, this is formulated on a Dynasty 30 statue from the Karnak Cachette. The statue owner voices an ‘appeal to the living’ describing proper behaviour vis-à-vis people of the past (and their statue forms):20

16

19

17

18

See, for example Bothmer 1969, 25, on ‘realistic’ early Saite faces. For example, in a review of terms applied to statues, Ramadan (2004, 61–6) makes no mention of mỉtt. For an excellent discussion, see Riggs 2014, esp. 144–51. The term stwt is discussed usefully by Henry Fischer (1963, 27–8).

‘He says: O initiates of Amun, lector priests and pure priests of the sacred place, you shall say “He shall favour you” for this statue; you shall bend your arms for it, under a bouquet for His Majesty. I am a man who is acted for, I did likewise for the ancestors’ (ỉnk s ỉr n=f ỉr=ỉ mỉtt n ỉmyw-ḥꜢt).

This appeal forms part of a rhetoric of reciprocity that is especially pronounced in Late Period non-royal monumental texts, exhorting passers-by to do good for the deceased and anticipating that the deceased will in turn do good for others (Jansen-Winkeln 1999, 45–93; Perdu 2000, 185–91; Price 2011a, esp. 229–37). Certain statues of individuals of the past might receive particular attention; archaism thus aimed to emulate the forms of (and the attendant cultic actions towards) these ‘successful’ statues, in order to encourage similar encounters (Price 2017, 398–9). An eloquent expression of this relationship to past forms occurs in the extensive dedication inscription on a markedly archaising group statue from Memphis, of Dynasty 25 date; the dedicant, a Chief of Sculptors named Ankh-sheshonq, describes the statue of his son in the family group:21 ‘His image as perfect as the ancestors to whom offeringloaves are given in the temple of Ptah’ (kỉ=f nfr mỉ tpyw-ꜥ dỉw n=f snw m ḥwt-ptḥ).

20 21

CG 583 + CG 835: Sethe 1906, 1820–1. JE 37883: Jansen-Winkeln 2001, 125, 388; Price 2011a, 48–9. JE 36728: Vernus 1976, 5–11; Price 2011a, 87–8; 2017, 402–3.

30

C. PRICE

The key issue is that Late Period individuals could achieve a certain perfection by making their statue forms resemble ancient ones, of proven effectiveness as a durable image that might still attract a regular offering service in the temple. At no point was there any requirement or desire that this form be a portrait likeness from life; rather to be ‘perfect’ (nfr) is to be like the enduring statues of the ancestors. As Jones (1990, 36) notes of Tja-iset-imu’s statue, there is no apparent intent to deceive the viewer about the antiquity of the piece and the inscriptions do not appear to be a later addition onto an already ancient statue. Why else go to this effort but for an expected reaction from a viewer — living, dead or divine? Conditions of seeing The text on the base of Tja-iset-imu’s statue — albeit with somewhat unclear grammar — concludes on the concept of (and implies the importance of) the act of ‘seeing’, ‘looking’ (mꜢꜢ). Such a desire for the living to look and notice the statue — aesthetic appreciation — is often underestimated in context; the oft-quoted ‘functional’ intention of a statue does not reduce this. The impact of statues like that of Tja-iset-imu on living viewers would have been dependent on their physical setting and what other kinds of material surrounded them (Price 2016b, 498–9; 2017, 396). The expectations of the modern museum-goer are not necessarily the same as someone frequenting an ancient temple full of statuary, like Karnak, and a number of factors would have affected the conditions of visibility;22 optimal display conditions need not always have been achieved. Of course, we cannot fully account for what we lack from such a setting. We must also be aware of the artificially global view of modern researchers — with ready access to detailed photographs of thousands of well-lit pieces — as compared to the realities of gaining access to statues in a functioning temple. For example, there might have been countless wooden statues that have not survived, as implied by Herodotus (II, 143; Rawlinson 1997, 199–200; Moyer 2002, 70–90), which would have altered a viewer’s

22

Compare, for example, the comments of Harry James (1962, 169–70), on whether texts could be seen in a temple context and actually read, and those of David Klotz (2012, 144) on deliberate pleasure in creating obscure phrases and cryptograms.

impression of Tja-iset-imu’s statue had they been set up surrounding his. An expectation existed that the living — i.e. contemporary and future generations with privileged access to sacred space — would get to actually ‘see’ (mꜢꜢ) temple statues. The hope for such visual interaction is expressed neatly by a text on the contemporary kneeling naophorous statue of Psamtiksaneith, asking passers-by to: ‘Pay attention (lit. ‘give your eyes’) to this statue (ỉmỉ ỉrty=tn r snn pn).’23 To see (a statue) was both a privilege and a ritualised act, especially in the context of restricted access to sacred space. ‘Seeing the god’ is listed in temple liturgies such as the Daily Cult Ritual (e.g. Eaton 2013, 62–3; cf. Van der Plas 1989, 4–35), and to see the statue of the deceased was a key part of the activation ritual of ‘Opening of the Mouth’ (Fischer-Elfert 1998). Also relevant, especially in relation to engaging with and responding to the past, are visitor graffiti — chiefly of the New Kingdom — from the Memphite necropolis. These are typically formulated as the actions of people coming ‘to see (r mꜢꜢ)’ (e.g. Navrátilová 2007, 166), with an implication of appreciation and respect for standing monuments already considered ancient. Tja-iset-imu’s statue thus enters into this dialectic to an unusual degree, through its archaising form and carrying as it does an apparently unique written reflection on the function of the statue in context. Bibliography Ashmawy, A. and Raue, D. 2017. Ägyptisch-deutsche Ausgrabungen in Heliopolis im Frühjahr 2017. Sokar 34, 64–5. Bahrani, Z. 2014. The infinite image: Art, time and the aesthetic dimension in antiquity. Chicago. Bassir, H. 2014. Image and voice in Saite Egypt: Selfpresentations of Neshor named Psamtikmenkhib and Payeftjauemawyneith. Tucson. Berman, L. 2014. The priest, the prince, and the pasha: The life and afterlife of an ancient Egyptian sculpture. Boston. Borchardt, L. 1934. Catalogue général des antiquités égyptiennes du Musée du Caire. Statuen und

23

Philadelphia, Museum of Archaeology and Anthropology 42-91, left side of naos: Ranke 1943, 113.

A PERFECT ‘LIKENESS’? VIEWING LATE PERIOD ARCHAISING SCULPTURE IN CONTEXT

Statuetten von Königen und Privatleuten im Museum von Kairo IV. Berlin. Bosse, K. 1936. Die menschliche Figur in der Rundplastik der ägyptischen Spätzeit von der XXII. bis zur XXX. Dynastie. Glückstadt. Bothmer, B. 1969. Egyptian sculpture of the Late Period. 2nd edition. New York. ———. 1985. The Brussels-Brooklyn statue of Bakenrenef (Membra Dispersa VI). In P. Posener-Kriéger (ed.), Mélanges Gamal eddin Mokhtar 1. Cairo, 99–103. ———. 2004a. Private sculpture of Dynasty XVIII in Brooklyn. In B. v. Bothmer, Egyptian art: Selected writings of Bernard V. Bothmer. New York, 167–98. ———. 2004b. Apotheosis in Late Egyptian sculpture. In B. v. Bothmer, Egyptian art: selected writings of Bernard V. Bothmer. New York, 249–78. Bresciani, E., Betrò, M. C., Giammarusti, A. and La Torre, C. 1988. Saqqara IV: Tomba di Bakenrenef (L.24). Attività del Cantiere Scuola 1985–1987. Pisa. Bresciani, E., Pernigotti, S., el-Naggar, S. and Silvano, F. 1983. Saqqara I. Tomba di Boccori: la galleria di Padineit, visir di Nectanebo I. Supplemento a Egitto e Vicino Oriente 3 (1980); Serie archeologica 2. Pisa. Charron, A. and Barbotin, C. (eds). 2016. Khâemouaset, le prince archéologue: Savoir et pouvoir à l’époque de Ramsès II. Arles. Cole, S. E. 2018. Statue of Tjayasetimu. In J. Spier, T. Potts, and S. E. Cole (eds), Beyond the Nile. Egypt and the Classical World. Los Angeles. Cooney, J. 1950. Three early Saite tomb reliefs. Journal of Near Eastern Studies 9 (4), 193–203. De Meulenaere, H. 1965. La statue du général Djedptah-iou-ankh (Caire 36949). Bulletin de l’Institut français d’archéologie orientale 63, 19–32. ———. 2011. Les desservants du culte des rois saïtes. In D. Devauchelle (ed.), La XXVIe dynastie, continuités et ruptures: Actes du Colloque international organisé les 26 et 27 novembre 2004 à l’Université Charles-de-Gaulle – Lille 3; promenade saïte avec Jean Yoyotte. Paris, 127–32. Den Doncker, A. 2017. Identifying copies in the private Theban necropolis: Tradition as reception under the influence of self-fashioning processes. In T. Gillen (ed.), (Re)productive traditions in ancient Egypt. Proceedings of the conference held at the University of Liège, 6th–8th February 2013. Liège, 333–70.

31

Devauchelle, D. 1994. Les stèles du Sérapéum de Memphis conservées au musée du Louvre. Egitto e Vicino Oriente 17, 95–114. Eaton, K. 2013. Ancient Egyptian temple ritual: Performance, pattern, and practice. Routledge Studies in Egyptology 1. New York. Emery, W. B. 1965. Preliminary report on the excavations at North Saqqara 1964–1965. Journal of Egyptian Archaeology 51, 3–8. Erman, A. and Grapow, H. 1926-1971. Wörterbuch der aegyptischen Sprache. Berlin. Firth, C. and Quibell, J. 1935. The Step Pyramid. Cairo. Fischer, H. 1963. Varia aegyptiaca. Journal of the American Research Center in Egypt 2, 17–51. Fischer-Elfert, H.-W. 1998. Die Vision von der Statue im Stein: Studien zum altägyptischen Mundöffnungsritual. Schriften der Philosophischhistorischen Klasse der Heidelberger Akademie der Wissenschaften 5. Heidelberg. Gourdon, Y. 2015. Rois vivants et rois défunts dans les inscriptions événementielles de la fin de l’Ancien Empire. In R. Legros (ed.), Cinquante ans d’éternité: Jubilé de la Mission archéologique française de Saqqâra. Mission archéologique de Saqqarah V. Cairo, 249–64. Hall, H. R. 1930. A general introductory guide to the Egyptian collections in the British Museum. London. Hassan, S. 1949. The Sphinx. Its history in light of recent excavations. Cairo. Hastings, E. A. 1997. The sculpture from the Sacred Animal Necropolis at Saqqara 1964–76. Egypt Exploration Society Memoirs 61. London. Ismail, M. 2011. Wallis Budge. Magic and mummies in London and Cairo. Kilkerran. James, T. G. H. 1962. Review of B. Bothmer, 1960, Egyptian sculpture of the Late Period. New York. Journal of Egyptian Archaeology 48, 169–71. Janes, G. 2002. Shabtis. A private view. Ancient Egyptian funerary statuettes in European private collections. Paris. Jansen-Winkeln, K. 1999. Sentenzen und Maximen in den Privatinschriften der ägyptischen Spätzeit. Berlin. ———. 2001. Biographische und religiöse Inschriften der Spätzeit. Ägypten und Altes Testaments 45. Wiesbaden. ———. 2007. Drei Statueninschriften einer Familie aus frühptolemäischer Zeit. Studien zur Altägyptischen Kultur 36, 49–79.

32

C. PRICE

Jones, D. 2000. An index of ancient Egyptian titles, epithets and phrases of the Old Kingdom. 2 vols. BAR International Series 866 (1–2). Oxford. Jones, M. (ed.). 1990. Fake? The art of deception. London. Jurman, C. 2010. Running with Apis. The Memphite Apis cult as a point of reference for social and religious practice in Late Period elite culture. In L. Bareš, F. Coppens and K. Smoláriková (eds), Egypt in transition: Social and religious development of Egypt in the first millennium BCE. Proceedings of an international conference, Prague, September 1–4, 2009. Prague, 224–67. Kaiser, W. 1987. Die dekorierte Torfassade des spätzeitlichen Palastbezirkes von Memphis. Mitteilungen des Deutschen Archäologischen Instituts, Abteilung Kairo 43, 123–54. Klotz, D. 2012. The peculiar naophorous statuette of a Heliopolitan priest: Hannover, Museum August Kestner 1935.200.510. Zeitschrift für Ägyptische Sprache und Altertumskunde 139 (2), 136–44. ———. 2014. Replicas of Shu: On the theological significance of naophorous and theophorous statues. Bulletin de l’Institut français d’archéologie orientale 114 (2), 291–337. Kozloff, A. and Bryan, B. (eds). 1992. Egypt’s dazzling sun: Amenhotep III and his world. Cleveland. Laboury, D. 2017. Tradition and creativity: Toward a study of intericonicity in ancient Egyptian art. In T. Gillen (ed.), (Re)productive traditions in ancient Egypt. Proceedings of the conference held at the University of Liège, 6th–8th February 2013. Liège, 229–58. Leclant, J. 1959. Quelques monuments peu connus de l’art égyptien dans les collections du Japon. Bulletin de la Société française d’égyptologie 28–9, 27–33. ———. 1961. Montouemhat: Quatrième prophète d’Amon, Prince de la ville. Bibliothèque d’étude 35. Cairo. van Leer, W. A. 1957. Egyptische oudheden. Mededelingen en Verhandelingen van het VooraziatischEgyptisch Genootschap ‘Ex Oriente Lux’ 12. Leiden. Lloyd, A. 2007. A commentary on Herodotus: Books I–IV. Oxford. Mathieson, I. and Dittmer, J. 2007. The geophysical survey of North Saqqara, 2001–7. Journal of Egyptian Archaeology 93, 79–93.

Moyer, I. 2002. Herodotus and an Egyptian mirage: The genealogies of the Theban priests. Journal of Hellenic Studies 122, 70–90. el-Naggar, S. 1986. Étude préliminaire d’un ciel voûté de l’hypogée de Bakenrenef (L.24) à Saqqara. Egitto e Vicino Oriente 9, 15–38. Navrátilová, H. 2007. The visitors’ graffiti of Dynasties XVIII and XIX in Abusir and Saqqara. The Visitors’ Graffiti 1. Prague. Nicholson, P. 2016. The Sacred Animal Necropolis at North Saqqara: Narrative of a ritual landscape. In C. Price, R. Forshaw, A. Chamberlain and P. T. Nicholson (eds), Mummies, magic and medicine in ancient Egypt. Multidisciplinary essays for Rosalie David. Manchester, 19–31. Ockinga, B. 1984. Die Gottebenbildlichkeit im Alten Ägypten und im Alten Testament. Ägypten und Altes Testaments 7. Wiesbaden. Perdu, O. 1996. L’avertissement d’Aménirdis Ière sur sa statue Caire JE 3420 (= CG 565). Revue d’égyptologie 46, 43–66. ———. 2000. Florilège d’incitations à agir. Revue d’égyptologie 51, 175–93. ———. 2012. Les statues privées de la fin de l’Égypte pharaonique (1069 av. J.-C.–395 apr. J-C.), tome I: Hommes. Paris. Petrie, W. M. F. 1909. Memphis I. British School of Archaeology in Egypt and Egyptian Research Account 15. London. Pieke, G. 2017. Lost in transformation: Artistic creation between permanence and change. In T. Gillen (ed.), (Re)productive traditions in ancient Egypt. Proceedings of the conference held at the University of Liège, 6th–8th February 2013. Liège, 259–304. Pischikova, E. 2017. Recarved Kushite head from the tomb of Karakhamun? (Further remarks on the dating of the South Asasif necropolis). In E. Pischikova (ed.), Tombs of the South Asasif necropolis: New discoveries and research 2012– 14. Cairo, 259–79. Pressl, D. 1998. Beamte und Soldaten: Die Verwaltung in der 26. Dynastie in Ägypten (664–525 v. Chr.). Frankfurt am Main. Price, C. 2011a. Materiality, archaism and reciprocity: The conceptualisation of the non-royal statue at Karnak during the Late Period (c. 750–30 BC). Unpublished PhD thesis, University of Liverpool. ———. 2011b. Ramesses, ‘King of Kings’: On the context and interpretation of royal colossi. In

A PERFECT ‘LIKENESS’? VIEWING LATE PERIOD ARCHAISING SCULPTURE IN CONTEXT

S. Snape and M. Collier (eds), Ramesside studies in honour of K. A. Kitchen. Bolton, 403–11. ———. 2016a. On the function of ‘healing’ statues. In C. Price, R. Forshaw, A. Chamberlain and P. T. Nicholson (eds), Mummies, magic and medicine in ancient Egypt: Multidisciplinary essays for Rosalie David. Manchester, 269–83. ———. 2016b. Archaism and filial piety: An unusual pair statue from the Cachette (JE 37136). In L. Coulon (ed.), La Cachette de Karnak. Nouvelles perspectives sur les découvertes de Georges Legrain. Cairo, 485–503. ———. 2017. ‘His Image as Perfect as the Ancestors’: On the transmission of forms in First Millennium BC non-royal sculpture. In T. Gillen (ed.), (Re)productive traditions in ancient Egypt. Proceedings of the conference held at the University of Liège, 6th– 8th February 2013. Liège, 395–410. Ramadan, W. 2004. Les désignations des statues dans l’Égypte ancienne. Discussions in Egyptology 59, 61–6. Ranke, H. 1935. Die Ägyptischen Personennamen I: Verzeichnis der Namen. Glückstadt. ———. 1943. Eine spätsaitische Statue in Philadelphia. Mitteilungen des Deutschen Instituts für Ägyptische Altertumskunde in Kairo 12, 107–38. ———. 1952. Die Ägyptischen Personennamen II: Einleitung. Form und Inhalt der Namen. Geschichte der Namen. Vergleiche mit anderen Namen. Glückstadt. Rawlinson, G. 1997. Herodotus. The Histories. London. Riggs, C. 2014. Unwrapping ancient Egypt. London. Robins, G. 1994. Proportion and style in ancient Egyptian art. Austin, TX; London. ———. 1997. The art of ancient Egypt. London. Rodrigo, A. de. 1999. A priestly family of Busiris in the Saite period. Chronique d’Égypte 74 (148), 240–56. Russmann, E. R. 1973. The statue of Amenemope-emhat. Metropolitan Museum Journal 8, 33–46. ———. 2001. Eternal Egypt: Masterworks of ancient art from the British Museum. Berkeley, CA. el-Sadeek, W. 1984. Twenty-sixth Dynasty necropolis at Gizeh: An analysis of the tomb of Thery and its place in the development of Saite funerary art and architecture. Veröffentlichungen der Institute für

33

Afrikanistik und Ägyptologie der Universität Wien 29. Vienna. Sethe, K. 1906. Urkunden der 18. dynastie. Abteilung IV. Leipzig. Simpson, W. K. 1971. Three Egyptian statues of the seventh and sixth centuries BC in the Boston Museum of Fine Arts. Kêmi 21, 17–33. Smith, H. S. 1974. A visit to ancient Egypt. Warminster. Smoláriková, K. 2010. The phenomenon of archaism in the Saite period funerary architecture. In L. Bareš, F. Coppens and K. Smoláriková (eds), Egypt in transition: Social and religious development of Egypt in the first millennium BCE. Proceedings of an international conference, Prague, September 1–4, 2009. Prague, 431–40. Spanel, D. 1988. Through ancient eyes: Egyptian portraiture; an exhibition organized for the Birmingham Museum of Art, Birmingham, Alabama. Birmingham, AL. Stammers, M. 2009. The elite Late Period Egyptian tombs of Memphis. BAR International Series 1903. Oxford. Strudwick, N. 1985. The administration of Egypt during the Old Kingdom. The highest offices and their holders. London. Tassie, G. J. 2015. ‘I’m Osiris, no I’m Osiris, no I’m Osiris’: Hairstyles and the afterlife. In P. Kousoulis and N. Lazaridis (eds), Proceedings of the Tenth International Congress of Egyptologists, University of the Aegean, Rhodes, 22–29 May 2008, II. Leuven, 1873–95. Usick, P. and Manley, D. 2007. The Sphinx revealed: A forgotten record of pioneering excavations. British Museum Research Publication 164. London. Van der Plas, D. 1989. ‘Voir’ dieu: Quelques observations au sujet de la fonction des sens dans le culte et la dévotion de l’Égypte ancienne. Bulletin de la Société française d’égyptologie 115, 4–35. Verbovsek, A. 2015. Reception and perception. In M. K. Hartwig (ed.), A companion to ancient Egyptian art. Chichester, 141–54. Vernus, P. 1976. Inscriptions de la Troisième Période Intermédiaire (III). Bulletin de l’Institut français d’archéologie orientale 76, 1–15. Zivie-Coche, C. 1991. Giza au premier millénaire: Autour du temple d’Isis Dame des Pyramides. Boston.

SMALL DIVINE STATUETTES: OUTFITTING RELIGION1 Marsha HILL

Abstract With the accrual of new archaeological finds and new textual analyses, and with growing attention to the range, registers and economic aspects of 1st millennium religious practices, it seems a propitious time to reconsider the practice of offering small divine statuary. Religious activities are surveyed briefly to gauge their similarities and distinctions in terms of participation and associated practices. Then the history of the practice of providing small statuary, specifically cupreous metal statuary, is surveyed. From the pharaonic period there is evidence from the inscriptions that divine statuettes naming private provider/beneficiaries might be associated with economic benefits provided to the temples, while material evidence highlights the fact that statuary donations were not guided by the volition of a large pious population regardless of temple requirements, but were intended to play a role. * * * Widely held views surrounding provision of small cupreous divine statuary go back to early periods of discovery when great numbers of statuettes without apparent chronological indications seemed to speak of a corresponding number of individual offerers. Application of the term ‘votive’, a term that has been defined as designating optional offerings, that is, offerings whose characteristic feature is their reflection of individual motivations (Pinch and Waraksa 2009, 2), has cooperated in partitioning the actual extant innumerable statues from broader social or functional contexts. And ancient ritual prescriptions, in their focus on interaction with a single image, do nothing to counter

1

I am grateful to Neal Spencer, Aurélia Masson-Berghoff, Daniela Rosenow and Marcel Marée of the British Museum for inviting me to take part in the stimulating conference at which an early version of this paper was presented, and to Aurélia Masson-Berghoff for her careful reading and references. Tine

a disconnect between the extant multiples and any fabric of ancient practices (Hill 2007). In recent decades, glimpses of different, more complex structures around donation practices have emerged. But the evidence is sporadic and the problem appears multiform. In this discussion I want to view evidence through three different frames that cumulatively suggest small statuary provision was not a matter of individual volition or personal piety, but was essentially structured by the requirements of given temple or cult practices. The first goal is to envision the range of religious practices in the 1st millennium, trying to briefly summarise recent views of each, as the context within which to view the practice of providing small divine statuary. Secondly, I look at the phenomenon of small statuary provision within the trajectory of its own long history. Thirdly, the meaning of certain physical evidence is interrogated. Many questions will, of course, remain: there is no one ‘moment’ in the 1st millennium that can represent the whole, and other types of cupreous statuary need to be integrated. I allude to some of these questions. But these considerations, I hope, can help to evolve a fairly flexible framework for understanding the provision and role of cupreous divine statuary as this complex phenomenon is further analysed. The range of religious practices in the 1st millennium The practice of providing small cupreous statuary evolved alongside other activities related to sustaining and participating in religion, many likewise newly visible in the 1st millennium. All their registers of use (i.e., are they connected with formal or informal

Bagh, Ny Carlsberg Glyptotek, and Olivia Zorn, Ägyptisches Museum, Berlin, kindly provided photographs of pieces in their care. At the Metropolitan Museum of Art, I would like to thank Diana Craig Patch for her continuing support, and Deborah Schorsch for her important input on metalworking.

36

M. HILL

religion); domains (i.e. are they connected with temples, shrines, animal temples and catacombs, tribune activities, waterways or other sacred locales); geographic parameters; and the overlap or distinctiveness of these practices require characterisation. While there are many questions, our understanding of the constellation of practices has implications for the understanding of any one of them. A selection of these that suggest that range, and offer points of reflection with regard to understanding the offering of cupreous statuary, can be briefly reviewed.

non-priestly and non-elite sources might occur, such sources are not often identifiable in the documentation (Scalf 2015, 369–72).

Temple building through the agency of officials

Letters to gods

During the 26th through 30th dynasties there is considerable involvement of officials in temple building and temple maintenance pursuits, which are no longer only couched behind the figure of the king but may be openly claimed (Spencer 2010).

These are found at animal shrines, and seem in general to be from priests (Scalf 2015, 369–71).

Land donations to temples by individuals Donation of land to a temple by an individual, often in support of maintenance of the individual’s funerary cult, and overseen by a temple functionary, was particularly strong from the Third Intermediate Period through the Saite period, when this form of documentation appears to end (Meeks 1979; Moreno Garcia 2013). Animal mummy provision Provision and burial of animal mummies in very large numbers was a widespread practice from Dynasty 26 into the Roman period. Interpretations often accept the numbers as direct reflections of numbers of offerers. However, most recently, analysis by Frédéric Colin (Colin, Adam and Pranjic 2014, esp. 44–51), based on 1st century BC texts from Kom Ombo and some earlier documents, points to the understanding of the mummies as gods and to involvement of public officials and priests in performing and guaranteeing the practice, underscoring arguments that this is not a ‘popular’ practice. A review of texts associated (or potentially associated) with the animal mummies reveals that, while financial donations or prayers from

2

Though Thomas and Villing 2015, 11 and fig. 9 report on pieces that were possibly thrown into the river as offerings.

‘Corn mummy’ provision Images of Osiris or Sokar were prepared from mainly organic materials by temple staff yearly for certain festivals of Osiris, the image from the previous year being buried or otherwise ritually disposed of (Coulon 2015).

Public festivals Great community festivals are increasingly acknowledged as features of Egyptian religious practice in the 1st millennium (Bryan 2014; Thomas 2015), and are reasonably regarded as the generators of large corpora of non-formal styles of images that appear to be associated with myths of certain gods, often associated with the New Year and flood cycles (Bulté 1991, 95–121; Thomas 2015, 42–52; Hill 2016; Goddio and MassonBerghoff 2016, 139–218). How the images might have actually functioned in relation to a festival is unknown,2 as are the ways in which the objects were acquired by participants. On the other hand, some classes of items are also found used in relation to shrines and temples of gods that deal with practical concerns, and also in settlements and cemeteries, as studies of Third Intermediate Period material from the Eastern Delta and Late Period material from Naukratis have noted (Bulté 1991, 119–21; Thomas 2015, 62–5). Oracles The seeking of oracles was an old practice, but becomes increasingly obvious at numerous sites (Volokhine 1998, 82–95). In the Bes chambers housed inside the temenos wall of the Saqqara Anubeion, figurines were found that are typologically similar to some

SMALL DIVINE STATUETTES: OUTFITTING RELIGION

of those also associated with festival culture, possibly ex-votos left by those seeking or receiving dreams. Pilgrimage Studies of pilgrimage in Egypt before the 1st millennium have emphasised that long travel for the purpose of visiting gods was the exception: travel for many people was not simple, and Egyptians tended to adhere to their local gods. To some extent this may have changed in the later 1st millennium, possibly with the integration of foreign groups fascinated by Egyptian gods and remarkable features, or with the growing interest in oracles that often came to be associated with certain locales (Yoyotte 1960; Volokhine 1998). This categorization and survey are admittedly nonsystematic, and the chronological matches imprecise; moreover, each category has its own developmental history and range. Nevertheless, based on the survey, the following can be remarked as potentially helping to situate — by parallel or contrast — cupreous statuary donation practices: •





3

4

A considerable area of formal religious practice, including temples and sacred animal temples and necropoleis, shows support from officials, newly apparent actors in temple provisioning. Integrated among formal religious practices are some that are not familiar, for example, provision of multiples of animal mummies, or yearly renewal of the organic figures — corn mummies — at the centre of the Osiris cult. Sacred space includes different kinds of locales, often more than one at the same time, ranging from temples as houses for maintenance of divine images to shrines or holy spots for encounters with divinities or aspects of divinities who particularly concern themselves with ‘practical’ matters (Budde 2011, 33–4, 299–302). Crossover cannot be excluded, but the different sorts of beliefs have their own cultural coherence that should always be acknowledged as a factor.

Interpretation of private statuary bearing divine images should be considered separately. In many other cultural discussions the term ‘votive’ is linked with specific understandings as ex-voto and expresses an individual’s volition without regard to usability or permanence

37

Providing divine temple statuettes Turning to the practice of providing divine temple statuettes, most often cupreous, in the 1st millennium, we encounter a cultural manifestation that evolved from practices going far back into Egyptian history. Essentially, before the 1st millennium, kings are the ones credited with endowing temples with any divine images, although there are hints of non-royal agency for some images of kings as gods that suggest wide and complex economic underpinnings for the practice (Fitzenreiter 2008; Moreno Garcia 2013, 4–6). Over the course of the 1st millennium divine statuettes proliferate in the material record; these are virtually always inscribed for the general benefit (life, occasionally protection) of a non-royal donor from the represented god, if they are inscribed at all, and are usually cupreous if they are preserved.3 The focus here, then, is on cupreous divine statuettes. Two divergent models for the accrual of the statuary are argued by the two modern excavators of sites that bring these practices to the fore, both sites where offering of statuettes and animal mummies is intertwined. I sketch the models only briefly, not to evaluate the excavator’s own understandings of their sites, but to set forward their propositions and the types of evidence that lie behind them. One model follows an understanding of the statuettes as votives.4 For example, at the Sacred Animal Necropolis at North Saqqara, statuettes and animal mummies are envisioned as dedications presented directly at the shrines and temples by a wide range of individuals in a manifestation of personal piety. Evidence from the larger site, including dream texts, oracle appeals and the like addressed by individuals to the sacred animals and associated deities, is cited in support of this view, although none of these documents were directly associated with the statuettes or animal mummies. Most recent accounts have also introduced the possibility of festivals as occasions for offerings, and pondered social compulsions that might exist (Davies and Smith 1997; Davies 2007).

(Rouse 1908; Votives Project 2017; Weinryb 2017). The Egyptian evidence relating to divine statuary is silent about the first, and the usual cupreous material, its form, and its value do not accord well with the second (Kemp 1991, 35).

38

M. HILL

Another model sees the statuettes and animal mummies as evidence of participation by elite and priestly classes in rituals associated with festivals related to the ibis cult. Evidence at Tuna el-Gebel points to preparation and placement of provided statuettes and mummies as being highly restricted without any participation by large populations or demonstration of personal piety. Rather, the model at Tuna envisions priestly offerings of the animal mummies and cupreous images, and believes that by Ptolemaic times ibis cult organisation is influenced strongly by royal cults and festivals. Interestingly, the excavator has recently suggested that certain festivals might be the occasion for accrual of large numbers of bronze statuettes, discussed further below (Kessler 2008).5 Attention to the history of the practice of donating small statuary (the majority of it preserved being cupreous metal), and to features of the cupreous statuary itself can make a contribution. Historical considerations Most cupreous statuary of divinities from all periods either does not retain inscriptions that were originally on separable bases, or did not have inscriptions originally; it is difficult to know which might be the case, and whether this changes over time. Moreover, even when inscriptions do exist, the use of inscriptions for dating depends on datable persons, name types or other established criteria. Similarly, assignment to a date by stylistic criteria is only possible in some cases (Hill and Schorsch 2007, 3–4, 117, 125), and is only of use in evaluating the nature of persons providing a statue when the statue also retains an inscription. All disclaimers aside, it is possible to say a few things about some of the people involved in statue provision and about changes through time based on two partial inscriptional corpora (Jansen-Winkeln 2007; De Meulenaere 1990; 1993). Numbers in square brackets in the following discussion refer to a Concordance to these corpora at the end of this paper. Owing to the large numbers of statuettes and problems of accessibility and dating, it is scarcely feasible to give a fuller

5

Other reviews incorporating the Tuna evidence with other evidence offer somewhat modified views of Tuna and the practice of providing cupreous divine statuary (Weiss 2012, 463–81; Fitzenreiter 2014), while recent excavations reveal details of a settlement of Tuna priestly attendants (Flossmann 2014).

picture, although more and more statues with their inscriptions are becoming available.6 Pre-Saite cupreous statuary For the Third Intermediate Period, including the Kushite period, a small set of eight dated inscribed divine statuettes can be analysed as belonging to one of four categories: 1. They bear royal and quasi-royal names alone [2, 3, perhaps 6]. None are of known kings, but kinglets, chiefs and God’s Wives are known. Particularly in the case of the last, it is possible or even likely that a second stage of an elaborate base existed that would place the statue in the next category. 2. They may bear royal names, usually as beloved of the god represented by the statuette as honorees with titled persons named as the beneficiary. Except in one instance (Fig. 1), the royal person honoured is a God’s Wife, and in all those cases the beneficiary is clearly a member of the God’s Wife’s retinue [1, 4, 5, 7]. 3. They may bear only the names of relatively highstatus beneficiaries [8]. 4. They may retain no inscription at all. A similar range of possibilities is seen in preserved precious metal statuary, which at this time is represented mainly by temple pendants: four kings/kinglets/ chiefs, one high-ranking woman (Hill 2015, 36–7; to which add Schulz 2006). Gods are Amun, Amenemope, Mut and Khonsu, and by adding in temple pendants Herishef, Bastet, Nephthys. None of these, of course, are gods associated with personal piety. Saite cupreous statuary In the Saite period there is quite an increase in statuary, and there is a related increase in statuettes datable by inscriptions, of which seventy-three are noted in the reference corpora used here. The examples are discussed here in the same four categories relating to the

6

Gradually the Metropolitan Museum of Art is putting photographs and inscriptional records for its metal statuary into the online database.

SMALL DIVINE STATUETTES: OUTFITTING RELIGION

Fig. 1: Mut, honouring Osorkon I, with the provider/ beneficiary the King’s-Son-of-Ramesses and Infantryman Djedbastetiufankh, born of the woman Ta(net)ptah. H. 24cm. Copenhagen, Ny Carlsberg Glyptotek, A.E.I.N. 1008. Photograph: Ole Haupt.

beneficiaries/providers of the statuary, as these offer points of reflection. 1. They may bear royal names alone (uncertain, see the next point). 2. The statuettes may a) mention or honour the king and a provider/beneficiary; or b) mention or honour a God’s Wife with members of her administration or retinue in the provider/beneficiary role (Fig. 2).

7

The examples with private names probably also date to Dynasty 26 (Delvaux 1998, 561), but Delvaux did not broach the matter and Jansen-Winkeln did not include them.

39

Fig. 2: Amunrekamutef, honouring the God’s Wife Ankhnesneferibre, and the provider/beneficiary the Chancellor of the God’s Wife Tjahorkhepesh, son of the Chancellor of the God’s Wife Ibi and Ibet. H. 21.5cm. London, The British Museum, EA 60042. © Trustees of the British Museum.

Seven of the seventeen statues mentioning the king belong to the group of very large bronzes of Wadjet and Horus of Buto, probably from Sais (Delvaux 1998); these seven retain more or less clear names of Saite kings on the side of thrones, but one pictures a non-royal offerer [14] and another names a private provider/beneficiary [10], while many private providers/beneficiaries are named among the related group.7

40

M. HILL

Fig. 3: Horus of Hebenu, with the provider/beneficiary Padiaset son of Ankhor born of Teneta... and an agent Chief... of Horus of Hebenu, Djedhor son of Djedasetiufankh. H. 24.4cm. New York, The Metropolitan Museum of Art, Rogers Fund, 1945, 45.2.11.

Therefore, it is difficult to be sure that original complete statues did not all include private providers. Other statuettes naming kings alongside provider/ beneficiaries include Neith, a Bes-Pantheos image named as Horus the Elder, Isis and Horus, Amun, Harpachered and Osiris. A number of these are of unusual size or intricacy. The nine statuettes naming Divine Wives and provided by those in their administration or in their retinue continue the Dynasty 25 tradition. The represented gods include Amun, Amunkamutef, Harpachered-the-Great, the Ancient, the first of Amun (Traunecker 2002), but also Bastet, and Isis with Horus. Most of these statuettes were originally elaborately inlaid like those of the Third Intermediate Period. 3. The remainder mention individuals only, dated: a) by factors such as the inclusion in their names of a Dynasty 26 royal name (Jansen-Winkeln 2007, IV, LXIX–LXX); or b) by the fact that they include an agent of the gift, a practice datable to Dynasty 26 (Fig. 3).8 Class 3b constitutes a formula first noted and explicated by Herman De Meulenaere (De Meulenaere 1990). He saw a structural identity with donation stelae (Meeks 1979) that helped to explain what had often been very confusing texts: the texts refer to an agent who is a member of the ancillary temple staff — door keeper, dancer, trumpet player — who in the case of the bronzes makes the donation on behalf of the provider. The mark of the agent is the locution that follows the give life formula; this locution is generally introduced by the arm and bread-loaf sign for dỉ(t) with possibly following prepositions. Subsequently Frédéric Colin (Colin 1998, 341–54) published a donation bronze base for Thoth, given by a Great Chief of the Two Desert Countries Amenhotep, son of the Great Chief of the Two Desert Countries Rerutek, with the agency of the servant of the ibis Amasis [68]. In this study Colin paid close attention to the difference in the phrases describing agency on donation stela and donation bronzes, realising that, in conjunction with the fact that the agents in both come from the same milieus within temples, they suggest a closed circle of actions: the land donor places land under the responsibility of a temple employee and that temple employee makes it his

8

This group includes two silver, rather than cupreous metal, pendants depicting Nefertum [60, 61].

SMALL DIVINE STATUETTES: OUTFITTING RELIGION

or her responsibility to set up a divine statue for the benefit of the land donor. No example of a land donor who is also named as the one benefiting from the blessing of the god on a statuette has come to light, but that is hardly surprising. Colin saw the land donations as incorporating general support for the cult of the statue, and pointed out that there might be other sorts of donations besides land donations that effectively constituted economic benefit/ cult support and warranted a statue dedication on an individual’s behalf. Simply the mention of an agent alerts us to a world of potentially more complicated relationships behind the presentation of the cupreous statuary, which should not, however, be surprising (Fitzenreiter 2008; Agut-Labordère 2013; Moreno Garcia 2013). Already, the donation of bronzes at Thebes by the circle of the God’s Wives (group 2b) is clearly connected in some way with membership in that circle, and temples were the centre of a thriving network of economic arrangements crisscrossing through society in Dynasty 26 (Agut-Labordère 2013). The identity of the gods included in Saite group 3 encompasses a considerable range. Group 3a includes, selectively: Neith, Osiris, Imhotep, Apis, child gods Harpachered and Penefernehem, Isis and Sopdu. Group 3b includes, selectively: Neith, Nekhbet, Nefertum, Bastet, Harpachered of Athribis, HorusMysis, Horus of Hebenu, Horus the Great, Hormerty, Khonsu and Thoth. It would be pressing too hard on the highly contingent nature of the available evidence, depending as it does on the rare published collections or purposeful compilations of material, to make much of this. But one can remark that there is not strong evidence among this named group for gods who might be associated with personal piety. As for the provider/ beneficiary, there are few titles overall. In group 3a is found a son of a general [30], three priests [28, 29, 39] and a daughter of a temple door’s opener(?) [40]. Group 3b comprises a general [46], a son of a general [45], a priest [59], a son of a singer in the temple [62], and a Great Chief of the Two Borders [68]. 4. Statuettes certainly exist that do not preserve inscriptions, or that did not originally have any inscription at all.

9

For example, Metropolitan Museum of Art acc. no. 42.2.3, visible on the collection database.

41

Some are datable by style to this period, as to earlier or later periods. Again, for those without inscriptions it is generally not possible to be sure that they did not have a separate inscribed base of metal or wood, although there are types definitely without inscriptions.9 But it is necessary to always keep this potentially large class in mind in trying to understand the larger phenomenon. Post-Saite cupreous statuary What happened after Dynasty 26 is difficult to bring into focus. Fewer statuettes can be firmly assigned to the period of Dynasty 27–30 or to the Ptolemaic period, as rulers are seldom mentioned or referred to in names, and, inherent in the dispersal of manufacture through networks of different temples with different relationships to any centralised image, it is very difficult or impossible to consistently differentiate Dynasty 26, 30 and Ptolemaic bronzes by style as already discussed. The institution of the Divine Wives has withered and there are no more bronzes of their officials. A few cupreous metal statuettes with inscriptions have been noted for Dynasty 27 (Raven 1992, 530; Thiem 1996, 101 fn. 2, 104–5). There is also a remarkable archaeological discovery of about four hundred bronze Osirises in use context at Ayn Manawir datable to the late 5th or early 4th century BC, although they bear no inscriptions and nothing survives to indicate whether these might have had inscribed wooden bases (Wuttmann, Coulon and Gombert 2007; GombertMeurice in this volume). For Dynasty 30 into the early Ptolemaic period, a fine Harpachered termed Lord of Hebyt provided by the son of Harsiese, the Great One (priestly title) of Netjery, has a style appropriate to the 4th century (Hill and Schorsch 2007, 122–3, 210, cat. no. 49) (Fig. 4). For the Ptolemaic period, a very elaborate statue of Horus-Ashakhet 29.2.3 (Fig. 5) can be dated to the 4th–2nd centuries by the names of the individuals on a bronze offering tray, and the deposition of a child god in the Falcon Catacomb at North Saqqara can be dated to about 89 BC (Hill and Schorsch 2007, 195–7, 210, cat. 50; 180, 182, cat. 58). According to the general understanding established by Herman De

42

M. HILL

Meulenaere, there are by that time no more bronzes of class 3b as described above that refer to an agent (De Meulenaere 1990). His limitation of this formula to Dynasty 26 is based essentially on the limitation of donation stelae, which in the form discussed by Meeks basically end with Dynasty 26. However, De Meulenaere does make it clear that the land donation system appears to continue thereafter, although recorded differently. Potentially, too, statue provision still accompanies a land donation but is noted differently, or no longer marked. It is also possible that the statue formula lasts longer than the stela formula, although this is speculation. One instance can be noted where the agent formula appears on a Harpocrates from North

Saqqara Falcon Catacomb Gallery 20, which was stacked between 341 and 245 BC, although it was not recognised as the agent formula at the time of publication. However, the statuette itself closely resembles bronzes securely datable to Dynasty 26 (Hill and Schorsch 2007, 121 cat. no. 47 Amasis) and so it is almost certainly earlier and has been included above among the Dynasty 26 examples of the agent formula [69].

Fig. 4: Harpachered Lord of Hebyt, with the provider/ offerer son of the Great One of Netjery Harsiese. H. 18.2cm. Baltimore, The Walters Art Museum, 54.554.

Fig. 5: Bes-image of Horus-Ashakhet, with the provider/ beneficiary Ibi, son of Pediastarte and Tadiese.... H. 16.8cm. New York, The Metropolitan Museum of Art, Rogers Fund, 1929, 29.2.3.

Discussion Evidence from Dynasty 26 suggests in a number of cases economic benefits — perhaps in connection with

SMALL DIVINE STATUETTES: OUTFITTING RELIGION

membership in the Divine Wife’s circle at Thebes or with a funerary foundation, or other special benefits provided to the temple — which are a reason a divine statue might be installed in one’s name (Agut-Labordère 2013, 1009–26 is suggestive). What is the meaning of the inscribed bronzes with neither Theban divine state connections nor agents of the donation noted? Could it be another way of doing the same thing, perhaps reflecting regional differences? These do not emerge strikingly in the non-systematic and non-comprehensive evidence so far. There is nothing obvious about the gods addressed to suggest the providers are seeking accessible gods, as is sometimes envisioned, and these are not the immediately reflexive tokens of pious visitors as the section on material considerations will indicate. Do these inscriptions imply a closer connection to the temple? Could they be priests, even though not named as such? Or could the providers have offered some other kind of economic benefit to the temple that did not require recourse to recorded quasi-contractual formulations? Festivals or occasions appear to be marked on a few statuettes [29] or equipment (Davies 2007, 184) and might figure somehow in the provider’s motivation. And what is the meaning of the putative statuary without any inscription, group 4? If we refer to ‘Ayn Manawîr (Wuttmann, Coulon and Gombert 2007), to the Serapeum (Ziegler 1981, 41–2; Gombert-Meurice in this volume) or to North Saqqara (Davies 2007, 184), there is always a large residue of bronzes where no ‘source’ is credited. This stratum certainly always exists. One possibility is that much or all should be credited to the fulfilment by temple workshops of the requirements of temple needs not met by ‘subscription’, though other possible economic mechanisms can be considered as possible underwriters. As already noted in connection with Tuna el-Gebel, it has been hypothesised that yearly replacement/renewal of images of the Osirian cycle of gods took place there (Kessler 2008, 160–2; noted Coulon 2015, 297 fn. 9). Particular great divine festivals are imagined to have provided the specific occasion.10

10

Evolution of practices through time and/or cultural melding is also a possibility, but we are ill-equipped to assume or evaluate

43

Material factors Production of the inscriptions, too, provides indications about the offering process. Inscriptions on cupreous statuettes appear generally to have been made in the same way as the surface decoration, that is, on the wax model before casting. At that point it was possible for artists, scribes and artisans to work simply and most directly with metal or even wooden tools to decorate the model. The alternative would be to inscribe the cast metal surface itself, which would require iron tools. Iron tools, as well as iron core supports, had been employed on cupreous statuary by Egyptian artisans very occasionally since at least the mid-9th century BC, but were reserved for repairs or alterations, i.e. cases in which there was no choice but to work directly on the metal surface (Schorsch 2007, 91). Leaving aside the observation of an opposition to the adoption of ferrous metallurgy in ancient Egypt, it is simply far easier to produce a well-written inscription in the yielding surface of a wax model than on a metal one, even using iron tools. A small statue of Amasis in the Metropolitan Museum of Art is an instance where the original inscription on the rear of the belt had been made in the wax model and was superseded by an inscription on the front of the kilt made subsequently directly onto the cast surface (Schorsch 2007, 91). By inference, the latter had to have been done using an iron tool. It would be useful to be able to distinguish visually between linear details and inscriptions that were chased or scored onto a wax model and those made by cupreous or ferrous tools on ancient cast metal surfaces according to objective criteria, but unfortunately, due to accumulations of archaeological corrosion and their subsequent degradation by mechanical and chemical cleaning, these surfaces do not easily lend themselves to such distinctions. Statuettes were almost certainly, then, inscribed at the same time they were cast in the workshops. This implies they were prepared under some measure of temple control, certainly not chosen from an array and then inscribed in front of the donor like a t-shirt at a booth

when our knowledge of ancient religious practice is precarious and subject to distortions as I have discussed.

44

M. HILL

in a county fair. Any inscribed statuette intended for presentation would have to be ‘pre-ordered’. One can comprehend, then, how some kind of priestly/temple need with regard to all this statuary might have been implemented. What would that need have been? Use in particular is a difficult topic to evaluate. There is plenty of representational evidence that small statuary of gods was employed in rituals, but whether the represented statuary includes the small cupreous statues — many with private names — that we actually have or something we more mystically refer to as ‘cult statuary’ is, of course, nowhere written. Textual evidence mainly describes interaction with a specific statue, again usually termed by us a cult statue, which could be some of these statues (since our definitions of cult statue are probably too restrictive), but need not be. But certain kinds of evidence suggest uses of these statuettes in temple activities. For example, one of the large Sais/Buto statuettes referred to retains integrally cast loops for carrying rods that accord with a use in connection with storied oracles of the Butic gods (Delvaux 1998, 565) (Fig. 6). And at ‘Ayn Manawîr statuettes of Osiris were discovered still in the arrangement they had while the temple was active, organised in a temple room around a larger image and with open paths for the circulation of officiants (Wuttmann, Coulon and Gombert 2007, 171).

In sum, there is a need to consider cupreous divine statuary as a culturally acknowledged need in order to sustain and serve temple cult, and mechanisms for its provision as woven into the fabric of the culture and economy as it evolved, and to actively seek explanations for the 1st millennium phenomenon in that direction, rather than envisioning a corpus founded in personal piety and to that extent serving no integrated function.

Conclusion Religious practices in the 1st millennium ranged from those associated with maintaining the god in his/ her temple to public festivals and oracles, and different practices met felt needs ranging from maintenance of cultural norms to personal encounters with the divine. Provision of temple statuettes seems to belong more in the former arena. In addition, there are a whole range of actors in Egyptian religion: the activity of officials and elites in supporting temples is more obvious than heretofore, and there is evidence that small statuary installation could be associated with economic benefactions to the temple. Evidence that the statuary itself was manufactured and inscribed at a locale apart from the provider/beneficiary of the donation further indicates that religious/temple needs, and not personal inclination, were more likely to influence the statuary. Cases in which privately donated or multiples of cupreous divine statuary was arranged and used within the temple for cult underscore the statuary’s purposeful relationship to the temple.

Fig. 6: Lion-headed Horus of Buto, with the provider/ beneficiary Horwedja son of Neithrudj. H. 58cm. Berlin, Ägyptisches Museum, Inv.-Nr. 13788. Photograph: Jürgen Liepe. Copyright SMB Ägyptisches Museum und Papyrussammlung.

SMALL DIVINE STATUETTES: OUTFITTING RELIGION

Concordance of statuettes The statuettes forming the corpus discussed in the text are listed below. The description given here comprises the god, any honoured royal/quasi-royal individual, any provenance information when provided by the sources, and references to the sources (with IS referring to Jansen-Winkeln 2007). The Dynasty 26 non-royal providers/beneficiaries and the agents of the provider/ beneficiary are not specifically listed, but the classes discussed in the text are reflected in the subsections. Third Intermediate Period – Dynasty 22–24 1. Mut, honouring Osorkon I. Copenhagen, Ny Carlsberg Glyptotek, AEIN 1008. IS II, 69, no. 13.47 (see Fig. 1). 2. Amun, honouring Tefnakht. Florence, Museo Archaeologico, 1777. IS II, 270–1, no. 28.11. 3. Amun, honouring Thothemhab. London, British Museum EA 11015. IS II, 367, no. 37.1. Third Intermediate Period – Dynasty 25 4. Base naming Amun, honouring Amenirdis I / Shepenwepet II / Pabatma. Singapore, Collection Loh. IS III, 293–4, no. 51.51. 5. Base naming Khonsu in Thebes Neferhotep, honouring Shepenwepet II. Cairo, Egyptian Museum, JE 37970. IS III, 304, no. 51.70. 6. Montu on a base naming Amun-Ra, honouring Amenirdis I / Shepenwepet II. Paris, Musée Rodin 234. IS III, 304, no. 51.72.11 7. Base naming Khonsu, honouring Amenirdis I / Shepenwepet II. Hanover, Kestner Museum 3494. IS III, 334, no. 51.127. 8. Amenemopet. Cairo, Egyptian Museum, CG 38480. IS III, 565, no. 52.405. Dynasty 26 (Saite period) Kings honoured or named with non-royal beneficiaries/providers 1. Base for a group with Neith, honouring Psamtek I. Berlin, Ägyptisches Museum 16785. IS IV, 38–9, no. 67.

11

Cynthia Sheikholeslami will discuss this bronze in a forthcoming study: she will demonstrate that the base and figure did not originally belong together, and that, while the base honouring

45

2. Neith, honouring Psamtek I . Berlin, Ägyptisches Museum 16784. IS IV, 39, no. 68. 3. Bes-pantheos named as Horus the Elder, honouring Psamtek I. Paris, Louvre 11554. IS IV, 216–7, no. 359. 4. Horus of Buto, name of Nekau II. Private collection. IS IV, 269, no. 5. 5. Base for Neith, honouring Nekau II. Boston, Museum of Fine Arts 1970.637. IS IV, 270, no. 9. 6. Isis and Horus, honouring Nekau II. Leiden, Rijksmuseum van Oudheden, A.B.2. IS IV, 282, no. 49. 7. Amun, honouring Nekau II. Munich, Ägyptisches Sammlung, ÄS 6978. IS IV, 282, no. 52. 8. Base for Neith, honouring Psamtek II. Leiden, Rijksmuseum van Oudheden, E XVIII.27. IS IV, 302, no. 14. 9. Harpachered, honouring Psamtek II. Vienna, Kunsthistorisches Museum 4173. IS IV, 350, no. 135. 10. Horus of Buto, name of Wahibre, Sais. Cairo, Egyptian Museum, CG 38598. IS IV, 816, no. 141. 11. Wadjet, names of Apries. Bologna, Museo Civico Archaeologico 294. IS IV, 351, no. 1. 12. Wadjet, name of Wahibre, Sais. Cairo, Egyptian Museum CG 39080. IS IV, 355, no. 7. 13. Atum or Apries?. Moscow, Hermitage I.1.a.5591 (3783). IS IV, 382, no. 79. 14. Wadjet, names of Apries. Vatican, Museo Gregoriano Egizio 37881. IS IV, 387, no. 107. 15. Osiris, honouring Apries, Serapeum. Cairo, Egyptian Museum CG 38245. IS IV, 404–5, no. 141. 16. Wadjet, names of Apries. Art market. IS VI, 415, no. 5. 17. Wadjet, names of Amasis, Sais. Cairo, Egyptian Museum no number. IS IV, 418–9, no. 17. God’s Wives honoured with their circle as providers/ beneficiaries 18. Isis and Horus. New York, Metropolitan Museum of Art 45.4.3. IS IV, 677, no. 62. 19. Base for Harpachered-the-Great, the ancient, the first of Amun, honouring Nitocris. Paris, Hoffman Collection 366. IS IV, 711, no 82. 20. Base for Amun, honouring Nitocris. London, British Museum EA 60040. IS IV, 712, no. 84.

the God’s Wives named remains relevant to the discussion here, the figure attached to it was no doubt Amun (also shown regularly in a striding pose).

46

M. HILL

21. Amun, honouring Nitocris. London, British Museum EA 63581. IS IV, 748, no. 120. 22. Bastet, honouring Ankhnesneferibre. London, Petrie Museum, UC 36443. IS IV, 748, no. 121. 23. Amunrekamutef, honouring Ankhnesneferibre. London, British Museum EA 60042. IS IV, 750, no. 124 (see Fig. 2). 24. Harpachered-the-Great, the ancient, the First of Amun, honouring Ankhnesneferibre. London, British Museum EA 41607. IS IV, 750–1, no. 125. 25. Amun, honouring Nitocris daughter of Amasis and Ankhnesneferibre. Chicago, Oriental Institute Museum 10584. IS IV, 753, no. 132. 26. Amun, honouring Nitocris, Ankhnesneferibre, etc. Paris, Louvre, A.F.1670. IS IV, 754–5, no. 133. No honoree, non-royal as provider/beneficiaries 27. Harpachered. Cairo, Egyptian Museum CG 38164. IS IV, 816, no. 142. 28. Osiris Wennefer. Copenhagen, Ny Carlsberg Glyptotek AEIN 156. IS IV, 816, no. 143. 29. Bastet. Zurich, Koradi-Berger Collection KB 4016. IS IV, 851, no. 208. 30. Sopdu, private coll. IS IV, 852, no. 212. 31. Apis, Serapeum. Cairo, Egyptian Museum CG 38589. IS IV, 925, no. 316. 32. Falcon, Serapeum. Cairo, Egyptian Museum CG 38128. IS IV, 925, no. 317. 33. Osiris-Ioh-Thoth, Serapeum. Cairo, Egyptian Museum CG 38428. IS IV, 925–6, no. 318. 34. Penefernehem, Serapeum. Cairo, Egyptian Museum CG 38124. IS IV, 926, no. 319. 35. Imhotep, Serapeum. Cairo, Egyptian Museum CG 38047. IS IV, 926, no. 320. 36. Imhotep, Serapeum. Cairo, Egyptian Museum CG 38048. IS IV, 926, no. 320. 37. Osiris Wennefer, Saqqara. Cairo, Egyptian Museum CG 38266. IS IV, 929, no. 332. 38. Neith, Saqqara. Cairo, Egyptian Museum CG 38957. IS IV, 930, no. 334. 39. Mut. Cairo, Egyptian Museum CG 38917bis. IS IV, 1063–4, no. 570. 40. Bes-masked Min-Harnakht, possibly from Thebes. Cairo, Egyptian Museum CG 38836. IS IV, 1100, no. 648. 41. Neith and child god. Formerly in F. G. Hilton Price collection. IS IV, 1160, no. 771. 42. Harpare. Cairo, Egyptian Museum JE 30335. IS IV, 1160, no. 772.

43. Isis. Fecamp, Musée Fecamp AE 01. IS IV, 1167, no. 796. 44. Hathor. Cairo, Egyptian Museum CG 38979. IS IV, 1168, no. 807. No honoree, non-royal as provider/beneficiary, agent on temple staff named 45. Neith, Memphis. New York, Metropolitan Museum of Art 08.202.9. IS IV, 930, no. 333; De Meulenaere 1990, 64, no. 3. 46. Nekhbet. Paris, Louvre E27210. IS IV, 1144, no. 718; De Meulenaere 1990, 64, no. 1. 47. Silver Nefertum, but Bastet his mother is the god invoked. Formerly Resandro Collection. IS IV, 1160, no. 773. 48. Silver Nefertum, but Bastet his mother is the god invoked. Art market. IS IV, 1160, no. 774. 49. Onuris depicted, named as Amun. Formerly Resandro Collection. IS IV, 1168–9, no. 808. 50. Bastet, Bubastis probably. Paris, Louvre E2533. IS IV, 60, no. 110; De Meulenaere 1990, 68, no 12. 51. Child god presumably wrongly on base naming Neith. Brussels, Musées royaux d’art et d’histoire, E6914. De Meulenaere 1990, 64, no. 2. 52. Neith. Baden-Baden, private collection. De Meulenaere 1990, 65, no. 4. 53. Horus-Mysis of Hebenu with antelope beneath and adorant. Baltimore, Walters Art Museum 54.1013. De Meulenaere 1990, 65, no. 5. 54. Horus-Mysis with antelope beneath, Zawiyet el Amwat. Cairo, Egyptian Museum CG 60359. De Meulenaere 1990, 66, no. 6. 55. Harpachered lord of Athribis. Tübingen, Ägyptische Sammlung 435. De Meulenaere 1990, 66, no. 7. 56. Harpachered, connected with Athribis, from Saqqara. Cairo, Egyptian Museum CG 38204. De Meulenaere 1990, 66–7, no. 8. 57. Harpachered, connected with Athribis. Paris, Louvre without number. De Meulenaere 1990, 67, no. 9. 58. Harpacheredeniese on lotus, cult of this god located in the Fayum area – Hawara. Hildesheim, Roemer Pelizaeus-Museum 60. De Meulenaere 1990, 67, no. 10. 59. Khonsupachered, Serapeum. Paris, Louvre N 5137. De Meulenaere 1990, 68, no. 11. 60. Nefertum. Durham, Oriental Museum 1971/61. De Meulenaere 1990, 69, no. 13.

SMALL DIVINE STATUETTES: OUTFITTING RELIGION

61. Nefertum. Cairo, Egyptian Museum CG 38077. De Meulenaere 1990, 69, no. 14. 62. Khonsupachered. Brooklyn TL 70.401.2. De Meulenaere 1990, 69–70, no. 15. 63. Horus the elder crushing an antelope. Leiden, Rijksmuseum van Oudheden A 971. De Meulenaere 1990, 70, no. 16. 64. Harmerti (associated with Horbeit) crushing an antelope. Cairo, Egyptian Museum CG 38618. De Meulenaere 1990, 70–1, no. 17. 65. Sematwypachered on a lotus. Paris, Louvre E3831. De Meulenaere 1990, 71, no. 18. 66. Base for Thoth [ibis] and offerer. Paris, Art Market. De Meulenaere 1990, 71–2, no. 19. 67. Harpachered. London, British Museum WA 132908. De Meulenaere 1990, 73–4, no. 20. 68. Base for Thoth as an ibis and offerer. Paris, Art Market. Colin 1998, 341–2. 69 Harpachered, North Saqqara Falcon catacomb, FCO 462. Davies and Smith 2005, 122–3, pl. LXIIa–b, fig. 28. 70. Thoth as an ibis and offerer. New York, Metropolitan Museum of Art 58.125.4a–c. Previously unpublished. 71. Horus of Hebenu spearing antelope under his feet. New York, Metropolitan Museum of Art 45.2.11 (see Fig. 3). Previously unpublished. 73. Harpachered (associated with Athribis). De Meulenaere 1993, col. 631. 73. Bes-masked Harpachered, Mitrahina. Baltimore, Walters Art Gallery 54.412. Steindorff 1946, 113– 4, no. 625, pls XCV and CXIX. Bibliography Agut-Labordère, D. 2013. The Saite period: The emergence of a Mediterranean power. In J. C. Moreno Garcia (ed.), Ancient Egyptian administration. Handbook of Oriental Studies = Handbuch der Orientalistik. Section 1, Ancient Near East, Vol. 104. Leiden; Boston, 965–1027. Bryan, B. M. 2014. Hatshepsut and cultic revelries in the New Kingdom. In J. M. Galán, B. M. Bryan and P. F. Dorman (eds), Creativity and innovation in the reign of Hatshepsut. Studies in Ancient Oriental Civilization 69. Chicago, 123–69. Budde, D. 2011. Das Götterkind im Tempel, in der Stadt und im Weltgebäude: Eine Studie zu drei Kultobjekten der Hathor von Dendera und zur Theologie der Kindgötter im Griechisch-Römischen

47

Ägypten. Münchner ägyptologische Studien 55. Darmstadt. Bulté, J. 1991. Talismans égyptiens d’heureuse maternité: ‘Faïence’ bleu vert à pois foncés. Paris. Colin, F. 1998. Les fondateurs du sanctuaire d’Amon à Siwa (Désert Libyque): Autour d’un bronze de donation inédit. In W. Clarysse, A. Schoors and H. Willems (eds), Egyptian religion, The last thousand years: Studies dedicated to the memory of Jan Quaegebeur. Orientalia Lovaniensia Analecta 84. Leuven, 329–55. Colin, F., Adam, F. and Pranjic, I. 2014. Harpocrate au Chien et les cadavres de Qasr ῾Allam. Perspectives sur le statut rituel des inhumations animales dans l’Égypte ancienne. Archimède 1, 32–63. . Coulon, L. 2015. Du périssable au cyclique: Les effigies annuelles d’Osiris. In S. Estienne, V. Huet, F. Lissarrague and F. Prost (eds), Figures des dieux: Construire le divin en images. Rennes, 295–318. Davies, S. 2007. Bronzes from the Sacred Animal Necropolis at North Saqqara. In Hill and Schorsch 2007, 174–87. Davies, S. and Smith, H. S. 1997. Sacred animal temples at Saqqara: New discoveries and recent research. In S. Quirke (ed.), The temple in ancient Egypt. London, 112–31. ———. 2005. The Sacred Animal Necropolis at North Saqqara  : The falcon complex and catacomb: The archaeological report. EES Excavation Memoir 73. London. Delvaux, L. 1998. Les bronzes de Sais. Les dieux de Bouto et les rois des marais. In W. Clarysse, A. Schoors and H. Willems (eds), Egyptian religion: The last thousand years: Studies dedicated to the memory of Jan Quaegebeur. Orientalia Lovaniensia Analecta 84. Leuven, 551–68. De Meulenaere, H. 1990. Bronzes égyptiens de donation. Bulletin des musées Royaux d’art et d’histoire 61, 63–81. ———. 1993. Review of Sylvia Schoske – Dietrich Wildung, Gott und Götter im Alten Ägypten. Mainz am Rhein, Verlag Philipp von Zabern. Bibliotheca Orientalis 50 [1992] cols 629–31. Fitzenreiter, M. 2008. Statuenstiftung und religiöses Stiftungswesen im pharaonischen Ägypten.

48

M. HILL

Notizen zum Grab des Pennut (Teil V). In M. Fitzenreiter (ed.), Die Heilige und die Ware zum Spannungsfeld von Religion und Ökonomie, Workshop vom 26.5. bis 28.5.2006. Internet-Beiträge zur Ägyptologie und Sudanarchäologie VII. London, 233–63. ———. 2014. Funktion und Kontext ägyptischer Kleinbronzen. In M. Fitzenreiter, C. E. Loeben, D. Raue and U. Wallenstein (eds), Gegossene Götter. Rahden/Westf., 169–76. Flossmann, M. 2014. Les maisons-tours de l’association religieuse de Touna el-Gebel. In S. Marchi (ed.), Les maisons-tours en Égypte durant la Basse Époque, les périodes ptolémaïque et romaine, Nehet 2. Paris, 9–31. Goddio, F. and Masson-Berghoff, A. 2016. From myth to festivals. In F. Goddio and A. Masson-Berghoff (eds), Sunken cities: Egypt’s lost worlds. London, 139–218. Hill, M. 2007. Lives of the statuary. In Hill and Schorsch 2007, 152–9. ———. 2015. A gilded-silver pendant of Nephthys naming Mereskhonsu. With an appended technical examination by D. Schorsch. Revue d’égyptologie 66, 33–49. ———. 2016. Tribal dynamics, child gods, festivals, and the faraway goddess: Mingling in the Egyptian Delta in the Third Intermediate Period. In J. Aruz and M. Seymour (eds), Assyria to Iberia: Art and culture in the Iron Age. Metropolitan Museum of Art Symposia. New York, 154–67. Hill, M. and Schorsch, D. (eds). 2007. Gifts for the gods: Images from Egyptian temples. New York; New Haven. Jansen-Winkeln, K. 2007–. Inschriften der Spätzeit. 4 vols. Wiesbaden. Kemp, B. 1991. How religious were the ancient Egyptians? Cambridge Archaeological Journal 5 (1), 25–54. Kessler, D. 2008. Einwickeln und unterirdische Ablage von Bronzen im Tierfriedhof von Tuna El-Gebel. In A. Spiekermann (ed.), ‘Zur Zierde Gereich...’ Festschrift Bettina Schmitz zum 60. Gerburtstag am 24. Juli 2008. Hildesheimer ägyptologische Beiträge 50. Hildesheim, 153–63. Meeks, D. 1979. Les donations aux temples dans l’Égypte du Ier millénaire avant J.-C. In E. Lipiński (ed.), State and temple economy in the ancient Near East II. Proceedings of the international conference organized by the Katholieke Universiteit

Leuven from the 10th to the 14th of April, 1978. Orientalia Lovaniensia Analecta 6. Leuven, 605–87. Moreno Garcia, J. C. 2013. Land donations. UCLA Encyclopedia of Egyptology. . Pinch, G. and Waraksa, E. A. 2009. Votive practices. UCLA Encyclopedia of Egyptology. . Raven, M. 1992. A catalogue project of bronzes in Leiden. In Sesto Congresso Internazionale di Egittologia: Atti. Turin, 529–34. Rouse, W. H. D. 1908. Greek votive offerings: An essay in the history of Greek religion. Cambridge. Scalf, F. 2015. Resurrecting an ibis cult: Demotic votive texts from the Oriental Institute Museum of the University of Chicago. In F. Haikal (ed.), Mélanges offerts à Ola El-Aguizy. Bibliothèque d’étude 164. Cairo, 361–88. Schorsch, D. 2007. The manufacture of metal statuary: ‘Seeing the workshops of the temple’. In Hill and Schorsch 2007, 189–99. Schulz, R. 2006. Ein neuer Prinz Shoschenq? In E. Czerny, I. Hein, H. Hunger, D. Melman and A. Schwab (eds), Timelines: Studies in honour of Manfred Bietak. Orientalia Lovaniensia Analecta 149. Leuven; Paris; Dudley, MA, vol. I, 307–8. Spencer, N. 2010. Sustaining Egyptian culture? Nonroyal initiatives in Late Period temple building. In L. Bareš, F. Coppens, and K. Smoláriková (eds), Egypt in transition: Social and religious development of Egypt in the First Millennium BCE. Proceedings of an international conference: Prague, September 1–4, 2009. Prague, 441–90. Steindorff, G. 1947. Catalogue of the Egyptian sculpture in the Walters Art Gallery. Baltimore, MD. Thiem, A.-C. 1996. A brief note concerning the filiation sꜢ nꜥ[-n]-f-ỉꜥḥ msỉ ḥpt.tỉ. Göttinger Miszellen 153, 101–5. Thomas, R. 2015. Egyptian Late Period figures in terracotta and limestone. Naukratis: Greeks in Egypt. Online at: . Thomas, R. and Villing, A. 2015. The harbour of Naukratis, ‘Mistress of Ships’, The British Museum Naukratis Project’s fourth fieldwork season at Kom Ge’if, Egypt (Beheira MSA site no. 100253) April–May 2015. Online at: . Traunecker, C. 2002. À propos de l’Harpocrate amonien. Online at: . Volokhine, Y. 1998. Les déplacements pieux en Égypte pharaonique: Sites et pratiques culturelles. In D. Frankfurter (ed.), Pilgrimage and holy space in late antique Egypt. Religions in the Graeco-Roman World 134. Leiden; Boston, 51–97. Votives Project = The Votives Project: offerings to the gods from Antiquity to the present. 2017. . Weinryb, I. 2017. Votives: Material culture and religion. Material Religion 13 (1), 98–112. Weiss, K. 2012. Ägyptische Tier- und Götterbronzen aus Unterägypten. Untersuchungen zu Typus,

49

Ikonographie und Funktion sowie der Bedeutung innerhalb der Kulturkontakte zu Griechenland. 2 vols. Ägypten und Altes Testament 81. Wiesbaden. Wuttmann, M., Coulon, L. and Gombert, F. 2007. An assemblage of bronze statuettes in a cult context: The temple of ‘Ayn Manâwir. In Hill and Schorsch 2007, 167–73. Yoyotte, J. 1960. Les pèlerinages dans l’Égypte ancienne. In Les Pèlerinages : Égypte ancienne, Israël, Islam, Perse, Inde, Tibet, Indonésie, Madagascar, Chine, Japon. Sources Orientales 3. Paris, 19–74. Ziegler, C. 1981. Une découverte inédite de Mariette, les bronzes du Sérapeum. Bulletin de la Société française d’égyptologie 90, 29–45.

II PRODUCTION: TECHNOLOGY AND WORKSHOPS

ORIGINS OF METALS FOR COUNTLESS BRONZES Aurélia MASSON-BERGHOFF and Ernst PERNICKA

Abstract

From Dynasty 26 (664–526 BC) onwards, votive or ritual statuettes in copper alloy began to be produced

in large quantities in Egypt. Caches and other deposits containing hundreds, if not thousands, of them have been discovered all over Egypt, usually inside or in the vicinity of temple complexes or sacred animal necropoleis (Weiss 2012, 381–462; Hill in this volume). Sometimes, the statuettes were carefully individually wrapped in linen and ritually placed around a larger cult statue, as illustrated by a cache in North Saqqara dated to the first half of the 4th century BC (Emery 1970; Davies 2007) and a temple deposit in ‘Ayn Manawîr in the oasis of Kharga dated around the end of the 5th century BC (Wuttmann, Coulon and Gombert 2007; Gombert-Meurice in this volume). Several deposits yielded damaged finds, such as a late 5th- to early 4th-century BC cache partially excavated by Petrie in the trading port of Naukratis (Petrie 1886, 41–2; Masson 2015; Masson-Berghoff 2019). Such practices indicate that even after the end of their uselife, whether pristine or in a poor state of preservation, these bronzes seem to have retained their sacred nature as gifts to the gods or ritual objects. For that reason their metal would not be recycled (Ogden 2000, 156; Spencer 2007, 29), at least in most documented cases. P. Wilson (2019) recently argued for the existence of a foundry reusing numerous sacred bronzes at Sais (see also discussion by Schorsch in this volume).1 With this implicit non-recycling policy in Egypt,2 the upsurge in the production of sacred metal statuary must have required a steady provision of metals, either in the shape of bronze scrap, or as fresh supplies of raw metals that entered the alloy composition of these statuettes. This paper will first present the results of compositional analyses of Late Period statuary in copper alloy, discussing the potential use of recycled material in their production. Textual and archaeological evidence as well as recent lead isotope analyses can provide some insight regarding the provenance of metals

1

2

The proliferation of votive and ritual bronzes during the Late Period would have required a steady supply of metals, in the shape of either bronze scrap or raw materials — not only copper, but also lead, which often formed a high percentage of their alloys. While the use of recycled material in their production remains a possibility, it is not always corroborated through compositional data. With copper and lead remaining essential commodities after the Bronze Age, Egypt’s access to raw materials and the trade in these metals represents a crucial research topic that has been little investigated so far. In addition to a few written sources, recent and ongoing surveys and excavations provide new insights into the exploitation of mines in the Mediterranean world during this period. Scientific analysis finally allows us to determine the origin of the copper and/or lead ores with good probability, as sufficient comparative data from ore deposits of raw copper and lead of the Eastern Mediterranean and the Near East are now available. Recent compositional and isotopic analyses have been carried out on a wide range of Late Period finds, including the Egyptian metal statuettes on which this paper puts a particular emphasis. The lead isotope analyses (LIA) indicate a variety of imported sources. The samples taken from Egyptian bronze statuettes and other sacred finds are consistent with Laurion in Attica, the northern Aegean and western Anatolia, but also possibly Cyprus, Faynan in the Wadi Arabah, the Sinai Peninsula and Iran. The historical implications of such results are tentatively put forward. * * *

The Sais find consists of a mass of melted and broken copper alloy containing many fragmentary statuettes and figure-fittings, alongside bits of burnt mud brick.

The melting down of votive or ritual material to produce new sacred statuettes is attested elsewhere in the ancient world, notably in the Greek cultural area (e.g. Lindström and Pilz 2013, 269).

54

A. MASSON-BERGHOFF AND E. PERNICKA

— principally copper and lead — used in a wide range of copper alloy objects, including Late Period votive or ritual metal statuary. They reveal a rather complex metal trade between Egypt and the Eastern Mediterranean world. However, the limited data available and the difficulty of interpreting the results of some scientific analyses mean that the conclusions reached in this paper are preliminary, and further research on a larger scale will be required.

produced in Middle Egypt show much lower tin and higher lead contents compared with those attributed to from Upper Egypt and even more so with those from Lower Egypt (Table 1). Table 1 Comparing concentrations (in %) of major elements in copper alloy Osiris statuettes according to regions (based on Riederer 1981, 241).4

Alloy recipes for Late Period statuettes: on workshops and the provision of metals Compositional analyses Numerous compositional analyses have been carried out on Late Period cupreous statuary.3 Noteworthy are the early, but substantial, analyses by atomic absorption spectrophotometry that J. Riederer carried out on hundreds of Egyptian statuettes, the bulk of which dated between 700 and 300 BC (see, e.g., results of analyses of 980 statuettes in Riederer 1981). Most of his analyses, however, concerned statuettes from private collections with little information on their findspot. High variations were recorded in the proportion of the main elements — copper (Cu), tin (Sn) and lead (Pb) — with copper varying between 65% and 100%, tin between 0% and 15%, and lead between 0% and 30%. Variations in concentration of trace elements — notably arsenic (As), silver (Ag), antimony (Sb) and iron (Fe) — proved discriminant in grouping Egyptian statuettes together, suggesting in some cases that they could have been produced in the same workshop. Riederer claimed that his analyses revealed regional trends in the concentrations of copper, tin and lead (Riederer 1981, 240–1). A wide regional provenance was attributed to about 180 Osiris statuettes on stylistic grounds, notably on the relative position of the arms of the god. Osiris represents a major theme in metal statuary across Egypt and his statuettes can display a great deal of variation in terms of alloy. However, calculating the average of each group, Riederer observed some key differences. The Osiris statuettes supposedly

3

The non-destructive analyses obtained by a portable instrument (pXRF) recently carried out on a large collection of Egyptian bronzes from the August Kestner Museum (Schulze and Lehmann 2014) are not included in the discussion here, since

Copper (Cu) %

Tin (Sn) %

Lead (Pb) %

Lower Egypt

81.2

6.9

11.1

Middle Egypt

78.7

3.8

16.1

Upper Egypt

82.0

5.1

12.3

Region of origin

Variable access to fresh supplies of metal could represent a first possible interpretation of such results. Tin is usually depleted when bronze scrap is remelted instead of using raw metals (see below). The generally low tin content in Middle Egypt bronze might indicate a higher dependence on bronze scrap in the production of some statuettes, while the Delta and Memphite regions, with their high connectivity with Mediterranean trade during the Late Period, probably had easier access to raw metals. The generally assumed low value of lead in comparison to that of tin could be an alternative explanation. A word of caution on these results and their interpretation is, however, necessary. Iconographic and stylistic features sometimes allow the attribution of an Osiris figure to a specific region in Egypt, but a geographic typology solely based on the position of the arms is now challenged (see for example the numerous types of Osiris discovered in ‘Ayn Manawîr temple context: Gombert-Meurice in this volume). From his analyses, Riederer concluded that ‘it would be highly improbable that [in] one place alloys of such a wide variation were used’ (Riederer 1981, 240). Yet, recent compositional analyses performed on material to which a find-spot can be assigned indicate that the situation is probably a bit more complex, with more variations attested on a single site and even within a single context. A first example is that of a small workshop discovered in a tomb at Qubbet el-Hawa, a necropolis located to the west of modern Aswan. It yielded remains related

4

such a method presents the risk of providing the composition of the corroded or altered surface (Nørgaard 2017; MassonBerghoff et al. 2018, 327). Trace elements are not reproduced in Riederer’s table.

ORIGINS OF METALS FOR COUNTLESS BRONZES

to the production of metal statuary, including wax models and failed castings (Auenmüller and Fitzenreiter 2014; Auenmüller 2014), dated to the 6th– 5th century BC. Two leaded bronze statuettes representing Nefertum and Isis (or Hathor) as well as six moulds mainly used in the production of Osiris statuettes were analysed (Schwab and Willer 2016, 80, table 6.1). Strong discrepancies appear in alloy recipes, including within the same category of finds (i.e. moulds for Osiris statuettes): the content of copper varied from 75% up to 91%, lead from 7.7% up to 21% and tin from 0.6% up to 6.1%. A great disparity of compositions was also revealed for Egyptian votive or ritual bronzes discovered in Naukratis, Egypt’s international harbour town founded in the late 7th century BC (on the site, see Villing et al. 2013–19). Their analyses were conducted in the context of the (Re)sources Project, a collaborative pilot study between the British Museum and the Curt-Engelhorn-Centre for Archaeometry in Mannheim, which looked at a wide range of objects made out of copper alloy, lead or faience (Masson-Berghoff et al. 2018).5 Several types of statuettes and ritual equipment were analysed. Metal statuary consisting of high-quality tin bronze can have a variety of lead and tin ratios, with concentrations of lead below 10% and tin above 4%. The Naukratis tin bronzes include: a figure of Neith of which only the inscribed base and the two feet are preserved (Fig. 1 – Sn: 9%; Pb: 3.8%); an ichneumon figure which used to contain animal remains (Fig. 2 – Sn: 7%; Pb: 9%); and a large votive box topped by the figure of an eel (Fig. 3 – Sn: 4.8%; Pb: 7.3%). Other sacred bronzes with significant tin contents are: a figure in the shape of a human-headed cobra depicting the god Atum (Fig. 4 – Sn: 10.6%; Pb: 2.5%); a wig which forms the figure-fitting of a medium-sized statuette (Fig. 5 – Sn: 8.1%; Pb: 2.5%); and the base of a situla’s model (Fig. 6 – Sn: 9.3%; Pb: 8.8%). All the tin bronzes listed above, except the figure-fitting and the situla, can be attributed to a late 5th–early 4th century BC cache of Late Period bronzes (Masson 2015; Masson-Berghoff 2019). Since this cache is a secondary deposit, some of

5

We would like to thank the Gerda Henkel foundation for their generous grant (AZ07/V/15) which allowed us to carry out this study.

Fig. 1: Statuette of Neith from the ‘cache of bronzes’ in Naukratis. British Museum EA 27577. © Trustees of the British Museum.

Fig. 2: Figure (mummy-case) of an ichneumon from the ‘cache of bronzes’ in Naukratis. British Museum EA 16040. © Trustees of the British Museum.

55

56

A. MASSON-BERGHOFF AND E. PERNICKA

its elements could have been produced long before their burial between the end of Dynasty 27 and beginning of Dynasty 30, but probably not before Dynasty 26 when Naukratis was founded. That cache also yielded an arsenical copper figure of superior quality depicting the male lion-headed deity Mahes (Fig. 7). Arsenical copper was abundantly used in the 4th and early 3rd millennium BC, but is unusual in the Late Period. Accordingly, the high level of arsenic (As: 3.2%; Pb: 2.2%) may indicate that the arsenic was added intentionally to the alloy (possibly by adding speiss, see Rehren, Boscher and Pernicka 2012). The presence of an arsenical copper is somewhat exceptional, since it is seldom found in copper-based alloys used for statuary produced after the New Kingdom (Ogden 2000, 153). It appears, nonetheless, that the choice of such an alloy, which is rather light in colour, remained in favour for certain classes of statuary made of copper alloy. High arsenic concentrations are regularly observed in statuettes depicting the child deity Harpocrates as well as cats and cat-headed deities (e.g. Riederer 1984; 1988). Mahes, a lion-headed deity and son of Bastet, fits rather well within such groups. As Ogden has already noted, compositional analyses of Egyptian bronzes have determined, to some extent, that

Fig. 3: Large eel votive box from the ‘cache of bronzes’ in Naukratis. British Museum EA 27581. © Trustees of the British Museum.

Fig. 4: Statuette of Atum from the ‘cache of bronzes’ in Naukratis. British Museum EA 27597. © Trustees of the British Museum.

ORIGINS OF METALS FOR COUNTLESS BRONZES

Fig. 5: Figure-fitting of a wig from Naukratis. British Museum EA 27599. © Trustees of the British Museum.

Fig. 6: Model of a situla on its base from Naukratis. British Museum EA 27587. © Trustees of the British Museum.

57

the alloy recipe tends to vary depending on the subject represented and that the different compositions could signal ‘traditions of, and constraints on, the local metalworkers as well as aesthetic and colour considerations’ (Ogden 2000, 148–9 with numerous bibliographic references). Finally, low-tin bronzes with a high lead content (above 15%) have been identified among the votive and ritual bronzes found at Naukratis. Often, these specimens are of lesser quality and have multiple comparable examples at the site. None can be given a definite find-spot: despite the cache of bronzes yielding many of their types, they could have been found in other sacred deposits in Naukratis (Masson-Berghoff 2019). It is also challenging to date them precisely — somewhere between the Late and the Ptolemaic periods is the best we can say. These bronzes include: an Osiris statuette (Sn: 3.1%; Pb: 18%); an Isis nursing Horus statuette (Sn: 3.5%; Pb: 15%); the statuette of a sacred bull (Sn: 1.58%; Pb: 15%); a small votive box surmounted by the figure of a cobra (Sn: 2.53%; Pb: 15%); and a decorated model of a situla (Sn: 0.95%; Pb: 22%) (Figs 8–12). Despite the great variability in the alloy recipes, high levels of lead alongside low tin contents are usually deemed typical for statuary copper alloys dated to the Late and Ptolemaic periods (Riederer 1988; Wuttmann, Coulon and Gombert 2007; Schwab and Willer 2016). The high lead proportion (up to c. 30% in some cases)

Fig. 7: Statuette of Mahes from the ‘cache of bronzes’ in Naukratis. British Museum EA 27594. © Trustees of the British Museum.

58

A. MASSON-BERGHOFF AND E. PERNICKA

has sometimes been explained as a way to improve and facilitate the casting of complex bronzes, because a lead-rich alloy combines a greater fluidity with a rather low melting point (Davies and Cowell 1987; Fitzenreiter, Willer and Auenmüller 2016, 133). However, the addition of only a few percent of lead to the metal — as little as 2 or 3% — suffices to achieve such properties (Philip 1991; Klein and Hauptmann 1999). Furthermore, it is clear from the Naukratis example that lead-rich alloys are particularly found in crude, small-sized sacred bronzes, usually produced in series, while higher levels of tin — or even arsenic — were noticed in more exceptionally crafted statuary. It is quite possible that the addition of high levels of lead was due to economic reasons, too (Masson-Berghoff et al. 2018, 327), because lead — as a byproduct of silver

production — was always a rather cheap commodity compared to tin. Thus, statuettes of leaded bronze would have been cheaply produced.

Fig. 8: Statuette of Osiris from Naukratis. British Museum EA 49132. © Trustees of the British Museum.

Fig. 9: Statuette of Isis nursing Horus from Naukratis. British Museum EA 49132. © Trustees of the British Museum.

Recycling material versus using fresh supply Even though sacred bronzes were often withdrawn from metal circulation by ritual deposition, this does not preclude the possibility that recycled metal was used in their production. In the light of the rich body of compositional analyses carried out on Late Period and Ptolemaic cupreous statuary, it seems that the practice existed, but not to a large extent. According to Fitzenreiter, Willer and Auenmüller (2016, 133), the absence of a standard alloy recipe for the statuettes produced in the small workshop at

ORIGINS OF METALS FOR COUNTLESS BRONZES

Qubbet el-Hawa implies that the bronzes contain a very high proportion of recycled metal. They propose, for example, that the copper proportion (91%) found in one of the Osiris moulds — rather high compared with any other analysed samples — may be accounted for by the addition of recycled copper pulp. A widely accepted way to spot the use of scrap in the production of bronzes is a tin proportion inferior to 2%, since such low tin concentrations have practically no recognisable effect on the properties of the alloy (e.g. Rovira and Montero 2003; Figueiredo et al. 2010; Valério et al. 2010; Ashkenazi, Iddan and Tal 2012). It is harder to cast bronze containing less than about 2% of tin, because molten copper can dissolve oxygen and give it off on solidification, producing gas bubbles in the cast (Tafel and Wagenmann 1951). With higher tin concentrations, this is less likely to happen because tin serves as an antioxidant. A high concentration of lead, instead of tin, can partially serve the same function, but tends to make the alloy a less attractive greyish colour. Three out of eight of the analysed samples from the Qubbet el-Hawa workshop contain less than 2% of tin, and only two of them had a tin proportion above 5% (Schwab and Willer 2016, 80, table 6.1). Accordingly, the use of bronze scrap seems likely in the manufacturing of some of the analysed finds. Fitzenreiter, Willer and Auenmüller (2016, 133) suggested: ‘Kleinere Werkstätten werden nur relativ wenig frisches Kupfer zur Verfügung gehabt haben und streckten die Schmelze mit Altmaterial’. The geographic position and/or the type of the workshops where statuettes were produced are probably relevant to procurement strategies of metals. Naukratis, a major Late Period international trading harbour, should have had easier access to fresh supplies of metal. Among the thirty finds in copper alloy analysed in the context of the (Re)Sources Project, five had a low tin content, below 2%, and three of these are ritual or votive bronzes. They include the statuette of Mahes (Sn: 0.38%), the decorated situla model (Sn: 0.95%) and the sacred bull (Sn: 1.58%). Furthermore, the two first cases present compositions which are unusual within the Naukratis corpus of analysed samples. The high levels of arsenic (As: 3.2%), antimony (Sb: 0.34%) and iron (Fe: 0.59%) identified in the Mahes statuette are not consistent with the use of bronze scrap, since all these elements tend to be depleted in the recycling of copper alloys. The low tin concentration, therefore, could possibly result from impurities either from the workshop or

59

Fig. 10: Statuette of a sacred bull (Apis?) from Naukratis. British Museum EA 27598. © Trustees of the British Museum.

Fig. 11: Small votive box topped with a raised cobra figure, from Naukratis. British Museum EA 27579. © Trustees of the British Museum.

Fig. 12: Model of a decorated situla from Naukratis. British Museum EA 27602. © Trustees of the British Museum.

60

A. MASSON-BERGHOFF AND E. PERNICKA

from the ore itself. As for the decorated situla, its high proportion of iron (Fe: 5.14%) is unparalleled within the Naukratis assemblage. If tin was added in the form of cassiterite, a tin oxide mineral, the use of a low-quality concentrate would have introduced more iron than tin. The coexistence of a metal recycling economy alongside the supply of fresh metal has been recently highlighted for the major New Kingdom metallurgical workshops of Qantir, a well-connected international trading port in the Late Bronze Age (Rademakers, Rehren and Pernicka 2017). A similar situation probably existed in Naukratis. However, according to the analysed finds so far, reliance on metal scrap does not appear to have been that widespread in the manufacturing of sacred bronzes. This conclusion finds support in the compositional analyses carried out by Riederer on hundreds, mostly unprovenanced, Late Period and Ptolemaic cupreous statuettes (Riederer 1984; 1988). A large majority contained more than 2% of tin, which would imply that most of them used raw materials. What were the origins of the metals in their production? Sources of metals in the production of sacred bronzes

Other texts of the period relate to metal trade in Western Asia. They include two neo-Babylonian tablets from Uruk (c. 550 BC) that mention copper and iron from Yamana (Oppenheim 1967, 237; Briant and Descat 1998, 96–7). Scholars have proposed different translations for Yamana, either Cyprus (Briant 2002, 383; Kassianidou 2012, 235) or Ionia (Oppenheim 1967, 241 and n. 14a; Van Alfen 2002, 173 n. 615). In a biblical passage (Ezekiel 27:13), copper or bronze artefacts are listed among a variety of goods shipped in and out of Tyre in the early 6th century BC (Van Alfen 2002, 26–7). They are said to come from Yawan (or Javan), Tubal and Meshech, which various specialists have assumed to be regions of Anatolia: Ionia, Cilicia and Phrygia respectively (Diakonoff 1992, 174 n. 29, 181, 185; Greenberg 1997, 551; Van Alfen 2002, 233 n. 839; Stager 2003, 240). Muhly, however, has contested the interpretation of Yaman or Yawan as ‘Classical Ionia’, arguing instead for the regions of Lycia, Cilicia and Cyprus (Muhly 2009, 25–8). A few less ambiguous testimonies corroborate the importance of Cypriot copper, such as a passage in the Odyssey mentioning copper from Temese (identified as Tamassos) and a late 4th-century BC inscription from Eleusis recording the construction of the Stoa of Philon with copper from Marion in the district of Paphos (both texts quoted in Kassianidou 2012, 236–7).

Textual evidence The import of metals in large quantity to Egypt is evidenced by the customs account from Elephantine. The text, probably dating to around 475 BC under the Persian king Xerxes, reports that three Phoenician ships transported 1,764 tonnes (21,000 karsh) of copper (designated as nḥsh sḥlmy’ on the papyrus), alongside other raw metals such as tin and iron (on this text, see Yardeni 1994; Briant and Descat 1998, 66–7; Stager 2003, 242). There has been some discussion as to whether the term sḥlmy’ defines an unknown toponym or should be understood as a label of ‘good quality’ (Briant and Descat 1998, 72 n. 52; Van Alfen 2002, 173 n. 615). Could sḥlmy’ correspond to Salamis, a city-state located on the Cypriot eastern coast (sa-lami-ni-o-se in Ancient Cypriot: O. Masson 1961, nos 392–3)? Iacovou (2002, 79) suggested that this important port of trade must have ‘undertaken the management of the export of the copper of Tamassos’ — a copper-rich region of Cyprus — already in the Archaic period. At the time of the Elephantine customs account, however, Salamis was not controlling the region.

Isotope analyses Lead consists partly of isotopes that are products of radioactive decay of uranium (U) and thorium (Th) that decay into the lead isotopes 206Pb, 207Pb and 208Pb. Lead consists of these three isotopes and a fourth, 204Pb, that is not produced by radioactive decay. Lead deposits can vary in their isotope composition, depending on their geological age and the U/Pb and Th/Pb ratios of the geological reservoir that supplied the lead. When a lead deposit is formed, these elemental ratios are changed by many orders of magnitude so that any further contribution by the decay of uranium and thorium becomes insignificant and the lead isotope composition becomes fixed. It is also then extremely unlikely that the lead isotope ratios will be altered by any of the physical and chemical processes that occur during the metallurgical process from ore to finished artefact, save for the mixing of lead of different origins. The major advantage of lead isotope ratios is that they do not change on the way from ore to artefact. Regardless of the processes involved in the treatment

ORIGINS OF METALS FOR COUNTLESS BRONZES

of ores or metal — whether it is roasting or smelting, cupellation or melting, alloying, dissolution or corrosion — the isotopic composition remains constant. Secondly, sample heterogeneity (which is notorious for making many chemical analyses difficult to interpret) is of no relevance for the isotopic composition. Accordingly, the lead isotope ratios of even a single artefact can be compared with those of ore deposits. If they are different then it can be concluded that the artefact does not derive from that specific ore source. Conversely, it is not possible to regard the provenance of an artefact as proven, even if it shares the same isotopic signature as an ore deposit. The reason for this is that although the variation of lead isotope ratios in ore deposits is much smaller than that of trace element concentrations, there exists the possibility that another deposit has the same lead isotope ratios. As Ben-Yosef underlined in a recent paper, metal provenance studies based on lead isotope analyses (LIA) for ancient Egypt are surprisingly rare (BenYosef 2018, 208–9, fig. 1). The three studies he specifically mentions for the Late Period include a rather early use of LIA by Fleming (1982) on a range of Dynasty 25 finds from Nubia and a few Dynasty 27 finds from Memphis, including statuettes.6 The second one, carried out by Schulze and Lehmann (2014) on eighteen Egyptian statuettes, unfortunately proved methodologically unsound (see below). The third one presented the results of the (Re)sources Project (Masson-Berghoff et al. 2018). It explicitly investigated the question of sources of copper and lead used during the Iron Age and later, particularly in Late Period Egypt. The analysed objects included Egyptian and Egyptianising bronze statuettes and other votive or ritual objects, but also slag, lead ores, weights, arrowheads, stamp-seals, mirrors, vessels, a dowel and faience amulets. To this short list can be added two other recent studies which provide data on lead sources used for the previously mentioned 6th–5th-century BC leaded bronze statuettes from Qubbet el-Hawa (Schwab and Willer 2016) and on 4th-century BC curse tablets in lead likely from Egypt (Vogl et al. 2016).

