This page intentionally left blank [606314]

Licență

This page intentionally left blank [606314]

Byadmin ianuarie 1, 2024

This page intentionally left blank

The Court and Court Society in Ancient Monarchies
Monarchy was widespread as a political system in the ancient world.
This volume offers the ﬁrst substantial discussion of ancient monar-chies from the viewpoint of the ruler’s court. The monarchies treated
are Achaemenid and Sasanian Persia, the empire of Alexander, Romeunder both the early and later Caesars, the Han rulers of China andEgypt’s Eighteenth Dynasty. A comparative approach is adopted to
major aspects of ancient courts, including their organisation andphysical setting, their role as a vehicle for display, and their place inmonarchical structures of power and control. This approach is broadlyinspired by work on courts in later periods of history, especially earlymodern France. The case studies conﬁrm that ancient monarchies cre-ated the conditions for the emergence of a court and court society. Theculturally speciﬁc conditions in which these monarchies functioned
meant variety in the character of the ruler’s court from one society toanother.
a. j. s. spawforth is Professor of Ancient History at the
University of Newcastle. His publications include Hellenistic and
Roman Sparta: A Tale of Two Cities (with Paul Cartledge, 2nd edition,
2002), The Complete Greek Temples (2006) and co-editing, with Simon
Hornblower, The Oxford Classical Dictionary (3rd edition revised,
2003).

The Court and Court Society
in Ancient Monarchies
Edited by
a. j. s. spawforth

CAMBRIDGE UNIVERSITY PRESS
Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo
Cambridge University Press
The Edinburgh Building, Cambridge CB2 8RU, UK
First published in print format
ISBN-13 978-0-521-87448-9ISBN-13 978-0-511-35457-1© Cambridge University Press 2007
2007Information on this title: www.cambridge.org/[anonimizat]
This publication is in copyright. Subject to statutory exception and to the provision of
relevant collective licensing agreements, no reproduction of any part may take place
without the written permission of Cambrid ge University Press.
ISBN-10 0-511-35457-6
ISBN-10 0-521-87448-3
Cambridge University Press has no responsibility for the persistence or accuracy of urls
for external or third-party internet websites referred to in this publication, and does not
guarantee that any content on such websites is, or will remain, accurate or a ppropriate.Published in the United States of America by Cambridge University Press, New York
www.cambridge.org
hardbackeBook (EBL)
eBook (EBL)
hardback

Cont ents
List of ﬁgures [page vi]
List of tables [viii]
Notes on contributors [ix]
Acknowledgements [xi]
List of abbreviations [xii]
Introd uction [1]
tony sp awforth
1N ew out of old? Court and court ceremonies in Achaemenid
Persia [17]
maria brosius
2King, court and royal representation in the Sasanian
empire [58]
josef wieseh ¨ofer
3T h e c ourt of Alexander the Great between Europe and Asia [82]
tony sp awforth
4F r i e nds in high places: the creation of the court of the Roman
emperor [121]
jeremy paterson
5The imperial court of the late Roman empire,
c.AD300– c.AD 450 [157]
rowlan ds m ith
6The imperial court in Han China [233]
hans van ess
7Court and palace in ancient Egypt: the Amarna period and later
Eighteenth Dynasty [267]
kate spence
Bibliography [329]
Index [354]

Figures
1.1 Staircase of the Apadana (after F. Krefter) page 28
1.2 Audience scene from Persepolis. Photo: M. Brosius 32
1.3 Plan of Pasargadae. Drawing by Marion Cox 48
1.4 Plan of Persepolis 50
2.1 Relief of Shabuhr I at Naqsh-i Rajab near Persepolis.
Photo: J. Wieseh ¨ofer 68
2.2 The deer hunt from the relief of Khusro II at Taq-i Bustan.
Photo: J. Wieseh ¨ofer 76
5.1 The ‘palace’ of Diocletian at Split (artist’s reconstruction byE. H ´ebrard. Photograph from E. H ´ebrard and J. Zeiller,
Spalato, le Palais de Diocl´ etien ,Paris 1912) 190
5.2 The ‘palace’ of Diocletian at Split (plan, from D. S. Robinson,Greek and Roman Architecture ,Cambridge, 1929) 192
5.3 Theodosius I and his court in the Hippodrome inConstantinople 214
5.4 The ‘Missorium of Theodosius’ (Real Academia de Historia,Madrid). Photo by Visipix.com 217
5.5 Leaf of an ivory diptych commemorating the entry of
Probianus into his term of ofﬁce as Vicarius of Rome (from
L. von Sybel Christliche Antik´ e,volume 2. Marburg, 1909) 218
7.1 Royal procession from the Narmer Palette 270
7.2 Akhenaten and his family in chariots (reproduced by kindpermission of the Egypt Exploration Society) 292
7.3 Huyawith the royal family (reproduced by kind permission
of the Egypt Exploration Society) 296
7.4 Plan of el-Amarna 304
7.5 The palace of Amenhotep III at Kom el-‘Abd 306
7.6 The King’s House at el-Amarna (reproduced by kindpermission of the Egypt Exploration Society) 307
7.7 The North Palace at el-Amarna 308
7.8 The reception of foreign tribute (reproduced by kindpermission of the Egypt Exploration Society) 314

List of ﬁgures vii
7.9 Akhenaten and Nefertiti rewarding the ofﬁcial Parennefer
(reproduced by kind permission of the Egypt ExplorationSociety) 315
7.10 Reward scene from the tomb of Ay at el-Amarna (reproduced
bykind permission of the Egypt Exploration Society) 321

Table
2.1 Table showing the rulers of the Sasanian dynasty page 80

Notes o nc o n t r i butors
maria brosius is Reader in Ancient History at the University of New-
castle upon Tyne. Her publications include Women in Ancient Persia (559–
331 BC) (1996, repr. 1998, 2002 ),The Persian Empire from Cyrus II to Artax-
erxes I (2000 ), and The Persians: An Introduction (2006 ). She has edited a
volume on AncientArchives: Concepts of Record-Keeping in the Ancient World
(2003 ), and (with A. Kuhrt) Persian Studies: Essays in Memory of D. M. Lewis
(1998).
jeremy paterson is Senior Lecturer in Ancient History at the Univer-
sity of Newcastle upon Tyne. He is primarily a social and economic historian
of the Roman world, with a particular interest in the wine trade. He recentlyedited, with Jonathan Powell, Cicero the Advocate (2004) and is currently
working on a book about the ancient economy. His interest in the nature of
power has already appeared in a study of emperors’ jokes and will reappearin the research he is undertaking on the relationship between Christians andRoman emperors.
rowlan ds m ith is Lecturer in Ancient History in the School of Histor-
ical Studies at Newcastle University. In his research he works chieﬂy on thecultural and social history of the high and late Roman empire, and on paganresponses to Christianisation: he is the author of Julian’s Gods: Religion and
Philosophy in the Thought and Action of Julian the Apostate (1995). His recent
publications include a study of patronal imagery in fourth-century Romeand a chapter in the Blackwell Companion to the Roman Empire (2006) on
the representation of the Roman past in the historiography of the imperialage.
tony sp awforth is Professor of Ancient History at the University of
Newcastle upon Tyne. His publications include (with P . A. Cartledge) Hel-
lenistic and Roman Sparta: A Tale of Two Cities (1989; 2nd edition, 2002),
the third edition (with S. Hornblower) of The Oxford Classical Dictionary
(3rd edition revised, 2003), and many contributions on aspects of the Greek
world under Roman rule. He is currently preparing a book on the Roman
reshaping of Greek cultural identity in the Principate.

x Notes on contributors
kate spence is an Afﬁliated Lecturer in the Faculty of Oriental Studies
at the University of Cambridge and a Fellow of the McDonald Institute forArchaeological Research, also at the University of Cambridge. Her research
focuses on the monumental and domestic architecture of ancient Egypt.She has excavated at several sites in Egypt including Akhenaten’s capitalat el-Amarna where she was involved in the re-examination of the NorthPalace.
hans van ess has held a chair in Chinese Studies at the University of
Munich since 1998. His main interests include the history of Chinese thought
with a special emphasis on Confucianism, early Chinese historiography,
early Chinese poetry and the history of relations between China and centralAsia. Among his publications are two books on the introduction of Con-
fucianism as a state orthodoxy under the Han and on the Neo-Confucianmovement under the Sung, as well as an introduction to Confucianism forageneral readership. Currently he is working on a comparison of the two
ﬁrst Chinese standard histories.
josef wieseh ¨ofer is Professor of Ancient History at the University
of Kiel (Germany) and director of its Institute for Classical Studies. He isamember of the Centre for Asian and African Studies at Kiel University,
editor of Oriens et Occidens and co-editor of Asien und Afrika ,Achaemenid
History and Oikumene .His main interests lie in the history of the ancient
Near East and its relations with the Mediterranean world, social history, the
history of early modern travelogues and the history of scholarship.

Acknowledgements
The editor and contributors warmly acknowledge their major debt to the
British Academy which funded the bulk of the costs of a workshop in 2004,
which brought all the contributors together in Newcastle. The idea for aninterdisciplinary project of this kind was stimulated by the collaborativeatmosphere of the School of Historical Studies at Newcastle; Jeremy Boulton,Head of School, also gave welcome additional funds for the costs of the
workshop. For helpful advice in the course of the project T ony Spawforth
would like to thank Yuri Pines, Naomi Standen, Roel Sterckx, Rosalind
Thomas and Philip van der Eijk. For practical support the project is gratefultoSarah Francis and the staff in the School Ofﬁce. All contributors owe a
debt of gratitude to Michael Sharp at Cambridge University Press for hisconstructive interest in the project from its inception, and to the Press’s two
anonymous assessors, whose comments improved an earlier draft of the text.TonySpawforth, ﬁnally, would like to thank his fellow-contributors for their
magniﬁcent engagement with the courts project, and for their good-naturedresponsiveness to editorial nagging.
To ny Spawforth

Abbrev iations
AchH ist Achaemenid History
AJA American Journal of Archaeology
AJP American Journal of Philology
Akk. Akkadian
ANET AncientNear Eastern Texts relating to the Old Testament ,e d .
J.B.Pritchard, 3rd rev. edn. Princeton (1969)
AnSt Anatolian Studies
AntAfr Antiquit´ es Africaines
Bab. Babylonian
BEFAR Bull´etin de l’Ecole fran¸ caise ` aRome: Antiquit´ e
CahDAFI Cahiers de la D´ el´egation Arch´ eologique Fran¸ caise en Iran
CIL Corpus Inscriptionum Latinarum (1863–)
CQ Classical Quarterly
CR Classical Review
DB Darius (Inscriptions of) Bisitun
DF E. Gibbon, Decline and Fall of the Roman Empire ,e d .
O.Smeaton, vols. I–VI, London (1910)
DHA Dialogues d’Histoire Ancienne
DN Darius (Inscriptions of) Naqsh-i Rustam
DOP Dumbarton Oaks Papers
DP Darius (Inscriptions of) Persepolis
DS Darius (Inscriptions of) Susa
EA Egyptian Archaeology
Elam. Elamite
EncIr Enciclopedia Iranica ,ed. E. Y ar-Shater (1985–)
FGH C. and T. M ¨uller, Frag menta Historicorum Graecorum
(1841–72)
FGrH F. Jacoby, Frag mente der griechischer Historiker (1923–)
Gr. Greek
GRBS Greek, Roman and Byzantine Studies
grI Greek version of the royal inscription(s)
HHS HouHan-shu
HS Han-shu

List of abbreviations xiii
ILS H. Dessau, Inscriptiones Latinae Selectae (1892–1916)
IrAnt Iranica Antiqua
J¨OB Jahrbuch des ¨ osterreichischen Byzantinistik
JRA Journal of Roman Archaeology
JWCI Journal of the Courtauld and Warburg Institutes
L¨A Lexikon der ¨Agyptologie
MAAR Memoirs of the American Academy in Rome
MCAAS Memoirs of the Connecticut Academy of Arts and Sciences
MDAI(R) Mitteilungen des deutschen arch¨ aologischen Instituts: R¨ omische
Abteilung
mpI Middle Persian version of the royal inscription(s)
Neh. Book of Nehemiah
NPi (Inscription of) N(arseh from) P(aikul)i (ed. H. Humbach
and P . O. Skjærvø, 1978–83)
OCD3S. Hornblower and A. Spawforth (eds.), The Oxford Classical
Dictionary ,3rdedn. Oxford (2003)
paI Parthian version of the royal inscription(s)
PdP Parola di Passato
PF R. T. Hallock, Persepolis Fortiﬁcation Tablets .Chicago (1969)
PFs R. T. Hallock, ‘Persepolis Fortiﬁcation texts’ , CahDAFI 8:
109–36
PFS Persepolis Fortiﬁcation Seals
RE A. Pauly, G. Wissowa and W. Kroll, Real-Encyclop¨ adie der
klassischen Altertumswissenschaft (1893–)
SAA State Archives of Assyria
SC Ssu-ma T’an and Ssu-ma Ch’ien (eds.), Shih-chi (Records of
the Scribe) .Peking (1959)
ˇSH (Inscription of) Sh(abuhr I from) H(ajjiabad)
SHA Scriptores Historiae Augustae
SIG3W. D ittenberger, Sylloge Inscriptionum Graecarum ,3 r de dition
(1915–24)
ˇSKZ (Inscription of) Sh(abuhr I on the) K(a‘ba-i) Z(ardusht at
Naqsh-i Rustam), ed. P . Huyse (1999)
SO Symbolae Osloenses
SSC Shi-san ching chu-sh¨ u(The Thirteen Classics)
StIr Studia Iranica
XP Xerxes (Inscriptions of) Persepolis
YClS Yale Classical Studies
ZA W Zeitschrift f¨ ur alttestamentliche Wissenschaft
ZPE Zeitschrift f¨ ur Papyrologie und Epigraphik

Introd uction
tony sp awforth
Taking the ﬁeld of ancient history as a whole, including the Middle and Far
East as well as Europe and the Mediterranean, the monarchical court cannot
be said to have occupied centre-stage in a way that might seem justiﬁed by theprevalence of monarchy as a system of power in antiquity.
1The reasons for
this relative neglect are complex and cannot all be unpacked here.2But o ne,
certainly, is the sense of the court as a ‘moribund social formatio[n]’3which
has permeated western consciousness since the French Revolution. Back-stairs inﬂuence, intrigue and ﬂattery: these generic phenomena of courtshave earned themselves a bad reputation in western democracies whichpose as the mirror-opposites of ‘old-regime’ arrangements of power, and inthe study of ancient monarchies they are often sidelined, or their associa-tion by ancient writers with ‘bad’ or ‘weak’ rulers, or with whole societiesclassed as degenerate, as the ancient Persians were by the ancient Greeks(Brosius in this volume), taken at face value.
4Inthose nineteenth-century
European monarchies which survived, especially after 1848, constitutional-
ism was the order of the day, as it had been (at least notionally) in the UnitedKingdom since 1688. Perhaps unsurprisingly, then, in the nineteenth and
for much of the twentieth century the students of monarchical institutionsin the ancient world have often been preoccupied with modernist attemptstodeﬁne their legal basis – as with the Macedonian kings, say, or the early
1Iamgrateful to the two anonymous assessors of this book, and above all to my collaborators in
this project, for comments which have helped to deﬁne and reﬁne the issues raised in this
Introduc tion.
2Afurther indication – apart from this book – that a change is afoot is the conference on ‘Royal
Courts and Capitals’ (Istanbul, October 2005 )which compared ancient, Islamic and European
courts and included two Roman historians among the speakers: see the conference report:
Mansel 2006 .
3Elias 1983 :8.Cf. ibid.: ‘As far as [courts] still exist in developed countries, they have lost much
of their earlier power and prestige. As compared to the time of their apogee, the court societiesof our day are mere epigones. The representatives of rising social formations usually regard theseremnants of a past era with mixed feelings.’ In the UK this last point is probably more true now
than in the period after the First World War when Elias wrote.
4Inthe UK media some newsprint journalists activate a depreciatory stereotype of the court to
criticise government when it is seen to act in an insufﬁciently open and democratic way: thus ‘thecourt of King T ony [Blair]’ , the ‘courtiers’ in cluding unofﬁcial advisors like Alistair Campbell.

2 tony spawforth
Roman emperors. In the legacy of earlier generations of scholars, there are
obvious exceptions to this sidelining of the court, such as the researches of
L. Friedl ¨ander on early imperial Rome or H. Berve on Alexander the Great,
and these have proved mighty bulwarks for at least two of the chapters in thisbook (Spawforth; Paterson). But even these studies tend to be descriptiverather than analytical. As the contributors to this book have found out, the
court as a central entity within the monarchies which they study has often
been taken for granted (Brosius; Spawforth), or scarcely conceptualised atall (Wieseh ¨ofer; van Ess; Spence). A related reason for this neglect is the tra-
ditional separation in western scholarship of the ‘trappings’ from the ‘sub-stance’ of power. The ‘trappings’ involve aspects of monarchy which scholarstrained to focus on the history of events or institutions have traditionally
found hard to take seriously as objects of study – ceremonial, say.
5They also
involve the crossing of interdisciplinary boundaries: the study of palaces,gardens, costume and iconography is traditionally the ﬁeld of art historiansand archaeologists; the study of feast s–as u b j e c tw h i c ho f t e nc o m e su pi n
this book – is arguably more the domain of the social scientist; and so on.Finally, this interdisciplinary complexity is nowadays being reinforced by thegendered approaches which are reappraising the role of women in rulers’courts, both ancient and modern. The prejudice in most if not all ancient
societies against women rulers (see Spence in this volume on EighteenthDynasty Egypt) has meant that powerful royal women have often been pre-sented as ‘unnatural’ by ancient (male) writers, from Achaemenid queensand princesses to the younger Agrippina, when in fact this sort of inﬂu-ence is a sociological phenomenon arising more or less naturally from theembedding of a ruler’s exercise of power in the domestic setting of the court.
What, then, is a ‘court’? All the chapters in this book have sought to
answer this question with reference, ﬁrst, to contemporary perceptions.Insome ancient societies a word for ‘court’ is harder to track down than
in others, and here the issue may require more lexicographical research(Spence and Wieseh ¨ofer in this volume). Elsewhere, with Chinese ch’ao ,s a y ,
meaning literally ‘morning audience’ , a modern dictionary may retrojecttoday’s sense of ‘court’ , a meaning of which the Han Chinese may not have
been fully conscious (van Ess in this volume). It is clear, though, that theancient Greeks and Romans had conceptualised the court at least to somedegree.
6Thus the Greek word from which the modern neologism ‘aulic’
derives, aul¯e,along with the Latin equivalent aula,isused by ancient writers
5There are, of course, signiﬁcant exceptions to this, e.g. Price 1984 ;Cannadine and Price 1987 .
6Winterling 1997 :151 n. 1. Quite how much is debated: Herman 1997 :204–5; Gregor 1997 :31.

Introd uction 3
both of the ruler’s dwelling as a physical entity and in a more abstract sense of
the people to be found there – ‘those peri t ¯enaul¯en’(see Spawforth, Paterson
and Smith in this volume).
This idea, that ‘the court’ is both the spatial framework of the ruler’s exis-
tence and also the social conﬁguration with which he shares that space, is
fundamental in modern attempts to deﬁne and analyse the court. It mightbe argued that the word ‘household’ then starts to look like a perfectly ade-quate alternative to ‘court’ , in the sense of the members of the ruler’s familycohabiting in his dwelling, along with their domestic attendants and body-
guards. There is some cogency in this viewpoint. But a concept is neededwhich takes account of more than simply household personnel. For a start,the dwellings of ancient rulers were the focus for decision-making and gov-ernance in monarchies which were all (in the case of the ones studied inthis book) personal and more or less absolutist. Factored into the ruler’sentourage in this kind of ‘Weberian’ patrimonial monarchy, therefore, mustbe the comings-and-goings of political ‘helpers’ and, in some cases (the lateRoman and Han Chinese empires, say: see Smith and van Ess in this vol-
ume), a ﬂedgling state bureaucracy. The apocryphal saying of Louis XIV,‘L’´etat c’est moi’ , could not be more misleading about the complex reality
of the exercise of royal power in pre-Revolutionary France, where ‘ideas,practices and even institutions’ did much to limit – albeit not control – theFrench king’s ‘theoretically formidable’ authority (Antoine 1989 :175–9).
Some such formulation, albeit with less emphasis (perhaps) on ‘institu-tions’ , could be put forward in summary of the powers of most, if not all,the monarchies studied in this book (the nomoi or ‘customs’ , say, which
restrained the Macedonian king). For this reason, the ruler’s space was also
the potential site of exchanges between ruler and all manner of subjects whowerenot necessarily members of his household or even in any sense his
ofﬁcials. This space, ﬁnally, was also where envoys of foreign powers werereceived. The attempts by each side to control or manage these exchanges
gave rise to the theatricality often thought of as characteristic of courts: pompand circumstance on one side; a carefully controlled demeanour, incorpo-rating deference, ingratiation and ﬂattery, on the other. The semantic ﬁeld
‘court’ best conceptualises the idea of a social conﬁguration characterised
bythese distinctive modes of communication.
The chapters which follow highlight the state elites as the key-group of
subjects in the workings of ancient courts. The study of monarchical courts,then, is unavoidably a form of elite history, since no ancient monarchies (orat least, none of those studied here) conceived themselves as instrumentsof ‘people power’ , even if gift-giving to the masses (the ‘bread and circuses’

4 tony spawforth
of imperial Rome) was one of ‘the necessary costs of stable autocratic gov-
ernment’ in antiquity.7The court in its physical sense was not entirely iso-
lated from the people – the linkage of the late Roman imperial palace tothe hippodrome is the most striking example of the quasi-integration ofordinary subjects into the spatial conﬁguration of the courts studied here(Smith in this volume). More often, however, as the following chaptersshow, the court turns out to be a place where issues of access to the rulerseem mainly to focus on the elites: these were the people whose face-to-faceencounters with the ruler Akhenaten’s Window of Appearance (Spence inthis volume), say, or Alexander’s state tent (Spawforth in this volume), nottomention feasts and hunts (Smith in this volume), were – to differing
degrees – intended to manage. Ancient monarchs, like those of later peri-ods, relied on trustworthy servants with whom they were obliged to sharetheir power if territories were to be administered, armies commanded, andother functions of ancient-world governance discharged. From the ruler’spoint of view, management of relations with his elites was critical, since itwas this group which provided both his key helpers and, as often as not, themost potent source of attempts to supplant him. The chief spatial and socialsetting in which both ruler and elites sought to manage their mutual interac-tion was the court. In the following chapters the manner of this interactionis shown to be culturally speciﬁc in different ancient societies in a range ofways; in all chapters, however, it is a recurrent and a major theme. Mostchapters distinguish an ‘inner’ from an ‘outer’ court, the former comprising
the ruler and those whom service or kinship kept more or less permanentlyin his vicinity, the latter denoting members of the elite who were a moreintermittent presence, in part by virtue of the coming-and-going betweencentre and periphery imposed on them by the delegated power with which
the ruler entrusted them.
Ancient courts, then, were complex entities. In negotiating this com-
plexity, contributors to this volume were able to take their bearings in partthanks to the ediﬁce of current scholarship on medieval and early moderncourts inspired, as even his critics concede, by the German social historian
Norbert Elias.
8Elias was the ﬁrst scholar to impart scholarly rigour to the
study of monarchical courts. His doctorate was ﬁrst published in 1933. AsDie h¨ oﬁsche Gesellschaft (‘The Court Society’) it only became widely known
ageneration later, however, following the publication of a new (and differ-
ent) German edition in 1969. This was translated into French in 1974, withEnglish translation by Edmund Jephcott following in 1983.
7Paterson 2004 .8E.g. the comment of Duindam 2003 :7 .

Introd uction 5
Elias centred his work on the court of Louis XIV and his successors at
Versailles, for his picture of which h ewas heavily reliant on the brilliant
but partisan memoirs of a courtier, the duc de Saint-Simon (1675–1753).
The text of Elias is rich and dense. One of his central propositions, thatthe court society of pre-Revolutionary France marked a necessary stage inthe ‘socio-genesis’ of the nineteenth-century French state, does not directlyconcern this book. Another, however, is more relevant. Basing himself on
Saint-Simon, Elias posed the question: How do hereditary monarchies ofthe patrimonial type reproduce themselves from one ruler to the next, oftenover a period of centuries? He argued that the crucial relationship in this
type of mo narchy was between the ruler and the elites (see above), and
that the court was the means by which each sought to bring inﬂuence tobear on the other – its fulcrum. In his case study, Louis XIV was cast asaconserving ruler, third king of the Bourbon line, whose chief aim was to
maintain his inherited position of power. Louis achieved this by constructingthe ‘hyper-palace’ of Versailles. Here he ensnared the kingdom’s elite – theFrench high nobility – by means of an elaborate system of etiquette. This kept
grandees in their place by conferring or withholding prestige-fetishes, such
as the notorious privilege of holding the candlestick at the royal coucher.The Versailles system worked for Louis XIV, according to Elias, because itmanipulated the aristocratic outlook of noble courtiers and in particulartheir obsession with honour and distinction. For the monarchy, Versaillesoffered the means to replace the real power of the high nobility with honoriﬁcfunctions.
Elias’ arguments have not escaped a revisionist assault from historians
in recent years. In particular, it is argued that he exaggerated the abso-lutism of Louis XIV and underplayed the fact that Louis and his courtierswereengaged in a mutual negotiation, in which the latter exchanged their
attendance and their deference for royal patronage and the wherewithal tomaintain traditional aristocratic status-goals;
9and, even if Louis XIV can
be said to have dominated his court, this was less obviously the case withhis two successors, where the reverse could seem more the case.
10Even so,
the work of Elias still remains perhaps the richest source of reﬂections onthe ruler’s court as a social conﬁguration. One of its strengths is that it con-stantly stands back and risks general observations about power, monarchy
9Duindam 1995 ;2003 :esp. 7–10 with earlier refs.
10E.g. Louis XVI’s ‘tr `es-arri `ere-cabinet’ , with the comment of Verlet 1983 :525: ‘Le superlative dit,
`al u i seul, le d ´eveloppement du mal dont souffre le souverain, repouss ´ec h e zl u ip l u sl o i n par la
pression de ses courtesans, dans ce ch ˆateau [Versailles] o `ui l´etouffe.’ But see Hours 2002 for an
Eliasque reappraisal of Louis XV’s ‘strategy of control of his Court’ .

6 tony spawforth
and social structure. The interest of these for scholars of ancient courts can be
gauged by the number of citations of Elias in the chapters which follow. Forstudents of ancient courts, Elias is also highly stimulating for his analysis –less fresh in 2007, of course, than at the time when he wrote it – of the ‘sub-stance’ of monarchical rule by means of the ‘trappings’ , and for insisting onthe importance of sociological concepts, notably conspicuous consumptionand status, in trying to understand the workings of courts. Less remarkablenow, to be sure, than when he originally wrote, this kind of approach, whileit may seem commonsensical these days in some ﬁelds of history-writing,cannot be said to be taken for granted in the study of antiquity.
Inrecent years the work of Elias has started to attract historians tending
one particular corner of antiquity, namely Greece and Rome. In the 1970sthe potential of the Eliasque approach was recognised by Keith Hopkins,well known for using methodology informed by the social sciences to break
new ground in studies of the Roman empire.
11Inthe mid-1990s, in his
chapter on the court of the Julio-Claudian emperors for the Cambridge
Ancient H istory Andrew Wallace-Hadrill acknowledged his debt to Elias,12
as did Gabriel Herman in an original study of Hellenistic court politicsand court etiquette (see below).
13Morerecently there has been a spate of
German studies of ancient (classical) courts within the Elias tradition ledbythe wide-ranging historian Aloys Winterling, who worked ﬁrst on the
court of the early modern electors of Cologne
14before turning to ancient
(classical) courts, above all the Roman imperial court.15This volume is
greatly indebted to the contributions of Winterling and his colleagues. That
said, their work is not well known in Anglophone scholarship.
Turning to the present volume, it was born out of a common conviction
among the contributors that our understanding of ancient monarchies couldbe usefully improved by viewing them through the optic of the court. A ﬁrstaim is to explore ways of adding conceptual rigour to an aspect of ancientrulership which, as noted earlier, has tended to be taken for granted, if it has
been considered at all. A second aim is to take the Elias-inspired debate aboutthe court to a range of ancient societies. T o an extent this approach has beenanticipated by Winterling ( 1997 ), in a collection which looked comparatively
at the ruler’s court across the sweep of classical antiquity. The Newcastleproject, however, sought to take the discussion of ancient courts out of thisclassical ‘box’ and to adopt a cross-cultural perspective. Egypt, Persia and
11Hopkins 1978 :181. See Smith in this volume.12Wallac e-Hadrill 1996 :285 n. 13.
13Herman 1997 .14Winterling 1985 .
15Winterling 1997b ;1998 ;1999 .Inthe later stages of the preparation of this volume another
collective work of German scholarship has appeared which explicitly seeks to ‘theorise’ the
court: Butz, Hirschbiegel and Willoweit 2004 .

Introd uction 7
China were chosen for a number of reasons. Diversity is one. Han China
was clearly not inﬂuenced by the practices of the classical world; nor wasEgyptinthe period studied by Spence in this volume. Along with Persia, all
three were ancient societies where little or no explicit attempt has so far beenmade to model the ruler’s court, and this chance to experiment was a furtherattraction. Eighteenth Dynasty Egypt, moreover, was at best a proto-historicsociety, and Spence’s investigatio nnecessarily addresses the methodological
challenge of melding historiographic models with archaeological evidence to
afar greater extent than other chapters in the book. Within classical antiquity,
Alexander the Great was an attractive ﬁgure because on the one hand hiscourt ceremonial is relatively well attested, while on the other he tends to be
seen chieﬂy as a military, not an aulic, ﬁgure. As already noted, the court ofthe Roman emperors is far from being virgin territory. But we felt that therewas still room for reviewing the imperial court in an explicitly comparativeway. The Roman Principate allows the initial stages in the creation of acourt society to be followed in some detail by the usual standards of ancient
evidence. The two chapters on Rome by Jeremy Paterson and Rowland Smith
not only debate with Elias, but also take account of recent German workas yet relatively unknown in the Anglosphere. With all the ancient statesunder review, in one respect the project sought to compare like with like.States were chosen which combined strong monarchies with empire in theperiods of their history examined in this book, because these seem to be theconditions in which, historically, court culture has tended to ﬂourish.
The Newcastle project explicitly adoped an interdisciplinary approach,
and all contributors are aware of, and in different ways have been stimulatedby,contemporary court studies. Inevitably, the ﬁgure of Elias looms large.
Itis important to stress, however, that the project was not intended to be
merely reactive to Elias and modern court studies. Indeed, the collection isnot touting a methodology as such, n or does it espouse any one model of
the court. In fact, despite the impact of The Court Society on modern work
on the court, currently there appears to be no authoritative model of a courttowhich early modern historians all subscribe.
16
Inorder to strike a balance between diversity and the coherence of theme
required to sustain a multi-authored approach, the editor provided an initial
brieﬁng paper for contributors to consult when writing their ﬁrst drafts. Allcontributors attended a workshop in Newcastle in May 2004, where each
pre-circulated draft was discussed among the group. The workshop con-cluded by subscribing to a common agenda around which ﬁnal contributions
16NoteDuindam 2003 :318–20.

8 tony spawforth
could cohere.17Various issues were agreed to merit discussion, although it
was up to individual contributors to decide where the emphasis should
lie.
Itwas proposed that each paper in its ﬁnal form should reﬂect on two
broad questions: whether it was legitimate to talk of a ‘court’ in the speciﬁcmonarchy being discussed, and how crucial the ruler’s court was for under-standing the machinery of power in the double sense of actual decision-making and power’s ‘representation’ . T o pursue these questions, furtherissues were identiﬁed which should, or could, be considered.
1Can the concept of a ‘court’ and of ‘courtiers’ be identiﬁed in contempo-
rarythought and language?
2Whobelongs to the court? How does membership break down into dif-
ferent status groups? Is it helpful to think in terms of an inner and outercourt of, respectively, people permane ntly in the ruler’s personal vicinity
and others whose presence is temporary?
3Howarethe people closest to the ruler recruited? How far is the ruler’s
freedom to recruit these people tempered by, for instance, establishedcareer-structures, where these can be said to exist at all?
4What structures of communication characterise the court? How is phys-
ical access to the ruler articulated? In the interaction between ruler andcourtiers, what part is played, and for what ‘structural’ reasons, by ﬂattery
and servile opportunism, or intrigue, or faction?
5What is the relationship of the ruler’s domestic setting or household
function to decision-making? How far, and for what particular reasons,do members of the household, including relations and domestic func-tionaries, inﬂuence decision-making?
6Does the ruler’s residence function as a site of representation? ‘Represen-tation’ embraces the whole range of forms of symbolic communication,such as art and architecture, ceremony and costume, which legitimate theruler’s authority.
7Does the court display ‘stateliness’ , in the sense of clearly amounting toan institution in its own right, acquiring organisational complexity, orcoming to function of its own volition, or inﬂuencing social attitudes, or
generating an autonomous ‘court society’?
8Howimportant is the court as a symbol of social order, a place where
social ranking is put on display and different elite groups are integratedbymeans of ceremony?
17Particular thanks are owed to Josef Wieseh ¨ofer, who kindly offered the workshop a preview of
Butz et al. 2004 ,avolume which, at the time, was still in press.

Introd uction 9
Inconsidering this wide range of issues, contributors were of course left
free to place the emphasis where they thought ﬁt. But the working deﬁnition
of the court and court society was intended to ensure a measure of rigourwhen contributors sought to assess – as they were encouraged to – whetheragenuine court phenomenon could be diagnosed for a given society, that is,
an entity which clearly went beyond the inevitable ‘group dynamics aroundleaders’ (Duindam 2003 :318; but see Spence in this volume).
From the outset, it was clear that the ancient sources would be a critical
constraint in the investigation of these issues. All contributors were therefore
asked to make explicit the strengths and drawbacks of the source material attheir disposal. It cannot be emphasised enough how limited these sourcesare when compared to those, for example, available to the modern historianstudying the court of Louis XIV. In this last case, the material includes worksof literature and reports by observers (Saint-Simon’s memoirs; the missivesof foreign ambassadors), courtiers’ journals (the marquess de Dangeau’snotably
18), ofﬁcial records of court departments and court ceremonies,
royal artefac ts, and of course the royal residences themselves, which, even if
destroyed or (as in the case of Versailles) severely altered since the Revolu-tion, are still copiously documented by the surviving archives.
The ancient societies viewed here exemplify enormous diversity in the
different sorts of limitations imposed by the evidence. For classical antiqui-tyweare heavily reliant on the artful works of a literary elite predisposed –
bycultural tradition as much as by political feeling – to sing the praises of
personal autonomy and view monarchy with ambivalence. In Alexander’scase this literary bias is further skewed by the fact that the extant Alexander-historians wrote, at the earliest, some three centuries after Alexander’s death;inevitably, they import the cultural colouring of their (Roman) time. AncientChinese historiography provides the basis for the study of the Han Chinesecourt and displays similar difﬁculties for the modern historian (van Ess in
this volume): the great Ssu-ma Ch’ien was critical of imperial policy; theBook of the Later Han was composed some four centuries after the start of
the Later Han period; and so on. The most striking feature of the sourcesfor the Achaemenid Persian period is the authority of a body of contempo-rarywriting from a largely hostile, neighbouring culture (Greece), which
used (and arguably distorted) representations of the Achaemenid empire
18Along with the memoirs of Dangeau’s grandson, the duc de Luynes, this journal suggests the
enormous interest which a court society takes in what might seem to modern historians to be
the trivia of the king’s day, including details of the king’s drinking, sleeping and hunting habits,recalling Alexander’s journal ( eph¯emerides ), a work held by some modern scholars to be an
ancient ﬁction (Hammond 1983 :5–10).

10 tony spawforth
as a means of constructing the cultural ‘other’; this can be offset to a lim-
ited degree by contemporary documents and archaeology, given full weightbyBrosius in this volume. Sasanian Persia, likewise, has left behind no
contemporary literature of its own about the court, although there is rele-
vant material in the literary culture of a hostile neighbour (the late Romanempire), not to mention much later Iranian traditions with the usual prob-lems of distortion and colouring. Inscriptions play an important part inoffsetting the dearth of other kinds of written testimony in the cases of Per-sia and Egypt. These texts are not, of course, without problems of their own:Egyptian funerary inscriptions address eternity; ofﬁcial inscriptions of the
Achaemenid or Sasanian rulers belong to the ﬁeld of royal representation
and must be understood as such. As for material remains, for one of the con-tributions, on Alexander, they are simply not available: modern knowledge
of the physical setting of Alexander’s court is dependent on written descrip-tions. For the other ancient societies studied in this volume there is, indeed,
archaeological evidence, although the modern tradition of archaeologicalexploration is much stronger for some parts of what was once the ancient
world, such as Egypt, than for others, such as the Sasanian empire (respec-
tively Spence and Wieseh ¨ofer in this volume). Eighteenth Dynasty Egypt
and Achaemenid Persia are particularly remarkable for the survival of a richbody of ‘royal’ art depicting the ruler and the court; these to some extentcompensate for the absence of much (Egypt) or anything (Achaemenid Per-
sia) in the way of indigenous written observations about the ruler’s court.
The chapters in the book illustrate this diversity of ancient source-material
and also show how this diversity limits, in different ways, our attempts toanalyse ancient courts. At one extreme, that of Sasanian Persia, simply todelineate the court in broad brush-strokes, on the basis of a fragmentary andproblematic palette of evidence, is an achievement. At the other extreme,the court of the fourth- and ﬁfth-century Roman emperors is attested by acopious body of evidence across a wide range of media. The varying nature
of the surviving evidence inevitably makes for difference in the length ofchapters in this book. The one on Sasanian Persia, a ﬁrst attempt to document
the Sasanian court, is relatively brief. The length of the chapter on the laterRoman empire (Rowland Smith), by contrast, is justiﬁed by the richness of
the evidence and the consequent complexity of the topic.
What can be said to have emerged from this book? Some general remarks
can ﬁrst be made about the Eliasque approach and its relevance to ancientcourts. In terms of the physical arenas for courtly behaviour, none of the
ancient courts studied here turns out to have been like Versailles in the sense

Introd uction 11
of being housed in a palace which also provided lodgings for elite courtiers.19
Partly this seems to have been a function of the absence of a tradition of
domestic attendance on ancient rulers by high-status members of the elite –
except in the case of Achaemenid Persia and Macedon (both societies, sig-
niﬁcantly, without large slave populations). But there were certainly courtcapitals like Persepolis, early imperial Rome, the palace-cities of the Han,or Akhetaten. Another practice of Versailles, the system of entr ´ees which
controlled the admission of members of the elite to the ruler’s presence ﬁrst
thing in the morning, has ancient parallels (Paterson). This has long beennoticed, and may be understandable in broadly similar terms: ‘attendanceon the prince’s ﬁrst public appearance has a sound practical reason: to ﬁndout what he intends to do, to try to remain in his company and hope to enjoyits attendant advantages’ .
20The disproportionate inﬂuence at court of low-
status household attendants, especially those whose personal services madethem intimates of the ruler, turns out to be a recurrent feature of courtsboth ancient and modern. Imperial freedmen and eunuchs in Roman andChinese antiquity (Paterson; Smith; van Ess), royal valets at Versailles:
21the
amalgam of private household and ‘stateliness’ which constituted the ruler’scourt had the effect of politicising quite lowly domestic roles if they con-
ferred frequent access to the ruler. The ruler’s promotion of such peoplehas also been seen as structural rather than a function (as contemporariesclaimed) of personal ‘weakness’: a device, that is, to offset over-relianceon an entrenched military and political elite (Smith). Along with eunuchs,the harems of some ancient courts offer a dimension to court life whichis obviously missing from more recent European courts. This undoubtedlyreal difference might seem to limit the applicability of an Eliasque approach
toancient courts (van Ess), although even this is debatable. Harems and
eunuchs were also very much a part of the Achaemenid Persian and thelate Roman courts, which Brosius and Smith in this volume both see, albeitfrom different perspectives, as complex court societies ‘in something likethe sense Elias proposed for Versailles’ (Smith).
Looking at the chapters more closely, Brosius argues that the hierarchical
practices of the Achaemenid court were rooted in the tribal social structureof the Persian people. The quest for legitimacy and stateliness prompted thearriviste Persian monarchy to absorb and adapt the older courtly traditions ofancient Mesopotamian kingship. As the rulers of a territorially vast empire,
19Although it is now recognised that even in an early modern context Versailles was atypical in
this respect.
20Baillie 1967 :171–2.21Da Vinha 2004 .

12 tony spawforth
the Achaemenid kings evidently felt the need for a strong representation
of power and stability at the centre of the state. T o this end, ceremonial,palaces and royal image harmonised a presentation of the ruler’s person asthe divinely ordained embodiment of the imperial polity. In AchaemenidPersia, far more strongly than in the other ancient monarchies considered in
this book, one arguably has the sense of the ‘King of Lands’ as anticipatingJean-Marie Apostolid `es’ early modern vision of ‘le roi-machine’ , a ruler
whose physical body was somehow understood as the symbolic incarnationof the state.
22Asthe place where the king and his entourage resided, the
Achaemenid court seems also to have constituted the chief physical setting
for playing out the relationship between the king and the Persian nobility, theelite group which provided the dynasty with its ascriptive pool of top-level‘helpers’ , mainly as satraps and generals. This politically pivotal relationship
was continually being renegotiated through gift-giving, household ofﬁce,participation in ceremonial, and royal activities such as feasting and hunting.
Atthe other end of Iranian antiquity, Josef Wieseh ¨ofer breaks new and
difﬁcult ground by surveying the evidence for the court of the Sasanian kings.Although very fragmentary, it is sufﬁcient to show a generally well-knowncorrelation between the social prestige of the ruler’s court and the ebb and
ﬂow of monarchical authority. The relative weakness of the Sasanian kingsdown to the late ﬁfth century AD is reﬂected in the very limited integrationof the great landowners as courtiers. The monarchy became much strongerfor a time in the ﬁfth and sixth centuries AD. During this period the prestigeof the court grew correspondingly, as shown, for instance, by its control oversartorial distinctions of rank.
The borrowing of one monarchy’s court culture by another monarchy
return sa sat h e me in the next chapter ,b yT o n yS p a wforth, on Alexander the
Great. In a manner not so different from Cyrus the Great, founder of the
Achaemenid empire whom he professed to admire, the Macedonian con-
queror presided over a vast territorial expansion of a hereditary kingdom.This in turn prompted him to adopt the court culture of the former rulingpower (the Achaemenids) as a device of representation and elite-integrationin the altogether new and unprecedented circumstances in which, asMacedonian king, he now found himself. The approach of Elias permits a
muchmore realistic assessment of ceremonial at Alexander’s court as a mech-
anism for ranking and integrating the multi-ethnic elites of the empire andalso of Alexander’s measured use of magniﬁcence – especially his imposingroyaltents. It also becomes much clearer how Alexander’s courtly practices
22Forthis formulation see Apostolid `es1981 .

Introd uction 13
have been distorted by the classical (espe cially Greek) critique of monarchi-
cal splendour as barbaric which colours the surviving literary accounts ofAlexander.
Inhis chapter on the creation of an imperial court under Augustus and his
dynastic successors, Jeremy Paterson argues that the changed arrangements
of power resulting from the transition from monarchy to empire made theemergence of a court in imperial Rome more or less inevitable. He delineatesthe negotiation of new types of relationship between the Roman elite and theemperor as the former adapted their traditional quest for power, status andwealth to the new political circumstances and the latter experimented with
different approaches to monarchical representation. In economic terms theemperor’s court is seen as the centre of distribution and brokerage, wherethe emergence of a courtly discourse characterised by ﬂattery and lack offrankness was an inevitable consequence of the intense elite competition forrewards (see also Spawforth on the ‘Alexander-ﬂatterers’).
Inthe period studied by Rowland Smith ( c.AD300– c.AD450), the
available sources across a range of media consistently present the emperor’s
court as a much more formal and ritualised institution in ways which cannot
be explained away as simply a subjective matter of shifting perceptions withinthe upper-class literary milieu. This change is attributed in the main to thethird-century decline in imperial reliance on the senatorial class, and therise of a new and numerically greatly expanded ‘service aristocracy’ as a
result of the administrative reforms of Diocletian and his successors. On
the one hand the late antique emperors were freed from the dictates ofcivilitas –the pretence of equality which the old senatorial class had valued
so highly. This removed earlier constraints on the ceremonial elevation ofthe person of the emperor and enhanced the importance of the court asthe focus of imperial ceremonies. On the other, the creation in effect ofanew aristocracy enhanced the role of the court as the chief place where
social ranking was displayed and the complex hierarchy of ofﬁce and honourintegrated. Borrowings by one court culture from another – in this case theSasanians – may also have helped to give the late Roman court its distinctivetimbre, as did its militarisation (although a strong military identity cropsup in other ancient courts, e.g. Spawforth and Spence in this volume).
Powerful dynastic women and household servants were also elements of
continuity in the conﬁguration of the imperial court from early to late an-
tiquity. In the ﬁfth century, Roman empresses combined with child-emperors, eunuchs and generals to rule the empire in a way strikingly sim-ilar to the picture of the later Han Chinese emperors in the analysis of vanEss. In Han China, a sacralised, hereditary emperor as often as not reigned

14 tony spawforth
rather than ruled, especially in Later Han times. Van Ess argues that the
Later Han court survived for as long as it did because it acted as a kind of
quasi-constitutional fulcrum, serving to balance the complementary inter-ests of powerful clans related to the imperial women on the one hand, andof the imperial bureaucracy on the other. As Smith suggests for the ﬁfth-century Roman imperial court, an important role of the court in this kind of
arrangement of power may have been to maintain ‘“the set of symbolic formsexpressing the fact that [a governing elite] is in truth governing”’ (quoting
Clifford Geertz). In his description of the Han court and its physical setting,van Ess also draws attention to the story of the founding emperor of the Hanline, Han Kao-tsu, a man of humble origins, who has to be told by a supporterwhy the emperor needs to build imposing palaces: ‘Without great and elegant[buildings], you will not [be able to display] your authority and majesty.’This appears to be one of the rare items of evidence for the ancient rationali-sation of monarchical display, to be compared with the claim of Louis XIVin his M´emoires :‘The people over whom we rule, unable to see to the bottom
of things, usually judge by what they see from outside.’
23
The ﬁnal chapter ,bySpence, takes us furthest back in time, to ancient
Egypt, with its focus the later Eighteenth Dynasty ( c.1390–1295 BC). Spence
traces the origins of a court-style institution to the predynastic period. Echo-
ing Brosius on the social roots of ranking at the Achaemenid court, she seesalink between the innate stratiﬁcation of ancient Egyptian society and the
pharaonic court, a place where the ranking of the elite was put on displayand where a sacred ruler was marked off ritually from lesser humanity. TheEgyptian court generated culturally sp eciﬁc royal ceremonial, of which the
use of the Window of Appearance for rewarding courtiers, studied here bySpence, is a distinctive feature of the reign of Akhenaten. More generally,prostration formed part of the ritual of approaching the ruler in Egypt as,later, it did in Achaemenid Persia. Like Persia, Egypt probably borrowedfeatures of its court culture from the even older royal traditions of theancient Near East. It is tempting, indeed, to see something of a courtlykoine in this part of the ancient world, with shared features including sacred
monarchy (the Achaemenid ruler seems to have been a quasi-sacred ﬁg-ure, albeit not divine as such), a common language of gesture, and prac-tices such as sartorial distinctions for courtiers based on royal gifts. UnderPhilip and Alexander, this style of court culture penetrated further west-ward than it had ever done previously, and it did so again in late (Roman)antiquity.
23M´emoires II, p. 15, ed. Dreyss; Elias 1983 :117–18.

Introd uction 15
Asis becoming clear, the religious underpinnings of ancient monarchy,
along with their expression in court art and ceremonial, is another recurrent
theme of this volume. If they have not received more emphasis in the Intro-duction, this is because the advance in the scholarly understanding of thisphenomenon over the last quarter-century has been considerable, certainlywhere Greco-Roman antiquity is concerned. It is perhaps not facetious tosay that ancient historians have at last caught up with the eighteenth-centuryFrench chancellor d’Aguesseau, who took the sanctity of supreme power so
muchfor granted that he saw ‘r ´egimes de droit divin’ even in republican
Holland or Venice.
24
The Newcastle project never aimed for comprehensive coverage of the courtsof the ancient world – an impossible task between one set of covers. Thecourts of the Greco-Macedonian dynastic states which between them came
torule over the tri-continental imperial state of Alexander the Great have
received considerable scholarly attention in the past and are not accorded
separate treatment here. But two important studies published independentlyof each other at much the same time, and explicitly acknowledging Elias,must be mentioned. Gabriel Herman (Herman 1997 )used the contempo-
raryGreek historian Polybius (second century BC) as the basis for a pene-
trating study of the social formations which grew up around the Ptolemies
in Egypt and the Antigonids in Macedon and which clearly reproduce fea-tures of a complex court society. In this period the political dominance ofmonarchy within the enlarged Greek world created by Alexander’s conquestis shown by the appearance of terms for ‘court’ and ‘courtiers’ in contempo-rary Greek writers.
25Herman is mainly concerned with diagnosing the type
of the courtier in the leading members of the royal entourage, the ruler’sso-called ‘friends’ , entrepreneurial individuals, overwhelmingly Greek, whowereentirely dependent on his favour, served him in various capacities as
agoverning elite in return for material rewards, and belonged socially to
larger networks connecting the court to Greek cities within and beyondthe kingdom’s frontiers. An entourage constituted in this way in importantrespects was anticipated by the court of Alexander and, indeed, that of Philip
before him (Spawforth in this volume). Herman shows that competition forroyalfavour among the ‘friends’ was intense, and seeks to demonstrate that
these entourages generated their own rules of behaviour or ‘etiquette’ whichallow them to be identiﬁed as ‘court societies’ in the Eliasque sense.
26
24Quoted by Antoine 1989 :170.25E.g. Men.897; FGrH 81F4 4 =Athen.142b.
26Herman 1997 :204, 219, 223–4, although it is not clear that the behaviours adduced are not
sometimes simply the normal standards of civility in upper-class Greek society of the time.

16 tony spawforth
Inalong chapter which appeared in the same year, Gregor Weber covered
similar terrain, likewise basing himself on the written sources and con-
centrating on the social structure and styles of communication within the
Hellenistic courts, with recognition, but only a brief discussion, of their rep-
resentative function (Gregor 1997 ). Weber emphasises the evolution of the
forms of interaction between kings and courtiers from the ethos of equalityand informal friendship in the opportunistic atmosphere of the early Hel-lenistic courts, to the formal hierarchies of ‘friends’ ranked as ‘ﬁrst’ , ‘second’and so on which emerged in the second century BC. In this period, the Hel-lenistic kingdoms were contracting territorially and from an Aegean point ofview were no longer the lands of opportunity of earlier times. In this harsher
climate, honoriﬁc titles at court were part of the struggle to retain the loyalty –no longer guaranteed – of the leading ﬁgures serving in the royal army andadministration. This ‘frozen formalism’ (Herman 1997 :223) characterised
the Hellenistic courts which survived into the period from the later secondcentury BC, and inﬂuenced the households of leading Roman aristocrats
when they began to resemble courts in miniature, with their segregation ofcallers into ‘ﬁrst’ and ‘second’ friends.
27
The summaries offered above of the chapters which follow hardly do
justice to their depth or richness. The contributors hope that the varietyof modern approaches and ancient evidence, and the broad span of bothplaces and periods, will offer the opportunity to observe a range of ancientcourts, and stimulate debate on the usefulness of ‘the court’ as an analytical
category in the study of ancient monarchy.
27Sen.de Ben .6.34.2, cited by Gregor 1997 :71n.192. See Paterson in this volume.

1 Newout of old? Court and court ceremonies in
Achaem enid Persia
maria brosius
Introduction
The need for a systematic study of the courts of ancient Near Eastern monar-
chies, including the court of the Achaemenid empire, is only now becoming
clear.1Am´elie Kuhrt identiﬁed these courts as elements of Near Eastern
kingship and the expression of power (Kuhrt 1995 ). Briant ( 1996 ,2002 )
offered a descriptive account of the Achaemenid court, although he fellshort of providing a historical context or adopting a theoretical approachtothe court as a political institution in the sense ﬁrst deﬁned by Norbert
Elias. Most recently Wieseh ¨ofer ( in press (c) )has discussed the Achaemenid
palace and its importance for the king.
2
Elias’ analysis asked how the social position of monarch was perpetuated
over numerous generations and dynasties and over considerable time peri-
ods. He identiﬁed the court as a grouping of people who played a key part inthis phenomenon and had an immediate interest in preserving the monarch.King and court existed in a relation of interdependence, in which each used
the other constantly to reafﬁrm their position within a strict hierarchicalorder. Both the king and the court used court ceremonies and court etiquetteas vehicles for expressing this interdependence. While the king used them toemphasise his unique position and his social distance from his courtiers, thecourtiers used them to display their own position within the hierarchical
order of the court. This system led t othe creation of a self-perpetuating
‘court society’ . This chapter attempts to identify a court and court society in
Achaemenid Persia, operating on the principles of interdependence between
king and court as argued for by Norbert Elias.
Inthe case of Achaemenid Persia, it needs pointing out that to a large extent
our knowledge depends on classical, and chieﬂy Greek, sources. Archaeo-logical evidence from the Achaemenid empire sheds light on the material
1Forone such study of palace institutions in the ﬁrst millennium BC see now Nielsen ( 2001 ). As
far as I am aware, as yet there is no systematic discussion of ancient Near Eastern courts and
court societies. But see Spence and Wieseh ¨ofer in this volume; also Joann `es2004 :94–8 on the
Assyrian court.
2Iwish to thank Josef Wieseh ¨ofer for making his article available to me before publication.

18 maria brosius
manifestations of the Achaemenid court and kingship. But no Achaemenid
literature survives which would allow us to study the problem from a Persianperspective. T o an overwhelming degree the Greek sources were hostile toAchaemenid Persia. Their accounts of Achaemenid court society are often
ideologically imbued, serving to construct the lifestyle of the king, and byextension all Persians, as opulent and decadent.
3The Greek sources there-
fore need careful handling if we are to use them to identify the elements ofacourt society and to explain the workings of the Achaemenid court. This
deﬁcit in the literary sources is offset to a limited extent by the administra-tive texts from the Persepolis archive.
4These provide independent evidence
for the hierarchy of administrative ofﬁcials and their interaction with thePersian king.
Anydiscussion of an ancient court must begin with the problem of
deﬁnition. How are we to deﬁne the ‘court’ of the Achaemenids? In theintroduction to the volume Hofund Theorie ,Butz and Dannenberg ( 2004:
4) distinguish between the court as a social grouping and as an institution.Thus, ‘court’ describes on the one hand the people surrounding the king,and on the other the institutional context within which the king operates,that is, the centre of his political, administrative, judicial and military power.Afurther ambiguity of the term ‘court’ becomes apparent from the fact that
it refers to a social conﬁguration of groups of people extending beyond theking’s household – those people, that is, who permanently accompany theking, wherever he is. In order to distinguish between these different groupswho can all be considered a part of the court entity, scholarship differentiatesbetween the close or inner court and the wider or outer court (Butz andDannenberg 2004 :12). This distinction allows us to separate out those
members of the court who constitute the permanent entourage from therest.
This chapter investigates the perso nnel and the institutional character of
the Achaemenid court and the appropriateness for the Achaemenid monar-chyofthis distinction between an inner and outer court. Before doing so, we
should remember that ‘court’ refers not only to the ‘Personalverb ¨ande’ struc-
tured around the king, but also to a physical space within which ceremoniesand forms of etiquette are performed. The palaces and palace complexesof the Achaemenid kings bear witness to the importance of a designatedspace within which the court – meaning the groups of people around the
3On the problem of the Greek sources see Wieseh ¨ofer 2001 :11–12; Sancisi-Weerdenburg 1987 :
33–45; Briant 1989a :33–77.
4See Hallock 1969 ,1978 .

Court and court ceremonies in Achaemenid Persia 19
king – operated. The importance of the palace as a ‘court’ is also mirrored
in the design, size and splendour of the royal tent which accommodated thecourt on its journeys between royal residences and on campaigns (see also
Spawforth in this volume).
Historical background
When the Persians settled in eastern Elam, in the province of Persis (modern
Fars), around 1000 BC, their society was tribal. Each tribe consisted of
anumber of clans, which in turn were made up of extended families. It
can be assumed that the heads of these families formed a hierarchy amongthemselves and thus created a natural ranking order within each tribe. Familyhierarchies may have been based upon the number of generations in a familylineage; or on a family’s size, gauged by the number of children, especially
sons, on whom depended a family’s chances of extending its inﬂuence overothers; or on a family’s wealth, in terms of livestock and land. The shaping ofthese hierarchies may also have been inﬂuenced by the qualities of the headof the household as a leader, qualities revealed by success in the politicaland military sphere. Two social issues were the particular concern of thehead of a family: the protection and welfare of his family, including thebetterment of its way of life; and, beyond the family, the welfare of his tribe.Self-preservation; natural ambition within one’s social environment; a wayof life characterised by an awareness that an individual’s actions affected thewell-being of the tribe as a whole; and pr otection from outside threat: all
these factors contributed to the proﬁle of the head of a family and shaped hisambitions for himself, and for his sons and daughters. They also contributedtohis family’s standing within the tribe, in turn affecting the tribe’s standing
among other tribes.
5
Herodotus knew of nine Persian tribes, the leading Pasargadae, Mara-
phians and Maspians, the settled tribes of the Panthialaeans, the Derusiansand the Germanians, and the nomadic tribes of the Mardians, Dropiciansand Sagartians (Hdt.1.125). The tripartite division of these tribes alreadyreﬂects a ranking order, within which the Pasargadae were identiﬁed as the
most notable. They had been the leading tribe for three generations, whenin the fourth generation Cyrus II the Great (559–530 BC) emerged as theleader of the Persians and the founder of the Persian empire.
6
5Forthe importance of the family in Persian society see for example DB IV: 60–1.
6The ﬁrst Persian leader attested in the sources is T eispes ( c.650–620 BC), who was succeeded by
Cyru sI(c.620–590 BC) and Cambyses I ( c.590–559 BC) (see Cyrus Cylinder line 21).

20 maria brosius
The identiﬁcation of several Persian tribes allows the assumption that
the concept of a social ranking order was in place at the time of the emergence
of the Persians as a political power. It may thus be postulated that the ele-ments of a hierarchically structured court society in their earliest form wentback to the extended-family formation of the tribe. Protection of a group,aspiration to leadership and competition among equals affected the head ofthe family, the leader of the tribe, the leader of the dominant tribe and, on theultimate level, the king and his nobility. Within this structure a key aspect ofcourt society can be detected: namely, the group’s support for an individual
and that individual’s reciprocal dependence on the group to maintain hisstatus (cf. Elias 1983 :3 ;H e r m a n 1997 :200). The relationship between the
individual and the group of supporters evolved from the group’s supportfor the head of a tribe, then for the head’s son and successors, before movingfrom this lineage of tribal chiefs to a dynasty of kings.
The kings of the ﬁrst Persian dynasty, the Achaemenids, named after the
eponymous founder of the empire, Achaemenes, ruled their empire from560 to 330 BC. Despite succession struggles
7and internal revolts, royal
power remained with the Achaemenid dynasty throughout this time. Theycontrolled the empire through vast provinces, so-called satrapies, which
weremade up of conquered kingdoms and territories. Each province was
governed by a satrap, who was installed in ofﬁce by the king, and who wasoften a close relation of the king, especially in important satrapies suchas Bactria (roughly modern Afghanistan) and Egypt. Owing both to theirfamilial link with the king and to the importance of the ofﬁce itself, satrapsweremembers of the royal court. They represented the king at local level
and had their own palaces, satrapal parks and estates, as well as their owncourts modelled on the royal court. Their ofﬁce gave them a key role as
intermediaries between the king and his subjects (cf. Herman 1997 :200).
One of the most signiﬁcant political achievements of the Persian monarchywas the creation of the satrapal system, which proved an effective means ofgovernment for over two centuries, and was even adapted by the Macedoniansuccessors of the Achaemenids, the Seleucids.
8
External inﬂuence on the establishment of the Persian court
Persian court society was not generated solely by the internal structure of
Persian tribal society. It was also subject to external inﬂuence from the other
7Especially following the murders of Xerxes and his heir Darius in 465, of Xerxes II in 424 and of
Artaxerxes III in 338.
8On satraps and the satrapal division of the empire see Wieseh ¨ofer 2001 :59–62; Jacobs 1994 ;
Tuplin 1987 .Onthe Seleucid empire see Sherwin-White and Kuhrt 1993 .

Court and court ceremonies in Achaemenid Persia 21
Near Eastern monarchies which helped to shape Persian kingship. These
kingdoms included Elam, as well as those of Media and Urartu, Lydia, Baby-
lonia and Assyria. Since the formation and the development of a court societygo hand in hand with monarchy and the expression of kingship, it is morethan likely that the courts of these neighbouring monarchies inﬂuenced thecourt society of Achaemenid Persia. The fact that their political and cultural
inﬂuence can be traced on other levels adds weight to the idea of a Persianadaptation of their court organisations. But it is difﬁcult to assess the extentof this inﬂuence, since the evidence for their own court societies is scarce.Their inﬂuence on the development of the Persian court can be grasped onlytoalimited extent.
The Neo-Elamite kingdom gave the Persians their ﬁrst encounter with
the concept of kingship and a political entity headed by the king and hiscourt. Elam was centred on two royal capitals, Susa, west of the Zagros
mountains, and Anshan, east of the Zagros, in the province of Persis. Anancient civilisation, Elam had a history of monarchic rule dating back tothe third millennium BC.
9Owing to the scarcity of historical sources for
Elam, however, it is well-nigh impossible to grasp Elamite court life. Giventhe duration of Elamite royal rule, the existence of a court society support-ing the kingship can certainly be assumed. In Elam there is evidence foradynastic succession, for the expression of kingship in the representation
of the royal couple on reliefs and in sculpture, as also in the constructionbykings of public and religious buildings, as attested by numerous build-
ing inscriptions. Local dynasts, too, expressed their kingship in rock reliefsdepicting scenes of prayer or religious processions composed of selected(?)groups of retainers, led by the king.
10The reliefs clearly reﬂect a need to
express kingship visually for the beneﬁt of a (court) society of high-ranking
individuals close to the king.
Following the destruction of Susa in 646 BC by the Assyrian king Assur-
banipal, the Elamite kingdom came under Assyrian control until the fall ofNineveh in 612 BC. A Neo-Elamite dynasty arose brieﬂy in Susa; meanwhile
the Persians gradually established their power in the region of Anshan, even-tually annexing Susa under Cyrus II (see Vallat 1998 :311). Anshan was, if
not the actual, then certainly the ideological centre of the Persian rulers, whoclaimed to be the political heirs of the Elamite kings.
11There i sn od oubt
that the early Persian kings adapted aspects of Elamite kingship, as well as
9On the history of Elam see Potts 1999 ;V a l lat1998 ;C a r t e r and Stolper 1984 .
10The reliefs at Kul-e Farah III ( c.1100–1000 BC) are an example of the visual expression of a king
performing a (religious) ceremony in the presence of a group of people (see Potts 1999 :pl. 7.9).
11This is most apparent in the Persian adaptation of the Elamite royal title, according to which
the Persian kings were called ‘king of Anshan’ (see Cyrus Cylinder lines 20–1).

22 maria brosius
taking over administrative and possibly military practices from Elam. It is
likely, then, that the court of the Elamite kings likewise inﬂuenced the palaceorganisation of the early Persian kings.
Other inﬂuences might have come from neighbouring Media and from
Urar tu,
12though in both cases we know almost nothing about their courts.
The Medes, like the Persians, were an Iranian people. They had settled innorth-west Iran around the same time as the Persians, and soon becameapolitical and military force to be reckoned with by their Mesopotamian
and Lydian neighbours. Median political organisation appears to have been
aloose federation of kings centred on different cities.
13Out of this group-
ing a dominating dynasty emerged under Deioces of Ecbatana. Herodotuscredited him with the introduction of court etiquette and court procedures:
Itwas Deioces ﬁrst who established the rule that no-one should come into the
presence of the king, but all should be dealt with by the means of messengers; that
the king should be seen by no man; and that it should be in particular a disgracefor any to laugh or to spit in his presence. He was careful to hedge himself withall this state in order that the men of his own age (who had been bred up withhim and were nobly born as he and his equals in manly excellence), instead ofseeing him and being thereby vexed and haply moved to plot against him, might byreason of not seeing him deem him to be changed from what he had been.
(Hdt.1.99)
Even if the historicity of Herodotus’ Median logos must be regarded with
caution, the portrayal of the king’s recognition of his unique position, which
set him apart from people who had been formerly his peers, reﬂects a cred-ible historical development from tribal or military leader to kingship. Theidea that the king had to distance himself from his subjects is echoed inXenophon’s description of Cyrus’ court:
Cyrusconceived a desire to establish himself as he thought became a king, but he
decided to do so with the approval of his friends, in such a way that his public
appearances should be rare and solemn and yet excite as little jealousy as possi-ble…S oC y r u s stationed a large circle of Persian lancers about him and gave orders
that no-one should be admitted except his friends and the ofﬁcers of the Persiansand the allies.
(Xen. Cyr.7.5.37, 41)
Limited access to the king meant that the person of the king was exalted
above other members of the court. In order for the king to maintain his
12For evi dence on Urartu and its court see Kuhrt 1995 :559–60.
13On the question of the Median kingdom see Sancisi-Weerdenburg 1988 .See also Kuhrt 1995 :
653–6; Rollinger 2003 .

Court and court ceremonies in Achaemenid Persia 23
extraordinary position, he could not seem to be one of the nobles: he had to
be a singular ﬁgure, removed from the nobility. Those who were admitted to
the royal presence were privileged above all others. This access to the king setthem apart from the majority who were denied it. Thus a noble’s closenesstothe king meant closeness to power, which in turn affected his standing
among his peers. For most, access to the king was not direct: anyone seekingan audience had to be announced and accompanied by an intermediary.This was the royal messenger, or staff-bearer, who was one of the mostimportant ofﬁcials at the court, holding a key position between the kingand his subjects.
14
The Babylonian and Assyrian courts
Abuilding text of Nebuchadnezzar II (604–562 BC) gives us a glimpse of
Babylonian court society. The text lists a number of functionaries in theroyalhousehold (Bab. maˇsennu ), including the chief baker, the chief of
thek¯asiru ,the superintendent of the palace, the major-domo, the maˇsennu
of the House of the Palace Women, the scribe of the House of the PalaceWomen, the master of ceremonies(?), the chief barber(?),
15the overseer of
slave-women, the cup-bearer, the chief singer, the secretary of the crownprince, the chief of cattle, the chief of boat-men, and the chief of the king’smerchants. These are followed by a list of names of the ‘great ones’ , i.e. thegovernors, then by a list of ofﬁcials, and ﬁnally by a list of the royal cityofﬁcials including the city kings of the Levantine coast ( ANET 307–8; cf.
Kuhrt 1995 :605–7).
The social groups revealed by the text in turn point to the existence of a
court hierarchy. First come the ofﬁcials immediately serving the king; then
the ofﬁcials responsible for the women’s quarters in the palace. All theseattendants operated in the immediate vicinity of the king. Distinct fromthem are the holders of political ofﬁce – the governors (Bab. ˇsaknu and
p¯ıh¯atu)inthe land of Akkad, who are referred to as ‘the great ones’ . They in
turn are placed at the top of the administrative structure, above ofﬁcials ofsmaller districts and governors of the cities and towns. Their status madethem members of the court, but, unlike the ﬁrst group, they belonged to theouter court, i.e. to the grouping of courtiers who were not constantly in theroyalvicinity.
The principal elements of kingship and court organisation are already
apparent in the court of the Assyrian empire. The Assyrian king resided in
14On the royal staff-bearer see Lewis 1977 :16; Briant 2002 :259.
15Or ‘chief engineer’ , or ‘chief of couriers’ .

24 maria brosius
the palace complexes of the royal capitals. Royal inscriptions attest to his
concern for religious buildings as well as the beautiﬁcation of the city. Palace
reliefs demonstrate his prowess in war, his military excellence and his skill
with weapons, both on horseback and in a chariot. The king was depicted
honing his military skills as a successful hunter of wild beasts, including lionsand panthers. Dynastic concerns found expression in the king’s nominationof an heir from among his sons; also in the staging of his funeral – the heir’sﬁrst duty was to observe his father’s funerary rites.
The existence of an order of rank amongst members of the Assyrian court
emerges in a text in which the loyalty of the courtiers is being questioned:
[Will any of the] ‘eunuchs’ (and) the bearded (ofﬁcials), the king’s entourage, or
(any) of his brothers or uncles
[his kin], his fa[ther’s line], or junior members of the royal line, or the ‘third men’ ,
chariot drivers (and) chariot ﬁghters,
[or the recruitment ofﬁcers, or] the prefects of the exempt military, or the prefects
of the cavalry, or the royal bodyguard, or his personal guard,
[or the keepers] of the inner gates or the keepers of the outer gate, or th e…‘ e u n u c h s ’ ,
[or . . .], or the palace superintendents, the staff-bearers (and) the wa(tch)men, or
the mounted(?) scouts (and) the trackers,
[or the lackeys, tailor]s, cup-bearers, cooks, confectioners, the entire body of crafts-
men,
or the Itu’eans and the Elamites, the mounted bowmen (?), the Hittites (or) the
Gurreans, or the Aramaeans, [or the Cimmerians, o]r the Philistines, or the
Nubians (and) the Egyptians or the Shabuqeans, [or the ‘eunuchs’ who b]ear
[arms], or the bearded (ofﬁcials) who bear arms and stand guard for the king.
(SAA 4no.142; Kuhrt 1995 :529–30)
AsKuhrt has pointed out, the high-ranking ofﬁcials operating around the
king enjoyed considerable wealth and social standing at court (cf. Kuhrt1995 :531; Joann `es2004 :96). They were given freehold and tax-exempt
estates, and are attested as landowners across the empire.
16Apart from the
household staff, there is also evidence for the presence at court of scholarlyadvisors who counselled the king on matters of science, religious ritual andastronomy.
Asin Babylonia, a household for royal women also existed at the Assyr-
ian court. The kings were polygamous and there is slight evidence that atleast some royal women descended from the Assyrian nobility. Presumablythere was some kind of ranking amongst them, with the highest rank prob-ably assigned to the mother of the designated heir. Beyond that, their family
16See Kuhrt 1995 :531 for the example of the chief reinholder of Assurbanipal.

Court and court ceremonies in Achaemenid Persia 25
background may well have had a bearing on their standing within the group.
Animportant observation can be made regarding the status of these royal
women, which in turn provides a clue as to the palace organisation. Assyrian
queens owned estates and controlled large households. They had residencesin the royal capitals (cf. Kuhrt 1995 :526–7; Joann `es2004 :96). At the head
of the queen’s household was the ˇsakintu ,afemale ofﬁcial who herself con-
trolled considerable wealth, and her deputy. Other members included a
female scribe of the queen’s household, a cook, and a male and female con-fectioner, as well as a woman referred to as sekretu ,‘enclosed woman of the
house’ (Akk. MUNUS.ERIM. ´E.GAL). This was either a term used for women
of the palace other than the king’s wives, or a reference to female servants.
Asall the evidence shows, a court society was well established in
Mesopotamia and northern Iran at the time of the Achaemenids’ rise to
power. Even though there is no clear proof that the Persian king modelledhis court on the Assyrian and Babylonian examples, the extant sources allowus to note different components within the court societies of Mesopotamiaand northern Iran. The following discussion will explore how far these samecomponents, including the different levels of personal attendants, a hierar-
chyofofﬁcials holding political and administrative ofﬁce in the vicinity
of the king, a court of the royal women, and a wider royal court, can beidentiﬁed in the organisation of the court of the Achaemenids.
The Achaemenid court
Achaemenid royal inscriptions give a term for the palace, OP vi /p113-(Bab.
bitu-/ Elam. ulhi)(cf. Wieseh ¨ofer, in press ). The term is ambiguous, as
Wieseh ¨ofer notes: ‘When Darius asks the god [ Ahuramazda ,M .B . ]t o protect
the Persians and his ‘house’ he is, of course, not referring to vi/p113-asabuilding,
but to the people living and working in the royal household’ (Wieseh ¨ofer,
in press ;cf. DB I: 61–71). In other inscriptions, the buildings of Persepolis
are refer red to as halmarraˇ s(DPf), with individual palaces more speciﬁcally
identiﬁed as ta¸cara (‘suite of rooms’) (DSd) and hadiˇ s(DSj) ‘seat of power’
(cf. Wieseh ¨ofer, in press ).17InGreek terminology we ﬁnd terms such as ta
basileia (e.g. Hdt.1.30, 98, 178, 181) and ta oik ¯emata (Ctesias, FGrH 688 F
9.13). These terms are equally ambiguous, denoting both a building complexand also the people operating within it.
17Wieseh ¨ofer points out that the Babylonian term used for these buildings is bitu-rather than
ekallu .

26 maria brosius
Inthe ﬁrst instance the king’s court was deﬁned by a group of attendants
and ofﬁcials whose duties kept them permanently in the immediate vicinity
of the king. T ogether with members of the king’s immediate family theyformed the inner court. The king’s attendants included his personal ser-vants, ointment bearers, the cup-bearer, the king’s parasol-bearer, the royalcharioteer and the stool-bearer. Members of this group took care of the
king’s daily needs, prepared and served his meals, and were in attendancewherever the king happened to be – not just within the palace complex,but during his migrations between royal capitals and on campaign. Somewereeunuchs, and Greek sources emphasise that eunuchs held particular
positions of trust at the court. According to Xenophon Cyrus II appointedeunuchs as his personal attendants because of their loyalty;
18and Xerxes I
was said to have entrusted the safe return of his sons to his ‘most honoured’eunuch Hermotimus (Hdt.8.104).
Among the king’s personal attendants depicted in the palace doorways
of Persepolis we ﬁnd parasol bearers and attendants variously holding theﬂy-whisk above the king’s head and carrying perfume, ointment bottlesand towels. The very fact of their depiction in these prominent locationssuggests the importance attached to the presentation of the king withinhis own court. These courtly scenes showed the king in his own palace,‘performing’ ceremonies connected with court life and court ritual.
Other attendants were responsible for ensuring the king’s comfort on
foot, on horseback or in the royal chariot. Their depiction on the reliefs ofthe staircases leading up to the Throne-Hall shows that the king regardedthem and their functions as an expression of his kingship.
The attendants carried carpets and rugs, ready to place before a king
whose foot was never supposed to touch the ground:
Through their ( the bodyguards’ )court also the king would go on foot, Sardis carpets,
on which no-one else but the king ever walked, having been spread on the ground.
Andwhen he reached the last court he would mount his chariot, or sometimes his
horse; but he was never seen on foot outside the palace. (Athen.12.514c)
The king’s stool-bearer ensured that the king never mounted or descended
from his chariot without the aid of a stool, which in turn always stood on a
18‘Asfor servants, he considered the most trusted as well as the most loyal men to be eunuchs: he
selected eunuchs for every post of personal service to him, from the door-keepers up’
(Xen. Cyr.7.5.65). Ctesias seems to conﬁrm Xenophon’s assertion that eunuchs were among the
courtiers close to the king. They enjoyed the king’s trust, and were given vital tasks, such as
supervising the return of the king’s body to Persepolis (Ctesias, FGrH 688 F 13). On eunuchs at
the Persian court see Briant 2002 :268–77; in the (late) Roman and Han Chinese courts: Smith
and van Ess in this volume.

Court and court ceremonies in Achaemenid Persia 27
special rug. According to Dinon:
whenever the king descended from his chariot . . . he never leaped down, although the
distance to the ground was small, nor did he lean on anyone’s arms; rather, a goldenstool was always set in place for him, and he descended by stepping on this; and theking’s stool-bearer attended him for this purpose.
(Dinon ap. Athen.12.514)
Given their frequent and close access to the king, complete trust and loy-
alty was required from these servants. T o ensure this, they were recruitedchieﬂy from among the members of the Persian nobility.
19Non-Persians
could also serve in these posts, as shown by the well-known example of
the biblical Nehemiah, who served as Artaxerxes’ cup-bearer. Eunuchs toowerefrequently recruited from a non-Persian background (cf. Hdt.3.92, 97;
Athen.12.514d). The allocation of these seemingly servile tasks was prob-
ably done in such a way as to create a hierarchy among the Persians andnon-Persians involved.
20
Atthe second level up were those Persians who held a court ofﬁce which
gave them immediate access to the king. These people included the king’sspear-bearer, his bow- and axe-bearer, the heads of the king’s bodyguard(hazarapatiˇ s), palace administration and royal treasury, the chief scribe, the
keeper of the gate, and the priest(s), along with the Persian nobles serving as
the King’s Councillors, as Royal Judges, and as the King’s Eye. In a separatecategory were the royal physicians, required to be permanently in attendanceon the king and his family, but who were of non-Persian origin.
The king’s spear-bearer and the axe- and bow-bearer occupy a prominent
position behind the king in the reliefs at Bisitun and Naqsh-i Rustam, as wellas on the audience relief from Persepolis. These positions were held by high-ranking Persian nobles who enjoyed the king’s trust. In the Naqsh-i Rustam
relief they are identiﬁed as ‘Gobryas, the Patischoraean, spear-bearer of
Darius the king’ (DNc), and ‘Aspathines, the bow-bearer, (who) holds theking’s axe’ (DNd).
21Both were members of the Persian nobility, and Gobryas
19See for example Patiramphes son of Otanes, who was Xerxes’ chariot driver (Hdt.7.40); see
Briant 2002 :310.
20Itis hard to know whether to classify the attendants of the king as courtiers or merely as
members of the household staff. While they may not have been courtiers in the sense of holdingan ofﬁcial position of authority and being able to take part in the decision-making process, thefact that they were of noble descent must mark them as members of the court. Historically,personal service by nobles is a hallmark of the courtier; cf. the Macedonian court under Philipand Alexander (Spawforth in this volume).
21Aspathines is thought to have succeeded Ardumani ˇsinofﬁce soon after Darius’ accession to the
throne. It would be interesting to know whether Gobryas and Ardumani ˇsheld their positions
as the king’s spear-bearer and bow-bearer/axe-bearer at the time of Darius’ accession, andshould therefore be identiﬁed on the Bisitun relief. Darius himself had been a spear-bearer ofCambyses II in Egypt (Hdt.3.139.2).

Figure 1.1 Staircase of the Apadana (after F. Krefter)

Court and court ceremonies in Achaemenid Persia 29
had been a fellow-conspirator in Darius’ coup in 522 (see below). The haza-
rapatiˇ swas the head of the king’s personal bodyguard, the One Thousand,
also known as the Apple-bearers after the golden apples on the butts of their
dress-spears.22This elite guard can probably be identiﬁed with the rows of
spear-bearing soldiers on the reliefs decorating the staircase of the Throne-Hall in Persepolis (Fig. 1.1). They formed a part of the 10,000 Immortals,
household troops forming a constant presence around the king.
23According
toHeracleides of Cumae, a court in the palace complex was named after the
One Thousand as the ‘Court of the Apple-bearers’ . The same writer saysthat these troops were all Persian by birth (Heracleides, FGrH 689F2 ;a p .
Athen.514b). Pseudo-Aristotle sums up the appearance of these courtiers
and bodyguards in the Achaemenid palace as follows:
The king himself, they say, lived in Susa or Ecbatana, invisible to all, in a marvellous
palace ( basileion oikon )with a surrounding wall ﬂashing with gold, electrum and
ivory; it had a succession of many gate-towers, and the gateways, separated by
many stades from one another, were fortiﬁed with brazened doors and high walls;outside these the leaders and most eminent men were drawn up in order, some aspersonal bodyguards ( doryphoroi )and attendants ( therapontes )tothe king himself,
some as guardians of each outer wall, called Guards ( pyloroi )and Listening-Watch
(¯otakoustai ), so that the king himself, who had the name of Master and God ( despot ¯es
kai theos )might see everything and hear everything.
(Ps.-Aristotle, de Mundo 398a; cf. Wieseh ¨ofer, in press )
While the hazarapatiˇ sheaded the military presence at court, there was also
ahead of the administration. Under Darius I this ofﬁcial was Parnaka son
of Arsames, an Achaemenid and the uncle of the king (cf. Lewis 1977 :7–8).
Parnaka oversaw the administration of Persepolis and its province, Persis.24
Hisposition at court gave him direct access to the king and he received orders
from the king in person relating to the distribution of foodstuffs from the
royalstorerooms. On Parnaka’s authority, the king’s orders were recorded
in writing before being carried out. A typical letter-order would begin withthe formula ‘T ell PN, Parnaka spoke as follows’ , followed by a reference to
22Rendered chiliarch in Greek, the term was once thought to designate a second-in-command,
but, as D. M. Lewis has demonstrated, the term describes a military ofﬁce (Lewis 1977 :17–20;
cf. Hdt.7.41.1).
23Cf. Xen. Cyr.7.5.41. Cyrus II is credited with the creation of the 10,000 Immortals
(Xen. Cyr.7.5.66–8).
24Neither Parnaka nor Zi ˇsˇsawi ˇsisknown to have borne a title. The idea that Parnaka may have
been the satrap of Persis or the ‘grand vizier’ of Persepolis must be rejected (see Lewis 1977 :
8–9). As the Persepolis texts reveal, Parnaka was permanently based at Persepolis, and so must
be regarded as the chief administrator of the royal capital. On the subordinates of both Parnakaand Zi ˇsˇsawi ˇss e eL e w i s 1977 :10–12.

30 maria brosius
the king having given the order (see below, p. 34). Parnaka’s command
was then sealed with his personal seal. The ofﬁcials addressed by Parnakawerethe chiefs of departments heading different sections of the administra-
tion: the chiefs of the workers (Elam. kurdabattiˇ s), of the cattle (Elam.
kasabattiˇ s)and of the wine (Elam. W.GEˇSTIN. kutira ). The high status of the
head of administration was reﬂected in the amount of daily rations he wasentitled to receive from the royal storehouse: 90 quarts of wine, 180 quartsof ﬂour, and 2 sheep.
25
Among his subordinates Zi ˇsˇsawi ˇscan be identiﬁed as Parnaka’s chief
assistant. He carried out similar administrative tasks, including the issuingand recording of orders to other ofﬁcials.
26Other ofﬁcials serving under
Parnaka were the palace scribes. They wrote in Elamite, Babylonian and
Aramaic, and themselves were probably headed by a chief scribe.27The
royalmessengers needed the king’s permission (Elam. halmi )b efore they
could travel the empire in his name; probably only a few had direct access
tothe king. Besides Parnaka, there was a chief treasurer (Elam. kapnuˇ skira ),
operating the ﬁnancial section of the palace administration, who also mayhave had access to the king.
King’s Councillors, Royal Judges and the King’s Eye
Members of both the royal family and the nobility acted as political advisors
tothe king. They formed an unspeciﬁed group of Persians of unknown size;
nor do we know when and how often they advised the king. Its members wereprobably not treated as a homogeneous group, but were ranked individuallyaccording to the importance of their services to the king (Lewis 1977 :22).
Royal Judges were consulted in legal matters.
28Their number, it seems, was
uneven, possibly to ensure a majority vote in a judgement (Diod.15.10.1).The Royal Judges held permanent ofﬁce at the court. Their level of juris-diction was different from that of the legal personnel of the satrapies andsub-districts, where local judges, bailiffs and magistrates dealt with everydaylegal affairs. Permanently on call, the Royal Judges, like the Royal Council-lors, belonged to the inner court of ofﬁcials in the king’s immediate service.
Ashis title suggests, the King’s Eye was charged with intelligence-
gathering, reporting to the king. No doubt there were many informantsin the king’s service. They probably reported to the chief informant, the
25See PF 665, 662, 668.
26On the identiﬁcation of Zi ˇsˇsawi ˇswith Tithaios (Hdt.7.88.1) see Lewis 1977 :592.
27Plut. Al.18.8 refers to a ‘chief of correspondence’; see Briant 2002 :258.
28Notealso the king’s consultation of the admirals of his ﬂeet before Salamis (Hdt.8.67).

Court and court ceremonies in Achaemenid Persia 31
King’s Eye, who uniquely had access to the king.29Their task was to gather
information on any signs of seditious activity in the empire which could
jeopardise the Persian peace, whether by individuals or groups.
The most important group in the inner court was the king’s immediate
family. This included the king’s mother and his wife or wives, the heir tothe throne, other royal sons and daughters, and those royal siblings whoresided at the court. The king’s mother and wife and the heir to the throne
held the highest rank at court. They were allowed to dine with the king(cf. Plut. Art.5.3) and to be present during royal audiences. Owing to their
familial closeness to the king, the royal women were able to act as mediatorsbetween the king and members of the nobility (see Brosius 2002 :116–19).
30
The audience relief from Persepolis shows the heir to the throne taking up
aposition immediately behind the king’s throne (Fig. 1.2). The relief offers
aspatial expression of the position of the heir as second-in-command. That
such a position formally existed is suggested by an inscription from Persepo-lis, in which Xerxes, the crown-prince, describes his rank as ‘the greatest afterthe king’ . As he puts it: ‘Other sons of Darius there were, but it was the desireof Ahuramazda that my father Darius made me the greatest after himself’(XPf §4). A similar position could be awarded to the king’s brother.Ariamenes, a brother of Xerxes, was discouraged from his plan to challenge
his brother’s succession with the promise of lavish gifts and the position of‘the greatest at court’ should Xerxes become king (Plut. Mor.173B–C).
31
Ac o u r tf o rr o y a lw o m e n ?
Amore difﬁcult question to answer is whether the women of the royal family
and the nobility formed a court in their own right. As well as female royaltyand noblewomen, this grouping would have included their attendants; royalconcubines; and administrative personnel. As noted above, an institution
called the ‘House of the Palace Women’ is attested at the courts of the Assyr-ian and the Babylonian kings. Was this tradition continued at the Persiancourt? Comparison of the status of Persian royal women with their Assyr-
ian and Babylonian counterparts is certainly justiﬁed. As in Mesopotamia,the mother of the king (Bab. ummi ˇ sarri ,Elam.
∗sunki ammari )and the
king’s wife (Bab. aˇsˇsat ˇsarri ;Elam. sunki irtiri )held the highest positions at
court, followed by other royal wives, their daughters and the concubines.
29On the King’s Eye see Wieseh ¨ofer 2001 :62; 266.
30Forthe importance of Assyrian women in court politics see Brosius 2002 :105–19.
31See also Sancisi-Weerdenburg’s interpretation of Masistes as the ‘greatest after the king’
(Sancisi-Weerdenburg 1980 :48ff; 122ff).