6

Fleming performed a series of PIXE (proton-induced X-ray emission spectroscopy) and LIA on twenty Egyptian metal objects — statuettes and other sacred finds, tools and weapons (Fleming and Crowfoot-Payne 1979; Fleming 1982). Fourteen objects in bronze and two lead ones originated from two Nubian

61

The provenance of tin is not discussed in this paper (on the difficulty of tracing the provenance of tin from tin isotopes, see most recently Berger et al. 2018). The origins of tin used in the Mediterranean world during antiquity are highly debated (many references are given in Van Alfen 2002, 170–2); supplies could come from eastern sources in Central Asia (Pigott 2011; 2012) or western sources such as Sardinia, Iberia or Cornwall (e.g. Muhly 1973; Valera and Valera 2003; Kassianidou 2012). Suggested sources for copper Data on the sources of copper specifically used in the production of copper-based statuary is very scarce. When compositional analyses reveal a lead content over 4%, it usually means that lead was added to the alloy and does not relate to a natural occurrence of the copper ore. And, as we have seen, numerous Late Period and Ptolemaic bronzes have a rather high lead percentage, often well above that 4% threshold. This means that LIA will in most cases reveal the sources of lead and not that of copper (for a few examples from Naukratis, see Fig. 13). Cyprus Cyprus is often considered to be the main copper supplier of Egypt, although most evidence — be it archaeological, iconographic, textual or scientific — concerns the New Kingdom (e.g. Kassianidou 2009). The production of copper in Cyprus became substantial after the Late Bronze Age. Already in 1982, Koucky and Steinberg had identified the period between the Late Cypro-Archaic and early Hellenistic periods as ‘one of the peaks in Cypriot metal production’ (Koucky and Steinberg 1982, 128, fig. 2). Since then, extensive surveys and recent excavations carried out on mining and smelting sites on the island — and often coupled with radiocarbon dating — have significantly expanded our knowledge of copper mining activities in Cyprus in this period (Kassianidou 2012, 232–5; 2013; 2016;

sites, Temple T at Kawa dated to Dynasty 25 (760–656 BC) and the Treasury of Sanam Abu Dom, occupied around 670 BC, and four bronze finds came from the Palace of Apries in Memphis, dated to Dynasty 27 (526–404 BC).

62

A. MASSON-BERGHOFF AND E. PERNICKA

Peege 2018). Exploitation of the copper-rich central region of Cyprus was especially intensive at the time (e.g. Ben-Yosef et al. 2011). Yet recent assertions regarding the Cypriot origin of copper in numerous Late Period statuettes were made on the basis of flawed data (Schulze and Lehmann 2014). Most of the analysed bronzes were rich in lead, and not only does Cyprus have no known lead deposits to date but its copper is distinctively low in lead (Constantinou 1982; Stos-Gale et al. 1997; Renzi et al. 2018, 278–81). Statuettes in stone and terracotta as well as amphorae and mortaria from Cyprus were discovered in abundance at Naukratis (Thomas 2015; in this volume; Villing 2015). We were, therefore, expecting to identify Cypriot copper in some of the Naukratite finds analysed in the context of the (Re)Sources Project. One unalloyed arrow-head (Cu: 99%) had copper most likely of Cypriot origin (Fig. 14). This leaf-shaped arrow-head belongs to an Egyptian type — which

means that it was probably produced in Egypt — and it can be dated rather early in the Late Period, probably 630–526 BC, considering its context of discovery (Thomas 2017). Another object that seems to match the lead isotope ratios on Cyprus is the statuette of Mahes. However, its lead content (2.16%) would be too high for Cypriot copper. Copper ores from Spain could also offer a good match for the copper of the Mahes statuette. Metal extraction in the Iberian Peninsula and the range of its trade before the Roman period is still poorly understood, although it seems to have been pretty significant (see McConnell et al. 2018 on the intensification of mining in that region in the 1st millennium BC, according to measurements of lead pollution in deep Greenland ice and atmospheric modelling). Some scholars have suggested that Phoenicians who settled in the Iberian Peninsula in the Early Iron Age were interested in its abundant metal resources, especially silver, which would have then been exported to the Eastern Mediterranean (i.e. Aubet 2009;

Fig. 13: Suggested origins of the copper used in sacred bronzes from Naukratis. Map by Aurélia Masson-Berghoff. © Trustees of the British Museum.

ORIGINS OF METALS FOR COUNTLESS BRONZES

Murillo‐Barroso et al. 2016). Very few analyses, however, were conducted to support or refute such a hypothesis.7 To what extent some of this metal could have ended up in Egypt is impossible to tell in the current state of research. We have also discussed the possibility that recycled metal might have been used in the manufacturing of the Mahes statuette, which would introduce a further complexity: in such a case, it becomes even more uncertain from which ore the lead isotope ratios derive.

Fig. 14: Leaf-shaped arrow-head from Naukratis. British Museum EA 27509. © Trustees of the British Museum.

7

The silver of a c. 500 BC diobol from Miletos matches well with the rich deposits of argentiferous galena in Linares, in southeastern Spain (Desaulty et al. 2011; Murillo‐Barroso et al. 2016).

63

Sinai and Eastern Desert Because of the geographic proximity of these regions, copper deposits in the Sinai Peninsula and the Eastern Desert should be considered, even if the extent of their exploitation during the Late Period remains unknown (as already underlined in Schwab and Willer 2016, 76). Resources of copper (and lead) ores available in the Sinai and in the Eastern Desert along the Red Sea coast were not unknown to the ancient Egyptians (Ogden 2000, 149–51). Control over the copper mines in the Sinai, exploited since the earliest times of the Pharaonic civilisation (e.g. Tallet, Marouard and Laisney 2012), was of crucial importance. Their exploitation is predominantly attested in the Bronze Age, especially during the Old and Middle Kingdoms, though Sinai copper was still very much in use during the New Kingdom (Rademakers, Rehren and Pernicka 2017). A recent survey has detected later activity in some copper mines in the southwestern Sinai (Bir Nasib and Wadi Homr), as well as in the northern and central Eastern Desert (Gebel Dara and Wadi Hamama), but they seem to occur during the Third Intermediate Period, the Ptolemaic and/or Roman periods (Abdel-Motelib et al. 2012). Most of these deposits are not as rich in copper as other districts such as Cyprus or the Wadi Arabah (see below), except for Bir Nasib (Hauptmann 2007). The significance of this huge smelting site is rarely recognised (Stos 2009). The lack of evidence for activity during the Late Period could be deceptive, considering how little investigated these regions are. None of the statuettes for which a copper provenance could be established were found to be consistent with what we know so far of the isotopic fields of the Sinai and the Eastern Desert. Still, the results of the analyses carried out in the context of the (Re)sources Project revealed that the copper of a couple of objects has affinities with copper sources in the Sinai. Their lead concentrations are also really low — below 0.1% — which is consistent with the lead concentrations found in copper ores from the Sinai and the Eastern Desert (Abdel-Motelib et al. 2012). One of them is a tin bronze plaque (Pb: 0.04%), discovered in a foundation deposit

64

A. MASSON-BERGHOFF AND E. PERNICKA

dated to Psamtek I (664–610 BC) at Tell Dafana (beneath one of the four corners of the ‘Qasr’, casemate building A: Leclère and Spencer 2014, 54, pl. 17), a settlement in the Eastern Delta (Fig. 15).

(Levy, Ben-Yosef and Najjar 2012). Faynan copper was widely traded, as a number of LIA have recently revealed. It was identified in copper ingots discovered in a Late Bronze Age/Early Iron Age shipwreck sunk off the Carmel coast (Israel) (Yahalom-Mack et al. 2014). It was used in a variety of Early Iron Age metal finds from a Phoenician sanctuary in Sidon (Lebanon) (Vaelske and Bode 2018), as well as in numerous Protogeometric tripod cauldrons dedicated in Greek sanctuaries between 950 and 750 BC (Kiderlen et al. 2016; Kiderlen and Bode forthcoming). Interestingly, tripod cauldrons dated to the Late Geometric (760–700 BC) no longer contain Faynan copper but a fahlore-type copper that could originate from the Alps (Kiderlen and Bode forthcoming). The Late Period statuettes discussed in this paper are even more recent in date. Isotopically, the copper ores from Timna do not really fit the copper-based objects analysed in the (Re)Sources Project, while the copper ores from Faynan provide one of the best overall compatibilities with most of them. These objects include, among others, several Egyptian bronzes from Naukratis (see Fig. 13) — statuettes of Neith, Mahes and Atum and the

Fig. 15: Plaque in tin bronze from a foundation deposit in Tell Dafana. British Museum EA 23556,d. © Trustees of the British Museum.

Wadi Arabah or northwestern Anatolia? Schwab and Willer (2016, 76) have dismissed the possibility that the copper used in the Qubbet el-Hawa statuettes originated from Timna (Israel) or Faynan (Jordan) in the Arabah valley, on the basis that no mining activity has so far been recorded at the time of their manufacture (the 6th–5th century BC). Extraction of copper in the Wadi Arabah during the Third Intermediate Period was definitely significant, with increased activities now recognised in the Early Iron Age at Timna (Ben-Yosef et al. 2012) and Faynan

Fig. 16: Hathoric sistrum from the sanctuary of Aphrodite in Palaepaphos (Cyprus). British Museum 1888,1115.19. © Trustees of the British Museum.

ORIGINS OF METALS FOR COUNTLESS BRONZES

figure-fitting (the wig) — as well as an undoubtedly Egyptian-made sistrum deposited in the sanctuary of Aphrodite in Palaepaphos (Cyprus) (Fig. 16). However, there is, so far, no known copper mining activity in Faynan during the Archaic and Classical periods, and the same copper-based objects can also be found consistent with Anatolian copper, especially with northwestern Anatolian deposits. Now, the copper resources from this region cannot compare with the previously discussed copper-rich regions in Cyprus and the Wadi Arabah. Nonetheless, Greek trading cities located in western Anatolia, which played an active role in trading with Egypt during the Late Period, could have bolstered metal trade from neighbouring mines. Some of the LIA that give us information about lead provenance seem to validate that view (see below).

65

Suggested sources for lead LIA have shown that diverse lead sources were used for the production of lead-rich Egyptian bronzes (for Naukratis statuettes and other sacred bronzes, see Fig. 17).

Not Egypt None of the analysed metal samples match with the isotopic field of the Eastern Desert (Fleming 1982; Schwab and Willer 2016; Masson-Berghoff et al. 2018), although the dearth of LIA data from that region should not be forgotten. Similar results were reached for LIA carried out on Bronze Age metal objects (Shortland 2006).

Fig. 17: Suggested origins of the lead used in sacred bronzes from Naukratis. Map by Aurélia Masson-Berghoff. © Trustees of the British Museum.

66

A. MASSON-BERGHOFF AND E. PERNICKA

Yet, the exploitation of local galena is documented in Dynasty 26. An inscription found at Gebel el-Rosas along the Red Sea coast records a mining expedition sent by the Mayor of Thebes Montuemhat in the year 14 or 15 of the reign of Psamtek I (Vikentiev 1957). Furthermore, two lead ore samples found in Dynasty 26 contexts at Tell Dafana were found to be consistent with Egyptian galena, which is characteristically highly radiogenic (Masson-Berghoff et al. 2018, 331). One of the two samples was buried in one of the foundation deposits of Psamtek I beneath the casemate building A (Leclère and Spencer 2014, 54, pl. 17) (Fig. 18), like the previously mentioned tin bronze plaque for which copper from the Sinai or the Eastern Desert has been suggested.8 Galena extracted from the Eastern Desert mines seems not to have been smelted for lead, but found other applications (notably for cosmetic and medical purposes or as pigment: Hallmann 2009). In the majority of the Late Period metal objects so far analysed, including Egyptian statuettes, lead can be attributed to Attic, north Aegean or Anatolian deposits.

Laurion Lead deriving from the Laurion mines in Attica has been identified in a number of Egyptian bronzes found in Naukratis9 and further south in Qubbet el-Hawa.10 Laurion lead has been determined through LIA for a number of other types of objects, such as early 6th-century BC faience scarabs produced in Naukratis (Masson-Berghoff et al. 2018, 332–3) and 4th-century BC curse tablets in lead (Vogl et al. 2016).11 The import of lead from Laurion to Egypt is also supported by recent discoveries of ingots at ThonisHeracleion — a major harbour town, located to the east of Alexandria, which controlled and guarded the entry point of the Canopic branch of the Nile from Dynasty 26 (Robinson and Goddio 2015). Nineteen lead ingots figured among the numerous ingots discovered during the underwater excavation (Van der Wilt 2010). One of them is a complete loaf-shaped specimen, weighing 23kg and bearing four Greek-style stamps. Comparable stamped lead ingots were retrieved from the Porticello wreck, which sank around the late 5th to early 4th centuries BC (Eiseman and Ridgeway 1987, figs 4-19 and 4-20; Parker 1992, 332–4, no. 879; Van der Wilt 2010, 161–2). LIA have also suggested the mines of Laurion as the source for the Porticello wreck’s ingots (Eiseman and Ridgeway 1987, 57, 107). Northwestern Anatolia and the northern Aegean

Fig. 18: Sample of lead ore from a foundation deposit in Tell Dafana. British Museum EA 23556,l. © Trustees of the British Museum.

8

9

The presence of samples of various materials, notably of semiprecious stones but also of various types of metal, is a common feature in Late Period foundation deposits: Weinstein 1973, 297. They include a statuette representing Isis nursing, the large votive box surmounted by an eel figure and the little situla base (see Fig. 17).

Regarding Egyptian bronzes found in Naukratis, sources located in the northwest region of Anatolia formed the best match for the lead used in the production of the ichneumon figure, while lead sources in Balya in northwestern Anatolia or on Thasos in the northern Aegean were identified for the Osiris and bull statuettes as well as for the votive box topped with a cobra (see Fig. 17). As for Qubbet el-Hawa, the LIA results for a statuette representing Nefertum and samples taken from moulds used to produce Osiris bronzes

10

11

That is the case for a standing Isis or Hathor statuette and samples taken from two different moulds producing Osiris statuettes (Schwab and Willer 2016, 77–9). Even though it was suggested that the curse tablets — which were bought in Egypt — were imported to Egypt from Attica (Vogl et al. 2016, 15–6), it is also possible that they were produced in Egypt using imported raw material.

ORIGINS OF METALS FOR COUNTLESS BRONZES

cluster well with those of lead or lead-rich copper alloy objects from Naukratis, for which the lead was attributed a provenance in Balya or Thasos (MassonBerghoff et al. 2018, fig. 13a). An Anatolian origin was also provisionally assigned to the lead of a sample taken from yet another mould producing Osiris statuettes (Schwab and Willer 2016, 78). While Schwab and Willer recognised a likely Aegean provenance for most lead-rich bronzes, they felt uncomfortable with the Anatolian attribution since they were not aware of any lead exports from that region during the 6th and 5th centuries BC (Schwab and Willer 2016, 80). Lead of northwestern Anatolian or northern Aegean origins was also identified in arrow-heads and a dowel from Naukratis analysed in the (Re)Sources Project, Dynasty 27 arrow-heads from Memphis (Fleming 1982) and arrow-heads of the Persian or early Hellenistic periods from the Southern Levant (Yahalom-Mack et al. forthcoming). Exploitation of lead-silver mines in the northern Aegean and northwestern Anatolia, including in Balya and Thasos, is particularly well-attested between the Archaic and Hellenistic periods (Pernicka et al. 1984; Vavelidis et al. 1988; Kassianidou 2012, 245–6). The vitality of exchanges between Egypt and these regions is illustrated by the numerous silver coins from the Thraco-Macedonian region found in Late Period treasures (Möller 2000, 209; Masson 2016). East Greek traders are often credited with transporting this silver to Egypt in exchange for grain and other commodities (Moreno 2007, 309–15; Van der Wilt 2010). Most of the cities and islands involved in the ‘foundation’ of Naukratis — as reported by Herodotus (Histories 2.178–79) — are situated on the eastern littoral of the Aegean.12 They were all major trading and seafaring powers at the time, and the vast amount of material imported from western Anatolia found at Naukratis is a tribute to their activities at the site (Möller 2000, 75–88; Villing 2013). That material includes ceramics from the Aeolis and the Troad regions (Villing et al. 2013–19), in the northwest of Anatolia, in the hinterland of which the mine of Balya is located. In such a context, the import of lead and possibly copper from this region to Egypt would be consistent with the trading dynamics at the time.

12

From north to south: Mytilene, Phokaia, Chios, Klazomenai, Teos, Samos, Miletos, Halikarnassos, Knidos and Rhodes.

67

Iran Finally, the decorated situla from Naukratis contained lead that seems to originate from Iran (see Fig. 17). It has been found to be notably consistent with either the large deposit of lead and silver at Nakhlak in the western Dasht-e Kavir in central Iran — which has been exploited since the 4th millennium BC and continues to be today — or with litharge (lead oxide from cupellation for silver production) in Arisman south of Kashan in Iran (Pernicka et al. 2011). The exploitation of lead-zinc deposits intensified in Iran during the Achaemenid period (550–330 BC) along with those of gold, silver and copper (Ghorbani 2013, 67–8). LIA results recently obtained on Scythian-type arrow-heads from the southern Levant have found one specimen to be consistent with Iranian lead (Yahalom-Mack et al. forthcoming). Import of Iranian metals to Egypt is a possibility, especially during the periods of Achaemenid rule in the country — during Dynasty 27 (525–404 BC) and Dynasty 31 (343– 332 BC). It would be, however, the first time that such a trade link had been attested in terms of metal. With the possibility that recycled metal was used in the production of this situla (see above) and since recycling (especially repeated recycling) tends to erase information on the sources of metal, LIA results need to be interpreted with extreme caution in this case.

Conclusions The increase in the production of copper-based statuettes in the Late Period is a striking phenomenon in Egypt, especially as this type of (sacred) object was not prone to recycling and has left a great amount of evidence in the archaeological record. The proliferation of such statuettes — alongside the continuous use of copper alloys in a wide range of other commodities — raises the question of the provision of metal for their manufacture. The important compositional fluctuation of Late Period bronzes might indicate variability in metal supply, which included recycling of metal (a priori probably not remelted sacred bronzes) as evidenced by some sacred bronzes with very low tin contents. Alongside local manufacturing and iconographic traditions, economic costs and access to raw metal or recycled material must have played a role in the production of bronze statuettes. Easy access to raw metals versus reliance on recycled metal could have depended on

68

A. MASSON-BERGHOFF AND E. PERNICKA

parameters such as the location and size of the workshops where the statuettes were produced. Many of the recent compositional and isotopic analyses, combined with some textual and archaeological evidence, still suggest a significant supply of raw material in Late Period Egypt, and seem to argue in favour of a diversity of sources of metal. The use of exogenous lead is dominant in the production of statuettes and other ritual or votive bronzes, with suggested sources located in Laurion, the northern Aegean and northwestern Anatolia, and possibly even as far away as Iran. In no cases were local lead ores in the Eastern Desert used. As for the copper, very few data presently exist since the addition of lead in many Late Period sacred bronzes — often in great proportion — prevents the identification of the copper’s origin. According to a few relevant LIA, the statuettes could have used copper from northwestern Anatolia, or possibly from Faynan in Wadi Arabah, although mining activities at the time are so far attested only in the former. The copper in other analysed Late Period metal finds was also found consistent with the isotopic and chemical values in Cyprus, and in the Sinai or Eastern Desert. These results highlight to a certain degree the involvement of Greeks and Phoenicians (as well as Cypriots?) in the metal trade. The corpus of isotopic analyses for Egyptian metal finds dating to the Late Period remains very limited and some mining regions are still poorly investigated. Further analyses from contexts across Egypt and the wider Mediterranean world would be needed to unveil more conclusively the patterns of metal ore exploitation and circulation during the Archaic and Classical periods. Bibliography Abdel-Motelib, A., Bode, M., Hartmann, R., Hartung, U., Hauptmann, A. and Pfeiffer, K. 2012. Archaeometallurgical expeditions to the Sinai Peninsula and the Eastern Desert of Egypt (2006, 2008). Metalla (Bochum) 19, 3–59. Ashkenazi, D., Iddan, N. and Tal, O. 2012. Archaeometallurgical characterization of Hellenistic metal objects: The contribution of the bronze objects from Rishon Le-Zion (Israel). Archaeometry 54 (3), 528–48. Aubet, M. E. 2009. Tiro y las colonias fenicias de Occidente. 3rd edition. Barcelona. Auenmüller, J. 2014. Die Ergebnisse der bildgebenden Verfahren. In M. Fitzenreiter, Ch. E. Loeben,

D. Raue and U. Wallenstein (eds), Gegossene Götter: Metallhandwerk und Massenproduktion im Alten Ägypten. Rahden, 113–26. Auenmüller, J. and Fitzenreiter, M. 2014. Eine Gusswerkstatt auf der Qubbet el-Hawa? In M. Fitzenreiter, Ch. E. Loeben, D. Raue and U. Wallenstein (eds), Gegossene Götter: Metallhandwerk und Massenproduktion im Alten Ägypten. Rahden, 101–6. Ben-Yosef, E. 2018. Provenancing Egyptian metals: A methodological comment. Journal of Archaeological Science 96, 208–15. Ben-Yosef, E., Shaar, R., Tauxe, L., Levy, T.E. and Kassianidou, V. 2011. The Cyprus Archaeomagnetic Project (CAMP): targeting the slag deposits of Cyprus and the Eastern Mediterranean. Antiquity 85 (330). . Ben-Yosef, E., Shaar, R., Tauxe, L. and Ron, H. 2012. A new chronological framework for lron Age copper production at Timna (Israel). Bulletin of the American Schools of Oriental Research 367, 31–71. Berger, D., Figueiredo, E., Brügmann, G. and Pernicka, E. 2018. Tin isotope fractionation during experimental cassiterite smelting and its implication for tracing the tin sources of prehistoric metal artefacts. Journal of Archaeological Science 92, 73–86. Briant, P. 2002. From Cyrus to Alexander: A history of the Persian Empire. Winona Lake. Briant, P. and Descat, R. 1998. Un registre douanier de la satrapie d’Égypte à l’époque achéménide (TAD C3,7). In N. Grimal and B. Menu (eds), Le commerce en Égypte ancienne. Bibliothèque d’étude 121. Cairo, 59–104. Constantinou, G. 1982. Geological features and ancient exploitation of the cupriferous sulphide orebodies of Cyprus. In J. D. Muhly, R. Maddin and V. Karageorghis (eds), Early metallurgy in Cyprus, 4000– 500 B.C. Nicosia, 13–23. Davies, S. 2007. Bronzes from the Sacred Animal Necropolis at North Saqqara. In Hill and Schorsch 2007, 174–87. Davies, W. V. and Cowell, R. M. 1987. Catalogue of Egyptian antiquities in the British Museum. VII, Tools and weapons. 1, Axes. London. Desaulty, A.-M., Telouka, Ph., Albalata, E. and Albarèdea, F. 2011. Isotopic Ag–Cu–Pb record of silver circulation through 16th–18th century Spain. Proceedings of the National Academy of Sciences 108 (22), 9002–7.

ORIGINS OF METALS FOR COUNTLESS BRONZES

Diakonoff, I. M. 1992. The naval power and trade of Tyre. Israel Exploration Journal, 168–93. Eiseman, C. J. and Ridgeway, B. S. 1987. The Porticello shipwreck: A Mediterranean merchant vessel of 415–385 B.C. Nautical Archaeology Series 2. Austin. Emery, W. B. 1970. Preliminary report on the excavation at North Saqqara 1968–9. Journal of Egyptian Archaeology 56, 5–11. Figueiredo, E., Silva, R. J. C., Senna-Martinez, J. C., Fátima Araújo, M., Braz, F. M. and Inêz Vaz, J. 2010. Smelting and recycling evidences from the Late Bronze Age habitat site of Baiões (Visey, Portugal). Journal of Archaeological Science 37 (7), 1632–4. Fitzenreiter, M., Willer, F. and Auenmüller, J. 2016. Beobachtungen zu Technologie und Werkverfahren. In M. Fitzenreiter, F. Willer and J. Auenmüller (eds), Materialien einer Gusswerkstatt von der Qubbet el-Hawa. Bonner Sammlung von Aegyptiaca 7. Berlin, 118–45. Fleming S. J., 1982. Lead isotope analyses of Late Period Egyptian bronzes. MASCA Journal 2 (2), 65–9. Fleming, S. J. and Crowfoot-Payne, J. 1979. PIXE analyses of some Egyptian bronzes of the Late Period. MASCA Journal 1 (2), 46–7. Ghorbani, M. 2013. The economic geology of Iran: Mineral deposits and natural resources. Dordrecht. Greenberg, M. (trans.) 1997. Ezekiel 21–37. The Anchor Bible. New York. Hallmann, A. 2009. Was ancient Egyptian kohl a poison? In J. Popielska-Grzybowska, O. Białostocka and J. Iwaszczuk (eds), Proceedings of the Third Central European Conference of Young Egyptologists. Egypt 2004: Perspectives of research. Warsaw 12–14 May 2004. Pułtuskpp, 69–72. Hauptmann, A. 2007. The early metallurgy of copper. Evidence from Faynan, Jordan. Natural Science in Archaeology. Heidelberg; Berlin; New York. Hill, M. and Schorsch, D. (eds). 2007. Gifts for the gods: Images from Egyptian temples. Metropolitan Museum of Art. New York. Iacovou, M. 2002. From ten to naught: Formation, consolidation and abolition of Cyprus’ Iron Age polities. Cahier du Centre d’études chypriotes 32, 73–97. Kassianidou, V., 2009. ‘May he send me silver in very great quantities’, EA 35. In D. Michaelides, V. Kassianidou and R. Merillees (eds), Egypt and Cyprus in antiquity. Oxford, 48–57.

69

———. 2012. The origin and use of metals in Iron Age Cyprus. In M. Iacovou (ed.), Cyprus and the Aegean in the Early Iron Age: The legacy of Nicholas Coldstream. Nicosia, 229–59. ———. 2013. The exploitation of the landscape: Metal resources and the copper trade during the age of the Cypriot city-kingdoms. Bulletin of the American Schools of Oriental Research 370, 49–82. ———. 2016. Copper metallurgy in Iron Age Kition. Cahiers du Centre d’études chypriotes 46, 71–88. Kiderlen, M. and Bode, M. Forthcoming. Greek copper markets: Modelling the imports to Greek central places ca. 1100–700 BC. In Proceedings of the 19th International Congress of Classical Archaeology in Cologne and Bonn, 22–26 May 2018. Kiderlen, M., Bode, M., Hauptmann, A. and Bassiakos, Y. 2016. Tripod cauldrons produced at Olympia give evidence for trade with copper from Faynan (Jordan) to south west Greece, c. 950–750 BCE. Journal of Archaeological Science: Reports 8, 303–13. Klein, S. and Hauptmann, A., 1999. Iron Age leaded tin bronzes from Khirbet Edh-Dharih, Jordan. Journal of Archaeological Science 26, 1075–82. Koucky, F. L. and Steinberg, A. 1982. The ancient slags of Cyprus. In J. D. Muhly, R. Maddin and V. Karageorghis (eds), Early metallurgy in Cyprus, 4000–500 BC. Nicosia, 117–42. Leclère, F. and Spencer, J. 2014. Tell Dafana reconsidered. The archaeology of an Egyptian frontier town. British Museum Research Publication 199. London. Levy, T. E., Ben-Yosef, E. and Najjar, M. 2012. New perspectives on Iron Age copper production and society in the Faynan region, Jordan. In V. Kassianidou and G. Papasavvas (eds), Eastern Mediterranean metallurgy and metalwork in the second millennium BC – A conference in honour of James D. Muhly, Nicosia, 10th–11th October 2009. Oxford, 197–214. Lindström, G. and Pilz, O. 2013. Votivspektren. In I. Gerlach and D. Raue (eds), Sanktuar und Ritual. Heilige Plätze im archäologischen Befund. Rahden, 267–74. Masson, A. 2015. Cult and trade. A reflection on Egyptian metal offerings from Naukratis. In Robinson and Goddio 2015, 71–88. ———. 2016. Interpréter le matériel grec et chypriote dans un contexte religieux et thébain: L’exemple

70

A. MASSON-BERGHOFF AND E. PERNICKA

du quartier des prêtres de Karnak – des consommateurs égyptiens de produits grecs et chypriotes. In G. Gorre and A. Marangou (eds), La présence grecque dans la Vallée de Thèbes. Rennes, 25–43. Masson-Berghoff, A. 2019. Naukratis: Egyptian offerings in context. In A. Masson-Berghoff and R. Thomas (eds), Naukratis in context. British Museum Studies in Ancient Egypt and Sudan 24. London, 127–58. . Masson-Berghoff, A., Pernicka, E., Hook, D. and Meek, A. 2018. (Re)sources: Origins of metals in Late Period Egypt. Journal of Archaeological Science: Reports 21, 318–39. Masson, O. 1961. Les inscriptions chypriotes syllabiques. Études chypriotes I. Paris. McConnell, J. R., Wilson, A. I., Stohl, A., Arienzo, M. M., Chellman, N. J., Eckhardt, S., Thompson, E. M., Pollard, A. M. and Peder Steffensen, J. 2018. Lead pollution recorded in Greenland ice indicates European emissions tracked plagues, wars, and imperial expansion during antiquity. Proceedings of the National Academy of Sciences 115 (22), 5726–31. Möller, A. 2000. Naukratis: Trade in Archaic Greece. Oxford. Moreno, A. 2007. Feeding the democracy: The Athenian grain supply in the fifth and fourth centuries BC. Oxford. Muhly, J. D. 1973. Copper and tin: The distribution of mineral resources and the nature of the metals trade in the Bronze Age. Hamden, CT. ———. 2009. The origin of the name ‘Ionian’. In V. Karageorghis and O. Kouka (eds), Cyprus and East Aegean: Intercultural Contacts from 3000 to 500 BC. Nicosia, 23–30. Murillo-Barroso, M., Montero-Ruiz, I., Rafel, N., Hunt Ortiz, M. and Armada, X.-L. 2016. The macroregional scale of silver production in Iberia during the first millennium BC in the context of Mediterranean contacts. Oxford Journal of Archaeology 35 (1), 75–100. Nørgaard, H. W. 2017. Portable XRF on prehistoric bronze artefacts: Limitations and use for the detection of Bronze Age metal workshops. Open Archaeology 3, 101–22. Ogden, J. 2000. Metals. In I. Shaw and P. Nicholson (eds), Ancient Egyptian materials and technology. Cambridge, 148–76.

Oppenheim, A. L. 1967. Essay on overland trade in the first millennium BC. Journal of Cuneiform Studies 21, 236–54. Parker, A. J. 1992. Ancient shipwrecks of the Mediterranean & the Roman provinces. Oxford. Peege, C. (ed.) 2018. Agia Varvara-Almyras. An Iron Age copper smelting site in Cyprus. Oxford. Pernicka, E., Adam, K., Böhme, M., Hezarkhani, Z., Nezafati, N., Schreiner, M., Winterholler, B., Momenzadeh, M. and Vatandoust, A. 2011. Archaeometallurgical researches at Arisman in central Iran. In A. Vatandoust, H. Parzinger and B. Helwing (eds), Early mining and metallurgy on the central Iranian plateau. Report on the first five years of research of the Joint Iranian-German Research Project. Archäologie in Iran und Turan 9. Mainz am Rhein, 633–705. Pernicka, E., Seeliger, T. C., Wagner, G. A., Begemann, F., Schmitt-Strecker, S., Eibner, C., Öztunalı, Ö. and Baranyi, I. 1984. Archäometallurgische Untersuchungen in Nordwestanatolien. Jahrbuch des Römisch-Germanischen Zentralmuseums Mainz 31, 533–99. Petrie, W. M. F. 1886. Naukratis I, 1984–5. Egyptian Research Account 3rd Memoir. London. Philip, G., 1991. Tin, arsenic, lead: alloying practices in Syria-Palestine around 2000 B.C. Levant XXIII, 93–104. Pigott, V. C. 2011. Sources of tin and the tin trade in southwest Asia: Recent research and its relevance to current understanding. In P. P. Betancourt and S. Ferrence (eds), Metallurgy: Understanding how, learning why. Studies in honor of James D. Muhly. Philadelphia, 273–91. ———. 2012. On ancient tin and tin-bronze in the Asian Old World: Further comments. In V. Kassianidou and G. Papasavvas (eds), Eastern Mediterranean metallurgy and metalwork in the second millenium BC. Oxford, 222–35. Rademakers, Fr. W., Rehren, T. H. and Pernicka, E. 2017. Copper for the pharaoh: Identifying multiple metal sources for Ramesses’ workshops from bronze and crucible remains. Journal of Archaeological Science 80, 50–73. Rehren, Th., Boscher, L. and Pernicka, E. 2012. Large scale smelting of speiss and arsenical copper at Early Bronze Age Arisman, Iran. Journal of Archaeological Science 39 (6), 1717–27. Renzi, M., Georgakopoulou, M., Peege, C., Fasnacht, W. and Rehren, T. 2018. Technology of copper

ORIGINS OF METALS FOR COUNTLESS BRONZES

smelting at Agia Varvara-Almyras. In Peege 2018, 269–94. Riederer, J. 1981. Analysis of Egyptian bronzes. Revue d’archéométrie 1, 239–43. ———. 1984. Die naturwissenschaftliche Untersuchung der ägyptischen Bronzen des PelizaeusMuseums in Hildesheim. Berliner Beiträge zur Archäometrie 9, 5–16. ———. 1988. Metallanalysen ägyptischer Bronzestatuetten aus deutschen Museen. Berliner Beiträge zur Archäometrie 10, 5–20. Robinson, D. and Goddio, F. (eds). 2015. ThonisHeracleion in context. Oxford Centre for Maritime Archaeology Monograph 8. Oxford. Rovira, S. and Montero, I. 2003. Natural tin-bronze alloy in Iberian Peninsula metallurgy: Potentiality and reality. In A. Giumlia-Mair and F. Lo Schiavo (eds), The problem of early tin. Oxford, 15–22. Shortland, A. J. 2006. Application of lead isotope analysis to a wide range of Late Bronze Age Egyptian materials. Archaeometry 48 (4), 657–69. Schulze, M. and Lehmann, R. 2014. Alte und neue Analyseverfahren an den Bronzen des Museums August Kestner und was sie uns verraten. In M. Fitzenreiter, C. E. Loeben, D. Raue and U. Wallenstein (eds), Gegossene Götter. Metallhandwerk und Massenproduktion im Alten Ägypten. Rahden, 133–54. Schwab, R. and Willer, F. 2016. Die Zusammensetzung und mögliche Herkunft der Gusslegierungen. In M. Fitzenreiter, F. Willer and J. Auenmüller (eds), Materialien einer Gusswerkstatt von der Qubbet el-Hawa. Bonner Sammlung von Aegyptiaca 7. Berlin, 71–81. Spencer, N. 2007. The Gayer-Anderson cat. British Museum Objects in Focus. London. Stager, L. E. 2003. Phoenician shipwrecks in the deep sea. In N. C. Stampolidis and V. Karageorghis (eds), Ploes. Sea routes… Interconnections in the Mediterranean, c. 1600–600 BC. Proceedings of the international symposium held at Rethymnon, Crete, September 29th–October 2nd, 2002. Athens, 233–47. Stos, Z. A. 2009. Across the wine dark seas… sailor, tinkers and royal cargoes in the Late Bronze Age Eastern Mediterranean. In A. J. Shortland, I. Freestone and Th. Rehren (eds), From mine to microscope. Advances in the study of ancient technology. Oxford, 163–80.

71

Stos-Gale, Z. A., Maliotis, G., Gale, N. H. and Annetts, N. 1997. Lead isotope characteristics of the Cyprus copper ore deposits applied to provenance studies of copper oxhide ingots. Archaeometry 39 (1), 83–123. Tafel, V. and Wagenmann, K. 1951. Lehrbuch der Metallhüttenkunde I. Leipzig. Tallet, P., Marouard, G. and Laisney, D. 2012. Un port de la IVe dynastie au ouadi el-Jarf (mer Rouge). Bulletin de l’Institut Français d’Archéologie Orientale 112, 399–446. Thomas, R. I. 2015. Cypriot figures in terracotta and limestone. In Villing et al. 2013–19. . ———. 2017. Tools and weapons. In Villing et al. 2013–19. . Vaelske, V. and Bode, M. 2018. Early Iron Age copper trails: First results of a pilot study at Sidon. Archaeology and History in Lebanon 48–9, 130–3. Valera, R. G., and Valera, P. G. 2003. Tin in the Mediterranean area: History and geology. In A. Giumlia-Mair and F. Lo Schiavo (eds), The problem of early tin. BAR International Series 1199. Oxford, 3–14. Valério, P., Silva, R. J. C., Soares, A. M. M., Araujo, M. F., Fernandes, F. M. B., Silva, A. C. and Berrocal-Rangel, L. 2010. Technological continuity in early Iron Age bronze metallurgy at the south-western Iberian Peninsula – a sight from Castro dos Ratinhos. Journal of Archaeological Science 37, 1811–9. Van Alfen, P. G. 2002. Pant’agatha: Commodities in Levantine-Aegean trade during the Persian period, 6–4th c. BC. PhD dissertation. The University of Texas at Austin. Van der Wilt, E. 2010. Lead weights and ingots from Heracleion-Thonis: An illustration of Egyptian trade relations with the Aegean. In A. Hudecz and M. Petrik (eds), Commerce and economy in ancient Egypt. BAR International Series 2131. Oxford, 157–64. Vavelidis, M., Gialoglou, G., Pernicka, E. and Wagner, G. A. 1988. Die Buntmetall- und Eisen-ManganLagerstätten von Thasos. In G. A. Wagner and

72

A. MASSON-BERGHOFF AND E. PERNICKA

G. Weisgerber (eds), Antike Edel- und Buntmetallgewinnung auf Thasos. Der Anschnitt, Beiheft 6, Deutsches Bergbau-Museum. Bochum, 40–58. Vikentiev, V. 1957. Les trois inscriptions concernant la mine de plomb d’Oum Huetat. Annales du Service des antiquités de l’Égypte 54, 179–89. Villing, A. 2013. Naukratis: A city and trading port in Egypt. In Villing et al. 2013–19. . ———. 2015. Egyptian-Greek exchange in the Late Period: The view from Nokradj-Naukratis. In Robinson and Goddio 2015, 229–46. Villing, A., Bergeron, M., Bourogiannis, G., Johnston, A., Leclère, F., Masson, A. and Thomas, R. (eds). 2013–19. Naukratis: Greeks in Egypt. British Museum Online Research Catalogue. London. . Vogl, J., Rosner, M., Curbera, J., Peltz, U. and Peplinski, B. 2016. Lead isotope analysis in magic artefacts from the Berlin museums. Archaeological and Anthropological Sciences 110 (5), 1111–27. Weinstein, J. M. 1973. Foundation deposits in ancient Egypt. PhD dissertation. University of Philadelphia, UMI. Weiss, K. 2012. Ägyptische Tier- und Götterbronzen aus Unterägypten: Untersuchungen zu Typus, Ikonographie und Funktion sowie der Bedeutung

innerhalb der Kulturkontakte zu Griechenland. Ägypten und Altes Testament 81. Wiesbaden. Wilson, P. 2019. Gateway to the underworld: The cult areas at Sais. In A. Masson-Berghoff and R. Thomas (eds), Naukratis in context. British Museum Studies in Ancient Egypt and Sudan 24. London, 341–64. . Wuttmann, M., Coulon, L. and Gombert, F. 2007. An assemblage of bronze statuettes in a cult context: The temple of ‘Ayn Manâwir. In Hill and Schorsch 2007, 167–73. Yahalom-Mack, N., Galili, E., Segal, I., Eliyahu-Behar, A., Boaretto, E., Shilstein, S. and Finkelstein, I. 2014. New insights into Levantine copper trade: Analysis of ingots from the Bronze and Iron Ages in Israel. Journal of Archaeological Science 45, 159–77. Yahalom-Mack, N., Herzlinger, G., Bogdanovsky, A., Tirosh, O., Garfinkel, Y., Dugaw, S., Lipschits, O. and Erel, Y. Forthcoming. Combining chemical and lead isotope analyses with 3-D geometric– morphometric shape analysis: A methodological case study of Scythian arrowheads from the Southern Levant. Yardeni, A. 1994. Maritime trade and royal accountancy in an erased customs account from 475 BCE on the Aḥiqar scroll from Elephantine. Bulletin of the American Schools of Oriental Research 293, 67–78.

EVIDENCE SUGGESTING ANOTHER BOSTON DYAD OF MENKAURA AND A QUEEN Florence Dunn FRIEDMAN with assistance from Michelle PISA

The greywacke dyad of King Menkaura and a queen (Fig. 1, MFA, Boston 11.1738)1 was excavated in 1910 by George A. Reisner, Director of the Harvard University–Boston Museum of Fine Arts Expedition, from a ‘Thieves hole’ in a southwestern storeroom of Menkaura’s valley temple at Giza (Reisner 1931, 37, Jan. 18–9; 110 (17), pls 54–60, with slightly different data

in the two entries). Where the dyad was originally intended to appear in the valley temple is unknown, since, like the triads, it left no trace of emplacements or other evidence of installation. Reisner found the dyad, still with a good deal of polychromy (Reisner 1931, 110 [17]), atop a severely damaged triad (Fig. 2f). Miraculously intact, except for a tiny chip from the end of the king’s beard,2 the dyad was a fortunate rescue from the vandalism inflicted on much of Menkaura’s statuary — vandalism that in several campaigns of destruction left scattered remains (mostly alabaster) as far as Khafra’s complex and beyond (Friedman forthcoming, 112). The Boston dyad, unlike the triads, lacked its final carving. The ancient sculptors apparently did not have time to carve the striations on the king’s nemes or kilt, or the queen’s wig, or, most importantly, to inscribe the base, where the triads’ inscriptions appear (Reisner 1931, pl. 46 a–f). The six greywacke triads Reisner found (Reisner 1931, 109–10, pls 36–46) were again not in their originally intended location, which was probably the valley temple’s open court (cf. Seidel 1996, 43–7). I have previously proposed that the ancient designers possibly once considered inserting them into the compound mud-brick niches located to the north and south sides of the court (Friedman 2011a, 18–9, though speculative; 2011b, 99–103). Such an idea, however, could not have been realised, given that the larger triads exceed the dimensions of the niches. The theme of the triads, I think, is signalled by the mekes (Bothmer 1950, 12, fig. 5, 15–6; Seidel 1996, 31) in the left hand of the standing Menkaura in Fig. 2e (Fig. 3a, b). The mekes is a soft document holder compressed at the centre so that the ends flare out slightly; it is a piece of iconography associated with the ritual rejuvenation of the king at the heb-sed and normally

1

2

Abstract For over a century the greywacke dyad (or ‘pair’ statue) of King Menkaura and a queen has been recognised as one of the greatest masterworks of ancient Egyptian art. Originally part of the statue programme of Menkaura’s Dynasty 4 valley temple, the dyad resides today at the Museum of Fine Arts, Boston. While previously researching the dyad and a group of related Menkaura triads, I offered evidence (a small fragment) for a second dyad, but one smaller than the Boston example. At that time, however, I did not recognise that additional fragments might suggest yet a third dyad that would duplicate the size and scale of the Boston masterwork. Using laser-scanned and 3D-printed reproductions of several of these fragments, this paper presents evidence for that additional ‘Boston’ dyad, suggesting that Menkaura’s statue programme once included at least three dyads and probably a fourth, for symmetry. Multiples would, in fact, be expected, considering Menkaura’s use of multiple greywacke triads, multiple alabaster seated statues and at least two alabaster colossi. The purpose of the dyads, I suggest, was to legitimise the king in the embrace of the queen mother assimilated to Hathor, at his eternal heb-sed, celebrated symbolically in his valley temple. * * *

All Museum of Fine Arts, Boston objects mentioned in this article can be found through their accession numbers at and at The Giza Archives at .

Harvard-MFA Expedition Diary pages, vol.01.p.010: The Giza Archives at .

74

F.D. FRIEDMAN

Fig. 1: Menkaura dyad, MFA 11.1738*. Photography: Michael Fredericks. *Courtesy, Museum of Fine Arts, Boston.

Fig. 2: a. JE 40678; b. JE 46499; c. JE 40679; d. MFA 11.3147*; e. MFA 09.200*; f. MFA 12.1514*. Photography: Michael Fredericks. Grey areas are reconstruction by Florence Dunn Friedman, drawn by Michelle Pisa. For reconstruction of MFA 12.1514, see Friedman 2015b, esp. 31–2. *Courtesy, Museum of Fine Arts, Boston.

EVIDENCE SUGGESTING ANOTHER BOSTON DYAD OF MENKAURA AND A QUEEN

75

Fig. 3: a. King holding mekes in Menkaura triad, MFA 09.200*; b. Alternate view of king holding mekes in MFA 09.200*; c. underground relief panel of Djoser running with mekes. Drawing after Friedman 1995, fig. 14; d. Reconstructed image of Sneferu standing with mekes. Drawing after Fakhry 1961, fig. 120. Photography: Michael Fredericks. *Courtesy, Museum of Fine Arts, Boston.

appears in the left hand of kings, who are usually shown running. Earlier reliefs show the mekes in the left hand of Djoser running the heb-sed race, from his underground panels at the Step Pyramid complex, and in the left hand of Sneferu, who, however, stands (like Menkaura) in a fragmentary scene in what we now know was his heb-sed ‘valley temple’ (Stadelmann 2011) (Fig. 3c, d). Neither the reliefs (though fragmentary) nor the Menkaura triad (see Fig. 2e) mention the heb-sed in their associated texts; rather, the hebsed is signalled solely by iconography, including the mekes. The mekes is not the same as the rounded objects that project from either fist of the king and other figures (Fischer 1975, 20) (Fig. 4a, b), including two female nome personifications on the triads (Cairo JE 40679, 46499). Though possibly representing abbrevi-

ated staves (theory summarised in Fischer 1975, 11), a bolt of cloth (Fischer 1975, 14) or amuletic rolls of cloth (Arnold 1999, 67), the meaning of these curved projections remains enigmatic (Labbé-Toutée and Ziegler 1999, 273; Ziegler 1999, 269). The triads, each with a different nome personification, were related to the heb-sed, based on further comparison with earlier material at Djoser’s heb-sed court at the Step Pyramid complex at Saqqara and at Sneferu’s heb-sed ‘valley temple’ court at the Bent Pyramid at Dahshur (Friedman 2011b, 105–11, 125; 2015b, 22). I have suggested that the triads together with the dyad were part of a heb-sed ritual, probably intended for Menkaura’s open valley temple court. Menkaura’s triads and dyad, however, are physically small (well under life-size: Fig. 5) relative to the size of the court, a point that could have been compensated for in part if

76

F.D. FRIEDMAN

Fig. 4: a. King holding enigmatic object in Menkaura triad, JE 40679; b. King holding enigmatic object in Menkaura dyad, MFA 11.1738*. Photography: Michael Fredericks. *Courtesy, Museum of Fine Arts, Boston.

Fig. 5: a. JE 40678; b. JE 46499; c. JE 40679; d. MFA 11.3147*; e. MFA 09.200*; f. MFA 12.1514*; g. MFA 11.1738*. Human figure at 5ft 3in. (160cm) for comparison. Photography: Michael Fredericks. Grey areas are reconstruction by Florence Dunn Friedman, drawn by Michelle Pisa. For reconstruction of MFA 12.1514, see Friedman 2015b, esp. 31–2. *Courtesy, Museum of Fine Arts, Boston.

there were originally more statues than have survived, and if they were elevated in some way in situ. That there were originally more greywacke statues than the surviving examples is apparent from about thirty-five bits and pieces of fragmentary greywacke sculpture at the MFA, Boston. Reisner, who found them, believed that they belonged to more triads, and I initially assumed the same. Though discussed at length in other publications, features of the triads must for clarity’s sake be repeated here for comparison with those of the dyad. Six triads, shared between Boston and Cairo, survive (see Fig. 2), but only one dyad. Of the six surviving triads, two were severely vandalised. One (Fig. 2f), as noted, was found

in the hole below the dyad, and the other (Fig. 2d) was found below the surface in the valley temple court. The triads illustrated in Figs 2a, b, c, e were found lined up on two sides of a corridor that appears to have served as a statue cache (Corridor III, Room 4: Reisner 1931, 35; for correction to Reisner: Friedman 2011a, fig. 3). Salient features include the following: Type 1 triads show the king striding in the centre flanked by Hathor, the king’s divine mother and wife, who is always to his right, and a male or female nome personification, to his left. Hathor and the nome personifications look outward to greater or lesser degrees (first noted in Cairo JE 46499 by Arnold 1999, 67; see photos of the other triads’ outward-gazing figures, in Friedman 2011b,

EVIDENCE SUGGESTING ANOTHER BOSTON DYAD OF MENKAURA AND A QUEEN

fig. 7), maintaining the forward-looking king as the statues’ focus. Type 2 has Hathor seated in the centre as the statue’s focus (the king never sits), flanked by a standing king at her left, and a nome personification to her right. Only the female nome survives, but a male personification could have been used for other Type 2 triads, as in my reconstruction for Fig. 2f (for use of the red crown and male personification on Fig. 2f, see Friedman 2015b). Type 1 triads have slightly varied heights, with the reconstructed triad fragment in Fig. 2d being noticeably taller than the other three Type 1 triads. The intact Type 2 triad in Fig. 2e, with Hathor in the centre, and the reconstructed Type 2 triad in Fig. 2f are somewhat smaller than the Type 1 triads and even seem to vary slightly in height from one another. In both Type 1 and 2 triads, the king always wears the white crown. All intact triads have high, wall-like back slabs and are inscribed on the base with an offering text for the king from an unidentified ‘I’ who should be understood as Hathor acting through the vehicle of the nome personification, in front of whom the inscription appears (Friedman 2011a, 11; 2015a, 98, n. 15, 99–100). What the inscription might have been on the dyad’s base is unknown. Expanding on an idea of Wendy Wood ‘that the triads are not representative of all the nomes but only those in which the cult of Hathor had been established with the support of royal patronage’ (Wood 1974, 86), I suggest a further point: the nomes chosen are those with estates, real or symbolic, founded by Menkaura and other kings for their divine mother, Hathor, and from which she then ‘feeds’ her son, the king, most notably, for his heb-sed. The estates in Menkaura’s triads’ nomes, probably like the estates in Sneferu’s nomes at his heb-sed ‘valley’ temple, were especially meant for provisioning the heb-sed (Friedman 2011a, 105–11; Friedman 2015b, 18–24). The dyad — a statue form with roots going back to the Predynastic — is much larger than the triads and its back pillar is only shoulder-high (see Fig. 1, right). The king wears a nemes without uraeus, which is unusual, and is accompanied not by Hathor, but by a queen who, in contrast to the triads, stands to his left. Surely assimilated to Hathor (Friedman 2008, 136–41), the queen, who is probably his mother (see below), embraces her royal son about his waist with her right hand, while clasping his upper left arm with her other hand, a combination of gestures that appears in mirror form between Hathor and the king in the single surviving Type 2 triad (Fig. 2e). The queen in the dyad appears most

77

specifically to correlate with the Hathor in the Boston Type 2 triad, the triad most overtly linked to the hebsed, referred to by the mekes in the king’s left hand (Friedman 2008, 137; see also 132, 134 for other hebsed features). Hathor’s gestures in the dyad, as well as in the triads, point not just to affection and protection, but legitimation, in effect, saying: ‘This is my son, king now and forever’, the theme of the Sed Festival. Fig. 5 shows how the dyad towers over the triads, while still being diminutive beside a 5ft 3in. (= c. 1.6m) human figure. Like the triads, the dyad(s) were most probably elevated for ritual use when installed in the temple. The triads were made in multiples, though none fully replicates another, showing variations in stylistic detail, size, format (Types 1 and 2), gestures, and the gender and insignia of the nome personifications. Similarly, the four alabaster seated statues found in situ in the offering hall were made in multiples (Reisner 1931, pl. 47a; reconstruction of one example in Fig. 6d), but with variations in stylistic detail, size, type of head covering, and extent of inscription (Friedman forthcoming, with citations). Even the (now restored) colossal alabaster seated statue in Boston (Fig. 6e), probably the major cult statue in the pyramid temple (Lacovara and Reeves 1987, 111–5, fig. 1, pls 3–4; for probable location of colossus, authors in n. 29 refer to Reisner 1931, plan 1 [8]), seems to have had at least one colossal companion somewhere in the complex, as evidenced by similarly sized alabaster fragments at the MFA, Boston, which, however, lack any find-spots beyond ‘Giza’ (Giza Archives; Friedman forthcoming, n. 9). One fragment is a right big toe (MFA 47.2078) that turns out to be even larger than the preserved right big toe on the restored Boston colossus. That big toe, together with several other alabaster fragments, appears to belong to at least another alabaster colossus, possibly the major cult statue in the valley temple’s sanctuary (Reisner 1931, plan IX, 2; Friedman forthcoming) by analogy with the placement of the Boston colossus in the pyramid temple. These two alabaster colossi, one in the pyramid temple and one in the valley temple, would underscore the major cultic focus of the pyramid complex, Menkaura and his cult of kingship. Consideration of the use of multiples in the Menkaura statue programme would suggest that the greywacke Boston dyad was not alone, but would have been one of at least two such figures, if not more; and that as with the greywacke triads and seated alabaster figures, variations in size and detail would have appeared among the different dyad examples.

78

F.D. FRIEDMAN

Fig. 6: a. MFA 09.200* (H. 84.5cm); b. JE 46499 (H. 95.8cm); c. MFA 11.1738* (H. 142.2cm); d. MFA 09.202* (statue base) reconstructed with head JE 40704 (H. 160cm); e. MFA 09.204* (H. 235cm). Photography: Michael Fredericks. Grey areas are reconstruction by Florence Dunn Friedman, drawn by Michelle Pisa. *Courtesy, Museum of Fine Arts, Boston.

About the greywacke fragments Reisner believed that the bits and pieces of greywacke sculpture, now stored in the Boston Museum of Fine Arts, belonged to more triads. A problem for the modern researcher, however, is that while Reisner recorded which fragments were found in the valley temple, he rarely noted their precise findspots in that temple. Consequently, we do not know which, if any, of the greywacke fragments were found near the dyad or triads. But since the fragments are critical for speculating about the full body of sculpture that once filled the valley temple, I researched a selection of fragments with the data available. It is these fragments that stand at the heart of this paper.3 Although Reisner thought they all originated from

3

For access to the fragments and other material, I thank, at the Museum of Fine Arts, Boston, Rita E. Freed, John F. Cogan, Jr. and Mary L. Cornille, Chair of the Department of Art of the Ancient World, Lawrence M. Berman, Norma Jean Calderwood,

triads, could some of the greywacke fragments belong to one or more dyads? This is one question that emerged from the research. It appears that many of the fragments go to lost triads as Reisner suspected. Some valley temple fragments, based on their iconography and our current data, can only go to triads (Fig. 7). The fragment with part of a male torso with non-royal kilt and belt tab can only go to a male nome in a triad (as in Fig. 2a and d) and the fragment with clasped hands can only go to another triad in which Hathor holds the king’s hand (as seen in the triads in Fig. 2b and d), since there is no handholding in the dyad. Another greywacke fragment that probably belongs to a triad is what appears to be part of an Upper Egyptian white crown (which I incorrectly assigned to the Fig. 2f triad in earlier work). The white

Senior Curator of Ancient Egyptian, Nubian, and Near Eastern Art; and especially Denise Doxey, Curator of Ancient Egyptian, Nubian, and Near Eastern Art, who was extremely generous with her time.

EVIDENCE SUGGESTING ANOTHER BOSTON DYAD OF MENKAURA AND A QUEEN

79

Fig. 7: Left to right: MFA 47.1720* male nome torso fragment; MFA 47.1719* clasped hands fragment; MFA 24.2796* white crown fragment. Photography: Michael Fredericks. Images to scale. *Courtesy, Museum of Fine Arts, Boston.

My initial goal was to reconstruct lost and broken triads from fragments. This process began with measuring,

photographing and drawing the fragments.4 The drawings were then inserted to scale in hypothetical Type 1 triad renderings (Friedman 2011a, fig. 6; 2011b, fig. 3). These renderings, however, produced triads that were too large in comparison with the extant Boston and Cairo examples, being in fact as tall as the dyad (142.2cm). That the hypothetical triads might be large relative to the extant examples was partly supported by the fact that Menkaura’s statue programme included statuary of significantly different sizes, with triads of varied heights and alabaster seated statues that are much larger than the triads or dyad (see Fig. 6). I also tried to reconstruct the Type 2 triad in Fig. 2f.5 Building on ideas generously shared with me by Peter Lacovara, I used the white crown (see Fig. 7, right), a royal beard, part of a wig and two pieces that respectively look like the king’s face and a nome’s face

4

5

crown is indeed the king’s only surviving headgear in the triads, while he wears the nemes on the dyad. The absence of clear find-spots for these fragments remains, however, an issue. The white crown’s provenance, for example, as given by Reisner, is not the valley temple, but the ‘Eastern Cemetery, [Avenue G 2], between mastabas G 7210–7220: G 7220 and G 7230–7240: G 7230’, according to The Giza Archives. Given that some fragments from the Menkaura complex migrated far afield from their original source over time, it is, nevertheless, still possible that this fragment belongs to one of his triads. Early attempts at reconstruction

For the photographing and drawing of the fragments, I must thank here MFA, Boston, staff, as well as Michael Fredericks (photographer) and Shawn Kenney (artist).

I direct the reader to the citations that follow, where photographs, which could not be included in this article, elucidate the discussion.

80

F.D. FRIEDMAN

(Friedman 2011a, fig. 5f; 2011b, fig. 2f). But this reconstruction was wrong, for when I later held 3D fragment reproductions over missing areas of the triad in Fig. 2f, the ‘fragments’ were clearly disproportional. Three-dimensional reproductions proved to be useful in assessing which reconstructions were likely and which were not (for the new theoretical reconstruction of the Fig. 2f triad that does not include any fragments, see Friedman 2015b, esp. 31–2). While the accuracy of any single reconstruction is uncertain, there were definitely more triads than the six extant examples, and some hypothetical triads could possibly be as tall as the dyad (142.2cm). Some of the fragments — such as a female face, a male fist, and a section of a thigh (?) (Fig. 8) — could belong equally well to a triad or a dyad. I have already suggested that a fragment from the valley temple came

from a queen in a slightly smaller dyad than the existing Boston dyad (Friedman 2008). That fragment was part of an advanced left leg of a striding woman (a high-status stance for a female) with the hem of her dress pulled forward (MFA 47.1747). The fragment could not go to a triad because it is from a female figure with advanced left leg at the proper left of a statue (viewers’ right), a combination of features that does not appear on any triad, but which appears in the Boston dyad (Friedman 2008, 141–4, with photos). The fragment also had an incised line that defined the back of the leg (Friedman 2008, fig. 27b), showing that the dyad from which it came included finished carving in that area, unlike that area in the unfinished Boston dyad (Friedman 2008, fig. 28a). To test the existence of one or more unknown greywacke dyads from the Menkaura valley temple, I started looking at the greywacke face,

Fig. 8: Top left: MFA 47.1756* female face fragment; top right: MFA 47.1765* fist fragment; below: MFA 47.1751* first thought to be a thigh fragment. Photography: Michelle Pisa. Images not to scale. *Courtesy, Museum of Fine Arts, Boston.

EVIDENCE SUGGESTING ANOTHER BOSTON DYAD OF MENKAURA AND A QUEEN

fist and leg section that were too large for any of the extant triads, including even the largest of them (Fig. 2d). Since these fragments were candidates for coming from a dyad, they needed to be compared with features of the Boston dyad.

Reisner assumed the face fragment was part of a triad (Reisner 1931, 110 [15]) now lost. For this fragment, he recorded the exact find-spot: ‘Under the walls of the last series of houses (I–5 to I–10) above the southern wall and on the surface of decay of the First Temple; Pl. 64g’ (Reisner 1931, 110 [15] [b]).

Latest attempts at reconstruction Two-dimensional renderings of the greywacke fragments were insufficient to extrapolate on three-dimensional statues. It was thus first necessary to hold the original stone fragments beside the Boston dyad for comparison. Two valley temple pieces were excluded in that study: a section of hair with carved striations (MFA 47.1721) that came from a partly finished statue, and a right big toe (MFA 47.1746) that seemed to fit the king’s foot but was ultimately unconvincing. Three greywacke fragments that were too large to fit any of the extant triads were finally selected. They included a battered face, a male fist gripping the so-called enigmatic object and a piece with a modelled surface that looked at first like part of a thigh (see Fig. 8). Reisner’s information on these fragments is limited, in accordance with the fact that he rarely described the fragments he uncovered, gave little information on their find-spots and seems generally to assume that they came from triads (e.g., Reisner 1931, 108 [3]; 110 [15]). They are currently identified as triad fragments in The Giza Archives,6 whose information is set out below, though the present paper aims to show that they come from at least one dyad; new dimensions have been added by the authors of this article.7

81

2 Fragment of greywacke triad: fist with rust-coloured pigment (MFA 47.1765) Find-spot: Menkaura valley temple (MVT) Dimensions: W. 7.5 × L. 10.2 × D. 4.4cm 3 Fragment of greywacke triad: smooth surface with some modelling, possibly from thigh (MFA 47.1751) Find-spot: Menkaura valley temple (MVT) Dimensions: L. 11.4 × W. 7.2 × D. 1.6cm

1 Fragment of greywacke triad: portion of face (MFA 47.1756) Find-spot: Menkaura valley temple (MVT); Dimensions: L. 7.8 × W. 7.8 × D. 2.6cm

While all three of these fragments came from the valley temple, the exact find-spot within the temple is given only for the face fragment, and that find-spot is not helpful in attributing it to a triad or dyad. For each fragment, The Giza Archives states: ‘This object was excavated by the Harvard University–Boston Museum of Fine Arts Expedition but was not recorded in any object register book. Excavated by the Harvard University–Museum of Fine Arts Expedition; assigned to the MFA in the division of finds by the government of Egypt’. The second and third fragments have the letter ‘E’ written on them, either glued to a tag on the back (E.19 on the fist) or directly painted on the piece (E.3114 on one corner of the thigh (?))8 But the meaning of the E designation remains unknown.9 The face fragment has lost its upper portion, including the eyes and much of the proper right side, and what remains is much damaged (see Fig. 8, top left). The cheekbone, nose, centre of the mouth and chin are

6

9

7 8

. The dimensions were taken separately by Michelle Pisa. I thank Susan J. Allen, Research Associate for Egyptian Expedition Archives, Art of the Ancient World, Museum of Fine Arts, Boston, for informing me that this piece is E.3114 and not E.31, as I read it through the faded paint. Another fragment (MFA 47.1764) also bears red paint, like that found on the fist and calf, also has an E designation, E.18, and was also found in Menkaura’s valley temple (MVT), according to the Giza Archives (). Still, I can neither identify what this piece is nor how it might fit into a figure or other object, much less a dyad.

Susan J. Allen, Research Associate for Egyptian Expedition Archives, Art of the Ancient World, Museum of Fine Arts, Boston, writes in a an email of 1 June 2016: ‘I can’t find anything that explains the E. numbers… I can only guess that the E. nos. may be early field numbers or else they may refer to a plan of the MVT.’ Peter Der Manuelian, Philip J. King Professor of Egyptology, Director, Harvard Semitic Museum, suggested in an email of 30 May 2016 that E could be for ‘Egyptian’, meaning the Boston Museum of Fine Arts’ Egyptian Department. He notes that the ‘older, unprovenanced Way Collection objects often had things like P for Pottery, W for Wood on them’. I thank Susan J. Allen and Peter Der Manuelian for their communications.

82

F.D. FRIEDMAN

especially battered. The damage along the upper lip and centre of the lower lip obscures the original contour of the mouth, especially the original outline of the upper lip. The fullness of the cheek and jawline remain. The vandals’ violent attack on the sculpture is suggested by the sheer, flat break on the back. Following Reisner, I originally thought the face fragment belonged to a triad, specifically a Type 1 in which Hathor always appears on the proper right with face turned slightly outward, as noted above (Friedman 2011b, 97–105). However, when holding the stone face fragment over the face of the queen on the Boston dyad, it appeared to fit the face well without suggesting any outward gaze (Fig. 9). The lips on the fragment look less full than the queen’s lips on the dyad, but this may be a result of damage across the fragment’s lips or due to a stylistic variance from one statue to another. Hathor’s lips on the triad Cairo JE 40678 (Fig. 10, far left), for example, are more compressed than those of the other three triad Hathors (also see photos in Seidel 1996, pl. 11) — Hathor being the goddess with whom the queen in the Boston dyad is assimilated. Stylistic variations are also seen in the facial features of Menkaura on the triads (Seidel 1996, pl. 10). Holding the greywacke face fragment over the queen’s face in the dyad highlighted a key logistical problem, however: it was impossible to place the stone fragment against the dyad itself, making a true fit impossible to determine. A three-dimensional reproduction that could be held close to the statue was needed to determine what really fit or did not. Making reproductions One-to one-replicas with a non-abrasive surface were made for the face, fist and ‘thigh’ fragments by first laser scanning them and then 3D printing them in a whitish, non-abrasive plastic material that could not damage the stone statues against which they would be placed.10 The first one-to-one reproduction of the face had the entire solid interior at the back (Fig. 11), which did not permit the fragment to be held flush against the queen’s

10

For the creation of the fragment reproductions, I thank Walter Gilbert, Chair of the Society of Fellows, and Carl M. Loeb University Professor Emeritus at Harvard University, who made all the work possible; Neil Gershenfeld, Director of the Center for

Fig. 9: MFA 47.1756* held over queen’s face on MFA 11.1738* dyad. Photography: Michelle Pisa. *Courtesy, Museum of Fine Arts, Boston.

face. The solution was to digitally create a second replica of just the surface ‘skin’ of the stone face fragment with the interior removed, before realising its 3D printing. The reproduction of just the outer contour of the face fragment enabled us to see just how closely the fragment fit the face of the dyad’s queen. Being sure that the dimensions of the reproduction of the face matched the original as exactly as possible was essential. Michelle Pisa measured the face and the other fragments with a Digital Caliper (Mitutoyo ABSOLUTE Digimatic 500-197-20) (Fig. 12). Nonetheless, due to the vagaries of printing, some margins of the original (Fig. 13, bottom centre) slightly exceeded those of the reproduction. With the interior of the

Bits and Atoms at The Massachusetts Institute of Technology, and his assistant, John DiFrancesco; Rita Freed, Larry Berman, and especially Denise Doxey; and Michelle Pisa, who contributed to every facet of this work.

EVIDENCE SUGGESTING ANOTHER BOSTON DYAD OF MENKAURA AND A QUEEN

83

Fig. 10: Faces of Hathor from the four intact triads: JE 40678, JE 46499, JE 40679 and MFA 09.200*. Photography: Michael Fredericks. *Courtesy, Museum of Fine Arts, Boston.

Fig. 11: Solid reproduction of MFA 47.1756*. Photography: Michelle Pisa. *Courtesy, Museum of Fine Arts, Boston.

object now completely gone, however, it became possible to lay the thin outer surface of the reproduction over the original stone fragment (see esp. Fig. 13, lower right). The overall reproduction of the original stone face is also excellent (see Fig. 13, top and bottom left), with a surface contour that replicates that of the original stone fragment. With this non-abrasive fragment reproduction, it is easy to show how it could not fit a Type 2 triad with Hathor seated at centre (on the distinction between the triad’s Types 1 and 2, see supra): its form swallowed the face of Hathor in this, the

smallest of the triads (Fig. 14). Laid atop the queen’s face in the dyad, it fit flush (Fig. 15). This snug-fitting reproduction of the outer contour of the original face fragment, as opposed to a solid three-dimensional reproduction or a two-dimensional rendering, confirmed that the face fragment goes to a front-facing figure and not, as I originally thought, an outward-gazing flanking Hathor in a Type 1 triad where she stands at proper right. Using physical reproductions of the outer contour of the original stone fragments, which could be ‘tried on’ the statuary, turned out to make all

84

F.D. FRIEDMAN

Fig. 12: Measuring MFA 47.1756*. Photography: Michelle Pisa. *Courtesy, Museum of Fine Arts, Boston.

Fig. 13: MFA 47.1756* and reproduction. Photography: Michelle Pisa. *Courtesy, Museum of Fine Arts, Boston.

EVIDENCE SUGGESTING ANOTHER BOSTON DYAD OF MENKAURA AND A QUEEN

Fig. 14: Reproduction of MFA 47.1756* held over Hathor’s face on MFA 09.200* triad. Photography: Michelle Pisa. *Courtesy, Museum of Fine Arts, Boston.

Fig. 15: Reproduction of MFA 47.1756* held over queen’s face on MFA 11.1738* dyad. Photography: Michelle Pisa. *Courtesy, Museum of Fine Arts, Boston.

85

86

F.D. FRIEDMAN

Fig. 16: MFA 47.1765*. Photography: Michael Fredericks and Michelle Pisa. *Courtesy, Museum of Fine Arts, Boston.

the difference in making matches. New findings were emerging that confirmed or rejected earlier hypotheses, and it began to appear that another dyad of the same size as the Boston dyad might once have existed. The same laser scanning and reproduction process was used on the fist fragment (Fig. 16), an object that bears a remarkable amount of surviving paint in a bright red rust colour, the same colour that a bright light reveals around the king’s ear, neck, face, nostrils and between his lips on the dyad (Reisner 1931, 110 [17]), and which was the colour intended for all flesh areas of the king’s body. The survival of the paint on this fist fragment and not on others may be due to its having been protected from exposure to deleterious elements such as moisture, to which the other fragments

11

My thanks to Mimi Leveque, Conservator, Peabody Essex Museum, for this explanation.

were apparently exposed.11 The fist grasps the so-called enigmatic object that is typically seen in the hands of male figures, including the king in the triads and the dyad, and even in the hands of two female nome personifications in the triads (Cairo JE 40679, 46499). The size of the fist already suggests it cannot come from a triad of the size of the six extant examples, thus making it a candidate for a dyad. Holding the stone fist fragment beside the king’s fist in the Boston dyad (Fig. 17) shows how similar the two are in size. After laser scanning the fragment, the interior was digitally removed before three-dimensional printing, yielding just the surface contour of the fist in a non-abrasive plastic material (Fig. 18), and when we slipped the reproduction over the hand of the king in the dyad, the reproduction slid comfortably into place (Fig. 19). The last of the three fragments (Fig. 20, top left) bears tiny traces of reddish paint scattered over the whole surface, which suggests that this fragment, like

EVIDENCE SUGGESTING ANOTHER BOSTON DYAD OF MENKAURA AND A QUEEN

Fig. 17: MFA 47.1765* held next to king’s hand on MFA 11.1738* dyad. Photography: Michelle Pisa. *Courtesy, Museum of Fine Arts, Boston.

Fig. 18: MFA 47.1765* and reproductions. Photography: Michael Fredericks. *Courtesy, Museum of Fine Arts, Boston.

87

88

F.D. FRIEDMAN

Fig. 19: Reproduction of MFA 47.1765* held over king’s hand on MFA 11.1738* dyad. Photography: Michelle Pisa. *Courtesy, Museum of Fine Arts, Boston.

Fig. 20: MFA 47.1751* and reproduction. Photography: Michelle Pisa. *Courtesy, Museum of Fine Arts, Boston.

EVIDENCE SUGGESTING ANOTHER BOSTON DYAD OF MENKAURA AND A QUEEN

89

Fig. 21: MFA 47.1751* fragment (left) and reproduction (right) held over king’s calf on MFA 11.1738* dyad. Photography: Michelle Pisa. *Courtesy, Museum of Fine Arts, Boston.

the fist, was part of a statue that was originally painted. Again, only the outer skin was reproduced (see Fig. 20). Identifying this object was not as straightforward as for the other two pieces. It appears that it did not belong to a thigh, as suggested in The Giza Archives. Instead, the piece matched the king’s extended left calf in the Boston dyad (Fig. 21, left). It was only with the laserscanned and 3D-printed reproduction with interior removed, however, that we could check our suggestion by now safely holding the reproduction against the king’s leg (see Fig. 21, right), raising it slightly higher on the calf and rotating it slightly counter clockwise. Thanks to this technology and process, I can suggest that there once existed another dyad of the same size as the Boston dyad and which, like the Boston dyad, was also painted (Fig. 22). Although these three fragments could belong to more than one dyad, a conservative estimate would limit the suggestion to one. Despite the possibility that they could fit a much larger triad (as hypothesised in earlier publications: e.g. Friedman 2011b, fig. 3, ‘Hypothetical Type 1’s’), nothing can so

Fig. 22: Proposed dyad incorporating MFA 47.1756*, MFA 47.1765*, and MFA 47.1751*. Grey areas are reconstructions by Florence Dunn Friedman, drawn by Michelle Pisa. *Courtesy, Museum of Fine Arts, Boston.

90

F.D. FRIEDMAN

far corroborate that suggestion, whereas all three threedimensional fragment reproductions match perfectly the Boston dyad. An additional note: just as the multiple triads have varied gestures among the figures, multiple dyads may also have had varied gestures. Thus, we might speculate, for example, that the hand-holding fragment in Fig. 7 (centre) came not from a triad as in Fig. 2b, d, but from a now lost dyad where the king and queen held hands. Conclusion If there once existed a second dyad of the same size as the Boston dyad, as well as a slightly smaller dyad as posited earlier (Fig. 23), then there were at least three dyads originally, and I would add a fourth for symmetry (Fig. 24). What does this hypothesis mean for the statue programme of Menkaura’s valley temple? First of all, the dyad as a statue type had more prominence than originally thought, though the focal figure in the dyad is somewhat ambiguous. Both king and queen take centre stage in different ways: the king is slightly taller than the queen and strides more forcefully than she does. However, the queen’s frontality gives her some prominence compared to the king, who glances to his right (Arnold 1999, 68). Furthermore, her size and extended left leg also denote unusually high female status. Identifying which high-ranking queen is represented in the dyad(s) has been a problem. She should probably not be identified as Menkaura’s wife, Khamerernebty II (if she was his wife: cf. Lehner 2015, 272, no. 179 with refs), as Reisner believed (Reisner 1931, 110 [17]), but as his mother (Arnold 1999, 68). Some of the features of the sculpture — slightly sagging breasts (Arnold 1999, 68), tummy roll and slightly broadened hips (Friedman 2008, 135) — suggest an older mortal woman. While Menkaura’s mother has traditionally been identified as Khamerernebty I (Friedman 2008, esp. 117–23, 135, 141, with references; Callender 2011, 115–6), Mark Lehner offers a persuasive hypothesis, based on new archaeological data, that Menkaura’s mother may have been Khentkawes I (Lehner 2015, 267–8, 271–4; but cf. Callender 2011, 149), whose settlement appears ‘conjoined’ with Menkaura’s valley temple (Lehner 2015, 271 says the word ‘dyad’ come to mind; also 267). Menkaura and his mother, striding forward, would, in multiples, mean that three to four such dyads of this political duo presented the king in the legitimising embrace of Khentkawes I, assimilated to Hathor.

The possibility of four dyads would amplify the themes of Menkaura’s legitimation, feeding and regeneration that are found in the triads. What the triads add are the nomes (and the unnamed but understood estates in those nomes) from which the king’s heb-sed was provisioned (Friedman 2011a, 105–11; Friedman 2015b, 18–24). Menkaura’s provisioner in the triads — the agent by which the provisions moved from real or imagined estates in the nomes to the king — is his divine mother Hathor. Menkaura’s provisioner in the dyad(s) is his human mother (identified with Hathor), who is understood to feed him from her body. Both divine and mortal mothers who gave birth to him, symbolically or in fact, also provide regeneration. Menkaura’s greywacke statues thus offer the king legitimation, provisioning and regeneration through his divine and earthly mothers (Friedman 2008, 139; on rebirth through Hathor in the triads, see Seidel 1996, 47). If Lehner is correct that Khentkawes I ‘may have been venerated within Menkaure[’s] valley temple in a close association with Menkaure’ (Lehner 2015, 215), then these multiple dyads of mother and son/queen mother and king provided a double cultic focus in the valley temple and, I would add, probably at his hebsed. The specific location for the dyads may have been the valley temple’s Annex, the extension on the eastern side of Menkaura’s valley temple (Lehner 2015, 246–7, fig. 21) that was a mere 15–20m from the Khentkawes settlement (Lehner 2015, 243, fig. 19). Lehner speculated that the Menkaura valley temple Annex was where ‘Khentkawes I was worshipped within the valley temple of Menkaura in the guise of Hathor’ (Lehner 2015, 273). If there were indeed four dyads, they may have flanked an opening or niche in an Annex room, with the two taller ‘Boston’ dyads on the inner side of an opening and the two smaller dyads on the outer (see Fig. 24). As a final observation, the statue programme of the valley temple included not only greywacke dyads and triads, but also four alabaster seated statues in the offering room, flanking the door to the sanctuary (Friedman forthcoming), as well as probably the alabaster seated colossus in the sanctuary itself, based on fragments that parallel those of the colossal statue fragments from the pyramid temple (Reisner 1931, plan VIII). This assemblage of statuary, acting much like the dramatis personae in a ritual drama — and possibly augmented by smaller excavated statues in different media, sizes and types (e.g., Reisner 1931, pl. 62, esp. j = MFA, Boston 11.735; pl. 63 except e, f; pl. 64a, c, f, i) — formed

EVIDENCE SUGGESTING ANOTHER BOSTON DYAD OF MENKAURA AND A QUEEN

91

Fig. 23: MFA 11.1738* dyad and proposed smaller dyad using MFA 47.1747* advanced left leg fragment. Photography: Michael Fredericks (MFA 11.1738) and Michelle Pisa (fragment). Grey areas are reconstruction by Florence Dunn Friedman, drawn by Michelle Pisa. *Courtesy, Museum of Fine Arts, Boston.

Fig. 24: a. MFA 11.1738* dyad and proposed duplicate incorporating MFA 47.1756*, MFA 47.1765*, and MFA 47.1751*. b. proposed smaller dyad based on MFA 47.1747* fragment and fourth dyad suggested for symmetry. Photography: Michael Fredericks (MFA 11.1738) and Michelle Pisa (fragments). Grey areas are reconstruction by Florence Dunn Friedman, drawn by Michelle Pisa. *Courtesy, Museum of Fine Arts, Boston.

92

F.D. FRIEDMAN

a grand theatrical production for the afterlife, little of which has survived intact and less of which made its way to the final installation. Bibliography Arnold, D. 1999. When the pyramids were built. New York. Bothmer, B. V. 1950. Notes on the Mycerinus triad. Bulletin of the Museum of Fine Arts 48 (271), 10–7. Callender, V. G. 2011. In Hathor’s image I: The wives and mothers of Egyptian kings from Dynasties I– VI. Prague. Fakhry, A. 1961. The monuments of Sneferu at Dahshur II: The valley temple, Pt. 1. Cairo. Fischer, H. G. 1975. An elusive shape within the fisted hand of Egyptian statues. The Metropolitan Museum Journal 10, 9–21. Friedman, F. D. 1995. The underground relief panels of King Djoser at the Step Pyramid complex. Journal of the American Research Center in Egypt 32, 1–42. ———. 2008. The Menkaure dyad(s). In S. E. Thompson and P. D. Manuelian (eds), Egypt and beyond: Essays presented to Leonard H. Lesko upon his retirement from the Wilbour Chair of Egyptology at Brown University June 2005. Providence, 109–44. ———. 2011a. Reading the Menkaure triads, Part I. In R. Gundlach and K. Spence (eds), Palace and temple. 5th Symposium on Egyptian royal ideology. Wiesbaden, 23–41. ———. 2011b. Reading the Menkaure triads: Part II (Multi-directionality). In N. Strudwick and H. Strudwick (eds), Old Kingdom, new perspectives: Egyptian art and archaeology 2750–2150. Oxford; Oakville, 93–114. ———. 2015a. The cultic relationship of the Menkaure triads to the small step pyramids. In F. Coppens, J. Janák and H. Vymazalová (eds), Royal versus divine authority: 7th Symposium on Egyptian royal ideology. Wiesbaden, 95–107.

———. 2015b. Economic implications of the Menkaure triads. In P. Der Manuelian and T. Schneider (eds), Towards a new history of the Egyptian Old Kingdom: Perspectives on the Pyramid Age. Proceedings of a symposium at Harvard University 26 April, 2012. Harvard Egyptological Studies 1. Leiden; Boston, 18–59. ———. forthcoming. The names of Menkaure. In K. O. Kuraszkiewicz, E. Kopp and D. Takács (eds), Old Kingdom art and archaeology. Warsaw, 111–30. Giza Archives, The. (last accessed 01/08/2018). Labbé-Toutée, S. and Ziegler, C. 1999. Triad of King Menkaure, catalog no. 68. In D. Arnold, C. Ziegler and K. Grzymski (eds), Egyptian art in the age of the pyramids. New York, 272–3. Lacovara, P. and Reeves, C. N. 1987. The colossal statue of Mycerinus reconsidered. Revue d’égyptologie 38, 111–5. Lehner, M. 2015. The monument and the formerly socalled valley temple of Khentkawes I: Four observations. In F. Coppens, J. Janák and H. Vymazalová (eds), Royal versus divine authority: 7th symposium on Egyptian royal ideology. Wiesbaden, 215–74. Reisner, G. A. 1931. Mycerinus: The temples of the third pyramid at Giza. Cambridge, MA. Seidel, M. 1996. Die königlichen Statuengruppen I. Hildesheimer ägyptologische Beiträge 42. Hildesheim. Stadelmann, R. 2011. The heb-sed temple of Senefru at Dahshur. In M. Bárta, F. Coppens and J. Krejci (eds), Abusir and Saqqara in the year 2010. Prague, 736–46. Wood, W. 1974. A reconstruction of the triads of King Mycerinus. Journal of Egyptian Archaeology 60, 82–93. Ziegler, C. 1999. King Menkaure and a queen, catalog no. 67. In D. Arnold, C. Ziegler and K. Grzymski (eds), Egyptian art in the age of the pyramids. New York, 268–71.

III VISIBLE STATUES: TEMPLES, PALACES AND HOUSES

IN TEMPLE AND HOME: STATUARY IN THE TOWN OF AMARA WEST, UPPER NUBIA Neal SPENCER

Abstract Amara West, founded as a new administrative centre for Pharaonic-occupied Kush around 1300 BC, provides the case study for exploring statue provision in a small, urban, though ideologically charged, environment. Across temple and domestic contexts, the small assemblage elucidates how inhabitants engaged with figural representations and alerts us to their relative rarity. An overview of statuary from other Nubian sites reminds us to expect changing emphases in display and commemoration, and distinct and local approaches to statue provision. *

*

*

As a Pharaonic foundation conceived as the new centre for the administration of Upper Nubia, with a decorated sandstone temple and a Residence for the Deputy of Kush, we might expect Amara West to have been furnished with numerous statues. The foundation of the town and its temple, subsequent periods of investment, or state activity such as tribute gathering and military campaigns, would seem to be apposite moments for the dedication of statuary, whether royal or private. Rather, the five modest statues recovered from Amara West principally reflect the dedication of humble monuments, whether in the temple or house, fashioned from materials available nearby. Given the preservation of the site, and the thorough and extensive excavations undertaken in 1938–9 and 1947–8 (Egypt Exploration Society [EES]) and then from 2008 to 2017 (British Museum), the absence of evidence for a statuary programme is here taken as primarily reflecting an ancient phenomenon rather than a product of the vagaries of preservation. The Amara West statues are presented here through the lens of the distinctive environments and social contexts in which they were used and then discarded or abandoned: temple and house. A brief discussion of the cemetery is also included. Thereafter, the question of whether the corpus is representative, rather than the result of a systematic removal of portable monuments,

is addressed, before evidence from other sites in Nubia is considered. The prevalence, or otherwise, of statuary within Egypt’s colonies has received surprisingly little attention, however, beyond the colossi of Abu Simbel and studies of statuary from Sai. Hein’s study of Ramesside building projects in Nubia devoted little attention to the question of statuary (1991). This case study prompts considerations upon the availability of materials, and whether the function (and changing nature) of these Pharaonic towns contributed to the presence or lack of statuary programmes. Furthermore, divergent approaches to statuary become evident in Lower and Upper Nubia, and a further distinction is apparent between the nature of Dynasty 18 and Ramesside non-royal dedications in temple spaces. Temple statuary Representations of statues The temple of Amara West is one of the best-preserved Ramesside cult buildings outside the Theban area. Cleared of sand and hastily recorded during one field season (12 November 1938 to 4 March 1939), the monument has only recently been published, on the basis of archival documentation, by Patricia Spencer, in volumes dedicated to the architecture (1997, 27–97, pls 15–74) and decoration (2016). Located in the northeastern corner of the walled town (Fig. 1), the temple covered an area of 41 × 16.5m, with ancillary buildings surrounding it on three sides, the front being against the northern town wall; an outer forecourt (27 × 20m) projected north beyond that wall. The temple core comprised a peristyle court, hypostyle hall, broad vestibule and three contiguous sanctuaries, with a staircase leading to the roof off to the right as one approached these sanctuaries. The decoration principally dates to the reign of Ramesses II, with additions in the reign of Merenptah, Amenmose, Ramesses III, Ramesses VI and Ramesses IX. The last king is represented by a large-scale inscription, dated to year 6, around the peristyle court, beneath the scenes of Ramesses II (P. Spencer 2016, 10–2, pls 20–35). A number of

96

N. SPENCER

Fig. 1: Amara West: plan of town with find-spots of statuary marked in red: 1: Ram statue; 2: Block statue of Amenemhat; 3: Statue of a man; 4: Ceramic statue with nemes-wig; 5: Anthropoid bust. Exact locations of each object were not recorded by the EES; for the find-spot of Cat. 5, the anthropoid bust, see Fig. 8. Courtesy of the Trustees of the British Museum (Amara West Research Project).

inscriptions were added to the temple by high officials, particularly viceroys and deputies of Kush, providing further detail on the periods of decoration, refurbishment and adjustments within the temple. These include the viceroys Hori I (temp. Siptah/Sethnakht), Hori II (Ramesses III), Siese (Ramesses VI), Ramsesnakht, Naherhu and Wentawet (all Ramesses IX; see also Auenmüller 2018). It thus seems that a temple built and decorated under Ramesses II was maintained and embellished throughout Dynasties 19 and 20, though without significant architectural changes. In many ways, this setting appears typical of New Kingdom towns, yet not one stone royal statue has been recovered, nor any processional statuary such as sphinxes. However, a scene on the west jamb of the doorway between the hypostyle hall and vestibule (Fig. 2), thus in a prominent location, depicts a figure

of ‘Ramesses who resides in Pr-Rꜥmss-mry-Ἰmn’ addressing a human-headed ‘Amun-Ra lord-of-thethrones-of-the-Two-Lands’, with the accompanying text reading: The great gate ‘User-maat-ra is sacred (ḏsr) and terrifying (šfyt), may he be granted life’. He made as his monument a ‘living statue’ (ẖnty ꜥnḫ) upon earth. He made the august temple in beautiful white stone, resembling (stw) the horizon of heaven. (P. Spencer 2016, pl. 38a–b)

This attests to the provision of statuary in the reign of Ramesses II, and specifically a form of statue (ẖnty) perceived as being portable, whether in reality or symbolically (Loeben 2001, 128, 130–1). A close parallel for this scene, on a doorway at Soleb, features Amenhotep III striding with arm outstretched and wearing the blue crown, labelled as the ‘son of Ra, Nebmaatra,

IN TEMPLE AND HOME: STATUARY IN THE TOWN OF AMARA WEST, UPPER NUBIA

97

beloved of his living statue upon earth (ẖnty ꜥnḫ tp tꜢ)’, before an image undoubtedly of Nebmaatra, lord of Nubia (Schiff Giorgini 1998, pl. 195). The sanctuary and vestibule

Fig. 2: Epigraphic copy of decoration on the north (front) face of the west jamb of the gateway between the hypostyle hall and the vestibule. Courtesy of the Egypt Exploration Society.

Royal and ritual statues are also depicted in the central sanctuary of the Amara West temple. A large-scale barque bears a veiled shrine, embellished with ritual figures; the pedestal names the god within as Amun-Ra lord-of-the-thrones-of-the-Two-Lands. Beneath, amongst piles of offerings including cattle, flowers and jars of liquids, are depicted small standing royal statues, all identified by cartouches (Ramesses II) and shown standing on rectangular bases atop sleds (P. Spencer 2016, pls 138–46). Such statues could be made of stone, metal or a composite of materials such as wood and various inlays. The central sanctuary was fitted with a barque support, against which was placed a small pedestal with steps and a socket on its upper surface (P. Spencer 2016, pl. 136b). This likely held the central cult image, perhaps housed in a wooden shrine and/or barque, as suggested by the reliefs upon the sanctuary walls (P. Spencer 2016, pls 138–46). A crystal eye assumed to be a statue inlay (Sudan National Museum 3107 = AW134; P. Spencer 1997, 49) was recovered from the temple’s west sanctuary, though in an unclear context; the eye could also have been destined for a wooden coffin, as eye inlays are attested in the elite cemetery D.1 In the right (east) sanctuary, a sandstone image of a ram head upon a plinth was found (Cat. 1, see Figs 13–6). The form of the statue, with curved horns, wig and a short beard, mirrors the form found in relief and painted representations (Spiegelberg 1921, no. 307). Provided with a hole cut into the top of the head, the statue could have been fitted with a headpiece such as a sun-disc and uraeus, as depicted on various representations of these ram images (Spiegelberg 1921, nos 280, 307). The additional holes drilled into the side of the Amara West statue’s head may have been to support horizontal horns, perhaps of wood or metal; similar holes are found on a ram statue from Dokki Gel (Bonnet and Valbelle 2018, 84, fig. 64 [294]). The base of the Amara West ram statue evokes temple architecture through its shape and the cornice, again consistent

1

A stone eye inlay, presumably from a coffin, was found in tomb G322 (F8973).

98

N. SPENCER

with 2- and 3D depictions of such images, some of which more explicitly echo the architectural vocabulary of Pharaonic temples (e.g. Capart 1944, fig. 22). The Amara West ram was once brightly coloured, consistent with depictions in Theban tombs (see Fig. 3) or on statues such as that of Penshenabu/Penmerneb from Deir el-Medina (Trapani 2015, 121 [131]), which combines both blue and black paint for the striated ‘wig’ of the ram, with the beard also painted black. A scene on the plinth of the Amara West ram (see Fig. 16),2 of a private individual driving a bovid, may evoke the eternal offerings destined for the god, similar to larger compositions on temple and tomb walls. In these, cows are led towards a deity (Epigraphic Survey 1994, pls 105–6) or are included in funerary processions (Davies and Davies 1963, pl. 1) and representations of revenues from estates (Davies and Davies 1933, pl. 8) or agricultural work (N. de G. Davies 1943, pl. 39). The clothing worn by the man is familiar from relief depictions at Amara West of officials and priests throughout the Ramesside era (P. Spencer 1997, pls 150b, 156, 159). Another ram statue (Bonnet and Valbelle 2018, 84, fig. 64), in sandstone, painted with yellow skin and red eyes, and with fittings for a headdress and horizontal horns, was found on the floor of a Thutmoside temple sanctuary at Dokki Gel, with a ram-headed standardfitting recovered nearby. This Dokki Gel ram statue is somewhat smaller (22cm in height), and thus more portable, than the Amara West example (43cm in height); it also sits on a more simple cuboid base. Fine ram sculptures were already present at Kerma in earlier times: a quartzite head of a ram from Tumulus III (Classic Kerma) may have been one of the objects brought from Egypt as loot (Gabolde 2018, 95–7, fig. 4). In Egypt itself, a statue type that combines the ram image with its donor, whether standing or seated (e.g. Poole 2015, 93–4 [91]; Connor 2016, 12–6), can reference forms of Amun in the inscriptions (Connor 2016, 12–3). A life-size statue of Sety II dedicating such an image was set up at Karnak (Russmann 2001,

178–9 [90]). Ram images appear in New Kingdom graffiti in the Theban mountains, usually identified with a form of Amun-Ra.3 The identity of the deity represented, in both the Dokki Gel and Amara West examples, is thus almost certain: a form of Amun-Ra.4 This god was often depicted as a ram, or ram-headed, since early Dynasty 18 (Davies 2017a, 71, figs 6–7). In the Amara West temple Amun-Ra ‘lord-of-the-thrones-of-the-TwoLands’ is commonly represented, though in anthropomorphic form (e.g. P. Spencer 2016, pls 44–5, 66).5 Amun-Ra is depicted as a ram on a number of stelae at Amara West and elsewhere in Nubia (Bonnet and Valbelle 2018, 83–4, fig. 63). One example, with the ram in full zoomorphic form before a brazier with the donor depicted below, designates the god as AmunRa the ram (pꜢ rhny EES obj. 224: P. Spencer 2016, pl. 206g). Another, found in a house, designates the ram as Amun-Ra lord-of-the-thrones-of-the-Two-Lands (N. Spencer 2009, 50, pl. 6 [now Sudan National Museum 35822]). Such images were obviously believed to be effective manifestations of the divine, given that graffiti depict them receiving offerings which can include a stand, ḥs-jars for cool water, bread loaves and floral bouquets (Spiegelberg 1921, no. 280; Spiegelberg 1927, fig. 1). A small private stela now in Leiden (Salvador 2015, 367 and 551–2 [VI. 72]) depicts an individual kneeling in a pose of prayer before such a ram image, while the late Dynasty 19 tomb of the ‘wꜥb-priest of Amun, and wꜥb-priest of Amenhotep (I), the image (pꜢ ỉbỉb) of Amun, Huy’ depicts such a ram image in a double shrine with carrying pole (Fig. 3) — the god is named as ‘Amun-Ra lord-of-the-thrones-of-the-Two-Lands’ (Betrò, del Vesco and Miniaci 2009, 98–100, fig. 70 and pl. 2). At Deir el-Medina, such images were set up inside the Hathor temple (Bruyère 1952, 44, 59, fig. 125). Where could the Amara West ram statue have been installed, originally? The statue was found in one of the subsidiary sanctuaries, so we must consider whether

2

4

3

A very unusual example of plinth decoration combines a ramimage (on plinth) before a standing figure of Taweret (Bianchi 2016). For example, Amun-Ra lord-of-Opet (Spiegelberg 1921, nos 307, 349), Amun-Ra lord-of-the-thrones-of-the-Two-Lands (Betrò, del Vesco and Miniaci 2009, pl. 2), simply Amun-Ra (Spiegelberg 1921, no. 280) or ‘the perfect ram Amun’ pꜢ rh(n)y nfr Ἰmn (Darnell and Darnell 2014, 47, pls 3B, 5B).

5

That ram-heads could be dedicated to Khnum and the Elephantine triad is attested with the sandstone statue of Ken, possibly from Sehel, representing him offering a ram-head to Khnum (Andreu 2000, 24, fig. 2). ‘Amun-Ra king-of-gods high-of-the-šwty-crown’ is invoked upon a finely decorated house door-lintel (N. Spencer 2014b, 50, figs. 9–10).

IN TEMPLE AND HOME: STATUARY IN THE TOWN OF AMARA WEST, UPPER NUBIA

99

it was what Fairman described as ‘a cultus image’ (1944, 143).6 This room yielded a small stela depicting Ramesses II (P. Spencer 1997, 45 [140]), and a crystal inlay (Sudan National Museum 3107 = AW134; P. Spencer 1997, 49), perhaps from a statue. However, given that there are a number of objects of rather domestic character (rings, scarab, inlays, net sinker, spear-heads), we can also expect that some of the objects found in the sanctuary are the result of secondary or tertiary deposition. The central sanctuary was fitted with a shrine pedestal (P. Spencer 2016, pl. 136b) and scenes of the barque of Amun-Ra (P. Spencer 2016, pls 136–47). One might expect a larger cult image to have been placed on this pedestal, or perhaps a shrine containing the cult image. The two subsidiary chapels bore decoration which focused on other gods, though Amun-Ra does appear. In the western chamber, Satet is shown Fig. 3: Ram-image housed in shrine, receiving offerings. proffering life to the king, pharaoh Tomb of Huy (TT14), Dra Abu el-Naga. Photograph: offering to an ibis-headed moon-god, Marilina Betrò. University of Pisa Mission at Dra Abu el-Naga. a ram-headed Herishef and gods with crocodile- and jackal-heads; one scene showed the king nostic piece with incised decoration. This featured before a seated Amun-Ra lord-of-the-thrones-of-thea column of hieroglyphs, with signs that could be read Two-Lands (P. Spencer 2016, pls 147–56). The eastern as n nb (‘of the lord of…’) or n Kš (‘of Kush’), flanked sanctuary, where the sandstone ram was recovered, feaby a decorative pattern of strokes and U-shapes to tured scenes of the king before the First Cataract triad, evoke feathering. The reverse surface is notably less a seated human-headed Amun-Ra, and standing figures of Amun-Ra lord-of-the-thrones-of-the-Two-Lands, [Mut] lady of Asheru, Ra-Horakhty and Ptah (P. Spencer 2016, pls 126–35). All three sanctuary spaces might have been suitable settings for the ram statue. Around fifty fragments of gold leaf were recovered from the eastern sanctuary, badly crumpled and folded (British Museum EA 86286, EES obj. 244, Fig. 4). Conservation of these fragments yielded a larger diag-

6

Given that Khnum is also represented in the Amara West temple (see below), it remains possible that the ram statue embodies that divinity, but this is rather unlikely. Fairman (1944, 143) interpreted the statue as a representation of Khnum, but the abundant evidence cited here indicated images of ram busts were forms of Amun.

Fig. 4: Gold leaf with incised decoration. British Museum EA 86286 (EES obj. 244), from the east sanctuary of the temple. Photograph: Manuela Lehmann. Courtesy of the Trustees of the British Museum (Amara West Research Project).

100

N. SPENCER

smoothed: could this gilding have been applied to a statue or figure, perhaps of wood or stone? The feathered motif is found on images of Ptah and certain goddesses (Hill 2016, 270–72). Alternatively, it may have been applied to another form of object, such as a wooden box. Statuary and furnishings, in a number of colours and materials, evidently once adorned the inner rooms of the temple. Statuary in the vestibule and hypostyle hall The small scale of the sandstone ram image, but especially the relief scene depicting a private individual, suggest this ram sculpture was a donation by an individual, and as such more likely to have been set up elsewhere in the temple. The purpose of donating such an image must have been twofold: to expound one’s piety and standing in society, but also to encourage intercession by the deity. A wide range of sources reflect the healing, hearing nature of Amun-Ra (Guglielmi and Dittmar 1992), which surely prompted the deposition of more modest clay figurines of ramheads, including at nearby Sai,7 or found alongside other animal forms such as crocodiles and vultures at Abydos (O’Connor 1969, 38). That small statues were displayed within the temple at Amara West is clear from the survival of two small stepped pedestals, one in the vestibule (P. Spencer 2016, pl. 123b), with another set against the barque/shrine pedestal in the central sanctuary (P. Spencer 2016, pl. 136b). More enclosed settings were also possible: a small naos found installed in the hypostyle hall (P. Spencer 2016, 16, pls 62–5) was dedicated under Amenmesse, and later redecorated under Ramesses VI, with a scene showing the viceroy Siese. The scale of this is not recorded, but given the depth of the adjoining buttress (60cm; P. Spencer 1997, 40), its interior cavity is likely to have been no more than 50cm deep and 50cm wide. A published photograph (P. Spencer 2016, pl. 62) suggests that the interior floor featured a narrower slot to fit the statue inserted inside. The form and size of the cavity would work well with this ram statue. The frontal focus to the statue (summarily modelled sides, and back, relief decoration on front only) might further support the suggestion it was housed in a niche or shrine. That this shrine is one possible setting for the ram statue, or a similar one, is suggested by the reference 7

SAV1N 2221 cited by Budka (2017a, 159) but not illustrated.

to ‘Amun-Ra king-of-gods’ within the scene added by Siese. He is shown worshipping the cartouches of Ramesses VI, the accompanying large-scale inscription giving the name and epithet ‘Amun-Ra king-of-gods, lord-of-heaven ruler-of-Thebes who creates eternity’. The smaller-scale hieroglyphs on the shrine’s jambs bear the royal titulary with epithets invoking Amun-Ra lord-of-the-thrones-of-the-Two-Lands and Thoth (P. Spencer 2016, pl. 63). The west side of the shrine bears two private dedications, by the Deputy of Kush Usermaatra-nakht and the Viceroy Wentawet (temp. Ramesses IX; P. Spencer 2016, pls 64–5). Another shrine was installed in the hypostyle hall: inscribed for Ramesses II, it was set on a larger pedestal to the right of the temple axis between two columns and embellished with small obelisks bearing the royal titulary (P. Spencer 1997, pls 41a, 42; P. Spencer 2016, 20, pl. 105). The underlying pedestal was richly decorated: beneath a cavetto cornice and torus moulding was a royal titulary surmounting a sema-tawy scene. The shrine could have housed a figure up to 65cm in height and 35.5cm in width, so the Amara West ram statue presented here would have comfortably fitted within. The depositional environment at Amara West precludes the preservation of wooden objects, owing to termite activity, so any wooden shrines (or statuary) installed in the temple would have left no trace. The ram statue, given its scale, could have been easily relocated, but was probably too heavy to have been regularly used in processional rituals. Nonetheless, that such images could be used as processional statues is clear from a Theban tomb depiction of carrying-poles supporting a shrine containing such a ram image (see Fig. 3; see Betrò, del Vesco and Miniaci 2009, pl. 2). An ostracon (Fig. 5; British Museum EA 85675 = EES obj. 203) found at Amara West provides further evidence that such ram images could be cult objects, and perhaps be portable. The ostracon bears an image of such a ram, wearing a striated headdress, a sun-disc, uraeus and horizontal horns. A schematic rendering of two flexed arms offers up the image; even more summary renderings of arms are found in the Theban graffiti (Spiegelberg 1921, no. 788). A further sun-disc, uraeus and ram horns combination beneath, in a thicker line, might have been part of a similar image. At Thebes, examples are also known of two ram-heads shown in close proximity (see Spiegelberg 1921, pl. 41). Two further elements that appear above the ram on the Amara West ostracon — a bird and an oval object — seem unrelated, with two lines of literary hieratic above.

IN TEMPLE AND HOME: STATUARY IN THE TOWN OF AMARA WEST, UPPER NUBIA

That the ostracon was found (as were two others) in E.14.9, the second of four vaulted magazines built east of the hypostyle hall (P. Spencer 1997, 56–7, pl. 16), seems fitting, given that sacred images and rituals would have been familiar to those circulating in these spaces. What is clear is that some inhabitants at Amara West were familiar with ram-heads as objects of cultic focus, much like their counterparts at Thebes.

101

halls and before temple pylons, generally along the temple axis (Bernhauer 2010, 97–98). The Amenemhat statue was found in the peristyle court, though the EES excavators did not record the exact position. The inscription down the front of the statue states that the individual was to benefit from ‘all that comes forth from the offering table of Amun-Ra’, underlining a primary intention behind such dedications: to benefit from the reversion of offerings. Amun-Ra is qualified with the epithet ‘foremost of the northern limits8 of Shaat [Sai]’ on the front, with Ra-Horakhty and AmunRa lord-of-the-thrones-of-the-Two-Lands also cited. The reference to Sai might suggest the individual was also active there, perhaps around the time of the transition of the Deputy’s Residence between Sai and Amara West. Ancillary spaces: magazines, corridors and chapels

Fig. 5: Ostracon, British Museum EA 85675 (EES obj. 203), from E.14.9. Drawing: Claire Thorne.

Peristyle court Perhaps appropriately, it is in the outer part of the temple where we encounter images of non-royal individuals. The fine, if small, block statue of Amenemhat (Cat. 2, see Fig. 17), the scribe of counting gold, is of a form prevalent in New Kingdom temples (Bernhauer 2010, 94) owing to its solidity and provision of considerable yet attractive surfaces for inscription. Block statues are known to have been set up in the columned

8 9

pḥ.ww, see Wb I, 538, 4 and 5–6. At Amara West, officials associated with granaries — such as Horhotep — were recorded on decorated doorways (Spencer, Stevens and Binder 2014, 19, 34).

The Amara West temple was flanked by a series of mud-brick magazines, subsidiary chapels and ancillary spaces, which had a complex history of reuse and repurposing (P. Spencer 1997, 53–74). One chapel (P. Spencer 1997, 50–1; 2016, 27–8, pls 163–76) dedicated to the Theban triad was decorated with scenes of pharaoh offering before a pedestal inscribed for AmunRa ‘lord-of-the-thrones-of-the-Two-Lands’, which must have supported the divine barque. Indeed, a small barque stand (190 × 90cm) still stood in the centre of the chapel, embellished with a cavetto cornice. This location for a barque stand indicates that statues were expected to move around the sacred spaces of the temple: these were not static images (see Loeben 2001). Evidence from Egypt reveals that magazines and ancillary rooms could also house statues. A naophorous statue of Sebty was found in a magazine room at Karnak-North (Jacquet 1994, 19; Jacquet-Gordon 1999, 114–9 [69]), within a layer of dismantled masonry. We need not assume this was a later dump of statues (suggested by Bernhauer 2010, 97–8), as a depiction in a Theban tomb of the reign of Ramesses II (TT 218, Ipuy) shows a statue upon a plinth within a granary, while two Dynasty 18 statues bear inscriptions indicating they were intended for granaries (Mougenot 2012).9 Another private statue (Cat. 3, see Figs 18–9) was found in the corridor between the hypostyle hall and a suite of four magazines, along the eastern side of the temple. The excavators describe it being found in ‘fill’ (P. Spencer 1997, 60–1, [169]), so we cannot be certain which phase it relates to, or whether it was ever set up

102

N. SPENCER

in this space rather than, say, in the peristyle or hypostyle courts. It remains difficult to interpret the original form of this modest statue (perhaps no more than 30cm in height when complete), made of poor-quality, perhaps local, sandstone. The form of the body and head10 are consistent with a block statue, but the area of broken stone beneath the chin (see Fig. 19) suggests it was another statue form, assuming we are not seeing the remains of a never-completed statue. It is also unlikely to be a statue depicting a man holding a basin (Bernhauer 2010, 68–71), as the composition is not typical of these, which favour arms extending out horizontally from the shoulders to grasp the basin (Bernhauer 2010, pl. 34), flexed (Bernhauer 2010, pls 35–6) or projecting horizontally forward (Bernhauer 2010, pl. 33). Furthermore, the chin does not typically rest on the basin itself. It may have been a theophorous statue, as these can feature a direct interface between the image of the god held and the donor’s face (for example, Bernhauer 2010, pl. 2 [1.4-17]), or a ‘nursing’ statue of the type favoured by Senenmut (Bernhauer 2010, 250–75, pls 21–9), though this form is rare after Dynasty 18. A further option is a mendicant statue, known from the Ramesside period, where the figure holds a hand before the mouth, palm upturned to receive water offerings from passers-by (Clère 1995; see also Frood in this volume). The statue may have held an emblem or element in front and to the right of them, as found with seated harpists11 and men playing two flutes or pipes.12 Finally, at this scale, it could have acted as a doorblock, as found with a number of distinctive Ramesside statues with a projecting element at the back, to hold doors open or shut (Rondot 2011). The ancillary buildings on the opposite side of the temple yielded another statue, perhaps the most

10

11

12

The simple wig is unusual. A block statue — incidentally one with a nursed child’s head — of Benermut, features a similar shape, though with the wig carved with details (Bernhauer 2010, 266–7, pl. 28 [6.15-42]; see Legrain 1909, 37–8, pl. 35 [CG 42171]). For a Ramesside short wig see the mendicant block statue of Sedjememwaw from Terenuthis (Schulz 1992, 56–7, pl. 34), or the much larger and finer Montuhirkheoeshef from Bubastis (Schulz 1992, 88–9, pl. 9). A statue of a harpist in Chicago (OI E11073) is only 6cm high, and perhaps of New Kingdom date. I would like to thank Angela Tooley for this suggestion. [accessed 3 February 2019]. These are usually dated to the Late Period, and small in size. See for example Petrie Museum UC4530 (I would like to thank

interesting of the Amara West group (Cat. 4, see Figs 20–1), a small-scale representation of a male figure wearing the nemes-wig. It was found in E.14.1, one of a small suite of rooms immediately west of the peristyle court (P. Spencer 1997, 70–2, pl. 62), with an additional suite of rooms added to the west (E.14.1A, E.14.1B, E.14.1C) and accessed via a doorway cut into the magazine enclosure wall. These rooms were built of mud brick, fitted with stone doorways, one inscribed for Ramesses II. Objects found within include hieratic ostraca, a copper nail and point, as well as stone rings and a pendant. These objects are described as being from the ‘upper level’, and there is no stratigraphic detail that allows us to understand whether they relate to the same use phase, or represent post-occupation deposits. What form of statue does this ceramic head come from? It may have been a representation of pharaoh, perhaps striding; an alternative is that the head formed part of a sacred standard or staff. As a statue, it would have had a partly hollow core, and perhaps been made out of several adjoining elements. Ceramic statues of pharaoh are very rare, though perhaps significantly the known examples are all of New Kingdom date. Three were found in the Karnak Cachette, including two royal heads in ceramic, of around ¾-life-size scale. Datable to the reign of Amenhotep III on the basis of the distinctive style, these include one statue (JE 38594)13 with nemes and uraeus (and again, no beard), partly covered in a red slip. The other (JE 38597)14 wears the blue crown, with a fine modelled uraeus. In the latter case, the crown is painted a vibrant red colour. That these formed parts of full-length sculptures is indicated by Legrain’s notes, in which fragments of other body parts with a red slip were found, including a back pillar inscribed for a king named Amenhotep.15

13 14 15

Aurélia Masson-Berghoff for drawing my attention to this statue). [accessed 3 February 2019]. [accessed 3 February 2019]. Legrain stated: ‘[n]ombreux fragments de statues de grandeur naturelle en terre cuite avec […] couverte en rouge vif’ and ‘j’ai signalé l’an passé la découverte de nombreux fragments de grandes statues de terre cuite, recouvertes d’une épaisse couche de rouge vermillon. Un morceau de pilier de statue semblable, trouvé cette année, portait la mention du [sꜢ-r῾ (Jmn-ḥtp)|], fils du Soleil Aménôthès. Ces monuments peuvent donc être datés au moins de la XVIIIe dynastie’. (these have yet to be located in the Egyptian Museum [accessed 3 February 2019]).

IN TEMPLE AND HOME: STATUARY IN THE TOWN OF AMARA WEST, UPPER NUBIA

103

A 46cm-high statue of a striding Ramesses IV, also recovered from the Karnak Cachette, made of steatite or slipped ceramic (‘red fired clay, with white slip’,16 CG 42151; Legrain 1909, 16–7, pl. 14), is of very similar dimensions to the Amara West head. It depicts the king with hands resting on a projecting kilt, wearing a nemes modelled with incised detail, provided with a head-band and uraeus with tail running across the top of the head. The king wears a beard, with wavy striated lines running down it; a finely modelled face is prepared to take inlays for the eyes and eyebrows, and has ear-holes for earrings, perhaps of precious stone, faience, metal or glass. Traces of blue are noted in the inlay cavities. Further detail is found on the kilt and belt, with a cartouche incised on each upper arm, and the prenomen on the belt. The statue strides upon a cuboid base, inscribed with the royal titulary, with the epithet ‘beloved of Amun-Ra king-of-gods’, while the back pillar designates the statue as a twt, distinct from the portable ẖnty-statue depicted on the door from the hypostyle hall to the vestibule at Amara West (Fig. 2). All three of these ceramic statues from the Cachette are of extremely high-quality work, with an attention to facial form, application of a surface slip and in one case (Ramesses IV), prepared for inlay. That these statues were carefully deposited in the Cachette underlines the esteem in which they were held by the priests; these were not lower-class objects made by those without access to ostensibly more expensive material such as stone, wood or precious metal. The function of such statues, as opposed to the stone statues that embellished courtyards, pylons and processional axes, or the cult statues in precious material, requires further elucidation.17 As the ceramic fabric of the Amara West statue appears consistent with locally produced pottery, it is likely to have been made on site. A number of ‘bread ovens’,18 and also the small pottery kiln found in the earliest phase of occupation (N. Spencer 2017), would have been suitable places for the firing of such a statue. While there is no clear evidence of faience production at the site, a number of fired clay objects found at the site — including a ceramic shabti (G301, Fig. 6) and

EES obj.945 (Fig. 7) from a room near the Deputy’s Residence — might indicate a local production of figurines and statuettes. Note that no ceramic coffins have been found at Amara West, though examples are known from Nubia.19 The statue may, of course, once have been fitted with inlays (e.g. eyes) in more precious materials, as with the Ramesses IV example discussed above. If not a representation of pharaoh, what other statue forms could such a head come from? It remains possible that the Amara West ceramic head formed part of a small sphinx; unfortunately the diagnostic area, such as the curve of the back, or the ‘tail’ of the nemes, is not preserved. Two fragments of ceramic sphinx statues were found at Soleb, with integrated bases, hollow interiors and ceramic walls of 6–7cm thick (Schiff Giorgini 2002, 428).20 The iconography of the nemes is also associated with the ‘souls of Pe and Nekhen’, statues of which were part of the New Kingdom ritual sphere, as depicted on processional barques, including upon the east wall of the central temple sanctuary at Amara West (P. Spencer 2016, pls 140–1). Statues of divine figures, fashioned in ceramic, have also been discovered in New Kingdom temples at Thebes and Koptos, though their date is uncertain. In the Ramesseum, over 250 fragments of statues, covered in a red slip, have been recovered (Lecuyot 2001– 2002). Some can be reconstructed as standing mummiform deities — perhaps Ptah-Sokar-Osiris — with a tripartite wig and hands crossed over the chest; one (S.10) bears a hole in the top of the head for a crownfitting, and another hole on the brow for the uraeus. Lecuyot interprets these as cult images, and some of the fragments were recovered from within the sanctuary area. Ceramic statues have also been found north of the temple of Amenoptep son of Hapu (face with striated wig, Robichon and Varille 1937) and at Medinet Habu. In the latter case, the two mummiform figures, 80 and 116cm in height, were found in the crypt of the main temple, perhaps representing Osiris and Ptah (Daressy 1906, 68, pl. 13 [CG 38234]; 124, pl. 27 [CG 38465]); their date remains unclear (Lecuyot,

16

19

17

18

[accessed 3 February 2019]. Trad and Mahmoud (1993, 45) suggested their portability would have made them suitable as sculptor’s models, but the careful application of painted detail might argue against this. It is likely such ‘bread ovens’ could also be used for the production of small fired clay objects (see Kemp and Stevens 2010, 483).

20

Cotelle-Michelle (2004) catalogues ceramic coffins from Dabod, Amada, Aniba, Toshka, Deberia and Soleb. See also Smith and Buzon 2018, 211–2 fig. 7. No photos or drawings have been published, and the present location of these is unknown.

104

N. SPENCER

Fig. 6: Ceramic shabti (F8004 [SNM 34559]) from shaft of tomb G301. Courtesy of the Trustees of the British Museum (Amara West Research Project).

Fig. 7: Fired clay statue of a man (EES obj. 945), from E.13.2.Y9(d). Present location unknown. Drawing on 1948–49 object card. Image: Egypt Exploration Society.

IN TEMPLE AND HOME: STATUARY IN THE TOWN OF AMARA WEST, UPPER NUBIA

105

2001–2002, 135–6).21 A ceramic head, in a very different, less formal, iconographic tradition (with bulbous eyeballs, and strips of clay representing hair) was also found at Medinet Habu (Oriental Institute Museum 15554, Teeter 2003, 68 [33], with a datation to the Third Intermediate Period). Unfired images of Horus and Osiris were recovered from a funerary chapel at Deir el-Medina (Bruyère 1929, 76, fig. 39). Outside Thebes, a group of ceramic sculptures from Koptos now in the Petrie Museum (Petrie 1896, 5, pl. 5; Adams 1986, 33–8, pls 4, 21–4) features hollow interiors, and hands holding sceptres. Finally, near the Djer tomb at Abydos, a 77cm-tall statue embellished with pigment (blue eyes, black resin on the wig) and gold foil (Louvre E13951; Lecuyot 2001–2002, 12–3, 135, pl. 17b), further underlines the value ascribed to some ceramic statues. No ceramic statue fragments have been recovered from Qantir (H. Franzmeier, pers. comm.). An unusual group of ceramic statues was found at Saqqara, in a rock-cut space near the Khaemwaset monument, partly sealed with blocks from it (Yoshimura and Kawai 2002, 22–9). These comprise four standing lioness-headed Sekhmets, two seated lions and humans with unusual body positions (hands to ears); small statues of Pepy I are found on the left side of the leg of two Sekhmets, and the Horus name of Khufu, but the statues have been posited as Ramesside by Lecuyot (2001–2002, 137). Other ceramic statues of the New Kingdom include sculptor’s models,22 statues of nonroyal individuals,23 shabtis and anthropomorphic canopic jar lids (Dorman 2002, 30–6). In Lower Nubia, C-Group cemeteries preserve evidence of a tradition of local manufacture of fired clay figurines (Williams 1983, 97–9, pls 102–3). One other possibility remains for this small head: could it have been a fitting for a divine standard? Depictions of the sacred barque of Amun-Ra at Luxor show a series of standards beneath the barque pedestal, typically including ram-, woman- and falcon-headed examples, i.e. the Theban triad, but also one with a male head wearing a nemes-wig (Van de Walle 1952). These objects were the subject of offerings, as explicitly stated in the Medinet Habu offering calendar (Van

de Walle 1952, 126–7). It is generally assumed they were made of wood and/or precious metals; a bronze ram head from Dokki Gel might be such a standardfitting (Bonnet and Valbelle 2018, 84, fig. 64). The New Kingdom ‘standard-bearer’ statues often depict royal heads atop the standards, associated with the royal ka (Chadefaud 1982, 153–8, figs 9–10). This iconography is known from Nubia, including the colossal engaged depictions of Ramesses II at Wadi es-Sebua (Chadefaud 1982, 11–2, 24–7), and a depiction within the tomb of Penniut at Aniba in Lower Nubia names such a statue as ‘pꜢ twt of Ramesses, ruler of Heliopolis, son of Amun, beloved of Horus of Miam [Aniba]’, with the inscriptions recording lands assigned to support the cult of the statue (Chadefaud 1982, 72–3). The scale of the Amara West head might be consistent with such a function, and the form of the hollow on its underside would have provided a method of fitting the head to a staff.

21

23

22

See also the example in the North Carolina Museum of Art G.74.2.7; [accessed 3 February 2019]. E.g. British Museum EA 21820, Dorman 2002, 33, pl. 5A.

Statuary in houses Excavations at Amara West under the EES (1938– 39, 1947–50) and British Museum (2008–18) revealed several houses, often complex multi-phase buildings with long histories of occupation and reuse (N. Spencer 2015). Evidence for figural sculpture being employed in these spaces is limited, ranging from unfired clay figurines of women, to fittings from composite statues and one complete statue found in situ. Secondary evidence includes decorated niches that may have housed statuary. The Deputy’s Residence (E.13.2: P. Spencer 1997, 163–5, 167–70, 179–83; pls 112–8, 128–32) can in many ways be perceived as a large house, though one that must have fulfilled official needs also. It comprised formal reception areas, private apartments and service suite (N. Spencer 2017, 329–34, fig. 7). A staircase for access to the roof, or an upper storey, is only attested in its latest incarnation, built from reused architectural elements (P. Spencer 1997, 179, pls 116, 128b–c). The building was fitted with decorated stone doorways and columns topped with palmiform capitals (see P. Spencer

E.g. the statue base from the Karnak Cachette (CG 42136, Legrain 1906, 87; [accessed 3 February 2019]), or a small stelaphorous statue, painted red, of an official of Ramesses II (Petrie UC.10713; Page 1976, 85 (94)).

106

N. SPENCER

1997, pls 117, 129–32), yet no statue fragments were found other than a ‘bronze uraeus wearing the double crown’ found in the ‘upper layers’ of E.13.2K, a room on the southern side of the Residence in its last incarnation. It should be noted, however, that few objects were recovered from the Residence. Architectural settings for statues were not evident, though again caution is needed as the archive plans leave many ambiguities about the layout of the building at different phases. An anthropoid bust in the back room of a house A pair of adjacent houses, formed from the division of a larger house (E13.3), in the northwestern corner of the walled town (Fig. 8), yielded an in situ domestic statue, but also secondary evidence that might relate to

statuary, and examples of clay figurines. After division of the original house through construction of a long dividing wall during or shortly after the reign of Ramesses III, the two resulting houses each comprised a linear sequence of rooms, with four in the northern house (E13.3-N), and three in the adjacent dwelling. Both had distinctive back rooms in terms of find assemblages and deposits, a ‘main’ room with hearth preceding it, and staircases, while sharing a close that provided access to the alleyway (see N. Spencer 2014a, 469–80). Neither house was furnished with a mastaba, an unusual omission at Amara West. An anthropoid sandstone bust (Cat. 5, see Figs 22–3) was found set on a pedestal (Fig. 9) in the back room (E13.3.26), a type typically designated as an ‘ancestor bust’. This fits well within the corpus in

Fig. 8: Plan and elevation (inset) showing location of bust F4182 within house E13.3-S. Courtesy of the Trustees of the British Museum (Amara West Research Project).

IN TEMPLE AND HOME: STATUARY IN THE TOWN OF AMARA WEST, UPPER NUBIA

terms of the schematic rendering of the bust,24 the apparent depiction of a male given the red skin and shape of the wig,25 and the application of a white wash to the bust.26 The closest parallel in form to the Amara West bust may be DM 4040, from a rubble deposit at Deir el-Medina, which features a very similar shape, and a painted collar; unfortunately the head is lost (Keith 2011, 168–9). Several other busts, assigned a Deir el-Medina provenance, display both a similar shape and a painted collar27 (Keith 2011, 206–7 [DMC 2]; 230–1 [Prov. 639]). Yet the Amara West bust is rare in several ways. It is unusual in eschewing the tripartite wig that is common at Deir el-Medina. No examples are known with the same wig as represented on the Amara West statue, the tripartite wig or no wig being favoured (Keith 2011, 48–52). Its material, sandstone, is also unusual within the known corpus, which is overwhelmingly of limestone, given the Theban provenance of so many of the busts.28 Furthermore, few are so securely dated. The Amara West bust was set on its pedestal within or after the reign of Ramesses III, on the basis of a scarab (F4561, see N. Spencer 2014b, 471, fig. 8) buried in the floor beneath the pedestal. Of course, the statue may have pre-dated its installation on this pedestal. The Amara West bust also provides new insights into how statuary may have been employed in a domestic setting. While over 190 such statues are known, most provenanced examples (Keith 2011, 7–25) are from Deir el-Medina (77) and Tell el-Amarna (4). For the former, Bruyère states that the majority come from the first room of houses and notes that niches may have been designed to accommodate them (1939, 171). However, only eleven were found in houses (Keith 2011, 11), of which one is tentatively associated with a niche, having been found on the floor beneath a niche. Three examples are attributed to the first room by the front door, and one to the second room of the

24 25

26 27

As is the case with nearly all known examples (Keith 2011, 2). Examples of such busts bearing a dedication to a female individual are known (Keith 2011, 70–4, 85), and depictions of women with red skin are not uncommon in the Ramesside period (see Keith 2011, 86–8; and a painted coffin lid from Amara West: Binder 2017, 598, fig. 7). See Keith 2011, 31. Parallels are known for busts with complex collars (e.g. Trapani 2015, 127 [139]), similar to those depicted on contemporary coffins with contiguous rows of strung beads and amulets. The paint on the Amara West example (see Fig. 23) is insufficiently preserved to ascertain if the collar was embellished with floral

107

Fig. 9: Sandstone anthropoid bust F4182, as found upon pedestal 4273 during excavation. Courtesy of the Trustees of the British Museum (Amara West Research Project).

small houses (Keith 2011, 7–16). Others were found amongst the small temples and chapels at Deir elMedina, in rubble deposits near the houses, and in the grand puits; of course it is likely a proportion of these were discarded from houses. At Tell el-Amarna, faience ‘amulet busts’ are also found, with provenances across

28

elements (see Keith 2011, 57–68), but the schematic collar on an example assigned to Deir el-Medina (Keith 2011, 230–1 [Prov. 639]) might represent a good model for imagining the original appearance of the Amara West statue. The only other example from Nubia was found at Sesebi, Brooklyn Museum of Art 38.545 (Kaiser 1990, pl. 61.4). Made of sandstone, the statue was coated in plaster and brightly painted, with dark red skin, a blue cap or headdress and red body. The head and body were found separately in a private house at Sesebi but there is no doubt that they belong together; [accessed 3 February 2019].

108

N. SPENCER

various areas of the city (Keith and Stevens 2011, 18–21). The majority of busts from other sites have been found in temple or tomb contexts (Keith 2011, 24), perhaps simply reflecting the greater intensity of archaeological investigation at these types of sites. Most recently, an early Dynasty 18 house at Tell Edfu yielded a bust amongst other fittings and objects redolent of domestic cult (Lerner 2019). While the specific positioning within the houses of Deir el-Medina is known for only a small number of busts, none are securely assigned to a room as far back in the house — thus likely to have been more private — as the Amara West example. The statue had been placed upon a pedestal, itself built over a deposit that had accumulated on the first floor of the back room of a newly subdivided house (N. Spencer 2014a, 473–4, fig. 10, pl. 16). As was typical for these back rooms at Amara West, the room was not provided with a hard clay floor, but rather an accumulation of de facto surfaces; other back rooms in houses feature schist slabs (to support vessels?) and vessels buried in the floor. It seems the bust — although it is possible, if unlikely, that one or more other busts were placed on the pedestal before the surviving one — was in use through the laying of the next two floors, the latest floor leaving the bust sitting only 44cm above floor level. Thereafter a layer of brick rubble and other debris accumulated, and the door into the room was bricked up and plastered over (N. Spencer 2014a, 478– 9, pl. 24). Was the rubble part of a roof and/or wall collapse, when the room was deemed surplus to requirements? In any case, those who blocked the door were clearly aware of the bust’s presence. Was there a different family or group now in the house, to whom the bust had no special significance? Should we consider the possibility that the Amara West bust was being ritually enclosed or entombed within the space after its active ritual life had ceased? The Deir el-Medina busts, most from rubble deposits and the grand puits area, may include many discarded by the occupants of houses (Keith 2011, 16–8). The widespread assumptions about a consistency of practice around these statues are thus questionable, and the Amara West example provides important evidence of where and how such statues could be installed and then removed from active service. It was located in a cluttered space: the back room of the house at Amara West is notable for the large number of finds recovered from within (Spencer 2014a, 48–9, pl. 23). This included jewellery, but also tools such as flint blades,

hammerstones and reused sherds. A clay figurine F4553 (Fig. 10) is the only other object evidently associated with household cult in our modern typologies, and raises the intriguing possibility that the bust and figurine could have been used in association; it also reminds us of the range of scales and materials deployed for figurative representations. At Deir el-Medina, busts were found with other objects related to household cult: Ꜣḫ n ỉḳr stelae, an offering table and a libation basin (Keith 2011, 12). Unfortunately, the excavation records from Deir el-Medina are not detailed enough to see if these objects were within the same occupation phase of the house. An ex-voto for Renenutet was found in the same room as bust DM 4009 (15.1.35 Sj, house SO.III; Bruyère 1934–35, 323 no. 6), and busts also appear in association with Taweret at Deir elMedina and nearby (Keith 2011, 15). In Mit Rahina, a limestone bust was found within a layer of debris, alongside a cobra and three female figures (Keith 2011, 291–2).

Fig. 10: Clay figurines found in house E13.3-N (F4245, above) and E13.3–S (F4553, below). Drawings: Alice Salvador. Courtesy of the Trustees of the British Museum (Amara West Research Project).

IN TEMPLE AND HOME: STATUARY IN THE TOWN OF AMARA WEST, UPPER NUBIA

109

House E13.3-S at Amara West was small and thus multi-purpose use of different spaces was likely. While the back room might have housed a domestic cult image, it may also have been used for storage and the gradual build-up of rubbish and discarded items. A hearth in the middle room — located in front of the room with the bust — suggests a focal point to the dwelling, whereas the front room underwent repeated refurbishment, whitewashing of the walls and the relocation of its grinding emplacement and basin. We must imagine inhabitants, when engaging with the anthropoid bust or other figural representations, stepping around and through other activities and their debris. We are far from the simple uncluttered spaces imagined in many reconstructions of domestic cult settings. Who did this statue represent? It was not inscribed, as is the case with nearly all surviving examples (Keith 2011) and any painted text has not been preserved.29 The statue is mute in terms of ownership, and this of course allowed flexibility of use, whether within the same household and/or if there was a change of occupancy. I have here followed Keith (2011, 2–4) in not using the term ‘ancestor bust’. The commemoration and interaction with ancestors — family or otherwise — was only one function such busts may have fulfilled. The busts can also represent the deceased in other ways, whether associated with a deity — Renenutet, Taweret, Hathor — and/or to perform protective functions (Keith 2011, 89–104). The repainting of busts suggests their use over an extended period (Keith 2011, 32–3), and perhaps reinterpretation as representing other individuals. The repeated repainting of a niche above the mastaba of a house (E13.7) at Amara West (Spencer, Stevens and Binder 2014, 27–8) provides an example of an architectural setting being refurbished and possibly reimagined. The adjacent house (E13.3-N) was created at the same time as that with the bust, but had an additional room. Here, no bust was found in the back room, nor a pedestal to support one, but rather a niche (4287) in the back wall (4167). Approximately 70cm wide and 70cm deep, and at least 90cm tall, the niche may simply have

been a storage cupboard, but a survey of other houses reveals at least four with niches, often decorated, with some consistency in location. The back room of house E13.3-N, again, yielded a clay plaque-like figurine (F4245, see Fig. 10), amongst an assemblage of finds not dissimilar from the back room next door: quernstone fragments and other stone tools, clay lids or stoppers, a bone point and a small number of beads.

29

30

A small area of red paint was noted on the underside: was the underside painted, is this a splash of paint, or could it be part of an inscription naming the donor or the individual represented (see Keith 2011, 204–6 [DMC 1])?

Statues, figurines and settings in other houses The most elaborate setting created within an Amara West house — which may have held a stela or small statue — was the decorated niche above the mastaba in the main room of house E13.7 (N. Spencer 2014b, 50–1, fig. 14b). Reconstructed from diagnostic painted and moulded mud fragments found tumbled on the floor in front of the wall within which it was set, the niche took the form of a cavity framed by a projecting architrave, later transformed into a cavetto cornice and torus mouldings, brightly painted with white gypsum, red and yellow ochre, Egyptian blue and bitumen black. Given that the wall within which the framed niche was placed was only one brick length thick, the niche must have been no more than 35cm deep, with surviving fragments indicating the cavity was at least 45cm wide and 50cm tall. The repeated repainting and refurbishment of the cavity, including its use of architectural idioms drawn from temple architecture, supports its identification as a setting for a small cult focus, perhaps a stela or statue; the latter was perhaps more likely.30 Niches of various sizes are also common in Amarna houses, some with apparent cultic function (Stevens 2006, 237–48). In two Amara West houses, cuboid cavities were cut into the rear of the house, into the town wall. One of these we might describe as an inner room (E.12.1), though it is without mastaba. It has two storage/back rooms off it to the north, and is the most formally laid out space in this modest house, being provided with a vaulted roof and whitewashed walls. The cavity cut into the town wall (P. Spencer 1997, 175, pl. 111b) can reasonably be associated with domestic ritual, given its

A late Dynasty 13 or early Dynasty 18 shrine at Askut (Smith 1990, 102, pls 15–8) featured a niche for a small stela or statue.

110

N. SPENCER

moulded cavetto cornice painted red, the black torus moulding and then further red pigment across the body of the shrine, with thick bands of yellow and black beneath, running onto the adjacent wall. This was a feature that was meant to be the centre of attention. The cavity was only 28cm in height, so suitable for a small statue or stela. The second wall niche, in the east town (D14.13a: P. Spencer 1997, 154, pls 78, 100c) had an applied clay roll above the door and a schist lintel, with a cavity of around 125cm square. While this is the back room, it is a small two-room dwelling (at least in terms of its ground floor plan), so the room may have been a focal point of the house, more akin to a mastaba room. Such niches would have been suitable places for the domestic display of statues or stelae. The latter type of object has been found in houses at Amara West, yet very few compared to the rich assemblage deposited in the temple (see P. Spencer 2016). One was found reused as a cover for a buried storage vessel, in the back room of house E13.9 (F4096: see N. Spencer 2009, 51, pl. 6). The depiction of Amun-Ra as a ram might suggest these were originally intended for the temple, given finds in both the Amara West and Dokki Gel temples (see above). Otherwise, two stelae may originally have been intended for houses. The EES discovered ‘in [room] D.14.7, on a late floor near stairs leading to a high level later room… a small votive stela showing Ramesses II offering wine to Satis’ (EES obj. 360; British Museum EA 68675; P. Spencer 1997, 157). The summary style of the sunk relief on this piece is reminiscent of stela F5808 (= Sudan National Museum SNM 35823) found in occupation deposits beneath the floor of house E13.6 (room 1). The clay figurines mentioned above have been found in significant numbers at Amara West, falling into two distinct traditions: one schematised into plaque form, the other depicting a pronounced representation of the female anthropomorphic form (Stevens 2017). The latter might represent a more local tradition, the former being more consistent with the material culture at contemporaneous Egyptian sites. Plaque-form figures are found in Dynasty 18 contexts at Tombos (Smith and Buzon 2018, 215–6, fig. 12) and Sai (Doyen 2015; Budka 2017a, 158–9, fig. 85), with Doyen positing that some of the incised decoration could reflect Nubian traditions (Doyen 2015, 146). That they could be used for display, albeit in a modest sense, is suggested by an example with a defined base (F4471, see Stevens 2017, 417, fig. 2). One might question whether the anthropomorphic forms, perhaps following a more localised

tradition, were used rather differently, as none of these would have stood unsupported. The temporal and spatial distribution of these figurines at Amara West, a site with systematically excavated housing zones and a temple, is suggestive of a distinction between the two forms (Stevens 2017, 413–6, fig. 6). These cluster in the three excavated residential areas, and a rubbish mound, but are almost entirely absent from the temple; the anthropomorphic forms seem weighted towards later deposits. Stevens cites examples of figurines found in deposits seemingly associated with the occupation phase of the houses in area E13, noting instances in the front rooms of houses and small back rooms (including that with the anthropoid bust), but also one from an outside space, perhaps swept out of a nearby house. Some of these figures have been lightly fired, whether intentionally or not (Stevens 2017, 412), but a very unusual ceramic statue represented a different tradition, more loyal to the reality of the human form as shown in larger wooden and stone statuary. Described as a ‘baked clay figurine’ (see Fig. 7, EES obj. 945; P. Spencer 1997, 202), it was found in room E.13.2.Y(d), part of a warren of small rooms in the northeastern corner of the Deputy’s Residence (P. Spencer 1997, 194–200, pls 114–5), perhaps best interpreted as small dwellings built into a courtyard space from an earlier phase. The figure is known only from a small sketch on the object card, the scale indicating it is around 22cm in height, and depicts a standing figure with kilt, perhaps a bald head, and arms held at its sides. It is stylistically unique at Amara West. Two other objects, found in the back room of the small house E13.8, initially constructed in midDynasty 19 (site phase II), relate to statues. These copper alloy objects were shaped as snakes, featuring the cobra head, the flaring front incised with detail, and then a tail curving to allow fitting to a statue head. One (F5633) was found in a mixed surface deposit, so might not relate to the house occupation, while the better-preserved one (F5693) was found in an occupation layer beneath. It seems these would have formed parts of composite statues, perhaps of wood, metal or other material, presumably of a king or god. While these might not relate directly to the house, it is not impossible that small statues were being manufactured in dwellings. House E13.8 is also directly adjacent to the remains of a metalworking area (E13.17), so it is possible that these are discards, maybe connected with the foundation of the temple. Finally, the drilling of

IN TEMPLE AND HOME: STATUARY IN THE TOWN OF AMARA WEST, UPPER NUBIA

111

small holes into the edges of doorjambs, attested on a jamb from villa E12.10, was presumably done in order to suspend something, perhaps small figurines or amulets.31 Funerary contexts Two burial grounds were used to inter the dead at Amara West: the elite cemetery C upon the desert escarpment overlooking the town from the north, and nearby cemetery D set on a low alluvial terrace adjacent to a wadi (Binder 2017). Surface erosion has removed evidence for the tomb superstructures in many cases, but both pyramids and tumuli are preserved. The pyramids were each fronted by a small mud-brick chapel, rectangular in plan. Such tomb monuments would have provided suitable settings for commemorative statues, perhaps in a niche at the rear of the chapel or in settings on the face of the pyramid. No evidence for such emplacements, or of any statues, has been encountered in the cemetery. The tombs were furnished with painted coffins, offering vessels and an array of other funerary objects, including jewellery, scarabs and amulets, and ‘cosmetic’ items (mirrors, razors, tweezers, wooden boxes). Shabtis were rarely placed in tombs, with eighteen fragments recovered from three pyramid tombs in cemetery D (G301, G320, G322). The shaft fill of tomb G320 yielded several shabtis of the Deputy of Kush Paser (temp. Ramesses III), an individual attested in the town (Fig. 11)32 and one of a singer of Amun Tyia.33 Other shabti fragments found in the same tomb, but without the owner’s name preserved, likely belong to the set of Paser, on the basis of scale and material.34 These are typical Ramesside shabtis just under 11cm in height, with black painted detail: a column of hieroglyphs, hands holding hoes, basket over the shoulder, a broad wesekh-collar and tripartite wig. The glaze ranges from a very pale green to a very pale blue, though often preserved only as white. Nearby pyramid tomb G322 yielded a shabti fragment similar in style and manufacture to those of Paser, with the name Iy-bay (F8966), presumably also a high official at

Amara West. While it is possible further shabtis of wood were placed in the tombs (as suggested at Tombos, Smith 2018, 81), no fragments have been found, in contrast to the large number of diagnostic funerary bed fragments. Further west of the three large pyramid tombs, though still in cemetery D, lies G301, a more modest pyramid and chapel above two chambers. The only find within the disturbed fill of the shaft was a single ceramic shabti (F8004 [SNM 34559], see Fig. 6). This quite striking sculpture is 16.6cm high, with body in Osiride form, and arms crossed over the chest; there is no tripartite wig. No other tools or implements can be discerned. The upper part of the body and head has been covered in red paint. This unusual shabti reflects a tradition found at other New Kingdom sites, for example at Gurob (Manchester Museum 1432.b: Janes 2012, 140–1 [69]) and Rifeh (Petrie 1907, 22, pl. 27C), all most probably locally produced. Other figural forms were not generally part of tomb assemblages — clay figurines were not taken to the grave (Stevens 2017, 418) — though tomb G216 is

31

34

32 33

A practice well-attested in temples (Perdu 2003). F8465 (SNM 36980), F8543 (SNM 36967), F8546 (SNM 36969) and F8548 (SNM 36971). F8545 (SNM 36968).

Fig. 11: Faience shabtis of Paser (F8543 [SNM 36967], F8548 [SNM 36971], F8577 [SNM 38142]) from shaft of tomb G320. Courtesy of the Trustees of the British Museum (Amara West Research Project).

F8547 (SNM 36970), F8549 (SNM 36972), F8550 (SNM 36973), F8551 (SNM 36974), F8552 (SNM 36975), F8576 (SNM 38144), F8577 (SNM 38142), F8591 (SNM 38146), F8593 and F8980 (SNM 38145).

112

N. SPENCER

a notable exception. In the northern part of the lower wadi cemetery, comprising a shaft with an offset burial niche (Binder 2011, 46, pls 12–3), it was used for the burial of at least six individuals, of which three were inserted into the same painted coffin (F9457-9458). Three amulets were found in the tomb (Fig. 12). However, as the coffin and bodies had been disturbed by looting, it is not possible to confidently associate these amulets to specific burials. The first amulet is a dark blue faience figure of Isis (F9466 [SNM 34572]), seated on a chair (with incised hatched decoration) and nursing the child Horus; she wears a modius, sun-disc and cow horns. A pale blue-green Pataikos (F9467 [SNM 34573]), with typical bandy legs, is busy with details, with dorsal pillar, scarab above head, holding two knives and perhaps devouring snakes. The dorsal pillar is incised with pseudo-hieroglyphs. The third amulet is of a notably different quality, a masterpiece in miniature sculpture (F9459 [SNM 34570]). Carved from a pale beige ivory, the figure squats slightly on a thin base, with legs flexed, hands resting on distended belly and a head that in its schematic rendering is suggestive of a mask. Two holes have been drilled through the chest, providing a secondary means of attachment:

perhaps it was sewn to a textile or garment? The hole for suspension was carved as part of the original figure, behind the top of the head. Social context, dating and production Beyond the dedication of a statue by the scribe of gold counting Amenemhat (Cat. 2, see Fig. 17) — the only inscribed statue found at Amara West — who else had the means, or permission, to dedicate monuments in the temple? Other than the king, this is likely to have fallen principally to the viceroy of Kush, the resident Deputy of Kush, and a small number of senior officials. The temple at Amara West preserved several other forms of commemoration on behalf of individuals: stelae (some set into the walls, or on stands set on the floor; see P. Spencer 2016, pls 201b, 202–6), hieratic graffiti (Spencer 1997, 40), and small sunk-relief scenes added to, or below, the temple decoration (see checklist in Auenmüller 2018). The latter typically show an individual in a pose of pious adoration (e.g. P. Spencer 2016, pl. 95b); none of these are found in the preserved parts of the sanctuary, but rather the outer parts of the temple. The individuals dedicating these monuments

Fig. 12: Figurative amulets from tomb G216: ivory Bes (three views) F9459 (SNM 34570), faience nursing Isis F9466 (SNM 34572), faience Pataikos F9467 (SNM 34573). Courtesy of the Trustees of the British Museum (Amara West Research Project).

IN TEMPLE AND HOME: STATUARY IN THE TOWN OF AMARA WEST, UPPER NUBIA

113

include viceroys and the resident Deputy of Kush, but also the ‘head of the gateway of Amun’ Nebdjefau (P. Spencer 2016, pl. 205b) and the Second Prophet of Amun Hori (P. Spencer 2016, pl. 205d). In the houses, stelae were more common, alongside the decorated doorways.35 Those holding the titles of ‘scribe of reckoning gold’ and ‘scribes counting all that comes from southern regions’ (Müller 2013, 271) were important actors in the state apparatus that sought to extract, process and ship gold to Egypt (Vercoutter 1958, 142–9; Müller 2013, 75–9). A Ramesside stela from Sai is inscribed with the name of a ‘scribe of reckoning gold and overseer of the gold countries of Kush’ and references the king’s sons (viceroys) Setau and Paser (Vercoutter 1958, 156–7 and pl. 45b); holders of this title were found buried in Tombos and Aniba. At the former, funerary cones attest to the burial in a pyramid tomb of the scribe of reckoning gold, Siamun (Tombos 2016). The Aniba cemetery was the location for the burial of Messwy (Steindorff 1937, 21, pl. 7 [6]). These officials are also attested in the gold-mining regions themselves. A scribe of the treasury who reckons gold, Khaemwaset, in year 49 of Ramesses II (Černý 1947, 55 [27]), and Anupemheb, another Ramesside scribe with this title (Černý 1947, 55 [30]), left their names in the Wadi Allaqi. Thus the Amara West block statue is most likely an attestation of an elite individual stopping at the town for a period of time as part of his inspection duties, dedicating a fine statue brought with him from Egypt. It would not have been an arduous task to ship this 35cm-tall statue, sculpted from stone extracted in quartzite quarries around Aswan or the Cairo area (Klemm and Klemm 2008, 215–31), upstream from those places. Did this reflect a preconceived plan on the part of its owner? Statuary left by other elite individuals at sites such as Sai may also reflect the temporary presence of these persons at the colonial towns (see Auenmüller 2018, 246–7). The other Amara West statuary (and stelae) relies on local, rather poor-quality, sandstone. It is somewhat less white and powdery than that of the temple and West Gate at the site, and might have been quarried at

Sai, as with a finely carved door lintel installed in a house (N. Spencer 2014b, 48–50, figs 9–10). The statues, though varying in accomplishment, all follow the contemporaneous style and canons of sculpture in the round within Egypt. Were they made by itinerant craftsmen? Smith (2014) has argued for this model as regards potters in Nubia, and it might be that, with such low demand, Amara West had no call for dedicated sculptors. At certain periods, particularly when the town was founded, or during significant decorative programmes in the temple, specialist craftsmen may have been present for carving and decorating the temple and its fixtures and fittings. Those individuals may then have moved on, taking their skills with them. It is relevant here that other material remains one might associate with temple decoration — the processing of pigments, and the creation or repair of metal tools — cluster in the early phases of the town’s history. Statues seem to have been rare at Amara West, and the tactile, visual and auditory experience of these images being produced was probably very limited. The stelae in the temple, which do seem to span the site’s history, are almost entirely very portable objects, and some may have been brought by visiting craftsmen and officials. In seeking physical vectors for ritual expression and commemoration, especially in the houses, inhabitants turned to the abundant and malleable clay. Both fired and unfired clay figurines and statues were produced. The clay figurines (Fig. 3), in particular, remind us of the changing, dynamic and unpredictable nature of the town. Stevens argues persuasively for these being community products, lacking a consistency of style that one would expect if certain individuals were making most of them (2017, 417). A shift in tradition, with more prevalence of anthropomorphic female figurines, can be posited; something not possible with the larger statuary, with only the anthropoid bust from a reliable stratigraphic setting. Faience figural amulets and shabtis, as found in the tombs, may have been made locally, though evidence for the production of glazed composition materials has yet to be identified at Amara West. Finally, the masterfully carved ivory Bes statue could well have been made locally, given the ivory- and bone-working attested at Amara West.36

35

36

The majority of stone doorway elements encountered in excavations were reused or uninscribed.

Manuela Lehmann is currently studying this material, which includes furniture fittings and inlays, jewellery and gaming boards.

114

N. SPENCER

Abandonment: systematic clearance of statuary? Does the absence of statuary reflect an ancient, systematic, clearance of the temple of objects perceived to be valuable or important? Patricia Spencer (2016, 38) assumed as much: the almost complete absence of statuary — only one headless statue of an official (object 69 of the Scribe, Amenemhat) was discovered — indicates that the clearance of the temple was undertaken systematically, with little left behind to be buried or re-used.

The private statuary (Cats 2–3) may already have been broken and deemed dispensable if specific ancestral cults had ceased. The ceramic head (Cat. 4) may also have been broken, but why would a complete ram image (Cat. 1) not be taken? Such a model is, to my mind, overly predicated on the assumption that the town, and temple, were abandoned by Egyptian residents desiring to return to Egypt as the Pharaonic state lost political and military control of the region. New fieldwork at Amara West has illustrated how other factors, notably deteriorating climatic conditions, are a more likely cause of an abandonment process that was probably gradual (Woodward et al. 2017). That burials continued in the cemetery (Binder 2011) strongly suggests the population had not moved far, perhaps to the opposite bank of the Nile, which would remain the principal settlement area from the 1st millennium BC through to the present day. Further upstream, the temple of Amenhotep III at Soleb was the source of a number of statues moved to Gebel Barkal, to furnish the Meroitic temples. Could such a phenomenon have occurred at Amara West? No evidence has been recorded of Pharaonic statues moved to Amara East, for example, though this temple was almost completely destroyed in the mid-19th century AD. Various taphonomic processes evidently greatly reduce the amount of statuary preserved in a given archaeological environment. At Amara West, as elsewhere, metal is likely to have been a precious commodity, and perhaps prone to reworking. That some removal of architectural stone occurred is suggested by the discovery in villa E12.10, a late New Kingdom extramural house, where sandstone doorjambs and lintels had been stacked in the large courtyard (see Spencer 2009, fig. 4, pls 7–8), as if in preparation for removal. I would interpret such an arrangement as reflecting the value of large sandstone blocks for reuse elsewhere, rather than the removal of elements back to Egypt because they

were significant from an ideological or religious perspective (none of the stacked architectural elements were inscribed). Another piece of evidence that calls into question the notion of systematic removal is the large number of stelae left in the temple. Several are royal commissions, including the Dream Stela and Marriage Stela set up in front of the temple gate (P. Spencer 2016, pls 6c–9), but also fine, large and well-preserved stelae of Ramesses II (P. Spencer 2016, pls 200a–b, 201c). If a systematic clearance of important sculpture was undertaken, surely such monuments would also have been included. The stela of the viceroy Usersatet found at Amara West (P. Spencer 2016, pl. 201b), carved before the foundation of the site, was evidently considered important enough to move, perhaps in the early Ramesside period, from its previous location. As such, I propose that the number of statues ever set up at Amara West is surprisingly low, or at least lower than we assume on the basis of the major sites in Egypt, specifically Thebes, Memphis and Elephantine. None of the sculpture approaches life-size, let alone colossal representations, though the temple architecture, and indeed the royal stelae set up in the forecourt, are of considerable scale (P. Spencer 1997, pl. 25). Despite this scale, the temple (see N. Spencer 2017) did not include the processional avenues, grand pylons, or architectural facades we are familiar with from Luxor and elsewhere. This would have reduced the number of settings in which statues might be installed. The West Gate would once have been an imposing monument, principally within its brightly painted passageway, yet the exterior façade was somewhat modest, inscribed with a column of royal titulary on either side of the door. As the western suburb developed, the gateway approach became more convoluted, requiring stepping down into a small courtyard before reaching the entrance. Statuary in Nubia under Pharaonic rule (c. 1500–1070 BC) Exploring statuary at other New Kingdom sites in Nubia is informative, yet frustrating. Many of the statues do not come from controlled excavations, and there has been relatively little excavation in housing areas beyond Amara West: the houses at Sesebi have never been published (Blackman 1937, 149–51, pl. 19; Fairman 1938, 152, pl. 8) leaving two zones at Sai (Azim 1975; Doyen 2017). None, to my knowledge, yielded any evidence for stone sculpture, though plaque-shaped

IN TEMPLE AND HOME: STATUARY IN THE TOWN OF AMARA WEST, UPPER NUBIA

clay figurines were found in the northern area at Sai (Doyen 2015). Rather, it is the great temple complexes that have been excavated; from these sacred contexts a number of statues have been recovered. Statuary formed an important part of the royal building programme from the foundation at Sai, with the near-life-size sed-festival statues of Ahmose II and Amenhotep I, though it remains debated whether these were posthumous, and in what type of chapel they were installed (L. Gabolde 2012, 117–20, figs 1–2). Thereafter, the early and mid-Dynasty 18 witnessed the provision of other royal statuary, including a statue of a god’s wife most recently identified with Merytamon, a queen of Amenhotep I (L. Gabolde 2012, 126– 8, figs 11a–d). A number of non-royal statues have also been found at Sai, including those of the viceroy Nehy (temp. Thutmose III) and a Dynasty 18 mayor (Auenmüller 2018, 244–5, table 2). The smashed-up assemblage of statues, several of which name Usersatet, the viceroy of Amenhotep II (Davies 2017b)37 is striking for both the number and range of statue forms and materials, though no block statues or standing statues are included. The majority of the finds from this cache are in granodiorite, a stone not available near Sai. They comprise seated figures, a stelaphorous statue, kneeling images holding a vessel or basin and possibly a cultobject, and a ‘green granitic’ stela. The group also includes a ‘greywacke (?)’ kneeling figure holding a votive object. If this stone identification is correct, the monument must have been quarried in the Wadi Hammamat: another example of an official bringing a monument from Egypt for dedication at a temple in Upper Nubia. Finally, several statue heads and part of an offering table make up a group of sandstone monuments, perhaps sourced locally. Davies (2017b, 145) proposes that the granodiorite examples, which are larger and very fine, are the products of a royal workshop (not on Sai), perhaps part of a coherent group, most likely set up in the mid-Dynasty 18 temple in the southern part of the town. The provisioning of a Pharaonic town with figural monuments is also suggested by a group of mid-Dynasty 18 stone shabtis at Sai (Minault-Gout 2012; Budka 2017b, 75–7, pl. 5) and Tombos (Smith 2018, 80–1).

37

On statue assemblages of individuals, see Bernhauer 2010, 25, 154–7. Note that the viceroy Usersatet is known from Amara West, as a fine monumental stela was moved to the temple (P. Spencer 2016, 10, pl. 201b) at some point after the town was

115

Sesebi, one of the other major early Dynasty 18 centres (Spence 2017), was furnished with stone plinths inside the West Gate, perhaps for statues or offering tables (Fairman 1938, 152). A black granite royal Thutmoside head (Stockholm E.1449; Blackman 1937, 147, pl. 16 [1]), found in the debris excavated from the temple area, attests to royal statuary provision at the site. A sandstone statuette from a house cellar (Fairman 1938, 152) may be the once brightly coloured, anthropoid bust (Kaiser 1990, pl. 61.4).38 This would be an interesting locus for such a statue, if placed there intentionally, given the evidence for the Amara West bust being left ‘entombed’ in a blocked-off room. Smaller figurative objects from Sesebi included steatite rams and a kohl-pot held by a monkey (Blackman 1937, 147 pl. 18.3). The major refurbishment and rebuilding of the site under Akhenaten has not left any statuary, though reliefs in the temple crypt provide evidence for the arrival of Amarna-style art (Spence et al. 2011, 36). We do not know if this crypt was intended to house sculpture, or indeed if Amarna-style statuary was set up at the site. Further upstream, amidst the heartland of the defeated Kerma state, a new Pharaonic town and sacred complex was built at Dokki Gel, with a trio of temples extended and refurbished throughout the New Kingdom (Bonnet and Valbelle 2018). While no statues were found in situ, the excavators posited a small sanctuary in one temple, on the basis of a brick pedestal or setting in the middle of the small room, and a drain leading away from it (Bonnet and Valbelle 2018, 41–3, fig. 25). This was also furnished with royal and private statuary in imported stone: a greywacke Ramesside statue (Valbelle 2001, 233–4, fig. 7; Bonnet and Valbelle 2018, 276–7); a grey granite head attributed to Thutmose IV, found within a cache of Napatan statuary (Valbelle 2003, 298–9, fig. 9; Bonnet and Valbelle 2018, 122, fig. 98); and statues of a vizier, an overseer of cavalry (Valbelle 2003, 294–5, fig. 5; 299, fig. 10) and an overseer of southern lands (Valbelle 2005, 252). Small votive stelae are also a part of the private donations to the temple (Bonnet and Valbelle 2003, 290–2, figs 3–5); the ram statue discussed above may have been a royal or private donation. Private stelae in

38

founded. This may have been commissioned for the temple at Sai. [accessed 3 February 2019].

116

N. SPENCER

greywacke (Valbelle 2009, 112–4, figs 6–7) and limestone (Valbelle 2009, 109, fig. 5) indicate that monuments made in Egypt were also being transferred to the Kerma area. Downstream from Kerma at Tombos, no evidence of statuary — other than shabtis — survives from the New Kingdom and Napatan tombs. Figural representation was present, however, in the ceramic and painted wooden coffins, and ceramic canopic jar lids (Smith 2018, 79–82, pl. 11). It is worth emphasising that the housing and temple areas of the town are difficult to access, lying beneath the modern town. The most upstream major Pharaonic centre was Gebel Barkal, which witnessed a series of temple construction programmes throughout Dynasty 18 and early Dynasty 19. Royal and private statuary, in a range of stone types (Kendall 2017), was set up in the temple. It is likely that some of these statues had been moved from Soleb. These include the lower part of a black granite statue of Thutmose III (MFA 23.737 and fragments, see Kendall et al. 2017, 164 n. 12), a granite base from a lion/sphinx with cartouche of Amenhotep II (in debris S of B502) and a greywacke (‘slate’) statue fragment inscribed for Thutmose IV. Statues of officials include the viceroy Merymore (black granite, from B500), an unnamed individual, a diorite Djehutymose (reign of Akhenaten, buried in Napatan cache), an overseer of southern land Hekaemsa-sen (B700), and that of Bakenwewel with the cartouche of Ramesses IX. A gneiss block statue found at el-Ghazali monastery, perhaps originally set up at Gebel Barkal, may be of New Kingdom date.39 Moving back downstream to Soleb, we see the last great programme of statuary provision in Upper Nubia, at the grandiose temple named Kha-em-Maat, constructed in the reign of Amenhotep III and dedicated to Amun-Ra but also Nebmaatra-Lord-of-Nubia. As with this king’s temples at Thebes, the temple was richly furnished with statuary known from fragments (Schiff Giorgini 1998, pls 331–4; Schiff Giorgini 2002, 415–28; Schiff Giorgini 2003, 249–64), and also referred to in the temple decoration (the ‘living statue upon earth’, ẖnty ꜥnḫ tp tꜢ, cited above). These include royal statues in quarzitic diorite and sandstone,

39

Ghazali F. No. 384; to be published by Jalina Tschernig and Laura Haupt. I am grateful to Johannes Auenmüller for drawing my attention to this statue.

a sandstone statue of Anuket, granodiorite ram-sphinxes from the processional approach to the temple, two large falcon statues in grey granite, an alabaster sphinx and fragments in granodiorite from other statues. Further statuary, moved to Gebel Barkal in Meroitic times, includes a hawk-statue, Sopdu and a uraeus (all in black granite), and a ‘green slate’ Amenhotep III. A dyad of Amenhotep III, in granodiorite and first seen by F. W. Green north of the Dal Cataract, seems also to come from Soleb, on the basis of the inscription it bears (Davies 2014). It is notable that, in contrast to Amara West, the layout of the temple (Schiff Giorgini 2002) provided suitable spaces in which to display monumental and processional statuary, and it continued to receive attention throughout the reign in terms of extension and re-envisioning of the temple’s layout and decorative programme (Bryan 1992, 106–10). The reign of Amenhotep III is remarkable, of course, both for the quantity and innovative types of statuary commissioned as part of temple construction programmes, notably at Thebes. The range of statuary, and their material, attested at Sai, Soleb and Gebel Barkal all suggest the provision of statuary from workshops in Egypt, as part of a sustained programme of investment in the temples. It is notable that this investment is largely restricted to Dynasty 18, and to the three major temple sites of that era. Dokki Gel/Kerma, Soleb and Gebel Barkal provided an appropriate architectural arena for the display of statuary, against pylons, around peristyle courts and especially along processional avenues. Sai, with its modest temple (Azim and Carlotti 2012), may have offered less opportunity for display, though of course we remain ignorant of where the Ahmose and Kamose sed-festival statues were installed (L. Gabolde 2012, 117–20). No evidence for statues has been found at Sedeinga. In Dynasty 19, attendant with the change in appearance of towns from large enclosures to more confined ones (Aksha, Amara West), a reduction in the space available for royal statuary becomes apparent, and there is also less imported stone apparent at these sites. Stelae remain important means of engaging with the divine, with both royal and private examples continuing at Amara West through Dynasties 19 and 20: twenty-seven stelae are known from the site (see P. Spencer 2016). The viceroy Setau commissioned a group of seven stelae for the temple at Wadi esSebua, with careful consideration of their semi-public

IN TEMPLE AND HOME: STATUARY IN THE TOWN OF AMARA WEST, UPPER NUBIA

setting and role in processional rituals and festivals (Frood 2016, 84). Dynasty 19 and the early part of the following dynasty are times of considerable statue production in Egypt, both royal and private, and complemented by the reinscribing of earlier statues, yet this does not seem to have been the case in Upper Nubia. Buhen echoes this picture, with an abrupt fall-off in statue dedication after Dynasty 18. Both hard- and softstone private statues have been found in the temple, such as that of the son of the prince of Teh-khet Amenemhat (Randall-MacIver and Woolley 1911, 108–9, pl. 36; 110, pl. 37), a granite statue of the scribe Ahmose (Randall-MacIver and Woolley 1911, 111, pl. 37), and a granite statue of the mayor Kamose (British Museum EA 1022; Edwards 1939, 3–4, pl. 4). But despite a large number of Ramesside stelae and inscribed doorways being found at Buhen (Emery, Smith and Millard 1979, 94–156), the only statue of this period is that of the viceroy Setau (KRI III: 108 [55]). Further north, the situation is different: from Abu Simbel northwards, statuary remained an important part of the appearance of the sequence of temples erected by Ramesses II. The massive engaged colossal statues of the two temples at Abu Simbel were complemented by processional statuary (PM VII, 108). As at Gerf Hussein (PM VII, 34) and Wadi es-Sebua (PM VII, 53–7), the ram/falcon-sphinxes, sphinxes and engaged statuary (Hein 1991, 9–11) were carved from local sandstone, in contrast to the granite deployed by Amenhotep III at Soleb. These temples also had statues in the cult niches, carved from the living rock. Engaged statues are also found at Gebel Dosha in Upper Nubia, representing Thutmose III, Amun-Ra and Satet (Davies 2016, 25–6, pl. 19). Private statuary continues to be donated to these Lower Nubian temples, notably a ram theophorous statue of the viceroy Paser at Abu Simbel (British Museum EA 1376; Bierbrier 1982, pls 46–7) or statues of Setau at Wadi es-Sebua (Hein 1991, 10, 19). This stark dissonance between the northern and southern sites in the Ramesside period poses important questions. To what extent does the increasingly heterogeneous nature of Amara West as a town — which through architecture and material culture can be demonstrated to develop away from the formal planned town at first envisaged (N. Spencer 2017) — prompt a reduced reliance on statuary for commemoration, display, ritual and indeed cultural expression? Was that also the case elsewhere in Ramesside Upper Nubia?

117

Conclusion Statuary was present within Pharaonic foundations in Nubia from earlier periods of colonial control, for example in the Middle Kingdom forts, and more significantly with the looting and relocation of sculpture to Kerma, deposited both in the grandiose tumulus tombs and also in the temple area at Dokki Gel. In the latter case, the context and function of reuse for the five Middle Kingdom statue fragments remains unclear (Valbelle 2011; Bonnet and Valbelle 2018, 39): they may originally have been placed in the Kerma cemetery. Middle Kingdom sculpture was also in circulation at Sai (Thill 2012) and Buhen (Emery, Smith and Millard 1979, 149–51, n. 28). However, sculpture never became a core element of elite Kerma culture, to our knowledge, in contrast to the Napatan and Meroitic eras. The reintroduction of life-size statues to this landscape, as part of the Pharaonic conquest and ensuing town-building programmes, may have been one of the most charged manifestations of Pharaonic power, alongside the stone temples themselves and their brightly coloured graphic decoration. Intriguingly, that reintroduction may have provided the inspiration for the royal imagery of the Napatan and Meroitic eras. The evidence from New Kingdom Nubia suggests peaks in statuary dedication in the early and mid-part of Dynasty 18, further considerable dedication of statues with the grandiose building projects of Amenhotep III at Soleb, and then a noticeable fall-off in Upper Nubia. The founding of Amara West dovetails with this fall-off in the early part of Dynasty 19. There is no lack of state investment at Amara West, given that a large stone temple was built and decorated, but the focus was less on display and grandiose architecture than in creating a compact town. Royal stelae were set up either side of the temple entrance, but there is no evidence for a statue programme of significant scale, and dedications were perhaps only possible when skilled artists passed through or were briefly stationed in the town (perhaps mostly when temples were being decorated or refurbished). I have illustrated elsewhere how the character of the town changed markedly from its early form, with an increasing number of houses and a more heterogeneous layout likely to have been shaped by individual/household agency (N. Spencer 2014a; 2015; 2017). Perhaps that dynamic also had a bearing on the provision of statuary, as the town became less shaped by the Pharaonic state’s intentions?

118

N. SPENCER

Acknowledgements This paper draws extensively on both objects from the site and the experience of fieldwork within the Amara West Research Project. The Project was conducted under the auspices of the National Corporation for Antiquities & Museums (Sudan), who also provided access to the storerooms of the Sudan National Museum. I would like to thank Dr Abdelrahman Ali Mohamed, Director General, archaeologist Mohamed Saad and curators Ikhlas Abdel Latif and Shadia Abdu Rabo. The anthropoid bust was excavated by Réné van Kertesz; further thanks are due to Elisabeth Greifenstein and Claire Thorne for their recording of the statue, and Susie Green for her 3D model. For access to the Egypt Exploration Society archives, I would like to thank Cédric Gobeil and his predecessor as Director, Chris Naunton; Carl Graves (Deputy Director) and Patricia Spencer (former Secretary) also deserve thanks. I would also like to thank Marilina Betrò (University of Pisa mission to TT14), Yekaterina Barbasha (Brooklyn Museum of Art). Discussions with Anna Stevens, Marie Vandenbeusch, Johannes Auenmüller, Mathew Dalton, Marcel Marée, John Taylor, Vivian Davies and Angela Tooley helped shape my thoughts on these statues. At the British Museum, further thanks are due to Michela Spataro (Department of Scientific Research), Manuela Lehmann (Amara West Project), Daniel Pett, Loretta Hogan (Department of Conservation) and photographer Kevin Lovelock.

Catalogue of statues 1. Ram statue (Figs 13–6) Sudan National Museum 3068, excavation object 121. Material: Sandstone, plaster, paint. Dimensions: height 43cm; width 17cm; depth 16.7cm. Provenance: From ‘the eastern room of the Sanctuary group came a small and rough statue, the head and shoulders of a ram’ (Fairman, diary AW II.1, 17–8).40 As discussed by Patricia Spencer (1997, 45 n. 154), this likely refers to the left sanctuary (as one approaches the back of the temple), though the object card refers to the

40

These diaries are house in the Lucy Gura Archive of the Egypt Exploration Society.

Fig. 13: Sandstone statue of a ram. Sudan National Museum 3068. Photograph: Neal Spencer. Courtesy of the National Corporation of Antiquities & Museums.

‘(local) southern’ room, i.e. that on the other side. The excavation number (121) is marked in ink on the right side of the base. Bibliography: Fairman 1939, 143; P. Spencer 1997, 46–7. Description: This pale yellow-beige sandstone statue depicts the bust and head of a ram set upon a base. The soft stone is badly eroded in places, with the eyes almost imperceptible, though the snout is well modelled, and the mouth is summarily indicated through an incised line. A straight and short, somewhat blocky, beard extends below the chin. A thick horn curls around each side of the face. Below the head, the bust is summarily rather than naturalistically modelled, with a vertical recess between base and chin. The sides and back of the statue are less carefully modelled, with a somewhat bevelled surface left by the sculptor. A socket (2.8 × 3.4cm; 4.0cm deep) in the top of the ram’s head

IN TEMPLE AND HOME: STATUARY IN THE TOWN OF AMARA WEST, UPPER NUBIA

119

Fig. 14: Sandstone statue of a ram; (left to right) right side, left side, rear. Sudan National Museum 3068. Photographs: Neal Spencer. Courtesy of the National Corporation of Antiquities & Museums.

Fig. 15: Sandstone statue of a ram; detail of head. Sudan National Museum 3068. Photograph: Neal Spencer. Courtesy of the National Corporation of Antiquities & Museums.

Fig. 16: Sandstone statue of a ram; detail of relief decoration on the front of the base. Sudan National Museum 3068. Photograph: Neal Spencer. Courtesy of the National Corporation of Antiquities & Museums.

provided a fitting for a crown. Additional holes are drilled into the side of the head. Remnants of a pale white-grey plaster survive on the right horn, the upper part of the bust, across the front recess and on the lower right part of the carved scene upon the front of the altar. It is most likely the whole statue was coated in white plaster; in places this is up to 5mm thick. Colourful paint was then applied to the plaster: the proper right back of the head has remnants

of black and blue lines, with more blue pigment surviving beneath the right horn. The left eye is outlined in black, and yellow paint survives around this. As such, the head of the ram is likely to have been yellow, with details (eyes, mouth) in black, and the remainder painted blue. The sub-cuboid base of the Amara West statue, rounded at the back, is embellished with a projection along the top, evoking a cavetto cornice. A small

120

N. SPENCER

sunk-relief scene of a non-royal individual striding towards the right is carved on the front of the base (see Fig. 16). He wears a long flaring kilt and a simple short wig and holds his left hand up, palm open. His right hand is bent to hold an object over his shoulder, perhaps a staff. The man strides alongside a bull, embellished with a necklace with lotus flower. Both man and bovid walk upon a simple register line that does not extend to the edge of the altar. No trace of inscription survives, either in this scene or elsewhere on the statue. The underside of the base is summarily carved to a relatively even surface allowing the statue to stand unsupported. The statue is largely complete, other than the original crown and some loss to the surface, especially the proper left horn. A part of the proper right front of the altar and the right horn have been lost to damage. 2. Block statue of Amenemhat (Fig. 17) Brooklyn Museum of Art 39.426. Material: Quartzite. Dimensions: Height 35cm, width 18.3cm, depth 28 cm. Provenance: Found ‘in sand fill of colonnade’ (Fairman diary, AWII.1, 9; 4 December 1938). Bibliography: PM VII, 159; P. Spencer 1997, 38; P. Spencer 2016, 10, pl. 201a; [accessed 3 February 2019]. Description: A very fine, though small, block statue representing a male figure seated upon a small cushion, set on a rectangular base. The statue is finely polished, and decorated with sunk-relief inscriptions that run around the base, down the front of the body and down a dorsal pillar. The head and parts of the right shoulder are missing; the remaining upper surface of the statue preserves parts of two crossed hands, somewhat summarily sculpted in low relief. While the sculptor has excelled in skilfully evoking the thighs and forearms of the individual, as if cloaked in a garment, the individual who carved the inscription was less careful, and several hieroglyphs had to be omitted from the column of text on the dorsal pillar to ensure it would fit (e.g.

41

pḥ.ww, see Wb I, 538, 4 and 5–6.

the ‘m’-sign in Amenemhat, and the shortened n kꜢ [n]). The stone itself features several large inclusions that range from light yellow to dark red. Inscriptions Front: prr.t nbt ḥr wḏḥw n Ἰmn-Rꜥ ḫnty pḥww n šꜢꜥ.t n kꜢ n sš ḥsb nbw Ἰmn-m-ḥꜢt ‘All that comes forth from the offering table of Amun-Ra foremost of the limits41 of Shaat [Sai], for the ka of the scribe of reckoning gold, Amenemhat’.

Left: ḥtp dỉ nsw Rꜥ-Ḥr-Ꜣḫty dỉ.f ḥꜥ.w rnp.w ỉrty ḥr dgꜢt ꜥnḫ.wy ḥr sḏm bw nfr n kꜢ n sš pr-ḥḏ [?] Ἰmn-m-ḥꜢt ‘An offering that the king gives for Ra-Horakhty, he gives the limbs/body of a youth, my two eyes seeing, my two ears hearing, (in) beautiful place, for the ka of the scribe of the pr-ḥḏ [?] Amenemhat’.

Right: ḥtp dỉ nsw Ἰmn-Rꜥ nb nswt tꜢwy nṯr.w nbw tꜢ-Stỉ r dỉ.sn ꜥnḫ wḏꜢ snb ḥꜥ.w rwḏ mn m ḥswt [n kꜢ n] sš ḥsb nbw Ἰmn -m-ḥꜢt ‘An offering that the king gives for Amun-Ra lord-of-thethrones-of-the-Two-Lands (and) all the gods of the BowLand, so that they may give l.p.h., strong and enduring limbs, in praise [to the ka] of the scribe of reckoning gold Amenemhat’.

Back: … ḏỉ(?).f nḏm-ỉb m ṯꜢ ḥww.t rꜥ nb ỉwty Ꜣbw n kꜢ (n) sš Ἰmn-[m]-ḥꜢt ‘… that he may give sweetness of heart in every day, without fail, for the ka (of) the scribe Amenemhat’.

3. Statue of a non-royal male (Figs 18–9) Sudan National Museum 3066. Excavated as object 169, marked as ‘422’. Material: sandstone. Dimensions: height 18.5cm; width 12.0cm; depth 16.5cm. Provenance: E.14.12, a corridor between the hypostyle hall and a suite of four magazines (E.14.7–10) (P. Spencer 1997, 60–61, [169]). Object found in ‘fill’. Bibliography: Fairman 1939, 143; PM III, 159. Description: Upper part of a male figure, shown seated with arms drawn up over knees, and shrouded in a garment that hides the arms and legs proper, though the shape of them is hinted at through the modelling. The head is covered with an unusual simple wig, with eyes and eyebrows modelled in a somewhat crude sfumato-manner, with a triangular nose and a broad,

IN TEMPLE AND HOME: STATUARY IN THE TOWN OF AMARA WEST, UPPER NUBIA

121

Fig. 17: Statue of Amenemhat, Brooklyn Museum of Art 39.426. Images from . Copyright: Brooklyn Museum of Art.

almost straight, mouth, while the proportions of the face appear somewhat flattened and wide. The most intriguing part of this statue is the broken area beneath and in front of the chin, which suggests a feature sculpted in front of the mouth. The body is better modelled than the head, with a relatively even curving surface, that evokes the width of the shoulders, and tapering down towards the waist; as mentioned above, both

arms are depicted as if stretching through a garment. Viewing the statue from below, the break exhibits a noted inwards curvature on the proper right, in contrast to the straight-sided left, suggesting an object, emblem or figure was held somewhat asymmetrically off to the proper right of the male figure. Carved from poor-quality sandstone with banding visible in places, the statue bears no evidence of any

122

N. SPENCER

Fig. 18: Sandstone statue, Sudan National Museum 3066; (clockwise from top left) front, left side, right side, back. Photographs: Neal Spencer. Courtesy of the National Corporation of Antiquities & Museums.

paint or plaster treatment of the surface, though the object is quite eroded. Patches of black across the wig and on the back left shoulder, and white on the underside of the proper right arm, seem to be later accretions. Furthermore, there are no traces of inscriptions, which may have been restricted to the base, now lost. The rough surface on the top of the right arm may even suggest the statue was never completed. 4. Ceramic head with nemes-wig (Figs 20–1)

Fig. 19: Sandstone statue, Sudan National Museum 3066; detail of face. Photograph: Neal Spencer. Courtesy of the National Corporation of Antiquities & Museums.

British Museum EA 85611; excavation numbers 94 and 95. Rejoined from two fragments: excavation objects nos 94 and 95; the latter is the fragment of the proper left edge of the headdress.

IN TEMPLE AND HOME: STATUARY IN THE TOWN OF AMARA WEST, UPPER NUBIA

123

Fig. 20: Head from a ceramic statue, British Museum EA 85611; (clockwise from top left) front, left side, right side, rear. Photographs: Kevin Lovelock. Courtesy of the Trustees of the British Museum.

Fig. 21: Head from a ceramic statue, British Museum EA 85611; top (left) and underside (right). Photographs: Kevin Lovelock. Courtesy of the Trustees of the British Museum.

124

N. SPENCER

Material: Ceramic. An unlevigated Nile silt fired to pale red with grey core, with inhomogeneous clay, infrequent elongated voids and rounded white inclusions (limestone/calcareous?), consistent with local pottery fabrics AW3–4, equivalent to Nile B2 (see Spataro, Millet and Spencer 2014).42 Dimensions: height 8.7cm; width 9.4cm; depth 8.1cm. Provenance: E.14.1, an extension to the temple magazines. Both fragments were found in the ‘upper level’. Bibliography: P. Spencer 1997, 71. 3D model: [accessed 3 February 2019]. Description: This fragment represents part of the head from a ceramic figure wearing a nemes-headdress, with the face entirely destroyed other than the lower jaw and the neck. The nemes is almost entirely preserved on the proper left side,43 showing the flaring side and lappet, with the distinctive banding indicated through lines incised prior to firing. The upper part of the figure is lost, including where one might expect a uraeus perhaps as a separate applied element in clay (or a hole for a further fitting above the headdress?). On the back of the head, the lines of the nemes converge at the base, representing the tie of the headcloth, with a small horizontal incision marking the top of the knot. The right side of the nemes is largely lost. Below the almost entirely destroyed face, where only the black core of the interior is visible, the interface between the neck and lower jaw is visible, with part of the right cheek surviving. Folds of flesh on the neck are suggested by the incising of two near-parallel lines.44 The surface of the neck has a noticeable curve to it, as it nears the interface with the lower part — or bottom — of the figure. This area of the preserved object is important, as it makes it unlikely the face was embellished with a beard, as one would expect this beard to rest against the neck, with negative space between beard and neck filled with ceramic material. The underside of the object is distinctive: a void is partly preserved, with the original fired surface visible,

Present location unknown, excavation number F4182. Material: sandstone, paint. Dimensions: height 29.2cm; width 23.2cm; thickness 11.2cm. Provenance: E13.3.26, rear room of a three-roomed house (see Fig. 8). Revealed within (4272), a deposit of windblown sand, resting upon a pedestal of mud brick and sandstone (4273, see Fig. 9). Bibliography: Spencer 2014a, 473–4, fig. 10 and pl. 16; N. Spencer 2015, 49, figs 12–3. 3D model: [accessed 3 February 2019]. Description: This small statue represents the bust of a figure wearing a black-coloured shoulder-length wig, with remnants of painted detail, including a composite collar and red-hued skin. The eyebrows, eyes, nose and mouth — the last with a noticeable if subtle smile — are visible despite the eroded surface, but the ears are not indicated and there is no evidence for a beard; much of the nose has been broken away. The bust is finely if schematically carved, with an elliptoid crosssection that tapers out towards the base.45 The base is sculpted flat, allowing the bust to stand unsupported on a flat surface.

42

44

43

Fabric description provided by Michela Spataro (British Museum, Department of Scientific Research), on the basis of examination with a 10× hand lens. The joining fragment has now been re-adhered by British Museum conservator Loretta Hogan.

indicating the object was not a solid piece. The shape of this void is roughly hemispherical, while none of the edges at the base of the object are original, but rather display the black core of the ceramic; i.e. the object once extended further below the face than it currently does. The head is hand-formed in clay, though the nemeswig seems to have been modelled separately and applied to the core before firing. The piece has been carefully considered, with both detail of the wig and folds of flesh indicated, and proportions consistent with statuary of the time. The surviving surface areas bear no slip, nor any traces of painted decoration. 5. Anthropoid bust of a male (Figs 22–3)

45

A feature highlighted in other New Kingdom statuary, for example the wooden figure of an official, British Museum EA 32772 (Russmann 2001, 182–83 [92]). ‘Mound-shape’ (Keith 2011, 36–38, figs 4a–b).

IN TEMPLE AND HOME: STATUARY IN THE TOWN OF AMARA WEST, UPPER NUBIA

Fig. 22: Sandstone anthropoid bust F4182, current location unknown. Photograph: Neal Spencer. Courtesy of the Trustees of the British Museum (Amara West Research Project).

125

126

N. SPENCER

Fig. 23: Sandstone anthropoid bust F4182, current location unknown. Drawing: Elisabeth Greifenstein / Claire Thorne. Courtesy of the Trustees of the British Museum (Amara West Research Project).

While the surface is affected by spots of black discoloration, presumably from the depositional environment — a small surface crack running diagonally along the front of the bust may be the result of insect action — remnants of painted decoration survive across the bust itself, hinting at an originally brightly coloured appearance. The bust seems to have been covered with a thin white wash, and a composite broad collar was then depicted in paint; areas of red pigment survive on both the front and back. Small areas of blue beneath the left shoulder, and under the wig on the rear, indicate that the collar ran around the bust and comprised at least two rows of beads. A small area of red paint was also noted on the underside of the base. Abbreviations KRI = Kitchen, K. A. 1969–90. Ramesside inscriptions, historical and biographical. Oxford. PM = Porter, B. and Moss, R. 1927–51. Topographical bibliography of ancient Egyptian hieroglyphic texts, reliefs and paintings. Oxford. Wb = Erman, A. and Grapow, W. 1926-1931. Wörterbuch der ägyptischen Sprache. Berlin.

Bibliography Adams, B. 1986. Sculpted pottery from Koptos in the Petrie Collection. Warminster. Andreu, G. 2000. La collection égyptienne du musée Dobrée (Nantes). Bulletin de la Société française d’égyptologie 148, 18–30. Auenmüller, J. 2018. New Kingdom towns in Upper Nubia: Sai, Soleb and Amara West in prosopographical perspective. In J. Budka and J. Auenmüller (eds), From microcosm to macrocosm: Individual, households and cities in ancient Egypt and Nubia. Leiden, 239–60. Azim, M. 1975. Quatre campagnes de fouilles sur la Forteresse de Saï, 1970–1973. 1ère partie: l’installation pharaonique. Cahiers de recherches de l’Institut de papyrologie et d’égyptologie de Lille 3, 91–125. Azim, M. and Carlotti, J.-F. 2012. Le temple A de l’île de Saï et ses abords. In Doyen and Devauchelle 2012, 11–65. Bernhauer, E. 2010. Innovationen in der Privatplastik. Die 18. Dynastie und ihre Entwicklung. Philippika, Altertumskundliche Abhandlungen 27. Wiesbaden.

IN TEMPLE AND HOME: STATUARY IN THE TOWN OF AMARA WEST, UPPER NUBIA

Betrò, M. C., del Vesco, P. and Miniaci, G. 2009. Seven seasons at Dra Abu El-Naga. The tomb of Huy (TT 14): preliminary results. Pisa. Bianchi, R. S. 2016. Statue der Göttin Thoeris beim opfern eines Widderkopfes. In N. Kehrer (ed.), Ramses. Göttlicher Herrscher am Nil. Petersburg, 232. Bierbrier, M. L. 1982. Hieroglyphic texts from Egyptian stelae, etc., in the British Museum, Part 10. London. Binder, M. 2011. The 10th–9th century BC – new evidence from Cemetery C of Amara West. Sudan & Nubia 15, 39–53. ———. 2017. The New Kingdom tombs at Amara West. In Spencer, Stevens and Binder 2017, 591–613. Blackman, A. M. 1937. Preliminary report on the excavations at Sesebi, Northern Province, AngloEgyptian Sudan, 1936–37. Journal of Egyptian Archaeology 23, 145–51. Bonnet, C. and Valbelle, D. 2018. Les temples égyptiens de Panébès ‘Le jujubier’ à Doukki Gel – Soudan. Mission archéologique suisse-francosoudanaise de Kerma / Doukki Gel. Paris. Bruyère, B. B. 1929. Rapport sur les fouilles de Deir el Médineh (1928). Fouilles de l’Institut français d’archéologie orientale du Caire 6.2. Cairo. ———. 1934–1935. Rapport sur les fouilles de Deir el Médineh, 1931–1932. Fouilles de l’Institut français d’archéologie orientale 10. Cairo. ———. 1939. Rapport sur les fouilles de Deir el Médineh (1934-1935), III: Le village, les décharges publiques, la station de repos du col de la Vallée des Rois. Fouilles de l’Institut français d’archéologie orientale du Caire 16. Cairo. ———. 1952. Rapport sur les fouilles de Deir el Médineh, 1935 à 1940. Fouilles de l’Institut français d’archéologie orientale du Caire 20. Cairo. Bryan, B. 1992. Designing the cosmos: Temples and temple decoration. In B. M. Bryan, A. P. Kozloff and L. M. Berman (eds), Egypt’s dazzling sun: Amenhotep III and his world. Cleveland, 73–120. Budka, J. (ed.). 2017a. Across Borders I: The New Kingdom town of Sai Island, Sector SAV1 North. Contributions to the Archaeology of Egypt, Nubia and the Levant 4. Denkschriften der Gesamtakademie 82. Vienna. ———. 2017b. The 18th Dynasty on Sai Island – new data from excavations in the town area and cemetery SAC 5. Sudan & Nubia 21, 71–81.

127

Capart, J. 1944. Sur un texte d’Hérodote. Chronique d’Égypte 19, 219–27. Černý, J. 1947. Graffiti at the Wādi el-’Allāḳi. Journal of Egyptian Archaeology 33, 52–7. Chadefaud, C. 1982. Les statues porte-enseignes de l’Égypte ancienne (1580 – 1085 avant J. C.). Signification et insertion dans le culte du Ka royal. Paris. Clère, J. J. 1995. Les chauves d’Hathor. Orientalia Lovaniensia Analecta 63. Leuven. Connor, S. 2016. Le statue del Museo Egizio. Modena. Cotelle-Michel, L. 2004. Les sarcophages en terre cuite en Égypte et en Nubie de l’époque prédynastique à l’époque romaine. Dijon. Daressy, G. 1906. Statues de divinités. Catalogue général des antiquités égyptiennes du Musée du Caire. Cairo. Darnell, J. C. and Darnell, D. 2014. Theban Desert Road Survey. Sixteenth (2007–2008) field season final report. Annales du Service des antiquités de l’Égypte 87, 29–59. Davies, N. de G. 1943. The tomb of Rekh-mi-rē at Thebes. Publications of the Metropolitan Museum of Art Egyptian Expedition 11. New York. Davies, N. M., and Davies, N. de G. 1933. The tombs of Menkheperrasonb, Amenmose, and another (nos. 86, 112, 42, 226). Theban Tomb Series 5. London. ———. 1963. Scenes from some Theban tombs: (nos. 38, 66, 162, with excerpts from 81). Private tombs at Thebes 4. Oxford. Davies, W. V. 2014. A statue of Amenhotep III rediscovered. In J. R. Anderson and D. A. Welsby (eds), The Fourth Cataract and Beyond. Proceedings of the 12th International Conference for Nubian Studies. 1–6 August 2012, London. British Museum Publications on Egypt and Sudan 1. Leuven, 399–408. ———. 2016. Recording Jebel Dosha: The chapel of Thutmose III. Sudan & Nubia 20, 18–27. ———. 2017a. Nubia in the New Kingdom: The Egyptians at Kurgus. In Spencer, Stevens and Binder 2017, 65–105. ———. 2017b. A statue-cache from Sai: Putting the pieces together. In Spencer, Stevens and Binder 2017, 133–48. Dorman, P. F. 2002. Faces in clay: Technique, imagery, and allusion in a corpus of ceramic sculpture from ancient Egypt. Münchner Ägyptologische Studien 52. Mainz am Rhein.

128

N. SPENCER

Doyen, F. 2015. Du trait élémentaire à la perception vitale: Quelques figurines féminines de la ville pharaonique de Saï. Cahiers de recherches de l’Institut de papyrologie et d’égyptologie de Lille 30, 133–57. ———. 2017. The architecture of SAV1 North. In J. Budka 2017a, 23–118. Doyen, F. and Devauchelle, D. (eds). 2012. Fouilles sur l’île de Sai (Soudan) 2005–2010. Cahiers de recherches de l’Institut de papyrologie et d’égyptologie de Lille 29 (2011–2012). Lille. Edwards, I. E. S. 1939. Hieroglyphic texts from Egyptian stelae e.t.c. in the British Museum 8. London. Emery, W. B., Smith, H. S. and Millard, A. 1979. The fortress of Buhen. The archaeological report. Excavation Memoir 49. London. Epigraphic Survey. 1994. Reliefs and inscriptions at Luxor Temple, 1: The festival procession of Opet in the colonnade hall. Oriental Institute Publications 112. Chicago. Fairman, H. W. 1938. Preliminary report on the excavations at Sesebi (Sudla) and ’Amarah West, Anglo-Egyptian Sudan, 1937–9. Journal of Egyptian Archaeology 24, 151–6. ———. 1939. Preliminary report on the excavations at ‘Amārah West, Anglo-Egyptian Sudan, 1938–9. Journal of Egyptian Archaeology 25, 139–44. Frood, E. 2016. Role-play and group biography in Ramessid stelae from the Serapeum. In R. Landgráfová and J. Mynářová (eds), Rich and great: Studies in honour of Anthony J. Spalinger on the occasion of his 70th feast of Thoth. Prague, 69–87. Gabolde, L. 2012. Réexamen des jalons de la présence de la XVIIIe dynastie naissante à Saï. In Doyen and Devauchelle 2012, 115–37. Gabolde, M. 2018. Insight into the perception of royal and divine powers among Kushites and Egyptians. In M. Honegger (ed.), Nubian archaeology in the XXIst century. Proceedings of the Thirteenth International Conference for Nubian Studies, Neuchâtel, 1st–6th September 2014. Publications de la Mission Archéologique Suisse à Kerma 1. Orientalia Lovaniensia Analecta 273. Leuven; Paris; Bristol, CT, 91–103. Guglielmi, W. and Dittmar, J. 1992. Anrufungen der personlichen Frommigkeit auf Gans- und WidderDarstellungen des Amun. In I. Gamer-Wallert and

W. Helck (eds), Gegengabe, Festschrift für Emma Brunner-Traut. Tübingen, 119–42. Hein, I. 1991. Die ramessidische Bautätigkeit in Nubien. Göttinger Orientforschungen, 4. Reihe: Ägypten 22. Wiesbaden. Hill, M. 2016. Ptah’s profile. In J. van Dijk (ed.), Another mouthful of dust. Egyptological studies in honour of Geoffrey Thorndike Martin. Orientalia Lovaniensia Analecta 246. Leuven, 251–307. Jacquet, J. 1994. Karnak-Nord VII. Le Trésor de Thoutmosis Ier. Installations antérieures ou postérieures au monument. Fouilles de l’Institut français d’archéologie orientale du Caire 36. Cairo. Jacquet-Gordon, F. 1999. Karnak-Nord VIII. Le Trésor de Thoutmosis Iᵉʳ. Statues, stèles et blocs réutilisés. Fouilles de l’Institut français d’archéologie orientale du Caire 39. Cairo. Janes, G. 2012. The shabti collections 5: A selection from the Manchester Museum. Cheshire. Kaiser, W. 1990. Zur Büste als einer Darstellungsform ägyptischer Rundplastik. Mitteilungen des Deutschen Archäologischen Instituts, Abteilung Kairo 46, 270–85. Keith, J.-L. 2011. Anthropoid busts of Deir el-Medineh and other sites and collections. Documents de fouilles de l’Institut français d’archéologie orientale du Caire 49. Cairo. Keith, J.-L. and Stevens, A. 2011. Amarna. In Keith 2011, 18–21. Kemp, B. and Stevens, A. 2010. Busy lives at Amarna: Excavations in the Main City (Grid 12 and the house of Ranefer, N49.18), I: The excavations, architecture and environmental remains. Excavation Memoir 90. London. Kendall, T., Mohamed, H. A., Wilson, H., Haynes, J. and Klotz, D. 2017. Jebel Barkal in the New Kingdom: An emerging picture. In Spencer, Stevens and Binder 2017, 159–92. Klemm, R., and Klemm, D. D. 2008. Stones and quarries in ancient Egypt. London. Lecuyot, G. 2001–2002. Statues en terre cuite retrouvées dans le secteur du sanctuaire du Ramesseum. Memnonia 12–13, 123–40. Legrain, G. 1906. Statues et statuettes de rois et de particuliers, I. Catalogue général des antiquités égyptiennes du Musée du Caire. Cairo. ———. 1909. Statues et statuettes de rois et de particuliers, II. Catalogue général des antiquités égyptiennes du Musée du Caire. Cairo.