Figure 1.2 The audience relief, Persepolis (photo: M. Brosius)

Court and court ceremonies in Achaemenid Persia 33
They were part of the royal entourage accompanying the king on his travels
tothe royal capitals and on campaigns. Women were part of Xerxes’ army
(Hdt.9.76.1) and accompanied the train of Darius III (Arr. An.2.11.9–10;
Curt.3.8.12; Plut. Al.24.1).32When they travelled with the king, separate tents
wereput up for them in a designated area of the camp (see Spawforth in this
volume). Like the royal women in Mesopotamia, Achaemenid royal women
controlled considerable wealth. As the Near Eastern and Greek sources cer-
tify, they owned estates, land and workshops across the empire, and con-trolled their own workforce. Administrators acted on their behalf, and their
personal seals show that royal women were able to give orders in their ownright.
33
So far no structure has been identiﬁed at Persepolis which could have
served as the women’s quarters. This makes it difﬁcult to give a ﬁrm answertothis question of a separate court for the women.
34Itis certainly wrong
tospeak of a ‘harem’ – a term implying that the women were secluded and
their freedom of movement restricted. But some kind of designated spacefor the royal women, comparable to the Assyrian expression ‘House of thePalace Women’ , seems likely. Greek sources frequently refer to the women’s
quarters at the Persian court. It is possible that an appropriate space was setaside for formal appearances by the royal women, just like the king’s morepublic rooms, where visitors were received and banquets held. Literary andarchaeological evidence shows that royal women had the right both to attendthe king’s audiences and to hold audiences of their own. Seals dated to theAchaemenid period show audience scenes in which high-ranking women
receiveafemale visitor.
35There can be no doubt as to their typological
association with the royal audience scene from the Persepolis reliefs. TheBible shows that the king’s wife could attend royal audiences (Neh.2:6), and
32See also Brosius 2002 :87–91.
33Twoofﬁcials are known to have acted as administrators for Irdabama, a princess: Ra ˇsda and
Uˇstana were responsible for distributing rations to her workforce at Shiraz (see Brosius 2002 :
134). Both Irdabama and Irta ˇsduna/Artystone are known to have used their own seals (PFS 51
and PFS 38 respectively).
34See Root 1979 :101–3. The identiﬁcation of one building on the royal terrace as a ‘harem’ is
tenuous, and has to be discarded completely if we accept the theory that the royal terrace was in
fact not used to accommodate the court (see below pp. 44, 47).
35Aseal from Persepolis, PFS 77∗(Garrison and Root (forthcoming)), crafted in Neo-Elamite
style, depicts a woman with a bobbed hairstyle seated on a throne. She is attended by a female
servant holding a ﬂy-whisk. In front of her, standing before an incense burner, is a femalevisitor, dressed and coiffed similarly. The audience scene on a chalcedony cylinder seal depicts a
woman seated on a throne wearing a Persian dress and a mural crown, a veil hanging from it at
the back. Her feet rest on a footstool. She is approached by a female attendant offering a birdand a lotus ﬂower – perhaps gifts brought by the Persian woman standing behind theattendant. This woman’s crown, veil and a Persian dress similar to the seated ﬁgure identify herhigh rank (see Spycket 1980 :ﬁ g . 7).

34 maria brosius
also mentions separate feasts celebrated by the women of the court (Esther
1:9–12).36The Persepolis Fortiﬁcation texts record the distribution of wine
and foodstuffs for the king’s wife by order of the king, showing that the kingfooted the bill for the queen’s entertainments. The quantities involved –some 2,000 litres of wine and 100 sheep – indicate occasions involvingseveral hundred guests:
Tell Y amak ˇsedda the wine-carrier, Parnaka spoke as follows: ‘2,000 quarts of wine
are to be issued to Artystone (Elam. Irtaˇsduna )thedukˇsiˇs(princess). It was ordered
bythe king.’ Month 1, year 19. Ansukka wrote the text. Maraza communicated the
message. (Fort.1795)
Tell Harrena the cattle chief, Parnaka spoke as follows: ‘Darius the king ordered me,
saying: “100 sheep from my estate are to be issued to Artystone the dukˇsiˇs.”’ And
now Parnaka says: ‘As the king ordered me, so I am ordering you. Now you are to
issue 100 sheep to Artystone the dukˇsiˇs,aswas ordered by the king.’ Month 1, year
19. Ansukka wrote the text. Maraza communicated the message. (Fort.6764)
The court ranking of royal women was determined by degrees of kinship
tothe king. Persian noblewomen at court were ranked by the standing of
their male relations, unless they had married into the royal family.
Royal physicians
The king employed the most eminent physicians from Egypt and Greece.They, too, were members of the court. One of the earliest known was theEgyptian Udjahorresnet, who attended Cambyses II:
HisMajesty assigned to me the ofﬁce of chief physician. He caused me to be beside
him as a companion administrator of the palace. I made his royal titulary, his name
being the King of Upper Egypt and Lower Egypt Mesuti-Re ( Son of Re ).
(Brosius 2000 :n o . 20)
Greek physicians such as Democedes of Croton, Apollonides of Cos and –
according to his own testimony – Ctesias of Cnidus lived at the court and
attended the king and members of the royal family. Greek historians creditthem with close and frequent access to members of the royal family, and they
enjoyed a high status at court. For example, Democedes received rich giftsafter successfully treating Darius I, including a house, and was invited tojoin the King’s Table (Hdt.3.132.1). Quite clearly, non-Persians whose skills
36The depiction of a funerary banquet on a stele from Dascyleium, showing a female in Persian
style dress seated on a stool next to a man reclining on a couch, both holding cups, is evidencefor the continuation of a motif known from the Assyrian court (see Boardman 2000 :ﬁg.5.61;
cf. Winter 1986 ): Assurbanipal reclines on a couch, with his queen enthroned beside him, both
holding drinking cups (Strommenger 1962 :pl. 241).

Court and court ceremonies in Achaemenid Persia 35
beneﬁted the king were granted the same privileges as the Persian nobles.
Apart from foreign physicians, the king also made substantial gifts to Greek
political exiles, among them the Spartan king Demaratus (Hdt.6.70), theAthenian tyrant Hippias (Hdt.5.96) and the Athenian statesman Themisto-
cles (Thuc.1.138.5). That said, their non-Persian ethnicity may have rankedthese Greeks rather low in the hierarchy of the court.
The outer court
The outer court included individuals and groups of people who belongedtothe court on either a permanent or a temporary basis, but who did not
normally operate in close physical proximity to the king. These courtiers canbe separated into several different groups. First, there were the membersof the extended royal family: that is, royal kin holding high ofﬁce in thesatrapies, along with their families. They resided in the satrapal centres, butjoined the king when ofﬁcial occasions required their presence at the royalcourt. Both their descent and their high rank gave them access to the king.
Satraps who were members of the royal family held a higher rank within thearistocracy than those who were merely members of the nobility, thoughthere will have been distinctions even among these last – some nobles, forexample, could claim closer afﬁnity with the king than others, thanks to a
marriage alliance with a member of the royal family.
37The hierarchy among
the satraps also had a bearing on their position at the King’s Table, and inthe king’s entourage.
Among the most important members of the outer court were the satraps,
governors of the lands of the empire. They could be members of the imme-diate and extended family of the king: brothers, sons, illegitimate sons (i.e.Darius son of Artaxerxes I, who was satrap of Hyrcania before succeedingtothe throne as Darius II), or brothers- and sons-in-law; or members of
the nobility with no known family tie to the royal house. Appointmenttohigh ofﬁce was intended to give employment to close members of the
royalfamily, and also to share power with Persian nobles linked to the royal
court by marriage. In this way a complex network was established which
helped to create a court society. If a hierarchical order existed amongst thesatraps, it was probably based on family ties to the king. This, of course,was subject to change: that is, when the king decided to honour a satrapwith the marriage of one of his daughters, as is attested in the case of
37Formsofgreeting conveyed differences of rank. According to Herodotus, Persian peers kissed
each other on the mouth, Persians of different ranks kissed each other’s cheeks, while a much
greater social disparity required the one to bow before the other (Gr. proskynesis )(Hdt.1.134.1).
Cf. Spawforth in this volume.

36 maria brosius
the satraps Pharnabazus (Plut. Art.27.7), Tissaphernes (Diod.14.26.4) and
Orontas (Xen. An.2.4.8; 3.4.31).
Satraps represented the king in the royal provinces. They had to execute
the king’s law, ensure the collection of royal taxes, administer the province
and levy troops at the king’s request. They were also responsible for thecelebration of royal ceremonies at local level and the observation of the royal
religious cult of Ahuramazda, as well as of other Persian gods. Major political
and military ventures were subject to the king’s approval. Satraps who hadsuccessfully fought the king’s cause could expect rewards of various kinds.
Local ofﬁcials and dynasts of the cities under Persian control were not
able to approach the Persian king directly but had to go through the satraps.One of the best examples of the satrap’s role as intermediary is the case ofArtaphernes, satrap of Sardis, who negotiated the Persian campaign against
Naxosonbehalf of Aristagoras of Miletus (Hdt.5.31–2). After Darius had
approved the campaign, Artaphernes assembled a naval force of 200 triremesunder the command of his cousin Megabates, while Aristagoras, the ruler ofMiletus, levied the infantry (Hdt.5.33.1). Other cases highlight the satrap’s
authority to act in lieu of the king. For example, when Sparta and Athensbegan investigating the possibility of Persian support to resolve the Pelo-ponnesian war, it was the satraps of Lydia and Phrygia who negotiated thepossible alliance with one or the other;
38and when Evagoras of Salamis
rebelled against Persian rule, the king ordered the satraps of those provinces
closest to Cyprus to deal with the matter (cf. Diod.15.2.2). As this policyreveals, the king delegated political and military issues to the satraps. On his
command, and with his consent, they were required to quash rebellions at
local level.
39
The second group of the outer court included the Persian nobility, deﬁned
as the families of the six nobles who had joined Darius I in 522 BC againstthe usurper Bardiya/Gaumata (DB IV: 80–6). According to Herodotus,these nobles were given the privilege of direct access to the king’s cham-bers (Hdt.3.84). In principle the Persian nobles were permanent membersof the court, though how much time they spent in the king’s immediatevicinity must have depended on what ofﬁce they held within the imperial
38Foranarrative of the negotiations between Sparta and the Persian satraps between 412 and 411
see Lewis 1977 :88–107.
39Only in exceptional cases did the king take part personally in military campaigns. After Darius’
punitive campaign against Athens ended in failur e, Xerxes personally led his army against
Athens. After several failed attempts to recover Egypt for the empire after the second rebellion
of 400, a campaign led by Artaxerxes III returned Egypt to Persian control in 343/2. Finally, ittook the defeat of the Granicus and the dawning recognition of Alexander’s military power
before Darius III led his army in person at Issus and Gaugamela.

Court and court ceremonies in Achaemenid Persia 37
and satrapal organisation.40Butnoble rank did not of itself secure the Per-
sian noble’s position at court. On the contrary, this position was always
subject to the king’s scrutiny. Behaviour judged improper by the king couldlead to punishment and loss of status; rebellion resulted in exile or deathfor the noble himself as well as other male members of his family. Thiscreated a complex hierarchy among the nobility, in which the standing ofindividual nobles could shift to reﬂect particular deeds of loyalty, or theextension of royal favours and privileges. For example, among the six nobles
around Darius I, Otanes and his family were honoured with special giftsand a lifelong position of privilege which placed them above all other nobles(Hdt.3.83.3).
Inthe Bisitun Inscription Darius lists these nobles in the following order:
Intaphernes, Otanes, Gobryas, Hydarnes, Megabyxus and Ardumanish
41
(DB IV: 80–6). Intaphernes, the ﬁrst Persian named in Darius’ list, hadbeen appointed commander of an army to quash the rebellion of Arakha inBabylon (DB III: 76–92). He was one of Darius’ trusted ofﬁcials, expressedin the Old Persian term bandaka which may be translated as ‘bondsmen’
or ‘dependants’ .
42According to Herodotus, Intaphernes fell out of favour
at court and was suspected of rebellion when he made use of his right toenter the king’s chamber directly: ‘He desired to enter the palace and speakwith the king; for this was the law, that the rebels should come into the
king’s presence without announcement given, if the king were not with oneof his wives’ (Hdt.3.118.2). His entry was regarded as a breach of trust andIntaphernes was charged with treason. He and the male members of his
family were punished with death.
Gobryas, Darius’ spear-bearer, held a high rank at court, judging from
his daily ration of 100 quarts of wine (PF 688), which places him above
40Speaking through Cyrus’ general Chrysantas, Xenophon describes the role of the nobility at the
time of Cyrus II as follows: ‘They [ the Persian nobles ]passed a resolution that the nobles should
always be in attendance at court, and be in readiness for whatever service Cyrus wished until heshould dismiss them. And as they then resolved so even unto this day those who are the subjectsof the great king in Asia continue to do – they are constantly in attendance at the court of theirprinces…A c c o r dingly, the nobles came to Cyrus’ court with their horses and their spears for
so it had been decreed by the best of those who with him had made the conquest of thekingdom’ (Xen. Cyr.8.1.6–8).
41Ardumani ˇswould soon be replaced by Aspathines, who appears on Darius’ relief at Naqsh-i
Rustam. Herodotus names him as one of the conspirators of 522. His order of nobles differs
from the one in DB. According to Herodotus (3.70), Otanes ﬁrst recruited Aspathines andGobryas, and then Intaphernes, while Gobryas roped in Megabyxus, and Aspathines Hydarnes.
42See Wieseh ¨ofer 2001 :34. The bandaka included loyal ‘king’s men’ as well as mere subjects of
the king. Briant suggests that those loyal to the king were initiated into the circle by an ofﬁcialceremony in which the candidate swore an oath of allegiance to the king (see Briant 2002 :325).

38 maria brosius
Parnaka.43According to Herodotus (7.2.2) Darius had married a daughter
of Gobryas (she remains unnamed) before he became king, while Gobryas
himself was given a sister of Darius in marriage (Hdt.7.5.1). Their son,Mardonius, married Darius’ daughter Artazostre (cf. PFa 5, Hdt.6.43.1).
Thus we ﬁnd Gobryas’ family linked to the king in a triple alliance. Thiswas part of a deliberate royal policy of establishing family alliances. Themarriages served to perpetuate royal links with this noble family beyond theﬁrst generation. Gobryas held high ofﬁce under Darius, while Mardoniuswas to become a commander in the Persian army, serving in Ionia after therebellion of 499/8–493/2. Their case provides a splendid example of the way
in which the nobility was bound to the king and thereby to the royal court.
The Gobryas family was by no means unique. In a similar fashion Darius
established close ties with Otanes, who may himself have been the son of theAchaemenid Pharnaspes. Darius married a daughter of Otanes, Phaidyme,
when he succeeded to the throne (Hdt.3.88.4), while a sister of Darius wasmarried to Otanes himself (Hdt.7.82). A daughter, Amestris, was to becomethe wife of Darius’ son and heir Xerxes (Hdt.7.61.2). Like Gobryas and hisfamily, the house of the Persian noble Otanes was inextricably linked to theroyalfamily. His own status as the father of the wife of the heir to the throne
will have put him among the leading grandees at court.
44
Whether or not similar marriage alliances were concluded with the nobles
Megabyxus and Hydarnes, they certainly remained closely connected with
the court. The homonymous son of Hydarnes was a commander of theguards under Xerxes (Hdt.7.83.1) and later became satrap in Phrygia.Artaxerxes II (424–404) was married to a daughter of the younger Hydarnes,
Stateira (Ctesias, FGrH 688 F 15), while the royal sister Amestris was mar-
ried to Hydarnes’ son T eritouchmes. Tissaphernes, the satrap of Sardis, most
likely also a member of this family, married one of Artaxerxes II’s daughters(Diod.14.26.4). Megabyxus’ son Zopyrus held military ofﬁce under Darius I,and his grandson, also called Megabyxus, was one of Xerxes’ chief marshals.He,too, was given a royal daughter in marriage called Amytis (Ctesias, FGrH
688 F 13). Under Artaxerxes I he commanded the Persian troops in Egyptduring the revolt of 460–454 BC.
These alliances point to a royal policy aimed at establishing and main-
taining the support of the Persian nobles. Do they give us an indication ofaranking order amongst the families? Certainly Gobryas and Otanes, and
possibly Ardumani ˇs/Aspathines too, were ranked highest during the reign
43Asthe highest ofﬁcial at Persepolis, Parnaka received 90 quarts daily (see above). The amounts
of the daily rations of wine and grain or ﬂour are an indicator of the status of the recipient (see
above, p. 30).
44On the marriage alliances see Brosius 2002 :52–4, 60–3.

Court and court ceremonies in Achaemenid Persia 39
of Darius, and the military ofﬁces held by the sons of all these nobles show
the efforts made by the kings to keep the families closely linked to the court.AsArtaxerxes II’s marriage to the daughter of Hydarnes (II) indicates, the
family of Hydarnes (I) also was considered of sufﬁciently high rank for theirdaughter to be elevated to the status of the king’s wife.
Itwas a special honour when the king bestowed the status of a King’s
Friend.
45This status was expressed through royal gifts including land, estates
and cities. Other gifts were weapons (daggers), jewellery (bracelets, torques,earrings), horse bridles and a special robe. The design and quality of themetal and precious stones used to produce these items, and the colour andembroidery of the textiles used, as well as the quality of craftsmanship, allidentiﬁed the gift as a royal one, singling out its recipient among his peers.
46
Insome cases, the gift took the form of an offer of marriage to a royal
daughter. A King’s Friend was a member of the court and allowed to jointhe King’s Table, that is to say, he was privileged to dine with the king at theking’s invitation. It is important to note that even within the group of theKing’s Friends a social ranking order existed (Wieseh ¨ofer 1980 :13).
47
Structures of communication
Itwas one thing to be a member of the court, but quite another to have
direct access to the king. As the analysis so far has revealed, the limitation ofaccess to the king was a prerequisite of kingship, and, at least under Darius I,only the six Persian nobles were exempt from it. The physical separation ofthe king from those permitted to dine at the King’s Table also suggests thatonly a select few members at court were allowed to address the king. In thislight, Plutarch’s claim that the king’s wife and the king’s mother, as well asthe heir to the throne, were allowed to dine with the king gains considerablesigniﬁcance, highlighting the importance of these few members as mediatorsbetween the king and the royal and noble courtiers.
Arguably only a few members of the close court were allowed to address
the king directly. As we saw, information was ﬁltered through to the chief ofﬁ-cials who were entitled to approach the king. Those permitted to come intothe presence, be they satraps, high administrators, councillors, commanders
45There was a distinction between a King’s Friend and a King’s Benefactor, though it seems that
the Greeks did not always comprehend it (see Wieseh ¨ofer 1980 :10–11). The status of a
Benefactor was bestowed upon individuals and groups of peoples in recognition of acts ofloyalty (Wieseh ¨ofer 1980 :15). In my view Benefactors were not members of the court.
46See Wieseh ¨ofer 1980 .Onthe creation of a ‘service aristocracy’ of the king see Wieseh ¨ofer 2001 :
37. On gift-giving see Sancisi-Weerdenburg 1989 .
47Notleast in the distinction between Persians and non-Persians.

40 maria brosius
of the royal bodyguard and of the army, probably had to follow a strict pro-
tocol, only speaking with the king’s express permission. Thus, for example,
Nehemiah, the king’s cup-bearer, expressed his concerns only after the king
invited him to speak (Neh.2:2). The head of the administration, Parnaka,who took direct orders from the king, may only have spoken directly withthe king when summoned or after requesting an audience through a royalmessenger.
Those who wanted to approach the king from outside the palace depended
on the keeper of the gate to be admitted. A well-known case is that of Sylosonof Samos, who took up his place at the palace gates to obtain an audiencewith Darius I (Hdt.3.140), which was duly granted. However, in the case
of Intaphernes’ wife, who requested a royal audience so as to intercede onbehalf of her brother, communication with the king was conducted via theroyalmessenger (Hdt.3.119).
The hierarchy of ofﬁce determined a courtier’s chances of taking part
in political, administrative and military decision-making. Members of theroyalfamily, especially the king’s mother and the king’s wife, are described as
being able to pass judgement with the king’s permission (cf. Ctesias, FGrH
688 F 14). Satraps were authorised to make decisions at local level, butmatters of state could only be decided in consultation with the king. In fact,the satraps were the crucial link between the king and his subjects (Herman1997 :200). However, it is hard to reach a clear-cut view of the courtiers’
participation in political decision-making, mainly owing to the fact thatthe hierarchy of ofﬁces did not necessarily coincide with that of rank. Forinstance, in the hierarchy of the court, where are we to place Parnaka, whowas the head of administration, but who also was the uncle of the king?Equally, lineage could make high ofﬁce quasi-hereditary. This happened inthe case of the satrapy of Phrygia, which remained within the dynasty ofPharnabazos. On the available evidence, it is hard to determine how thesetwo strands – ofﬁcial status and court ranking – interacted. In addition, thehierarchies of ofﬁce and of rank were in constant ﬂux as the king conferredor withdrew special favours and positions in individual cases. Overall thecourtly hierarchy was clearly very complex.
Royal representation and royal centre
The self-presentation of the sovereign within his court was intimately boundup with the people – the courtiers – with whom the king chose to surroundhimself. The presence of these courtiers was therefore required at all ofﬁcialevents, including the funerary rites for a dead king, the royal investiture, royal

Court and court ceremonies in Achaemenid Persia 41
audiences, and the king’s migration between the different royal capitals. The
death of one king and the accession of another were the most importantevents of any reign, requiring the presence of the Persian nobility as an
afﬁrmation of their loyalty to the king and the Achaemenid dynasty. At thedeath of a king, it was his successor’s duty to proclaim the ofﬁcial empire-wide mourning period. The king’s body was returned to Persepolis and
prepared for burial in one of the tombs of the Achaemenid dynasty at nearbyNaqsh-i Rustam.
48Carved in the rock face, the cross-shaped fac ¸ades h eld a
varying number of tombs. The proximity of these tombs to the royal centre,the fact that the king’s body had to be returned to Persepolis, and equally,the consistent burial of the later Achaemenid kings at the same site, togethershow that the royal funeral was a central event for the royal dynasty, requiringthe presence of the court to perform the necessary ceremony and ritual.
Equally, the ceremonial investiture of the new king demanded the pres-
ence of the courtiers so that they could conﬁrm their allegiance to the newking. From the time of Darius I on, the royal investiture was celebrated atPasargadae – thus paying homage to the founder of the empire, while also
offering a reminder of the humble origins of the Persians. Before the kingcould don the royal robe, tiara and other paraphernalia of kingship, he had to
put on the ancient robe worn by Cyrus before he became king, eat terebinthand drink sour milk (Plut. Art.3.1–2). Only then would he receive his royal
staff, special shoes, the king’s seal and the lotus ﬂower (Polyaen.7.7.17).
49
The ceremony was observed by the king’s family as well as by the courtiers,the king’s bodyguard, the nobility and the Immortals.
Itis usually assumed that the proclamation of the heir was marked by an
ofﬁcial celebration. At this point, the heir would adopt an ofﬁcial throne-name and be presented with the kitaris and the lotus ﬂower as symbols
of his status. He was also now allowed to stand behind the king at ofﬁcialaudiences, as discussed above (see Brosius 2006 ).
Among other events which doubtless required the presence of the court
was the ofﬁcial celebration of the king’s birthday.
50
Royal banquets and royal hunt
Feasting and hunting enjoyed a long history as royal pursuits in the courts
of ancient Mesopotamia and are well attested in archaeological and literary
48On the royal tombs see Schmidt 1970 .49See Wieseh ¨ofer 2001 :32; Brosius, 2006 .
50See Hdt.1.133: ‘The day which every man most honours is his own birthday. On this he thinks
it right to serve a more abundant meal than on other days; before the rich are set oxen or horsesor camels, or asses, roasted whole in ovens; the poorer serve up the lesser kinds of cattle.’

42 maria brosius
sources. Private and ofﬁcial banquets were an expression of kingship, as
was the display of foodstuffs, tableware of precious plate, and the luxuriouscouches on which the king and dignitaries reclined. An invitation to the royal
banquet was an ofﬁcial declaration of royal favour, high rank and privilege.Dining with the king was a privilege for any member of the court who wasinvited; when the king was on campaign, it also became a privilege for thegenerals and commanders, who were invited to the royal table (see Sancisi-Weer denburg 1989 :133–5). According to the Cyropa edia,oncampaign the
king took his meals with his army: that is, he was not isolated from themen, but shared the same food and had his tent pitched in the middle ofthe encampment, surrounded by the tents of his army.
Andfor himself Cyrus had a tent made big enough to accommodate all whom he
might invite to dinner. Now he usually invited as many of the captains as he thought
proper, and sometimes also some of the lieutenants and sergeants and corporals;and occasionally he invited some of the privates, and sometimes a squad of ﬁvetogether, or a squad of ten, or a platoon, or a whole company in a body. And he also
used to invite individuals as a mark of honour, whenever he saw that they had donewhat he himself wished everybody to do. And the same dishes were always placedbefore those whom he invited to dinner as before himself.
(Xen. Cyr.2.1.30)
According to Heracleides of Cumae the king’s dinner was a court ceremony
following a strict protocol:
All who attend upon the Persian kings when they dine ﬁrst bathe themselves andthen serve in white clothes, and spend nearly half the day on preparations for dinner.Of those who are invited to eat with the king, some dine outdoors, in full sight ofanyone who wishes to look on; others dine indoors in the king’s company. Y et eventhose do not dine in his presence, for there are two rooms opposite each other, inone of which the king has his meal, in the other the invited guests. The king can seethem through the curtain at the door, but they cannot see him. Sometimes, however,on the occasion of a public holiday, all dine in a single room with the king, in thegreat hall. And whenever the king commands a symposium (which he does often),
he has about a dozen companions at the drinking. When they have ﬁnished dinner,that is, the king by himself, the guests in the other room, these fellow-drinkers aresummoned by one of the eunuchs; and entering they drink with him, though eventhey do not have the same wine; moreover, they sit on the ﬂoor while he reclines onacouch supported by feet of gold; and they depart after having drunk to excess.
Inmost cases the king breakfasts and dines alone, but sometimes his wife and some of
his sons dine with him. And throughout the dinner his concubines sing and play thelyre; one of them is the soloist, the others sing in chorus . . . For one thousand animalsare slaughtered daily for the king; these comprise horses, camels, oxen, asses, deer,

Court and court ceremonies in Achaemenid Persia 43
and most of the smaller animals; many birds are also consumed, including Arabian
ostriches…g e e s ea n dc o c k s . (Heracleides, FGrH 689 F 2; ap. Athen.4.145)
People invited to dine with the king were by no means a limited group of
nobles. According to Xenophon the king used the seating order as a means
of singling out and honouring individual guests:
Cyrusthusmade public recognition of those who stood in his ﬁrst esteem, beginning
even with the places they took when sitting or standing in his company. He did not,
however, assign the appointed place permanently, but he made it a rule that by noble
deeds any one might advance to a more honoured seat, and that if anyone shouldconduct himself ill he should go back to one less honoured. And Cyrus felt it a
discredit to himself, if the one who sat in the seat of highest honour was not alsoseen to receive the greatest number of good things at his hands.
(Xen. Cyr.8.4.5)51
Royal feasts were no small matter: they catered for several thousand guests.
According to Polyaenus ( Strat.4.3.31–2) Alexander found the following list
of foodstuffs for the king’s breakfast and dinner inscribed on a bronze
pillar:52
400artabai –aMedian artabe is an Attic medimnos –ofpure wheat ﬂour,53300 a.
of second-grade ﬂour, 300 additional a. of third-grade ﬂour: a total of 1,000 a. of
wheat ﬂour for dinner. 200 a. of pure barley ﬂour, 400 of second-grade and 400 ofthird-grade: a total of 1,000 a. of barley ﬂour. 200 a. of rye, 10 a. of the ﬁnest barleyﬂour made for a drink. X a. of ground cardamon, sifted ﬁne. 10 a. of peeled barley.
1
3a. of mustard seed.
400 male sheep, 100 oxen, 30 horses, 400 fatted geese, 300 turtle doves, 600 small
birds of all kinds, 300 lambs, 100 young geese, 30 gazelles.
10maries –amaris is ten Attic choes –offresh milk, 10 m. of sweetened whey. A
talent by weight of garlic.1
2t. by weight of pungent onions, 1 a. of phyllon (silphium
fruit?), 2 mnai of silphium juice, 1 a. of cumin, 1 t. by weight of silphium,1
4a. of
oil of sweet apples.1 a. of posset from sour pomegranates,1
4a. of oil of cumin, 3
t. of black raisins, 3 mnai of anise ﬂowers,1
3a. of black cumin, 2 kapeties of seeds
ofdiarinon ,10a.ofpure sesame, 5 m. of gleukos from wine, 5 m. of cooked round
radishes in brine, 5 m. of capers in brine, from which they make sour sauce, 10 a. of
salt, 6 m. of Ethiopian cumin, 30 mnai of dried anise, 4 k. of celery seed, 10 m. ofsesame oil, 5 m. of cream, 5 m. of terebinth oil, 5 m. of acanthus oil, 3 m. of sweetalmond oil, 3 a. of dried sweet almonds, 500 m. of wine.
When he was in Babylon or Susa, half was palm wine and half grape wine. Two
hundred waggon loads of green wood, 100 waggon loads of wood ( hule), 100 square
51See also Wieseh ¨ofer, in press .
52See Spawforth in this volume for Alexander’s imitation of Persian royal dining.
53a. = artabe (about 52.5 litres); m. = maries (about 33 litres); t. = talent; k. = kapeties (about 1.1
litre); mna (about 437 grams).

44 maria brosius
cakes of liquid honey, weighing 10 mnai. Whenever he was in Media he distributed
the following items. Three artabes of safﬂower seed, two mnai of saffron. Theseitems were used for drinks and for breakfast.
Hedistributed 500 a. of pure wheat ﬂour, 1,000 a. of pure barley ﬂour, 1000 a.
of second grade barley-meal; 500 a. of semidalis, 500 m. of groats made from olyra ,
20,000 a. of barley for the animals, 10,000 waggons of chaff, 5,000 waggons of straw,200 m. of sesame oil, 10 m. of vinegar, 30 a. of ﬁnely chopped cardamon. All thesethings listed he distributed to the soldiers(?). This is what the king consumes in aday, including his ariston ,hisdeipnon ,and what he distributes.
(Polyaen. Strateg .4.3.31–2; see Lewis 1987 :82–5)
The preparation of a royal feast is depicted on the ceremonial staircases
leading up to Darius’ palace, the ta¸cara.Servants carry dishes, wineskins
and animals, sheep and goats: the victuals for the royal feast (see Sancisi-Weer denburg 1989 :133). This fact alone, by the way, must make us wary
of accepting the notion of a ‘private palace’ in the case of Darius, or indeedXerxes and Artaxerxes I. The structures often identiﬁed as such are perhaps
better seen not as living quarters, but as the private dining space of the king,where he ate in isolation from his guests at the King’s Table. Greek sourcesfrequently state that the king dined alone or with his immediate family,while guests dined in a separate room. The palaces of the Achaemenid kingsmust have provided a dining space for the royal family which met these
requirements.
The king’s banquets were imitated by the satraps who represented the
king at local level. Banquet scenes depicted on reliefs, funerary stelai, ivoriesand seals found in many different parts of the empire show that the rep-resentation of the dignitary encouchant ,accompanied by his wife, was a
recognised artistic motif for the expression of the courtly life.
54
Like banqueting, hunting too was a royal Achaemenid pursuit, as in other
ancient court societies studied in this volume (see the chapters by Spawforth,Wieseh ¨ofer, van Ess and Spence). Hunting took place in royal enclosures
stocked with wild beasts and tended by gamekeepers, or in the wild, whereroyalendurance and physical and mental toughness were tested even more.
With its reliance on lances and the bow, hunting was seen as a form of military
training. Fighting against the most ferocious animals, such as boars, lions,
panthers and bears, as well as hunting game like gazelle, deer and birds suchas ostriches, exercised the skill of the participants and displayed it to one’speers:
When the king goes out hunting, he takes out half the garrison; and this he does
many times a month. Those who must take the bow and arrows, and, in addition
54See above, n. 36.

Court and court ceremonies in Achaemenid Persia 45
tothe quiver, a sabre or scimitar, they carry along also a light shield and two spears,
one to throw, the other to use in case of necessity in hand-to-hand encounter . . .
and as their king is their leader in war, so he not only takes part in the hunt himselfbut sees to it that the others hunt too. The state bears the expense of the hunting forthe reason that the training it gives seems to be the best preparation for war itself.
(Xen. Cyr.1.2.9–10)
Participation in the royal hunt was hedged with etiquette. The chief rule was
that the king reserved to himself the right to strike the ﬁrst blow. When the
king’s son-in-law (Megabyxus: see above, p. 38) disobeyed this rule in goodfaith, thinking that the king was about to be attacked by a lion, he was bannedfrom the court and stripped of his court privileges. Only the interventionof the king’s mother Amestris, and the king’s daughter Amytis, Megabyxus’wife, se cured his pardon, eventual return to the court and restoration of his
privileges – he was permitted once more to join the King’s Table (Ctesias,FGrH 688 F 40–1).
55Itis likely that, on occasion, the women of the court
joined the hunting party.
The royal entourage
Members of the court, obviously enough, formed an essential part of the
king’s entourage. Whenever the king travelled across the empire, migratingbetween his royal capitals or on campaign, the court travelled with him.The royal entourage was an expression of kingship, conveying power, orderand control. The precise order in which the entourage travelled was care-fully planned. This emerges from the order of march described in threesources: Xenophon’s Cyropa edia (for the order of the entourage of Cyrus
II), Herodotus (for the army of Xerxes in 480 BC), and Curtius Rufus(Curt.3.3.8–16), who listed the order of the train of Darius III as follows:
1.ﬁre carried on 2 silver altars in front
2.magi
3.365 men in purple robes
4.white horses pulling the chariot of Ahuramazda
5.1h o r s e of the sun
6.10 chariots
7.horsemen of 12 nations
8.10,000 Immortals
55Butcompare this to Plut. Mor.173d: ‘[Artaxerxes] was the ﬁrst to issue an order that any of his
companions in the hunt who could and would might throw their spears without waiting for
him to throw ﬁrst.’

46 maria brosius
9.15,000 royal kindred ( cognati )56
10. spear bearers ( doryphoroi )
11. the king’s chariot
12. 10,000 lancers
13. 200 king’s relatives on right and left
14. 30,000 infantry
15. 400 king’s horses
16. the chariots of the king’s mother and king’s wife
17. the women of the queen’s household on horseback
18. the king’s harmamaxa for children and eunuchs
19. 365 concubines
20. 600 mules, 300 camels
21. the wives of the King’s Relatives and Friends
22. the troops of sutlers and batmen
23. the band of light-armed troops.
Centrally positioned, the king was surrounded by spear bearers and lancers,
cognati and kin, the 10,000 Immortals and 10,000 lancers. These were com-
plemented by the cavalry and 400 of the king’s horses. Then there followedthe royal women, children and concubines, as well as the wives of the King’sRelatives and Friends.
The physical space of the court: Pasargadae, Persepolis and Susa
The physical manifestation of the court chieﬂy took the form of the palaces inthe royal capitals of the Achaemenid kings. The palace was an expression ofkingship, royal power and political control.
57Spaciousness and monumental
architecture were as important as luxurious building materials, artistic style,and interior and exterior design. Cyrus II recognised that monarchical powercould be physically expressed in the foundation of a new royal capital – an
idea with a long Near Eastern tradition. The creation of new settlements,
58
along with a new royal palace with its buildings projecting the royal image,was the visual expression of a monarch’s command of resources.
56Forthese, see Spawforth in this volume.
57On the Persian ‘royal house’ (OP vi/p113-) and the palaces of Darius and Xerxes see above,
pp. 25–6; also Wieseh ¨ofer, in press .
58Foranexcellent analysis of Near Eastern cities see van de Mieroop 1997 ,who cautions against
seeing royal power as the sole determinant of civic government. As he points out, we need to
distinguish between the king’s roles as a ‘national’ leader and as head of the palatialorganisation, his own household (1997: 119).

Court and court ceremonies in Achaemenid Persia 47
Insupport of his role as national leader the king could rely on an elaborate palatial
organization that provided him with administrators and warriors. Palaces were
among the most prestigious buildings of all Mesopotamian cities, and the palatialsector was at times of utmost prominence in society.
(V an de Mi errop 1997 :120)
The palace with its extensive building complexes was the dominating
feature of the royal city thanks to its position within the city, which was
normally both elevated and fortiﬁed, though not necessarily central. Theroyalpalace was a manifestation of the unique position of the king, of his
exalted position above all other levels of society, yet at the same time it
asserted his presence among his subjects. A new royal complex required thecreation of a whole palatial organisation, from the king’s personal attendantstoother kinds of royal servant, along with ofﬁcials, administrators and
guards. As the palace became a manifestation of kingship, so did the royalcourt and its society.
The conquest of Media in 559 BC provided Cyrus II with his ﬁrst opportu-
nity to manifest his power in this way, with the foundation of the ﬁrst Persianroyalcapital, Pasargadae (Elam. Batrakataˇ s). Named after Cyrus’ tribe, the
city was testimony to the value placed on the pre-eminent Persian tribe. Itslocation in the plain of Marv Dasht in Persis was said to commemorate thehistoric battle between the Median king Astyages and the army of Cyrus II.
The full extent of Pasargadae is unknown and will only be revealed by
further excavation. However, the recent suggestion of R. Boucharlat deservesserious consideration, namely, that the city’s inhabitants may have continuedtolive traditionally as nomadic past oralists, erecting their tents around the
ofﬁcial buildings.
59These buildings, intended for royal representation, were
built in stone and decorated with black and white marble. One was Cyrus’‘private’ palace, a hypostyle hall based on a rectangular plan and set within a
royalgarden, hidden from the public palace, where the king held audiences.
The visitor was admitted to the palace complex via a gate. Here guardscontrolled access and only allowed visitors to enter the royal compound
after approval by a courtier. If admission was granted, the visitor would beguided to the royal audience hall (Fig. 1.3).
Asfor their architectural design, the buildings at Pasargadae show their
artistic debt to the kingdoms conquered by Cyrus. Median inﬂuence can bedetected in the construction of columned halls.
60But in the construction of
his palaces at Pasargadae, Cyrus II went further than merely acknowledging
59www.achemenet.com/recherc he/sites/pasargades/pasargades.htm; Boucharlat 2001a and2001b .
60Our earliest archaeological evidence for these comes from Godin T epe and Nush-e Jan; see
Stronach 1987 (with ﬁgs. 6 and 7 for plans of Godin T epe and Nush-e Jan).

48 maria brosius
Figure 1.3 Plan of Pasargadae (drawing by Marion Cox)

Court and court ceremonies in Achaemenid Persia 49
aMedian connection. The remaining wall decorations of the palaces, carved
in relief, show an adherence to the style of decoration of the Assyrian and
Babylonian palaces. Though the reliefs at Pasargadae are only partly pre-served, we can still identify the Assyrian smiting-god and lion demon, aswell as the Assyrian ﬁsh-garbed man and the bull-man, though the pairing
of these last two departs from Assyrian tradition (see Stronach 1997 :44).
The entrance gate to the royal compound depicts one of the most intriguingﬁgures in Persian art, the winged genius: a ﬁgure commonly depicted inthe art of the Levantine coast, with two pairs of wings, clothed in Elamitedress and an Egyptian headdress. While the ﬁgure of the winged geniusultimately derives from Assyrian models known from the palace relief atDur-Sharrukin (Khorsabad),
61the Elamite dress is a reminder of the Per-
sian link with Elamite culture (cf. Stronach 1997 :43).62
The achievement of Cyrus at Pasargadae cannot be emphasised too
strongly. He (and his advisors) integrated architectural features from thelands which the king had conquered to create his royal capital. In one spaceit offered a visual expression of Cyrus’ conquest of the western kingdoms. Itwas this space which gave a visitor – whether a foreign ambassador, a localpetitioner or a satrap – access to the king. Cyrus’ idea was innovative: thephysical seat of monarchy embodied the power of the king and reﬂected thetriumph of his conquests. It was simultaneously an expression of kingship,
of power, of control and of political integration. Darius I and his succes-sors were to base their concept of kingship on Cyrus’ ingenious idea. Itfound physical expression in the royal terrace of Persepolis and the tombsat Naqsh-i Rustam.
What does the palace complex at Pasargadae reveal about the king and
kingship and about the establishment of the Persian court? Primarily, theconstruction of this walled palace complex with its entrance gate established
an important feature of kingship: controlled access to the king. This vettingof access to the king turned him into a ﬁgure remote from his subjects. Fromleader of his tribe, from ﬁrst among equals, Cyrus II had established his royalpower through the conquests of kingdoms. Subsequently he took on thetrappings of kingship and of monarchical representation. The audience hall
was separate from the private palace of the king, itself set apart by a structuredgarden and accessible, we can assume, only to the king’s attendants, hisbodyguard and members of his family.
61The city was founded by Sargon II (721–705 BC).62See Herrmann and Curtis 1998 .

50 maria brosius
Figure 1.4 Plan of Persepolis
Persepolis and Susa
The palace complex of the city of Persepolis, called Parsa by the Persians,
epitomises royal power and Achaemenid kingship (Fig. 1.4).63
The city itself has not yet been excavated, the archaeological focus so
far being the royal terrace. Exploiting the natural rock of the Kuh-e Ramat,which was extended by a retaining wall of massive masonry, the royal terracetowered18metres above ground level. The complex is surrounded by a
fortiﬁcation wall which incorporates the hill of Kuh-e Ramat as well. Accesstothe terrace was provided by a large double staircase which led to the Gate
63ForPersepolis see Schmidt 1953 ,1957 ;R o o t 1979 .

Court and court ceremonies in Achaemenid Persia 51
of all Lands. Its broad, shallow steps ensured two things: ﬁrst, a visitor could
progress only by slow steps onto the royal terrace. Second, the steps allowedthe ascent and descent of horses, and possibly even chariots. Doubtlessthis would have been a privilege permitted only to an exceptional few –perhaps to the king alone. Once on the terrace, the visitor entered the Gateof all Lands, a square building with four columns at its centre. Here, thevisitor had to sit and wait – stone benches were provided for this purpose –
until a guard or royal messenger had announced his arrival to the king andpermission was given to advance onto the terrace.
64Asin Pasargadae, the
entrance gate controlled access to the king. Its attendants, guards and royalmessengers ensured that no one could enter the palace complex unnoticedor unannounced.
The ﬁrst buildings constructed by Darius I included the palace of the king,
the so-called Treasury and the administrative complex. The most impor-tant one, however, is the vast Throne-Hall, the Apadana .Elevated from the
ground level of the rest of the complex by a further 3 metres, the Apadana
was built on a square ground plan, with thirty-six columns arranged in rowsof six, each 20 metres high, their bul l-headed capitals supporting a cedar-
wood roof. The two double staircases with the famous reliefs depicting the
inner court and the Immortals, as well as the peoples of the empire, led tothe hall. The visitor would be led up to the audience hall via one of thesetwo staircases.
65
Darius’ successors completed his public and administrative buildings and
added their own palaces also. Under Artaxerxes I (465–424 BC) the complexunderwent a signiﬁcant change. A new gate was added, giving access to anew court in turn leading to a second throne-hall, the Hall of One HundredColumns. This was not raised up in the manner of the ﬁrst throne hall, and
the only public access was through a porticoed entrance. But there can beno doubt as to the representational func tion of the hall, which processions
would reach from the Gate of all Lands by means of the new gate and across
acourtyard.
AsCyrusIIhad done in Pasargadae, Darius I incorporated many architec-
tural elements into the design of the palaces and halls of Persepolis inspiredbythe regional traditions of the empire. But unlike the mere borrowing of
64See above, pp. 39–40.
65Itwould be intriguing to know whether the two staircases implied a hierarchy amongst those
who used it. Who was allowed to ascend which staircase? Was one reserved for the king only,
the other for visitors? As for the double staircase itself, the reliefs imply that courtiers andbodyguards used one side, gift-bearers of the lands the other. But we cannot be sure that thestaircase reliefs depict a reality or an ideal.

52 maria brosius
stylistic elements at Pasargadae, Darius and his advisors created a coherent
new royal style, which was consistently adhered to in the construction ofall royal palaces and copied in the satrapal centres of the empire. This stylecombined Assyrian, Babylonian, Median, Urartean, Egyptian and Greek ele-
ments in the construction of columns and column bases, gateways, window
frames and blind windows. This was a deliberate decision to create a palacecomplex articulating in its architecture the idea of an all-inclusive empire.
Itﬁnds clear echoes in Darius’ building inscription from Susa, listing the
origins of the building materials used in the construction of the palace there,along with the multi-ethnic workforce.
The cedar timber was brought from a mountain called Lebanon. The Assyrian people
brought it to Babylon. From Babylon the Carians and Ionians brought it to Susa.The sissoo -timber was brought from Gandara and from Carmania. The gold which
was worked here was brought from Sardis and from Bactria. The precious stonelapis lazuli and carnelian which was worked here was brought from Sogdiana. Theprecious stone turquoise, which was worked here, this was brought from Chorasmia.The silver and the ebony were brought from Egypt. The ornamentation with whichthe wall was adorned was brought from Ionia. The ivory which was worked herewas brought from Ethiopia, and from India and from Arachosia. The stone columnswhich were worked here were brought from a village called Abiradu, in Elam. Thestone-cutters who worked the stone were Ionians and Sardians. The goldsmiths whoworked the gold were Medes and Egyptians. The men who worked the wood were
Sardians and Egyptians. The men who worked the baked brick were Babylonians.The men who adorned the wall were Medes and Egyptians.
(DSf §4)
The same palatial conﬁguration of an entrance gate, throne-hall, palace and
administrative rooms was copied at Susa, where Darius I began constructionsoon after his accession. Passing through the entrance gate (40 ×28 m),
the visitor entered the elevated royal complex; with the palatial buildingsopposite, he then had to turn right into the courtyard to approach theApadana .AsatPersepolis, this was based on a square plan,
66though at
Susa neither the throne-hall nor the palace was raised above ground-level.
The palace itself was rectangular, with three courtyards leading to royalapartments.
67The surviving architecture of the palace complex of Persepolis
66Each side was 109 metres long, with three columned porticoes. Six row of six columns
20 metres high held up the wooden ceiling. The throne was probably placed on the stonefoundation in the northern row (Boucharlat 1997 :59).
67Royal palaces were also built in Ecbatana and Babylon. Current excavations at Ecbatana have
not yet revealed an Achaemenid palatial space; but the Achaemenid palace at Babylon is wellattested (see Kuhrt 2001 ).

Court and court ceremonies in Achaemenid Persia 53
and the other royal cities nowadays only hints at the former splendour of
the Achaemenid court, its rituals and its ceremonies.
Ways of re cruitment
The question as to how courtiers were recruited is not easily answered forthe Achaemenid court. Clearly, many came from the extended royal familyitself; as we have seen, male relations of the king held high ofﬁce as satraps,administrative heads and military commanders. Holders of other ofﬁcesin the inner and outer court were appointed by the king, and while someofﬁces acquired a hereditary character, others were always ﬁlled by the kingwhen they fell vacant. At this point a number of factors inﬂuenced the king’spatronage, such as the desire to promote a speciﬁc noble in return for loyaldeeds bringing him to the royal notice. The same factors could inﬂuencethe appointment of non-Persian ofﬁcials too. It is likely too that childrenof the nobility were educated at the court, although we have no detailedinformation about this practice.
68
Itis clear enough, however, that the Achaemenid court was an institu-
tion of considerable complexity. Different hierarchies of tribe, rank andhigh ofﬁce were interwoven to create a court organisation. Importantly,these hierarchies did not exist independently of one another: a ﬂexible sys-temallowed changes within and between them. The king controlled these
changes as the fount of court patronage. Court life was a system of inclusion
and exclusion, of separate groups around the king who enjoyed closer orlesser proximity to the king. Court life gained complexity because withineach group, be it the members of the King’s Table, or the King’s Friends,among the councillors and military leaders, a ranking order existed. While,on the one hand, this created a clear structure, on the other it created compe-tition, ambition and strife among courtiers eager to maintain and improvetheir position. Servile opportunism and ﬂattery were undoubtedly part ofeveryd aylife at court.
This complex grouping of courtiers became a unity through ceremony
and ritual, which included royal feasts, hunting, birthday and New Y ear cel-ebrations, investiture, royal funerals and commemorative rituals. For some,if not all, courtiers, this unity was also a matter of sharing the same court
68Foranoverview of Persian education see Briant 2002 :327–30.

54 maria brosius
language, Old Persian, and the celebration of the cult of the Persian royal
gods, especially Ahuramazda.
Manifestation of social ranking through royal gifts
This leads us to the ﬁnal questions. How did the court work? Why was thesystem, set in place by Cyrus II, so successful over so many reigns? Elias’judgement on the French court applies here, too. A court society is not ananonymous group of people, but a ﬁguration of individuals (see Elias 1983 :
8; Herman 1997 :200). It is likely that the establishment of court ceremony
and ritual created a highly stratiﬁed ranking order among attendants, themilitary, the administrators, the royal household, the women and the nobil-ity. Each individual would then have identiﬁed closely with his or her rolewithin this ranking order. Rank was recognised and acknowledged among
the Persian nobles themselves (see Hdt.1.134). Even the satrapal rebellionsin Asia Minor, which disrupted the ‘ pax Persica ’during the mid-fourth cen-
tury, could not overturn the kingship. Why? The answer, in my view, liesin the king’s emphasis on the individual. Everyone involved in court lifewas related or known to the king, marriage alliances were carefully con-structed, and high positions were ﬁlled by individuals trusted by the king.Loyalty had to be proven, and was rewarded with royal gifts, land and estates,with appointment to high ofﬁce, and sometimes with the honour of inter-
marriage with the royal family. Within this highly personalised meritocracy,one aspect was of vital importance: the ruler’s personal knowledge of eachsubject whom he honoured:
Andwhenever Cyrus wished to honour anyone it seemed to him proper to address
him by name. Furthermore it seemed to him that those who were conscious of
being personally known to their general exerted themselves more to be seen doingsomething good and were more ready to abstain from doing anything bad.
(Xen. Cyr.5.3.47–8)
Showing a similar concern for the army, Xerxes is said to have addressed his
soldiers before the army’s march to Greece as follows:
Riding in a chariot past the men of each nation he questioned them, and his scribes
wrote all down, till he had gone from end to end of the infantry and cavalry. This
done…X e r x e s alighted from his chariot into a ship of Sidon, sitting wherein under
agolden canopy he was carried past the prows of the ships, questioning of them in
the manner as of the army and making the answers to be written down. (Hdt.7.100)

Court and court ceremonies in Achaemenid Persia 55
The key to the successful workings of the Achaemenid court was personal
recognition by the king and his bestowal of gifts to reward service. According
toXenophon, the importance of gift-giving as a political instrument was
already recognised by Cyrus II:
Andbesides this he used to reward with gifts and positions of authority and seats
of honour and all sorts of preferment others whom he saw devoting themselves
most eagerly to the attainment of excellence; and thus he inspired in all an earnestambition, each striving to appear as deserving as he could in the eyes of Cyrus.
(Xen. Cyr.8.1.39)
The king had created a system of royal favour and patronage in which the
constant royal scrutiny of the achievement of individual courtiers gave them
in turn the chance to increase their wealth and status and to enhance theirofﬁcial position and degree of personal closeness to the king. T erms like‘the greatest after the king’ , the ‘most honoured’ and the ‘most trusted’
allow a glimpse of the pyramidal formation which characterised the dif-ferent courtly ranks. Importantly, the system never stagnated, but was inconstant ﬂux, with royal favour and gift-giving acting as a continuous spur
toindividual ambition. Equally, privileges could be revoked, and therefore,
time and again, the nobles had to reafﬁrm their loyalty in order to con-ﬁrm their worth and status. Failure to do so led to the withdrawal of royalfavour. No court ofﬁcial was exempt from punishments for disloyalty, run-ning from loss of status and landed wealth to banishment from the court,and even death (potentially for al lthe male members of his family too).
Thus the ethos of this ‘service aristocracy’ was based not only on personalambition, but also on fear of loss of status. Accordingly, it fostered loyaltyon the one hand, but also intrigue and rivalry on the other, with ambitiousnobles and royal scions becoming allies in the plotting of potential palacecoups.
69
Each noble, satrap and commander was an integral part of the monarchy.
This is at the heart of the Achaemenid king’s creation of a court society.The king was an absolute monarch, but he reigned with the support of hiscourtiers, who had an active interest – stimulated by both ambition and
fear – in maintaining the stability of empire. As satraps they had a share ofpolitical power, since they represented the king in the provinces – where the
69Herman’s observation about Hellenistic courts seems equally valid for Achaemenid Persia:
‘Within the court were taken the decisions which shaped the destiny of the kingdom, and
within it were found the most formidable groupings both of the king’s supporters and of his
opponents’ (Herman 1997 :200).

56 maria brosius
court system was repeated on a smaller scale – and formed the link between
the king and his subjects. As nobles in high ofﬁce close to the king, they
contributed to creating and maintaining the king’s exalted position, while
their relative proximity to the monarch gave them prestige and status amongtheir peers.
Importantly, members of the non-Persian elite were also integrated into
the system, being selected for (high) ofﬁce both at the court and – more fre-quently – at provincial level. Drawing the circle even wider, the peoples of thelands of the empire, too, were assigned an integral place in the presentationof kingship. At any rate, this is the message conveyed on the door-jambs ofArtaxerxes’ palace at Persepolis and on the tomb reliefs of Naqsh-i Rustam.
The king, enthroned, is held up by an even mightier throne, supported by thepeoples of the empire. Royal consideration for the individual who honouredthe king, however humble his background,
70underlay the court system with
its intricate human networks, and this system in turn helped maintain thestability of the empire.
Conclusion
The court had long been an integral part of monarchical rule in the ancientNear East. Concepts of court, court ceremony and court ritual therefore were
well established when the Persians rose to power, and the Achaemenids were
able to integrate these older court styles into the existing tribal and militarystructure of their own society. That said, it remains hard to identify majorchanges in the court structures of the ancient Near Eastern monarchies over
time, and we have no means of knowing whether aspects such as admissiontothe court, appointment to high ofﬁce and the ranking procedures among
courtiers changed in the course of Achaemenid rule.
With the establishment of Persian kingship, the creation of a highly organ-
ised court became a necessary element of royal self-representation. Theinterdependence of king and court revealed itself as the king felt obligedtoemphasise his unique position, becoming remote from his peers and his
subjects, while at the same having to remain a highly visible ﬁgure. Thecourt society responded to both these aspects of Achaemenid kingship: as
70See the story told of Artaxerxes’ attitude towards gifts: ‘it was no less the mark of a king and
lover of his fellow men to accept small gifts graciously and with a ready goodwill; and so, on atime when he was riding by, and a simple labourer, possessed of nothing else, took water from ariverinhis two hands and offered it to the king, he accepted it pleasantly and with cheerful
smile, measuring the favour by the ready goodwill of the giver, and not by the service renderedbythe gift’ (Plut. Mor.172B).

Court and court ceremonies in Achaemenid Persia 57
an institution it emphasised the king’s exalted status, while the courtiers
themselves became intermediaries between the king and his subjects. Theking needed the court as a ‘theatre of power’ which emphasised his uniqueposition through pomp and court ceremony. The stability of the empire,or at least the impression of a stable empire, depended on the king and hiscourt.
71
71Inmyview, this becomes evident in the breakdown of the relationship between king and satrap
under Alexander the Great, who adapted the political structure of the satrapies, but failed to
recognise the need to honour individuals and integrate them in his court. See Brosius 2003 ;also
Spawforth in this volume, with a different interpretation of Alexander’s court politics.

2 King, court and royal representation in the
Sasanian empire
josef wieseh ¨ofer
Introduction1
With the exception of the (late) Roman imperial court,2theoretical and his-
torical approaches to the princely court as ‘the most general arrangement of
power in pre-modern society’3are mostly a phenomenon of medieval and
early modern, rather than ancient, history.4Asfor the civilisations of the
ancient Near East, here we are more or less treading on virgin ground.5Ye t
the monarchies of the Assyrians, Babylonians, Egyptians, Persians, Parthi-
ans and Sasanians stand absolute comparison with their early modern Euro-pean counterparts in terms of their impact on their subject-populations andneighbours. This does not mean that no scholar until now has ever studiedthe life, institutions and modes of communication in ancient Near Easterncourts.
6Butsuch work has seldom been based on a theoretical model or
comparison with the court phenomenon in other societies or cultures.7As
for the Sasanian empire, we do not even possess detailed descriptions orantiquarian studies of the court as a centre of decision-making and gov-ernance, as the focus of social networks, as a stage for royal or aristocraticrepresentation, as a centre of consumption, or as a control centre for trans-
regional communication.
8Norhas any expert in the Iranian languages ever
addressed the question as to whether there was a Middle Persian or Parthianequivalent of Latin curia ,English court or German Hof.
Unfortunately, the speciﬁc nature of the Sasanian source material (see
below) makes it extremely difﬁcult to rise above the merely descriptive andantiquarian approach. Despite the difﬁculties, this chapter aims to provide
1Iwould like to thank T ony Spawforth for his kind invitation to take part in the courts workshop
at Newcastle and all my fellow-participants for helpful comments on a ﬁrst draft of this chapter.
Ashort outline of Sasanian history and culture can be found in Wieseh ¨ofer 2001.
2See Smith in this volume; Winterling 1998 ,1999 ;Scholten 1995 .
3Duindam 2003 :302.4Butsee Winterling 1997b ;H e r man 1997 .
5The proceedings of a Copenhagen colloquium on ‘Palace, King and Empire’ , held in 2000, arestill in press (Larsen, in press).
6See, e.g., Briant 2002 :passim for the Persian empire.
7Butsee now Brosius and Spence in this volume.
8There is some relevant material in Wieseh ¨ofer 2001 (with the older literature).

King and court in the Sasanian Empire 59
an initial ‘problematisation’ of the fo llowing aspects of Sasanian arrange-
ments of power:9the social constitution of the ‘court’ and the ﬂuctuating
presence there of various different groups of people; structures of commu-
nication at court (court politics in other words), as well as the problemof order and conﬂict at court; the function of the court in royal decision-making and as a site of royal (and aristocratic) representation (including theexemplary role of the court in matters of manners and culture); methods of
recruiting the people closest to the ruler and of safeguarding their loyalty;
the ‘stateliness’ of the court, and its importance as a setting for the displayof social ranking.
The chapter does not pretend to answer all these questions. But it strives
tocut a swathe through the source material for the court so as to improve
our understanding of power and ‘state-building’ in Sasanian Iran.
Ashort history of the Sasanian empire
Aswith the Parthians, we have only patchy evidence for the aims of Sasanian
foreign policy and military campaigning. The focus of this material is the his-toryof Sasanian encounters, friendly or bellicose, with western neighbours.
Under Ardashir (AD 224–241/2), the founder of the Sasanian empire, all
the former Parthian provinces except Armenia fell into the Sasanids’ hands.Under the same king, an offensive policy against Rome is also discernible.
Shabuhr I (240–271/2) was more successful than his father: his campaignsnot only saw the conquest of Armenia, but shook the Roman empire toits foundations. Sasanian armies overran Syrian Antioch and Cappadocia,and for the ﬁrst time a Roman emperor (Valerian) fell into the hands ofthe enemy. Despite later setbacks (e.g. the successful resistance of Odae-nathus of Palmyra), Shabuhr left an empire extending from Mesopotamiain the west to Peshawar in the east. Quarrels for the throne and Diocletian’sequally aggressive eastern policy caused the Sasanians to lose Armenia andeven some of their territories east of the Tigris. It was only Shabuhr II in
the fourth century who was able to set aside the terms of the ignominioustreaty of Nisibis (298). Shabuhr not only fended off the Roman emperor
Julian in a battle outside the Sasanian royal residence of Ctesiphon, but
also made successful use of diplomacy and warfare to wrest much of the
9Forthese ‘Weberian’ constituents of a court see Winterling 2004 .Although originating in
attempts to deﬁne the ideal type of the early modern European court, Winterling’s typology has
proved helpful for the analysis of the Sasanian court undertaken in this chapter. Other usefulconcepts: Duindam 2004 ;Butz and Dannenberg 2004 .

60 josef wieseh ¨ofer
lost Sasanian territory from the Roman emperor Jovian, Julian’s successor
(AD 363). These wars were marked by harsh persecutions of the Sasanianempire’s Christian subjects, who were regarded by the Christian emperorstothe west as under Roman protection, and by the Sasanian authorities as a
pro-Roman ﬁfth column. In 387 the eastern part of Armenia again becameaSasanian province.
Over the next century the Hephthalites (‘White Huns’) gave the Sasanians
muchmore trouble than the Romans, with whom the kings of Iran came to
amutual agreement around 400. The Hephthalites were a kindred people of
the steppes who had gradually penetrated central Asia, conquering Sogdia,Bactria, the western parts of the Tarim basin and north-west India. They notonly inﬂicted two crushing defeats on the Sasanian king Peroz (465/484),but were also able to force tributary status on the Sasanian empire. T ogetherwith famines, these defeats brought the empire to the verge of ruin. Under
the inﬂuence of the ethical and religious ideas of one Mazdak, the peasantsrebelled, weakening the landowning aristocracy, to whom the vast majority
owed services and obligations. At ﬁrst supporting these ideas, King Kawad
and his son Husraw ended by brutally suppressing them. However, theyboth tried to take advantage of this weakening of the nobility to initiate fun-damental social, economic and military reforms. On the basis of a generalland survey, a new system for exacting the land-tax was devised. T ogetherwith agricultural reforms, this system aimed to support small farms from
which taxes could easily be extracted, and to prevent the growth of hugeestates with powerful owners who could resist the royal tax-collectors. Inaddition, the empire was divided into four military districts, and specialforces were raised for the purpose of surveillance and border control.
With domestic stability re-established, Husraw I reopened hostilities with
Rome (Byzantium) by breaking the ‘eternal peace’ (540). Military success
on the western and south-Arabian fronts meant that he was eventually abletoraise the level of Byzantine tribute-payments to Iran. In the east, Husraw
destroyed the Hephthalite empire with the help of the Western Turks ( c.560).
Hisreign also constituted the cultural climax of Sasanian civilisation.
Under Husraw, known as the ‘wise king’ , Iran became the hub of a reciprocal
transfer of knowledge between east and west. But as early as the reign of his
son Ohrmezd (Hormizdas) IV, the conﬂict between king and aristocracyresurfaced (after 579); serious attacks by the Turks worsened the situation.
Matters seemed to improve when Ohrmezd’s son Husraw II was able to put
down the rebellion of Wahram Chobin, a general descended from the oustedArsacid dynasty; in a war with Byzantium he succeeded in campaigning as
far into Byzantine territory as Egypt (619) and the vicinity of Constantinople

King and court in the Sasanian Empire 61
itself (626). But a counter-attack of the emperor Heraclius forced Husraw to
giveup his newly conquered territories, and the king himself was overthrown
byarebellion of nobles, and eventually killed (628). After an ensuing phase
of anarchy marked by frequently changing regencies, the noble faction ofRustam put Y azdgerd III on the throne. However, the last Sasanian king
proved unequal to the task of defending his empire against the Muslimarmies, weakened as it was by particularism and by the dissolution of the pro-Sasanid Lakhmid ‘buffer state’ in Arabia. After heavy defeats at Qadisiyyain Iraq (636) and Nihawand in Media (642), Y azdgerd retreated to easternIran, where he was killed at Merw (651). His empire was soon incorporated
into the caliphate.
The sources10
Dealing with the Sasanian court as a centre of communication and a place ofrepresentation is not an easy undertaking. Most of our information is not of
adocumentary character. Instead, the sources from the centre of the empire,
like the royal inscriptions, the works of royal art (rock reliefs, silver bowls,etc.) and the extant works of Middle Persian literature, present an ideolog-ically biased image of the king and court society. They normally stress theexemplary role and behaviour of both parties, sometimes describing the fatal
consequences when courtly norms were transgressed. By contrast, Chris-
tian and Manichaean writers and the Romano-Byzantine authors presentthe Sasanian kings and their ofﬁcals either as persecutors of their faiths oras military opponents. The early Islamic historians situate pre-Islamic Iraninto the larger ‘history of salvation’ which started with Muhammad’s mis-sion. Finally, many late Middle and New Persian or Arabic texts preserveallegedly authentic proclamations bySasanian rulers or reports of Sasanian
court life. But in reality these are later (Islamic) views of a glorious early
Sasanian past, or belong to the ‘legendary cycles’ of the semi-ofﬁcial ‘IranianNatio nal History’ .
Asaresult of this conﬁguration of source-material, two moments in the
history of the Sasanian court and Sasanian royal representation are muchbetter attested than others. The ﬁrst is the age of the empire’s foundersArdashir (Artaxerxes) I, Shabuhr (Sapor) I, Wahram (Bahram) I and Narseh
(third century AD), the second the reigns of the reformer king Husraw
10Foranoverview see Wieseh ¨ofer 2001 :153–64, 283–7. See also Abka’i-Khavari 2000 and Cereti
1997 .

62 josef wieseh ¨ofer
(Khusro/Chosroes) I and his successors Ohrmezd (Hormezd/Hormizdas)
IV and Husraw II (sixth and seventh centuries AD).
Forthe early Sasanian court, the royal inscriptions are the most important
source. First, some mention the major ofﬁcials and dignitaries of the empireand at court.
11Second, the way in which they present the kings with their
titles and pronouncements embodies a speciﬁc idea of kingship as the quasi-divine authority interceding with the gods, in particular Ohrmezd (AhuraMazda), on behalf of the king’s subjects. Third, some royal inscriptions are
in effect records of royal achievements or resgestae ,s e r v ing as a corrective to
the Roman and the late Arabic sources. Of course, these inscriptions must beviewed as extremely one-sided evidence for the royal patron’s self-image. In
their original versions on rock fac ¸ades and buildings at sacred sites, they were
meant to proclaim the king’s intimate relationship with his putative ances-tors and with the gods. In the now-lost copies which once circulated, they
aimed to prove the legitimacy of Sasanian royal rule to the outside world.
Inawell-judged choice of site, Shabuhr I had his resgestae inscribed on an
ancient building (Ka‘ba-i Zardusht) of Achaemenid date at Naqsh-i Rustamin the vicinity of Istakhr and Persepolis, in front of the rock-cut tombs of theAchaemenid kings. Emulating these earlier Persian kings, he had his deeds
inscribed in three languages: Middle Persian, Parthian and Greek.
12For
Roman historians, the ResGestae Divi Saporis (ˇSKZ)aremainly interesting
as evidence for Shabuhr’s campaigns against the Roman east. But they arenot just a record of the deeds of the ‘King of Kings of Eran and Non-Eran’:they also provide an insight into the royal court of the third century AD andits numerous governmental and court ofﬁces.
Inabilingual (Middle Persian and Parthian) inscription on the Paikuli
monument in Iraqi Kurdistan (NPi), Shabuhr’s son Narseh describes atlength the military challenge to his succession from his rival and great-nephew Wahram III, and his subsequent recognition and coronation by thegreat men of the empire.
13Like his father’s resgestae ,Narseh’s inscribed
achievements also list royal dignitaries and functionaries and mentionnumerous titles and ofﬁces.
Highly hagiographic in character, the Manichaean texts concerning the
death of Mani in a Sasanian state prison give some further details aboutmodes of communication at the court of the early Sasanians.
14The royal
11Forap rosopography of the early Sasanian empire see the current online project of Weber.
12Huyse ( 1999 )o f f e r sa ne x c e l l e nt edition of the ResGestae Divi Saporis .
13Edition with commentary: Humbach and Skjærvø 1978 –83.
14See Sundermann 1981 (edition and commentary); Weber, s.v. ‘Kerd /p180r, Sohn des Ardaw ¯an’ (for
the traditions concerning Mani’s death).

King and court in the Sasanian Empire 63
reliefs are closely linked to the royal inscriptions both as to ﬁnd-spots and
subject-matter.15They combine cosmic with worldly events, and have a
compound and cumulative character. They offer eloquent witness to the
self-representation of the Sasanian rulers, to their efforts to establish their
legitimacy, and to their relations with the gods, foreign enemies (the Romanemperors), and family-members and dignitaries.
Our most important evidence for the court of Husraw I and his successors
comes from the Byzantine and Islamic historical tradition, the latter depend-
ing heavily on late Sasanian works – notably the X
wad¯ay-n ¯amag (‘Book of
Lords’), a semi-ofﬁcial and semi-legendary history of Iran from the ﬁrstworld king Gayomard to the time of Husraw II, compiled under Y azdgerd III,
the last Sasanian monarch.
16Aswell as these ‘historical’ texts, there existed
in Sasanian times a number of works dealing with institutions, court pro-tocoland etiquette, the arts and entertainment – A’
/p180n-n¯amag(s) (‘Book(s) of
Rules (of Propriety)’) – which from an early date were being translated into
Arabic. Other genres of Middle Persian literature are normative or didactic
works,andarz (‘wisdom’) and courtly (partly epic) texts. Only a small part
of the literature of late Sasanian times has come down to us.17The main
problem with what survives is the difﬁculty of separating historical fromlegendary material and the descriptive from the prescriptive.
Inorder to construct as full a picture as possible, in this chapter I shall deal
with the court society, royal representation and courtly life of both early and
late Sasanian times. For the reigns of Shabuhr I and Narseh, the survivingepigraphic and artistic evidence puts the stress on court society and royalrepresentation. For the sixth and seventh centuries, the emphasis of the
available evidence shifts to court ceremonial, etiquette and entertainment,
and also to the normative styles of courtly communication.
The Sasanian court
The constitution of the court
Andalso here, by means of this inscription, we founded: a ﬁre (temple), Husraw-
Shabuhr by name, for Our soul and (Our) glory; 1 ﬁre (temple), Husraw-Adur-
Anahid by name, for the soul and glory of Adur-Anahid, the Queen of Queens, Our
daughter; 1 ﬁre (temple), Husraw-Ohrmezd-Ardashir by name, for the soul and
15Most of the reliefs have been published in the excellent series ‘Iranische Denkm ¨aler’ (Berlin:
Reimer).
16See Y arshater 1983 ;see also Huyse, in press.
17Anexcellent short history of Middle Persian (Pahlavi) literature was published by Cereti 2001 .

64 josef wieseh ¨ofer
glory of Ohrmezd-Ardashir, the Great King of the Armenians, Our so n…A n dw e
ordered that (the following) should be done [i.e., offerings be made]: for our soul 1
lamb, 1 gr/p180w(and) 5 h¯ofan bread, 4 p¯aswine; for the souls (of) Sasan, the Lord; and
Pabag, the King; and Shabuhr, the King, the son of Pabag; and Ardashir, the King of
Kings; and Xwar(r)anzem, the Queen of the Empire; and Adur-Anahid, the Queen
ofQueens;and Denag,t h eQ u e e n ;a n d… (ˇSKZ mpI 22–5)
The pre-eminence of the royal family within the Sasanian court is always
apparent. The resgestae of both Shabuhr I ( ˇSKZ)and Narseh (NPi) contain
lists of court personnel graded by rank. These give ﬁrst place to the membersof the royal family, including queens and other ‘ladies’ (mpI b¯an¯ug).
18The
genealogical section of ˇSKZ is in three parts. The ﬁrst (mpI 23–4/paI 18–
19/grI 39–43) is reserved for Shabuhr himself, his daughter Adur-Anahid,the ‘queen of queens’ ( b¯ambiˇ sn¯an b ¯ambiˇ sn), and his three sons Ohrmezd-
Ardashir (the later king Ohrmezd I), Shabuhr and Narseh. The second
(mpI 25/paI 20/grI 46) names the royal ancestors: Sasan, the founder ofthe dynasty, the kings Pabag, Shabuhr and Ardashir (I), and X
war(r)anzem,
the ‘queen of the empire’ ( ˇsahr b ¯ambiˇ sn), probably Ardashir’s wife. The third
section (mpI 25–7/paI 20–2/grI 47–52) lists the living members of the royalhouse, again starting with Adur-Anahid and ending with the consorts ofNarseh. In this second list of King Shabuhr’s descendants, members of the
royalhouse are not graded by rank, but probably by age. It has rightly been
stressed that social, not family, status dictated a man’s or a woman’s rankingboth in the royal genealogy and in the royal household. Female members ofthe royal family appear on the royal reliefs as well as on coins; they are alsoimmortalised on gems and seals of their own. These artworks, together withthe ranking of queens, consorts and princesses attested in the inscriptions,show the social prominence of the royal women. It is not surprising to ﬁndwomen even ascending the throne shortly before the fall of the empire: this
was the case with Boran and her sister Azarmigdukht (in the absence of malecandidates, it must be admitted).
Inaddition, the rule of succession was strictly patrilineal and restricted to
members of the Sasanian family. The succession crises of the third centuryAD (Narseh against Wahram III), the fourth century AD (Ardashir II againstShabuhr III), and the sixth century AD (Wistahm against Husraw II) alldemonstrate the difﬁculty of circumventing this rule. But birth and selectionbythe preceding sovereign were only two of the necessary prerequisites for
rulership. There is also the idea that the future king should have divine
grace ( xwarrah ), best understood in modern terms as the ‘charisma’ of
18Manyof these people are now included in Weber’s online prosopography.

King and court in the Sasanian Empire 65
kingship.19Inthe inscriptions of the early kings, legitimacy can also be
claimed by emphasising kinship with preceding rulers: Shabuhr refers to his
father Ardashir, his grandfather Pabag, the eponymous Sasan, and even tothe former great kings of Iran (the legendary Kayanids?).
20
Third, the title m¯azd¯esn bay k ¯eˇcihr az yazd ¯an(‘Mazdean divine Lord,
whose origin [is] from the gods’) borne by the reigning ˇs¯ah¯an ˇs¯ah(‘King of
Kings’) in ˇSKZ shows that the Sasanian kings stressed the Mazdean quality
of their royal power and their own divine nature – although this differedfrom that of the yazd ¯an,i.e. Ohrmezd and the other gods.
21Other royal
males do not share this title with the kings.
Fourth, Shabuhr I founded ﬁre temples ‘for his own soul and glory’ ( pad
am¯ar u w ¯an ud pann ¯am)and for the souls and the glory of his relations
and deceased ancestors, endowing their upkeep. These endowments servedsocial functions, providing material help for relations and friends and aspecial ‘pension’ for the founder’s descendants. But they were also meant toconfer prestige on their royal donor, to foster his subjects’ trust and loyalty
and to reinforce the existing social order. Deceased members of the royalfamily also became objects of organised worship, in a manner similar to theGreek cult of dead heroes.
22The ﬁre temples were normally named after
their founders and benefactors. Th us the ﬁre temple founded by Shabuhr I
for his own soul and glory was given the name Husraw-Shabuhr (‘Glorious
is Shabuhr’).
Finally, the Sasanians practised close-kin marriage ( xw¯ed¯odah ), deemed
meritorious by the Zoroastrian theologians. This served not only to keepproperty within the family, but also to retain kingship within the clan.
23
However,not all royal marriages were incestuous: exogamy for political
reasons is also attested.
Asboth the royal inscriptions and Manichaean texts show, not all mem-
bers of the royal household were permanent members of the royal court. Inparticular, the king’s adult sons and other important relations were in per-sonal proximity to the ruler only when their administrative duties or excep-tional events, such as festivities or wars, required their presence at court,or if the ‘travelling king’ with his entourage happened to visit a prince’sprovince. Thus, we may distinguish a nuclear or inner court of permanentmembers from an extended or outer court of temporary courtiers. In early
19Cf. Gnoli 1999 .
20Fordifferent attempts to identify these ‘ancestors’ and ‘forefathers’ see Daryaee 1995 ,2001 –2;
Shahbazi 2001 ;Wieseh ¨ofer 2002; Huyse 2002 ;K e t t e nhofen 2002 .
21Panaino 2003 :276–83.22Macuch 1992 ,1994 ;Stausberg 2002 :219–20.
23Wieseh ¨ofer 2001 :180, 291.

66 josef wieseh ¨ofer
Sasanian times the inner court mainly seems to have comprised members of
the royal family and the royal household. The great aristocratic landownersand magnates belonged to the outer court, since they were mainly preoc-cupied with managing their estates and their dependants, the peasants andtenant farmers (see below).
Social ranking in the Sasanian empire and court and the recruitment
of ‘courtiers’
This is the range of the arrow shot by Us, the Mazdean divine Lord Shabuhr, the King
of Kings of Eran and Aneran, whose origin is from the gods, the son of the Mazdeandivine Lord Ardashir, the King of Kings of Eran, whose origin is from the gods, thegrandson of the divine Lord Pabag, the king. And when we shot this arrow, we were
shooting before the kings (landholders ( ˇsahrd ¯ar¯an)), the princes ( wispuhr ¯an), the
grandees ( wuzurg ¯an)and the nobles ( ¯az¯ad¯an).
(ˇSH1–6)
Inhis inscription at Hajjiabad,24Shabuhr I mentions four groupings of
aristocrats in order of political prominence and prestige and according to
the dictates of lineage. His son Narseh also refers to these groups in hisPaikuli inscription, showing that they were the most important people to
require royal acknowledgement. Members of the ﬁrst group were called
bythe (singular) title ˇsahrd ¯arin Middle Persian and ‘ruler of the peoples’
(despot ¯es t¯on ethn ¯on)inGreek. They comprised local dynasts and those
sons of the ‘King of Kings’ to whom he had entrusted the government ofkeyparts of the empire. The group ranked second (mpI wispuhr ¯an,g r I hoi
ek basile ¯on)c o m prised members of the royal clan who did not belong to the
ruler’s immediate family (‘princes of the blood’ in more recent parlance).
The third grouping (mpI wuzurg ¯an)was made up of the heads of the seven
most important noble families – the Waraz, the Parthian clans Suren andKarin, the Spahbed, Mihran, Spandiyad and the ‘lords of Andegan’ . Last
came the ¯az¯ad¯an,orrump of the Iranian nobility.
25
Since these groups shared with the king the obligation to sacriﬁce for
the souls of both the living and the dead, Shabuhr’s resgestae list their
membership at the time of inscription both by name and, if appropriate, bycourt or governmental function. But we need to bear in mind that in ˇSKZ a
mighty king is making his report on the state of his empire and court. On theother hand, the Narseh inscription makes clear that the ‘King of Kings’ and
24The inscription describes an archery contest between the king and his suite, in which Shabuhr
‘brok ear e c o r d ’ .
25The ranks of nobility are also mentioned in ˇSKZ 22/17/39f. and NPi 2–3 (85).

King and court in the Sasanian Empire 67
the aristocracy were linked by a network of mutual obligations and common
interests; but also that a ruler in need of political and military support had torespect aristocratic privileges and was heavily dependent on noble consent
and cooperation. In other words, in early Sasanian times social ranking
certainly manifested itself at court. But, as far as the nobility was concerned,
it was not only the king who dictated rank: lineage could be as importantas royal favour. The Sasanians did not create these ‘structures of standing’themselves, but inherited them from the Parthians – while at the same timeenhancing the rank of the Persian, i.e. south-west Iranian, aristocracy. Thisis shown by the ending of the ‘formula’ for presenting the different groupsof nobility in the Paikuli inscription: ‘The landholders and the princes,the grandees and the nobles and the Persians and the Parthians.’
26Loyal
Parthian clans had not had to give up their leading position, but were now
complemented by Persian clans. At a later period, other ‘clans’ rose into the
rank of the magnates.
Asar esult of their high social, political and economic standing, the
leading families were also made to play an advisory and validating partin the proclamation of a new king: for Narseh and his predecessors, we canassume a token consultation of the highest dignitaries of the empire. Thisafﬁrmed the nobility’s ancient role in determining, or at least conﬁrming, thesuccession.
27
While this was going on at the court of Constantine . . . Antoninus [a Roman
deserter] was led into the winter residence of the king [Shabuhr II, J.W.] and was
received there with open arms. He was distinguished with the dignity of the tiara, an
honour by which you may sit at the royal table and by which deserving men amongthe Persians are allowed to give advice and voice their opinions at assemblies.
(Amm.Marc.18.5.6)
Asthe extract clearly shows, under a powerful king outsiders and new men
had a good chance of promotion to a prominent position in the inner court
byarbitrary acts of royal patronage. An outstanding example is the ambitious
Zoroastrian ‘priest’ Kerdir. In the period from the reign of Shabuhr I to that
of Wahram II, he rose to great prominence at court and was even in aposition to advertise his promotion by means of inscriptions. These werecarved onto the fac ¸ades or walls of important royal sites and monuments,
as in this example: ‘The King of Kings Ohrmezd [i.e. the son of Shabuhr I,J.W.] b estowed on me the tiara ( kul¯af)and the belt ( kamar ), and he raised
26NPi 3 (§5) e.a.
27See Skjærvø in Humbach and Skjærvø 1983: 13 and Sundermann 1983 :84ff. A ‘king’s council’
is mentioned in NPi §68, the ‘sham consultation’ in NPi §§73 and 75.

68 josef wieseh ¨ofer
Figure 2.1 Relief of Shabuhr I at Naqsh-i Rajab near Persepolis
myposition ( g¯ah[‘throne’ , i.e. the place near the king, J.W.]) and my dignity’
(pth.ˇsly).28Kerdir is the clearest instance of a dignitary who rose from being a
rather humble ‘courtier’ to the highest position. He owed this success to his
special talents and to his skill in making himself indispensable to each new
king in turn. Close proximity to the king (i.e. the position of one’s g¯ahat
royalpronouncements, audiences and banquets) and function announced
aperson’s standing at the early Sasanian court, and distinctive dress then
advertised this position to a broader public. Among the most prominentmarkers of status were the tiaras ( kul¯af), with certain colours and heraldic
symbols indicative of particular ranks or distinctions. Belts ( kamar )s tudded
with gems and earrings played a similar role.
29
Forthe Iranian aristocracy, however, for a long time the real criterion of
social grandeur was less a title or royal distinction than lineage, and in timesof crisis or under ‘weak’ kings the higher nobility could even force a ruler toput lineage ﬁrst. The Byzantine historian Procopius shows this happeningin the crisis-prone reign of King Kawad (ﬁfth–sixth century), father of thefamous Husraw I: ‘He [Kawad] was mindful of the rule that did not allowthe Persians to transfer any ofﬁces ( archai )t os t r a n g e r s ,b u to n l yt os u c h
28KKZ 4/KNRm 9f. /KSM 5 (Edition: Gignoux 1991 ).29Peck 1992 .

King and court in the Sasanian Empire 69
men who were entitled to the respective position of honour ( tim¯e)through
their lineage.’30
Next he [Husraw/Khusro I] ordered the heads of the leaders of the Mazdakites
tobe chopped off and their possessions to be shared out among the poor and
needy. He killed a large number of those people who had conﬁscated other people’s
possessions, and restored these possessions to their original owners . . . He decreedthat, where those responsible for the upbringing of the children of leading familieshad died, he himself would be responsible for them. He married the girls amongthem to their social equals and provided them with their bridal outﬁt and necessitiesout of the state treasury; and he gave the youths in marriage to wives from noblefamilies, presented them with money for dowries, awarded them sufﬁcient riches,and ordained that they should be members of his court so that he might call uponthem for ﬁlling various of his state ofﬁces . . . He made enquiries about the cavalrymenof the army ( al-as ¯awira ), and those lacking in resources he brought up to standard by
allocating to them horses and equipment, and earmarked for them adequate ﬁnancialallowances . . . He selected good administrators, tax ofﬁcials, and governors, andgavethe persons appointed to these functions stringent orders.
(Tabari 1.897.1ff., trans. C. E. Bosworth)
Aswesaw, the rank of a Parthian or Persian nobleman remained more
or less independent of royal favour until the end of the ﬁfth century AD.
Down to that time, the unruly heads of the great noble houses (the Suren,Karin, Lords of Andegan, and others) admitted only a nominal allegiance
tothe central power. In their hereditary territorial domains they were vir-
tually independent of the king, and royal power and inﬂuence dependedtoalarge degree on effective control of the provincial governors – mostly
members of the royal clan – and on the active support of the majority of thehigher nobles. This state of affairs changed only in the late Sasanian period.Symptomatic is the way in which the wearing of belts, rings, clasps and
other sartorial distinctions now required royal approval. As the Byzantineauthor Theophylactus (1.9) maintains, ofﬁcial rank now acquired greaterprestige than lineage.
31This strengthening of royal power resulted from the
great crisis of state and empire starting in the mid-ﬁfth century.32Crucial
elements in the crisis were the disastrous defeats of Peroz I (AD 459–484)
30Procop. Pers.1.6.13; cf. 1.13.16 (Mihran is in fact the name of a noble clan, J.W.).
31‘Since it is a well-known habit of Persians to bear names according to distinguished positions,
as if they disdained to be called by their birth names’ . See Procop. Pers.1.17.26–18 (a Mihran is
punished by being deprived of a golden hairband): ‘For in that country no one is allowed to
wear a ring or a belt, a clasp or any other object of gold without royal bestowal.’ For other
examples see Tabari 1.990.16f. de Goeje; Theophyl.Sim.3.8 and Dinawari 85,6f. Guirgass.
32That the ﬁfth and sixth centuries AD were crucial times in Sasanian history is proved by themany important developments in domestic politics during that era, such as the emergence of a

70 josef wieseh ¨ofer
against the Hephthalites in the east, resulting in tribute-payments to the
Hephthalite ‘state’ , and several years of drought and famine. Meanwhile, the
twofold burden imposed on the peasants by landlords and state taxes onthe one hand, and by the Hephthalite occupation of parts of the kingdomon the other, prompted a rural exodus and peasant revolts. The peasantsdrew religious and moral inspiration for such actions from the social teach-ings of Mazdak, especially his call for communal ownership. The nobilityhad also been weakened by war casualties, encroachments by the poor, andinternecine strife over how best to handle the crisis. History leaves unclearwhether the chief victims of these uprisings were the great landowners orthe lower nobility.
33Whatever the case, the subsequent reforms by Kawad
Iand his son Husraw I were fundamental.34They extended direct taxation
tothe estates of the landed aristocracy. By establishing a new order for the
nobility and the army, they also tried to change the empire’s social structureand the position of the ruler vis- `a-vis the aristocracy: both the restoration
tothe nobility of its old lands and the reassignment of unclaimed estates
weremeasures carried out at the behest of the king. In addition, new noble
groups were created. The ‘knights’ (mp asw¯ar¯an)were a military nobility
obliged to follow the king on campaign. They seem to have been meant toreplace the old units of aristocratic retainers (who had borne their own arms
and had never really been at the king’s disposal). Arab authors also intro-duce a new (or newly emerged) lower nobility, the dehk ¯an¯an,w h o as the
richest local landowners – sometimes owning entire villages – took over the
administration of a village. They owed their elevation to the king, who hadgranted them land, money and other assistance. They were to serve as local
counterweights to the high aristocracy, who were critical of royal policy, and
the potentially rebellious peasantry, and could also, if necessary, offer theking military support.
35Tabari’s report of Husraw’s reforms, quoted above,
states unambiguously that the late Sasanian court changed too. As we sawearlier, the inner court had previously consisted of members of the king’spersonal household (family members and domestic staff), with other highernobles being included only as temporary visitors to the (outer) court. ButHusraw’s inner court now comprised both his relations and members of
akind of service nobility ( Dienstadel ). Hand-picked and promoted by the
ruler himself, these last were more loyal to him than to their clans of origin.
clerical hierarchy in Zoroastrianism modelled on the state hierarchy, and a new emphasis in
dynastic legitimation on the mythological Kayanid link, etc.
33Forthe Mazdakite movement see Sundermann 1977; Guidi and Morony 1991 ;Wieseh ¨ofer
2007a.
34Rubin 1995 .Arather different picture is drawn by Howard-Johnston 1995 .
35Forthe ‘knights’ and dehk ¯an¯ansee Altheim and Stiehl 1954 :129–74; 1957 :57–81.

King and court in the Sasanian Empire 71
Itis this kind of court which is mirrored in most of the Middle Persian
literary works (see below).
Under Husraw’s immediate successors, however, tensions between king
and high aristocracy reappeared. It has been suggested that the king rapidly
lost control of the knights, who reverted to being retainers of the great –and virtually independent – landowners. And right from the start the king’ssupreme military commanders must have been powerful territorial lords.
36
Asthe political inﬂuence of the great landowners revived, they began to
re-create their own armed retinues of ﬁghting men and to make their own
domains ﬁscally independent. As in ear ly Sasanian times, these powerful and
ambitious nobles once more became only occasional visitors to the court.Bycontrast, members of the king’s inner court risked losing their political
clout under a ‘weak’ ruler and becoming ‘courtiers’ in the strict sense of theword.Brieﬂy checked in their ambitions by Husraw II’s centralising of the
ﬁnancial administration, the landed and military aristocracy still managedtoconspire against him, reproaching the king for his tyrannical treatment of
the nobility, his ruinous exaction of land taxes and his bloody wars againstByzantium. After Husraw’s death, kingship became the instrument of differ-
ent factions of the aristocracy. The rapid advance of the Muslim army andthe sudden collapse of Sasanian sovereignty in Iran are eloquent testimonytothe paralysis induced by the particularist interests of the leading classes
of the empire in this last phase of Iran’s pre-Islamic history.
The ‘stateliness’ of the court and structures of
communication at court
(Offerings shall be made for the beneﬁt of the souls of those who live under the
reign of Shabuhr, the King of Kings:) For Ardashir, king of Adiabene; for Ardashir,
king of Kerman; Denag, the queen of Meshan, the dastgerd of Shabuhr; Hamazasp,
the king of Iberia; Walakhsh, the prince ( wispuhr ), the son of Pabag; Sasan, the
prince, brought up (in the house of) Farragan; Sasan, the prince, brought up (inthe house of) Kadugan; Narseh, the prince, the son of Peroz; Narseh, the prince, theson of Zadsprakhm; Shabuhr, the ‘viceroy’ ( bidaxˇ s:‘second in command’); Pabag,
the chiliarch ( haz¯aruft ); Peroz, the ‘master of the cavalry’ ( aspbed ); Ardashir (from
the house of) Waraz; Ardashir (from the house of) Suren; Narseh, the lord ofAndegan; Ardashir (from the house of) Karin; Wohnam, the ‘commander-in-chief’
(framadar ); Frig, the satrap ( ˇsasab )ofWeh-Andiyok-Shabuhr; Sridoy (with the
surname) Shahmust; Ardashir (with the surname) Ardashir-Shnom; Pachihr (with
36Rubin 2000 :657.

72 josef wieseh ¨ofer
the surname) Tahm-Shabuhr; Ardashir, the satrap of Goyman; Chashmag (with the
surname) New-Shabuhr; Wohnam (with the surname) Shabuhr-Shnom; Tirmihr,the commander-in-chief of the fortress of Shahrkerd; Zig, the master of ceremonies(ay¯en
/p180g,g r I deipnokl ¯etor:‘caller to the meal’); for Ardawan (from) Demawend;
Gundifarr, the son of Ewag (?); Pabig (?) (with the surname) Peroz-Shabuhr, the
son of Shanbid; Warzan, the satrap of Gay; Kerdsraw, the ‘viceroy’; Pabag, the sonof Wisfarr; Walakhsh, the son of Seluk; Y azadbed, the ‘adviser’ ( handarzbed )o f
the queens; for Pabag, the ‘sword-bearer’ ( ˇsafˇs¯el¯ar); Narseh, the satrap of Rind;
Tiyanag, the satrap of Hamdan; Wardbed, the ‘master of the servants’ ( paristagbed );
Djoymard, the son of Rastag; Ardashir, the son of Wifr; Abursam, the son of Shabuhr,the ‘head of the palace guards’ ( dar
/p180g¯an s¯al¯ar); Narseh, the son of Barrag; Shabuhr,
the son of Narseh; Narseh, the ‘master of supplies’ ( grastbed ,g r I ho epi t ¯esann¯on¯es);
Hormezd, the ‘master of scribes’ ( dib/p180rbed), the son of Hormezd, the ‘master of
scribes’; Nadug, the ‘master of the state prison’ ( z¯end¯an/p180g); Pabag, the ‘master of the
gate’ ( darbed ); Pasfal, the son of Pasfal; Ewakhsh (?), the son of the ‘commander of the
fortress’ ( dizbed ); Kerdir, the ‘priest’ (¯ehrbed ); Rastag, the satrap of Weh-Ardashir;
Ardashir, the son of the Viceroy; Mihrkhwast, the ‘treasurer’ ( ganzwar ); Shabuhr,
the ‘commander’ ( fram ¯ad¯ar); Ashtad, the ‘scribe of contracts’ ( dib/p180r/p180Mihr¯an)f r o m
Ray; for Sasan, the eunuch ( ˇs¯abest ¯an), the son of Sasan; Wiroy, ‘the supervisor of
exchange’ ( w¯az¯arbed ); Ardashir, the satrap of Niriz; Baydad, the son of Wardbed;
Kerdir, the son of Ardawan; Zurwandad, the son of Bandag; Winnar, the son of Sasan;
Sasan, the judge ( d¯adwar ); Wardan, the son of Nashbed (?); Gulag, the ‘master of
the wild boars’ ( w¯ar¯azbed ).(ˇSKZ mpI 31–5)37
Inhisresgestae ,Shabuhr I lists the dignitaries, ofﬁcials and aristocrats of his
empire whose proximity to the ruler, if only as occasional members of the
outer court, entitled them to have offerings made on behalf of their souls.Other inscribed lists of this kind have come down to us, one of them in theresgestae of the second Sasanian king, and several more in Narseh’s Paikuli
inscription. They are all similarly arranged, starting with the members ofthe royal house, followed by members of the (seven) most important nobleclans and ending with other dignitaries and ofﬁcials. As far as Shabuhr’slist ( ˇSKZ)isconcerned, the arrangement of names seems to result from a
particular combination of personal and political considerations. In otherwords, the list is evidence both for the dignitaries’ personal relationship to
the king and for Shabuhr’s appointment of people to ofﬁce on the basis ofdecisions of character and/or talent.
Ardashir, the king of Adiabene, heads the sixty-seven dignitaries of
Shabuhr’s court. As he is only mentioned in ˇSKZ,wecannot be sure if
he owed his outstanding position to a personal relationship with the king
37Manyof these individuals appear in Weber’s prosopography.

King and court in the Sasanian Empire 73
or to the importance of his province at that time – or both. The consoli-
dation of royal power under the ﬁrst two Sasanids probably explains theconsiderable increase in the size of the outer court. The court of King Pabag
(Shabuhr’s grandfather) had consisted of a mere eight members; Ardashir Iappointed thirty-one dignitaries; Shabuhr I doubled their number. In otherwords, empire-building generated greater complexity in the court.
38
He[the prophet Mani, J.W.] went to Belabad (B ¯etL¯ap¯at./W¯eh-Andiy ¯ok-ˇS¯abuhr),
the place of suffering . . . They [the Magi, J.W.] . . . charged him (Mani) before Kardel
[Kerdir, the mowbed ,J . W .]. And Kardel told it to the synkathedros .T h e n they both
went to see the counsellor to the king ( magist ¯or)and told him about the charges. And
themagist ¯orbrought them forward to the king. After he had heard about them . . .
he sent for my lord. On Sunday, Mani entered Belabad. On Monday, he was accused.On Sunday, the king gave out an order and sent for him.
(Manichean Homilies, 45.9–23 Polotsky)
Andhe[Wahram I, J.W.] stood up from his meal; and, putting one arm around the
queen of the Sakas and the other round Karder [Kerdir, J.W.], the son of Ardawan,he came to the Lord [Mani, J.W.].
(Manichean T ext from Turfan M 3.15–20)
Although we cannot be sure about the exact functions implied by the terms
synkathedros and magist ¯or,itisclear from this passage that in order to get
an audience with the king in early Sasanian times, people had to go throughthe proper channels. King Wahram, who is dining with two close ‘friends’(as shown by the royal gesture of embrace), orders Mani to wait. At the endof the meal, he goes over to the waiting ‘prophet’ and makes clear that he isnot welcome.
Anelaborate courtly protocol was also part of late Sasanian diplomatic
procedures.
39Intheir self-representation, the two great powers of Iran and
38Sadly there is no documentation for the Arsacid court comparable to these early Sasanian lists.
Ifthere were, we might be able to recognise the special Sasanian traits of court ofﬁce and court
society. Even more regrettably, we lack a comparable account of the inner and outer court oflate Sasanian times: Byzantine historians give detailed reports of Persian diplomatic missions tothe Roman emperor (see Constantin.Porphyr. de Caeremoniis 1.89f. Bonn), but show little
interest in the diplomatic trafﬁc in reverse. Even Menander the Guardsman, who has much tosay about Byzantine–Sasanian diplomacy, gives no description of Husraw’s court. And theIranian reports either are of literary rather than historiographical character (the contemporary
worksofMiddle Persian literature, see below), or are New Persian or Arabic adaptations of
late-Sasanian works (Firdausi, Shahnameh ,etc.). These have to be approached with great
caution, since they are not simply translations, but in fact are epic or historiographical textsrecast in the idiom of their time of composition and with distinctive Islamic touches
(Abka’i-Khavari 2000 ,auseful collection of sources, is quite uncritical in this respect).
39Forlate Sasanian diplomacy see G ¨uterbock 1906 ;S a k o1986 ;Winter 1988 ;L e e 1993 ;
Wieseh ¨ofer, 2007c .

74 josef wieseh ¨ofer
Rome/Byzantium rhetorically stressed their respective superiority. In prac-
tice both sides had to recognise their equal might and achieve a modus
vivendi. Unsurprisingly, peace treaties between the two powers were drawn
up with particular ceremony. In the preamble to a Sasanian document ofratiﬁcation ‘the divine, good, father of peace, ancient Chosroes [Husraw I],
king of kings, fortunate, pious and beneﬁcent, to whom the gods have givengreat fortune and a great kingdom, giant of giants, formed in the image of
the gods’ calls his Roman opposite n umber ‘Justinian Caesar, our brother’ .
Even if the titulature given to the Byzantine emperor is clearly briefer than
his own, the form of address (‘brother’) nevertheless reveals the ‘king whoreigns over kings’ and the ‘victor of wars’ conceding to the ‘lord of all
things and of the world’ (Amm.Marc.19.2.12) equal rank in a diplomaticcontext. This emerges with particular eloquence and colour in the words
which Byzantine authors such as Petrus Patricius and John Malalas ascribetoSasanian kings and their diplomats. The two empires are described as
two lights, which, ‘like eyes, are adorned by each other’s light’ (Petr.Patr. fr.13 M ¨uller), or as two divinely ordained centres of civilisation, ‘the moon of
the west’ and ‘the sun of the east’ (Ioh.Mal.18.44 (p. 449) Thurn). Equally,Rome/Byzantium granted parity of rank and dignity to the Sasanians, even
though many Romans in fact wished total annihilation on the eastern foe(Amm.Marc.23.5.19), not least because its existence challenged Roman pre-tensions to universal empire. It was also customary for the two great powers
toannounce new reigns to each other by means of a special envoy, and to
respond with messages of congratulation,
40as well as exchanging gifts and
asking after the well-being of the royal ‘brother’ when each received theother’s embassies in solemn audience.
41
Royal representation: manners and the cultural example of the court
When Ardashir was 15 years old, news reached Ardawan [the last Parthian king,
later on deposed by Ardashir] that Pabag had a son who was accomplished in the
chivalrous arts . . . [Ardawan asks Pabag to send him his son, and the latter does not
dare disobey the command]. When he [Ardawan] saw Ardashir, he rejoiced, heldhim in esteem, and ordered that he should go hunting and to the ball game [polo]with his sons and knights every day. Ardashir did so. With god’s help he became
40Ioh.Mal.18.34, 36 (pp. 445, 448) Thurn; Men.Prot. fr.9.1 Blockley; Th.Sim.3.12;
Theophan. Chron .250 de Boor; Chr.Pasch .735 Dindorf.
41Petr.P atr.apud Constantin. Porphyrogen., de Caeremoniis 1.89, 90 Bonn; Procop. Pers.1.24; gifts
of the Augusta to the Persian queen: Ioh.Mal.18.61 (p. 467) Thurn.

King and court in the Sasanian Empire 75
more competent and more clever than all of them in the ball game, in riding, in
chess and in backgammon. (K¯arn¯amag /p180Ardaxˇs/p180r/p180P¯abag ¯an 2.5, 10–12)
The ‘Ardashir romance’ , written in the late Sasanian period and subsequently
revised,42retrojects the social conditions of its date of composition into the
period of the empire’s founder. For this reason it is sometimes seen basicallyas a description of the court of the last Sasanians.
43Itis certainly true that
after Husraw’s reforms and the creation of a service nobility, there was anew stress on good breeding in the social milieu of the king, where the sonsof this nobility spent time both as hostages for their father’s loyalty and as‘courtiers’ . Obedience, elegant manners, culture, games and hunting were
expected and practised. It is unsurprising, then, that late Middle Persian
andarz texts (‘wisdom literature’) or their Arabic translations include works
which employ royal utterances and speeches from the throne to discuss orprescribe the proper character, behaviour and appearance at court (at meals,special festivities, audiences, etc.) not only of the king, but also of his sub-jects ( bandag ¯an). In their original form these works were probably meant
tofoster belief in a divinely sanctioned political and social hierarchy in the
empire generally and, more speciﬁcally ,atthe royal court. The unique posi-
tion of the king
44manifests itself in his dress, jewellery, headgear, crown and
throne, in his luxurious habits45and, last but not least, in the splendour and
architectural plan of his residential palaces. It is clear from Arabic trans-lations that the Arab conquerors of Iran in the seventh century AD werehugely taken with the enormous crown of Husraw II
46and with the giant
royalcarpet known as ‘Khusro’s spring’ in the winter palace at al-Mada’in.47
Middle Persian texts like the famous Husraw /p180kaw¯ad¯an ud r ¯edag- ¯e
(‘Husraw and his page’)48present the court of Husraw I and his succes-
sors as the home of savoir vivre .Waspuhr, a poor and unemployed young
man, presents himself to the king, asking to be tested on his knowledge ofluxurious living, which embraced ﬁne cuisine, including the preparation ofjellied meats, ragouts, preserves and stewed fruit; music; ﬂowers and their
42Foranexcellent edition with a French translation and commentary see Grenet 2001 .
43Forlate-Sasanian court culture see Altheim 1962 :195–212.
44Forthe titulature of the kings see above.
45Forthe respective sources see Abka’i-Khavari 2000 .
46Forthe famous crown of Husraw II see Tabari 1.2446.11ff. de Goeje.
47Tabari 1.2452.7ff: ‘Sixty times sixty yards as a single carpet by the dimension of its surface, on
which the paths formed ﬁgures, the separating parts rivers, the intervals between them hills. On
its border earth sown with spring growth out of silk against branches of gold, and its blossomsof gold, silver and the like.’ See Morony 1989.
48Edition: Unvala 1921 .For the character of the text (and other similar texts) see Cereti 2001 :
178–84.

76 josef wieseh ¨ofer
Figure 2.2 The deer hunt from the relief of Khusro II at Taq-i Bustan
scents; and connoisseurship of women and steeds for riding. This text also
itemises the various different arts of warfare, every kind of board game,and seventeen different types of quarry on the royal hunts.
49This last list
includes game birds, showing that the ‘sport’ par excellence of the Iranian
rulers served not just as a test of strength, but also as a source of victuals.50
Husraw’s interest in foreign games like chess51relates to his promotion of
scholarship and the arts, a well-known aspect of royal self-representation.
Chosroes [Husraw I] has been praised and admired quite beyond his deserts not
just by the Persians but even by some Romans. He is in fact credited with being alover of literature and a profound student of philosophy and somebody is supposedtohave translated the works of Greek literature into Persian for him. It is rumoured
49This interesting list comprises bulls, wild asses, stags, boar, young camels, calves, buffaloes,
gazelles, hares, rabbits, partridges, pheasants, larks, cranes, bustards, ducks and peacocks.
50Forthe royal hunt see the famous hunting reliefs of the Taq-i Bustan grotto near Kirmanshah
(Tanabe 1983 ).
51The Middle Persian text Wiz¯ariˇsn ¯ıˇcatrang ud nihiˇ sn ¯ın ¯ew-ardaxˇ s¯ırtells the story of chess’s
introduction to Iran (edition and commentary: Panaino 1999 ;see also Daryaee 2002 ).

King and court in the Sasanian Empire 77
moreover that he has absorbed the whole of the Stagirite [Aristotle] more thoroughly
than the Paeanian orator [Demosthenes] absorbed the works of the son of Olorus[Thucydides], that his mind is ﬁlled with the doctrines of Plato the son of Aristonand that not even the Timaeus, bristling as it does with geometrical theorems andscientiﬁc speculations, would elude his grasp, nor for that matter the Phaedo or theGorgias or any other of the polished and more intricate dialogues, as for instancethe Parmenides. Personally, I could never bring myself to believe that he was soremarkably well-educated and intellectually brilliant.
(Agathias 2.28.1–3.1, trans. J. D. Frendo)
Inaspirit of hyper-patriotism Agathias challenged the admiration for the
enemy king found in Syriac literature, where Husraw was described as a
‘wise king’ or one ‘who has read all the books of the philosophers’ .52Healso
calls into question the claims and motives of Husraw himself, as found inthe royal resgestae (k¯arn¯amag ):
Wehave made inquiries about the rules of the inhabitants of the Roman empire and
the Indian states . . . We have never rejected anybody because of his different religion
or origin. We have not jealously kept away from them what we afﬁrm. And at thesame time we have not disdained to learn what they stand for. For it is a fact that tohave knowledge of the truth and of sciences and to study them is the highest thingwith which a king can adorn himself. And the most disgraceful thing for kings is to
disdain learning and be ashamed of exploring the sciences. He who does not learnis not wise.
(Ibn Miskawaih 206.4ff. Caetani)
Despite the unmistakable note of self-praise here, the king’s efforts to pro-
mote higher learning are undeniable. Agathias himself (2.30f.) reports thatHusraw offered a welcome to the Neoplatonic philosophers, homeless after
the closure of their school in Athens. When their disappointment with Iranand its inhabitants led them to return home, he insisted on their exemp-tion from punishment by the Romans during his peace negotiations withByzantium.
53The king’s discussions with Zoroastrian, Christian and other
experts about questions of cosmogony and the end of the world, about God,
primary matter and the elements are famous.54Husraw’s interest in the east
is shown by his commissioning a translation of the Indian book of fables,thePanchatantra ,which the physician Burzoy had brought from India.
55
Besides philosophy, theology and statesmanship, Husraw was also inter-ested in foreign contributions to law and medicine. In medicine, Iranianand Indian as well as western traditions were assimilated – they are reported
52Barhebraeus Chron.Eccl .3.91.8f. Abbeloos-Lamy; Michael Syr.366 v. 11f. Syr; 2.339 trans.
Chabot. See Cameron 1969 /70, esp. pp. 172–4.
53Cf. Hartmann 2002 .54Wieseh ¨ofer 2001 :217, 299.55de Blois 1990 .

78 josef wieseh ¨ofer
byBurzoy, himself a physician from Nishapur, in his introduction to the
collection of fables. According to one Arabic source, Husraw I even wrote
amedical book himself, a compilation based on Greek and Indian works.
Through Sasanian–Middle Persian mediation not only medical and phar-
maceutical literature from both east and west, but also Romano-Byzantineagricultural writings and the Almagest of Ptolemy, found their way into
Arab ic literature.
56
Ingeneral, the late Sasanian period witnessed a literary renaissance, much
of it sponsored by the royal court. In particular, Husraw I Anushirwan andhis successors are credited with the promotion of literature: Weh-Shabuhr,the head of the Zoroastrian clergy under Husraw I, is said to have publishedthe twenty-one nasks of the Avesta; in addition the X
wad¯ay-n ¯amag (‘Book
of Lords’), the semi-ofﬁcial ‘Iranian National History’ , evidently acquiredan initial authoritative form under Husraw, before being repeatedly revised(and continued) in later times.
57Finally, numerous compilations of andarz
texts, as we have seen, are attributed to Husraw and his entourage, and even
the composition of such texts by the king himself.
Conclusion
Sadly the Sasanian empire and monarchy are still neglected by ancient his-torians.
58Although the source material, as we have seen, is unbalanced and
difﬁcult to handle, this preliminary contribution aims to show that the Sasa-nian court has a part to play in the further development of ancient courtstudies. Sasanid Iran shaped Near Eastern history and cultures for morethan four hundred years as the great opponent and neighbour of Romeand Byzantium. It also heavily inﬂuenced neighbouring cultures to botheast and west, and in turn was greatly inﬂuenced by them. Earlier stud-ies of Sasanian ‘feudalism’ all too often drew unjustiﬁed and inappropriate
56Forthe intermediary role of Sasanian Iran in law and science see Gignoux 2006 ;Ullmann 1978 ;
Richter-Bernburg 1998 ,1999 ,2000 .
57Itwas from a New Persian prose translation of the ‘Book of Lords’ , collected by Zoroastrians in
eastern Iran and compiled by Abu Mansur Ma’mari, secretary of the municipal head of Tus,
that Abu’l-Qasim Mansur Firdausi drew his material when, between the years 975 and 1010, hedecided to collect the stories about pre-Islamic Iran into an epic poem, the Shahnameh .
However,although the poet’s Sasanian kings are particularly vivid and complex characters, and
although the Shahnameh givesdetailed information about king, court and royal representation
in the Sasanian empire, we have to see it not as a historical source for Sasanian institutions andconditions, but as a complex mix of reality and ﬁction, of historical information and
contemporary attitudes towards the past.
58Cf. Wieseh ¨ofer 2006 .

King and court in the Sasanian Empire 79
parallels between Sasanian Iran and the medieval European monarchies.
Nonetheless, the theoretical parameters of modern court-studies, based on
late medieval and early modern courts, have proved useful in rationalisingthe source material for the Sasanian court and power and ‘state-building’in Sasanid Iran. However, much remains to be done. On the problem of thesources, we urgently need a fresh analysis of the Arabic and New Persiantexts in the light of the extant late Sasanian, Byzantine and Syriac litera-
ture. The possibility of the mutual inﬂuence of Iranian and Byzantine courtinstitutions merits closer examinatio n(see Smith in this volume). The ety-
mological ﬁelds ‘court’ , ‘rank’ and ‘dignity’ in Middle Persian and Parthiandeserve more research. And ﬁnally, more archaeological research on palacearchitecture and royal representation is needed.

Table 2.1. Rulers of the Sasanian dynasty
Number Date AD Name Indigenous name Genealogy Commentary
1 224–241/2 Artaxerxes (Artaxares) I mp Arda ˇs/p180rS o n of P¯abag Founder of the Sasanian empire
2 239/40–270/2 Sapor(es) I mp ˇS¯abuhr Son of 1 Co-regent with 1 until 241/2
3 270/2–273 Hormisdas (Hormizdes) I mp Ohrmezd-Arda ˇs/p180rS o n of 2
4 273–276 Wahram (Va(ra)ranes) I mp Wahr ¯am So no f2
5 276–293 Wahram (Va(ra)ranes) II mp Wahr ¯am So no f4
6 293 Wahram (Va(ra)ranes) III mp Wahr ¯am Son of 5 Contest for the throne with 7
7 293–302 Narses mp Narseh Son of 2 Contest for the throne with 6
8 302–309 Hormisdas II mp Ohrmezd Son of 7
9 309–379 Sapor(es) II mp ˇS¯abuhr Son of 8
10 379–383 Artaxerxes (Artaxares) II mp Arda ˇs/p180rS on (brother?) of 9
11 383–388 Sapor(es) III mp ˇS¯abuhr Son of 9
12 388–399 Wahram (Va(ra)ranes) IV mp Wahr ¯am Son (brother?) of 11
13 399–421 Y azdgird I (Isdigerdes) mp Y azdgerd Son of 1214 421–439 Wahram (Va(ra)ranes) V mp Wahr ¯am (G ¯or) Son o f1 3
15 439–457 Y azdgird II (Isdigerdes) mp Y azdgerd Son of 14

16 457–459 Hormisdas III mp Ohrmezd Son of 15
17 459–484 Peroz(es) mp P ¯er¯oz Son of 15; brother of 16
18 484–488 Balas (Blases) mp Walax ˇsS on of 15; brother of 17
19 488–496; 499–531 Kabades I mp Kaw ¯ad Son o f1 7
20 496–498 Zamasphes (Zames) mp Zam ¯asp Son of 17; brother of 19
21 531–579 Chosroes I mp Husraw (Xusr ¯o) Son of 19 Epithet An ¯oˇsirw¯an
22 579–590 Hormisdas IV mp Ohrmezd Son of 2123 590–628 Chosroes II mp Husraw (Xusr ¯o) Son of 22 Epithet Abarw ¯ez
24 590–591 Wahram (Va(ra)ranes) VI mp Wahr ¯amˇC¯ob/p180nR ival claimant to the throne of 23
25 628 Kabades II mp Kaw ¯ad Son o f2 3
26 628–630 Artaxerxes (Artaxares) III mp Arda ˇs/p180rS o n of 25
27 630 Shahrbaraz mp ˇSahrwar ¯az
28 630 Chosroes III mp Husraw (Xusr ¯o) Nephew of 23
29 630–631 Boran mp P ¯ur¯an Daughter of 23 Queen
30 631 Azarmidukht mp Azarm /p180gduxt Daughter of 23; sister of 29 Queen
31 631–632 Hormisdas V mp Ohrmezd Grandson of 2332 631–633 Chosroes IV mp Husraw (Xusr ¯o)
33 633–651 Y azdgird III (Isdigerdes) mp Y azdgerd Grandson of 23

3 The court of Alexander the Great between
Europe and Asia
tony sp awforth
Introduction
Inabook on ancient courts, this chapter1arguably is an oddity, since alone
of the contributions its focus is a single ancient ruler, who himself was
scarcely typical, given not only the freakish scale of his conquests, but alsohis plans, novel in a Greek context, for stabilising his dominion over them.ButAlexander is simply too important a ﬁgure to leave out. Partly this is
amatter of the ancient evidence. Given the ancient preoccupation with
Alexander’s military achievements, it can come as a surprise to ﬁnd that heis the only Macedonian or Hellenistic ruler about whose ‘court’ ceremonieswehave some detailed knowledge (Weber 1997 :43). This evidence leaves us
in no doubt that Alexander attached real political signiﬁcance to ‘holdingcourt’ . It is clear too that, like his father, Philip II, or Augustus in a later
age, Alexander was a consummate ‘master of self-representation’ .
2Here a
reappraisal is offered of the evolving role which Alexander assigned to his
‘court’ and its ceremonies in the period after his invasion of the Persian
empire (334 BC), with particular emphasis on a neglected topic, namely,the physical space constituted by the royal quarters, including his feasting-and audience-tent ( tente d’apparat ). The underlying assumption throughout
is that Alexander from the outset was intent on retaining his conquests inAsia. This is shown by his maintenance of Persian structures of control
and exploitation as soon as Persian territory began to fall into his hands(Arr.1.17.1): that is to say, whatever his ultimate aims may have been whenhe ﬁrst invaded, he was consciously engaged from the start in what modernhistorians would call ‘state-building’ . It is in this light that the role of his‘court’ seems worth reappraising.
1This chapter has been immeasurably improved by interaction with participants at the Newcastle
workshop ‘The Court and Court Society in Ancient Monarchies’ in May 2004. I am particularly
grateful to Maria Brosius, Simon Hornblower and John Moles for discussion and comments on
earlier drafts. Responsibility for the ﬁnal content, of course, is mine alone.
2Carney 2000a :275, of Philip.

The court of Alexander the Great 83
Issues of deﬁnition
Itis nothing new for modern scholarship on Macedonian kings to refer
to‘the court’ and to ‘courtiers’ , although it is fair to say that the assump-
tions underlying these usages tend not to be spelt out. The only detailed
study of Alexander’s ‘court’ is that of Helmut Berve, who assigned to ‘derk¨onigliche Hof’ the ﬁrst third of his tripartite study of the institutions of
Alexander’s empire ( 1926 ,I:3–100). Since he came too early to be inﬂuenced
bythe more theoretical and sociological approaches to court studies trig-
gered ultimately by Norbert Elias, his study may now seem to lack conceptualrigour (Herman 1997 :201). That said, and quite apart from the fact that his
researches provide the basis for all further work, Berve also took for granted
the historical importance of the ‘court’ in Alexander’s monarchy. This isbyno means always the case in Alexander-studies. There is a long-standing
tendency, reaching back to antiquity indeed, to sound a negative note in con-
fronting the aulic side to Alexander’s reign. Symptomatic is the dismissalof Chares, Greek author of a lost history which took a marked interest inAlexander’s ‘court’ , as a ‘triﬂer’ by one distinguished historian of Alexander(T arn1948 ,II: 70).
3This viewpoint seems to reﬂect the traditional censori-
ousness of modern historiography in the face of the nowadays ‘moribund’social formation of the court (Elias 1983 :8;Winterling 1997 :151–2). For
rather different reasons, the court as an analytical category was omitted from
the most exhaustive study of Macedonian royal institutions yet published(Hatzopoulos 1996 ), whose author was committed to a ‘constitutional’ view
of the Macedonian monarchy.
4Inthe long-running modern debate about
the nature of Macedonian kingship, Hatzopoulos’ position runs counter tothe tendency in recent years, with which this chapter is in essential agree-ment, to emphasise the informal nature of royal power, the absence of muchin the way of ‘constitutional’ checks and balances, and the conduct of day-to-day politics as a ‘power game’ between the king, the Macedonian elite
and the people (V ¨olcker-Janssen 1993 :33; Errington 1996 ). On this view,
the king was ‘basically an autocrat limited principally by his relative powervis-`a-vis the Macedonian nobility’ (Anson 1985 :316; cf. Carney 2000b :6).
Afocus on the court of the Argeads, as Alexander’s dynastic lineage was
known, may offer a new perspective on this debate, given that historically‘court culture has often been associated with absolutism’ (Asch 2003 :80).
3The implicit criticism seems to be of the subject-matter which Chares focused on. As a recorder
of fact, Chares was certainly not above criticism: see Badian 1981 :50–1.
4See the reviews of Archibald 1999 and, more critical of the main thesis, Borza 1999 and L ´evy
2001 .

84 tony spawforth
So far I have preferred to use scare-quotes around ‘court’ . How appro-
priate, in fact, is a court-centred analysis to the study of the fourth-century
Macedonian monarchy, and in particular to Alexander and his (after 334
BC) usually roving entourage? Few would dispute that fourth-century BCMacedon was a state in which power and decision-making were centred
on the ruler’s house (or tent when, as commander-in-chief, the king ledthe Macedonian army to war). Alexander’s royal entourage in Asia exhibitsmost, if not all, the characteristics of a court, as deﬁned in modern theory(Winterling 2004 :89–90). It can be divided into an inner and an outer court
(Butz and Dannenberg 2004 :12), the ﬁrst centred on the ‘service’ ( thera-
peia)which followed him in Asia and the other on people with whom he
socialised on a daily basis (see below). The second comprised high-statusindividuals within the empire who were not normally a physical part ofthis immediate entourage, such as members of the royal family left behindin Macedon, or elite-administrators (his provincial governors or satraps,notably). It also included visitors of standing – increasing in numbers, nodoubt, as Alexander’s successes mounted – such as the numerous ‘peo-ple sojourning’ ( parepid¯ emountas )w h ow e r e included by Alexander in the
wedding festivities at Susa (Appendix, especially B, 7). Some of these vis-
itors were prominent members of communities under Alexander’s sway,such as the arms-dealer ( hoplophulax )Gorgus, a leading citizen of Greek
Iasos in south-west Asia Minor, said to have ‘passed time ( diatrib¯ on)with
Alexander’ at Ecbatana in 324 BC, when he successfully brokered a dealon behalf of his Iasian fellow-citizens.
5Ap r i v ileged category (to which
Gorgus may or may not have belonged) comprised Alexander’s ‘ritualisedfriends’ ( idioxenoi ), who were given seats of honour at the Susa wed-
dings (Appendix, A, 1–2). These high-stat us people were non-Macedonian
(Berve 1926 ,I:62–3), and were likely to include non-Greeks as well as
Greeks (Herman 1987 :12), and in the nature of this particular type of
social relationship their home-communities would be far away (Herman1987 :31).
Inthe classic fashion of rulers’ courts, Alexander’s close court incorpo-
ratedrepresentatives of the upper classes of his kingdom among his per-
sonal servants. The ‘royal boys’ ( basilikoi paides ), drawn from the leading
5SIG3312.7–8; Bosworth 1988 :253 with refs.; Herman 1987 :85–6. I take this meaning of
hoplophulax tobe implied by his alleged offer to Alexander of ten thousand suits of armour,
along with catapults and other weapons, at Ecbatana in 324 BC, which presumably Gorgus
intended to provide from his own resources: Ephippus, FGrH 126F5=A t h e n.12.538b. His
huge wealth obviously reﬂects the vast fortunes to be made from supplying Alexander’s
war-machine. Heisserer ( 1980 :170) took the post to be an ofﬁcial one in Alexander’s army.

The court of Alexander the Great 85
Macedonian families (at least one royal prince, the future Amyntas II, was
said to have taken his turn: Ael. VH12.43), performed a mix of domestic
services, such as waiting on the king at table, and outside duties, notably
attendance on the royal hunt, which, taken together, justify the conven-tional equation with the royal pages of more recent European monarchies.The institution was clearly intended to promote ideals of personal serviceand loyalty to the monarch, as suggested by its function ‘as a kind of semi-nary’ for Macedonian ofﬁcers and generals (Curt.8.6.6).
6Under Alexander,
and possibly earlier Macedonian monar chs, domestic servitors of the king
included adult members of the Macedonian elite (see below, on Ptolemy).Itneeds stressing that service of this kind, glossed in Ptolemy’s case as ‘dis-
tinguished and honourable’ ,
7was profoundly alien to the Greeks and later
the Romans, for whom the pages were comparable to, or even mistaken for,chattel slaves (Ael. VH12.43; Curt.8.6.2).
The other key group in this inner court comprised the Hetairoi (‘Com-
panions’), Macedon’s traditional ofﬁcer class, since Philip’s time recruitedbythe king personally, but continuing to include members of established
families whose lineages and, perhaps, landed bases gave them a certain inde-pendence from the king.
8Socially this group dominated the daily entourage
of the king. It furnished him with his personal bodyguard, under Alexander ahand-picked and particularly favoured group normally seven-strong, trans-formed by their privileged access to the king into brokers of royal graces –Hephaestion, Alexander’s childhood friend, is the chief exhibit here (refs.
at Berve 1926 ,I:63).
9The Companions also provided the pool from which
Alexander selected his daily companions at the royal table (below). Sincethese two, socially interconnected, groups of royal pages and Companionsprovided the chief internal opposition under both Philip and Alexander, itis likely that their incorporation into the inner court served a monarchi-cal strategy of promoting the personal allegiance of the Macedonian elitestothe king. On one level, Alexander in particular seems to have used the
routines of his domestic and social life to try to manage these groups in a
manner reminiscent of the Eliasque reading of the Bourbon court – thus a
6Hammond 1990 ;Bosworth 1995 :90–3.
7See Athen.4.171b–c citing Artemidorus: epiphan¯ es kai entimos h¯ ec h r eia,o ft h e post of edeatros
held by Ptolemy (see below).
8Kienast 1973 :258; V ¨olker-Janssen 1993 :32–7.
9Notetoothe courting of Alexander’s ‘friends’ with rich gifts by Orsines, satrap of Persis, in
325/4 BC (Curt.10.1.24). Such brokers already existed at Philip’s ‘court’ in 336 BC, when a
Macedonian (almost certainly), his name lost, was honoured by the Athenians for ‘taking care of
Athenians visiting Philip’: T od 1948 :no.181 lines 12–15. I am grateful to Simon Hornblower for
this reference.

86 tony spawforth
royalinvitation to dinner could serve to dissemble a leading Companion’s
imminent fall from grace.10On another level, Alexander’s relations with
his retinue conform to the revisionist model of early modern courts, which
emphasises how ‘the exercise of authority [at court] was always in somesense a negotiation’ (Adamson 1999b :38). This observation seems borne
out in Alexander’s case by his habit of sounding out the opinion of leadingMacedonians on controversial issues over food and wine; in this way the
royaltable served as ‘the ground on which one’s relationship with the king
and others was constantly being tested’ .
11Aswill be seen in the ‘obeisance’
affair (below), Alexander did not always have his way.
The communicative structure typical of relations between an absolutist
ruler and his courtiers was a feature of Alexander’s entourage. On the one
hand an ancient tradition presents ‘frank speaking’ or parrh¯ esia asac o n –
vention in relations between Macedonian kings and their subjects,12and
this convention is evident in exchanges between Alexander and individ-ual Companions.
13On the other, praises, ﬂattery and submissiveness were
increasingly the norm under Alexander (Plut. Al.23.4; Arr.4.8.2–3), and the
type of the courtier, entirely dependent on his sovereign, is well captured
in Alexander’s alleged reproof of his favourite, Hephaestion (a Compan-ion), as ‘a fool and a madman for not knowing that without Alexander’sfavour he was nothing’ (Plut. Al.47.6). The uneasy concurrence of two such
antithetical styles of communication r eﬂects a transformation under Philip
and Alexander in the character of the Companion-elite from people withafair degree of social and economic independence from the Macedonian
monarch into courtiers,
14in the Polybian sense of hoi peri t¯ en aul¯ en,‘ a t
thenodo ft h ek i n g…a to n em o m e n tu n i v e r s a l l ye n v i e da n da tt h en e x t
universally pitied’ (5.26.3, cited by Herman 1997 :205). This change had less
todo with the moral defects of individuals, pace the ancient writers, than
the way in which, under Alexander above all, the huge aggrandisement inroyalpower created competition for the ruler’s graces on a scale previously
unknown in Macedon, as is suggested by the cohort of ‘Alexander-ﬂatterers’ ,as they were known (Chares, FGrH 125 F 4 = Athen.12.538f), who were in
the habit of giving Alexander ‘extravagant presents’ . The favour-curryingmentality behind such gifts is suggested by the tale of Hagnon of T eos inwestern Asia Minor, a Companion of Greek extraction who sought to please
10Curt.6.8.16 (Philotas).
11Borza 1983 :54; V ¨olker-Janssen 1993 :46–7; cf. Davidson 1997 :286–8 (‘power dinners’).
12Polyb.5.27.6; see Anson 1985 :314–15; Murray 1996 .
13E.g. Coenus at the Hyphasis river: Arr.5.27.2–9.
14On this change see V ¨olker-Janssen 1993 :32–48.

The court of Alexander the Great 87
Alexander by ‘writing to him that he wanted to buy Crobylus, whose beauty
was famous in Corinth, as a present for him’ .15
Off the battleﬁeld, Alexander’s ‘court’ was the chief site of the represen-
tation or image-management which contributed to creating his outwardidentity as a monarch (Butz and Dannenberg 2004 :24). In another clas-
sic function of courts, it was also the place where the ruler used ceremo-nial to construct and display the relative ranking of elite groups within thestate. These two aspects of Alexander’s ‘court’ receive particular emphasisin this chapter, where their role in Alexander’s state-building enterprise isre-examined.
Other features of courts are missing or not strongly present under
Alexander, in part as the result of his itinerancy. There was no central palace,although this feature of the court phenomenon was present in Macedon atleast from the reign of Archelaus (413–399 BC); and by the time of Philip IIat the latest was physically ﬁxed at Pella (Nielsen 1994 :81). But even under
Philip the ‘court’ must have often been migratory, since he was famous forcampaigning throughout the year (Dem.9.50); Alexander’s use of elaborateroyaltents may well have gone back to Philip (see Appendix). By the same
tokenthe sustained physical proximity of Alexander to the Macedonian
ofﬁcer class during the campaigns in Asia should not be seen as completelywithout precedent. Even so, Alexander’s city of tents must have gone much
further in developing the features of a centre of governance as well as power;the presence of the royal archive, housed in the tent of Alexander’s secretary(Plut. Eum .2), is surely suggestive here. The women of the royal family, in
particular Alexander’s mother and full sister, did not accompany him intoAsia, although they can be considered a part of the outer court (above),
and in the case of Olympias brought inﬂuence to bear (or sought to) byletter (Berve 1926 ,I:63; II: no. 581); on the other hand, new royal women
entered Alexander’s household in Asia, including two wives and a harem(below).
The ancients themselves probably did not see Alexander’s peripatetic
entourage as qualitatively different from the ﬁxed power-centres of otherancient monarchs. At any rate, aulic language was certainly applied by laterancient writers to describe his household, as in ‘the Macedonian courtiers’ of
15Plut. Al.22.2. Plutarch’s Alexander indignantly rejects the offer, although, as has been pointed
out (Berve 1926 ,I:10–11), such an offer (if historical) must have been thought likely to win
royalapproval. In passing, to see relationships of reciprocity between Alexander and his
entourage (so Hatzopoulos 1996 ,I:335–6) seems misguided: the essence of reciprocity, at least
as the ancient Greeks understood it, lay in the approximate parity not only of the gifts’ value
but also of the donors’ status: G. Herman, OCD3,s.v. ‘reciprocity (Greece)’ .

88 tony spawforth
the king, hoi peri t¯ en aul¯ en Makedones (Diod.Sic.17.101.3), or the malefactor
whom Alexander expelled ‘from his aul¯e’(Plut. Al.70.3); these usages con-
ceivably go back to the primary sources on which these later writers drew.
From a comparative perspective, ﬁnally ,this itinerant retinue meets all the
criteria of the early modern court: the place where the ruler lived; the seat of
governance; and ‘the venue where the ceremonial and ritual that surroundedthe prince could be properly observed’ (Adamson 1999a :10).
Insum, it seems safe to drop the scare-quotes and refer straightforwardly
from now on to the Macedonian, and speciﬁcally Alexander’s, court.
Bias in the ancient evidence
When approaching the ancient writers on Alexander, a basic truth must
ﬁrst be acknowledged. The Alexander which they have bequeathed to us toaconsiderable extent is a ‘Roman’ Alexander. Their construct, to be sure, is
based on earlier accounts, albeit to an extent, and by routes, which remainhotly contested.
16Butthis material, acquired by whatever means, had to
be recast so as to hold the interest of Roman-period audiences, living atatime when monarchy was the dominant state-form of the Greco-Roman
world and when the past could serve as a discreet vehicle for discourse about
‘good’ and ‘bad’ rulers and the vicissitudes of court life. In the case of Quintus
Curtius, writing in Latin under the Caesars, his portrayal of Alexander is
the most obviously coloured by Roman-elite concerns about the exercise ofimperial power (Spencer 2002 ;see Paterson in this volume). Beyond that,
all these texts are cast in a consciously literary vein, so that their artfulness,or ‘craftedness’ , can hardly be overestimated.
17
When it comes to court life, Alexander is presented paradoxically in these
sources. On the one hand, and especially in those texts aimed chieﬂy at aGreek readership, he is made to resemble a well-behaved Athenian gentle-
man in his personal disdain for royal luxury and his self-control in the faceof sensory temptations such as Darius’ beautiful wife, or Asiatic gastro-nomy.
18On the other hand, as a matter of record he adopted more and
more of the imposing ceremonial of the Persian court from 330 BC onwards,and his court eventually became a byword for conspicuous consumption,as shown by a long section ‘concerning the luxury ( truph¯e)ofAlexander’
16Forexample, Hammond 1983 ,esp. 1–11.
17See, for instance, Moles 1985 ,onthe intertextuality in just one passage (admittedly, the
important ‘second preface’) of Arrian.
18Especially in Plutarch: e.g. Al.20.8; 21.4; 22.4.

The court of Alexander the Great 89
included by the third-century AD Greek writer Athenaeus of Naucratis in
hisDeipnosophists (537d–540a). Ancient writers themselves were aware of
the paradox. The preferred solution, which suited the moralising tendencyin classical historiography as well as reﬂecting the unwillingness or inabilityof most ancient writers on Alexander to rationalise royal splendour, was topersonalise the issue in terms of Alexander’s corruption by his eastern con-quests.
19This was a particularly tempting line of explanation in Alexander’s
case, since it allowed Greek and Roman writers to play with the classicalstereotype of ‘oriental/barbarian monarchy’ (e.g. Hall 1989 ), to which the
‘corrupted’ Alexander could be assimilated when he was not being proferred
tothe ancient reader as its morally superior – because ‘Greek’ – antithesis.
20
Our fullest accounts of ceremonial at Alexander’s court come in the form
of anecdotal material deriving ultimately from Chares and (probably) Ephip-pus (App.) and relating to his stay at Susa (324 BC). Detailed descriptionssurvive of his use during this time of a palatial tent-complex to celebrate the‘mixed marriages’ , as well as for his routine public audiences. As a reminder
of the extremely partial coverage in the ancient sources generally of ceremo-nial at Alexander’s court, it is instructive to compare the coverage of theseevents in the anecdotal tradition about Alexander on the one hand and in
the primary Alexander-narratives on the other. These last make no mentionat all of Alexander’s audience-giving at Susa. As for the marriage feast, itssetting goes unreported by either Diodorus (17.107.6) or Arrian (7.4.4–8);Plutarch notes it sparingly ( Al.70.2, with a few details); the coverage of Susa
byCurtius, who might have entered into more detail, is lost (it probably fell
in the lacuna at the end of 10.1). T o judge from this one example, it is areasonable assumption that the aulic aspects of Alexander’s reign are gener-
ally under-reported in the surviving ancient tradition about Alexander. Oneexplanation for this state of affairs may well have to do with the predomi-
nantly apologetic presentation of Alexander, at least in Plutarch and Arrian:thick description of ‘barbarian’ display and pageantry would have tended toundermine the case which they make for Alexander’s heroic Greekness. Thedangers of this ancient bias for modern historians are suggested by modernevaluation of the splendours of Susa. Illuminated as these splendours are by
the chance spotlight of the anecdotal material, did they really constitute asigniﬁcant ratcheting up of Alexander’s use of pomp and circumstance in324 BC, as has been assumed (Bosworth 1980 :8)?
19E.g. Arr.4.7.4; Diod.Sic.17.77.4; Curt.6.6.1–3; Justin 12.3.8–12.
20Briant ( 2002 )gives an insightful discussion of narratological sub-texts to the classical portrayal
of Alexander.

90 tony spawforth
The court of Philip II
Before looking further at Alexander’s court, something more must ﬁrst be
said about that of his father, Philip II. The evidence is poor, but makes clearnonetheless that the Macedonian court even before Alexander’s conquestswas developing along new lines. Philip II (360/59–338 BC) was a heredi-tary ruler who succeeded after a prolonged period of dynastic weakness.Politically and militarily Philip had laid the foundations for Alexander’s
achievements by transforming Macedon from a minor Balkan kingdominto a multi-ethnic empire or ‘Vielv ¨olkerstaat’ , one which united Mac-
edonians, Thracians and Greeks under Philip’s personal rule in a fashionnot so unlike that of Macedon’s giant imperial neighbour, the ‘supra-ethnic’ Achaemenid Persian state (Kienast 1973 :248). The centre of this
new state-entity was Philip’s court at Pella, and most scholars are agreedthat Macedon’s new-found power under Philip was linked to importantchanges at court, both in its organisation (Kienast 1973 :251–68) and more
broadly in ‘the public presentation of monarchy in Macedonia’ (Carney2000a :275).
On this last point, it is clear that Philip, well before Alexander started
todo so, was increasing the association of royal power with display and
ceremonial. The Macedonian court has acquired a rather homespun image
in modern scholarship as a place where the ‘women of the royal familymade the daily meals and wove the clothing of their menfolk’ (Hammond1989 :31). But display had been an adjunct of monarchy in Macedonia
at least since the reign of King Archelaus (413–399 BC), said to have hiredZeuxis, a leading Greek painter, to decorate his palace.
21Since Greek authors
weredisinclined to view royal luxury, in Macedonia any more than in Per-
sia, as ‘une marque et un symbole de la splendour royale’ (Briant 2002 :
356), Philip’s conspicuous consumption is reported in negative terms (‘insa-tiable and extravagant’ , apl¯estos kai polutel¯ es:the hostile Theopompus, FGrH
115 F 224 = Athen.4.167a).
22Buthis lavish hospitality impressed Athenian
ambassadors in 346 BC,23and Philip may well have devised the ‘hundred-
seater’ dining tent which Alexander is found using at Dium as early as335/4 BC, scarcely two years into his reign (Appendix). As his son would
21Ael.VH14.17; Nielsen 1994 :81.
22The evidence of the ‘T omb of Philip’ at Vergina for the lifestyle of the Macedonian court is left
out of this discussion, since doubts continue to surface about the dating and the identity of theoccupants (Palagia 2000 :189–98). The question in the end may only be decided, if at all, with
the full publication of the ﬁnds, pottery included.
23Hammond 1989 :142 citing Aeschin.2.41–2; 47; 51–2.

The court of Alexander the Great 91
be later, Philip was an impresario of monarchy, as shown by the spectacular
royalwedding which he organised at Aegeae in 336 BC. This included a
procession of statues of the twelve gods made for the occasion, ‘wroughtwith great artistry and wondrously adorned with the brilliance of wealth’ , to
which, with a showman’s touch, Philip boldly added his own statue, ‘suitablefor a god’ (Diod.Sic.16.92.5).
Twodetails in the evolving ceremonial at Philip’s court are of particular
interest because they raise the question of Persian inﬂuence at Pella prior toAlexander’s succession. When Arrian is describing the duties of the ‘royalboys’ he includes their mounting the king on his horse in ‘the Persian style’(Arr.4.13.1). The clear implication of the passage is that this custom wentback to Philip (rather than being an innovation of Alexander). Bosworth(1995 :92–3) rightly saw the custom as a ‘mark of honour for the reigning
king’ . The straightforward inference would be that Philip took over thisPersian royal practice from knowledge of customs at the Persian court (see
below). Bosworth however denied this, on the grounds that the practice wasknown earlier in the fourth century BC to the Athenian Xenophon, whorecommends it for the old and inﬁrm. But there is no obvious reason why
the custom should have been imported to Pella via the Greeks rather thandirectly from the Persian empire. And it was presumably the royal Persianassociations which made ‘the Persian style’ an appropriate way of honouringthe Macedonian king; so a direct Macedonian borrowing from the Persiancourt is probably in question here.
Secondly, by 336 BC at the latest, Philip had taken to using a special chair
tosignify his elevated rank, a thronos or ‘throne’ , on public occasions. The
evidence comes from a fragmentary reference to ‘the Macedonians around
the throne’ ( peri thron[ou] )atthe moment of his assassination in the theatre
at Aegeae.
24Itis therefore not quite true that the royal throne was ‘unknown
as such to the Macedonians’ (Fredricksmeyer 2000 :159), although this seems
to be its ﬁrst ﬁrmly dated appearance.25It is te mpting to see the practice
as another innovation which Philip borrowed from the Persian court, thethrone having for long been the ‘primary symbol of royalty in the Near East’(2000: 159).
26
24P.Oxy.15.1798; Hammond 1991 :401; re-edition by Grzybek 1999 .
25Alavish marble throne was discovered at Vergina in the ‘T omb of Eurydice’ . It is probably a
royal tomb, withaﬁ r m terminus post quem of 344/3 BC, but both the sex and the identity of its
occupant are disputed: Carney 2000b :242–3 with refs.
26Alexander’s use of a golden throne, certainly after his return from India, is well documented:
Fredricksmeyer 2000 :152, 159–60, 161. For the Persian capital crime of sitting on the king’s
throne, and Alexander’s changing attitude to this taboo: ibid.

92 tony spawforth
Moregenerally, Dietmar Kienast ( 1973 )has argued that many of Philip’s
reforms at court and in the administration were inspired by neighbouring
Persia. Even if this argument can be contested here and there in the detail,27
the general thesis of Persian inﬂuence should hardly be controversial, unless,
as Kienast observes ironically, one believes that the only outside inﬂuence onfourth-century Macedon came from Greece ( 1973 :269). Even the Athenians
weresusceptible to their hereditary enemy Persia’s culture and way of life
(Miller 1997 ). When it came to monarchy, the paramount prestige of the
Persian court made it an inevitable model for lesser courts on its periphery.
28
Macedon in addition had a long history of Persian links. For thirty years in the
ﬁfth century BC it had been a Persian vassal state (Kienast 1973 :269); under
Philip there was continued contact, including the Persian envoys famouslyentertained in his father’s absence by the young Alexander (Plut. Al.5.1),
and the prolonged presence at the court for seven years of a high-rankingPersian satrap, Artabazus, along with his entourage.
29Asthe substance of
royalpower increased under Philip, there was a parallel elaboration of its
trappings, in conformance to the sociological proposition that ‘the most
real way of asserting one’s rank is by documenting it through an appropriate
social appearance’ (Elias 1983 :64). If this view of Philip is essentially correct,
then two striking characteristics of Alexander’s reign, his interest in theappearances of power and his imitation of the Persian royal court, can besituated on a larger trajectory rooted in his father’s ‘Macedonian revolution’.
Alexander’s court before 330 BC
Although Alexander’s court in this period is ill attested, there is enoughevidence to show that his adoption of Persian court ceremonial in 330 was
far from being a Damascene conversion to royal pomp. His early under-standing of conspicuous consumption as a facet of royal prestige is shownbythe festival which he staged in Macedonian Dium on the eve of his expe-
dition (335/4 BC), including a sumptuous feast. ‘He erected a tent to hold ahundred couches and invited his friends and ofﬁcers, as well as the ambas-
sadors from the cities, to the banquet. Employing great magniﬁcence, heentertained great numbers in person besides distributing to his entire forcesacriﬁcial animals and all else suitable for the festive occasion, and put his
27Bosworth 1995 :92–3; Briant 2002 :924–5.
28Briant 2002 :670–3, on the Xanthian dynasts of Lycia.
29Berve 1926 ,II: no. 152; Kienast 1973 :270–1.

The court of Alexander the Great 93
army in ﬁne humour’ (Diod.Sic.17.16.4). Such events formed a key part of
that distinctly non-Greek ‘culture of generosity’ whereby, ‘through lavishgift-giving and hospitality’ , Macedonian kings ‘obliged men into their ser-
vice’(Dalby 1996 :153–4). In the ﬁrst few years of the expedition, when its
success hung in the balance, Alexander used display to make a show of force.Hisprogress was marked by a series of armed parades at Ephesus (334 BC),
Soli (333 BC) and Memphis (332/1 BC), presumably aimed at local audi-ences likely to include Persian spies, but no doubt intended to promote cohe-sion and team-spirit among the men themselves (Arr.1.18.2; 2.5.8; 3.4.5);as will be seen later, this military dimension would remain fundamental toAlexander’s evolving self-representation in Asia. We begin to hear of moreelaborate feasts as the victories accumulated: after Issus, a dinner in the mag-niﬁcent state tent of Darius, now Macedonian booty (Curt.3.11.21; 3.12.2;Plut. Al.20.6–21.1); lavish feasts laid on by the Babylonians after Gaugamela
(Diod.Sic.17.64.4); and then the junketing at Persepolis (Diod.Sic.17.22.1;Curt.5.7.2). Throughout this period Alexander also gave audience to envoys,
such as the Lycians in 334/3 BC (Arr.1.24.5), or the ‘many embassies fromGreece’ at Memphis in 332/1 BC (Arr.3.4.5); but there is as yet no hint in
the sources that these occasions were particularly marked by pomp.
Alexander’s initial adoption of Persian court
ceremonial (330 BC)
The sources are in agreement that Alexander began to hold court with new
splendour after the sack of Persepolis (early 330 BC), and they associate theﬁrst moves in this direction with his passage through the adjacent regionsof Hyrcania and Parthyaea in northern Iran, later in the same year. Alexan-der’s feasting- and audience-tent, of which more shortly, makes its ﬁrstalleged appearance in Hyrcania (Polyaen. Strat .4.3.24); his personal adop-
tion of a form of Persian royal dress is placed by Plutarch ( Al.45.1) a little
later (autumn 330 BC), when he had moved on into Parthyaea (Bosworth1980 :5–6). Prior ( pr¯oton)tothis sartorial innovation, according to Diodorus
Siculus (17.77.4–7), he had already introduced Asian ushers ( rhabdouchoi )
at court and given himself a new Persian bodyguard, including Darius’brother Oxyathres. Later ( eita), along with his own assumption of a new
royalcostume, he gave purple cloaks to the Companions and Persian har-
nesses for their horses, as well as acquiring a harem
30with its accompanying
30Accept ed e.g. by Bosworth 1980 :5;Fredricksmeyer 2000 :155.

94 tony spawforth
staff of eunuchs (Curt.6.6.8). Both the cloaks and the harnesses were Persian
status-indicators; the former were worn by the Persian king’s courtiers.31The
ushers marked the introduction of Persian court etiquette, which stressedthe mystique and deference surrounding the king, and required in turn themaintenance of order and policing of behaviour on formal occasions; in328 BC these ushers were the object of a complaint to Alexander that ‘Mac-edonians had to beg Persians in order to approach the king’ (Plut. Al.51; see
below). Their appearance in 330 probably also heralds Alexander’s appoint-ment of an ‘introducer’ ( eisangeleus ), a chamberlain who managed ofﬁcial
receptions and audiences, his duties modelled on those of Persian counter-
parts.
32The only eisangeleus known by name was a Greek, Chares of Mytilene
(Plut. Al.46).
Moreelaborate tented accommodation, not to mention the harem, would
have added considerably to Alexander’s baggage train, and on the face of ittheir acquisition at this juncture is at odds with the claim in one ancientsource (Curt.6.6.15–17) that the king burnt his transport carts before thenext leg of his march, towards Bactria (modern Afghanistan). Althoughaccepted by some moderns (e.g. Bosworth 1988 :99), this claim is dubious,
however, not least because the same episode is assigned a different date inother ancient writers.
33Indeed, Pierre Briant (2003: 361–5) has questioned
whether such a measure was ever taken, pointing to the exemplary, moral andrhetorical context of these passages, which implicitly contrast Alexander’s
commitment to military discipline with the luxurious practices on campaign
of the Persian kings. Whatever the case, it seems clear that Alexander’s armyremained heavily encumbered during the Bactrian campaigns. Already in
328 BC the ‘royal service’ ( basilik¯ etherapeia )was sufﬁciently important
tobe in the charge of a leading Companion, Pithon son of Sosicles, and
large enough to have to be left behind while Alexander campaigned againstSpitamenes (Arr.4.16.6; Berve 1926 ,I:25).
Alexander’s tented quarters
It is t ime now to consider Alexander’s tented accommodation in more
detail (the ancient evidence is set out in the Appendix). One of his tentswas an ancient marvel, repeatedly described by ancient authors at leastuntil the third century AD (Appendix, A, nos. 1–3). In the early Principate
31Bosworth 1980 :5;Savalli-Lestrade 1998 :300–1; Fredricksmeyer 2000 .
32Berve 1926 ,I:19–20; Jacoby, FGrH 125 F 2, commentary.
33On the eve of the invasion of India, 327 BC: Plut. Al.57.1; Polyaen. Strat .4.3.10.

The court of Alexander the Great 95
two of its tent-poles were thought worthy of display outside the imperial
palace at Rome (Appendix, A, no. 10). The fullest descriptions of this tent,going back to the lost Alexander-historians Chares and (probably) Ephippus(Appendix, B, nos. 1, 3), describe its use by Alexander at Susa in 324 BC. Herethe tent’s magniﬁcence arose from its sheer size as well as its copious use ofprecious materials, including costly embroideries, precious stones, and goldand silver; on this occasion it formed a tripartite complex comprising theoikos or ‘house’ , a vast forecourt ( aul¯e), and a surrounding enclosure wall
(peribolos ).
Alexander’s use of imposing tents can be traced further back. A luxurious
tent had been with the king in India, where he used it (or elements from it)
in 326/5 BC as a luxurious andr¯ onfor feasting local leaders (Appendix, A,
no. 9). This or another tent appears earlier on the Indian campaign. Fol-lowing a life-threatening wound while storming a town of the Malli people,Alexander had displayed himself by having ‘his tent’ ( tabernaculum )s e t
up on two ships lashed together for this purpose on the River Hydraotes(Appendix, A, no. 8). That this tent was an imposing affair seems con-ﬁrmed by its appearance earlier, at the battle of the Hydaspes river (326BC): Alexander ‘ordered his tent ( tabernaculum )tobepitched . . . the unit
usually in attendance on him to stand guard before it, and all the sumptuoustrappings of royalty ( omnem apparatum regiae magniﬁcentiae )tobedelib-
erately ﬂaunted before the enemy’s eyes’ (Appendix, A, no. 7). Two yearsearlier, in Bactria, the erection and striking of a royal tent on the banks ofthe Oxus river in 328 BC required the oversight of a dedicated Macedonianofﬁcial with specialist tent-pitchers ( ho epi t¯ on str¯ omatophulak¯ on), whose
excavations for the tent-poles dug deep enough to strike oil.
34The existence
of a royal tent for audience-giving by this date is alleged by Polyaenus, asecond-century AD Macedonian writer, according to whom Alexander ‘hadit made’ for use as a courtroom ‘when deciding cases among the Bactrians,Hyrcanians and Indians’ . Although this author’s reputation for veracity is
mixed, some of the entries in his collection of stratagems being ‘historicallyvaluable, others ﬁctitious’ ,
35the particular of this tent’s use as a ‘courtroom’
(dikast¯ erion )isnot self-evidently the invention of Polyaenus (see below).
Asit happens, the mention of Hyrcania ﬁts with the text of Diodorus, who
implies – as already noted – an initial introduction of court ceremony intwo phases late in 330 BC, with an earlier phase preceding Alexander’spart-adoption of Persian dress in Parthyaea.
34Appendix, A, no .6=P l u t . Al.57.4; cf. Arr.4.15.7; Berve 1926 ,I :1 8 .
35J. B. C ampbell in OCD3,s.v.

96 tony spawforth
Asfor the origin of the Susa tent, scholars have often assumed that it was
the same tent as the one used by Alexander at Dium in 335/4 BC,36since both
are described as ‘one-hundred couch tents’ (Appendix, A, nos. 1–5). This
tent in turn may well have been a legacy from Philip, who perhaps acquired
it during his years of intensive campaigning in Thrace. It could well haveaccompanied Alexander to Asia in 334 BC: his dinners and symposia oncampaign are attested as early as the following winter (Plut. Al.17.5 (Phaselis
in Lycia)), and tented accommodation for them must have been anticipated.On the other hand, in key respects Alexander’s Susa tent sounds decidedlyPersian: the richly embroidered tapestries and rugs, reminiscent of the tent
of Mardonius (Hdt.9.82.1); the sky-like canopies ( ouraniskoi ), obviously
derived from the circular ouranos of Persian royal tents (Appendix, A, nos. 3,
11); and the ﬁfty supporting columns, recalling the hypostyle halls of Susa orPersepolis (Miller 1997 :51). Its Persian character is most straightforwardly
explained on the view that it was, in fact, a Persian royal tent (so Miller 1997:
51): if not one of those captured after Issus in 333 BC (below), then another,perhaps the tent which accompanied Darius on his ﬁnal, fatal, march in330 BC (Curt.5.19.3); or, if not, then an older, Macedonian, royal tent was,at the least, now transformed by the incorporation of Persian elements(a version which would better accommodate the account of Polyaenus). Asatemporary structure made up of different components, this tent could
have been assembled in different ways, and its ﬁttings could easily have seenfurther evolution – for instance, its linen hangings in 324 BC (Appendix, A,no. 1: othonia )recall the earlier gifts to Alexander of Indian linen (Curt.9.8.1,
linea vestis ).
Nowadays the well-documented tents of the Ottoman Turks perhaps best
conveyjust how luxurious Alexander’s royal tents may have been (Miller
1997 :50). Like the ancient Persians, the Turks in origin were Asian nomads,
and in both societies the rulers’ tents seem to have developed strikinglysimilar forms. The imperial tents of the sultans likewise depended for theirimposing effect on size and magniﬁcent decoration, including rich embroi-deries using precious metals. According to a European writer, in 1673 sixhundred camels were needed to carry the sultan’s two campaign tents. There
are speciﬁc similarities to Alexander’s Susa tent: the sultans’ tents were alsosurrounded by a screen wall; they too were used for audiences, banquetsand other festivities.
37
36C. Bradford Welles in the Loeb Diodorus at 17.16.4; Murray 1996 :16; Savalli-Lestrade 1998 :
299 n. 38.
37Forpictures of Ottoman Turkish tents see www.turkishculture.org/tents/tents.html (with
accompanying notes by Professor Nurhan Atasoy) (visited June 2004). Note in particular the

The court of Alexander the Great 97
Itremains to ask whether these various royal tents known to have been
used by Alexander between late 330 and 324 BC were one and the same. The
camp of Darius III at Issus (333 BC) included a complex of at least three royaltents (cf. Curt.3.13.11): one, in which Alexander took a bath, was for Dar-
ius’ personal use (Plut. Al.20.6); another, ‘worthy of admiration for its size
and height’ , was used for feasting (Plut. Al.20.6); a third, next door, was for
the use of Darius’ wife and mother (Curt.3.12.3), and had its own forecourt(3.12.8; 10). It is argued below (pp. 99–101) that, in a development parallel toAlexander’s initial adoption of Persian court ceremonial, from late in 330 BChis feasting style also underwent a Persian-style elaboration. Alexander also,from this date, was accompanied by a Persian-style harem (above). Thewhole spatial arrangement of Alexander’s royal quarters must have under-gone reorganisation and enlargement to accommodate these changes. Itmakes sense to assume no longer just one, but a whole complex of royal tents,some or all of them in some sense Persian in origin. It is conceivable thatCurtius reﬂects this development when, after describing Alexander’s initial
adoption of Persian ceremonial in 330 BC, he thereafter invariably describesAlexander’s itinerant accommodation with a new and more imposing term,‘regia ’o r‘ r o y a lq u a r t e r s ’ ,o f which Alexander’s personal tent was just one
compon ent.
38
Feasts and audiences
What use did Alexander make of his feasting- and audience-tent in
particular? One scholar has recently expressed surprise that Alexander ‘chosetoorganise his festivities in a tent even though the Persian palaces were
still standing’ (Briant 2002 :258). This paradox is starkest for the tent’s
most famous outing at Susa in 324 BC, for the celebration of the ‘mixedmarriages’: seemingly odd because Susa possessed a perfectly good royalpalace which Alexander had left unscathed in 331 BC (Fredricksmeyer 2000 :
148). In this particular case, an explanation will be canvassed later. But
miniature ‘T ents at Okmeydani, Surname-i Vehbi, 1720’ , illustrating the imperial tent complex
for a circumcision festivity: a hierarchy of imperial tents forming a complex, including four ofgraded size enclosed by circular screen walls. The largest has a separate entrance-tent.
38ForAlexander’s personal tabernaculum within the regia note 8.1.47; 2.5; 9.5.22 (the Cleitus
episode at Maracanda, although Arr.4.8.9 and Plut. Al.51.4; 6 place this within a building);
9.5.22; 30 (Alexander’s Mallian camp in 326 BC). In fairness, it cannot be excluded that Curtius’switch to ‘ regia ’isrhetorical, to emphasise the more ‘royal’ demeanour of Alexander which
Curtius had just described disapprovingly (6.6.1–9) before the word makes its ﬁrst appearance
todescribe Alexander’s quarters (the Philotas episode in Drangiana: 6.7.6–8.23 passim ).

98 tony spawforth
the tent’s purpose must anyway be seen in the broader chronological con-
text of the use to which Alexander had put it since (on the view argued
above) late in 330 BC. Its hundred-couch capacity made it suitable for hisdaily dining, which normally took place in the company of sixty or sev-enty of his inner circle, mostly Companions (Berve 1926 ,I:13). As for state
functions, one of these, certainly, was feasting, for which it had been usedin India; this last was a special diplomatic event, noted as such by Curtius(Appendix, A, no. 9).
Inaddition, and no less importantly, Alexander used the tent for an activ-
ity denoted in the ancient sources by the Greek verbs chr¯ematizein ,‘transact-
ing business’ , and dikazein ,‘sitting in judgement’ . The former verb is used in
ancient descriptions (Appendix, A, nos. 3–5) of the tent’s use for audiencesat, once again, Susa in 324 BC. The use of the imperfect tense ( echr¯ematizen )
suggests the volume of business which Alexander transacted on this occasionalone. It is unlikely that Susa was exceptional in this regard. At Babylon inthe following year, according to Ephippus ( FGrH 126 F 4 = Athen.12.537d),
he also transacted business ( echr¯ematise ), this time in a royal park
39–
again in state, sitting on a gold throne, ﬂanked by couches with silver legsfor the Companions. Berve ( 1926 ,I:12) assumed that this ‘business’ mainly
meant military affairs. But Plutarch ( Al.23.2) paired ‘attending to military
business’ ( suntatt¯ on ti t¯ on polemik¯ on)with ‘judging’ ( dikaz¯ on)inhis descrip-
tion of Alexander’s daily routine when not at war, and it was the requirementsof ‘judging’ , speciﬁcally among ‘barbar ians’ , which for Polyaenus explained
the audience-tent’s creation.
By‘judging’ we should probably understand, as well as administrative
activity speciﬁc to the ﬁeld of ‘laws’ , the whole gamut of administrative busi-ness which must have been put before Alexander for a decision. ‘Petition-and-response’ is a model of rule applied to the Roman imperial monarchy:‘the whole nature of the imperial entourage, with its remarkably limited
resources and increasingly peregrinatory character, was shaped by the pres-
sure to respond to initiatives from below’ (Millar 1977 :617). This model
seems to be applicable, at least to some extent, to the Persian empire too.Audience-giving was a fundamental aspect of royal governance under the
Achaemenid kings, to the extent that the image of the king in audience has
been claimed as the summation of Achae menid rulership. Whenever they
left their palaces, the Persian kings were the targets of petitioners.
40Since
what was taking place as Alexander progressed through the eastern provincesof the former Persian empire was nothing less than the replacement of one
39Forthe location note Arr.7.25.3.40Briant 2002 :191 citing Xen. Cyr.8.3.19–23.

The court of Alexander the Great 99
sovereignty with another, Alexander is likely to have found himself, willy-
nilly, the heir to this traditional role of the Persian king. These audiences –one can hypothesise – would have given Alexander the opportunity tocultivate the local elites of central Asia, to exchange presents, and to presentthe public face of the new ruler, based on a carefully crafted self-image incor-porating the tent and his newly adopted Persian royal ceremonial, alongwith heavy military show (below). That Persian royal audiences provided a
speciﬁc model for Alexan der, ﬁnally, is suggested by a (hostile) fragment of
the Greek historian Ephippus. For Alexander’s audience-giving he recordsthe use of scented wine and unguents to perfume the ﬂoor, of incense, anda‘holy stillness ( euph¯ emia )and silence’ ( FGrH 126 F 5 = Athen.12.538a).
Censers placed before the enthroned monarch were a feature of Persian royalaudiences (Briant 2002 :240), and it is likely enough that all these practices
which Ephippus records were taken over from Persian court etiquette. Thepicture presented by Polyaenus, then, of Alexander ‘deciding cases amongthe Bactrians, Hyrcanians and Indians’ , in spite of its Roman colouring,
41
has the ring of truth.
Upgrading the royal table
Another aspect of Persian rule which Alexander seems to have imitated
from 330 on was the Persian king’s famously lavish table. Despite the allegedsimplicity of his personal eating habits, paradoxically Alexander becamerenowned for the magniﬁcence of his dinners (Plut. Al.23). As we have seen,
the tradition of lavish royal dining in Macedon went back to Philip. But underAlexander its lavishness reached new heights by an incremental process, theoutlay ‘increasing with his successes until it reached the sum of 10,000drachmai’ per dinner (Plut. Al.23.6). Dating the stages of this elaboration
is hard. It was marked by the increasing numbers involved in preparationand service: we hear of ‘ofﬁcials in charge of the bakers and meat-cooks’(Plut. Al.23.3) and waiters under the orders of a trapezokomos ortrapezopoios
(Berve 1926 ,I:41with n. 6). Luxury in terms of expanding numbers of
domestics was already marked among Alexander’s Companions before theend of 330 BC, as shown by the masseurs and chamberlains acquired byPhilotas before his execution (Plut. Al.40; Berve 1926 ,I:42), and in these
matters Alexander surely led rather than followed.
41Polyaenus, a syndic in the imperial courts at Rome, makes Alexander sound a bit like a Roman
proconsul on his assize tour.

100 tony spawforth
Other signs suggest that this increasing elaboration of the royal table coin-
cided with Alexander’s adoption of Persian court ceremony late in 330 BC.
There is slight evidence for honoriﬁc positions in the king’s table-serviceheld – in the fashion of more recent courts, as already noted – by mem-bers of the Macedonian elite. As well as a ‘chief cup-bearer’ , attested in324/3 BC in the person of a young son of Alexander’s viceroy in Europe(Plut. Al.74), we also hear of the edeatros or taster of royal dishes (Chares,
FGrH 125 F 1 = Athen.4.171b–c). He in fact was a kind of lord high steward
in charge of the whole service of the table.
42This post was held by none
other than Ptolemy, Alexander’s childhood friend. Since Chares mentionedthis appointment relatively early on in his history of Alexander, in the thirdof at least eleven books, it is reasonable to suppose, with Jacoby,
43that the
appointment belonged to Alexander’s initial elaboration of his court in 330.Finally, Plutarch ( Al.50.2) describes the arrival of ‘some people bringing
Greek fruit from the coast’ in 328 BC when Alexander was in Maracanda
(Samarkand). By this date at the latest, we can see the establishment of anaspect of the luxury of Alexander’s table on which elsewhere Plutarch com-ments generally: the supply for royal delectation of ‘the rarest fruits or ﬁshfrom the coast’ ( Al.23.5–6).
This aspect of Alexander’s dinners is strongly reminiscent of the dining
practices of the Persian court. Greek contemporaries certainly saw similari-ties between the two styles of royal dining, and relative costs were explicitlycompared (Ephippus, FGrH 126 F 2 = Athen.146c–d). Other details suggest
conscious imitation: on one occasion at least (Susa) Alexander graded his
guests at a feast according to royal Persian protocol: the more favoured ateinside, the less favoured in the courtyard;
44the circular seating at the Opis
feast in the same year was likewise a Persian arrangement (below). The ofﬁceofedeatros was believed in later antiquity to render by a Greek word what in
fact was a Persian usage.
45Reasonably, then, some have seen here a Persian
ofﬁce,46or, if not, the reinvention of a traditional Macedonian ofﬁce along
Persian lines (Berve 1926 ,I:39–40). Going back to the rare foodstuffs offered
toAlexander, likewise the Persian king was supplied with food and drink
42Arte midorus of Tarsus (ﬁrst century BC) cited by Athen. Deip n.; Aelius Dionysius (p. 151
Schwabe) cited by Eust.1403.40; Berve 1926 ,I:39–40.
43FGrH 125 F 1 and F 4, commentary.
44Appendix, A, nos. 1–2; B, no. 7; Athen.12.538c; Briant 2002 :258. See below.
45to men onoma Hell¯ enikon, h¯ ed ec h r e i aP e r s i k ¯ e,according to the second-century AD
grammarian Aelius Dionysius (p. 151 Schwabe), cited by Eust.1403.40; see Jacoby, commentary
toFGrH 125 F 1.
46Briant 2002 :263; Berve 1926 ,I:39–40 for the view that only its titular quality under Alexander
was Persian-inspired.

The court of Alexander the Great 101
from all over the empire: subjects ‘anxious to curry regal favour would delib-
erately send delicacies to tempt the royal palate’ (Strong 2002 :8). Plutarch’s
evidence, cited above, could at a pinch be argued to show that Alexander
became heir to this Persian royal tradition, although Plutarch mentions onlyone part of the empire, the Greek ‘coast’ ( thalassa ), by which presumably
he meant the seaboard of Asia Minor. Recent scholarship has emphasisedthe political and symbolic overtones of the Persian king’s dinner: ‘the sup-ply of food and drink from all over the empire highlighted and embodiedthe domination of the king over his subjects’ (Sancisi-Weerdenburg 1993 :
298). Was Alexander aware of, and did he also seek to perpetuate, this ideo-logical dimension to Persian royal dining? It is possible. At any rate, whenAlexander decided to create a Persian-style court, he evidently decided – andfor the same reasons – to re-create some of the display and aulic deferencecharacteristic of the Persian royal style of dining as well.
47
Timing and context of the adoption of Persian ceremonial
This section takes fuller stock of Al exander’s timing in the initial elabora-
tion of his court on Persian lines from late 330 BC onwards. The standardview is that the decision was a rational one, prompted by the murder of
Darius III and the news shortly afterwards of the usurpation of the Persianthrone by Bessus, an Achaemenid (Bosworth 1980 ). The political decision
byar uler to adopt an imposing court as a state-building device has par-
allels in other periods and places (see van Ess in this volume); for whatit is worth (since Roman colouring may have crept in), ancient sourcesexpress the view that Alexander on this occasion was motivated by state
interest (Plut. Al.47.3; Arr.7.29.4). What needs stressing is how far he took
this elaboration of the trappings of his court. As well as the adoption ofPersian ceremonial, Alexander’s use of imposing Persian-style tents dates
from this time, from which seems to date, as well, the growing splendour,Persian-style, of his table. Taken together, these initiatives seem more than
a‘limited gesture’ (Bosworth 1980 :8): they suggest a major turning point
in Alexander’s evolving strategy of state-building in Asia.
On the speciﬁcs of timing, if weight can be placed on the language of
Diodorus, Alexander had already begun to introduce these changes before he
adopted Persian-style dress in Parthyaea; and if Polyaenus is taken au pied de
47On the practical level, captured specialists from the service of the Persian king were available to
help (Athen.12.608a).

102 tony spawforth
la lettre ,his feasting- and audience-tent was already in use in Hyrcania. The
point may seem a small one: but these scraps can be construed to show that
Alexander’s decision was not only bigger in scope but also more premeditatedthan is sometimes allowed: at any rate, it was not just a contingent responsetothe news, arriving in Parthyaea, of Bessus’ usurpation of Darius’ throne
(as argued by Bosworth 1980 :5–6). The true catalyst seems to have been the
discovery of the murder of Darius some six weeks earlier, and Alexander’sdecision, made public soon after, to pursue the conquest of the eastern partsof Darius’ empire (Bosworth 1988 :96–7). But it is hard to credit that he and
his friends had never discussed future plans rather earlier: it had been overayear ago, following his victory at Gaugamela (331), since Alexander had
taken a new royal title, ‘king of Asia’ (Fredricksmeyer 2000 ), and over a year
since he had appointed his ﬁrst non-western satrap of a newly conqueredprovince (the Persian Mazaeus, put in charge of Babylonia: Arr.3.16.4) –the ﬁrst step in a sustained attempt by Alexander to develop a hegemonicrelationship with the ruling elite of the Persian empire (of which more
shortly).
Social ranking, court ceremonial and obeisance
Inam onarchical state with a pyramidal social structure, one recognised
function of courts is ‘rank allocation’ at the top end of the social hierar-chy.Atpublic ceremonies, strict rules governing the order of precedence in
such matters as seating served to display, and indeed construct, the relativeranking of the highest social strata (Winterling 1997 :153). From 331 BC
Alexander’s entourage had started to incorporate a tiny but growing num-ber of high-status Persian/Iranian collaborators. This number was increasedfrom late in 330 BC by Alexander’s recruitment of high-ranking Persianstoform a guard of honour at court.
48Participation by high-ranking Per-
sians in the daily life of the court is revealed in the ancient accounts of thedrinking party at which Alexander attempted to introduce obeisance in 327(Arr.4.12.2; Curt.8.7.22; see below). Indeed, the social, as well as political,dimension to Alexander’s admission of Persians to his court should not beunderestimated, given that ‘Macedonian and Persian nobles had in com-mona na r i s t o c r a t i cw a yo fl i f e…a n dv a l u e s ’ ;
49long ago Berve estimated
that ‘the number of leading orientals taking part in the campaigning must
48Fredricksmeyer 2000 :151 and 152 n. 45; see above.
49Briant 1985 :183 cited by Gregor 1997 :3 5n .3 1 .

The court of Alexander the Great 103
have been very important’ ( 1926 ,I:80). Once Persian ceremonial had been
introduced at court late in 330, it could serve to display the relative ranking
of these Persians alongside the Companions and other elite groups withinthe hegemonic class of his empire. Asit happens, the clearest instance of
ceremonial functioning in this way comes in 324 BC at the vast banquet for
9,000 diners (Arr.7.11.9) which Alexander celebrated at Opis on the Tigrisriver.On this occasion, according to Arrian, dining couches were organised
in concentric circles centred on Alexander himself: the nearest circle com-prised his ‘Macedonians’ , the next ‘Persians’ , the third ‘any persons from theother peoples who took precedence for rank or any other high quality’ .
50
Inthis hierarchical seating plan, the order of precedence radiated outwards
from Alexander, with the ‘Macedonians’ assigned the places closest to himand thus the highest rank. Two points about these arrangements need moreemphasis than they normally receive. First, feasting in concentric circlesseems to have been a Persian custom, which Alexander on this occasion(and no doubt others, if we were better informed) had taken over. AfterAlexander’s death, in 317/16 BC, it appears again in the detailed descrip-tion of a feast at Persepolis hosted by Peucestas, the Macedonian satrap ofPersis famed for his adoption of Persian ways.
51Second, not only did the
seating arrangements at Opis, as others have noticed, single out the Per-sians in Alexander’s entourage by assigning them highest ranking after the‘Macedonians’ , but they also conferred no special distinction on Greeks qua
Greeks; high-status Greek guests on this occasion (with the exception, no
doubt, of ethnic Greeks among the Companions) sat in an undifferentiatedmass with the ‘other peoples’ .
52
Alexander’s concern to honour the Persians, on this occasion over and
above ‘even’ the Greeks, invites reappraisal of a more problematic episodewhich had occurred three years earlier: Alexander’s attempted introduc-tion of ‘obeisance’ ( proskun¯ esis). This much-discussed episode (Bosworth
1995 :68–90) took place at a drinking party at court in spring 327 BC,
when Alexander, in collusion with sympathetic members of his entourage,experimented with the introduction of this Persian custom at court for per-
formance by his Macedonian and Greek, as well as his Persian, courtiers.‘Obeisance’ ( proskun¯ esis)was a term used by Greek authors to describe two
quite different things (which the Greeks then confused). First, it denoted thegesture (or range of gestures) of veneration which Greeks performed beforeagalmata or images of deities in temples. Second, it embraced a variety of
50Arr.7.11.8–9; Badian 1958 :429–30.51Diod.Sic.19.22.1–3.
52Asnoted by Brunt 1983 :240–1 nn. 7–8.

104 tony spawforth
Persian gestures of greeting which differed according to the relative rank
of those involved (Hdt.1.134). In this second context, for Greeks the word
evoked above all the obeisance offered by subjects before the Persian king.
Persian evidence conﬁrms that obeisance was an essential preliminary to a
royalPersian audience (Briant 2002 :222–3). The later Greek tradition took
Alexander’s attempted introduction of obeisance as a bid to be worshippedas a god; not only this, but at least one Greek on the spot, Alexander’s ofﬁ-cial historian Callisthenes, refused to perform obeisance on this occasion,and there are grounds for assuming that his objection was religious.
53Some
powerful modern voices share this ancient Greek view, which has effectivelyhijacked the debate, both ancient and modern, about Alexander’s inten-tions on this occasion.
54On an alternative view, ‘it is plausible to suppose
that Al[exander] sought to introduce the practice among all his courtiers,only in order to create a greater measure of equality between Persian andwestern notables and thus to conciliate the Persians’ .
55Heretwo points are
emphasised. First, even if Alexander was aware of the religious signiﬁcanceof obeisance for Greeks (as opposed to Macedonians), it is scarcely crediblethat Greeks were his ﬁrst priority in 327. As the whole logic of his adop-tion of Persian court forms in 330 suggests, and seating arrangements atOpis conﬁrm, Alexander was chieﬂy concerned in these years to use rankallocation to honour Persian nobles (and Asiatic elites more generally: see
below on the Susians) who chose to collaborate with the invader. More-over, this innovation was certainly more than a ‘a purely technical reform’
(Badian 1981 :52, paraphrasing the view of others): it went to the heart of
the evolving representation of his state-building enterprise since late in 330(more of this in the concluding section). Although the experiment tookplace in the relative informality of a routine drinking party, Alexander (onthis argument) intended the act, if accepted, to be performed mainly onformal occasions, such as audiences, as in the old Persian court, when theimplied parity of standing between Macedonian and Persian notables wouldhave been witnessed by people outside the court.
The second point concerns the Macedonian (as opposed to the Greek)
view of obeisance in the third quarter of the fourth century BC. The Mac-
edonian Companion Leonnatus, who is cited as ‘off-message’ on this occa-sion, mocked the act – not an obviously religious reaction.
56It is wo rth
asking, then, how readily fourth-century BC Macedonians would have seen
53Badian 1981 :51–2.54Badian 1981 :51–2; Bosworth 1995 .
55Brunt1976 –83, I: 539; other refs.: Badian 1981 :40.
56Leonnatus in Arr.4.12.2; Polyperchon (wrongly) in Curt.8.5.22–4 (Badian 1981 :52n.38).

The court of Alexander the Great 105
obeisance in religious (as opposed to more broadly cultural) terms. The case
for a widespread Macedonian adherence at this date to the veneration of reli-gious statues remains to be proven.
57Major temples to house such images
are conspicuously absent in the fourth-century BC homeland, not least atDium, the pan-Macedonian religious centre (Spawforth 2006a: 34). Mac-edonian cult-buildings which pre-date Philip’s reign have been excavated atVergina and Dium. All three are small and, to judge from the remains as
published, architecturally modest. The ‘temple of Eukleia’ at Vergina com-prised a porch and a small inner building with bases for statues in each ofthe two far corners; the position between these on the central axis, wherebythis date a principal agalma would normally be found in a Greek temple,
was reserved for an offering table (Andronicos 1984 :49–51). Two structures
comparable in size and plan to the ‘temple of Eukleia’ , dated to the late sixth
century BC, have been found at Dium in a sanctuary of Demeter near that
of Zeus; ‘inside were wooden benches on which were placed dedications’;no traces of agalmata are reported from either building (Pandermalis 1997 :
17–18; 1999 :61–2). None of these three buildings supports the view that the
familiar Greek triad of naos ,agalma and b¯omos (see Burkert 1985 :50) was
central to Macedonian religion before Alexander.
58Turning to the sanctuary
of Zeus at Dium, Polybius (4.62.2) implies that at the time of the Aetoliansack in 219 BC the roofed elements were limited to stoas; the excavationshave uncovered a huge altar, but no sign of a temple. The enthroned Zeuson Alexander’s coinage ‘was most certainly not a particular statue’ (Price1991 ,I:30), although here, if anywhere, one might expect to ﬁnd the Zeus
of Dium, had such a statue existed at the time.
59
Insum on this point, the planners of the attempted introduction of obei-
sance in 327 BC certainly seem to have anticipated Macedonian objections,but they may not necessarily have expected these to be on religious grounds,if, as suggested here, Macedonians in 327 had not been in the habit of per-forming proskun¯ esisbefore agalmata back home. From a court perspective,
57Forthis Greek religious phenomenon see Scheer 2000 .
58Pandermalis 1999 :44–51. Philippson claimed a temple for Dium, but cited no evidence: REV. 1
(1903) col. 833.
59Alife-sized Demeter, an Artemis and a Core, said to be fourth-century BC, were found in the
sanctuary of Demeter near Lete in 1936, but the precise archaeological context is unknown, and
until or unless they can be shown to pre-date Philip and to have been what are traditionallycalled ‘cult-statues’ (as opposed, e.g, to offerings), their existence does not refute the argumentpresented here (Hatzopoulos 1994 :48, 123–7). The implications of this argument, if any, for
Macedonian Hellenism cannot be pursued here; but Dodona, another ethnos -sanctuary on the
Greek periphery, was likewise slow to adopt the cult of statues (Mee and Spawforth 2001 :379,
382). Whatever one makes of Philip’s play with agalmata at Aegeae in 336 BC, he was chieﬂy
concerned to impress the Greeks present on that occasion (Diod.Sic.16.91.6).