IN TEMPLE AND HOME: STATUARY IN THE TOWN OF AMARA WEST, UPPER NUBIA

Lerner, L. 2019. Ancient urban villa with shrine for ancestor worship discovered in Egypt. [Accessed 27 January 2019]. Loeben, C. E. 2001. Beobachtungen zu Kontext und Funktion königlicher Statuen im Amun-Tempel von Karnak. Leipzig. Minault-Gout, A. 2012. La figurine funéraire Saï inv. S. 964 (SNM 23424) et un groupe de quatre chaouabtis de la XVIIIe dynastie de même type. In Doyen and Devauchelle 2012, 189–200. Mougenot, F. 2012. Au plus près des offrandes: Des statues de particuliers dans le grenier du dieu au Nouvel Empire. Revue d’égyptologie 63, 201–7. Müller, I. 2013. Die Verwaltung Nubiens im Neuen Reich. Meroitica, Schriften zur altsudanesischen Geschichte und Archäologie 18. Wiesbaden. O’Connor, D. 1969. Abydos and the University Museum. Expedition 12, 29–38. Page, A. 1976. Egyptian sculpture. Archaic to Saite. From the Petrie Collection. Warminster. Perdu, O. 2003. Des pendentifs en guise d’ex-voto. Revue d’égyptologie 54, 155–66. Petrie, W. M. F. 1896. Koptos. London. ———. 1907. Gizeh and Rifeh. British School of Archaeology in Egypt / Egyptian Research Account 13. London. Poole, F. 2015. The materialisation of status in the New Kingdom. In C. Greco (ed.), Museo Egizio. Modena, 90–103. Randall-MacIver, D., and Woolley, C. L. 1911. Buhen. University of Pennsylvania, Egyptian Department of the University Museum. Eckley B. Coxe Junior Expedition to Nubia 7/8. Philadelphia. Robichon, C., and Varille, A. 1937. Thèbes. Fouilles de l’Institut français d’archéologie orientale. Chronique d’Égypte 12, 174–80. Rondot, V. 2011. De la fonction des statue-cubes comme cale-porte. Revue d’égyptologie 62, 141–57. Russmann, E. R. 2001. Eternal Egypt. Masterworks of ancient art from the British Museum. Berkeley, CA. Salvador, C. 2015. VI.72. Votive stela for Amun. In P. Giovetti and D. Picchi, Egypt. Millenary splendour. The Leiden collection in Bologna. Milan, 551–2.

129

Schiff Giorgini, M. 1998. Soleb V: Le temple. Bas-reliefs et inscriptions. Edited by Nathalie Beaux. Bibliothèque générale 19. Cairo. ———. 2002. Soleb III: Le temple. Description. Edited by Nathalie Beaux. Bibliothèque générale 23. Cairo. ———. 2003. Soleb IV: Le temple. Plans et photographies. Edited by Nathalie Beaux. Bibliothèque générale 25. Cairo. Schulz, R. 1992. Die Entwicklung und Bedeutung des kuboiden Statuentypus: eine Untersuchung zu den sogenannten ‘Würfelhockern’. Hildesheimer ӓgyptologische Beiträge 33/34. Hildesheim. Smith, S. T. 1990. Askut in Nubia: The economics and ideology of Egyptian imperialism in the second millennium B.C. Studies in Egyptology 1. London. ———. 2014. A potter’s wheelhead from Askut and the organization of the Egyptian ceramic industry in Nubia. Journal of the American Research Center in Egypt 50, 103–21. ———. 2018. Colonial entanglements. Immigration, acculturation, hybridity in New Kingdom Nubia (Tombos). In M. Honegger (ed.), Nubian archaeology in the XXIst century. Proceedings of the Thirteenth International Conference for Nubian Studies, Neuchâtel, 1st–6th September 2014. Publications de la Mission Archéologique Suisse à Kerma 1. Orientalia Lovaniensia Analecta 273. Leuven; Paris; Bristol, CT, 71–89. Smith, S. T. and Buzon, M. R. 2018. The fortified settlement at Tombos and Egyptian colonial strategy in New Kingdom Nubia. In J. Budka and J. Auenmüller (eds), From microcosm to macrocosm: Individual, households and cities in ancient Egypt and Nubia. Leiden, 205–25. Spataro, M., Millet, M. and Spencer, N. 2014. The New Kingdom settlement of Amara West (Nubia, Sudan): Mineralogical and chemical investigation of the ceramics. Archaeological and Anthropological Sciences, 1–23. Spence, K. 2017. Sesebi before Akhenaten. In Spencer, Stevens and Binder 2017, 449–63. Spence, K., Rose, P. J., Bradshaw, R., Collet, P., Hassan, A., MacGinnis, J., Masson, A. and Van Pelt, P. 2011. Sesebi 2011. Sudan & Nubia 15, 34–8. Spencer, N. 2009. Cemeteries and a late Ramesside suburb at Amara West. Sudan & Nubia 13, 47–61. ———. 2014a. Amara West: Considerations on urban life in colonial Kush. In J. R. Anderson and

130

N. SPENCER

D. A. Welsby (eds), The Fourth Cataract and Beyond. Proceedings of the 12th International Conference for Nubian Studies. 1–6 August 2012, London. British Museum Publications on Egypt and Sudan 1. Leuven, 457–85. ———. 2014b. Creating and re-shaping Egypt in Kush: Responses at Amara West. Journal of Ancient Egyptian Interconnections 6 (1), 42–61. . ———. 2015. Amara West: House and neighbourhood in Egyptian Nubia. In M. Müller (ed.), Household studies in complex societies: (Micro)archaeological and textual approaches. Oriental Institute Seminars 10. Chicago, 169–210. ———. 2017. Building on new ground: The foundation of a colonial town at Amara West. In Spencer, Stevens and Binder 2017, 323–55. Spencer, N., Stevens, A. and Binder, M. 2014. Amara West. Living in Egyptian Nubia. London. ———. (eds). 2017. Nubia in the New Kingdom. Lived experience, Pharaonic control and indigenous traditions. British Museum Publications on Egypt and Sudan 3. Leuven; Paris; Bristol, CT. Spencer, P. 1997. Amara West I. The architectural report. Excavation Memoir 63. London. ———. 2016. Amara West III. The scenes and texts of the Ramesside temple. Excavation Memoir 114. London. Spiegelberg, W. 1921. Ägyptische und andere Graffiti (Inschriften und Zeichnungen) aus der Thebanischen Nekropolis. Heidelberg. ———. 1927. Der heilige Widderkopf des Amun. Zeitschrift für ägyptische Sprache und Altertumskunde 62, 23–7. Steindorff, G. 1937. Aniba II. Glückstadt. Stevens, A. 2006. Private religion at Amarna: The material evidence. BAR International Series 1587. Oxford. ———. 2017. Female figurines and folk culture at Amara West. In Spencer, Stevens and Binder 2017, 407–27. Teeter, E. 2003. Ancient Egypt: Treasures from the collection of the Oriental Institute. Oriental Institute Museum Publications 23. Chicago.

Thill, F. 2012. Statuaire privée égyptienne de Saï. In Doyen and Devauchelle 2012, 253–95. Tombos 2016a = Scribe-reckoner of the Gold of Kush. [Accessed 13 March 2019]. Trad, M. and Mahmoud, A. 1993. Aménophis III au Musée Égyptien du Caire. Dossiers d’archéologie 180, 40–7. Trapani, M. 2015. Popular and household cults at Deir el-Medina. In C. Greco (ed.), Museo Egizio. Modena, 120–39. Valbelle, D. 2001. Kerma. Les inscriptions et la statuaire. Genava 49, 229–34. ———. 2003. Kerma. Les inscriptions et la statuaire. Genava 51, 291–300. ———. 2005. Kerma. Les inscriptions et la statuaire. Genava 53, 251–4. ———. 2009. Kerma. Les inscriptions et la statuaire. Genava 57, 109–19. ———. 2011. Les statues égyptiennes découvertes à Kerma et Doukki Gel. In D. Valbelle and J.-M. Yoyotte (eds), Statues égyptiennes et kouchites démembrées et reconstituées. Paris, 13–20. Valbelle, D. and Bonnet, C. 2003. Amon-Rê à Kerma. In N.-C. Grimal, A. Kamel and C. May-Sheikholeslami (eds), Hommages à Fayza Haikal. Bibliothèque d’étude 138. Cairo, 289–304. Van de Walle, B. 1952. Le pieu sacré d’Amon. Archiv Orientalni 20, 111–35. Vercoutter, J. 1958. Excavations at Sai 1955–57. Kush 6, 144–69. Williams, B. B. 1983. Excavations between Abu Simbel and the Sudan frontier, Part 5: C-Group, Pan Grave, and Kerma remains at Adindan Cemeteries T, K, U, and J. Oriental Institute Nubian Expedition 5. Chicago. Woodward, J., Macklin, M., Spencer, N., Binder, M., Dalton, M., Hay, S. and Hardy, A. 2017. Living with a changing river and desert landscape at Amara West. In Spencer, Stevens and Binder 2017, 227–57. Yoshimura, S. and Kawai, N. 2002. An enigmatic rockcut chamber. Recent Waseda University finds at North Sakkara. KMT: A Modern Journal of Ancient Egypt 13/2, 22–9.

EVIDENCE FROM AMARNA RELIEFS CONCERNING ROYAL STATUES AND THEIR CONTEXTS: HOW ACCURATE WERE THE RELIEFS?1 Kristin THOMPSON

Abstract Any investigation of the appearance of the ancient city of Akhetaten (modern Tell el-Amarna) is likely to draw to some extent upon the depictions of its buildings and other features in the reliefs on the walls of tombs and royal buildings at the site. The question then arises as to how accurate and useful such depictions are. In seeking evidence for the number and placement of statues in the royal buildings, one discovers that they are relatively rarely shown in reliefs. When they are shown, only a small number are present, primarily in the Great Aten Temple, with none at all shown in the Small Aten Temple or, with one possible exception, in the palaces. Yet early excavations at Amarna uncovered numerous statues in royal buildings. The large number of statuary fragments discovered, or rediscovered in dumps, by the current expedition to Amarna further indicates that in fact these buildings contained numerous statues, even though their decoration was in most cases incomplete. This paper demonstrates that, although the reliefs do not offer an indication of the number of statues finished and installed, some useful evidence for poses and placements within the buildings can be gleaned from them. * * * One innovative aspect of Amarna art was the extensive depiction of buildings in reliefs on the walls of tombs and royal buildings. Numerous rooms and courts are laid out in almost blueprint fashion, and details of their interiors — including furniture, columns, doorways, stored supplies, busy or lounging servants and priests preparing offerings — all give a sense of a relatively realistic presentation. A particularly large image of the Great Aten Temple (GAT) in the tomb of Meryra extends the length of one long wall and even turns the

1

Many thanks to Marianne Eaton-Krauss, Marsha Hill, Barry J. Kemp and Christian Bayer for help and suggestions in the preparation of this essay.

corner onto an adjacent, shorter wall (Davies 1903–8, I, pl. XXV). Such reliefs give the impression that these buildings are shown in their entirety, though they are likely to be simplified views of finished structures or aspirational views of buildings as planned but never completed. Such detailed and precise depictions of buildings, at least relative to earlier Egyptian artistic conventions, tempt the archaeologist and the art historian to seek clues as to the original layouts of buildings and the ways in which they were decorated. For the layouts of the buildings, the reliefs have in some cases proven helpful, when interpreted cautiously. To what extent can these reliefs be used as evidence for the number, scales, poses, placements and contexts of statues in the city’s royal buildings? The only way to answer that question is to compare the images with the statuary finds from the represented buildings. Such a comparison has become more feasible in recent years. Since the current expedition began work in 1977, many hundreds of statuary fragments have been recovered from a dump left in the mid-1930s by the Egypt Exploration Society (EES) team behind its expedition house, a deposit referred to as the North House Dump. Evidence suggests that most or all of them came from the Great Palace (GP). Hundreds more were found in the current expedition’s excavations at the Kom el-Nana, Small Aten Temple (SAT) and GAT. These have more precise recorded find-spots within the buildings than do the North House Dump pieces. Added to the pieces from previous excavations at Amarna now in museums, these finds make it clear that all the temples contained some statuary, and that a considerable number stood in the GP. Even having pieces representing only a small portion of the royal statues reveals more about poses and scales than can be learned from the reliefs. The convention of representing important places and persons on a larger

132

K. THOMPSON

scale than other elements of a scene means that the scale of statues in reliefs cannot confidently be judged. The Osiride statues depicted in several reliefs as standing outside the Sanctuary of the GAT are small compared to nearby human figures and altars. Several fragments that probably came from those statues, as well as similar statues from the East Karnak Aten temple (Manniche 2010) and the Broad Hall of the GP (Pendlebury 1951, vol. 1, 52, 66 and vol. 2, pls XV, 2 and LXVII, 7 and 8), indicate that they were colossi. Similarly, numerous diagnostic pieces reveal the commoner poses, such as members of the royal family holding offering tables, Akhenaten standing in an Osiride pose, or princesses holding an object to their breasts, as well as some innovative poses from the reign. Certainly for determining the number of statues in most of the buildings, the reliefs are largely useless. This is particularly true of the palaces. The GP is the largest single source of statuary finds in the entire site, and yet not a single relief can confidently be identified as showing that huge complex. There seem to have been no identifiable fragments of statues found in the North Palace. At least two statuettes in poor limestone adorned the façade of the North Riverside Palace, thought to have been the main residential palace for the royal family. Nearly all of that building went under the cultivation before excavations began at the city, and thus any other statuary fragments are lost. The relatively small palace or palaces shown in several tomb reliefs contain no statues at all, and thus they need not be considered here. The one unprovenanced block that apparently depicts a statue in a palace will be discussed below. The main information that the reliefs offer concerns the placement and context of statues. The usefulness of the reliefs’ information depends on how accurately they reflect the actual layout and appearance of buildings which have yielded fragments of royal statues. The GAT Sanctuary had an entrance that included limestone Osiride colossi, and pieces of numerous smaller hard-stone statues have been discovered that must have stood in or around it and in the Stela Emplacement to its west. No statues are depicted at the front end of the GAT, though some fragments of statues have been found there. The SAT similarly contained statues of granite, quartzite and indurated limestone, including at least one granite colossus. Two of the several outlying sunshade temples have been excavated. The Kom el-Nana, associated with Nefertiti, was decorated with life-size or smaller statues of the royal

family in quartzite, indurated limestone and granodiorite. Complex MII in the Maru-Aten contained chips from bases of at least four substantial granite statues and four or five quartzite ones. Surface finds from a possible sunshade temple at el-Mangara included a few fragments of quartzite (now in the government magazine at el-Ashmunein), one of them part of a statue base. A possible additional sunshade, the Stone Chapel adjacent to the Desert Altars, yielded evidence for a few hard-stone statues. Of these temples, only the GAT and SAT seem to have been depicted in the tomb reliefs. Stéphane Pasquali has analysed the depictions of the temples in the tombs and found criteria for distinguishing which is which. He points out that some tombs depict the king’s reward to the tomb owners from the Window of Appearances (and in one case a temple courtyard). He interprets the culmination of the ceremony to be a chariot trip by the honoree to the SAT. That, he argues, is the temple depicted in the tombs of Tutu, Pentu and Parennefer (Pasquali 2013, 208, 212, 214). One of the indications Pasquali uses to identify the SAT is the absence of any depictions of statues. Those reliefs that show Osiride colossi between the columns at the entrance to the Sanctuary, and in some cases a seated statue at the Stela Emplacement, depict the GAT. The presence of statues does seem to identify a relief as depicting the GAT. Their absence, however, does not necessarily imply that the temple shown is the SAT. The temple in Tutu’s scene has a courtyard with a stela emplacement and slaughter yard, features that occur in the GAT but apparently not in the SAT (Pasquali 2013, 209–10). Pasquali also assumes that the large central altar in the first court of the SAT differentiates it from the GAT. Following recent excavations of the Long Temple, Kemp has pointed out, ‘The loss of the wide strip running down the axis of the temple leaves it uncertain as to whether this has been an open area or had supported a larger and longer offering-place, as is hinted at in the tomb pictures’ (Kemp 2013, 30; on the large altar in the GAT depicted in the tombs of Meryra and Panehesy, see Davies 1903–8, I, pl. XXV and II, pl. XVIII). The point remains, however, that images of the SAT are irrelevant here. The lack of statuary in images of the SAT probably cannot be explained by positing that the building was unfinished and hence undecorated. The SAT was one of the earliest buildings commenced upon the founding of the city (Mallinson 1995, 169, 212). It is the only royal building in which all the surviving Aten

EVIDENCE FROM AMARNA RELIEFS CONCERNING ROYAL STATUES AND THEIR CONTEXTS

cartouches are in their early form (Gabolde 1998, 117). Given that statues are known to have stood in the SAT, it follows that the reliefs do not always depict statues in buildings where they existed. In general, a very small number of the dozens, perhaps hundreds, of statues that stood in royal buildings were shown in the reliefs. This survey begins with an examination of the reliefs in situ in tombs, proceeds to isolated blocks without specific provenance, and to one from the GP. It concludes with a discussion. Representation of statues in the tomb reliefs 1. Meryra, first view (Davies 1903–8, I, pl. XI) The GAT is presented vertically, with its entrance at the bottom. Davies’ pl. XI contains the GAT’s back half. No statue appears by the benben at the Stela Emplacement, but four Osiride colossi stand in the pillared entrance to the Sanctuary. Three of these have a small female figure beside the king. No actual pieces of statuary have been found that would indicate whether this figure is Nefertiti or one of her daughters. A block discovered at Abydos, however, shows a similar scene. On its right side, Akhenaten wears the white crown, with a smaller female figure behind him, wearing a tripartite wig and tall double plumes. To the left are

Fig. 1: A 1936 photograph of Akhenaten’s right shoulder with a flail held against it, from a limestone colossus. It was found south of the temenos wall in the Sanctuary area of the Great Aten Temple. Courtesy of the Egypt Exploration Society.

133

similar figures of Akhenaten with the red crown and a queen with a double-plumed crown, while at the far left Akhenaten is depicted again but wearing the white crown, with only the plumes of a smaller figure surviving (Simpson 1995, 76–7, fig. 138). The plumes mark the smaller figure almost certainly as Nefertiti or possibly Tiye. Despite the unusually large disparity between the heights of Akhenaten and Nefertiti, the small figures are not princesses, and the same applies to the scenes in the Amarna tombs. The front half of the GAT (Davies 1903–8, I, pl. XII) shows no statues. The Osiride statues likely relate to substantial fragments of colossi made of ordinary limestone found by William Flinders Petrie and later the EES in the dumps to the south of the Sanctuary. The alternation of red and white crowns remains consistent across all representations of these colossi in tomb reliefs, but fragments so far discovered include no recognisable portions of crowns. Petrie lists ‘an ear, a toe, and a piece of the chest’ (Petrie 1894, 18). The only records of the EES finds are two photographs of a chest fragment (Fig. 1) and a group of smaller pieces including a false beard and some fists (Fig. 2). Given that they were photographed in the GAT spoil heaps and were not assigned numbers, they were probably left in the field and have disappeared. J. D. S. Pendlebury lists them as: a shoulder, collarbone and flail (see Fig. 1; 53cm in height); a knee and edge of a kilt, a foot in a sandal; three left

Fig. 2: A 1936 photograph of other fragments from at least three limestone colossi, found in the same area. Courtesy of the Egypt Exploration Society.

134

K. THOMPSON

fists with flail handles; two fragments of false beards (see Fig. 2; the larger 42cm long), and part of a stomach and kilt with an early Aten cartouche (Pendlebury 1951, vol. 1, 13).2 2. Meryra, second view (Davies 1903–8, I, pl. XXV) The upper half of the east and north walls depicts the GAT horizontally, stretching around a corner. The Stela Emplacement includes a seated statue of Akhenaten. Again, four Osiride colossi stand between pillars in the Sanctuary’s entryway but with no small statues beside them. There are no other statues in the temple. The fact that the seated statue by the Stela is absent in one temple scene and present in the other might be explained if the statue was installed during the interval between the carving of the two scenes. The temporal gap would be impossible to determine, but both reliefs contain the late forms of the Aten’s cartouches, and in each case there are four princesses participating, driving to the temple and then shaking sistra while their parents make offerings there.3 3. Panehsy (Davies 1903–8, II, pl. XIX) This plate (Davies 1903–8, II, pl. XIX) contains the back half of a long horizontal image of the GAT. The benben stela shares a base with a seated statue representing Akhenaten wearing the blue crown. Uniquely, the entrance to the sanctuary shows eight Osiride statues of Akhenaten beside eight pillars, whereas all other depictions show only four. There are no small figures beside them. No other statues are shown, with the possible exception of a recumbent headless bull or possibly a sphinx on a table beside the entrance; Davies has marked this object with a question mark. No other evidence for this statue, if that is what it was, is known. The front half of the temple (pl. XVIII) contains no statues. The three images of the seated statue in Meryra’s and Panehsy’s tombs may explain the origins of a group

2

3

The chest that Petrie mentions may be the same as the shoulder and flail found by the EES. Petrie identifies the pose of the colossal limestone statues as ‘with crossed arms holding the crook and flail’ (Petrie 1894, 18) which he could not have known from the other pieces he lists. Marc Gabolde has pointed out that the nobles’ tombs provide little evidence for dating their reliefs and inscriptions.

of polished diorite statuary fragments found by Howard Carter in the dumps south of the temenos wall at the sanctuary end of the GAT. These, now in the Metropolitan Museum in New York, include an ankle with a profile compatible with a seated statue (21.9.433), some toes (21.9.494, 563, and 574), and a possible left shoulder (21.9.335). Referring to the EES excavation of the Stela Emplacement area in 1933, Pendlebury mentions ‘Many fragments of purple sandstone [i.e., quartzite] from the stela and of black granite [i.e., granodiorite or diorite] from the statue’ (Pendlebury 1951, vol. 1, 12). During a 2012 re-excavation of the Stela Emplacement area, similar pieces of diorite were found, some of them in the EES spoil heaps from the 1933 excavation of the area. Since these pieces were not numbered and may have been left on the site, it is possible that some of the 1933 finds may have been among the ones discovered in 2012. They include an articulation, perhaps the side of a knee (S-7684), a stretch of body surface with pleats (S-7841), and a beaded area from the ridge of a blue crown (S-7840). The crown fragment resembles some pieces of the same stone in the Metropolitan Museum (21.9.533 and 537). Marsha Hill comments: ‘These relate closely in size and other indications to fragments from the Amherst collection in New York that may well be from this area, and indications from both sources suit remarkably well the statue depicted in the tombs of Meryra and Panehsy, although the size is slightly over-life size rather than colossal, as the tomb depictions could imply’ (Hill 2012, 6; for the archaeological context of these finds, see Shepperson 2012, particularly 24–5). Thus in this case, the three images of the seated colossus provide significant evidence for a probable pose and context for this important statue. On the other hand, fragments of smaller quartzite statues were found in the same vicinity, as well as pieces of reliefs, parapets and balustrades. As Mary Shepperson writes, these ‘point to a considerably more crowded and varied area than the simple, if impressive, installation suggested by the tomb depictions’ (Shepperson 2012, 25).

Princesses, particularly Maketaten, who are known from other evidence to have died by a certain time, are still depicted as if they were alive (Gabolde 1998, 110–2). Nevertheless, the fact that most of the temple scenes include only three princesses may hint at a later date for this particular image.

EVIDENCE FROM AMARNA RELIEFS CONCERNING ROYAL STATUES AND THEIR CONTEXTS

4. Huya, first view (Davies 1903–8, III, pls VIII, IX and X) A general view shows Akhenaten leading Tiye into her temple, identified as ‘the Sunshade of the King’s mother and Great Royal Wife Tiye’. This is the only relief that shows a building containing a substantial number of statues and apparently also the only one depicting a temple other than the GAT or the SAT. Behind the second pylon, there is a court with a ramp and altar, surrounded by a colonnade. Between the columns are probable pair statues, alternatively of Akhenaten’s father, Amenhotep III, alongside Tiye and of Akhenaten himself also alongside Tiye. Tiye wears her typical Hathoric crown featuring a disk, horns and double feathers. The pairs stand upon what seems to be a single base, their hands at their sides and their left legs advanced, as if walking into the temple. Davies identifies each pair as two separate statues side by side (Davies 1903–8, III, 21). There were originally eight pair statues on each side of the colonnade; portions of fifteen survive, with one missing apart from fragments of its inscription.

5. Huya, second view (Davies 1903–8, III, pl. XI) Pl. XI (Davies 1903–8, III) shows a detail of the temple’s inner shrine. Just inside the third pylon there is a short colonnade with three pair statues on either side, in this case holding offering tables. Amenhotep III alternates with Akhenaten, while Tiye again accompanies both men, wearing the Hathoric crown. A single pair statue of the same type, with Amenhotep III and Tiye, is depicted in a small room past the next doorway, for a total of seven. To the left of the next doorway is a table heaped with offerings, on either side of which there appears to be a statue of a king, facing away from the table. Their top halves are missing, but they probably also held offering tables, based on two parallel scenes: a relief in the tomb of Ahmes and Roeder block P. C. 260/Boston 63.961, both discussed below. The left edge of Huya’s relief contains the innermost shrine’s doorway and large offering table, approached by a low set of steps. On either side of the steps four individual statues are lined up: from the ‘top’ (i.e., to the left of the steps), Akhenaten, Tiye, Amenhotep III and Tiye again, all holding offering tables with support panels underneath. The same arrangement is repeated on the ‘bottom’ (to the right).

135

There are four figures on the steps to the innermost shrine. These might represent statues, but their feet — and possible bases — are missing. There is no parallel from Amarna for such a placement of standing statues on a staircase. W. Raymond Johnson and F. J. Gilles interpret these figures as representations of Akhenaten, Nefertiti, Amenhotep III and Tiye themselves, entering the shrine (Johnson 1996, 73; Giles 2001, 68). Davies assumed that the inauguration of the temple occurred too late in the reign for Amenhotep III to be alive and concluded that the figures are statues (Davies 1903–8, III, 23). The remains of the relief indicate the presence of thirty-two statues in all or perhaps thirty-six, assuming the figures on the steps to be statues. This seems a plausible number to have stood in a large but peripheral temple. There is, however, no way to confirm such a quantity. Kemp posits that this temple might be the ‘Lepsius building’, which is entirely lost beneath the modern cultivation and an irrigation canal. No statuary fragment traceable to that building has been identified (Kemp 1995, 412–4). Another candidate for Tiye’s sunshade temple is the small chapel to the west of the Desert Altars on the plain between the North Suburb and the cliffs containing the North Tombs. The chapel was built of limestone, most of which had later been removed for reuse, leaving behind broken bits of relief. As exposed by the EES in late 1931 and only cursorily recorded, its remains consisted of a rectangular gypsum foundation revealing little about the building’s layout. Its enclosure wall was 68 × 112m. The excavators estimated the chapel itself to have been ‘some ten m. square’. They also noticed traces of a similar building to the north (Frankfort and Pendlebury 1933, 101). Kemp confirms that these traces remain and suggests that it ‘would be worth cleaning again and planning in detail since this was something which was not done’ (Kemp 1995, 451). A small number of hard-stone objects were found there, one bearing the name of Amenhotep III and another the name of Tiye. Fragments of a granite bowl contain part of Amenhotep III’s prenomen cartouche, as well as a late cartouche of the Aten (EES 31/449; Frankfort and Pendlebury 1933, 102, pl. XLVII, 2; Liverpool 1973.1.472). Pieces of a statue’s offering table have Akhenaten’s nomen cartouche and part of Tiye’s cartouche on opposite sides (EES 31/44; Frankfort and Pendlebury 1933, 102, pl. XLVII, 3; Cairo 57210). Considering the temple’s scale as depicted in Huya’s tomb, it seems likely that the image shows

136

K. THOMPSON

Tiye’s sunshade as it was planned to look, while in reality it had not progressed to the stage where its full complement of statues had been installed. The relief nevertheless provides our only source for analysing this sunshade’s intended place in the overall statuary programme at Akhetaten. 6. Huya, third view (Davies 1903–8, III, pl. XVIII) The scene shows a sculptor’s workshop, with master sculptor Iuty working on a statue of Beketaten, Tiye’s daughter. She wears a dress and stands with one arm held down at her side and the other holding an object to her breast. That pose instantiates a well-attested Amarna statue type. Most statues of princesses show them nude, but there are clothed examples. There is no statue of Beketaten in the relief of Tiye’s sunshade, and no statue fragment identifiably representing Beketaten has been found. 7. Ahmes (Davies 1903–8, III, pl. XXX) Four colossal Osiride statues are depicted in the pillared entrance to the GAT Sanctuary. Three have small female figures beside them. Inside the Sanctuary there is a large central offering platform. On either side, facing away from the platform, is a pair statue of Akhenaten and Nefertiti (or individual statues of each) holding offering tables. He wears the blue crown, while she sports her tall, flat-topped crown. This is the only tomb relief which depicts statues inside the GAT sanctuary. 8. The royal tomb, first view (Martin 1989, pls 34 and 35) A relief on Wall A in Room Alpha shows the GAT at sunrise. The Sanctuary’s entrance shows eight columns, only four of which originally had Osiride colossal statues of the king beside them. Traces of two of them survive. At least one had a small female figure beside it (see detail, pl. 35).

4

Since Roeder lists the relief blocks by category, the same block may be described multiple times. I have cited the text which

9. The royal tomb, second view (Martin 1989, pls 47 and 59) A similar but less well-preserved image from Room Alpha, Wall C, contains faint traces of the four colossi (see detail, pl. 59, where the bent arm of one statue and part of its hips are indicated). 10. The royal tomb, third view (Martin 1989, pl. 68) A scene of the royal family mourning the death of the second oldest daughter, Meketaten, contains her figure standing in a kiosk. This could be a statue or her mummy in an anthropoid coffin. She holds her open hands at her sides and wears a dress and sandals. Representations of statues on relief blocks without specific provenance The tombs at Amarna contain the most extensive in situ images of the city’s major buildings. The many smaller relief fragments on talatat taken from the city for reuse elsewhere show that scenes similar to those in the tombs were carved on the walls of royal buildings. These blocks probably originated in all of the stone buildings in the city, and they became mixed during transit. Thus their provenance cannot be definitively traced to a specific building. A small number of blocks bear images of statues, which differ considerably from those in the tombs. They do not represent the familiar four (or eight) colossi from the GAT Sanctuary entrance, and only a few of them depict offering tables. This difference may suggest the artists who decorated the royal buildings were not the same ones who decorated the tombs and that the royal artists had greater access to parts of buildings where other types of statues stood. These blocks are discussed and illustrated in Roeder 1969.4 1. Block 84-VI (Roeder 1969, 164, pl. 27) ‘In der linken unteren Ecke einer grossen Darstellung, unmittelbar nahe der Umrahmung, steht auf

gives either the fullest description of the subject or the one most relevant to its depiction of statuary.

EVIDENCE FROM AMARNA RELIEFS CONCERNING ROYAL STATUES AND THEIR CONTEXTS

einem Sockel die Statue eines knienden (Kap. V J 3) Königs.’ Only the legs and hips survive. There is no hint as to the context for this statue, since decorative border elements occupy the rest of the block. Fragments of small kneeling statues of Akhenaten and Nefertiti similar to the one depicted survive (EES 32/29, from inside the southwest corner of the GAT temenos wall, San Diego 14736; EES 32/23, from outside the west side of the GAT temenos wall, Cambridge – Fitzwilliam Museum E.75.1933). 2. Block 775/VIII (Roeder 1969, 164, pl. 51) The right edge of a scene, bordered by vertical decorative bands. In the upper register at the left there is a statue base supporting bare legs; the figure survives only up to mid-thigh. Roeder takes this to be Akhenaten, but there appears to be no kilt, making it more likely that the statue represents a nude princess. Again, not enough of the image survives to suggest its context. 3. Block P. C. 260 (Roeder 1969, 164, pl. 207 and Cooney 1965, 100–1; Boston 63.961) This scene probably depicts the sanctuary of either the GAT or SAT. At the left are two doorways with broken lintels, each framing an offering table. In the centre is a large table heaped particularly high with offerings. There is a statue on its left, facing left, and one to its right, facing right. Both hold offering tables. Roeder takes the one on the right to be Nefertiti, probably on the basis of what looks like the outline of a floor-length dress. Nevertheless, the right-hand figure appears to have a faint groove that may depict the waistband of a kilt, suggesting that the statues both may depict Akhenaten. Cooney (1965, 100) also identifies both statues as Akhenaten. Enough of this scene survives to suggest that statues in this common offering-table pose were placed near altars in the temples. A fragment from the GP, discussed below, offers further evidence for this possibility. 4. Block 421–VII (Roeder 1969, 164, pls 21–2) At the left are two laden offering tables with a small stand between them. The right half is occupied by three evenly spaced columns. Between the middle and righthand columns stands a statue of Akhenaten, a royal female, and a child (on the identity of the woman, see

137

entry 5 below). At the far right there is a similar statue. Akhenaten and the woman appear to be holding hands, with her also holding the child’s hand. Small group statuettes and fragments of larger statues involving various combinations of the royal couple and the princesses holding hands survive (Paris E15593, Berlin 20494). Similar poses are also evident in the statues accompanying the boundary stelae. The image offers valuable evidence that sets of similar statues were placed between large columns in rows in royal buildings. The EES excavation of the Broad Hall showed that there had been plans to alternate columns, marked with squares on the floor plaster, and colossal statues of Akhenaten, indicated by rectangular stone foundations. Pendlebury concluded: ‘Although this area was originally laid out as a gigantic colonnade well over 150 metres in length, the scheme was never carried out, and the “parade ground” was bordered by statues alone’ (Pendlebury 1951, vol. 1, 52). The placement of statues between columns is shown in the relief of Tiye’s sunshade, discussed above. The peristyle hall in Amenhotep III’s mortuary temple at Kôm el-Hettan placed quartzite and granite Osiride colossi of the king between columns as well (on-site lecture from Hourig Sourouzian, February 2016). Thus the alternation of columns and statues may have been a common pattern for Amarna royal buildings. 5. Block 454-VII (Fig. 3; Roeder 1969, 38, pls 20–1) The left side of the block is blank, bordered at the right by a vertical band representing a wall. This band takes a ninety-degree turn at the bottom edge and continues across to the right side, representing a floor. A column at the right edge rests on the floor, as does the base of a group statue that occupies much of the right half of the block. A king stands at the right with his hands at his sides; beside him, a woman in a Nubian wig stands with her left arm around the king’s waist. She holds hands with a little girl, presumably a princess (Fig. 3). The identities of the three figures are uncertain. The king is probably Akhenaten. The woman does not appear to be Nefertiti. The rough texture of the stone and minor surface damage around the heads makes it difficult to tell, but she seems not to wear a uraeus, as Nefertiti would. Moreover, the wig appears possibly to have been re-carved from a Nubian one to a princess’s sidelock. Such an alteration is observed in a few relief blocks where images of Akhenaten’s secondary wife,

138

K. THOMPSON

Fig. 3: Amarna block from Hermopolis depicting a group statue of Akhenaten, Kiya (?) and a child on a base. Courtesy of the municipal archive, Hildesheim.

Kiya, have been altered to transform them into Meritaten (e.g., Copenhagen ÆIN 1776 and 1797). The depth of the carving and the harsh shadows make it difficult to ascertain whether such an alteration has occurred.5 Whether or not it did, Block 454-VII may provide rare evidence for the existence of one or more statues which included Kiya and may have been altered in favour of Meritaten.6 The most plausible place for such statues to have stood is the Maru-Aten, building complex MII, where a number of palimpsest inscriptions changing a name and titulary from Kiya to Meritaten

were found, notably on a fragmentary parapet (Oxford 1922.141).7 This image resembles the one in Roeder Block 421– VII, immediately above, in that it shows a group statue of Akhenaten, Kiya (?) and a princess. Here the single column may stand at the end of a row in a scene similar to the one above.

5

6

An examination of the original block seems impossible. The photograph in Fig. 3 was taken in 1939. Another, taken in 1959, shows the block with severe damage, with the upper right area entirely broken away; the king’s figure is missing from the waist up and the woman’s from the chest up. An undated drawing done before the block was damaged shows the woman wearing what appears to be a wig with no uraeus. These images are all in the municipal archive, Hildesheim, in the Roeder collection. My thanks to W. Raymond Johnson for the suggestion that the woman in this image might represent Kiya (pers. comm.).

6. Block 458/VII (Roeder 1969, 45, 52, pl. 22) The right half of the block shows a walled courtyard and half of a broken lintel from a doorway. At the left

7

A fragment of an offering table inscribed with the name of Kiya was purchased from George Fraser by the British Museum in 1891 (BM EA 26814). Its source at Amarna seems plausible, but its specific find-spot is unknown. Similar palimpsest inscriptions were found in the North Palace, but while fragments of hardstone statues were found in the MaruAten, none was discovered in the North Palace (see the section on the Maru-Aten in Hill and Thompson forthcoming).

EVIDENCE FROM AMARNA RELIEFS CONCERNING ROYAL STATUES AND THEIR CONTEXTS

is an interior room of a temple with the upper twothirds of a column depicted. At the far left a statue of Akhenaten faces left and holds an offering table. To the right of the column is a similar statue of Nefertiti wearing a short wig. Both figures survive down to about mid-thigh. Roeder notes that they stand with feet together. Short support panels are depicted beneath the offering tables, confirming that these figures are statues rather than the king and queen themselves. The presence of a column between the two statues fits with the layout of the two previous blocks, further suggesting that in royal buildings statues were sometimes placed between the columns in colonnades. 7. Block 494-VII (Roeder 1969, 170, 202, pl. 58) At the upper left corner of this image there is a statue of a royal woman, probably Nefertiti, facing left and holding an offering table supported by a pillar. The depiction is almost complete, save her head. A princess shaking a sistrum stands behind her mother on the base. A stretch of flat surface below separates them from a heap of offerings, presumably sitting on an offering table that has broken away at the bottom of the block. This statue’s context differs markedly from all the others described so far. Immediately behind the statue is a wall with a door, and beyond that another wall with another door. These appear to be the rectangular doorways used in representations of palaces rather than the broken-lintel ones shown in temples. Roeder identifies the setting as ‘einem Hof des Palastes nahe dem Harim’ (Roeder 1969, 202). In the room represented at the right side of the block, a female musician playing a harp sits on an odd round object, perhaps a mat or bolster-like cushion — or the circular object might be a tambourine placed beside her seat. Roeder (1969, 202) calls it ‘einem kreisrunden Hocker’. Below her is a shallow rectangular object that resembles the base under the statue at the left, but is probably something else. There is no evidence for any statue of a musician at Amarna. In addition, the figure of the harpist is larger than that of Nefertiti’s statue, a breach of decorum that would hardly be permitted. This block may be a unique case of an image of a statue standing in a palace. The musician’s larger size suggests that the statue is relatively small.

8

An image of this piece can be found online: (accessed 10 April 2018).

139

Pieces of small granite statues of the king and queen holding offering tables were found in the North Harem and in the southern area of the GP’s Broad Hall adjacent to it. The most notable one with a firm provenance is a granite fragment depicting the hips and thighs of the king, with a narrow broken area running down the centre where the table’s support panel would have met the body (EES 34/180; Brussels E 7210; for a complete list of the pieces, see Hill and Thompson forthcoming). The Kofler-Truniger Collection contains a similar unprovenanced portion of a granite statue of Nefertiti on the same scale and with a similar broken area.8 No archaeological evidence supports the notion that a statue of a princess holding a sistrum existed, but such a possibility should be entertained, especially given the anomalous arm positions indicated by the breaks on the torsos of some surviving princess statues. The sistrum would presumably have been made in metal and inserted into a hole in the hand. Representation of a statue on a relief from the Great Palace One depiction of a statue on a block from a royal relief was found in situ in the GP. It appears to show a statue of Akhenaten facing leftward and holding an offering table (Fig. 4; Toronto 966.81.11; Pendlebury 1951, vol. 2, pl. LXVIII, 8; Green 2009, 27–9). The slight uncertainty arises from the fact that the lower legs of the figure are missing, eliminating a possible base that the statue would have stood on. Nevertheless, the figure of the king is small in comparison with the altar standing in front of him and with the large Aten hands reaching down into the scene. These features suggest that this cannot be a depiction of the king himself, as does the lack of an Aten ray holding an ankh to the figure’s nose. A foreground arm and a torso in profile also frequently indicate that a statue is being represented (see, for example, the well-known depictions of sculptors carving a seated and a standing colossal royal statue in the tomb of Rekhmire: Davies 1944, pl. LX). A narrow band extends diagonally down and forward from the back of the table. The band may be meant to depict the support panel for the table, though fragments from actual offering-table statues show that the

140

K. THOMPSON

Fig. 4: A small statue of Akhenaten holding an offering table and standing beside a real offering table. Photograph: the author. Digital tracing by Christina King. Courtesy of the Royal Ontario Museum.

triangular panels tapered downward and back toward the royal figure’s feet. Five large Aten hands reach down from the top of the block, four of them toward the real offerings on the large table and one somewhat smaller one toward the statue’s offerings. The size of the altar suggests that the statue is relatively small, perhaps half life-size. The fragment was discovered in 1935 in the Central Halls, near the southern end of the GP (EES 35/544; Pendlebury 1951, vol. 1, 70–1). Given that this depiction of a statue has a known findspot, it is worth examining the subjects of some of the other relief fragments found in the Central Halls. While the reliefs on the walls, balustrades and columns of the vast Broad Hall seem to have been largely confined to offering scenes, those from the Central Halls display a broad range of localities and activities in the city. There were scenes with chariots and horses, on scales both small (EES 35/551; Dunedin E.27.22) and large (EES 36/__; Fig. 5, negative TA_36-37_O_

9

Online at , right-hand column, last photograph (last accessed 02/08/2018); whereabouts unknown.

Film_0058; whereabouts unknown). One block shows steps and two mooring posts from a large quayside scene (35/544; see Pendlebury 1951, vol. 2, pl. LXVIII, 8), and three blocks contain smaller-scale depictions of channels of water and riverine plants (unnumbered, negative TA-36-37_O_Film_0063).9 At least one largescale offering scene adorned the walls, as the middle block in Fig. 5 shows, though the Toronto block is clearly from a different, smaller one. A bull on another block may be in a procession, heading toward sacrifice, or part of an agricultural scene (35/535; Brooklyn 36.883). Another shows servants working or gossiping near a palace (33/38, Boston 37.658). There are also intimate scenes including the royal family. One block (unnumbered, Fig. 6, negative TA_36-37_O_Film_0060; whereabouts unknown) is reminiscent of the well-known painting of Akhenaten, Nefertiti, and their daughters (Oxford 1893.1-41 (267)). Here the slight curve under the upper foot suggests a large cushion, like the one Nefertiti sits on in the painting, while a smaller figure, either a princess or servant, stands at the left. Another intimate scene shows a princess being breastfed (36/20, Brooklyn 37.405). Musicians depicted on blocks may have been shown entertaining the royal family, including a relief of three musicians, one playing a lyre (EES 35/507, Pendlebury 1951, vol. 1, 70; 2, pl. LXVII, 4; Brooklyn 36.882). In the discussion of Roeder Block 494-VII (number 7 above) mention was made of fragments of half-life-size granite statues of the royal couple holding offering tables, found in the North Harem area of the GP. Might these have stood near altars? It is not clear whether Pendlebury’s team excavated the large, central area of the Broad Hall. Perhaps they did not, as all the finds from the Broad Hall are listed as having come from the east and south edges. There is no known evidence for altars anywhere in the GP, and Barry Kemp’s interpretation of it as a space for public ceremonies and banqueting suggests that there would have been none (Kemp 2012a, 145–6). The statue of Nefertiti in block 7 is not shown near an altar. There is no reason to think that the Toronto block depicts part of the GP; it more likely shows either the GAT or SAT, the only buildings which contain archaeological evidence for offering altars. A granite fragment depicting the lower legs of a lifesize statue of Akhenaten holding an offering table (S-8335; Fig. 7; Kemp 2014, 14) was found at the back of the colonnade of the Long Temple of the GAT. The statue might have stood near one of the many rows of altars in this part of the temple. The scene in the

EVIDENCE FROM AMARNA RELIEFS CONCERNING ROYAL STATUES AND THEIR CONTEXTS

141

Fig. 5: A 1936 photograph of three unregistered block fragments from the Central Halls of the Great Palace, from left to right: two pots, Aten hands from a large offering scene, and horses’ hooves, probably from a chariot scene. Whereabouts unknown. Courtesy of the Egyptian Exploration Society.

Fig. 6: A 1936 photograph of three unregistered block fragments from the Central Halls of the Great Palace, from left to right: the back of a wig, Nefertiti’s foot from a large relief, and a large foot on a cushion. Whereabouts unknown. Courtesy of the Egypt Exploration Society.

142

K. THOMPSON

Fig. 7: The lower legs, just above the ankle, of Akhenaten from a life-size granite statue of the king holding an offering table. The protruding broken area is the remains of the support panel beneath the table. Photograph: the author. Courtesy of The Amarna Project.

Toronto block suggests the possibility that such an arrangement existed. Lyn Green has posited that such a statue’s function would be ‘to stand in for the ruler when he or she cannot physically be present to make the offerings’ (Green 2009, 29). Representations of statuettes on altars in tombs, boundary and royal stelae Another type of statuette is depicted on some of the boundary stelae, in four tombs and on a fragment of a royal stela. This motif — a statuette of the king kneeling on the front edge of the altar and usually holding a large conical bread loaf as an offering — has generated little discussion. The lunettes of only three of the boundary stelae — A, N and U — retain enough of their reliefs to contain these statuettes, but some of the others probably had them as well (Murnane and Van Siclen III 1993, pl. 18A; Davies 1903–8, V, pl. XXXIII; Murnane and Van Siclen III, 1993, pl. 29). In each scene, the statuette kneels on the left side of an elaborate altar, while on its right side Akhenaten and Nefertiti stand holding their arms up to the Aten in adoration and two princesses shake sistra.

10

Online at (last accessed 05/07/2018); current whereabouts unknown.

The examples in the tombs show Akhenaten and Nefertiti holding up offerings rather than making adoring gestures. A well-preserved example appears at the entrance to the tomb of Apy (Davies 1903–8, IV, pl. XXXI). The two offering tables held up by Akhenaten and Nefertiti contain the cartouches of the Aten, supported by figures that may also be statuettes: two small standing images of the king as a child on his table and a seated one of Nefertiti on hers. These may, however, represent figurines of a material other than stone, though no surviving examples of such figurines are known. The kneeling statuette occurs in Parennefer’s tomb (Davies 1903–8, VI, pl. II); it is the only image where the figure’s uraeus is visible. In both the Apy and Parennefer reliefs, the statuette kneels facing away from the royal family on the opposite edge of the altar from them. This is a double offering scene, and the conical loaf also appears in the left-hand offering scene, but damage has eliminated the statuette. The same arrangement occurs in two separate offering scenes in the tomb of Mahu (Davies 1903–8, V, pls XV and XVI). A third offering scene on Mahu’s false door, presumably mistakenly, places the statuette on the same side of the altar as the family, facing the king (Davies 1903–8, V, pl. XXIII). Another variant appears in the tomb of Panehesy. There, two offering scenes include the standard placement of the statuette on the far side of the altar (Davies 1903–8, II, pls VII and VIII). On the lintel, however, there is a double offering scene centred on a single altar. The kneeling figure appears on both edges of the altar, meaning that the left-hand statuette belongs to the right-hand family group and vice versa. Finally, a small fragment of a granodiorite stela found in the Central Courts of the GP shows part of a laden altar and a clear image of the kneeling statuette (EES 35/452).10 Cyril Aldred remarked upon statuettes on altars, citing the ones depicted in the tombs of Apy and Mahu, and wrote: ‘the first examples that the writer has been able to trace are of the Amarna period’ (Aldred 1988, 46; for a discussion of later depictions of similar statuettes, see Hill 2004, 126–8). One surviving statuette with some unusual features might be an example of this type of kneeling figure (Fig. 8; Berlin 21637). The statuette kneels with arms outstretched. There was no back pillar, though there was a panel of connective stone between the missing arms and another between the feet. A conical loaf or other offering could not have been carved from the same block as the figure, since

EVIDENCE FROM AMARNA RELIEFS CONCERNING ROYAL STATUES AND THEIR CONTEXTS

143

the knees contain no broken surface. Most notably, there was no base under the figure. Instead the space between the bent legs has been carved out, and the bottoms of the legs and of the surviving curved foot are flat (see Fig. 9). The statuette has no provenance, though it almost certainly came from Amarna. Ludwig Borchardt bought it in the nearby town of Mallawi in 1914, the last year in which his German expedition was excavating at Amarna. Without its head, it is 12cm high, which makes it a plausible size to match the depictions in the reliefs. Discussion We now know that the major royal buildings in the central city, the GAT, the SAT and particularly the GP, were populated with dozens of statues, including, at least in the case of the GP, ten or eleven colossal statues and probably more.11 Why, given how many details of the buildings the reliefs show, do they depict so few statues? If standard depictions of the temples were devised early on, they might show them in an early state of decoration, before many statues had been added. The large Osiride colossi in mediocre limestone that apparently fronted the sanctuary of the GAT might have been among the first statues added, reflecting an effort to create an impressive approach during the earliest period of the building’s use. As mentioned above, one of its fragments contained an early Aten cartouche. One might also posit that artists devising the images of the temples did not have access to the innermost area, the interior of the sanctuary, where most of the statues stood. For that reason, the artists included only statues in the area in front of the temple: the colossi and the seated statue by the large benben stela. This does not explain, however, why Ahmes’ tomb uniquely did contain images of statues within the sanctuary. The question of why statues are entirely lacking from depictions of the Long Temple area at the front of the GAT is complicated. Excavations from 2012 to 2015 confirmed Pendlebury’s identification of two major building stages in this front area (Pendlebury 1951, vol. 1, 5–8) and explored the indications of these

11

The western side of the Broad Hall is under the cultivation. Given the number of foundations for colossal statues shown on the east and south sides of the plan of the Hall (Pendlebury 1951, vol. 2, pl. XIV) and assuming that the Hall was symmetrical, the intention was to erect a total of 144 such statues around the area’s perimeter.

Figs 8 and 9: Right profile and view of the underside of a kneeling limestone statuette. Photographs: the author. Courtesy of the Ägyptisches Museum und Papyrussammlung, Staatliche Museen zu Berlin.

144

K. THOMPSON

stages more thoroughly. The second stage began in or after year 12 of Akhenaten’s reign and involved raising the temple’s floor level with rubble, sealed on top by a mud-plaster floor. The rubble included a hieratic label from year 12 (Kemp 2013, 28), as well as many hardstone fragments from the earlier phase of the temple. These were apparently broken deliberately and thrown into the mixture of fill, as described by Kemp: During the cleaning, many fragments of carved stone were found, either in the material that had collapsed from the side baulks and spread over the gypsum foundation layer or actually within the baulks as they were cut back for short distances. The fact that they came from material that was the original ancient fill put down as part of the construction for the final phase of building the temple, shows that the breakage did not take place after the Amarna period. The most likely explanation is that the pieces belonged to stonework from the first phase of building which had occupied the area at the west end of what became the first court of the Long Temple. The fragments came primarily from architectural elements in travertine, indurated limestone, conventional limestone and granite. There were also inlays in granodiorite and red quartzite, mostly from a large cavetto cornice the background of which was made from indurated limestone. One or two pieces (one of them made from basalt) seemed to come from balustrades (Kemp 2013, 31).

As this description indicates, the earlier decoration that was broken up and buried was not of inferior material being replaced with better stone. A few notable pieces of statuary from the earlier phase were found, including a fragment of a life-size statue in indurated limestone, depicting Nefertiti’s stomach and hips (S-8264; Hill 2014, 14–6). This was under the mud pavement and thus undoubtedly from the earlier phase. The granite lower legs from a statue of Akhenaten holding an offering table (S-8335; see Fig. 6, discussed above) were found near the back of the foundations of the Long Temple colonnade. A considerable number of other statuary fragments were recovered during the removal of the 1930s EES excavation spoil heaps. To which phase of the temple these belonged cannot be determined. Possibly none of the new statues which might have been intended to replace the first-stage ones in the Long Temple area had been installed by the end of the reign. The depictions of this part of the GAT show the second-phase stone pylon and large rows of columns, which have been confirmed by recent excavations. Thus one might expect to see statues there if any had been installed. So far no statuary fragments

identifiably from the second phase of the front of the GAT have been discovered. Finally, the nobles to whom the tombs belonged might have been interested in depicting those aspects of the temples that most closely related to their own duties. Many of them might have had responsibilities for providing offerings for those temples, and hence the heaped offering tables that are one of the main features in the reliefs would display their work. For example, Panehesy, who supervised the slaughter of cattle for the temple, shows two slaughter-yards and numerous carcasses appear in the GAT as depicted in his tomb — more than in any other image of that temple. None of these nobles apparently was responsible for the statuary, which would have been made in royal workshops such as the one discovered north of the GP. Only Huya, who supervised the workshop of Queen Tiye and was responsible for statuary of her and her family, displays her sunshade temple as crowded with statues. Conclusion Whatever the reasons for the relative paucity of images of statues, the tomb reliefs provide a few useful clues about the placement of statues. They situate a seated statue, probably associated with known fragments, at the Stela Emplacement. They reveal where the colossi, of which only a few fragments were found dumped outside the temple precinct, had originally stood. In such cases, they confirm or clarify the archaeological evidence. The isolated blocks may hint at specific information, such as the possible existence of a group statue including Kiya, and they may offer tantalising hints concerning the spaces where certain types of statues might have been displayed. On the whole, however, it is the actual remains of the statuary that provide the most reliable information about how many statues stood in the royal buildings and about their scale and poses. In many cases their provenance attests to where the statues originally stood. Still, it is striking how few statues are depicted in the tomb reliefs. Going only by these scenes, we would never imagine there was a Great Palace containing beautiful hard-stone colossi comparable to the betterpreserved sandstone ones of Akhenaten from his East Karnak temple. We might conclude that the SAT contained no statues at all. The archaeological evidence tells us far more about the representations of statues in the reliefs than the reliefs tell us about actual statues that stood in the buildings they depict.

EVIDENCE FROM AMARNA RELIEFS CONCERNING ROYAL STATUES AND THEIR CONTEXTS

Bibliography Aldred, C. 1988. An early image-of-the-king. In J. Baines, T. G. H. James, A. Leahy, and A. F. Shore (eds), Pyramid studies and other essays presented to I. E. S. Edwards. London, 41–7. Cooney, J. D. 1965. Amarna reliefs from Hermopolis in American collections. New York. Davies, N. de G. 1903–8. The rock tombs of el-Amarna, 6 vols. London. ———. 1944. The tomb of Rekh-mi-rē‘ at Thebes. New York. Frankfort, H. and Pendlebury, J. D. S. 1933. The city of Akhenaten. Part II: The North Surburb and the Desert Altars. London. Gabolde, M. 1998. D’Akhenaton à Toutânkhamon. Lyons. Giles, F. J. 2001. The Amarna age: Egypt. Warminster. Green, L. 2009. Ten Amarna blocks in the Royal Ontario Museum. Journal of the Society for the Study of Egyptian Antiquities 36, 17–35. Hill, M. 2004. Royal bronze statuary from ancient Egypt: With special attention to the kneeling pose. Leiden. ———. 2012. The stela site: Finds. Horizon 11, 5–6. ———. 2014. Part of a statue of Nefertiti (fig. 11). In Kemp 2014, 14–7. Hill, M. and Thompson, K. Forthcoming. Royal statuary from Amarna. London. Johnson, W. R. 1996. Amenhotep III and Amarna: Some new considerations. Journal of Egyptian Archaeology 82, 65–82 and pls IV–VIII. Kemp, B. J. 1995. Outlying temples at Amarna. Amarna reports VI, 411–62.

145

Kemp, B. 2012a. The city of Akhenaten and Nefertiti: Amarna and its people. London. ———. 2012b. Tell el-Amarna, 2011–12, Journal of Egyptian Archaeology 98, 1–26. ———. 2013. Tell el-Amarna, 2012–13. Journal of Egyptian Archaeology 99, 1–34. ———. 2014. Tell el-Amarna, 2014. Journal of Egyptian Archaeology 100, 1–33. Mallinson, M. 1995. Excavation and survey in the Central City, 1988–92. Amarna reports VI, 169–215. Manniche, L. 2010. The Akhenaten colossi of Karnak. Cairo. Martin, G. T. 1989. The royal tomb at el-ʽAmarna, Vol II. London. Murnane, W. J. and Van Siclen III, C. C. 1993. The boundary stelae of Akhenaten. London. Pasquali, S. 2013. Un jardin au petit temple d’Aton de Tell el-Amarna? Égypte nilotique et méditerranéenne 6, 205–31. (last accessed 05/07/2018). Pendlebury, J. D. S. 1951. The city of Akhenaten. Part III: The Central City and the Official Quarters. Vol. 1: Text; Vol. 2: Plates. London. Petrie, W. F. M. 1894, repr. 1974. Tell el-Amarna. Warminster. Roeder, G. 1969. Amarna-Reliefs aus Hermopolis. Hildesheim. Shepperson, M. 2012. The Stela Emplacement. In Kemp 2012b, 19–26. Simpson, W. K. 1995. The inscribed material from the Pennsylvania–Yale excavation at Abydos. Publications of the Pennsylvania–Yale Expedition to Egypt 6. New Haven.

RECONSTRUCTING THE STATUARY OF THE COURTYARD OF THE TEMPLE OF KHNUM ON ELEPHANTINE Tobias KRAPF

Abstract The Ptolemaic to Roman courtyard of the temple of Khnum on Elephantine preserves unique evidence for the heterogeneous assemblage of sculptures, stelae and cult equipment that once adorned the main sanctuary of the island. During its systematic study between 2011 and 2016 over 300 fragments were catalogued. Alongside the recording of sculptures and statue bases, nearly thirty traces of the placement of statues on the floor of the courtyard itself were identified. This paper offers a first account of the assemblage of statues, which covers the chronological horizon from Thutmose II to the Roman imperial period and includes royal statuary as well as sculptures of officials, rams and lions or sphinxes. What emerges is the picture of a courtyard crowded with these diverse sculptures, some of which were obviously kept from an earlier phase of the sanctuary, and which coexisted with obelisks, altars, large offering tables and architectural elements. The arrangement of the sculptures and the existence of several tree pits underline the importance of the main axis of the yard, which was certainly used for processions to and from the sanctuary. The question that remains is who was actually allowed to enter the courtyard and consequently see all the sculptures; but it seems increasingly probable that at least on some occasions larger crowds were permitted to enter. *

*

*

The destruction, reuse and resulting poor preservation of the Late Period to Roman sanctuary of Khnum on Elephantine (Fig. 1) have offered significant opportunities for research. It has been possible not only to excavate parts of its previous architectural phases (von Pilgrim 2001) and reconstruct the building history from the Middle Kingdom onwards, but also to locate through systematic excavation some of its equipment, since the site’s poor condition did not encourage thorough clearing in the 19th and early 20th century. In five study campaigns from 2011 to 2016 over 300 fragments of statues, bases, stelae and offering tables that were once installed in the temple courtyard were

identified in the storerooms, lapidaries and the site itself. This work is based on previous studies initiated by Susanne Bickel, to whom I am deeply grateful for entrusting me with this material and for handing over her documentation. In this paper, a first account of this rich evidence is presented, establishing the picture of a mixed assemblage of sculptures and other items that decorated the courtyard of the temple of Khnum during the Ptolemaic and Roman periods. The setting The ruins of Elephantine, some of them now lost, were documented by the Napoleonic expedition, but clearing was quite limited and systematic excavation was not undertaken until the early 20th century, when the site was divided between a French (1906–11) and a German (1906–08) mission. Both concessions included parts of the sanctuary of Khnum. Interest was mostly triggered by the discovery of papyri. During this period, the ram cemetery was completely excavated (Delange and Jaritz 2013). As usual, finds were shared and many objects were sent to different French museums, including sculptures from the temple of Khnum such as a seated statue of Thutmose III now in Nantes (Musée Dobré, AF 795) and an unfinished royal statue now in Dijon (Musée des Beaux-Arts, AF 798, Delange 2012). A photograph dated to 1922 attests to the clearing of the courtyard of the sanctuary by that year. Subsequently, in the first half of the 20th century, the Middle Kingdom Heqa-ib sanctuary and its sculptures were excavated by the Egyptian inspectorate of antiquities (Habachi 1985, 15). Important topographical work at the temples of Khnum and Satet was undertaken by H. Ricke in the 1930s and 1950s (Ricke 1960). In 1969 the joint German (Deutsches Archäologisches Institut, Abteilung Kairo) and Swiss (Schweizerisches Institut für ägyptische Bauforschung und Altertumskunde in Kairo) project marked a new start for the exploration of the Elephantine island. Besides the excavation of the ancient town, important results came from the thorough exploration of the two main sanctuaries of the island, those of Khnum and Satet.

148

T. KRAPF

Fig. 1: Plan of the sanctuary of Khnum on Elephantine during the Roman period. 1) temple (Nectanebo II); 2) courtyard; 3) terrace; 4) nilometer; 5) ram cemetery; 6) Roman contra-temple. After Arnold 2005, 42, fig. 6.

The construction of the first temple of Khnum dates to Dynasty 12, probably the reign of Senwosret I, while Khnum was previously venerated alongside Satet in the goddess’s sanctuary, which can be traced back to the 4th millennium BC. The execution of the decoration of the newly established sanctuary of Khnum continued throughout the entire Middle Kingdom, when Khnum took over the role of the main deity of Elephantine (Eder 2000). A statue of Senwosret III/Amenemhat III dedicated to Khnum and found in the debris of the sanctuary of Heqa-ib most probably belongs to this temple (Habachi 1985, 113 no. 103; Eder 2000, 8). The temples of both Khnum and Satet were rebuilt in the New Kingdom. As the New Kingdom temple of Khnum was most likely erected on a level higher than that of the Ptolemaic to Roman courtyard, only foundation trenches and reused blocks, as well as some sculptures, have survived. A foundation trench between two columns, where no structural part of the architecture would be expected, indicates the presence of a statue, whereas two trenches inside the entrance are good evidence for sphinxes as part of the original layout (von Pilgrim 2001, 43, fig. 6, and 45). The construction of a new, much larger temple began under Nectanebo II. The New Kingdom temple

was dismantled beforehand, as some of its blocks were reused in the foundations of the new complex. A small provisional temple in the north assured the continuity of cult activity and was destroyed when the new precinct wall was built after the completion of the temple of Nectanebo II (Kaiser et al. 1999, 145–8). The construction was probably interrupted during the second Persian invasion, when the New Kingdom courtyard and Thutmoside pylon might have still existed (none of their blocks were reused in the foundations), and both might have been in use contemporaneously with the new temple (von Pilgrim 2001, 48), the decoration of which was continued into the Ptolemaic period (e.g. the gate decorated by Alexander IV). A pronaos was added under Ptolemy VI and VIII, while during the latter’s reign work on a new courtyard, replacing the last New Kingdom structures, was probably initiated. Bearing in mind this late replacement of the New Kingdom courtyard, the reuse of several New Kingdom statues, and especially those of Ramesses II, in the Ptolemaic to Roman courtyard is hardly surprising. Furthermore, two blocks of a gate of earlier date remained visible on the main axis of the courtyard, probably flanked by two sphinxes, as two rectangular depressions of c. 2 × 1m in the floor indicate. This central axis was also marked

RECONSTRUCTING THE STATUARY OF THE COURTYARD OF THE TEMPLE OF KHNUM ON ELEPHANTINE

by two column bases with the name of Ramesses II (Fig. 2; Junge 1987, 52 no. 4.3.1.5) and a series of trees whose pits were documented during the excavation (Kaiser et al. 1982, 311–4). The pylon and terrace in front of the courtyard are attributed to Augustus (Jaritz 1980) and the decoration of the columns of the courtyard continued until the Antonine period (Laskowska-Kusztal 1996, 10). The open space of the courtyard, which was flanked by thirteen columns on the long side and eight columns on the short, measured c. 25 × 47m and offered enough space to accommodate a large crowd, as will be argued below. The temple of Khnum was burnt during the rise of Christianity, towards the end of the 4th century AD. Its precinct was repurposed from the 5th century AD onwards for domestic use and many workshops are attested, including one for the production of stone mortars that flourished until the 9th century AD. The artisans reused the diorite sculptural and architectural fragments as raw material, a fact that explains the numerous tiny sculpture fragments found during the excavation (Arnold 2005). In the courtyard of the temple, about fifty habitation units for soldiers were built and a church was established in the pronaos. The temple itself was dismantled and used as a quarry, e.g. for the construction of the city wall of Syene in the 6th century AD (Arnold 2005, 44 and Niederberger 1999, 107–12). As the heavy and hard granite was of no interest, the roof and monolithic shrines fell into the pit where once the temple stood and remained there up to the present day. A considerable number of fragments of sculptures and bases have been found reused in the walls of those Late Antique buildings in the courtyard, such as the head and throne of a statue of Psamtek I or II (Grossmann 1980, pl. 19bd; Junge 1987, 65; Bickel 1995) and the base of a Roman bronze statue in the pronaos church (Grossmann 1980, pl. 21a). The feet of a statue of Ramesses II (Fig. 3 [in front of the rightmost column]), whose upper part was already discovered in the early 19th century and today is in the British Museum (British Museum EA 67, Fig. 4), were found during the Swiss and German excavations in house T8A, located in the courtyard, built against the pronaos (Sourouzian 1998, 281–4). The fragment K12955 of a large diorite offering table was extracted in February 2013 during the work on this same project from the foundation of the south wall of house T52, near the ram cemetery, and joined to two other fragments from the temple of Khnum found in previous campaigns (Fig. 5).

149

Reassembling the equipment of the courtyard The aim of this project is to reassemble all these fragments, as well as further objects and evidence concerning the decoration and equipment of the courtyard. More than 200 fragments of sculptures were identified that belong – certainly or probably – to the courtyard of Khnum. They can be attributed to about eighty different sculptures based on their material, dimensions and sculptural type. For instance, there are twenty-eight different back pillars. The size of the fragments ranges from a few centimetres to almost complete over-lifesize statues, and their date ranges from the New Kingdom to the Roman period. Nine of the sculptures have at least parts of their plinth preserved and more than a dozen blocks are clearly identifiable as statue bases. Some of those bases actually correspond to the size of plinths of statues and they were reassembled and reerected in front of the north colonnade (see Fig. 3). In general, these bases do not bear inscriptions. Exceptions are a pink granite bilingual base with a Greek and Demotic dedication of a priest of Elephantine, Biggeh and Philae to Ptolemy VI Philometor, Cleopatra II and their son, that has holes in its upper surface for accommodating the statues of the royal family (Maehler and Thissen 2010), as well as a sandstone base with a demotic graffito (K12912) that, however, did not belong to the Ptolemaic to Roman courtyard as it was found in the pronaos foundations. A large base for a plinth of 177.5 × 59.5cm with a rounded front part most probably served for a sphinx (see Fig. 2). Most interesting is the fact that the floor pavement preserves today more than thirty traces of statue placements (Figs 6 and 7), either as depressions (along the axis of the courtyard) or as raised surfaces of different sizes left spared at the moment of the final preparation of the floor (in front of the columns). This indicates that the placement of diverse statues in front of the columns was planned in the original design of the courtyard (Kaiser et al. 1984, 184–5). These statues, therefore, already existed. Wherever the surface in front of the columns is preserved traces of statue placements are visible. Their size varies from a few centimetres to almost 1m, ideal for accommodating either large standing or smaller striding or even enthroned statues. Five pink granite bases with holes for large bronze statues can be dated to the Roman period, thus adding five more statues to the assemblage (Fig. 8). Further fragments belong to small altars and bases without an

150

T. KRAPF

Fig. 2: Courtyard of the temple of Khnum with the gate of Alexander IV, to its right the base for a sphinx and in the foreground one of the column bases of Ramesses II and the central axis marked on the pavement. Photograph: the author.

Fig. 3: Several sculpture bases set up on the statue placements in front of the north columns of the courtyard, including the feet and plinth of a statue of Ramesses II (to the right) and an unfinished striding statue (in the centre). Photograph: the author.

RECONSTRUCTING THE STATUARY OF THE COURTYARD OF THE TEMPLE OF KHNUM ON ELEPHANTINE

151

opening for a plinth. A large such block, assembled from at least fourteen fragments, is decorated on all four sides by Psamtek II, like the aforementioned seated royal statue (Junge 1987, 67, 77–8 and pl. 40e and 46e, no. 6.3.2 and 7.1.4). Two obelisks, of which a top fragment was found, flanked the front of the pronaos (Kaiser et al. 1984, 184). The fragments of fourteen offering tables were also found in the area of the sanctuary of Khnum (Fig. 5), but only two of them – both in the shape of the nilometer – were found in situ, placed against a tree pit in the small Roman contra-temple against the back wall of the temple of Khnum (Dreyer et al. 2005, 56 n. 157) and adjacent to a platform close to the western precinct wall (Kaiser et al. 1997, 188 and pl. 29b–c). A stela from Elephantine, whose exact find-spot is unknown, shows Hathor using an offering table in front of a ram mummy (Delange and Jaritz 2013, 203–4). Indeed, five more offering tables were found in the ram cemetery (Delange and Jaritz 2013, 188–9). In a niche of the northern wall of cemetery B a sandstone statue of a seated couple was placed next to an offering table (Delange and Jaritz 2013, 52–3, pl. 21a). The graffiti of offering tables on the floor pavement of the courtyard of Khnum (Kaiser et al. 1982, pl. 66c) clearly attest to the use of such equipment in the courtyard. Additionally, a certain number of stelae can be attributed to the sanctuary of Khnum. Larger stelae might have been placed between the columns, as some traces on the floor indicate (Kaiser et al. 1984, 187, fig. 5).

A heterogeneous assemblage of sculptures Fig. 4: Upper part of a red granite colossal statue of Ramesses II. British Museum EA 67. Photograph: © Trustees of the British Museum.

Fig. 5: Offering table K12955 + K12959 + K12603. Photograph: Axel Krause.

Most sculptures were found highly fragmented, especially owing to the workshop that produced stone mortars. Some larger pieces were deliberately broken in order to be reused in later walls; for example the right leg of a striding over-life-size sculpture shows the holes for the pegs used to break it (K12512, Fig. 9). Another example is a seated statue of Thutmose II (Fig. 10), whose head was found in a later wall near the Roman terrace of the temple of Khnum (Kaiser et al. 1973, 191). The find-spot of the torso is unclear, but it is said to have been found on the site of the old museum (Kaiser et al. 1973, 191 n. 131). The restoration of the statue became possible when the base was found in 1981 in a wall of the church in the pronaos; it could be joined to the two large fragments of the

152

T. KRAPF

Figs 6 and 7: Statue placements in the courtyard of the temple of Khnum. Photographs: the author.

RECONSTRUCTING THE STATUARY OF THE COURTYARD OF THE TEMPLE OF KHNUM ON ELEPHANTINE

153

Fig. 8: Roman base for a bronze sculpture (K12615), set up on the rectangular traces in the pavement for the placement of a statue base. Photograph: the author.

Fig. 9: Leg of an over-life-size sculpture (K12510 + K12512). Photograph: Axel Krause.

Fig. 10: Statue of Thutmose II in the New Elephantine Museum. Photograph: the author.

154

T. KRAPF

throne found earlier in the area of the sanctuary of Khnum and deposited in one of the lapidaries. Furthermore, two leg fragments found in 1978 among the debris around the sanctuary also joined (Dreyer 1984, 489). The dimension of the base (0.55 × 0.97m) would actually allow this statue to be located on some of the statue placements in front of the columns of the courtyard. The sculpture shows the distinct marks made when it was broken into reusable blocks. This is evidence that the sculpture was still visible in Late Antiquity when it was broken for reuse as building material. Very different is the case of an enthroned statue of Merenptah, which is still (at the time of writing) halfburied in the modern village (Kaiser et al. 1973, 191– 2). Under these circumstances, it becomes evident that some fragments listed here might not originally have belonged to the sanctuary of Khnum but to another cult site such as that of Satet, but this cannot alter the picture developed below. From these diverse examples, it already becomes clear that the sculptures cover a long period ranging from at least Thutmose II to the Roman emperors. Furthermore, they include not only royal representations, but also animals and gods, as well as officials. The oldest statue is the abovementioned sculpture of the seated Thutmose II, dedicated by Hatshepsut, whose name has been erased by Thutmose III. A second, similar torso most probably was also dedicated by Hatshepsut but no inscription is preserved. The restored statue bears the inscription mrjj-ẖnmw, ‘beloved by Khnum’, and it must remain an open question whether the second was dedicated to Satet or also to Khnum (Dreyer 1984, 492–3). In chronological order follows the seated statue of Thutmose III that is today in the Musée Dobré at Nantes, as well as a diorite statue of Amenhotep II with a triangular kilt (Junge 1987, 36). Several statues represent Ramesses II, such as the aforementioned torso in the British Museum (EA 67) to which a pair of feet at Elephantine can be attributed. A second pair of feet in pink granite (K12602) with a ceremonial bull tail between the feet, as well as an inscribed back pillar with the name of Ramesses II (of analogous dimensions), but with no preserved plinth, belonged to a second statue forming a pair of Osirian statues with EA 67 (Sourouzian 1998, 283–4). At least two more sculptures, in sandstone, can be attributed to Ramesses II: a near-life-size standing statue found in the filling of the nilometer (K148, Junge 1987, 53–4, pl. 34a–c no. 4.3.4.1) and an enthroned statue (Junge

Fig. 11: Reconstructed statue of Psamtek I or Psamtek II. After Bickel 1995, 102, fig. 10.

1987, 54, pl. 37a–b no. 4.3.4.2). The latest New Kingdom statue is the enthroned Merenptah mentioned above, still half-buried in the modern village. Late Period activity is attested by a sculpture of Psamtek I or Psamtek II (Fig. 11), who were both militarily active in the south of Egypt. Susanne Bickel (1995) proposed that a head found in the early 20th century outside the gate of Alexander IV, and now exhibited in the Aswan Museum, may belong to a throne found much later in the pronaos church and inscribed with the name Psamtek in a cartouche. This attribution can now be further strengthened thanks to the discovery of a back pillar fragment that joins at head level and whose inscription has the same width as the one on the throne and bears signs that are of exactly the same dimensions. Five new joining fragments of the throne and the legs were similarly identified and restored during this project. There are no Ptolemaic royal sculptures except an unfinished 1.98m tall (preserved height) striding gran-

RECONSTRUCTING THE STATUARY OF THE COURTYARD OF THE TEMPLE OF KHNUM ON ELEPHANTINE

ite statue (now in the Musée des Beaux-Arts at Dijon; AF 798) whose feet and plinth were found in 1988 (Delange 2012, figs 639–42). A second, smaller unfinished statue of uncertain date was also found on Elephantine. The plinth of this second statue can be seen today positioned once more on a base in the courtyard (see Fig. 3; Kaiser et al. 1984, pl. 58a–b). Finds from the Roman period include not only the aforementioned bases of bronze sculptures, but also a pink granite royal head with the typical locks visible below the crown. It is now commonly agreed that the famous Primaportatype bronze head of Augustus found at Meroë and today exhibited in the British Museum (British Museum 1911,0901.1, Fig. 12) can be connected to Strabo’s mention of a Meroitic rebel attack on the Roman cohorts stationed in Syene involving the looting of sculptures at Syene, Philae, and on Elephantine (Strabo, Geography, 17. I. 54; see Matić 2014). Similarly to Elephantine, bases for bronze statues were also uncovered at Syene, making both places possible candidates for the origin of the head. In addition to those royal statues, and the many more specimens preserved in fragmentary state, there are a small diorite ram (Delange and Jaritz 2013, 204–5, fig. 156, pl. 74a; plinth 19 × 45.3cm) and a sandstone ram protome (Delange and Jaritz 2013, 205–6, fig. 157, pl. 74b), both of which can be directly linked to the cult of Khnum. Parts of a third ram, in sandstone, are published in Delange and Jaritz 2013, 204–5, fig. 155. The diorite ram is yet another example of an unfinished sculpture. Unfortunately, its head is missing. The back part of a lion or sphinx, also in diorite, has been identified. Both lion and sphinx are represented by several fragments in the studied assemblage, such as a small diorite sphinx with a wig (Fig. 13) and a sandstone lion that is today in the garden of the Elephantine Museum and whose exact find-spot on Elephantine, as for many sculptures, is unknown. None of these sculptures is large enough for the pink granite base of a sphinx preserved in situ in the courtyard of the temple of Khnum. The fragments of at least three sculptures of officials are datable to the Graeco-Roman period by their typical fringed clothing (see Bianchi 1978). They testify to the placement of private statues in the sanctuary of Khnum during this period and accordingly to the accessibility of the courtyard (Figs 14 and 15). These statues were manufactured in black diorite and, in one case, of finegrained grey granite. They can be compared with a similar statue of an official found in the vestibule of the temple of Tebtunis together with two Ptolemaic royal

Fig. 12: Bronze head from an over-life-sized statue of Emperor Augustus found in Meroë. British Museum 1911,0901.1. Photograph: © Trustees of the British Museum.

Fig. 13: Small diorite sphinx (K12543). Photograph: the author.

155

156

T. KRAPF

sculptures (Rondot 2004, 136–42 and pls 99–111). A series of small, mainly seated sculptures from the Elephantine assemblage might equally be connected to private offerings. Conclusions

Fig. 14: Graeco-Roman statue of an official (K12527). Photograph: Axel Krause.

Fig. 15: Graeco-Roman statue of an official (K12529). Photograph: Axel Krause.

The picture that emerges is that of a Ptolemaic to Roman colonnaded temple courtyard with a highly diverse assemblage of statues, stelae, altars and bases, offering tables, architectural relics and tree pits. Some sculptures were clearly reused from the New Kingdom courtyard, while other equipment was repurposed for the construction of its later replacement. For instance, a stela of Osorkon II was used as frame slab of one of the tree pits in the courtyard (Kaiser et al. 1982, 329– 34). Such large assemblages are normally only known from temple cachettes, as at Luxor. When all this evidence is ultimately integrated and considered together with the architecture and other evidence from the site, such as dozens of graffiti on the paved floor, as is planned for the final publication with Felix Arnold and Jitse Dijkstra, it will become possible to reconstruct the diverse activities that took place at the site, the people involved, and their perception of the place. The main axis of the processions can easily be redrawn, but the temple of Khnum on Elephantine island is a rare case, where a more complete setting of the festivals can be reconstructed, rather than one based solely on depictions and texts. According to the statuary, different groups of peoples were involved: priests, the royal court, officials and, most probably, larger crowds. In fact, those who did not have the means to pay for a proper sculpture or even a small sandstone offering table would still scratch the visual equivalent into the pavement. In the Roman imperial period, the emperors not only dedicated Egyptian-style granite royal sculptures, but they also used the temple courtyard as a ground for state propaganda through the placement of bronze sculptures. Strabo’s text underlines that the significance of these sculptures both for the Roman state and the local population, although in different terms for each, cannot be underestimated. At Karnak, a small temple for Roman imperial cult with several such bases was built directly in front of the first pylon (Lauffray 1971, 118–21). In any case, such sculptures would not have been set up there, if this had not been a highly visible place. According to Felix Arnold (2005, 41), the courtyard could accommodate up to 5,000 people. The graffiti on the temple terrace

RECONSTRUCTING THE STATUARY OF THE COURTYARD OF THE TEMPLE OF KHNUM ON ELEPHANTINE

above the Nile, attesting to the presence of stands for beer and other commodities, provide a glimpse into the festivities (K.-Th. Zauzich in Jaritz 1980, 78). Acknowledgements I am deeply grateful to the Schweizerisches Institut für ägyptische Bauforschung und Altertumskunde in Kairo and especially its director Cornelius von Pilgrim as well as Beatrice von Pilgrim for supporting my work and helping me in identifying all the fragments on the site and in the storerooms. I would also like to thank all the other researchers on Elephantine who helped me through discussions during my stays on the island and in Kairo, especially Felix Arnold and Jitse Dijkstra, as well as Uros Matić who shared with me his research on the Meroë head in the British Museum. Systematic photographic work was undertaken with Axel Krause. But it was Susanne Bickel who kindly handed this project to me and who came for this reason especially to Elephantine. Bibliography Arnold, F. 2005. Götterdämmerung in Ägypten, Auf Elephantine kann der Untergang eines pharaonischen Tempels nachverfolgt werden. Antike Welt 36.6, 39–46. Bianchi, R. E. 1978. The striding draped male figure of Ptolemaic Egypt. In H. Maehler and V. M. Strocka (eds), Das ptolemäische Ägypten, Akten des internationalen Symposion 27.–29. September 1976 in Berlin. Mainz am Rhein, 95–102. Bickel, S. 1995. La statue d’un roi Psammétique reconstituée. Bulletin de l’Institut français d’archéologie orientale 95, 93–102. Delange, E. (ed.). 2012. Les fouilles françaises d’Éléphantine (Assouan) 1906–1911, Les archives Clermont-Ganneau et Clédat. Paris. Delange, E. and Jaritz, H. 2013. Elephantine XXV. Der Widderfriedhof des Chnumtempels. Archäologische Veröffentlichungen des Deutschen Archäologischen Instituts, Abteilung Kairo 105. Wiesbaden. Dreyer, G. 1984. Eine Statue Thutmosis’ II. aus Elephantine. Studien zur Altägyptischen Kultur 11, 489–99. Dreyer, G., Bommas, M., Budka, J., Duttenhöfer, R., Jeuthe, C., Jones, J., Kopp, P., Kroll, H., LaskowskaKusztal, E., Pasternak, R., von Pilgrim, B., von

157

Pilgrim, C., Raue, D., Schaten, S., Seidlmayer, S. J. and Ubertini, Ch. 2005. Stadt und Tempel von Elephantine, 31./32. Grabungsbericht. Mitteilungen des Deutschen Archäologischen Instituts, Abteilung Kairo 61, 13–138. Eder, Ch. 2000. Einige Bemerkungen zum ChnumTempel des Mittleren Reiches auf Elephantine. Göttinger Miszellen 178, 5–29. Grossmann, P. 1980. Elephantine II. Kirche und spätantike Hausanlagen im Chnumtempelhof. Archäologische Veröffentlichungen des Deutschen Archäologischen Instituts, Abteilung Kairo 25. Mainz am Rhein. Habachi, L. 1985. Elephantine IV. The sanctuary of Heqaib. Archäologische Veröffentlichungen des Deutschen Archäologischen Instituts, Abteilung Kairo 33. Mainz am Rhein. Jaritz, H. 1980. Elephantine III. Die Terrassen vor den Tempeln des Chnum und der Satet, Architektur und Deutung. Archäologische Veröffentlichungen des Deutschen Archäologischen Instituts, Abteilung Kairo 32. Mainz am Rhein. Junge, F. 1987. Elephantine IX. Funde und Bauteile, 1.–7. Kampagne, 1969–1976. Archäologische Veröffentlichungen des Deutschen Archäologischen Instituts, Abteilung Kairo 49. Mainz am Rhein. Kaiser, W., Andraschko, F., Bommas, M., Jaritz, H., Niederberger, W., von Pilgrim, C., Rodziewicz, M., Seiler, A. and Ziermann, M. 1997. Stadt und Tempel von Elephantine, 23./24. Grabungsbericht. Mitteilungen des Deutschen Archäologischen Instituts, Abteilung Kairo 53, 117–93. Kaiser, W., Arnold, F., Bommas, M., Hikade, Th., Hoffmann, F., Jaritz, H., Kopp, P., Niederberger, W., Paetznick, J.-P., von Pilgrim, B., von Pilgrim, C., Raue, D., Rzeuska, T., Schaten, S., Seiler, A., Stalder, L. and Ziermann, M. 1999. Stadt und Tempel von Elephantine, 25./26./27. Grabungsbericht. Mitteilungen des Deutschen Archäologischen Instituts, Abteilung Kairo 55, 63–236. Kaiser, W., Avila, R., Dreyer, G., Jaritz, H., LaskowskaKusztal, E., Seidlmayer, S. and Ziermann, M. 1984. Stadt und Tempel von Elephantine, Elfter/ Zwölfter Grabungsbericht. Mitteilungen des Deutschen Archäologischen Instituts, Abteilung Kairo 40, 169–205. Kaiser, W., Avila, R., Dreyer, G., Jaritz, H., Rösing, F. W. and Seidlmayer, S. 1982. Stadt und Tempel von Elephantine, Neunter/Zehnter Grabungsbericht.

158

T. KRAPF

Mitteilungen des Deutschen Archäologischen Instituts, Abteilung Kairo 38, 271–344. Kaiser, W., Bidoli, D., Grossmann, P., Haeny, G., Jaritz, H. and Stadelmann, R. 1973. Stadt und Tempel von Elephantine, Dritter Grabungsbericht. Mitteilungen des Deutschen Archäologischen Instituts, Abteilung Kairo 28(2), 157–200. Laskowska-Kusztal, E. 1996. Elephantine XV. Die Dekorfragmente der ptolemäisch-römischen Tempel von Elephantine. Archäologische Veröffentlichungen des Deutschen Archäologischen Instituts, Abteilung Kairo 73. Mainz am Rhein. Lauffray, J. 1971. Abords occidentaux du premier pylône de Karnak. Le dromos, la tribune et les aménagements portuaires. Cahiers de Karnak 4, 77–144. Maehler, H. and Thissen, H.-J. 2010. Eine zweisprachige Weihinschrift aus Elephantine. Mitteilungen des Deutschen Archäologischen Instituts, Abteilung Kairo 66, 175–7. Matić, U. 2014. Headhunting on the Roman frontier: (Dis)respect, mockery, magic and the head of

Augustus from Meroë. In M. A. Janković, V. D. Mihajlović and S. Babić (eds), The edges of the Roman world. Newcastle, 117–34. Niederberger, W. 1999. Der Chnumtempel Nektanebos’ II. Architektur und baugeschichtliche Einordnung. Archäologische Veröffentlichungen des Deutschen Archäologischen Instituts, Abteilung Kairo 96. Mainz am Rhein. von Pilgrim, C. 2001. Stratigraphie d’un temple: Le temple de Khnoum à Éléphantine du Nouvel Empire à la période Ptolémaïque. Bulletin de la Société française d’égyptologie 151, 35–53. Ricke, H. 1960. Die Tempel Nektanebos’ II. in Elephantine und ihre Erweiterungen. Beiträge zur ägyptischen Bauforschung und Altertumskunde 6. Cairo. Rondot, V. 2004. Tebtynis II, Le temple de Soknebtynis et son dromos. Cairo. Sourouzian, H. 1998. Raccords Ramessides. Mitteilungen des Deutschen Archäologischen Instituts, Abteilung Kairo 54, 279–92.

EGYPTIAN AND CYPRIOT STONE STATUETTES IN CONTEXT AT LATE PERIOD NAUKRATIS Ross THOMAS

Abstract Naukratis held a privileged position as a controlled port of trade and base for traders from its founding at the end of the seventh century BC until the establishment of Alexandria. This study will focus on the Cypriot and Egyptian stone statuettes that dominate the figurine assemblage during this important period of Naukratis’ history. Drawing upon the hundreds of stone statuettes found during successive excavations at the site since 1884, this study will consider the context of the statuette assemblage and what this may tell us about the religious practices of different communities at Naukratis. *

*

*

Naukratis was established in the late 7th century BC and held a privileged position as a controlled port of trade and base for Eastern Mediterranean traders during the Late Period, from its founding around 630–610 BC until the establishment of Alexandria after 331 BC. Fieldwork undertaken by Petrie (1886), Gardner (1888), Hogarth (Hogarth, Edgar and Gutch 1898–99; Hogarth, Lorimer and Edgar 1905), Coulson (1996) and Leonard (1997; 2001), and since 2012, the British Museum (Thomas and Villing 2013; forthcoming; Thomas 2015e), have revealed tens of thousands of artefacts found within discrete areas of the settlement that can be broadly grouped into: ‘Greek’ sanctuaries (the sanctuaries of Apollo, Aphrodite, Dioskouroi, Hera and the Hellenion); Egyptian religious areas (the sanctuary of Amun-Ra Baded and the ‘cache of bronzes’); and the (primarily) Egyptian domestic and industrial areas of the ‘town’, the ‘south site’1 and Kom Hadid (Fig. 1). The British Museum Naukratis Project (Villing et al. 2013–19), which has been reappraising artefacts and archive material from the old

1

2

These labels were created by Petrie and Hogarth to describe specific areas of the site. I am excluding architectural sculpture, full scale or colossal sculpture of the Greek tradition, and of the Hellenistic and

excavations, has enabled the detailed study of many of these finds, including 431 stone figures and the contexts within which they were found. The catalogue of all figures (comprising free-standing statuettes or figurines and relief carved figure plaques) and sculpture from the old excavations is published with a full typological discussion in Thomas 2015a; 2015b; 2015c; and Thomas and Higgs 2017, where a full discussion of date, type and previous misinterpretations have been explored, negating the necessity to duplicate those discussions here. These can now be complemented by recent discoveries from new fieldwork by the British Museum. This study will focus on the Cypriot and Egyptian stone statuettes2 that dominate the figurine assemblage during the period that starts around c. 630–610 and finishes around 330 BC (although some forms continue to c. 200 BC). This paper will focus on about 350 of the stone statuettes that represent the main forms distributed across Naukratis, starting with 204 examples from the Egyptian assemblage, then moving to the corpus of 145 Cypriot examples. The study will consider figures made of other materials, where they inform the use of those made of stone, and the conclusion will focus on comparisons between the groups and what this may tell us about the religious practices of different communities at Naukratis. Egyptian statuettes The Egyptian carved limestone figures come from a long-lived industry from the Memphite area and are relatively common across Lower Egypt. The fine soft white Memphite limestone was perfect for carving and retaining painted decoration. Rare poor-quality limestone, siltstone, sandstone and mudstone figurines found at Naukratis possibly represent other unidentified

Roman periods (Thomas and Higgs 2017) and Egyptian sculpture (Masson forthcoming), as these represent traditions, practices and periods that are outside the scope of this paper.

160

R. THOMAS

Fig. 1: Map of Naukratis, showing location of the Nile during the Late Period, with the distribution of Egyptian figurines (green) and Cypriot figurines (yellow). Map of Naukratis by Ross Thomas, incorporating all previous fieldwork and preliminary geophysics results. © The Naukratis Project, Trustees of the British Museum.

EGYPTIAN AND CYPRIOT STONE STATUETTES IN CONTEXT AT LATE PERIOD NAUKRATIS

161

production places. Two distinct styles are represented within the assemblage. The earliest are crude plaques and figures, often with only the simplest features with traces of red and black paint. A second style of more elaborate (and diverse) free-standing figures often appears more ‘naturalistic’ in execution. The Egyptian assemblage can be broadly separated into eight groups based on subject (Table 1; Figs 2–5), a detailed typology of which is published (Thomas 2015c) and will be referred to in the following as ‘Thomas type’. There is a range of animals and fantastic creatures including a series of upright, human-headed birds and cats (‘Libyan sphinx’ or ‘Bastet sphinx’, Thomas type L-H.1.1–2). Rams, falcons and doves as well as baboons and jackals (Thomas type L-A) were associated with specific Egyptian deities such as Amun, Horus, Thoth and Anubis respectively (Thomas 2015c, 27). Crudely carved anthropomorphic figures include kneeling or crouching captives (Thomas type L-C) with bound limbs that may have been components of magic spells (Thomas 2015c, 27–8, 32, 52), as well as unidentified humanoid specimens which may also be magical ingredients, or have had another function (Thomas type L-O). Putative festival-goers are represented by a range of musicians (often nude, wig-wearing females, possibly representing the Isis-Hathor figures discussed below) playing a drum, tambour or a lyre/harp (Thomas type L-M.1–2), as well as bearded sympotic figures depicted reclining and drinking (Thomas type L-S.1). The majority of figurines depict the Egyptian deities Horus-the-child, Isis-Hathor and their protector Bes. These come in the form of ‘erotic groups’ and riders, as well as nude female and macrophallic nude male figures. Large limestone plaques depicting Bes brandishing a sword and holding a snake closely resemble the decoration on the walls of the Bes Chambers at Saqqara (Quibell 1907, 13; Martin 1981, 27–89; Jeffreys et al. 1988, 33–63), which may have been a small chapel (off the temenos wall in Saqqara) where the conception and/or birth of Horus-the-child was celebrated (Thomas 2015c, 56–7) and where numerous ‘erotic’ and phallic figurines were found.3 Brightly painted red-pink macrophallic youths were exceptionally popular figurines at Naukratis. The most common form, with eighty-one examples, and accounting for 39% of all the Egyptian figurines, is the mac-

rophallic child-god Hor-pa-khered (Horus-the-child, Thomas types L-I.1–9), with sidelock. In these playful depictions, he is shown smiling, sometimes carrying a wine amphora or a frog, or playing a musical instrument – all symbols of the inundation flood and its festivals. A smaller group of nine pieces depict a macrophallic youth engaging in sexual intercourse with his consort as an ‘erotic group’, also known as ‘symplegma’. The earliest, c. 630–400 BC, figure plaques (Thomas type L-E.1) are less common than the subsequent c. 400–200 BC more intricate examples (Thomas type L-E.2). Another popular figurine, with seventeen examples, depicts Hor-pa-khered with his distinctive sidelock (and two additional pieces where he is depicted macrophallic) on horseback (Thomas type L-R.1–2). Nude female ‘Hathoric figures’ were also popular at Naukratis (forty-nine examples). These came in four main types, all nude: woman reclining on a bed (Thomas type L-F.1),4 often alongside one or more children (presumably just born); woman giving birth (Thomas type L-F.3);5 woman standing in a shrine or kiosk (Thomas type L-F.2); woman free-standing (Thomas type L-F.4). The first two may refer to the birth of Hor-pa-khered (Thomas types L-F.1 and L-F.3), while the latter two may represent the goddess entering or leaving the birth chamber or mammisi (Thomas types L-F.2; L-F.4). However, the far earlier Dynastic tradition of producing and using figures of nude females on beds, which were associated with the cults of Isis, Mut, Hathor and Anuket, may equally relate to medicomagical practices surrounding childbirth (Thomas 2015c, 54, 61, 80). It is unclear precisely how these were used, even when a circular motif (a protective symbol?) is incised on the back (see Fig. 4). The majority of Late Period Egyptian stone figurines (as is also the case with terracottas) depict events concerning the conception – represented by the hieros gamos, the sexual unions between gods (Barrett 2011, 105, 308–9) – and the nativity (birth) of Horus-thechild alongside the representation of his protector, Bes. Such events were celebrated within specific chapels or kiosks, in the case of the conception, within the ‘place of drunkenness’ (also known as the columned porch of drunkenness’ or the ‘hall of roaming marches’: Jasnow and Smith 2010–11, 42–3; Klotz 2012, 395), and the

3

4

This chamber was, however, probably constructed in the 4th century BC.