106 tony spawforth
the issue for the Macedonians arguably boiled down to status: the implied
equality with the defeated Persian nobility, to be displayed before the worldon formal occasions, turned out to be an unacceptable assault on their senseof their social worth.
Conclusion: court and state-building
Inassessing the increasingly symbolic weight which (on the view presented
here) Alexander assigned to his court and its trappings from late 330 BC,one can sense the realities of the Macedonian position, or at least Alexan-der’s perception of them, once the death of Darius highlighted his campand army as the real centre of power for most of the former Persian empire.Morethan ever, Alexander from now on was engaged in the process of
state-building even as he continued his conquest of the remaining Per-sian satrapies and sought out the ‘usurper’ Bessus. He was succeeding amonarchy which traditionally relied heavily on the projection of its claimstodomination through a lavish court (Brosius in this volume). The com-
pulsion on Alexander to present his newly evolving successor-r ´egime in
comparable terms must have been strong. The projection of ‘stateliness’ ,
another classic function of monarchical courts, seemed to be required atthis juncture, to be achieved in Alexander’s case by a new level of aulic dis-play, which placed before Asian observers images of military might, royalluxury and the ruler’s symbolic pre-eminence, this last constituted by ‘trap-pings’ and by the ceremonial deference of a high-status entourage, includingleading Persians. Anthropology has recognised this political dilemma: ‘Thestate is invisible: it must be personiﬁed before it can be seen, symbolizedbefore it can be loved, imagined before it can be conceived.’
60One might
go further and argue that the increasing resemblance of Alexander’s courttothat of the Persian kings was a function of the pressure to adopt a simi-
lar approach to the challenge of ‘ruling’ huge territories with rudimentarytechnologies of control, so that hill-tribes such as the Uxii in the Zagros
mountains, under Alexander as much as under the later Achaemenids, maynever have been properly subdued.
61The Persian monarchy had evolved
central symbols, ceremonial and a supporting ideology which together pro-
jected a royal image of divinely ordained omnipotence (Briant 2002 ). In
some ways Alexander’s self-representation from late 330 BC increasinglyheaded in a similar direction, at least in its use of display and ceremonial
60Clifford Geertz, cited by Muir 1997 :230.61NoteBadian 1985 :441–2 n. 2.

The court of Alexander the Great 107
tocreate ‘a strong impact on the senses’ aimed at ‘convey[ing] the essence
of power’ to the gamut of spectators, local and from further aﬁeld, at these
courtly performances (Duindam 2003 :183).
Inwarming to the Persian model, Alexander’s longer-term political con-
ditioning was surely also a factor. He had grown up in an increasinglycosmopolitan court, and watched his father come to rule with the help
of a multi-ethnic Companion elite (Theopompus, FGrH 115 F 224 =
Athen.4.167b) over a state structurally similar to the Persian empire in that
both were an agglomeration of ‘peoples’ (Kienast 1973 :esp. 247–51). From
the outset, Alexander may well have been unable to imagine his rule overAsia except in broadly similar terms.
Asis well established, Alexander certainly had good political reasons for
seeking to coopt the Persian nobility – a process begun in 331 BC with hisappointment of Mazaeus (above). The Macedonians were a tiny minority inthe former lands of the Persian king and some kind of consensual rule basedon the winning over of the central and regional elites of the old empire wasthe only long-term hope for the stabilisation of the Macedonian conquest.The message implicit right from the start of his adoption of Persian courtceremonial in 330 was made explicit in 324 BC, when Alexander used a
court festivity (the Opis banquet again) to proclaim publicly ‘harmony and
partnership in rule’ ( homonoia kai koin¯ onia t¯ es arch¯ es,Arr.7.11.9) between
Macedonians and Persians (Bosworth 1980 ).
The adoption of Persian court ceremonial not only offered people a sen-
sory representation of Alexander’s commitment to this ideal of ‘partner-ship in rule’ . Along with Alexander’s adoption of an increasingly Persianstyle of feasting, it also sent out the clear signal that Persians were wel-come at Alexander’s court. The evidence for high-ranking Persians in his
entourage has already been noted. Their physical presence was important ifAlexander was to be in a position to imitate the famed ‘open-handedness’orpolud¯ oria of the Persian kings, which served as a fundamental lubri-
cant in the Persian court system (Brosius in this volume). Alexander wasfamously generous (Plut. Al.15.3; 39–41; 63), to the extent that he suppos-
edly deﬁned royal friendship as the acceptance of his gifts (‘he could notregard as his friends those who wanted nothing of him’ , Plut. Phoc.18.4);
his recorded gifts include land, houses, incense, apparel and gold plate, nottomention appointments. Alexander certainly used gift-giving for speciﬁc
political ends: he was said to have ‘conciliated with gifts’ the Macedonianswho disapproved of his adoption of Persian customs ( tais d¯ oreais ethera-
peuen ,Diod.17.78.1). It is inconceivable that Persian nobles who attended
his court were not rewarded, and seen to be rewarded, in this way, even if the

108 tony spawforth
speciﬁc evidence (except in the case of appointments) is hard to come by.
Alexander’s deﬁnition of friendship, if true, suggests that his generosity alsohad a symbolic value in creating, or cementing, ties and obligations with hiselites (Asiatics included). One might compare early modern courts, wheregifts were ‘a means of enhancing the status of the recipent and of creating a
social bond between ruler and the noblemen attending his court’ (Asch 2003 :
88–9).
Alexander’s increasing remoteness as a result of the introduction of Per-
sian court ceremonial surely also had a practical purpose. Access to Alexan-der had never been free. Routinely, it seems, rank-and-ﬁle Macedonianshad to wait outside Alexander’s tent to be escorted in by a Companion (in330 BC: Curt.6.7.17), or by the duty bodyguard from the select group ofseven mentioned earlier (in 326 BC: Arr.4.13.7). The Companions seem tohave had a general right of access (Arr.5.28.3); but this privilege was prob-ably linked to ofﬁcial duties, since even as high-ranking a ﬁgure as Philotaswas said to visit the royal tent only twice daily at the time of his arrest(Arr.3.26.3; Berve 1926 ,I:19). The adoption of Persian court ceremonial
late in 330 BC clearly transformed the rules of access. This development wassufﬁciently marked by 328 BC that a leading Companion (Cleitus), as notedabove, could complain that Macedonians had to beg Persians in order toapproach the king (Plut. Al.51): the reference was to Alexander’s Persian ush-
ers. By 324 BC, it could be claimed, ‘the number of his friends and servitorsbeing so great, no one dared approach Alexander; such was the majestyassociated with his person’ (Phylarchus, FGrH 81 F 44 = Appendix, A,
no. 3). In absolutist monarchies well-deﬁned mechanisms controllingadmission to the royal presence have served historically as a coping mecha-nism in the face of pressure from courtiers seeking access to the ruler. As theempire of which he was the human centre inexorably expanded, along withthe business which it generated, including, crucially, the increasing num-bers hoping to beneﬁt from his favour (cf. Winterling 2004 :80), Alexander
would have faced similar problems. The volume of business for his audiences
suggested by the imperfect echr¯ematize at Susa has already been noted. One
result was (with little doubt) an increase in the brokering role of Alexander’s
inner circle (p. 85 above). Another was to compel the introduction ofgreater controls in the organisation of audience-giving. That Alexander
(and his chamberlains) were doing just this is shown by the ‘ingenious andmethodical arrangements’ , as Bradfor dWelles called them, for Alexander’s
audiences at Babylon in 324. Diodorus alone preserves these interestingdetails: a schedule ( apograph¯ e)was drawn up, according to which the king
heard ( echr¯ematise )the embassies in order of subject matter: religion ﬁrst;

The court of Alexander the Great 109
then bearers of gifts; then disputes with neighbours; and so on. Signiﬁcantly,
although many embassies came from places outside Alexander’s dominions,many communities from territory under Alexander’s direct rule were alsorepresented, including Macedonian envoys and ones from ‘the peoples and
cities, as well as the local rulers, of Asia’ . That is to say, notwithstandingthe administrative division of the kingdom into satrapies and (as Macedoneffectively was) a viceroyalty, Alexander’s subjects expected to approach theirking in person.
62
Going back to Alexander’s Persian-style tents, their acquisition reﬂects
the functional attractiveness of Persian ‘court nomadism’ , which provided averyspeciﬁc, and apt, model for Alexander and his itinerant centre. When
the Persian king was on the move, ‘the royal ten t…b e came the centre of
power’ , and in his travels the king ‘showed off the might and wealth of thecourt and army’ (Briant 2002 :esp. 186–91).
63All this was precisely what
Alexander seems to have tried to do from late in 330 as he moved furthereast. Indeed, one reason for accepting Polyaenus’ implied date for the cre-ation of the feasting- and audience-tent is that it answers the question: wherewas Alexander’s new court ceremonial performed? It required a setting ofcommensurate grandeur and stateliness which nonetheless could be con-
jured up in an army camp: a state tent `al aP e r s e was the obvious solution.
AsPolyaenus, with more than a touch of Greek condescension, puts it: a
‘brilliant courtroom suitable for a general, astonishing the barbarians . . . by
[its] appearance’ (Appendix, A, no. 4).
64
This emphasis by Polyaenus on Alexander ‘the general’ speciﬁcally glossed
the marked projection of military authority in the audience-tent ceremonialwhich he had just described. Since it was embedded in an army camp, Alexan-der’s itinerant court inevitably had a pronounced military character; andanyway the routine presence of bodyguards at Alexander’s feasts (Arr.4.8.8–9; Plut. Al.51.5–6) reminds us that the physical security of the king was a
traditional aspect of the Macedonian royal court – quite apart from any addi-
tional measures felt necessary in the context of Asia. But the ancient accountsof the audience-giving at Susa reveal the extent to which a formidable displayof the powers of coercion at Alexander’s disposal had become integral to his‘image-management’ . This is hardly surprising: army processions, as notedearlier, were already a feature of Alexander’s progress through Asia Minor
62Diod.Sic.17.113.1–4, citing the editor on p. 453, n. 3 of the Loeb edition.
63Nurham Atasoy (above n. 36) makes similar points about Ottoman imperial nomadism.
64Alexander’s itinerant use of this tent for audiences and feasts, on the reconstruction offered
here, affords a rather more dynamic picture of his interaction with the Iranians from that putforward in the most recent discussion of Alexander and the Persians (Brosius 2003 ).

110 tony spawforth
and Egypt; and right from the start Alexander had openly proclaimed the
spear as the legitimation of his assault on Persian territory (Diod.Sic.17.17.1–2). By the time of Susa, Alexander’s command of the military resources ofAsia allowed for a more nuanced symbolism. The array of elephants, to be
sure, may have had straightforward overtones of ‘shock and awe’ . But theparading of Persian and other Iranian troops in and around the audience-tent conveyed a consensual, multi-ethnic feel to Alexander’s army, one which
chimed both with local memories of the ethnically composite army of the
Achaemenid kings, and also with Alexander’s self-proclaimed quest for ‘har-
mony and partnership in rule’ with the Persians in particular (above). Finally,although the military parade at Susa reﬂected the increasing momentum ofAlexander’s enrolment of Iranian manpower in the last years of the reign,his ﬁrst recruitment of a Persian guard of honour went back (as seen earlier)tolate 330 BC, and presupposes the ceremonial deployment of these troops
from that time onwards.
The broad aims of Alexander’s adoption of court ceremonial are easier
toevaluate than its consequences. It is clear, for a start, and this is a key
point which the chapter seeks to make, that Alexander did not waver in theimportance, indeed centrality, which he attached to his court ceremonial asapolitical tool in this respect. He delayed until 327 BC the attempted intro-
duction of obeisance (above), which, on the view taken here, was intendedas a ceremonial representation of the koin¯onia or commonality of the Mac-
edonian and Persian components in Alexander’s new imperial elite. Thisdelay presumably reﬂected Alexander’s heedfulness of the hostility to theIranian rapprochement in some Macedonian quarters, and the consequent
need for a graduated approach (cf. Diod.Sic.17.77.7). After the obeisanceepisode Alexander proceeded to the invasion of India, returning to the cen-tral regions of his empire in 325/4 BC. The use of the court as a medium
for cultivating the Iranians now resumed. In 324 BC he conferred on ‘selectPersians’ the title of ‘kinsman’ ( sungen¯ es), a Persian distinction, along with
the equally Persian privilege of kissing the king.
65Butthe weddings between
Alexander and the Macedonian Companions and brides of royal and noble
65Arr.7.11.1; 6–7; Fredricksmeyer 2000 :156–7. The usual view that ‘Kinsman’ was a title taken
over from the Persian court has been questioned by Savalli-Lestrade 1998 :395–8, who rejects
the authenticity of the alleged 15,000 ‘ cognati regis ’inDarius III’s army in 333 BC (Curt.3.3.14)
on the grounds that they were too many to be real kinsmen. But this number of royal kinsmen
generated by a polygamous lineage in power by this date for over two centuries is not inherentlyimprobable when compared with the estimated 73,418 members of the Chinese imperial clanin 1915, after nearly three centuries of Qing rule: Rawski 1998 :94. Briant ( 2001 :108) does not
agree that the passage should be ‘evacuated from the Achaemenid dossier’ .

The court of Alexander the Great 111
Persian stock, celebrated at Susa with a great feast in the state tent, are the
eye-catching item here.66
Itis on this occasion that we are given a unique glimpse of the integrating
function of the court at the level of Asian local, rather than central, elites. On
this occasion Alexander made gifts of purple clothing to 500 men of Susa –local notables presumably – and assigned them places in the audience-ceremonial (Appendix, A, no. 4). In passing and in the same connection,
one would dearly like to know more about Alexander’s hunts in Asia, whereboar (Curt.8.6.7) and lions were among his quarry. The royal lion-hunt wasatraditional symbol of royal power both in Macedon and in Asia.
67Inother
periods and places, rulers’ hunts have be en an aspect of central negotiation
with local elites, providing lower-ranking notables with opportunities to
come into close contact with the ruler and his court by following the hunt
themselves.68Wasthis a traditional aspect of Persian royal hunts, and was
it one which Alexander’s hunts in Asia took over?
Insum, it is arguably hard to ﬁnd a much clearer afﬁrmation from classical
antiquity of the political importance attached by a clear-sighted monarch tohis court as an instrument of power. In a book on courts, it would be usefultobe able to point to a clear consequence of this emphasis by Alexander
on the court. It is clear that, in the years immediately prior to his death,circumstances had led him to reduce – at least for the time being – his useof Iranian satraps (Bosworth 1980 :9,13). However, whether in some larger
sense his construction of his state as a joint Macedonian and Persian projecthad ‘succeeded’ or ‘failed’ at the time of his death is sufﬁciently unclear in thesources for modern historians to have opposing views. Just to canvass somerecent scholarship, on one view ‘[h]is success was particularly remarkable
with Medes and Persians’ (Hammond 1989 :226); on another, ‘the vision
of a new empire based upon a Macedonian–Persian elite failed’ .
69What is
66Alexander’s avoidance of the Persian royal palace when holding court at Susa (above), and
preference on this occasion for his state tent, may perhaps be explained by his lingering weathereyeonGreek opinion, since for Greeks Susa had always been far more a symbol of the old
Persian power than Persepolis.
67Palagia 2000 :181–4; Briant 2002 :297–9.
68Compare Rawski 1998 :22oncontacts between the lower-ranking members of the Inner Asian
elites and the Qing emperors during imperial hunts. The same point could be made aboutBourbon royal hunts in the ˆIle-de-France.
69Brosius 2003 :192. This position is in danger of becoming a topos, not always in accordance
with the evidence. For instance, the mixed Macedonian/Persian marriages at Susa have been
claimed as at best a ‘limited success’ (2003: 176–7). But of the ‘around eighty’ Iraniannoblewomen married off to Macedonian Companions in this way (Arr.7.4.6), precisely one(Amastris: Berve 1926 ,II: no. 50) is certainly known to have been abandoned by her
Macedonian husband (Craterus), although, since she ended up as the wife of Lysimachus,

112 tony spawforth
undeniable is that within seven years of the ﬁrst introduction of Persian court
trappings Alexander was dead. Historiographical justice probably requires
that we register a non liquet as to how his state-building would have fared
had he ruled for as long as, say, his great predecessor in Asia as imperialconqueror, Cyrus the Great – around twenty-seven years.
Appendix: Alexander’s state tents
A. The ancient texts
[1]The wedding tent. Chares, FGrH 125 F 4 = Athen.538b-539a (late third
century BC):
/p67/p134/p114/p104/p118 /p100 /p150/p110 /p116 /p128 /p166/p100/p101/p107/p134/p116 /p128 /p166/p116/p228/p110 /p112/p101/p114/p170 /p37/p108/p148/p120/p97/p110/p100/p114/p111/p110 /p172/p115/p116/p111/p114/p105/p228/p110 ‘/p194/p116/p101,’/p106/p158/p115/p105/p110 ,‘/p101/p179/p108/p101
/Delta1/p97/p114/p101/p177/p111/p110 ,/p103/p134/p109/p111/p117/p118 /p115/p117/p110/p101/p116/p148/p108/p101/p115/p101/p110 /p151/p97/p117/p116/p111/p211 /p116/p101 /p107/p97/p170 /p116/p228/p110 /p138/p108/p108/p119/p110 /p106/p169/p108/p119/p110 ,/p150/p110/p101/p110/p158/p107/p111/p110/p116/p97
/p107/p97/p170 /p100/p197/p111 /p113/p97/p108/p134/p109/p111/p117/p118 /p107/p97/p116/p97/p115/p107/p101/p117/p97/p115/p134/p109/p101/p110/p111/p118 /p150/p110 /p116 /p128 /p228/p97/p201/p116 /p128/p228/p116/p187/p112/p119/p131 ./p167/p110 /p100/p149 /p111/p173/p107/p111/p118 /p151/p107/p97/p45
/p116/p111/p110/p116/p134/p107/p108/p105/p110/p111/p118 ,/p150/p110 /p128/p230/p151/p107/p134/p115/p116/p104 /p167/p110 /p107/p108/p169/p110/p104 /p107/p101/p107/p111/p115/p109/p104/p109/p148/p110/p104 /p115/p116/p111/p108 /p128 /p166/p103/p97/p109/p105/p107 /p128 /p166/p101/p173/p107/p111/p115/p105 /p109/p110/p228/p110
/p136/p114/p103/p117/p114/p142./p161/p100/p149/p97/p201/p116/p111/p211 /p99/p114/p117/p115/p187/p112/p111/p117/p118 /p167/p110 ./p115/p117/p109/p112/p97/p114/p148/p108/p97/p98/p101/p110 /p100/p149 /p101/p171/p118 /p116/p188 /p115/p117/p109/p112/p187/p115/p105/p111/p110
/p107/p97/p170 /p116/p111/p198/p118 /p171/p100/p105/p111/p120/p148/p110/p111/p117/p118 /p140/p112/p97/p110/p116/p97/p118 /p107/p97/p170 /p107/p97/p116/p148/p107/p108/p105/p110/p101/p110 /p136/p110/p116/p105/p112/p114/p111/p115/p219/p112/p111/p117/p118 /p151/p97/p117/p116 /p128 /p228/p116 /p101/p107 /p97 /p170
/p116/p111/p177/p118 /p138/p108/p108/p111/p105/p118 /p110/p117/p109/p106/p169/p111/p105/p118 ,/p116/p159/p110 /p100/p149 /p108/p111/p105/p112/p159/p110 /p100/p197/p110/p97/p109/p105/p110 /p112/p101/p122/p158/p110 /p116/p101 /p107/p97/p170 /p110/p97/p117/p116/p105/p107/p159/p110 /p107/p97/p170 /p116/p111/p198/p118
/p112/p114/p101/p115/p98/p101/p169/p97/p118 /p107/p97/p170 /p116/p111/p198/p118 /p112/p97/p114/p101/p112/p105/p100/p104/p109/p111/p211/p110/p116/p97/p118 /p150/p110 /p116 /p128 /p166/p97/p201/p108 /p128/p166./p107/p97/p116/p101/p115/p107/p101/p197/p97/p115/p116/p111 /p100/p149 /p190 /p111/p178/p107/p111/p118
/p112/p111/p108/p117/p116/p101/p108/p228/p118 /p107/p97/p170 /p109/p101/p103/p97/p108/p111/p112/p114/p101/p112/p228/p118 /p172/p109/p97/p116/p169/p111/p105/p118 /p116/p101 /p107/p97/p170 /p189/p113/p111/p110/p169/p111/p105/p118 /p112/p111/p108/p117/p116/p101/p108/p148/p115/p105/p110 ,/p203/p112/p188
/p100/p149 /p116/p97/p211/p116/p97 /p112/p111/p114/p106/p117/p114/p111/p177/p118 /p107/p97/p170 /p106/p111/p105/p110/p105/p107/p111/p177/p118 /p99/p114/p117/p115/p111/p117/p106/p148/p115/p105/p110 ./p116/p111/p211 /p100/p149 /p109/p148/p110/p101/p105/p110 /p116/p159/p110 /p115/p107/p104/p110/p159/p110
/p203/p112/p148/p107/p101/p105/p110/p116/p111 /p107/p169/p111/p110/p101/p118 /p101/p171/p107/p111/p115/p97/p112/p158/p99/p101/p105/p118 /p112/p101/p114/p169/p99/p114/p117/p115/p111/p105 /p107/p97/p170 /p100/p105/p134/p108/p105/p113/p111/p105 /p107/p97/p170 /p112/p101/p114/p105/p134/p114/p103/p117/p114/p111/p105 .
/p112/p101/p114/p105/p101/p98/p148/p98/p108/p104/p110/p116/p111 /p100/p149 /p150/p110 /p116 /p128 /p228/p112/p101/p114/p105/p98/p187/p108/p119/p131 /p112/p111/p108/p117/p116/p101/p108/p101/p177/p118 /p97/p201/p108/p97/p177/p97/p105 /p122/p119/p131/p119/p116/p111/p170 /p107/p97/p170 /p100/p105/p134/p99/p114/p117/p45
/p115/p111/p105,/p107/p97/p110/p187/p110/p97/p118 /p155/p99/p111/p117/p115/p97/p105 /p112/p101/p114/p105/p99/p114/p197/p115/p111/p117/p118 /p107/p97/p170 /p112/p101/p114/p105/p114/p103/p197/p114/p111/p117/p118 ./p116/p166/p118 /p100 /p97/p201/p108/p166/p118 /p167/p110 /p116/p188
/p112/p101/p114/p169/p109/p101/p116/p114/p111/p110 /p115/p116/p134/p100/p105/p111/p105 /p116/p148/p115/p115/p97/p114/p101/p118 .’
Chares in the tenth book of his Histories of Alexander says: ‘When he over-
came Darius, he concluded marriages of himself and of his friends besides,constructing ninety-two bridal chambers in the same place. The structure
was large enough for a hundred couches, and in it every couch was adornedwith nuptial coverings, and was made of silver worth twenty minae; but his
owncouch had supports of gold. He also included in his invitation to the
banquet all his personal friends and placed them on couches opposite himselfand the other bridegrooms, while the rest of his forces, both land and naval,he entertained in the courtyard with the foreign embassies and tourists.Moreover the structure was decorated sumptuously and magniﬁcently with
another of Alexander’s Macedonian Companions and subsequently a king (Brosius 2003 :177
n. 13 with refs.), even she is hardly evidence for a failed experiment.

The court of Alexander the Great 113
expensive draperies and ﬁne linens, and underfoot with purple and crim-
son rugs interwoven with gold. T o keep the pavilion ﬁrmly in place therewerecolumns thirty feet high, gilded and silvered and studded with jewels.
The entire enclosure was surrounded with rich curtains having animal pat-ternsinterwoven in gold, their rods being overlaid with gold and silver. The
perimeter of the courtyard measured four stadia.’ (Loeb translation by C. B.Gulick, 1933)
[2]The wedding tent. Aelian, Varia H istoria 8.7 (early third century AD):
/p37/p108/p148/p120/p97/p110/p100/p114/p111/p118 /p194/p116/p101 /Delta1/p97/p114/p101/p177/p111/p110 /p101/p179/p108/p101 ,/p103/p134/p109/p111/p117/p118 /p101/p172/p115/p116/p169/p97 /p107/p97/p170 /p151/p97/p117/p116/p111/p211 /p107/p97/p170 /p116/p228/p110 /p106/p169/p108/p119/p110 .
/p150/p110/p101/p110/p158/p107/p111/p110/p116/p97 /p100/p149 /p167/p115/p97/p110 /p111/p172 /p103/p97/p109/p111/p211/p110/p116/p101/p118 /p107/p97/p170 /p171/p115/p134/p114/p105/p113/p109/p111/p105 /p116/p111/p197/p116/p111/p105/p118 /p111/p172 /p113/p134/p108/p97/p109/p111/p105 ./p167/p110
/p100/p149 /p190 /p136/p110/p100/p114/p228/p110 /p190 /p203/p112/p111/p100/p101/p99/p187/p109/p101/p110/p111/p118 /p107/p97/p170 /p151/p115/p116/p105/p228/p110 /p97/p201/p116/p111/p198/p118 /p151/p107/p97/p116/p111/p110/p116/p134/p107/p108/p105/p110/p111/p118
./p107/p97/p170
/p151/p107/p134/p115/p116/p104 /p107/p108/p169/p110/p104 /p136/p114/p103/p117/p114/p187/p112/p111/p117/p118 /p167/p110 ,/p161/p100/p149/p97/p201/p116/p111/p211 /p99/p114/p117/p115/p187/p112/p111/p117/p118 ,/p107/p97/p170 /p107/p101/p107/p187/p115/p109/p104/p45
/p110/p116/p111 /p112/p142/p115/p97/p105 /p137/p108/p111/p117/p114/p103/p111/p177/p118 /p107/p97/p170 /p112/p111/p105/p107/p169/p108/p111/p105/p118 /p172/p109/p97/p116/p169/p111/p105/p118 /p203/p106/p166/p118 /p98/p97/p114/p98/p97/p114/p105/p107/p166/p118 /p109/p101/p103/p97/p116/p169/p45/p109/p111/p117./p115/p117/p109/p112/p97/p114/p148/p108/p97/p98/p101 /p100/p149 /p101/p171/p118 /p116/p188 /p115/p117/p109/p112/p187/p115/p105/p111/p110 /p107/p97/p170 /p116/p111/p198/p118 /p171/p100/p105/p111/p120/p148/p110/p111/p117/p118 /p107/p97/p170 /p107/p97/p116/p148/p107/p108/p105/p110/p101/p110
/p136/p110/p116/p105/p112/p114/p111/p115/p219/p112/p111/p117/p118 /p151/p97/p117/p116 /p128 /p228./p150/p110 /p100/p149 /p116 /p128 /p166/p97/p201/p108 /p128/p166/p101/p172/p115/p116/p105/p228/p110/p116/p111 /p97/p175 /p116/p101 /p138/p108/p108/p97/p105 /p100/p117/p110/p134/p109/p101/p105/p118 ,/p97/p172
/p112/p101/p122/p97/p170 /p107/p97/p170 /p97/p172 /p110/p97/p117/p116/p105/p107/p97/p170 /p107/p97/p170 /p111/p172 /p105/p112/p112/p101/p177/p118 ,/p107/p97/p170 /p97/p172 /p112/p114/p101/p115/p98/p101/p177/p97/p105 /p100/p149 /p151/p105/p115/p116/p105/p228/p110/p116/p111 /p107/p97/p170 /p111/p172
/p112/p97/p114/p101/p112/p105/p100/p104/p109/p111/p211/p110/p116/p101/p118 /p237/p69/p108/p108/p104/p110/p101/p118 .
When Alexander captured Darius he celebrated his own marriage and that
of his friends. The number of people marrying was ninety, and the bridalchambers equal in number. The hall for the reception and banquet had one
hundred couches. Each couch had silver feet, except his own, which had gold;
they were all decorated with purple or embroidered cloth, of a weave muchprized among the barbarians. He took his personal guests from foreign statestothe banquet and had them seated facing him. In the courtyard there was
afeast for the other forces, the infantry, marines, and cavalry. Ambassadors
and Greeks resident locally were at the feast. (Loeb, translation by N. G.Wilson, 1997)
[3]The audience tent. Phylarchus, FGrH 81 F 44 = Athen.141f–142f (third
century BC):
/p116/p135/p118 /p100/p149 /p99/p114/p117/p115/p142/p118 /p112/p108/p97/p116/p134/p110/p111/p117/p118 /p107/p97/p170 /p116/p159/p110 /p99/p114/p117/p115/p166/p110 /p138/p109/p112/p101/p108/p111/p110 /p203/p106 /p236/p165/p110 /p111/p172 /p80/p101/p114/p115/p228/p110
/p98/p97/p115/p105/p108/p101/p177/p118 /p150/p99/p114/p104/p109/p134/p116/p105/p122/p111/p110 /p112/p111/p108/p108/p134/p107/p105/p118 /p107/p97/p113/p158/p109/p101/p110/p111/p105 …/p150/p108/p134/p116/p116/p119 /p106/p104/p115/p170/p110 /p190 /p70/p197/p108/p97/p114/p99/p111/p118
/p106/p97/p169/p110/p101/p115/p113/p97/p105 /p116/p166/p118 /p107/p97/p113 /p236/p161/p109/p148/p114/p97/p110 /p151/p107/p134/p115/p116/p111/p116/p101 /p103/p105/p110/p111/p109/p148/p110/p104/p118 /p112/p97/p114 /p236/p37/p108/p101/p120/p134/p110/p100/p114/p119/p131 /p100/p97/p112/p134/p110/p104/p118 .
/p167/p110 /p103/p135/p114 /p97/p201/p116/p111/p211 /p161 /p115/p107/p104/p110/p159 /p107/p108/p105/p110/p228/p110 /p114 /p233,/p99/p114/p117/p115/p111/p177 /p100/p149 /p107/p169/p111/p110/p101/p118 /p110 /p233/p107/p97/p116/p101/p177/p99/p111/p110 /p97/p201/p116/p158/p110 .
/p111/p172 /p100/p149 /p203/p112/p101/p114/p116/p101/p169/p110/p111/p110/p116/p101/p118 /p111/p201/p114/p97/p110/p169/p115/p107/p111/p105 /p100/p105/p134/p99/p114/p117/p115/p111/p105 /p112/p111/p105/p107/p169/p108/p109/p97/p115/p105/p110 /p150/p107/p112/p101/p112/p111/p110/p104/p109/p148/p110/p111/p105/p112/p111/p108/p117/p116/p101/p108/p148/p115/p105/p110 /p150/p115/p107/p148/p112/p97/p122/p111/p110 /p116/p188/p110 /p138/p110/p119 /p116/p187/p112/p111/p110 ./p107/p97/p170 /p112/p114/p228/p116/p111/p105 /p109/p149/p110 /p80/p148/p114/p115/p97/p105 /p106 /p233/p109/p104/p108/p111/p45
/p106/p187/p114/p111/p105 /p112/p101/p114/p170 /p97/p201/p116/p159/p110 /p150/p110/p116/p188/p118 /p101/p172/p115/p116/p158/p107/p101/p115/p97/p110 /p112/p111/p114/p106/p117/p114/p97/p177/p118 /p107/p97/p170 /p109/p104/p108/p169/p110/p97/p105/p118 /p150/p115/p113/p166/p115/p105/p110 /p150/p120/p104/p45/p115/p107/p104/p109/p148/p110/p111/p105 /p109/p101/p116/p135 /p100/p149 /p116/p111/p197/p116/p111/p117/p118 /p116/p111/p120/p187/p116/p97/p105 /p116/p188/p110 /p136/p114/p105/p113/p109/p188/p110 /p99/p169/p108/p105/p111/p105 ,/p111/p172 /p109/p149/p110 /p106/p108/p187/p103/p105/p110/p97

114 tony spawforth
/p150/p110/p100/p101/p100/p117/p107/p187/p116/p101/p118 ,/p111/p172 /p100/p149 /p203/p115/p103/p105/p110/p111/p98/p97/p106/p166 ,/p112/p111/p108/p108/p111/p170 /p100/p149 /p107/p97/p170 /p107/p117/p134/p110/p101/p97 /p101/p178/p99/p111/p110 /p112/p101/p114/p105/p98/p187/p108/p97/p105/p97 .
/p112/p114/p111/p101/p105/p115/p116/p158/p107/p101/p115/p97/p110 /p100/p149 /p116/p111/p197/p116/p119/p110 /p136/p114/p103/p117/p114/p134/p115/p112/p105/p100/p101/p118 /p77/p97/p107/p101/p100/p187/p110/p101/p118 /p112/p101/p110/p116/p97/p107/p187/p115/p105/p111/p105 ./p107/p97/p116/p135 /p100/p149
/p109/p148/p115/p104/p110 /p116/p159/p110 /p115/p107/p104/p110/p159/p110 /p99/p114/p117/p115/p111/p211/p118 /p150/p116/p169/p113/p101/p116/p111 /p100/p169/p106/p114/p111/p118 ,/p150/p106/p236/p111/p213 /p107/p97/p113/p158/p109/p101/p110/p111/p118 /p150/p99/p114/p104/p109/p134/p116/p105/p122/p101/p110
/p190/p37/p108/p148/p120/p97/p110/p100/p114/p111/p118 /p116/p228/p110 /p115/p119/p109/p97/p116/p111/p106/p117/p108/p134/p107/p119/p110 /p112/p97/p110/p116/p97/p99/p187/p113/p101/p110 /p150/p106/p101/p115/p116/p104/p107/p187/p116/p119/p110 ./p155/p120/p119/p113/p101/p110 /p100/p149
/p107/p197/p107/p108/p119/p131 /p116/p166/p118 /p115/p107/p104/p110/p166/p118 /p116/p188 /p116/p228/p110 /p150/p108/p101/p106/p134/p110/p116/p119/p110 /p138/p103/p104/p109/p97 /p100/p105/p101/p115/p107/p101/p117/p97/p115/p109/p148/p110/p111/p110 /p150/p106/p101/p105/p115/p116/p158/p107/p101/p105/p107/p97/p170 /p77/p97/p107/p101/p100/p187/p110/p101/p118 /p99/p169/p108/p105/p111/p105 /p77/p97/p107/p101/p100/p111/p110/p105/p107/p135/p118 /p115/p116/p111/p108/p135/p118 /p155/p99/p111/p110/p116/p101/p118 ,/p101/p178/p116/p97 /p109/p197/p114/p105/p111/p105 /p80/p148/p114/p115/p97/p105 ,/p116/p187
/p116/p101 /p116/p159/p110 /p112/p111/p114/p106/p197/p114/p97/p110 /p155/p99/p111/p110 /p112/p108/p166/p113/p111/p118 /p101/p171/p118 /p112/p101/p110/p116/p97/p107/p111/p115/p169/p111/p117/p118 /p167/p110 ,/p111/p179/p118 /p37/p108/p148/p120/p97/p110/p100/p114/p111/p118 /p155/p100/p119/p107/p101
/p106/p111/p114/p101/p177/p110 /p116/p159/p110 /p115/p116/p111/p108/p159/p110 /p116/p97/p197/p116/p104/p110 ./p116/p111/p115/p111/p197/p116/p104/p110 /p100/p149 /p192/p110/p116/p119/p110 /p107/p97/p170 /p116/p228/p110 /p106/p169/p108/p119/p110 /p107/p97/p170 /p116/p228/p110 /p113/p101/p114/p45
/p97/p112/p101/p117/p187/p110/p116/p119/p110 /p111/p201/p100/p101/p170/p118 /p150/p116/p187/p108/p109/p97 /p112/p114/p111/p115/p112/p111/p114/p101/p197/p101/p115/p113/p97/p105 /p37/p108/p101/p120/p134/p110/p100/p114/p119/p131
./p116/p111/p105/p111/p211/p116/p111/p110 /p150/p103/p101/p103/p187/p110/p101/p105
/p116/p188 /p112/p101/p114/p170 /p97/p201/p116/p188/p110 /p136/p120/p169/p119/p109/p97 .
Moreover, the famous plane-trees of gold, even the golden vine under which
the Persian kings often sat and held court . . . appeared to be of less worth, saysPhylarchus, than the expense lavished daily on all occasions at Alexander’scourt. For his pavilion alone contained a hundred couches and was sup-
ported by ﬁfty golden uprights. The canopies stretched over the upper parttocoverthe whole were elaborately worked with gold in sumptuous embroi-
deries. Inside, all round it, stood ﬁrst of all ﬁve hundred Persians, Apple-bearers, with gay uniforms of purple and quince-yellow; after them bow-men to the number of a thousand, some dressed in ﬂame-colour, others incrimson; but many, too, had mantles of dark blue. At the head of these stoodﬁve hundred Silver-Shields, Macedonians. In the centre of the pavilion wasplaced a golden chair, sitting on which Alexander held court with his body-guard stationed close on all sides. Outside the tent the elephant-divisionwas posted near in a circle with full equipment, also a thousand Macedo-nians in Macedonian uniform, next ten thousand Persians, and the largebody, amounting to ﬁve hundred, who wore the purple; for Alexander hadgranted them the privilege of wearing this garment. And the number of his
friends and servitors being so great, no one dared to approach Alexander;such was the majesty associated with his person. (Loeb translation by C. B.Gulick, 1933)
[4]The audience tent. Polyaenus, Strategemata 4.3.24 (mid-second century
AD):
/p37/p108/p148/p120/p97/p110/p100/p114/p111/p118 /p150/p110 /p109/p149/p110 /p116/p111/p177/p118 /p77/p97/p107/p101/p100/p187/p115/p105/p110 /p163 /p150/p110 /p116/p111/p177/p118 /p237/p69/p108/p108/p104/p115/p105 /p100/p105/p107/p134/p122/p119/p110 /p109/p148/p116/p114/p105/p111/p110 /p107/p97/p170
/p100/p104/p109/p111/p116/p105/p107/p188/p110 /p155/p99/p101/p105/p110 /p116/p188 /p100/p105/p107/p97/p115/p116/p158/p114/p105/p111/p110 /p150/p100/p111/p107/p169/p109/p97/p122/p101/p110 ,/p150/p110 /p100/p149 /p116/p111/p177/p118 /p98/p97/p114/p98/p134/p114/p111/p105/p118 /p108/p97/p109/p112/p114/p188/p110
/p107/p97/p170 /p115/p116/p114/p97/p116/p104/p103/p105/p107/p187/p110 ,/p150/p107/p112/p108/p158/p115/p115/p119/p110 /p116/p111/p198/p118 /p98/p97/p114/p98/p134/p114/p111/p117/p118 /p107/p97/p170 /p116 /p128 /p228/p116/p111/p211 /p100/p105/p107/p97/p115/p116/p104/p114/p169/p111/p117
/p115/p99/p158/p109/p97/p116/p105 ./p150/p110 /p103/p111/p211/p110 /p66/p134/p107/p116/p114/p111/p105/p118 /p107/p97/p170 /p235/p85/p114/p107/p97/p110/p169/p111/p105/p118 /p107/p97/p170 /p73/p110/p100/p111/p177/p118 /p100/p105/p107/p134/p122/p119/p110 /p101/p178/p99/p101 /p116/p159/p110
/p115/p107/p104/p110/p159/p110 /p230/p100/p101 /p112/p101/p112/p111/p105/p104/p109/p148/p110/p111/p110 ./p161/p115/p107/p104/p110/p159 /p116/p188 /p109/p148/p103/p101/p113/p111/p118 /p167/p110 /p107/p108/p105/p110/p228/p110 /p151/p107/p97/p116/p187/p110
./p99/p114/p197/p115/p101/p111/p105
/p107/p169/p111/p110/p101/p118 /p203/p112/p101/p116/p169/p113/p101/p110/p116/p111 /p97/p201/p116 /p128 /p166/p112/p101/p110/p116/p158/p107/p111/p110/p116/p97 ../p203/p112/p101/p114/p116/p101/p169/p110/p111/p110/p116/p101/p118 /p111/p201/p114/p97/p110/p169/p115/p107/p111/p105 /p100/p105/p134/p99/p114/p117/p115/p111/p105 ,

The court of Alexander the Great 115
/p112/p111/p105/p107/p169/p108/p109/p97/p115/p105/p110 /p150/p107/p112/p101/p112/p111/p110/p104/p109/p148/p110/p111/p105 ,/p116/p188/p110 /p138/p110/p119 /p116/p187/p112/p111/p110 /p150/p115/p107/p148/p112/p97/p122/p111/p110 ./p80/p148/p114/p115/p97/p105 /p109/p149/p110 /p112/p114/p228/p45
/p116/p111/p105 /p112/p101/p110/p116/p97/p107/p187/p115/p105/p111/p105 /p109/p104/p108/p111/p106/p187/p114/p111/p105 /p112/p101/p114/p170 /p116/p159/p110 /p115/p107/p104/p110/p159/p110 /p150/p110/p116/p188/p118 /p175/p115/p116/p97/p110/p116/p111 /p112/p111/p114/p106/p117/p114/p97/p177/p118 /p107/p97/p170/p109/p104/p108/p169/p110/p97/p105/p118 /p150/p115/p113/p166/p115/p105/p110 /p150/p120/p101/p115/p107/p104/p109/p148/p110/p111/p105 ./p109/p101/p116/p135 /p100/p149 /p116/p111/p198/p118 /p109/p104/p108/p111/p106/p187/p114/p111/p117/p118 /p116/p111/p120/p187/p116/p97/p105 /p116/p188/p110 /p173/p115/p111/p110
/p136/p114/p105/p113/p109/p188/p110 /p155/p99/p111/p110/p116/p101/p118 /p116/p97/p177/p118 /p150/p115/p113/p166/p115/p105 /p100/p105/p158/p108/p108/p97/p116/p116/p111/p110
./p111/p172 /p109/p149/p110 /p103/p135/p114 /p97/p201/p116/p228/p110 /p106/p108/p187/p103/p105/p110/p97 ,/p111/p172 /p100/p149
/p107/p117/p134/p110/p101/p97 ,/p111/p172 /p100/p149 /p203/p115/p103/p105/p110/p111/p98/p97/p106/p166 /p112/p101/p114/p105/p101/p98/p148/p98/p108/p104/p110/p116/p111 ./p116/p111/p197/p116/p119/p110 /p112/p114/p111/p115/p116/p97/p110/p116/p111 /p77/p97/p107/p101/p100/p187/p110/p101/p118
/p136/p114/p103/p117/p114/p134/p115/p112/p105/p100/p101/p118 /p112/p101/p110/p116/p97/p107/p187/p115/p105/p111/p105 /p116/p228/p110 /p109/p101/p103/p169/p115/p116/p119/p110 /p136/p110/p100/p114/p228/p110 ./p107/p97/p116/p135 /p100/p149 /p116/p188 /p109/p148/p115/p111/p110 /p116/p166/p118
/p115/p107/p104/p110/p166/p118 /p190 /p99/p114/p117/p115/p111/p211/p118 /p155/p107/p101/p105/p116/p111 /p113/p114/p187/p110/p111/p118 ,/p150/p106/p236/p111/p213 /p112/p114/p111/p107/p97/p113/p158/p109/p101/p110/p111/p118 /p150/p99/p114/p104/p109/p134/p116/p105/p122/p101/p110./p111/p172 /p115/p119/p109/p97/p45
/p116/p111/p106/p197/p108/p97/p107/p101/p118 /p150/p106/p101/p115/p116/p158/p107/p101/p115/p97/p110 /p151/p107/p97/p116/p148/p114/p119/p113/p101/p110 /p116/p111/p211 /p98/p97/p115/p105/p108/p148/p119/p118 /p100/p105/p107/p134/p122/p111/p110/p116/p111/p118 ./p150/p110 /p107/p197/p107/p108 /p128 /p228/p116/p166/p118
/p115/p107/p104/p110/p166/p118 /p116/p188 /p116/p228/p110 /p150/p108/p101/p106/p134/p110/p116/p119/p110 /p138/p103/p104/p109/p97 /p100/p105/p101/p115/p107/p101/p117/p97/p115/p109/p148/p110/p111/p110 /p150/p106/p101/p115/p116/p158/p107/p101/p105 /p107/p97/p170 /p77/p97/p107/p101/p100/p187/p110/p101/p118/p99/p169/p108/p105/p111/p105 /p115/p116/p111/p108/p135/p118 /p77/p97/p107/p101/p100/p111/p110/p105/p107/p135/p118 /p155/p99/p111/p110/p116/p101/p118 ./p150/p112/p170 /p100/p149 /p116/p111/p197/p116/p111/p105/p118 /p112/p101/p110/p116/p97/p107/p187/p115/p105/p111/p105 /p83/p111/p197/p115/p105/p111/p105
/p112/p111/p114/p106/p117/p114/p111/p115/p99/p158/p109/p111/p110/p101/p118 ,/p107/p97/p170 /p109/p101/p116/p135 /p116/p111/p197/p116/p111/p117/p118 /p150/p110 /p107/p197/p107/p108/p119/p131 /p112/p134/p110/p116/p119/p110 /p80/p148/p114/p115/p97/p105 /p109/p197/p114/p105/p111/p105 (/p80/p101/p114/p45
/p115/p228/p110)/p111/p172 /p107/p134/p108/p108/p105/p115/p116/p111/p105 /p107/p97/p170 /p109/p148/p103/p105/p115/p116/p111/p105 ,/p107/p101/p107/p97/p108/p108/p119/p112/p105/p115/p109/p148/p110/p111/p105 /p112/p97/p110/p116/p170 /p107/p187/p115/p109/p119/p131 /p80/p101/p114/p115/p105/p107 /p128 /p228,
/p112/p134/p110/p116/p101/p118 /p136/p107/p105/p110/p134/p107/p97/p118 /p155/p99/p111/p110/p116/p101/p118 ./p116/p111/p105/p187/p110/p100/p101 /p167/p110 /p37/p108/p101/p120/p134/p110/p100/p114/p111/p117 /p116/p188 /p100/p105/p107/p97/p115/p116/p158/p114/p105/p111/p110 /p150/p110 /p116/p111/p177/p118
/p98/p97/p114/p98/p134/p114/p111/p105/p118 .
When deciding legal cases among the Macedonians or Greeks, Alexander
preferred to have a modest and common courtroom, but among the bar-barians he preferred a brilliant courtroom suitable for a general, aston-ishing the barbarians even by the courtroom’s appearance. When decidingcases among the Bactrians, Hyrcanians and Indians he had a tent madeas follows: the tent was large enough for one hundred couches; ﬁfty goldpillars supported it; embroidered gold canopies, stretched out above, cov-ered the place. Inside the tent ﬁve hundred Persian Apple Bearers stoodﬁrst, dressed in purple and yellow clothing. After the Apple Bearers stoodan equal number of archers in different clothing, for some wore ﬂame-coloured, some dark blue, and some scarlet. In front of these stood Mac-
edonian Silver Shields, ﬁve hundred of the tallest men. In the middle ofthe tent sat the gold throne, on which he sat to give audiences. Bodyguardsstood on each side when the king heard cases. In a circle around the tentstood the corps of elephants Alexander had equipped, and one thousandMacedonians wearing Macedonian apparel. Next to these were ﬁve hun-
dred Susians dressed in purple, and after them, in a circle around them all,tenthousand Persians, the handsomest and tallest (of Persians), adorned
entirely with Persian decorations, and all carrying short swords. Such wasAlexander’s courtroom among the barbarians. (translation by P . Krentz andE. Wheeler)
[5]The audience tent. Aelian, Varia H istoria 9.3 (early third century AD):
/p97/p201/p116 /p128/p228/p100/p149/p37/p108/p101/p120/p134/p110/p100/p114/p119/p128 /p161 /p109/p149/p110 /p115/p107/p104/p110/p159 /p167/p110 /p107/p108/p105/p110/p228/p110 /p151/p107/p97/p116/p187/p110 ,/p99/p114/p117/p115/p111/p177 /p100/p149 /p107/p169/p111/p110/p101/p118 /p112/p101/p110/p45
/p116/p158/p107/p111/p110/p116/p97 /p100/p105/p101/p105/p108/p158/p106/p101/p115/p97/p110 /p97/p201/p116/p159/p110 /p107/p97/p170 /p116/p188/p110 /p192/p114/p111/p106/p111/p110 /p97/p201/p116/p166/p118 /p136/p110/p101/p177/p99/p111/p110 ,/p97/p201/p116/p188/p118 /p100/p149 /p190

116 tony spawforth
/p192/p114/p111/p106/p111/p118 /p100/p105/p134/p99/p114/p117/p115/p111/p118 /p167/p110 /p107/p97/p170 /p150/p107/p112/p101/p112/p187/p110/p104/p116/p111 /p112/p111/p105/p107/p169/p108/p109/p97/p115/p105 /p112/p111/p108/p117/p116/p101/p108/p148/p115/p105 ./p107/p97/p170 /p112/p114/p228/p116/p111/p105
/p109/p149/p110 /p80/p148/p114/p115/p97/p105 /p112/p101/p110/p116/p97/p107/p187/p115/p105/p111/p105 /p111/p172 /p107/p97/p108/p111/p197/p109/p101/p110/p111/p105 /p109/p104/p108/p111/p106/p187/p114/p111/p105 /p112/p101/p114/p170 /p97/p201/p116/p159/p110 /p150/p110/p116/p188/p118 /p101/p172/p115/p116/p158/p107/p101/p45/p115/p97/p110 /p112/p111/p114/p106/p117/p114/p142/p118 /p107/p97/p170 /p109/p104/p108/p169/p110/p97/p118 /p160/p115/p113/p104/p109/p148/p110/p111/p105 /p115/p116/p111/p108/p134/p118
./p150/p112/p236/p97/p201/p116/p111/p177/p118 /p100/p149 /p116/p111/p120/p187/p116/p97/p105 /p99/p169/p108/p105/p111/p105 ,
/p106/p108/p187/p103/p105/p110/p97 /p150/p110/p100/p101/p100/p117/p107/p187/p116/p101/p118 /p107/p97/p170 /p203/p115/p103/p105/p110/p111/p98/p97/p106/p166./p112/p114/p188 /p100/p149 /p116/p111/p197/p116/p119/p110 /p111/p172 /p136/p114/p103/p117/p114/p134/p115/p112/p105/p100/p101/p118
/p112/p101/p110/p116/p97/p107/p187/p115/p105/p111/p105 /p77/p97/p107/p101/p100/p187/p110/p101/p118 ./p150/p110 /p109/p148/p115/p104 /p100/p149 /p116 /p128 /p166/p115/p107/p104/p110 /p128 /p166/p99/p114/p117/p115/p111/p211/p118 /p150/p116/p169/p113/p101/p116/p111 /p100/p169/p106/p114/p111/p118 /p107/p97/p170
/p150/p112 /p97/p201/p116 /p128 /p228/p107/p97/p113/p158/p109/p101/p110/p111/p118 /p37/p108/p148/p120/p97/p110/p100/p114/p111/p118 /p150/p99/p114/p104/p109/p134/p116/p105/p122/p101 ,/p112/p101/p114/p105/p101/p115/p116/p219/p116/p119/p110 /p97/p201/p116 /p128 /p228/p112/p97/p110/p45
/p116/p97/p99/p187/p113/p101/p110 /p116/p228/p110 /p115/p119/p109/p97/p116/p111/p106/p117/p108/p134/p107/p119/p110 ./p112/p101/p114/p158/p128/p101/p105 /p100/p149 /p116/p159/p110 /p115/p107/p104/p110/p159/p110 /p112/p101/p114/p169/p98/p111/p108/p111/p118 ,/p155/p110/p113/p97 /p167/p115/p97/p110
/p77/p97/p107/p101/p100/p187/p110/p101/p118 /p99/p169/p108/p105/p111/p105 /p107/p97/p170 /p80/p148/p114/p115/p97/p105 /p109/p197/p114/p105/p111/p105 ./p107/p97/p170 /p111/p201/p100/p101/p170/p118 /p150/p116/p187/p108/p109/p97 /p231/p97/p131/p100/p169/p119/p118 /p112/p114/p111/p115/p101/p108/p113/p101/p177/p110.
/p97/p201/p116 /p128/p228/p112/p111/p108/p198 /p103/p135/p114 /p167/p110 /p116/p188 /p150/p120 /p97/p201/p116/p111/p211 /p100/p148/p111/p118 /p136/p114/p113/p148/p110/p116/p111/p118 /p203/p112/p188 /p106/p114/p111/p110/p158/p109/p97/p116/p111/p118 /p107/p97/p170 /p116/p197/p99/p104/p118
/p101/p171/p118 /p116/p117/p114/p97/p110/p110/p169/p100/p97 .
Alexander’s own tent could accommodate a hundred beds. Fifty gold pil-
lars divided it and supported the roof, which was gilded and expensivelyembroidered. Inside it stood in line ﬁrst of all ﬁve hundred Persians, calledthe apple bearers, wearing cloaks of purple and quince yellow; then cameathousand archers dressed in ﬂame colour and scarlet. In front of these
werethe ﬁve hundred Macedonians with silver shields. In the middle of
the tent was a golden throne, on which Alexander sat to transact business,surrounded on all sides by bodyguards. An enclosure wall around the tentwas manned by a thousand Macedonians and ten thousand Persians. Noone dared approach him without good reason, as he aroused great fear; hispride and good fortune had raised him to the position of a tyrant. (Loeb,translation by N. G. Wilson, 1997)
[6]Alexander’s tent on the River Oxus, 328 BC. Plutarch, Life of Alexander
57.4 (around AD 100):
/p190/p103/p135/p114 /p150/p112/p170 /p116/p228/p110 /p115/p116/p114/p119/p109/p97/p116/p111/p106/p117/p108/p134/p107/p119/p110 /p116/p101/p116/p97/p103/p109/p148/p110/p111/p118 /p136/p110/p159/p114 /p77/p97/p107/p101/p100/p219/p110 ,/p192/p110/p111/p109/p97
/p80/p114/p187/p120/p101/p110/p111/p118 ,/p116/p128/p166/p98/p97/p115/p105/p108/p105/p107 /p128 /p166/p115/p107/p104/p110 /p128 /p166/p99/p219/p114/p97/p110 /p189/p114/p197/p116/p116/p119/p110 /p112/p97/p114/p135 /p116/p188/p110 /p236/p87/p120/p111/p110 /p112/p111/p116/p97/p109/p188/p110
/p136/p110/p101/p107/p134/p108/p117/p102/p101 /p112/p104/p103/p159/p110 /p203/p103/p114/p111/p211 /p108/p105/p112/p97/p114/p111/p211 /p107/p97/p170 /p112/p105/p109/p101/p108/p219/p100/p111/p117/p118
.
The Macedonian, namely, who was set over those in charge of the royalequipage, Proxenus by name, as he was digging a place for the king’s tentalong the river Oxus, uncovered a spring of liquid which was oily and fatty(etc.). (Loeb translation by B. Perrin, 1949)
[7]The royal tent at the Battle of the river Hydaspes, 326 BC. Curtius, History
ofAlexander the Great 8.13.20 (?ﬁrst century AD):
Alexander in diversa parte ripae statui suum tabernaculum iussit assue-
tamque comitari ipsum cohortem ante id tabernaculum stare et omnemapparatum regiae magniﬁcentiae hostium oculis de industria ostendi.

The court of Alexander the Great 117
Alexander ordered his tent to be pitched elsewhere on the river bank, the unit
usually in attendance on him to stand guard before it, and all the sumptuoustrappings of royalty to be deliberately ﬂaunted before the enemy’s eyes.
(translated by J. Y ardley, 1984)
[8]Ar o y a lt e nt aﬂoat on the River Hydraortes, 326 BC. Curtius, History of
Alexander the Great 9.6.1 (?ﬁrst century AD):
duobus navigiis iunctis, statui in medium undique conspicuum tabernacu-
lum iussit.
[Alexander] ordered two ships to be lashed together, and his tent to be set
up in the centre.
[9]Aroyal tent used in a banquet for Indian tribes, 326/5 BC. Curtius,
History of Alexander the Great 9.7.15 (?ﬁrst century AD):
Invitatis deinde ad epulas legatis gentium regulisque, exornari convivium
iussit. c aurei lecti modicis intervallis positi erant, lectis circumdederat aulaea
purpura auroque fulgentia, quidquid au tapud Persas vetere luxu aut apud
Macedonas nova inmutatione corruptum erat, confusis utriusque gentis
vitiis, in illo convivio ostendens.
Alexander then invited the ambassadors and petty kings of the two tribes to
abanquet and had a sumptuous feast prepared for them. A hundred gold
couches were set out a short distance from each other, around which he
had set tapestries glittering with purple and gold. In that banquet he put onshow all the decadence that had long existed among the soft-living Persiansor had been recently acquired by the Macedonians, thus combining the vicesof the two peoples. (translated by J. Y ardley, 1984 )
[10] Elements of the tent in Rome. Pliny, Historia Naturalis 34.48 (ﬁrst
century AD):
Alexandri quoque Magni tabernaculum sustinere traduntur solitae stat-uae, ex quibus duae ante Martis Ultoris aedem dicatae sunt, totidem anteregiam.
Itis also said that the tent of Alexander the Great was regularly erected with
{four}statues as tent-poles, two of which have now been dedicated to stand
in front of the temple of Mars the Avenger and two in front of the RoyalPalace. (Loeb, translation by H. Rackham, 1952; ‘four’ is not in the Latin)

118 tony spawforth
[11] The ouranos of Persian royal tents. Photius, Lexicon ,s.v./p111/p201/p114/p97/p110/p187/p118 (ninth
century AD):
/p80/p148/p114/p115/p97/p105 /p100/p149 /p116/p135/p118 /p98/p97/p115/p105/p108/p101/p169/p111/p117/p118 /p115/p107/p104/p110/p135/p118 /p107/p97/p170 /p97/p201/p108/p135/p118 ,/p230/p110 /p116/p135 /p107/p97/p108/p197/p109/p109/p97/p116/p97 /p107/p117/p107/p108/p111/p116/p101/p114/p166 ,
/p111/p201/p114/p97/p110/p111/p198/p118 .
InPersia the royal tents and courts, with circular ceilings, (like) skies.
B.Commentary
1. Chares, Alexander’s chamberlain ( eisangeleus ), from whom Aelian’s
account [2]patently derives, described in the tenth book of his Histories
the tent used for the wedding feast at Susa in 324 [1].P h y larchus, Polyaenus
and Aelian (again), based ultimately on a primary source which is lost,
described Alexander’s audience tent [3–5] .T h e two tents were one and the
same, as seen long ago (Jacoby’s commentary to Chares, F 4).
2. The use of the imperfect /p150/p99/p231/p104/p109/p134/p116/p105/p122/p101 of Alexander holding audience
shows that the original text which passages 3–5reﬂect was not describing a
speciﬁc audience, but a speciﬁc run of audiences. The ﬁve hundred Susians(Polyaenus) were presumably recruited during Alexander’s stay at Susa in324. So this is a description of Alexander’s audience-giving while at Susa, asseen by Bosworth ( 1980 :8).
3. Jacoby (ibid.), following Droysen, assumed that book 10 of Chares’
lost history was the source for the descriptions of both tents. If so, Chares
must have described the tent’s ﬁttings at Susa twice in the same book: once
in his account of the wedding, and once in a description of Alexander’saudience-giving. Anything is possible, but this repetition seems a bit odd.Moreover, while Chares gives the height of the tent’s columns but not their
number, the description drawn on by Phylarchus, Polyaenus and Aelian
does the reverse. Therefore this description probably derived from anotherauthor. The obvious candidate is Ephippus, who also described Alexandergiving audience at Babylon ( FGrH 126 F 4). Its predictable association with
tyranny in Aelian’s version, if not his own interpretation, would suit this
attribution, given Ephippus’ marked hostility to Alexander (Pearson 1960 :
63). By contrast, the Macedonian patriotism of Polyaenus seems to havetoned down this judgement.
4. Polyaenus includes two details about the audience-tent not found in
Phylarchus or Aelian: (1) that Alexander had it specially made, and (2)that it served as his ‘court-room’ among ‘the Bactrians, Hyrcanians andIndians’ . Here Polyaenus refers sketchily to the years 330–326 BC, dur-
ing which Alexander did indeed campaign in Hyrcania (330 BC), Bactria

The court of Alexander the Great 119
(329–327 BC) and India (327–326 BC). Plutarch ( Al.45.1) places his initial
adoption of Persian court-ceremonial in autumn 330, immediately after his
campaign in Hyrcania and during his stay in Parthyaea (Bosworth 1980 :
5–6). The acquisition of the tent at this juncture would then have coincidedwith Alexander’s initial elaboration of his court. Miller ( 1997 :51) pointed
out that the Susa tent ‘so well parallels the Achaemenid tradition in palacearchitecture that it may have been the tent of the Persian king’ . For the severalPersian royal tents which fell into Alexander’s hands, see above, pp. 96–7.
5. Bosworth ( 1980 :8)doubted whether ‘such a mammoth structure could
have accompanied Alexander on all his travels’ . The whole point about tents,however, is that they can be disassembled. Moreover, the royal baggagetrain did not follow Alexander everywhere. Later in 330, for instance, in
the campaign against the Tapuri, he separated from ‘the waggons and thebaggage-train’ (Arr.3.23.2); and in 328 BC, during the second invasion ofSogdiana, ‘the royal service’ ( basilik¯ etherapeia )was left behind at Zariaspa
(Arr.4.16.6; Berve 1926 ,I:25). On the other hand, the ‘royal equipment’
(basilik¯ ep a r askeu¯ e), along with women, children and transport animals,
was sufﬁciently indispensible that it followed Alexander on his gruellingmarch across the Gedrosian desert in 326/5 BC (Arr.6.25.5).
6. Polyaenus illustrated the use of the audience-tent in 330–326 BC with
adescription speciﬁc to 324 BC. In this sense, at any rate, ‘his information
is garbled to some extent’ (Bosworth 1980: 8). One would like to knowwhether this was the only detailed description of the audience-tent to whichhe had access. The tent itself may have been refurbished after the return fromIndia, especially if it had formed part of the royal baggage train ‘swept away’
byﬂood-water in the Gedrosian desert (Arr.6.25.5). As suggested above
(p. 89), it was not necessarily the case that the ceremonial for the audi-ences at Susa was more impressive than in the early days of the tent’s use,although the associated show of military force had now evolved to includelarge numbers of Persian troops in parade-uniforms, as well as Indian ele-phants. Then again, there were any number of eye-witnesses at Susa, Greek aswell as Macedonian. The source of Polyaenus’ information, both the detailed
description and the tent’s prior history, is uncertain. W ¨ofﬂin (1860; cited by
Krentz and Wheeler 1994 )thought of Phylarchus. But the speciﬁcity of
Polyaenus’ ‘Susians’ , where Phylarchus simply has a ‘large body’ , suggests
that Polyaenus was closer to the original than Phylarchus. Polyaenus washimself a native of Macedonia, where interest in Alexander was marked inthe early Roman empire (Spawforth 2006b: 17–18), and presumably he hadaccess to a fair amount of the earlier Greek literature on the king. Aelian’sversion, while simpliﬁed, has a detail not found in the other two, namely,

120 tony spawforth
theperibolos or screen wall which enclosed the tent, a fact which seems
to exclude his direct use of Athenaeus ( pace N. G. Wilson, Loeb, p. 285,
note a). Since the same material about the luxury of the Companions pre-
cedes the tent description in both authors (Athen.539b–d; Ael. VH 9.3),
both may have adapted an unknown, earlier, anthology (see N. G. Wilson,pp. 10–11 of the Loeb).
7. Gulick translated tous parepid¯ emountas of Chares as ‘tourists’ , noted
Aelian’s version ( [2]), and (p. 434 note) took these ‘tourists’ to be ‘the
Greeks who happened to be visiting the city’ (i.e. Susa); Wilson has ‘Greek
residents locally’ . Berve ( 1926 ,I:80–1) correctly saw that these were visitors,
not to Susa as such, but to Alexander’s court, including them among his‘voluntary courtiers’; pace Aelian, their Greekness should not be assumed.
Berve took /p150/p110 /p116 /p128/p166/p97/p201/p108 /p128/p166with/p116/p111/p198/p118 /p112/p97/p114/p101/p112/p105/p100/p104/p109/p111/p211/p110/p116/p97/p118 alone, ‘the visitors to
the court’ (in the abstract sense). But aul¯ehere has its concrete meaning,
serving to locate Chares’ list of additional diners not eating in the tent(as in Gulick’s translation). The word returns later in the same extract,when Chares gave the dimensions which explained the courtyard’s hugecapacity. Aelian’s version mistakenly took /p150/p110 /p116 /p128/p166/p97/p201/p108 /p128/p166with the feasting troops
only, leaving the ambassadors and Greek visitors, by implication, to feastelsewhere.
8. The ancient texts may give some clues as to the shape of the tents.
Ancient etymology derived ‘tabernaculum’ [8–9] from ‘taberna’ , a structure
built from boards or tabulae (Lewis-Short, s.v. ‘tabernaculum’ , citing Festus).
Thus ‘tabernaculum’ is assumed to indicate a rectangular tent (e.g. Perron1990: 220). The gold-threaded ouraniskoi or ‘little skies’ [3–4] ,c onstitut-
ing the feasting- and audience-tent’s ceiling, evoke the circular canopies orouranoi which were a feature of Persian royal tents [11] and which Alexander
evidently imitated (Lavagne 1988 :96–7). Modern scholars have speculated
that this tent was ‘au d ˆome d’or’ (Lavagne 1988: 96–7); that the ouraniskos
was a central feature in an otherwise rectangular tent (Perron 1990: 220);and that it formed the canopy of a royal dais or baldachino, supported bythe statue-like columns of which four were known to Pliny [10],with the
royalthrone placed underneath (Goukowsky 1978, I: 191–2; Perron 1990:
220 n. 16). A cosmic meaning has been assumed: ‘symbole c ´eleste d’une
domination universelle’ (Goukowsky 1978, I: 191–2).

4 Friends in high places: the creation of the court of
the Roman emperor
jeremy paterson
Introduction
The end of the ﬁrst century BC witnessed a profound transformation of the
Roman respublica from a political system in which power was diffused across
arange of democratic and aristocratic institutions to one where power was
seen to reside primarily in the hands of one man. This period of change andexperiment provides a unique opportunity to see a court and court society
coming into being, a process of interest not just to Roman historians but
also to all those who seek to understand the nature of court societies.
The ﬁrst impetus for the creation of any court society is the recogni-
tion of the ruler as the monopoly, or near-monopoly, possessor of power.Iwould argue that allsocieties in which power becomes the monopoly of
the ruler create the conditions for the emergence of some form of ‘courtsociety’; in other words this is a truly sociological, or even anthropolog-ical, phenomenon.
1Norbert Elias’ famous study of court society (Elias
1983 )starts by claiming a speciﬁcally sociological approach as opposed
toahistorical one; however, he ends up by representing his chosen court
society, that of Louis XIV in France, as a stage in the formation of the mod-ern state, which is in essence a historical argument.
2Inclaiming that the
court is a sociological phenomenon I am not suggesting that all court soci-
eties develop in the same way or in accordance with some law of nature.3
The evolution of a court and court society around the Roman emperorhas many features historically speciﬁc to that period of Rome’s develop-ment. For example, although Romans were inﬂuenced by the model set
1See Herman 1997 :206: ‘The emergence of court societies has more to do with universal social
conﬁgurations than with concrete historical circumstance: it is a sociological rather than a
historical phenomenon.’
2See the critique of Elias by Duindam (1995: passim ). For a major new attempt to bring rigour to
theorising about courts see now the contributions in Butz, Hirschbiegel and Willoweit 2004 ,
which exploit a range of approaches well beyond the Weberian analytical techniques of Elias,
including New Institutional Economics.
3See Savalli-Lestrade ( 1998 :289), who notes that the existence of a group of faithful followers is a
characteristic of all absolute monarchs; but the particular form it takes in any one place and time
can be in any of a great many different conﬁgurations.

122 jeremy paterson
bythe Hellenistic rulers they encountered in the eastern Mediterranean,
there were signiﬁcant differences, most notably the fact that, whereas in the
Hellenistic monarchies all political institutions were centred on the ﬁgure
of the ruler, in Rome the senate and magistracies, which already existedbefore the emergence of the Principate, to some extent continued to pro-vide potential sources of power and inﬂuence not entirely dependent on the
emperor.
Inreality no monarch can rule alone. The Roman emperor was no excep-
tion to this. He needed trusted and reliable helpers, who expected rewardsfor their loyalty. For their part his subjects queued up to represent them-selves as worthy recipients of rewards, favours and careers. It follows fromthis that the court and court society are a negotiation between the ruler andthe subject; the court is not simply imposed by the ruler.
Courtly behaviour is the ‘language’ of this dialogue; it becomes the
medium through which loyalty, favour, deference and respect are expressedboth by the ruler and by his courtiers. Above all, as will be highlighted later,the conventions and customs of the court provide an element of reassuranceand conﬁdence in a relationship which is inherently unstable and fraughtwith uncertainty. The early Roman Principate was an age of experiment,
in which ruler and ruled combined in a lively contemporary debate aboutthe consequences of the identiﬁcation of power with the emperor and hishousehold. Contemporaries had a variety of models to work with. Therewas the tradition about the early kings of Rome. There were also the archaicGreek mod el of the tyrant (Greek: turannos ), the traditions about the kings
of Persia (see Brosius in this volume), and the near contemporary encoun-ters with the Hellenistic monarchs of the Greek east. The interesting point
about all these models is their ambivalent nature, depending on whether thetyrannical and autocratic aspects or beneﬁ cial rule, respect and civility are
emphasised. The relationship between the ruler and his subjects could becharacterised as either that between master and slave or that between father
and child.
4
Wher eas problems about the legal powers of the Roman emperor have
absorbed the attention of generations of modern scholars (though less soin the past thirty years), contemporaries of the emperors tended to takethe emperor’s all-encompassing power as a given and were concerned farmore on the one hand with the behaviour of the emperors, their lifestyle,attitudes, social values and relationships with others, and on the other withthe etiquette and appropriate behaviour of the subject who is seeking to gain
4See the useful discussion of Roller ( 2001 :213–87).

The creation of the court of the Roman emperor 123
ofﬁce, reward or beneﬁt from the emperor – in other words with the issues
which are central to all court studies.
Powerisareality, but it is also invisible, like the wind.5Butthe emperor
needed to be seen tohave power. For their part courtiers also had to be able
todemonstrate to the rest of the world that they had access to, and beneﬁted
from, the ruler’s power. Hence the ceremonies, the etiquette of courts, andthe honours granted to and by grateful subjects ought to be interpreted asways of making power manifest. It follows that the emperor’s court shouldbe at the heart of our understanding of the Roman Principate. So it is allthe more surprising that Brian Jones in an excellent brief section on thenature of the imperial court in his biography of Domitian could commentas recently as 1992 that ‘detailed discussions of the imperial court are rare’(Jones1992 :202 n. 3). In fact the subject has not been ignored so much
as simply taken for granted as being largely unproblematic.
6One reason
for this comparative neglect is that the ancient sources themselves at ﬁrstsight do not contain any full-blown study of the emperor’s court. It has beensuggested that the reason for this is the secrecy which the historian CassiusDio (53.19) argued surrounded the decision-making processes in the Prin-cipate.
7Butthis is to miss the point. It is true that the processes by which the
emperor came to major decisions were often obscure because the debate, ifthere was any, was usually behind closed doors, or informally among closefriends of the emperor. But under most emperors the court was relativelyopen and accessible, certainly to the majority of those who chose to writethe history of Rome in this period.
8Itis much more likely that the sources
tended to take for granted the everyday ceremonial and courtesies which
from the start were part of their life at court, and concentrated instead onthe examples of exceptionally arrogant or demeaning behaviour. We do nothave a Saint-Simon for the Roman Principate. However, once we get our eyein and know what we are looking for, it is clear that commentary on the court
5The comparison of power with the wind and ceremonial with snow was made by David
Cannadine (Cannadine and Price 1992 :1).
6See Wallace-Hadrill ( 1996 :285) and also the lively, useful, but under-theorised study by Turcan
(1987 ). It is interesting to note that the situation until recently has been the same with regard to
the historical study of the Hellenistic kings. For Alexander see now Spawforth in this volume; forthe Hellenistic kings, Herman ( 1997 )presents a ground-breaking study explicitly inﬂuenced by
Elias and brieﬂy discussed in this volume’s Introduction.
7Wallac e-Hadrill 1996 :284; it is important to note that Dio’s primary concern in the passage is a
historical one about the reliability of the information in his sources.
8Pliny the Y ounger in his panegyric of the emperor Trajan states, ‘Y our meals are always taken inpublic and your table open to all’ ( Pan.49.5), and Pliny himself ( Ep.6.31) recalls with pride his
attendance at informal dinners with the emperor after a hard day’s work in his council(consilium ).

124 jeremy paterson
and court life is everywhere in our sources – from Tacitus’ concerns about
the servility shown by many senators and the rise in the public inﬂuence offreedmen from the imperial household, for example, to Suetonius’ anecdotesabout the manners, conversation and public appearances of emperors.
9The
discourses of Epictetus on ‘How should we behave towards tyrants?’(1.19)and ‘On Freedom’ (4.1) certainly draw on a long, conventional literary andphilosophical tradition about the nature of ﬂattery and the distinction ofthe ﬂatterer from the true friend – one might compare Plutarch on ‘Hav-ing many friends’ ( Mor.93B–95B) and on ‘How to distinguish a ﬂatterer’
(Mor.48F–74E).
10Inthe case of Epictetus, they take on an added resonance
and contemporary signiﬁcance once it is realised that the author, as a slavein the service of an inﬂuential freedman at the court of Nero, was writing asacourt ‘insider’.
11All these sources come alive once it is realised that they
reﬂect a vital contemporary debate about how an emperor should behave
and how his subjects should behave towards him – that is, in large measure,ad e b a t ea b o u t the court and court life.
ptMuch of our source material about courtly society in the early
Principate comes from people who, to a greater or lesser extent, were part of
that society. Tacitus, Pliny the Y ounger and the Greek historian Cassius Dioweresenators with successful public careers which depended on the favour
of emperors. What men like this had to do to rest easy at night was aboveall to concoct an interpretation of events which saw monarchy in Rome asan unavoidable consequence of the failure of Romans to govern themselvescollectively under the late Republic.
12On this view there was no alternative
tomonarchy and this in turn justiﬁed the members of the elite in their par-
ticipation in public life and in their willingness to adopt courtly behaviourtosolicit the emperor for ofﬁce and favours. They had to believe that it was
possible to have a successful public career and take part in the life of thecourt, while avoiding degrading servility or compromising their integrity.
Tacitus saw a model in Marcus Aemilius Lepidus, the consul of AD 6.
Lepidus was described by a contemporary as a man who stood ‘very close to
the name and fortune of the Caesars’ (Velleius Paterculus 2.114.5); indeed hehad a very distinguished Republican ancestry and was related to the emperorAugustus himself. According to an anecdote, the emperor on his deathbed
9Another reason for the neglect of the court as a s ubject may be the modern scepticism about
anecdotes as historical evidence (see Saller 1980 ). However, Laurence and Paterson ( 1999 :195)
discuss the ways in which sayings by and about emperors may be used to illustrate
contemporary perceptions of the emperor’s role, despite the impossibility in many cases of
conﬁrming their historicity.
10See Fraisse 1974 :409–11.11Millar1965 and Starr 1949 .12Syme 1986 :439–54.

The creation of the court of the Roman emperor 125
had pronounced Lepidus ‘fully capable of becoming emperor ( capax
imperii )but considering the position beneath him’ . All the more surprising
then that a cynical historian like Tacitus should pick such a man out as the
archetype of the good courtier. In Tacitus’ opinion Lepidus was a serious andwise man, who enjoyed consistent inﬂuence and favour with the emperor
Tiberius; but he was no toady – rather he used his inﬂuence to do good
bysuccessfully countering the savage proposals against individuals made
byﬂatterers of the emperor.
13This assessment leads Tacitus on to wonder
whether success in winning the emperor’s favour was a matter of luck orwhether a person could make their own luck by behaving with integrity and‘bysteering a path, clear of intrigue and danger, between the perils of open
deﬁance and the degradation of servility’ . Tacitus here reﬂects the centralpreoccupation of the Roman elite under the Principate, which was to expectthe emperor to create the conditions in which they could acknowledgethe fact of his pre-eminence without feeling uncomfortable, and couldmaintain their belief that their success and favour with the emperor wasthe natural recognition of their true abilities, rather than the mere accidentor whim of the ruler.
14So the insiders in their histories praise emperors
who make life easy for men of their kind by not lording it over them intooobvious a manner. The elite also expected reasonable and direct access
tothe emperor. Tacitus and others reserve their ﬁercest criticisms for the
members of the imperial household, who were their social inferiors butbecause of their position within the court could come between the senatorand the emperor, such as Claudius’ freedmen.
Suetonius and Epictetus were ‘insiders’ in a somewhat different sense.
They had careers within the imperial households. They were part of theadministrative system of the emperor that kept the Principate going. Sueto-nius, in particular, tended to judge the effectiveness of the emperor and thecourt by the practical criteria of the extent to which they got the job done
successfully. He was largely unencumbered by sentimental longing for thelost Republic or touchiness about the issue of ‘freedom’.
The ultimate ‘insider’ , of course, was the emperor himself. For all the
remarkable insights within the Meditations of Marcus Aurelius, there is
comparatively little on the court. He recognised that life in court was poten-
tially corrupting for the monarch. Y et he insisted that ‘in a court too it isstill possible to live well’ in accordance with his chosen Stoic philosophy.
13Tac.Ann.4.20.
14Pliny Ep.8.6.13 on the emperor honouring those who ‘deserved reward’ ( praemia merentes ), a
passage discussed by Brunt ( 1988 :39).

126 jeremy paterson
Of his teacher, Sextus of Chaeronea, Plutarch’s nephew, he said, ‘Simply to
be in his company was more delightful than any ﬂattery, while at the sametime those who enjoyed his company looked upon him with greatest rev-erence’ ( Med .1.9.1) – it was an ideal model for a philosopher king. Marcus
(Med .1.17.3) thanked the gods that he had learned that it was possible to
live in court yet do without the trappings and pomp of bodyguards, spec-tacular costumes, torch-lit parades, statues and the like (an interesting listof what could in the second century AD constitute court ceremonial). Inshort, Marcus sought to avoid becoming ‘Caesariﬁed’ ( Med .6.30).
Apoet like Statius has been categorised as a sycophantic insider because
of the half dozen poems to Domitian in his collection, Silvae .Itistruethat
Statius was always keen for signs of imperial favour – quite naturally giventhat this was the route to success and reward; but it would be wrong to dismisshis poem on being invited by Domitian to a large formal dinner in the palace(Silvae 4.2) as mere ﬂattery. A man who does not walk the corridors of power
regularly is unlikely to be blas ´eorcynical about attendance at a banquet in
the presence of the emperor. On the contrary he would expect, indeed wouldexult in, a level of dignity, pomp and splendour – otherwise what is the point
of being part of the occasion? And if the emperor condescended to minglewith his diners, so much the greater would be the impact of the occasion.
Formal court poetry it may be; but it reﬂects a reality.
Itis important to realise that our sources are not going to give a single, clear
and consistent account of the experience of being part of courtly society. Itdepends very much on who the source is.
When it comes to the analysis of the court of the Roman emperor, for
once in the study of Roman history Theodor Mommsen did not succeed insetting the agenda for future generations. He concentrated on the internalorganisation of the emperor’s household and the access of his friends tothe court. But he offered little by way of dynamic analysis; indeed his mainaccount (‘Hof und Haushalt’) is startlingly brief and is fundamentally ﬂawedbyafalse distinction between the ‘state’ aspect of the emperor’s court and the
court as the emperor’s private household.
15The account to which all stud-
ies since have been indebted is that of Ludwig Friedl ¨ander in his monu-
mental study of Roman life and manners, originally published in 1862(Friedl ¨ander 1921 ). He also details the bureaucrats, freedmen and slaves
who made up the imperial household along with the emperor’s friends andcompanions; but his most important innovative contribution was his study
of court ceremonial, which has only recently been superseded. Aspects of the
15Mommsen 1887 Bd II 2, 3, 833–9. For a wide-ranging study of the earlier research on the
Roman imperial court see Winterling ( 1999 :12–38).

The creation of the court of the Roman emperor 127
Roman court continued to receive close study, for example the members of
the emperor’s household (e.g. Weaver 1972 ), but rarely, if ever, were they set
in a more general discussion of court society. The grand synthesis of Millar
(1977 )contains vast amounts of material of relevance, but signiﬁcantly the
term ‘c ourt’ does not appear in the index. Wallace-Hadrill ( 1996 )cites Elias
in his account of the imperial court for the Cambridge Ancient History ,but
this is really almost the ﬁrst sign that ancient historians were beginning totake account of the emergence of aulic studies among historians of the earlymodern period, which had followed the publication of Elias. As a resultancient historians are bound to be indebted for a long time to come to thescholarship of Winterling ( 1999 ;and for the later Roman period Winterling
1998 ). He is at the heart of the groups of European scholars who are in the
process of bringing intellectual rigour to the theories about the nature ofcourt society, though their focus has been largely, but not exclusively, on the
courts of early modern Europe.
Inall this it may be salutary, for a moment, to listen to the occasional note
of scepticism about whether the reason historians have not explicitly dealtwith the court and court society of the Roman emperors is that there was
in reality no court of the Roman emperors. Gag ´e(1971 :191ff) questions
whether we should talk about an imperial court – at least in the perioddown to the Severi. It is of interest that his principal reason is that Romansenators and their families did not live permanently in the imperial palace.Inother words he takes Versailles under Louis XIV as the deﬁning template
for a court. However, in other court societies, such as the Habsburg courtin Vienna, the nobility were not constantly present at court. In any case, inRome it was a regular expectation that senators, unless excused, attended the
morning greeting ( salutatio )ofthe emperor at the palace on the Palatine.
Nevertheless, Gag ´e’s doubts force us to attempt to deﬁne what is meant by
the court of the Roman emperor.
Language can often be a key indicator of change. A clear sign that there
was a new phenomenon comes from the appearance in Latin of the term‘aula’for the court. It scarcely appears during the Roman Republic.
16But it
becomes the normal noun for the imperial court from at least the mid-ﬁrstcentury onwards. The term is Greek in origin and its primary meaning is
the hall or courtyard of a house. It was borrowed from the Hellenistic world,where it was regularly used both for the physical palace of a monarch andfor the collective members of that court (see this volume’s Introduction).Itis an indication that the Romans had models of Hellenistic kingship in
16Only in Cic. Fam .15.4.6 (where signiﬁcantly it is used of the court of the Hellenised king of
Cappadocia).

128 jeremy paterson
their minds when the Roman Principate came into being, because the Greek
termbecame current rather than its Latin equivalent, ‘ atrium ’. T h e G r e e k
terms ‘ to basileion ’o r‘ ta basileia ’for palace in the same way came to mean
not just the palace as a building but the more abstract ‘court’ . In this case,however, the Hellenistic term was not adopted, perhaps because of its tooovert reference to kingship (Greek: basileus ); instead Palatium, the name of
the hill on which Augustus’ residence was located, came to mean ‘palace’ andalso by extension those within the palace.
17Hence Dio (53.16) explained:
‘The royal r esidence ( ta basileia )iscalled Palatium, not because it was ever
decided that it should be called this, but because Caesar lived on the Palatineand had his headquarters there . . . Even if the emperor resides somewhereelse, the place where he stops is called the Palatium.’ The palace is whereverthe emperor is.
Another indication of a changed world is the way words in the new con-
text can take on a signiﬁcance they did not have previously. The emperor
Tiberius publicly rebuked someone for addressing him as dominus (‘master’)
(Suet. Tib.27 and Tac. Ann.2.87). In this, as in so much else, he was following
the precedent of Augustus (Suet. Aug.53.1). In private life, the term ‘ dominus ’
was a formal, respectful address between social equals (roughly equivalentto‘mylord’). ‘ Vale domine frater ’(‘Farewell, master and brother’) is the way
acorrespondent signs off a letter found at Vindolanda near Hadrian’s Wall
in the North of England.
18Pliny the Y ounger regularly addressed Trajan as
‘domine ’inhis ofﬁcial correspondence with the emperor.19But in public in
the context of the court this form of address became problematic, because‘dominus ’isalso the term regularly used for the owner of slaves – a connota-
tion which emperors would be anxious to avoid.
20Forms o faddress become
critical and difﬁcult issues. In the context of court life ‘the path of languageis narrow and slippery’ ( angusta et lubrica oratio ), as Tacitus ( Ann.2.87) so
memorably put it.
The genesis of the emperor’s court at Rome
Toleave on one side for the moment the question of exactly who constituted
the emperor’s court, it is illuminating to consider how and why the courtcame into being in Rome. In so far as they have considered the matter at
17See K. Ziegler, REXVII.3, col.6ff.18Tab.Vindol .II 295; cf. Tab.Vindol .II 234 and 248.
19Pliny Ep.10passim .
20This is well discussed by Roller ( 2001 :254–8). A full discussion of dominus by M. B ang in
Friedl ¨ander 1921 :82–8.

The creation of the court of the Roman emperor 129
all, historians have tended to take one of two approaches. The ﬁrst, which
may be dubbed ‘from Haus to Hof’ , is to emphasise the continuity fromthe Republic to the Principate by identifying many of the features of theimperial court as already existing in some form in the houses and house-holds of the great members of the Roman Republican elite. So the palaceand court of the Roman emperor are different in scale but not in kind fromthese precedents.
21The alternative is to see the courts of the Hellenistic kings
as providing conscious models for the construction of a system which wasessentially novel for Rome and in some r espects offensive to the traditions
of the Roman Republic. The solution suggested here is that both elementsplay a role.
22However,our understanding of the nature of the Roman impe-
rial court is fundamentally aided by concentrating on those features of the
Principate which contemporaries considered novel:
1Powerand patronage were perceived as residing ultimately in the person
of one man.
2Asac onsequence those immediately around the emperor – his family,
household and close companions – were perceived as deriving power andinﬂuence in their own right from the emperor.
3One result of this was that in some cases individuals from outside the socialand political hierarchy were recognised as having abnormal power andinﬂuence, e.g. women, imperial freedmen, some of the slaves attendanton the emperor, and eunuchs in the later empire; while court societiesfrequently reinforce already existing social hierarchies, they also can havethis subversive effect.
4Access to and recognition by the emperor became the key to a successful
and secure public life for the elite.
5The obverse was, if anything, even more important: exclusion from thecourt frequently spelt political, social and often actual death (either by
suicide out of despair at the end of one’s career or as a result of beingattacked in the courts by opponents who were encouraged to do so as aresult of the emperor’s indication of the withdrawal of his favour).
23
21Wallac e-Hadrill 1996 :290.
22Anticipated in a slightly different context by Crook ( 1955 :21), in his discussion of the friends
of emperors: ‘The institution of amici principis derives from two sources, the purely Roman
political tradition of the Republic and the hierarchy of the Hellenistic courts. It is important
not to overemphasize one at the expense of the other.’
23So, for example, Sextus Vistilius, who had been accepted into the entourage ( cohors )o ft h e
emperor Tiberius, fell out of favour after the emperor took offence over something he wasaccused of saying, and committed suicide after being excluded from the emperor’s company(convictu principis prohibitus )( T a c . Ann.6.9).