5

Sometimes erroneously called ‘concubines’. Also known under the mistaken term of ‘baubo’.

162

R. THOMAS

Table 1: Types of Egyptian stone figure. Types of Egyptian stone figure

Date

Number

Animals Fantastic hybrid animals

(Thomas 2015c, type L-A.1–2) (type L-H.1–2)

c. 630–330 BC c. 630–400 BC

10 9

Captives Unidentified humans/humanoids

(type L-C.1) (type L-O.1)

c. 630–330 BC c. 630–330 BC

3 7

Musicians Symposiasts

(type L-M.1-2) (type L-S.1)

c. 630–400 BC c. 550–400 BC

7 3

Bes plaques

(type L-B.1–2)

c. 630–330 BC

7

(type L-F.1.1–3) (type L-F.2) (type L-F.3) (type L-F.4)

c. c. c. c.

BC BC BC BC

34 6 4 5

‘Erotic groups’

(type L-E.1–2)

c. 630–200 BC

9

Macrophallic child-gods

(type L-I.1–9)

c. 630–200 BC

81

Male youths on horses

(type L-R)

c. 550–330 BC

19

Nude females:

Reclining In shrine Giving birth Free-standing

630–400 600–250 630–400 630–200

Fig. 2: Common categories of Late Period Egyptian limestone figures, comprising (top row, from left): captives (L-C.1, Ashmolean Museum, Oxford, AN1886.462); animals (Thomas 2015c, type L-A.1, British Museum, 1886,0401.1483); symposiasts (L-S.1, Ashmolean Museum, Oxford, AN1886.461); Bes (L-B.2, British Museum, 1886,0401.1579); (bottom row, from left): unidentified humanoids (L-O.1, British Museum, 1886,0401.1491, from the Apollo sanctuary); fantastic hybrid animals (L-H.1.1, Ashmolean Museum, Oxford, AN1896-1908-G.72, possibly from a well); musicians (L-M.2.1, Ashmolean Museum, Oxford, AN1896-1908-G.71); and riders (L-R.1, British Museum, 1886,0401.1486). Photographs © Trustees of the British Museum or © Ashmolean Museum, University of Oxford, respectively. Photography by British Museum staff.

EGYPTIAN AND CYPRIOT STONE STATUETTES IN CONTEXT AT LATE PERIOD NAUKRATIS

163

Fig. 3: Examples of the most common categories of Late Period Egyptian limestone figures (c. 620 BC–300 BC), comprising: (top row, left to right): nude females reclining (Thomas 2015c, type L-F.1.1, Fitzwilliam Museum, Cambridge, E.191.1899); nude females in shrines (L-F.2, British Museum, EA 68861); nude females giving birth, or ‘baubo’ (L-F.3, British Museum, 1965,0930.954); and free-standing nude females (L-F.4, Fitzwilliam Museum, Cambridge, E.SU.147); (middle row, from left): ‘erotic groups’ (Thomas 2015b, type L-E.1, Egyptian Museum, Cairo, JE33557; L-E.2, Fitzwilliam Museum, Cambridge, E.SU.150); and (bottom row, from left): macrophallic child-gods (L-I.2, British Museum, 1965,0930.929; L-I.3, British Museum, EA 90337; L-I.5, Egyptian Museum, Cairo, JE33598; L-I.6.2, British Museum, EA 90351). Photographs © Trustees of the British Museum, © The Fitzwilliam Museum, Cambridge or © Egyptian Museum, Cairo respectively. Photography by British Museum staff, except for Egyptian Museum, Cairo.

164

R. THOMAS

Fig. 4: Limestone figure plaque of woman on a bed (Thomas 2015c, type L-F.1.1), with a child at her feet and with detail of an incised (protective?) symbol scratched on the back, dated c. 620–400 BC. Fitzwilliam Museum, Cambridge, E.191.1899. Photograph © The Fitzwilliam Museum, Cambridge. Photography by British Museum staff.

mammisi (or birth chapel: Arnold 2003, 33; Knoblauch and Gill 2012, 9–13). A single chapel may have served to celebrate both events, as both were decorated with papyriform or palm columns and Hathor-headed capitals, which is what may be depicted behind the nude female figure on numerous stone and terracotta figurines from Naukratis and elsewhere (Thomas 2015c, 56–7, 79). Egyptian figures in houses and Egyptian sanctuaries Previous excavators recognised the prevalence of Egyptian limestone figures within the town area (Petrie 1886, 22, 36; Edgar 1905, 129; see Fig. 1). Yet only 20% of Egyptian stone figurines (41 of 204) have any contextual information recorded. Of those with contextual information, 68% came from the town. This is confirmed by the assemblage of Egyptian figurines in stone and terracotta (although faience and bronze have a very different distribution), where 62% were found within the town, 25% within Egyptian sanctuaries (where faience and bronze are more commonly found, Masson 2018; Masson-Berghoff 2019) and 13% within

6

For example: ‘large number of indecent types … but a discussion of their types is profitless’ (Gutch 1898–99, 95, nos 271– 80); ‘the site […] has produced a great quantity of indecent statuettes of various ages and in various materials’ (Edgar 1905,

Greek sanctuaries. Egyptian figurines accounted for 92% of all figurines found within the town. Within Egyptian religious spaces (comprising the sanctuary of Amun-Ra and the cache of bronzes), 100% of the figures (in all materials) were of Egyptian types, although only the specific forms were found there and the assemblage is dominated by faience and bronze material (see Masson 2018, 87, 91; Masson-Berghoff 2019, 127–40). The poor contextual information available for these figurines is due to two factors. Firstly, the contexts in which they were found (mainly Egyptian houses) were not the research focus of the previous excavators, who were concentrating on Greek and Egyptian sanctuary areas. Many were acquired, or found unstratified, so the excavators had no knowledge of their original findspot, and even when a domestic context was excavated, there was little interest in recording or publishing the details of houses. Secondly, there is a complete absence of recorded contextual information for limestone phallic or erotic figurines (Thomas 2015c, types L-I and L-E) by Petrie, Gardner, Hogarth and anyone working for them. These were labelled ‘indecent’,6 and the excavators seemed too embarrassed to mention these

130); ‘squatting indecent terracotta, but without head or end of penis’ (Hogarth Diary 24–25 April 1903); ‘Sebakh stuff – including bearded head, & realistic circumcised penis’ (Hogarth Journal, 26 April 1903).

EGYPTIAN AND CYPRIOT STONE STATUETTES IN CONTEXT AT LATE PERIOD NAUKRATIS

figures in their publications and even in their notebooks. This is remarkable considering that this is the single largest group of figurines found at Naukratis and there is a distinct male gender bias present within the assemblage (male 57%, female 34%, animals and other 9%). Only a single phallic limestone figure had its ‘town’ context recorded, and that is because it was from a more recent survey (Coulson 1996, 139–45, no. 20, pl. XVII-6, where it was misidentified as female, because the phallus had broken off). Indeed, many pieces were not even studied until 2012 (Thomas 2015b, 6–9). It is, however, clear that numerous phallic terracotta figures were discovered in the town, including examples found during the recent British Museum excavations. The remaining types depicting nude females (which were not considered ‘indecent’), riders and animals (40 of 109 examples) did not suffer such discrimination. Better contextual details were recorded for the nude female figures,7 particularly limestone female figure plaques concerning childbirth (Thomas 2015c, types L-F.1 and L-F.3). These seem to have been found in two distinct contexts: firstly with votive offerings (especially faience amulets and bronze reptile cases) buried in pits excavated in 18858 outside the western temenos wall of the Amun-Ra Baded sanctuary and in a putative Egyptian chapel (Masson-Berghoff 2019, 140–54) next to the northwest corner;9 and secondly in mud-brick buildings in the southern site town area close by,10 as found across the town by Petrie11 within mud-brick structures (probably houses,

7

8

9

‘They are figures in white limestone, rudely carved, or merely blocked out; one is a human headed cat (?) seated & the other a standing female figure, with a pilaster generally up the left side & nothing on the right. They are always very rough… There are also two seated figures with a drum on the knees of the same style’ (Petrie Journal, 13 January 1884, 74–5). For a full discussion of this context and the possible location of chapels or shrines in this area see Masson-Berghoff 2019, 140–54. For example in area ‘24b’: ‘outside enclosure wall on W[est] side …. nude G[ree]k female figure of the type described by M. W. Petrie in his last report’ (Griffith Journal, 7 January 1885; the drawing actually depicts Egyptian type Thomas L-F.1.1). Late he reports ‘today in 24a at a depth of about 14 f[eet] a stone female figure larger than usual broken’ (Griffith Journal, 9 January 1885). Site 24a is described as a pit. Hogarth also found a large number of bronze, faience and four (and possibly more) reclining female limestone figures (presumably type L-F.1) just outside the pylon (Masson-Berghoff 2019, 141–2). ‘In the main dig, finished the chamber which abuts on Great Wall face … Bits of early Nauk. wares - & several bits stone

165

see Fig. 1). Few limestone figures were placed within Egyptian sanctuaries, shrines and caches (and along their boundary walls), but faience and bronze amulets were abundant in these contexts. The British Museum excavations in 2015–16 revealed on the riverfront stone, terracotta and wooden phallic, rider and female figures from 5th- and early 4th-century BC contexts (Thomas and Villing forthcoming). As these were found alongside domestic pottery, they may have originated from domestic dumps. Alternatively these may have been intentionally deposited within the river, perhaps during inundation festivals, as many such figures were found within the river deposits at Mendes (Redford, Sternberg-el-Hotabe and Redford 1991, 67–9; Redford 2004, 129–30). Four distinct patterns can be seen within the Egyptian limestone figurine assemblage within three zones of Naukratis: the Egyptian sanctuaries; the Greek sanctuaries; and the Egyptian houses and general town area.12 1. Egyptian stone figures were rarely found within Greek sanctuaries, and those few that were are unique and of unusual and sometimes unidentifiable types (Thomas type L-O, see Fig. 2) not found elsewhere at Naukratis (with the exception of rider type L-R.1, discussed below). 2. Stone Egyptian figurines were rarely found within Egyptian sanctuaries or caches, although they were sometimes found within ritual deposits immediately outside of sanctuaries. Those that were found inside

10

11

12

figures (4 frags. child births) from 1 ft. down’ (Hogarth Journal, 3 May 1903). Also he describes a group of chambers … (by )… Great Wall … Baubo figure – complete (Hogarth Journal, 6 May 1903). Alternatively these chambers and walls may belong to houses. ‘Pits E(ast). of main S(outh). site … Evidently shop & house remains … 3 small stone heads… 1 childbirth (1/2)’ (Hogarth Journal, 6 May 1903). Hogarth possibly mentions an ‘erotic figure’ from this area, although it is not clear if it is of limestone or terracotta, or even if it comes from the excavations undertaken that day within a: ‘group of chambers … so far the stuff looks like houses … 1 bit symplegma relief’ (Hogarth Journal, 5 May 1903). Thus part of the south site was made up of domestic structures, whilst others areas had Egyptian religious functions. Petrie discusses the contents of the houses: ‘and two or three more shapely figures with disc-shaped heads, long hair, and the hands folded over the stomach’ (Petrie 1886, 36; may refer to Ashmolean Museum, Oxford, AN1886.453). I am excluding the cemetery as no Egyptian or Cypriot figures are known from the cemetery of Naukratis to date.

166

R. THOMAS

sanctuaries usually depict specific animal or mythological hybrid animal types associated with Egyptian cults (Thomas type L-A; L-H, see Fig. 2), or are rare high-value pieces (Thomas type L-F.4, see gilded example in Fig. 5). 3. Most Egyptian forms were only found within the town (sometimes we know they were specifically found within houses), down wells (within the town area), or on the river bank within domestic rubbish or within river deposits. These comprise all figurines representing the conception and birth of Horus-the-child and the associated inundation festivities (Thomas types L-I; L-E; L-F.1–3, see Fig. 3; L-R; L-M, see Fig. 2). 4. The only Egyptian limestone figurine type found in all three zones (Egyptian religious contexts such as the cache of bronzes, the Greek Hellenion sanctuary, in the town and down wells) were rider figurines (Thomas type L-R.1, see Fig. 2). These seem to have been used differently by distinct communities in Naukratis. Cypriot figures It was a Cypriot alabaster figure, offered to Petrie in 1883, that led to the discovery of Naukratis (Thomas 2015b, 2; Fig. 6, N-1.4b). The limestone and (less frequently) alabaster Cypriot figures account for 145 of the c. 271 Cypriot figurines found at Naukratis (the remainder were terracotta), receiving much scholarly attention (Edwards 1885, 262; Petrie 1886, 13–4, 36; Gardner 1888, 55–9; Smith 1892, 103–27; Gutch 1898–99, 67–97; Pryce 1928, 183–4; Möller 2000, 157–60; Jenkins 2001, 166–73; Nick 2006, 59; Höckmann 2007, 176–7; Thomas 2015b).13 Scientific analyses suggest the limestone is Cypriot, probably from quarries around Idalion and Larnaka (Thomas 2015b, 2, 11). The alabaster has been positively identified as gypsum alabaster (calcium sulphate dihydrate) which is not available in Egypt, and probably from Cyprus (Jenkins 2001, 166–7, 173; Höckmann 2007). Although some scholars in the past have suggested

13

The comprehensive typologies developed by Nick (2006) and Höckmann (2007) are used here in a revised form published in Thomas 2015b, which incorporates recent revisions of dating based upon excavations undertaken at Samos, Knidos and Miletos. They will be referred to in the following as ‘Nick type’ and ‘Höckmann type’.

Fig. 5: Limestone figure of nude standing woman or goddess, with painted and gold leaf decoration (Thomas 2015c, type L-F.4, British Museum, EA 68814), dated c. 400–200 BC. Photograph © Trustees of the British Museum.

EGYPTIAN AND CYPRIOT STONE STATUETTES IN CONTEXT AT LATE PERIOD NAUKRATIS

167

Fig. 6: Main Cypro-Archaic figure types as published in Nick 2006 (N-) and Höckmann 2007 (H-), expanded in Thomas 2015c. Female forms: standing (N-2.1a, British Museum, 1886,0401.1386; N-2.1b, Head N15.10.5.10 and feet N15.10.5.2 from British Museum excavations of Dioskouroi sanctuary); sitting (N-3.1, British Museum, 1888,0601.31). Male forms: standing (N-1.1, Petrie Museum, London, UC16472); striding (N-1.4a, Petrie Museum, London, UC16469); sitting (N-1.6, British Museum, 1888,0601.32); and musician (N-1.2, British Museum, 1888,0601.28). Alabaster striding form (N-1.4b, British Museum, 1886,0401.1382). Alabaster kouros (hero?) with lion (N-1.5, British Museum, 1886,0401.1381). Kouroi in alabaster striding (H-1.3, Egyptian Museum, Cairo, CG27425); ‘adorant’ form (H-2.3, British Museum, 1888,0601.14); wearing mantle (H-3.3, British Museum, 1886,0401.1490). Alabaster vessels: kore alabastra (H-5.3, British Museum, 1888,0601.15); tripod cosmetic dish (H-5.4, British Museum, 1886,0401.1508). Animals: lion (N-4.1, British Museum, 1888,0601.33); hawk (N-4.2, Bolton Museum, 1966.114.A); group with bull and man (N-3.2, British Museum, 1886,0401.1390). Photographs © Trustees of the British Museum, © The Fitzwilliam Museum, Cambridge, Bolton Library and Museum Service, © Egyptian Museum, Cairo or © Petrie Museum of Egyptian Archaeology, UCL respectively. Photography by British Museum staff, except for Egyptian Museum, Cairo.

168

R. THOMAS

Cypriot (or Cypro-Ionian) figurines were made in Naukratis, there is no evidence that either limestone or alabaster figurines were made there (see discussion in Thomas 2015b, 11). They display a mixture of Assyrian, Phoenician, Egyptian, Persian and Greek (Aegean or Ionian) features, which is not surprising considering their distribution across Cyprus and the Eastern Mediterranean, including Samos, Rhodes (particularly Lindos, Kamiros and Vroulia), Knidos, Delos, Chios, Ephesos, Miletos and the Phoenician cities of Byblos, Amrit/Marathos and Sidon (Thomas 2015b, 10–1). The majority of these limestone and alabaster figures date between c. 610 and 525 BC (Thomas 2015b, 4, 10; see Table 2), comprising (earlier) ‘Egyptianising’ and ‘Cypriot’ styles of 610–570 BC and the later ‘CyproIonian’ style dated c. 600 or 580–525 BC (also known as the ‘Aegean class’), which reflects changing fashions comprising the increasingly prevalent depictions of youth (either kouroi or korai) and the increased use of alabaster over time. The Cypriot figurines usually represent both female and male worshippers, often finely painted, sometimes with vermilion, including red and black paint depicting the eyes, lips, hair, clothing designs and jewellery, and usually carrying objects for dedication or animals to sacrifice. Animals, riders, chariots, figure groups, fantastic creatures and deities are rarely represented. The assemblage includes six main types (Figs 6–7; Table 2). Table 2: Types of Cypriot stone figure. Types of Cypriot stone figure

Date

Number

Cypro-Archaic limestone figures of women

c. 610–525 BC

53

Cypro-Archaic (mainly) limestone figures of men

c. 610–540 BC

27

Cypro-Archaic limestone figures of animals and groups

c. 600–550 BC

9

Cypro-Ionian (mainly) alabaster kouroi and heroes

c. 600–525 BC

37

Cypro-Ionian alabaster korai figure vessels and cosmetic dishes

c. 580–525 BC

10

Cypro-Classical limestone figures14

c. 510–300 BC

9

14

Female figure plaques usually have unworked backs and are depicted wearing a chiton, with long hair, most frequently (but not always) covered by a veil lifted above the forehead and falling behind the ears down the back. They wear earrings, bracelets and one or two necklaces with a central medallion. They are typically depicted with one arm crossing the torso and holding an offering: a bowl, saucer, lamp, bird, lotus flower, miniature calf, tambour or disc. They come in three (overlapping) chronological type styles. The earliest groups (Nick type 2.1a, dated 610–575 BC; Nick type 2.1b–d, dated 600–550 BC) were common in the Aphrodite, Apollo (and the adjacent ‘town’ area) and Dioskouroi sanctuaries. However, the latest variant (Nick type 2.1e, dated 560–525 BC) was only found within the Aphrodite sanctuary. Rare female figure variants include one kneading dough (Nick type 2.2) and the enthroned kourophoros, thought to be a Cypriot adaptation of the traditional Egyptian iconography of Isis-Hathor nursing Horus (Nick type 3.1). A small group of complex figure groups including a bull with two attendants, riders, and chariot groups as well as a range of animal figurines in the form of lions and falcons (Nick types 3.2 and 4) were found in both the Apollo and Aphrodite sanctuaries. Male figures were depicted standing, striding, or (rarely) seated. The standing form has three variants: plain, musician with a lyre and a ram-bearing youth. The most common are of standing bearded males sporting a conical cap with cheek-guards or other headdress over long hair, and wearing a chiton (a simple tunic or gown-like garment), covered by a tasselled mantle or wrap (sometimes called a himation, following the Greek term for the garment) crossing diagonally from the left shoulder to the right knee (Nick type 1.1; type 1.7; seated variant type 1.6). They were found in both the Apollo and Aphrodite sanctuaries. A later variant dated c. 575–550 BC comprises two young male lyre players wearing a long chiton covered by an ependytes (a thigh-length over-garment, see Nick type 1.2) and was also found in both the Apollo and Aphrodite sanctuaries. Striding figures, often depicted wearing an Egyptian ‘apron and kilt’ and a Cypriot cap, are dated c. 600–560 BC (Nick type 1.4; 1.4.2),15 and have been found within the Apollo sanctuary, while ram-bearing

This refers to a later stylistic group that are generally of CyproClassical date, but were produced and used over a longer period 15

than scholars generally apply to the Cypro-Classical period (of 475 or 450BC until approximately 300BC). Two examples carved in alabaster have no provenance.

EGYPTIAN AND CYPRIOT STONE STATUETTES IN CONTEXT AT LATE PERIOD NAUKRATIS

169

Fig. 7: Main Cypro-Classical limestone figure types, comprising: temple boy (A, Ashmolean Museum, Oxford, AN1886.457; B, Ashmolean Museum, Oxford, AN1896-1908-G.1042); female standing figures (C, British Museum, 1911,0606.6; D, City Art Gallery & Museum, Bristol, H446.c) and inscribed base from a statue of Herakles by Sikon from Cyprus (E, British Museum, 1900,0214.22). Photographs © Trustees of the British Museum, © Ashmolean Museum, University of Oxford and © Bristol Museums, Galleries & Archives respectively. Photography by British Museum staff.

variants (Nick type 1.3.1) were found in the ‘town’ next to the Apollo sanctuary. The ‘hunter’, a later 575– 540 BC variant of this striding form – wearing apron, kilt and Cypriot cap (Nick type 1.3.2) and carrying hares and pigs – was found near the altar within the Aphrodite sanctuary. Cypriot kouroi appear closely related to Greek sculpture and were perhaps made for a Greek market. Some examples preserve traces of their original paint: reddish-brown on the body, vermilion on lips and hair band, and black on hair, eyes, eyebrows, moustache and pubic hair. Distinct carving styles have been labelled ‘Egyptianising’ (group 1; 600–575 BC), ‘Cypriot’ (group 2; 580–570 BC) and ‘Cypro-Ionian’ (group 3; 580–550 BC and group 4; 550–525 BC), although the cultural labels represent chronological features of fashion, rather than distinct cultural groups. Striding kouroi (twenty-eight examples) came in three distinct forms, all four (broadly chronological) styles and in both limestone and alabaster materials. The earliest styles (groups 1 and 2), c. 600–570 BC, were only

made in the simplest kouros form (Höckmann type 1) with arms flat against his side. These were made of limestone in 80% of cases and only attested in the Apollo sanctuary. Kouroi carved in the later 580–525 BC (groups 3 and 4) style came in a wider variety of forms including the ‘adorant’ form (Höckmann type 2) with one arm across his chest, and the ‘mantle kouros’ (Höckmann type 3), wearing thin, closely fitting clothes as well as the simple form (Höckmann type 1). Sculptors carving in this later style preferred alabaster 82% of the time. These later forms were found in the Aphrodite sanctuary and in the ‘town’ area next to the Apollo sanctuary. Four finely carved kouroi holding lions (Nick type 1.5) in front of them (two in limestone and two in alabaster), often considered representations of a hero (possibly Herakles) were found in the ‘town’ (near Apollo), Aphrodite sanctuary, and one at the ‘south site’ next to the Great Temenos of Amun-Ra. Speculation on the significance and meaning of the kouroi at Naukratis have been overstated and based on incomplete or erroneous contextual or provenance

170

R. THOMAS

information. This study confirms they represent at most 26% of the assemblage16 and the new contextual information presented here suggests they were widely distributed across the site.17 A related group from this 580–525 BC ‘CyproIonian’ alabaster industry comprises both korai18 alabastra (vessels in the form of a woman) and tripod cosmetic dishes with zoomorphic or sphinx feet (Höckmann type 5; style group 3-4). Both were attested in the Apollo and Aphrodite sanctuaries. The final group comprises the disparate group of nine fragments from Cypro-Classical limestone sculptures (see Fig. 7): (up to)19 five female standing figures representing the continuation of the Archaic form discussed above; (up to) three crouching child (so-called ‘temple boy’) statuettes; and the base and feet of a Classical-style sculpture of Herakles. Although limited in number, these few finds confirm that Cypriot figurines continued to reach the Aphrodite, Apollo and Hellenion sanctuaries into the 4th century BC. This brief description highlights the abundance of Cypriot stone figures reaching all the Greek sanctuaries from the earliest phases of Naukratis throughout the Late Period.

In stark contrast to the Egyptian figures, a remarkable 63% (91 of 145 total) of the Cypriot stone figurines have good contextual information recorded. Of those with context, 88% were found within Greek sanctuaries (fifty-three Aphrodite; twenty-three Apollo; two Hellenion; two Dioskouroi). The remainder probably came from disturbed contexts (ten ‘town’; one ‘south site’), for the ‘town’ context given by Petrie was an area of disturbed deposits immediately to the east of the Greek temples. It is quite likely that many objects found by Petrie in this ‘town’ area originally came from the Apollo (or another adjacent Greek) sanctuary, as only the lowest strata within the Apollo sanctuary were left undisturbed before excavations commenced. The example Hogarth found in 1903 within the large area of Naukratis that Hogarth called the ‘south site’

has no further details. It is possibly from a disturbed deposit that related to the Aphrodite sanctuary. The gender pattern for Cypriot limestone figures is roughly equal, and it remains balanced when considering Cypriot figurines in all materials (limestone, alabaster and terracotta, where 44% were of women, 43% men, 2% baby boys and 11% animals or fantastic creatures). Both male and female figurines were found in sanctuaries of both male and female deities, and all the major types were represented in both the Aphrodite and Apollo sanctuaries. There are, however, some gender biases within the assemblage found within each sanctuary. For example, theres is a complete absence of early kouros forms from the Aphrodite sanctuary, which seems unusual given their popularity within the Apollo sanctuary and considering the fact that the Aphrodite sanctuary produced the most Cypriot figurines and was the best preserved. It should also be noted that while female figure plaques were the most common figure type in all Greek sanctuaries, there was a greater variety of male figure forms found at Naukratis. Female figures (and figure vessels) account for 70% of stone figurines from the Aphrodite sanctuary and only 45% at the Apollo sanctuary. However, in both sanctuaries, the terracotta figure gender profile seems to have balanced out the overall male–female ratio. Partial contextual information can be reconstructed from the find year (when compared with the pattern of types found in each sanctuary) for some of the fiftyfour without context. Twenty came from the Egypt Exploration Fund excavations of 1884–86. Of these, five came from the second season, and were probably found within the Aphrodite sanctuary, as all other Cypriot figurines found that season were found within that sanctuary (and this may explain why 60% of them are female figures). The nine that came from the first season were probably found in the Apollo sanctuary or the ‘town’ area immediately to the east (like all others found that year), which may explain why 67% of these were male. Of the eleven found during the 1899 and 1903 seasons, no specific context can be ascertained, as Hogarth was certainly acquiring pieces and finding them within the rubbish heaps in the ‘centre of the

16

18

Cypriot figurines and Greek sanctuaries

17

Only half of these have find-spot information and nine have uncertain or dubious provenances. Kouros figurines found or acquired in Egypt are often assumed to have come from Naukratis. See Höckmann 2007.

19

No definitive korai figures have been found in context at Naukratis (see putative example in Höckmann Type 4.4; Bonn, Akademisches Kunstmuseum B 173). It is not always easy to distinguish with this late style some fragments of temple boy figures from the female figures.

EGYPTIAN AND CYPRIOT STONE STATUETTES IN CONTEXT AT LATE PERIOD NAUKRATIS

town’,20 near all the main Greek sanctuaries. The specific context of Cypriot figurines is explained below for the Dioskouroi, Hellenion, Apollo and Aphrodite sanctuaries. The Dioskouroi sanctuary was discovered by Petrie and excavated in 1884–86. However, only 123 objects from these excavations have this provenance, including only a single terracotta figurine vessel. Since 2014, British Museum excavations have revealed part of the 6th-century BC sanctuary, including a small mud-brick platform, next to which two Cypriot female figure plaques, dated c. 610–560 BC, were found with three inscribed dedications to the Dioskouroi, burnt animal remains of pig, sheep, goat and cattle, as well as Egyptian, Greek, Cypriot and Phoenician ceramics (Thomas and Villing forthcoming). The ashy context is currently tentatively dated 550–500 BC on the basis of associated finds and is thought to represent the final Saite phase of the sanctuary, directly beneath the early 5thcentury BC substantial rebuild of the immediately adjacent Hellenion gate. Because this is clearly a (disturbed) secondary or tertiary deposit, there is a limit to what this context can tell us, other than that Cypriot figurines were present within the Dioskouroi, as they were in the Apollo and Aphrodite sanctuaries. The small sample of figurines now known from this context totals just five (two Cypriot, two Greek and one Egyptian). The Hellenion sanctuary revealed 115 figurines in all materials (5% were Cypriot, 69% Greek and 26% Egyptian) of which only two are Cypriot stone and four are Cypriot terracotta figures. The Hellenion was the centre of extensive East Greek dedicators’ activity, including dedication to the ‘Gods of the Greeks’, a very early representation of a common Hellenic identity (Höckmann and Möller 2006, 18). British Museum excavations in 2014–16 revealed the western gateway and a c. 1m-high casemate platform over which all the sanctuaries were built. However, all figurines known from this structure were discovered by Hogarth in 1899 and 1903. Cypriot limestone sculpture was limited to two Cypro-Classical figurines: a c. 450–400 BC crouching temple boy and a 4th-century BC sculpture of Herakles, with the inscription ‘Sikon from Cyprus

20

171

made (and) Aristion (dedicated) to Herakles’ (Johnston 2015, 6). Three Egyptian limestone rider figures (Thomas L-R.1) were found in this area, including one with the inscription ‘Nymph... I belong to...’ (Johnston 2015, 6), suggesting this common Egyptian Horusthe-child representation was used (and understood?) in a very different way in this ‘Greek’ context to how it would have been used in an Egyptian house. The absence of early Cypro-Archaic forms within the Hellenion may relate to preservation conditions (the Archaic levels seem to have been disturbed and buried under the substantial 5th-century BC construction) and excavator’s bias (Hogarth concentrated on the more productive Persian period levels, leaving, perhaps by accident, Saite levels that the British Museum were able to excavate). Yet a clear pattern emerges with an abundance of Rhodian, Ionian and locally made terracottas (in both Greek and Egyptian forms), particularly the sixty-four exclusively female protome busts (Thomas 2015d, 7) associated with dedications to Aphrodite Pandemos within one area of the Hellenion. This confirms a significant ritual pattern within the Hellenion which is different from that of the other Greek sanctuaries of Naukratis. The Apollo sanctuary (Fig. 8; Table 3) revealed fifty-five figurines in all materials (56% were Cypriot, 11% Greek and 33% Egyptian), of which twenty-three were Cypriot stone and eight Cypriot terracotta figures. When Petrie excavated the Apollo sanctuary, he identified a sequence of construction layers, surfaces and votive pits from c. 620 BC through to the Hellenistic period (Thomas 2015b, 5; see Fig. 8). The corrected drawing of the section and plan of the Apollo sanctuary (shown in Fig. 8), can add some additional information for a reappraisal of the temple sequence discussed by Koenigs (2007, 327, 340–1) and the stratigraphy discussed by Villing and Thomas (2015, 5; Thomas 2015b, 5–7). I now suggest the following sequence: 1. The sanctuary was active by 620–600 BC; the votive offerings of this phase are represented within a (series of) votive pit(s) called the ‘trench with bowls’. This trench21 is reopened periodically during the life of Temple 1.

Hogarth Notebook, 29 April 1903: ‘The rubbish heaps (in the centre of the town) … 1 Cypriote stone figure’; and on 25 April 21

1903: ‘sebakhin’s stuff … curious little adoration scene? & bit of early statuette’. Or at least a number of pits in the same location.

172

R. THOMAS

Table 3: Matrix of Apollo sanctuary with distribution of Cypriot figure types per phase. Date of context

Bothros

500–330 BC

530–510 BC

AΠ1–2

CyproIonian vessels

1

5

5

1

2

Temple 1

2. Temple 1, an Ionic temple with limestone features erected c. 560–550 BC. The foundation for this was cut through sterile ‘basal mud’ and built on a sandbox foundation (AΠ 190 and AP20);22 and remains of the limestone rubble (AP30, AP40, AP50, see Fig. 9) from its creation and subsequent robbing out23 (prior to the construction of Temple 2) are all that remains.24 3. Temple 2, an Ionic temple with Ephesian marble features erected c. 530–510 BC. Of particular relevance is the stratified sequence recorded by Petrie of five fills within a single trench dug through the sterile soil on which the first limestone temple of Apollo was built. Petrie’s ‘trench with bowls’

24

CyproIonian kouroi

AΠ5

620–590 BC

23

CyproArchaic animals

AΠ6–7

AΠ3–4

22

CyproArchaic men

Temple 2

620/600–560 BC

>620 BC

CyproArchaic women

AΠ8

575–525 BC

580–550 BC

surface

The latest material found within this sand deposit is dated 570– 560 BC (British Museum 1886, 0401.1299), closely matching Koenig’s date for the construction of Temple 1. There are currently no finds identified from stratum AP3, AP60 or below that can be dated to later than 560–530 BC. However, identifying the precise depth of the robber pit, based on the state of the archives and the small number of objects with context numbers, is unlikely to be decisive. Petrie assumed (1886, 13) that Temple 2 was built immediately over Temple 1 because of the identification of fragments in that location (1886, 13). If this were the case limestone fragments from the construction of Temple 1 would not have been found to the east at a greater depth than the sterile ‘basal mud’ layer that covers the whole of Naukratis at approximately the same level (Pennington and Thomas 2016). The deposits to the east

is a series of votive pits dug and re-excavated both before and during the life of Temple 1. A pit (or trenches) filled with cleared votives and other ‘sacred rubbish’ is also known as a bothros, or a favissa.25 Many Cypriot figurines were found within votive pit deposits numbered AΠ1–5 that relate to activity associated with the first temple of Apollo of the Saite Dynasty (620–525 BC). Of these deposits AΠ3–4 contained all of the (exclusively limestone) figures (that we have specific contexts for) in a highly varied group comprising five Cypro-Archaic female figures, five CyproArchaic male figures, two Cypro-Ionian kouros figures and one group depicting a bull being led to sacrifice. This group excluded alabaster figures and figure vessels, which were only found26 above a hard clay

25

26

can only be explained by the foundation and subsequent robbing pit of a limestone structure (although this could conceivably be another structure using the same material as Temple 1). The presence of limestone rubble from Temple 1 to the west, under Temple 2, merely represents the recycling of architectural rubble and elements from Temple 1 within the foundation and superstructure of Temple 2. The relocation of Temple 2 may have been to make it more visible from the river, which migrated over 80m to the west during the occupation of Naukratis (Pennington and Thomas 2016). See the discussion and definition of these terms, with bibliography, in Osborne 2004; Lindström and Pilz 2013, 268–9, 274; Thomas 2015b, 5. Alabaster korai alabastron fragments (dated 575–550 BC; Höckmann Type 5; British Museum 1886,0401.1393).

EGYPTIAN AND CYPRIOT STONE STATUETTES IN CONTEXT AT LATE PERIOD NAUKRATIS

173

Fig. 8: Apollo sanctuary plan and section (dotted line on plan). Note: the scale of depth for the section is exaggerated in order to display detail. Drawing by author, based on Petrie (1886, pls 41 and 44), but corrected for actual location.

174

R. THOMAS

Table 4. Matrix of Aphrodite sanctuary with distribution of Cypriot figure types per phase. Date of context 400–200 BC

West Φ13

South

CyproArchaic animals

CyproIonian kouroi

CyproIonian vessels

Φ14

Φ11

Φ9

Φ8

Φ10

8

Φ7

Φ4

6

1

1

3

Temple 2 Φ3

Φ6

Φ5

630–525 BC Φ1 >630–600 BC

CyproArchaic men

Temple 3

>500 BC 630–520 BC

CyproArchaic women

North

Φ12

>320 BC 500–300 BC

East

Φ2

ΦA ΦB

1

1

2

Temple 1

surface and further layers that covered the area (AΠ6– 7) over which the second Apollo temple was built of Ephesian marble in c. 530–510 BC. A single CyproClassical female figure plaque dated c. 425–375 BC (Bristol, City Art Gallery & Museum, H446.c) was probably found within the Apollo sanctuary, although the precise find-spot was not specified. The sanctuary of Aphrodite (Fig. 9; Table 4) was excavated stratigraphically by Petrie in 1885, during the second season of excavations funded by the Egypt Exploration Fund, which was directed by Gardner. The Aphrodite sanctuary revealed 167 figurines in all materials (77% of which were Cypriot, 11% Greek and 13% Egyptian),27 of which fifty-three were Cypriot stone and seventy-five Cypriot terracotta figures. Substantial unpublished contextual information is recorded in Petrie and Gardner’s notebook, including two lists of contexts, many of which relate to context numbers written on the objects (including twenty-four of the Cypriot limestone figurines), or on forty-seven of the eighty-two boxes within which objects were shipped from Egypt in 1886. Many of the fifty-three Cypriot limestone figures and 128 Cypriot terracotta figures

have these contexts recorded on them. The stratigraphic sequence is not refined, but does provide us with seven phases (see Table 4). First the surfaces (Φ1–2) over which Temple 1 was built sometime after 625–600 BC, then the use and modification of Temple 1 (Φ3-7, ΦB) over the course of the 6th century BC. Then Temple 2 was built around 500 BC, abutted by contexts containing 6th-, 5th- and 4th-century BC pottery (Φ8–11). Finally Temple 3 was built, probably during the 4th century BC, which was abutted by fills of the 4th century BC (Φ12–14). Ten figurines were found within deposits abutting Temple 1 contexts (Φ3–7, ΦB), which were associated with pottery dated 630/610 BC to 560/520 BC.28 They include two female figures down well ΦB (possibly another bothros deposit), three female figure plaques in northern area Φ4, two female figure plaques in southern area Φ3 as well as one female plaque, one kouros and one male plaque by the altar Φ7. Contexts (Φ8–11) abutting Temple 2 were associated with pottery dated 620/600 BC to 550/525 BC. However, the 6th-century pottery and all the figurines found in this phase must have been disturbed with residual redeposited finds,

27

28

For figurines that could be from a Saite phase (dated c. 620–525 BC), the Cypriot bias is more apparent, with 81% Cypriot, 12% Egyptian and 7% Greek.

Three early 5th-century BC sherds found within Temple 1 Φ5 and Φ6 must be contamination or disturbance from the construction of the Temple 2 phase immediately above.

EGYPTIAN AND CYPRIOT STONE STATUETTES IN CONTEXT AT LATE PERIOD NAUKRATIS

175

Fig. 9: Aphrodite sanctuary plans and sections with context numbers, comprising Temple 1 and Temple 2 phases with east–west section and north–south section (dotted line on plans). Drawing by author, based on Gardner 1888, pls 1–3.

176

R. THOMAS

since numerous 5th- and 4th-century BC pottery sherds were found within this phase, which relates to the use of the Temple 2 phase during the Persian period. This would explain the presence of two female figure plaques, one figure vessel and one male plaque in the northern area Φ10, that of five female plaques, two female figure vessels and one animal from around the altar Φ8, as well as one kouros and one female figure plaque from the entrance to the temple and altar area Φ8/9. These finds must have been moved after they were first dedicated within this temple. Female figures (and figure vessels) account for 70% of stone figurines from the Aphrodite sanctuary; however, the high occurrence of male terracotta figures within the Aphrodite sanctuary suggests there was no gendered pattern. Indeed, one of the female figures was offered by a Greek man called Polyermos to Aphrodite. A male archer or hunter figurine wearing an Egyptian shendyt kilt and a Cypriot cap was dedicated by the Greek Kallias to Aphrodite. The terracottas were made of fabrics distinctive of the regions around Salamis, Paphos, Kition and Amathus in Cyprus, which may inform our interpretation of the potential origin of some of the stone figures. Some of the figurines were of types associated with Cypriot ‘Great Goddess’ sanctuaries. As with the stone figurines, Cypriot terracotta figures suggest the sanctuary continued to be in direct contact with Cyprus until c. 300 BC. The unusual longevity of this contact (unlike that of Archaic Ionian sanctuaries) and the lack of nonCypriot figurines from the Aphrodite sanctuary site (compared with the Hellenion and Apollo sanctuaries), and the fact that the majority of Cypriot figurines from Naukratis were found here suggests that maybe this was an important sanctuary specifically for Cypriots and Greeks travelling from Cyprus. Conclusions on figure form, context and meaning Egyptian and Cypriot figurines were contemporary, primarily anthropomorphic figure representations mostly made of limestone, and both were present within a decade or two of the founding of Naukratis. Here the similarities stop, for the two assemblages could not be more different. The two groups of artefacts represent the different traditions of distinct communities, one that may be best described as Egyptian (primarily Lower Egyptian or Deltaic tradition) and a diverse Eastern Mediterranean community may be defined as broadly Cypro-Ionian (being both Greek,

mostly from the Ionian coast, and Cypriot). The CyproIonian label is, however, problematic as its academic use is associated with style (represented as a subgroup of the whole Naukratis assemblage), and also because this does not cover the use of such figurines by Phoenicians (which is attested in Phoenician sites, if not proven at Naukratis). The different characters represented (gods or mortals) represent different religious practices (Greek-Cypriot and Egyptian), and had different votive practices that were performed in different places (the sanctuary or the home). The Egyptian figurines represent important deities (or animals associated with deities) and/or are associated to festivals celebrating the inundation of the Nile. The Cypriot figurines represent people dedicating objects or animals, or the animals being sacrificed (and only rarely at Naukratis, specific deities). Despite being found primarily in secondary or tertiary contexts, distinct distributions across the settlement can be recognised. Fortunately, the final deposition of the figures remained close to the location of their original use (either within the house or nearby dumps in the town, or within the temenos of the sanctuary where they were originally dedicated). The Cypriot figures appear to have been used exclusively as dedications within ‘Greek’ sanctuaries, having been found by altars, in wells, redeposited in bothroi following sanctuary clearances and in tertiary deposits resulting from later sanctuary building developments. This is entirely consistent with how these figurines were used when found in other Greek (mainly East Greek) and Cypriot sanctuaries, which is the main context within which such parallels for the Naukratis examples come. When found within Greek sanctuaries, Cypriot limestone figures (of various types and genders) were usually found within mixed construction fills or bothros deposits deposited between 630 BC and 560 BC. These include the little temple of Vroulia (Nick 2006, 21) and the acropolis of Lindos (Blinkenberg 1931, 5), the Heraion in Samos (Schmidt 1968, 98; Hendrich 2007; Henke 2011, 211–7), the Aphrodite sanctuary at Miletos (Senff 2009, 219–20; Senff pers. comm 2015; Henke 2017, 54–6, 303), and the sanctuary at Emeçik near Knidos (Berges 2006, 55–7; Tuna et al. 2009, 232). This contrasts with the traditional view based on observations of votive practice within sanctuaries in Cyprus itself. Despite the hindrance of few of the seventy-seven Cypriot sanctuary sites discovered during the 19th century that having been scientifically excavated, recorded and published,

EGYPTIAN AND CYPRIOT STONE STATUETTES IN CONTEXT AT LATE PERIOD NAUKRATIS

Ohnefalsch-Richter observed a clear gendered patterning within Cypriot sanctuaries with either male or female figurines (in both stone and terracotta) dominating each sanctuary depending upon the deity worshipped there (Ohnefalsch-Richter 1893, 322–3, 531; Ulbrich 2001, 93; Ulbrich 2008, 65–102). Almost exclusively male figures were found at the ‘ReshefApollo’ sanctuaries at Tamassos-Frangissa and Idalion and the ‘Apollo’ sanctuary at Voni (Ohnefalsch-Richter 1893, 322–3, 531, pls IV–X), while almost exclusively female figurines were found within the ‘Aphrodite’ sanctuaries at Idalion, Achna and Arsos (OhnefalschRichter 1893, 531, pl. VII; Ulbrich 2008, 305–6, 313–4, 447–8; Averett 2011, 135). This pattern was confirmed by the restudy of the ‘Apollo’ sanctuary at Idalion (Senff 1993, 58, pls 53–4) and recent excavations at Kition-Bamboula which revealed mainly (if not exclusively) male figures (Yon 2015, 296–8, 304), as did the rural sanctuary at Athienou-Malloura (dominated by male figurines of charioteers, riders, warriors and their horses: Averett 2011, 135–43). Yet, this gendered pattern was not found in all sanctuaries on Cyprus, as already noted for Limniti and another shrine between Achna and Xylotimbo (Ohnefalsch-Richter 1893, 12, 20, 322, sanctuary nos 12 and 52) and more recently at Marion, where the Marion-Peristeries sanctuary had primarily (but not exclusively) female figures and the Marion-Maratheri sanctuary contained both male and female figurines within its assemblage (Smith, Weir and Serwint 2012, 168 and 171). At Salamis a number of sanctuaries previously thought to be gendered, actually have figurines of both genders represented (but not necessarily in equal proportions) and it is possible that gendered patterning at other sanctuaries may have been the product of excavation sampling strategy (Kiely pers. comm 2018). The lack of a gendered pattern within the Naukratis sanctuaries would then suggest a closer relationship with the East Aegean Greek sanctuaries, or the coastal Cypriot sanctuaries at Xylotimbo, Marion, Limniti and Salamis. While we can only claim that two Cypriot figures were certainly dedicated by Greeks (based on their inscriptions), while Cypriot or other peoples may have dedicated other Cypriot figures within the sanctuary, the pattern of dedication is consistent with primarily Greek East Aegean sanctuaries and those of coastal Cyprus. Unlike the Ionian sanctuaries, however, the continued contact between Naukratis and Cyprus throughout the 6th and until the 4th century BC, particularly within the Aphrodite sanctuary, as represented by the figurine

177

assemblage (see Fig. 7), is not present at the East Aegean sanctuaries, where the dedications of Cypriot figurines were concentrated in the 7th and early 6th centuries BC (Henke 2017; Thomas 2015b). The possible motives and mechanics of how these figurines reached Naukratis and the ritual practice performed there is suggested by the anecdote of the Greek trader Herostratos of Naukratis, who having sailed round many lands, […] touched also at Paphos in Cyprus. There he bought a statuette of Aphrodite, a span high, of archaic style, and went off with it to Naukratis [...] and having sacrificed to the goddess, and dedicated the image to Aphrodite, he called his friends and relations to a banquet in the temple itself. (Athenaios 15.675–6, quoting Polycharmus of Naukratis; Gardner 1888, 55; Jenkins 2001, 166–73)

That this is more than a literary conceit is suggested by the fact that Archaic figurines, sometimes numerous examples, have been found in the cargoes of contemporaneous merchant shipwrecks (Parker 1992, 17, 527, see 84 on Cadiz F and 438 on Tyre G; Panvini 2001, 35, 60–1, fig. 40; Özdaş and Kızıldağ 2016, 81–2; 2017, 41–5 on the Bozburun Archaic shipwreck). Although the Naukratis Cypriot figurines were found in most of the ‘Greek’ sanctuaries, for both the Saite and Persian periods there are clear distinctions between these sanctuaries. The majority of Cypriot figurines were found within the Aphrodite sanctuary, and very few were found within the Hellenion (which was dominated instead by Ionian and Rhodian types, Thomas 2015d, 7–10, 17), suggesting different practices and audiences for the two sanctuaries. Owing to the small sample size, preservation and limited excavation area, it is not currently possible to compare the Dioskouroi or Hera sanctuaries with the better-known Apollo, Aphrodite and Hellenion sanctuaries. Yet, despite their differences, the Greek sanctuaries were more similar to each other than the assemblages that came from the town, which was dominated by Egyptian figurines (where local terracotta examples were most popular). Clearly a significant population practised a form of Egyptian domestic religion in Naukratis, which, despite the disappearance of the limestone tradition in the early Ptolemaic period, persisted in the terracotta industry into the Ptolemaic and Roman periods (Thomas 2015a). However, the scarce contextual information recorded by the excavators for the Egyptian figurines limits any detailed analysis of

178

R. THOMAS

distribution across the site. There was a small and specific range of forms which were found exclusively within the Egyptian sanctuaries. However, there seems to be some overlap in common Egyptian forms found within houses, down wells, and just outside the sanctuary precinct of Amun-Ra (possibly from small chapels, Masson-Berghoff 2019, 145–6). There is only one type of stone figurine which is found in Egyptian and Greek religious and domestic contexts: the Egyptian rider figurine. This seems remarkable in an ethnically diverse settlement such as Naukratis. Rather than representing some common or shared practice, the Greek inscription on the example found within the Hellenion suggests these were instead used and interpreted differently by Greeks and Egyptians. It is also possible to use these assemblages as a rough proxy for understanding secular activities, such as trade and contact between regions (and various ethnic groups). For example, it is clear that the majority of Cypriot figures were imported during the Saite Dynasty, after which there was a steep decline (yet continued presence within the Apollo, Aphrodite and Hellenion sanctuaries) until the beginning of the Hellenistic period (Thomas 2015a; 2015b). Still, the significant number of limestone figures brought from the Memphite region to Naukratis reminds us that Egyptian residents and visitors to Naukratis were also mobile. The Egyptian figurines were clearly present throughout the Late Period, but disappear early in the Ptolemaic period. The pattern of both Egyptian and Cypriot figures would suggest a significant reduction in contact between Naukratis and Cyprus, and between Naukratis and Memphis (representing also contacts with Upper Egypt, the Red Sea, and beyond) just after, around 300 BC. Although both figurine types were used differently, by different communities, and subtle variations may be recognised between the sanctuaries of diverse foreign communities at Naukratis, all relied upon the privileged economic and political role of Naukratis as a hub of trade connected to the Mediterranean Sea, but also Egypt and beyond. Bibliography Arnold, D. 1999. Temples of the last pharaohs. Oxford. Arnold, D. 2003. The Encyclopedia of Ancient Egyptian Architecture. Princeton. Averett, E. W. 2011. The ritual context of the Malloura terracotta figurines. In M. K. Toumazou, P. N. Kardulias and D. B. Counts (eds), Crossroads and

boundaries: The archaeology of past and present in the Malloura Valley, Cyprus. The Annual of the American Schools of Oriental Research 65. Boston, 133–48. Barrett, C. E. 2011. Egyptianizing figurines from Delos: A study in Hellenistic religion. Leiden. Berges, D. 2006. Knidos, Beiträge zur Geschichte der archaischen Stadt. Mainz. Blinkenberg, C. 1931. Lindos. Fouilles de l’Acropole 1902–1914, I. Les petits objets. Berlin. Coulson, W. D. E. 1996. Ancient Naukratis. II/1 The survey at Naukratis. Oxford. Edgar, C. C. 1905. Naukratis 1903, G. – minor antiquities. In Hogarth, Edgar and Lorimer 1905, 123–36. Edwards, A. B. 1885. Terra-cottas of Naukratis (first notice). Academy 702, 261–2. Gardner, E. A. 1888. Naukratis. Part II. Sixth Memoir of the Egypt Exploration Fund. London. Gutch, C. 1898–99. The terracottas. In Hogarth, Edgar and Gutch 1898–99, 67–97. Hendrich, C. 2007. Die Säulenordnung des ersten Dipteros von Samos. Bonn. Henke, J.-M. 2009. Cypriot terracottas from Miletus. In V. Karageorghis and O. Kouka (eds), Cyprus and the East Aegean: Intercultural contacts from 3000 to 500 BC. Nicosia, 206–17. ———. 2011. New evidence for the definition of workshops of Cypriot terracottas at East Aegean findspots and its chronological background. Cahiers du Centre d’Études Chypriotes 41, 211–22. ———. 2017. Die zyprischen Terrakotten aus Milet: Neue Überlegungen zur Einordnung der archaischen zyprischen Terrakotten aus ostägäischen Fundkontexten und ihrer werkstattspezifischen Zuweisung. Milesische Forschungen 7. Berlin. Höckmann, U. 2007. Zyprisch-griechische Kleinplastik: Kouroi, andere Figuren und plastisch verzierte Gefässe. In U. Höckmann and W. Koenigs (eds), Archäologische Studien zu Naukratis II. Worms, 13–307. Höckmann, U. and Möller, A. 2006. The Hellenion at Naukratis; Questions and observations. In A. Villing and U. Schlotzhauer (eds), Naukratis: Greek diversity in Egypt. Studies on Greek pottery and exchange in the Eastern Mediterranean. British Museum Research Publication 162. London, 11–22. Hogarth, D. G., Edgar, C. C. and Gutch, C. 1898–99. Excavations at Naukratis. The Annual of the British School at Athens 5, 26–97.

EGYPTIAN AND CYPRIOT STONE STATUETTES IN CONTEXT AT LATE PERIOD NAUKRATIS

Hogarth, D. G., Lorimer, H. L. and Edgar, C. C. 1905. Naukratis 1903. Journal of Hellenic Studies 25, 105–36. Jasnow, R. and Smith, M. 2010–11. ‘As for those who have called me evil, Mut will call them evil’: Orgiastic cultic behavior and its critics in ancient Egypt (PSI Inv. [provv.] D 114a + PSI Inv. 3056 verso). Enchoria 32, 9–53. Jeffreys, D. G., Smith, H. S., Price, M. and Giddy, L. L. 1988. The Anubieion at Saqqâra. Egypt Exploration Society Excavation Memoir 54. London. Jenkins, I. 2001. Archaic kouroi in Naucratis: The case for Cypriot origin. American Journal of Archaeology 105, 163–79. Johnston, A. 2015. Greek and Latin inscriptions on stone. In Villing et al. 2013–19. . Klotz, D. 2012. The lecherous Pseudo-Anubis of Josephus and the ‘Tomb of 1897’ at Akhmim. In A. Gasse, F. Servajean and C. Thiers (eds), Et in Ægypto et ad Ægyptum: Recueil d’études dédiées à Jean-Claude Grenier. Montpellier, 383–96. Knoblauch, C. M. and Gill, J. C. 2012. Antecedents to the Ptolemaic mammisis. In C. M. Knoblauch and J. C. Gill (eds), Australasian Conference for Young Egyptologists, 4–6 September 2009, Melbourne. BAR International Series 2355. Oxford, 9–13. Koenigs, W. 2007. Archaische griechische Bauteile. In U. Höckmann and W. Koenigs (eds), Archäologische Studien zu Naukratis II. Worms, 311–84. Leonard Jr., A. 1997. Ancient Naukratis. Excavations at a Greek emporium in Egypt. I, The Excavations at Kom Ge’if. The Annual of the American Schools of Oriental Research 54. Boston. ———. 2001. Ancient Naukratis. Excavations at a Greek emporium in Egypt. II, The Excavations at Kom Hadid. The Annual of the American Schools of Oriental Research 55. Boston. Lindström, G. and Pilz, O. 2013. Votivspektren. In I. Gerlach and D. Raue (eds), Sanktuar und Ritual. Heilige Plätze im archäologischen Befund. Rahden, 267–74. Martin, G. T. 1973. Excavations in the sacred animal necropolis at North Saqqâra, 1972–3. Journal of Egyptian Archaeology 60, 15–29. ———. 1981. The Sacred Animal Necropolis at North Saqqâra, the Southern Dependencies of the Main Temple Complex. Egypt Exploration Society, Excavation Memoir 50. London.

179

Masson, A. 2018. Scarabs, scaraboids and amulets. In Villing et al., 2013–19. . ———. Forthcoming. Egyptian sculpture. In Villing et al., 2013–19. Masson-Berghoff, A. 2019. Naukratis: Egyptian offerings in context. In A. Masson-Berghoff and R. Thomas (eds), Naukratis in Context. British Museum Studies in Ancient Egypt and Sudan 24, 127–58. . Möller, A. 2000. Naukratis: Trade in Archaic Greece. Oxford. Nick, G. 2006. Zypro-ionische Kleinplastik aus Kalkstein und Alabaster. Archäologische Studien zu Naukratis I, ed. U. Höckmann. Möhnesee. Ohnefalsch-Richter, M. 1893. Kypros, the Bible and Homer. London. Osborne, R. 2004. Hoards, votives, offerings: The archaeology of the dedicated object. World Archaeology 36, 1–10. Özdaş, H. and Kızıldağ, N. 2016. An Archaic period terracotta statue found in Bozburun. TINA (Türkiye Sualtı Arkeolojisi Vakfı) 7, 80–3. ———. 2017. Marmaris, Bozburun Sualtı Kazısı 2017 Sezonu. TINA (Türkiye Sualtı Arkeolojisi Vakfı) 8, 40–7. Panvini, R. 2001. The Archaic Greek Ship at Gela. Palermo. Parker, A. J. 1992. Ancient Shipwrecks of the Mediterranean and the Roman Provinces. BAR International Series 580. Oxford. Pennington, B. and Thomas, R. I. 2016. Paleoenvironmental surveys at Naukratis and the Canopic branch of the Nile. Journal of Archaeological Science: Reports 7, 180–8. Petrie, W. M. F. 1886. Naukratis. Part I, 1884–5. Third Memoir of the Egypt Exploration Fund. London. Pryce, F. N. 1928. Catalogue of sculpture in the Department of Greek and Roman Antiquities of the British Museum. I.1. Prehellenic and early Greek. London. Quibell, J. E. 1907. Excavations at Saqqâra 1905– 1906. Cairo. Redford, D. B. 2004. Excavations at Mendes I. The royal necropolis. Leiden; Boston. Redford, D. B., Sternberg-el-Hotabe, H. and Redford, S. 1991. Three seasons in Egypt: The first season of excavations at Mendes (1991). Journal of the Society for the Study of Egyptian Antiquities 18, 1–79.

180

R. THOMAS

Schmidt, G. 1968. Samos, VII: Kyprische Bildwerke aus dem Heraion von Samos. Bonn. Senff, R. 1993. Das Apollonheiligtum von Idalion: Architektur und Statuenausstattung eines zyprischen Heiligtums. Studies in Mediterranean Archaeology 94. Jonsered. ———. 2009. Beasts, heroes and worshippers: Statuettes made of Cypriote limestone for the Aphrodite sanctuary of Miletus. In V. Karageorghis and O. Kouka (eds), Cyprus and the East Aegean: Intercultural contacts from 3000 to 500 BC. Nicosia, 218–28. Smith, A. H. 1892. Catalogue of sculpture in the Department of Greek and Roman Antiquities British Museum. Vol. 1. London. Smith, J. S., Weir, M. G. and Serwint, N. 2012. The sanctuaries of Marion. In W. A. P. Childs, J. S. Smith and J. M. Padgett (eds), City of gold: The archaeology of Polis Chrysochous, Cyprus. Yale, 167–85. Thomas, R. I. 2015a. Stone and terracotta figures from Naukratis: An introduction. In Villing et al. 2013– 19. . ———. 2015b. Cypriot figures in terracotta and limestone. In Villing et al. 2013–19. . ———. 2015c. Egyptian Late Period figures in terracotta and limestone. In Villing et al. 2013–19. . ———. 2015d. Greek terracotta figures. In Villing et al. 2013–19. . ———. 2015e. Naukratis, ‘Mistress of ships’, in context. In D. Robinson and F. Goddio (eds), ThonisHeracleion in context. Oxford Centre for Maritime Archaeology Monograph 8. Oxford, 247–65. Thomas, R. I. and Higgs, P. 2017. Greek and Roman sculpture. In Villing et al. 2013–19. . Thomas, R. I., and Villing, A. 2013. Naukratis revisited 2012: Integrating new fieldwork and old research. British Museum Studies in Ancient Egypt and Sudan 20, 81–125. ———. Forthcoming. Return to Naukratis: New fieldwork at Kom Geif, 2012–2014. Annales du Service des antiquités de l’Égypte. Tuna, N., Atıcı, N., Muşkara, U. and Sakarya, I. 2009. Some remarks on the limestone figurines recently found at the Archaic sanctuary of Apollo in the territory of Cnidos. In V. Karageorghis and O. Kouka (eds), Cyprus and the East Aegean: Intercultural contacts from 3000 to 500 BC. Nicosia, 229–43. Ulbrich, A. 2001. An archaeology of cult? Cypriot sanctuaries in 19th century archaeology. In V. Tatton-Brown (ed.), Cyprus in the 19th century AD: Fact, fancy and fiction. Papers of the 22nd British Museum Classical Colloquium, December 1998. Oxford, 93–108. ———. 2008. Kypris. Heiligtümer und Kulte weiblicher Gottheiten auf Zypern in der kyproarchaischen und kyproklassischen Epoche (Königszeit). Münster. Villing, A., with Thomas, R. I. 2015. The site of Naukratis: topography, buildings and landscape. In Villing et al. 2013–19. . Villing, A., Bergeron, M., Bourogiannis, G., Johnston, A., Leclère, F., Masson, A. and Thomas, R. (eds). 2013–19. Naukratis: Greeks in Egypt. British Museum Online Research Catalogue. London. . Yon, M. 2015. Les sculptures de pierre. In A. Caubet, S. Fourrier and M. Yon (eds), Kition-Bamboula. 6, Le sanctuaire sous la colline. Travaux de la Maison de l’Orient et de la Méditerranée. Série recherches archéologiques 67, 295–334.

IV BECOMING INVISIBLE: STATUE CACHES

A CLASSIFICATION OF ‘SACRED’ CACHES IN ANCIENT EGYPT Guillaume CHARLOUX and Mona Ali Abady MAHMOUD1

Abstract Although particularly spectacular, the caches of sacred artefacts discovered in Egyptian religious complexes, often discussed in the literature, have rarely been interrogated from a purely archaeological point of view. In this paper, we have tried to consider the context of discovery of the caches as a priority. Through this examination, we propose to identify among caches of sacred artefacts those that can really be characterised as ‘sacred’ (that is to say, having received a liturgical burial at the hands of the temple clergy), by establishing a classification according to common and objective criteria. * * * In the Egyptological literature, it is well-attested that sacred caches had the role of protecting, hidden from the ungodly gaze, the divine instruments and statues consecrated in the temples (Bonnet and Valbelle 2005, 174–6; Nicholson 2005, 12). Still invested with its magic character and power, the unusable or cumbersome religious furniture had to be preserved from any external intrusion; the artefacts were accordingly gathered in an enclosed space by the priests of the temple. Their burial was certainly accompanied by liturgical ceremonies, the exact nature of which escapes us in the absence of descriptive textual sources. The archaeological examination of these specific ensembles should, therefore, tell us more about the gestures of the buriers and the associated rites. In consulting the scientific literature on

1

Franco-Egyptian Centre for the Study of the Temples of Karnak (Cfeetk, Ministry of Egyptian Antiquities, CNRS/USR 3172), Egypt.

this subject, it must be admitted that it is just as difficult to characterise the deposits and make a typology, as to date them (e.g. Valbelle 2016). For this reason, we believe that the establishment of a preliminary typological classification according to the archaeological context alone will contribute to a better understanding of this type of cache and facilitate their interpretation (Fig. 1). Sacred caches and non-sacred caches of religious artefacts It should be noted at the outset of this paper that what we here call ‘sacred caches’ make up only a small part of the ‘caches of sacred (or religious) artefacts’ as a whole, and they are differentiated by us according to criteria which will be defined below, after a review of the archaeological documentation. This distinction, which aims to isolate a practice that was the subject of a religious process from the temple clergy, considers

Fig. 1: The discovery of a small sphinx in the favissa of the temple of Ptah in Karnak and on-site restoration of its front paw, 4 December 2014. © CNRS-CFEETK, J. Maucor.

184

G. CHARLOUX AND M.A.A. MAHMOUD

a set of deposits of religious artefacts in their secondary context only — votive or ritual furniture and figures coming from a religious space — whose origin and function are, most of the time, difficult to assess.2 Egyptological publications are full of discoveries of caches of sacred artefacts; as only a really complete inventory would make it possible to define the contours of the multiple practices, it is important to exclude from the typological examination of the sacred caches at this stage of the research the following categories: •







2

‘Treasure’, here assimilated to the ‘safeguarding deposit’ (dépôt de sauvegarde in Vernus 1989) which gathers — often in a container — a set of objects having a more economic than cultic value (ingots, coins, jewels) that are destined to be found again by the buriers. Also sometimes referred to as a ‘hoard’, it has no religious function. It is sometimes difficult to distinguish from a sacred cache, especially when the treasure includes divine statuettes, as is the case with the treasure of Athribis (Engelbach 1924, 181–5), or consecrated objects as in Toukh el-Qaramous (Edgar 1906, 207, 209) and Douch (Reddé 1989; Reddé, Gout and Lecler 1992). The identification of the intent of the burier — an interpretation fraught with risk — depends here on the deposit and its organisation. ‘Execration deposit’. This type of pit deposit (for which the term favissa is inappropriate), attested from the Old to the New Kingdom, was composed of small inscribed plaques or ostraca but also of animal or human substitutes, especially figurines of prisoners (Vila 1963, 146–7). The objects were used in rituals to annihilate the power of enemies through magic (Muhlestein 2008). The figurines were a support to texts of spell that sometimes covered them (Jambon 2010, §38). ‘Embalming deposit’. A collection in a pit or reserved area of the utensils, textiles and dishes used for the preparation of bodies before burial (Eaton-Krauss 2008; Knoblauch 2016). ‘Foundation deposit’, intended to commemorate, protect, purify and sacralise the constructions under which it is positioned (Weinstein 1973; Masson 2015). It is easily recognised by the use of artefacts produced for this purpose only — often small or

Furniture and statues in their primary contexts in the official religious spaces, in domestic sanctuaries or in tombs are, obviously, excluded from this contribution.



even miniature dishes — and usually the presence of royal cartouches. A foundation deposit is also buried at strategic locations of the building, such as under the corners of the walls and the door thresholds. Although it is a deposit of consecrated objects, we reject this type of cache from our study, since the artefacts are designed for the deposit itself and are therefore considered in primary use. ‘Chaufournier pit’. This is also a ‘safeguarding deposit’ but for agricultural or architectural purposes; it can gather elements of religious statuary in a pit dug near lime kilns, in which they have occasionally been discovered (Munro 2016, 49, n. 8). Well-known in the Mediterranean world, it is, however, rarely mentioned with significant archaeological information in Egyptology. It does not, of course, have any religious aspect.

The typological classification of sacred caches Without presenting here the list of caches considered, for lack of space (Charloux and Thiers in press), we introduce below several common criteria permitting the identification of sacred caches among caches of sacred artefacts, according to their shapes, the nature of the buried objects, and the location of the deposit, but also according to the internal arrangement of the filling (Fig. 2). A. The shape of the deposit We have chosen to give priority to the archaeological context of the deposit and, in particular, its shape and its relation to the architectural environment as a decisive criterion for this typological classification. It results in the elaboration of three distinct categories: A1 The deposit of artefacts in pit or favissa The favissa is, by definition, a pit dug specifically for burying consecrated objects (Fig. 3). We know that its location is rarely random: it is laid under pavements in open spaces, especially the courtyards of religious buildings (Jambon 2016, 156–7) and preserved within the temple precinct.3

3

The identification of favissae in domestic context can be debated.

A CLASSIFICATION OF ‘SACRED’ CACHES IN ANCIENT EGYPT

185

The shape of the deposit 1. The favissa

2. The “foundation cache”

3. The “prepared cache”

a. structures intentional built for the burial

b. pre-existing and reused structures / spaces

The internal organization of the deposit A. Unique stage

One artifact

B. Staging organization

Multiple artifacts

Multiple artifacts

The nature of the religious artifacts

1. Religious furniture

2. Statuary cache

3. Mixed material

Magical and physical protection

Cover

Guardian

Osirian figure

Mutilation of artifacts

Fig. 2: Schematic typology of ‘sacred’ caches in ancient Egypt. © CNRS-CFEETK, G. Charloux.

186

G. CHARLOUX AND M.A.A. MAHMOUD

A2 The ‘foundation cache’ It is under the term ‘foundation cache’ that we designate the deposits corresponding to the burial of statues in the foundation pits of buildings, which should be distinguished from the more famous ‘foundation deposits’.4 Artefacts do not have a specific architectural role, other than filling the monument’s foundation pit, or the pavement preparation layers (e.g. Ziegler 1981). The pit is not specifically dug to receive the artefacts, in contrast to favissae or foundation deposits. They can, nevertheless, present original arrangements in the same way as the other categories of caches, indicating specific care on the part of the priests at the time of the burial (Charloux 2012, 276–80) (Fig. 4). The ‘foundation caches’ occurred as part of liturgical ceremonies in connection with the construction of the monument.

Fig. 3: Artist’s drawing showing the organisation of the cultic furniture (copper alloy statue, ceramic lion and schist statue) in a favissa discovered in Hierakonpolis. After Quibell and Green 1902, pl. 47.

A

Fig. 4A and B [next page]: Original, and careful, positions — standing, lying and upside-down — of three statues found in the foundation pit of the Opet temple in Karnak, at three corners of the forecourt of the temple. © CNRS-CFEETK, G. Charloux.

4

Sometimes named ‘consecration deposit’, e.g. Leclère 2009; Coulon 2016, 32, n. 30, but without precise definition.

A CLASSIFICATION OF ‘SACRED’ CACHES IN ANCIENT EGYPT

Statue of Priest Horudja O.168065-1

Statue of Baboon O.168005-1

Block statue of Priest Nesmin O.168155-1

Legend

Statue of Ramesside Priest found in 1951 by A. Varille

187

188

G. CHARLOUX AND M.A.A. MAHMOUD

A3 The ‘prepared cache’ This category includes the depressions / basins / recesses / niches used to conceal a statue in a secondary context and probably having similar functions to the favissae. It might be suggested that there is a close relationship between an intentionally dug pit and a natural and an artificial recess created by removal of material or the reuse of a vacant space below the surface. The role of this cache remained the burying of the statue(s), to protect or hide them. The difficulty of interpretation lies here in distinguishing what comes out of the religious ritual. At this stage of the research, we would be inclined to differentiate two main groups of ‘prepared caches’: • A3a. Deposits installed in structures intentionally built for the burial (e.g. masonry recess, pit with builtup walls or niche/recess created for a statue, etc.); • A3b. Deposits in pre-existing and reused structures and spaces (e.g. well, basin, etc.) We know several variants of these caches: for example, a space reserved in the foundations of the temple at Ermant containing the heads of a colossus and of a priest statue, and a stela (Thiers 2014), or another one in the threshold of a gate at North Karnak (Barguet and Leclant 1954, pl. XXXV) (Fig. 5). In Tabo, C. Bonnet describes a recess dug in the basement of the temple — the walls of which were covered with reused blocks and some bricks, which allowed the preservation of

religious furniture (Bonnet 1986, 19). The Water Sanctuary of Meroe (Török 1997), consisting of a large masonry basin, could also serve as a statuary cache where statue(s) could be installed or thrown. This category also includes the statues or statuary fragments re-employed in the masonry of a building, filling empty spaces like the previous caches (Varille 1943, 23; Ben-Tor 2006, 6, no. 9). Having a priori an architectural role, the statue thus hidden in the foundations of the sanctuary also possibly possessed a magical or apotropaic function, like the previous category of cache. B. The internal organisation of the deposit The study of the internal organisation of the sacred caches makes it possible to envisage a further subdivision to the preceding structural groups. B1 The caches with a unique stage The deposit is created at one time and has only one stage: the homogeneous filling was put in place without discontinuity or interruption in the installation of artefacts. Its study can, however, show a sophisticated organisation, as for example at Deir el-Bahri or Lisht (Gautier and Jéquier 1902, 30) (Fig. 6), like the following category (B2). These caches include either a single artefact [B1a], which is the most basic category of cache — such as favissae encountered at Tabo or at the temple of Mentuhotep II at Deir el-Bahri (Arnold, Peek and Hauser 1979, 46–7, 49), or several artefacts [B1b] (e.g. Lisht, Kerma, Hierakonpolis, etc.). There is a subgroup specific to this second set: these are favissae (especially) including dozens or even hundreds of Osirian statuettes, often made of metal, not associated with other furniture or statuary elements, and for which we have little detailed contextual information (Benson and Gourlay 1899, 72; at Medinet Habu: Hölscher 1939, 40; Hölscher 1954, pl. 19g; Thomas 1995, 205). We might ask ourselves if there was sometimes a larger cultic statue(s) (in perishable material) in the pit, which could have been totally lost through decay in some cases. B2 The caches with organised ‘stages’

Fig. 5: Cube statue installed in a recess of a gate in the temple of Montu in Karnak. After Barguet and Leclant 1954, pl. XXXV.

This second group brings together the most famous Egyptian cache, the Cachette of Karnak, together with

A CLASSIFICATION OF ‘SACRED’ CACHES IN ANCIENT EGYPT

the favissae of Luxor, of Ptah in Karnak and certain caches of North Saqqara and Meroe in particular. The excavators’ observations and the old photographs make it possible to recognise or even to count ‘stages’, i.e. distinct levels within the filling of the caches. The study of these deposits reveals an elaborate internal arrangement, the examination of which shows a complex and well-thought-out burial practice. The favissa of the temple of Ptah is one of the most evocative examples of this category (Fig. 7). During the excavation of the Cachette of Karnak, Legrain also observed ‘filons intacts d’objets’ and mentioned levels with and without artefacts in the inner filling (Legrain 1906, 144–5). Saghir’s observations on the cache of the temple of Luxor are similar, mentioning series of statues that appeared in quantity and in stages, sometimes after several days of digging without any new discovery (Saghir 1992, 15). The information provided by the author makes it possible to restore the staging of

189

artefacts over at least four, perhaps even five, levels (Fig. 8). The existence of multiple fillings or multiple stages has sometimes led to the suggestion that the cache had been filled several times, following successive reopenings — based in particular on a practice ‘bien attestée dans le monde gréco-romain’ (Bonnet and Valbelle 2005, 181). However, the examination of several pits, including that of the favissa of the temple of Ptah in Karnak, clearly reveals no post-burial opening, and highlights the ephemeral nature of the burial operation of the objects. In the past, several Egyptologists have concluded the same, for example Quibell and Green (1902, 34–5). C. The nature of the religious artefacts in the caches The identification of the nature of sacred objects buried in the cache (cult furniture vs. statuary) seems to allow a subdivision of the previous ensembles. Several studies have indeed shown the regular absence of mixed material within the considered assemblages, furniture and statuary not having the same symbolic significance (Davies 2007, 178). However, this somewhat radical dichotomy must also be nuanced. C1 The cache of religious equipment The caches of furniture of the temple are composed of deposits of cultic/votive objects from the temple, other than statuary. They can include architectural columns, cultic vessels such as situlae, other ritual equipment, elements of sanctuaries (naoi), etc. The most notable have been found in North Saqqara, but others are attested in Ashkelon (Bell 2011; Stager, Schloen and Master 2011), Tell el-Herr (Valbelle 1995; Leclère 2007; Valbelle and Marchi 2012), Kawa (Macadam, Griffith and Kirwan 1955, 92–3, 125 [33/5], 158–9 and pl. 58c), Dendera (Petrie 1900, 28–9, 34, 63–4, pls 24, 36; Shore 1979, 158), and Douch (Wuttmann et al. 1996, 438, fig. 71). C2 The statuary cache

Fig. 6: Sketch of the arrangement of the statues of Senwosret I in the favissa of Lisht. After Gautier and Jéquier 1902, fig. 23.

The statuary sacred caches gather one or more statues, but also (or only) figurines and statuettes (Fig. 9). This is the case in the favissa of Luxor, for example (Fig. 10). In the deposit, the association of private, royal and divine statues in caches is neither rare nor systematic. Nor does it seem to govern the location of the statuary

190

G. CHARLOUX AND M.A.A. MAHMOUD

Fig. 7: Reconstruction in section of the location of artefacts found in the favissa of the temple of Ptah in Karnak, from a 3D model realised by photogrammetry after restoration of the objects. © CNRS-CFEETK, K. Guadagnini and G. Charloux.

deposit, for example under an enclosure wall or in a courtyard (contra Valbelle 2016). The private statuary offered to the temple cult is sacralised in the same way as the other statues, although it is likely that some form of hierarchy determines the position and the order of the deposit. Within the statuary caches, the analysis of statuary caches sometimes makes it indeed possible to envisage a form of hierarchy of buried effigies and the emergence of a central figure within the excavated or prepared space. In the favissa of the temple of Ptah, there

is no doubt that the statue of the god Ptah, surrounded by many figurines and laid flat at the bottom of the pit, is the raison-d’être of the deposit (see Fig. 7) (Charloux et al. 2017). The internal arrangement proceeds from the location of the divine statue, around and above which are distributed the other theriomorphic and anthropomorphic figures. It must be kept in mind, moreover, that the favissa is dug behind the naos of the deity, within his own worship area. Let us take a second example: on the second level of the favissa of the temple of Luxor, the figure of

A CLASSIFICATION OF ‘SACRED’ CACHES IN ANCIENT EGYPT

191

Fig. 8: Diagram summarising the different stages of the favissa of the temple of Luxor. © G. Charloux.

Amenhotep III, lying and facing west, is surrounded by three statues at the same level, to emphasise the central statue (Saghir 1992, 8, figs 13–4). This magnificent statue of Amenhotep III, which is most beautiful, certainly represents the heart of the cache (see Fig. 10). It is likely that other caches also exhibit this ‘statuocentrism’, but this practice does not seem systematic. C3 The mix of the artefact categories Although caches containing one category of material seem to prevail, it seems that many exceptions counterbalance this homogeneity. The Cachette of Karnak has indeed delivered a significant quantity of wooden furniture, associated with hundreds of stone statues and thousands of bronzes (Jambon 2009, 272), as well as many bones and remains of animal mummies (Legrain 1906, 146). When a mix of furniture (ritual equipment and statues) appears, the statues appear to be primary; they seem to be at the heart of the deposit’s intention and probably have a higher symbolic significance.

Fig. 9: Small masculine head with golden face from the favissa of the temple of Ptah at Karnak, at the moment of its discovery. © CNRS-CFEETK, J. Maucor.

192

G. CHARLOUX AND M.A.A. MAHMOUD

On the contrary, when bronze statuettes of Osiris are found in caches of ritual equipment, as is often the case, their presence does not modify the specific character of the deposit, in our opinion. It is simply due to the preponderance of the cult of Osiris, especially in the 1st millennium BC. It is, therefore, linked with the functional purpose of the sacred cache (see conclusion). D. The physical protection of the deposit The top part of the caches is sometimes characterised by the presence of a cover, for example the base of the statues of Horemheb and Atum in Luxor (Saghir 1992), the fragments of a naos in North Saqqara (cache 1) (Smith 1976, 16, pl. V.2), the stela of Sety I in the Cachette of Karnak (Legrain 1905, 63) or the statues’ bases in Kerma (Bonnet and Valbelle 2005, 76). This compact, flat element forms a level of symbolic and physical protection that seals the deposit (Fig. 11). When this ‘sealing’ element is installed in a courtyard or an open space of the temple, it is in

addition to an upper pavement which definitively closes the deposit. Like a foundation pit, this feature can be completed by the addition of a separation / purification layer of fluvial sand (Karnak) or of gypsum (Luxor) (Jambon 2016, 156–7). E. Recurring divine effigies It appears that several divine figures (guardian animal and votive [E1] or ritual figurines of Osiris [E2] in particular) are common in the caches (Fig. 12). We observe the presence of a guardian, sphinx, lion, but also a jackal (Anubis) within certain caches of religious artefacts (e.g. Emery 1967, pl. XXI/2; Török 1997, pl. 52; favissa of Ptah [see Fig. 1]). This role of the jackal/wild dog Anubis is well-known, protecting the remains of his feast buried in the ground, and thus becoming mythologically the protector of Osirian viscera to the point of appearing on the seal of the Theban necropolis. The Osirian figure, meanwhile, is omnipresent in the caches of sacred objects, where it is evoked or

Fig. 10: Second stage of the favissa of the temple of Luxor. In the centre of the deposit, the statue of Amenhotep III, according to Saghir 1992, fig. 14.

A CLASSIFICATION OF ‘SACRED’ CACHES IN ANCIENT EGYPT

193

present in almost a third of cases. In the favissa of Ptah, fourteen small and totally uninscribed figures of the deity have been recorded, making it, by far, the main group. In addition to the Osirian function of certain caches, this preponderance of Osirian figures is closely linked to the rise of the cult of Osiris in the 1st millennium BC. F. The question of the mutilation of objects The ‘mutilation’ of the statues — or at least the fact that they are invariably broken or damaged — seems to constitute a general principle of caches (e.g. Chassinat 1921, 56–7; Mond and Myers 1940, 49) (Fig. 13). Although it is usually not easy to identify the event that led to the breaks, it is undeniable that a major part of them occurred well before the deposition; this is, for example, the case of the statuette of Mut (O.179-8048-5) in the favissa of Ptah, which presents signs of ancient resFig. 11: Upper part of the favissa of the boat landing stage of the temple toration, visible notably by cracks filled of Amun at Karnak, creating a flat cover for the statuary deposit. with mortar. © CNRS-CFEETK, A. Bellod, 5899. Noting the variety of situations encountered pre-burial, we deduce that it is not so much the reasons for the mutilations that are decicaches. However, its consideration is important to sive as the systematism of this practice. Therefore, this explain the function of the deposit, since the buried is not a discriminating criterion for a typology of artefacts were certainly at the end of use or of ‘life’. The degradation, voluntary or not, the ‘death’ of the statue, was an indispensable element to the constitution of the deposit (on the de-activation of statues through mutilation, see Connor in this volume). Towards a typology of sacred caches

Fig. 12: Statue of recumbent jackal (Anubis) in a North Saqqara deposit. After Emery 1967, pl. XXI, no. 2.

In this short presentation, we wanted to survey our knowledge related to the practice of sacred caches and try to better define their intrinsic characteristics. The criteria proposed for this typology are all the more important to consider as they make it possible to rethink the repositories in sets as objectively as possible and to try to perceive, by category, functional and/or chronological logics. The identification of a religious or ritual intention in the deposit is still a hard task. This intention often remains unprovable by the field data in most scientific

194

G. CHARLOUX AND M.A.A. MAHMOUD

4. The presence of ‘protective’ images in the deposit. Although non-discriminating, these figures are an indubitable mark of prophylactic rituals and confirm the identification of the deposit from the moment the first three criteria are present. Caches that lack these criteria could also have been subject to ritual, but only those thus identified can be described as ‘sacred’ on an objective basis. As a result, there are few caches of religious artefacts that meet these criteria, which can, therefore, be described as sacred: •



• Fig. 13: Face of the sphinx recently discovered in Karnak, in the favissa of the temple of Ptah. The nose was probably damaged by a mason’s tool such as a chisel. © CNRS-CFEETK, J. Maucor.

publications, and even more so when these are absent or partial. With the utmost caution, however, it seems that one can confirm the existence of ‘truly’ sacred caches, with ritual intention, operated under the control of the clergy of the temple. Four discriminating criteria must be considered in order to identify this, in our opinion: 1. An important investment (time, effort and willingness) necessary for the installation of worship and votive furniture in reserved and prepared places, protected within the religious complex, and rendered inaccessible. 2. An original arrangement of religious objects, revealing an effort to establish some intrinsic organisation of the repository. Without this criterion, which is closely associated with the previous one, the ‘sacred’ nature of the deposit remains often difficult to establish. 3. The damaged state of most of the figures. The statues had to be, in one way or another, at the end of ‘life’ or use.



The caches with a ‘staged’ organisation, the different levels of which are prepared for the sake of staging, which aims to create a universe of its own (Cachettes of Karnak and Luxor, favissa of Ptah, etc.). The elaborate foundation caches, for example that of the temple of Opet in Karnak, which highlight the link between the religious monument and the filling of its foundation pit. Most of the prepared caches, which required a significant investment in connection with the cult building, for example that of Ermant or Tabo. Some caches with a unique but original filling, for example those of Deir el-Bahri, Kerma or the Temple of Ermant.

Note finally that, in the framework of a recently published article, these observations on the organisation of the caches, the symbolic protection of the deposit and the presence of votive figurines, allowed us to suggest that the sacred caches were, at least in certain cases, simulacra of Osirian tombs (Charloux et al. 2017; for more details see Charloux and Thiers in press). Acknowledgements We wish to thank Aurélia Masson-Berghoff for her invitation to the colloquium at the British Museum and for her careful reading of this paper, and also Christophe Thiers for his many useful comments. The project of study of the temple of Ptah in Karnak (dir. C. Thiers) is supported by LabEx Archimède from the programme ‘Investissement d’Avenir’ ANR-11LABX-0032-01. We particularly wish to thank the Ministry of Antiquities of Egypt (MoA), the CNRS and the French Ministry of Foreign Affairs for their financial, technical and scientific support.

A CLASSIFICATION OF ‘SACRED’ CACHES IN ANCIENT EGYPT

Bibliography Arnold, D., Peek, G. M. and Hauser, W. 1979. The temple of Mentuhotep at Deir el Bahari. New York. Barguet, P. and Leclant, J. 1954. Karnak-Nord IV. Fouilles de l’Institut français d’archéologie orientale 25. Cairo. Bell, L. A. 2011. Collection of Egyptian bronzes. In L. E. Stager, D. M. Master and D. Schloen (eds), Ashkelon 3. The seventh century BC. Winona Lake, 397–420. Benson, M. and Gourlay, J. A. 1899. The temple of Mut in Asher: An account of the excavation of the temple and of the religious representations and objects found therein, as illustrating the history of Egypt and the main religious ideas of the Egyptians. London. Ben-Tor, D. 2006. The sad fate of statues and the mutilated statues of Hazor. In S. Gitin, J. E. Wright and J. P. Dessel (eds), Confronting the past: Archaeological and historical essays on ancient Israel in honor of William G. Dever. Winona Lake, 4–16. Bonnet, 1986. La découverte de la statue. In C. Maystre (ed.), Tabo I. Geneva. Bonnet, C. and Valbelle, D. 2005. Des pharaons venus d’Afrique: La cachette de Kerma. Paris. Charloux, G. (ed.). 2012. Le parvis du temple d’Opet à Karnak. Exploration archéologique (2006– 2007). Travaux du Centre franco-égyptien d’étude des temples de Karnak. Bibliothèque générale 41. Cairo. Charloux, G., Thiers, C., Abd Al-Aziz, M., Ali Abady Mahmoud, M., Boulet, St., Bourse, C., Guadagnini, K., Laroye, J. and ElNasseh, A. S. 2017. The afterlife of Egyptian statues: A cache of religious objects in the temple of Ptah at Karnak. Antiquity 359, 1189–1204. Charloux, G. and Thiers, C. In press. Le temple de Ptah à Karnak III. La favissa. Travaux du Centre francoégyptien d’étude des temples de Karnak. Bibliothèque générale. Cairo. Chassinat, E. 1921. À propos d’une tête en grès rouge du roi Didoufrî. Monuments et Mémoires publiés par l’Académie des Inscriptions et Belles-Lettres, Fondation Eugène Piot 25, 53–75. Coulon, L. 2016. Les chapelles osiriennes de Karnak. Aperçu des travaux récents. Bulletin de la Société française d’égyptologie 195–6, 16–35. Davies, S. 2007. Bronzes from the Sacred Animal Necropolis at North Saqqara. In M. Hill and

195

D. Schorsch (eds), Gifts for the gods: Images from Egyptian temples. New York, 174–87. Eaton-Kraus, M. 2008. Embalming caches. Journal of Egyptian Archaeology 94, 288–93. Edgar, M. C. C. 1906. Report on an excavation at Toukh el-Qaramous. Annales du Service des antiquités de l’Égypte 7, 205–12. Emery, W. B. 1967. Preliminary report on the excavations at North Saqqâra 1966–7. Journal of Egyptian Archaeology 53, 141–5. Engelbach, R. 1924. The treasure of Athribis (Benha). Annales du Service des antiquités de l’Égypte 24, 178–85. Gautier, J.-E. and Jéquier, G. 1902. Mémoire sur les fouilles de Licht. Mémoires publiés par les membres de l’Institut français d’archéologie orientale 6. Cairo. Hölscher, U. 1939. The excavation of Medinet Habu II. The temples of the Eighteenth Dynasty. Oriental Institute Publications 41. Chicago. ———. 1954. Excavations at Medinet Habu V. Post Ramesside remains. Oriental Institute Publications 66. Chicago. Jambon, E. 2009. Les fouilles de Georges Legrain dans la Cachette de Karnak (1903–1907). Nouvelles données sur la chronologie des découvertes et le destin des objets. Bulletin de l’Institut français d’archéologie orientale 109, 239–79. ———. 2010. Les mots et les gestes. Réflexions autour de la place de l’écriture dans un rituel d’envoûtement de l’Égypte pharaonique. Cahiers ‘Mondes anciens’. Anthropologie et histoire des mondes antiques. . ———. 2016. La Cachette de Karnak. Étude analytique et essais d’interprétation. In L. Coulon (ed.), La Cachette de Karnak. Nouvelles perspectives sur les découvertes de G. Legrain. Bibliothèque d’étude 161. Cairo, 131–75. Knoblauch, C. 2016. A new group of Middle Kingdom embalming deposits? Another look at pottery dumps and repositories for building materials in Middle Kingdom cemeteries. Ägypten und Levante 26, 329–56 Leclère, F. 2007. Un sanctuaire d’époque hellénistique. In D. Valbelle (ed.), Tell el-Herr: Les niveaux hellénistiques et du Haut-Empire. Paris. ———. 2009. Le temple d’Osiris Coptite. In C. Thiers and M. Boraik (eds), Centre franco-égyptien

196

G. CHARLOUX AND M.A.A. MAHMOUD

d’étude des temples de Karnak. Rapport d’activité 2009. Luxor, 24–8. Legrain, G. 1905. Renseignements sur les dernières découvertes faites à Karnak. Recueil de travaux relatifs à la philologie et à l’archéologie égyptiennes et assyriennes 27, 61–82. ———. 1906. Nouveaux renseignements sur les dernières découvertes faites à Karnak (15 novembre 1904–25 juillet 1905). Recueil de travaux relatifs à la philologie et à l’archéologie égyptiennes et assyriennes 28, 137–61. Macadam, M. F. L., Griffith, F. L. and Kirwan, L. P. 1955. The temples of Kawa. Oxford University Excavations in Nubia. London. Masson, A. 2015. Foundation deposits. In A. Villing, M. Bergeron, G. Bourogiannis, A. Johnston, F. Leclère, A. Masson and R. Thomas, 2013–19, Naukratis: Greeks in Egypt, British Museum Online Research Catalogue. . Mond, R. L. and Myers, O. H. 1940. Temples of Armant. A preliminary survey. London. Muhlestein, K. 2008. Execration ritual. In W. Wendrich, J. Dieleman, E. Frood and J. Baines et al. (eds), UCLA Encyclopedia of Egyptology. Los Angeles . Munro, B. 2016. Sculptural deposition and lime kilns at Roman villas in Italy and the western provinces in Late Antiquity. In T. M. Kristensen and L. M. Stirling (eds), The afterlife of Greek and Roman sculpture. Late Antique responses and practices. Ann Arbor, 47–67. Nicholson, P. T. 2005. The Saqqara Bronzes Project. Journal of Egyptian Archaeology 91, 12–15. Petrie, W. M. F. 1900. Dendereh 1898. Egypt Exploration Fund 17. London. Quibell, J. E. and Green, J. 1902. Hierakonpolis II. London. Reddé, M. 1989. Le trésor de Douch (information). Comptes rendus de l’Académie des inscriptions et belles-lettres 133 (2), 427–45. Reddé, M., Gout, F. and Lecler, A. 1992. Douch IV. Le trésor. Inventaire des objets et essai d’interprétation. Documents de fouilles de l’Institut français d’archéologie orientale du Caire 28. Cairo. al-Saghir, M. E. 1992. Das Statuenversteck im Luxortempel. Mainz am Rhein. Shore, A. F. 1979. Votive objects from Dendera of the Graeco-Roman period. In J. Ruffle, G. A. Gaballa

and K. A. Kitchen (eds), Glimpses of ancient Egypt. Warminster, 138–60. Smith, H. S. 1976. Preliminary report on excavations in the Sacred Animal Necropolis, Season 1974– 1975. Journal of Egyptian Archaeology 62, 14–17. Stager, L. E., Schloen, J. D. and Master, D. M. 2011. Ashkelon 3. The seventh century BC. Winona Lake. Thiers, C. 2014. Armant: Recent discoveries at the temple of Montu-Re. Egyptian Archaeology 44, 32–5. Thomas, N. (ed.) 1995. The American discovery of ancient Egypt. Los Angeles. Török, L. 1997. Meroe city, an ancient African capital. John Garstang’s excavations in the Sudan. Egypt Exploration Society Occasional Publications 12. London. Valbelle, D. 1995. Les niveaux hellénistiques de Tell el-Herr. Bulletin de la Société française d’égyptologie 132, 30–42. ———. 2016. Statues enterrées, dépôts liturgiques et différentes catégories de favissae. In L. Coulon (ed.), La Cachette de Karnak. Nouvelles perspectives sur les découvertes de Georges Legrain. Bibliothèque d’étude 161. Cairo, 21–33. Valbelle, D. and Marchi, S. 2012. Un dépôt votif de la forteresse de Tell el-Herr. In C. Zivie-Coche and I. Guermeur (eds), Parcourir l’éternité. Hommages à Jean Yoyotte. Bibliothèque de l’école pratique des hautes études, Section des Sciences Religieuses 156. Turnhout, 1032–6. Varille, A. 1943. Karnak I. Fouilles de l’Institut français d’archéologie orientale 19. Cairo. Vernus, P. 1989. Caches au trésor dans l’Égypte pharaonique. Hathor 1, 35–40. Vila, A. 1963. Un dépôt de textes d’envoûtement au Moyen Empire. Journal des savants 3 (1), 135–60. Weinstein, J. N. 1973. Foundation deposits in ancient Egypt. Ann Arbor. Wuttmann, M., Bousquet, B., Chauveau, M., Dils, P., Marchand, S., Schweitzer, A. and Volay, L. 1996. Premier rapport préliminaire des travaux sur le site de ‘Ayn Manawir (oasis de Kharga). Bulletin de l’Institut français d’archéologie orientale 96, 385–451. Ziegler, C. 1981. Une découverte inédite de Mariette, les bronzes du Sérapéum. Bulletin de la Société française d’égyptologie 90, 29–45.

THOUSANDS OF OSIRIS: THE ARCHAEOLOGICAL CONTEXTS OF THE BRONZES FOUND IN THE TEMPLE OF ‘AYN MANAWÎR AND AT THE SERAPEUM OF MEMPHIS1 Florence GOMBERT-MEURICE

Abstract It is first argued that the group of finds, mainly consisting of Osiris bronze figures, that was discovered in the temple of Osiris-Iw at ‘Ayn Manawîr in 1994 attests to a lost liturgy. The statuettes were found in a chapel presumed to be still in use when it collapsed. The assemblage contains recurrent groups of two or four figures shaped from the same mould and likely to have been set on the same base. Such series of figures do not seem to relate to economic matters or pilgrim’s practices, but rather to an iconography specifically selected for a ritual. Another archaeological assemblage of bronze figures discovered in the 19th century, deposited in the foundations of the huge religious centre of the Serapeum at Saqqara, is also explored. A survey of the Osiris statuettes found on the site, and now partly preserved in the Louvre, raises the question of their interpretation as testimony of a liturgical assemblage. It appears that here, too, it is possible to detect some recurrent groups of statuettes. Some details that may stand as criteria for liturgical assemblages are suggested, in order to facilitate the future study of such assemblages among the numerous surviving bronze statuettes.

Thousands of bronze figures Deposits of bronze figures are widely attested in the temples of the 1st millennium BC, but proper documentation of their precise context of discovery, date or function is often lacking. The archaeological context of most bronze figures remains vague essentially because they have been discovered around temples and near walls, in rooms undefined from an architectural point of view. Furthermore, they were found in discouragingly large numbers during the 19th and first half of the

1

I would like to deeply thank Marie Millet, Astrid Emerit and Aurélia Masson-Berghoff who have kindly read this article.

20th century, when huge areas provided a generous number of objects that were often scattered among various art dealers, especially the small and valuable ones (e.g. Daressy 1897, 170; Bisson de la Roque 1927, 102–3, 119; Winlock 1941, 43; Chevrier 1949, 255; Chevrier 1950, 437–8; Bénazeth 1991). This dearth of information constrains to a general approach and not to a precise one. This is why bronze figures are usually classified by the god they depict (e.g. Page Gasser 1990; Grenier 2002; Tiribilli 2018), as in the excellent books on the subject published by Günther Roeder (1937; 1956). Nevertheless, a few accurate works on inscribed bronze figures give an idea of the dating, the provenance and part of the function of some of them, especially on the so-called bronzes de donation (De Meulenaere 1988; Colin 1998, 339; Hill in this volume). Some studies also oriented towards a more theological approach (e.g. on Saïs bronze statues, see Delvaux 1998). A geographical approach is also promoted by Katja Weiss, who brought together all the published bronze figures originating from the Delta (Weiss 2012), and some deposits have also been reassessed and recontextualised thanks to archival research and new excavations at the sites (at Naukratis for example: Masson 2015). North Saqqara’s Sacred Animal Necropolis (SAN) bronze figures — more than 1,800 of which were discovered between 1964 and 1976, and 600 more in 1995–6 — have been studied in their context as a comprehensive assemblage (Davies and Smith 2005; Davies 2007). Sue Davies has stressed the fact that numerous bronze caches are generally found in relation to new constructions, or at least at ‘the level of foundations of buildings’ (Davies 2007, 181); this does not mean that they were used as foundation deposits, but rather that they were most probably hidden after the clearing of chapels when new constructions were erected. She has also

198

F. GOMBERT-MEURICE

suggested that a great proportion of this material was ‘dedicated in connection with public or private celebrations’ (Davies 2007, 185) as well as for prophylactic reasons (Davies and Smith 2005, 32 and 62). In a general article about the bronzes discovered in a chapel of the small temple of Osiris-Iw on the site of ‘Ayn Manawîr in the Kharga oasis (Wuttman, Coulon and Gombert 2008),2 it was also suggested that they were probably not a votive production offered by pilgrims — which is the usual interpretation for this type of object3 — but that they were rather ‘involved in certain ceremonies in a secondary manner’ (Wuttman, Coulon and Gombert 2008, 172). More recently, Dieter Kessler and Nasr el-Dine supported a similar interpretation for the bronze statuettes discovered in a sacred animals’ cemetery at Tuna el-Gebel (Kessler 2008; Nasr el-Dine 2010; Nasr el-Dine 2015). Dieter Kessler pointed out the relation between the deposits of bronzes in the funerary galleries and the Osirian festivals, an interpretation that had already been partly promoted by Pierre Koemoth (Koemoth 1993). Significantly, he based his analysis on the relationship between bronze figures and mummies as religious objects on the archaeological site. In order to better understand the rituals behind the deposits of bronze figures, I would like to focus on the archaeological contexts and to highlight the importance of analysing groups of bronze figures as a whole even when they seem to be secondary (at least) deposits. Interesting lines of enquiry arise from a comparison of the group of bronzes found at ‘Ayn Manawîr, in the small village temple active during the Persian period, with the one discovered in the so-called Serapeum of Memphis, one of the most important religious complexes of Egypt where worship flourished over centuries.

In 1994, the excavations of the temple of ‘Ayn Manawîr under the supervision of Michel Wuttman (IFAO) provided 389 bronze statuettes,4 374 of which represented Osiris or were fragments belonging to Osiris figures.5 The settlement is well documented, with studies of the surrounding environment showing that the village depended on qanat (Wuttmann et al. 1996; 1998).6 Furthermore, the rich demotic documentation on pottery ostraca gives a lot of evidence about the organisation of the village. Information about the community, its temple and the history of some families is provided in 461 texts. They also date the village to the first Persian domination (from the 5th to 4th century BC with 390 BC as a terminus) and allocate the temple to Osiris-Iw. This Osiris-Iw is the main god of the Roman temple at Douch and the town itself was called Pr-Wsir-Iw (the domain of Osiris-Iw). The overwhelming presence of Osiris figures in this temple should not, therefore, be a surprise even if no inscription gives the precise identity of the figures represented. The statuettes were deposited in two lateral chapels (chapels F and F’), one on top of the other, which were directly connected to the hypostyle courtyard and to another chapel (chapel E) located at the back of the temple (Wuttmann et al. 1996, 394, fig. 4). A total of 375 figurines of Osiris were spread over the 3.09 × 1.73m chapels, covered by a type of Nubian vault. According to the latest archaeological and architectural study conducted by Arnaud Gigante (publication in preparation), their access was not straightforward. The partly underground chapels could only be accessed by short staircases, which themselves were

2

5

3

4

For the first study of these bronzes by Peter Dils see Wuttmann et al. 1996, 432–5. See, for instance, Ziegler 1985, 43: ‘les innombrables statuettes étaient probablement offertes par les pèlerins dont la mémoire demeurait ainsi auprès des dieux de Memphis. Leur accumulation à l’intérieur des lieux saint posait un cas de conscience aux prêtres qui trouvèrent cette solution pour s’en débarrasser sans les profaner’. I would like to thank Laurent Bavay, Director of the IFAO, for his authorisation to publish some results of my study of this corpus here.

‘Ayn Manawîr bronze figures, a reference assemblage?

6

Other types are rare and include: two Isis nursing Horus, two unidentified goddesses, one man in an offering posture and another one bearing offerings, as well as a cat with a recumbent ram, one bull pendant and another pendant in the shape of a dog-headed god. See also Michel Chauveau (EPHE–IVe section) and Damien Agut-Labordère (CNRS-ArScAn) on the Achemenet site: – ArScAn-HAROC, Nanterre. Hosted by TGIR Huma-Num, Paris.

THOUSANDS OF OSIRIS

199

Fig. 1: Chapel F’ of the temple of ‘Ayn Manawîr and the access to the hypostyle courtyard in 1994. © Institut français d’archéologie orientale, Le Caire.

Fig. 2: Naos located in the northwest corner of the lateral chapel of ‘Ayn Manawîr temple, with a statue of Osiris in wood and filled up with bronze statuettes of the same god, in 1994. © Institut français d’archéologie orientale, Le Caire.

reachable only through a wooden trapdoor. It seems that the room was still in service when it collapsed. Indeed, the accesses to the hypostyle courtyard and to the other chapel (E) situated behind were not blocked.7 This means that the bronze figures were discovered in a position which reflects their last liturgical use, a rather unusual situation for this type of material. Since the chapels are clearly connected with the rest of the building, the bronze figures do not represent the usual deposit buried to clear the ritual space prior to a new (re-)construction. Religious reasons, such as a reference to the tomb of Osiris, probably explain their

underground location (Fig. 1). In this respect it is important to note the concentration of the statuettes at the northwest corner of the chapel (Fig. 2). The Osiris statuettes seem to have been deposited in a concentrated and organised way (Fig. 3), on and around the main statue of Osiris and its wooden naos. Some conclusions can be drawn from the first analysis of the corpus of the statuettes found there. On the one hand, the assemblage shows a great variety of style and technologies, even if most of the bronze figures are solid cast — except 17 — and have a very flat profile. Their posture varies widely, which challenges, once

7

Statues were also found in an area (D) at the bottom of the temple.

200

F. GOMBERT-MEURICE

Fig. 3: Three Osiris figures stuck together by concretions, in 1995 before cleaning. ‘Ayn Manawîr inv. no. 3668: H. 18.7cm, W. 5.5cm, D. 0.3cm, H. from shoulders to heels 10.5cm; inv. no. 3669: H. 12.8cm, W. 3.3cm, D. 0.2cm, H. from shoulders to heels 7.1cm; inv. no. 3670: H. 10.3cm, W. 2.8cm, D. 0.2cm, H. from shoulders to heels 5.5cm. © Institut français d’archéologie orientale, Le Caire.

8 9

No. 3384A,B,C,D. An interesting parallel and technological explanation is provided by finds made in a small foundry workshop discovered in a tomb at Qubbet el-Hawa, where, for example, very flat Osiris statuettes were produced alongside a larger Harpocrates solid cast figure (Fitzenreiter et al. 2014).

more, Roeder’s theory about arm position as a criterion for the geographic origin of bronze statuettes (Coulon 2016a, 513; see also Ziegler 1978, 33 who suggested that bronze figures were transported by travellers, which could explain the variety of the god’s positions, an interpretation that is not well followed). The height from shoulders to heels — which, after study, I defined as the best dimension for the categorisation of Osiris bronze figures ranges from 3cm to 29cm with a majority of statuettes measuring between 5 and 7cm high. Most of the statuettes (354) show a solar disc on the atef crown. Four Osiris bronze figures discovered on the same wooden base share significant similarities (Fig. 4).8 Their height including pins varies from 6.6cm to 6.9cm, but they are strictly equal in height without the pin (H. 3,5 cm). They have clearly been produced from the same mould. Sorting by size and type allows the creation of specific categories of Osiris bronze figures, determining those that were produced according to the same model or even from the same mould (Fig. 5).9 Finally, interrelating some criteria, it is possible to distinguish thirty-one groups of two to eight Osiris figures which seem to come from the same mould, or at least were produced following the same model. It concerns 116 Osiris figures. Precise observation shows that some of them were worked again after the first mould impression of the wax, in order to highlight details, such as the uraeus, the eyes, the feathers or the sceptres. Some have been modelled on the rear side to give the shape of the buttocks; others have been plainly twisted at the level of the bust, in a longitudinal direction or at the rear of the neck, to give the illusion of the volume of a body (Fig. 6). These meticulous observations establish altogether specific types.10 Why were two to four copies made of some statuettes of Osiris, which were then put on the same base or at least in the same deposit? Can it be a simple question of the wealth of the donor? The inscribed bronze figures of the Saite period (664–525 BC), called ‘bronze de donation’ by De Meulenaere, inform us that the benefit of the cult was returned through bronze figures to the member of the clergy who was in charge of the cult.11 Should we link the repeated small figures

10

11

A typology will be published for that assemblage in order to give some simple criteria for comparison of one set with another. For the characterisation of the clergy in charge of bronze deposits, especially in oracular settings, see Kessler 2008, 161.

THOUSANDS OF OSIRIS

201

with the multiplicity of low clergy? One could interpret each bronze figure, even without inscription, as a special dedication done for the profit of a priest (important or not), a kind of declension of the bronze de donation. A critical issue is to find out the frequency of those deposits of bronze figures: were they produced year after year or all at one time? It is yet to be determined when and how the priests were involved in rituals and deposits of bronzes. However, the assemblage can at least be related to a specific liturgy around Osiris, carried out by priests, with some groups of two or four Osiris figures most likely dedicated together. As Dieter Kessler has already pointed out for the Tuna el-Gebel assemblage (see above), it seems that Fig. 4: Four bronze figures on the same wooden base discovered the hypothesis of pilgrims of varying in chapel F of ‘Ayn Manawîr (inv. no. 3384 A, B, C, D), after cleaning. wealth dedicating those statuettes does H. (max, including pin) 6.9cm, W. 8cm, H. from shoulders to heels 3.5cm. not fit the ‘Ayn Manawîr bronze finds. © Institut français d’archéologie orientale, Le Caire. The re-duplication of small bronze figures of the same iconography on the details. Were those details related to a specific ritual same base might be a question of liturgy, the reflection and feast in accordance with Osiris-Iw or to a local of a theology operating during specific feasts, like the Osiris form? one invoking different forms of Osiris (Coulon 2009; In any case, the classification of the Osiris figures 2015). This multiplication of Osiris figures on one of ‘Ayn Manawîr certainly supports a new reading of base, which does not correspond to any of the numerthe rituals to which they were related. Besides, the preous categories established by Roeder, is also wellcise stylistic and technological approach of the ‘Ayn attested at other sites such as Edfu, Medinet Habu or Manawîr assemblage provides criteria for comparison Saqqara.12 Sometimes, small pendants show two to five Osiris figures cast together and sharing a base (Fig. 7). with other sets of bronze figures discovered outside So there is no doubt that it is a special iconography. the oasis, such as in Karnak (Gombert-Meurice 2018a). Furthermore, the type of production discovered in ‘Ayn A comparison of this group with other sets discovered Manawîr is close to the one found in the Hibis temple, all over Egypt can also shed light on the practice of 13 Kharga oasis (Winlock 1941). A great majority of the depositing bronze figures, especially Osiris figures, figures wear a solar disc on their crown, and figures and especially during the first Persian domination. The with loops are rare. The Serapeum bronze figures, for finds at the Serapeum of Memphis are one of these instance, present highly different proportions for these sets.

12

See, for example, three Osiris figures of different scales (maximum height of 7.2cm) and sharing the same base, discovered in the Serapeum of Memphis (Daressy 1906, pl. XIX, CGC 38357). Some double figures of Osiris, yet to be published, have an inscribed base and some bases without statues may also give new information.

13

While the context is different, the objects are similar (about them: Winlock 1941). I would like to thank Marsha Hill for giving me the opportunity to see some of these statues.

202

F. GOMBERT-MEURICE

Fig. 5: Two figures of Osiris manufactured from the same mould, discovered in the temple of ‘Ayn Manawîr. Left (inv. no. 3781): H. 9.8cm, W. 2.2cm, D. 0.25cm, H. from shoulders to heels 5.4cm. Right (inv. no. 3615): H. 9.8cm, W. 2.2cm, D. 0.2cm, H. from shoulders to heels 5.4cm. © Institut français d’archéologie orientale, Le Caire.

Fig. 7: ‘Ayn Manawîr inv. no. 3488: H. 3.4cm, W. 1.5cm, D. 0.6cm, H. from shoulders to heels 2.3cm. © Institut français d’archéologie orientale, Le Caire.

Fig. 6: Two Osiris figures impressed on the face and twisted on the rear to give the illusion of the volume of a body. Left (inv. no. 3306): H. 8cm, W. 1.9cm, D. 0.7cm, H. from shoulders to heels 4.5cm. Right (inv. no. 3318): H. 7.1cm, W. 1.8cm; D. 0.7cm, H. from shoulders to heels 4cm. © Institut français d’archéologie orientale, Le Caire.

THOUSANDS OF OSIRIS

A group of bronze figures of Osiris found in Saqqara (Louvre)

203

The Serapeum is probably one of the most important archaeological sites for bronze objects. With all the discoveries made during the multiple excavations, more than 4,000 bronze figures have been registered, some of which are inscribed (Maspero 1882; Davies 2007). The objects found by Mariette in the Serapeum have been mainly published by Christiane Ziegler (Ziegler 1978; 1985).14 Since then, a more complete copy of the Inventaire Mariette has appeared,15 including more than 6,331 entries with ten, twenty, thirty or hundreds of lots of objects including 1,627 bronze figures or cult instruments. It gives an account of the excavations that is substantially different from the first publication edited by Maspero in 1882. That publication was a collection of Mariette manuscripts put together in chronological order, without reflecting the changes in Mariette’s perspective over the period in which he produced them. Allusions to the same discovery of bronze figures can be found in different chapters of the book, summarised in different ways (e.g. Maspero 1882, 38 and 77). Parts of the text seem to be a sort of notebook written during the excavations themselves and other parts seem to be done from memory, and with some flights of fancy.16 Because of their more or less literary style they made a great impression, and were honoured with a mention in the main official accession register of the Musée du Louvre, the Inventaire E. One century after the discovery, in 1944, forty-two objects were registered and said to have been found under the pavement of the way going to the Serapeum and sent to the

Louvre by Mariette. It is a short and misleading piece of information.17 It is also mentioned that they were despised in antiquity because of their crudeness and, as a result, they remained non-inventoried in the storerooms of the Louvre. Table 1 below, which follows information given in the Inventaire Mariette, reveals that the bronzes were found in many more locations. Four main areas yielded bronze figures: Nectanebo’s pylon, the southern precinct wall, the northern wall and the foundations of the pavement located in the western pylon.18 The register also provides the date of the finds, with information on the distribution of each group of bronze figures. For example, in the area called ‘Pylon’, Mariette discovered, on 12 May 1851, seven Harpocrates statuettes; twelve days later, two statuettes of Thoth, five of Anubis and sixteen figures of Apis; and, finally in June, nine statuettes of Isis. At the foot of the southern wall of the pylon of Nectanebo, he found thirty-three statuettes of Harpocrates and thirty-five figures of Apis, etc. The inventory shows, therefore, that the finds were not randomly mixed in terms of types, but more or less grouped by deities.19 In total, there are at least forty-six different gods or types of ritual bronze objects quoted, with 460 objects described as ‘divers bronzes’.20 This situation seems to be a continuation of the one observed in the North Saqqara SAN, where 75 percent of the pieces have been discovered in hidden caches each containing from five to more than two hundred objects (Davies 2007, 178). In both cases the bronze figures show a wide array of types, although Osiris still dominates the corpus with about 500 statuettes in the SAN and about 468 figures discovered in three main places during Mariette’s excavations.

14

17

15

16

Her study was based on the ‘Catalogue provisoire fait en Égypte’ written by Mariette (register MS 7DD10) which includes 3,054 entries registered from October 1851 to February 1852. MS 7DD10bis. The title of the register written by Mariette himself is ‘Registre pour servir à l’inscription des monuments sous les numéros qu’ils avaient à leur départ du Sérapéum pour Paris. Cahier n°4’. It has been completed by a note by Théodule Devéria ‘M. Mariette possède à Boulaq le double de ce cahier’. It includes objects discovered from 28 October 1850 to 19 January 1854. ‘En soulevant le dallage du dromos, nous nous sommes aperçus que le terrain sur lequel on l’a posé, est parsemé de statuettes de bronze, représentant des divinités. Tantôt les statuettes sont isolées, le plus souvent on les trouve par tas énormes et confusément entassés les unes sur les autres. Un de ces tas nous en donne 260, un autre plus de 300’ (Maspero 1882, 38).

18 19

20

Inventaire E of the Louvre Museum: ‘Les numéros E 17125 à 17167 ont été retrouvés par Mariette sous le dallage de l’allée conduisant au Sérapéum et envoyés par lui au Musée du Louvre. Méprisés dans l’antiquité à cause de leur grossièreté, ils avaient étés, pour la même raison laissés dans les réserves du Louvre sans être inventoriés. Ils avaient été décapés par M. André et inventoriés à la suite conformément à la décision du Comité du 30 Novembre 1944’. See Maspero 1882, 35, pl. E. We assumed that the date inscribed is really the date of the discovery, but it could also be the date when Mariette registered the objects. All of those ‘various’ bronze objects were discovered in 1853 in a hurry, which could explain why Mariette was not as precise as previously, when he gave the name of the god represented. 419 of the ‘divers bronzes’ came from the ‘Mur d’enceinte, à 30 mètres [au] Nord du Pylône’.

204

F. GOMBERT-MEURICE

Table 1: Find-spots of bronze figures recorded in the Inventaire Mariette. Find-spots of bronze figures

Year 1851

‘Pylône de Nectanébo’ ‘Mur d’enceinte du nord’

Year 1852

Year 1853

Total

756

756

18

18

‘Mur d’enceinte, à 15 mètres Nord du Pylône’

1

1

‘Mur d’enceinte, à 30 mètres Nord du Pylône’

423

423

‘Mur d’enceinte du sud’

372

372

‘à 30 m au sud du Pylône’

20

20

‘Dans les fondations du trottoir construit du temps d’Ouaphrès. – côté Sud contre la colonnade de Ramsès I’

4

4

‘Dans les fondations du mur d’enceinte de l’Est – à 30m du Pylône’

3

3

21

21

1

1

‘Dans les fondations du trottoir dallage situé à l’Ouest du Pylône’ ‘Dans les fondations du trottoir construit du temps d’Ouaphrès’ ‘Dans le sable, près de la Chambre bleue. Pastophorium’

1

‘Sous les fondations des gros murs trouvés au sud de la Chambre bleue’

1

1

2

3

1

1

5

1,149

3

475

1,627

Location not indicated in the register Total

Furthermore, one of the SAN caches shows tight organisation, with some statuettes surrounding a larger statue of Osiris in a naos (Davies and Smith 2005, 188; Davies 2007, 178). In this deposit, statuettes are not thrown together but arranged in an almost geometrical order, organised around the main statue. This situation echoes the ‘Ayn Manawîr bronze figures’ organisation. According to the groups of bronze statuettes still stuck together by sandy concretion that I have been studying, this was probably also the case for some Osiris and Isis statuettes discovered in the Serapeum. Is it possible to identify some of those secondary (or more) bronze deposits as parts of a preserved ritual by comparing them with the ‘Ayn Manawîr findings, which we assume to be still involved in a liturgy during the Persian period? Examples such as the bronze figures found within a vase discovered in the Osiris Wennefer Neb-Djefau chapel in Karnak, as parts of an Osirian ritual, invite us to pay attention to the multiplicity of cases that could occur in such ‘secondary’ contexts (Coulon 2016b, 27–35). We are certainly still far from being able to produce a classification of bronze statuettes according to the original or later rituals to which they were linked, but it is one of the paths to explore further.

21

Sandy concretions characterise the whole lot, even sometimes after a (bad) cleaning. Of course, this criterion would not be pertinent outside the Louvre collection, which has restricted

1

Thanks to Mariette’s register, the labels, the documentation, and also a specific sandy concretion,21 it is now possible to identify 1,064 bronze figures kept in the Louvre storeroom as coming from the Serapeum (Gombert-Meurice 2018b, 48–57). Among them, there are 505 Osiris figures, more than the 468 registered in the Inventaire Mariette. Most of them were still in sandy concretions or stuck together and not registered (Fig. 8). The analysis of this group of Osiris figures provides some clear preliminary results. •





First, as has been pointed out before (Ziegler 1978, 41–2), the archaeological contexts prove that the bronze figures were mainly produced before Nectanebo I’s reign (380–362 BC) and that they were hidden at that time, while the onomastic approach tends to date most of them to Dynasty 26. According to the copy of the Inventaire Mariette 7DD10bis, 392 figures were found around Nectanebo’s pylon, 70 in the southern precinct wall and only 6 figures were discovered in the northern wall, which indicates how concentrated in specific places these bronze statuettes were. Their shoulder-to-heel dimensions show that around 90 statuettes of Osiris are under 5cm, around 190

archaeological provenances for bronze figures. For example, the concretions found on the Serapeum bronzes differ greatly from the earthy concretion of the Tanis group figures.

THOUSANDS OF OSIRIS

Fig. 8: Figure of Osiris with the head of another Osiris statuette stuck with concretion and corrosion. Inv. no. E.33193: H. 16.8cm; W. 4.2cm; D. 4.5cm, H. from shoulders to heels 9cm. © 2015 Musée du Louvre, dist. RMN-Grand Palais/Christian Décamps.





22 23

between 5 and 8cm, and around 40 between 8 and 11cm. Then, the number of Osiris figures per group reduces drastically, from about a dozen per size range down to just two statues of about 40cm high (always without the pin).22 Categories of Osiris figures can be determined, just as in the ‘Ayn Manawîr assemblage. From an iconographic point of view, it should be noted that only ninety Osiris statuettes have a solar disc on their atef crown, which is a significantly smaller proportion than was observed in the ‘Ayn Manawîr and Hibis assemblages. Another important difference between this group and the ‘Ayn Manawîr set concerns the presence of at least one loop on some sixty-eight Osiris statuettes at the Serapeum. Fifty-three of them have a loop on the base, and when a loop appears on the base, it is systematically on the right part, at the foot of the Osiris figure (Fig. 9). The most feasible explanation is that these loops relate to gestures performed with the figures, or positions they had to

N 5152 A, H. 40.2cm and N 5152 B, 41.2cm. This was one of the main surprises revealed by the radiography of thirty-six Osiris figures carried out by Benoît Mille from the

205

Fig. 9: Loop on the right side of the base of an Osiris bronze figure. Louvre inv. no. E 33234, H. 11.4cm, W. 3.2cm, T. 3.7cm, H. from shoulders to heels 6.9cm. © 2015 Musée du Louvre, dist. RMN-Grand Palais/Christian Décamps.





maintain, during rituals. The loops have previously been interpreted as possibly being used to fix textiles on to the statue (Koemoth 1993, 172). A lot of the statuettes which have not been restored still have linen impressions captured in corrosion products. Wrapping a statuette with linen certainly had a religious significance, perhaps for the sacralisation of the statuettes, or to imply that they were, like the god Osiris in the theology, in a situation of death before being reborn (Kessler 2008, 160; Widmer 2013). As for the technical aspect, there are a lot of solid casts with very thin Osiris figures, but also a great variety of techniques, many more than expected.23 This point is of primary interest, since it suggests that the statuettes were probably not produced in the same workshop at one time.

This brief overview of the Louvre bronze figures of Saqqara, summarised from more precise observations, tends to show a situation much more complex than the

C2RMF (Centre de Recherche des Musées de France). This study is yet to be published.

206

F. GOMBERT-MEURICE

one observed in the ‘Ayn Manawîr temple. In order to restore groups, various criteria of dimensions, iconography and technology must be interpreted in combination with each other. However, as has been observed for the oasis assemblage, in a more simple way, it reinforces the idea of the recurrence of groups, with series of Osiris figures of the same dimensions, and with some constants such as the position of the loop, and other typological details. Toward comparisons for a ritual interpretation Closer observation of the numerous groups or sets of repeated bronze figures, comparing their technology and typology with reference groups such as the ‘Ayn Manawîr one, and giving due attention to their archaeological context, will slowly but surely help to define their ritual involvement. The thousands of Osiris figures discovered all over Egypt give a general feeling of repetition. However, the actual levels of similarities and variations within that quantity of bronze figures needs to be studied as a subject in itself. To begin with, systematic standards should be applied in the analysis of bronze assemblages, standards I will partly propose in the publication of the ‘Ayn Manawîr bronze figures. Publishing Osiris bronze figures with their dimensions from shoulders to heels will probably help to compare the groups properly. Then, a general examination should indicate the presence or absence of a solar disc on the crown and of a pin beneath the feet and the existence and position of loops, as well as whether the statuette was produced as a solid or hollow cast. Whenever possible, additional technological details need to be checked for the smallest figures: was the wax just impressed on the front or the rear partly remoulded? With those very simple criteria it will be possible to share and compare assemblages from one excavation to another. Eventually, the catalogue of bronze figures will perhaps be replaced by a catalogue of feasts and details of local religious calendars and practices… Bibliography Bénazeth, D. 1991. Tod. Les objets de métal. San Antonio, TX. Bisson de la Roque, F. 1927. Rapport sur les fouilles de Médamoud (année 1926). Fouilles de l’Institut français d’archéologie orientale du Caire 4. Cairo.

Chevrier H. 1949. Rapport sur les travaux de Karnak. 1948–1949. Annales du Service des antiquités de l’Égypte 49, 241–67. ———. 1950. Rapport sur les travaux de Karnak. 1949–1950. Annales du Service des antiquités de l’Égypte 50, 429–67. Colin, F. 1998. Les Fondateurs du sanctuaire d’Amon à Siwa (désert lybique), autour d’un bronze de donation inédit. In W. Clarysse, A. Schoors and H. Willems (eds), Egyptian religion: The last thousand years. Studies dedicated to the memory of Jan Quaegebeur. Orientalia Lovaniensia Analecta 84. Leuven, 329–55. Coulon, L. 2009. Une trinité d’Osiris thébains d’après un relief découvert à Karnak. In C. Thiers (ed.), Documents de théologies Thébaines Tardives (D3T 1). Les Cahiers « Égypte Nilotique et méditerranéenne » 3, 1–18. ———. 2015. Du périssable au cyclique: Les effigies annuelles d’Osiris. In S. Estienne, V. Huet, F. Lissargue and F. Prost (eds), Figures de dieux. Construire le divin en images. Rennes. ———. 2016a. Les statuettes d’Osiris en pierre provenant de la Cachette de Karnak et leur contribution à l’étude des cultes et des formes locales du dieu. In L. Coulon (ed.), La Cachette de Karnak. Nouvelles perspectives sur les découvertes de Georges Legrain. Bibliothèque d’étude 161. Cairo, 505–65. ———. 2016b. Les chapelles osiriennes de Karnak. Aperçu des travaux récents. Bulletin de la Société française d’égyptologie 195–6 (Jun–Oct), 16–35. Daressy, G. 1897. Notice explicative des ruines de Médinet Habou. Cairo. ———. 1906. Catalogue général des antiquités égyptiennes du Musée du Caire N° 38001–39384, Statuettes de divinités, Vol. I. Cairo. Davies, S. 2007. Bronzes from the Sacred Animal Necropolis at North Saqqara. In M. Hill (ed.), Gifts for the gods: Images from Egyptian temples. New York, 174–87. Davies, S., and Smith, H. S. 2005. The Sacred Animal Necropolis at North Saqqara. The falcon complex and catacomb. The archaeological report. Egypt Exploration Society Memoirs 73. London. Delvaux, L. 1998. Les bronzes de Saïs, les dieux de Bouto et les rois des marais. In W. Clarysse, A. Schoors and H. Willems (eds), Egyptian religion: The last thousand years. Studies dedicated

THOUSANDS OF OSIRIS

to the memory of Jan Quaegebeur. Orientalia Lovaniensia Analecta 84. Leuven, 551–68. De Meulenaere, H. 1988. Bronzes égyptiens de donation. Bulletin des musées Royaux d’art et d’histoire 61, 63–81. Fitzenreiter, M., Loeben, C. E., Raue, D. and Wallenstein, U. (eds). 2014. Gegossene Götter. Metalhandwerk und Massenproduktion im Alten Ägypten. Rahden/Westf. Gombert-Meurice, F. 2018a. La statuaire de bronze des chapelles osiriennes de Karnak. Rapport préliminaire. Saison 2017. . ———. 2018b. Les bronzes du Sérapéum de Memphis. Une enquête et sa méthode. Grande Galerie. La Recherche au Musée du Louvre 2017, Hors série 48–57. Grenier, J.-C. 2002. Les bronzes du Museo gregoriano egizio. Aegyptiaca Gregoriana 5. Vatican City. Kessler, D. 2008. Einwickeln und unterirdische Ablage von Bronzen im Tierfriedhof von Tuna el-Gebel. Hildesheimer ägyptologische Beiträge 50, 155–63. Koemoth, P. 1993. Le rite de redresser Osiris. In J. Quaegebeur (ed.), Ritual and sacrifice in the ancient Near East. Orientalia Lovaniensia Analecta 55. Leuven, 157–74. Maspero, G. 1882. Le Sérapéum de Memphis par Auguste Mariette Pacha publié d’après le manuscrit de l’auteur. Paris. Masson, A. 2015. Bronze votive offerings. In A. Villing, M. Bergeron, G. Bourogiannis, A. Johnston, F. Leclère, A. Masson and R. Thomas, 2013– 19, Naukratis: Greeks in Egypt. British Museum Online Research Catalogue. . Nasr el-Dine, H. 2010. Bronze d’ibis provenant de Touna el-Gebel. Bulletin de l’Institut français d’archéologie orientale 110, 235–49. ———. 2015. À propos de Sekhmet à Touna el-Gebel. In N. Castellano, M. Mascort, M. Piedrafita and J. Vivò (eds), Ex Aegypto lux et sapientia. Hommage al professor Josep Padro Parcerisa. Nova Studia Aegyptiaca IX. Barcelona, 411–24. Page Gasser, M. 1990. Götter bewohnten ägypten. Bronzefiguren der Sammlung ‘Bible+Orient’ der Universität Freiburg Schweiz. Freiburg/CH; Göttingen.

207

Roeder, G. 1937. Ägyptische Bronzewerke. PelizaeusMuseum, Hildesheim. Glückstadt; Hamburg; New York. ———. 1956. Ägyptische Bronzefiguren. Mitteilungen aus der ägyptischen Sammlung, Königliche Museen zu Berlin 6. Berlin. Tiribilli, E. 2018. The bronze figurines of the Petrie Museum from 2000 BC to AD 400. Golden House Publication 28. London. Weiss, K. 2012. Ägyptische Tier- und Götterbronzen aus Unterägypten. Untersuchnungen zu Typus, Ikonographie und Funktion sowie der Bedeutung innerhablb der Kulturkontakte zu Griechenland. Part 1. Ägypten und Altes Testament 81. Wiesbaden. Widmer, G. 2013. À propos de quelques dédicaces sur lin de l’époque romaine: Une pratique votive méconnue? In E. Frood and A. McDonald (eds), Decorum and experience: Essays in ancient culture for John Baines. Oxford, 185–92. Winlock, H. E. 1941. The temple of Hibis in Khargeh Oasis. Part I. The excavations. The Metropolitan Museum of Art, Egyptian Expedition XIII. New York. Wuttmann, M., Barakat, B., Bousquet, B., Chauveau, M., Gonon, T., Marchand, S., Robin, M. and Schweitzer, A. 1998. Deuxième rapport préliminaire des travaux sur le site de ‘Ayn Manawir (oasis de Kharga). Bulletin de l’Institut français d’archéologie orientale 98, 367–462. Wuttmann, M., Bousquet, B., Chauveau, M., Dils, P., Marchand, S., Schweitzer, A. and Volay, L. 1996. Premier rapport préliminaire des travaux sur le site de ‘Ayn Manawir (oasis de Kharga). Bulletin de l’Institut français d’archéologie orientale 96, 385–451. Wuttman, M., Coulon, L. and Gombert, F. 2008. An assemblage of bronze statuettes in a cult context: The temple of ‘Ayn Manawir. In M. Hill (ed.), Gifts for the gods: Images from Egyptian temples. New York, 167–73. Ziegler, Ch. 1978. Une découverte inédite de Mariette, les bronzes du Sérapéum. Bulletin de la Société française d’égyptologie 81, 29–45. ———. 1985. Les Osiris-Lunes du Sérapéum de Memphis. In S. Schoske (ed.), Akten des Vierten Internationalen Ägyptologen Kongresses. Studien zur altägyptischen Kultur, Beiheft 3. Hamburg, 441–5.

LOOKING FOR CONTEXTS: RECENT WORK ON THE KARNAK CACHETTE PROJECT Laurent COULON*, Yves EGELS**, Emmanuel JAMBON*** and Emmanuel LAROZE****

Abstract

An incomplete archaeological process

The Karnak Cachette, excavated by Georges Legrain between 1903 and 1907, is one of the most fascinating discoveries of Egyptian archaeology and an exceptional source for the study of Egyptian statuary (e.g. Bothmer 1960, 151–3; De Meulenaere 1998). However, the recontextualisation of the finds remains a desideratum, as the huge quantity of statues found in this favissa can hardly be used as pieces of evidence for reconstructing the history and archaeological evolution of the temples of Karnak when deprived of any precise connection with their context beyond the mere label ‘found in Karnak’. Even this label should sometimes be considered with caution, as it may be a matter of discussion whether a statue belongs to the Cachette corpus. This contribution will present an overview of the different issues raised by the ‘contexts’ of statues, which have been addressed by the Karnak Cachette Project since its beginning in 2006.

As far as ‘old’ discoveries are concerned, a historiographical approach is a priority in order to understand the nature of the documentation which is available (and which is lacking) and to produce new data from it. As several historiographical surveys of the Cachette’s discovery and its scientific exploitation have recently been published (Azim and Réveillac 2004; Jambon 2009; Coulon and Jambon 2016), we will restrict ourselves here to a few summary statements, with a special focus on the uncertainties surrounding the delimitation of the corpus.

In October 1901, Legrain, following the instructions given by Gaston Maspero, began the systematic clearance of the courtyard between the southeast corner of the Great Hypostyle Hall and the seventh pylon. Legrain started by excavating the base of the walls of the courtyard and the feet of the seventh pylon itself. Within a few months, he found there two pillars of Senwosret I, two sphinxes of Amenhotep II, twelve statues and a stela which had originally adorned the north face of the seventh pylon, and many loose blocks, in particular those of a gate of Amenhotep I (Azim and Réveillac 2004, I, 254–73; II, 182–200). The depth reached by Legrain in this area is not clear but the level could be the same as that of the pillar of Senwosret I or the fragments of the gate of Amenhotep I, which is 72.34m (Legrain 1904a, 31). It is significant to note that this level seems to correspond approximately to the base of the foundation of the seventh pylon. Thus, by the end of Legrain’s seventh mission in Karnak, the southern third of the courtyard had been thoroughly excavated. During the next archaeological season (1902–3), Legrain devoted time to sorting out the blocks already discovered, especially those of the gate of Amenhotep I, which he thought he could rebuild in the courtyard (Azim and Réveillac 2004, I, 264–8; II, 189–94). However, some blocks were still missing, and Legrain obtained permission from Maspero to continue the excavation of the courtyard heading north in order to look for them. The work started there midDecember 1903, after the infiltration waters had receded. The discovery of the Cachette itself occurred on 26 December 1903, with the first group of finds including the stela of Sety I (see below, ‘The context

*

***

The historiographical context and the limits of the corpus

**

École Pratique des Hautes Études, Université Paris Sciences Lettres (EPHE, PSL), UMR 8546, Paris. École Nationale des Sciences Géographiques (ENSG) / Institut National de l’Information Géographique et Forestière (IGN), Paris.

****

Institut für die Kulturen des Alten Orients (IANES), Tübingen. Centre National de la Recherche Scientifique (CNRS), UMR 8167, Paris.

210

L. COULON, Y. EGELS, E. JAMBON AND E. LAROZE

of burial’). In the subsequent months, Legrain continued to excavate around this location and 5m south of these first finds, he reached, in June 1904, the level of 67.15m.1 The next season, the pit became a deep chasm: ‘We are now reaching 15 or 16 metres (it’s dizzying, when one is at the top!) and we are still making discoveries’ (Legrain’s letter to Maspero, 27 May 1905 [Ms. IdF 4027]). More than eight hundred statues, sixteen to seventeen thousand bronze statues and many other objects were found during four campaigns between 1903 and 1907 (Azim and Réveillac 2004, I, 275–335; II, 201–302; Jambon 2009). The excavations were brought to an end owing to the risks to which the workers were exposed and the low number of objects discovered during the 1907 season (Azim and Réveillac 2004, I, 283). The work was resumed in this area between 1955 and 1958 by S. Adam and F. el-Shaboury in order ‘to clear the whole courtyard down to virgin soil’, but it is unlikely that this goal was attained according to the poor documentation, mostly photographs, which is available to us (Biston-Moulin and Boraik 2017, 40, n. 9). Although Legrain himself, like other Egyptologists of his time, immediately began to study and publish some of the objects discovered in the Cachette (Coulon and Jambon 2016, 95–100), the legacy of this extraordinary discovery was somewhat disrupted by a series of unfortunate circumstances, of which the First World War and the untimely death of Legrain in Luxor (in August 1917) were the most serious. After that, and for more than three decades, there would be only isolated progress. It is only from the 1950s onward that the work really resumed, with the patient inquiry conducted by Bernard V. Bothmer and Herman De Meulenaere (Coulon and Jambon 2016, 105–12). Later, substantial progress was made at the beginning of the 21st century, the year 2004 being particularly auspicious, with an exhibition presented in Grenoble (Goyon and Cardin 2004) and the publication of the seminal work of Michel Azim and Gérard Réveillac on Legrain’s photographic archives (Azim and Réveillac 2004). However, the valuable catalogue of finds published by Azim and Réveillac remained incomplete, as it was

based only on the photographs available to the authors,2 a situation which was not remedied by other resources, such as Porter and Moss, Topographical Bibliography. To overcome this limitation, a database project was launched in 2006 at the Institut français d’archéologie orientale (IFAO), enhanced in 2008 by the signature of a cooperation protocol with the Egyptian Ministry of State for Antiquities, then the Supreme Council of Antiquities (SCA) (Coulon and Jambon 2016, 114–7).3 The creation of a database was a response to the need for a flexible tool to deal with this massive and somewhat polymorphic material. Only a modifiable database could, on the one hand, offer updated information for each object and, on the other, provide an overview of all items from the Cachette, allowing users to search this corpus according to various criteria. The Karnak Cachette database was put online in November 2009, and has been regularly updated thereafter. In its second version, which was put online in January 2012, more than eight thousand photographs were included, those taken by Bothmer for the Corpus of Late Egyptian Sculpture and those taken by our team at the Egyptian Museum in Cairo between 2008 and 2010. The overwhelming majority of the corpus consists of statuary, of which private and royal statues of various kinds and different periods today form the lion’s share.4 This results from the fact that most of the bronze statues, mainly statues of deities and especially of Osiris, mysteriously disappeared after the excavations (Jambon 2009, 271–5; for the few nowadays in Cairo Museum, see B-CK 858, 957, 1054). Beside statuary, the Cachette has also yielded many stelae (for example B-CK 32, 687, 813, 831, 950), but also pieces of sacred furniture from the temple, such as a series of five altars and one inscribed measure of capacity, all with the name of Thutmose III (respectively B-CK 234, 376, 1062–4 and 238). Damaged bronze hieroglyphic signs with cramps (B-CK 818–20, 857, 914–6) or small pieces of faience for inlay work (for example B-CK 855, 893 or 910) are also likely to be part of this same furniture, but it is, however, difficult to grasp the function in Karnak of a piece of jewellery such as the ring of Nefertiti (B-CK 721), or of a cosmetic jar in

1

3

2

The measurements were calculated through the photogrammetric processing of Lythgoe’s photographs, see below, ‘The context of burial’. For instance, only a part of the photographs kept in the Kuentz’s archives were included in this book. For the second part, see Coulon and Jambon 2016, 100, n. 46.

4

. The objects in this database are quoted here by their “B-CK” [Base Cachette de Karnak] number. See Legrain 1906a; Legrain 1909; Legrain 1914; Josephson and el-Damaty 1999; and the online database Cachette of Karnak: .

LOOKING FOR CONTEXTS: RECENT WORK ON THE KARNAK CACHETTE PROJECT

the shape of a hedgehog with a woman lying on it (B-CK 908).

211

Although the catalogue of Legrain’s finds in the Cachette has been much enriched by recent research, it must be admitted that to encompass the totality of the content of the Cachette is still (and is doomed to remain) an unreachable aim. Firstly, as already mentioned, the excavations were left unfinished in 1907 and the work carried out between 1955 and 1958 did not prove sufficient to complete them. It is, therefore, most probable that all the elements initially belonging to the Cachette have not been unearthed yet. As for those which were found during the excavations, some of them disappeared soon after their discovery. We have already alluded to the mysterious fate of the thousands of bronze statuettes excavated there, for which, at least, there is still a hope that, one day, they will be rediscovered. There can be no such hope for the ‘true shoal of wood furniture and statues’ that Legrain ‘fished’ from the muddy waters of the Cachette in May–June 1905. He had no technical way to preserve these fragile items and despite his efforts, these ‘ten cubic metres’ of wooden objects, once put in contact with fresh air, left nothing but a ‘thick brown mud on the fingers’ and ‘the regret of not having been able to save these fugitive wonders from the oblivion where they have instantly fallen’ (Legrain 1906b, 145–6). The absence of a precise register of the finds made by Legrain in the Cachette is also undoubtedly a major obstacle to its exact appreciation. For some objects, the provenance is doomed to be doubtful, as there is not sufficient evidence to assert that they were found in the Cachette. Thus, three stone statues of Osiris which were formerly part of the Fr. W. von Bissing collection and are now kept in Munich and Hanover (B-CK 1215–7) could have possibly been found in the Cachette, but a Medinet Habu provenance cannot be discounted (Coulon 2016b, 507–9). This absence of reliable inventory has been only partially filled through the recent rediscovery of some notebooks of the long-

lost (and still partially missing) Journal de fouilles of Legrain, which are now kept at the Louvre Museum.5 One of these diaries, entitled ‘Cachette’, which covers the period between 5 December 1903 and 3 April 1904, includes the list of the 162 first K numbers given by Legrain (Fig. 1). This list is essential to reconstruct the chronology of the finds and the assemblages of statues transferred from the temple to the Cachette. We had already, within the framework of the IFAO–SCA cooperation protocol, scrutinised the Cairo Museum registers, in particularly the Journal d’Entrée and the Temporary Register. Both contain valuable information on the history of the objects, especially regarding the ‘K numbers’ inventory, namely the list of the objects discovered by Legrain in Karnak mainly in the Cachette (Azim and Réveillac 2004, 293–5; Jambon 2009, 245–51, 256–64). This inventory can be completed, at least for the 162 first entries, by the list from Legrain’s notebook ‘Cachette’ (Jambon forthcoming). However, it should be kept in mind that the finds made by Legrain in the Cachette and in other places at Karnak and its immediate neighbourhood were registered by him in the same inventory, as exemplified by the first number of the list, ‘K1’, the group statue of Sn-nfr and Sn(t)-n(Ꜣ)y (Cairo CG 42126), which was found to the north of the Great Hypostyle Hall. In some cases, the data collected from the recently found archival documents allowed us to dispel the doubts concerning the provenance of an object with a ‘K’ number. Thus, the group statue of PꜢ-šrj-n-tꜢ-jswy, ʿšꜢ-jḫt and Nfrt-jj.w (K5, Cairo JE 36576), whose exact provenance was until recently uncertain, is now securely attributed to the Cachette based on the information provided by Legrain’s letters and diaries together with photographs and notes taken by Lady Cecil during her visit to him.6 In other cases, the attribution to the Cachette is an option, but remains hypothetical. Thus, the main fragment of a large statue of PꜢ-dj-Jmn-jpt (Cairo T.R. 27/1/21/1) has been included in the Karnak Cachette database (B-CK 1197), like many fragments registered as ‘Karnak Legrain’s finds’ in the Egyptian museum’s registers, but its exact provenance is unknown (Coulon 2016c, 99). Moreover, a wooden box

5

6

The fuzzy boundaries of the corpus

Briefly discussed in Coulon and Jambon 2015; detailed studies of some of these papers will be published in a forthcoming volume of the Cahiers ‘Égypte Nilotique et Méditerranéenne’.

See the commentary on B-CK 5: (last consulted 16 June 2018). For Lady Cecil’s testimony, see Cecil Reid 2016, 188–9, fig. 13.

212

L. COULON, Y. EGELS, E. JAMBON AND E. LAROZE

recently found in the basement of the Egyptian Museum in Cairo has yielded many stone fragments, most probably gathered by Legrain, which are possibly to be ascribed to the seventh pylon’s favissa (Selim 2016, 249, with ref.). In one case, a join was even made between one of these fragments and the statue Cairo JE 37143 found in the Cachette (Selim 2010, 275, 282–3). Though, in this case, it is very probable that the two parts of the statues have the same provenance, it should be noted that different fragments belonging to one statue were sometimes found inside and outside the Cachette (Bothmer 2004 [1981], 352, n. 8). The context of burial Obviously, the first type of context which is to be taken into account concerning the statues found in the Cachette is the cache itself. Legrain’s diaries have provided us with a precise evocation of the first days of the discovery, including some sketches of the disposition of the first objects during December 1903 and the beginning of January 1904. One of these sketches Fig. 1: A page of Legrain’s diaries with the beginning of the list of the statues discovered in 1903–4, mostly in the Cachette. Cahiers Legrain 1903–1904. (Fig. 2) shows the position of the large © DAE, Musée du Louvre. stela of Sety I (B-CK 686), laid on a layer of sand containing some silt covering a group of statues of various periods (Legrain 281–2; Coulon, Jambon and Sheikholeslami 2011, 27). 1904b, 272–3; the first three statues are B-CK 3, 17 Nonetheless, the somewhat blurred picture resulting and 18). However, our knowledge of the archaeological from these circumstances can be improved, by using context of the rest of the objects found, that is their various archival material. precise position inside the cache, and the contours of the pit itself, is limited. At the time of Legrain, stratiFrom archival photographs and plans to graphic methodology was not commonly used in excaphotogrammetry vations. Even if it had been, the very specific conditions of these digs, in the Nile Valley in the early 1900s As already mentioned, the spatial documentation (with not only the seasonally rising water table, but concerning the excavations in the Cour de la Cachette also the inundation itself still very active), would have is very poor. Though Legrain reported some measuremade its use very uncertain. Suffice it to recall that, ments or some sketches in his diaries, these indications when Maspero wrote a vivid report to explain to do not allow the stratigraphy of the Cachette to be the French educated public what was happening in the reconstructed clearly. In this respect, the photographs Cachette, he did it under the heading ‘Fishing for stattaken during Legrain’s clearances are a more signifiues in the Temple of Karnak’ (Maspero 1910 [written cant source of information. They provide insights into in February 1905]; see Azim and Réveillac 2004, I, the condition of the work — the number of workers,

LOOKING FOR CONTEXTS: RECENT WORK ON THE KARNAK CACHETTE PROJECT

213

could work in theory, this specific use of photos that were not originally intended for photogrammetry remains a challenging issue. To the best of our knowledge, it seems that such an exploitation of a photographic archive had never been tried before in archaeology. Concerns remain specifically in the lack of metadata relative to the pictures. Photogrammetry requires technical information such as the focal length and the size of the ‘captor’, which are now available in the EXIF (exchangeable image file format) files issued by digital cameras. Unfortunately, the exact nature of the photographic material used by Legrain and the other photographers who took pictures of the excavations is unknown (Azim and Réveillac 2004, I, 76). Moreover, the pictures which have been collected are of widely varying quality and mostly of poor resolution. In spite of these difficulties, the context of the Cachette courtyard was a perfect framework to undertake experimentation and to improve the technical process. The compilation and the detailed analysis of the pictures was the first step. All the pictures were documented by author and date, numbered, resized Fig. 2: Sketch showing the location of the first objects discovered in the Cachette. December 1903. Cahiers Legrain 1903–1904. and sorted chronologically. For each © DAE, Musée du Louvre. scene, the position and the orientation of the camera have been roughly determaterials, positions of the ramps, berms and archaeomined; the data were positioned on a map to highlight logical artefacts — and they are, moreover, a useful possible associations of pictures which could work as complement to the textual notes. At the present time, stereoscopic pairs. An automatic correlation system building on Azim and Réveillac’s catalogue, our team based on SIFT (scale-invariant feature transform) point has gathered fifty-two pictures related to the Cachette’s matching has also been tested to match up the images. excavations, taken by Legrain himself or, more often, This second process was much more efficient than the by visitors and tourists. This corpus of photographs is, first one by bringing into focus several groups of therefore, characterised by a great disparity of origin photos (Fig. 3). Among these results, a more precise and quality. In regard to the site as shown in the picanalysis allows us to detect two batches of pictures. tures, it appeared to us that some of the photographs The most interesting one consists of seven photocould be used as stereoscopic pairs. In other words, we graphs which were taken one after the other by the assumed that it could be possible to retrieve 3D inforsame photographer, A.M. Lythgoe, during the first mation through these sets of old pictures. While it campaign of excavations, between May and June 1904.7

7

The images are now kept at the Metropolitan Museum in New York. See Goyon and Cardin 2004, 16.

Fig. 3: The identification of corresponding points by the SIFT algorithm has helped to highlight some image groups with common points (green squares in the table). © YE.

214 L. COULON, Y. EGELS, E. JAMBON AND E. LAROZE

LOOKING FOR CONTEXTS: RECENT WORK ON THE KARNAK CACHETTE PROJECT

215

Fig. 4: Group of seven photos of the same scene. The identification of corresponding points by the SIFT algorithm has helped to bring out the fact that they could be arranged in three panoramic images. © A. M. Lythgoe, MMA New York.

Furthermore, the identification of corresponding points by the SIFT algorithm showed that this batch was in fact a set of overlapping photographs intended to create panoramas of the same scene, taken from three different positions (Fig. 4). The images were then stitched together to finally set up three panoramic views. The second group is made up of two photos taken on 21 January 1904 (Fig. 5).8 The two shots looking south were probably taken from the top of the northern wall of the Cour de la Cachette from two different positions. These two overhead views show a very

8

Published as figs 11 and 12 in Cecil Reid 2016, 187. We are indebted to the author for allowing us to include these photographs in this article.

interesting group of scattered blocks. In spite of an excellent overlap, the two images cannot be easily processed due to the lack of control points on both pictures. The walls of the Cour de la Cachette are visible at the top of one image only. This situation, therefore, makes the positioning of the model in the space impossible. Once the potential assessment of our archives had been completed, a linkage between the pictures and the field was required, which involved taking measurements in the field of some details visible on the old

216

L. COULON, Y. EGELS, E. JAMBON AND E. LAROZE

Fig. 5: Pair of images tested for photogrammetric process. The calculation of this set of pictures is problematic owing to the lack of control points on both images. The walls where these points can be measured are visible on only one of the shots. © Coll. A. Cecil Reid.

LOOKING FOR CONTEXTS: RECENT WORK ON THE KARNAK CACHETTE PROJECT

pictures. Fortunately, the context had not dramatically changed: the four large walls surrounding the courtyard have remained about the same over the past century. Basically, it was necessary to measure natural points previously spotted in the old pictures with a theodolite (total station). However, this task would have taken too long in the field and would have been restricted to a selection of pictures only. Therefore, it was thought preferable to make a complete textured 3D model of the context, as recent technological advances in photogrammetry allow topographic 3D models of any context to be produced quite easily. In September 2015, topographic data were collected during a short campaign carried out by the authors at Karnak. In the field, several hundred photographs were taken, not only in the Cour de la Cachette (58 × 48m) but also in other parts of the temple which are visible in the archive photographs, such as the standing obelisk or the first and eighth pylons. A topographic survey of the area was also completed and geo-referenced inside the Karnak temple system, which guarantees its permanent validity. The complete 3D model of the Cachette was computed with Photoscan software. The archival photographs were processed using a suitable photogrammetric method based on the Redresseur and Cumulus softwares developed by Yves Egels. The Redresseur software has been improved in order to deal with the underlying problem of the material used: the old cameras were equipped with decentered lenses, creating distortions that must be taken into account. Firstly, three parameters had to be determined for each image: the camera position, the orientation and the focal length. Secondly, automatic correlation was processed when the overlapping of the pictures was sufficient. For example, the berm facing north — clearly visible on one of Lythgoe’s pictures — has been successfully restituted in volume with this method. Curves were also calculated. Nevertheless, the calculation of a 3D model has been restricted to this single case. The coordinates of points were indeed mostly determined manually. This operation undertaken on the Redresseur software consists in linking common details — the socalled ‘tie points’ — visible on at least two pictures. Control points already determined in xyz are also required to scale and orient the model. After

9

The unique plan of the Cachette courtyard drawn by Legrain is entitled ‘plan des fouilles de 1901–1902 au VIIe pylône’ (digital version: Cfeetk n°61940).

217

calculation, the coordinates of the tie points were determined. The spatial information gained with this photogrammetric method were put together on plans (Figs 6–7) and sections (Figs 8–9). These synthetic documents were also completed with other data such as a plan,9 as well as sketches and notes, recorded by Legrain. They allow our knowledge of the Cachette stratigraphy to be understood at a glance. It has been observed that the diagrams provided by Legrain, although on the surface very simple, are in fact very accurate. For example, the measurements on the one which illustrates the letter sent to Maspero (written 19 April 1904; see Azim and Réveillac 2004, 289) are fairly consistent with the data provided by the photogrammetric process using Lythgoe’s pictures. A burial made in a single process Another important source for reconstructing the context of the cache is the aforementioned ‘K numbers inventory’, which made possible the partial reconstruction of the chronology of the excavations, and, conversely, the establishment of a rough ‘chronostratigraphy’. The latter shows, for example, that royal monuments were mostly found together in the Cachette, from which it can reasonably be inferred that they were buried together. More surprisingly, this list shows too that the ‘scribe statues’ belonged to the same excavated areas (Jambon 2016, 141–7). In general, this study has led to the conclusions that, although there were New Kingdom statues from the top to the bottom of the Cachette, the number of the most recent statues, from the Late Period to the Ptolemaic era, increased progressively as the excavation became deeper. A very interesting example of this is provided by the statue of Plato son of Plato (B-CK 608; Coulon 2001), which can be securely dated to 98 BC and was found in May 1905, very deep in the Cachette. Thus, if recent objects were buried together with more ancient ones, from the top to the bottom of the Cachette, it is difficult to suppose that there were several successive trenches in the Cachette, and it is most likely that this repository was created at one time, as Legrain himself concluded (Legrain 1905, 66; Jambon 2016, 136–9).

Fig. 6

LOOKING FOR CONTEXTS: RECENT WORK ON THE KARNAK CACHETTE PROJECT

The purpose and date of the Cachette To conclude this section on the context of burial, it may be worthwhile to say a few words concerning the raison d’être of the Cachette of Karnak. From the very first discoveries, Legrain sought to establish the date on which the Cachette had been created as well as its purpose. To these ends, he paid close attention to the condition and the relative position of the objects he found and took a great interest in their dating, in particular for the latest ones. From these observations, he suggested that the burial must have taken place in the last decades of the 1st century BC, when the Theban region had experienced major politico-military troubles. Legrain supposed that the Theban priests wanted to protect the treasures of the Karnak temple from a future attack or a siege (Jambon 2016, 131–2; Jambon forthcoming). Although the dating suggested by Legrain seems to be correct, no real ‘treasure’ has ever been found there, and therefore this scenario has to be corrected. The Cachette may rather be seen as a ritual deposit, a favissa, where objects that were damaged or had fulfilled their ritual purpose would have been discarded. As has been stressed elsewhere, the Karnak Cachette is perhaps exceptional in its wealth and its size, but is by no means a unique phenomenon (Valbelle 2016; Jambon 2016, 155–9; Charloux and Thiers in press). When comparing the Karnak Cachette with other

219

deposits, for example the Luxor Cachette, some common characteristics are observed, such as the presence of a flat object sealing the top of the cache (that is, in the case of the Cachette, the stela of Sety I), the choice of an open area at a strategic place within the temple, and some specific arrangements of parts of the same statue in different locations. In the light of this comparison, we came to the conclusion that, even if the constitution of the Cachette resulted obviously from removals related to concrete problems in the temple, and as such was probably part of a ritual purification, nevertheless it also has specific characteristics which connect it with prophylactic magic and funerary procedures (Jambon 2016, 159– 63). This last possibility is supported not only by the presence of the aforementioned huge quantity of Osiris bronze figures in the Cachette, which would fit this ritual purpose, but also by the discovery in December 2014 of a small cache near the temple of Ptah of Karnak. This hiding-place, carefully excavated, revealed a lot about the conditions of its ritual implementation, the authors reaching the conclusion that ‘the favissa of the temple of Ptah constitutes an exceptional example of the grave of a statue of a god situated close to its main place of worship’ (Charloux et al. 2017, 1202; see also Charloux and Mahmoud in this volume). Although we do not have, up to now, any Egyptian text describing the performance of such a ritualised

Fig. 6: Plan view of the Cachette courtyard with the spatial information determined by photogrammetric process or retrieved from Legrain’s diaries and reports. © E. Laroze, Y. Egels, L. Coulon and E. Jambon. Additional legends referring to the numbers on the plans (1) ‘The excavation has not been pursued as further north and west as would have been necessary to retrieve all the monuments which are still hidden there. By venturing further than I did, I would have risked the collapse of either the west wall, where the treaty of the Khetas is carved, or, on the northern side, of the corner of the hypostyle hall where is the Poem of Pentaour’ (after Legrain 1906b, 144). (2) ‘When, in May and June [1905], we reached a depth of nine and ten metres, we met, in the north part of the Cachette, a true shoal of wood furniture and statues, approximately ten cubic metres’ (after Legrain 1906b, 145). (3) ‘At a depth of eight and nine metres, we have found very brown bones (…). Among the most curious ones, I would mention the maxilla of a little child, the head of a carnivore and a great number of ovines’ bones’ (after Legrain 1906b, 146). (4) ‘The ground consists of a natural occurring deposit, made of sand layer alternating with clay layer deposited according to [Legrain] by the flood’ (after Azim and Réveillac 2004, 281). (5) ‘I think, I would ask you to tear down a part of the west wall, north of the door and south of the Khetas, for we have fourteen metres there, nearly vertical, and still more antiquities’ (from a letter from G. Legrain to G. Maspero, 27 May 1905, Ms. IdF 4027). (6) ‘We are now reaching 15 or 16 metres (it’s dizzying, when one is at the top!) and we are still making discoveries’ (after a letter from G. Legrain to G. Maspero, 27 May 1905, Ms. IdF 4027).

220

L. COULON, Y. EGELS, E. JAMBON AND E. LAROZE

Fig. 7: The central part of the Cachette courtyard (enlarged view of Fig. 6). © E. Laroze, Y. Egels, L. Coulon, E. Jambon.

Fig. 8: South–north longitudinal section of the Cachette courtyard. © E. Laroze, Y. Egels, L. Coulon, E. Jambon.

Fig. 9: East–west section of the Cachette. According to Legrain’s diaries, the western pit would have reached up to 15m in depth. © E. Laroze, Y. Egels, L. Coulon, E. Jambon.

LOOKING FOR CONTEXTS: RECENT WORK ON THE KARNAK CACHETTE PROJECT

burial, the general meaning of the hidden cache itself is sufficiently documented. P. Vernus, who has examined this question in detail (Vernus 1989; 2009; 2016), has convinc+ingly demonstrated that the creation of a cache in a temple must be viewed, beyond its practical motives, from the perspective of a tradition in which one can find what he calls ‘cachettes fabuleuses’. Different kinds of objects were supposed to be hidden in such mysterious places, such as, for example, relics of the body of the dead god Osiris or sacred books written by the gods themselves. Some of these mythological caches were supposed to be forever unfindable; some others, on the contrary, were established by the demiurge for a revelation to come. Though the priests of Karnak probably intended the Cachette of Karnak to be of the first category, Legrain, walking a path frequently followed by wise men in ancient Egypt, discovered this ‘secret place in the temple’ and offered back to his contemporaries a knowledge ‘taken away from humanity since ancient times’ (Vernus 2016, 17–9). The initial context of the statues The limits of textual indications on the settings of statues The question of the initial context of the statues found in the Cachette has been addressed in a recent thesis (Price 2011, 173–213), which allows us to limit ourselves to a brief account of the state of the art, before dealing with specific case studies. As a rule, ancient Egyptian sites have yielded a very large number of statues of all periods, but very few in their original context (e.g. Verbovsek 2007, 264–76), and Karnak is no exception. Of course, a few colossal representations of kings are still visible at the gates of temples, sometimes accompanied by the effigies of their courtiers, but smaller statues found in situ are in general not in their original setting. The texts inscribed on the statues frequently provide information concerning their initial location, as exemplified by one of the most common formulae: ‘May my statue dwell in such or such temple or part of the temple’; but generally, the indications are very imprecise.10 Sometimes the ousekhet (Price 2011, 180–5) or the entrance of the temple (Price

10

See e.g. Cairo JE 37173 (Dynasty 30): snn pn mn(w) m ḥwt-nṯr nt Jmn r ḥḥ ḏt (‘May this statue remain in the temple of Amun forever eternally’).

223

2011, 187–8; Klotz 2015, 67 and 71, n. [i]; JansenWinkeln 2016, 399) is mentioned as a desired place for the monument to be set up, but no precise indication is given which would allow us to determine its exact location. Nevertheless, in some very rare cases, the text is more explicit. Thus, on the back pillar of the statue Cairo JE 37452, found in the Cachette (B-CK 931), which belongs to a priestess called Takhybiat, the inscription reads: ‘May it [i.e. my statue] remain, may it remain in the temple of Osiris of Coptos foremost in the house of Gold!’ (Coulon 2008, 31). The ‘temple’ mentioned here is well-known. It is a chapel located in the northeastern part of the Temple of Karnak. In this particular case, it is possible to know from which part of the temenos of Karnak the statue was removed to be buried in the Cachette. This demonstrates that the objects were removed from different sectors of the temenos of Karnak, not only from the central part of the Amun temple. This removal was probably undertaken according to a systematic plan. Recently, by building on the results of the meticulously reconstructed chronology of the discoveries in the Cachette and by testing an old assumption of G. Maspero’s, it has been argued that groups of statues found in a limited period of time by Legrain had probably been collected from the same area of the temple (Jambon 2016, 139–47). Besides explicit mentions of the location of the statue, information concerning its setting in the temple can be deduced from the religious content of its inscriptions, as the ritual texts and the divinities inscribed on a statue were chosen according to the cultic context in which the monument was placed. This phenomenon can be exemplified by comparing different statues belonging to the same person but positioned (or at least supposed to have been positioned) in different locations inside the Theban temples. This is shown, for instance, by the monuments of a man called Ns-mnw son of PꜢ-dj-Jmn-nb-nswt-tꜢwy: two statues of this priest were discovered by Legrain in the Karnak Cachette: Cairo JE 37178 (Fig. 10a) and Cairo JE 37191 (Fig. 10b). Another one is London, British Museum EA 41561 (Fig. 10c), an osirophorous statue which may be of Saite date and subsequently reused by our Ns-mnw; the British Museum acquired it in 1905

224

L. COULON, Y. EGELS, E. JAMBON AND E. LAROZE

a

b

c

d

Fig. 10: a) Statue Cairo, JE 37178 (© CLES/IFAO); b) Statue Cairo, JE 37191 (© CLES/IFAO); c) Statue London, British Museum, EA 41561 (© Trustees of the British Museum); d) Statue Opet O. 168155-1 (© CFEETK).

LOOKING FOR CONTEXTS: RECENT WORK ON THE KARNAK CACHETTE PROJECT

from the famous antiquity dealer Mohamed Mohassib and it is not unlikely, as stated by PM II2, 165, that this statue had also been found in the Cachette and stolen soon after its discovery. A fourth statue of the same man was recently discovered during Emmanuel Laroze and Guillaume Charloux’s work on the Opet temple, in the southwestern part of Karnak. This is a block statue (Fig. 10d) whose upper part is preserved (see in this volume Charloux and Mahmoud, fig. 4). As was made clear by Guillaume Charloux in his excavation report, this statue was carefully placed in the foundation trench of the temple, when it was rebuilt during the Ptolemaic period. Other statues were also buried in the same way at different places next to the walls of the courtyard: another Late Period statue of a priest, a Ramesside statue of a priest, and a baboon, probably related to the neighbouring temple of Khonsu (Charloux et al. 2012, 277–80). The statue of Ns-mnw is still unpublished but we can make some basic observations from the preliminary description (Virenque, in Charloux et al. 2012,

225

210 and 226) and the photographs available. When compared to his two block statues from the Cachette, it appears that Ns-mnw’s Opet statue shares a common set of biographical formulations with them. For instance, on the three monuments, Ns-mnw insists on the length of his life: eighty-five years!11 Other formulaic statements are common to the inscriptions of all three statues. However, the statue found in the Opet temple has some distinctive features, which clearly indicate its relation to this sanctuary. One of them is the representation of gods on the wig, with captions: two forms of Osiris which are specific to the Opet temple, namely ‘(the one of) Ipet-weret’ and ‘master of the provisions-kꜢw’.12 Another such feature is the identity of the gods mentioned in the proscynema: Osiris of Ipet-weret, Horus son of Isis son of Osiris and Ptah Nefer-her. By contrast, the proscynemas of the Cachette statues only mention the Theban triad and more generally the gods and goddesses of Ipet-sut.

Table 1: The divinities on Ns-mnw’s statues Statues

Divinities invoked

Divinities depicted

Cairo JE 37178

(in the ḥtp-dj-nsw formula) – Amun-Ra Lord of the Throne(s) of the Two Lands, foremost of Ipet-sut, Primeval of the Two Lands, Sacred of arm, Kamutef, king of the gods – Mut, the Eye of Ra, mistress of the sky, mistress of the gods – Khonsu-in-Thebes-Nefer-hotep – Montu-Ra master of Thebes – all the gods and goddesses who are in Ipet-sut

Cairo JE 37191

(in the ḥtp-dj-nsw formula) – Amun-Ra Lord of the Throne(s) of the Two Lands, foremost of Ipet-sut, Primeval of the Two Lands, [Sacred of] arm, Kamutef, king of the gods – Mut, the Eye of Ra, mistress of the sky – Khonsu-in-Thebes-Nefer-hotep – Montu-Ra master of Thebes – Amunet who dwells in Ipet-sut

London BM EA 41561

(ḥsy n) Ptah-Sokar-Osiris

Opet O. 168155-1

(in the ḥtp-dj-nsw formula) – [Osiris] master of provisions (nb kꜢw) – Osiris Ipet-weret Wennefer, king of the gods, master of the sky, – [Isis] the divine mother the earth, the duat, the wat[er and the mountains]13 – [Osiris] Ipet-weret – [Isis] – [Ptah] Nefer-her – Horus son of Isis-son-of-Osiris – Ptah Nefer-her lord of Maat.

11

Statue Opet O. 168155-1, left side, x+2-x+3: fqꜢ.n.f tw(.j) m rnpt 85; statue Cairo JE 37178, back pillar: ꜤḥꜤ.f m Ꜥnḫ rnpt 85 n Ꜣb n šd jbd.f [...] (cf. Jansen-Winkeln 2001, 93, n. 14); statue Cairo JE 37191, back pillar: [fqꜢ].n.f tw(.j) m rnpt 85.

12

13

On this form of Osiris and its connection with the Opet temple, see Traunecker 2004, 59. On this epithet nb pt tꜢ dwꜢt mw ḏww, see LGG III, 627a-628c; Tillier 2016, 364–5.

226

L. COULON, Y. EGELS, E. JAMBON AND E. LAROZE

In this case, it is possible to observe that, even in a restricted geographical area, the decoration and the inscriptions of the different statues commissioned by a priest were conditioned by the context in which they were intended to be set up. The perspectives offered by the digital processing of inscriptions This kind of recontextualisation of the temple statues based on the content of their inscriptions has not been attempted on a large scale, mostly due to the lack of a tool which would allow us to study the corpus of statues without separating the inscriptions, the decoration of the statue and its layout, and the data we have concerning the prosopographical data, the context of its discovery, etc. This led to the idea of building a searchable electronic corpus of the texts inscribed on the objects from the Cachette, which would be an extension of the Karnak Cachette database. With the development of the digital humanities, a large set of possibilities is now open to create online corpuses which include both inscriptions and images and allow queries on both of them. The TEI (text encoding initiative), based on XML (extensible markup language), has offered us a very productive framework, as it had done already for Greek and Latin epigraphers, for instance, who created their own standard, Epidoc, now widely adopted. This project was developed in collaboration with Vincent Razanajao, who developed a user-friendly tool for the encoding of the transliterated hieroglyphic texts inscribed on the objects from the Karnak Cachette, Xefee; and Emmanuelle Morlock, who is currently working on the publication tool, the first part of the corpus, the Dynasty 26 statues, being currently prepared for publication by Sepideh Qahéri. In contrast to a database which breaks into pieces the content of an object, this marking up does not affect the continuity of the text. Every textual element remains embedded in its context, and each text is embedded in a header, providing metatextual data and references to bibliography and related images (Razanajao, Morlock and Coulon 2013). Eventually, it will become possible to systematically search for textual and metatextual data concerning the initial location of the statues. We have also worked on the implementation of TEI/Epidoc standards in ancient Egyptian epigraphy to serve as a basis for building bridges between text-edition projects within Egyptology and through the collaboration of other research teams involved in digital epigraphy in other fields.

Conclusion As highlighted in the preceding lines, the recontextualisation of the statues discovered in the Karnak Cachette is a wide-ranging undertaking because of the complex history of these objects: in ancient times, they experienced at least one initial installation context, reflected in the epigraphic evidence, and a ritual burial in the Cachette; however, the later burial context may in some way reproduce the first, if one considers a grouped removal from certain spaces of the sanctuary. In modern times, the turbulent history of excavation and conservation of objects has made their recontextualisation a major historiographical issue, which cannot be dissociated from the Egyptological study of these monuments. In both cases, it appears that the use of new photogrammetric technologies and digital humanities offers powerful tools that can partly compensate for the loss of excavation information or the dispersion of objects. Bibliography Azim, M., and Réveillac, G. 2004. Karnak dans l’objectif de Georges Legrain. Catalogue raisonné des archives photographiques du premier directeur des travaux de Karnak de 1895 à 1917. Vols I and II. Paris. Biston-Moulin, S. and Boraik, M. 2017. Some observations on the 1955–1958 excavations in the Cachette Court of Karnak. Cahiers de Karnak 16, 39–51. Bothmer, B. V. 1960. Egyptian sculpture of the Late Period. 700 B.C. to A.D. 100. New York. ———. 2004 [1981]. The block statue of Ankh-Khonsu in Boston and Cairo (Membra Dispersa V), with a contribution by Herman De Meulenaere. In M. E. Cody (ed.), Egyptian art: Selected writings of Bernard V. Bothmer. Oxford, 337–54. [First published in Mitteilungen des Deutschen Archäologischen Instituts, Abteilung Kairo 37]. Cecil Reid, A. 2016. The Princess and the Karnak Cachette. Princess Henry of Battensburg’s travels on the Nile – 1903/04 as described in Lady William Cecil’s Journal. In Coulon 2016a, 177–91. Charloux, G., Angevin, R., Marchand, S., Monchot, H., Oboussier, A., Roberson, J. and Virenque, H. 2012. Le parvis du temple d’Opet à Karnak. Exploration archéologique (2006–2007). Travaux du Centre franco-égyptien d’étude des temples de Karnak. Bibliothèque Générale 41. Cairo.

LOOKING FOR CONTEXTS: RECENT WORK ON THE KARNAK CACHETTE PROJECT

Charloux, G., Thiers, C., Abd Al-Aziz, M., Abady Mahmoud, M. A., Boulet, S., Bourse, C., Guadagnini, K., Laroye, J., and Sayed el Nasseh, A. M. 2017. The afterlife of Egyptian statues: A cache of religious objects in the temple of Ptah at Karnak. Antiquity 91 (359), 1189–204. Charloux, G. and Thiers, Chr. In press. Le temple de Ptah à Karnak III. La favissa. Travaux du Centre franco-égyptien d’étude des temples de Karnak. Coulon, L. 2001. Quand Amon parle à Platon (La statue Caire JE 38033). Revue d’égyptologie 52, 85–112. ———. 2008. La nécropole osirienne de Karnak sous les Ptolémées. In A. Delattre and P. Heilporn (eds), ‘Et maintenant ce ne sont plus que des villages ...’ Thèbes et sa région aux époques hellénistique, romaine et byzantine. Actes du colloque tenu à Bruxelles les 2 et 3 décembre 2005. Papyrologica Bruxellensia 34. Bruxelles, 17–31. ———. 2016a. La Cachette de Karnak. Nouvelles perspectives sur les découvertes de Georges Legrain. Bibliothèque d’étude 161. Cairo. ———. 2016b. Les statues d’Osiris en pierre provenant de la Cachette de Karnak et leur contribution à l’étude des cultes et des formes locales du dieu. In Coulon 2016a, 505–63. ———. 2016c. Padiaménopé et Montouemhat. L’apport d’une statue inédite à l’analyse des relations entre les deux personnages. In Ph. Collombert, D. Lefevre, St. Polis and J. Winand (eds), Aere Perennius. Mélanges égyptologiques en l’honneur de Pascal Vernus. Orientalia Lovaniensia Analecta 242. Leuven, 91–120. Coulon, L. and Jambon, E. 2015. Les cahiers de Georges Legrain. Grande Galerie. Le Journal du Louvre 32, 52–56. ———. 2016. L’exploitation scientifique de la Cachette de Karnak, de Georges Legrain à nos jours. Essai d’historiographie. In Coulon 2016a, 89–129. Coulon, L., Jambon, E. and Sheikholeslami, C. M. 2011. Rediscovering a lost excavation. The Karnak Cachette. KMT. A Modern Journal of Ancient Egypt 22, 18–32. De Meulenaere, H. 1998. The Cachette of Karnak. In F. Tiradritti (ed.), Egyptian treasures from the Egyptian Museum in Cairo. Vercelli, 334–55. Goyon, J.-Cl. and Cardin, Chr. 2004. Trésors d’Égypte: La ‘cachette’ de Karnak (1904–2004). Exposition en hommage à Georges Legrain à l’occasion du IXe Congrès International des Égyptologues,

227

Musée dauphinois, Grenoble, 4 septembre 2004–5 janvier 2005. Grenoble. Jambon, E. 2009. Les fouilles de Georges Legrain dans la Cachette de Karnak (1903–1907). Nouvelles données sur la chronologie des découvertes et le destin des objets. Bulletin de l’Institut français d’archéologie orientale 109, 239–79. ———. 2016. La Cachette de Karnak. Étude analytique et essais d’interprétation. In Coulon 2016a, 131–75. ———. Forthcoming. ‘Nous avons rencontré une sorte de cachette …’ Le Journal de fouilles retrouvé, Georges Legrain et la Cachette de Karnak (décembre 1903–avril 1904). In L. Coulon and V. Rondot (eds), Actes de la table ronde: Les cahiers de l’égyptologue. Notes et journaux de Georges Legrain. Paris, Musée du Louvre 9 décembre 2015. Cahiers Égypte Nilotique et Méditerranéenne. Jansen-Winkeln, K. 2001. Biographische und religiöse Inschriften der Spätzeit aus dem Ägyptischen Museum Kairo. Ägypten und Altes Testament 45. Wiesbaden. ———. 2016. Zu Kult und Funktion der Tempelstatue in der Spätzeit. In Coulon 2016a, 399–410. Josephson, J. and el-Damaty, M. 1999. Statues of the XXVth and XXVIth Dynasties. CGC 48601–48649. Cairo. Klotz, D. 2015. The cuboid statue of Ser-Djehuty, master sculptor in Karnak. Los Angeles County Museum of Art 48.24.8 + Cambridge University, Museum of Archaeology and Anthropology 51.533. Revue d’égyptologie 66, 51–109. Legrain, G. 1904a. Second rapport sur les travaux effectués à Karnak du 31 octobre 1901 au 15 mai 1902. Annales du Service des antiquités de l’Égypte 4, 1–40. ———. 1904b. Rapport sur les travaux exécutés à Karnak du 28 septembre 1903 au 6 juillet. Annales du Service des antiquités de l’Égypte 4, 265–80. ———. 1905. Renseignements sur les dernières découvertes faites à Karnak. Recueil de travaux relatifs à la philologie et à l’archéologie égyptiennes et assyriennes 27, 61–82. ———. 1906a. Statues et statuettes de rois et de particuliers, I. Catalogue général des antiquités égyptiennes du Musée du Caire. Cairo. ———. 1906b. Nouveaux renseignements sur les dernières découvertes faites à Karnak (15 novembre 1904–25 juillet 1905). Recueil de travaux

228

L. COULON, Y. EGELS, E. JAMBON AND E. LAROZE

relatifs à la philologie et à l’archéologie égyptiennes et assyriennes 28, 137–61. ———. 1909. Statues et statuettes de rois et de particuliers, II. Catalogue général des antiquités égyptiennes du Musée du Caire. Cairo. ———. 1914. Statues et statuettes de rois et de particuliers, III. Catalogue général des antiquités égyptiennes du Musée du Caire. Cairo. LGG = Leitz, C. (ed.), Lexikon der ägyptischen Götter und Götterbezeichnungen, vol. I–VIII. Orientalia Lovaniensia Analecta 110–116 and 129. Leuven, 2002–3. Maspero, G. 1910. La pêche aux statues dans le temple de Karnak. In G. Maspero, Ruines et paysages d’Égypte. Paris, 161–75. 2 PM = Porter, B., and Moss, R. L. B. 1960–. Topographical bibliography of ancient Egyptian hieroglyphic texts, reliefs and paintings. Second edition. Oxford. Price, C. 2011. Materiality, archaism and reciprocity: The conceptualisation of the non-royal statue at Karnak during the Late Period (c. 750–30 BC). Dissertation, University of Liverpool. Razanajao, V., Morlock, E. and Coulon, L. 2013. The Karnak Cachette texts online. The encoding of transliterated hieroglyphic inscriptions. Poster at The Linked TEI: Text encoding in the Web. TEI Conference and Members Meeting 2013, 2–5 October 2013, Rome. . Selim, H. 2010. Statue fragments from Karnak Temple in the basement of the Egyptian Museum in Cairo.

Bulletin de l’Institut français d’archéologie orientale 110, 275–88. ———. 2016. A fragment of the Hathor-headed naos sistrum from a sistrophorous statue of Senenmut from Karnak. In Coulon 2016a, 249–53. Tillier, A. 2016. Un linteau au nom d’Auguste. Karnak Varia (§4). Cahiers de Karnak 15, 357–69. Traunecker, C. 2004. Dimensions réelles et dimensions imaginaires des dieux d’Égypte: Les statues secrètes du temple d’Opet à Karnak. Ktéma 29, 51–65. Valbelle, D. 2016. Statues enterrées, dépôts liturgiques et différentes catégories de favissae. In Coulon 2016a, 21–33. Verbovsek, A. 2007. Befund oder Spekulation? Der Standort privater Statuen in Tempeln des Alten und Mittleren Reiches. In B. Haring and A. Klug (eds), Funktion und Gebrauch altägyptischer Tempelräume. 6. Ägyptologische Tempeltagung. Königtum, Staat und Gesellschaft früher Hochkulturen 3,1. Wiesbaden, 257–76. Vernus, P. 1989. Caches au trésor dans l’Égypte pharaonique. Hathor. Estudos de egiptologia 1, 33–40. ———. 2009. Cachettes. Cachettes fabuleuses. Cachettes royales. In P. Vernus, Dictionnaire amoureux de l’Égypte pharaonique. Paris, 175–202. ———. 2016. Cachettes dans la civilisation pharaonique: De la trouvaille d’un trésor à Deir el-Médina à la mobilisation de la crypte oubliée dans l’idéologie monarchique. In Coulon 2016a, 7–20.

TAHARQO AND HIS DESCENDANTS: A STATUE CACHE UPSTREAM OF THE FIFTH NILE CATARACT Julie Renee ANDERSON, Salaheldin MOHAMMED AHMED, Mahmoud SULIMAN BASHIR and Rihab KHIDIR ELRASHEED

Abstract Archaeological excavations at Dangeil, a Kushite site located approximately 350km north of Khartoum, Sudan, have focused on a temple of the 1st century AD dedicated to the god Amun. Recently, fragments of several early Kushite royal statues dating to the early 7th century BC were uncovered from within the temple. Several hypotheses have been put forth to account for their presence and subsequent disposition at Dangeil, and one possibility is that they originated in a cache that had been disturbed during the destruction of the later temple. Statue caches containing the same family of early Kushite rulers as those at Dangeil have been found previously at Gebel Barkal and KermaDokki Gel. The discovery of the Dangeil statues, upstream of the Fifth Nile Cataract, necessitates a reevaluation of the history of the early Kushite period and of the causal factors proposed to account for these caches. Dangeil, Sudan – temple and site Since 2000, a team from the National Corporation for Antiquities and Museums, Sudan, in cooperation with the British Museum, has been conducting excavations at Dangeil, a Kushite site upstream of the Fifth Cataract, roughly 350km north of modern Khartoum. The site is substantial, covering 12ha. Excavations thus far have focused on the centre of the site and revealed a large temple of the 1st century AD dedicated to the god Amun (Pls 1, 2 and Fig. 1). The temple, situated in the middle of a quadrilateral temenos enclosure, faces the Nile and measures approximately 131m east– west by 39.5m north–south across the first pylon. Both temple and enclosure wall were constructed of a mixture of materials including locally sourced ferricrete and quartzarenite sandstone, fired bricks and mud bricks, along with reused architectural elements, the origin of which remains uncertain. As found on many late Kushite sites, among them Hamadab and el-Hassa, Dangeil’s walls were set on fired brick foundations and

comprised a mud-brick core faced on the exterior with plastered fired bricks, with the exception being the sanctuary, where the columns were sandstone and the walls were faced with decorated sandstone blocks. Within the remainder of the temple, columns were constructed of standardised fired brick quarters or thirds mortared together to form drums, the column diameter varying between halls and being determined by galleting. The interior and exterior walls of the structure were plastered with lime, either applied as a wash on mud plaster or as a finishing layer on top of a white lime scratch coat, and then painted using hematite and goethite, the sources of which remain as yet unidentified, and Egyptian blue pigments. Many of the halls and much of the processional way were paved with sandstone flagstones, except in the peristyle court where the processional way contained a mixture of materials that varied in shape and size. The late Kushite Amun temple M250 in the Royal City of Meroe, though built almost entirely of sandstone, is of similar size with only temple B500 at Gebel Barkal being larger. Fragments of a sandstone altar bearing the cartouches of Queen Amanitore (1st century AD) were found in the sanctuary. Her prenomen and nomen also were inscribed on the bases of six pairs of ram statues, now fragmentary, that once flanked the processional way in the peristyle court, three pairs being situated to the west and three to the east of a central kiosk (Anderson and Mohammed Ahmed 2013). Additional evidence for another royal name, that of King Natakamani (1st century AD), was also found. It is likely that Amanitore and Natakamani constructed the temple as part of their extensive building programme, which saw palaces and temples constructed or renewed throughout their kingdom, notably for example in the Butana at Wad ban Naqa, Meroe and Naqa, but also further to the north at Gebel Barkal and Amara East (ancient Pedeme). Statue discovery at Dangeil Over the course of several excavation seasons, fragments of statues belonging to early Kushite kings,

230

J.R. ANDERSON, S. MOHAMMED AHMED, M. SULIMAN BASHIR AND R. KHIDIR ELRASHEED

Plate 1: Map showing the location of Dangeil, Sudan. Drawing: C. Thorne.

TAHARQO AND HIS DESCENDANTS: A STATUE CACHE UPSTREAM OF THE FIFTH NILE CATARACT

Fig. 1: Plan of the Amun temple, Dangeil. Drawing J. Dobrowolski. © Berber-Abidiya Archaeological Project.

Plate 2: Amun temple, Dangeil, facing westward towards the first pylon and the Nile. Photograph: Mohamed Tohami. © Berber-Abidiya Archaeological Project.

231

232

J.R. ANDERSON, S. MOHAMMED AHMED, M. SULIMAN BASHIR AND R. KHIDIR ELRASHEED

Plate 3: Dangeil statue fragments in situ in the southeast room of the Amun temple, facing east. © Berber-Abidiya Archaeological Project.

dating from between the 7th and 6th centuries BC, were discovered in Dangeil’s 1st century AD Amun temple. Initially, a granite fist holding a mekes case was uncovered in a pit in the temple’s southwest room. Excavation thereafter of the neighbouring southeast room revealed numerous, randomly distributed, large fragments of several different statues mixed together within the layers of the destruction debris of the temple. These statues included several early Kushite rulers, namely, Taharqo (690–664 BC), Senkamanisken (643– 623 BC), Aspelta (593–568 BC) and one as yet unidentified king (Pls 3 and 4).1 The statues primarily were broken at the neck, hands, knees and ankles; basically,

1

For a discussion of the discovery and description of the fragments found in the southeast room see Anderson and Mohammed Ahmed 2009; 2010a; 2010b; 2014.

the thinnest and weakest structural points. Among the fragments were the base and torso of both Taharqo and Senkamanisken, and the base and head of Aspelta. The faces and inscriptions do not appear to have been purposefully defaced; however, the chest of the Taharqo statue displayed peck-marks and a Meroitic graffito was scratched on the left thigh (Pl. 5). Each statue depicts a bare-chested, kilted ruler in a striding position, with arms at his sides and hands grasping mekes cases. The feet are shod with sandals. The ruler’s titulary was carved in hieroglyphs on the statue’s back pillar. Fragments of the early Kushite statues also were directly associated in the same disturbed archaeological

TAHARQO AND HIS DESCENDANTS: A STATUE CACHE UPSTREAM OF THE FIFTH NILE CATARACT

233

Plate 4: Torso of Senkamanisken in situ in southeast room of the Amun temple, Dangeil. © Berber-Abidiya Archaeological Project.

Plate 5: Torso of Dangeil’s Taharqo statue with peck-marks and graffito visible on the left thigh. © Berber-Abidiya Archaeological Project.

context with pieces from two late Kushite statues carved from sandstone. Other notable fragments discovered elsewhere within the structure included the left foot of the Taharqo statue, which was found in the southern stairwell of the second pylon in a secondary deposit (Anderson, Mohammed Ahmed and Sweek 2012, 74–5, pl. 10), and the lower torso of Aspelta (Anderson, Khidir elRasheed and Suliman Bashir 2017, 159–61) discovered in the peristyle court, in the fill of a pit cut through the processional way east of the kiosk. The

latter fragment joined others found in the southeast room, though was discovered over 70m away in an unrelated archaeological context. This fragment, consisting of the kilt and upper thighs, is approximately half life-size. The hieroglyphic inscription preserved on the back pillar begins with the end of Aspelta’s prenomen. The fragmentary inscription found on the back of the head begins with the Horus name of the king, of which only the top of the serekh is preserved. As the shoulders and upper torso are missing, it is difficult to determine if there was enough room to accommodate

234

J.R. ANDERSON, S. MOHAMMED AHMED, M. SULIMAN BASHIR AND R. KHIDIR ELRASHEED

both the Nebty and Golden Horus names of the king.2 When the extant fragments including the head are assembled, the inscription reads (Pls 6 and 7):3 [nfr-ḫꜤ] [nfr-ḫꜤ wsr-ỉb(?)] [nswt bỉty mry]-kꜢ-rꜤ sꜢ rꜤ ỉ-s-p-l-t Ꜥnḫ ḏt mry rꜤ-ḥr-Ꜣḫty [dỉ Ꜥnḫ] ḏd wꜢs nb ḏt [Whose appearances are beautiful] [Whose appearances are beautiful] (?) [Whose heart is strong] (?) [King of Upper and Lower Egypt] Ra whose Ka is loved, Son of Ra, Aspelta, May he live forever. Beloved of Ra-Horakhty, Given all life, stability and dominion forever.

The titularies of both Dangeil’s Taharqo and Aspelta statues describe the king as ‘Beloved of Ra-Horakhty’ presumably followed by a now absent or partial place name. This section is missing on the Senkamanisken statue. Taharqo’s back pillar reads (Pl. 8): [ntr nfr] nb tꜢwy nb ỉr ḫt nsw-bỉty ḫw-nfrtm-rꜤ sꜢ rꜤ tꜢ-h-r-ḳ [mry] rꜤ-ḥr-Ꜣḫty ḥr ỉb ms[t(?) dỉ Ꜥnḫ dd wꜢs nb mỉ rꜤ] dt [The perfect God]. Lord of the Two Lands. Lord of Action (ritual). King of Upper and Lower Egypt, Nefertum-KhuRa. Son of Ra, Taharqo [Beloved] of Ra-Horakhty who resides in Ms[t(?) Given all life, stability and dominion like Ra] forever.4

Plate 6: Head of Aspelta, Dangeil. © Berber-Abidiya Archaeological Project.

On the Taharqo statue, Ra-Horakhty is spelt phonetically using the Ꜣḫ bird followed by a single land sign.5 This spelling is somewhat unusual. Similar contemporary statues found at Gebel Barkal and in Kerma-Dokki Gel describe each of the rulers as ‘Beloved of Amun’ followed by their respective geographic locations, such as ‘Beloved of Amun of Pnubs’ (Kerma-Dokki Gel). The relationship between Dangeil and Ra-Horakhty is unknown, but based on the inscriptions found on the royal cult statues, it might be tentatively suggested that in addition to Amun, one of the important gods to which Dangeil was home may have been a form of RaHorakhty, or at least his veneration was perhaps more emphasised at Dangeil than at other sites. The Kushite kings were closely tied to Ra. Piankhy performs rituals

in the temple of Ra at Heliopolis (Eide et al. 1994, 99–100). In the 4th century BC, Nastasen ‘dances before Ra’ (Eide et al. 1996, 479–82), perhaps alluding to a ceremony in the dais rooms found in several Amun temples (for example, Kawa, Naqa, Dangeil etc.),6 but it is premature to draw any conclusions at this point. Though beyond the scope of this paper, it might be further considered as to whether the resident god of every ‘Amun-style’ temple in Kush was actually Amun (Kushite Amani) and how far the adoption of this temple form and associated rituals extended.7 While some buildings such as B500 at Gebel Barkal are without doubt temples to a form of Amun (there being Amun

2

5

3

4

For comparable titles see Eide et al. 1994, 228. The authors are grateful to Marcel Marée for discussions regarding this inscription and its reconstruction. For a discussion concerning potential toponymic suggestions for Ms[…] see Anderson and Mohammed Ahmed 2009, 81–2.

6 7

See further Anderson and Mohammed Ahmed 2009, 81, fig. 2. The authors thank Vivian Davies for his comments regarding this inscription. See further Anderson and Mohammed Ahmed 2006–7. For a discussion of usage of Egyptian concepts and ideology in the Kushite built environment see Török 2002a; 2002b.

TAHARQO AND HIS DESCENDANTS: A STATUE CACHE UPSTREAM OF THE FIFTH NILE CATARACT

of Napata) and this is confirmed textually, in the absence of such proof there is a danger in merely assuming such a dedication and extrapolating associated rituals often founded on Egyptian analogies, solely based upon what might be a superficial architectural likeness. For example, the discovery of the use of sorghum rather than wheat for the Amun temple offerings during the late Kushite period at Dangeil is one such

Plate 7: Back pillar of Aspelta, Dangeil. © Berber-Abidiya Archaeological Project.

Plate 8: Back pillar of Taharqo, Dangeil. © Berber-Abidiya Archaeological Project.

235

236

J.R. ANDERSON, S. MOHAMMED AHMED, M. SULIMAN BASHIR AND R. KHIDIR ELRASHEED

illustration of the dangers of assumption (Anderson and Mohammed Ahmed 2006; Anderson et al. 2007; Anderson and Mohammed Ahmed 2010c). Dangeil is some distance away from the centre of early Kushite power at Napata/Gebel Barkal; however, it seems likely that its statues originated in an earlier royal endowment on the site. Within the first hall, the trench in which the foundations of the temple’s fired brick columns were founded cut through part of a substantial mud-brick building (Pl. 9). Prior to the construction of the temple, this earlier structure had been systematically levelled to its lowest courses. In addition to being cut by the later temple’s foundation trenches, these earlier walls were also cut by its construction pits and associated scaffolding post-holes, and then again later by pits and ovens from the posttemple occupation.