130 jeremy paterson
6Aset of conventions or norms emerged which governed the social inter-
action between the emperor and those around him.24
7Flattery as a form of behaviour became important and prevalent, even
though emperors may have protested against it and subjects found itdemeaning.
‘The Roman senate was an assembly of kings.’ Such it is claimed (though
with variations in our sources) Cineas reported back to Pyrrhus, Hellenistic
king of Epirus, after an abortive attempt to negotiate peace with Rome in280 BC.
25The remark is most likely apocryphal. At the heart of the Roman
Republic there was a body of families who, particularly in the last two cen-
turies BC, accumulated extraordinary levels of wealth. Once the Roman eliteencountered the Hellenistic world on a regular basis, they began to take onthe trappings of the Hellenistic kings. As provincial governors, their pro-gresses around their provinces could seem very like the arrival of a Hellenistic
dynast, and the provincial communities responded by using the language
and etiquette which they had previously used about their Hellenistic rulers,praising them as founders, god-like, and establishing festivals in their hon-our (see Rawson 1975 ). Back in Italy and Rome the elite sought to make
their houses the equivalent of the Hellenistic palaces they had encounteredin the east. They raided the great libraries of the Hellenistic rulers.
26Their
entourages included poets and scholars. They ﬁlled their houses and theirgardens with the statuary of the Greek world.
27The need for their houses
tohave large public reception rooms was discussed in a famous passage
of Vitruvius (6.5), where he emphasises their requirements for vestibules,atria and peristyles on a grand scale, which beﬁtted their rank ( ad decorem
maiestatis ).
28The reason for this was that parts of the nobiles ’houses were
essentially public places, the setting for councils and for judicial activities
associated with the holding of the magistracies.
According to Seneca ( Ben.6.34), it was Gaius Gracchus and, a little later,
Livius Drusus the Y ounger who adopted the practice of classifying theircallers at the morning salutatio (segregare t urbam suam ), by receiving some
in secretum ,others in company and the rest enmasse .This was an overt
24On the importance of norms and rules of conduct in the creation of a court society see Asch
and Birk 1991 .
25Plutarch Pyrrhus 19, or ‘a city of kings’ according to Justin Epit.18.2.
26See Aemilius Paullus (Plut. Aem .28.6), Sulla’s appropriation of Aristotle’s library (Strabo
1.1.54), and Lucullus’ great collection (Plut. Luc.42) (Rawson 1985 :c h . 3).
27Flamininus in 194 BC (Livy 34.52), Marcellus after the capture of Syracuse in 211 BC, and mostnotoriously Mummius after the sack of Corinth in 146 BC.
28See the nuanced discussion of Wallace-Hadrill ( 1994 :10ff).

The creation of the court of the Roman emperor 131
adaptation of the practice at Hellenistic courts.29The practice of receiving
friends at the morning salutation in groups was continued by the emperors,
although some tried to modify or even abolish it (see Talbert 1984 :68–70).
Winterling ( 1999 :121) has argued that the practice may have been no more
than a practical arrangement of dividing those who came between differentrooms which the master of the house then toured. But whether or not the
head of the household, or later the emperor, intended a ranking or privilege,they were unable to prevent it from being taken in that way.
30
Thus many of the features of the courtly society under the Principate are
found in some form in the Republic. However, despite the competitivenessof the Roman Republican elite, social norms, often expressed in terms ofmorality, sought to keep all the members of the elite on a par. When powerfell into the hands of a single individual with Julius Caesar and the comingof the Principate, a continuous debate was initiated among the elite as to justwhat was ﬁtting and reasonable in terms of deference, display and behaviourtowards the monarch.
The ruler’s court is not the creation of the ruler alone or even his initiative.
Itis as much the means by which the subjects come to terms with the fact
that power is now the monopoly of the ruler, and the way they create a modus
vivendiwith that ruler. Some strands of this can be seen in the aftermath
of the civil war in the 40s BC and the dictatorship of Julius Caesar. Anextraordinary range and scale of honours were granted Caesar in 46 and
45 BC.
31They included the right to special dress, the use of a gilded chair, a
bodyguard of senators and equites ,the renaming of the month of his birth,
quadrennial games in his honour, the creation of a cult in his honour andmuchelse (many of these can be traced to Hellenistic precedents). But they
wereall initiated by the senate, usually in Caesar’s absence. They represent
less Caesar’s manoeuvring to create a new role for himself, and much morethe attempt by the senators and others to come to terms with the new realitiesof power. Cases like this raise the more general issue of who created courtceremonial – the emperor, the courtiers, or both? In the Roman context the
initiative should more often be ascribed to the courtiers rather than to theruler, whose main role was to attempt to restrain novel, excessive ﬂattery.
The extraordinary honours granted to Julius Caesar raise another more
general issue. There has been an endless and largely inconclusive debate from
29Forthe ranking at the court of Ptolemy V Epiphanes of the king’s ‘ﬁrst friends’ , etc. see Mooren
(1975 and 1977 ).
30See e.g. ILS1320, an inscription, now lost, of C. Caesius Niger, an eques who is described as ‘ ex
prima admissione ’inthe time of Augustus or Tiberius.
31See Gelzer 1968 :307, 315–16.

132 jeremy paterson
antiquity to the present about what Julius Caesar’s intentions really were and
what the motives behind the grants of honours were. However, the reasonwhy such investigations are bound to be inconclusive lies in the difﬁcultyof interpreting the signiﬁcance of any ceremonial practice. Whatever themotives of the proposers of these honours and whatever the intentions ofCaesar in accepting them, it was impossible for anyone, ruler or courtier, torestrict the meaning of any practice. Consider the notorious events of the
Lupercalia (15 February 44 BC). At this religious ceremony Julius Caesar,
recently designated dictator perpetuus ,appeared in public seated on a gilded
chair on the Rostra and for the ﬁrst time wearing the ceremonial dress of the
early Roman kings (purple toga and gold wreath).
32Itis Valerius Maximus
(1.6.13) who acutely points out that these honours had been granted tohim by the senate and that Caesar wore them on this occasion so as nottooffend the senate! The Luperci, priests dressed in goatskin loin-cloths,
wereled in their run round the Palatine by the consul, Mark Antony, who
on his return to the Forum, mounted the Rostra and attempted to place onCaesar’s head a diadem (a characteristic feature of the dress of Hellenistickings). Caesar refused, had the diadem taken to the Capitol to be dedicatedtoJupiter Optimus Maximus, and ordered that it should be placed on public
recordthat on that day Caesar had refused the kingship offered to him. The
refusal of honours offered is a key aspect of the dialogue between ruler and
subject. So Augustus in his account of his reign, the ResGestae (5, 6 and
10), lists the honours and posts he had been offered, but had refused. Oursources for the Lupercalia incident have every possible interpretation of whathappened and its meaning.
33Depending on the source, the people either
applauded or groaned when Antony tried to put the diadem on Caesar’shead, and again when Caesar refused it. The event has been interpretedeither as staged by Caesar to test the waters, or as an initiative by Antonyas an act of ﬂattery, or as an attempt to get Caesar publicly to renounce hisworrying ambitions (Dio 46.17 and 19), or even with more sinister intent –
‘You [Antony] signed Caesar’s death warrant at the Lupercalia’ ( tu…i l l u m
occidisti Lupercalibus ,Cic.Phil.13.41). This is not just a problem of our source
material; it is rather that the event could have at the same time any or all ofthese meanings depending on who you were. What matters here is not ‘whatwas happening’ , but ‘what was thought to be happening’ by the participantsand other observers.
32Cic.Div.1.119.
33Cic.Phil.2.85–7 – and elsewhere in the Philippics; Dio 44.11; Suet. DJ79.2; Livy per.116;
Nicol.Dam.F130, 71–5; Vell.2.56.4; Plut. Caes .61;Ant.12; Appian BC2.456ff; Florus 2.13.91.

The creation of the court of the Roman emperor 133
Inap erceptive comment, Cicero considered the possibility that Caesar
wanted to restore a Republican constitutional government, but he saw that
he would not be able to because he had bound himself to so many.34The
dictator was the prisoner of his entourage – those who had joined him inthe civil war in the hope of personal advancement and who had no interestin the return of a republic. Elias ( 1983 )sought to represent the court as a
mechanism created by the ruler to control the noble courtiers. But in hiscritique of Elias, Duindam ( 1995 )demonstrated that the development of a
court and court society was not necessarily an effective means of controlling
noble-born courtiers, that the court can be a source of power and inﬂuencefor its noble courtiers, and that the ruler can be as much in the effectivecontrol of the court as vice versa. Further, Caesar’s successors would have
noted that among his assassins were those whose hopes of advancement hadbeen disappointed.
35
Inthe last years of Julius Caesar’s life many features later associated with
emperors appear. Here was a court in the making. He was accompanied inhis tours by a huge entourage – 2,000 (probably mainly soldiers) by Cicero’sestimate in a famous account of Caesar’s visit to him in Campania ( ad
Att.13.52). Caesar himself became the focus of a deluge of petitions and
requests, a circumstance which itself could cause problems. On a visit to
Caesar’s house to present a petition, Cicero was kept waiting. According toMatius, a Caesarian supporter, Caesar remarked, ‘Can I doubt that I am
utterly loathed when Marcus Cicero has to sit and wait?’( ad Att .14.1.2 and
2.3). The creation of regular procedures is essential for a ruler to avoid thissort of offence. The ceremonial and etiquette which developed in the Prin-cipate may have been irksome to some, but to most was a conﬁrmation oftheir status and privilege.
36Itwas essential that the courtier knew what his
position was, how to approach the ruler, what to do and say to win favour.Finally, there was the issue of Caesar’s clementia ,his willingness to forgive
opponents and allow them to resume their public lives. Clementia was an
attribute of monarchs, so that the soliciting of forgiveness and the grant-ing of it were acts which acknowledged and conﬁrmed the primacy of theruler. Hence Cato’s ostentatious suicide as an act of refusal to recognise any
dependency on Caesar’s good will. But the majority, who were not preparedtotake such a drastic line, had to modify their behaviour in order to achieve
34Cic.ad Fam .9.17.2. Cf. Cicero’s similar assessment of Octavian in ad Att .14.12.2.
35Among them the great-grandfather of the emperor Galba, who despite serving as a legate with
Caesar in Gaul ascribed his defeat in the consular election to Caesar’s failure to support him
and so out of pique joined the conspiracy against him (Suet. Galba 3).
36Cf. Elias 1983 :c h . 5.

134 jeremy paterson
their goals in the face of the new situation in which Caesar had the power.
Thus at a meeting the whole senate appealed to Caesar to forgive M. Mar-cellus, the consul of 51 BC ( ad Caesarem supplex accederet ), while Marcellus’
cousin, Gaius, threw himself at Caesar’s feet ( sead Caesarem pedes abiecisset )
(Cic. ad Fam .4.4.3). Caesar gave in. But such behaviour was recognition of
where power now lay. Again, Cicero in his speeches before Caesar himselfon behalf of Q. Ligarius and later King Deiotarus chose to plead directlyfor mercy rather than mount a full defence: ‘I take refuge in your clemency.Iseek mercy for the offence. I beg forgiveness’ ( Lig.30); ‘Forgive, forgive,
Caesar’ ( Deio tar.12) This was the sort of plea ( deprecatio )which was to
become the norm, but which Quintilian ( Inst.5.13.5) recognised had no
real role previously in a Republican advocacy. Courtly behaviour is more a
creation of the ruled than an imposition by the ruler.
Modelling monarchy and the Dictator’s Dilemma
Inthe few years of the 40s BC in which Julius Caesar dominated Roman pub-
lic life, some of the characteristics of a court society were already emerging asanegotiated response to his pre-eminence. More signiﬁcantly indeed, in the
years that followed his assassination the attempt to restore the Republican
system was essentially a sideshow to the main struggles, which were aboutwho would be the next Caesar. Julius Caesar had opened up the possibilityof getting things done in a new way. He should be seen not as representingthe end of the Republican system, but as the beginning of the Principate.
Julius Caesar did not create a court around him; nor, I suggest, did the
early emperors. But a court grew up, and it is important to look at thefactors which encourage the creation of a court around autocrats. Weber’smodel of court society as ‘patrimonial’ in pre-modern societies – that is,centred upon the palace and personal household of the ruler, rather than
on a bureaucratic machinery of government – has obvious relevance to thecourt of the Roman emperors.
37Elias’ work ( 1983 )isessentially a Weberian
sociological study which concentrates on the connections between powerand social structures and values. It offers a picture of the French ancien r´ egime
in which the elites are offered distinc tion and privilege as compensation
for the loss of control of power. He posits an unavoidable interdependencebetween king and court. But these sociological approaches do not tackle fullythe fundamental question about why court societies arise around autocrats
37Forthe lack of bureaucratic machinery see Garnsey and Saller 1987 :c h . 2.

The creation of the court of the Roman emperor 135
in the ﬁrst place. One answer centres upon the recognition that autocrats are
often more effective redistributors of wealth than other political systems.38
According to Suetonius ( Aug.101), Augustus declared that no more than
150 million sesterces would come to his heirs, even though he had received1,400 million during the last twenty years of his reign simply from thewills of his friends, ‘because he had spent nearly all of it along with his
two paternal patrimonies and other inheritances for the beneﬁt of the state(in rem publicam )’ . On the basis of these ﬁgures Augustus retained about
10 per cent of this part of his income. This gets into context the issue of theemperor’s expenditure on buildings, show, personal luxury, etc., which couldfrom time to time excite adverse comment. It is an example of what may betermed the ‘Imelda Marcos shoe syndrome’ – her extraordinary expenditure
on shoes excited much comment and was powerful propaganda against her,but was only ever a small part of the wealth that came to the Marcoses. If theemperor is seen essentially as a redistributor rather than an accumulator,then rent-seeking becomes centred on the emperor. Access to him becomesessential for maintaining and increasing wealth, privilege and position forthe elite. The courtiers, however, also become brokers, interceding to winfavours for their clients (cities, communities and individuals), an essentialpart of the redistribution system.
39
Inthis way the court and court society becomes the context for this
process of brokerage, and the means by which this system is organisedand maintained. But there is one more central factor which structures thedynamics of the relationship between ruler and court – this is ‘the Dicta-
tor’s Dilemma’ .
40The autocrat can never be entirely sure that his subjects’
words and actions are truly motivated by loyalty and respect. As a result the
position of autocrat carries with it an inherent sense of insecurity, whichcannot be relieved either by the use of fear and repression or by the distri-bution of rewards. No Roman emperor could be unaware of this, given theassassination of Julius Caesar, which ushered in the Principate. The dilemmawas also felt by the subject: ‘Praise a tyrant and he takes no pleasure in it.Forhethinks that those who praise him feel otherwise’ (Dio Chrysostom
Orat .6.59).
38Formodern models of the economics of autocracy see Wintrobe 1998 .Heincludes discussion
(not always accurate) of Roman emperors. He argues that ‘dictatorships tend to redistribute
income more than democracies do’ (149–60). See also Paterson 2004 .
39On this see Kettering ( 1986 )onseventeenth-century France, cited as a useful comparison for
the Roman empire by Wallace-Hadrill ( 1989 :81ff).
40See Wintrobe ( 1998 :20–39) for a detailed analysis of the dilemma.

136 jeremy paterson
The dilemma was recognised in antiquity and analysis of it tends to centre
on discussions of how to distinguish the true friend from the ﬂatterer.41In
termsofrational-choice theory the ruler and his subjects have a mutual
‘signalling’ problem. Essentially the problem is how either party can make
‘credible’ commitments in situations when there is gross inequality between
the two, and promises and obligations on either side are unenforceable. The
ruler can never be sure that the subject is speaking frankly and the subject
has no means of holding the ruler to his word. Furthermore the dilemmais inescapable. It is a fundamental feature of autocracy.
42‘Grant to every
one who wishes to offer you advice on any matter whatever, the right tospeak freely and without fear. For if you are pleased with what is said, itwill beneﬁt you greatly, and if you are not persuaded, you have not suffered
any harm.’ Such was the advice to Augustus put in the mouth of Maecenasbythe historian Dio (52.33.6). Indeed, both rulers and subjects frequently
stated the need for free and frank discussion; but the situation they foundthemselves in made this end unattainable. When the ruler holds the powertoadvance you or hold you back, include you or exclude you, even ultimately
decide whether you lived or died, then as a subject your actions and wordsinevitably became something of a ‘performance’ put on for the beneﬁt ofthe ruler to convince him of your worthiness and loyalty.
43This involved
an element of second-guessing what the response to your words or actionsmight be from the ruler and modifying your own behaviour so as to generateaparticular response. This is why consistency and predictability are so prized
in rulers. Hence the criticism of Gaius Caligula – what was frightening abouthim was that no one knew what he would do next: this puts the courtier inan impossible position where he cannot calculate the emperor’s reaction tohis words or actions. Cassius Dio, a successful courtier, has some thoughtfulcomments on the difﬁculty facing courtiers when dealing with Tiberius,
who had a reputation for never saying what he really meant.
44
The competitiveness of court life also ensured that the courtiers were
bound to employ ﬂattery. They were playing for big stakes: ‘It is not Caesarhimself that anyone fears, but death, banishment, prison, loss of property,deprival of civic rights. Nor does anyone love Caesar himself, unless hehappens to be a man of great worth; but we love riches, a tribunate, a
41See Fraisse 1974 :409–11; Savalli-Lestrade 1998 :345ff; Plutarch Mor.93B–95B (‘On having
many friends’) and Mor.48F–74E (‘How to recognise the ﬂatterer’).
42Kuran ( 1995 )analyses the disjunction between ‘private beliefs’ and ‘publicly expressed
opinions’ of subjects under an autocracy. See also Scott ( 1990 )and Goffman ( 1967 :47–95).
43The ‘theatricality’ of court life has been splendidly analysed by Bartsch ( 1994 ).
44Dio 57.1.

The creation of the court of the Roman emperor 137
praetorship, a consulship. When we love and hate and fear these things, those
who have the disposal of them must necessarily be our masters’ (EpictetusDisc.4.1.60 (On Freedom)). So no one could afford to opt out from the
use of ﬂattery, because they could not ensure their rivals for rewards wouldnot use it. Indeed, the tendency will be to seek ever more extravagant andnovel forms of ﬂattery in order to outdo competitors for the favours of theemperor. Although philosophical critics tended to emphasise the degradingnature of ﬂattery, it was not simply a mark of servility; it was also a powerfultool in the hands of the courtier to achieve his ends.
45
There are two main strategies open to the ruler for trying to deal with the
dilemma: repression and reward.46These can be set out as the two axes of a
graph and the position of any ruler represented as a point which marks the
particular combination of repression and reward that is used. Repressioncan be effective; but its primary limitation is that it does not solve the Dic-tator’s Dilemma; rather it engenders it – the more the ruler represses, themore he increases his fear, because he cannot be sure of what his subjectsare really thinking and planning. From the start of the Principate, with theprecedent of the assassination of Julius Caesar before them, emperors hadtotake measures to protect themselves – so bodyguards, occasional wearing
of armour under the toga at public events, and personal searches becamethe norm. But emperors had to balance this against the expectation of beingopen and accessible to people. They faced a further problem, summed up inan aphorism ascribed to Domitian (Suet. Dom. 21): ‘He used to say that the
lot of emperors was a most miserable one, because no one ever believes inany conspiracy they discover unless it has resulted in their deaths.’
47So the
suppression of plots, such as that of Piso in AD 65, can lead to the creationof martyrs, whose reputation posthumously may be exploited to foster andmaintain opposition to the emperor. But the existence of men, often of inﬂu-ence and access to the emperor, who are ready to plot the emperor’s death,raises problems for the rest of the elite. Is ostentatious and violent opposition
totyranny the most desirable model for their own behaviour? Should they
feel guilty, or morally deﬁcient, if the yrefuse to join in the opposition and,
indeed, seek to continue their careers even under repressive tyranny? Theseproblems are neatly illustrated by Tacitus’ own unsympathetic account of the
45Scott 1990 :34: ‘We get the wrong impressio n…i fw ev i s ualize actors perpetually wearing fake
smiles …T od os oi st os e et h ep e r f o r mance as totally determined from above and to miss the
agency of the actor in appropriating the performance for his own ends.’
46Wintrobe 1998 :33ff.
47Transferred to Hadrian by SHA Avidius Cassius 2.4, on which see Laurence and Paterson ( 1999 :
192f).

138 jeremy paterson
Piso conspiracy ( Ann.15.48), in which he challenges the motives, behaviour
and bravery of most of the participants. The reason for his attitude is revealed
in his biography of his father-in-law, Agricola ( Agr.42), where he represents
Agricola as the model of an honourable man, who with moderatio andpru-
dentia steered a successful career under Domitian and did not provoke the
vengeance of the emperor upon him by deﬁant and futile parades of inde-
pendence ( libertas ). Tacitus declares: ‘Let those, whose habit is to admire
disobedience, recognize that men can be great even under bad emperors,and that obsequium andmodestia ,i fa l l i e dt o industry and energy, can reach
that peak of praiseworthiness, which most men attain only by following aperilous course, winning fame, without beneﬁting the state, by an ostenta-tious martyrdom.’ Tacitus was speaking for most of those around him whosought to live out honourable lives under the Principate. Here is the truecredo of the courtier. The continued survival of the Principate as a systemwas dependent upon Tacitus’ viewpoint being the general one.
Reward is a key strategy for rulers, or rather over-reward –that is, the
use of a price premium to stimulate loyalty and build trust. This can bedone by the distribution of rents, direct gifts to enable individuals to gain ormaintain status, the support for individuals’ public careers, the granting ofofﬁce, etc. There are innumerable examples of the Roman emperor’s givingof rewards and gifts both large and small.
48The need to appear generous, to
reward lavishly, remained despite the fact that the Roman emperors tended to
become the monopoly providers of most major goods (high ofﬁce, fundingfor major projects, etc.). Emperors could ill afford a reputation for stinginess.
According to Suetonius ( Tib.46) Tiberius was ‘ pecuniae parcus et tenax ’–
although it might be questioned whether this was necessarily a criticism –and Galba had a reputation for avarice ( Galba 12). On the other hand,
excessive lavishness and generosity were equally the subject of criticism,
because it was recognised that the emperor would need to fund this throughadditional exactions from the people (Suetonius Gaius 38). The fact that
the outcome of trials was often the conﬁscation of wealth and property,which frequently came directly or indirectly into the funds controlled by theemperor, created the view that the primary motivation for such trials wasto bene ﬁt the coffers of the emperor.
There is another key feature of rewards. They need to be given with the
prospect that they will be repeated in the future. The expectation of future
reward is a powerful incentive for the courtier to remain loyal (see Brosius in
this volume on the politics of gift-giving at the Achaemenid court). It is part
48Millar1977 :133ff and Kloft 1970 .

The creation of the court of the Roman emperor 139
of the process of building up trust on both sides, the creation of conﬁdence,
and the building of a reputation for not ‘cheating’ on the unspoken contractbetween ruler and ruled, despite the fact that such a contract was essentiallyunenforceable by either party. This is the fundamental explanation behindmuchof the court routine and ceremony – the invitation to the salutatio ,t o
theconvivium ,the daily greeting of senators and equites ,and the formal kiss
of the emperor. All these help to promote a regular way of demonstratingcontinuing conﬁdence on both sides in the relationship.
Finally, it is in the interest of the ruled to persuade the ruler to take a
long-term view of his rule. A ruler who only expects to be in power for ashort time is more likely to take all opportunities to enrich himself and hissupporters at the expense of the immiserisation of his subjects, because hedoes not care about the longer-term consequences. On the other hand theruler who seeks to remain in power is more susceptible to implementing
policies designed to encourage long-term loyalty, including being seen tobe of beneﬁt to his subjects. This is why from the very beginning of thePrincipate the succession to the Roman emperor is seen as a key issue not
just for the ruler but also for the ruled. The annual oath of loyalty sworn by allcommunities was to the emperor and his family. By transferring allegiance
tothe heir of the emperor, his subjects fostered the emperor’s willingness
totake a long-term view. Equally, among the Roman elite the succession
of the emperor’s legitimate heir held out the prospect of the continuationof favours currently enjoyed and, perhaps more importantly, prevented thepossibility of power falling into the hands of a rival member of the elite. AsElias ( 1983 :177–8 no. 15) notes, hereditary succession ‘assured each of the
leading groups that [the king] was not too one-sidedly bound to the interestsof an opposing group since, unlike a usurper, he had not needed to seek alliesamong the conﬂicting social groups in order to come to power. The lawfulorigin of the kings distanced them equally from all the social groups inthe country.’ In the case of the Roman emperors there was the interestingvariation of the current emperor designating a successor by adopting fromwithin the Roman elite, normally when no natural heir was available.
49It
is far from clear that adoption was obviously more acceptable to Romansenators than the continuation of a dynasty. Tacitus’ attitude is difﬁcult todiscern, as commentators have noted (see Chilver 1979 :76–7 and Hammond
1959 :ch. 1), and he also gives Mucianus a speech later which forcefully
rejects this approach ( Hist.2.76–7). The undesirability of any alternative to
the legitimate inheritance of the position of princeps was amply illustrated
49On Galba’s adoption of Piso see Tac. Hist.1.14ff; on Nerva’s of Trajan in October 97: Pliny Pan.8.

140 jeremy paterson
on those occasions when the system broke down – the civil wars of the
30s BC and of 69 AD and 194 AD.
Imper ial circles
Elias ( 1983 :122ff) notes that in crises charismatic leaders (in a Weberian
sense) emerge. They need to foster a cohesive group around them who areprepared to suppress their jealousies and rivalries to work to promote theirleader. In such situations talent and ability rather than lineage and nobil-ity are essential for the members of such a group. So such situations canoffer considerable opportunities for advancement which are not normallyavailable. However, once the monarchy has been established, then the abso-
lutist patrimonial monarch really wants to promote jealousies and rivalrieswithin his court in order to ensure the fragmentation which helps to prevent
the emergence of rivals. This model neatly ﬁts the emergence of the Prin-cipate. Augustus’ inner circle originally consisted of ‘new men’ rather thanRepublican nobility (the thesis of Syme 1939 ). Such men owed their careers
entirely to Augustus. On the other hand, one of the notable features of thesucceeding reigns is the number of trials in which the Roman elite sought todestroy the careers of rivals – a situation fostered and encouraged by emper-ors with rewards for successful prosecution and frequently the addition ofconﬁscated property to the imperial holdings.
Around the emperor there grew up a series of concentric circles contain-
ing groups and individuals who gained power and inﬂuence for themselvesbytheir perceived proximity and access to the emperor: the imperial family
(domus ), close conﬁdants, the imperial household ( familia )and ‘friends’
(amici ). These made up the bulk of the ‘court’ . As noted above, the Hellenis-
tic term ‘ aula’was adopted and widely used for the court surrounding the
emperor. The term covers both the sense of the physical environment – thepalace itself – and the people who surrounded the emperor. So the aula could
be said to favour a person (Tac. Hist.1.13: prona in eum aula –said of Otho).
What is less clear is how far the court extends. Marcus Aurelius ( Med .8.31)
described the court of Augustus as consisting of ‘wife, daughter, descendants(eggonoi ), members of his parents’ generation ( progonoi ), sister, Agrippa,
kinsfolk ( suggeneis ), members of the household ( oikeioi ), friends, Areius
[court philosopher], Maecenas, physicians, haruspices ( thutai )’ . So the court
usually covers family and household, but also at least some ‘friends’ – thosewho were regularly in the company of the emperor, particularly at leisuretimes. Suetonius ( Vit.4), for example, describes Aulus Vitellius as holding

The creation of the court of the Roman emperor 141
aprominent place at court ( praecipuum in aula locum tenuit ). He was the
familiaris of three emperors (Gaius, Claudius and Nero), because, accord-
ing to Suetonius, he shared Gaius’ passion for charioteering and Claudius’
obsession with gaming, and ﬂattered Nero’s lyre-playing.
Atthe heart of the court was the emperor himself. Proximity to power
endows individuals with power and inﬂuence in their turn. Thus the imperialdomus ,the family, frequently acted as power-brokers in their own right. The
most notable evidence of this phenomenon is the appearance under the Prin-cipate of powerful women (the mothers, wives, daughters and mistresses ofemperors) wielding an unprecedented public inﬂuence. Livia under Tiberiusis a case in point. There was her notorious support for Plancina, Piso’s wife,now conﬁrmed by the inscription recording the trial and punishment of Piso(T ac. Ann.3.15). Then there was the consul of 29, Gaius Fuﬁus Geminus, who,
according to Tacitus, was good at attracting women ( aptus adliciendis femi-
narum animis )and owed his success to the favour of Livia ( gratia Augustae
ﬂoruit ). All the members of the imperial family attracted supporters and
an entourage.
50Within the family group there is an interesting subset –
the children of particular favourites who live and are brought up alongsidemembers of the imperial household. For example the future emperor, Titus,son of Vespasian, who was closely associated with Narcissus, the powerfulfreedman at Claudius’ court, was brought up in aula with Claudius’ own
son, Britannicus, and had the same schooling, with the same masters. Hewas sitting next to Britannicus at the fateful dinner when Britannicus col-lapsed and died (Suet. Tit.2). Even more interesting were the sons of foreign
kings brought up in the court – they are sometimes misleadingly describedas ‘hostages’ . They are a group which deserves further study.
51
Close to the emperor there was a small number of individuals of varying
status, who acted as trusted conﬁdants, and whose whole career rested upontheir intimate connection with the princeps rather than any ofﬁcial ofﬁce
they held. Maecenas and, above all, M. Agrippa under Augustus and Sejanusunder Tiberius provide the model for these sociilaborum –‘partners in
mylabours’ . They too came to wield inﬂuence and power.
52The fact that
these individuals’ inﬂuence was based entirely on their connection withthe emperor and their being seen as the recipient of special favours fromhim is shown by the protests of senators after the fall of Sejanus. They
50See an excellent study of the circle around Antonia under the reign of Tiberius: Kokkinos 1992 .
51Iowemuchon these last two groups to discussion with Andrew Lipinski, my postgraduate
student.
52Forad etailed discussion of how the inﬂuence of such people was gained and employed see
Roddaz ( 1984 ).

142 jeremy paterson
wereanxious that their public demonstrations of favour towards him might
be the source of accusations now that he had been removed from power.
Tacitus ( Ann.6.8) has M. T erentius admitting that he was Sejanus’ friend,
‘that he had sought him out and had been delighted when his friendship was
accepted’; but he had done so because he had seen the honours and favourwhich Tiberius had bestowed on him. ‘We thought it marvellous if Sejanus’freedmen and doorkeepers recognised us.’ Friendship with Sejanus openedthe way to friendship with Tiberius ( ut quisque Seiano intimus ita ad Caesaris
amicitiam validus ). It was not Sejanus T erentius and others honoured, but
Tiberius, who had shown Sejanus such favour. It was not their role to reason
why: ‘The gods have given you, Tiberius, supreme control, to the rest of us isleft the glory of obedience ( obsequii gloria ).’ The whole passage is a powerful
illustration of the processes at work in the imperial court.
Around the emperor grew up the familia ,the household of slaves and
freedmen, some of whom also gained inﬂuence from their proximity tothe central power.
53The familia included the domestic staff of the imperial
residences, the doorkeepers, litter-bearers, ornamental gardeners, tasters,
chefs, court jesters, doctors, astrologers and the like. Most of these were
slaves; but once again proximity to the emperor could endow the individ-ual with inﬂuence in his own right. The post of cubicularius (controller
of the bedchamber), for example, had considerable inﬂuence on access tothe emperor. Gaius’ cubicularius ,Helicon, is described by Philo ( Leg.175)
as ‘playing ball with Gaius, exercising with him, bathing with him, sharingmeals, and was with him when he went to bed’ . It was a matter of concern toPhilo’s Alexandrian embassy that Helicon sided with their Greek opponentsin their dispute. People of course objected to the role which such individ-uals played as ‘gatekeepers’ to the emperor. ‘How should a man becomewise all of a sudden when Caesar puts him in charge of his chamber-pot?’ ,
remarked Epictetus.
54Epictetus, a court insider, himself a one-time slave
of Epaphroditus, Nero’s freedman alibellis ,recalled how his master sold
another slave because he was useless, only to ﬁnd the same man installed asshoe-maker to the emperor. Whereupon, ‘if only you could have seen howEpaphroditus honoured him now!’ It was not the emperor speciﬁcally who
gave such minor post-holders power and inﬂuence inconsistent with theirlow social status. Rather it was the courtiers who perceived the person aspotentially inﬂuential and as someone who might give them access to the
53On all of these see the excellent study by Weaver ( 1972 ).
54Epictetus Disc.1.19.7 (How should we behave towards tyrants?). See also Millar ( 1965 )and
Starr ( 1949 ).

The creation of the court of the Roman emperor 143
emperor. While the court in some ways reﬂected and reinforced the structure
of Roman society, in these ways it could also be perceived as subverting it.
The familia also included the slaves and freedmen who carried out the
day-to-day administrative tasks, dealing with funds, imperial property, etc.The hierarchy ran from what Weaver designates as ‘sub-clerical’ functionar-ies, up through ‘clerical grades’ to the senior administrative posts staffedunder the Julio-Claudians by freedmen. Again the posts could be a sourceof power and inﬂuence. The most notorious examples were the freedmen ofClaudius, Pallas, Narcissus and their like, whose very real power (for exam-ple representing the emperor on campaign) and access to the emperor weredeeply resented by the elite. That their inﬂuence was real is shown by thefact that these posts came to be held by equites in the second half of the ﬁrst
century. Members of the familia Caesaris attracted a much higher propor-
tion of wives of freeborn status than the rest of slave-born society, accordingtoWeaver’s analysis. At the pinnacle the ofﬁces could be the cause of great
social mobility – at least for a few. Augustus granted the equestrian gold ringtohis physician Antonius Musa (Dio 53.30.3). Under Claudius, Narcissus
received ornamenta quaestoria by se natorial decree and Pallas the ornamenta
praetoria .Under Vespasian, Hormus was advanced into the equestrian order
for the leading role he had played in the Flavian takeover (Tac. Hist.3.12: ‘ is
quoque inter duces habebatur ’,and 4.39).
Then there were the amici (‘friends’). ‘Good amici werethe most valuable
instrument of good government’ (Tac. Hist.4.7). However, these are a very
difﬁcult group to deﬁne. In a sense everyone of roughly the same social statusas the emperor, that is, of the senatorial and equestrian orders, could expectto be co nsidered an amicus ,unless the emperor speciﬁcally renounced that
friendship. Emperors regularly described individuals in letters or decrees as‘myfriend’ , both as a courtesy and to emphasise to others that the person had
his backing. The important thing was that ‘Caesar knows you’ . Thus Epictetus(2.14.1): ‘What else could anyone imagine you need? Y ou are rich; youhave children, possibly also a wife and many slaves. Caesar knows you. Y ouhave many friends.’ When the emperor was in Rome, it became the customfor senators and equites togo to greet him daily (these are the cotidiana
ofﬁcia ). These were separate from the more general salutatio .Acall to the
palace, a meeting with the emperor, the ceremonial kiss, all were sufﬁcienttoconﬁrm an individual as a ‘friend’ . However, it is clear that a closer group
of companions ( comites )and advisors formed an important sub-category
of ‘friends’ – the cohors amicorum ,the entourage of the emperor. One of
the things which distinguished them from other amici was that they would
travel w ith the emperor when he was away from Rome. The choice of such

144 jeremy paterson
companions as advisors and ofﬁcials was one of the most fraught decisions
any emperor had to make. It is no surprise that discussion of such decisions
takes up by far the greatest part of the debate on how to run the empire, whichis put in the mouths of Agrippa and Maecenas by the senatorial historianDio.
55The general principle advocated by Dio and others was to pick the
‘best’ citizens. By ‘best’ was meant those of the highest social classes, but
there was also recognition to be given for ability and talent. However, therewas no escape from the emperor’s dilemma: ‘The ruler who grants such giftsin the correct way clearly makes it his primary goal to weigh the merits ofeach individual. In this way he honours some and passes over others, withthe result that some have another reason for pride and others feel a newresentment, each being conscious of their own worth. If, however, the ruler
tries to avoid this result and tries to distribute these honours in a haphazard
way, then he will fail completely’ (Dio 52.12.5). The comites ,then, are those
amici who are called regularly to advise the emperor, to form his consilium ,
todine informally with him, and to enjoy relaxation and entertainment with
him. Such men rarely lived in the palace. They were called into the presenceof the emperor. The Elder Pliny would visit Vespasian before daylight (PlinyEp.3.5.9) before going on to carry out his ofﬁcial duties. It is clear that being
acomes was frequently a very demanding position.
56There were complaints
that it diverted them from their other duties – particularly advocacy. Cn.Cornelius Lentulus Augur, the wealthy conﬁdant of Augustus, complained
that Augustus ‘had not heaped on him nearly so much as he had lost bysurrendering the practice of eloquence’ (Sen. de Ben .2.27.2). Pliny ( Ep.4.24)
remarked of a contemporary that: ‘Friendship of the emperor has taken
him away from his civil duties.’ The duties of ‘those who were occupiedin the emperor’s entourage’ ( hi qui circa principem sunt occupati )could be
accepted as a legitimate excuse for not being able to fulﬁl a legal obligation(Dig.4.4.11.2).
Itwas exclusion from friendship which could be viewed as so serious.
When Vespasian gave offence to Nero by his lack of attention at Nero’s per-
formances in Greece, he found himself excluded not just from the emperor’sentourage ( contubernium )but also from the public salutatio ,ashediscov-
ered to his consternation ( trepidum )w h e nad o o r k e e p e r( quidam ex ofﬁcio
admissionis )put him out and told him to go to hell! He retired to a country
estate, where he was described as in hiding and terror until a grant of ofﬁce
55Dio 52.1–40. The debate, which reﬂects the views of Dio at the beginning of the third century
AD, has been described as ‘the most remarkable treatise on politics that survives from theprincipate’ (Brunt 1988 :40).
56Crook 1955 :26.

The creation of the court of the Roman emperor 145
signalled he was back in favour (Suet. Vesp.4 and 14). When Nero forbad
Thrasea Paetus from attending with the rest of the senate to congratulate
the emperor on the birth of his daughter, according to Tacitus Thraseatook this as ‘the announcement of his imminent death’ ( Ann.15.23). Again
Plutarch ( Mor.508A, ‘On talkativeness’) describes a certain Fulvius, whose
indiscreet gossip to his wife about Augustus caused the emperor embar-rassment. When Fulvius attended the morning meeting with Augustus, his
‘Hail, Caesar’ was met with the response, ‘Farewell, Fulvius’; he retired andcommited suicide.
57
Ritual and routine
The story of Vespasian shows that membership of the ‘court’ was not amatter of titles, posts or anything objective; you were a friend of the emperorbecause you were seen as a friend of the emperor. Y our status was dependenttosome extent on what others thought your status was. As Elias ( 1983 :94)
emphasises, a person belonged to courtly society, no matter what his title orposition, only as long as the others thought him a member. ‘Social opinionis the foundation of existence.’ From this comes the fact that our sourceson Roman imperial life are full of stories, rumours, signs of favour, etc.,because that was what made up courtly life; these were the ways in whichpeople judged what their own standing and that of others was in society andin relation to the emperor.
So the court becomes a stage on which the emperor can demonstrate his
favour and continuing approval of his courtiers and receive their expressionsof loyalty, while the courtiers can be seen to be the recipients of the emperor’sgood will and to be maintaining their loyal support for the emperor. It is acharacteristic of court societies that they develop routines and rituals. These
become ways of creating and reinforcing predictability and the conﬁdenceof all parties in the world of unenforceable promises which is at the heart ofthe relationship between the emperor and the court. These cotidiana ofﬁcia ,
daily duties, could be represented by both emperor and courtier as irksomeobligations. However, in reality their regularity and routine nature were areassurance and created conﬁdence in all who took part. It is a characteristic
of ‘bad’ emperors that they often neglected or disrupted these events. So Dio(78.17.3) complained that Caracalla ‘would send us word that he was goingtohold court or deal with some business directly after dawn; but then he
57Tac.Ann.1.5 has a different version associated with Paulus Fabius Maximus.

146 jeremy paterson
would keep us waiting until noon and often until evening, and would not
even admit us to the vestibule of the palace, so that we had to stand around
outside somewhere else. Then usually at some late hour he decided that he
would not even exchange kisses with us that day.’
There was another aspect of the routine ceremonial and ritual at court
which was a constant source of debate and negotiation between the emperorand the elite. Roman senators expected the emperor to be ‘one of us’ , asPliny put it,
58that is, to treat senators as far as possible as social equals.
The emperor was to be no more than le premier gentilhomme .B u t on the
other hand the dignity of the ofﬁce sometimes required the emperor toemphasise the distance between himself and members of the elite, and todemand deference. There was always a tension between those two aims – itwas another of the dilemmas of ofﬁce.
Marcus Cornelius Fronto, the orator, tutor and conﬁdant of the emperor
Marcus Aurelius, in a letter to the emperor ﬁnds a neat and clever way both
of complimenting the emperor and also of suggesting that his relationshipwith the emperor was one founded on true friendship, not simply on any
of the conventions. Why, he asks, does Marcus love his servant Fronto somuch?Itwas not as if he had ever held a major position. Indeed, he was not
even exceptional in discharging the cotidiana ofﬁcia ,daily duties. Far from
it. ‘For he does not come regularly to your house at dawn; nor does he takepart in the daily salutatio (greeting) nor does he follow you everywhere as a
comes (companion); nor is he a constant spectator of your activities.’
59It is
auseful list of the courtly routines of a Roman senator.
Itwas Augustus who initiated the requirement that, when he was resident
in Rome, senators should attend him in a body ﬁrst thing in the morning.From 12 BC Augustus did not hold the ceremony on days when the senate
met, so as to avoid holding up the meeting, and indeed when his health brokedown in his last years, he abandoned these public appearances altogether.
60
Whywould the emperor impose upon himself and the members of the
elite what could be a tiresome obligation on both sides? In reality it was apractical way to solve real problems. T o see the members of the senate and theprominent members of the equestrian order together enmasse prevented the
emperor from being pestered constantly by them individually for interviewsand meetings and also avoided the inevitable jealousies which could arisewhen one is seen to be favoured over another. It was a characteristic practiceof kings and the imitators of kings ‘to make divisions of their throng of
58Pliny Pan.2.4.59Fronto ad M. Caesarem 1.3 (Naber p. 5/6) (Loeb 1 pp. 83f.).
60Dio 54.30.1, 56.26.2.

The creation of the court of the Roman emperor 147
friends’ ( discribere populum amicorum )( S e n . de Ben .6.34.2). The strategy
played upon the strong sense of social equality among the upper classes. In
order to control the numbers, emperors adopted the Hellenistic practice ofdividing the visitors into groups ( prima admissio ,secunda admissio ), with
the whole event controlled by ofﬁcials.
61Itis extremely unclear on what basis
the division was made. Senators and equites could both ﬁnd themselves as
part of the cohors primae admissionis (see Sen. de Clem .1.10), and Winterling
(as noted earlier) may be right to see it as simply an arbitrary administrativeconvenience, rather than a division made with regard to social pre-eminence.
Further, it was inevitable that early morning would be chosen for these
gatherings. It was in any case the traditional time for the salutatio .I fthe
emperor had tried to hold the gathering at any other time, this would haveopened the way for people to try to jump the queue, to get to see the emperorearly and so on. By holding the gatherings early, both the emperor andcourtiers were freed to get on with their lives and work for the rest of the day.
Atthis morning reception both senators and the emperor were expected to
wear formal dress, the toga. Nero caused offence by appearing at the salutatio
for senators in informal dress. Before he was emperor, Severus Alexanderis supposed to have turned up at an imperial dinner in an inappropriatepallium ,G r e e kc l o ak; he was lent one of the emperor’s own togas, which was
taken as an omen of future greatness!
62
The emperor would greet senators with a kiss. There has been much
misunderstanding about the signiﬁcance of this kiss. It was the normal formof greeting between members of Greek and Roman elites.
63Arrian (4.11.3)
notes that ‘on greeting men receive a kiss’ . Seneca ( de Ira 2.24.1) suggests that
the kiss was a normal part of a greeting when he imagines a complaint that:‘that man did not give me a civil greeting; he did not return my kiss’ .
64Cicero
saw his enemies signifying their support for Clodius by kissing him (Cic. ad
Fam .1.9.10: Sest.111). In his epigrams Martial pictures a world full of kissing.
Indeed it was impossible to escape kissers ( basiatores )(Ep.11.98.1): there was
the man who insists on kissing everyone even in the cold of winter ( Ep.7.95);
at one point ( Ep.12.59) he gives a whole list of the undesirables who insist
on kissing you on your return from a long stay abroad. The prevalence ofkissing among the upper classes was demonstrated by an outbreak among the
61These ofﬁcials included assistants ( adiutores ), masters ( magistri ), announcers of names
(nomenclatores )(CIL3.6107; 14.3457; 6.8931; cf. 6.8698ff).
62SHA Hadrian 22.2; Dio 63.13; SHA Severus 1.7.
63Contra Kay ( 1985 :265) who wrongly asserts that kissing was not a usual form of greeting in the
ancient world. On the kiss see Kroll 1931 ;Friedlander 1921: 56–9 (M. Bang); K ¨uhn 1987 .
64‘Ille me parum humane salutavit; ille osculo meo non adhaesit .’

148 jeremy paterson
upper classes of mentagra ,ah erpes-like skin infection, which led Tiberius
toban the exchange of the daily kisses ( cotidiana oscula )a t thesalutatio
(Pliny NH26.2 and Suet. Tib.34.2). The kiss was a gesture towards equality,
asign of the emperor treating senators as his peers.65So ‘there was general
delight’ , claimed Pliny ( Pan.23.1), addressing the emperor Trajan, ‘when you
embraced the members of the senate, as they embraced you when you went
away’.Once again it is the arrogant, ‘bad’ emperor who does not engage in
this practice.66Pliny ( Pan.24.2) goes on to praise Trajan further: ‘Y ou do
not force your citizens to grasp your feet, nor do you simply proffer yourhand to their kiss; your lips keep their old courtesy now you are emperor;your hand respects its proper place.’ T o kiss someone on the hand was the
characteristic action of a client begging a favour – so candidates for ofﬁce
kissed the hands of those whose support they solicited.
67Tokiss someone’s
feet was the extreme act of a person begging for mercy, the characteristic of adefeated enemy or a helpless subordinate.
68Simply to hold out one’s hand to
be kissed thus emphasised the superiority of the ruler, while to allow your feettobe kissed was the mark of the tyrant. According to Suetonius ( Gaius 56.2)
the petty attempts of Gaius Caligula to demean Cassius Chaerea includedholding out his hand to be kissed. Later it was said that Maximinus Thraxacted in an arrogant way at the salutatio ,‘holding his hand out to be kissed
and allowing his knees and even his feet to be kissed’ (SHA Maxim .28.7). So
the kiss was an important gesture; but members of the elite expected morethan this. A lack of warmth to accompany the formalities could be takenas a sign of disfavour. According to Tacitus ( Agr.40), Agricola on his recall
from Britain was ordered to come to the palace by night, was received justwith the formal kiss by Domitian and then without a word of discussion
was dismissed to join the crowd in attendance.
Emperors also held salutationes on some days for the rest of the people
as well. These were the publicae orpromiscuae salutationes .
69From an early
hour ‘a crowd made up of almost all classes’ ( multitudo omnium fere ordinum ,
65Cf. the account by Chares of Mitylene ( FGrH 125 F 18a, quoted by Plutarch, Al.54) on the
introduction of kissing among Alexander and his Companions. At a symposium a friend, after
drinking from a cup given to him by Alexander, kissed Alexander, and all but one of the otherfriends did so as well. See Bickerman 1963 ;Spawforth in this volume.
66Dio 59.27.1, ‘Gaius Caligula used to kiss very few.’
67Seneca Ep.118.3; Epictetus Disc.4.10.20.
68Cic.ad Att .1.14.5 on Clodius; Sest.145 on his brother Quintus begging Cicero’s enemies for
support; ad Fam .4.4.3 on C. Marcellus. Seneca noted that when Gaius Caligula stretched out his
slippered foot to be kissed, the victim ‘lay a sasupplicant before him just as defeated enemies
grovel before their opponents’ ( de Ben .2.12); cf. Dio 59.27.1.
69Suet. Aug.53.2.

The creation of the court of the Roman emperor 149
Gell. NA4.1.1) thronged around the area outside the front of the palace –
thevestibulum .70When the announcement came that the emperor was
receiving, people divided up into their separate groups for admission.71
They then proceeded into the palace. Security required that those coming in
would be subject to searches, a potentially tricky issue for the emperor and
one which was likely to cause offence.72The way in which one was received
bythe emperor was a key indicator of one’s standing. For some merely to
be present at the salutatio was a high honour. Emperors were viewed with
approval when they knew your name without prompting from a nomen-
clator .Nero ‘greeted all classes instantaneously and from memory’ ( omnis
ordines subinde ac memoriter salutavit ,S u e t.Ner.10).
The invitation to dinner with the emperor was a particularly important
marker for the elite. The convivium raised in an acute way one of the inher-
ent contradictions of the Principate. Was the emperor one of the nobilityor above the nobility? As D’Arms ( 1990 )has pointed out, the convivium
had a long-established principle of reﬂecting a sense of equality among theguests, as a way of fostering amicitia .For example Plutarch in one of the
quaestiones convivales conveysthis idea: ‘let men understand that the din-
ner is a democratic affair and has no outstanding place like an acropoliswhere the rich man is to recline and lord it over less distinguished folk’(Mor.616d–f). The emperor might be expected to go out of his way to put
guests at ease. So Pliny ( Pan.49.4–6) has Trajan taking all his meals in public
with his table open to all. He joined in conversation in a relaxed and easy
way. The contrast with Domitian is always there. One of the signs of this‘equality’ was an element of freedom of speech by all and a willingness of
the host to take jokes against themselves in good part: ‘at dinner parties acertain freedom is not unseemly in people of humble rank ( humiles ), while
hilaria is becoming in all’ (Quint. Inst.6.3.28). T o some extent it was possible
toplead the effects of drink to avoid any lasting resentment. The niceties of
theconvivia werealways a matter of subtle negotiation – and considerable
anxiety when a guest overstepped the mark.
Augustus’ dinners were always formal; he was good at drawing out even
his most reticent guest, and enlivened the meals with music and performance(Suet. Aug.74). On the other hand, Dio (67.9) describes the feast given by
Domitian for senators and equites tocommemorate the dead in the Dacian
campaign. In a room decked out in black, with place-tags that looked like
70On the vestibulum see Gell. NA16.5.71Gell. NA20.1, ‘We were separated up.’
72Suet. Aug.27.4, Claud .35.1; Vespasian tried to create a good impression by abandoning them
(Suet. Vesp.12).

150 jeremy paterson
gravestones, and a menu of dishes offered to the spirits of the dead, and in
asilence broken only by Domitian’s comments on death, the guests might
be forgiven for wondering just whose deaths were at issue. Their terror was
only assuaged by the appearance of gifts from the emperor once they hadreturned home. Misconstrued ceremonial or macabre joke? Historical fact
or literary construct? Perhaps both. Let no one doubt that there can be veryreal terrors in dining with a tyrant (see Simon Sebag-Monteﬁore ( 2003:
passim )onStalin). Y et different reactions to Domitian were possible. In
Silvae 4.2 Statius offers a much more favourable account of the sacra cena in
the palace in Rome, to which he was invited. He is pleased that he got to seethe emperor close up and the guests were not required to stand as Domitiancirculated round them. Again this is no simple eye-witness account andsome might be tempted to dismiss it as ﬂattery by a court poet. However, itwas difﬁcult for guests to avoid the feeling that, while they were watching theemperor, they were also on display and the emperor was observing them. Itwas easy to construe this as malevolence on the part of the emperor.
73
These stories demonstrate two fundamental points about the study of the
court. First, there is no one single correct account of the court; there are many
narratives from many points of view, all of which have validity. Allied to thisis a second conclusion, which arises from the account of Domitian’s dolefuldinner. Whatever the intentions of the ruler or of the other participants incourt life, it was impossible for anyone, emperor or subject, to impose a
single meaning and signiﬁcance on an event.
The emperor was constantly solicited for favours. As already noted, gift-
giving was a key strategy for emperors to win loyalty and support. However,
receiving gifts from people was potentially problematic. As Roller ( 2001 :
193ff) shows, emperors were extremely reluctant to be seen to accept majorgifts or favours, because it would be thought to have put them under an
obligation to the giver. This in turn would be the source of jealousy on thepart of other aristocrats and could be seen to undermine the emperor’s pre-eminence. Turning down a gift or favour without causing offence was oneof the arts of imperial government. When the people of Rome presentedAugustus with the subscription they had raised to pay for the rebuilding
of his house after a ﬁre, he simply took a small amount, no more than adenarius from each pile. He did not want his house to be a burden to thepeople; on the other hand his action was an acknowledgement of their kindoffer (Suet. Aug.57.2). The difﬁculty of accepting major gifts because of the
73Pliny Pan.49.5 praised Trajan: ‘Y ou do not sit menacingly over your guests, watching and
marking all they do.’

The creation of the court of the Roman emperor 151
jealousy it might engender may well be part of the explanation of why the
emperor was regularly included in people’s wills and why he was prepared toaccept such inheritances. These gifts came after death and when the emperorcould not show favouritism towards any individual.
74
This problem of ‘equality’ is found mo regenerally in the dialogue between
emperor and the elite, as Grifﬁn ( 2003 )has stressed. While true equality was
impossible and the recognition of the emperor’s superiority was unavoid-able, the emperor could at least act as though the transactions between himand others were the result of true friendship. So, according to Seneca ( de
Ben.2.13.2), ‘the gifts that please are those that are bestowed by one who
wears the look of humanity, or at any rate of gentleness and kindness, by one
who, though he was my superior when he gave them, did not exalt himselfabove me, but with all the generosity in his power descended to my ownlevel, and banished all display from his giving’ .
Palatium
One of the principal contexts for courtly behaviour was the palace. The samesort of dialogue which was carried on about the nature of the court andcourtly behaviour also developed about what was appropriate as a residence
for the emperor. We have seen that the houses of the Roman nobility inthe Republic were thought of as ways of displaying and demonstrating theirstatus, of re-emphasising the social hierarchy. But they were also in partpublic spaces, accessible to friends and clients, where the noble also carriedon public business at the times when he held a magistracy. There was a livelydebate from at least the second century BC among the elite about what wasasuitable scale for such buildings. The debate took the form of concerns
about the moral effects of luxury, but also about the ways that increasingwealth, combined with the competitiveness of the elite, endangered the elite’s
sense of being a collective in which all one’s peers were in some ways equals.That debate took on a new twist when it became a question of what wasappropriate for the residence of the princeps .
Julius Caesar solved the problem by basing himself in the Domus Publica
on the Sacred Way, the ofﬁcial residence of the pontifex maximus .I n3 6B C
Augustus’ solution was to obtain a number of houses, including that of the
orator Hortensius and the notably lavish property of Q. Lutatius Catulus,
74Roge rs1947 .

152 jeremy paterson
on the south-west corner of the Palatine.75This house was partly rebuilt
after a ﬁre in 3 AD. What we have to imagine is a sort of imperial com-
pound of houses, including Augustus’ , the so-called House of Livia, andCatulus’ house. Here resided not just Augustus, but some at least of hisfavourites. In 29 BC he gave the house of Catulus to M. Agrippa, and Sueto-nius ( Gramm .17) describes Agrippa’s sons, Gaius and Lucius, being tutored
in the atrium of the Catulinae domus .Itisimportant to note that from
the start the public ‘ofﬁcial’ aspect of the residence was emphasised. Theoriginal houses were bought at public expense and the public got togetherasubscription to fund the rebuild after the ﬁre, although (as noted above)
Augustus carefully avoided using most of these funds. The whole complex
became state property (Dio 55.12). It contained a temple to Apollo withtwo great libraries attached. This became a place for Augustus to receiveembassies. Thus in 13 AD ‘Augustus took his seat in the temple of Apollo,in the Roman Library’ to receive ambassadors from Alexandria ( P.Oxy .2435
verso). Signiﬁcantly, in Augustus’ old age the senate was regularly called to
meet in the library and portico of the temple of Apollo (Suet. Aug.129). It is
clear that the practice continued under the Julio-Claudians, as we see fromthe Tabula Hebana: ‘in the palace in the portico which is near the templeof Apollo, in which the senate customarily meets’ (Thompson 1981 ). The
destruction in the ﬁre of 64 AD seems to have ended the custom.
Sueton ius ( Aug.72–3) makes much of the apparent limited scale of these
buildings and their modest decoration and furniture. But this is misleading.Suetonius must be comparing the house with the huge scale of Domitian’s
Domus Flavia created at the end of the ﬁrst century AD. It should rather
be seen as something quite exceptional for its time. It focused attention onthe Palatine (Favro 1996 :203–4) with the gleaming marble of the temple of
Apollo, the tower with Augustus’ private study, and the entrance to the whole
complex ﬂanked by laurel trees, surmounted by an oak crown, and with a
display of trophies. Ovid’s response to this ( Tr.3.1.33–4) is noteworthy: ‘I
beheld doorposts marked out from others by gleaming arms and a dwellingworthy of a god ( tectaque digna deo ).’ It is worth underlining that once again
this ofﬁcial adornment of entrances is an honour which goes back to theRepublic. From the time of Augustus, as we saw, the imperial residence had
taken over the name of the hill, Palatium.
The presence of Agrippa in a house within the compound is perhaps not
that surprising given his exceptional status. But the palace never developed
75Dio 49.15.5 and 55.12.4–5, Vell.Pat.2.81, Pliny NH17.2, Suet. Gramm .17. See Dumser 2002 s.v.
Domus: Augustus, Steinby 1993–2000 s.v. Domus: Augustus (Palatium) (I. Iacopi).

The creation of the court of the Roman emperor 153
along the lines of Versailles in seeking to provide apartments for all the
court nobility. Maecenas chose to have his residence somewhere within
the huge Horti M aecenati on the Esquiline. Augustus used to retire there
when he was ill. There is a signiﬁcant number of houses of Augustan datescattered over the Palatine and it is tempting to wonder how many cameinto the possession of particular favourites of Augustus and his successors.T.Statilius Taurus, consul along with Augustus in 26 BC, had a house on
the Palatine (Vell.Pat.2.14.3).
Succeeding emperors proceeded to annex much of the Palatine hill and
beyond for their palace residence. The motive for this is less the expandinghousehold and more the desire to impress and to make space for buildings,porticoes and gardens comparable to those found in the great horti of sub-
urban Rome. In doing so it was possible to cross the line between what wasgenerally acceptable and what was not. So Gaius came in for criticism forhis extension which took the palace to the edge of the Forum, ‘making theTemple of Castor and Pollux his vestibule’ (Suet. Gaius 22). Of course most
notorious of all was Nero’s expansion of the palace to link it up with thehorti of Maecenas and imperial properties on the Esquiline, ﬁrst with the
so-called domus transitoria and then with the elaborate and controversial
domus aurea ,w h i c hb rought horti down into the heart of Rome.
Vespasian ostentatiously rejected the Neronian experiment which had
caused such controversy, and on the whole avoided the Palatium as muchas possible, preferring to live in the HortiSallustiani (Dio 65.10.4). It was
Domitian who renewed the ‘century-long quest to ﬁnd a style of domesticarchitecture appropriate to the princeps’ (Zanker 2002 :107 in the best recent
discussion).
76Several features stand out. The new palace has two built-
up and imposing fac ¸ades – one overlooking the Circus Maximus and the
other on the north, facing out onto an impressive open space (the Area
Palatina )and down a long slope which descended to the Arch of Titus.
Itdominated the city: hence the appropriate reference to it as an arxor
citadel (see Klodt 2001 :37–62). People often assembled in the Area Palatina
toawait the salutatio (Gell. NA20.1.2). They were faced with an imposing
fac¸ade w hich did not have an entrance but had a high socle on which stood
acolonnade. It was possible to come out into this colonnade through doors
from the major public rooms; but there were no stairs down from thislarge podium into the Area .T h ee m p e r o r might appear in the colonnade
looking down on the people in the Area .This creates messages of majesty and
remoteness, quite different from the accessibility associated, for example,
76See also Tamm 1963 .

154 jeremy paterson
with Vespasian – ‘the doors of the imperial residences were open all day and
no guard was placed at them’ (Dio 66.10.5) – or Trajan, whose free-and-
easy approach the Y ounger Pliny ( Pan.47.5) contrasted unfavourably with
the gates and locked doors of Domitian’s palace.
The entrance to the palace complex probably lay off to the left of the Area
Palatina as you faced the palace. Here was likely to have been the vestibulum
where everyone stood round and chatted, while waiting for the chamber-lain to call them in to the salutatio .Zanker then plausibly reconstructs a
circuitous route through the peristyle of the residential part of the domus ,
into the peristyle of the Domus Flavia and then into the main reception
halls. As Dickmann ( 1999 )has demonstrated, this sort of tour was com-
mon in Pompeian houses and was designed to show them off to visitors.The three biggest rooms of the public part of the palace, the Aula Regia ,
the so-called Basilica and the elaborate Cenatio Iovis ,w e r e the site not just
for receptions but also for the feasts ( convivia ). Zanker suggests these three
rooms alone could have accommodated triclinia for some 500 guests. It was
precisely his convivia which were such a notable feature of Domitian’s reign
(Suet. Dom .21). Martial (8.39) notes the unprecedented scale of these rooms:
Previously there was no place that could contain
The banquet tables on the Palatine and the ambrosian feasts.
77
InNero’sexotic and bejewelled dining rooms in the Golden House, there
wereceiling panels which could open to shower ﬂowers upon the guests
and piping which sprayed perfume upon them. In such a place it was notclear whether the diners were the audience or the actors. Dining areas inprivate villas continued to be built to take advantage of the view – that is,the diners inside are able to look outwards while they are dining. But it hasbeen suggested that in Domitian’s palace the dining areas look inwards andare constructed like theatres, with the diners acting as the audience and theemperor’s triclinium functioning as a sort of theatre box, where he can be
observed and can observe.
78
The Domus Flavia declined in use over the succeeding centuries. This
was in part because emperors from the second century onwards were lessfrequently in Rome. In addition, emperors from Augustus onwards did notalways reside in the imperial palace when they were in Rome; they oftenstayed in either the urban properties of friends or other imperial properties
in the city. They, like the nobility, had their summer retreats in the hills
77‘Qui Palatinae caperet convivia mensae / Ambrosiasque drapes non erat ante locus’ .
78Bek 1983 :81ff.

The creation of the court of the Roman emperor 155
around Rome (see most notably Domitian’s Alban villa). They also had
properties scattered around Italy for their various tours. More distant tourswould take great organisation and planning.
The construction of a suitable palace as the environment for the emperor
and his court was subject to the same debate as with all other aspects of
court life. All the residences from Augustus onwards functioned in part as
public buildings, places where the emperor met with his court and othersin spaces designed to impress.
Autocracy and the court
The history of the Principate has often been represented as a gradual descentinto absolutism. The development of the court could be seen as part of a pro-cess which eventually led to the elaborate bureaucratic and hierarchic court
around the remote hieratic emperors of the fourth and ﬁfth centuries AD.However,that would be misleading. In reality Augustus had the power to be
as arbitrary an autocrat as any of his successors. From the very beginning,power was centred on the ﬁgure of the emperor and there were few, if any,constraints other than those which were self-imposed. So to some extent the
character of each emperor’s rule depended on the personality of the individ-
ual.
79However,it is not the nature of the emperor’s power which changes
over time; rather it is the court, the arena in which that power is presented,
which develops as a result of the dialogue between ruler and ruled. The char-acter of court society may depend less on the behaviour of the individualthan on certain evolving sociological processes. The courtier needs to ﬁndnew ways of standing out from his contemporaries so as to win the particularfavour of the emperor. Hence there is a tendency to make higher and higherbids in the ﬂattery stakes and to propose more and more exotic honours forthe ruler. For his part the emperor has to pick his path between the wish torestrain excessive ﬂattery and honours but not to offend the proposer. Even
more important was the need to be seen to be fair to all and not to be at thebidding of favourites. As a result the emperor tended more and more to bepositioned above the court rather than in the court, and to become a god-likeﬁgure invested in glorious costume and surrounded with formal ceremonial.The origins of this process were there from the start with Augustus’ righttowear triumphal dress.
80Insuch a position of pre-eminence the emperor
79Foraninteresting discussion of this see Potter ( 2004 :60–6).
80Crook ( 1996 :136–7) and the ground-breaking studies of Alf ¨oldi ( 1970 ).

156 jeremy paterson
could be seen as above the daily disputes and bickering of the court. At the
same time it was in the interests of both the emperor and the court to for-
malise procedures and behaviour so as to make it look as though there wasarationale, a fairness in the workings of the court, rather than the arbitrary
whim of the emperor. The creation of courtly behaviour provided increas-ing certainty and clarity in how things got done. So emperors moved fromproviding individual gifts and rewards for courtiers to the regular paymentof salaries for members of the increasing court bureaucracies; posts at courtbecame graded in an acknowledged hierarchy; courtiers’ positions withinthe hierarchy of the court were recognised by the increasingly elaborateassumption of overblown honoriﬁc titles. This process can be seen either asthe culmination of the development of the court in the early Principate oras a speciﬁc response to the conditions of the late empire, as discussed byRowland Smith in the next chapter .Iwould argue that the court of the late
empire is in many ways very different from that of the early Principate, butboth are rational responses to the social logic which leads to the creation ofall court societies.
Courtly life was a collaboration, a quest, a dialogue between ruler and
ruled. It was full of experimentation – not always accepted or long lasting.
The courtier was as important a constructor of courtly society as the ruler –perhaps more so.

5 The imperial court of the late Roman empire,
c.AD300– c.A D4 5 0
rowlan ds m ith
I, [an obelisk], reluctant once, am [now] commanded to obey the Lords
serene…[ f o r ]e v e r y thing yields to Theodosius and to his everlasting
offspring; hence I am conquered and mastered and raised up into thehigh sky.
(Inscription on an obelisk-base in the Hippodrome adjoining the
Great Palace at Constantinople, c.A D 390 ( ILS821))
OEmperor Augustus [Theodosius], if ever there was any one who was
justiﬁably in fear and trembling when about to speak in your presence, Iam assuredly he; I both feel it so myself, and perceive that this is how Imust seem to those who share in your council at court.
Pacatus, Panegyric of Theodosius ,A D 389 ( Pan.Lat .II.1)
‘Wesawin the papers that you had had a long talk with King Theodosius,’
myfather ventured. ‘Why, yes – the King, who has a wonderful memory
for faces, was kind enough to remember, when he noticed me in thestalls, that I had had the honour to meet him on several occasions at theCourt…A na i d e – d e-camp came down to bid me pay my respects to His
Majesty, whose command I naturally hastened to obey.’
(M. Proust, Al’ombre des jeunes ﬁlles en ﬂeurs
(trans. C. K. Scott Moncrieff))
The imperial court in the late Roman state: ‘absolutism’ ,
imperial ‘decline’ and the notion of ‘the court society’
Late in the third century, in the wake of several profoundly destabilising
decades of near-constant foreign and civil warfare, the Roman empire’smilitary and administrative structures were extensively reformed on theinitiative of the emperor Diocletian (AD 284–305). His reforming effortsreached even to the imperial ofﬁce: an innovative system of ‘tetrarchic’
collegiate rule was developed – a team of four co-reigning emperors, two of
senior rank to oversee the eastern and the western territories respectively,

158 rowlan ds m i th
each with a junior to assist him.1That particular conﬁguration was soon
undone, chieﬂy by the ambitions of Constantine. But Constantine was a
tetrarch’s son, and his own career was rooted in the system: he had spent
twelve years at Diocletian’s court, and had ﬁrst tasted power as a juniortetrarch, and many of his own military and administrative reforms would
later build on Diocletian’s. So, too, in his last years he devised a power-sharing arrangement reminiscent of the T etrarchy, with his sons placedin regional capitals as junior emperors and Constantine himself residingin a newly founded capital on the Bosporus. That conﬁguration, too, wassoon unravelled, but the basic innovation implicit in Diocletian’s system –the division of power between co-ruling emperors based in regionally demar-cated sectors – would endure; Constantinople became the administrativecapital for the empire’s eastern sector, and from the mid-fourth centuryuntil the institutional demise of the western empire late in the ﬁfth therewereusually two emperors co-reigning, one in the east and one in the
west.
Modern historians of the empire customarily treat Diocletian’s reforms
as harbingers of a new phase of imperial government, and in Anglophonescholarship it is conventional to denote the empire from around the end ofthe third century onwards as ‘the late Roman empire’ . That said, the mod-ern usage arises less from any verdict on an individual emperor’s work thanfrom a more generalised impression of transformation: in many other con-nections besides its military and administrative organisation – social andeconomic structures, artistic and religious culture, an ‘absolutist’ imperialideology and style – the ‘late’ empire of the fourth and ﬁfth centuries givesthe impression of a world distinctly different from the Principate of theﬁrst to early third. Among the deep changes at issue, the most obviously‘transformational’ are perhaps the Christianisation of Roman government
and society, and the ‘barbarising’ of the army – the increasing reliance ofthe authorities on barbarian ‘federates’ recruited to serve as rank-and-ﬁlesoldiers and military commanders.
2ToEdward Gibbon, of course, these
changes seemed pernicious and enfeebling, a ‘triumph of barbarism and
religion’ that contributed to the fall of the empire; and on these or other
grounds many historians since Gibbon’s time, notably Michael Rostovtzeff,
have represented the late empire as an era of manifest decadence and decline.Butinthe 1970s Anglophone scholarship took a new and radical turn in
the writings of Peter Brown, and that older consensus evaporated: in whatsense, if any, the late empire could properly be reckoned a world in decline
1Barnes 1982 :196.2MacM ullen 1988 :199–204.

The imperial court of the late Roman empire 159
became the focus of prolonged debate. Brown himself, revisiting the issue
in the mid-1990s, thought the fourth and early ﬁfth centuries could moreaptly be styled the ‘apogee’ of the Roman state, at least in so far as its abilityto‘make itself present’ to subjects and ‘induce particular habits of mind
and behaviour’ was concerned;
3and to another expert writing at that time
it seemed self-evident that ‘no responsible historian would [now] want toaddress the fall of Rome as either a fact or a paradigm’ .
4To b e s u re ,n o n e
would now be willing to apply any model of ‘decline and fall’ without exten-
sive reservations, but there are currently revisionists as well as radicals in thisdebate
5–and there are some facts that are hardly disputable: on a very basic
test, its capacity to hold and effectively control territory against external
forces, the Roman empire was a less powerful state by the end of the fourthcentury than it had been under the Antonines in the late second; and by the
end of the ﬁfth it had ceased to exist as a state in the west. In that sense, atleast, ‘decline’ in the long run was real enough; and the question whetherinternal factors as well as external pressures contributed signiﬁcantly to itremains a real one. The barbarian invasions of the late fourth and early ﬁfth
centuries, it is true, were linked to large-scale movements of various peo-
ples beyond the empire’s frontiers, and beyond the control of its rulers. ButRostovtzeff, like Gibbon, believed that long-term trends within the empire
impaired its capacity to respond to military crisis, and it is implicit in thechapter-title Rostovtzeff chose to head his sketch of the reforms of Diocletian
and Constantine – ‘The oriental despotism and the problem of the decayof ancient civilization’ – that in his view the new style of imperial rule theyintroduced was itself a signiﬁcant contributory factor; its authoritarian ide-ology and centralising tendencies, he thought, further undermined a patternof urban life and local civic patriotism that was already under strain, and itsproliferating bureaucracy encouraged venality and corruption in adminis-tration. As Rostovtzeff formulated it, the argument was vulnerable on more
than one count, and its underlying premises have often been challenged,but in a modiﬁed form it still has its modern supporters.
6Itis undeniable,
at least, that the late imperial style of rule emphasised absolutist principlesin its ideology and imagery, and that it fostered the growth of a centralisedand much enlarged bureaucracy in which ofﬁces were often bought andsold; and these two developments are closely bound up with a third – theemergence, from Diocletian’s time on wards, of a distinctively ‘late imperial’
court (we use the term generically here: at most times in our period, we have
3Brown 1997 :24; cf. Brown 1992 :17.4Bowersock 1996 :29.
5E.g. Liebeschuetz 2001 :233.6E.g. MacMullen 1988 :44, 190 and passim .

160 rowlan ds m i th
noted, there were at least two emperors co-reigning, and each had his own
‘court’).
The signiﬁcance of the court in this connection was barely touched on
byRostovtzeff, but Gibbon had observed it when he sketched the Constan-
tinian political system darkly (and in terms to which Rostovtzeff was clearlyindebted) as a ‘despotism’ in which ‘the simplicity of Roman manners wasinsensibly corrupted by the stately affectation of the courts of Asia’: ‘everyrank was marked with the most scrupulous exactness, and its dignity dis-
played in a variety of triﬂing and solemn ceremonies, which it was a study tolearn, and a sacrilege to neglect’; ‘a severe subordination of rank and ofﬁce’suffused the system, ‘from the titled slaves seated on the steps of the thronetothe meanest instruments of arbitrary power’; even the inauguration of
consuls ‘was [now] performed at the place of the imperial residence’ .
7In
Gibbon’s view, then, the late imperial court epitomised debasement, andbred corruption in government. On any view, it formed the hub aroundwhich the personnel of the expanded imperial bureaucracy lived and worked;ambitious men were drawn to its environs in increasing numbers from citiesfar and wide across the empire. Not only that: over the fourth century, thecourt itself came to wear a different face; its hierarchical structures and its
ceremonial and art mirrored and articulated the absolutist aspirations of
emperors in a pattern of ritualised behaviour that required the participa-tion of a complicit court elite. And these developments clearly had reper-cussions well beyond the conﬁnes of the court: court patronage brokereddeals, made careers, helped or baulked the efforts of individual petitionersand civic delegates to gain an audience with the emperor; and the imageof power evoked in court ceremonial ﬁltered through to the cities in otherways, more or less direct. When Brown stressed the fourth-century state’sability to ‘make itself present’ to subjects and to ‘induce particular habits ofmind and behaviour’ , he was thinking partly of just such things: the inﬂateddiction of imperial edicts, the grandiose portrait statues of emperors, theelaborate pomp and stylised acclamations that marked a touring emperor’sarrival at a city – and the echoes of these practices in the honours cities paidtoimperial governors and local benefactors.
8Asfor the other underlying
factor he had in mind, the wide reach of the state’s centralised bureaucraticapparatus, it was plainly in one sense a development that chimed with theabsolutist ideology emanating from the court: the expanded bureaucracy ofthe palace would facilitate a more thorough-going and detailed control ofthe empire from the centre. But there is also a paradox to observe in this
7DFvol. II, 89–91.8Rouech ´e1984 .

The imperial court of the late Roman empire 161
connection: courts, after all, entail court patronage; and a classic Weberian
typology treats bureaucracy, at least in its ideal ‘rational’ form, as a system at
odds with the workings of patronage. The palace-based bureaucracy of the
late empire was more ‘patrimonial’ than ‘rational’ , no doubt, but its growthmay still have made it harder in practice for emperors to rule as absolutemonarchs: there was a greater need to delegate power to subordinates; andas the bureaucracy developed it generated extensive documentary archives,and arguably a secretariat predisposed to conduct its business with textuallyestablished rules and precedents strongly in mind – which is to say, con-ducting it on terms that militated against the arbitrary exercise of autocraticpower.
9
The late Roman court was thus much more than the emperor’s home
and household and place of relaxation, and more than the place where hedeliberated with advisors on particular matters of policy. It was a socio-political institution that helped to r eshape the bureaucratic institutions of
the late empire, and a theatre in which a particular ideology of royal powerwas visibly enacted and made manifest to an elite within the court andbeyond it; and the image of power it constructed arguably impinged on thementality of subjects in broader political and cultural senses too, for goodor ill. On these counts, the court of the late empire is a subject that lateRoman historians must take seriously, and not just for its intrinsic interest;
its development needs to be studied with an eye to its bearing on the broaderdebate about ‘decline’ . The notion that its workings and ramiﬁcations mightsomehow have been detrimental to effective government in the late empireis not, after all, just a modern or Gibbonian notion; it surfaces in ancienttexts. In this connection, apophthegms and anecdotes can be particularly
interesting, since they will often be reﬂecting less an individual’s judgementthan a collective attitude or prejudice. T o the question of what it is thatmakes emperors evil, for instance, the author of the Augustan History had a
ready answer: ‘unscrupulous friends ( amici ), pernicious attendants ( satel-
lites), surpassingly greedy eunuchs, courtiers ( aulici )who are either fools or
knaves – and [the ruler’s] ignorance of public affairs’ . And he appealed toan expert witness to bear him out:
The emperor Diocletian, while still a c ommoner, declared that nothing was harder
than to rule well. Four or ﬁve men gather and form a plan to deceive the emperor,
then tell him what he must approve. The emperor, who is shut up in his palace,cannot know the truth. He is forced to know only what these men tell him; heappoints as judges men who ought not to be appointed, and removes from public
9Kelly 1994 :161–76.

162 rowlan ds m i th
ofﬁce men whom he ought to keep in ofﬁce . . . As Diocletian himself used to say,
“Anemperor – however virtuous, careful or excellent he may be – usually still gets
sold.”’10
One would hardly vouch either for the authenticity of this quote or for the
writer’s claim to have heard it from his own father, purportedly a contempo-
raryof Diocletian; the author of the Augustan History was almost certainly
ahoaxer writing very near the end of the fourth century, and writing light
entertainment rather than serious history. But if that is so, the whole passagemay seem illuminating of the terms in which the workings of the court ofthe 390s were apt to be perceived and caricatured in the popular opinionof the time. Its vignette of the emperor secluded in his palace, inattentivetoevents beyond it or misinformed about them by corrupt courtiers, and
routinely exposed to the guile of the eunu chswho ﬁlled the domestic ofﬁces
of the court, ﬁnds many parallels in other writings of those years: Synesiusdamned the eastern court of Arcadius as a crowd of petty-minded wastrelsswarming around a slothful ‘jellyﬁsh’ of an emperor; Claudian excoriatedits´eminence grise ,aeunuch who had made millions through bribe-taking
and intrigue and ended up a consul; and Ammianus recalled the eunuchs ofConstantius’ court as a scheming brood of vipers – the emperor, he point-
edly remarked, ‘had considerable inﬂuence’ with the chief snake.
11And the
impression the hoaxer conveys of the court as an emasculating, introvertedworld that made for lethargic rulers calls to mind a more precisely contex-
tualised anecdote of the early ﬁfth century, a story preserved in Procopius(Wars 3.2.39) about Arcadius’ brother, the western emperor Honorius
(AD 395–423). Honorius was remembered as a notably feeble sort, a puppetof his guardian and ministers (he had inherited the throne from his father atthe age of ten), and in this case the image of courtly decadence is conjoinedwith a celebrated emblem of imperial ‘decline’ – the capture of Rome in 410
byaband of Goths. Ravenna was by then the western capital, and it was there,
the story went, that Honorius learned of the disaster from one of the courteunuchs, a keeper of the royal poultry: when the eunuch announced to himthat ‘Rome has perished’ , Honorius’ ﬁrst thought was that his favourite bird,acockerel named ‘Rome’ , had died; ‘But he has just eaten from my hand!’ ,
he exclaimed – and then supposedly sighed with relief when the truth of thematter was explained to him. The story is transparent ﬁction, and the buttof the joke is an insipid individual rather than an institution; but an institu-tional malaise is implied, and the humour presupposes an audience inclinedtoview the court and its trappings as symptoms, at least, of the sickness. That
10SHA Aur.43.1–4.11Syn.De regno 14; Claud. in Eutropium i–ii; Amm.18.4.3–4.

The imperial court of the late Roman empire 163
view may have been quite fatuous, of course, and it plainly does nothing
toestablish that the workings of the late Roman court, taken in the round,
really did tend to be subversive of effective government. But by the same
token, if we wish to discuss the court in the round, it would be naive and
singularly unhelpful to seek to maintain any absolute distinction between
the court’s historical ‘reality’ and the perceptions that contemporaries hadof it. We must acknowledge at the outset that ‘the late Roman court’ isaconvenient shorthand expression for a complex historical category: the
underlying subject at issue is a distinc tively conﬁgured ﬁeld of collective
human activity and social experience, and the terms in which late Romans
understood and represented it are themselves an aspect of the subject. Onthat score even fatuous ancient representations of court life, if they spoke toasigniﬁcant strand of literate opinion at the time, may still be historically
signiﬁcant; ‘effective government’ in the late empire still required, amongother things, the cooperation of the authorities and the urban elites – andhence a certain degree of elite consensus: ‘in the fourth century, courtesywas still necessary’ .
12
Gibbon famously likened late Roman government to a ‘splendid theatre’ ,13
but he did not attempt any systematic study of the ritualised behaviour andceremonial of its court, and nor, for a long time, did modern historians of
the late empire. In the 1930s, Andreas Alf ¨oldi stressed the visual and sacral
aspects of imperial power in two trail-blazing studies that linked coinageand texts to trace the development of imperial ceremonies, insignia anddress styles; but he was not focusing principally on the late empire, norexclusively on court ceremonial, and until the 1960s late imperial court
ritual remained a topic usually left to historians of art or to Byzantinists. It
received no serious attention, for instance, in one of the great achievements
of modern scholarship on the late empire, A. H. M. Jones’ monumental‘social, economic and administrative survey’ of 1964; Jones’ view of the
historical importance of the matter is implicit in his crisp comment onthe innovations in court ceremonial introduced under Diocletian: ‘suchtriﬂes can hardly have made much practical difference’ .
14Butalready in the
early sixties some scholars were discussing the ritualised aspects of courtculture on very different presuppositions: there were papers of exceptionaloriginality by Ramsey MacMullen on the court’s formalised ‘bureaucracy-speak’ , and on the theatricality and sheer visual potency of late Romanceremonial; and Gervase Mathew, discussing the fusion of imperial ideology
12Brown 1992 :7–34 (quoting 25).13DFvol. II, 90.
14Jones1966 :29; cf. Jones 1964 :40.

164 rowlan ds m i th
and Christian symbolism in court ritual and art, linked it to the artistic tastes
and intellectual interests – the ‘Byzantine Confucianism’ , in his phrase – ofthe scholars who served in the upper levels of the late Roman bureaucracy.
15
Since then, inﬂuential works of the seventies and eighties have made the
‘representation’ of imperial power through formalised display a mainstream
subject for historians of the Roman empire. At one point in Bread and
Circuses Veyne went so far as to picture the entire City of Rome as mutating
in the early Principate into a vast ‘court for the emperors’ , a Versailles-like ‘shop-window’ in which ‘the courtier attitude’ shaped the outlook ofthe population at large, the plebs as much as ‘mandarin’ senators; thatparticular image is more provocative than persuasive, but ‘the power ofimages’ and of stylised behaviour to ‘construct’ and ‘represent’ imperialpower from the Augustan age onwards has become virtually an orthodoxyin the wake of Zanker’s work on Augustan (and more recently, Flavian) artand architecture.
16So too, the representational power of the late Roman
court’s ritualised behaviour and ceremonial has been illuminated through
studies of its panegyrics and sculpture;17few would now dispute that it
helped to shape a society ‘addicted to g randiose symbolic gestures’ in which
‘anacute rank-consciousness permeate d…s e lf-understanding’ .18
Ifthe modern consensus on this point owes a general debt to socio-
logical and art-historical approaches of the sort that inform the workof Veyne and Zanker, there is also, of course, a more particular socio-logical inﬂuence: Elias’ Die h¨ oﬁsche Gesellschaft .Since its publication in
English translation in 1983, Elias’ book has become a familiar point ofdeparture in Anglophone ‘court studies’ of late medieval and early mod-ern Europe; but even before the English version appeared, the modelof ‘court society’ that Elias adumbrated with reference to the Bourboncourt of seventeenth- and eighteenth-century France was being com-
mended to students of the Roman empire by Keith Hopkins, and it wouldlater be cited as a ‘fundamental analysis’ in an innovative account ofthe Julio-Claudian court by Wallace-Hadrill.
19Others are more scepti-
cal, it is true: the aptness of applying Elias’ model in the case of the lateRoman empire (or indeed, to any pre-modern state) has been disputed
20–
and so, for that matter, has the model’s intrinsic historical aptness evenin relation to Elias’ own chosen test-case, the court of Versailles.
21The
15MacM ullen 1962 (=1990: 67–77); 1964a (=1990: 78–106); Mathew 1963 :38–61.
16Veyne1990 :383–97 (Eng. trans. of 1976 French original); Zanker 1988 ,2002 .
17Panegyric: MacCormack 1981 ;sculpture: Smith 1985 and 1997 ;Kiilerich 1993 .
18McCor mick 1985 :14.19Hopkins 1978b :181; Wallace-Hadrill 1996 :285.
20Schlinkert 1996 :478–81.21Duindam 1994 ;2003 :7–11.

The imperial court of the late Roman empire 165
conceptual and methodological questions at issue have increasingly engaged
the attention of specialists in court studies in recent years, and by 2004 the
‘theorizing of the court’ was itself the dedicated subject of a colloquium at
which Elias’ model was discussed as a historical phenomenon.22For R oman
historians, the work done by Aloys Winterling in the later nineties wasalready pointing in this direction; besides writing a major study of thedevelopment of the Roman imperial court in the ﬁrst two centuries ADhe edited two collections of papers, one on institutional aspects of the late
antique court, another on ‘ancient courts in context’;
23and in one of his
owncontributions to the latter (essays on ‘ideal types’ and ‘comparative per-
spectives’) he adapted Weber’s typology to formulate a reﬁned ‘ideal type’of ‘the Court’ as a conﬁguration of power.
24
Itwas one of Elias’ central claims that the ‘absolutist’ court of the Bourbons
essentially functioned as an instrume nt for the ‘domestication’ of a feudal
warrior-nobility. But even if that held true for Versailles – and it is a claimwidely criticised by specialists in that ﬁeld
25–itcould hardly apply in the
case of the late Roman empire. In that case, the traditional senatorial aristoc-racy could certainly not be classiﬁed as eve nremotely a feudal warrior-elite;
indeed, on the supposition that the late Roman court served to ‘domesticate’elements of any ‘warrior-elite’ , that elite would be more plausibly associatedwith the new, non-aristocratic breed of ‘soldier-emperors’ that had emerged
in the mid to late third century (Diocletian himself being a notable instance).There are also certainly some notable aspects of court life omitted in Elias’account; he barely touched, for instance, on the signiﬁcance of the religionand the clergy of Versailles.
26Butperhaps his critics are inclined to ask too
muchof his model – and perhaps the rather abstract theoretical reﬁne-
ments and closely focused institutional studies of some ‘post-Elias’ courtspecialists do not always convey as clearly as Elias did the terms in whichlife at court was registered in the experience of contemporaries. Whetherhis picture of the Bourbon court is empir ically or methodologically cogent
is a question for historians of Bourbon France or for sociological theorists
22Duindam 2004 .
23Winterling 1999 (court in the ﬁrst two centuries); 1998 (late antique court); 1997b (‘court in
context’).
24Winterling ( 1997b :25;2004 :89) proposes a model under six broad headings: the process by
which different groups converge to ‘constitute’ a court; its ‘structures of communication’; its
roles in the formation of political policy and the representation of the monarch; the principles
on which those serving in intimate proximity to the ruler are recruited; the politicalsigniﬁcance of the court’s organisational structures, and of their connections with organisedpolitical groupings outside it; and the court’s role in the display of social rank in general.
25Winterling 1997b :12–13.26See now McManners 1998 :29–57.

166 rowlan ds m i th
–but whatever its conceptual shortcomings and empirical blind-spots, and
however doubtful its applicability to pre-modern social contexts, there are
nonetheless many features of Elias’ account of the Bourbon ‘court society’and its ‘sociogenesis’ that can hardly fail to strike students of the late Romancourt as suggestive, at least. The court that Louis XIV established at Versailles
reﬂected and promoted the ideology of monarchic absolutism in manifold
ways; it reconﬁgured the nobility in relation to the monarch, and ﬁxed itsranking order; it fostered a centralised state bureaucracy functioning in close
physical proximity to the court; its spatial conﬁguration and its ceremonialand pageants pictured Louis as a god-like Sun King, and his panegyristsprojected the image to a readership beyond the court in the orations theypublished. So too, the social and cultural values engendered at court rippledout to affect the outlook of a wider social elite (a key feature, surely, of a‘court society’ in any strong sense of the term). Transposed in art and lit-
erature, they helped to shape the representation of the age in the eyes ofcontemporaries – and before very long, its representation in a classic work
of historiography: when Voltaire wrote his history of ‘the spirit of men’ inThe Age of Louis XIV –‘the most enlightened that the world has seen’ –
he devoted three chapters in it to characteristic ‘incidents and anecdotes’in which the doings of the court ﬁgure prominently. Writers of Louis’ ownday could evoke the ‘Versailles attitude’ in various registers, grandiloquentand playful: in earnest poetic vein, ‘the age of Louis the Great’ is hailed asacultural match for that of Augustus; meanwhile, in Perrault’s fairy-tales,
aresourceful cat can prosper at court to win his master a princess’s hand
in marriage, and in La Fontaine’s Fables(a work dedicated to the Dauphin
in its ﬁrst edition) ‘sa majest ´elionne’ reigns unchallengeably in a court of
foxes, wolves and dogs, deigning to spare an unwittingly errant rat, or pro-nouncing his own claws too sacred to punish an unmannerly stag (it is a taskfor his wolves, he decides).
27On all these basic counts, we shall see, there
are counterparts to be observed in the case of the late Roman court. Eventhe folk-tales and fables of court life ﬁnd their ancient parallels: the wit andgood looks of a poor girl from Athens can charm an emperor’s heart andmake her his empress; a pauper’s gift of a giant apple sows discord at courtand wrecks an imperial marriage; an actress of surpassing beauty takes anempress’ brother as her lover, but then sees the light and becomes a holynun.
28Andwhether or not they assent to Elias’ particular model of the court,
27Fab.2.11; 7.7; 8.14.
28Holum 1982 :112–13 (‘Athenais’/Eudocia), 176–7 (apple of discord); Cameron 2000 :182–5
(actress).

The imperial court of the late Roman empire 167
it comes tellingly easily to historians of the late empire to draw comparisons
with the absolutist ideology and centralised bureaucratic administration of
seventeenth- and eighteenth-century France.29On the face of it, the late
Roman empire has the best claim of any ancient Greek or Roman state to be
construed as one in which power was conﬁgured and mediated through a
court whose scale and complexity can arguably bear comparison with Elias’
‘Versailles model’, and it is not unreasonable to think of the imperial court
of late Roman times as constituting a ‘court society’ in something like Elias’sense of the term.
Inwhat follows, then, it is not implied that Elias’ model is perfect or that
it is applicable to the late Roman case in all its features; it is simply thatone judges the similarities numerous enough and close enough to illumi-nate some key features of the late Roman case.
30Having sketched a basic
historical setting, and having outlined how the late Roman court tends to beviewed in modern scholarship (and how the views taken of it, ancient and
modern, may impinge on broader issues in the history of the late empire), wepass now to more detailed discussion. This discussion falls under three mainheadings: ﬁrst, a brief section on terms of deﬁnition and description, and onthe range of primary source material that historians of the late Roman courtcan exploit (pp. 167–71); next, a contextualising discussion of the socialand political factors that shaped the development of the late imperial court(in Elias’ terms, its ‘sociogenesis’) (pp. 172–87); then a discussion of somemore particularised aspects, a (highly selective) ‘anatomy’ of the court’sphysical, institutional and cultural faces (pp. 187–225). A brief closing sec-tion will offer some concluding observations in the light of the discussion(pp. 225–32).
‘The court’: late Roman terms of description and sources
of information
‘The late Roman court’ , we noticed earlier, is a generalising term for a com-
plex historical category. In English usage, a royal court can be variously
29E.g. Brown 1971 :42;1992 :13, 23; Veyne 1976 :343, 378, 449; MacMullen 1988 :99.
30Versailles is not the only case that could have been adduced, of course: we might as easily have
cited suggestive parallels, say, from the court world of the ‘shining princes’ of Heian Japan (see
Morris1964 ,abrilliant work); but as we have indicated, Elias’ work has made Bourbon
Versailles a familiar point of reference for historians of the Roman court – and it has a
particular resonance for the fact that Louis XIV’s own publicists and contemporaries imaginedthemselves subjects of a second Augustus, and took for granted a historical continuity of sortsbetween the ﬁfth-century emperors and the kings of France (Jones 2002 :2 ,6 ) .

168 rowlan ds m i th
construed: as a material object in a speciﬁc location (a distinctive architec-
tural complex within which the ruler resides); as a hierarchically ordered
human collectivity (the person of the ruler in company with his house-
hold, servants and ‘courtiers’ , wherever located); and as an institutionalisedsocial-political entity within which indi viduals and factions compete for
prestige and power, and through which the ruler and his ministers and advi-sors make and issue policy-decisions in his name. In late imperial Latinand Greek usage too, we shall see, the terms commonly used to refer to‘the court’ embrace all of these bands of meaning: basileion/basileia can be
Greek for palatium ,but it often implies more than a royal dwelling, and
aul¯e(and its Latin derivative aula)often denotes more than a hall; ‘those
around the aul¯e’means ‘court ofﬁcials’; ‘those powerful in the basileia ’a r e
grandees at court, the sort that Ammianus calls ‘holders of the ﬁrst place in
aula regia ’, o r ‘ habitu ´es of the summa aula’ ,orsimply ‘ potentes in regia ’.
31
Likewise, the consistory ( consistorium )was both the place in which elite
counsellors gathered in the emperor’s presence, and the collective name
for those counsellors as a corporate entity. An inner elite among them werestyled ‘consistory-companions’ ( comites consistoriani )o r‘ thecomites within
the consistory’ ( ILS1237); but all members of the consistory were also part
of a comitatus of court personnel in a broader sense. In ‘court-speak’ , the
comitatus also included the staffs of the imperial household and the various
ministries attached to the emperor’s person, and the soldiers of the imperialguard. Like ‘the court’ in English usage, then, comitatus often serves in Latin
todenote ‘the imperial court’ as a whole, when viewed as an institutionalised
human collectivity.
32
Although no author of the time offers a comprehensive description or
systematic account of the late imperial court as such, a great deal of infor-mation about it can be gathered from the extant historiographic and literarysources; in comparison to what is available for ancient courts in other peri-ods, the evidence is abundant. On top of that, there is a profusion of othertypes o ft e x t ual testimony, and the material evidence of art and archaeology.
31On the range of usage, Schlinkert 1996 :461; Eunapius fr. 67.1 Blockley (‘ hoi peri t¯ en aul¯ en’);
Lib.Or.1.94 (‘ dynatoi in the basileia ’); Amm.21.15.4; 16.8.11; 26.7.6.
32Forclarity, we may notice here an ambiguity in the term comes in late imperial usage. Comes
was not just a title held by high-ranking persons at the imperial court; it was also a title
attaching to a variety of administrative and military ofﬁces in the provinces, and it could begranted, moreover, as a purely honorary title; a man might thus be a comes without being a
member of the courtly comitatus .Acomes in this looser sense is to be classed rather as a
member of the general ‘Order of Companions’ , the comitiva ,within which he was ranked in one
of three ‘orders’ of status. But in practice, needless to say, there was very often an overlap: many‘comites of the ﬁrst order’ were ofﬁcials at the court, and hence also members of the comitatus .

The imperial court of the late Roman empire 169
Still, there are deﬁciencies in the evidence, and there are very few aspects
of the court and ‘court society’ of the late empire that one could hope todescribe in anything like the detail that historians can aim for in the caseof Bourbon Versailles, Elias’ chosen paradigm. For example, fourth-centuryarchaeological traces of two of the major palace sites chieﬂy at issue (Milanand Constantinople) are either nugatory or non-existent, and there are noancient equivalents of the archives of architectural drawings or the accountbooks detailing the occupancy of apartments within the palace complex thatexist for Versailles. As for the formalised daily routine of the emperor and his
retinue, it seems likely that a particular ministerial department, the scrinium
dispositionum ,was responsible for organising his timetable,
33and there are
scattered insights to be gleaned from our texts; Ammianus, for instance,remarks of the emperor Jovian that ‘he was anxious to imitate Constantius
[II], often occupying himself with serious business till after mid-day’ – fromwhich one infers that the afternoon was more usually devoted to rest andrecreation.
34InAmmianus, of course, we have an exceptionally valuable
witness, a historian writing expansively on the fourth-century empire with
personal experience of the workings of its court. So too in the speeches andletters of Julian the Apostate we can glimpse court life as it was seen throughan articulate emperor’s own eyes. There is also a large body of correspon-dence extant from persons who had dealings at one time or another with thecourt, sometimes close up (Ausonius, say, or Ambrose), sometimes from a
certain distance (Symmachus or Libanius). Nonetheless, one cannot attempt
adescription of the daily routine of the court remotely comparable in level
of detail to modern historians’ accounts of the punctiliously ordered hour-to-hour doings of Louis XIV from lever tocoucher –the twice-daily change
of clothes, the garden-walk, the hunt, the meals, the attendance at Mass, thesessions with councillors;
35no late Roman emperor was a memoirist after
the fashion of Louis XIV, and there is no real late Roman counterpart tothe voluminous ‘insider’ memoirs of a Saint-Simon at Versailles or a LordHervey(Pope’s ‘Sporus’ and ‘amphibious thing’) at the court of George II –
nor any, either, to the frank letters in which royal servants of Louis serving inadministrative posts away from court reported back to their court patrons.
36
Weare better placed so far as court culture, ideology and ceremonial are
concerned. Here the discrepancy in the quantity and type of evidence avail-
able is less marked and less vexing: an abundance of late imperial panegyric
33Jones1964 :578.34Amm.25.10.14, with Matthews 1989 :237.
35Elias 1983 :200; Jones 2002 :2–3; McManners 1998 :43.
36Brown 1992 :13, citing Beik 1985 .

170 rowlan ds m i th
is extant in prose and verse by authors quite the equals of Louis XIV’s Bishop
Bossuet in their hyperbolic praises of emperors as embodied expressions ofdivinity,
37and even texts that do not overtly treat court matters are often
revealing in this connection. The abbreviated summaries of Roman history
produced by Aurelius Victor and Eutropius, for instance, were works writtento win a ne m p e r or’s attention and favour by authors already ensconced as
bureaucrats at the court,
38and the opening sentences of each convey a lot
about the court’s cultural ethos: ‘In about the 722nd year of the city’ , Victorbegins, ‘the custom commenced at Rome of obeying one man alone’; ‘Inaccordance with the wish of your Clemency’ , Eutropius’ preface runs, ‘I havegathered a brief narrative . . . so that your Serenity’s divine mind may rejoicethat it has followed the actions of illustrious men.’ So too, even a Christiandevotional poem recounting a vision of heaven, it turns out, can inform ourunderstanding of court ideology and ceremonial.
39The visual evidence of
statue portraits and coin-images of emperors – interpreted with the requi-site methodological awareness
40–can enhance the picture (the minting of
gold coin, after all, became a monopoly of the court in the course of thefourth century);
41and for details of ceremonial it is possible also to draw
something from the Byzantine ‘Book of Ceremonies’ compiled by Constan-tine Porphyrogenitus – a tenth-century text, but one preserving some detailsfrom a sixth-century style-book by Petrus Patricius.
42
Asfor the administrative structure of the court and its bureaucratic appa-
ratus, there are two sources of outstanding value and importance: the Noti tia
Dignitatum and the Theodosian Code. The Noti tiaas we have it derives from
adocument compiled for a high-ranking ofﬁcial (the primicerius notari-
orum )connected to the consistory at the court of Milan in the early ﬁfth
century, and it essentially provides what its full title implies – ‘A Register
of All Ranks and Administrative Posts, both Civilian and Military’ in thelate empire. Some of the manuscripts transmitting it, moreover, includemany coloured illustrations of the ‘heraldic’ badges (the shield devices andinsignia) of the posts at issue, making it a rich source for ceremonial aswell.
43The Theodosian Code collects the laws and edicts (over 2,500 items)
issued by the emperors from Constantine to Theodosius II. The Code waspublished in 438 in sixteen books, Books One and Six of which collect
37MacCor mack 1981 ;Kantorowicz 1963 ;cf. Jones 2002 :3–4.
38Bird 1993 :xiii; 1994 :viii–ix.39Bremmer 1988.40Smith 1997 .
41Jones1964 :437.
42Av.Cameron 1987: 109–26; McCormick 1985 :4,for traces of earlier, ﬁfth-century ‘proc `es
verbaux’ .
43Kelly 1998 :163–5; Berger 1981 (insignia).

The imperial court of the late Roman empire 171
all edicts relating to state-ofﬁces and to the ranking of imperial bureau-
crats. Formally, it was compiled at Theodosius’ order, but what is knownabout the process of compilation is itself a testimony to the potent inﬂu-ence of the inner elite at his court, and appointment to the body oversee-ing the task was itself a mark of high honour – and a route to yet higherpromotion.
44Moreover, since the edicts were drafted by a leading court ofﬁ-
cial, the quaestor sacri palatii (an ofﬁce created by Constantine)45and were
then pronounced at meetings of the consistory, and since a date and placeare speciﬁed for each edict’s delivery, the Code also serves as our basic guidetothe whereabouts of the emperors and their court retinues at particular
points in time over a century and more (AD 313–438) – on which score,it shows beyond doubt that for most of the fourth century the court wasan itinerant body. The Noti tiaand the Code, then, provide a plethora of
evidence about the institutional face of the court; but the form in which
they provide it also raises problems of interpretation. It is often observed ofthe Code in general that where a particular matter recurs again and againin its laws, the behaviour they demand is probably more a pious aspirationthan any accurate reﬂection of what was actually happening;
46likewise, the
Code will to some extent distort a court reality messier than the impression
it may seem to project of a stable, hierarchical order overseen and sustainedbyimperial ﬁat: beneath its compilers’ systematised section-headings, the
diachronically ordered lists of laws in the Code disclose the court and itsbureaucratic structures in a complex process of development throughoutthe fourth century. It was only in the 370s, for instance, that a single com-prehensive ranking system universally applicable to the upper levels of allmilitary and civilian administrative ofﬁces became formalised in law, andwhat A. H. M. Jones observed of the order of precedence of ofﬁce-holdersin this connection could fairly serve as a more general motto for the late im-perial court’s administrative face: ‘it was immensely complicated andbecame progressively more so’ .
47
Noattempt can be made here to engage at length with the details of
the court’s evolution as an administrative entity: we shall attempt only toindicate the inﬂuences affecting the development of the late Roman court inits main lines (pp. 172–87), and then a selective description of its distinctiveformal features in terms that can convey something of its underlying social-political importance (pp. 187–225).
44Schlinkert 2002: 283.45Harr ies1988 :148.46Jones1964 :viii; MacMullen 1990 :67.
47Heather 1998a :188–9; Jones 1964 :534.

172 rowlan ds m i th
Elite reconﬁguration and the ‘sociogenesis’ of late Roman
‘court society’
Adiscussion that aims to contextualise the late Roman court with an eye to
Elias’ account of the court of Versailles can aptly open with a comparison
adduced by a leading Francophone sociologist-historian of imperial Rome.ToPaul Veyne, Constantine seemed in one respect a ﬁgure closer to Napoleon
than to a Louis of the ancien r´ egime :overhalf a century, between AD 260
and AD 310,
[a] new ruling caste [had] emerged, not from a political or social revolution [as
in France] but from a transformation of institutions and the army . . . From everystandpoint (including its literary culture), the [elite] political personnel of Constan-
tine’s time was as different from that of the early empire as the political personnelof nineteenth century France was from that of pre-Revolutionary times. So endedthe Hellenistic and Roman period of ancient history. Constantine meant to heapfavours on this new caste (like Napoleon creating barons and counts and makingthem wealthy). But [he] wished also to be reconciled with the Senate in the strictsense . . . (like Napoleon trying to reconcile the nobility of the ancien r´ egime ).
(V eyne1990 :449 n. 210)
Historians of the Roman empire (and of revolutionary France too, perhaps)
may think this a formulation in need of qualiﬁcation on various counts,
but few would dispute the basic premise that, notwithstanding any counter-vailing continuities, the composition of the fourth-century Roman imperialelite was markedly unlike that of the Julio-Claudian or Antonine elite. It isonly on this premise that it makes real sense to differentiate the ‘late Roman’court as a subject of historical discussion, for ‘court’ and ‘elite’ go closely
together: Veyne’s remarks imply a radical reconﬁguration of the court elite,
and construe it as an emblem – and maybe also an engine – of the broadershift from ‘high’ to ‘late’ empire.
Active initiatives on Constantine’s part were certainly central to this recon-
ﬁguration, but they need to be set against the T etrarchic background fromwhich the Constantinian dynasty emerged. Diocletian, too, we have stressed,was a proactive reformer, and his military and administrative innovations –not to mention the absolutist ideology of divine kingship promoted underthe T etrarchs – clearly set the stage for what followed. This point has par-ticular force in connection with court ceremonial; it is to Diocletian, weshall see, that the ancient sources tend to ascribe the introduction of a keyceremonial feature of the late imperial court, the so-called ‘adoration of the
purple’ . But in his case in turn there are continuities as well as innovations

The imperial court of the late Roman empire 173
toreckon with. T etrarchic ‘absolutism’ only ampliﬁed on a strand that had
been implicit in Roman imperial ideology from its Augustan beginnings,48
and Diocletian was certainly not the ﬁrst emperor to lay overt claim to status
as a god-emperor elected to rule by divinity, with all that that implies forthe formal intercourse between emperor and elite, and for the role of cere-monial as an enactment of absolutist id eology. As formal imperial titles, for
instance, dominus noster anddeus et dominus –titles that prompted moderns
tocoin the term ‘the Dominate’ to refer to the empire under the T etrarchs –
are quite frequently attested in pre-T etrarchic third-century usage; partic-ularly signiﬁcant, perhaps, is the precedent of the emperor Aurelian in the270s, who is advertised in his coinage as deus et dominus natus ,‘ b o r ng o d
and master’ (Alf ¨oldi 1970: 211), and reckoned (by a late source, admittedly)
tohave quelled a mutiny with the declaration that the soldiers erred if they
believed they governed the emperor’s fate, ‘for God alone could grant himhis position and decide the length of his reign’ .
49The same is the case with
sundry details of titulature and visual features in coin-images and portraitstatues that are attested for Diocletian and/or Constantine, and which aretaken in combination to mark a distinctively ‘absolutist’ late Roman styleof rule. Constantine, it should be granted, not Aurelian (the claims of lateliterary sources notwithstanding) was the ﬁrst Roman emperor to wear thejewelled royal diadem;
50but for many other titles and symbolic appurte-
nances there were clear precedents. The orb held by the emperor as anemblem of power over the world, for instance; or his claim to possess adivine comes (an especially piquant item for us, with its implication that
the emperor has recourse to a heavenly as well as an earthly comitatus );
or the routinised application of the epithet sacrum tothings pertaining to
the emperor and his doings, until in Constantinian ‘bureaucratese’ it haseffectively come to mean simply ‘imperial’ ( sacrae constitutiones for im-
perial legislative acts, sacrum palatium of the imperial palace, sacra largitio
of the imperial treasury, sacrum cubiculum of the imperial bedchamber, and
so on); even, perhaps, the ritual ‘adoration of the purple’ , or elements of it(see below, pp. 175–6) – each of these features ﬁnds pre-T etrarchic counter-parts in the mid to late third century.
51When Millar characterised that
period as ‘even on our inadequate evidence, one of the crucial periods inEuropean history’
52he was not thinking principally of the court, but if we
48Veyne1990 :292–306; Crook 1996 :113–23; Garnsey and Humfress 2001 :24.
49Petrus Patricius, fr. 10.6 (= FGH IV.197).50Alf¨oldi 1970 :267; Smith 1997 :177.
51Alf¨oldi 1970: 117–20 (orb); 1970: 216 (divine comes ;cf. Nock, 1972 :653–75); Alf ¨oldi 1970:
32–3 (routinised application of sacrum ); 1970: 53–9, with Stern 1954 :184–9 (‘adoration’).
52Millar1967 :248.

174 rowlan ds m i th
take an Eliasque view of the signiﬁcance of court ritual his judgement has a
special edge. In the symbolism of its ceremonial and titulature, at least, thelate Roman court clearly owes something substantial to third-century inno-vations in imperial style and ideology :ifadistinctively late imperial court
style can be said to have emerged under Diocletian and Constantine, then,it was constructed partly by a combining and reconﬁguring of pre-existingelements into a pattern that to some extent became systematised in theirreigns.
53
Toidentify third-century precedents for features of ideology and ritual
that show up in the fourth-century court does little in itself, however, toexplain why or how the ideology and ceremonial of that court took on
the shape it did; it really only shifts that question back in time. There willalmost always be continuities of a kind to observe, and on that basis onecould trawl further back again for ‘precedents’ of some sort till one ended
up, say, with the deus et dominus Domitian and his Domus Flavia .
54To g ive
explanatory force to the particular third-century precedents we have noticed,
one needs to look beyond them to their broader third-century historicalcontext, and to consider what background conditions and developments
could have caused an overtly absolutist imperial style, one that Tacitus or
Suetonius would have judged monstrous or mad, to come to appear quite
normal and proper features of court practice to a fourth-century eye. Thequestion is potentially a very large one: here it must sufﬁce to indicate certainkeydevelopments and to comment brieﬂy on their likely signiﬁcance. We
touch in turn on the emergence of ‘soldier-emperors’ in the third century; on
possible ‘oriental’ and Hellenistic inﬂuences on court style; on the politicalmarginalisation of the traditional senatorial elite; and on the rise of a non-aristocratic professional bureaucratic elite.
‘Crisis’ and ‘recovery’ were long the watchwords for historians of the
third-century empire. Over the last generation, the notion that a generalised‘crisis’ afﬂicted the empire overall in this period has been challenged on many
counts, but in the decades around the mid-century there was indisputably
amilitary crisis, and it is clearly more than a coincidence that imperial
ideology subsequently takes an overtly ‘absolutist’ turn in its emphasis onthe image of the emperor as soldier and guarantor of victory and its corre-spondingly diminished concern to maintain the style of the civilis princeps
53Only ‘to some extent’ , one ought to stress: in Eliasque terms, the ‘sociogenesis’ of late Roman
court society was a long-drawn-out affair that continued beyond Constantine’s reign, and
perhaps only crystallised with the emergenc eofConstantinople and Ravenna as permanent
imperial capitals and court residences in the late fourth and early ﬁfth centuries.
54Zanker 2002 .See Paterson in this volume.

The imperial court of the late Roman empire 175
as a senator-emperor closely identiﬁe dwith the traditional aristocratic elite
(see Paterson in this volume). We can note as particularly signiﬁcant on this
score the progressive decoupling of emperors from the social milieu of theRoman senatorial aristocracy, and ﬁnally from the city of Rome itself; early
in Diocletian’s reign, Rome was to cease to serve as an imperial residence.
55
From around the mid-third century, as is well known, many of the Roman
emperors were career-soldiers hailing from the Balkans who had risen topower through the ranks and will of the Danubian legions – emperors with-out family roots in the Roman senatorial milieu, and with less reason thantheir earlier counterparts had had to publicise themselves in conformity withthe old ideology of the civilis princeps ruling by senatorial consensus.
56The
intensiﬁed emphasis in their publicity on the symbolism of military author-ity and divine charisma was surely partly driven by their need to sustain thesupra-personal mystique of the imperial ofﬁce in symbolic terms that spoke
effectively to their key target-audience – the armies – and fostered a senseof awe and loyalty. The practical difﬁculty that ‘soldier-emperors’ faced onthis score is implicit in the proliferation and quick turnover of emperorsand pretenders in this period; they tended to rule a short time, and to die atthe hands of disaffected troops or in civil warfare against rival generals. Thiswould clearly seem a highly signiﬁcant background factor for our purposes:
the symbolism of military authority and divine charisma suffused the cere-monial and titulature of the T etrarchic and Constantinian courts, and bothDiocletian and Constantine’s own father, Constantius – and for that matterthe other two original T etrarchs also – were men from just the stable at issue:all were Balkan-born career-soldiers of low social status and obscure stockwho had won promotion through talent or luck.
57
Insome of its particular features, late Roman court ceremonial was con-
ceivably inﬂuenced by non-Roman cultural practices. A major constituent of
the military crisis of the third century was the rise of a new power in the eastlaying claim to the heritage of the Achaemenid empire, the Sasanians, and inthird-century imperial ideology Sasanid Persia became ‘the enemy’ par excel-
lence ,ap r ime focus of attention for Roman emperors in both warfare and
diplomacy (see Wieseh ¨ofer in this volume). It has often been suspected (it is
implicit in Rostovtzeff’s phrase ‘the Oriental Despotism’) that the symbolicelevation of late Roman emperors as socially remote, depersonalised beingsdisplaying themselves to subjects in motionless, imperturbable posture mayowemore than a little to an ‘oriental’ royal style encountered by the Romans
in this connection: Ammianus, after all, identiﬁed distinctive features of
55Potter 2004 :281.56Wallac e-Hadrill 1982 .57Barnes 1982 :30–8.

176 rowlan ds m i th
the court ceremonial and of the stylised costume and deportment of the
emperor – the ritual adoration of his purple robe of ofﬁce, and his wearingof jewelled clothing and footwear – as ‘Persian customs’ , and attributed theirintroduction among the Romans to a Diocletianic innovation.
58One might
even look for a stimulus in particular diplomatic and military episodes; in
387 (if the panegyrists are credited), Diocletian went to Syria to make atreaty with the Sasanian king which involved the latter prostrating himself
at Diocletian’s feet in a ritual proskynesis ,and in a campaign of 298 his junior
colleague Galerius captured the camp and household and much of the court
of King Narses – ‘his wives, sisters and children, a vast number of the Per-sian nobility and a huge amount of Persian treasure’ – and led them back intriumph to Diocletian.
59Admittedly, some moderns doubt the relevance of
the ‘Persian model’ ,60and Alf ¨oldi judged the hypothesis of direct external
inﬂuence from Sasanid (as opposed to old Achaemenid) Persia redundant:he traced the origins of the adoratio purpurae back to a ﬁrst-century blending
of a Roman tradition (the supplicatio )and Hellenistic royal ceremonial, and
argued on numismatic evidence that the ritual had already become ﬁrmlyestablished in the Roman imperial court at least twenty-ﬁve years beforeDiocletian’s accession.
61But Alf ¨oldi was probably misleading on that last
count, and in any event it is important to stress that, so far as the political
and social workings of the fourth-century court are concerned, the issue istodo not simply with factual realities but with fourth-century perceptions.
Whether or not ‘Persian customs’ were in fact a signiﬁcant inﬂuence on late
Roman imperial ceremonial and stylised deportment, it is telling in itself
that fourth-century sources talk as if they were; they look back on the reignof Diocletian as the point of entry of a new ‘despotic’ imperial style that theychoose to associate with foreign, Persian, practices.
62
Hellenistic court style was no doubt also important as a residual inﬂu-
ence, a sA l f ¨oldi supposed. The royal diadem that makes its ﬁrst appearance
on a Roman emperor’s head in coin-images of Constantine was modelledon the Hellenistic diadem;
63and from Augustus onwards, of course, cere-
monial ruler-cult rooted in a Hellenistic model was always an element in theself-representation of Roman emperors – an element that might be evokedovertly or by association, emphatically or ‘modestly’ , in different contexts. If
the ideology of civilitas had prompted ‘good’ emperors in the early empire to
58Amm.15.5.18; cf. Victor 39.2, Eutrop. 9.26.
59Pan.Lat .10.10.6, with Potter 2004 :292 n. 151; Eutrop. 9.25.
60Nixon and Rogers 1994 :51.
61Alf¨oldi 1970 :6–25, 45–73; cf. Kolb 2001 :39–40; but NB the response to Alf ¨oldi in Stern 1954.
62Matthews 1989 :244–5.63Smith 1997: 177.

The imperial court of the late Roman empire 177
affect modesty on this score, third-century soldier-emperors had less cause
toplay shy in their dealings with the senatorial elite, and features of cere-
monial once chieﬂy associated with provincial emperor-worship may easilyhave seeped into court ritual. A particular Hellenistic cultural icon arguablyspoke loudly in this connection. The image of Alexander the Great had ﬁg-ured in the repertoire of imperial self-publicity ever since Augustus gazedon Alexander’s mummy and chose to use his portrait head as his ofﬁcialseal,
64but the re-emergence of Persia as the enemy par excellence may have
encouraged third-century emperors and their publicists to take a sharperinterest in the ﬁgure of Alexander. Even before the rise of the Sasaniandynasty, the emperor Caracalla (211–217) was seeking to project himself as
aruler in the mould of Alexander; and by the 230s the formal titulature
of an African municipium could ﬂatteringly associate the reigning emperor
Alexander Severus, currently campaigning against the Sasanians, with hisgreat namesake.
65Inthe anonymous text extant as Alexander’s Itinerary we
have evidence of a lively interest in Alexander at the court of Constantius IIin the mid-fourth century: composed c.340, the Itinerary wasaw o r ko f
court literature dedicated to Constantius at a time when he too was ﬁghting
aPersian war; it measured his achievements against Alexander’s, and a good
case can be made for ascribing authorship of both the Itinerary and a near-
contemporaneous translation into Latin of the Alexander Romance to a
highly placed ﬁgure at court – Flavius Polemius, a consul of 338.
66Impe rial
publicists could likewise pick up on features of Alexander’s own court style –or on what was taken for such, on the strength of stories in the populartradition: Julian, for instance, wrote panegyrics to be delivered at court in
Constantius’ presence in which, like the author of the Itinerary ,heﬂattered
the emperor by comparing him to Alexander, notably for his generositytophiloi at court.
67Forthe philosophically minded, of course, Alexander
was hardly an impeccable model, and in the texts of moralists his desiretobe worshipped as a god had often ﬁgured as a topos for excessive pride.
Nonetheless, the image of Alexander could still have helped to disarm lin-
gering chauvinist suspicions that the ritual ‘adoration of the purple’ was anutterly alien ‘oriental’ practice unbecoming at the court of a Roman emperor.IfDiocletian, for instance, was indeed the emperor under whom the ritual
‘adoration of the purple’ was formalised, one wonders whether he would
have wished to be viewed (or to view himself) as importing a Persian custom
64Sueton ius,DA18; 50.
65Potter 2004 :142–4 (Caracalla); Gascou 1981 :231–40 on Giuﬁ (cf. Dio 80.18.3).
66Lane Fox 1997 :239–52.67Jul.Or.1.43c.

178 rowlan ds m i th
outright; it would seem unlikely, to judge by his excoriation of ‘the detestable
customs and depraved laws of the Persians’ in his celebrated Edict against
the Manichees .Alexander could offer a mediating precedent; the adoratio
plainly has a close connection with proskynesis –aP ersian custom originally,
but famously one that the Macedonian Alexander had taken up, and hence,in a sense, had ‘domesticated’ (see Spawforth in this volume). Alexander’scourt style, in any event, had more than one dimension that could be evoked
in imperial publicity; over against the conqueror who adopted Persian dressand demanded proskynesis as the son of Zeus-Ammon, there was a homelier
Macedonian side – the tough soldier-son of Philip who hunted and dined
at ease with a band of intimate philoi .This aspect may perhaps have been
in Julian’s mind when he set about purging the hypertrophied court of hispredecessor Constantius II. Julian was (on balance) an admirer of Alexan-der,
68not least because he had proverbially judged a ruler’s true wealth to
lie in his possession of true philoi (Jul.Or.2.86b), and as a junior emperor
in the 350s Julian had evoked the Alexander-image in his own publicity andcorrespondence; in his Self-Consolation on the Departure of Sallustius Julian
likened his own need for an honest herald to Alexander’s, and implicitlycast the friend departing from his western court as a loyal Hephaestion, andon one view a later coin portrait of Julian was modelled on a well-knowncoin-image of Alexander.
69
On the ideological plane, aggrandising features of court style could easily
be accommodated in the rhetoric of court panegyric and propaganda byrecourse to Hellenistic philosophic theories of kingship, as they later were
in the fourth century by Themistius and Synesius, and (with an overtlyChristian tilt) by Eusebius in his Tricennial Oration .
70So too, more gener-
ally, Greek intellectual currents of the period could impinge on educatedminds at court – the pagan henotheist strain that developed in third-century(neo-)Platonist ‘solar’ theology, for instance, with its emphasis on a tran-scendent unifying Godhead, depersonalised and far distant, working itseffects through a hierarchy of subordinate daemones .Quite what the lines
of relationship were between such philosophic ideas and the increasinglyabsolutist strain in imperial ideology and publicity is a matter we must leaveaside – but it is irresistible to observe that Plotinus himself enjoyed closecontact with the court of Gallienus, and that one of his star pupils, Cassius
Longinus, fetched up in Palmyra as a rhetor at the court of Odaenathus, and
68Smith 1995 :12–13.
69Alf¨oldi 1962: 403 (coin-image); Jul. Or8.250d–251c, with Smith 1995 :3 3( S a l l ustius’ departure).
70Euseb. Tric., esp 5.4–7; Baynes 1955 :168.

The imperial court of the late Roman empire 179
wrot eas p e e c ha b o u tt h e prince which was still being read in the late fourth
century.71
Wepass from questions of the court’s ideology and style to a more basic
matter: the progressive marginalisation of the traditional senatorial class in
army and administration in the course of the third century, and its implica-tions for the social composition of the imperial court. In the early empire, thesenatorial order had furnished the emperor with his generals and provincialgovernors and with the bulk of the amici and comites from among whom
he selected his advisory counsellors at court. The senatorial ethos of publicservice had been studiedly non-prof essional; senators were typically ‘all-
rounders’ who held both civilian and military ofﬁce in the course of careers
punctuated by long periods of residence at Rome as privati and habitu ´es
of the Palatine court. By the second century the expansion of equestrianofﬁces in the imperial administration had increased the pool of equestrianamici and comites from whom court counsellors could be drawn, but the
dominance of the senatorial class in the mixture had remained uncontesteduntil the reign of Commodus (180–192).
72The military and economic pres-
sures of the third century ﬁnally dissolved this pattern, further enhancingthe role of elite equestrians in the administrative and military branches, andmaking them a powerful presence at court: by the 270s, equestrian legateshad come to replace senators as legionary commanders and governors ofpraetorian provinces.
73The effective redundancy of the old senatorial elite
was underlined by the reforms of the T etrarchic age. Under Diocletian, theprovincial administration was reconstructed and the command structureof the army was further modiﬁed:
74the traditional provincial territories
weresubdivided into smaller units, effectively doubling the total number
of provinces to around a hundred, and these new units were subordinatedwithin a new hierarchical structure; sets of adjacent provinces were assigned
toone of a dozen ‘dioceses’ , with governors assigned at both levels, ‘vicars’ for
the dioceses, praesides in provinces, each with a substantial ofﬁce of support
staff – three hundred for a vicar, one hundred for a praeses .
75Inthe fourth
century, probably soon after Constantine’s death, a new administrative layer
would be established: the dioceses themselves were divided into three (and
later into four) units, each assigned to a praetorian prefect.76Atthe same
time, provincial governors were stripped of their former military functions,which were now transferred to a new order of provincial military ofﬁcials
71Rist 1967 :13; Potter 2004 :261.72Eck2000 :197–203.73Potter 2004 :229–32, 258.
74Jones1964 :42–9.75Jones1964 :592 (on fourth-century ﬁgures).
76Jones1964 :270.

180 rowlan ds m i th
(duces ). The net result was a vast increase in the total number of ofﬁces to be
ﬁlled in the military and provincial administration, the great bulk of which
weresimply no longer open to senators; military command was a profes-
sional preserve, and almost all of the new provinces were to be administeredbyequestrian governors. And there were repercussions, too, for the imperial
bureaucracy. As the number of military and provincial posts increased, sodid the number of bureaucrats needed to service them (and in the tetrarchicarrangement, furthermore, there were four emperors, each with his owncomitatus ). In the earlier third century there had been around 200 salaried
senior civil servants at most, equestrians working with the clerical assistanceof around 2,000 slaves and freedmen of the imperial household; estimatesof the total size of the bureaucracy in the fourth century, by contrast, put itat around 30,000–35,000, of whom perhaps as many as 6,000 were holdersof prestigious ‘upper-level’ posts.
77The proportionate increase is immense,
and an obvious question arises: how were the professional personnel of thisgreatly enlarged bureaucracy to be integrated and status-ranked in the social
hierarchy of the late empire?
Against this background, the exuberant development of the fourth-
century imperial court and its attendant bureaucracy – the sheer increase in
the number of personnel, the elaboration and systematising of ceremonialand protocols, the complexity of its hierarchy of ofﬁces – can plausibly beviewed as central both to the political and social integration of a new pro-
fessional elite and to the forging of a rapprochement of sorts between thesoldier-emperors and the traditional ideology of civilian society.
78‘Every
form of rule’ , Elias maintained, ‘is the precipitate of a social conﬂict: [itconsolidates] the distribution of power according to its outcome, [and] the
moment of consolidation . . . has a determining inﬂuence on its speciﬁcform.’
79Inthe case of the late empire, the long reign of Constantine is a key
period of consolidation, and for our purposes it would be hard to overstatethe importance of three Constantinian developments in particular.
First, Diocletian’s effective separation of the military and civilian career
ladders in the imperial service was completed and institutionalised underConstantine.
80Generals were from now on almost exclusively men of purely
military background appointed from the junior ofﬁcer ranks – and, increas-ingly, of Germanic ethnic extraction. The attention of the educated urbanclasses, and of at least some of the elite senatorial class, now ﬁxed on the im-perial court, and on ofﬁce-holding at one level or another in the departments
77MacM ullen 1988 :144, 264; Heather 1998a :189.78Matthews 1975 :47–54.
79Elias 1983 :146.80Jones1964 :101.

The imperial court of the late Roman empire 181
of its bureaucratic apparatus, as prime means of advancement; a successful
career in the bureaucracy could gain a non-senator senatorial status, not tomention wealth. Nor could those already possessed of senatorial status andwealth in abundance afford to ignore the court if they entertained politi-
cal ambitions; proximity to the emperor offered inﬂuence and powers ofpatronage, and appointment to the ranks of the comites of the consistory
and to the high ofﬁces at the pinnacle of the bureaucracy.
81The jargon and
trappings of the bureaucracy, it is telling, continued to evoke the aura of
the military units in which the senatorial and equestrian elites had formerlyserved as ofﬁcers and generals: departmental posts were formally styledmilitiae ,military ofﬁces, and those who occupied them were all technically
soldiers; their pay was received (or at least calculated) in kind, as ‘rations’and ‘fodder’ ( annonae and capitum ); ofﬁcials wore ‘uniform’ ( vestis )–a
military cloak ( chlamys )and belt ( cingulum )–and sported embroidered
patches ( segmenta )ontheir tunics as badges of ofﬁce; on entry to their
posts, they had their names added to the strength of a ﬁctive military regi-ment.
82Inpractice, though, the distinction between real military service
(militia armata )and a civilian ofﬁce in the bureaucracy ( militia ofﬁcialis )
was clear and sharp. T o gain entry to the latter, one used whatever means onecould; a decent level of education was assumed, and legal expertise could help
especially. But so could patronal inﬂuence, and so could cash: purchase ofofﬁces is very well attested for this period, and indeed became normalised.
83
This is as one would expect: in Weberian terms, we have observed, the lateRoman bureaucracy remained a ‘patrimonial’ bureaucracy – a system still
closely linked to the emperor’s household and court in which patronage andfavouritism could still trump criteria of seniority or ‘merit’ . Court servicepatently served as a key avenue of social mobility; ambitious men were will-ing to pay not just for an immediate appointment to ofﬁce at court, but eventohave their names, or those of their children, entered as supernumeraries
on waiting lists for posts years before they became vacant.
84
Second, Constantine created a new ‘Order of Imperial Companions’ – a
comitiva whose members were hierarchically ranked internally as comites of
the First or Second or Third Order.85The title comes was by no means new
in itself – it had long been used to designate the amici whoaccompanied the
emperor on his travels86–but the formalised subdivision into three grades
was a major innovation, and the rank was from now on ofﬁcially conferred
81Jones1964 :557.82MacM ullen 1990 :95–100; Kelly 1998 :168; Jones 1964 :566.
83Ste Croix 1954 :38; Veyne 1981 :339.84Jones1964 :570; Heather 1998a :196.
85Jones1964 :104–5; Scharf 1994 .86Eck2000 :206–7.

182 rowlan ds m i th
bycodicil. Moreover, as we have noted earlier, in the fourth century the
title of comes came to be awarded often as an honorary grant to men who
were not actually members of the emperor’s comitatus ,and who might
not be holding an ofﬁcial post at all. Constantine’s creation of this newly
formalised category of imperial comites can be viewed inter alia as a means
of reconciling and integrating the old senatorial aristocracy and the newbreed of professional administrator: the title could be bestowed on bothsenators and high-ranking equestrian ofﬁcials in imperial service; as a new,‘third’ order of nobility that overlapped with the established senatorial and
equestrian orders, it became a status-category to which members of boththose orders (and, for that matter, men who were members of neither)could aspire.
87Italso advertised the centrality of the court, of course, and
the practical importance of access to its ‘high authorities’ (Amm.28.6.9),either directly or through a court patron. Inasmuch as the rank of comes
was likely to be granted for what the emperor viewed as loyal and usefulservice, it illuminates a basic choice facing Rome’s traditional senatorialaristocracy in the fourth century:
88either to keep one’s direct contact with
the court to a minimum and to rest content with a life of otium and local
dignity in a historical ‘show-piece’ of a city that was rarely even visited bythe emperor, and fast becoming a political backwater; or else to accept thathigh birth now counted for less than it once had, and to make for the courtand seek high ofﬁce and advancement in imperial service. It was a choice towhich Symmachus frequently alluded in his letters. Symmachus himself wastemperamentally inclined to the former path; he was awarded the honorary
title of comes ,admittedly, when he visited the court of Valentinian I at Trier
on the senate’s behalf, to present the emperor with a gift of gold to celebratethe ﬁfth anniversary of his reign, but he remained only a ‘ comes of third rank’ ,
and never sought or held an ofﬁce in the imperial bureaucracy. His near-contemporary Petronius Probus, by contrast, the head of the great house
of the Anicii, followed the other path: ‘more at home amid the intriguesof court than the salons of Rome’ , he became ‘the greatest aristocrat andcourtier of his time’ , winning the praetorian prefecture four times over –
and immense riches in the process.
89
Third and last, Constantine set in motion a process that over the course of
the fourth century would utterly transform the senatorial order itself, bothin the numerical strength and the social composition of its membership,
87Jones1964 :106.88See Matthews 1975 :9–31.
89Cameron 1985 :164; Matthews 1975 :17, 187.

The imperial court of the late Roman empire 183
and in its basic relationship to the imperial court and bureaucracy.90Var-
ious Constantinian measures contributed to produce this transformation.
The most signiﬁcant, in the long term, was the practice of awarding senato-rial status as of right to numerous categories of equestrian ofﬁce-holders on
appointment – to all provincial praesides ,all diocesan ‘vicars’ and all praeto-
rian prefects, for instance – or when they retired; likewise, it also came to be
awarded routinely to retiring military ofﬁcers down to, and including, the
rank of tribune. Then there was the opening up again to hereditary senators
of some of the administrative posts that Diocletian had reserved exclusivelyfor equestrians. And on top of that there was the creation of a new im-perial senate for Constantine’s newly founded capital in the east. Aroundthe mid-century, the senate of Constantinople was raised to parity of dig-nity with Rome’s, and senators resident in the eastern provinces were trans-ferred from the Roman to the Constantinopolitan senatorial roll: whereasthe entire senatorial order under Diocletian still numbered around 600, asit had in Augustus’ day, by the 380s there were about 2,000 Constantino-politan senators, and probably as many again at Rome.
91Senatorial status,
moreover, remained hereditary, so growth would continue exponentially astime passed. This massive expansion was comparable to what had alreadyoccurred earlier in the case of the (non-hereditary) equestrian order, whosenumbers had been increased many-fold by the appointment of commoners
toaplethora of Diocletianic posts that conferred equestrian status on the
holder, and by retrospective grants of honorary equestrian status to rewardloyal service.
92The scale of increase in the case of the senatorial order,
and the transformation of the relationship holding between it and the im-perial bureaucracy, can be appeciated if we collate ﬁgures from the mid-thirdcentury and the start and end of the fourth; around AD 250 there were prob-
ably about 200 top-ranking bureaucrats, all equestrian; under Diocletian,wehave just noted, there were perhaps 600 senators overall, again effec-
tively excluded from the ofﬁces of the bureaucracy; by around AD 400, bycontrast, there were nearly 6,000 high-ranking posts in the imperial bureau-
cracy conferring senatorial status on the holder, either on appointment oron retirement.
93
Overall, then, both of the traditional aristocratic orders, senatorial and
equestrian, became grossly enlarged in the course of the fourth century.The predictable result was status inﬂation. In the case of equestrians, onlyenrolment in the inner elite category of perfectissimi now conferred any real
90Forwhat follows, see Heather 1998a :184–204; Potter 2004 :386–91.
91Heather 1998a :187.92Jones1964 :526–9.93Heather 1998a :190.

184 rowlan ds m i th
prestige; and as the holding of senatorial status became more widespread,
the lustre even of the equestrian ‘perfectissimate’ declined. So too with thesenatorial order itself: as possession of senatorial status per se became more
common, its prestige diminished, and a formalised hierarchy developed
within the order – in ascending order, the senatorial ranks of clarissimus ,
spectabilis andillustris ;overtime, effective membership of the senate was to
be conﬁned to the top-ranking band of senatorial illustres ,and the prestige
and privileges associated with the lower two bands were reduced. The broadeffect of this internal banding was ‘to transform the aristocracy from oneof birth to one of ofﬁce’;
94senatorial status as such was still hereditary, but
ahereditary senator, even if he was the son of an illustris ,only inherited
the rank of a clarissimus ;inorder to obtain the rank of a spectabilis or an
illustris for himself he needed either to win appointment to an ofﬁce in the
imperial bureaucracy that conferred it automatically, or else to be awardedthe rank by the emperor’s special grant. Either way, an entr ´ee at court, or the
help of a well-placed patron there, was normally a sine qua non .Inthe case
of the new imperial senate at Constantinople, the lesson was particularlyclear: its rapid growth from the mid-fourth century onwards discloses it asincreasingly an epiphenomenon of the imperial court and central bureau-cracy. Unlike its Roman counterpart, the new senate in the east was linkedtothe court from the ﬁrst, and the links became closer as Constantinople
developed into a permanent imperial residence and administrative capital:the emperor’s high ministers, top generals and elite comites ranked as illustri
in Constantinople’s senate; subordinate court ofﬁcials ranked as spectabiles
orclarissimi ,either while still in post or by grant of these statuses on retire-
ment. So too, in some circumstances, appointment to the new senate couldease the way to preferment at court. When Constantius II wished to enlargethe senate of Constantinople in the 350s, he looked especially for recruitsfrom the old propertied classes of the eastern cities;
95part of the attraction
of membership for men of this background was the immunity from localcurial burdens that it conferred; but once enrolled as ‘backbench’ clarissimi ,
they were well placed if they wished – and many were – to tap into thenetwork of court patronage and seek posts in the palatine ministries or asprovincial administrators. Like Constantine’s new ‘Order of Companions’ ,then, the transformed senatorial order of the fourth century undercut theolder status-categories, drawing its membership from a much broader bandwithin the propertied classes of the empire, and stimulating the development
94Jones1964 :529.95Heather 1998a :187–8.

The imperial court of the late Roman empire 185
of a new, court-dependent, ‘aristocracy of service’ .96Ifthe fourth-century
court was permeated by an acute rank-consciousness, it was also a power-
ful vehicle for social mobility. The poet Ausonius offers a striking example
of the way it could transform an individual’s political and social horizons:aGaul of respectable but quite modest family background, Ausonius had
prospered quietly for decades as a we ll-to-do professor of rhetoric in his
native Bordeaux before his reputation earned him the post of tutor to thefuture emperor Gratian at the court of Valentinian; once established there,he could strike up friendships with the likes of Symmachus, and he went onto beco me a comes as ‘Quaestor of the Sacred Palace’ , ‘the arbiter of petitions
and creator of laws’ , and ﬁnally a praetorian prefect and a consul.
97The case
of Ausonius is in some respects a special one, and luck played its part in hismeteoric rise, but it exempliﬁes a broader trend: ‘the real story of the fourthcentury’ , it has recently been asserted, lies in ‘the rise of the [new] senatorial
bureaucrat and the adaptation of the old aristocracy to new conditions’;moreover, since the court-centred bureaucracy drew many of its personnelfrom a class of men who would otherwise have tended to pursue local careersin their home cities, the transformation of the senatorial order ‘representedapolitical revolution not only at the imperial centre … but also in the
localities’.
98
Inasmuch as a sound education was normally a prerequisite for a good
career in the imperial bureaucracy, social mobility had its limits: the bulk ofthe ofﬁce-holders who won entry into the senatorial order through serviceat court would still come from the class that could afford to pay for a liberaleducation for its sons – around ten years’ worth of professors’ fees (Ausonius’family had managed that – with difﬁculty). All the same, if Libanius’ testi-mony is credited, service in the palace secretariat had on occasion launchedmen of much lower birth than Ausonius into spectacularly successful careersin the ‘aristocracy of service’: Flavius Philippus, who rose to hold two prae-torian prefectures and the consulate under Constantius II, and to be publicly
commended by that emperor as ‘Our parent and friend . . . [a great man]
celebrated on the lips of all civic communities and of diverse nations’ ,
99was
reputedly the son of a sausage-maker; so too, Datianus, whose father had
been a cloakroom-attendant at the public baths, ended up a top-rankingcomes of Constantius, with patrician status and a consulship to his name.
100
These are extreme cases and perhaps arise out of a particular mid-century
96Heather 1998a :191.
97Symm. Ep.1.23; Hopkins 1961 ;Matthews 1975 :51–5, 69–72; cf. MacMullen 1964b .
98Heather 1998a :209, 197.99Swift and Oliver 1962 :247–8.
100Lib.Or.42.23–5; Jones 1964 :129; Matthews 1975 :41.

186 rowlan ds m i th
political context; the impact of social mobility on the inner aristocracy at
court arguably peaked around that time, and was decreasing again by the
late fourth and ﬁfth centuries.101Butthey can still serve as an emblem of
the broad change in the social composition of the imperial court elite thatprompted Veyne, in the provocative passage we cited at the start of this sec-tion, to speak of the reign of Constantine as ‘the end of the Hellenistic andRoman period of ancient history’ .
Indiscussing the transformative impact of Constantine’s measures on the
fourth-century court, we have said nothing of his Christianity. The omis-sion may seem an odd one: Constantine, after all, did engage in high-proﬁleChristian devotional observances at the palace (a point on which we shallhave more to say later), and Eusebius claims that he ‘offer[ed] up regularprayers with all the members of his court’ , ordering the observance of theSabbath day;
102and Christian bishops, for their part, were clearly well aware
of the practical need to cultivate contacts and maintain agents at court as achannel of access to the ruler – and quick to exploit the possibilities in the
wake of Constantine’s conversion. ‘Bishops at court’ are a familiar presencefrom the beginning of the reign,
103and the court is the backdrop to some of
the most celebrated episodes that bear on Constantine’s engagement withthe clergy: the opening session of the Council of Nicaea was convened in thejudgement hall of the imperial palace, with the emperor presiding, bejew-elled, diademed and clad in purple, and after the business of the Councilwas done the bishops were entertained at a palace-dinner so magniﬁcentthat Eusebius was put in mind of the paradisal feasts of Christ and hisapostles; so too, Constantine’s much-discussed claim that he himself was a‘general bishop’ outside the Church was made at an imperial dinner party.
104
Nonetheless, our omission of the topic from the discussion has its logic, at
least as far as the fourth century is concerned: notwithstanding the fact thatevery fourth-century emperor from Constantine onwards except for Julian
professed the Christian religion, there is no real sign that Christian ideol-ogy or practice exerted any signiﬁcant formative inﬂuence on the imperialcourt’s ceremonial or administrative structures. At Louis XIV’s Versailles
there were around 200 ecclesiastical posts, chief among them that of thegrand aumˆ onier ,‘almost as ancient as the monarchy itself’ , an ofﬁce which
conferred on the holder by virtue of appointment the rank of commander in
101Heather 1998a :196; Jones 1964 :550–1.102Euseb. VC4.17.
103Hunt1998 :238–40; Drake 2000 ,chs. 6–8 passim ;cf. Kolb 2001 :70–2.
104Nicaea: Euseb. VC3.10, 3.15, with Barnes 1981 :315, 219; Const. as ‘ koinos episkopos ’: Euseb. VC
4.24, with Straub 1967 :37–56.

The imperial court of the late Roman empire 187
the order of the Saint-Esprit, the grandest of all France’s chivalric orders.105
There was no late Roman equivalent: no prestigious religious ofﬁce was
created at court to be held by eminent churchmen, and emperors did notyetneed priests to crown them. Insofar as there was a high priest at court,
the ‘sacred’ emperor himself fulﬁlled the role; it is well known that untilthe 380s the Christian emperors continued to take the traditional title ofpontifex maximus
106–and for a good century after that, deceased emperors
were st ill formally styled divi.A sf o rt h ee lite at court, the proportion of
individual Christians among them will clearly have been growing through-out the period; but construed as a collective group, the new governing elite‘owedlittle or nothing to Christianity’ in its conﬁguration and its ceremo-
nious practices and culture.
107The court elite, that is to say, still cohered as
aclass around a body of traditions and rituals which had developed directly
from a non-Christian imperial past – and the continuing power of those tra-ditions to affect the social and cultural outlook of Christian as well as pagan
subjects is witnessed not least by the very lines along which Christian artand architecture and liturgy developed in our period: ‘in the stratum of thecourt’ , it has been judged, ‘the architectural vocabulary remained anchored
ﬁrmly in the classical tradition, [and] emperors, courtiers, court bishopsand court architects remained the mainstays of the classical heritage of theChristian empire far into the ﬁfth century’.
108
Descriptive particulars: the late Roman court and
its workings
Wepass now from discussion of the context within which the late impe-
rial court took shape, and the conﬁguration and social composition of the
fourth-century court elite, to more particular aspects of the court and its
workings: ﬁrst, the physical setting of court life (pp. 187–96); then the organ-
isation and behaviour of its personnel (pp. 196–209); then its role as a theatrefor formalised self-display and royal ceremonial (pp. 209–25).
The court as a physical entity: late imperial ‘palaces’ and residences
‘The royal residence’ , a writer of the early third century observed, ‘is called
Palatium [‘palace’] not because it was ever decided that this should be so, but
105McManners 1998 :39.106Cameron 1968 .107Brown 1998 :651.
108Krau theimer 1986 :68–9 (quoted); cf. Grabar 1967 (iconography); Shepherd 1967 (liturgy);
Brown 1995 ;Salzman 1990 .Admittedly, the older scholarly consensus on this point has been
challenged recently – but hardly decisively (see Brown 1995 ,r e v i e w ing Mathews 1993 ).

188 rowlan ds m i th
because the emperor lives on the Palatium [i.e. the Palatine Hill at Rome]
and has his headquarters there. His house also gained somewhat in prestigefrom the hill itself, for Romulus had previously dwelt there; for this reason,if the emperor resides anywhere else, his stopping-place receives the nameofPalatium .’
109Under the T etrarchs, we have noticed earlier, Rome ceased
to serve as a nimperial residence, and if one excepts the few years when the
usurping son of a tetrarch, Maxentius (AD 306–312), based himself there,and Constantine’s three-month sojourn after his victory over Maxentius atthe Milvian Bridge, this remained the case throughout most of our period(Constantine only twice came to Rome thereafter, on brief visits
110). In Dio’s
sense of ‘stopping-places’ , though, palatia (in Greek, basileion/basileia )w e r e
scattered far and wide across the empire; thus Eusebius ( VC.1.22.2) could
speak of Constantine’s father dying in a basileia at Y ork in 306 (a build-
ing in, or close by, the legionary fortress, one assumes); and Julian on hismarch out east from Antioch on his Persian expedition of 363 could writein a letter of his stopping over at Batnae in a modest basileia made of wood
and clay.
111Butbythe early fourth century, linguistic usage was shifting;
rather than simply denoting an emperor’s headquarters, wherever he hap-
pened to be, palatium normally implied an actual imperial building, whether
or not it was currently occupied by the emperor. The shift indirectly dis-closes the emergence of a distinctively late Roman court, for the word thatcame to replace palatium/basileia to de note the imperial headquarters in
an institutional sense is comitatus ,aterm indicative of the migratory and
militarised character of the court. Its counterpart in Greek, stratopedon (lit-
erally, a ‘military camp’), continues in usage well after the founding of Con-stantinople as an imperial capital: thus Constantine, ordering councillorsfrom Tyre to Constantinople, can summon them ‘to the stratopedon of our
Clemency’ .
112
Palatium/basileia ,meanwhile, comes to be applied particularly to im-
perial residences in half a dozen or so cities of special importance: Trier inthe Rhineland; in Italy, Milan especially (the ﬁrst ‘capital’ to be spoken of as a
rival t oR o m e
113); Serdica, Sirmium and Thessalonica in the Balkans; Nico-
media – and later, of course, Constantinople – on the Bosporus; Antioch inSyria. The geographical disposition again marks out military considerations
as a driving factor, and the broad background to this development clearlylies with the Diocletianic creation of the arrangement under which several
109Dio 53.16.5; see Millar 1977 :30; also Paterson in this volume.
110Barnes 1982 :72, 77.111Jul.Ep.58 (98 Bidez) 401a.
112Millar1977 :42–3, citing Athanasius, Apol.Sec .86.
113Pan.Lat .III.11(AD 290/1).

The imperial court of the late Roman empire 189
emperors ruled simultaneously in allocated regional sectors of the empire,
each tetrarch residing with his own comitatus in camps or cities in his allot-
ted reg ion.114Most of these cities continued to function as regional imperial
residences or ‘capitals’ long after the T etrarchic age; fourth-century emper-
ors remained itinerant rulers by military necessity, engaging with barbariansin the Rhineland and the Danube regions and with Sasanians in the east,and, as we noted earlier, there was usually more than one emperor rulingat any given time in our period, each with his own regional responsibilities,‘palaces’ and comitatus .Itwas only in the late fourth and early ﬁfth cen-
turies that certain cities emerged clearly as permanent imperial residencesand administrative capitals for the eastern and western sectors of empire:in the east, the ‘New Rome’ , Constantinople; in the west, Milan (and then,from 402 onwards, Ravenna in its stead). The ceremonial and social andadministrative structures of the court could hardly have been unaffectedbythis development. It is surely signiﬁcant, for instance, that the Byzantine
tradition of composing treatises and sty le-books on court ceremonial begins
in the ﬁfth century.
115The stable landscape of the Great Palace and its sur-
rounding architecture at Constantinople would encourage a formalisation
of ritual practice; when ﬁfth-century emperors returned there from a jour-ney away, court ofﬁcials gathered at speciﬁed locations to welcome themback amidst chanted acclamations; particular routes were established forimperial processions from the Great Palace to the Great Church (the ﬁrst StSophia) and to the state granaries; the Hippodrome adjoining the Palace andthe Hebdomon, a military complex in the European suburbs, became primestages for imperial self-display (at the Hippodrome, the ceremonial pompa
circensisrecalled the ceremonial inauguration of Constantine’s ‘New Rome’
on 11 May 330; at the Kampos, a parade-ground in the Hebdomon, theemperor accompanied by his retinue of courtiers mounted a statue-girdedplatform 5 metres high to be acclaimed by his troops on his accession).Even the departures of emperors to the suburban palatia and to the nearby
ruralestates to which they withdrew to relax became occasions for proces-
sions (according to a catalogue prepared c.430, there were already by then
four other imperial palatia at Constantinople besides the Great Palace, and
several smaller imperial ‘mansions’ ( domus )).
116Itis all the more vexing,
then, that there are virtually no extant archaeological traces of the spatiallayout of the fourth-century palaces at Constantinople or Milan, or of the
114Barnes 1982 :47–87, listing known tetrarchic residences and stopping-places.
115McCor mick 1985 :4 .
116McCor mick 2000 :137, 157–9; Seeck 1962 :229–43; Noethlichs 1998 :22–6.

Figure 5.1 The ‘palace’ of Diocletian at Split (artist’s reconstruction)

The imperial court of the late Roman empire 191
early ﬁfth-century palace at Ravenna.117Only at Ravenna (and only on one
recent view) are there sparse material traces hinting at a complex adapted
from a military base within a perimeter wall, and at an enlargement of the
palace buildings under Valentinian III (AD 425–455).118
Todraw any picture of the inﬂuences and purposes at play in the archi-
tectural conﬁguration of these complexes that goes much beyond hints in
literary testimonies, we have to proceed by analogy from other palace sites.The best-preserved complex of likely relevance (admittedly, it was built inanon-urbanised setting as a retirement residence for a ruler who had abdi-
cated, rather than as the palace of a still-active ruler) is the seaside ‘Palace ofDiocletian’ at Split (Figs. 5.1and 5.2).
119The model of a military fort ( cas-
trum )and its commanding ofﬁcer’s administrative and residential quarters
(praetorium )isclearly inﬂuential in this case: the complex occupied a rectan-
gular area enclosed by a circuit of walls with protecting turrets, and dividedbyaxial colonnaded streets running east–west and north–south from three
monumental gates on the north, south and west sides. The northern halfof the enclosure contained service and military buildings; in the southwardhalf there were two large rectangular precincts, one housing the emperor’sownmausoleum, the other a temple, and beyond them, in the southernmost
sector, secluded and furthest distant from the gates, the residential palace-wing itself. The central area of this wing, accessible via a circular vestibule,
contained three large rectangular halls which presumably served as public
rooms. Of these, one of the two ﬂanking halls is apsed and seems likely to
have served as a basilica (audience-hall); the other was perhaps a triclinium
(dining-hall); beyond these, extending to the western and eastern extremitiesof the wing, lay the domestic rooms and bathing suites; and running alongthe entire southern edge of the complex there was a terraced promenadelooking directly out to se a–af e a t u r ep r e ﬁguring the outlook from Con-
stantine’s Palace at Constantinople.
120Details of design at Split, particularly
on the principal (northern) monumental gate, have often been construedtosuggest that eastern (Syrian) as well as Roman architectural inﬂuences
wereat play, perhaps echoing the style of palaces built earlier for Diocletian
in the east: there was one such at Antioch (a structure known only from
117The palace at Milan, ﬁrst developed as the residence of Diocletian’s colleague Maximian, is
only known from texts; at Constantinople the overall site of Constantine’s Great Palace islocatable, but later building has completely obscured the picture for the fourth century.Krau theimer 1983 :69–71 (Milan), 49 (Constantinople).
118McCor mick 2000 :136–7, citing Deichmann, 1989 .
119Ward-Perkins 1981 :454ff; Wilkes 1986 ;cf. Duval 1997 :143–7 (doubting its value as a model).
120Dagron 1974 :93.

192 rowlan ds m i th
Figure 5.2 The ‘palace’ of Diocletian at Split (plan)
adescription in Libanius), and another at Nicomedia which had become
established as his principal residence before his abdication in 305.121
From Diocletian’s time until the inauguration of Constantinople in 330,
Nicomedia remained the prime ‘capital’ of the eastern empire. In this case,
too,the palace is known only from texts – but they offer vivid glimpses of
the physical setting of court life there. An entry in the Suda discloses the
emperor Licinius (AD 308–324) relaxing beside a fountain in a courtyard
garden in the palace, attended by a retinue of courtiers and bidding one ofthem to cut down a bunch of grapes.
122Apassage in Lactantius, who had
121WardPerkins 1981 :458; Millar 1977 :50–1; Duval 1997 :140–1.
122Suda,s.v. ‘Auxentius’; for the Christian twist to this story, see Millar 1977 :52.

The imperial court of the late Roman empire 193
lived at Nicomedia in Diocletian’s reign, indicates the impact of the new
palace complex on the urban environment around it, and the ideology thatunderpinned its development: ‘Here there were basilicas, here a circus, amint, an arms factory, a house for his wife, and one for his daughte r…
such was [Diocletian’s] incessant mania for making Nicomedia the equalof Rome.’
123AsaChristian, Lactantius was a hostile witness in this case, of
course; more often, the tone of such remarks is laudatory. For instance, a
panegyric of 310 delivered in praise of Constantine at Trier dwells on theemperor’s generosity to the city: ‘I see a Circus Maximus to rival, I believe,that in Rome, I see basilicas and a Forum, royal buildings ( regia opera ),
and a seat of justice – I see these being raised to such a height that theypromise to be worthy neighbours of the sky and stars.’
124Here,atleast,
wehave a surviving structure to set beside the text, an exceptionally well-
preserved apsed basilica of the early fourth century, built adjacent to thepalace to serve as an aula regia ,animperial audience-hall. The building’s
design reveals an accomplished architect creating a theatre for the visualexpression of absolutist ideology; its prop ortions and internal optical devices
are ingeniously contrived to draw the eye to the apse, in which the ruler satenthroned against a background of richly coloured mosaic.
125
Specialists remain divided as to whether any single underlying archi-
tectural model generally determined palace structure in the late imperial
period.126On one view, there was a Hellenistic architectural model at play,
picked up on and mediated through the Romanised military layout of amid-third-century palace built for Philip the Arab at Syrian Philippopolis,and later at Split. That view is hard to sustain, and whether the Diocletianicsite at Split can be treated as a detailed template for others for which little orno material evidence exists is disputable: it is hardly clear, for instance, thatthe Great Palace at Constantinople, constructed over an expanse of around100 hectares, walled-off and terraced like Domitian’s Domus Flavia ,e v e r
replicated the tight, near-symmetrical castrum organisation of Domitian’s
‘palace’ at Split. That said, the material traces and panegyrists’ praises of
tetrarchic ‘capitals’ such as Trier, or Thessalonica, or Sirmium, show the
late imperial ‘palace’ as usually a unit standing in an intimate relation-ship to another architectural element so as to form a broader spatial com-plex encompassing a circus/hippodrome,
127and whether or not the Palace
at Constantinople was laid out in conformity with a speciﬁc architectural
123Lact. Demorte pers .7.9.124Pan.Lat .VI.22.5.125Ward-Perkins 1981 :444–5.
126Dagron 1974 :93; Duval 1997 :127–53 (a sceptical critique).
127Millar1977 :47; Ward-Perkins 1981 :442, 449–50; Duval 1997 :139.

194 rowlan ds m i th
model, there was surely a broader se nse in which the imperial residential
area on the Palatine at Rom e–ac o mplex bounded on one side by the Circus
Maximus, and overlooking the Forum and Senate-House on the other – was
ashaping inﬂuence. Constantine’s particular choice of site for the Great
Palace seems to have been inﬂuenced by the location of a pre-existing hip-
podrome at Constantinople – which is to say that one of the best-known
features of the complex, the arrangement giving the emperor and his retinuedirect access from the Palace to the imperial box ( kathisma )i nt h e Hippo-
drome, a prime locus for imperial self-display lying immediately adjacenttothe western edge of the Palace, replicated the pattern at Rome.
128So, too,
the Great Palace was from the ﬁrst conceived of as a focal element in a largermonumental centre; just as the Palatine complex stood close by the T empleof Palatine Apollo and looked out onto the Forum and Senate-House andJupiter Capitolinus’ temple, so the view from the Great Palace’s monumental
entrance, the ‘Brazen Gate’ ( Chalke ), looked out to what had formerly been
Byzantium’s civic centre in the high empire, the Tetrasto¨on–aspace now
reconﬁgured as the Augusteion ,avast porticoed square creating an arena for
imperial processions; at the square’s eastern edge was a senate-house for thenewly constituted imperial senate of Constantinople, and beyond that, onthe slope rising towards the old acropolis of the city, the church of St Irene(and later, from 360 onwards, St Sophia).
129The parallels suggest that in
the layout of his new eastern capital Constantine wished to appropriate andevoke the charismatic imperial aura of the Palatine residential complex and
its surrounding public space and architecture.
Inthe absence of material traces, the internal conﬁguration of the fourth-
century palace area at Constantinople eludes us. T exts indicate some of
its major component parts – but very little about how they were spatiallyrelated.
130The overall layout was perhaps less regular than the analogy
with the compact symmetrical suites of Diocletian’s ‘Palace’ would suggest;
arguably it resembled rather the sprawl of an aristocratic country-villa com-plex. In any event, we hear of a monumental entrance, the Chalke ;alarge
court, the ‘Tribunal’ (or Delphax ), with smaller meeting chambers attached;
theMagnaura ,anapsed basilican reception hall; two consistoriums, the
‘Summer’ and the ‘Small Consistorium’;
131agreat dining-chamber, the
128Krau theimer 1983 :49; Dagron 1974 :320–47; cf. Zanker 2002 :109 on the Domus Flavia .
129Dagron 1974 :98–9, 138–40.130McCor mick 2000 :140; Bardill 1999 .
131Aswehave indicated earlier, ‘ consistorium ’was a term that could refer not just to a body of
advisors but also to the place in which they assembled – in architectural terms, a structure akin
toabasilica, a meeting-chamber with an elevated recess from which the enthroned emperor
issued his judgements. On the relation of the two consistoriums, see Guilland 1969 :56–8.

The imperial court of the late Roman empire 195
‘Triclinium of the Nineteen Couches’ , capable of seating over 200 guests;
the imperial residential suite, the Daphne ,fromwhich a spiral staircase gave
access to the imperial box in the Hippodrome (above); kitchens and quarters
for the domestic staff; garden terraces and courtyards; several chapels;the barracks and stables and armoury of the imperial guard; an im-perial mint; and the ofﬁces of the various other departments of the imperialbureaucracy.
132
Intime, other palatia werebuilt around the Great Palace as mansions
for great courtiers. Immediately to the west of the Hippodrome, palaceswerebuilt in the ﬁrst half of the ﬁfth century for two high-ranking eunuch
praepositi (grand chamberlains), Lausus and Antiochus. The archaeological
remains of Antiochus’ palace are securely identiﬁed, and its dimensions do
something to suggest by comparison the massive scale of the chief roomsof the Great Palace: a stairway led up to a semi-circular entrance porticonearly 50 metres in diameter, beyond which stood a hexagonal reception-chamber, each side of which was over 10 metres wide; the identiﬁcation
of Lausus’ palace from material evidence is more problematic, but it toowas evidently very large; it housed a choice collection of sculptural master-pieces acquired by the Chamberlain, among them Pheidias’ statue ofOlympian Zeus, a piece 12 metres high.
133And a sw eh a v en o t iced earlier,
other imperial palaces and villas developed in the suburbs and outside thecity.
134Inthe west and the Balkans, too, named country villas are attested by
Ammianus as occasional imperial residences in the vicinity of regional capi-
tals or fortresses.135The archaeological evidence in these cases points to sub-
stantial, semi-fortiﬁed complexes, and to call such places imperial ‘retreats’perhaps carries misleading connotations. ‘Contionacum’ near Trier, say, hasbeen plausibly identiﬁed with the remains of an extensive villa complex at
132Kelly 1994 :162–3. These ofﬁces must have occupied many rooms; by the sixth century, for
instance, the contents of the archive of the Praetorian Prefect of the East alone seem to have
ﬁlled over twenty storage-rooms located under the Hippodrome.
133Krau theimer 1986 :71; McCormick 2000 :139; for the (disputed) dates of construction, see
Greatrex and Bardill 1996 :193–7; and for the location and artworks of Lausus’ palace, Bardill
1997 :67–9.
134E.g. Theodosius II’s sisters were fond of ‘Ruﬁnianae’ , a palace on the Asian shore, andemperors had country properties at which they could relax: a letter of Julian’s describes his gifttoone Evagrius (arguably a rhetor attached to Julian’s court) of a small villa and estate on the
Bithynian coast bequeathed by his grandmother – in Julian’s memory, ‘a most delightful
summer resort’ affording a view of Constantinople in the distance across the Bosporus.Callinicus, VH y p atii37.3; Julian, Ep.25(4 Bidez).
135‘Pistrensis’ , for instance, 20 miles or so distant from Sirmium, or ‘Murocincta’ (‘thewall-encircled villa’) near Carnuntum, or ‘Contionacum’ near Trier. Amm.29.6.7; 30.10.4;Matthews 1989 :401.

196 rowlan ds m i th
Konz that included an apsed audience-hall, and it ﬁgures in the Theodosian
Code as a place from which Valentinian I issued laws in 371.136These details
nicely illustrate a basic point already remarked on earlier: until quite late in
the fourth century, emperors were by and large still rulers on the move; evenin the ﬁfth, emperors based at Constantinople might move in some seasonstoanother nearby city (Thracian Eudoxiopolis, for instance, or Diocle-
tian’s old capital at Nicomedia) and in the west Rome began to ﬁgure againas a temporary imperial residence for the emperors based at Ravenna.
137
Andwherever he went, an emperor on his travels was accompanied by an
entourage of servants, soldiers, comites and bureaucratic functionaries. We
pass now to consider ‘the court’ in this aspect, as an ordered collectivity ofpersons.
The court as a human collectivity: the emperor’s comitatus
Inthe formal language of the Theodosian Code, the court personnel are
collecti vely thesacer comitatus or the domus sacra ordomus aeternalis ;
Ammianus usually calls them simply ‘the comitatus ’o r‘ thecomitatus of the
Augustus’ .138Avariety of expressions were used by Ammianus and others to
refer to its leading members: ‘those powerful in/holders of ﬁrst place in the
emperor’s hall ( regiaaula)’ , or more simply palatinae dignitates ,‘[holders of]
palace honours’ . In Greek, the expressions used by writers like Libanius andEunapius and Zosimus show a comparable range, speaking of the personnel
collectively as the ‘royal household’ ( basilik¯ eoiketia )oras‘those around the
aul¯e’,and of the leading members as ‘the dynatoi in the palace ( basileia )’ , or
‘those in the hetairia around the emperor’ (that is, his comites ).
139The ten-
dency of such locutions to continue to evoke a material object, a hall or court-yard,isnoteworthy – but so is the fact that they are often manifestly being
used to designate the court as a human collectivity or a political institution,without reference to any particular topographical location at a given time.
The comitatus of a fourth-century emperor probably numbered around
6,000 persons – on the move, a massive group requiring a great deal ofthemensores (the palatine ofﬁcials who supervised arrangements for the
billeting and supply of the court on its journeys): we must envisage roadspacked for miles with columns of soldiers, carriages carrying the imperialhousehold and its comites and high ofﬁcials, waggon trains loaded with
136Cod.Theod .4.6.4; 9.3.5; a drawing in Weitzmann 1979 :116 reconstructs the site.
137Jones1964 :366.138Schlinkert 1996 :460–1; Neothlichs 1998: 15.
139Schlinkert 1996 :460–1 for refs., adding Libanius Or.11.194 (on comites ).

The imperial court of the late Roman empire 197
clerks’ ﬁle boxes, coined money and gold bullion – for the imperial treasury
travelled with the ruler, and by the late fourth century the minting of gold had
itself become exclusively a matter for the comitatus .140Constantine himself
acknowledged the strains that the migratory life imposed on his bureaucrats:‘Nor are the palatini [clerks] strangers to the toils of the camp, [since] they
accompany Our standards and are always present at Our ofﬁcial acts, andas they devote themselves to their clerical duties they are exposed to lengthyjourneys and difﬁcult marches’ ( Cod.Theod .6.36.1). No extensive account
of the bureaucratic departments within which these clerks served can beundertaken here, but even a summary and highly simpliﬁed description ofthe divisions of the comitatus will sufﬁce to convey both the scale of the
emperor’s personal retinue and the striking degree to which the businessof imperial government had come to centre on a departmentalised palacebureaucracy under the control of an inner elite of courtiers and ministersappointed by the emperor.
For conve nience, we can divide the comitatus into four basic constituent
parts. First, soldiers. The infantry and cavalry regiments of the palace guard,thescholae palatinae (at Constantinople, there were originally seven such
regiments) accounted for roughly half of the comitatus –about 3,000 crack
troops (in the fourth century, at least), predominantly of Germanic origin,
drawn from the emperor’s ﬁeld army; and a band of forty picked men fromthese regiments formed the white-uniformed candidati ,the emperor’s elite
personal bodyguard.
141Bythe sixth century, the scholaehad become largely
decorative units and the regimental posts were up for sale; a new corps, theexcubitores ,had taken over their practical military functions. The duties of
thecandidati had likewise by then become mainly ceremonial; but in the
fourth they still functioned as a true bodyguard, and would accompanythe emperor on his campaigns.
142There was also an elite corps of ofﬁcer-
cadets attached to the court, the domestici et protectores :some were soldiers of
proven worth promoted from the ranks, others ambitious younger men whohad obtained their commissions by inﬂuence or purchase – sons of militaryofﬁcials, sons of German nobles, civilians in search of a sinecure.
143The
corpsofcadets – and military service at court in general – was accordingly
apotent force for social mobility; the father of the emperor Valentinian, for
instance, had been a Pannonian peasant-soldier whose strength and skill asawrestler had earned him promotion from the ranks as a protector ,f r o m
140Jones1964 :367, 437.141Jones1964 :613, 658.
142Frank1969 :127–42; Whitby 1987: 464, with Amm. 25.3.6 ( candidati with Julian in Persia).
143Jones1964 :612, 638–9.

198 rowlan ds m i th
which post he then rose further, ending up a military comes ;and Valentinian
himself, at the time of his accession, had recently been given command of a
scholaof the imperial guard (Ammianus 30.7.2–4; 26.1.5).
Second, the emperor’s personal household establishment, the sacrum
domesticum ,classiﬁed into three broad divisions: the cubicularii (eunuch
chamberlains), the castrensiani (domestic personnel such as cooks, pages
and hairdressers) and thirty silentarii ,acorps of court-ushers who attended
the emperor at meetings of the consistory.
Third, the ‘sacred’ consistory itself – a ‘privy council’ made up of high-
ranking ofﬁcials and imperial comites ,and served by a secretariat of notarii
under the headship of a chief secretary, the primicerius notariorum .144The
consistory had its origin in the informal consilium principis familiar to his-
torians of the earlier empire, but its very name hints at the changes wrought
bythe ‘absolutist’ strain in late imperial ideology: its members remained
standing ( consistentes )i n the emperor’s presence throughout its meetings,
and a good part of its business was formal or ceremonial; it functioned notsimply as an advisory cabinet, but as a council of state and a court of justice:it was at sessions of the consistory that civic delegations and foreign envoyswerereceived, ‘dignities’ and titles of high ofﬁce conferred, and imperial
largesses bestowed. In principle, moreover, formal sessions of the full con-sistory were one-way affairs, events technically termed silentia at which the
membership was supposed to listen quietly to the ruler’s announcements,with the silentarii in attendance brandishing golden wands to ensure silence.
The reality, as several passages in Ammianus indicate, by no means alwaysconformed to this solemn image; in the fourth century, at least, meetings
could still be the occasion of uninhibited debate about policy matters.
145
Butitseems clear that key decisions were often taken by a smaller circle of
intimate comites of the consistory meeting less formally, and then presented
toafull consistory-meeting as faits accomplis ;and in the ﬁfth century the
consistory turned into an essentially ceremonial body.
Fourth and last (here we gather a large and heterogeneous bureaucratic
category under one broad heading) there were the ofﬁcials and functionar-ies of the various departmental ministries and their sub-ofﬁces ( scrinia) ,
military and civilian. If we exclude the notarii of the consistory, already
mentioned, the central bureaucratic staff can be divided into four principalpalatine ‘ministries’ ( ofﬁcia ):
146
144For w hat follows, see Jones 1964 :333–41.145Matthews 1989: 268.
146Fordetailed discussion of the bureaucratic structures, see Jones 1964 :562–86; Noethlichs
1991 :1111–58 (= RAC ,s.v. ‘Hofbeamter’); Noethlichs 1998 :27–39.

The imperial court of the late Roman empire 199
1the staffs of the three sacra scrinia ,the imperial secretariats of the magistri
memoriae ,epistularum andlibellorum respectively, who drafted responses
toindividual and communal petitions, prepared legal cases and issued
certiﬁcates of enlistment to members of the imperial service;
2thelargitionales ,the staff of the sacrae largitiones ,the ministry of imperial
ﬁnance, which processed tax revenues, regulated the minting of coin and
was responsible (among much else) for the provision of the clothing ofcourtiers and bureaucrats;
3theprivatiani ,the staff of the res pri vata,administering the collection of
rents from all imperial property and land;
4the ‘regiment’ ( schola)o f theagentes in rebus ,acorpswhose members
began their careers as imperial couriers and then progressed to posts asimperial inspectors and agents of various sorts, some of whom remainedbased at the court.
Arough indication of the size of the central bureaucracy as it existed
c.AD400 can be got by calculating a notional total from the number of
men recorded in the Theodosian Code as serving in these various ministriesat various points in the late fourth or ﬁfth centuries in (usually) the easternsector of the empire: 520 notarii (AD 381); 130 clerks in the sacra scrinia
(AD 470); 224 largitionales (AD 399) (with over 600 more entered on the
waiting lists as supernumeraries); 500 privatiani (AD 399); 1,174 agentes
in rebus (AD 430) (not all of them, of course, present at court). The total
ﬁgure for the east alone is thus well in excess of 2,500.
147And on t op of the
bureaucratic divisions so far mentioned, one must allow for the presenceof the central staff of whichever praetorian prefect was attached to a givencomitatus –inthe case of Constantinople, the ‘Prefect of the East’ , in the case
of Milan, or Ravenna, the ‘Prefect of Italy’ .
148
To co nv eyaﬂ a v o u r of the modus operandi of the comitatus ,w ec a n pick
out here only a few of the key posts for comment. The members of theconsistory constituted an elite group. Most were appointed to it as comites by
the emperor’s personal grant, many after service as a governor of a provinceor diocese, some simply as valued imperial favourites at court (the low-bornDatianus, for instance, noticed earlier as a high-ranking comes of Constantius
II, had never held any ofﬁce in the bureaucracy). The four top civilianministers at court, however, were ex ofﬁcio members: the ‘Master of Ofﬁces’
(a post created by Constantine), who had overall charge of the three sacra
scrinia,the corps of the imperial couriers, and (later) the regiments of the
147Heather 1998a :189 draws up a table from ﬁgures in Jones 1964 :572–86.
148Jones1964 :370–1.

200 rowlan ds m i th
palace guard; the ‘Quaestor of the Sacred Palace’ (another Constantinian
innovation), who served as the emperor’s chief legal ofﬁcer and represen-tative at court and proclaimed his edicts to the consistory (he was usuallyaprominent lawyer or rhetorician: we have noted Ausonius as a case in
point); and the heads of the two ﬁnancial ministries, the comes largitionum
and the comes rei privatae ,who between them controlled the tax revenues,
the administration of imperial estates, mines and mints, and the payment ofsalaries. As well as these four, collectively styled the comites consistoriani ,the
Praetorian Prefect in attendance at court (the supreme civilian minister of
state overseeing the administration of the provinces and dioceses in his allot-tedregion of empire) was also a member ex ofﬁcio ;
149and so likewise were
several high military ofﬁcials: the commander of the ofﬁcer-cadets ( comes
domesticorum ); the commanders of the regiments of the palace guard; and
two ‘Masters of Soldiers in Attendance’ – frequently of non-Roman (usuallyGermanic) extraction – who commanded divisions of the mobile ﬁeld armyin the vicinity of the court. The consistory’s membership – in the fourthcentury, at least – is thus emblematic of the broader reconﬁguration of the
court elite as an ‘aristocracy of service’ that we have discussed earlier: some
men of traditional Roman senatorial background could turn up in it, eitheramong the ex-proconsular governors or ex- vicarii ,orsimply, like the grand
Nummius Albinus,
150as esteemed advisors; but the heads of the four great
Palatine ministries in this period were always ‘new men’ , and a good few of
them had risen through service as notarii :either emperors did not care to
place the traditional aristocracy in these four high posts at court, or else theold senatorial families viewed them as involving a form of personal service tothe emperor that they shunned as undigniﬁed.
151So too, the high military
posts that conferred membership ex ofﬁcio werenow no longer available
tothe old senatorial families, and while Roman aristocrats still show up
often as Praetorian Prefects of Italy, the Prefects in the East were usuallyparvenus (a conspicuous example, Flavius Philippus, allegedly the son of asausage-seller, has been noted earlier). In the consistory, then, a smatteringof traditional aristocrats kept company with ambitious ex- notarii ,Frankish
or Gothic or Burgundian generals, and comites of widely divergent social
background.
Among the corps of notarii who served as the consistory’s secretariat,
the longest-serving member, the ‘First Senior Notary’ ( primicerius notari-
orum ), emerged at court as an important ofﬁcial in his own right; by 381,
if not earlier, primicerii were formally ranked on a par with proconsuls –
149Gutsfeld 2004 :85–7.150cos.AD345; ILS1238.151Jones1964 :134.

The imperial court of the late Roman empire 201
and were hence members of the senatorial order.152The primicerius had
responsibility for the preparation of the codicils issued to all senior ofﬁ-
cials to mark their appointment, and of comprehensive lists of all holders
of high-ranking ofﬁces ( dignitates ); it was thus a primicerius for whom one
of our key sources, the NotitiaDignitatum ,was compiled. The corps of
notarii that he headed was perhaps originally staffed by men of low social
standing, but it soon came to form a prestigious company at court, and inthe course of the fourth century competition for appointment to it becameintense. By the mid-century, emperors were already entrusting notarii with
important tasks and missions as special agents, couriers and ambassadors,and some of them were subsequently raised to the highest ranks and ofﬁcesat court. As the corps’ prestige grew, its social composition changed; postsonce thought suitable only for uncultured clerks with short-hand skills, orunappealing because of the long years of service involved, came to attractwell-educated sons of the urban propertied classes – and by the ﬁfth cen-
tury, even some members of the old Roman aristocracy.
153And as the corps
became fashionable, the overall number of notarii grew: in AD 381, we
noticed earlier, there were apparently 520 in the East alone. The numberactually in service at court at any given time is another matter, and veryhard to judge: the ﬁgure of 520 will include not only men serving as im-perial agents in the provinces, but also a good number of persons who hadobtained the rank as a sinecure by inﬂuence or payment. But whatever thenumber, the corps of the notarii holds a special interest for us, and, like-
wise, the secretariats of the four main palatine ministries supervised by the
comites consistoriani –the Master of Ofﬁces, the Quaestor of the Palace, and
the two ﬁnancial comites .They were a point at which the privileged milieu
of the court and its bureaucracy clearly intersected with the wider world ofthe non-aristocratic but respectable propertied urban classes, and throughwhich the image and values of the emergent ‘court society’ of the fourth cen-tury could impinge directly on the consciousness and aspirations of thoseclasses. More than that, it could be argued that the social ethos of the ‘typical’upwardly mobile, non-aristocratic, notary – educated, technically capable,status-oriented, conscious of his own worth and rank but with an insider’sconsciousness also of the distance that lay between him and the lofty heights
of the inner elite at court, the celsae potestates
154–informs some of the
keysource material for our period. The fourth-century historical ‘epitoma-
tors’ Victor and Eutropius, we have noted earlier, were both bureaucrats in
the secretariats of either the consistory or the sacra scrinia ,and Ammianus
152Jones1964 :573–4.153Amm.29.1.1; Jones 1964 :127–8, 572–4.154Amm.28.6.9.