Excavations beneath the late Kushite levels in the southeast room of the temple revealed the remains of a mud floor, and several large mud-brick walls that belonged to an earlier building (Pl. 10). Here the later temple was built directly on top of some earlier walls, sharing their orientation, while in other places its foundations had been cut through them. Like those uncovered in the first hall, these walls had been uniformly levelled in preparation for the construction of the later building. It is evident that a sizeable mud-brick structure preceded the late Kushite temple. It seems not unlikely that this was an earlier Amun temple or sacred structure, and it is probable that the early Kushite royal statues originated from this building. Few artefacts were associated with this earlier phase, but those discovered included early Kushite material such as faience ram’s heads and lotus amulets and shallow wheel-made dishes (Anderson and Mohammed Ahmed 2014, 618–9, pls 19–21). The rate of expansion of the early Kushite rule southward from their capital at Napata remains unknown, as do its southern geographical limits. However, it seems clear that Dangeil was under direct royal control with the establishment of a royal statue cult on the site in the first part of the 7th century BC and, as discussed above, there is evidence to suggest that the 1st century AD temple was constructed on top of an earlier, presumably sacred, Kushite structure. The status of Meroe, the royal city of the later Kushite period located upstream of Dangeil, remains uncertain in the early 7th century BC. Evidence for demonstrably royal construction there appears in the latter half of the 7th century BC, increasing throughout the region during and after the reign of Aspelta (Pope 2014, 30–1). The earliest Kushite royal name found in the Royal Plate 9: Orthophoto of the first hall, Dangeil Amun temple, with mud-brick City at Meroe thus far, despite the foundations of an earlier structure visible, particularly in the centre of the image. Orthophoto: R. Hajduga. © Berber-Abidiya Archaeological Project. numerous excavations conducted

TAHARQO AND HIS DESCENDANTS: A STATUE CACHE UPSTREAM OF THE FIFTH NILE CATARACT

237

Plate 10: Southeast room of the Dangeil temple with earlier mud-brick walls visible, facing east. © Berber-Abidiya Archaeological Project.

there beginning with the University of Liverpool Institute of Archaeology expedition led by J. Garstang in 1909, is a single sandstone block bearing the name of Anlamani (623–593 BC) (Grzymski 2004, 168, no. 148), Aspelta’s predecessor. Early Kushite artefacts have been found around Meroe;8 the question, however, is whether the area is under direct royal control and when, or whether it is simply within Napata’s zone of influence. It is also interesting to note that, despite the increasing amount of archaeological work being conducted in the Butana, namely at el-Hassa, Muweis, Hamadab, Naqa, Wad ban Naqa, Meroe, Awalib, Abu Erteila and Musawwarat es-Sufra etc., the quantity of early Kushite material remains scant and largely is non-royal.

8

For a list and analysis of early Kushite material found at the Royal City and in the West and South Cemeteries see Török 1997a, 15–20, 25–32. For artefacts found elsewhere in the region see Dunham 1957; Vercoutter 1961; Dunham 1963;

It should be apparent that construction by the el-Kurru dynasts at Dangeil does not necessarily entail construction by the el-Kurru dynasts at Meroë farther south, though it may well signal their southward expansion toward Meroë. (Pope 2014, 31, fn. 187)

If, during the early 7th century BC, the early Kushite rulers were expanding their territory southwards, then Dangeil may have served as ‘a point of tangency between spheres of established royal hegemony and growing royal influence’ (Pope 2014, 32). It should be acknowledged, however, that the absence of evidence for something, in this case royal creations and artefacts, does in and of itself not constitute evidence.

Shinnie and Bradley 1980, 13–7, 313; Omar el-Sadig 2002; Edwards 2004, 126–8; and Anderson and Mohammed Ahmed 2009, 84–5, table 1.

238

J.R. ANDERSON, S. MOHAMMED AHMED, M. SULIMAN BASHIR AND R. KHIDIR ELRASHEED

Statue caches at Gebel Barkal and Kerma-Dokki Gel Two groups of broken statues, both of which included the same early Kushite rulers (Taharqo, Tanwetamani, Senkamanisken, Anlamani and Aspelta) as those discovered at Dangeil, have been found at Gebel Barkal and at Kerma-Dokki Gel. Both assemblages were associated with Amun temples and are thought to have originated in caches. George Reisner discovered two disturbed statue deposits at Gebel Barkal in 1916 (Reisner 1917; 1931; PM 1951, 213, 221; Dunham 1970). One was in Temple B800, room B904, while the other, found north of Amun temple B500’s first pylon, was designated B500A. Reisner suggested the statues originally had been erected in Amun temple B500 owing to its relative proximity to both deposits. Although they had been disturbed, the statues could be reassembled and largely were complete (Pl. 11). There was evidence for the careful deposition of the statue fragments, particularly in B904 where the heavy torsos were more or less regularly placed together with apparent care, as were the bases. Fragments from nine statues of the aforementioned five Kushite rulers were unearthed, as well as the statue of a Kushite queen Amanimalel (Sudan National Museum SNM 1843),9 possibly a contemporary of Anlamani. Fragments from individual statues were found in both caches, indicating that they had been created at the same time. For example, Tanwetamani (664–653 BC) (Toledo Museum 49.105) was reconstructed from fragments found in both caches with the right ankle and neck to left knee coming from B500A, and part of the left leg and base with right foot originating in B904 (Dunham 1970, 17). Similarly Senkamanisken (Museum of Fine Arts, Boston MFA 23.731) was likewise divided between the two deposits, with his head coming from B500A, and pieces of the body and base from B904 (Dunham 1970, 21). The Kerma-Dokki Gel (ancient Pnubs) cache, discovered in 2003 by Charles Bonnet, was found intact in a sealed pit east of the columned forecourt of the Amun temple (Bonnet, Honegger and Valbelle 2003; Bonnet and Valbelle 2004; 2005). There is also evidence to suggest methodical placement of the broken statue fragments. The bases, for example, were deposited last and were found in the upper layers of the pit Plate 11: Anlamani statue from Gebel Barkal B 500A and B 801B (SNM 1845). Photograph: Rocco Ricci. © The British Museum.

9

See further Ali Mohamed and Anderson 2012, 60.

TAHARQO AND HIS DESCENDANTS: A STATUE CACHE UPSTREAM OF THE FIFTH NILE CATARACT

239

(Bonnet and Valbelle 2005, 74). The Kerma statues are now housed in the Kerma Museum (Pl. 12). The statues from all three groups, Gebel Barkal, KermaDokki Gel and Dangeil, come from royal workshops and their iconography manifests the ideology of the royal cult and the godking. All are shown in striding position and wear the royal kilt. Three different forms of crown are depicted. The first has the four feathers of Onuris (Anhur) placed on top of a Nubian cap crown. Use of this crown is not restricted to one particular ruler or period within the groups. The second is the Nubian cap crown with double uraei, again not restricted to a particular ruler or period. This crown appears to be a ‘universal symbol of Kushite kingship, [while the addition of Onuris feathers on the Plate 12: Statues from the Kerma-Dokki Gel cache, now in the Kerma Museum. Photograph: J. Anderson. first crown appears associated with his mythical role as] the triumphant warrior “from the offerings to the god, in contrast with the situation in South”’ (Török 1997b, 287), wherein he successfully contemporary Egypt, as L. Török has suggested: ‘As returns from Nubia with Tefnut. The third type of headno intermediary statues of contemporary non-royal pergear shown is the double crown of Upper and Lower sons are known from Kush, it may be supposed that Egypt adorned with double uraei. This could be viewed popular religiosity, the pious contact between god and as a universal symbol of kingship, as specific geomen, was channelled through the cult of royal statues’ graphic associations with Upper and Lower Egypt (Török 1997b, 317). It might be envisioned that many would no longer be applicable after the reign of Tanof these cache statues originally were set up in the forewetamani, if indeed the Kushites even interpreted the court or outer courts of their associated Amun temples double crown in that fashion in the first place.10 The Kerma-Dokki Gel statue of Anlamani is unique so far, (as the statue bases within the Kerma-Dokki Gel temwith its addition of ram horns to the crown, but it cerple’s forecourt suggest, and which the number of statue tainly reflects the close association of the king with the fragments found, as for example in Gebel Barkal B500, divine Amun. also appears to indicate), or in front of the temples Two Senkamanisken statues depict the ruler wearing themselves. The forecourt and temple frontage could be the leopard-skin garb of the sem-priest (Kerma-Dokki a place open to petitioners who lacked access to the Gel statue and Gebel Barkal SNM 1842).11 The king is temple’s cult statue and more sacred interior spaces explicitly shown in the role of the priest who makes (Török 2004, 161–2; see further Török 2009, 349).

10

See Török 1997b, 198–214 for a discussion of early Kushite titulary and archaism.

11

Taharqo is also shown depicted as a priest on wall reliefs at Kawa: see Török 1997b, 316.

240

J.R. ANDERSON, S. MOHAMMED AHMED, M. SULIMAN BASHIR AND R. KHIDIR ELRASHEED

Colossal statues and stone types The cache statues were made of hard stones such as granite, which likely originated in quarries around the Third Cataract and in the area of Aswan, though this has yet to be demonstrated. Many are life-size and some colossal, though the latter has not been the case so far at Dangeil. The difference in colour of stone utilised for statues of the rulers of Dynasty 25, Taharqo and Tanwetamani, and that used for the kings that followed is notable. Stones used in the later statues appear lighter in colour in every instance. It seems that little hard stone is used for statuary after the reign of Aspelta, a time when the Royal City of Meroe rises in prominence. Of the identified royal hard Plate 13: Colossal late Kushite sandstone statues in front of the Sudan National Museum (SNM 23924). Photograph: J. Anderson. stone statues created in the period following Aspelta, none could be deemed colossal in size. There is one of Aramatelqo (Priese 1974; Berlin, Ägyptisches a reduced access to various hard stone quarries and/or Museum 2249), Aspelta’s successor, depicted in hebto craftsmen able to work it. However, there remain sed dress, and another of Akhratan (mid-4th century examples of small, portable figurines of gods in hard BC) from Gebel Barkal Amun temple B500 (Museum stone, such as the granite statuette of the ram-headed of Fine Arts, Boston MFA 23.735). There is also a fragAmun from Barkal Temple B700 dated between the 1st ment from the Royal City at Meroe of a granite statue and 3rd centuries BC (Kendall 2004; SNM 1844), and of Natakamani (1st century AD), 14cm in size, which the late Kushites do create large altars, statues and steis now housed in Liverpool (World Museum 49.47.709). lae in ferricrete sandstone (Pl. 14). Consequently, this As there are, however, numerous undated and unidentichange may be indicative of a different religious focus fied hard stone fragments from, for example, Gebel in late Kushite society, wherein large statues of rulers Barkal, the above general summary may not represent were not as important to ritual life as they had been a truly accurate picture. earlier. It is also possible that the later, smaller, portaFollowing the reign of Aspelta, colossal statues are ble hard stone figurines were created reusing hard stone still created, though not in hard stone and usually not from earlier sculptures.12 Large late Kushite statues primarily depict gods and depicting rulers. Two colossal sandstone statues from were executed in sandstone. The divine statue of the Tabo now standing in front of the Sudan National king appears to be replaced by colossal statues of gods, Museum (SNM 23982) have been variously identified particularly the hunter and warrior gods Arensnuphis as gods or as an unidentified king, possibly Natakamand Sebiumeker.13 Dating to the 1st century BC, ani (Pl. 13) (Ali Mohamed and Anderson 2012, 97) and Sebiumeker (Ny Carlsberg Glyptoteck, AEIN 1082) so these may prove to be the exception. Perhaps the and Arensnuphis (National Museums of Scotland, reduction in number of hard stone royal statues reflects

12

The authors are grateful to Aurélia Masson-Berghoff for this suggestion.

13

These gods may have played a role in validating the kingship. See further Török 2011, 210–11.

TAHARQO AND HIS DESCENDANTS: A STATUE CACHE UPSTREAM OF THE FIFTH NILE CATARACT

241

Plate 14: Late Kushite ferricrete sandstone frog statue from the hafir (a man-made earthen-walled reservoir which fills during the rainy season) at Basa, now in the Sudan National Museum. Photograph: J. Anderson.

Edinburgh, A.1910.110.36) (Pl. 15) from the Temple of Isis M600 in the Royal City of Meroe are two such examples. These later sculptures also appear to have been placed in accessible positions within or in front of the temples in a fashion similar to that of the earlier Kushite royal statues. ‘These divine images in all likelihood fulfilled the same cult functions that were earlier associated with the royal colossi’ (Török 2002a, 302). The creation of the caches – hypotheses Several hypotheses have been proposed to account for the subsequent disposition of Dangeil’s statues, one suggestion being that they originated in a cache that had been disturbed during the destruction of the later temple. If the statue groups at Dangeil, Gebel Barkal and Kerma-Dokki Gel were broken and buried in caches around the same time, and as the same family group of rulers is included in each, then there are several possible explanations. The first proposal is that an external agent was responsible.14 It has been posited

14

For Kushite statue transformations by external agents see also Schorsch, this volume.

Plate 15: Arensnuphis, now in the National Museums of Scotland, Edinburgh (A.1910.110.36). Photograph: J. Anderson.

242

J.R. ANDERSON, S. MOHAMMED AHMED, M. SULIMAN BASHIR AND R. KHIDIR ELRASHEED

that the invading Egyptian army of the Dynasty 26 pharaoh Psamtek II destroyed the statues during his campaign into Kush in the early 6th century BC (595– 589 BC). Following their destruction, the statue fragments were buried in caches (i.e. Bonnet and Valbelle 2005, 164–71; Bonnet 2011; Valbelle 2011, 31; 2012, 51; see further Morkot 2000, 303–4). Within this theory, the responsibility for the breakages and caches could either lie wholly with the invading army, or more likely, be divided between the Egyptians and the Kushites with the former being the perpetrators of the statues’ destruction and the latter the creators of the caches. Psamtek II’s invasion of Nubia is related on four stelae, two from Shellal, one from Karnak and one from Tanis (Cairo, Egyptian Museum JE 67095), and referred to in graffiti carved by Carian, Greek and Phoenician mercenaries, for example as found on the colossal statues of Ramesses II at Abu Simbel (PM 19722, 37 [135]; with discussion: Eide et al. 1994, 283–8 and Gozzoli 2017, 48–50, 107–16). The campaign ostensibly occurs during the reign of Aspelta, the latest king included in the statue groups. Briefly, one of the stelae at Shellal, dated to year three (593 BC) of Psamtek II’s reign, mentions that they reached Pnubs (Dokki Gel-Kerma) and that this was the location of an Egyptian victory. This is also related in the fragmentary stela from Karnak. Only the lunette of the other Shellal stela remains. Psamtek II’s Tanis stela (Cairo, Egyptian Museum JE 67095) also dates to year three, but the text is broken and its reconstruction not certain. It mentions that the Egyptian army fights in the land of Shas (šs), an unknown location that has, for example, been variously identified with Sai, downstream of the Third Cataract, or Sanam, which lies downstream of the Fourth Cataract (Vercoutter 1958, 147; Eide et al. 1994, 284–5). The army advances reaching Tergeb (trgb) and the residence of the kwr (qore), the Kushite ruler. Tergeb is situated near tꜢ dhnt. As Kushite rulers could have a number of palaces, reaching such a residence is not of great assistance in charting an army’s route. Tergeb has recently been identified as Tergedum as listed by Pliny and identified with modern Tergis/Soniyat in the southern Dongola Reach (Żurawski 2003, 25–6, 526) downstream of Gebel Barkal/Napata. TꜢ dhnt has been postulated as being Gebel Barkal (Kendall 1991, 308).

15

Primum visum ergo primum.

A Carian graffito records that the army reached just past Kerkis ‘as far as the river allowed’ (Eide et al. 1994, 288). While ‘as far as the river allowed’ seems to allude to a cataract or location where river navigation was difficult, the location of Kerkis itself is uncertain and has been hypothesised as being in the Second Cataract or alternatively near the Fourth Cataract, possibly at Korkos (Eide et al. 1994, 288–9; Gozzoli 2017, 60). The first statue cache discovered was at Gebel Barkal. It was viewed as corroborating evidence for Psamtek II’s inscriptions concerning his invasion of Nubia, an idea then furthered by the cache discovery at Pnubs, Kerma-Dokki Gel, a site mentioned in Psamtek II’s texts. If Psamtek II’s army had sacked the Kushite capital and reached the Great Temple of Amun, then it seems probable that this event would have been included, and perhaps indeed it was on the Tanis stela, but there is no extant evidence directly mentioning such an event. The palace of Aspelta at Gebel Barkal was destroyed by fire and many of the main temples there also display evidence of fire damage. The readiest explanation is that these structures were deliberately torched by a marauding Egyptian army. Regrettably, C14 dates performed on material from Aspelta’s palace were inconclusive (Kendall and Wolf 2007, 86–7). To fit the Dangeil statues into the established scheme of the Psamtek II hypothesis put forth to explain the earlier statue discoveries, increasingly elaborate (but not impossible) explanations are required to accommodate this new data, and as such, may be misleading interpretations of the later discovery. Dangeil is located far upstream from Gebel Barkal and Kerma-Dokki Gel. It is situated beyond the Hagr el-Merwa, the Egyptian frontier of the New Kingdom, and is south of the Fifth Cataract. Had the cache at Dangeil been the first to be discovered, it seems unlikely that a marauding Egyptian army would have been among the initial causal explanations put forward.15 It is also unknown as to whether there is yet another statue group to be discovered upstream of Dangeil. Another suggestion to explain the Kushite caches has been internal political pressures such as dynastic strife or disagreement. The damnationes memoriae of the names of Aspelta, the latest ruler present in the caches, on his election stela, an inscription that focused on his legitimacy as ruler, as well as on his banishment

TAHARQO AND HIS DESCENDANTS: A STATUE CACHE UPSTREAM OF THE FIFTH NILE CATARACT

stela (Eide et al. 1994, 232–58; Gozzoli 2017, 52–5) certainly suggest an official condemnation of him and, by extension, perhaps of his family. In the former, the names of Aspelta and those of his female relatives were erased. ‘Whoever attacked the Election Stela of Aspelta fully understood the importance of the female sisters within Kushite kingship’ (Gozzoli 2017, 53–4), thus suggesting an internal rather than external source for this event. In the banishment stela, dated to the second year of his reign, Aspelta expels a group of priests from the Amun temple in the Kushite capital Napata and executes them, ostensibly for interfering with the oracle of Amun. This animosity between Aspelta and the priesthood of Amun could very well explain the damnationes on the stelae; however, it should be noted that apart from incidental damage, his name and titulary are preserved on the majority of the cache statues, unlike the damnationes memoriae perpetrated on Dynasty 25 monuments in Egypt (Gozzoli 2017, 61–6). Dangeil’s Aspelta proves the exception as the lower left leg and back pillar appear to have been deliberately smashed. However, as the Dangeil statues were found scattered in disturbed late contexts no timeline for this event can be established. Yet another explanation put forward for consideration is temple refurbishment. This explanation is not reliant upon the statue groups being cached at the same time, but does not exclude it. The Kerma-Dokki Gel cache also contained the head of a Dynasty 18 pharaoh and part of a seated figure dating to the New Kingdom. Similarly, the Gebel Barkal caches also contained fragments of statuary of earlier date (Bonnet and Valbelle 2005, 188–9). The Dangeil statues were associated with fragments of late Kushite statues; however, again owing to the disturbed nature of the finds it is not possible to determine exactly when they became associated together. As far as it can be determined, very little royal statuary is created following the reign of Aspelta, so the early Kushite statues, particularly those of the Taharqo family group, would be the foremost ones present. These statues may have remained in the temples for several generations following Aspelta. The Amun cult ritual seems to change its focus over time, moving from an emphasis on the divine statue of the ruler,

16

The authors are grateful to C. Rilly for his comments regarding this graffito. See further Anderson and Mohammed Ahmed 2009, 82, pl. 7.

243

Plate 16: Detail of the Meroitic graffito on the left thigh of Dangeil’s Taharqo statue. © Berber-Abidiya Archaeological Project.

which may have served as intercessor for petitioners, to those of gods such as Arensnuphis, which may have then fulfilled this self-same role as discussed above. The cursive Meroitic graffito incised on the left thigh of Dangeil’s Taharqo statue raises some interesting questions (Pls 5 and 16).16 The earliest cursive Meroitic inscriptions known thus far are two graffiti from Kerma-Dokki Gel that date to the 2nd century BC (Rilly 2003, 41–55; Leclant et al. 2000, 1377–8). This might suggest that the statue was not put in a cache until at least this date. All three sites, Dangeil, KermaDokki Gel and Gebel Barkal, undergo extensive construction, remodelling and renovation during the late Kushite (Meroitic) period. At Kerma-Dokki Gel ‘the Classic Meroitic period sees an almost complete reconstruction of the religious and palatial complex’ (Bonnet and Valbelle 2004, 111). At Gebel Barkal, though much of the later Kushite structures remain to be excavated, Natakamani and Amanitore (1st century AD) enlarge and restore the sanctuary of the Amun temple (B500) and build a large palace. They carry out extensive building activities throughout their kingdom during their reign. At Dangeil, they also build or renovate

244

J.R. ANDERSON, S. MOHAMMED AHMED, M. SULIMAN BASHIR AND R. KHIDIR ELRASHEED

the Amun temple, as discussed above, in which the numerous early Kushite royal statue fragments were found. It is possible that the statues were cached at this time as they had outlived their earlier role and had no place in the later temples.17 Further, what this might suggest is that at sites where significant early Kushite temples remained in use and particularly where these temples were extensively remodelled or renewed during the late Kushite period, especially in Upper Nubia, there is a strong possibility of an associated statue cache. Acknowledgements The Berber-Abidiya Archaeological Project was codirected from 2000 to 2014 by Julie Anderson (British Museum, UK) and Salaheldin Mohammed Ahmed (National Corporation for Antiquities and Museums, Sudan) and from 2015 to the present by Julie Anderson (British Museum, UK), Rihab Khidir elRasheed (National Corporation for Antiquities and Museums, Sudan) and Mahmoud Suliman Bashir (National Corporation for Antiquities and Museums, Sudan). The project is grateful for the assistance and support from the following organisations and individuals: Archeology4All, Italy; British Museum, UK; British School in Rome, Italy; Dangeil Village, Sudan; El Salha Project (Central Sudan), Italy; Institute for Bioarchaeology, UK; National Corporation for Antiquities and Museums, Sudan; Nubian Archaeological Development Organization (Qatar-Sudan); Qatar-Sudan Archaeological Project (Qatar-Sudan); Royal Ontario Museum Petrographic Laboratory, Canada; Section Française de la Direction des Antiquités du Soudan (SFDAS-Khartoum); The Sorbonne, France; University of Dongola, Sudan; University of Southampton, UK; Acropole Hotel, Sudan; Dr D. Bird; Mr A. Giambrone; and all of the project’s team members. Bibliography Ali Mohamed, A. and Anderson, J. 2012. Highlights from the Sudan National Museum. Khartoum. [reprinted 2015, 2017].

17

The authors are grateful to Pawel Wolf for interesting discussions concerning this particular possibility.

Anderson, J., D’Andrea, A. C., Logan, A. and Mohammed Ahmed, S. 2007. Bread moulds from the Amun temple at Dangeil – An addendum. Sudan & Nubia 11, 89–93. Anderson, J., Khidir elRasheed, R. and Suliman Bashir, M. 2017. QSAP Dangeil 2016: ‘Aspelta, beloved of Re-Harakhty’ and tombs in the temple. Sudan & Nubia 21, 159–68. Anderson, J. and Mohammed Ahmed, S. 2006. Bread moulds and ‘throne halls’: Recent discoveries in the Amun temple precinct at Dangeil. Sudan & Nubia 10, 95–101. ———. 2006–7. The ‘throne room’ and dais in the Amun temple at Dangeil, Nile State, Sudan. Cahier de recherches de l’Institut de papyrologie et d’égyptologie de Lille 26, 29–39. ———. 2009. What are these doing here above the Fifth Cataract?!!: Napatan royal statues at Dangeil. Sudan & Nubia 13, 78–86. ———. 2010a. Dangeil: La découverte d’un nouveau temple d’Amon. Méroé, un empire sur le Nil aux confins de multiples cultures. Les Dossiers d’archéologie, Hors-série 18. Paris, 50–5. ———. 2010b. Une nouvelle cachette de temple: Napatéens et Méroïtes sur la 5e cataracte du Nil. In G. Andreu-Lanoë, M. Baud and A. Sackho-Autissier (eds), Méroé, un empire sur le Nil. Paris, 231–2. ———. 2010c. Bread, the staff of life. Recent discoveries at Dangeil, Sudan. In Between the cataracts 2. Proceedings of the 11th International Conference for Nubian Studies. Warsaw University 27 August–2 September, 2006. Polish Archaeology in the Mediterranean Supplement Series 2.2/1. Warsaw, 55–9. ———. 2013. Dangeil 2012: Sacred ram – Avatar of the god Amun. Sudan & Nubia 17, 70–7. ———. 2014. Early Kushite royal statues at Dangeil, Sudan. In J. R. Anderson and D. A. Welsby (eds), The fourth cataract and beyond. Proceedings of the 12th International Conference for Nubian Studies. 1–6 August 2010, London. British Museum Publications on Egypt and Sudan 1. Leuven, 613–20.

TAHARQO AND HIS DESCENDANTS: A STATUE CACHE UPSTREAM OF THE FIFTH NILE CATARACT

Anderson, J., Mohammed Ahmed, S. and Sweek, T. 2012. Meroitic building techniques: A few observations from Dangeil. Sudan & Nubia 16, 72–9. Bonnet, C. 2011. Les destructions perpétrées durant la campagne de Psammétique II en Nubie et les dépôts consécutifs. In D. Valbelle and J.-M. Yoyotte (eds), Statues égyptiennes et kouchites démembrées et reconstituées. Paris, 21–32. Bonnet, C., Honegger, M. and Valbelle, D. 2003. Kerma Soudan. 2001–2002, 2002–2003. Genava 51, 257–300. Bonnet, C. and Valbelle, D. 2004. Kerma, Dokki Gel. In D. A. Welsby and J. R. Anderson (eds), Sudan: Ancient treasures. London, 109–12. ———. 2005. The Nubian pharaohs: Black kings on the Nile. Cairo; New York. Dunham, D. 1957. Royal tombs at Meroë and Barkal. Royal cemeteries of Kush. Vol. IV. Boston. ———. 1963. The west and south cemeteries at Meroë. Royal cemeteries of Kush. Vol. V. Boston. ———. 1970. The Barkal temples. Boston. Edwards, D. 2004. The Nubian past. An archaeology of the Sudan. London. Eide, T., Hägg, T., Holton Pierce, R. and Török, L. 1994. Fontes historiae Nubiorum. Textual sources for the history of the middle Nile region between the eighth century BC and the sixth century AD. Vol. I. From the eighth to the mid-fifth century BC. Bergen. ———. 1996. Fontes historiae Nubiorum. Textual sources for the history of the middle Nile region between the eighth century BC and the sixth century AD. Vol. II. From the mid-fifth to the first century BC. Bergen. Gozzoli, R. 2017. Psammetichus II. Reign, documents and officials. London. Grzymski, K. 2004. Inscribed jamb fragment. In D. A. Welsby and J. R. Anderson (eds), Sudan: Ancient treasures. London, 168. Kendall, T. 1991. The Napatan palace at Gebel Barkal. A first look at B 1200. In W. V. Davies (ed.), Egypt and Africa. London, 302–13. ———. 2004. Statuette of the god Amun with ram head. In D. A. Welsby and J. R. Anderson (eds), Sudan: Ancient treasures. London, 147. Kendall, T., and Wolf, P. 2007. Excavations in the palace of Aspelta at Jebel Barkal, March 2007. Sudan & Nubia 11, 82–8.

245

Leclant, J., Heyler, A., Berger-el-Naggar, C., Carrier, C. and Rilly, C. 2000. Répertoire d’épigraphie Méroïtique. Corpus des inscriptions publiées I–III. Paris. Morkot, R. 2000. The black pharaohs. Egypt’s Nubian rulers. London. Omar el-Sadig, S. 2002. Some fragments of a statue of King Aspelta at Umm Dom (Khartoum province). Archéologie du Nil Moyen 9, 89–93. Pope, J. 2014. The double kingdom under Taharqo. Studies in the history of Kush and Egypt, c. 690– 664 BC. Leiden. PM 1951 = Porter, B., and Moss, R. 1951. Topographical bibliography of ancient Egyptian hieroglyphic texts, reliefs, and paintings. Vol. VII. Nubia, the deserts, and outside Egypt. Oxford. PM 19722 = Porter, B., and Moss, R. 19722. Topographical bibliography of ancient Egyptian hieroglyphic texts, reliefs, and paintings. Vol. II. Theban temples. 2nd ed. Oxford. Priese, K-H. 1974. Die Statue des napatanischen Königs Aramatelqo (Amtelqa), Berlin, Ägyptisches Museum Inv.-Nr. 2249. In Festschrift Ägyptisches Museum Berlin. Mitteilungen aus der Ägyptischen Sammlung, Vol. 8. Berlin, 211–32. Reisner, G. 1917. The Barkal temples in 1916. Journal of Egyptian Archaeology 4, 213–27. ———. 1931. Inscribed monuments from Gebel Barkal. Zeitschrift für Ägyptische Sprache und Altertumskunde 66, 76–100. Rilly, C. 2003. Les graffiti archaïques de Doukki Gel et l’apparition de l’écriture méroïtique. Meroitic Newsletter 30, 41–55. Shinnie, P. and Bradley, R. 1980. The capital of Kush 1. Meroitica 4. Berlin. Török, L. 1997a. Meroe city. An ancient African capital. John Garstang’s excavations in the Sudan. Vol. I. London. ———. 1997b. The kingdom of Kush. Handbook of the Napatan-Meroitic civilization. Leiden; New York; Cologne. ———. 2002a. The image of the ordered world in ancient Nubian art. The construction of the Kushite mind 800 BC–300 AD. Probleme der Ägyptologie 18. Leiden; Boston; Köln. ———. 2002b. Space, temple, and society. On the built worldview of the Twenty-Fifth Dynasty in Nubia. In I. Caneva and A. Roccati (eds), Acta Nubica. Rome, 231–8.

246

J.R. ANDERSON, S. MOHAMMED AHMED, M. SULIMAN BASHIR AND R. KHIDIR ELRASHEED

———. 2004. Sacred landscape, historical identity and memory: Aspects of Napatan and Meroitic urban architecture. In T. Kendall (ed.), Nubian Studies 1998. Boston, 157–67. ———. 2009. Between two worlds. The frontier region between ancient Nubia and Egypt 3700 BC–AD 500. Probleme der Ägyptologie 29. Leiden; Boston. ———. 2011. Hellenizing art in ancient Nubia 300 BC–AD 250 and its Egyptian models: A study in ‘acculturation’. Leiden. Valbelle, D. 2011. Les statues égyptiennes découvertes à Kerma et Doukki Gel. In D. Valbelle and J.-M.

Yoyotte (eds), Statues égyptiennes et kouchites démembrées et reconstituées. Paris, 13–20. ———. 2012. Les stèles de l’an 3 d’Aspelta. Cairo. Vercoutter, J. 1958. Excavations at Sai, 1955–7. Kush 6, 144–69. ———. 1961. Le sphinx d’Aspelta de Defeia (Khartoum Museum no. 11777). In Mélanges Mariette. Bibliothèque d’étude 32. Cairo, 97–104. Żurawski, B. 2003. Nubia II. Southern Dongola Reach survey 1. Survey and excavations between Old Dongola and Ez-Zuma. Warsaw.

V AFTERLIVES: REUSE AND DESTRUCTION

RITUAL METAL STATUARY IN ANCIENT EGYPT: ‘A LONG LIFE AND A GREAT AND GOOD OLD AGE’ Deborah SCHORSCH

Abstract Little archaeological evidence of metal recycling in ancient Egypt survives, and virtually none suggesting that temple ritual statuary was dispersed through official channels to be melted down and reused for ritual or profane purposes. Instead, there is ample evidence indicating that these cupreous and precious-metal figures, varying widely in size and ranging from basic to lavish in manufacture and embellishment, were retained in the temples over long periods, were occasionally altered to suit shifting political–religious ideologies and were respectfully buried within temple precincts upon retirement. Along with archaeological context, physical indications of alterations and reuse are extremely important for recognising extended lives and evolving functions, and thereby contribute to our understanding of ritual practices and the procurement and dispersal of sacred images. Technical investigations using typical museum laboratory techniques including visual examination, radiography and compositional analyses allow us to clarify function and characterise transformation. Examples discussed here include: Kushite royal figures that were defaced and reused by their Saite successors; statues of deities transformed through alteration of their regalia or later embellishment with precious materials; a small representation of Osiris slated for repair after it was damaged beyond recognition; and statues of royal and presumably high status non-royal personages that were altered for reasons that for now remain obscure. *

*

*

Examples of Egyptian stone sculpture usurped and reused during antiquity are plentiful; in royal contexts,

1

The term cupreous is used here to describe all copper-based formulations of unknown composition. In ancient Egypt, these would be unalloyed coppers, arsenical coppers, bronzes or leaded bronzes. The possibility of precocious technological discoveries and developments not immediately adopted, or only on a local scale, certainly exists, but to the best of our knowledge, brass was not used to cast statuary in ancient Egypt. Egyptian

such actions generally reflect changes in conjoined political and religious inclinations. A relief at Deir elBahri representing Hatshepsut before Amun is one potent example (Roth 2005, fig. 96): Hatshepsut’s name was replaced with that of her stepson Thutmose III towards the end of his reign, perhaps to assure a smooth succession for his own son; one century later during the Amarna period, the image of Amun was defaced, only to be recut later, sometime after Akhenaten’s death and the restoration of the Theban gods. Similarly, the alteration of insignia on Kushite royal ‘bronzes’ has been well documented (Hill 2004, 66–8).1 It is, however, specifically with regard to metal that another mechanism of reuse requires our consideration: a significant physical property of metals, not shared by most materials used by ancient artisans, and one that has surely been exploited since man took up pyrotechnical activities, is their ability to be melted and remelted and, therefore, cast and recast. Even hammered sheet can be cut down and with annealing, refashioned. This potential for reuse is just one reason why metals, even after their transformation into objects of beauty, cult or utility, generally retain an inherent worth; metals have always served as an enduring medium of exchange. The decision to retire a metal object may be motivated by several factors. Most simply, the owner wishes to exchange the metal itself for another commodity. Objects become damaged or old-fashioned, and they may represent the gains of looting, a fact that often must be obscured. Deposits of metalwork intended to be temporary, but never recovered, may reflect personal, institutional or illicit wealth lost or buried in haste. If they contain damaged and/or heterogeneous objects – often with raw materials, tools or further indications of a metallurgical industry – they might

copper-alloy statues said to contain zinc in greater than trace amounts need to be scrutinised carefully. In addition to errors or issues introduced by the analytical process, improper dating of statues in question, or failure to recognise that they are modern forgeries, as well as historic restoration processes that involve the use of zinc powder or foil, may be responsible in such cases.

250

D. SCHORSCH

represent the property of a craftsman or workshop engaged in remelting and repurposing metal for redistribution, which apparently was undertaken even in copper-rich Cyprus (Knapp 1988). Ironically, a wig fragment from a New Kingdom figure attests to the practice of metal recycling there (Schorsch 2007). In Egypt, deposits that may connote recycling – for the purposes of this article a term reserved strictly for processes that entail remelting and the complete destruction of the original work – have seldom been identified, at least very few dating prior to the Ptolemaic period (e.g. Edgar 1906; Wainwright 1925). Potential exceptions are the hoard of precious-metal objects, ingots and vessel fragments that was found inside a pottery vessel in a domestic context at Amarna (Pendlebury 1931, 236); gold jewellery, scrap, amulets and small precious-metal statuary deposited during the Third Intermediate Period discovered in the Bastet temple in Bubastis (Bakr and Brandl 2010, 45–7; Brandl and Wenzel 2010); and a deposit of foreign ingots and coins found in a Dynasty 27 context at Karnak (Masson 2016, 32–7). Defying clear interpretation are several large deposits of metal with royal associations found at Tôd (Stefanova 2015), Lisht (Arnold 1988, 99–101) and Tell Basta (Lilyquist 2012), all of which also have elements – e.g. intentionally damaged objects and raw materials – that might reflect recycling. Certainly in the case of the Lisht metal, as Arnold notes, the circumstances that placed a scrap-metal collection secured with a New Kingdom royal seal below the rubble of a Middle Kingdom temple remain a puzzle. There seem to be no Egyptian textual sources that mention metal recycling, although a few examples from neighbouring cultures can be cited (Philip 1988, 190; Ornan 2012, 448). Recycling seems an inevitable consequence of grave robbing, which is attested to in many texts and well supported by archaeological evidence. Thefts from the temples and their desecration have also been documented (Goelet 1996). With regard to ancient metallurgy as a whole, occasional analytical studies can be cited in which the use of recycled metal in a limited context has been

2

For a somewhat later find of a ‘bronze hoard’ at Athribis, see Kamel (1968). More than one hundred statues were found in a fill layer above tombs tentatively dated to the Saite period, near the Userkaf pyramid at Saqqara by el-Khouly (1978). Because of the funerary nature of the site, this group may

confirmed or convincingly proposed on the basis of bulk composition, trace elements or lead-isotope analysis (e.g. Bayley and Butcher 1995; Ponting 2002; Armbruster and Eilbracht 2006). In other contexts, researchers have explicitly raised the possibility and concluded that recycling was not practised (Ashkenazi, Iddan and Tal 2012; Gener et al. 2014). On occasion, recycling has been cited to explain analytical results that could otherwise not be satisfactorily interpreted (e.g. Attanasio, Bultrini and Ingo 2001). In fact, there is no analytical evidence demonstrating that ancient Egyptian ritual statuary was recycled, and the evidence of various temple statuary caches, albeit for the most part poorly excavated and documented, intimates the contrary, as do some more recent finds. Of the various terms – hoard, cache, votive, dedication and deposit – used to describe ancient assemblages of valuable artefacts and raw materials placed temporarily or permanently ‘out of sight’ in non-funerary contexts, the most neutral – deposit – proves most helpful for describing large groups of metal statues, primarily representing deities, found during early excavations in 1st millennium BC and early 1st millennium AD temple precincts at sites such as Sais (Daressy 1901, 130; Daressy 1917, 247−8), Saqqara (Lauer and Picard 1955; Ziegler 1981; Gombert-Meurice in this volume), Dendera (Abdalla 1995), Medinet Habu (Hölscher 1954, 30−3), Karnak (Jambon 2009), Hibis (Winlock 1941, 42−3, pl. XXVII), Koptos (Petrie 1896, 24, pl. XXI, 12–8), Naukratis (Petrie 1886, 41–2; Masson 2015) and elsewhere.2 Many thousands of unprovenienced metal statues of gods and goddesses found in collections around the world presumably originate from early excavations at these sites. In addition, most metal ritual implements lack find-spots, but some have been found within the temples, attesting to their long employment and peaceful retirements (Daressy 1902; Green 1987, 1–4). These depositions signal intentional acts and the ritual figures and implements were meant to remain in place for all time. Metal statuary and metal temple furnishings were not unique in this regard: ivory and faience animal figures dated to the Early Dynastic period found at Hierakonpolis, Abydos,

represent a secondary deposition previously buried in a temple context. Two ‘Osiris’ deposits found by Bisson de la Roque (1927, 102−3, 119) at Medamud might signify similar circumstances. For a nuanced discussion of the different types of sacred caches in Egypt, see Charloux and Mahmoud in this volume.

RITUAL METAL STATUARY IN ANCIENT EGYPT: ‘A LONG LIFE AND A GREAT AND GOOD OLD AGE’

Elephantine and Koptos constitute an early precedent for the temple deposits of metal statuary described here. Along with the physical evidence of extended service in the temples, new perspectives offered by more recent finds of 1st millennium BC metal statuary in the Sacred Animal Necropolis at Saqqara (Davies 2007) and at ‘Ayn Manawîr in the Kharga oasis (Wuttmann, Coulon and Gombert 2007) will be presented below. The ritual function of metal statuary in ancient Egypt The meaning of ritual and the manifestation of ritual practices are widely discussed in anthropological, sociological and theological literature. For our present purposes, the briefest of definitions proposed by Ian Hodder (1982, 159), ‘ritual refers to performance and the associated rules’, seems appropriate.3 In ancient Egypt, certain types of cupreous, silver and gold statuary are recognisable as instruments of temple ritual. There is no indication that these images were discarded, or that the metal itself was recycled through official temple channels; on the contrary, presented here are examples of statuary representing deities and elite personages that attest to the figures’ extended lives and sheltered afterlives. Although earlier examples exist, most surviving metal statuary associated with temple ritual dates to the 1st millennium BC. Ancient Egyptian deity figures in metal exist in great variety; a myriad national, local and foreign gods and goddesses survive, and representations of the same deity are often outfitted with different attributes or regalia. The figures vary in size and in the quality and sumptuousness of their execution. Revered figures of precious materials are known from texts: ‘electrum, lapis-lazuli, turquoise and every august and precious stone’ adorned an image associated with Tutankhamun’s refurbishing of the Ptah, South-of-his-Wall temple at Memphis (Davies 1995, 32). Such statues have often been identified as cult images, presuming that

3

4

The entire sentence reads: ‘Unlike both religion and magic, ritual refers to performance and the associated rules rather than to abstract concepts and beliefs’. Precious-metal representations of deities called cult statues include the gold figure of Amun in the Metropolitan Museum (26.7.1412) and the silver falcon-headed god in the Miho Museum (Roehrig 1996). Lorton (1999, 128, figs 2, 3 and 5) identifies as cult figures a late Predynastic or Dynasty 1 granite falcon, a Middle Kingdom black-bronze crocodile said to be

251

each inhabited a central shrine in the inner sanctum of its temple home, where it served as a corporeal manifestation of the sacred and was accessible only to the kings and priests who ministered to its needs through ritual acts and utterances (Lorton 1999; Robins 2005). Only very few figures have been found in archaeological contexts that allow unequivocal identification of this role, so such designations are usually based on the size of the statue in question, the costliness of its materials and the complexity of its decoration,4 although in smaller or provincial temples, many sacred images may well have been made of substantially less precious materials, such as wood that was gilded (Hill 2007d). Furthermore, it has been demonstrated that individual temples had not a single cult figure, but many, and different ritual functions and postures surely were taken up by different types (Eaton 2007). Most surviving metal ritual statuary is cupreous and in absolute terms less costly than the gold, silver and electrum images known primarily from texts. The more modest figures are frequently described as votive, said to reflect private acts of piety, but in the context of ancient Egypt, this model is problematic (Hill in this volume). Moreover, the copper and tin used to cast bronze were also valuable commodities and further richness resided in their refined workmanship and precious-metal inlays and cladding, as well as inlays of other materials such as semi-precious stones, glass, faience and ivory. Metal representations of kings in postures assumed during ritual intercourse with deity figures survive in small but significant numbers. In the corpus of Egyptian cupreous statuary, it is these royal figures that have been studied most systematically (Hill 2004). With arms and hands posed in gestures of offering or protection, kneeling and occasionally striding, the royal images were placed, for example, on wooden barques processing from the temples, where they interacted with deities seated in closed kiosks to protect them from the eyes of the populace, or in cult groupings

from Hawara, and a Late Period silver falcon. Based on dimensions of surviving wooden inner shrines, in which these figures are thought to have been housed, Lorton estimates that figures would necessarily have been small and that their crowns were removed before they were recessed for the night; according to his calculations, the extant shrines could not have accommodated the Miho Museum figure, which measures nearly 42cm without its crown.

252

D. SCHORSCH

within the temple. Such processions are well documented on temple walls starting in the New Kingdom, and on standards bearing processional barques (e.g. Leiden, Rijksmuseum van Oudheden AED 87) that replicate the deities’ royal and sacred entourages. Cult groupings of royal statues engaged in ritual activities were represented on temple walls and on naoi as well.5 The repertoire of metal implements connected to these rites includes protomes of gods in human or zoomorphic form and standards that served as fittings on sacred barques, censers and temple furnishings, some of which have been recovered in temple deposits (e.g. Green 1987). It is, therefore, telling that the earliest king represented in metal holding a ritual posture, Senwosret I (r. c. 1961–1917 BC), is prostrate on the lid of a censer (Hill 2007a, 9, figs 4–5; 201, cat. no. 1). Next in line chronologically, in a more conventional kneeling pose associated with temple ritual, is a late Middle Kingdom figure said to be from a group of statues found at Hawara (Ortiz 1996, cat. no. 37; Hill 2004, 12−6, 196, cat. no. 128), including the crocodile identified as a cult figure by Lorton. The find may reflect the simultaneous deposition of works produced over a period that extends over Dynasties 12 and 13 (Hill 2004, 12−6), which is particularly interesting in the context of the current study of extended service in the temples. In contrast to the more restrained poses held by the kneeling kings, some royal statues used in cult performances hold very active poses, for example a large striding figure of the Dynasty 23 king Pedubast I known from a torso fragment in Lisbon (Hill and Schorsch 2005, 181–4). Starting in the New Kingdom perhaps, statues of priests charged to carry out the functions in the name of the kings can be identified from their clothing and inscriptions (Hill 2007b, 29). As a rule, it is difficult to identify the role or multiple roles performed by a specific player. Physical evidence of extended life Interpreting physical evidence related to the extended life of statues that resided in the temples and their evolving functions contributes to our understanding of

5

6

For example, a group of statues (each shown on its own separate base) consisting of a kneeling king offering to two unknown gods is represented on the naos found by Naville (1887, pl. 5, second lowest register, left) at Saft el-Hina. The regnal dates of Shabaqo are currently under discussion, as well as his position respective to Shabitqo. If Shabitqo were to

temple ritual and the procurement and dispersal of sacred images. That royal figures remained active in ritual over long periods is demonstrated by the altered regalia often observed on statues of Kushite rulers (c. 750–656 BC) (Fig. 1). Although their distinctive double uraei, ram’s head pendants and headband with streamers were deemed unacceptable at some time during the reign of their Saite successors (664–525 BC), these royal statues remained active receptacles of sacral energy. Of the twenty figures of kneeling kings of Egyptian (as opposed to Nubian) manufacture and context attributed to Dynasty 25 by Hill, eleven show evidence of some form of alteration (Fig. 2a, b) (Hill 2004, 66−8, cat. nos 19, 36, 41, 62, 72, 90, 104, 223, 224, 243 and 255), and two require closer examination to clarify suggestive indications (cat. nos 37 and 294). Although well-attested for monuments and temple equipment, there is only one known metal representation of a Kushite ruler (Athens, National Archaeological Museum 6241) bearing the name of a later king (Hill 2004, 27, n. 27). Originally uninscribed, this ruler’s Kushite regalia were intentionally defaced and the figure was inscribed to Psamtek (Tourna 2007), a name used by the first, third and sixth rulers of the subsequent dynasty. If this figure does represent King Shabaqo, as Hill (2004, 60−2) posits on the basis of style and physiognomy, it must have remained in active service through the reigns of at least four and most likely six subsequent rulers;6 the military campaigns against the Kushites in Nubia under Psamtek II (reigned 595–589 BC), which led to widespread erasure of earlier monuments, may have provided a political context for inscribing the Athens figure with the name of a later king (Sauneron and Yoyotte 1952; Tourna 2007).7 Presumably, non-Kushite royal figures also continued to reside in the temples and retain a sacral character beyond the reign of the king they originally served, without the necessity of altering their regalia. It is worthwhile to consider briefly the modification of a kneeling bronze statue of the Saite ruler Amasis (Hill and Schorsch 2007, 210, cat. no. 47) (Fig. 3) that was not usurped; rather the alteration may reflect a change in the figure’s ritual role over the course of

7

be moved from the third to second position in Dynasty 25, the number of intervening reigns would be reduced by one. Like the royal inscription added in Saite times, the removal of both uraei, rather than just one, is atypical (Hill 2004, 60); the significance of ancient, intentional damages to the face, torso, right arm, and kilt also require consideration.

RITUAL METAL STATUARY IN ANCIENT EGYPT: ‘A LONG LIFE AND A GREAT AND GOOD OLD AGE’

extended use. Spelled out in gold inlays (described as precious metal in Hill and Schorsch 2007, 210, cat. no. 47) on the back of the king’s belt and on the front of his kilt are the two different cartouche names Amasis used throughout his more than forty-year reign (570–526 BC); until the more indistinct name on the belt was deciphered by James Allen (Hill 2004, 84, 166, cat. no. 31), the figure had been thought to represent an earlier Saite king usurped by Amasis (Bothmer 1960, 51). Although it cannot be concluded from the wording of the royal names that they were written at different times, the technical evidence convincingly suggests this. The channels containing the inlays on

Fig. 1: Kushite king. Egypt, Third Intermediate Period, Dynasty 25, c. 713–664 BC. Solid cast bronze, traces of gold leaf; H. 7.5cm. New York, The Metropolitan Museum of Art, Lila Acheson Wallace Gift, and Anne and John V. Hansen Egyptian Purchase Fund, 2002 (2002.8). © The Metropolitan Museum of Art.

253

Amasis’ belt were carved directly into the wax model used to cast the figure, which is the preferred method for introducing linear details to cast cupreous surfaces (Fig. 4a). The hieroglyphs on the kilt were inlaid into channels cut into the metal surface at some time after

Fig. 2a, b: Details showing sites of erased insignia on the chest and head of Kushite king (Fig. 1). © The Metropolitan Museum of Art.

254

D. SCHORSCH

the figure was cast, and although the workmanship is satisfactory, the inscription is awkwardly placed, spilling without regard across a seam and onto the central tab of Amasis’ kilt (Fig. 4b). Reassignment to a new ritual role may have necessitated a new form of the royal name, just as physically repositioning the figure respective to the cult’s sacred recipient or the human celebrant might have placed the original name outside of their line of sight. A motivating force behind the alteration of another bronze royal figure (Sotheby’s 2001, 26–7, cat. no. 28) is also difficult to explain. This kneeling king appears at first glance to be a conventional member of the large corpus of royal metal statuary; at the time of its appearance on the New York art market it was attributed to Dynasty 21/22 (c. 1075–712 BC). The facial features and body proportions do support a Third Intermediate Period date (Fig. 5), as does the double lozenge belt (Hill 2004, 46−7, 125−6). Although consistent in form with this date, the undersized nemes and its placement are jarring, and it initially raised suspicions about the figure’s authenticity. Still, the archaeological corrosion on the statue’s surface, when observed under magnification, spoke emphatically for its antiquity.8 An explanation for the disparity between the figure and its headdress, if not a clarifying narrative or underlying motive(s), is seen in radiographs (Fig. 6). The statue is hollow cast with a cavity that extends through its torso, head and legs. The solid arms were cast separately and attached mechanically in a conventional manner using tenons extending from the body, one of which was repaired in antiquity. Most significantly, the nemes was cast separately and placed over a rounded skull. The nemes is tight fitting, and a radiotransparent line that tracks the edge of the headdress implies that some cutting back at the temples was necessary to accommodate it. There is a single hole in the nemes extending into the forehead for attaching a separately cast uraeus. The back of the skull is unusually flat, but this probably represents its original shape rather than an alteration: as seen in the profile radiograph, the generous cavity between the back of the skull and the inner wall of the nemes indicates that such a modification would not have been necessary. There are a few examples of separately cast royal headdresses that were presumably intended for statu-

8

The figure was examined in the Department of Objects Conservation at The Metropolitan Museum of Art in 2001.

Fig. 3: King Amasis. Egypt, Late Period, Dynasty 26, reign of Amasis (570–526 BC). Solid cast bronze, gold leaf and inlay; H. 11.0cm. New York, The Metropolitan Museum of Art, Gift of Edward S. Harkness, 1935 (35.9.3). © The Metropolitan Museum of Art.

ary, as well as some puzzling small ones, and of course, separately made headdresses could have been of other materials. Nonetheless, extant figures of kings wearing separable headwear – such as a nemes or a red or white crown, which substantially cover the skull – are rare, and the known examples have incomplete heads.

RITUAL METAL STATUARY IN ANCIENT EGYPT: ‘A LONG LIFE AND A GREAT AND GOOD OLD AGE’

Fig. 4a, b: Details of royal names on the front and back of King Amasis (Fig. 3). © The Metropolitan Museum of Art.

A royal torso in the Ortiz Collection (Hill 2004, 15, 196–7, cat. no. 129), and a striding king in Munich (Schoske 1988, 207–10), both from the Middle Kingdom group found at Hawara, and a kneeling figure inscribed to Osorkon Meriamun (Hill 2004, 133, cat. no. 237, pl. 19), all employed a similar solution: the head was recessed to accommodate a headdress placed onto it, with a tang extending into the core cavity. The kneeling royal figure from the Hawara group (Hill 2004, 12−6, 196, cat. no. 128) is similar, except that the head is solid. Often during the Third Intermediate Period, provincial dynasts, such as Peftjaubast and Nesbanebdjedet, and priests, such as Menkheperra, assumed a degree of power not attainable in periods of greater political unity, and as such it is not easy to evaluate the station or status of the figure’s original owner or what type of headdress, if any at all, he might have worn. As represented by a statue in Rio de Janeiro, Menkheperra, a prominent early Third Intermediate Period high priest of Amun, wears the skullcap of his office but also a type of kilt reserved for royalty, and had his name written with royal titles in a cartouche (Kitchen 1990, 188–9). The Libyan dynast Nesbanebdjedet, in a small faience or glazed steatite statue in the Brooklyn Museum (37.344E) dating from c. 755–730 BC, had himself shown kneeling, holding nw pots, just as the phraseology of his inscription is that generally reserved for kings. On a bronze figure in the Museum of Fine Arts, Boston, Peftjaubast, who ruled at Herakleopolis

255

256

D. SCHORSCH

Fig. 5: Kneeling king. Egypt, Third Intermediate Period (c. 1070‒713 BC). Hollow cast bronze; H. 15.9cm. Private Collection (Sotheby’s 2001, cat. 28).

Fig. 6: Computed radiograph, profile view of kneeling king, detail with head (Fig. 5). © The Metropolitan Museum of Art.

as a contemporary of Piy (reigned c. 750–722 BC), wears a cap resembling the typically Kushite royal crown and modified versions of other Kushite regalia (Russmann 1981, 155; Hill 2004, 157−8, cat. no. 15). In theory, therefore, an individual of higher or lower status may have added at some point a royal headdress to his own or someone else’s figure. How the statue ended up with this nemes that is much too small, and if it was ‘upgraded’ for its owner or usurped by someone else, is not certain, but it is relevant to the current investigation because it raises the possibilities that some figures wore different headdresses at different times, that some headdresses were removed on certain occasions9 and that ritual statues and/or their attributes were retained and might be subject to subsequent (re) mixing and (mis)matching.

The dating of the figure remains problematic. The round skull visible in the radiographs is one feature sometimes seen on later Third Intermediate Period nonroyal statues representing Kushites or that echo Kushite visual traditions, e.g. the kneeling figure of Peftjaubast (Russmann 1981), but in spite of this, the facial features are not typically Kushite, and there is no evidence in radiographs of the ex-Sotheby’s king that the figure had worn a Kushite cap or its Egyptian derivative. Evidence of any other typically Kushite regalia is absent. Finally, the tabs in front of the king’s ears are angular and therefore typically Egyptian, and unlike the large, rounded tabs seen on Kushite rulers or contemporary dynasts such as Peftjaubast and Iuput II (Russmann 1979, 51). As noted earlier, the usurpation of royal figures by subsequent rulers occurred not infrequently in ancient Egypt, just as there are instances in which private individuals reused royal possessions. A striking bronze

9

Lortet hypothesised that crowns were removed from deity figures when they were placed in their shrines; see above, n. 4.

RITUAL METAL STATUARY IN ANCIENT EGYPT: ‘A LONG LIFE AND A GREAT AND GOOD OLD AGE’

figure in the Metropolitan Museum perhaps illustrates the reverse situation: the figure of a high-status individual ‘upgraded’ to represent a king (Hill 2012) (Fig. 7). The figure is solid cast, and having lost its separately made arms and legs, it now measures 21.0cm. Small fragments of gold track parallel lines articulated in corrosion running up both sides of the torso and continuing up the neck and head (Fig. 8a), marking the locations of narrow channels in the metal that were used to mechanically secure gold foil to the surface. The statue exhibits typical features of the Kushite period: a powerful body with slender waist, a rounded head with small ears, narrow eyes and a relaxed smile. With his shaved pate, this man is clearly a priest, and the long cords with tassels falling on each side of the apron belong to a priestly costume favoured in Kushite and early Saite times; a group of processing priests on the Saite period Oracle Papyrus in the Brooklyn Museum (47.218.3a–j) wear the same apparel. A panther head viewed from above, centred on the front of the figure just below his belt, alludes to the panther skin often worn by Egyptian clerics and kings; in fact, no figures of priests that sport a panther head without also having a pelt cloaking their shoulders and back can be cited. In this case, there is no physical evidence indicating that the front of the figure was cut back to accommodate the placement of the animal head. Furthermore, it also does not appear that the figure had a separately fashioned pelt previously or originally had a pelt delineated on its surface that has been obscured by subsequent abrasion or another mechanical or chemical process. The figure does not wear a royal kilt, but rather one that apparently was altered to suit royalty. The two streamers flanking each side of the apron were unquestionably executed directly on the statue (rather than in the wax model) in antiquity, presumably with the intention of altering its attributes and perhaps the identity of the man represented. Even with the unaided eye, one discerns in the rough surfaces below the streamers the original contours of the apron, and with magnification, the chisel marks present along the cut edges are visible. (Fig. 8b). The absence of lines separating the streamers from the apron and from each other contributes to the garment’s awkward appearance. Whereas the possibility exists that the absent limbs were original to the figure, there are reasons to suppose that they were ancient replacements for arms and legs that originally were integrally cast. For one, a solid cast

257

Fig. 7: Kushite priest. Egypt, Third Intermediate Period, Dynasty 25, c. 713–664 BC. Solid cast leaded bronze, traces of gold leaf; H. 21.0cm. New York, The Metropolitan Museum of Art, Purchase, Gift in Memory of Manuel Schnitzer and Lila Acheson Wallace Gift, 2010 (2010.259). © The Metropolitan Museum of Art.

258

D. SCHORSCH

metal figure of this size with separately cast arms is unusual, particularly arms held close and parallel or nearly parallel to the torso, the position that statues of priests generally display. Each of the missing arms had a square tenon fitted into corresponding mortises on each side of the torso at shoulder height. These joins tempted gravity, since no metal ‘pegs’ were inserted through the shoulders to secure the tenons, nor were the mortises notched from above, two common strategies used when the mortises are let into the body rather than the arms.10 More significantly, it is unprecedented that the legs on a solid figure such as this one were cast separately, which even more than the arms would be difficult to attach securely using the mechanical means typically employed in ancient Egypt. Circumstances that might have imposed these alterations to the limbs and garment hem are difficult to confirm. The arms and legs may have been damaged

accidentally, but it is difficult to imagine that all could have been substantially affected in a single mishap. The arms may have been replaced in order to alter their positioning, but this is unlikely to apply to the legs – the positioning of the hips and thighs show clearly that figure was striding forward – and it seems equally unlikely that the arms and legs were removed at different times. It is possible that the kilt was shortened, and that the legs were removed at the same time.

Fig. 8a, b: Details showing kilt of Kushite priest (Fig. 7); a) channel used for securing gold cladding. b) arrows indicate marks left by chisel(s) used to cut back the figure’s apron to form streamers. Note the absence of lines separating the streamers from the apron and each other. © The Metropolitan Museum of Art.

10

To judge from the mass of surviving armless or semi-armless torsos, all of the mechanical methods used by the Egyptians to attach upper limbs were inadequate for eternity. However, even over the short term, some arrangements seem more effective than others. An example of the most common arrangement, in which the tenons extend from the body and are inserted into mortises running through the top of the arms, is the black-bronze figure of Thutmose III in the Metropolitan Museum (1995.21).

A less common method, whereby tenons extending from the arms were set into mortises notched from above and held in place by gravity, is seen on the kneeling figure of Dynasty 22 King Pimay in the British Museum (EA 32747), or even more occasionally into mortises notched on the back of the shoulder, as on a Third Intermediate Period kneeling king in the Walters Art Museum (54.2094).

RITUAL METAL STATUARY IN ANCIENT EGYPT: ‘A LONG LIFE AND A GREAT AND GOOD OLD AGE’

The figure is the focus of on-going Egyptological and technical investigations that will document all alterations and hopefully can provide a basis for knowing its attributes and its posture when initially created, and the roles taken up by its first and possible subsequent owners (Hill and Schorsch, in preparation). Textual sources point to the great wealth of the Egyptian temples but, except for a rare few, the fates of these precious cult figures are unknown. A reclining falcon of hammered copper sheet with a gold head and crown and obsidian eyes is one such surviving example (Fig. 9). Found in a temple at Hierakonpolis in 1897 by Quibell (1900, 11), and carefully studied, documented and assembled by Christian Eckmann and Saher Shafik at the Egyptian Museum in Cairo nearly one hundred years later, the figure attests to successive alterations implying continued use in temple ritual over an extended period and a respected retirement when its cult activities were terminated (Eckmann and Shafik 2005; Hill 2007a, 8−9). Tentatively identified as Horus of Nekhen, this large figure was discovered where it had been carefully interred under a floor slab in the central chamber of a temple that had been established in Predynastic times (Quibell and Green 1902, 27). The reassembled fragments of the falcon and its original stand are now reunited with the gold head and doublefeather crown (CG 14717 and 52701), long on display in the Museum, that were found in the same spot (Eckmann and Shafik 2005, pls 43−5; 52, fig. 35) along with other elements of an ‘ensemble’. Among these, probably most significantly, is a cupreous figure of a striding king (JE 38849, formerly JE 32158) that had been mounted onto the base under the falcon’s beak. Rectangular and round perforations on the top and sides of the falcon’s base mark its original location; the royal figure itself has not been seen for decades (Eckmann and Shafik 2005, 54−7). In the absence of any direct parallels, as Eckmann and Shafik point out, this heterogeneous ensemble has been dated by various scholars to the Predynastic period through the New Kingdom, and they appropriately conclude that the disparities observed may best be explained by a series of alterations carried out while the cult figure remained in active use; three phases are identified. In its earliest manifestation, dating prior to Dynasty 6 (c. 2323–2150 BC), the falcon was made of wood and was finished with a layer of gesso, probably painted, that survives in traces on the inside of the copper shell, as do fragments of wood. According to Eckmann and Shafik, it was likely during Dynasty 6

259

that the wooden body was clad with several overlapping sheets of copper and capped with the skilfully hammered and delicately chased gold head. In fact, the head may have been applied sometime after the copper cladding, because there are gaps between the gold and copper sheets around the neck, through which the wood substrate would have been visible, that are otherwise difficult to explain. It is possible that the gold head had previously been used elsewhere. The distinctive assembly of the copper sheets recalls the method used to produce figures of the Dynasty 6 King Pepy I (reigned c. 2289–2255 BC) and his anonymous companion – the earliest surviving examples of metal figural statuary from Egypt – found in a nearly adjacent chapel. The final addition was the gold double-feather crown, which clearly was crafted with less skill than the falcon head. That it replaces an earlier, possibly white, crown seems likely, in view of a non-functional

Fig. 9: Cult statue of a falcon. Egypt, Hierakonpolis. Dynasty 6, c. 2323–2150 BC. Hammered copper and gold sheet, obsidian, H. c. 55‒65cm. Egyptian Museum, Cairo, various JE & CG numbers. (Eckmann and Shafik 2005, fig. 43).

260

D. SCHORSCH

cutout on the top of the falcon head (Eckmann and Shafik 2005, 67−8, and n. 127). Furthermore, the feathers were not directly attached to the crown, nor are there slits in the head itself that would have accommodated narrow tangs extending from the bottom of the feathers. Attaching the crown would have required temporarily removing the head from the body, and the mixture of gold and copper nails used for reattachment and the isolated hammer marks near the edges of the falcon head’s otherwise highly polished surface may also date to this event (Eckmann and Shafik 2005, 58−60). No conclusions can be based on the qualitative gold analyses cited by Eckmann and Shafik (2005, 68, n. 129), although the solder, possibly silver-based, that was noted on the feathers during the technical examination may be relevant for dating them. Lucas (1962, 215−6) reports silver solder on the bed of Hetepheres (Dynasty 4), but there is otherwise little evidence of solder having been used in Egypt at such an early date. This of course, may reflect lacunae in the archaeological record as well as a dearth of technical investigation. Although the royal figure has been used as justification for a New Kingdom date for the falcon, Hill (2007a, 8−9) counts it among the instances where metal statuary took innovative steps, developing forms and collocations first seen in stone sculpture only centuries later. How long this falcon cult figure was an active participant or recipient in ritual activities can only be estimated, but available archaeological evidence does suggest that it was interred at the end of the Old Kingdom, or at the latest, before the temple was modified in early Dynasty 12 (Hill 2004, 8−9; Eckmann and Shafik 2005, 68). Physical indications of ancient alterations, historical evidence and archaeological context all attest to an extended life span for a large figure of Seth in the Ny Carlsberg Glyptotek (Schorsch and Wypyski 2009) (Fig. 10). Wearing a long striped kilt and a double crown, Seth is shown in anthropomorphic form, with the slender, drooping snout of the imaginary Seth-animal. His other major feature – upright ears that widen and square off at the top – has been purposefully removed and replaced with ram’s horns (Fig. 11a, b). The figure dates to the late New Kingdom, when Seth gained prominence as a national god under the Ramessides, who ruled during Dynasties 19−20 (c. 1295−1070 BC). During the Third Intermediate Period, Seth fell gradually out of favour; his name no longer appeared in theophoric appellations, his temples

declined and no new monuments were built for his veneration. The active destruction of Seth’s image and erasure of his name are attested first during the reign of Osorkon II (c. 874−850 BC) (Sourouzian 2006, 337−8). The alterations of the Glyptotek figure probably date earlier, when the god was neglected, but not yet vilified. Regardless, the long period during which it served as a sacred image – as Seth or as Amun, whom the

Fig. 10: Seth. Egypt, New Kingdom, Dynasty 19‒20, c. 1295‒1070 BC. Solid cast unalloyed copper, arsenical copper, bronze, with silver and copper alloy inlays, H. as restored 67.7cm. Ny Carlsberg Glyptotek, Copenhagen (ǢIN614). Courtesy of the Ny Carlsberg Glyptotek / Ole Haupt.

RITUAL METAL STATUARY IN ANCIENT EGYPT: ‘A LONG LIFE AND A GREAT AND GOOD OLD AGE’

261

Fig. 11a, b: Details of Seth (Fig. 10) showing: a) original position of an upright Seth ear removed during the Third Intermediate Period; b) ram’s horns attached during the Third Intermediate Period (proper left) and in modern times. Courtesy of the Ny Carlsberg Glyptotek / Ole Haupt.

figure came to represent – is demonstrated by the circumstances of its deposition. The figure was found in Saqqara in the late 19th century with three large and several small metal statues. The small figures can longer be traced, but the three large statues are in the Louvre (Ziegler 1996, 34–5, figs 10−2). After decades of scholarly dissent regarding the dating of the male figure that was traditionally called Pachasou, Yoyotte (1958) was able to identify its owner as a Libyan – ‘Bepeshes, l’enfant du chef des étrangers’ – and place it in Dynasty 22 (c. 945−710 BC). Stylistic analysis has allowed a more precise dating of Bepeshes and the second, uninscribed, male figure to the later years of that dynasty.11 Because of Seth’s association in burial with these two figures, his lifetime in the temple must have lasted at least several hundred years. A figure of Nefertem in Leiden (Schneider and Raven 1997, 28−9, cat. no. 18; Hill 2007c, 52–5) (Fig. 12) appears to have been similarly transformed in ancient times, although presumably for different, unknown reasons. Because of its style and features of manufacture, the statue is attributed to the Third Intermediate Period, a time when three-dimensional representations of deities in metal were still relatively infrequent, and metal statuary in general was eccentric in manufacture (Hill and Schorsch 2016, 277‒80, fig. 18).

11

For citations of studies proposing this more refined dating, see Schorsch and Wypyski 2009, 183, n. 26.

In addition to physical evidence relating to the alteration of iconographic details, the long history of the figure, going back to the time it was acquired by the Rijksmuseum van Oudheden in 1826, speaks in favour of the antiquity of the alterations. The son of Ptah and Sekhmet, Nefertem is a solar deity associated with the blue lotus, which spreads its fragrant scent when it opens at dawn; his aggressive aspect comes perhaps from the association with his feline mother and other leonine goddesses. With a change of regalia, the figure of Nefertem came to represent Montu. Nefertem is typically shown in human form in a short kilt, bearded, with a uraeus and a lotus blossom flanked by menits and two feathers on his tripartite wig. Here he is wearing a feather corselet and holds his left hand raised to his chest. In this pose, Nefertem often carries an ostrich plume fan. His separately cast beard, no longer extant, is evidenced by holes under his chin and on his chest and by the straps indicated on his cheeks. A hole in the centre of the wig above his brow shows the placement of a separately cast uraeus. Three-dimensional figures of Nefertem, even examples as large as this one, often have attachment loops, an as-yet-unexplained phenomenon (Patch 2007; Hill 2007e). Relatively few representations of Montu in metal survive. A local Theban god who rose to prominence in the Middle Kingdom, Montu is generally shown with a falcon’s head, wearing a sun disc, a doublefeather crown and a double uraeus; initially a war god, Montu became a solar deity through his association

262

D. SCHORSCH

with Ra and Amun. Like Nefertem, Montu has a more or less latent aggressive aspect. On the Leiden figure, the front of Nefertem’s double loop was trimmed to accommodate Montu’s feather crown, which itself appears to have been cut down on the proper left bottom edge to fit over Nefertem’s wig (Fig. 13a, b). Clearly, instances can be cited in which ancient figures have been outfitted in modern times with matching or mismatching ancient regalia, but in this case the nature of the workmanship associated with the alterations, which also entailed the crude cutting of a hole on the top of Nefertem’s head in order to remove the lotus with menits and feathers, as well as the appearance of the cut surfaces on both the Nefertem figure and the Montu crown, strongly support the hypothesis that they were executed in ancient times.12 Differences in the appearance of the surfaces on the figure and the feather crown are presumably due to the specific interactions of their different alloys with their common burial environment, even if shared, and chemical cleaning, which they appear to have undergone. Ancient tool marks centred round the broken-off tenon on the figure’s forehead (Fig. 13c) may reflect the intentional removal of Nefertem’s uraeus. The circumstances leading to this modification are entirely unknown, and the feather crown may have come from an existing figure or was made, albeit imperfectly, to attach to the Nefertem figure. Hill (2007b, 52−5) posits that the feather corselet – often worn by Montu – may have supported the reassignment of the figure (Fig. 13d). A final example of reuse of a temple statue, or in this case, intended reuse, is a particularly interesting fragment – the feet and legs of a modest mummiform figure from a provincial context – that was apparently damaged and slated to be repaired. Still invested with a wax model compensating for its lost torso and head, the fragment was found in an eccentric deposit at Qubbet el-Hawa, in the cliffs on the west bank opposite Aswan. There, in a chamber containing Late Period (664–332 BC) burials, excavators found artefacts related to the production of metal statuary, including wax models, some already invested; clay and wood forms for moulding wax models; and failed castings of individual figures and series of figures (Auenmüller and Fitzenreiter 2014; Auenmüller 2014). While it was

12

The figure was examined in the Department of Objects Conservation at the Metropolitan Museum in 2008.

Fig. 12: Nefertem with headdress of Montu. Egypt, Third Intermediate Period, Dynasty 21‒24, c. 1070‒713 BC. Hollow cast leaded bronze, H. 36.5cm. Leiden Rijksmuseum van Oudheden (H.III.M…-2). © Rijksmuseum van Oudheden, Leiden.

RITUAL METAL STATUARY IN ANCIENT EGYPT: ‘A LONG LIFE AND A GREAT AND GOOD OLD AGE’

Fig. 13a, b, c, d: Details of Nefertem/Montu (Fig. 12) showing: a) the placement of Montu double-feather crown; b) Nefertem’s altered attachment rings; c) ancient tool marks associated with the removal of Nefertem’s uraeus; d) Nefertem’s feather corselet. © Rijksmuseum van Oudheden, Leiden.

263

264

D. SCHORSCH

first thought to belong to this group of failed castings in some stage of rehabilitation, it is now recognised that this particular fragment is the surviving portion of what was a fully executed work – insofar that it was gilded – that was to be completed for reuse by casting on the missing section (Meinel and Willer 2016, 104). Discussion In addition to the numerous early finds of metal ritual statuary within the sacred precincts of the Egyptian temples referenced in the introduction, more recent studies by Harry Smith and Sue Davies of numerous figures excavated at the Sacred Animal Necropolis at Saqqara from 1964 to 1976 provide a more nuanced view of these deposits (Davies 2007). Specifically pertinent to this study are the observation that many of the statues were already damaged when they were placed in deposits (with the implication that damaged ritual statuary was retained in sacred precincts and not recycled) and the conclusion that at least some of the caches discovered must be linked to refurbishment of the cult. Most significant is the discovery of two royal figures (Cairo JE 91436, JE 91520) dated to the Third Intermediate Period (or perhaps, in one case, even earlier) in a gallery of the Falcon Catacombs that was in use from 341–245 BC (Davies and Smith 2005, 40, 100, FCO258; 108–9, FCO-318), and of a situla from the reign of Psamtek I in a sector dated to the 4th century BC (Davies 2007, 183–4). The respect accorded to ritual statuary that was no longer intact is evident elsewhere: with one exception, the statuary produced over an extended period of years and placed with intent and care in a favissa in the Ptah temple at Karnak during the second half of the Ptolemaic period was in damaged condition at the time of its interment (Charloux et al. 2017). The case studies presented here support such archaeological evidence pointing to the extended lifespans in the temples apparently enjoyed by many of these figures; a more recent discovery in a provincial temple in the Kharga oasis provides clues to how they might have spent those decades and centuries in service. In 1993, a mud-brick temple at ‘Ayn Manawîr consecrated to Osiris-Iw (‘Osiris-has-come’) was excavated by the Institut français d’archéologie orientale (IFAO) within a modest settlement inhabited from the early 5th century until around 370 BC (Wuttmann, Coulon and Gombert 2007; Gombert-Meurice in this volume). In chapels adjacent to the sanctuary, excavators found

nearly four hundred bronze statuettes, nearly all modest representations of Osiris. Also recovered was a large section of a wooden naos still containing the lower part of a wooden mummiform figure. What makes ‘Ayn Manawîr special is that it represents an accumulation or assemblage – rather than a deposit – of statuary built up over time as the result of many acts, as the individual pieces were brought to the temple and pressed into the service of the cult. It is the working life of the statuary. Evidence of ancient alterations that has been noted on several additional metal figures deserve attention in the future. Furthermore, a closer look at early finds of deposits that contained ritual statuary of disparate dates, in so far as they can be reconstructed, is also warranted, even when the date of the actual deposition is unknown (Heinz and van der Wilt 2016; GombertMeurice in this volume). None of the hundreds of kneeling royal figures studied by Hill was cast with an integral base (Hill 2004, 131–5; Schorsch 2004), which allows one to suppose that they might have been moved from one separable base or support to another as role or function necessitated. Furthermore, not one single kneeling figure survives with a separable base, which could have been of wood or stone as well as metal (Hill 2004, 136). There are also few extant bases without associated royal figures. How then were these statues supported during cult rituals, while displayed and when stored? Were some of the figures ‘permanently’ mounted on separable bases, and if not, how might one otherwise explain damage sustained to their arms, hands, legs, feet and tangs (e.g. Hill 2004, 225, cat. no. 241) intimating that at some point they were forcibly separated from their supports? The figure of Amasis (see Fig. 3), which was modified by the addition of a second inscription, is one of the many statues that sustained damage to the wrists (rendered invisible by restoration), just as one of the Kushite kings cited earlier with regard to his modified regalia (Hill 2004, 226, cat. 243) suffered substantial damage to his arms, tangs and feet (see Fig. 1). Would this have occurred at the time when the statue was reinscribed (in the first case) or modified (in the second case) and then repurposed? Alternatively, do the damages stem from the time when the figures were forcibly removed from permanent bases and retired from active duty? Further Egyptological and technical attention to bases associated with other kinds of metal ritual temple statuary may provide explanations for some of these confusing and seemingly contradictory observations.

RITUAL METAL STATUARY IN ANCIENT EGYPT: ‘A LONG LIFE AND A GREAT AND GOOD OLD AGE’

Acknowledgements Many, many thanks to my munificent colleague in the Metropolitan Museum, Marsha Hill, for her countless contributions to my technical investigations of Egyptian metal statuary. Thanks also to Maarten Raven, Rijksmuseum van Oudheden; Mogens Jørgensen (formerly) and Rebecca Hast, Ny Carlsberg Glyptotek; Christian Eckmann, Römisch-Germanisches Zentralmuseum (Mainz) and Richard Keressey, New York, for generously providing access and sharing information.

Bibliography Abdalla, A. 1995. Finds from the sebbakh at Dendera. Göttinger Miszellen 145, 19–28. Armbruster, B. and Eilbracht, H. 2006. Technological aspects of the Viking age gold treasure from Hiddensee, Germany. Historical Metallurgy 40, 27–41. Arnold, D. 1988. The pyramid of Senwosret I. Publications of the Metropolitan Museum of Art Egyptian Expedition 22; The South Cemeteries of Lisht 1. New York. Ashkenazi, D., Iddan, N. and Tal, O. 2012. Archaeometallurgical characterization of Hellenistic metal objects: The contribution of the bronze objects from Rishon Le-Zion (Israel). Archaeometry 54, 528–48. Attanasio, D., Bultrini, G. and Ingo, G. M. 2001. The possibility of provenancing a series of bronze Punic coins found at Tharros (western Sardinia) using the literature lead isotope database. Archaeometry 43, 529–47. Auenmüller, J. 2014. Die Ergebnisse der bildgebenden Verfahren. In M. Fitzenreiter, Ch. E. Loeben, D. Raue and U. Wallenstein (eds), Gegossene Götter; Metallhandwerk und Massenproduktion im Alten Ägypten. Rahden, 113–26. Auenmüller, J. and Fitzenreiter, M. 2014. Eine Gusswerkstatt auf der Qubbet el-Hawa? In M. Fitzenreiter, Ch. E. Loeben, D. Raue and U. Wallenstein (eds), Gegossene Götter; Metallhandwerk und Massenproduktion im Alten Ägypten. Rahden, 101–6. Bakr, M. I. and Brandl, H. 2010. Precious metal hoard from Bubastis. In M. I. Bakr, H. Brandl, with F. Kalloniatis (eds), Egyptian antiquities from Kufur Nigm and Bubastis. Museums in the Nile Delta 1. Berlin, 43–53.

265

Bayley, J. and Butcher, S. 1995. The composition of Roman brooches found in Britain. In S. T. A. M. Mols, A. M. Gerhartl-Witteveen, H. Kars, A. Koster, W. J. Th. Peters and W. J. H. Willems (eds), Acta of the 12th International Congress on Ancient Bronzes: Nijmegen 1992. Nederlandse Archeologische Rapporten 18. Amersfoort, 112–9. Bisson de la Roque, M. F. 1927. Rapport sur les fouilles de Médamoud (1926). Fouilles de l’Institut français d’archéologie orientale du Caire, 4, 1. Cairo. Bothmer, B. V. 1960. Egyptian sculpture of the Late Period: 700 B.C. to A.D. 100. Brooklyn. Brandl, H. and Wenzel, G. 2010. The 1992 hoard called the ‘new treasure’ of Bubastis (catalogue entries 66–83). In M. I. Bakr, H. Brandl, with F. Kalloniatis (eds), Egyptian antiquities from Kufur Nigm and Bubastis. Museums in the Nile Delta 1. Berlin, 207–43. Charloux, G., Thiers, C., Abd el-Aziz, M., Boulet, S., Bourse, C., Guadagnini, K., Laroye, J. and elNasseh, A. M. S. 2017. The afterlife of Egyptian statues: A cache of religious objects in the temple of Ptah at Karnak. Antiquity 91, 1189–1204. Daressy, G. E. J. 1901. Rapport sur des fouilles à Sa el-Hagar. Annales du Service des antiquités de l’Égypte 2, 230–9. ———. 1902. Une trouvaille de bronzes à Mit Rahineh. Annales du Service des antiquités de l’Égypte 3, 139‒50. ———. 1917. Rapport sur des fouilles à Sag el-Hagar. Annales du Service des antiquités de l’Égypte 17, 245‒60. Davies, B. G. 1995. Egyptian historical records of the later Eighteenth Dynasty: Fascicle VI. Translated from W. Helck, Urkunden der 18. Dynastie 22. Warminster. Davies, S. 2007. Bronzes from the Sacred Animal Necropolis at North Saqqara. In Hill and Schorsch 2007, 174–87. Davies, S. and Smith, H. S. 2005. The sacred animal necropolis at North Saqqara: The falcon complex and catacomb. The archaeological report. Egyptian Exploration Society, Excavation Memoir 73. London. Eaton, K. 2007. Types of cult-image carried in divine barques and the logistics of performing temple ritual in the New Kingdom. Zeitschrift für ägyptische Sprache und Altertumskunde 134, 15–25.

266

D. SCHORSCH

Eckmann, Ch. and Shafik, S. (eds). 2005. ‘Leben dem Horus Pepi’: Restaurierung und technologische Untersuchung der Metallskulpturen des Pharao Pepi I. aus Hierakonpolis. Mainz. Edgar, C. C. 1906. Report on an excavation at Toukh el-Qaramous. Annales du Service des antiquités de l’Égypte 7, 205–12. Gener, M., Montero Ruiz, I., Murillo-Barroso, M., Manzano, E. and Vallejo, A. 2014. Lead provenance study in medieval metallic materials from Madinat al-Zahra (Medina Azahara, Córdoba). Journal of Archaeological Science 44, 154–63. Goelet Jr., O. 1996. A new ‘robbery’ papyrus: Rochester MAG 51.346.1. Journal of Egyptian Archaeology 82, 107–27. Green, C. I. 1987. The temple furniture from the Sacred Animal Necropolis at North Saqqâra, 1964–1976. London. Heinz, S. and van der Wilt, E. 2016. Buried bronzes: Caches of the Sacred Animal Necropolis. Egyptian Archaeology 49, 43–6. Hill, M. 2004. Royal bronze statuary from ancient Egypt, with special attention to the kneeling pose. Egyptological Memoirs 3. Leiden. ———. 2007a. Charting metal statuary: The archaic period through the pre-Thutmoside New Kingdom (ca. 3100–1479 B.C.). In Hill and Schorsch 2007, 7–17. ———. 2007b. Shifting ground: The New Kingdom from the reign of Thutmose III (ca. 1479–1070 B.C.). In Hill and Schorsch 2007, 23–37. ———. 2007c. Heights of artistry: The Third Intermediate Period (ca. 1070–664 B.C.). In Hill and Schorsch 2007, 51–63. ———. 2007d. Lives of the statuary. In Hill and Schorsch 2007, 153–9. ———. 2007e. Loops and metal statuary. In Hill and Schorsch 2007, 87–8. ———. 2012. Statuette of a Kushite priest, adapted for a king. Metropolitan Museum Bulletin 70 (1), 8. Hill, M. and Schorsch, D. 2005. The Gulbenkian torso of King Pedubaste: Investigations into Egyptian large bronze statuary. Metropolitan Museum Journal 40, 163–95. ———. (eds). 2007. Gifts for the gods: Images from Egyptian temples. New York. ———. 2016. Ptah’s profile. In J. van Dijk (ed.), Another mouthful of dust. Egyptological studies in honour of Geoffrey Thorndike Martin. Orientalia Lovaniensia Analecta 246. Leuven, 251‒307.

———. (in preparation). Becoming royal. Hodder, I. 1982. The present past, an introduction to anthropology for archaeologists. New York. Hölscher, U. 1954. The excavation of Medinet Habu, Volume V: Post–Ramessid remains. Ed. and trans. by Elizabeth B. Hauser. Oriental Institute Publications 66. Chicago. Jambon, E. 2009. Les fouilles de Georges Legrain dans la Cachette de Karnak (1903–1907): Nouvelles données sur la chronologie des découvertes et le destin des objets. Bulletin de l’Institut français d’archéologie orientale 109, 239‒79. Kamel, I. 1968. A bronze hoard at Athribis. Annales du Service des antiquités de l’Égypte 60, 65–71. el-Khouly, A. 1978. Excavations at the pyramid of Userkaf. 1976: Preliminary report. Journal of Egyptian Archaeology 64, 35–43. Knapp, A. B. 1988. Hoards d’oeuvres: Of metals & men on Bronze Age Cyprus. Oxford Journal of Archaeology 7, 147‒76. Kitchen, K. A. 1990. Catalogue of the Egyptian collection in the National Museum, Rio de Janeiro. Rio de Janeiro. Lauer, J.-Ph. and Picard, Ch. 1955. Les statues ptolémaïques du Sarapieion de Memphis. Publications de l’institut d’art et archéologie de l’université de Paris 3. Paris. Lilyquist, C. 2012. Treasures from Tell Basta: Goddesses, officials, and artists in an international age. Metropolitan Museum Journal 47, 9–72. Lorton, D. 1999. The theology of cult statues in ancient Egypt. In M. B. Dick (ed.), Born in heaven, made on earth: The making of the cult image in the ancient Near East. Winona Lake, IN, 123–210. Lucas, A. 1962. Ancient Egyptian materials and industries. 4th edition, revised and enlarged by J. R. Harris. London. Masson, A. 2015. Cult and trade. A reflection on Egyptian metal offerings from Naukratis. In D. Robinson and F. Goddio (eds), Thonis-Heracleion in context. Oxford Centre for Maritime Archaeology: Monograph 8. Oxford, 71–88. ———. 2016. Interpréter le matériel grec et chypriote dans un contexte religieux et thébain: L’exemple du quartier des prêtres de Karnak – des consommateurs égyptiens de produits grecs et chypriotes. In G. Gorre and A. Marangou (eds), La présence grecque dans la vallée de Thèbes. Rennes, 25–43. Meinel, D., and Willer, F. 2016. Röntgen-Mikro-Computertomographie (μCT): Virtuelle Freilegung

RITUAL METAL STATUARY IN ANCIENT EGYPT: ‘A LONG LIFE AND A GREAT AND GOOD OLD AGE’

innenliegender Strukturen. In M. Fitzenreiter, F. Willer and J. Auenmüller, Materialien einer Gusswerkstatt von der Qubbet el-Hawa. Bonner Aegyptiaca. Berlin, 82‒117. Naville, E. 1887. The shrine of Saft el Henneh and the land of Goshen (1885). Egypt Exploration Fund Memoir 5. London. Ornan, T. 2012. The role of gold in royal representation: The case of a bronze statue from Hazor. In R. Matthews and J. Curtis (eds), Proceedings of the 7th international congress on the archaeology of the ancient Near East: 12 April–16 April 2010, the British Museum and UCL, London, 3 vols. Wiesbaden, I, 445–58. Ortiz, G. 1996. In pursuit of the absolute: Art of the ancient world from the George Ortiz collection. Revised edition. Bern. Patch, D. C. 2007. Nefertem. In Hill and Schorsch 2007, 143–5. Pendlebury, J. D. S. 1931. Preliminary report of excavations at Tell el-’Amarna 1930–1. Journal of Egyptian Archaeology 17, 233–44. Petrie W. M. F. 1886. Naukratis I, 1984–5. Egyptian Research Account 3rd Memoir. London. ———. 1896. Koptos. London. Philip, G. 1988. Hoards of the Early and Middle Bronze Ages in the Levant. World Archaeology 20, 190–208. Ponting, M. J. 2002. Roman military copper-alloy artefacts from Israel: Questions of organization of ethnicity. Archaeometry 44, 555–71. Quibell, J. E. 1900. Hierakonpolis I. Egyptian Research Account 4th Memoir. London. Quibell, J. E., and Green, F. W. 1902. Hierakonpolis II. Egyptian Research Account 5th Memoir. London. Robins, G. 2005. Cult statues in ancient Egypt. In N. H. Walls (ed.), Cult image and divine representation in the ancient Near East. Boston, 1–12. Roehrig, C. A. 1996. Cult figure of a falcon-headed deity. In Ancient art from the Shumei Family Collection. New York, 4–7. Roth, A. M. 2005. Erasing a reign. In C. H. Roehrig, R. Dreyfus and C. A. Keller (eds), Hatshepsut – from queen to pharaoh. New York, 277‒81. Russmann, E. R. 1979. Some reflections on the regalia of the Kushite kings of Egypt. In F. Hintze (ed.), Africa in antiquity: The arts of ancient Nubia and the Sudan. Proceedings of the symposium held in conjunction with the exhibition, Brooklyn, September 29–October 1, 1978. Berlin, 49–53.

267

———. 1981. An Egyptian royal statuette of the eighth century BC. In W. K. Simpson and W. M. Davis (eds), Studies in ancient Egypt, the Aegean, and the Sudan: Essays in honor of Dows Dunham on the occasion of his 90th birthday, June 1, 1980. Boston, 149–55. Sauneron, S., and Yoyotte, J. 1952. La campagne nubienne de Psammétique II et sa signification historique. Bulletin de l’Institut français d’archéologie orientale 50, 157–207. Schneider, H. D., and Raven, M. J. 1997. Life and death under the pharaohs: Egyptian art from the National Museum of Antiquities in Leiden. The Hague. Schorsch, D. 2004. Appendix 3: Technical examination of Catalogue no. 242. In Hill 2004, 259–60. ———. 2007. Fragment of a wig. In Hill and Schorsch 2007, 32–3. Schorsch, D. and Wypyski, M. T. 2009. Seth, ‘figure of mystery’. Journal of the American Research Center in Egypt 45, 1–24. Schoske, S. 1988. Berichte der Staatlichen Kunstsammlungen – Neuerwerbungen: Staatliche Sammlung Ägyptischer Kunst. Münchner Jahrbuch der bildenden Kunst. Dritte Folge 39, 203–12. Sotheby’s. 2001. Egyptian, Classical, and Western Asiatic antiquities, including property from the collection of the late Marion Schuster, Lausanne. Sale cat., December 7, 2001. New York. Sourouzian, H. 2006. Seth fils de Nout et Seth d’Avaris dans la statuaire royale ramesside. In E. Czerny, I. Hein, H. Hunger, D. Melman and A. Schwab (eds), Timelines: Studies in honour of Manfred Bietak, 3 vols. Orientalia Lovaniensia Analecta 149. Leuven; Paris; Dudley, MA, vol. I, 331–54. Stefanova, M. 2015. Vessels, ingots, and chains from the El-Tod treasure. In A. Oppenheim, Do. Arnold, Di. Arnold and K. Yamamoto (eds), Ancient Egypt transformed: The Middle Kingdom. New York, 287–8. Tourna, E. 2007. Kushite king. In Hill and Schorsch 2007, 92–4. Wainwright, G. A. 1925. A hoard of silver from Menshah, Girga Mudiriah. Annales du Service des antiquités de l’Égypte 25, 120–34. Winlock, H. E. 1941. The temple of Hibis in el Khārgeh oasis Part I: The excavations. Publications of the Metropolitan Museum of Art Egyptian Expedition 13. New York.

268

D. SCHORSCH

Wuttmann, M., Coulon, L., and Gombert, F. 2007. An assemblage of bronze statuettes in a cult context: The temple of ‘Ayn Manâwir. In Hill and Schorsch 2007, 167–73. Yoyotte, J. 1958. Le dénommé Mousou. Bulletin de l’Institut français d’archéologie orientale 57, 81–9.

Ziegler, C. 1981. Une découverte inédite de Mariette, les bronzes du Sérapéum. Bulletin de la Société française d’Égyptologie 90, 29‒45. ———. 1996. Jalons pour une histoire de l’art égyptien: La statuaire du métal au Musée du Louvre. Revue du Louvre et Musées de France 46 (1), 29–38.

STATUES IN LATE ANTIQUE EGYPT: FROM PRODUCTION AND DISPLAY TO ARCHAEOLOGICAL RECORD Troels Myrup KRISTENSEN

Once relatively neglected, the study of late antique statuary is now a flourishing field (for a recent overview, see Witschel 2015). Indeed, immediately prior to the British Museum’s Annual Egyptological symposium on ‘Statues in Contexts’, two major studies of the life and afterlife of sculpture in Late Antiquity appeared. First, the Oxford-based Last Statues of Antiquity project published its final results as a print volume containing interpretive essays exploring different aspects of the decline of the so-called ‘statue habit’, accompanied by a database compiling both archaeological and textual evidence for newly erected sculpture between AD 284 and 620 (Smith and Ward-Perkins 2016).1 Second, in The Afterlife of Greek and Roman Sculpture, the outcome of two seminars held in Aarhus, Denmark, an international group of scholars investigates the wide range of late antique responses and practices that have

produced the archaeological records in which a very large number of Classical statues have been found (Kristensen and Stirling 2016a). These two volumes apply different, but laudably supplementary, approaches to late antique sculpture and help us to understand the sculptural landscapes across the late antique Mediterranean as they developed during a crucial period of social and religious change in which three-dimensional representation was slowly but surely losing ground. Yet at the same time, neither of these new books offers much in-depth discussion of Egypt, the focus of this particular volume.2 This paper explores two aspects of Egyptian statues in Late Antiquity: first, the relatively limited corpus of late antique sculpture from Egypt and its display in public, religious and funerary contexts; and second, the re-display or modification of earlier sculpture, a much wider but equally important topic, not least when we consider the large quantities of older statuary that have been discovered in late antique contexts. Why the practice of dedicating statues, an integral part of Roman society and culture (Stewart 2003), came to an end is a fundamental question, but our discussion should not end there. Although new statues were not produced in Egypt after the 5th century, old statues continued to be part of the fabric of cities and sanctuaries as well as more private settings, such as houses and villas, for centuries, in some cases even millennia (for discussion of some of this evidence, see Kristensen 2013, 107– 95). A dual focus on both new and old statuary is, therefore, essential in order to understand the contexts in which sculpture was put to use in Late Antiquity. Understanding the responses that ‘residual statues’ solicited indeed remains one of the most fundamental challenges that any contextual study of statuary in the late and post-antique Mediterranean must confront.3

1

2

Abstract This paper discusses the viability and ultimate decline of the statue habit in late antique Egypt. The first section provides a brief overview of the corpus of newly erected late antique sculpture from Egypt and attempts to outline some of the different contexts in which it was displayed, although such efforts are hindered by the fact that solid archaeological data is hard to find. The second section considers the multiple ‘afterlives’ of sculpture, i.e. the re-display or modification of earlier sculpture in late antique Egypt, and suggests that future scholarship would benefit by adopting the life history approach in its work on the numerous cases of ‘residual’ statuary that was present in Late Antiquity (as well as later periods). * * *

The accompanying database is available at and is cited below by LSA numbers. Entries for Egypt are authored by Marianne Bergmann, Ulrich Gehn and Julia Lenaghan.

3

The one exception is the essay by Gehn and Ward-Perkins 2016, which devotes six pages to the evidence of newly erected statuary in late antique Egypt. I here borrow the phrase ‘residual statuary’ from Lea Stirling, see Kristensen and Stirling 2016b.

270

T.M. KRISTENSEN

Production and display of new statuary This first section looks at the production and display of new statuary in late antique Egypt (Riggs 2015). Late antique Alexandria was a major urban centre that must have had an extraordinarily rich assortment of Egyptian, Hellenistic and Roman statuary on display. However, too little of this survives to present a coherent picture of the placement of statuary within public and private spaces.4 The growth of the modern city and the haphazard nature of much early exploration have simply not provided ideal conditions for a contextual study of statuary. The few pieces of late antique statuary that have been found outside Alexandria (Athribis, Crocodilopolis) similarly come with virtually no useful information that would allow us to associate them with a specific archaeological context. The data collected by the Last Statues of Antiquity (LSA) project makes it easy to assess the production and display of statuary in late antique Egypt in quantitative terms. For a period of roughly 340 years (AD 284–620), LSA identified twenty-four bases and twenty statues that can be linked to Egyptian contexts (Gehn and Ward-Perkins 2016). In addition, the LSA database includes three literary references to newly erected sculpture across this period, all related to Alexandria. Egypt was clearly not a province where the Classical ‘statue habit’ thrived, in contrast to (for instance) Asia Minor, where there is evidence of about ten times as many statues and bases for the same period (Lenaghan 2016). It must be said that there are some limitations to the corpus collected by LSA: while inscribed bases have the great advantage of being relatively well dated, as a medium for understanding the ‘life history’ of a statue they only tell us of one particular moment or stage in its life. Furthermore, in Late Antiquity it is not always the case that an inscription was contemporary with the statue that it adorned, since reuse and re-carving became increasingly common during this period. While this may not have mattered to contemporary viewers, it does say something important about the increasing difficulties of sourcing entirely new statuary. Even more importantly, the inscribed bases relate to only one

4

In particular when compared with many late antique cities in Asia Minor, such as Aphrodisias (Smith 2016) and Ephesus

particular type of sculpture, namely honorific portrait statuary, which was certainly not the only genre produced in late antique Egypt even though it was a key medium for the purpose of public representation. Cult statuary and other types of sculpture set up in temples and private houses were not supplied with such useful aids for future archaeologists. For the dating of these statues, as well as those whose bases are not extant, we rely on stylistic criteria that have often attracted controversy or remain relatively crude assessments of quality, especially in a provincial setting. Scholars of late antique sculpture are thus faced with considerable problems of chronology that also affect quantitative assessment (Stirling 2005, 91–137; Witschel 2015). In the case of Egypt we are even worse off, as research traditionally has focused on the pharaonic period and much less frequently on later periods. LSA has furthermore refrained from including mythological statuary, such as the statuettes that belong to the so-called Sidi Bishr group, consisting of thirteen sculptures, some of which are now on display in the Bibliotheca Alexandrina Antiquities Museum (Fig. 1). Some scholars have dated a statuette of Aphrodite that belongs to this group to the late 4th century (suggesting a date of c. 400, Hannestad 1994, 123–6; Stirling 2005, 100–1). The details of the group’s archaeological context (they were supposedly deposited together in a well), and even its discovery, are sketchy, which makes it very difficult to interpret the significance of what is nonetheless an important piece of evidence for the kind of statuary that was available to late antique Alexandrian customers for the decoration of their houses. With these provisos in mind, we can return to the evidence compiled by LSA to explore some of the new statuary that was erected in late antique Egypt. The surviving statuary compiled by LSA is all categorised as portraits and made of a variety of different materials: ten are porphyry, the precious local stone that was reserved for imperial portraiture which remained in production until at least the middle of the 4th century; eight are marble, and thus imports into Egypt (and as such evidence of continued trading links); and limestone and bronze account for a single item each, although for the latter material, inscriptions attest to

(Auinger and Sokolicek 2016). An exception is the statuary from the Polish Kom el-Dikka excavations: Kiss 1988.

STATUES IN LATE ANTIQUE EGYPT: FROM PRODUCTION AND DISPLAY TO ARCHAEOLOGICAL RECORD

several images that have since been lost.5 None of the surviving statues provides much in terms of archaeological data that would allow us to place them within a specific spatial setting, although many of the imperial portraits were clearly intended for display as part of public monuments. Among the private portraits, a himation-clad statue of a young man, dated to the late 3rd or early 4th century, comes from a funerary context, as it was discovered in a necropolis at modern Miniet el-Basal that was in use from the Hellenistic period onwards.6 Unfortunately, its precise relationship to a specific burial is not clear. A now headless togate portrait statue, dated to the 3rd to early 4th century, may also have come from a necropolis in the Sidi Gabir district; but in this case we once again lack precise archaeological data.7 When it comes to the chronology of the epigraphic evidence for newly erected statuary in Egypt, we are generally on firmer ground. In a number of cases, inscriptions can even be placed within an archaeological context. The majority of the inscriptions (totalling twenty of twenty-four cases) are dated to the Tetrarchic and Constantinian periods and offer testimony to a small number of relatively high-profile public relations projects that included the dedication of portrait statues in specific contexts, where they served to underline both imperial and local claims to authority (Gehn and Ward-Perkins 2016, 109). About half of the inscriptions come from Alexandria, where they adorned the bases of both imperial statues and portraits of civic officials, such as the praetorian prefect Maternus Cynegius, who had received imperial permission to set up a bronze togatus in an unspecified loco celeberrimo at some date between 384 and 387 (Gehn and WardPerkins 2016, 113–4).8 Again, we are faced with the complex situation of modern Alexandria that has left us with virtually no solid archaeological data to place these inscriptions in a broader public setting. We are much better off in the case of the other half of the corpus of late antique statue bases, some thirteen examples in all, which come from the military camp built inside the Luxor temple during the reign of Diocletian.9 One of the bases found in 1888 can still be

5

Porphyry: LSA-836, 846, 1003, 1005, 1007-1009, 1051, 2425, 2561, 2673. Limestone: LSA-846. Marble: LSA-1560, 1561, 1663, 1664, 2101, 2129, 2130, 2676. Bronze: LSA-1028 (suspected to be a modern copy), also attested in at least two inscriptions (LSA-872, 877), and see Gehn and Ward-Perkins 2016, 113.

Fig. 1: Aphrodite from Sidr Bishr © Bibliotheca Alexandrina Antiquities Museum/ photograph by M. Mounir.

6 7 8 9

LSA-2101. LSA-2129. LSA-872. LSA-1179, 1180, 2621–2631.

271

272

T.M. KRISTENSEN

seen in situ in front of the so-called imperial cult room (Fig. 2).10 The bronze statue that originally stood on top of the base was set up in honour of Constantine by Valerius Rometalca, the dux or commander of troops in Egypt and Libya. The exceptional collection of statuary at Luxor (even if sadly only glimpsed through their inscriptions) is testament to the continued role of statuary in creating and maintaining links of authority on an imperial scale. The fact that the language of the inscription in this and many other cases (in Thebes as well as Alexandria) is Latin highlights the fact that the target audience was not the local population. After Constantine, evidence for new honorific statues becomes very thin on the ground, although the statue habit did experience a small ‘renaissance’ in the late 4th century, albeit with only three examples documented by LSA (Gehn and Ward-Perkins 2016, 113). The latest evidence for the erection of a portrait statue in Egypt comes from a literary source, Theophanes the Confessor, who noted that the people of Alexandria set up a statue of Anastasius in 501.11 This comes almost one hundred years after the last statue dedication documented by a dated statue base. Bearing in mind that Theophanes (c. 758/760–817/818) was writing several hundred years after the event that he describes, we should perhaps be careful how much precision we read into his account, although it certainly cannot be ruled out that such an imperial portrait was erected at this time. Furthermore, it is a quite different matter if the portrait was ‘new’ in the sense of being a newly produced piece of sculpture. This short overview of the data recently compiled by LSA makes it clear that the late antique Egyptian statue habit was confined to very particular contexts, such as the installation of the Diocletianic military camp in the Luxor Temple as well as a small number of private funerary monuments in Alexandria. When the system of government in which statues constituted a significant form of capital broke down, so did the statue habit in public and military contexts (Anderson 2016; Liverani 2016; Ward-Perkins 2016). It is much more difficult to assess how these changes affected statuary in private contexts, however, and here we are confronted with the intractable problems of chronology and the lack of proper archaeological documentation. Hopefully future work will be able to remedy this situation.

10

LSA-1180. On the wall paintings in the imperial cult room, see Jones and MacFadden 2015.

Fig. 2: Valerius Rometalca base, Luxor temple. Photograph: Troels Myrup Kristensen.

Old statues in new contexts While new statuary was rare in late antique Egypt, old statuary remained on view in a range of different settings and experienced a myriad of different fates. Arguably, much statuary that had been on display was ignored and simply forgotten over time. Other pieces represented a useful resource that could be exploited for different purposes. Marble, for example, was easily burned into lime for building materials, a practice that continued into very recent times in Egypt (Greenhalgh 2009, 448–68; 2012, 104–7; Munro 2016; Bonnie 2016). Other statues were more ideologically charged, such as cult statuary in temples that, for some Christian viewers, represented the epitome of pagan practices. Christian attacks on such statues are a staple in much late antique Egyptian literature, although these accounts can never be taken at face value (Kristensen 2013, 76–89; 2014; Dijkstra 2015). However, they do pro-

11

LSA-2754.

STATUES IN LATE ANTIQUE EGYPT: FROM PRODUCTION AND DISPLAY TO ARCHAEOLOGICAL RECORD

vide important insights into how Christians imagined and represented performances of iconoclasm, such as the famous case of the Alexandrian Serapeum whose cult statue was subjected to a fierce attack in 391/392 (on Christian destruction of pagan statuary: Kristensen 2013; 2014; 2015a; forthcoming). After the cult statue had been torn to pieces, the fragments were distributed across the city, where they were put to the fire. The archaeological record in Egypt (and other parts of the late antique world, it should be noted) suggests a range of different Christian responses to pagan sculpture, including deliberate destruction as well as many more subtle appropriations and adaptations (on the transformation and mutilation of statues across time in Egypt, see also Connor in this volume). Erasure of the thousands of relief images that adorned the walls of temples converted into churches required considerable effort and determination by the attackers, and was consequently rare (Sauer 2003; Kristensen 2013; most recently Wong 2016, which looks in detail at the destruction and mutilation of reliefs at Edfu and Dendera). Selective destruction of particularly potent body parts of images was therefore widespread, both in Egypt and in other parts of the late antique world

(Kristensen 2013, 175–92). Evidence for this practice has also been noted outside the Egyptian context (for recent examples from the Athenian Agora, see Martens 2015 and Riccardi 2015). Behind such selective and ideological responses lay fundamental discussions of divine presence and the media of representation that are important to approach from both a textual and a material dimension (Kristensen 2013, 175–92; Peppard 2015). However, work on this issue is also complicated by the difficulty of identifying Christian agency in each case, let alone dating individual responses, as in most cases we have very little solid archaeological data to work with (for a recent discussion of some of this complexity: Anderson 2017 on the case of the Parthenon; for a more general account of the complexities of working with the Egyptian evidence: Kristensen 2013, 109– 18). While this may lead to a certain degree of speculation, it is important to ask questions to stimulate further research and hopefully also inspire better documentation in the future. The so-called ‘life history approach’ provides us with one method to explore the issue of what happened to the old statues that continued to be present in the cityscapes and sanctuaries of late antique Egypt. Fig. 3

Construction matter ma e ials

Quarries

Lime kiln

Raw Materials

Primary display

Continued primary display

? Pass ssiive re retten enttition Secondary S d display di l

Carv r ing

Systemic Use

Destroy, discard, or deposit p

Archaeological Record (invisible)

273

Recarv r ing; mutilation; cross-marking Destroy, discard, or depossitt

Reclamation

Fig. 3: Flow diagram showing the life histories of sculpture. Diagram by L. Stirling, illustration: J. Troost, after Kristensen and Stirling 2016b, 7, fig. 2.

274

T.M. KRISTENSEN

is thus an attempt to illustrate all the possible stages in the ‘life’ of a statue, based on work by the behavioural archaeologist Michael Schiffer, who developed his more general model to illustrate what he called the life history of objects. In broad terms, the adaptation of Schiffer’s model (taken from the recent The Afterlife of Greek and Roman Sculpture) follows the life of a statue from quarrying, carving and primary dedication to continued display, where it could be subjected to re-carving, mutilation, and so on, until its ‘death’ by means of destruction, discarding or deposition, and thus entry into the archaeological record. The advantage of the life history approach is that it gives equal attention to all the contexts in which people engaged with statues, and thus not only production or primary display — which are what archaeologists and art historians are most often interested in. The main challenge that the model presents is that it can be very difficult indeed to find solid archaeological data for the chronology and meaning of individual responses. This becomes abundantly clear when we consider the Egyptian evidence in more detail. First of all, it should be noted that reuse was not a new phenomenon to Late Antiquity. For example, recent excavations by the Supreme Council of Antiquities in the Sphinx Avenue at Karnak discovered that bodies of statues of sphinxes had been used already as bedding materials in various Roman period structures (Boraik 2014, 211). However, as temples and other buildings went out of use in Late Antiquity, more and more materials could be repurposed in a number of different contexts. For instance, a pragmatic case of reuse has been documented at Aswan, where relief-decorated blocks were taken from a temple, clearly no longer in use, and repurposed in the construction of the late antique town wall (Müller 2014, 61).12 It is also clear from epigraphy that by Late Antiquity certain statues had lost their original identities, making them ripe for reuse. At least one Egyptian case thus attests to a relatively cheap alternative to starting from scratch: a red granite statue base from Antinoopolis.13 On one side (which would have been the base’s original front) is an inscription dedicated to Antinoos, dated to the 130s, and on its top, a pair of ‘footprints’ for the placement of a monumental bronze statue is visible. At some time between 388 and 390, the base was turned

around and reinscribed with a dedication by the clarissimus Flavius Eutolmius Tatianus to the emperors Valentinian II, Theodosius I, Arcadius and Honorius. Whether the accompanying statue (now lost) was recarved or simply given a new identity by means of an inscription is not clear. It is also unclear which of the first emperors was represented. The name of the dedicant was later erased, providing testimony of a third stage in the life of this particular monument. A more ideologically driven kind of response is demonstrated in a number of cross-marked statues that have been found in Egypt (Fig. 4) (Kristensen 2012, 59, cat. no. B10). One of these, a basalt portrait of Germanicus, was prominently featured in the British Museum exhibition on post-pharaonic Egypt (O’Connell 2015, 96; discussed in more detail in Kristensen 2013, 93–6).

12

13

It is interesting to note the number of outward-facing reliefs in this case.

Fig. 4: Portrait of Germanicus with cross-marking on its forehead, British Museum. Photograph: Troels Myrup Kristensen.

LSA-876. It is unclear if a marble torso (now sadly lost) found near the base belonged to its second phase of use.

STATUES IN LATE ANTIQUE EGYPT: FROM PRODUCTION AND DISPLAY TO ARCHAEOLOGICAL RECORD

A cross has been carved on its forehead, and I have previously discussed the meaning of this response in considerable detail. From Alexandria come another three sculptures that have been treated in a similar way (Kristensen 2012, 55, cat. nos A8–9). The most prominent among these is an over-life-size cuirassed statue of Marcus Aurelius, found in Horeya Street some time around 1872 (Fig. 5). It was found as part of a cache of sculptures, although this was the only one to have been furnished with crosses.14 Sadly the find circumstances of this remarkable statue are anything but clear, which makes it difficult to place this response within its archaeological context. More broadly speaking, the practice of cross-marking sculpture is an empire-wide phenomenon of which I have compiled some fifty-seven examples (Kristensen 2012).15 The reason why a particular statue was targeted in this way is likely to have differed from case to case, depending on local circumstances of Christianisation and specific episodes of violence or simply appeasement. The relatively high number of examples from Alexandria could point to a more orchestrated campaign, although it is unfortunate that we know so little of the archaeological contexts to which the cross-marked sculptures belong. At any rate, we must take these examples of cross-marked statues as evidence for a particular modus operandi in which some pagan sculptures remained on view in public spaces after some adjustments were made to mark them out as Christian bodies. As such the cross-marked statues constitute examples of continued primary display, as outlined in the model of the life history approach in Fig. 3. As for understanding how statues entered the archaeological record, Egyptian evidence is sadly hard to come by. The previously mentioned Sidi Bishr group that was discovered in 1973 has been interpreted as a so-called ‘cachette’, a collection of sculptures that were buried together. Unfortunately, the archaeological evidence for this interpretation is non-existent, and it is very difficult to say how (and even if) the sculptures came to be buried together.16 If the interpretation holds up, the pristine condition of these sculptures makes it

14

15 16

Other sculptures in this ‘cache’ include a nude torso, an IsisTyche and a possible portrait of Alexander the Great. See addenda to Kristensen 2012 on Academia.edu. Important for better understanding of this find is the stillunpublished work on so-called ‘cachettes’ by Silviu Anghel, who notes that the assemblage of the Sidi Bishr sculptures may not be ancient.

275

likely that we should see their deposition as a neutral form of response. In closing, it is useful to take a slightly closer look at two cases in which we can begin to reveal some of the complexity that was involved at the different stages in the life histories of Egyptian statues over the course of Late Antiquity, not least considering that they provide us with a certain amount of archaeological detail.

Fig. 5: Alexandria Marcus Aurelius, Alexandria Graeco-Roman Museum. Photograph: J.F. Gout, after Empereur 2000.

276

T.M. KRISTENSEN

The first case comes from the Dutch–American excavations at Berenike on the Red Sea (Sidebotham, Wendrich and Hense 1996, 229–42; Sidebotham 2011, 64–6). In the so-called ‘Shrine of the Palmyrenes’, built in the early 3rd century AD, excavators found 143 fragments belonging to a nearly life-size bronze statue of a female deity originally placed on a gypsum base (Fig. 6). Further finds include a number of other gypsum objects, an offering table and incense burners, indicating that we are firmly within a cultic context. The fragments were found in levels that could be dated by ceramic evidence to the late 4th or early 5th centuries, a time when the temple no longer served its original function. The statue’s high degree of fragmentation suggested to the excavators that it had been subject to gradual decay

Fig. 6: Reconstruction drawing of the so-called ‘Shrine of the Palmyrenes’, Berenike. Courtesy of Martin Hense.

rather than a single episode of destruction. However, no fragments of the head, right arm and right half of the upper body were uncovered, suggesting that specific parts of the statue may have been targeted at some point. This recent find, although difficult to interpret, demonstrates some of the potential of new data when it comes to fully recovering the chronology and mode of response in Late Antiquity. The second case presents less in terms of archaeological documentation, but is valuable nonetheless, especially from the perspective of the life history model. This is the well-known ‘Philosopher’s Circle’ at Saqqara, which provides further insights into the continuous life of a monument across at least seven centuries (Fig. 7). The original Hellenistic monument included nine life-size statues made from a locally sourced limestone, set up in a semi-circle next to the dromos, the major processional way through the sanctuary of Serapis (Lauer and Picard 1955; Ridgway 1990, 131–4). The sculptures are still displayed outdoors at the archaeological site, where they have weathered badly since their discovery in the 19th century. The original arrangement has attracted considerable interest among scholars, but what is more interesting in the present context is the evidence for the way in which these sculptures were handled over time. For instance, on the back of one of the statues of Herakleitos there is a small carving of a man (Fig. 8). This shows one of the ways in which people engaged with statues, which is very similar to the way public art is occasionally treated in modern times. We can imagine the scene of a spectator waiting for or watching the parade on the processional way, keeping himself busy carving a small relief in a hidden spot, or perhaps even depicting a member of the procession. Obviously, such informal and ephemeral acts of response are very difficult to date. Even more intriguing are the limestone blocks that propped up the sculptures when they were originally excavated. To the publishers of the monument, this suggested that the statues had been taken down at some point in Late Antiquity and then eventually been put back up again. They suggested that this was most likely to have happened during the reign of Julian the Apostate — or alternatively under the influence of Neoplatonism at the beginning of the 4th century (Lauer and Picard 1955, 6–7). Although this chronology is sketchy, and much work is still needed to understand what happened here, the case of the Philosophers’ Circle suggests that many statues in late antique Egypt experienced an incredibly complex afterlife that

STATUES IN LATE ANTIQUE EGYPT: FROM PRODUCTION AND DISPLAY TO ARCHAEOLOGICAL RECORD

incorporated a rich variety of responses, positive as well as negative. Other cases, such as perhaps our sanctuary at Berenike, may be suggestive of another, more neutral option (simple forgetfulness or neglect) that was responsible for the ultimate entry of statuary into the archaeological record. Conclusion The two cases discussed at the end of this article resonate in important ways with the overall model of the different responses to sculpture represented in Fig. 3. As the life history model suggests, the story only really begins once a statue has been erected Fig. 7: Philosopher’s circle, Saqqara, as currently displayed. and people start doing things with and Photograph: Troels Myrup Kristensen. around it. While some new statues were clearly erected in late antique Egypt, most of the sculptural landscape consisted of old, retained statues that triggered a variety of responses which in turn reveal many different stories of late antique religious and social change. It is my hope that this volume will stimulate future work on Egyptian sculpture that addresses not only production and primary display, but also the entire life span of statues. More work is also needed to dig deeper into the literary functions of Christian narratives of idol destruction as topoi, such as Shenoute’s encounter with the cryptopagan Gesios and his collection of pagan sculptures. These textual accounts of iconoclasm played a very particular role in the Christian imagination in relation to the ontology of images as well as in the construction of identity and memory (on Shenoute’s attacks on the idolatry of Gesios, see Brakke and Crislip 2015, 193– 297). Indeed, even though three-dimensional representation certainly declined during this period, Classical motifs lived on in Egypt in other media, as evident from, for example, the rich evidence of textiles with mythological scenes that have been found in funerary contexts (see most recently Kristensen 2015b; Thomas 2016). Lastly, we should keep in mind that the story of Egyptian statuary does not end with Late Antiquity. Obviously, a variety of later practices and responses constitute important components in the making of the archaeological record. The widespread reuse and Fig. 8: Philosopher’s circle. After Lauer and Picard 1955, 141, fig. 80. destruction of antiquities (including statuary) in

277

278

T.M. KRISTENSEN

19th-century Egypt is well documented (see e.g. Greenhalgh 2012, 72–4, which investigates this process through the lens of travel reports and guidebooks, sources with which Egypt is especially well supplied from as early as the 1860s. See also Kristensen 2013, 110–1; Greenhalgh 2016). These later episodes are equally important to fully document the entire life histories of Egyptian statuary, from production and primary display to deposition in the archaeological record. Bibliography Anderson, B. 2016. The disappearing imperial statue: Toward a social approach. In Kristensen and Stirling 2016a, 290–309. ———. 2017. The defacement of the Parthenon metopes. Greek, Roman and Byzantine Studies 57, 248–60. Auinger, J. and Sokolicek, A. 2016. Ephesus. In Smith and Ward-Perkins 2016, 160–73. Boraik, M. 2014. SCA excavations at Luxor: New discoveries from the first millennium AD. In E. O’Connell (ed.), Egypt in the first millennium AD. Perspectives from new fieldwork. British Museum Publications on Egypt and Sudan 2. Leuven, 207–13. Bonnie, R. 2016. Thrown into limekilns: The reuse of statuary and architecture in Galilee from Late Antiquity onwards. In J. Day, R. Hakola, M. Kahlos and U. Tervahauta (eds), Spaces in Late Antiquity. Cultural, theological and archaeological perspectives. London, 190–211. Brakke, D. and Crislip, A. 2015. Selected discourses of Shenoute the Great: Community, theology, and social conflict in late antique Egypt. Cambridge. Dijkstra, J. 2015. ‘I wish to offer a sacrifice to God today’: The discourse of idol destruction in the Coptic Life of Aaron. Journal of the Canadian Society for Coptic Studies 8, 61–75. Empereur, J.-Y. 2000. A short guide to the GraecoRoman Museum, Alexandria. Alexandria. Gehn, U. and Ward-Perkins, B. 2016. Egypt, the Near East, and Cyprus. In Smith and Ward-Perkins 2016, 109–19. Greenhalgh, M. 2009. Marble past, monumental present. Building with antiquities in the mediaeval Mediterranean. Leiden. ———. 2012. Constantinople to Córdoba. Dismantling ancient architecture in the East, North Africa and Islamic Spain. Leiden.

———. 2016. Travelers’ accounts of Roman statuary in the Near East and North Africa: From limbo and destruction to museum heaven. In Kristensen and Stirling 2016a, 330–48. Hannestad, N. 1994. Tradition in late antique sculpture. Conservation, modernization, production. Aarhus. Jones, M. and MacFadden, S. (eds). 2015. Art of empire. The Roman frescoes and imperial cult chamber in Luxor Temple. New Haven. Kiss, Z. 1988. Alexandrie IV. Sculptures des fouilles polonaises à Kôm el-Dikka 1960–1982. Warsaw. Kristensen, T. M. 2012. Miraculous bodies: Christian viewers and the transformation of ‘pagan’ sculpture in Late Antiquity. In S. Birk and B. Poulsen (eds), Patrons and viewers in Late Antiquity. Aarhus, 31–66. ———. 2013. Making and breaking the gods. Christian responses to pagan sculpture in Late Antiquity. Aarhus. ———. 2014. Using and abusing images in Late Antiquity: Column monuments as topoi of idolatry. In S. Birk, T. M. Kristensen and B. Poulsen (eds), Using images in Late Antiquity. Oxford, 268–82. ———. 2015a. Iconoclasm. In E. Friedland, M. G. Sobocinski and E. K. Gazda (eds), The Oxford handbook of Roman sculpture. Oxford, 667–80. ———. 2015b. Dressed in myth: Mythology, eschatology, and performance on late antique Egyptian textiles. In H. Leppin (ed.), Antike Mythologie in christlichen Kontexten der Spätantike. Berlin, 263–96. ———. Forthcoming. Statues. In W. Caraher, T. Davis and D. Pettegrew (eds), The Oxford handbook of early Christian archaeology. Oxford. Kristensen, T. M. and Stirling, L. M. (eds). 2016a. The afterlife of Greek and Roman sculpture. Late antique responses and practices. Ann Arbor. ———. 2016b. The lives and afterlives of Greek and Roman sculpture: From use to refuse. In Kristensen and Stirling 2016a, 1–24. Lauer, J.-P. and Picard, C. 1955. Les statues ptolémaïques du Sarapieion de Memphis. Paris. Lenaghan, J. 2016. Asia Minor. In Smith and WardPerkins 2016, 98–108. Liverani, P. 2016. The sunset of 3D. In Kristensen and Stirling 2016a, 310–29. Martens, B. A. 2015. Sarapis as healer in Roman Athens: Reconsidering the identity of Agora S 1068. In M. M. Miles (ed.), Autopsy in Athens.

STATUES IN LATE ANTIQUE EGYPT: FROM PRODUCTION AND DISPLAY TO ARCHAEOLOGICAL RECORD

Recent archaeological research on Athens and Attica. Oxford, 51–65. Müller, W. 2014. Syene (ancient Aswan) in the first millennium AD. In E. O’Connell (ed.), Egypt in the first millennium AD. Perspectives from new fieldwork. British Museum Publications on Egypt and Sudan 2. Leuven, 59–69. Munro, B. 2016. Sculptural deposition and lime kilns at Roman villas in Italy and the western provinces in Late Antiquity. In Kristensen and Stirling 2016a, 47–67. O’Connell, E. 2015. Living with the monumental past. In C. Fluck, G. Helmecke and E. R. O’Connell (eds), Egypt: Faith after the pharaohs. London, 92–7. Peppard, M. 2015. Was the presence of Christ in statues? The challenge of divine media for a Jewish Roman God. In L. M. Jefferson and R. M. Jensen (eds), The art of empire: Christian art in its imperial context. Minneapolis, 225–69. Riccardi, L. A. 2015. Homage and abuse: Three portraits of Roman women from the Athenian Agora. In K. F. Daly and L. A. Riccardi (eds), Cities called Athens. Studies honoring John McK. Camp II. Lewisburg, 321–50. Ridgway, B. S. 1990. Hellenistic sculpture I. The styles of ca. 331–200 BC. Bristol. Riggs, C. 2015. Egypt. In E. Friedland, M. G. Sobocinski and E. K. Gazda (eds), The Oxford handbook of Roman sculpture. Oxford, 552–68. Sauer, E. 2003. The archaeology of religious hatred in the Roman and early medieval world. Stroud.

279

Sidebotham, S. E. 2011. Berenike and the ancient Mediterranean spice route. Berkeley, CA. Sidebotham, S. E., Wendrich, W. Z. and Hense, A. M. 1996. Statuary and cult objects. In S. E. Sidebotham (ed.), Berenike 95. Preliminary report of the 1995 excavations at Berenike (Egyptian Red Sea Coast) and the survey of the Eastern Desert. Leiden, 229–43. Smith, R. R. R. 2016. Aphrodisias. In Smith and WardPerkins 2016, 145–59. Smith, R. R. R. and Ward-Perkins, B. (eds). 2016. The last statues of antiquity. Oxford. Stewart, P. 2003. Statues in Roman society. Representation and response. Oxford. Stirling, L. M. 2005. The learned collector. Mythological statuettes and Classical taste in late antique Gaul. Ann Arbor. Thomas, T. (ed.) 2016. Designing identity. The power of textiles in Late Antiquity. Princeton, NJ. Ward-Perkins, B. 2016. The end of the statue habit, AD 284–620. In Smith and Ward-Perkins 2016, 295–308. Witschel, C. 2015. Late antique sculpture. In E. Friedland, M. G. Sobocinski and E. K. Gazda (eds), The Oxford handbook of Roman sculpture. Oxford, 323–39. Wong, J. Y. 2016. Raze of glory: Interpreting iconoclasm at Edfu and Dendera. Journal of Late Antiquity 9 (1), 89–31.

KILLING OR ‘DE-ACTIVATING’ EGYPTIAN STATUES: WHO MUTILATED THEM, WHEN, AND WHY? Simon CONNOR1

Abstract This contribution to The British Museum Annual Egyptological colloquium 2016 proposes to approach the final step of the ‘life’ of Egyptian statues: their ‘death’. As demonstrated by several scholars and other papers in this volume, Egyptian images can be considered as powerful, meaningful, active agents. One of the best proofs of their importance in ancient Egyptian society is the very fact that they so often show signs of intentional mutilation, in specific spots on the figures. This article aims to review the different factors which may be at the origin of the state of damage of Egyptian statues, as we can see them today in museum collections or in archaeological sites. * * * Destroying images is a strong act. In all cultures producing statues, and in all countries where monuments are part of the landscape, mutilating them consists of a meaningful action, with the intention to shock, to put an end to a cultic activity and to symbolically erase specific records of the past. We all have in our minds the destructions of antiquities in the Middle East that have occurred during the last twenty years. The defacing of images is probably as old as the beginning of their production, and is attested in Egypt throughout its whole history. The visitor who walks through the galleries of a museum will notice that most of the Egyptian statues surrounding him or her have been — more or less deeply — damaged, in such a way that finding an intact statue is, in fact, quite exceptional. This can be due to a vast variety of factors, which this article intends to review.

1

I thank The British Museum’s staff for their kind invitation to participate in this colloquium. This paper presents a first overview of research carried out in parallel to my work as a curator in the Museo Egizio, Turin, and which I will continue within the framework of a Andrew W. Mellon Fellowship at The Metropolitan Museum of Art (2017–18). The preliminary research and

Statues must be considered as both artworks and archaeological artefacts, since the traces that they bear of manufacturing, of use, of modification and of damage, as well as the architectural contexts in which they were found, tell us a lot about the treatment to which they were subjected and about the different phases of history through which they passed. Careful observation of their physical surfaces allows a great deal of information to be gathered. As is well-attested and well illustrated by the different contributions in this colloquium volume, statues were certainly considered as powerful and meaningful objects through all Egyptian history. During the pharaonic period, they were conceived as receptacles for the ba, were active as substitutes of the represented individuals or entities, and were the subjects of rituals. During the Christian and Islamic periods, they were thought of as relics of ancient paganism, and as such, were either damaged — as in the case of the Serapeum in Alexandria, which was burned to the ground with all its statues in AD 391/392— or later regarded as guardians of a sort, as attested by numerous 12th- to 15th-century Arab authors (see point 2.2 below). So, even if today they are mainly regarded as precious works of art with a high historical and material value, traded on the art market and constituting major attractions in museums, in ancient times they were first and foremost active agents. 1. Accidental vs. intentional Various sources may be at the origin of a statue’s fragmentary state. Causes may, of course, be accidental. The course of time as well as natural factors such

documentation necessary for the lecture presented at The British Museum were rendered possible thanks to a post-doctoral grant from the Institut français d’archéologie orientale. My thanks also go to Klara Dietze, Nicola Barbagli, Marsha Hill and Niv Allon for their helpful comments and corrections.

282

S. CONNOR

as earthquakes can cause statues to fall down from their bases, and masonry blocks falling from the ceiling or the upper part of walls can hit statues and damage them. This is perhaps what has happened, for example, to the statue of Ramesses II in Turin, for no signs of deliberate smashing are visible.2 The statue was found fragmented into pieces, apparently in the ruins of the Temple of Amun-who-hears-the-prayers in Karnak by Jean-Jacques Rifaud (Rifaud 1830, 348 [9]; PM II, 214). The numerous fissures still visible recall that stage of fragmentation, but no element seems to be missing — except the extremity of the uraeus’ head, one of the most fragile parts of the statue — and the surface of the statue displays no traces of deliberate smashing. Its original position in the area of the eastern temple cannot be reconstructed so far, but one may assume that it fell from its base and broke on hitting the ground. Accidental breakages of statues must have occurred quite commonly during the pharaonic period. This is attested by the frequent restorations visible on statues, for example on the sides of the throne of the Dynasty 12 colossus reused by Ramesses II (Berlin ÄM 7264, currently in New York, MMA L.2011.42: Oppenheim et al. 2015, 300–4, cat. 221) or the western seated colossus in front of the pylon of Luxor (Fig. 1). A factor may be transportation, as perhaps in the case of the statue of Ramesses II, protected by the falcon god Herun, found in Tanis in a mud-brick chapel (Cairo JE 64735: Montet 1935–7, 11–4, pls 10–1; Simpson 1982, 267). The statue is carved from a block of granodiorite, while the falcon’s face consists of a separate piece made of dark limestone, carefully cut in order to be inserted in the god’s figure. Such a dark limestone seems to have been chosen in order to fit with the granodiorite body and to be as discreet as possible; it is most probably a repair, rather than an intentional composite sculpture. The statue may have been damaged at some point in its history, perhaps during its transportation to Tanis (from Piramesse?). Intentional damage is unlikely, since the rest of the statue is perfectly preserved.

2

Turin, Cat. 1380 (Connor 2017). In the first inventory of the Drovetti collection, written for the transport of the objects from Egypt to Livorno in 1819, the statue is described in these terms: ‘Re guerriero, con un elmetto, di grandezza naturale, che esige delle ristorazioni per riunire i pezzi staccati.’ The statue was only reassembled a few years later at Champollion’s instigation, as attested in a letter that he wrote in 1824: ‘J’ai enfin obtenu

Fig. 1: Western colossus of the pylon gate of Luxor Temple: detail of the throne repaired in antiquity. Photograph: author.

Natural catastrophes such as earthquakes may have caused the destruction of many monuments, like for example the statues of Amenhotep III’s Temple of Millions of Years in Kôm el-Hettan. According to the results of the recent research led by the team of H. Sourouzian, the site suffered from at least one important earthquake between 1200 and 900 BC, which destroyed the architectural structures and perhaps caused the destruction of many colossi, whose size naturally renders them more vulnerable (Sourouzian et al. 2011, 273–327; Karakhanyan, Avagyan and Stadelman 2014, 61–90). This is most probably the case with the two standing colossi of the North Gate, which were recently re-erected close to the site, for no evident sign of intentional mutilation is visible: the noses, eyes and beards suffered from no damage other than erosion. These colossi probably cracked and collapsed owing to their own weight and to the instability of the ground. The same phenomenon may have caused the breaking of the quartzite standing colossi of the solar court of the same temple. The Colossi of Memnon themselves suffered from an earthquake, and were repaired apparently by Septimius Severus at the beginning of the 3rd century AD (Letronne 1833, 40–56; Fournet 1996, 145).

qu’on assemblât les morceaux de la statue de Sésostris [= Ramesses II], dont je parle dans ma première Lettre. Il n’y manque rien, et quand je considère la beauté et l’admirable perfection de cette figure colossale, je regrette de n’en avoir pas assez dit dans ma Lettre, en faveur de l’art égyptien’ (Hartleben 1909, 104).

KILLING OR ‘DE-ACTIVATING’ EGYPTIAN STATUES: WHO MUTILATED THEM, WHEN, AND WHY?

Both colossi always remained on view, and, probably for that reason, also bear marks of intentional breakage: their faces and arms, as well as the faces of the queens standing at either side of their legs, bear what seems to be traces of severe, deliberate smashing (Fig. 2). Distinguishing these two different situations — intentional and accidental damage — may be difficult and it will not always be possible to find out whether a defacement is ancient and intentional or not. In the case of the Sphinx of Giza, for example, is the break of the nose only due to natural erosion, the giant head of the statue suffering from wind and sand for 4,500 years, or was the nose deliberately cut to annihilate the power of this pagan image? M. Lehner has suggested an intentional mutilation, according to some tool marks visible at the level of the missing nostrils (Lehner 1991, 179– 80). The persistent legend of the cannon bullet of Bonaparte is, of course, unfounded. So how can we securely identify intentional and meaningful mutilations? Protruding parts, such as the uraeus, the nose, the beard, the hands, the sceptres or other attributes are by their nature the most fragile and susceptible to breakage. However, when all of these parts, which are also the most symbolic parts of the statue, bear traces of damage, while the rest of the surface is intact, one may seriously question the possibility of accidental breaks (see also the various interpretations offered for the Dangeil royal statues, which were broken at their thinnest and weakest structural points: Anderson et al. in this volume). Among many other examples, we may cite the granodiorite dyad of Horemheb and Mutnedjemet (Turin Cat. 1379; Fig. 3), which shows breaks on each arm, on the breast of the queen, on each of the eyes, the ears (and earrings), on the nose, mouth, chin and on the two uraei, while the head of the king is completely missing and his beard apparently hammered out. The indurated limestone dyad of Amun and Mut in Luxor Temple (Fig. 4), from the same period, shows severe damage on the legs; the arms are almost completely missing; the mouth, eyes and ears of both deities seem to have received repeated blows, as well as the uraeus of the goddess. The god’s beard is entirely missing and its break follows the outlines of the chest and neck, while the surrounding surface has remained intact, which renders an accident improbable. The same situation can be

3

Two of them, from which remain only the torso and kilt, are still in situ (Postel 2014, 118, fig. 5). One is in Cairo, Egyptian Museum, TR 18.4.22.4 (Evers 1929, pl. 83); the three others are

283

Fig. 2: Southern Colossus of Memnon, detail of the queen. Kôm el-Hettan. Quartzite. Photograph: author.

observed at the nose of Amun, while Mut’s nose, now missing, was originally a separate piece inserted in the face, probably a repair of an old break. Another postAmarna indurated limestone statue, showing a king embraced by Horus, Osiris and Isis (Cairo JE 49536; Fig. 5), displays breaks in various spots that do not seem to be accidental: all faces have been literally erased; only Horus still has some parts of his head, but his beak and eyes were the targets of several hits; the same fate befell the chests of the goddess and the sovereign, and the beard and sceptres of Osiris, while the legs and front base have been hacked out, apparently with some care. Comparison of a large number of pieces, and more particularly of statues belonging to coherent series and archaeological contexts, allows us to highlight the parts of the statue that are damaged in such a systematic manner that no coincidence is conceivable. The series of Senwosret III’s ‘praying’ statues from Deir el-Bahri illustrates this phenomenon perfectly.3 All six statues were found in the depression located to the southwest of the temple of Mentuhotep II. They were probably thrown down from the terrace and may have partially broken due to the shock of the fall, but they were first cut and smashed in specific zones of the body. The

in London, British Museum, EA 684, 685, and 686 (Wildung 1984, 202, fig. 176; Polz 1995, 235, pl. 48a; Strudwick 2006, 90–1).

284

S. CONNOR

Fig. 3: Dyad of Horemheb and Mutnedjemet. From Karnak. Granodiorite. H. 139cm. Turin, Cat. 1379. Photographs: Pino and Nicola Dell’Aquila. © Museo Egizio.

Fig. 4: Dyad of the gods Amun and Mut. Luxor Temple. Indurated limestone. Photograph: author.

arms, which are emphasised because of pose (and action) of the king, were systematically hammered, as well as the cobra uraeus, which was thought to magically protect the king, and the nose, which allowed the statue to breathe. In two cases, even an eye was smashed. The series of quartzite colossi from Herakleopolis Magna, usurped from late Dynasty 12 sovereigns for Ramesses II (Connor 2015), also had their faces smashed at some point in history. The nose has been completely removed, as well as the uraeus and the beard, which has been erased rather carefully on the colossus Cairo JE 45975. Even the queens on either side of the legs had their heads completely removed. All these statues show damage on precisely the same zones of the body, which are not necessarily the most fragile and thus can hardly have suffered from accidental breaks. Therefore, these parts can be considered symbolically meaningful, since they are those which apparently needed to be erased when the statues had to be ‘de-activated’. Breakages to these parts would seem thus to demonstrate a coherent practice of ritual damaging.

KILLING OR ‘DE-ACTIVATING’ EGYPTIAN STATUES: WHO MUTILATED THEM, WHEN, AND WHY?

285

Since mutilation of statues is indeed an attested practice in the ancient world, let us try to understand the motives that led ancient inhabitants of the Nile Valley to deface the images present in their surroundings. Let us take into consideration different possible sources and reasons for ancient intentional breaking or mutilation of statues: iconoclasm during the Christian and Islamic periods; looting; official proscription and personal animosity; magic/ritual mutilation; or war damage. 2. Mutilating statues 2.1 Christian iconoclasm One of the most frequently mentioned reasons to explain the state of fragmentation of statues and reliefs is the iconoclasm carried out against pagan objects and monuments during Late Antiquity (Kristensen in this volume) and the Islamic era. At least in their earlier Fig. 5: Statuary group of a post-Amarna king, with Horus, Osiris and Isis. stages, the revealed monotheistic reliFrom Abydos. Indurated limestone. H. 172cm. Cairo, JE 49536. Photographs: author. gions prohibited the production of any likeness of a living being, and this is, in many cases, the main reason for the defacement of preand 5th centuries, these attitudes of violence and Christian images. However, although the practice of destruction towards ancient images in the Christian damaging pagan monuments is well-attested, particuperiod do not necessarily reflect everyday behaviour, larly during Late Antiquity, one should be careful and probably have more to do with occasional events, before attributing the state of fragmentation of initiated by some radical groups. In many other situapharaonic monuments too quickly and systematically tions, it seems that the people of Late Antiquity preto anti-pagan iconoclasm. served testimonies of the past, or simply forgot them Christian iconoclasm is well documented by a large and let them disappear under dust and sand (Hannestad number of texts of early Christian authors of the 4th 1999; Kristensen 2010). Different responses towards and 5th centuries. They describe the violent conflicts pagan monuments are, therefore, attested. which marked the rise of Christianity in Alexandria — A possible way of ‘exorcising’ or ‘neutralising’ the best documented being the destruction of the Serapagan monuments may have been the incision of peum in AD 391/392 (Kristensen 2010). The accounts a cross or a Christian star, of which we find innumerable examples on tombs and temple walls. On statuary, of Theodoret, the bishop of Cyrrhus (Historia Ecclesihowever, it has remained relatively exceptional. Only astica 5.22: Schaff 1892, 148–9) and Rufinus (Historia a few examples can be cited, and, as far as I know, all Ecclesiastica 11.23: Amidon 1997, 81–2) relate how of them are statues from the Roman period: a marble the chryselephantine colossal statue of Serapis was statue of Marcus Aurelius from Alexandria, which has publicly smashed, dismembered and burned, in order to a large cross carved on the abdomen (Alexandria, show the pagans that this ‘idol’ contained no power. Graeco-Roman Museum, inv. 3250: Tkaczow 1993, The head was then exposed to the mockery of the peo248–9, no. 169; Kristensen 2012, cat. B10; Kristensen ple of Alexandria. Although such phenomena are 2013, 126–7, fig. 2.7); a head attributed to Germanicus, attested throughout the whole Roman Empire in the 4th

286

S. CONNOR

without known provenance but carved in an Egyptian stone, with a cross on the forehead (London, British Museum 1872,0605.1: Kristensen 2012, cat. A10; Kristensen 2013, 93–4, fig. 1.17; Fluck, Helmecke and O’Connell 2015, 98, cat. 102); and a few other examples (Dölger 1930, 280–4; Langmann 1985; Hjort 1993; Kristensen 2012, 98, fig. 1.20). Incising this cross may have been a ‘soft’ alternative to complete destruction, although these statues bear other mutilations, which could be — but are not necessarily — contemporary with the cross incision. The absence of archaeological context prevents us from dating these actions. A more radical reaction would be the transformation of a statue into a cross: literary sources inform us that the torso from the wooden cult statue of Kronos in Alexandria was re-carved in the shape of a cross in AD 324, when the temple was rededicated as the church of Theonas (Haas 1997, 209–10). 2.2 Islamic iconoclasm The deactivation or ‘killing’ of pagan images is wellattested in the 4th and 5th centuries AD, but Islamic iconoclasm is more difficult to point out. In some cases, archaeological remains, ancient photographs or drawings show the level of the ground during the last centuries. In the case of Luxor Temple, for example, the Description de l’Égypte (vol. III) and the illustrations of David Roberts show that the heads of the two seated colossi at the entry of the temple were easily accessible from the ground, which may have facilitated their defacing. However, dating these mutilations to the Islamic era according to that argument is barely more than a hypothesis, as on the other bank of the Nile, the south colossus of Memnon shows the same mutilation, though the ground level there never reached the upper body. It remains, in fact, difficult to date such mutilations. The information provided by textual sources of the Islamic period illustrates diverse reactions towards pagan monuments, especially in the 13th and 14th centuries (Reid 2002, 30). Some describe, for example, an intentional damaging of the Sphinx in the 14th century,4 while other texts from the same period tell that

4

The author Ahmed al-Maqrizi (1364–1442), in his work alMawa‘iz, reports the action of an activist Sufi called Mohamed Sa‘id al-Su‘ada, who went in 1378 to the pyramids plateau in order to disfigure the Sphinx (Haarman 1980; McGregor 2013, 184).

the Sphinx was considered a monumental talisman, protecting the cultivated area from the sands.5 The pyramids were partially dismantled to build the walls of medieval Cairo, but they were also part of the Islamic cultural landscape (Smith 2007). In fact, a large number of medieval Islamic authors praise the fact that their rulers never destroyed these monuments. In contrast to the Christian period, the Egyptian population of the Islamic period was no longer in direct contact with pagan cult, and the pharaonic monuments, which were often already dismantled and buried under the sand or alluviums, could no longer be seen as dangerous records of idolatrous practices, but more as testimonies of the accounts written in the books of revelation.6 There would thus have been no need for ostentatious mutilation or systematic destruction of pagan monuments. In this period, therefore, mutilation or damage carried out on pharaonic images may be in many cases simply due to their reuse as ‘spolia’ in medieval monuments, instead of meaningful and ritual deactivations of statues. The greywacke sphinx of Senwosret II (Cairo JE 37796 – TR 16.2.21.6: Sourouzian 1996) is one among many examples. This statue, which comes from Heliopolis according to its inscriptions, was found in the early 20th century in the masonry of Mottaher mosque in Old Cairo. Lacking the head and the front paws, the body of the lion formed a quite convenient parallelepiped block to be reused in an architectural structure. Other examples of statues recut to be reused in Islamic constructions are the quartzite sphinxes inscribed for Amenemhat V (Sekhemkare) and Ramesses II, also from Heliopolis as indicated by their inscriptions, which were found reused in a postern of the medieval walls of Cairo; today they are on display in a modern reconstruction of the postern on a square, just north of Bab el-Nasr (Fig. 6). The practice of ‘spolia’ is attested at all periods and examples are very numerous. Such reuses were already frequent in pharaonic times. The excavations of the pavement and foundations of the Ptolemaic temple of Medamud, for example, brought to light statues and architectural fragments of Senwosret III, several Dynasty 13 rulers and Sobekemsaf I. Much earlier, the architects of

5

6

See Ahmed al-Maqrizi (cf. supra) and Abdelrahman al-Suyuti (1445–1550), in his work Husn al-mubadara (McGregor 2013, 178). See the author Abd el-Latif al-Baghdadi (1162–1231), The Eastern Key (McGregor 2013, 179).

KILLING OR ‘DE-ACTIVATING’ EGYPTIAN STATUES: WHO MUTILATED THEM, WHEN, AND WHY?

287

Fig. 6: Reconstructed postern of the northern wall of medieval Cairo, near Bab el-Nasr, with reused sphinxes of Amenemhat V and Ramesses II. Quartzite. L. of each sphinx: 157cm. Photographs: author.

Merenptah’s Temple of Millions of Years employed fragmented gigantic sphinxes and jackal statues. Just as in the case of the Christian period, as Kristensen has pointed out, different reactions may have occurred contemporaneously. The existence of some cases of iconoclasm does not mean that this reaction towards pharaonic antiquities was common among the population. The large number of statues or reliefs which have always remained visible, yet have not been defaced, speaks in favour of a certain tolerance of these ancient images — or an ignorance of their significance. 2.3 Looting One reason for the destruction of some statues seems to have been the plundering of burial chambers. Perhaps in order to avoid the curse of the tomb owner, statues from the funerary chapel or the serdab were often found intentionally smashed, possibly by the robbers. The statues of Iteti, from Giza (Fig. 7, Turin S. 1876: Connor 2016, 40, 86), and of King Djoser, from

Fig. 7: Head of the statue of Iteti. From Giza. Limestone. H. 117cm (whole statue). Turin, Suppl. 1876. Photograph: Pino Dell’Aquila. © Museo Egizio.

288

S. CONNOR

his pyramid complex in Saqqara (Cairo JE 49158: Sourouzian 1995, 149–52) are probably both victims of such vandalism. In the case of Iteti, the head has been cut off; the torso and feet were not found during the excavations of Schiaparelli. The faces of both statues show evident traces of intentional breakage, produced by a series of blows, made with a hammer and chisel: the nose has been mutilated, as well as the area of the eyes, and, in the case of Djoser, the beard has been partially cut off. It is difficult to know whether a symbolic or a material reason caused such destruction (the two are not incompatible). A practical motive may be the process of extraction of the precious materials used for the inlaid eyes (metal and rock crystal) or for features ornamenting the statue. That might have been the case of the statue of the commander-in-chief Hor, son of Tithoes (Berlin ÄM 2271, 1st century BC–1st century AD), which was once ornamented with insignia and necklaces, later removed and erased (Lembke and Vittmann 1999).7 Nevertheless, the damage inflicted on the noses and the beard, and the beheading of Iteti’s statue, speak in favour of a proper intention to render them unable to act. Concerning the beheading, it must be kept in mind that when dealing with the head of a statue or a headless sculpture, it is often difficult to identify the reason for its decapitation. As we will see later, as the head is considered the most important part of the body, decapitation is a commonly attested practice in ancient societies as well as in art, as a radical and symbolic way of killing and depriving the individual of his identity. However, statues may have also been damaged, mainly beheaded, in modern times (i.e. in the last centuries), by antiquities robbers. Indeed, a head is much easier to transport, to hide and to sell than a complete statue, and remains a very valuable piece. This might partly explain why sculpture collections include so many heads — and, consequently, so many headless bodies. As we will see, the nose and the beard are the most frequently damaged zones on royal and divine statues. Concerning the eyes and nose, we may propose that the reason was to make the owner of the tomb blind and unable to breathe, while the robbers were plundering

7

I thank Nicola Barbagli for bringing this statue to my attention. The man’s face, otherwise well preserved, has also been heavily disfigured by what seems to be a double knock on the nose.

the tomb. Such looting is hard to date, but it might have occurred early in the pharaonic period, before the site was covered with sand. In the case of a looting situation, the damage suffered by a statue would therefore be intentional, but not necessarily meaningful — or at least not only — and a purely material motivation might be at the origin of a defacement: the removal of precious parts. 2.4 Official proscription and personal animosity A well-attested reason for damaging statues is the damnatio memoriae, or proscription. As pointed out by T. M. Kristensen, in some cases, this type of damage is obvious, ‘reminding viewers that they should “remember to forget” the disgraced’ figure or ‘at least maintain a very particular kind of memory of the individuals subjected to this punishment’ (Kristensen 2015, 670). For political, religious or personal reasons, some individuals had their names erased from the monuments, and sometimes their representations destroyed, in order to cancel their memory, and, in the case of Egyptian imagery, to magically incapacitate the representations of these persons and to render them inactive. Proscription or damnatios memoriae have been studied by several authors, especially in the cases of royal proscription. A well-known case is that of Hatshepsut (Fig. 8), who suffered a systematic proscription at the

Fig. 8: Granite kneeling statue of Hatshepsut (front) and limestone heads of jubilee pillars (back). From Deir el-Bahri. New York, The Metropolitan Museum of Art, 29.3.1. H. 261.5cm. Photograph: author.

KILLING OR ‘DE-ACTIVATING’ EGYPTIAN STATUES: WHO MUTILATED THEM, WHEN, AND WHY?

end of the reign of Thutmose III (Dorman 2005; Roth 2005, 280–1; Bryan 2012, 365–9). At Deir el-Bahri, the statues of the queen were removed from the temple, systematically altered and thrown into two large holes in front of the temple. The damage was clearly intentional and concentrated in specific areas, with stone hammers used to inflict it: the statues were usually beheaded; the hands and feet were often hacked, and, for the standing statues, the base separated from the body; the face was usually less damaged, apart from the uraeus, an emblem of royalty and divine protection that was systematically erased (Arnold 2005). However, the inscriptions and names are still present on these statues. A different treatment was given to the reliefs, the figures remaining intact and her name being substituted by those of Thutmose I or II. Her statues were ritually mutilated and buried in front of her former monument, in order, as said by B. Bryan, to ‘incapacitate her spirit and thereby prevent any potentially hostile activity’ (Bryan 2012, 369). One century later, during the phase of the ‘Amarnaiconoclasm’, another large-scale proscription was carried out by royal order against the traditional gods, in particular Amun. Most probably mainly for political reasons, Akhenaten ordered the name of Amun to be erased (even within the cartouches containing the name of his father Amenhotep III) and his statues destroyed throughout the whole country. This iconoclastic campaign, and the restorations which followed at the end of Dynasty 18 and during Dynasty 19, are visible on statuary. Particularly representative cases are Cairo CG 42052 (Lindblad 1984, 51, cat. 4), CG 42066 (Laboury 1998, 224–6, cat. C 66), CG 42065 (Sourouzian 1991, 69–70, fig. 23) and Turin Cat. 767 (Connor 2017, 50–2), showing respectively Thutmose I, Thutmose III, Amenhotep II and perhaps Amenhotep III with the god Amun, and, in the case of the Turin statue, with Amun and Mut. On each of these groups, the figure of Amun was hammered, while the king remained intact. When restored in the post-Amarna or Ramesside periods, the Amun figure was either cut, in a reduced size, in what remained of the damaged piece, or completed with additional sculpted fragments to fill the gaps and replace the too-damaged parts of the body. After his reign, Akhenaten himself suffered from a damnatio memoriae: his name was erased everywhere, his monuments were dismantled, their blocks were reused as construction material and the statues of the royal family smashed (Fig. 9) (e.g. Thompson 2011; Bryan 2012). Still associated with the Amarna

289

sovereign, Tutankhamun and Ay, too, were subject to a quite systematic usurpation during the reign of Horemheb. The statue in the Louvre, inv. E 11609 (Fig. 10) (Barbotin 2007, I, 130–2, cat. 73; II, 200–5), which represents the god Amun protecting the child king, shows, among other mutilations, an erasure of some signs of his cartouches, leaving only the god’s names, in order to render Tutankhamun unidentifiable: [twt-῾nḫ]-Ἰmn [nb-ḫpr.w]-R῾. It is, however, difficult to date the other damage done to the piece: both figures were beheaded (the head of the god has been re-fixed), and the hands and feet were smashed, as well perhaps as the god’s beard. Are these mutilations contemporary with the hammering of the cartouches? This might justify why the statue was ‘de-named’ instead of being usurped, but remains difficult to explain, when reusing the statue would have been quite easy. Proscription also applies to non-royal individuals. Reliefs, wall paintings, as well as statues in Memphite mastabas (see numerous examples cited in Kanawati 2003) and Theban tombs contain an innumerable corpus of mutilations. In many cases, defacement affects most of the human figures, and could be due to various factors; but in others, only some figures, clearly identified by their inscriptions or their iconography, are defaced. A well-known case is that of the paintings of Rekhmire (TT 100); the figures of the deceased and of his son, even in the scenes of the rituals for the mummy, are the object of a meticulous and systematic hammering, leading to the supposition that some kind of disgrace fell on the family, or that some personal hatred was expressed against its members, perhaps from political opponents, sufficiently powerful to lead to the disfigurement of their representations on the tomb’s walls. Such a destructive action must have taken place, at the latest, shortly after the disappearance of this family, when the remembrance was still sufficiently vivid to lead to the destruction of their images. In the current state of knowledge, it is difficult to ascertain if the destructive action came from a royal order or from a private initiative. As regards non-royal sculpture in the round, a probable example of proscription is that of a quartzite triad of early Dynasty 13 (Louvre A 47; cf. Fig. 11), which once represented three generations of high priests of Ptah, grandfather, father and son (Delange 1987, 81–3). The figure of the son, on the proper left of the block, has been carefully cut out. The rest of the statue is intact, except for the noses of the figures. In this case,

290

S. CONNOR

Fig. 9: Heads of statues of the Amarna royal family. New York, The Metropolitan Museum of Art, 2005.363 (partly restored), 57.180.79, 21.9.2, 21.9.487. Indurated limestone and quartzite. From el-Amarna. Photographs © The Metropolitan Museum of Art.

KILLING OR ‘DE-ACTIVATING’ EGYPTIAN STATUES: WHO MUTILATED THEM, WHEN, AND WHY?

291

it is not necessarily due to an intentional mutilation, since they were the most protruding parts of this sculpture. According to the inscription, the statue was ordered by the central figure, Nebpu, for himself, his father and his son. The name of the son is actually still visible, but his figure was carefully cut out, and the angle even re-carved, in order to transform the triad into a dyad. The characters are not known from any other source. Cases of political proscription and personal hatred seem therefore to differ: in the first case, the memory of the targeted individual needs to be eradicated (the names of Hatshepsut, Akhenaten, Tutankhamun, are erased everywhere, and their statues severely mutilated or, more rarely, reused); when personal hate (or disfavour?) occurs, the individual’s name is not erased, but only his figures are hacked out, so as to remind readers of the inscriptions of the discredit fallen on him. 2.5 Magic, ritual mutilation

Fig. 10: Detail of the statue of the god Amun protecting Tutankhamun in worshipping position. From Karnak or Luxor? Granodiorite. H. 220cm. Paris, Musée du Louvre, E 11609. Photograph: author.

In some cases, the defacing of statues seems to have been a ritual performed without the intention of destroying them, but as an act which properly allowed them to serve their particular functions. This would seem to be the case with the very peculiar group of the so-called ‘reserve heads’. These limestone heads do not consist of fragments of statues, but proper statues originally conceived as heads. Around thirty of them are known. All have been found in shafts or burial chambers of Dynasties 4 and 5 in the Memphite necropolis. They were apparently (although none was found in its original position) conceived as being part of the underground part of the tomb, therefore close to the body of the deceased. Almost all of these heads show evident traces of hacking on the ears. The necks and skulls, also, often have several incisions, which have been interpreted by R. Tefnin as perhaps the symbol of decapitation or trepanning (Tefnin 1991a; 1991b). Several theories have been formulated concerning the presence of these heads in some Old Kingdom burials. N. Picardo recently proposed that they could be seen as a kind of negation of the decapitation (Picardo 2007). In the same way as dangerous hieroglyphs can be represented cut or injured, in order to prevent any negative action, these decapitated heads would have been magically rendered inoffensive. It would constitute a kind of negation of the risk of being beheaded. The process of damaging these disembodied heads would thus be a kind of ‘execration

292

S. CONNOR

Fig. 11: Fragmentary triad of high priests of Ptah of late Dynasty 12 or early Dynasty 13. Unknown provenance. Quartzite. H. 92cm. Paris, Musée du Louvre, A 47. Photographs: author.

magic’, or what Picardo calls a ‘semantic homicide’, preventing the deceased from becoming headless, which is one of the fates most greatly feared during the post-mortem transition, according to numerous representations in later funerary texts. The reserve heads would have interceded magically and conceptually against the loss of the deceased’s head. It would thus probably be the result of a kind of ritual, which perhaps occurred during the funeral. Another, much later, case is that of the statue of the deified prince Ahmes-Sapair (Louvre E 15682: Fig. 12), which shows several traces of apparent ritual damage, completed by ancient Egyptians themselves (Vandersleyen 2005; Barbotin 2007, I, 32–4; II, 8–15). According to its style, the statue belongs to late Dynasty 17, and is therefore approximately contemporary to the prince himself, and may have been sculpted for his grave — although its size, 103.5cm high, is unusually tall for this period. Its inscriptions describe the sadness of the royal family confronted with the prince’s death. The cult of this prince continued in the Theban region during the whole New Kingdom, as attested by a large

series of stelae (Vandersleyen 2005, 23–6, 38–45). For some reason, at some point in its use, numerous blows were inflicted upon the statue to break the arms and legs of the prince, although neither his face nor his nose were broken. Furthermore, a few holes were cut in the throne and painted red. It has been proposed that the intention of this damage was to magically bind him to his seat, without cancelling his capacity to act as a deified figure (Vandersleyen 2005, 15–6). Another hole appears more difficult to explain: a hole cut into the back of the head, within the curls of the wig. Does it have to be considered as a kind of trepanation? Do we deal again with a way to render the statue and the represented prince harmless, but without killing him, since he was still supposed to act as a deity? Or, more simply, was it intended to help in fixing a metallic ornament onto the wig, as suggested by C. Barbotin (2005)? In these cases, which still depend greatly on our interpretation, we might not be dealing with a ‘killing’ or a ‘deactivation’ of these sculptures in the round, but properly with their ‘activation’ in a specific role that they were supposed to fulfil.

KILLING OR ‘DE-ACTIVATING’ EGYPTIAN STATUES: WHO MUTILATED THEM, WHEN, AND WHY?

293

Fig. 12: Statue of prince Ahmes-Sapair. Unknown provenance. Limestone. H. 103cm. Paris, Musée du Louvre, E 15682. Photographs: author.

2.6 War Another possible reason for the mutilation of statues might be war damage. It has been suggested in the past that the innumerable statues found in the Cachette of Karnak were buried owing to breakages caused by Assyrian or Persian incursions; nevertheless, recent research has shown that the Cachette dates from a

much later time, at the end of the Ptolemaic period, i.e. several centuries after any foreign invasion in Thebes. Therefore this thesis is considered less and less probable today (see Jambon 2016, 154 with bibliography; Coulon et al. in this volume). However, the royal statues found in the Cachette of Dokki Gel may have been buried, according to C. Bonnet, as a consequence of the military expedition of Psamtek II in Nubia. If so, it

294

S. CONNOR

must be noted that their dismemberment was quite carefully made, by an experienced sculptor, seemingly with an intention to cause minimum damage to the sculptures (Bonnet 2011), which would suggest that this (ritual?) mutilation occurred not during a violent raid, but during a planned and codified procedure, perhaps even a ceremony. J. Anderson compares the Dokki Gel discovery with similar finds from Gebel Barkal and Dangeil, and she proposes a temple refurbishment as an alternative explanation to the burial of the Kushite kings’ statues (Anderson et al. in this volume). The case of the sanctuary of Heqaib in Elephantine is perhaps clearer. This small temple shows signs of violent destruction before being buried and covered with successive levels of occupation, which sealed its ruins and its mutilated statues (Habachi 1985, 159). Its destruction can be dated to the Second Intermediate Period, around 1700–1600 BC, i.e. at a time when the enemy in the south, the kingdom of Kerma, was particularly powerful and aggressive. Contemporarily, in Sudan, the kings of Kerma were buried under large tumuli, together with sacrificed human bodies, bucrania and fragments of Egyptian statues, most of them now part of the collections of the Museum of Fine Arts, Boston. Several theories have been proposed to explain the presence of these fragments of Egyptian statues in these Sudanese royal tombs. Only one of them was found integrally preserved, the statue of Sennuy (Boston 14.720: Reisner 1923, 34–5, cat. 32). The other ones are not only damaged, but in most cases they consist solely of fragments, which are impossible to join together. It is noteworthy that on the face fragments found in Sudan, the nose is often present, and there are no traces of smashing on the arms, nor on the chin. We seem to be dealing with a different process of damaging, therefore, than in the cases found in Egyptian contexts. To explain the presence of these hundreds of

isolated statue fragments, it may be suggested that these pieces came to Kerma already in that state. Indeed, military incursions of Kerma armies into Egypt are well-attested by texts during the Second Intermediate Period (see for example the texts in the tomb of the governor Sobeknakht II in Elkab: Davies 2003). Might the Kerma soldiers have attacked the sanctuary and its sculptures during a raid, bringing back fragments of statues as a kind of trophy to Sudan, where they would accompany the king to his grave, as symbols of victory over Egypt?8 This remains in the realm of conjecture, and would need further evidence, such as a join between material found in Kerma and that still in place in Egypt.9 In front of their destroyed temple and its broken statues, the inhabitants of Elephantine might then have chosen, instead of repairing the site, to bury it and, before that, ritually render inactive what remained of its sculptural material, which would explain why the statues found inside by Habachi, even those of which the rest of the body is still well preserved, almost systematically lack their noses. Only the statue of ImenyIatu (Habachi 1985, 64, pl. 103–10, cat. 37) is intact, as well as (except for a knock to the chin beard), the statue of Khema (Habachi 1985, 43–4, pls 39–45, cat. 15).

8

9

O’Connor 1974, 30–1; Wenig 1978; Bonnet 1997; Davies 2003; Valbelle 2004; Davies 2004, 101, no. 75; Davies 2005, 50, 55, nn. 18–9; Minor 2012. Another case in Nubian territory of probable symbolic victory over Egypt or even the Roman Empire is the head of Augustus found in Meroe (London, BM 1911,0901.1), which was found in 1910 by Garstang’s team buried in the ground at the entry of a temple or shrine. That structure, decorated with scenes of Meroitic military triumph, might be interpreted as a victory shrine (Opper 2014, 20–9). When entering it, people necessarily had to walk on the ground containing that head, which it is very tempting to see as a symbolic treading down.

2.7 The case of the Cachettes or favissae: deactivation of statues? Finally, mutilations might perhaps be linked, at least in certain cases, to some still unknown ritual in the case of temples’ ‘Cachettes’ or favissae, pits containing statues, which were found in numerous temples throughout Egypt, under the pavement of a courtyard. Sculptures were sometimes found reduced to small fragments, as in the case of the two statues of Amenhotep III found

Such as a joint between a headless statue found in the sanctuary of Heqaib in Elephantine (Habachi 1985, 44, cat. 16, pls 46–8), and a head discovered at Kerma, now in Boston MFA 20.1207 (Reisner 1923, 40, cat. 66). Thanks are due to the staff of the Egyptian Department of Boston Museum, who kindly provided me with a cast of the break of the head. I am currently in the process of asking for access to the lower part of the statue, in the storage of Elephantine Island, in order to check that joint. Should it be confirmed, this connection would bring a decisive light on the membra dispersa found in Nubian contexts, and a new interpretation of the documents taken away as fragments from Egypt to Kush, by Kerma military expeditions, during the Second Intermediate Period.

KILLING OR ‘DE-ACTIVATING’ EGYPTIAN STATUES: WHO MUTILATED THEM, WHEN, AND WHY?

295

in two circular pits under the pavement of Montu’s temple in Karnak (Barguet and Leclant 1954, 46–7, figs 78–9; Valbelle 2016, 21–3). In most cases, statues have been found in a fragmentary state, and the recurrence of damaged parts of the body, as well as the way the fragments are displayed inside the pit, speak in favour of intentional mutilations (Jambon 2016, 148–50).10 It remains difficult to date this damage precisely and to ascertain that it specifically occurred before the burial of the statues. Were they buried because they were damaged, or ritually mutilated because they had to be ‘deactivated’ before being buried in the sacred pit? That question remains difficult to answer with certainty, notably because, although most of the statues found in cachettes were found in a fragmentary state, many were also intact (see, for example, the statues from the Cachette of the Luxor Temple: el-Saghir 1991); it remains clear, however, that the pieces were not thrown inside the pit, but were carefully buried and associated in coherFig. 13: ‘Ozymandias colossus’. Ramesseum. Granite. ent groups, to ‘sleep’ inside the sacred Photographs: author. area (Goyon and Cardin 2004, 19–20; Jambon 2016; see also Charloux and Mahmoud in this volume about the statue from acting, by depriving it of its feet (capaca recently discovered and carefully excavated cache in ity to walk), arms (capacity to act) or head (capacity to Karnak). Further research is needed concerning these hear, to breathe, to see, to speak). favissae, taking into consideration their differences of time and context, which suggest that there may have 2.8 Multiple cases been many different reasons for their digging and for the breaking/cutting/mutilation of some of the statues Often, a single monument may attest different phases found inside. of damage or reuse. In the Ramesseum, the enormous If statues were mutilated precisely just before being colossus that inspired Shelley’s ‘Ozymandias’ poem buried in a sacred pit — which still needs confirmation, and that once sat in the first court, may have initially since it would in any case have been far from systemcracked and collapsed accidentally owing to an earthatic — the idea would, surely, not have been to ritually quake, but it also bears traces of intentional cuts, as one ‘kill’ the represented person or the memory of him or can see on the large scar on the face, witness of an her, since the inscriptions generally remained intact, unfinished sawing action, probably in order to produce but maybe more to ‘deactivate’ the object, to prevent blocks for masonry or for millstones (Fig. 13). Before

10

Some statues cited by E. Jambon (2016: Cairo CG 42027, 42059, 42148, 42149) show traces of cutting, which cannot be accidental.

296

S. CONNOR

Figs 14–15: Standard-bearer statue of Sety II. Turin, Cat. 1383. Sandstone. H. 516cm. Photographs: Pino Dell’Aquila © Museo Egizio.

this cutting action for the purpose of reuse, the statue apparently suffered a ritual mutilation: the uraeus was not just damaged, but also carefully cut off, and the whole tail was completely erased; this cannot be by accident. The snake-goddess who was thought to magically protect the king is indeed one of the parts that needed to be systematically hammered when a statue was supposed to be deactivated. The jubilee statuepillars in the courtyard also bear traces of systematic

mutilation: the colossal faces show clear tool marks on the eyes, nose, mouth and chin. Several phases of modification and disfiguring may be the reflection of different meanings and practices through Egyptian history. The Turin statue of Sety II (cat. 1383, Figs 14–15) shows several traces of damaging, which attest several successive responses towards ancient monuments. Before being transported to Livorno and then to the Piedmont’s capital in 1824, this

KILLING OR ‘DE-ACTIVATING’ EGYPTIAN STATUES: WHO MUTILATED THEM, WHEN, AND WHY?





colossal sculpture was rising, together with its twin (Louvre inv. A 24: Barbotin 2007, I, 100–2; II, 142–5, cat. 48) at either side of the axial entrance of the barque chapel that Seti II erected around 1200 BC in front of the Amun temple. The bases that once supported them are still in situ and allow reconstructing the original position of both colossi. Several marks allow following a series of episodes through the long ‘life’ of these statues:

11

Until then, although a dangerous god of disorder, Seth was also considered a necessary deity, notably in the fight against Apophis during the nocturnal journey of the sun, and as such received a cult, which seems to have developed especially in the Ramesside period. Personal names including the name of Seth disappear after Dynasty 20, as well as any restoration or building of a temple dedicated to this god. The growth of the personal cult of Osiris during the 1st millennium, as well as the

297

The edges of the base are severely blunted and even form small cavities, which are clearly due to a repetitive action of scratching. These concavities, called ‘cupules’, are usually interpreted as the result of a ritual activity that consisted in scratching the stone, in order to take off some powder that those who could access the sacred area were probably conserving as kind of magic souvenir or amulet. Such activity is attested at least as early as the New Kingdom, as attested by the columns of the Temple of Amun-who-hears-the-prayers, carved during the reign of Thutmose III, which show several of these cupules. They were later covered with a layer of plaster during the reign of Ramesses II, on the occasion of the enlargement of the oratory temple (Gallet 2013, 5–6, figs 5–6). On the statues of Sety II, no elements allow us to date these cupules, since the practice of removing ‘temple powder’ seems to have continued long after the pharaonic period, as is suggested by the position of such cupules, very high on the walls of certain temples, when the level of the ground in the courtyards or the forecourts had already risen several metres owing to the accumulation of layers. On the lateral sides of the base and on the back pillar of the Turin statue, the ‘Seth’ hieroglyph was erased inside every cartouche of the king (sth.y mr(y).n ptḥ). The hammering concerns only the god’s sign, while the rest of the cartouche was left intact, as was the complete other cartouche, with the king’s throne name. The target was thus obviously not the individual Sety, but simply the hieroglyph representing the god Seth, inside his name. This cancellation probably occurred at some point in the 1st millennium BC, when the god Seth became particularly unwelcome in a sacred space.11 It is noteworthy that even in the earlier case of Sety I, at the beginning of Dynasty 19, the hieroglyphic figure of Osiris sometimes replaced that of Seth in the cartouche of the king, for example on the wall decoration of his tomb. Strangely, this

assimilation of Seth with the hated invaders, led to a persecution of Seth, who became the incarnation of the enemy par excellence (Soukiassian 1981; te Velde 1984, 910–1). Hammering the name of the god from the inscriptions in temples may have been a way to fight against the mythological enemies who were threatening the cosmos, and at the same time, to fight against the earthly enemies of Egypt.

298

S. CONNOR

cancellation of the Seth sign appears only on the Turin statue, while the Louvre one conserved it. I cannot see any other explanation beside the fact that perhaps the Louvre statue — or at least its inscriptions — were less visible at the time of this antiSeth iconoclasm. A figure of Amun was once carved at the top of the standard that the king holds against his left arm. On both statues, this figure was heavily mutilated — but not the god’s name on the inscriptions. No persecution against the god Amun is known after the Amarna period; the breaking of the god figure has thus probably to be related to anti-pagan iconoclasm, in Late Antiquity. On the Louvre statue only, the nose and the beard of the king were carefully cut away, as well as the head of the uraeus. As we will see, these three spots were precisely the main targets of the hammerers who wanted to ‘deactivate’ a statue. Such actions are attested throughout the whole pharaonic period and probably still in Late Antiquity, when minds and patterns of thought, although modified by the adoption of the new religion, were still impregnated with ancient Egyptian beliefs. Strangely, the nose and beard of the Turin colossus are still intact, for a reason that I cannot explain so far. Finally, a rectangular hole was cut in the front side of the base of the Turin statue. It damaged part of one of the cartouches, but it does not seem to have been an action against some hieroglyphic signs — the hole is too wide, too high and too deep. Most probably, this cavity was hollowed in this solid block of stone in order to fix a beam, perhaps for the installation of houses or structures in the courtyard of the temple, at some point in Late Antiquity or the medieval period.

The previous pages reviewed some possible causes for damage to statues. As art historians and archaeologists, we are accustomed to reconstructing the original

appearance of a piece in order to appreciate its aesthetic value, to study its stylistic features, to put it back in its architectural context, to re-enact the rituals which may have surrounded it. This is, of course, necessary; however, focusing on the actual damage to these artworks can also be extremely revealing and tell us a lot about the role that a statue may have played. The first step is to try, by looking at the archaeological context and the relationship with other pieces, to differentiate accidental breaks from intentional mutilations; then, to distinguish intentional but not necessarily meaningful breaks (as sometimes in the case of the reuse of blocks or the looting of precious material) from intentional and meaningful mutilations, which, as we have seen, seem to focus on the parts of the body which would allow the statue to be in full possession of senses (the nose, the eyes, the mouth, the ears), those permitting it to act (the arms, especially in the case of praying statues, or often the wrists in the case of seated statues, and legs), and those which would confer some symbolic or magical power (the uraeus, the sceptres, but also the beard, often carefully erased so as to deprive the figure of its male capacity). As noted by B. Bryan, statues were often treated in the same way as human bodies sometimes were in the Predynastic period,12 or in the late Middle Kingdom, in Nubia:13 broken, disjointed, decapitated. The removal of their hands, feet and heads left them functionless. Mutilating an image, like mutilating a body, would impede its potential for action (Bryan 2012, 364, 366, 368). Once the intentional nature of a mutilation has been acknowledged, detecting the date and reason of this defacing may be possible from different clues. Sometimes, the identification of the target, such as Hatshepsut or Akhenaten, allows us to suggest a date for the proscription, due to the political context (in those cases, the late reign of Thutmose III or the post-Amarna period). In other cases, the archaeological context can help us. The reserve heads, for example, seem to have been ritually damaged perhaps even during their production, or in the process of their installation in the tomb. In the case of sites that are covered with sand or debris, the upper levels provide a terminus ante quem

12

13







Conclusion and further avenues for research: when and why mutilate a statue?

As attested at the sites of Hierakonpolis and Gerza, where traces attest intentional dismembering of bodies either before or shortly after burial: head, hands and feet were removed, seemingly to hinder the movements of the dead (Petrie, Mackay and Wainwright 1912, 8–15; Dougherty and Friedman 2008).

The head and the headless body of a sacrificed individual were found associated with a ritual deposit (Vila 1963, 145–7; Vila 1973, 628–9; Jambon 2010, 4–5).

KILLING OR ‘DE-ACTIVATING’ EGYPTIAN STATUES: WHO MUTILATED THEM, WHEN, AND WHY?

for the smashing of the sculptures. This is how we can date, for example, the destruction of the statues of the sanctuary of Heqaib to the Second Intermediate Period, i.e. at the time of the royal tumuli of Kerma. The smashing of the genitals or the breast of a statue may be due to Christian prudishness in Late Antiquity, as pointed out by N. Hannestad (Hannestad 2001). In some cases, as we saw, a single statue can bear traces of successive episodes of damage, repairs and again mutilations, as in the case of the group statues of a king and the god Amun, restored by the successors of Akhenaten, and defaced again at some point in Egyptian history. Several elements can, therefore, be taken into consideration and help us to interpret damage to statues: anti-pagan iconoclasm, war damage, ritual deactivation in order to render the image harmless because of proscription, or because they did not serve anymore and had to be buried, or, in still other cases, because they were to be reused as masonry. When looking at a broken piece, its fragmentary state is an integral part of its history. Its mutilations can tell us a lot about the originally represented character, about the periods during which the piece existed, about the factors that may lead to iconoclasm, and about the material value and power that people always conferred upon images.

Bibliography Amidon, P. R. 1997. The Church History of Rufinus of Aquilea. Books 10 and 11 (translated by Philip R. Amidon). New York; Oxford. Arnold, D. 2005. The destruction of the statues of Hatshepsut from Deir el-Bahri. In C. H. Roehrig, R. Dreyfus and C. A. Keller (eds), Hatshepsut: From queen to pharaoh. New York, 270–6. Barbotin, C. 2005. Un intercesseur dynastique à l’aube du Nouvel Empire. La Revue du Louvre 4, 19–28. ———. 2007. Les statues égyptiennes du Nouvel Empire. 2 vols. Paris. Barguet, P. and Leclant, J. 1954. Karnak-Nord IV. Fouilles de l’Institut français d’archéologie orientale du Caire 25. Cairo. Bonnet, C. 1997. The kingdom of Kerma. In D. Wildung (ed.), Ancient kingdoms of the Nile. Paris, 89–95. ———. 2011. Les destructions perpétrées durant la campagne de Psammétique II en Nubie et les

299

dépôts consécutifs. In D. Valbelle and J.-M. Yoyotte (eds), Statues égyptiennes et kouchites démembrées et reconstituées. Paris, 21–32. Bryan, B. 2012. Episodes of iconoclasm in the Egyptian New Kingdom. In N. N. May (ed.), Iconoclasm and text destruction in the ancient Near East and beyond. Oriental Institute Seminars 8. Chicago, 363–94. Connor, S. 2015. Quatre colosses du Moyen Empire ‘ramessisés’ (Paris A 21, Le Caire CG 1197, JE 45975 et 45976). Bulletin de l’Institut français d’archéologie orientale 115, 85–109. ———. 2016. Le statue del Museo Egizio. Turin; Modena. ———. 2017. La statua di Ramesse II. Turin; Modena. Davies, W. V. 2003. Sobeknakht of Elkab and the coming of Kush. Egyptian Archaeology 23, 3–6. ———. 2004. Stone vessel. In D. A. Welsby and J. R. Anderson (eds), Sudan Ancient Treasures: an exhibition of recent discoveries from the Sudan National Museum. London, 101. ———. 2005. Egypt and Nubia: conflict with the Kingdom of Kush. In C. H. Roehrig, R. Dreyfus and C. A. Keller (eds), Hatshepsut: From Queen to Pharaoh. New York, 49–56. Delange, E. 1987. Catalogue des statues égyptiennes du Moyen Empire (2060–1560 av. J.-C.). Paris. Dölger, F. J. 1930. Antike und Christentum Kultur- und Religionsgeschichtliche Studien, II. Munster. Dorman, P. F. 2005. The proscription of Hatshepsut. In C. H. Roehrig, R. Dreyfus and C. A. Keller (eds), Hatshepsut: From queen to pharaoh. New York, 267–9. Dougherty, S. and Friedman, R. F. 2008. Sacred or mundane: Scalping and decapitation at Predynastic Hierakonpolis. In B. Midant-Reynes and Y. Tristant (eds), Egypt at its Origins 2. Proceedings of the international conference ‘Origin of the State. Predynastic and Early Dynastic Egypt’, Toulouse (France), 5th–8th September 2008. Orientalia Lovaniensia Analecta 172. Leuven; Paris; Dudley, MA, 311–38. Evers, H. G. 1929. Staat aus dem Stein: Denkmäler, Geschichte und Bedeutung der ägyptischen Plastik während des Mittleren Reichs I. Munich. Fluck, C., Helmecke, G. and O’Connell, E. R. (eds). 2015. Ein Gott: Abrahams Erben am Nil – Juden, Christen und Muslime in Ägypten von der Antike bis zum Mittelalter. Berlin.

300

S. CONNOR

Fournet, J.-L. 1996. Inscriptions grecques inédites de la rive ouest d’Assouan: Du nouveau sur le colosse chantant de Memnon? Bulletin de l’Institut français d’archéologie orientale 96, 143–70. Gallet, L. 2013. Karnak: The temple of Amun-RaWho-Hears-Prayers. In W. Wendrich (ed.), UCLA Encyclopedia of Egyptology, Los Angeles. . Goyon, J.-C. and Cardin, C. (eds). 2004. Trésors d’Égypte: La ‘Cachette’ de Karnak, 1904–2004: Exposition en hommage à Georges Legrain à l’occasion du IXe Congrès international des Égyptologues, Musée dauphinois, Grenoble, 4 septembre 2004–5 janvier 2005. Grenoble. Haarmann, U. 1980. Regional sentiment in medieval Islamic Egypt. Bulletin of the School of Oriental and African Studies 43, 55–66. Haas, C. 1997. Alexandria in Late Antiquity: Topography and social conflict. Baltimore; London. Habachi, L. 1985. Elephantine IV. The sanctuary of Heqaib. Archäologische Veröffentlichungen 33. Mainz. Hannestad, N. 1999. How did rising Christianity cope with pagan sculpture? In E. Chrysos and I. Wood (eds), East and west: Modes of communication. Proceedings of the First Plenary Conference at Merida. Leiden, 173–203. ———. 2001. Castration in the baths. In N. Birkle (ed.), Macellum. Culinaria archaeologica. Robert Fleischer zum 60. Geburtstag von Kollegen, Freunden und Schülern. Mainz, 66–77. Hartleben, H. (ed.). 1909. Lettres de Champollion le Jeune. Bibliothèque égyptologique 30–1. Paris. Hjort, Ø. 1993. Augustus Christianus – Livia Christiana: Sphragis and Roman portrait sculpture. In L. Rydén and J. O. Rosenqvist (eds), Aspects of Late Antiquity and early Byzantium. Stockholm, 99–112. Jambon, E. 2010. Les mots et les gestes. Réflexions autour de la place de l’écriture dans un rituel d’envoûtement de l’Égypte pharaonique. Cahiers ‘Mondes anciens’ 1: Écritures rituelles. Anthropologie et histoire des mondes antiques. Online at: . ———. 2016. La Cachette de Karnak: Étude analytique et essais d’interprétation. In L. Coulon (ed.), La Cachette de Karnak. Nouvelles perspectives sur les découvertes de Georges Legrain. Bibliothèque d’étude 161. Cairo, 131–77.

Kanawati, N. 2003. Conspiracies in the Egyptian palace: Unis to Pepy I. London; New York. Karakhanyan, A., Avagyan, A. and Stadelmann, R. 2014. Geological and archaeoseismological investigations at the Colossi of Memnon from 2009 to 2011–2012. Annales du Service des antiquités de l’Égypte 87, 61–90. Kristensen, T. M. 2010. Religious conflict in late antique Alexandria: Christian responses to ‘pagan’ statues in the fourth and fifth centuries CE. In J. Krasilnikoff and G. Hinge (eds), Alexandria: A cultural and religious melting pot. Aarhus Studies in Mediterranean Antiquity 9. Aarhus, 158–75. ———. 2012. Miraculous bodies: Christian viewers and the transformation of pagan sculpture in Late Antiquity. In S. Birk and B. Poulsen (eds), Patrons and viewers in Late Antiquity. Aarhus Studies in Mediterranean Antiquity 10. Aarhus, 31–66. ———. 2013. Making and breaking the gods: Christian responses to pagan sculpture in Late Antiquity. Aarhus. ———. 2015. Iconoclasm. In E. Friedland, M. Grunow and E. Gazda (eds), The Oxford handbook of Roman sculpture. Oxford, 667–80. Laboury, D. 1998. La statuaire de Thoutmosis III: Essai d’interprétation d’un portrait royal dans son contexte historique. Aegyptiaca Leodiensia 5. Liège. Langmann, G. 1985. Ein Kaisertaufe (?) in Ephesos. Jahreshefte des Österreichischen Archäologischen Institutes in Wien (Vienna) 56, 65–9. Lehner, M. 1991. Archaeology of an image: The Great Sphinx of Giza. Dissertation, Yale. Lembke, K. and Vittmann, G. 1999. Die Standfigur des Horos, Sohn des Thotoes (Berlin, Ägyptisches Museum SMPK 2271). Mitteilungen des Deutschen Archäologischen Instituts, Abteilung Kairo 55, 299–313. Letronne, A.-J. 1833. La statue vocale de Memnon étudiée dans ses rapports avec l’Égypte et la Grèce. Étude historique faisant suite aux recherches pour servir à l’histoire de l’Égypte pendant la domination des Grecs et des Romains. Paris. Lindblad, I. 1984. Royal sculpture of the early Eighteenth Dynasty in Egypt. Medelhavsmuseet Memoir 5. Stockholm. McGregor, R. 2013. Sufi iconoclasm and the problem of comparative religion. In G. Cecere, M. Louvet and S. Pagani (eds), Les mystiques juives, chrétiennes et musulmanes dans l’Égypte médiévale.

KILLING OR ‘DE-ACTIVATING’ EGYPTIAN STATUES: WHO MUTILATED THEM, WHEN, AND WHY?

Recherches d’archéologie, de philologie et d’histoire 35. Cairo, 173–87. Minor, E. 2012. The use of Egyptian and Egyptianizing material culture in Nubian burials of the Classic Kerma period. UC Berkeley Electronic Theses and Dissertations. . Montet, P. 1935–7. Les fouilles de Tanis en 1933 et 1934. Kêmi 5, 1–18. O’Connor, D. 1974. Political systems and archaeological data in Egypt: 2600–1780 BC. World Archaeology 6 (1), 15–38. Oppenheim, A., Arnold, Do., Arnold, Di. and Yamamoto, K. (eds). 2015. Ancient Egypt transformed: The Middle Kingdom. New York. Opper, T. 2014. The Meroe head of Augustus. London. Petrie, W. M. F., Mackay, E. J. H. and Wainwright, G. A. 1912. The Labyrinth, Gerzeh and Mazguneh. British School of Archaeology in Egypt and Egyptian Research Account 21. London. Picardo, N. 2007. ‘Semantic homicide’ and the socalled reserve heads: The theme of decapitation in Egyptian funerary religion and some implications for the Old Kingdom. Journal of the American Research Center in Egypt 43, 221–52. PM = Porter, B., and Moss, R. 19722. Topographical bibliography of ancient Egyptian hieroglyphic texts, reliefs, and paintings, II. Theban temples. Oxford. Polz, F. 1995. Die Bildnisse Sesostris’ III. und Amenemhets III. Bemerkungen zur königlichen Rundplastik der späten 12. Dynastie. Mitteilungen des Deutschen Archäologischen Instituts, Abteilung Kairo 51, 227–54. Postel, L. 2014. Le paysage monumental. In F. Morfoisse and G. Andreu-Lanoë (eds), Sésostris III. Pharaon de légende. Gand, 114–26. Reid, D. M. 2002. Whose pharaohs? Archaeology, museums, and Egyptian national identity from Napoleon to World War I. Cairo. Reisner, G. A. 1923. Excavations at Kerma, Parts IV–V. Harvard African Studies 6. Cambridge, MA. Rifaud, J.-J. 1830. Tableau de l’Égypte, de la Nubie et des lieux circonvoisins; ou itinéraire à l’usage des voyageurs qui visitent ces contrées. Paris. Roth, A. M. 2005. Erasing a reign. In C. H. Roehrig, R. Dreyfus and C. A. Keller (eds), Hatshepsut: From queen to pharaoh. New York, 277–81. el-Saghir, M. 1991. The discovery of the statuary cachette of the Luxor Temple. Sonderschrift des

301

Deutschen Archäologischen Instituts, Abteilung Kairo 26. Cairo. Schaff, P. 1892. A Select library of Nicene and postNicene fathers of the Christian church. Second series, Tome III  : Theodoret, Jerome Gennadius, Rufinus: Historical writings, etc. New York. Simpson, W. K. 1982. Egyptian sculpture and twodimensional representation as propaganda. Journal of Egyptian Archaeology 68, 266–71. Smith, M. 2007. Pyramids in the medieval Islamic landscape: Perceptions and narratives. Journal of the American Research Center in Egypt 43, 1–15. Soukiassian, G. 1981. Une étape de la proscription de Seth. Göttinger Miszellen 44, 59–68. Sourouzian, H. 1991. A bust of Amenophis II at the Kimbell Art Museum. Journal of the American Research Center in Egypt 28, 55–74. ———. 1995. L’iconographie du roi dans la statuaire des trois premières dynasties. In Kunst des Alten Reiches: Symposium im Deutschen Archäologischen Institut Kairo am 29. und 30. Oktober 1991. Sonderschrift des Deutschen Archäologischen Instituts, Abteilung Kairo 28. Mainz, 133–54. ———. 1996. A headless sphinx of Sesostris II from Heliopolis in the Egyptian Museum, Cairo, JE 37796. In P. Der Manuelian and R. Freed (eds), Studies in honor of W. K. Simpson. Boston, vol. II, 743–54. ———. 2003. Conservation of statuary. In Z. Hawass (ed.), Egyptology at the dawn of the twenty-first century. Proceedings of the Eighth International Congress of Egyptologists Cairo 2000. Cairo; New York, 406–13. Sourouzian, H., Karakhanyan, A., Avagyan A. and Stadelmann, R. 2011. Fifth report on excavation and conservation work at Kôm el-Hettan from 9th to 12th seasons (2007–2010) by the Colossi of Memnon and Amenhotep III Temple Conservation Project. Part I. The Colossi of Memnon and archaeoseismology. Annales du Service des antiquités de l’Égypte 85, 273–327. Strudwick, N. 2006. Masterpieces of ancient Egypt. London. Tefnin, R. 1991a. Art et magie au temps des pyramides: L’énigme des têtes dites de ‘remplacement’. Monumenta Aegyptiaca 5. Brussels. ———. 1991b. Les têtes magiques de Gizeh: Prothèse et mutilation. L’énigme des têtes dites de remplacement. Bulletin de la Société française d’égyptologie 120, 25–37.

302

S. CONNOR

Thompson, K. 2011. How to destroy an Amarna statue. Ancient Egypt 12 (2), 38–44. Tkaczow, B. 1993. The topography of ancient Alexandria (an archaeological map). Travaux du Centre d’archéologie méditerranéenne de l’Académie polonaise des sciences 32. Warsaw. Valbelle, D. 2004. The cultural significance of iconographic and epigraphic data found in the kingdom of Kerma. In T. Kendall (ed.), Nubian studies, 1998: Proceedings of the Ninth Conference of the International Society of Nubian Studies, August 21–26, 1998, Boston, Massachusetts. Boston, 176–85. ———. 2016. Statues enterrées, dépôts liturgiques et différentes catégories de favissae. In L. Coulon (ed.), La Cachette de Karnak. Nouvelles perspectives sur les découvertes de Georges Legrain. Bibliothèque d’étude 161. Cairo, 21–33.

Vandersleyen, C. 2005. Iahmès Sapaïr, fils de Seqenenrê Djéhouty-Aa (17e dynastie) et la statue du Musée du Louvre E 15682. Connaissance de l’Égypte ancienne 8. Brussels. te Velde, H. 1984. Seth. In W. Helck and E. Otto (eds), Lexikon der Ägyptologie V. Wiesbaden, 908–11. Vila, A. 1963. Un dépôt de textes d’envoûtement au Moyen Empire. Journal des savants, July–Sept, 135–160. ———. 1973. Un rituel d’envoûtement au Moyen Empire égyptien. In M. Sauter (ed.), L’homme hier et aujourd’hui. Recueil d’études en hommage à André Leroi-Gourhan. Paris, 625–39. Wenig, S. 1978. Africa in antiquity: The arts of ancient Nubia and the Sudan. Brooklyn. Wildung, D. 1984. L’âge d’or de l’Égypte: Le Moyen Empire. L’Âge d’or des civilisations. Paris.