202 rowlan ds m i th
himself, if his career had taken another turn than it did, might quite easily
have become a notary in the bureaucracy; he had begun his career as anofﬁcer-cadet at court, and the cadet corps was a well-established route to apost as a notary. As for the Noti tiaand the Theodosian Code (along with
Ammianus, the fundamental texts for any reconstruction of this period),
both were documents produced by the court-based bureaucracy, and thor-oughly imbued by its values and outlook: they were drawn up for the primi-
cerius notariorum and the Quaestor of the Sacred Palace respectively – which
is to say, of course, drawn up by the notaries in their ofﬁces.
Within the section of the comitatus serving in the emperor’s domestic
household, the dominant functionary was (or soon became) the praeposi-
tus sacri cubiculi ,the ‘Grand Chamberlain’ , who had overall charge of the
cubicularii ,the staff of palace eunuchs who acted as personal servants to the
emperors (and to empresses: in time, indeed, the empress came to possessadesignated praepositus of her own). The praepositus was himself a eunuch,
selected by the emperor and serving at his pleasure rather than for a ﬁxedperiod; praepositi thus tended to outlast many of the high ofﬁcials at court,
whose tenure of a particular ofﬁce would normally be limited to a year orso. Under the praepositus wereother eunuch high ofﬁcials, among them the
‘Superintendent of the Bedchamber’ ( primicerius sacri cubiculi ), the ‘Count
of the Sacred Wardrobe’ ( comes sacrae vestis )and the ‘Steward ( castrensis )o f
the Palace’ , in charge of the non-eunuch staff of the imperial household.
155
The presence of eunuchs as domestic servants in the imperial householdwas nothing new per se (they were already there in Julio-Claudian times),
156
but the high visibility of the eunuch staff at court in the fourth and ﬁfthcenturies, and the high functions and titles assigned to them, constitute
one of the most obviously distinctive features of the late Roman court ascompared to (say) that of the Antonine age. It is not just that the eunuchs
now routinely served at the royal table, prepared beds and royal clothing,and locked the emperor safely in his bedroom at night: they were frequentlyassigned as tutors to the royal children (Julian, for instance, rememberedhis own eunuch tutor notably fondly);
157and by the mid-ﬁfth century, at
least, the posts of captain of the emperor’s personal bodyguard (the spathar-
ius)and keeper of the privy purse (the sacellarius )both went to eunuchs.158
Aboveall, one is struck by the political inﬂuence, the formal honours and
the sheer wealth accruing to the eunuchs who served as praepositi .L i ving in
155Butn.b. Costa 1972 :358–87, arguing that in the early fourth century the castrensis was
superior to the praepositus ,and not originally himself a eunuch.
156Suet. Claud .28.157Misop .352a–d; Smith 1995 :24–5.
158Jones1964 :567–70; Delmaire 1995 :165–72; cf. Guilland 1955 :65; Noethlichs 1998 :37–8.

The imperial court of the late Roman empire 203
close physical proximity to the ruler and effectively controlling access to him
in a private setting, the praepositus could often determine the success or fail-
ure of an approach made by a delegate or petitioner; Ambrose, for instance,ascribed his failure to secure his objective on a diplomatic visit to the courtof the usurper Maximus in the 380s partly to the fact that the praepositus
Gallicanus would not grant him a private audience with the emperor, butonly a reception in the consistory.
159Bythe late fourth century, the post of
praepositus had come to confer on its holder the highest senatorial rank of
illustris automatically on retirement, and in one notorious case (Eutropius,
in AD 399) a praepositus even won the ultimate accolade of a consulship –
and thereby immortality of a sort in a classic work of court literature, asthe butt of a vitriolic invective by Claudian (even the most grotesque freakof nature, the poet declared, paled in comparison to a consular eunuch).
160
According to one observer (admittedly, a highly disaffected one, and prob-
ably not to be trusted on the point), some men were so determined to winEutropius’ patronage that they had themselves castrated, ‘los[ing] their wits
and their testicles’ .
161That is hard to believe, but the post of praepositus cer-
tainly became a byword for fabulous wealth corruptly acquired through thesale of ofﬁces and bribe-taking: in 432, for instance, when Bishop Cyril ofAlexandria wished to gain the support of Theodosius II against the Nestor-ians, he sent 200 pounds of gold to the emperor’s praepositus ,and another
50 to the empress’s, to ease the way for his representative at court.
162Bythat
time, we have noticed, praepositi at Constantinople could afford to build
sumptuous palaces on their own account, and to ﬁll them with extensivecollections of classical sculpture.
Itis important to emphasise that on the face of things the exercising of
such inﬂuence and power at court by eunuchs was an extraordinary devel-opment. In Roman eyes, eunuchs in general were a thoroughly contemptiblespecies; Roman law banned castration on Roman soil as a monstrosity, andin society at large eunuchs were usually encountered as domestic attendantsof aristocratic women, or (when young) as catamites for masters with exoticsexual tastes. The imperial eunuchs at court, then, were slaves or ex-slaves,and typically barbarians to boot, slaves imported as castrated children fromPersia, Armenia or the Caucasus.
163Inthe eyes of fourth-century writers, the
acquisition of political inﬂuence and power at court by such persons was anoutrage: they ascribe it to the eunuchs’ artful ﬂattery of weak emperors and
159Ambrose, Ep.24; Scholten 1998 :54.160Claudian in Eutrop .1.22; Cameron 1970 :126–37.
161Eunapius Fr. 65.7 Blockley.162Brown 1992 :17; Holum 1982 :180.
163Matthews 1989 :274–7. Not allcame as children: at least one praepositus had ﬁrs ta r r i v e da sa n
adult emissary from the Persian court; for the context, Greatrex and Bardill 1996 :173.

204 rowlan ds m i th
totheir inordinate greed as a type, and the praepositi and their underlings
are accordingly damned by the likes of Ammianus and Claudian as guileful,
malevolent freaks – so many ‘lizards and toads’ , bare-bottomed ‘apes’ in east-ern silks, sallow-skinned ‘snakes’ with squeaky, boyish voices, a tribe of ‘bats’who, once outside the protective ambit of the court, hide themselves awayin dark places to avoid public odium.
164These prejudicial images of deceit-
ful minions hood-winking impressionable rulers were to shape Gibbon’sview of ‘eunuch power’ as a monstrous oriental import that degraded and
enfeebled the Roman state: ‘If we examine the general history of Persia,India and China, we shall ﬁnd that the power of the eunuchs has uniformly
marked the decline and fall of every dynasty.’
165There are certainly interest-
ing parallels to be drawn in this connection with the role of court-eunuchs ineastern settings – in Han China, say, as discussed by van Ess in this volume –but the propensity of the ancient writers and of Gibbon to account for theprominence of eunuchs in the late Roman court as a situation arising fromthe ‘oriental’ guile of a succession of individual praepositi and the vanity or
weakness of individual emperors does not constitute an adequate account of
astate of affairs that lasted centuries; the role of the eunuchs at court needs to
be viewed and explained as a normalised, ‘systemic’ phenomenon in whichprovision was made, say, for a trainee corps of adolescent eunuchs.
166
‘Eunuchology’ currently ﬂourishes as a specialised sub-ﬁeld in the study
of antiquity, and of associated later periods too: there are ﬁne explorations ofthe ‘construction’ of a third gender in the ﬁgure of the eunuch at Byzantiumand its traces in eunuchophiliac Christian texts.
167The institutionalised
status of the late Roman court eunuchs has been closely studied in recentyears,
168but the key account remains a brilliant paper by Keith Hopkins ﬁrst
published in the sixties.169Hopkins saw the key to the explanation of the
inﬂuence of the praepositi not in the weakness of individual emperors, but
in a deep structural tension inherent in an absolutist monarchic system –the tension between the emperor and the political and military aristocracy,however conﬁgured. ‘Absolute’ or not, no emperor could ever in practicerule the empire (or his portion of it, when two were co-reigning) by himself;
he needed to delegate power to others as generals and administrators, and
164Basil, Ep.115; Claudian in Eutrop .1.303; Amm.16.6.17; 19.7.7; 18.4.4.
165DFvol. II, 177.
166Matthews 1989 :276, inferred from Amm.18.4.4.
167Ringrose 2003 (third gender); Mullett in T ougher 2002 (eunuchophilia); Marmon 1990 .
168Schlinkert 1994 ;Scholten 1995 ;1998 :51–74.
169Hopkins 1963 (=1978: 172–96); qualiﬁcation of his stress on eunuchs as ‘outsiders’ in T ougher
2002 .

The imperial court of the late Roman empire 205
the corollary was the recognition and maintenance of an elite grouping of
some sort. But while the delegation of power was unavoidable, the poweraccruing to the elite helpers selected always constituted a potential threattothe monarch, and emperors sought to limit it by various means – partly
byconciliating the elite to keep it loyal, partly by maintaining a balance of
power within its constituent parts. In the early empire, the threat had lainmainly with the traditional senatorial aristocracy from whom the emperorschieﬂy drew their executive helpers, and emperors had checked it partly by
adopting a conciliatory ideology of civilitas that represented senators as the
emperor’s social peers, partly by restricting the length of senators’ tenuresof military and civic ofﬁces and by fostering a counterbalancing element ofnon-senators, equites ,deployed as praetorian prefects and as functionaries
in administrative roles that senators regarded as menial. By our period, theposition was much changed: the image of the emperor as a divine, or god-like,power had consigned the ideology of civilitas tothe sidelines, and the social
composition of the senatorial order had been transformed, the old senatorial
aristocracy having been quite eclipsed by the rise of the equestrians; by thelater fourth century, we observed earlier, the two old orders had effectivelycoalesced in a reconﬁgured and much enlarged senatorial order in which
new senators were constantly being created by virtue of ofﬁce-holding in theemperor’s service; senatorial status was now itself a reward for loyalty. Butthe basic problem facing emperors still remained; to control the armies andraise the taxes from the provinces effectively, they still had to delegate power;
and the elite class thus created, however it was conﬁgured, still potentiallythreatened the centralising power of the monarch: it was still apt to aspire tothe privileged social and economic status of a traditional landed aristocracy,and Hopkins pointed in this connection to a worrying ‘centrifugal’ tendencyin the west, with the emergence of proto-feudal estate owners ( potentiores )
resistant to taxation and to the levying of military recruits from their work-
forces.
170Andinsome ways the problem was accentuated by the absolutist
ideology which elevated rulers as ‘sacred’ beings far removed from even theirmost elite subjects. Whereas the style of the civilis princeps had facilitated
social interaction between the ruler and his privileged elite, the absolutistethos emphasised the social distance between them. In practice, though, achannel of contact with the elite was stil lessential for both parties; emper-
ors needed to know (and to be seen to know) what inﬂuential subjects werethinking in order to control them; they needed to play powerful individualsoff against one another with a sense of their likely reactions, rewarding the
170Hopkins 1978 :188.

206 rowlan ds m i th
loyal and punishing or accommodating the disaffected – and to be seen to be
doing so. Conversely, the members of the elite required a channel of contactwith the emperor to mark out and maintain their privileged status in the
eyes of others.
Insome historical periods (see Brosius and Spawforth in this volume),
the rendering of personal service to the ruler by nobles at court could help tomeet these needs and fostered a degree of social intimacy; at Versailles, say,domestic posts in the ruler’s household such as the premiers gentilhommes
de la chambre or the grand maˆ ıtre de la garde-robe ,were honoriﬁc charges
awarded to dukes.
171Inthe case of the late Roman court, there was no
‘domestication’ of the nobility in the literal fashion that Elias claimed for
Versailles; the prevailing aristocratic ethos precluded it. Some other means
of bridging the divide was required by the ‘sacred’ emperor, and in Hopkins’view it was principally the corps of eunuchs serving in the imperial house-
hold who came to function as the key intermediary channel, ‘a lubricantpreventing too much friction between the emperor and the other forces ofthe state that threatened his superiority’ .
172Atcourt, the eunuchs’ ease of
movement around the palace kept the emperor well informed of goings-on, and they provided (for a price) access to the emperor’s person – butaccess meted out in such a way that it could not be taken for granted, wouldnot deprive the imperial ofﬁce of its aura of power and grandeur in theminds of petitioners, and would not require the emperor to award honoursand ﬁnancial beneﬁts more liberally than he wished. Deployed away fromcourt as envoys on special missions, they could keep a check on, and if need
be take pre-emptive action against, military commanders, governors andleading bishops. Although no ancient source explicitly says that emperorsconsciously raised the eunuchs up as a means of controlling the elite, it is
hard to think they were blind to their potential value on that score. At quitewhat point in the fourth century the eunuchs’ corporate role at court wasdecisively enhanced is also unclear, but the praepositi were well ensconced
bythe mid century. The key innovation at issue was surely the consistent
allocation to eunuchs of a tradit ional post in the imperial household – the
chamberlain’s – and it is tempting to follow Hopkins in associating this
change with the elaboration of court ritual that began with Diocletian, and
tosuggest that the implication in the ancient sources that an oriental model
was at play may be well founded; we had occasion to notice earlier that in298 the junior tetrarch Galerius captured and led back to Diocletian theharem and household and many of the courtiers of the Sasanian king – and
171Elias 1983 :162; Antoine 1989 :214–15; Jones 2002 :16.172Hopkins 1978 :180.

The imperial court of the late Roman empire 207
the eunuchs who attended the king’s wives and children were presumably
also taken along with them.
Eunuchswere not the only category of court functionary that could be
deployed to control the elite, of course, but on various counts they wereparticularly well suited for this purpose from the emperor’s point of view.Since they were wholly dependent for their inﬂuence and power on the
favour of the emperor who had picked them as his servants, their own self-interest would tend to keep them loyal to him. And in the nature of thecase, with eunuchs there were no inherited family allegiances or ambitionsharboured for sons to undercut that personal loyalty. Their barbarian andservile origins, the impossibility of their fathering upwardly mobile sons, andthe widespread odium attaching to their name and physical appearance allmeant that eunuchs could not hope to be other than socially marginalisedpersons; unlike, say, the recruits to the corps of the palace notarii whose
social composition we discussed earlier, the imperial eunuchs could neverbe assimilated into the ranks of the new aristocracy. An emperor mightbe prepared to permit a valued praepositus to acq uire riches and badges of
dignity on a scale that might have raised his suspicions in an aristocrat’scase, but no matter how much wealth a eunuch amassed, or how high theranks and titles the emperor awarded him, he would always be an upstart
and outsider in the eyes of the aristocracy. Even the opprobrium attachingtoeunuchs generally could work to the ruler’s advantage; by delegating the
implementation of unpopular measures to the agency of eunuchs he coulddeﬂect criticism and unpopularity from himself, and if need be he could atone and the same time mollify the elite and emphasise his autocratic powerbythrowing an unpopular praepositus tothe wolves, and then redistribute
his wealth to favoured comites .Arcadius chose to do that in the case of
Eutropius in 399, only months after creating him a consul; and the exem-
plary lesson that contemporaries could draw from Eutropius’ dramatic fallis vividly witnessed in the homily that the bishop of Constantinople deliv-
ered in its wake: ‘Where now, [Eutropius], are the brilliant trappings ofyour consulship? …W h e r ei st h ea p plause which once greeted you in the
city, where the acclamations in the Hippodrome, the ﬂatteries of spectators?Gone, all gone . . . mere dreams.’
173The position of even the greatest of the
praepositi was thus always intrinsically far more precarious than that of a
superﬁcially comparable ﬁgure in the courts of sixteenth-century Europe,theprivado or ‘minister-favourite’; the inﬂuence of court grandees such as
Olivares or Richelieu may have depended heavily on the monarch’s favour
173Holum 1982 :62–3; Buck 1988 :28–30; J. Chrysostom Hom.in Eutrop .1.

208 rowlan ds m i th
and provoked elite resentment, but the favourites’ social origins were emi-
nently respectable, and often impeccably aristocratic.174
On several counts, then, the role served by the imperial eunuchs at court
seems nicely calculated to ‘lubricate’ the interactions of the ruler and theelite. A more radical tactic was potentially available to the ruler, of course:tocut back the exuberant growth of the court personnel that secluded him
from easy social contact with subjects and revert to a simpler pattern ofbehaviour that resurrected elements of the older, more accessible style ofthecivilis princeps .Julian arguably had a ‘reactionary’ purpose of that sort
in mind when he purged his predecessor’s palace staff at Constantinople,expelling ‘hordes’ of barbers, cooks, butlers and waiters, eunuch cubicularii
‘more numerous than ﬂies in the spring’ , and most of the ‘villainous’ and
overb earing notarii and the agentes in rebus serving as imperial agents and
couriers.
175Intheir stead, he summoned back ‘upright men’ of culture to
be his companions at court, and the appeal of this accessible imperial stylein the eyes of elite subjects is plain from the compliments paid to Julianbycourt panegyrists: ‘As Emperor, he maintains in the same affection all
whom he accepted in friendship as a private citize n…n oo n ei sd e b a r r e d
from access to him, the doors of the palace are closed to none.’
176But it
is plain, too, that the tactic was riskily double-edged: the well-to-do classesin the fourth century might still value the old imperial virtue of civilitas ,
but emperors were nonetheless expected to project an aura of grandeurand power, and unguarded shows of affability could easily deﬂate it. Julianhimself did not escape criticism on this score, even from his admirers; hedemeaned the dignity of the imperial ofﬁce, Ammianus judged, when he‘leapt from his judgement seat’ in the senate house, ‘forgetful of who he was’ ,
and ran out to greet a philosopher-friend with a reverent kiss, as if they weresocial equals.
177Signiﬁcantly, Julian’s ‘reactionary’ experiment – if such it
was – was not repeated by later emperors. They were aware, of course,that a semblance of civilitas was appreciated by elite subjects, and they still
engaged on occasions in judiciously modulated displays of openness andbonhomie which panegyrists were pleased to register as marks of imperialvirtue: thus Pacatus can afﬁrm that Theodosius had not merely ‘summoned
“Friendship” ( amicitia )tothe palace, but [had] arrayed [it] in purple and
installed [it] on the throne’ , and Claudian can recall how, on a state visit paidtoRome in 389, Theodosius had ‘played the citizen’s part . . . suffering himself
174Elliott and Brockliss 1999 :1–7, 297–9.
175Libanius Or.18.130–5; Amm.22.4.10; Dvornik 1955 :71–81; Smith 1995 :44–6, 170.
176Mamertinus [ Pan.Lat .III], 25.5–26.4.177Amm.22. 7.1–4; cf. Socrates HE3.1.

The imperial court of the late Roman empire 209
toexchange jests with the people . . . and to visit the houses of senators and
the doors of private subjects, setting aside the haughtiness of rank’ .178But
before this bonhomie there had been a spectacular ceremonial procession (an
adventus ,ofwhich we say more below), and the images of royal ‘friendliness’
in these texts were topoi that came easily to the lips of speakers who hadread Pliny’s Panegyric of Trajan or Dio of Prusa’s orations On Kingship .
The fourth-century panegyrists’ variations on them, moreover, are artfullynuanced to emphasise the distance between the ruler and his subjects;
179
in his inestimable goodness, Theodosi us takes Friendship to his bosom by
lifting it up to his high throne, and in his awesome power he permits himselftodescend ﬂeetingly to the houses of senators like some kindly deus praesens ,
asight to be adored and wondered at. As the ‘reluctant’ obelisk erected by his
order in the Hippodrome at Constantinople was compelled to acknowlege,‘everything yields to Theodosius and to his everlasting offspring’ ( ILS821);
and when he died, the poets avowed, Theodosius would reascend to hisproper home in the Heavens: the Bear and Orion would vie ‘to welcomethe new star, each wondering …w h i c hc o n s t e l l a t i o nh e thought worthy of
his presence’ .
180Late Roman emperors, that is to say, were celebrated above
all else as embodiments of an irresistible and potentially terrifying sacredpower: Christian emperors after Constantine did not overtly claim to begods in the manner of Diocletian, who had identiﬁed himself with Jupiter,but it is well known that in a modiﬁed form the adoration of the image ofthe emperor in ruler-cult persisted well into the ﬁfth century.
181Wepass
now to the projection of this image of embodied sacredness in the pompand ceremony of the court.
The ‘sacred’ emperor’s self-display: court ceremonial and protocol
Amemorable passage in Gibbon likens the system of late Roman government
toan intricate stage-play: ‘by a philosophic observer, [it] might have been
mistaken for a splendid theatre, ﬁlled with players of every character anddegree, who repeated the language, and imitated the passions, of their orig-inal model’; and it has been nicely observed that the tendency to represent
178Pacatus [ Pan.Lat .II], 16.2; Claudian VI Cons.Hon .590–62.
179Konstan 1997: 124–30 (topoi); Kelly 1998 :148–9 (distance); but n.b. the qualiﬁcations as to
the ‘inaccessibility’ of the emperor in Garnsey and Humfress 2001 :311–13, citing Valentinian’s
evocation of old primus inter pares style in a law of 371 (Cod.Theod.8.5.32).
180Claudian II Cons.Hon .171–3.
181Cult: Bowersock 2000 :53–6; Brent 1999 ;Salzman 1990 :130–46, 178–81; on ofﬁce of emperor
as divine, Kolb 2001 :63–72.

210 rowlan ds m i th
the late Roman emperors as star-actors in a courtly masque minutely stage-
managed to advertise their resplendent power through gesture and costumegoes back to Ammianus.
182Anabundance of textual and visual material
could be adduced to illuminate the late Roman court in its ceremonious
face.183Herewecan only pick out for comment some especially character-
istic forms of ritualised display: imperial processions and acclamations; theritual ‘adoration of the purple’; the ceremonies at which appointees to ofﬁce
in the emperor’s service were formally invested at court with their codicilsand symbolic appurtenances of ofﬁce; and (brieﬂy) religious observances,royalbanqueting and hunting.
InAD 357, Constantius II and his mobile court paid a rare state visit to
Rome. Ammianus, writing thirty years afterwards as one who had himself
witnessed Theodosius’ state visit of 389, describes his passage into the city
as follows:
Hecame on [along the ViaFlaminia ], preceded by standards on both sides, sitting
alone in a golden chariot, shining with all kinds of brilliant precious stones which
seemed to spread a ﬂickering light all round. The chief ofﬁcers who went before himalso were surrounded by dragons embroidered on various kinds of tissue, fastenedtothe jewelled points of spears, the mouths of the dragons open to catch the wind,
which made them hiss as if inﬂamed with anger …A f t e r these marched a double
rowofsoldiers glittering with brilliant light, clad in radiant breastplates . . . and
the cavalrymen in cuirasses whom they call clibinarii [‘Oven-men’], iron-masked
and breast-plated, girdled with iron belts – you might take them not to be men, butrather statues polished by Praxitele s…T h ee m p e r o ra sh ep r o c e e d e dw a ss a l u t e da s
‘Augustus!’ by voices of good omen, the mountains and shores re-echoing the shouts
of the people, amid which noise he preserved the same immovable countenance thathe was accustomed to display in his provinces. For though he was a man of shortstature, yet he bowed down when entering through the city’s high gates, lookingstraight ahead, as if he had his neck in a vice; he turned his eyes neither to rightnor left, as if he were a graven image of a man; nor did he sway when jolted bythe wheel of his chariot, nor was he ever seen to spit or wipe or rub his face ornose, or to move his hands about. And though this calmness was an affectation, yetthese and other portions of his inner life were indicative of a most extraordinarypatience – a patience granted, as it was given to be appreciated, to him alone [ofmen].
(Amm.16.10.8–10)
This set-piece description of the adventus (ceremony of arrival) of an
emperor is justly famous as an illustration of the transmission of im-
perial ideology in visual form through gesture and display; in stressing the
182DFvol. II; Matthews 1989 :247.
183Forlate imperial visual imagery in art, Brilliant 1963 :163–211; Weitzmann 1979 .

The imperial court of the late Roman empire 211
imperturbability of the emperor in his absolute power, the spectacle implic-
itly depersonalises the individual ruler and projects an emblem of eternalpower and victoriousness; as Ammianus notes, it was meant to evoke atriumphal procession.
184The same ideological turn can be discerned in
the new style that comes to the fore in imperial statue portraiture in thefourth century; unless an accompanying dedicatory inscription is extant, lateimperial portrait-statues are notoriously difﬁcult to identify (the so-called‘Barletta’ statue, for instance, has been identiﬁed by different scholars withhalf a dozen emperors), because the sculptors have little interest in repro-ducing any individual emperor’s facial features; what they aim to portrayis not an individual ruler’s likeness but an idealised type – not anemperor
buttheemperor, stylised as a ‘pantocratic superman’ .
185There is a ne v i dent
evocation of this brand of imperial portraiture in Ammianus’ account of the
adventus of 357; Constantius is transmuted into an unmoving statue – a con-
dition to which emperors were explicitly advised to aspire by an intellectualprot ´eg´eofConstantius who delivered a speech in his presence at Rome on
this very occasion. This was Themistius, leading senator (and later prefect)of Constantinople, advocate of the theory of the emperor as ‘Divine LawIncarnate’ , court-philosopher and propagandist to a succession of fourth-
century rulers, tutor to an emperor’s son – a ‘born survivor’ whose adroitness
in negotiating the hazards of dramatic regime-change makes him a ﬁttingancient match for Talleyrand.
186What is being displayed for gloriﬁcation in
the ritual adventus ,then, is not so much an individual emperor as a deperson-
alised, absolute power inhabiting the body of an emperor, an ‘eternal pres-ence’ deriving from the heavens.
187InAmmianus’ passage there is one point
only at which Constantius breaks his self-willed rule of regal immobility,and it is tellingly symbolic; although he is physically a man of short stature,the power embodied in the emperor has rendered him a god-like giant whomust duck his head to pass under the high-arched monumental gates of
Rome. Around his carriage and bodyguard, by contrast, a crowd of onlook-
ers were jostling and shouting excitedly – but even on that score there wasan established etiquette for the occasion in the cries saluting the Augus-tus. Chanted acclamations of the emperor ﬁgured in many settings – attheadventus ,i n the Hippodrome at the epiphany of the ruler before his
184Matthews 1989 :231–4; MacCormack 1981 :17–89, esp. 41–2; Amm.16.10.2.
185Brilliant 1963 :163; Smith 1985 :219–20; 1997 :194–201.
186MacM ullen 1990 :85–6, 303 n. 18 (emperor as eikon ), cf. Themistius Or.5.64b (Law Incarnate);
MacCor mack 1981 :4 0( T h e m . Or.3 delivered at Rome); on Themistius’ charmed career as an
imperial propagandist, ‘the Talleyrand of his day’ , Heather 1998b :125–49.
187MacCor mack 1981 :55.

212 rowlan ds m i th
people, and quite often in the senate. Although the custom was well estab-
lished before our period, it predictably now takes on a more baroque appear-ance.
188The most celebrated example occurred at a meeting of the Roman
senate on 25 December 438, on the occasion of the promulgation of theTheodosian Code: after a reading of the edict ordering its compilation, thesenators chanted a sequence of forty-three ritual acclamations, each onemuchrepeated: ‘Augustuses of Augustuses, greatest of Augustuses’ (chanted
eight times); ‘God gave you to us, God gave you to us’ (twenty-eight times);and on, and on – the whole sequence, it is reckoned, would have taken agood hour to complete.
189
Ammianus’ account of the adventus focuses on the magnetic presence
of the emperor, but the spectacle in the ceremony it describes was not justConstantius II, but also a body of several thousand people, the itinerant
court enmasse ,marching into Rome in ranking order in a choreographed
procession that must have taken hours to pass under the high gates. Weshould think of the event Ammianus described as a visual projection notonly of the sacred power invested in the individual monarch, but of theprinciples that ordered the hierarchy of the whole ‘court society’ over whichhe presided. Ammianus, for instance, does not mention the empress Eusebiain his account of Constantius’ adventus ,but he later discloses that both she
and his sister Helena had accompanied him on his journey from Milan toRome, and we know from Julian that upon her arrival at Rome she was
formally received by the senate, and that she presided at a distributionof largesse to representatives of the plebs.
190Itis hardly conceivable that
the managers of imperial ceremonial had neglected to train her, too, inthe etiquette of gesture and deportment called for by the occasion, and thesame holds true for the ranks of comites and their staffs; they themselves,
and the shrewder among the spectators, would both take good note of wherethey ﬁgured in the line as it processed.
Aslong as emperors were itinerant, the adventus was associated particu-
larly with the arrival of the emperor in a provincial city (for most of whoseinhabitants, it would most likely be the only time they encountered theemperor in the ﬂesh), but as Constantinople developed as a permanentresidence the ceremonial associated with the adventus also became ‘inter-
nalised’ in the imperial processions regularly staged in the capital city. Wehave noticed earlier the processions to the Great Church and the granaries,
188Alf¨oldi 1970 :82–8; Rouech ´e1984: 181–99; MacMullen 1990 :81.
189Matthews 1989 :248.190Amm.16.10.18; Jul. Or.3.129b.

The imperial court of the late Roman empire 213
but there was another that holds a special interest for its symbolic reso-
nance. Constantine had marked the inauguration of the city in 330 with agrand procession: the emperor and his courtiers gathered in the Forum at
the foot of a porphyry column to witness the placing on its top of a goldenstatue of Constantine crowned radiantly as Apollo-Sol. From there they hadprocessed to the imperial box of the Hippodrome adjoining the Palace toacclaim a uniquely privileged charioteer: a golden image of Constantineborne in a golden chariot completed a circuit round the race-track, accom-panied by troops of the imperial guard in full regalia; in his right hand thegolden emperor held another golden image, the Tych¯ eof his City.
191For
centuries afterwards, the ceremony of the pompa circensis was re-enacted on
the anniversary of its ﬁrst performance, the emperor of the day in the im-perial box honouring at once the founder of his capital and the eternal powerof imperial ofﬁce in company with his people. And from 390 on, moreover,the imperial entourage gathered for the occasion looked out across the trackonto a near-reﬂection of itself, sculpted in relief on the base of the giantobelisk that Theodosius erected in the spina of the Hippodrome that year.
The panel facing them showed the emperor, ﬂanked by his ‘everlasting off-spring’ Arcadius and Honorius, standing on a platform below the kathisma
with a victory wreath in his hand, surrounded by high ofﬁcials and rows of
guards; beneath them, assembled spectators watch musicians and dancers(and underneath the panel itself stands the commemorative inscription wequoted from to open this chapter) (Fig. 5.3). The exact symbolic associations
are debated (on one view, the Hippodrome is symbolic of the cosmos, andthe obelisk at its centre represents the sun), but their basic thrust is hardly indoubt: the panel celebrates the eternal victory and triumph of the emperor,the embodiment of imperial majesty. The Hippodrome adjoining the GreatPalace thus emerged as the locus par excellence of the divine power attaching
tothe ruler, the point at which royalty’s aura was most densely concentrated;
bythe late ﬁfth century, it is telling, it was coming to supplant the military
Kampos of the western suburbs as the setting for the ceremonial accession
of the emperor; the emperor was now acclaimed not by the soldiers, but byhis people gathered at the symbolic centre of the world.
192
Atthepompa circensis ,asthe golden images of Constantine and his city’s
Tych¯ ewereparaded around the Hippodrome, the reigning emperor came
forward from the imperial box and prostrated himself before it as it passed.
191Dagron 1974 :37–41, 307–9; Kantorowitz 1963: 149–62 (solar imagery in Byzantine court
ritual).
192Kiilerich 1993 :37, 48–9; Geyssen 1998 ;Dagron 1974 :314–16.

214 rowlan ds m i th
Figure 5.3 Theodosius I and his court in the Hippodrome in Constantinople, from the
base of the Obelisk of Theodosius
On one very special symbolic occasion, then, the emperor performed an act
of obeisance comparable to those that his su bjects routinely performed for
him. In the ceremony of the ‘adoration of the purple’ , the subject approachedand prostrated himself before the emperor’s dais in the consistory and wasoffered the gold-embroidered hem of the ruler’s purple robe to raise to hislips and kiss – a ‘fetishising’ of the purple, so to speak, as a material expres-sion of depersonalised imperial power. The Theodosian Code discloses thatthe possession of a purple robe or the acquisition of purple dye by anyunauthorised person was an act of treason punishable by execution, andan episode relating to ‘Diocletian’s Palace’ at Split suggests that this law wastaken seriously. Ammianus tells how, in 356, an innocent man was arraignedon a charge of treason after his estranged wife claimed that he had stolenthe purple coverlet draped over Diocletian’s tomb in his Mausoleum;
193half
acentury after his death, then, Diocletian still wore the purple. The gar-
ment’s potency as a visible manifestation of the ‘everlasting divine [power]bywhich the Roman State will ever stand unshaken’ is strikingly conveyed
193Amm.16.8.4.

The imperial court of the late Roman empire 215
in Ammianus’ account of the ceremonial investiture of Gratian as emperor
at the hands of his father: Valentian ‘adorned his son with a crown and therobes o fthe supreme fortuna ’and displayed him before the ranks of the
army present; ‘Behold’ , he said, ‘the imperial robe [now placed on you] bythe wish of myself and my soldiers . . . Gird yourself now, then, as a colleagueof your father.’
194
The earliest explicit reference to the ‘adoration of the purple’ occurs in
alawdated to 354 in the Theodosian Code (8.7.4), in which it is implicitly
treated as already a well-established practice. It was quite likely a Diocletianic
innovation:195wehave noticed earlier that the adoratio was at least thought
byAmmianus and other fourth-century writers to owe something to Persian
practice, and to go back to Diocletian. For our purposes it is less important toﬁx an exact date for its introduction at court than to stress the degree to whichit became routinised in the fourth century as a key feature of court ritualwhich – to judge from Ammianus and the Theodosian Code – was not viewedas ‘un-Roman’ in the sense that it entailed any dishonourable or slavish self-abasement on the part of the adorer: rather, they imply that it was regardedas an honour and privilege and mark of favour for the adorer as well as theadored. In this connection, it has an evident counterpart in the older customof the morning salutatio of the emperor in earlier centuries (see Paterson in
this volume); at meetings of the consistory, the members kissed the purple inan established order of rank.
196Men app ointed to an imperial ofﬁce routinely
performed the ritual at their formal investiture, and Ammianus, who hadbeen enrolled as an ofﬁcer-cadet in Constantius’ court, must himself havekissed the purple. He does not speak of his own case in his history, butas an ofﬁcer in the service of the general Ursicinus he witnessed and laterdescribed a particular adoratio in 355 which for our purposes is highly
instructive. Ursicinus had been under a cloud – suspected of complicityin a revolt, and slandered at court. When news that the rebel leader haddeclared himself emperor reached the court one evening (signiﬁcantly, thewould-be usurper had donned a makeshift purple robe made up of strips
of silk from a military standard
197), Constantius had to decide immediately
whether Ursicinus was innocent and should be rehabilitated in the eyes ofthe court, or condemned by association. Ammianus records the outcome:‘He summoned an emergency meeting of the Consistory’ , and ‘all the nobleshastened to the palace . . . Ursicinus was sent for by the praepositus ,which
194Amm.27.6.1; 27.6.11–12.195Avery 1940 ;S t e r n 1954 .
196Winterling 1998 :c h . 7passim (salutatio ); Avery 1940 :69(kissing in order of rank).
197Amm.15.5.17.

216 rowlan ds m i th
is the most honourable kind of summons, and as soon as he entered the
consistory-chamber he was offered the purple much more graciously than
on any previous occasion.’198ForUrsicinus, then, the invitation to perform
theadoratio was a demonstration before the assembled ‘lofty powers’ at court
that the emperor held him in high regard as a loyal and trustworthy general –and Ammianus is content to report the episode in that spirit. The passage alsoimplies that the adoratio was performed fairly routinely at court, and there
is a well-known case from Egypt to show that persons much less grand thanUrsicinus had their chances to kiss the purple – and that for them, too, it was
abadge of esteem that could be used to ‘pull rank’ to practical advantage. In
340, a petition reached Constantius’ palace from one Flavius Abinnaeus, anarmy ofﬁcer in Egypt who had been passed over for promotion by his localcommander in favour of another candidate. The rival had used local patronal
inﬂuence, but Abinnaeus had a better card to play: some years earlier, serviceon a military escort had taken him to the court of Constantinople, and bychance into the presence of the emperor himself – on which occasion, ‘Y our
Divinity ordered me to adore the venerable purple.’ Now, years later, hepetitioned the emperor to intervene in order to secure him promotion tothe post in dispute: ‘for it is clear that [the letters by which my rival procuredthe post] were advanced by patronal inﬂuence ( suffragium ), whereas I was
advanced by your sacred decision [to have me kiss the purple] . . . mayYour Clemency therefore vouchsafe that I be the man appointed’ .
199This
petition is known to have succeeded – an eloquent witness to the mystiqueevoked bytheadoratio in an ofﬁcers’ mess thousands of miles distant from
Constantinople; and at a more elevated level, it has an analogy in an imperial
law of 387 enhancing the ranking order of the emperor’s ofﬁcer-cadet corpsat court by granting a privilege that the ruler deemed undeniable ‘to thosewho are thought worthy to touch our Purple’ .
200
Atthe formal investiture of a palace ofﬁcial or provincial administra-
tor,the appointee performed the adoratio and received a commemorative
notice of his appointment in the form of a scroll or bound book inscribedwith an acclamation to the recipient wishing him success in the name of
the emperor. If the post awarded was a high-ranking one, the appointeeusually also received an ivory diptyc h–a hinged pair of gold-edged ivory
leaves, decorated on the outer sides with carved reliefs of the emperor, andwith his codicils of ofﬁce enclosed. The ceremoniousness of the moment
198Amm.15.5.18, with Matthews 1989 :244–5.
199Abinnaeus Archive, Pap.Abinn.1.11–14; Jones 1964 :637.
200Cod.Theod .6.24.4; Matthews 1989 :247.

The imperial court of the late Roman empire 217
Figure 5.4 The ‘Missorium of Theodosius’ (AD 388)
of investiture is well conveyed in a magniﬁcent work of late fourth-century
art, the ‘Missorium of Theodosius’ – a large silver plate representing aninvestiture of a high ofﬁcial, and presented to him as a commemorative gift(Fig. 5.4). Three emperors – Theodosius I, ﬂanked by two juniors, Arcadius
and Valentinian II – sit enthroned under arched lintels, attended by im-perial guards, and dressed in ﬁnery: belted tunics with ornamental borders,imperial cloaks fastened at the shoulder with jewelled ﬁbulae, and jewelledsandals. Each of them wears a diadem made up of two bands of pearls withalarge jewel at the crest, and each is nimbed. The two ﬂanking emperors
hold symbolic orbs, while Theodosius himself, sitting upright and lookingimpassively out towards the viewer, extends his right hand to slip a codicil ofappointment into a fold in the cloak of a kneeling ofﬁcial (clearly a very high-ranking subject, perhaps a Praetorian Prefect); below the tableau, mean-
while, a reclining female ﬁgure holds a cornucopia. The imagery elevates

218 rowlan ds m i th
Figure 5.5 Leaf of an ivory diptych commemorating
the entry of a Vicarius of Rome into his term of ofﬁce
(about AD 400?)

The imperial court of the late Roman empire 219
the emperors as quasi-divine universal rulers, the source of all riches and
honours, glorying in the fecundity of the ‘everlasting’ Theodosian house.201
High o fﬁce, now, was a gift plucked out from an imperial horn of plenty,
and graciously bestowed by the sacred hand extended downwards by theenthroned emperor. On this count, the representation of Theodosius onthe Missorium conforms to a standard iconographic type; it has an evi-dent parallel in a visual image preserved in Filocalus’ ‘Codex-Calendarof 354’ of Constantius II dispensing largesse: bejewelled and nimbed, theruler sits on a raised dais within a curtained aedicula ,with coins pour-
ing down from his outstretched right hand – and a line of silver lanxes(ceremonial plates) stacked up behind his throne to be distributed to elitesubjects.
202
The carved ivory diptychs of appointment are objects well known to his-
torians of late antique art (Fig. 5.5). They were highly prized as markers of
status, and the recipients kept them afterwards on display in their ofﬁces,on tables covered in blue cloth. In the case of high posts in the civil admin-istrative service, a set of candlesticks and a large ceremonial inkstand werealso awarded, and likewise displayed in the post-holder’s ofﬁce.
203So too, it
pleased an appointee to a provincial governorship to be represented with hiscodicil of appointment in his hand in a commemorative statue.
204Forus, the
particular interest of these appurtenances lies in the style of communicationthey imply. In the earlier empire, the style of the civilis princeps had encour-
aged emperors to communicate with high ofﬁcials with a degree of opennessand informality; in the fourth century, that pretence of social ‘equality’ isreplaced by a highly stylised pattern of communication in which gesture,
gift-giving and chanted acclamations combine to emphasise the asymmetry
of the relationship between imperial majesty and even the highest-rankingimperial subjects. The obvious textual analogue to this visual grammar ofdominance and submission lies in the contorted periphrases adopted in theletters of petition addressed to emperors by the likes of Flavius Abinnaeus,and in the emperors’ own de haut en bas pronouncements in the Theodosian
Code, as drafted by the secretariats of the palace ministries for promulgation
through the consistory. Petitioners have recourse to the ‘optative of cour-tesy’ , asking favours of ‘Y our Clemency’ , ‘if it should please Y our goodness’;
emperors announce their laws or wishes in the royal plural through themouth of the Quaestor in the consistory, referring to themselves as ‘Our
201Kiilerich 1993 :19–23.202Weitzmann 1979 :79(no. 67); Matthews 1989 :232.
203Matthews 1989 :255.204Smith 2002 .

220 rowlan ds m i th
Serenity’ , ‘Our Clemency’ , ‘Our Wisdom’ , ‘Our Eternity’ , ‘Our Tranquillity’ ,
‘Our Imperial Divinity’ .205
Processions, acclamations and the formal ceremonial of the consistory
werenot the only occasions for imperial self-display at court, of course; the
emperor, and often members of his family, were on show in a variety of othersettings too. Two secular communal activities – banqueting and hunting –will serve as our prime examples here, but religious observances performed
in the presence of courtiers within the palace demand a word: they couldplainly serve to underpin the sacred emperor’s claim to a unique degree ofafﬁnity with divinity, and in this connection the devotional behaviour ofrulers keen to promote a sectional religious interest could take on particular
exemplary signiﬁcance. The most obvious cases in point are Constantine
and Julian. Constantine, Eusebius insists, had ‘modelled his palace into achurch of God and afforded an exemplar of piety to those gathered within
it’ , offering up regular prayers with all the members of his court; he orderedSundays to be observed there, ‘instruct[ing] his bodyguard in the practice
of piety’; each day ‘at a stated hour’ he would retire to a private chapel ‘insolitary converse with his God’; at Eastertide, ‘in imitation of the Saviour’sgracious acts, he lavished abundant bounties’; and his preparations for the
Persian campaign he was planning in his last years included the making of
‘atent of great splendour in the shape of a church’ to serve as a mobile
chapel for the emperor and his entourage during the campaign.
206Julian
did virtually the same, but in a pagan register, performing a daily morning-sacriﬁce ‘under the trees in the palace garden’ before a crowd of attendants,dedicating a shrine to Helios (or Mithras?) ‘in the middle of the [Great]Palace’ at which he presided at initiations, compelling the palace guard to
sacriﬁce in his presence on their pay-days.
207
The repercussions at court of Constantine’s ‘conversion’ were not con-
ﬁned to the court’s devotional activities: we noticed earlier that it broughtanew breed of guest, the Christian bishop, to the palace dining-rooms.
Banqueting is a familiar topic for historians of courts (as the other chaptersin this volume attest). It can be construed both as a mode of conspicu-ous expenditure that can emphasise the ruler’s power and as a communalactivity that can foster a sense of social cohesiveness among the privilegedgroup invited to dine in the ruler’s company. Within that group, a comes ’
place in the dining-hall could serve as a visible demonstration of his place
205Cod.Theod .15.1.26; 15.1.30–31; 16.4.1–4; MacMullen 1990 :67–77, esp. 71–3; Matthews 1989 :
255.
206Euseb. VC417–18, 22, 56.207Lib.Orr.1.121;18.127; Sozomen 5.17.

The imperial court of the late Roman empire 221
in the ranking order at court, depending on how closely he was placed
tothe emperor’s own couch, but any invitation to dine at court with the
emperor was a mark of high honour and potentially a path to preferment.Themistius, for instance, was quick to l et friends know when Constantius
invited him to dinner – and in his case the invitations to the palace wenthand in hand with his rapid rise to eminence as a Constantinopolitan sen-ator in 355. Themistius’ friends in their turn could hope that crumbs fromthe emperor’s table would fall their way. Libanius’ reply on hearing of theinvitation speaks volumes on this score: ‘Y our presence at [the emperor’s]table’ , he wrote, ‘denotes a greater intimacy – so that, inasmuch as yourprofessions arise from a concern for your friends, anyone you mention isbetter off.’
208Libanius had particular cause around this time to wish to be
spoken of favourably in Constantius’ presence – and to appreciate what aplace at the emperor’s table could accomplish. In 354 he had resettled at hisnative city, Antioch. Antioch was at that time the seat of the junior emperorGallus, and a local rival of Libanius ha dsullied his name at Gallus’ court
bybribing a young man ‘whose good looks had earned him many a din-
ner’ there to accuse Libanius of using black magic against both Gallus andConstantius: an absurd charge, Libanius insisted – ‘but still, it was believed
that the emperor thought me disloyal, and would show it when he appearedin public by not sparing me a glance’ .
209
Asfor ‘conspicuous consumption’ , there was a balance to be struck:
the luxurious gourmand was a moralising topos – ‘Where now are yourdrinking-parties and suppers . . . the manifold dainties prepared by yourcooks?’ , it was mockingly asked of the fallen eunuch Eutropius
210–and
self-control ( continentia )was traditionally an imperial virtue. For Ammi-
anus, an emperor’s attitude to the pleasures of the table offered a tellingtest-of-thumb: he praises Julian for sharing the simple fare of his soldiers
while in Gaul, and Valentinian for choosing to dine elegantly rather thanextravagantly; by contrast, Jovian’s excessive love of food and wine had
impaired his ‘imperial dignity’ , and his castrensis (the ofﬁcial in charge of the
services at table) is scornfully styled ‘the minister of bellies and throats’ .
211
Imperial panegyrists were accordingly inclined to play down the luxury
of the imperial table, indulging in conventional topoi that surely mini-mise the true level of expenditure involved. Pacatus, say, praises the fareon offer at Theodosius’ palace-feasts as wholesomely modest by compar-ison with the excesses of Neronian times – but a hostile source insists on
208Lib.Ep.52 Norman (= 66.2 Forster).209Lib.Or.1.98–9.
210J.Chrysostom Hom.in Eutrop .1.1.211Amm.16.5.3; 30.9.4; 25.10.15; 26.8.5.

222 rowlan ds m i th
their unprecedented luxury.212The hostile source is probably here the better
guide, even if it exaggerates: on Ammianus’ and Libanius’ testimony, at any
rate,the Great Palace at Constantinople was already awash with cooks and
waiters in Constantius’ day, and the existence there of the grand ‘Tricliniumof the Nineteen Couches’ suggests provision for state banquets on a scalethat easily matched the feasts at Domitian’s Domus Flavia .
213Aprecisely
contextualised story in Eunapius, moreover, reports a particularly grand
banquet provided by none other than Th eodosius – a feast for a company of
Gothic ‘federates’ and their leader, the ‘Master of Soldiers’ Fravitta, whichhad notoriously ended in drunken mayhem.
214
Lastly, we touch on the hunt as a setting for imperial self-display. Hunt-
ing was deemed a ﬁtting relaxation for emperors, and their prowess in theﬁeld was a standard theme in court poetry; even Theodosius II, reputedlyan effete and bookish character who rarely left the palace, was praised for‘draw[ing] the bow like [Homer’s] T eucer’ , and he certainly did hunt: he
died from an injury incurred when he fell from his horse while hunting.
215
Totreat ‘the hunt’ abstractly as a stage for the display of imperial virtue and
power is perhaps to discuss the issue in an over-schematised way; emperorsneeded to relax, after all, and hunting was well established as the favouredpastime not only of kings but of aristocrats in general. Hunting scenes pro-liferate in the art produced for the late imperial elite – in the mosaics andtapestry-curtains of their villas, on the glassware and silver plates they dinedoff, on their sarcophaguses, on their very clothes: in Ammianus, the tunicsof Roman aristocrats shimmer ‘with the shapes of many animals’ , and to afourth-century Christian homilist’s eye the vanity of the rich was manifest intheir love of garments emblazoned with ‘lions and leopards, bears, bulls anddogs, forests and rocks and hunters’ .
216Still, the hunting done in the parks of
the emperors’ rural retreats was a form of communal activity that made for abond between a ruler and the courtiers who rode with him (cf. Spawforth inthis volume, on Alexander). The symbolism of the staged hunting-shows inthe capital at which the emperor presided over the dispatch of lions and wildbeasts conveyed much the same lesson as the ‘once reluctant’ obelisk erectedin the Hippodrome, now ‘conquered and mastered’ and hoisted skywardsagainst its will: ‘all things [must] yield to Theodosius and his everlasting
212Pan.Lat III.14; Zosimus 4.33.1.
213Lib.Or.18.130; cf. Amm.22.4.10; Zanker 2002 (Domus Flavia ); for the persistence of the high
imperial ‘triclinium’ couch-arrangement at late imperial state banquets, Dunbabin 1991 .
214Eunapius Fr.59 (Blockley).
215Holum 1982 :101, 208; Cameron 1982 :228–9 (quoting from Cyrus’ panegyric of c.435).
216Amm.14.6.9; Asterius Hom .1,PG.40:165 (quoted from Maguire 1999 :242).

The imperial court of the late Roman empire 223
offspring’ , the ‘serene lords’ of the world ( ILS821). On both these counts,
analogies with the role of the Royal Hunt in other antique settings come
easily to mind. The Persian tradition of the royal lion-hunt went back to theAchaemenids (see Brosius and Wieseh ¨ofer in this volume): its representation
on Sasanian reliefs and fabrics was certainly later a stylistic inﬂuence on theByzantine iconography of the hunt,
217and it perhaps contributed something
tothe association of emperors in Late Roman art with the ﬁgure of the ‘Great
Huntsman’ , whose spearing of a lion symbolised the triumph of imperial
Virtus over all obstacles and enemies.218Inany event, the symbolism of the
hunt could certainly evoke the image of the emperor as an instrument of
the Eternal Victory that ensured the defeat of Rome’s enemies: hunting hadstrong military associations – sometimes, indeed, large-scale hunts seem tohave been organised with a practical military purpose in mind, as training-exercises in manoeuvring for the imperial cavalry
219–and on this score the
troops that Julian led out against the Persians in 363 enjoyed a symbolic
victory, at least, when they encountered a hunting-park of the Persian king
stocked with lions and bears: ‘our cavalry forced the gates of the enclosureand killed them all with hunting-spears and arrows’ .
220InAmmianus’ eyes,
accordingly, the ruler’s skill in the hunting ﬁeld was another test of his char-acter – but here again, the creation of a favourable impression was a matterof deportment as much as expertise: whereas Constantius’ horsemanshipand skills with the spear and the bow combined with a digniﬁed presencetodo him credit, Gratian’s passion for shooting wild beasts in the imperial
enclosures bespoke an over-eagerness to impress his companions ( proximi )
and struck Ammianus as excessive.
221InGratian’s case, at any rate, there
wereprivate beast-shows staged for the court in the palace-complex at Milan
as well as public shows in the amphitheatre: a Christian source recalls anoccasion when Bishop Ambrose went to the palace to bring an urgent mattertoGratian’s attention and only gained a hearing by joining the hunters in a
purpose-built enclosure in the palace grounds.
222
Asdistinct from the staged slaughter of ﬁerce beasts in the arena, there was
the mounted hunting of wild game in the vicinity of imperial estates – anenergetic pastime less trammelled by the strictures of court ceremonial andprotocol, and perhaps valued all the more as a setting that still allowed forspontaneous human contact between the ruler and a select band of courtiers.Butevenhere, behaviour could be stylised, and for some the prospect may
217Grabar 1936 :60–1; Patlagean 1992 :257–63.
218Brilliant 1963 :183–8, esp. 184 (medallion-image of Constantine II).
219Rance 2000 :254–7.220Amm.24.5.2, with Smith 1999 :97.
221Amm.21.16.7; 31.10.19.222Sozomen 7.25.

224 rowlan ds m i th
have carried echoes of Alexander the Great relaxing with his intimate philoi .
Wenoticed earlier that in the third and fourth centuries interest in the ﬁgure
of Alexander as a paradigmatic warrior-king seems to have intensiﬁed as
the Sasanians emerged as an energetic power in Persia, and this sharpenedinterest is reﬂected in images of Alexander ﬁghting and hunting on horsebackin late imperial art: a giant fourth-century cameo seems to associate himwith a contemporary emperor (Constantius II perhaps?) in the guise of a
diademed cavalier galloping in triumph over the bodies of slain barbarians;and like the emperors Alexander could be associated too with the talismanicﬁgure of the ‘Huntsman’ who secures the triumph of good over evil inthe world.
223Butondecorated table-ware the mounted Alexander is also
depicted hunting stags and boars in a more informal mode, under a legendwishing him long life with his family and friends;
224Julian, for one, admired
this homelier, Macedonian side to the ‘Alexander-style’ , and perhaps had it inmind when he added a post-script in his own hand to a letter summoning atrusted friend to his court in Gaul in his days as Constantius’ junior emperor:
‘There is good quarry of deer [here]’ , he remarked.
225For som ee m p e r ors,
then, the hunt may have indeed appealed as a means of ‘bonding’ with one’scomites .That said, the ‘friendliness’ of emperors could never be taken for
granted, in this or any other sphere: their power was ‘god-like’ , and like gods
they could grow angry. Indeed, exemplary anger in some measure was alwaysintegral to their projection of power. Around the very time he summoned‘upright men’ to be his friends at court, the ‘friendly’ Julian ordered several
prominent courtiers of Constantius to be burned alive, and in Valentinian’scase the sudden bouts of rage were notorious and sometimes lethal to thosein his entourage (and ﬁnally to himself: Valentinian was struck down byan apoplectic ﬁt at a meeting of the consistory, infuriated by the speechof a delegation of German tribesmen who had come to sue for peace).
226
InAmmianus’ view, ‘a semblance’ of imperial ‘clemency’ was the most that
Valentinian could manage, and his behaviour on the hunting ﬁeld and at the
beast-shows revealed as well as anything the choleric nature underneath themask: an attendant who spoiled his day’s hunting by releasing a hound toosoon was summarily beaten to death, and his household functionaries hadreason to fear two notably uncivil courtiers privileged to live close beside his
bedchamber – ‘Morsel’ and ‘Innocence’ , a pair of pet bears, both ferociousman-killers. They were presumably set loose for exemplary purposes inthe palace’s hunting pits, and ‘Innocence’ , at least, was ﬁnally rewarded
223Weitzman 1979 :91, 83.224Weitzman 1979 :89.225Jul.Ep.6 (9 Bidez).
226Amm.22.3.11 (Julian); 30.6.3 (Valentinian).

The imperial court of the late Roman empire 225
bythe emperor with the freedom of the forests for years of murderous
‘servi ce’.227
Concluding remarks
Wehave aimed to indicate how the late Roman imperial court came to take
on its distinctive shape, and to describe its workings in terms that convey
something of the court’s social-political importance. It was not central toour purpose to debate the merits or shortcomings of any particular modelof ‘court society’ on this score, and the question whether Elias’ ‘Versaillesmodel’ is apt or ‘true’ in its correspondence to Roman (or for that mat-terBourbon) historical realities is not, for our purposes, a crucial one: the
‘court society’ is a historical paradigm, not a historical fact, and inasmuch as
paradigms make universalising claims they will always be vulnerable to crit-icism in speciﬁc historical cases. We adduced the case of Versailles, rather, asastarting-point that could offer suggestive counterparts to the late Roman
case in certain telling features. On that count, our discussion has surelyestablished that the social-political milieu in which late Roman emper-ors lived and worked can safely be reckoned a complex ‘court society’ insomething like the sense that Elias proposed for Versailles; the late imperialcourt can certainly be said to have ‘conﬁgured’ and mediated monarchic
power in a particular form through its reconﬁguration of the court elite,its palace-centred bureaucratic apparatus, its ‘absolutist’ ceremonial, and itstransmission of ‘courtly’ social and cultural values to a wider elite in the
empire at large through literature and art.
Wehave emphasised as a background factor the ideological and structural
changes associated with the emergence in the third century of emperors from
within a non-aristocratic military elite of career-soldiers, rulers with social
origins in the provincial (often Balkan) peasantry; the traditional image ofthe virtuous emperor as a citizen-princeps ruling in close cooperation withap e e r-group of Roman senatorial aristocrats did not utterly fade away, but it
was sidelined as imperial ideology took an overtly authoritarian, ‘absolutist’ ,turn which celebrated emperors as gods or vessels of divine power; and thegreat bulk of the military and administrative ofﬁces of the empire came
tobe ﬁlled by non-senators. The structural and ceremonial distinctiveness
of the late imperial court and its expanded bureaucracy betokens a rap-prochement of sorts between the ethos of the new ‘professional’ governing
227Amm.30.8.2; 29.3.3; 29.3.9; see Matthews 1989 :260.

226 rowlan ds m i th
class and an older aristocratic social ethos – a progressive accommodation
of the new elite within the traditional structures of Roman ‘civil’ societyin a new ‘aristocracy of service’ which greatly expanded the membership –and quite transformed the social composition – of the senatorial order.Court ceremonial articulated and normalised these ideological and struc-
tural adjustments. Ceremonies such as the adventus and the ‘adoration of
the purple’ emphasised the sacred emperor’s remoteness from the commonrunofhumanity, but at the same time they enabled a reconﬁgured and
considerably enlarged governing elite – the emperor’s comites especially – to
interact with the ruler in a setting whic hfostered a collective sense of social
identity and established an internal ranking order. There was no seriouseffort made (except, on one view, in Julian’s case) to revert to the older styleof rule that had fostered an illusion of social equality between ruler and elite,but the image of the civilis princeps could still be evoked to conjure a sem-
blance of openness; Constantius, we have seen, could publicly commend ahigh ofﬁcial as a ‘parent and friend’ , and when Valentinian I reafﬁrmed therightofRoman senators to travel to the court at the state’s expense by means
of the imperial carriage-service, ‘because We very often desire to see [them]on account of the merits of our confraternity ( collegium )’ , hee c h o e dt h e
style of the ‘citizen-king’ .
228Even in the later fourth century, there was still
asense in which the workings of the ‘court society’ could help, if need be,
tolegitimise and ‘domesticate’ particular emperors in the eyes of the elite.
On one recent view, for instance, the court-panegyrists’ intricate praises inthe 380s of the ‘serene lord’ Theodosius I as the founder of an everlastingdynasty, a paragon of imperial virtue who had only reluctantly accepted the
crown at the behest of the then-senior emperor Gratian, masks the rise ofamilitary usurper.
229Or there is Gratian’s father Valentinian, who hailed
from the same humble Pannonian stock as many of the ‘soldier-emperors’ ofthe third century (his own father, we noticed earlier, was a talented peasantsoldier who had risen to become a protector in the comitatus ,and ﬁnally
amilitary governor). In Ammianus’ judgement, Valentinian’s violent rages
(not to mention his pet bears) disclosed a temperament that was ‘patentlyferocious’ .
230ButAmmianus also judged that he ‘strove as best he could to
restrain his ﬁerce urges’ to counter his reputation for harshness, and at least
‘affected a friendly style’ at times, and his obituary-notice of Valentinian por-
traysacultivated side to the man: he wrote and spoke with eloquence and
ﬂuency, painted and sculpted gracefully, dined elegantly – and radiated ‘royal
228Cod.Theod .8.5.32, with Garnsey and Humfress 2001: 31.
229Sivan 1996 .230Amm.27.7.4.

The imperial court of the late Roman empire 227
majesty’ in his appearance and deportment.231And as w enoticed earlier,
he selected a notably cultivated teacher for Gratian, in the shape of the poet
Ausonius. These are not qualities one readily associates with third-century
soldier-emperors, and the difference is surely due in part to Valentinian’sexposure to a court society in which ‘inﬂuences of environment and edu-
cation counted for as much as those of heredity and regional origin’;
232
his father the protector had clearly had the young Valentinian reasonably
well tutored, and the Pannonian supporters who prospered at Valentinian’s
court were not illiterate troopers: they were mainly educated professionals,
lawyers and clerks with a background in the court bureaucracy.233
These are particular cases, and it would certainly be over-schematic to
ascribe to the late Roman court in general a ‘domesticating’ or ‘civilising’function of the sort that Elias sought to establish in his ‘Versailles model’ .Tocontemporary observers such as Ammianus, the court looked more like
a‘snake-pit’ in which individuals and cliques – high ofﬁcials, bureaucratic
functionaries, eunuchs, empresses – were constantly intriguing for inﬂu-ence or riches or power, and the leading courtiers lived precarious lives:seemingly unshakeable grandees like the eunuch Eutropius, or Honorius’guardian and generalissimo Stilicho, could suddenly fall from favour and besummarily exiled or beheaded; innocent men could be gruesomely torturedon trumped-up charges of seditious magic, condemned at show-trials andstrangled or burned alive; informers and spies hovered in the palace’s cor-ridors and dining rooms, reporting back to sinister imperial agents with
maﬁoso-like nicknames: under Constantius II, the prosecutor ‘Paul theChain’ would link up a sequence of charges with ingenious malice, andthe network of informers controlled by Mercurius the ‘Count ( comes )o f
Dreams’ seemed to give him access almost to one’s private thoughts.
234
Constantius, in Ammianus’ view, was a morosely suspicious individual all
tooready to believe himself surrounded by conspirators and to act on that
belief,235and the personalities of individual emperors will certainly have
affected the complexion of the imperial court at given times; some wereforceful and energetic, some were insipid and lazy, some only children whenthey took the purple. But even the capacity of a strong-willed emperortoimpose his personality on the court was necessarily limited. The dele-
gation of power to functionaries in the enlarged court bureaucracy, wenoticed earlier, was likely to generate textually established precedents and
231Amm.27.7.4; 30.8.2; 30.9.4–6.232Matthews 1975 :48.233Matthews 1975 :44–5.
234Amm.29.1.5–2.20 (show-trials and terror under Valens, AD 370–1); 15.3.4–5 (Paul and
Mercurius).
235Amm.21.16.9.

228 rowlan ds m i th
routines that reduced the scope for arbitrary autocratic interventions, and
while court ceremonial might ascribe an aura of omnipotence the monarch,
in reality the ‘remote’ emperor at the apex of the court hierarchy was nec-essarily dependent on his agents and advisors for his knowledge of howthings stood within his comitatus :what they chose to tell him must often
have been coloured by their own interests, and emperors who failed to checkthe truth of their reports were liable to be hoodwinked.
236Tosafeguard his
position, then, the monarch too had to engage shrewdly as a player at thecourt, calculating his interests and options with an eye to the impression
his actions would make on the minds of courtiers. The smooth running ofthe court, for instance, required regular procedures to be observed – butan emperor who allowed them to run wholly undisturbed risked conveyingthe impression that power lay effectively with the secretariats of the palacebureaucracy; selective interventions would always be needed to emphasisethat, if he wished, the ruler could disrupt and override the regularities. Fea-tures of late imperial court practice that on the face of things seem arbitraryor irrational or corruptive of effective government – the trumping of theregular, seniority-based arrangements for promotion in the imperial service
byacts of royal favour, the sudden and apparently arbitrary withdrawals of
support from individual grandees, the overlapping of bureaucratic func-tions within different sections of the palace secretariat and the resultinginterdepartmental jealousies – need to be appraised in that light: from themonarch’s viewpoint, they could help to prevent power devolving to anunsettling degree upon particular ministers or sectional interests withinthe court. A ‘rational’ bureaucracy, that is to say, would have served theruler’s own interest less well than the ‘patrimonial’ arrangement that in fact
obtained. Whether the way the palace bureaucracy worked had signiﬁcantlydeleterious consequences for imperial government in the round is anotherquestion. ‘Corruption’ in the form of ‘tips’ ( sportulae )and the sale of ofﬁces
was undoubtedly rife, both at a high level and in the poorly paid lowerechelons, for whose staff bribe-taking offered an obvious means of sup-plementing their salaries, but whether the cost of running the bureaucracyand the private initiatives in which its functionaries indulged did signiﬁcantdamage in the end to the government’s capacity to fund military recruitmenteffectively in the face of barbarian invasion is much less clear, and remainsacontroversial matter: A. H. M. Jones, at any rate, judged on balance that
the bureaucracy, despite its manifest ﬂaws, played an important part in the
236Amm.29.1.20 (of Valens); cf. the story about Diocletian in SHA Aur.43.1–4 (quoted earlier,
pp. 161–2).

The imperial court of the late Roman empire 229
preservation of the empire – and if his verdict has since been doubted, it has
not been conclusively disproved.237
Like the Bourbon kings of pre-Revolutionary France, the late Roman
emperors espoused an ‘absolutist’ ideology. But in practice, we have seen,no aspiring roiabsolu could rule an empire without delegating some power
toothers, and those who obtained a share of it would always potentially
pose a threat to the monarch’s personal supremacy and security. The prob-lem that the emperors had constantly to juggle with in their dealings withthe elite is a problem inherent in monarchy, a dilemma nicely conveyed to aneighteenth-century readership in Swift’s classic satire on the absolutist pre-tensions of the European monarchs of his day: ensconced with his airborne
court on his ﬂoating island, the king of Laputa ‘would be the most absolute
prince in the universe, if he could but prevail on a ministry to join with him;but these, having their estates below on the continent, and considering thatthe ofﬁce of a favourite hath a very uncertain tenure, would never consent tothe enslaving [of] their country’ . The late Roman empire, too, had its clergy-men who mocked the vanity of courtly pomp (and also its landed aristocratswho disliked paying tax), but on balance the court was an institution thatserved the interests of both the ruler and the elite quite well. ‘At the politicalcentre of any complexly organized society’ , it has been said, ‘there is both
agoverning elite and a set of symbolic forms expressing the fact that it is
in truth governing.’
238Inthe late Roman empire, the elite in question was
centred in the imperial court, and the symbolic forms expressing its capacity
togovern were expressed through a ‘court society’ (in a fashion analogous to
the Later Han arrangements analysed by van Ess in this volume). The cere-monious delegation of high ofﬁces through the emperor’s ‘sacred’ person inthe court consistory, and the grandiose symbolism of imperial self-displayin general, will certainly have helped to sustain in the minds of inﬂuentialsubjects a presupposition basic to the survival of any dynastic monarchy –the sense that power was ultimately invested in a family that was somehowdestined to rule. The emergence in the late fourth and early ﬁfth centuries ofchild-emperors who by deﬁnition ‘reigned but did not rule’ is testimony to
the strength of that sentiment: Honorius was ten when he took the purple,Theodosius II seven, Valentinian III only six. It is highly signiﬁcant that, insuch circumstances, it was not the elite ofﬁcials of the consistory who inpractice ruled collectively in the emperor’s name, but usually an empress –
237Jones1964 :601–6, 1056–64; cf. Garnsey and Humfress 2001 :47–9; for a darker view,
MacM ullen 1988 :171–97 (impact on military resources), MacMullen 1990 :68(vocabulary of
corrup tion).
238Geertz 1983 :124.

230 rowlan ds m i th
either the mother or an older sister of the emperor – or else a general who
sought to secure his position by a marriage link with the imperial family, asStilicho did. Even in the fourth century, empresses could wield considerablepolitical inﬂuence within the imperial household (Julian had good reason tobe grateful to Constantius’ wife Eusebia for her support in the 350s),
239but
in the ﬁfth their prominence and polit ical importance at court is especially
notable: they acquired their own households and designated praepositi ,and
managed their own revenues; they had coins issued in their own names;they dispensed largesse; they exercised patronage; their statues were placedin public places alongside images of emperors, and adored in their com-pany.
240The enhanced role of the ﬁfth-century ‘Theodosian empresses’
would appear to derive from a lively conviction on the part of Theodo-
sius I that the absolutist claims of imperial ideology must be underpinnedbybasic blood-ties of dynastic kinship: the majority of the ﬁfth-century
empresses were themselves daughters of emperors, women brought up inthe court society from birth, and most of them took the title Augusta on
the occasion of their marrying an emperor, or bearing him a child. In mostcases, moreover, they lived longer than the men-folk of their generation. Theprominence of the empresses at court thus afﬁrmed a dynastic continuityand transmitted a dynastic tradition. Strictly speaking, their formal stand-ing as Augustae was still a boon bestowed by the emperors, but in practice
the inﬂuence of a determined empress could by now determine particularimperial policies. It was Theodosius II’s empress-sister Pulcheria, not theemperor himself, who planned and devised a ‘crusade’ against Persia in 421towinher languid brother credit as a ‘master of victory’ , and after his death
in 450 Pulcheria reigned alone for a month – and then effectively decided thesuccession herself, by marrying a serviceable military tribune and personallyinvesting him with the imperial robe and crown. The bride was ﬁfty years oldand sworn to holy virginity, so the marriage remained unconsummated –but that merely renders it all the more telling as a witness to the symbolicrole that the royal women of the court had come to play as vessels of ‘sacred
power’ .
241
In431, Pulcheria’s actions had probably also been decisive in securing
the condemnation of the ‘heretic’ Constantinopolitan patriarch Nestoriusagainst the wishes of her brother’s own counsellors and eunuchs. That wascertainly the belief of Orthodox Christians: after her own death, Pulcheria
was revered as a saint in the company of the Virgin Theotokos ,the Mother of
239Aujolat 1983 .240Holum 1982 ;StClair1996 :158–62; McCormick 2000 :146–7.
241Holum 1977 ;1982 :208.

The imperial court of the late Roman empire 231
God.242Herbrother, too, notwithstanding their disagreement over Nesto-
rius, was renowned for his piety, and their cases could plainly be exploited
toshed light on the evolving relationship of the court and the ecclesiastical
authorities in the late empire. In the course of our discussion we have occa-
sionally touched on particulars that bear on this matter: Eusebius’ dressing-up of the ideology of monarchic absolutism in Christian theological termsin his Tricennial Oration (a panegyric of Constantine delivered at court
in the emperor’s presence in the palace at Constantinople in 336); BishopCyril’s payment of a massive bribe in his efforts to win over Theodosius II’s
praepositus ;Bishop Ambrose entering a hunting enclosure in the gardens
of the palace at Milan to seek out Gratian; Constantine’s daily prayers inadedicated chapel in the palace; the presence of bishops as guests at his
dinner-table, and of bishops sent as observers to his court from major sees.Apart from that, we have left the church aside: our discussion has leaned
particularly towards the itinerant court of the fourth century, for which theevidence is particularly plentiful and suggestive, and we judged earlier that
in the fourth century, at least, Christianity did not exert a signiﬁcant shapinginﬂuence on the structures and ceremonial of the court. But the late imperialcourt was not a static entity, and although we have not aimed to describe the
ﬁfth-century courts of Constantinople and Ravenna separately or in detail,wehave still tried to convey something of the way the court came to function
in the earlier ﬁfth century, too, at a time when these cities were emergingas ‘permanent’ capitals in the eastern and western zones of the empire.Our selection of c.AD450 as an end-date was not meant to suggest any
sudden mid-century demise or radical transformation of the ‘court society’in either of these cities; there was nothing of the sort. Nonetheless, ‘ c.450’
seems a reasonable closing point on several grounds, one of which touchesagain on the question of Christianity. The deaths of two long-reigningemperors in east and west, Theodosius II (408–450) and Valentinian III(425–455), mark the end of a dynasty that had lasted seventy years, and inthe west the institutional framework of imperial rule on which a complex‘court society’ depended was patently unravelling well before the deposi-
tion of the ‘last Roman emperor’ in 476; for the previous twenty years,barbarian generals had been blatantly ruling in the west in all but name.AtConstantinople, of course, a ‘court society’ would continue to ﬂourish
for centuries under the Byzantine emperors – but under a guiding ideologyin which the ruler’s sacral aura came to rest closely on his afﬁliation withthe Christian church and his standing as an ‘imitator’ of Christ; and in the
242Holum 1982 :163–5; 226–7.

232 rowlan ds m i th
eastern Roman empire there were developments in the mid and late ﬁfth
century that seem to presage the change to come: in 450, when Theodosius
II’s successor was solemnly invested with crown and purple robe, the cer-emonial was performed by an empress who had dedicated herself to holyvirginity as a bride of Christ, and before the century was out, the patriarch
of Constantinople was crowning late Roman emperors.
243Inthe long run,
then, Christianisation, which changed so much else in the Roman empire,would indeed come to change the face of the imperial court.
243Kantorowicz 1963 ;C a m e r o n 1976 :163, 178; Dagron, 1996 ,chs. 4–5 passim (on evocation in
Byzantine ideology of OT model of priest-king); Holum 1982 :208.

6 The imperial court in Han China
hans van ess
Introduction
Abook on the imperial court of ancient China’s Han dynasty has yet to
be written. Most specialists working in the early China ﬁeld see the court
as one speciﬁc part of central administration and descriptions subsumeit under this category.
1Thus, the court is seen as the central institution
which some ofﬁces belonged to, whereas the heads and senior membersof other ofﬁces and ministries went there from time to time in order todiscuss matters which were important for the empire. As well as being con-stituted by the imperial household, the Han court was also the place wherehigh ofﬁcials were convened from time to time, especially when court audi-ences took place at which, at least in theory, people could speak up freely.Although Michael Loewe has described the Han concept of sovereignty, hedid not stress the importance of the court as an institution.
2Of course,
there are historiographical descriptions of political intrigues going on atspeciﬁc times.
3Hans Bielenstein has characterised the history of the Later
Hanperiod in terms of power struggles between two or three large factions,
each consisting of a clan, which in turn represented several families inter-related by marriage. These factions alternately dominated the court, making
it work for the interests of the clan, which had its power-base far away fromthe capital. The question which this chapter wishes to address is whetherthe court as such, and beyond the struggles of personal interest, did actuallyplay an important role in stabilising the Chinese empire.
The court of the Han as an institution was closely related to the palace,
erected shortly after the dynasty rose to power. Hence the Han court, at leastmost of the time, was not an itinerant one. Most emperors did not travel,but instead held audience every morning in their palace. In fact, ‘ch’ao’ , thewordwhich our dictionaries translate as ‘court’ , means nothing else than
‘morning-audience’ . Pronounced ‘chao’ , the same character used for writing
the word ‘ch’ao’ means ‘morning’ .
1See e.g. Bielenstein 1980 .Anabridged version of Bielenstein’s argument can be found as
Biele nstein 1986a ;see also Bielenstein 1979 .Compare also Loewe 1986b .
2Loewe 1986a .3See for example Loewe 1974 and de Crespigny 1975 .

234 hans van ess
Asin other cultures, it seems useful to introduce for Han China the dis-
tinction between an inner and an outer court. The inner court, the enlarged
household of the ruler, was for a long time not the real centre of power.Only during the last decades of the Han did this change. In contrast to thepractice at the court of an absolutist ruler, power was generated through theinteraction of the inner court and its permanent members with the chang-ing members of the outer court, who were not necessarily present every day.Most of the ofﬁcials who came to an audience belonged to the outer court.
Thus, the court of the Han in the ﬁrst place was a ﬁxed locality within
the palace where ofﬁcials and the nobility had to come from time to time inorder to report on their duties and on other matters relevant for government.Besides the emperor himself, permanent members of the court included hiswomen and a great number of eunuchs. In addition, there were male palace-
attendants and other gentlemen-of-the-palace .These were positions usually
giventoyoung men who had ﬁrst to pass a test assessing their suitability
for a career in government. In fact, they often owed their position to thestanding of their fathers. These were only temporary members, who sooneror later were promoted to other positions, often in the provinces, far awayfrom the capital, where they had to prove their talent once again. Eventually,asuccessful ofﬁcial would come back and serve as minister or bureau-head
in one of the ministries, which would bring him into contact with the courtagain. Among these ministries, some were responsible for court affairs. Buttheir heads, too, belonged to the civil bureaucracy and usually held theirpositions for a limited time, only to receive other positions at the end of theirtermof ofﬁce. Nevertheless, this system ensured that to be at court was seen
as the crowning of a successful career by every member of the bureaucracy.
4
The Han court in a historical perspective
The period of more than 400 years dominated by the rule of the Han dynastysaw many changes. These meant that the imperial court of Kao-tsu, thefounder of the Han, did not resemble that of the last emperors of the LaterHan. Therefore, a short historical overview is needed, before we can turn to
the court itself.
4Winterling ( 1999 :35) points to the doubts recently raised, especially by French historians, as to
whether we can really talk about a ‘court’ as far as imperial Rome is concerned. They reject the
concept on the grounds that the Eliasque type of the courtier, residing in the ruler’s palace, is not
found in the Rome of the early Caesars (see Paterson and Smith in this volume). The samecriticism could, of course, apply to Han China as well. Nevertheless, as Winterling has pointedout, there is sufﬁcient justiﬁcation for using the word‘court’ for the ancient monarchies. It is his
deﬁnition which this chapter follows.

The imperial court in Han China 235
Ssu-ma Ch’ien, the major historian for the period, describes the rise
of the Han to power as unprecedented. Prior to the Han, for as long as
there had been rulers and subjects, it had taken decades or sometimes evencenturies for a ruling dynasty to be toppled by a rival family.
5According
tothe historian, since the era of the Y ellow Emperor all the families ruling
China had belonged to one and the same clan-system. There was a dynasticcycle: a dynasty normally began with a founding father, who excelled bymeans of the virtuous behaviour inherited from his father and grandfather.The ﬁrst rulers of a dynasty normally could build on what the founder hadaccomplished, thus establishing a splendid house which would bring peaceeverywhere. Subjects would come to court on their own initiative, convinced
that nobody under Heaven disposed of as much virtue and power as theruler. In the course of time, however, this virtue declined and the ruling
house degenerated. Often a dynasty would experience a so-called ‘middle-resurrection’ , until ﬁnally a virtuous founder established a new dynasty,
supported by most of the leading families under Heaven.
This paradigm could be observed with the Chou dynasty, which took
power from the Shang sometime in the middle of the last century of the sec-ond millennium BC. Their rule gradually declined until the dynasty almostbecame extinct in the ninth century BC. It was restored to power by two vir-tuous rulers, before a gradual decline recommenced in the eighth century.Powerful states began to arise in all parts of the world. Although modern
scholarship tends to see in these states independent units with no real tiesbinding them to the nominal sovereign, the ancient historiographers saythat these states were nominal vassals of the Chou; their ancestors had beeninvested by the founder of the Chou, but their power now overshadowed thecourt of the Chou king. Nevertheless, so long did the virtue of the Chou’s
founding father last that several hundred years lapsed before a new dynasty,the Ch’in, could ﬁnally unify the whole world under its own rule. As a result,according to Ssu-ma Ch’ien’s tale, in 221 BC the Ch’in ruler assumed the
role of an emperor (ti), a title which implied divine power. The Chou rulers
had been only ‘kings’ (wang), but in the centuries prior to their ﬁnal demiseso many other rulers had also called themselves ‘king’ that it seemed logicaltoﬁnd a new title for a ruler whose power was unchallenged.
The First August Emperor of the Ch’in (Ch’in Shih Huang-ti)
6thought
that he had learned the lesson of history. In order to prevent a new periodof bloodshed he decided that henceforth there should be no more feudal
5Preface to Table no. 4, ‘The Table of [Events] During the Time of the Change from Ch’in to
Ch’u, [arranged] by Months’ , SC16/759.
6Ch’in Shih Huang-ti thought that his dynasty would last for 10,000 generations. Therefore his
posthumous name was simply ‘First-Generation Ruler of Ch’in’ .

236 hans van ess
lords. He renounced the investing of members of his own family with terri-
tory,declaring that in future everyone under Heaven would be governed by
functionaries owing their position to their knowledge of law and order, nottohereditary privilege. China became a bureaucratic state, with the conse-
quence that there was no hereditary nobility at the court of the Ch’in. Y et,although the First Emperor thought that this reform would preserve ruleover the world within his own family for 10,000 generations, after only a
dozen years ﬁghting broke out again. The rivals for the throne came fromthe small ‘lanes and alleys’
7–that is, they were mere commoners, a phe-
nomenon previously unseen in Chinese history. This lack of a noble pedigreeinﬂuenced their behaviour.
Liu Pang, later canonised as Han Kao-tsu, defeated his competitors as well
as the peers who had helped him after a bloody struggle for power, doing sothrough a mixture of a peasant’s slyness and the ability to listen to peoplewho were more intelligent than himself. Allegedly he then did away withthe cruel laws of the Ch’in. However, what else he should do with his newlyacquired power was not immediately clear to him. On the one hand he wasunable single-handedly to rule a territory the size of China, which even thenwas huge, and, on the other, his family was not numerous. His solution wastosplit up the eastern part of China into eleven large kingdoms. With the
exception of Ch’ang-sha on the southern periphery, these kingdoms were
ruled by his relatives, since most of the old nobility had been killed off in
the wars leading up to the ﬁrm establishment of Han rule. These relativesreceived the title of ‘kings’ , and as soon as they had been invested they had
toleave the capital of Ch’ang-an for their kingdoms. Their relations with
the central court in western China, where the bureaucratic structure ofthe Ch’in had survived, were strained. The kings rarely came to court, andwhen they did, there were debates as to whether they should be allowedtoenjoy familial intimacy with the emperors, or be treated and comport
themselves as real subjects. Several wars during the second century betweenthese kings and the emperors ﬁnally led to a structure in which the formerwerereduced to little more than titular rulers with hardly any actual power.
Until the end of the Han period the kings were required to live in their
owncapital cities in the provinces. Usually, they were not allowed to play a
role at the imperial court. Rather, their own courts were miniatures of the
imperial court. Therefore we do not need to treat them as part of the courthere. They came as visitors to the imperial palace from time to time, at mostannually.
7SC16/759.

The imperial court in Han China 237
AfterEmperor Kao’s death, for several years his widow, Empress L ¨u, ruled
on behalf of several under-age emperors, but without making institutional
changes relevant to our topic. Empress L ¨uwas ousted from power by senior
members of the Liu family who feared that she might establish a dynastyof her own. The reign of her successor, Emperor Wen, is usually seen asan uneventful time of imperial recovery after a long period of unrest. Thisperiod of recovery was only brieﬂy interrupted by the revolt of several king-doms against the central authority in 154 BC, under Emperor Wen’s succes-sor. Their defeat marked the end of provincial lords able to challenge the rulerat Ch’ang-an. Emperor Wu acceded to the throne in 141 BC and brought thepower of the Han to its peak. He expanded imperial territory far into centralAsia and southern China, a region inhab ited by non-Chinese people. His
successor, Emperor Chao (87–74 BC), however, reigned as a minor and diedwithout an heir shortly after reaching manhood. This same problem would
recur frequently later on, resulting in a series of succession crises. At the same
time, the second half of the ﬁrst century BC saw a thorough Confucianisa-tion of many aspects of life, including the prevailing attitudes at court andin the imperial living quarters. In AD 9 this Confucianisation culminated inthe usurpation of Wang Mang, who fourteen years later was killed by forcesloyal to the Han. Kuang-wu, the founder of the Later Han dynasty, shiftedthe capital from the city of Ch’ang-an in western China to Lo-yang in theeast. New palaces and ritual structures had to be built there, and existingpalaces enlarged. Life at court, however, was modelled on the example ofthe Former Han. This worked well until the reign of the third ruler of theLater Han, Emperor Chang (75–88 AD). The ensuing period of more thanahundred years is generally seen as a time of decline. Government was no
longer in the hands of the emperors – usually minors – and their ofﬁcials,but fell into those of eunuchs and the families of imperial wives. This wasprobably the period when the Han court changed most dramatically.
The palaces and the establishment of the court
Asthe Liu family of the Han, as well as their helpers, had lowly origins,8
Liu Pang had no idea of what being emperor involved. Ssu-ma Ch’ien andPanKutell us that it took some time until the emperor understood why it
8Apparently the Liu started to claim to be descendants of the mythical Emperor Y ao (traditional
dates 2356–2255 BC) only during the ﬁrst century BC. The ﬁrst hint of this claim is to be found
in the biography of Sui Hung in Han-shu 1962 ,75/3154. Ssu-ma Ch’ien in his Shih-chi draws
attention to the lowly origin of the Liu several times.

238 hans van ess
was necessary to invest in the creation of a capital and a seat of power larger
than his immediate needs required.
Atﬁrst he saw this investment as an unnecessary luxury, and one of his
closest followers had to explain its purpose to him:
Inthe second month [of his seventh year, i.e. 200 BC, the emperor] went to Ch’ang-
an. Hsiao Ho was building the Wei-yang Palace, and was erecting the Eastern Portal,
the Northern Portal, the Front Hall, the Arsenal, and the Great Granary. The Emperorsaw their greatness and elegance and was very angry. He said to [Hsiao] Ho, ‘Theworld is full of tumultuous cries; I have toiled and suffered for many years; my success
or failure cannot yet be known – why are you building these palaces and halls beyondmeasure?’ [Hsiao] Ho replied, ‘The world is not just yet subjugated – for that reasonweshould take this opportunity to complete the palaces and halls. Moreover the Son
of Heaven has the four seas [and all within them] for his household. Without greatand elegant [buildings], you will not [be able to display] your authority and majesty.Weshould not moreover let it be that later generations should ﬁnd anything to be
despised.’ The Emperor was delighted, removed from Y ¨ueh-yang, and established
his capital at Ch’ang-an. He established the ofﬁce of the Superintendency over theImperial House to arrange the precedence among his nine [classes of] relatives.
9
This anecdote describes the creation of the physical structures of the Han
court at Ch’ang-an. It is clear that an important motive for building the
palaces was the need to express the splendour of the dynasty. For this pur-pose, palaces had to be constructed. T extual sources as well as excavationsshow that the Han erected two major and several minor palaces at Ch’ang-an. The two main palaces were the Palace of Prolonged Joy (Ch’ang-lokung), where the empress dowager lived, and the Eternal Palace (Wei-yangkung), where the emperor himself resided along with his harem. Althoughlittle of these palaces survives, excavations have shown that the total areaof the Palace of Prolonged Joy was about 6 square kilometres, which wasone-sixth of the total area of the city. Interestingly, the Eternal Palace, withan area of 5 square kilometres, was slightly smaller than the palace of theempress-dowager. Both palaces lay in the southern part of the city, theEternal Palace in the west and the Palace of Prolonged Joy in the east. In
addition there were two smaller palaces and two markets within the citywalls.
10Unfortunately, we do not know where the ofﬁcials and the nobility
lived, but we must assume that, like the commoners, they had their resi-dences outside the walls, although they may have had houses or roomsfor temporary residence within the palace, too. With the exception of the
9Han-shu 1962 ,1B/64, trans. Dubs 1944 ,118ff. Compare also San-fu huang-t’u chiao-cheng,
Hsi-an, 1980, p. 35.
10Wang Zhongshu 1982 :5 .

The imperial court in Han China 239
armoury, which had to be within easy reach of the palaces in an emergency,
it also seems that government ofﬁces were sited some distance from thepalaces.
11
Weare slightly better informed about Lo-yang, the capital of the Later
Han. Lo-yang also had two palaces, namely the Southern and the Northern
Palace. There was no clear functional division between these two palaces.
Rather it seems that the emperors moved freely between them. The SouthernPalace covered an area of 1,300 metres from north to south and 1,000 metres
from east to west. Most of the time it seems to have been the imperialresidence, although some rulers preferred the Northern Palace. The palaces
wereinterconnected by covered passageways. T extual sources tell us that the
palaces mainly served a dual function. They housed the living quarters of theemperor, his family and the harem. These were the ‘forbidden apartments’ towhich, at least in theory, nobody besides the emperor, the closest membersof his family and the eunuchs had access. Secondly, there was a public spaceconsisting of several audience halls, the most important being the Main
Audience Hall or Front Hall. Other audience halls seem to have been used
for less formal occasions. An important part of the Southern Palace housedschools for the sons of the elite and imperial libraries. There were at least twolibraries in the Southern Palace and one in the Northern Palace. In addition,there were several palace ofﬁces.
12
Within the city walls and south-east of the Southern Palace, so that their
ofﬁcials were within easy reach of the emperor, were located the ministries ofthe three highest administrative bodies of the state. The residential districtsof the high ofﬁcials and aristocrats were also located within the city walls, aswerethe Grand Store-house and (as seen earlier) the Armoury and several
markets. Commoners had to live outside the walled city.
13Unfortunately,
Lo-yang was burnt down at the end of the Later Han and, as the sourcesmake clear, its destruction was complete.
Soon after his ﬁnal victory, the founder of the Han was faced with the
problem of unruly peers arguing over their respective merits in what canonly have been a provisional palace, where they got drunk, abusing eachother and furiously hacking the pillars of the building with their swords.Only then did the emperor understand the need to introduce a propercourt ceremonial which established a ﬁxed etiquette for his followers.
14This
ceremonial prescribed the position and comportment of participants at the
11Butcompare Giele 2006 :76–81. On p. 77 Giele states that some records suggest that some of
the ofﬁces of the executive council members and ministers of the Former Han may have been
situated within the palace precincts.
12Most of the evidence has been assembled by Bielenstein ( 1976 :22–33).
13Wang Zhongshu 1982 :32–5.14SC99/2722.

240 hans van ess
emperor’s annual New Y ear audience for the nobility and the higher echelons
of the bureaucracy. It is clear that the primary aim behind the establishmentof this court ceremonial was to achieve order.
This same episode of the squabbling entourage prompted the creation of
the Han nobility and the birth of the outer court as a centre of the state.
15
Inaceremony of enfeoffment which, by the end of the Han, would take
the form of handing out a small bag ﬁlled with mud from a central altarof earth,
16Liu Pang gave all of his followers ﬁefs and the rank of a ‘hou’ , a
title usually translated as ‘marquess’ .17The practice of ennobling meritorious
subjects continued throughout the Han period, although during later reignsterritorywas not necessarily involved when a man was ennobled. In many
cases the person concerned received a title alluding to his merit and a stipend,consisting of the taxes of a certain number of households. In addition,
Emperor Wu introduced the principle of real division of the land. Thisallowed kings to divide up their territories among their sons who, with theexception of the eldest, also received the title of marquess. Until Wu’s reign
primogeniture had held sway, with the eldest son inheriting his father’s titleand territory. Although our sources claim the opposite, the new measurewas not passed out of pity for the impoverished offspring of the Liu family.Rather it has to be seen as part of the successful attempts by the dynasty toreduce the size of the kingdoms.
Under Kao-tsu and during the following two reigns of his son Emperor
Hui(194–187 BC) and widow Empress L ¨u(187–180 BC), the marquesses –
unlike the kings – were allowed to stay in the capital. The main positions inthe bureaucracy inherited by the Han from their Ch’in predecessors weredivided among them. Most of the second level of the administration, theso-called nine ministers, was also ﬁlled with former military supporters ofLiu Pang. They comprised the Grand Master of Ceremonies, the Super-intendent of the Imperial Household, the Commandant of the Guards, theGrand Coachman, the Commandant of Justice, the Grand Herald, the Direc-
torofthe Imperial Clan, the Grand Minister of Agriculture and the Privy
15Cf. Butz and Dannenberg 2004 :11–12: ‘The court appears now as a centre of power which
served the aim of attaching those persons to the prince whose support he necessarily needed for
his self-preservation.’
16HouHan-shu 1965 ,chih 9/3202f, com. 8. The passage has been translated by M ¨uller ( 1979 :
310–19).
17The question as to whether the system of the early Han should be described as ‘feudal’ is, ofcourse, disputed. In the light of the source mentioned in the previous footnote, my own view is
that we should be allowed, for the sake of simplicity, to speak in terms of ‘ﬁefs’ and ‘enfeofﬁng’as far as this period is concerned. Compare the detailed guidelines for enfeofﬁng of a feudallord in Po Hu T’ung, trans. Tjan Tjoe-som 1949 and 1952 ,ptVII, 410–12.

The imperial court in Han China 241
Treasurer.18All these ministers belonged to the outer court: that is to say,
they were working in localities which (if the archaeological data provided by
Wang Zhongshu can be trusted) were clearly separated from the palace itself,
although from time to time they had to come to the emperor’s audiences todiscuss important measures. These ofﬁcials, then, on the face of it were not‘courtiers’ as in the Versailles-based analysis of Norbert Elias. Their primary
function was bureaucratic.
Among the nine ministries some deserve special mention. The Super-
intendent of the Imperial Household was responsible for security within thepublic parts of the palace compounds and for protecting the emperor whenhe left the palace. The private apartments were protected by the eunuchs.
19
Under the superintendent’s authority fell the palace-attendants and other
palace gentlemen who were armed and served as imperial bodyguards. Inaddition to this protective function, the superintendent and his staff had toprovide the emperor with advice and criticism. Finally, the superintendentwas also in charge of overseeing seventy internuncios or receptionists. Onbehalf of the emperor these people undertook various activities, such asoffering condolences on the death of less important ofﬁcials. All in all,within the palace precincts there were probably between several hundred
and over a thousand ofﬁcials under the superintendent’s authority.
Some of his duties overlapped with those of the Commandant of the
Guards, whose ofﬁce under the Former Han was also located within the
palace precincts. The guards controlled the palace entrances and apparentlypatrolled within the palace too. Under the Later Han, this ministry lay outsidethe palace and was staffed with about 3,000 guards.
Aloys Winterling points out that typically the medieval European courts
wereorganised around four departments, namely the treasury, the provision
of food and of drink, and the stables.
20Although we cannot ﬁnd exactly the
same structure in Han China, all four aspects were important there as well.The function of the Grand Coachman was to supervise the imperial stablesand at the same time the horses for the army. There were several stables andcoachhouses in the capital and at least one or two in the palace itself, and we
can assume once again that large numbers were employed there. The ofﬁceof the Privy Treasurer was particularly important. He was the only ministerserving the emperor alone, supervising all those who – mostly as eunuchs –waited on the emperor inside the palace.
21Inaddition to the emperor’s
private purse, he also oversaw payments for his food, drink, clothing, etc.
18Iamfollowing the translations of Bielenstein ( 1980 ).
19Biele nstein 1980 :23.20Winterling 2004 :83.21Biele nstein 1980 :47.

242 hans van ess
The management of these court activities must have been a complex task,
not least because their cost was funded by the emperor from his own purse,separate from the government treasury.
22
Itis not possible and – since the work done by Bielenstein ( 1980 )–
certainly not necessary to enumerate and describe all the ofﬁces at the Han
court. The important point is that under the Former Han there were two
major divisions, one between the nobility and the commoners, the otherbetween the military and civilians. As was said above, during the reigns ofthe ﬁrst three Han sovereigns (the third being the only Han empress whoactually ruled
23), nobles were allowed to stay at court. The fourth emperor,
Wen(180–157 BC), who like Emperor Hui was a son of the founder of the
dynasty, tried to oust the faction formed by his father’s military supporters
byordering the marquesses to leave the capital for their ﬁefs, which were
often far away. Even so, the nobility created by the founder of the dynastycontinued to play an important role until the accession of Emperor Wu in
141 BC. Full marquesses were expected to reside in their ﬁefs, and we knowof cases where the heir to a marquessate renounced his title so as to play a rolein government. Y et, exceptionally, some enfeoffed nobles were permitted tolive in the capital.
The reign of Emperor Wu as a watershed in the development
of the Han court
Emperor Wu introduced the custom of ennobling the highest member of the
bureaucracy, the chancellor, as a marquess. At this time the many descen-dants of the former military supporters of Kao-tsu by far outnumberedthe marquesses enfeoffed under his successors,
24and still dominated life at
court. Succeeding as a mere sixteen-year-old, Emperor Wu sought to infuse
22The Privy Treasurer was responsible for the management of the emperor’s purse during the
Former Han, whereas under the Later Han the Grand Minister of Agriculture performed this
function. Bielenstein 1980 :4 6and 55.
23The case of Empress L ¨uiscomplex. As the chief wife of the founder of the dynasty she was very
powerful. In name, two child-emperors followe dEmperor Hui on the throne. Y et, Ssu-ma
Ch’ien and Pan Ku devoted an imperial annals section to Empress L ¨uinher own name,
obviously because the two minors either were not taken seriously or were seen as illegitimate.This matter is not easy to decide, since L ¨uisone of only two empresses in Chinese history who
managed actually to rule (the other one being Wu Tzu-t’ien of the T’ang). Both women havebeen described as cruel and ruthless.
24SC18 lists 143 men who were ennobled by Kao-tsu, whereas SC19 has only three men
ennobled by Emperor Hui, 31 by Empress L ¨u(most of whom where deposed upon her death
and the subsequent ousting of her faction), 29 by Emperor Wen and 30 by Emperor Ching.

The imperial court in Han China 243
new blood into his bureaucracy. Probably he was trying to free himself from
the forces in power over the last sixty years, who did not want a strong-willed emperor. The court not only offered the ruler a means of controlling
and structuring the families of people who were his nominal subjects, butalso controlled the emperor himself.
25One important constraint on the
emperor came in the form of young ofﬁcials serving as palace-attendantsand thereby temporarily forming part of the inner court; their main dutywas to remonstrate. Some may have done so independently, but many wereclosely connected to the households of senior members of the outer court,that is, marquesses and high ofﬁcials. They not only protected the emperor,but also guided him.
Forastrong emperor who wanted to free himself of such restraints it
was crucial to recruit fresh blood into the nobility and the bureaucracy.Emperor Wu sought to achieve this aim by two means. First, he introducedan examination system requiring many (but by no means all) candidatesfor ofﬁce either to master one of six disciplines attached to the Confuciancanon, or to become experts in the legal code. Taking almost a century tobecome ﬁrmly established, this system at ﬁrst was regarded with disdain byour major historiographical source for the period, since it promoted pettyscholars lacking the broad experience needed to govern even a provinceor a provincial district, let alone a state. They were tools in the hands ofthe emperor, who could manipulate them for his own ends.
26Although in
theory the ruler always occupied the pinnacle of the hierarchy, many saw thecourt as the centre for the brokerage of power. On this view the emperor was
not all-powerful, but, as the nominal head of the system, had to be drivenin the desired direction.
Secondly, under Wu the bureaucracy as well as the nobility was trans-
formed by the fact that some people were now admitted to the court throughtheir personal relationship with the emperor. This development will beexplored in more detail in the next section.
When Emperor Wu came to the throne, the empire enjoyed peace and had
recoveredits natural prosperity after a period of decline. This was an ideal
situation for a young ruler with many ambitions. He wanted to expand thefrontiers of his empire in all directions. Since this enterprise involved a long
25As Biele nstein ( 1980 :143) put it, the Han emperors were no autocrats.
26There is as yet no scholarly consensus on the posit ion of Ssu-ma Ch’ien. The interpretation
givenabove is the one I am going to argue for in a forthcoming book which compares the two
major sources for this period, namely Shih-chi and Han-shu (see Appendix). However, many
scholars tend to the opinion that Ssu-ma Ch’ie nwas actually a Confucian and that he agreed
with many of Emperor Wu’s measures. See, e.g., Hardy 2000 .

244 hans van ess
period of war, it obviously ran counter to the interests of the old nobility,
whose taxes had to fund imperial ambit ions. A series of depositions of
feudal lords began. In many cases the heirs of the founding nobility weresimply accused of some crime, and then either beheaded or allowed tobuy themselves off, becoming commoners or serving in the army. The ﬁefsreverted to the state. By the end of Wu’s reign almost none of the old nobility
survived. They were replaced by military ofﬁcers on the basis of merit. Asthe wars continued, however, many of these new nobles were deposed aswell. Therefore, there was great instability in the composition of the outer
court. Nobody could be sure that he would retain his position from one day
tothe next. Whereas some commoners rose to important positions rapidly,
their fall from grace could be even quicker.
Manyscholars entering the bureaucracy and thus becoming members of
the outer court under Emperor Wu came from the eastern empire. Theychallenged the position of the old elite in the capital, Ch’ang-an, which lay
in western China. Among the social obligations of the old elite, its mem-bers were expected to assert their inﬂuence and status by entertaining hugecrowds of so-called guests and retainers. As with Elias’ portrayal of the nobil-ity of medieval Europe, they were under pressure to spend money on theirentourages and build up courts of their own. Under Wu there was a markedchange: the emperor viewed with suspicion any subject who displayed too
muchwealth and power.
27Thus, the social role of the nobility changed
radically. They no longer dared to entertain lavishly.
Of course, the historians draw this picture for one period only; if a true
picture, it certainly did not last. Although the examination system was nom-inally based on merit, once ﬁrmly established and accepted throughout theempire it became a convenient means for a small group of rich families tosecure their local power. It is highly unlikely that the Han Chinese exam-ination system encouraged a high degree of social mobility. The sons ofimportant families received an education in a certain set of texts, therebyproving that they were ﬁt to occupy the most important positions in thestate. Although a few poorer men may have risen into the ruling class bythis route, education was much too expensive to be available for all. Thisbecomes very clear when we look at the history of the Later Han. In thisperiod, mastery of a canonical discipline had become the chief means ofadmission into the bureaucracy. But upward mobility remained virtuallyimpossible for individuals without the protection of one of the dominant
27SC111/2946, HS55/2493. On the phenomenon of the reduced social role of the new elite cf.
also Fu Lo-ch’eng 1977 :23. I have argued for this in more detail (van Ess 2004 ).

The imperial court in Han China 245
two or three factions, which together comprised no more than ten or twelve
big clans.
However,as a result of the reforms and changes under Emperor Wu, some
forty or ﬁfty years after his death the rhetoric of power had become markedlydifferent. Whereas the old elite at court had entertained lavishly, often creat-ing small courts of their own, in later times such conspicuous consumptionbecame much less common. We often ﬁnd inﬂuential persons claiming apoor background, although we know that they were not poor at all.
28These
claims are contradicted by descriptions of the wealth and afﬂuence of theinﬂuential families found in many sources.
29Confucian rhetoric demanded
modesty from the elite, although social reality was certainly very differentfrom the ideal; as a result, interpretation of our sources can be problematic.Furthermore, after Wu’s reign noble rank no longer guaranteed inﬂuence at
court. Rank usually followed power, rather than conferring it.
Role and function of the emperor at court
Asmentioned already, the word ‘emperor’ , used posthumously of a ruler,
implied a divine nature. The same is suggested by the title ‘Son of Heaven’ ,bywhich formulation subjects often referred to the living emperor. The
emperor was not supposed to mingle freely with the common people,although there are many popular stories about emperors doing so. Nor-mally the emperor was seen outside the court only on solemn occasions of aritual and religious, or a ceremonial, nature. The two main types of sacriﬁces
reﬂected the fact that the emperor at one and the same time ruled over the
whole empire and was the head of the imperial family. First, there were thesacriﬁces of the state cult, offered on behalf of the empire, and second, thosefor the imperial ancestors. The ancestors were buried in funerary moundsoutside the capital; the sacriﬁces for the state-cult took place at religiouscentres inherited by the Han from the Ch’in dynasty. These sites lay some
100 kilometres to the north and west of the capital – further than the tombs,but still within the region of the capital.
30
Apart from sacriﬁcing, the emperor’s main duty was to hold audiences,
which allowed his ofﬁcials to present governmental and legal business. Han
28Biele nstein 1979 :88. Here we see a big difference from the grandiose self-representation which,
according to Norbert Elias, ruined sections of the European nobility in the seventeenth and
eighteenth centuries.
29Foraconvenient overview of these questions of wealth, see Ch’ ¨uT’ung-tsu 1972 :181–5.
Compare also Y ¨uYing-shih 1967 .
30Cf. Loewe 1992 .

246 hans van ess
intellectuals were strongly of the opinion that the emperor should inter-
fere as little as possible in public affairs. Assessing the rule of Empress L ¨u,
Ssu-ma Ch’ien alludes to the concept of Lao-tzu: the emperor rules theempire without acting and without leaving his house. The people never seehim, yet the empire is well governed.
31The same is true for military affairs.
Although the emperor discussed these with his generals, military theoristsof the time believed that as a non-specialist with no military training thesovereign ought to leave all decisions in the ﬁeld to the general-in-chief. Assoon as he left his palace and entered the camp, he was no longer the supremecommander but had to bow to specialist knowledge.
32Yet, despite the fact
that ofﬁcials worked out everything, the emperor nevertheless retained averyimportant function: his was the last word. Every political decision,
no matter who drew it up, had to be issued in the form of an edict fromthe emperor himself. In other words, if the high ofﬁcials exerted executivepower, the emperor fulﬁlled the legislative role in governance.
Because of his many functions, the life of the emperor was highly ritu-
alised. The chapter of the Li-chi called ‘Jade Ribbons’ (Y ¨u-ts’ao) describes
the different robes worn by the emperor on different occasions during theyear. At the routine morning audiences he had to wear a leather cap.
33He
changed his robes for breakfast and for lunch. A scribe sitting to his left
recorded the emperor’s acts when he was seated, whereas a scribe sitting to
his right did this when he was up and about. Many other texts prescribe themorning audience as obligatory for the emperor. Tso-chuan ,ahistorical text
from the turn of the fourth and third centuries BC, says: ‘The gentleman hasfour times: In the morning he listens to matters of government, at daytimehe makes visits and inquiries, in the evening he improves the ordinancesand at night he reposes.’
34
Shuo-y¨ uan,acollection of anecdotes compiled under the Former Han,
several times praises rulers who rose early in the morning to hold audience.35
31The quotation comes from SC9/412. Lao-tzu chapter 47: ‘During the time of the Hsiao Hui
Huang-ti (The August Emperor Hui the Filial) and Kao-hou, the black-haired [common]
people were able to part from the suffering of the Warring States, and lords and vassals all
intended to rest in non-action. For this reason, Hui-ti let fall his robes and folded his hands,Kao-hou as a female ruler announced [she would issue] the imperial decrees and managed to
govern without going out of her chambers, and the world was peaceful’ (trans. Nienhauser1994 –2002 ,II: 137). At the end of the treatise on the rites, as well as in the biography of his
father and his autobiography forming the last chapter of the Shih-chi ,Ssu-ma-Ch’ien stresses
the idea that the emperor should not leave his palace ( SC23/1173 and 130/3292).
32See for example SC57/2074, 102/2757–9.
33SSC 1473C–1474A, Couvreur, Li Ki, 678.
34Tso-chuan ,D u k eC h a o ,ﬁ r s ty e a r .S e eL e g g e 1861 –72, V: 580.
35Chapter 1, ‘The Way of the Ruler’ .

The imperial court in Han China 247
The didactic nature of these anecdotes may raise doubts as to whether all
emperors really complied with their ceremonial and ofﬁcial obligations.The common-place admonitions of Confucian ofﬁcials include warning anemperor against neglecting his duties, which often could be said to presage
even the fall of the dynasty. Ofﬁcials were obliged to remonstrate with their
rulers,
36and often did so if we believe our historical sources. Hence an
audience was not necessarily a pleasant experience for the emperor, althoughthese occasions could also be dangerous for ministers who transgressedcertain unwritten but nevertheless well-deﬁned boundaries. Because of these
risks, few chancellors under the Former Han died a natural death.
Intheory, the emperor held a court audience for his ofﬁcials every morn-
ing at day-break, before breakfast. On these occasions he received ofﬁcialswho had already handed in memorials; these had been checked in the interimbythe responsible authorities, who had deemed them worthy of discussion.
Usually several high ministers were present at these court debates. Some
proposals were passed on for deliberation by a larger circle, comprisingnot just the higher echelons of the bureaucracy but also specialists, such asexperts in canonical scriptures, much in demand in matters of ritual. Some-
times a minister requested a private audience, which in exceptional caseswas granted. The other discussants would be ushered out, and the ministercould speak openly. Once a high ofﬁcial managed to convince the emperor
that an opponent had been beheaded, even though the emperor had justseen this person in the previous audience.
37Court ofﬁcials were supposed
tobe present daily, but every ﬁfth day was a so-called bathing-day, when
they could go home and look after family affairs.
Under the Later Han, when most emperors were children, court audiences
wereoften held by the empresses, who listened to ofﬁcials from behind
acurtain. Once a year the kings and marquesses were obliged to attend
an audience and bring presents. On such occasions a drinking party wasorganised, and the dynasty celebrated itself and put the splendour of theimperial house on display. However, when Emperor Wu needed money forhis wars, he ordered each marquess to present a golden sacriﬁcial vessel, madetoan exacting standard, at the annual audience. Many of those who came
could not afford this and were deposed accordingly.
38Thus, the emperor
did not always use conspicuous consumption to reveal the power of thedynasty: instead, at times he used his position to collect the resources which
he needed.
36Tjan Tjoe-som 1952 :463.37SC101/2742 and HS49/2273 and 2300f.
38SC30/1439.

248 hans van ess
Foreign guests were received in a ceremonial audience too. Whether these
‘barbarians’ should be given precedence in seating arrangements over the
Hannobility was a hotly debated matter.39Notonly this question but many
others relating to court ceremonial were subject to constant change and
heated discussions. A compromise on many of these issues was agreed underEmperor Chang, the third ruler of the Later Han, following a long discussionin the so-called Hall of the White Tiger. Pan Ku is said to have been thecompiler of the decisions surviving under the name of ‘Comprehensive
Discussions in the White Tiger Hall’ ( Po-hu t’ung-i ). Although this is one of
our best sources for the actual treatment of ritual questions under the Han,probably it was the solution only of a given historical moment.
If we accept the evidence of such works as the Po-hu t’ung-i or several
chapters of the Li-chi ,wehavetoconclude that the minutiae of the life of the
Hanemperors were no less prescribed than those of Louis XIV as evoked by
Elias (following Saint-Simon). It was not just their daily routine which seemstohave followed ﬁxed rules. In addition, their duties also included roles in
ceremonies attached to several festivals during the year. The beginning of the
chapter ‘Monthly Ordinances’ (Y ¨ueh-ling) of the Li-chi ,o n l yo ne of several
comparable texts attached to the calendar, serves as a good example of how
ancient normative texts prescribe imperial behaviour:
Inthe ﬁrst month of spring . . . the Son of Heaven occupies the apartment on the left
of the Ch’ing-yang (Fane); rides in the carriage with the phoenix (bells), drawn by
the azure-dragon (horses), and carrying the green ﬂag; wears the green robes, andthe (pieces of) green jade (on his cap and at his girdle pendant). He eats wheat andmutton. The vessels which he uses are slightly carved, (to resemble) the shooting
forth (of plants).
Inthis month there takes place the inauguration of spring. Three days before this
ceremony, the Grand Recorder informs the Son of Heaven, saying, ‘On such and
such a day is the inauguration of the spring. The energies of the season are fullyseen in wood.’
40On this the Son of Heaven devotes himself to self-puriﬁcation, and
on the day he leads in person the three ducal ministers, his nine high ministers, thefeudal princes (who are at court), and his Great ofﬁcers to meet the spring in theeastern suburb . . .
Inthis month the Son of Heaven on the ﬁrst day prays to the Powers on high for
agood year: and afterwards, the day of the ﬁrst conjunction of the sun and moon
having been chosen, with the handle and share of the plough in the carriage, placedbetween the man-at-arms who is its third occupant and the driver, he conducts
39Discussed, e.g., by Chia I. See HS48/2240.
40Inthe Chinese system of Five Elements (Wood, Fire, Earth, Metal and Water), wood is the
element of spring since it is associated with the green colour of the young shoots.

The imperial court in Han China 249
his three ducal ministers, his nine high ministers, the feudal princes and his Great
ofﬁcers, all with their own hands to plough the ﬁeld of the Powers [on high]. TheSon of Heaven turns up three furrows, each of the ducal ministers ﬁve, and the otherministers and feudal princes nine. When they return, he takes in his hand a cup inthe great chamber, all the others being in attendance on him and the Great ofﬁcers,and says, ‘Drink this cup of comfort after your toil.’
41
Similar regulations follow for every single month of the year. At ﬁxed inter-
vals the emperor has to change his robes from green (or blue) to red, the
colour of summer, and yellow, the colour of the centre and midsummer,
in turn switching to white in autumn and black in winter. Of course, it isunlikely that the Han rulers followed all these prescriptions.
42
Another task of the emperor, the tour of inspection, was handled in dif-
ferent ways over the four centuries of Han rule. Whereas the founder of thedynasty came to know the empire from his wide-ranging wars, his succes-
sors did not travel very much. Emperor Wen grew up in a frontier provinceadjoining the territory of the Hsiung-nu and returned several times to thescene of his childhood. But his son, Emperor Ching, stayed in the vicinity ofthe capital throughout his life, as did Emperor Wu during the ﬁrst thirty yearsof his reign. In the years following 110 BC, however, Wu often travelled forthe sake of a ritual at Mount T’ai in eastern Shan-tung. We should certainlytake seriously the emperor’s religious motivation on these occasions, whichformed part of a larger search for immortality. But we should not forget thatthese journeys also provided an opportunity to get to know the empire muchmore thoroughly than his predecessors. Y et the historian criticises Wu’s trav-els for the damage which they did. He claims that two provincial governorscommitted suicide because they were unprepared for the imperial visit and
had not been able to procure the huge amount of provisions needed to feedthe imperial entourage.
43Nolater emperor travelled as much as Emperor
Wu,despite the fact that Po-hu t’ung-i justiﬁes imperial tours of inspection
and prescribes them at least every ﬁve years.44Once again this formula may
reﬂect an attempt to reconcile the advocates of a strong imperial presence in
the provinces and those in favour of the emperor staying in the capital. Thelatter faction was to retain the upper hand under the Later Han and during
41Li-chi ,Y¨ueh-ling, trans. Legge 1885 :249–55, slightly altered.
42Weknow that a ritual structure cited in the work just quoted, the so-called ‘Hall of Light’
(Ming-t’ang) where governmental orders were proclaimed, was not built in the capital until the
end of the Former Han, although some Confucian scholars had tried to convince Emperor Wuof its necessity much earlier, around 140 BC. A hall of this type evidently existed throughoutthe Later Han at the new capital of Lo-yang.
43SC30/1438.44Po-hu t’ung nos. 128–30, especially 130, Tjan Tjoe-som 1952: 497.

250 hans van ess
muchof imperial China’s subsequent history. Keeping the emperor in his
palace saved money and gave the bureaucratic class the opportunity to exert
power unhindered. More importantly for understanding the signiﬁcanceof the court as an institution, when the emperor stayed at home, access tohim was limited to inner and outer court members. Other channels wereclosed.
The emperor a tl e i s u r e
The abundance of material on ritual should not mislead us into thinkingthat the emperor had no life of his own. The sources provide a vivid pic-ture of leisure at court, especially for Former Han times. The two majorancient histories of the Former Han period, The Records of the Historian
(Shih-chi )and The Book of Han (Han-shu ), tell us that Emperor Wen used
toattend private parties at the home of his favourite T eng T’ung
45and that he
once climbed a mountain together with his favourite concubine, savouringthe view and pointing out her home-town.
46However,in this connection
wealways ﬁnd scholar-ofﬁcials who remonstrate with the emperor and
tell him that he should stick to the etiquette requiring the concubine to
remain a step lower than the empress, despite the emperor’s love for her.
Emperor Wu is well known for having left the palace incognito, and fromlater times there is a whole genre of literature telling how he went to eat-ing houses and mingled unrecognised with the common people. When hewas young, just after his accession, one of his greatest pleasures was hunt-ing. On these outings he would roam far and wide with the best ridersand archers, both from the palace and from whichever region he wasvisiting:
Some time before this, in the third year of the chien-y ¨uan era [138 BC], the emperor
ﬁrst began going out on incognito expeditions . . . The emperor was accompanied by
the gentlemen in constant attendance and arme driders, along with the sons of good
families from Lung-hsi and Pei-ti who had been assigned to await his command andwho were skilled at horsemanship and archery. They were instructed to rendezvousat the gate of a particular palace . . . The group would set out in the evening, at thetenth notch of the water-clock, announcing themselves as the party of the marquess
of P’ing-yang,
47and by dawn would reach the foot of the Southern Mountains.
There, they would pursue and shoot at deer, boar, fox, and rabbit, and wrestle bare-handed with black and tawny bears, galloping wildly through the ﬁelds of rice and
45SC96/2683.46SC102/2753.47This was the ﬁef of Emperor Wu’s sister.

The imperial court in Han China 251
grain while the common people all shouted and screamed curses at them. Once
the people got together and reported the matter to the magistrate of Hu-tu, who
went and asked for an interview with the marquess of P’ing-yang. The horsemen
wereabout to set upon the magistrate with their whips, but the magistrate in a rage
instructed his clerks to shout at the hunters and command them to halt. Several ofthe horsemen were accordingly detained, and it was only after they had shown himthe carriages and other equipment and revealed the true nature of the expeditionthat they were at last able to obtain release . . . The emperor went on a number ofsuch trips, but he was still somewhat intimidated by his grandmother, the empressdowager, and did not dare to venture very far away.
48
Later, Wu is said to have become aware of the trouble which his hunts
caused the common people and to have established the Shang-lin park,where imperial hunting parties could take place without disruption to rurallife. Pavilions and pleasure seats were erected in the countryside, staffedwith servants, maidens and even concubines. Wild animals were introduced
into the park, along with exotic fruits from central Asia. However, althoughritual texts depict hunting as a necessary exercise since it served as a military
training,
49literati often criticised emperor sfor their hunting expeditions,
claimed as dangerous and expensive.50Once again we see here the tendency
to reduce the emperor to his role as a performer of sacriﬁces and the focus
for discussions about policy. The emperor was needed to provide a centre,but he did not have to take an active part in decision-making. A centre hadtobe there, hollow though it may have been.
Other pleasures were considered less harmful. We know of the existence
of jesters at court, although once again the written sources stress that oneof their functions was to remonstrate with the ruler, albeit in a humorousway. Slabs from Han tombs carry reliefs on which jesters are clearly visible,suggesting that we should not take our sources too seriously in this context.Itseems very likely that jesters were employed as professional entertainers
for the frequent court banquets.
Music should not be overlooked either. Ritual texts stress the impor-
tance of music in several matters, including such different contexts asthe ceremonies in the temple of the ancestors and educating the people.Archaeological ﬁnds, including complete sets of bells and chime-stones,
suggest that music was also an important part of entertainments at court.Emperor Wu fell in love with dancing girls at least twice; the same befell
48HS65/2847, trans. Watson 1974 :83–4.
49This was very much the case with the nomadic people who occupied China in later times, and it
is a feature recognisable in other cultures, e.g. ancient Greece.
50SC101/2740 against racing, SC117, Ssu-ma Hsiang-ju change from hunting to literature.

252 hans van ess
Emperor Ch’eng. Lady Fu, one of Wu’s palace ladies, came from a fam-
ily of musicians and singers. Her elder brother had been castrated afterbreaking the law. When she came into the palace, he too was summoned,becoming an imperial favourite. He was given the task of introducing newmusic in place of the old style, now considered out-of-date and dull.
51
Forcourt entertainments Wu also established a Bureau of Music. How-
ever, this was abolished a century later because its melodies were considered
licentious.52
All in all, the early emperors seem to have been much freer in their daily life
than the later ones. Literature was the only imperial pleasure which remainedunchallenged until the end of the Han, replacing hunting and music in thelater reigns, to judge from the sources.
53Under emperors Ching and Wu we
know of several kings who tried to become Maecenas-like patrons of poetsand philosophers. They invited these people to their courts either for simplepleasure or because they wanted to comply with the old Chinese ideal ofthe sage king who governs by means of virtue and who exerts a civilisinginﬂuence on his people. These efforts were not always welcome to the centralgovernment or the emperor himself because they were seen in some casesas signs of intolerable ambition.
Anearly example of such a patron-king was the king of Huai-nan, the
compiler of the ﬁrst summa of Taoist thought. Accused of an attempted
rebellion, he was forced to commit suicide. Somewhat later, another example
concerns a king who asked the imperial government to obtain for him a copy
of Ssu-ma Ch’ien’s Records o fthe Historian .The request was refused on the
grounds that this text contained too many strategies and tactics from the time
before the empire’s uniﬁcatio n–ac lear sign that historical literature was
deemed to be simultaneously both precious and subversive.
54And whe nthe
Later Han disintegrated at the end of the second century AD, local warlordsarose who competed to attract the best literary talents as a kind of brains-trust. It should not be forgotten that the recitation of poems was a form
of entertainment for rich kings and the emperor himself. But poetry – atleast the serious kind – was always seen as a means of indirect criticismand remonstrance. There were long discussions about the poetic tensionbetween wishing to criticise and wishing to write beautiful lines of artfulpoetry. Criticism is what was asked for in theory, but in practice the aspect
of entertainment was becoming more and more important.
55
51125/3195.52Biele nstein 1980 :52; Loewe 1974 :193–210.
53Inthis respect, the Han court again resemble dthe medieval European court. Butz and
Dannenberg 2004 :25.
54HS80/3324f.55Yang Hsiung.

The imperial court in Han China 253
The role of the harem
One factor which guaranteed a family the acquisition of inﬂuence and power
at court was the successful introduction of one or more daughters intothe imperial harem. The role of the harem in Han times can hardly beoverestimated. During the Later Han it was probably the most important
court institution because – as will be explained – it formed the ‘hinge’
between inner and outer court. However, once again we need to realise thatthe institution underwent major changes over time. Normative texts fromChina’s early times on stress the function of a king’s chief wife as the ﬁrst ladyof the palace.
56They celebrate the virtue of a young woman who is selected
to become the ruler’s chief wife. The chapter on marriages in Records of
the Rites (Li-chi )saysthat of old the Son of Heaven had one main wife
(hou), three ladies (fu-jen), nine imperial concubines (pin), twenty-sevenchamber wifes (shih-fu) and eighty-one ‘women to sleep with’ (y ¨u-fu).
57
The ruler was regarded as the father of the empire and his chief wife as itsmother. Therefore, her position was extremely sought-after. The family andconnections of a woman who became empress could expect to enjoy wealth
and favour. The passage from the Record so ft h e Rites probably has to be seen
as an attempt to systematise a much more complicated reality prevailing atthe Han court – where indeed there was, however, a bureaucratised haremsystem. At the beginning of the Han there were six ranks of palace ladiesbelow the empress. Emperor Wu adde dfour additional ranks, and Emperor
Y¨uan (48–33 BC) a ﬁfth. At the end of the Former Han the number of harem
women is said to have totalled about 3,000.
58Manyof these women from
time to time were released and married off, especially after an emperor’sdeath.
59
Kao-tsu is well known for his sexual adventures. Fearing for the inﬂuence
and the very survival of her own clan, his chief wife, Empress L ¨u, threatened
tokill every concubine whom the emperor favoured. This was one reason
why it was difﬁcult to ﬁnd a suitable candidate for the throne after her owndeath. Fortunately, a concubine whom Liu Pang had once favoured hadleft the harem to follow her son to a dangerous border region where hehad become king when still a minor. Another woman had originally beenaconcubine of a king whose territory the emperor once visited. Offered to
the emperor on this occasion, she fell pregnant and bore a son. But shortly
56E.g. the ﬁrst poem in the canonical Book of Odes: Legge 1861 –1872 ,IV: 1–4.
57SSC 1681C.
58Biele nstein 1980 :73f.59See, e.g., HS4/123.

254 hans van ess
afterwards she was thrown into prison when the king in question fell out of
favour. This behaviour by the founder of the dynasty later made his son bythis poor concubine the source of serious problems.
60
The father of Emperor Wu changed his chief wife three times. The his-
torian Ssu-ma Ch’ien gives us all the details about the intrigues leading
tothe deposition of the crown prince, who was replaced by the son of
the third empress. The imperial women, and probably also their families,forged alliances in order to succeed in this extraordinarily important mat-ter.Emperor Wu himself had inherited the sexual proclivities of his great-
grandfather Kao-tsu. He preferred commoner girls over ones from good
families. Ssu-ma Ch’ien and Pan Ku report the following about the man-ner in which the second empress of Emperor Wu gained admission to thepalace:
Empress Wei’s name was Tzu-fu. She was of very humble birth.61Herfamily bore the
name of Wei and came from the village of the marquess of P’ing-yang, the husband
of Princess P’ing-yang. Tzu-fu was a singing girl in the household of Princess P’ing-yang. After Emperor Wu ascended the throne several years passed and he was still
without a son. Princess P’ing-yang, his elder sister, had sought out ten or more
girls of good family whom she dressed in ﬁne clothes and kept in her house, and
when Emperor Wu was returning from a sacriﬁce at Pa-shang and stopped to pay avisit with her the princess showed him the beautiful young ladies she had gathered
together to serve in her house. The emperor, however, was not impressed by any
of them. After wine had been served the singing girls came forward to perform.Looking them over, the emperor found only Wei Tzu-fu to his liking. In the courseof the day, when he rose and left the hall to change his clothes, Tzu-fu went alongtowait on him and assist him with his robes, and in the passageway he bestowed
his favour on her. The emperor returned to his seat, looking exceedingly pleased,and presented the princess with a thousand catties of gold. The princess accordinglyoffered him Tzu-fu, and it was arranged that she should be sent to live in the palace.
When Tzu-fu was about to climb into the carriage Princess P’ing-yang patted her
on the back and said, ‘Go! Eat well and take care of yourself. If you should becomehonoured, do not forget me!’
Tzu-fu remained in the palace over a year, but was never in that time favoured
bythe emperor. The emperor in the meantime had decided to select those ladies in
waiting who were performing no useful function and dismiss them and send themhome and, when Wei Tzu-fu was granted an audience with him, she wept and beggedto be se nt away with the group. But the emperor felt sorry for her and favoured
her once more. In time she became pregnant and the honour and privilege which
60Hewas obliged to commit suicide after an attempted rebellion ( SC118/3083–94, Watson 1961 ,
II: 359–87).
61Probably a pun. The surname Wei is a homophone of ‘wei’ , which means ‘small’ , ‘lowly’ .

The imperial court in Han China 255
she enjoyed increased daily. Her older brother Wei Ch’ang-ch ¨un and her younger
brother Wei Ch’ing were summoned and made palace attendants. After this Tzu-fu
enjoyed extraordinary favour with the emperor. She bore him three daughters andas o n .
62
Because the empress was childless she was later deposed and replaced by
WeiTzu-fu. Interestingly in this narrative, even the Princess of P’ing-yang,
asister of Emperor Wu, seems to have been very eager to ﬁnd a suitable wife
for her brother. She certainly did not do so out of altruism. The idea wasrather that, if successful, she would be able to inﬂuence decision-making
at court. But the fact that Wu promoted a woman of low origins to therank of empress can also be explained by precisely the same reasoning. By
selecting a wife without prominent kin he could ﬁnd allies who would sharehis own ideas rather than pursue their own interests. The brother of WeiTzu-fu soon became generalissimo of the Han armies even though – as
Ssu-ma Ch’ien intimates – he had neither sufﬁcient military experience noran education justifying his promotion. T ogether with the brother, a wholeclique was raised up. Again, the historian seems to imply that one aim ofthe wars in central Asia which began only a few months after Wei Tzu-fubecame empress was to give the new elite a chance to shine, so that EmperorWucould then enfeoff them. However, one can also see the whole story the
other way round: the old elite would probably not have followed EmperorWuin his military adventures. Interestingly, twenty years later, long after
Empress Wei had fallen out of favour, the pattern repeated itself. Wu againtook a strong liking for a singing-girl, and again her brother was chosen as
general-in-chief of a campaign in central Asia. T o rely on people recruitedbythe emperor in person was a means of bypassing the regular court, whose
members might have challenged imperial policy.
63
Emperor Wu is an exception both in terms of the length of his reign and
in his autocratic style of rule. Late in his reign, he deposed his empressonce again. Her son, the crown prince, had to commit suicide after a failedrebellion against his father in 91 BC. In order to see what the regular pattern
of power at court looked like, we have to look at the following reigns indetail as well. When Emperor Chao ascended the throne in 87 BC he wasstill a seven-year-old minor. His reign was dominated by Huo Kuang, a
62SC49/1978f, trans. Watson 1961 ,I;also HS97A/3949.
63Compare the remarks by Winterling ( 1998 :11): ‘Die Bevorzugung von Personen niedriger
Herkunft in der engsten kaiserlichen Umgebung l ¨aßt auf eine latente Bedrohung der Kaiser
durch Personen hohen aristokratischen Prestiges schließen.’ (The preferential treatment of
persons of a lowly background in the closest vicinity of the emperor suggests that the emperorswere potent ially threatened by persons who enjoyed a high aristocratic prestige.)

256 hans van ess
regent whose father had once had an affair with a sister of Empress Wei. Huo
Ch’¨u-ping, the second most important general of the Hsiung-nu campaigns,
was born as a result of this affair, which in turn brought his half-brother,
HuoKuang, into the palace. However, as a half-brother Huo Kuang was not
directly related to the Wei family, a fact which probably had saved his lifein 91 BC. Huo Kuang arranged a marriage between the daughter of one ofhis close allies and Emperor Chao, but managed to get rid of the family inquestion when it became too independent. The girl was merely ﬁve yearsold when she was ofﬁcially made empress in 83 BC. Emperor Chao diedonly three years after receiving the cap of manhood in 77 BC, a ceremonywhich in theory allowed him to rule on his own. His empress was thirteenor fourteen at that time and the two had failed to produce an heir. This factallowed Huo Kuang to manipulate the succession and to continue to rulefor some years until his own death.
Afteracabal led by Huo Kuang successfully prevented one son of Emperor
Wu from ascending the throne on the grounds of alleged misconduct in
aritual matter, a great-grandson of Wu-ti became emperor at the age of
seventeen. He had previously lived among the common people and wasprobably selected by the regent because he had no prior ties to the imperialclan. But the new emperor was not grateful. Only three years later, he oustedthe Huo clan from power. He was the last emperor of the Former Hantorule without the support of an empress’ powerful family. The reign of
his son, Emperor Y ¨uan (49–33 BC), was dominated by the clan of his wife,
Empress Wang. Emperor Ch’eng (33–7 BC) is portrayed as a weak ruler. LikeEmperor Wu, he is known to history for stripping power from his empress,member of one of the great clans, deposing her, and installing a new empress,the famous Chao Fei-yen.
64The Han-shu tells us that when Chao Fei-yen
was born, her parents – apparently too poor to raise her – exposed her forthree days, taking her back only because she turned out to be still alive. Shewas trained as a singing-girl and a dancer, eventually performing before theemperor, who took a great liking for her. It is said that when Emperor Ch’engplanned to install Chao Fei-yen as his new empress, he had ﬁrst to enfeoffher father so as to allay the doubts of the empress-dowager, who consideredthe family too lowly.
65
The reign of Emperor Ch’eng is usual ly seen as the beginning of the decline
of the Former Han. For traditional Chinese historiography the story of ChaoFei-yen has always been a lesson in the devastating impact on a ruling house
64Chao the Flying Swallow, a name which was given her because of her graceful appearance.
65HS97B/3984; for Chao Fei-yen see HS97B/3988f.

The imperial court in Han China 257
when an emperor’s love strayed from his chief wife. In later times this story
was subject to hundreds of poems and other literary reworkings. In Mingtimes the Chao Fei-yen material even formed the basis for a pornographicnovel.
66Butthe impression remains that in reality the story is the histori-
ographer’s attempt to gloss over the fact that the powerful Wang clan triedtoforestall a challenge to its position f romthe family of Emperor Ch’eng’s
new empress. The negative tone to the story of Chao Fei-yen as told in theHan-shu stems partly from the fact that the emperor’s erotic adventures
failed to produce an heir, thereby further strengthening the Wang clan. Butit also reﬂects the fact that a female member of the clan of the historianhimself had unsuccessfully aspired to the position of empress.
The dominance of the Wang clan was only brieﬂy interrupted when
Emperor Ai (7–1 BC) came to the throne. Although he tried to promotethe clans of his own relations, he too failed to produce an heir. For this mis-fortune the historian blamed the fact that his reign was dominated by a malefavourite who slept in the emperor’s bed and received all kinds of beneﬁts asaresult of this intimacy. Imperial protection of male favourites was seriously
frowned upon by all traditional Chinese historians,
67mainly because this
particular avenue of access to the emperor was denied to ordinary membersof the bureaucracy, to which the historians themselves usually belonged. Theheterosexual relations of the emperor also provided opportunities to bypassthe usual career pattern (see above), but these were deemed unavoidable inthe interests of dynastic reproduction. But same-sex liaisons were seen as anunnecessary luxury detrimental to all parties. First they offended the careerbureaucracy, at least in theory based on merit. Secondly, they challenged theraison d’ ˆetre for the presence of the great clans at court, which was closely
bound up with the provision of an imperial heir.
68
66Chao-yang ch’¨ u-shih ,trans. into German by Engler ( 1980 ).
67There is a chapter with biographies of male favourites in Shih-chi and Han-shu .HouHan-shu
lacks such a chapter, replacing it with one on eunuchs, for whom see below.
68Shih-chi relates the story of T eng T’ung, the male favourite of Emperor Wen, who ‘did not have
any other talent s… than caring for his body in order to bewitch the sovereign’ . Once, when
Emperor Wen suffered from an ulcer, T eng T’ung sucked it. When the emperor asked him who
loved him most, he answered that no one in the empire loved the emperor more than thecrown-prince. The emperor called the crown-prince in and ordered him to suck his ulcer,which he duly did, but with an expression of disgust on his face. Thereafter the crown-princehated the favourite and removed him as soon as his father had died ( SC125/3192f). The career
bureaucracy did its best to discourage the emperor from promoting T eng T’ung to too high aposition ( SC96/2683). However, it is not clear what exactly imperial relations with a male
favourite involved. For example, it is said about Li Y en-nien that he slept in the emperor’s bed(literally: ‘he slept and got up together with the sovereign’), although we know that he was aeunuch ( SC125/3195).

258 hans van ess
Aneight-year-old emperor linked to the Wang clan succeeded Emperor
Ai,but died only ﬁve years later without an heir. In AD 6 Wang Mang, the
leading male member of the Wang clan, put another child on the throne,
this time a one-year-old. The child was deposed only three years later byWang Mang himself, who reigned as a usurper until AD 23.
The Han dynasty was restored by a distant member of the founder’s family,
who shifted the capital to Lo-yang, a city in eastern China, much closer tohis origins in what today is northern Hu-pei, in southern-central China.Three factions dominated court politics during the next two centuries. Onewas an alliance between several families of rich landowners from the fertileplain of southern-central China. The other two factions, based in northernChina, had trading interests in central Asia. Pan Ku himself, the author oftheHan-shu ,belonged to one of these.
Hans Bielenstein has described how during the Later Han the pendulum
swung back and forth between expansionism, driven by the northern fac-tions, and peace advocated by the landowners from Ho-nan and Hu-pei. Aswing of the pendulum was always accompanied by the enthronement ofanew empress from one of these three competing factions.
69Wher eas the
founder of the Later Han, his son and his grandson were all adults capableof ruling in their own right, from around AD 75–100 the families of theempresses steadily gained in power. The rulers were children, and almost alldied before reaching twenty. The Book of the Later Han (Hou Han-shu) by
FanYeh (AD 398–446; see Appendix) several times implies, and at least once
openly suggests, that some of these child-emperors were poisoned. If theyhad children, their chief wives were not the mothers. Therefore, on severaloccasions sons were removed from less well-placed members of the haremand claimed as children of the empress. In such cases the real mothers wereusually killed and their family, if allowed to live, would be banished to placessufﬁciently removed from the capital for the crime to stay hushed up.
Often the empresses were older than the emperors. Nevertheless, they were
manipulated by the leading male members of their families, who becameregents, a position usually accompanied by the title of ‘generalissimo’ (ta
chiang-ch ¨un). Empresses who tried to assert their will were sometimes
deposed. The members of a particular clan most involved in governmentwereroutinely annihilated when the pendulum swung in favour of another
clan.
70Usually this happened when the regent or the empress-dowager died.
69See Bielenstein 1986b .
70Forthe empresses of the Later Han see Bielenstein’s condensed account in Bielenstein 1986b ;
also de Crespigny 1975 .

The imperial court in Han China 259
The resulting vacuum was then ﬁlled by the family of the new empress. While
some of the regents are described as honest men, others are said to have beenaddicted to all kinds of licentious behaviour – chess, dog- and horse-races,cock-ﬁghting, football, gambling, women and alcohol. According to our
sources, government increasingly fell into the hands of big men who werenot ‘virtuous’ .
Afterall this historical detail, the conceptual question to be asked is why
the position of emperor remained so important. As the great clans were sopowerful and the throne so weak, one might think that the dominant clan ofthe moment would not have needed the court. Was there a religious beliefin the sanctity of the emperor which overrode all political and sociologicalreasons for a central institution at the heart of the empire? Respect for the
institution of the court is clearly demonstrated by the desperate attempts ofthe great clans to dominate it. Were individual clans too afraid of the risksinvolved to dare to leave the court and attempt to create a new imperialcentre elsewhere? These questions are addressed again in the conclusion.
The eunuchs
Strong rulers used the harem to promote new members of the elite as allies
against the old elite. When the ruler was weak, as was the case for most of theLater Han period, he was dominated by the great clans. Domination of weakchild-emperors by important families was the rule, strong rulers the excep-
tion. Power at court was divided among the regent and his family, the careerbureaucracy and the throne itself, usually consisting of the emperor andhis empress. As described above, in theory this could sometimes mean thatall three groups belonged to a single faction. But in practice the system wasmuchmore complicated. Although one clan could try to inﬁltrate as many of
its own members into the scholar-bureaucracy as it could, it never managedcompletely to dislodge the other clans. If a ruler died and a child-emperor
succeeded, the empress-dowager often supported pre-existing policies, andit was not easy for the regent to go against her will.
Indifﬁcult matters court conferences had to be held and compromises
found, often through the mediation of scholars expert in the canonical textsand able to draw on them to construct convincing arguments. In the middleof the second century AD, scholars at an imperial academy in the capital,feeling that they were increasingly bypassed by the ruling clans, became astrong voice of opposition at the heart of the empire. According to the Hou
Han-shu the number of students grew to more than 30,000 at this time, and

260 hans van ess
they inﬂuenced the language and rhetoric of the court. The students were
opposed both to the regents, whose might hindered them from using theirnewly acquired knowledge to start a career at court, and to another groupwhich grew very powerful during the second century AD: the eunuchs.
From the most ancient times Chinese rulers relied on eunuchs for many
duties in the Inner Palace. The HouHan-shu is the ﬁrst standard history of
China to contain a chapter specially devoted to biographies of select eunuchs.Inhis preface to this chapter Fan Y eh writes:
The I-ching says: ‘Heaven has hung up images [namely the stars] and the sage
takes his model from this [in his government on earth].’71There are four Eunuch-
stars which are situated next to the position of the Emperor. Therefore, when the
[author] of the Rites o fthe Chou established the ofﬁces, he also ﬁlled this number.
‘The doorkeeper watches over the prohibitio ns of the middle doors [which separate
the inner from the outer].’ ‘The court eunuch is responsible for compliance with thewarnings regarding the women’s palace.’ Furthermore it says that ‘There are ﬁve menin the inner chamber of the king.’
72The Monthly Ordinances say: ‘In mid-winter he
orders the prefect-eunuch to inspect the doors and to take care of the bed-rooms.’73
IntheLesser Elegantiae of the Odes,too, there is a piece by a [eunuch called] Hsiang
Powhocriticizes the slanderers.74
This shows that eunuchs have since olden times been present at the court of the
king. Was this not due to the fact that their body was not endowed with the completephysical abilities and that their feelings and will were therefore concentrated andgood so that they could have contact with the palace ladies and were easy to put tolabour and to nourish?
75
Inan example of analogical thinking, the idea at the beginning of this passage
is that the structure of the ﬁrmament is duplicated on earth. Every star
corresponds to a position on earth. The pole star is the star of the ruler
and is surrounded by stars which represent court ofﬁcials. Therefore, theposition of the ruler during court audiences was always in the northernpart of the audience hall, whereas his guests were in the southern part. Onthe western side, facing east, stood the military ofﬁcers, and on the eastern
71I-ching ,Hsi-tz’u A/11 (Harvard Y en-ching concordance), Richard Wilhelm, trans. IGing ,296.
72The Chou-li is one of three canonical texts concerning the all-important subject of the Rites. Its
traditional composer was said to be the duke of C hou, one of the founders of the Chou dynasty
almost a thousand years before the rise of the Han. However, modern scholarship agrees that,
although the text may contain very early elements, it is probably a third-century BCsystematisation of the hierarchy of the Chou. For the passages quoted here see Chou-li ,SSC
686C–687A and 644C.
73SSC 1382C.74Legge 1861 –1872 ,III, The Book of Odes, no. 200, 346–9.
75HHS 78/2507.

The imperial court in Han China 261
side, facing west, the civilian ofﬁcers.76What is important for us is that in
this passage the eunuchs are considered not as a discrete group but as part
of the throne. If we look at the court as a centre of power composed ofdifferent interest groups, we probably have to adopt the idea that emperorsand eunuchs usually fought for the same interests against those of the clansand the career bureaucracy.
The historian clearly says that eunuchs were good for guarding the harem
but should not be given higher positions. As in later times, scholar-ofﬁcialsin Han China looked down on eunuchs as on male favourites. Castration wasthe most serious punishment after the death penalty.
77Although some poor
families apparently castrated their sons so that they could live in the palace,eunuchs were certainly seen as inferior creatures. It seems that eunuchs didnot play an important role in politics or at court for most of the FormerHan perio d. Interestingly, in the Han-shu ,which deals exclusively with this
period, the biographies of eunuchs are found in the chapter devoted to theemperors’ male favourites and do not have their own chapter.
This reﬂects the fact that although under the Former Han there were,
exceptionally, eunuchs inﬂuential in court politics, it was only during the
second century AD that they came to constitute a political faction in theirownright. Of course, this was due to the declining role of the emperor and
his empresses. Before the power of omnipotent regents, the only peoplethese poor imperial children could turn to for help were their eunuchs. Theyoung emperors did have teachers, who were usually well-known scholars.
The ofﬁce of the teacher of the heir-apparent – and later often the emperorhimself – was nominally one of the highest in the hierarchy, but it did notimply actual power. By contrast, the eunuchs did have access to the levers ofpower. They were the only attendants who could provide the young emperorswith real support.
During the second half of the second century AD tensions arose between
the great families and the eunuchs. Following a dispute over the successiontoEmperor Huan (AD 146–168), these tensions culminated in a show-
down between the forces of T ou Wu, the regent, and the eunuchs. Theeunuchs won, and as a result were able to dominate the entire reign ofEmperor Ling (AD 168–189). A career in the court bureaucracy becameimpossible for anyone without the support of the eunuchs. A butcher’sdaughter became empress – this was obviously a manoeuvre to exclude the
76The system goes back to Shu-sun T’ung who is said to have created the court ceremonial for Han
Kao-tsu. See his biography in SC99/2723. Compare also T’ung-chih of Cheng Ch’iao, 44/589A.
77Asalready noted, Ssu-ma Ch’ien was castrated. However, according to Han-shu he was given a
high-ranking eunuch ofﬁce afterwards ( HS62/2725).

262 hans van ess
big clans. However, this did not save the eunuchs. When Emperor Ling died,
the dynasty had almost come to its end, and the turmoil of rebellions anduprisings produced independent military leaders. One of them marched tothe capital and massacred the eunuchs. As a result they vanished from thepolitical scene, only to be replaced by military leaders.
78
The eunuchs became powerful just as the old system of power-brokerage
at court was breaking down. When the great families had lost their abilitytodominate the emperor, eunuchs ﬁlled the vacuum and became the repre-
sentatives of the court. Their rise was the last stand of the Han court. Whenthey fell, the inner court ceased to exist and consequently the outer court,namely the elite bureaucracy, lost its inﬂuence as well.
Conclusion
Asseen above, the initial raison d’ ˆetre of the court of the Han was to establish
order after the rise to power of a new elite of former nobodies. The courtceremonial was probably invented by members of the old elite, the survivors
of a long period of war, to domesticate a group of rough military leaders.Rituals and ceremonies served this goal of achieving order. The court wascertainly a ritualised space where a mistake in ceremonial matters could
easily cost a courtier his position. It soon became the centre of culture andreﬁnement as well. But the sociological model of the court as a distributor of
favours and as a cultural showcase does not sufﬁce to explain its longevity.
Italso seems that belief in the divine nature of the emperor grew with the
longevity of the dynasty. This shows that, despite the emperor’s role as a kindof high priest in religious matters, we should not think of Han China as atheocracy. The court did not exist simply because the emperor was god-likeand therefore had to be served by the elite.
Inexploring this question, it is more helpful to look at the institutional
aspect. The court was the institution where people struggled for power.‘Intrigue’ is an important term in accounts of the court in ancient Chinesehistoriography. Apparently there was a big difference between the Formerand the Later Han. As far as our historians are concerned at least, court lifeduring much of the Former Han period was dominated by humble peoplefrom a rustic background. But at the end of the Former Han and during theLater Han, an oligarchy of rich families emerged which controlled access
78This has been conveniently summarised by Burchard Mansvelt-Beck in Twitchett and Loewe
1986 :317–45. On the role of the eunuchs under the Later Han in general see Jugel 1976 .

The imperial court in Han China 263
tothe high bureaucracy. This oligarchy would keep its position of power
for many centuries to come.79Nevertheless, the institution of the court
remained the same and was accepted.
Ifwewant to understand why a court was needed and why it survived for
over four centuries, it is instructive to look at the Later Han period, when the
court continued to exist for a long time even without strong emperors. This
shows that the court cannot be seen merely as an instrument for the display
of power by an autocratic ruler. As was shown above, the emperor, and withhim the throne as an institution, constituted the head of the legislative body.Itseems that there was some kind of pre-modern division of powers. The
emperor and with him the permanent inner court had legislative power,either with a great clan, represented by an empress, or with eunuchs asthe real decision-makers. The outer court on the other hand, consistingmainly of the bureaucratic elite, served an executive function. This meansthat the emperor functioned as a counterbalance against the bureaucracy.Legal authority belonged to both sides: whereas the minister of justice had to
lead investigations and function as chie fprosecutor even against members of
the imperial clan, the emperor had to approve his decisions in all importantcases, after they had been submitted to the throne in the form of a memorial.
Precisely because of this balancing function, for a long time a consen-
sus existed that a court was the best possible mechanism for ordering theworld. Because the inner court was the heart of this mechanism, there was
the desire to control it. As long as no one was powerful enough to challengeaweak dynasty which still retained the support of other clans, and as long
as these clans in turn were not strong enough even if they were to act inconcert, the throne remained the centre of the empire. Its domination by
great families for more than half of the whole Han period is probably –
again – to be explained in institutional terms. The big families monopolisedexecutive power because their members dominated the bureaucracy. How-
ever, if a family introduced a daughter into the harem and by this means
gained direct access to the emperor, it added legislative power and the ﬁnalwordin juridical matters to its executive functions. The last point may also
explain why regents were often executed after an empress-dowager’s death:
the throne was being used to approve immediate measures against a factionwhose power had become much greater than that of the other factions. T oleave it in place threatened the system of checks and balances.
Norbert Elias has explained the phenomenon of court culture in sociolog-
ical terms. There are indeed some features of his model which ﬁt the Chinese
79Johnson 1977 .

264 hans van ess
case. Normative texts from the early Chinese past as well as descriptive texts
about the Later Han contain, for example, very detailed prescriptions fordistinguishing ranks, recalling the complicated hierarchies of the absolutistEuropean courts. Similarities with Europe are also to be found in the fact
that the daily life of the Han emperors was structured by a great many rules.Yet, overall Elias’ model does not really work in the Chinese case. The inner
court with its two main political factors, the harem and the eunuchs, has no
equivalent in late medieval Europe, and the outer court is clearly differentbecause its protagonists were members of a career bureaucracy. Certainly,the role of the court as a reﬁned centre of culture and literature was extremelyimportant in both parts of the world, serving the same purpose of present-ing the court as the high centre of the state. But in Han China Confucianinﬂuence prevented the ideal courtier from displaying his afﬂuence. There-fore, at least in theory, one of the main elements in the Eliasque model ofthe court – the aristocratic compulsion to assert status through conspicu-ous consumption – was missing from the Chinese case. For this reason, theconstitutional model suggested abov eoffers the best explanation for why
the Han needed a court.
Appendix: the sources
Our understanding of the history of the Han dynasty (207 BC–AD 220)is mainly shaped by three sources. The ﬁrst is the Record so ft h e Scribe
(Shih-chi )o r ,a s it was earlier known, the Documents of the Lord Grand
Scribe (T’ai-shih kung shu ). This work attempted to record the history of
China from its very beginnings under the mythical Y ellow Emperor (tradi-tional dates 2697–2597 BC) until the time of the authors. It was begun bySsu-ma T’an (died 110 BC) and almost ﬁnished by his son, Ssu-ma Ch’ien(c.145 BC– c.86BC).
80Howeve r,the bulk of the Shih-chi deals with the
history of the Ch’in (221–207 BC) and the ﬁrst century of the Han. Ssu-ma T’an and Ssu-ma Ch’ien both served as ‘Lord Grand Scribe’ (T’ai-shihkung) at the court of the sixth Han emperor, Han Wu-ti (reigned 140–87 BC). The main task of this ofﬁcial was to watch over and calculate theimperial calendar, to observe the movements of the stars and to interpretthem. Historiography was only a secondary duty, derived from his role as
80Afew passages had clearly been added after the death of Ssu-ma Ch’ien. But by and large the
book can be said to be the work of the two Ssu-ma.

The imperial court in Han China 265
an astronomer. Moreover, when reading the Shih-chi it must always be kept
in mind that Ssu-ma Ch’ien was opposed to many imperial policies. In fact,
he ﬁnished his work after suffering castration for defending a general whohad surrendered to the Hsiung-nu, a nomadic people living on the northernfrontier and China’s main external threat. Therefore, the Shih-chi is certainly
not an ofﬁcial history of the Han. Indeed, it is plausible that its version ofevents is coloured by Ssu-ma Ch’ien’s personal experiences. It is dangerous,
then, to take the Shih-chi at face value. This chapter has emphasised many
changes which took place under Emperor Wu. But the reader should be
aware that Wu’s reign may receive more attention here than it deserves –
reﬂecting the weight which Ssu-ma Ch’ien himself accorded it.
The account of the Han empire in the Shih-chi has strongly inﬂuenced
our second source, namely the Book of the Han (Han-shu ), another joint
venture by a father and his son, namely Pan Piao (AD 3–54) and Pan Ku
(AD 32–92). We are told that when Pan Ku began the continuation of hisdeceased father’s work, he was imprisoned on a charge of unauthorisedtampering with dynastic history. Fortunately for us, his younger brotherPanCh’ao, later an important general in central Asia, managed to convince
the emperor that Pan Ku was a loyal subject. Even so, Emperor Ming (AD57–75) of the Later Han only allowed Pan Ku to ﬁnish his father’s workafter he had proved his loyalty. He did this by writing the annals of thefounding emperor of the Later Han, Kuang-wu (reigned AD 25–57), andof several other persons, including at least one rival for Kuang-wu’s throne.The magnum opus to w hich he then returned began with the rise of Liu
Pang, posthumously named Kao-tsu (‘Founding Ancestor’ , reigned 206 or
202–195 BC), and went beyond the fall of the Former Han in AD 9 to theso-called ‘usurpation’ of Wang Mang (AD 9–23).
81The Han-shu is usually
said to be a more orthodox source than the Shih-chi .Y e ti tt o o has its own
agenda, namely to support the interests of the Pan family in central Asiaand the politics of expansion associated with Emperor Wu of the FormerHan. Pan Ku was executed in AD 92, when a peace faction at court sought to
end the frontier wars. This fact is relevant for Pan Ku’s treatment of certainprominent Han courtiers.
Thirdly there is the Book of the Later Han (HouHan-shu ). This, our major
source for the period AD 25–220, was written much later, albeit based onearlier sources. Its author, Fan Y eh (AD 398–446), also died a violent death.With this work too there are grounds for suspecting bias. As little work has
81See his biography in HHS 40A, 1333f.

266 hans van ess
been done so far on the authenticity of the HouHan-shu ,82its huge historical
scope means that for the time being we have to rely on it. A second history
of the Later Han, the HouHan-chi of Y¨uan Hung (AD 328–376), has largely
escaped the notice of modern scholars, probably because it was excludedfrom the semi-ofﬁcial canon of standard histories drawn up much later bytraditional Chinese historians.
82Butnote Bielenstein 1954 .The author concludes that the complex process of source-
composition which underlies the compilation of the HouHan-shu guaranteed a high degree of
objectivity. I am much less optimistic about this.

7 Court and palace in ancient Egypt: the Amarna
period and later Eighteenth Dynasty
kate spence
Ishall make Akhetaten for the Aten, my father, in this plac e…N o r shall
the Great King’s wife say to me ‘Look, there is a nice place for Akhetaten
in another place’ , nor shall I listen to her. Nor shall any ofﬁcial in mypresence – whether ofﬁcials ‘of favour’ or ofﬁcials ‘of the outside’ , or thechamberlains, or any people who are in the entire land – say to me: ‘Look
there is a nice place for Akhetaten in another place’ , nor shall I listen tothem.
Oath of Akhenaten (Murnane and van Siclen 1993 :40)
Around 1347 BC, King Akhenaten stated under oath his intention of found-
ing a new city as his residence and cult centre for the visible sun-disk or Aten,
whom he had recently promoted to the position of sole god, excluding themajority of the traditional Egyptian pantheon.
1Details of the foundation
and reafﬁrmation ceremonies were carved onto boundary stelae cut into thecliffs around the site and these are amongst the most important historicaldocuments for the period (Murnane and van Siclen 1993). The above state-ment emphasises the king’s resolve to stand by his chosen site for the city. Inpassing, he makes an unusual acknowledgement of the potential inﬂuenceof royal women and courtiers.
Ofﬁcial royal texts in Egypt rarely mention individuals other than the
king. Royal projects such as building temples, ﬁghting wars and performingritual were recorded as historical events but were usually presented as if
kings acted in isolation. The major exception occurs in a genre of literatureknown as K¨onigsnovelle (Hermann 1938 ;Spalinger 1982 :101–19; Loprieno
1996 ), to which this part of Akhenaten’s boundary text is related. In some of
these texts, the king decides a plan of action in the company of his advisors.The ofﬁcials offer sound advice which the king usually ignores, choosinginstead a course of action which often sounds over-ambitious, foolhardyor downright dangerous. The ofﬁcials agree with his decision because he is
1The author wishes to acknowledge valuable co mments on an early draft of this chapter made by
participants at the workshop in Newcastle University; Professor Barry Kemp read a later draft
and made some perceptive points which have been incorporated into the text. Any errors remainthe responsibility of the author.

268 kate spence
king. The activity then takes place and is reported to be a success. The reason
for the inclusion of these rather unusual narratives in royal records was toemphasise the divine insight of the king who was inspired and guided bythe gods. It was also presumably thought to show him as a strong leader,over-riding the conventional earthly wisdom of his advisors in a manner
that was clearly to be considered positive by the intended audience.
Inpractice, these texts also provide a rare reﬂection of life at court and the
process of decision-making. The king was an individual with great powerwhose decisions, however idiosyncratic, affected the lives of his household,the elite and the whole populace of Egypt; but he was also surrounded by agroup of people who sought to advise and inﬂuence him. He was as reliant
on them for information and for the promulgation of his authority as theywereon him for their position and prestige; these courtiers formed both his
social milieu and the highest ofﬁcials of his government. Underlying ancientEgypt’s pervasive ofﬁcial image of absolute royal authority, divine approval
and social stability were ﬂuid, symbiotic relationships between kings andthe courtiers who promoted this ideal.
The sources
Insharp contrast to most of the other periods discussed in this volume (but
see Wieseh ¨ofer on Sasanian Iran), we have no contemporary descriptive
accounts of life at the ancient Egyptian court prior to the Hellenistic period.Instead, information must be gleaned from royal inscriptions, literary and
administrative texts, private tomb autobiographies and representations, andarchaeology. There is much scope for study, but the record is fragmentaryand largely devoid of detailed evidence for the important personal dimen-sion of social interaction and intrigue that emerges from studies of better-documented examples as a fundamental aspect of court society (Elias 1983 :
70–1; Duindam 2003 :248–9). Despite the paucity of textual sources there
can be no doubt that a court existed: in private inscriptions, members of theelite list titles acquired during their career and, alongside those relating toroles in administration, are courtly ranking titles such as ‘sole companion
[of the king]’ and ‘royal acquaintance’ . Titles such as ‘palace administrator’and ‘overseer of the audience chamber’ attest to the importance of the palaceas an institution and the existence of a courtly society within it, while titlessuch as ‘cupbearer’ and ‘fan-bearer on the right hand of the king’ , held byindividuals of high status, show that some aspects of personal service for theking were provided by courtiers and were considered great honours. There

Court and palace in ancient Egypt 269
is hierarchy implied in the various courtly titles as well as in administrative
ones: for example, in addition to the title ‘companion’ , the ofﬁce of ‘headof the companions of the king’ is known ( L¨AII: 1237).
2There is also clear
evidence for the importance of etiquette and ritual at court ( L¨AII: 1229,
1237; Coulon 2002 ).
Evidence for the court can be drawn from a number of types of textual
source. Royal documents such as the K¨onigsnovelle texts have already been
mentioned. These tend to focus on the king and to be fairly formulaic;thus their value for illuminating the court is limited. Well-preserved lists oftitles and autobiographical inscriptions on objects, statuary and tomb wallshave formed the basis of a number of important prosopographical studieswhich have shed light on the structure of the administration and career pathsthrough it, as well as positions and roles at court (e.g. Helck 1954 ;B a e r 1960 ;
Kanawati 1977 ;S t r u d w i c k 1985 ;for the New Kingdom: Helck 1958 ;B r y a n
1991 :242–331; Kitchen 1982 :125–53; Murnane 1998 ). Autobiographical
texts found in private tombs often focus on a particularly close relationship
between an individual and the king (usually described in terms of unusualduties or assignments entrusted to the ofﬁcial, honours conferred and thereasons for favouritism) but they often avoid referring to peers at court as
this might detract from the presentation of this relationship as personaland exclusive (unless the king is said to distinguish the individual above hiscontemporaries). The king was rarely represented alongside humans in art
as this was not considered decorous, and scenes of life at court are foundinfrequently before the Amarna period. The primary exceptions to this areEarly Dynastic representations of the king with members of his court (suchas the scene from the Narmer Palette shown in Fig. 7.1)and images of
courtiers with the king displayed on the walls of pyramid temples of the Old
Kingdom.
3
Afew administrative documents survive which provide evidence for
studying the court. The most important of these is Papyrus Boulaq 18 whichprovides an account of palace income and expenditure for a period of twelvedays in the late Middle Kingdom (Quirke 1990 ). The question of the broader
applicability of conclusions drawn from this document is problematical: itderives from the Thirteenth Dynasty when the power and authority of theking were somewhat limited and it relates to a palace at Thebes that was notthe principal residence of the king. The short period covered also renders
2The God’s Father Ay held both titles in the reign of Akhenaten.
3See, for example, scenes from the pyramid temple of King Sahure (Borchardt 1913 :pls. 17, 19,
32–4).

Figure 7.1 The king inspects the decapitated bodies of enemies in the company o f senior courtiers and standard-bearers. From the
Narmer P alette.

Court and palace in ancient Egypt 271
it hard to assess broader trends, but it remains our only document detailing
the constitution of the royal household (albeit in a presumably limited form,as it was constituted on a royal visit) and is extremely informative. Thisdocument will be discussed in more detail later in the chapter.
Middle Kingdom stories provide our most vibrant illustrations of the
Egyptian court. The story of Sinuhe describes the ﬂight and self-imposed
exile of a courtier who fears he may be implicated in the assassination of
King Amenemhet I. The latter part of the story describes Sinuhe’s return to
Egyptand includes an audience at court and Sinuhe’s rehabilitation there,
which, although probably ﬁctional, must have been credible to its audiencegiventhe popularity of the text. Some of the stories, such as those in Papyrus
Westcar (Lichtheim 1973 :215–22), present rulers of the Old Kingdom in
aless than favourable light which is clearly intended to contrast with the
exemplary behaviour of the Middle Kingdom kings in whose courts they
werecomposed. These stories do not describe the court – they were written
for an audience already familiar with its structure and nature – but theincidental details of the texts are nonetheless valuable even though the storiesthemselves must again be treated as largely ﬁctional.
There is a strong tendency for discussions of courts to be based on textual
evidence, and there are good reasons for this: texts describe the relationships,
intrigue and behaviour on which much of the discussion is focused. How-ever, the Egyptian textual sources are not only limited and biased – invariably
deriving from the elite or court itself – but were also subject to a strict senseof decorum (Baines 1990 ). It seems likely that the extent to which decorum
has rendered the sources unrepresentative tends to be underestimated: a fewexamples will sufﬁce to illustrate the problem. First, interaction between the
king and his ofﬁcials and courtiers is rarely described (or represented in art)whereas in reality this must have formed the focus of life at court. Second,from the Fifth Dynasty royal sons are virtually invisible in the sources (withthe exception of the sons of Ramesses II) but they are present and preparedfor kingship when the throne is vacated: it thus seems unlikely that they hadno experience of court life before coming to the throne. Third, although weknow that kings of the New Kingdom were decidedly militaristic in outlook,the evidence for military inﬂuence at court is relatively limited. These areobvious examples; there must be many more, invisible to the modern reader.
Formuch of Egyptian history the court was the focus for the production
of the luxury artefacts (including the statuary, paintings and relief carvinglabelled today as ‘art’). The form and style of these artefacts spread outwardsfrom the residence city, inﬂuencing production throughout the countryand further aﬁeld, although regional styles are also apparent, particularly

272 kate spence
before the New Kingdom. Architecture can also reveal much about social
interaction, status and position at court. For most periods there is signiﬁcantevidence in the form of tomb chapels, but preservation of the domestic
architecture of royal settlements is limited outside the New Kingdom.
This chapter focuses on the Amarna period and its aftermath, equivalent
tothe latter part of the Eighteenth Dynasty (Akhenaten–Horemheb, c.1352–
1295 BC, with some discussion of the reign of Amenhotep III, c.1390–
1352 BC, which preceded that of Akhenaten). The late Eighteenth Dynastyyields a particularly broad range of relevant material, allowing discussion to
move beyond the limited historical sources and highlighting the importanceof the archaeological record. In addition to royal and elite inscriptions,there is an important corpus of diplomatic correspondence between Egyptand Near Eastern rulers surviving from this period – the Amarna Letters(Liverani 1990 ;M o r a n 1992 ;Cohen and Westbrook 2000 )–while the private
tombs at Amarna are decorated with an unprecedented set of representations
of Egyptian royal and courtly activity (Davies 1903 to1908b ). Horemheb
seems to have felt it necessary to justify his position in the post-Amarnaera and issued royal texts and decrees, which are particularly revealing withregard to life at court (Murnane 1995 :230–3, 235–40). The archaeological
remains of the city of Amarna provide concrete evidence for palatial and
ritual settings, as well as institutions and the houses of courtiers and other
members of society (Kemp 1989 :261–317); such archaeological evidence is
limited and fragmentary for other periods of Egyptian history. Even for thisperiod, however, the evidence remains thin as far as charting the nuances ofcourt politics and the shifting factions of courtiers is concerned, although
historians of the Eighteenth Dynasty have long been prone to speculation,particularly with regard to the Amarna period (Montserrat 2000 :12–54).
The historical context
Egyptwas ruled by around thirty dynasties of kings, from c.3000 BC when
the country was ﬁrst uniﬁed until the Roman conquest in 30 BC. Before theﬁrst millennium BC most of these rulers were of Egyptian descent;
4all were
resident in Egypt (as far as is known). Egyptian history is conventionally
divided into periods of centralised control under divine kings punctuated by
4The major exception being the Hyksos rulers of the Second Intermediate period
(c.1650–1550 BC) who resided at Avaris in the eastern Delta and were of Asiatic (probably
Syrian or Palestinian) descent (Bourriau 2000 :186–95).

Court and palace in ancient Egypt 273
‘intermediate periods’ in which the country was divided under competing
rulers.5The Egyptians themselves placed great importance on the concept
of unity, expressed prominently in the iconography of kingship throughimages of the uniﬁcation of the two halves of Egypt (Kemp 1989 :27–9)
and in the titles of the king ( L¨AIII: 641–59; Baines 1995 :125–8). Those
historical kings who had reuniﬁed Egypt after a period of ‘disorder’ (i.e.an intermediate period) were held in particular reverence (Redford 1986 :
35). The majority of the evidence for the royal court derives from periodsof unity, as it was only when control was centralised that the palace coulddraw on the resources of the whole country and divert this wealth into theproduction of the art, architecture and literature which form the bulk of theavailable source material for studying Egyptian culture.
Consideration of the historical setting raises a number of issues of rel-
evance to the treatment of sources. First, the limited evidence renders it
necessary to draw on material from many periods; such an approach isencouraged by the superﬁcially conservative nature of the Egyptian sources.However,alongside a degree of continuity there are changes in administra-
tion and in the composition of the court over time, even when one allows forthe patchy nature of the evidence (Leprohon ( 1995 )p r o v i d e sa useful sum-
mary). The Old Kingdom (until its latter years) produced extensive courtcemeteries in the Sakkara region, containing the burials of courtiers, admin-
istrators and members of the king’s household, attesting to a high degreeof centralisation and showing the extent to which wealth was concentratedaround the king (Roth 1993 ). From the Middle Kingdom we ﬁnd signif-
icant wealth displayed in provincial cemeteries, and the relationship andpower-balance between centre and provinces is clearly complex: provincialofﬁcials bore titles granted to them by the king and played an important partin the administration, but some of these provincial leaders built up theirownhouseholds into small court-like conﬁgurations with their own title-
holders and ofﬁcials.
6The New Kingdom court shows a signiﬁcant degree
of centralisation but there seems to be more emphasis on the military than
5Manley ( 1996 )provides a short overview of Egyptian history. Shaw ( 2000 )gives a more detailed,
multi-authored treatment. The period under discussion here (up to the end of the New
Kingdom) is usually divided into the Early Dynastic period (Dynasties 1–2, c.3000–2686 BC),
the Old Kingdom (Dynasties 3–8, c.2686–2125 BC), the First Intermediate period (Dynasties
9–early 11, c.2160–2055 BC), the Middle Kingdom (Dynasties 11–14, c.2055–1650 BC), the
Second Intermediate period (Dynasties 15–17, c.1650–1550 BC) and the New Kingdom
(Dynasties 18–20, c.1550–1069 BC). The dates follow Shaw (2000: 479–83). There remains
disagreement as to when the divisions between some of these periods should fall.
6The process of decentralisation can be traced back to the late Old Kingdom. See Franke ( 1991 )
for a nuanced discussion of the role of provincial rulers and their relationship with the court.

274 kate spence
formerly. Kings of the Eighteenth Dynasty used their armies to build up
an empire which brought signiﬁcant wealth to king and court. The courtappears, at least superﬁcially, to hav ebeen less introspective than its ear-
lier counterparts as a result of military campaigning abroad and diplomaticactivity. Our evidence for the various periods is not equivalent, however,and it remains likely that such broad generalisations are crude.
Second, this chapter focuses on the Amarna period on account of the
broad nature of the available evidence, but this is probably the most contro-versial episode in Egyptian history.
7Akhenaten’s decision to abandon the
traditional Egyptian pantheon and attendant religious practices in favour of
worship of the visible sun-disk was accompanied by changes in the style and
content of art and literature and also in architecture. Opinions differ as to the
extent of his changes and also the underlying motivation for them: to some
they derive from religious zeal while to others they represent a desire to curbthe rising power of the priesthood of the state god Amun-Re. Although pre-cursors of his religious changes can be identiﬁed in preceding reigns, therecan be no doubt that they had a signiﬁcant impact on life at court, particu-larly given Akhenaten’s decision to uproot the royal household and move to anewly constructed residence in Middle Egypt. However, although the atmos-phere at court may have been particularly bitter and difﬁcult during the earlyyears of his reign and many of the actors may have changed, the institution
of the court seems to have survived relatively intact: the king could not haveeffected and maintained the changes he made without his courtiers.
Although the Amarna period represents a particularly extreme example,
it also illustrates the fact that, in addition to the broader historical trends thatcan be isolated, there are likely to have been signiﬁcant differences betweenreigns. The constitution of the court and the nature of life there would
have been strongly inﬂuenced by the personality of each individual king,and there is no guarantee that a conclusion drawn from isolated materialdating to any reign is broadly applicable. Murnane’s detailed study of thegovernment of Egypt during the reign of Amenhotep III mentions some ofthe differences between the reigns of Amenhotep III and Akhenaten that canbe established on the basis of prosopographical and broader textual study(Murnane 1998 ).
7Van D ijk (2000 :272–94) gives a recent and brief outline of the period. More substantial
treatments are found in Aldred ( 1988 )and Redford ( 1984 ), the latter focusing particularly on
the early years of Akhenaten’s reign and his activities at Karnak. Baines ( 1998 )gives a nuanced
interpretation of Akhenaten’s changes within their historical context. See Montserrat ( 2000 :
esp. 12–54) for a discussion of modern interest in Akhenaten’s reign and its rendering in
historical studies, literature and art.

Court and palace in ancient Egypt 275
Third, evidence tends to derive from periods when writing was widely
used and power was centralised, but these parameters are not equivalent
tothose marking the existence of the royal court. Although archaeological
material can be difﬁcult to interpret, it can provide valuable informationon periods that textual sources do not cover and material on aspects ofthe court ignored or unclear in the texts. The importance of archaeologicalsource material is highlighted in the following discussion on the origin of theEgyptian court and in the treatment of Amarna palaces later in the chapter.
The origins of the Egyptian court
The origins of the Egyptian court are difﬁcult to pin down but must lie inthe predynastic period. In the fourth millennium BC increasing social com-plexity can be traced in the material record of southern Egypt. A group ofcompeting ‘proto-states’ emerged under the leadership of local chiefs (Kemp
1989 :31–5; Bard 1994 ;Baines 1995 ;Wilkinson 2000 ). These groups pro-
duced artefacts which gradually became widespread throughout the whole
country, cultural uniﬁcation thereby preceding political uniﬁcation which
probably occurred shortly before 3000 BC under leaders deriving from theThinite region, known primarily from their tombs at Abydos. It was aroundthis period of uniﬁcation that writing ﬁrst appeared in Egypt, providing theﬁrst examples of the names and titles of kings and ofﬁcials (Baines 1995 ).
Dating to this period, perhaps even depicting the decisive battle of the
political process of uniﬁcation, is a richly decorated siltstone palette dedi-cated by King Narmer and left as a votive offering in a temple at Hierakon-polis. The palette’s decoration strongly suggests that the Egyptian court waswell established before the country was uniﬁed (Fig. 7.1). The king wears
crowns, false beards and a kilt with an apron and a tail, and carries regalia, allof distinctive designs characteristic of the dress and regalia of later Egyptian
divine kings. Hierarchy of scale is used to show relative importance, withthe king’s ﬁgure dominating the ﬁelds in which he is represented. Behindthe king walks an individual depicted about half the size of the king whocarries the king’s sandals and a vessel; early hieroglyphs describe him onlyas a ‘servant of the ruler’ (Smith 1992 :244). Y et he is clearly an impor-
tant ofﬁcial: he wears a cylinder seal around his neck, perhaps suggestingarole in administration. In the scene shown in Figure 7.1(the upper ﬁeld
on the front of the palette), the king walks with his sandal-bearer behindhim from a rectangular object (probably a building) towards two rows ofheadless corpses. Preceding him are four quarter-size ﬁgures bearing divine

276 kate spence
standards and a half-size ﬁgure with long hair or wig, unusual costume and
accessories, and a title that may be related to that later used to designate thevizier or ‘chief minister’ (Wilkinson 1999 :137) although others identify him
as a priest (Baines 1995 :120) or ‘shaman’ (Helck 1986 :12–13). In the border
at the top of the object the king’s name is written within a rectangular framewith niches, thought to represent the early palace. We thus ﬁnd character-
istic examples of the architecture, costume, titles, ranking, subordinationof high ofﬁcials, divinisation of the king and ceremonial activities of latercourt life expressed in a nuanced way on this one artefact, and also on other
royal obj ects of the period (Gundlach 1998 :62–84).
Itshould come as no surprise that the court predates the historical period.
The existence of absolute power is not conﬁned to literate societies and,where very signiﬁcant political power is represented in an individual, thepossibility of advancement through personal favour rather than individualmerit or popular support encourages congregation around that individual.Attempts to gain inﬂuence give rise to cycles of behaviour involving ﬂattery
and intrigue, characteristic of court societies.
8Asagroup emerges around
aleader, ceremonial activity emerges or is imposed which serves further to
differentiate the leader from his courtiers and to perpetuate this symbioticrelationship. The problem with the early Egyptian evidence is that court
society is characterised primarily through ﬁgurations of individuals andparticular modes of behaviour and these are very difﬁcult to trace in ancientsocieties in the absence of historical sources. Duindam ( 2003 :318) argues
that the term ‘court’ should not be used ‘as a bland equivalent of groupdynamics around leaders’ , but deﬁning the precise moment at which these
group dynamics become a court proper is extremely difﬁcult, particularly
when the evidence is archaeological.
Cemeteries and objects deriving from tombs form the bulk of our evi-
dence for early periods in Egypt, and archaeological studies have identiﬁedanumber of factors of relevance to the development of the state (Wilkinson
2000 )that may also be of interest for charting the emergence of the court. At
anumber of sites in the late predynastic period there is evidence that some
elite individuals were buried in cemeteries isolated from the burial places ofthe majority of the local population, suggesting an attempt to differentiaterulers, and in some cases other members of the elite, from their subjects. The
concentr ation of imported and manufactured prestige goods in the graves
of these individuals suggests signiﬁcant control of trade and productionbut also shows the desire to express power and wealth through display of
8Butbynomeans unique to them (Duindam 2003 :318).

Court and palace in ancient Egypt 277
expensive luxury goods. The artistic motifs Bard ( 1994 :111) described as
‘state symbolism’ also occur in a few elite tombs from the Naqada III period
(from c.3200 BC) immediately preceding uniﬁcation, although Wilkinson
(2000 :380–1) suggests that such symbolism is found even earlier in the
Naqada I period (before c.3500 BC). The richness of burials at other sites
was gradually eclipsed by that of the tombs at Abydos, suggesting that the
authority of the rulers buried here was increasing at the expense of others,and that those not connected directly to the rulers at Abydos had morelimited access to wealth and prestige goods. A. M. Roth suggests that thephyle system into which staff of temples, royal mortuary cults and the palacewereorganised during the Old Kingdom derives from local groups or fam-
ilies (‘clans’) prominent during the pre-literate period. She proposes thatthis kinship-based system was then modiﬁed into the organisational sys-temrequired by the centralised government and bureaucracy (Roth 1991 :
esp. 205–16). The existence of such kinship groups was presumably essentialtothe early development and stability of the royal circle, but the breaking
of explicit links between kinship lines (hereditary nobility) and the kingin favour of non-hereditary ofﬁcial positions expressed through titles wasessential to the way the court developed in Egypt.
The origins of the Egyptian court thus clearly pre-date the historical
period and lie in the processes of social stratiﬁcation and in the nucleationof individuals around a ruler perceived to be absolute.
9Following uniﬁca-
tion, the Egyptian king wielded enormous personal power but was faced withgoverning and exploiting a large territory. Egyptian bureaucracy and admin-istration are thought to have developed in the Early Dynastic period as kingand court rapidly adjusted to governing the country from the royal residence(Wilkinson 1999 :109–49). Over the ﬁrst three dynasties there was a prolif-
eration of administrative titles as the bureaucracy took shape (Trigger et al.1983 :66). Large and well-equipped tombs of ofﬁcials of the First Dynasty are
found at Sakkara, adjacent to the new royal centre at Memphis. These, likethe funerary enclosures of contemporary kings at Abydos, were decoratedwith elaborate buttressing in a style known to Egyptologists as ‘palace fac ¸ade’ .
This is thought to link the tombs visually with the royal palace, constitutingan indicator of association with the king and of distinction from the generalpopulation, and thus a court style (Kemp 1989 :55; Wilkinson 1999 :225).
Interestingly, this ‘palace fac ¸ade’ niching seems be an example of the
conscious borrowing of ideas and imagery from the Near East, showing
9‘Courts are the most general arrangement of power in pre-modern society’ (Duindam 2003 :
302).

278 kate spence
that, despite the early and indigenous origins of kingship and the court, ideas
wereborrowed to aid the expression of power and the differentiation of king
and court (Smith 1992 ;cf. Brosius in this volume on Achaemenid-Persian
borrowings). The Narmer Palette again indicates that this borrowing is early:the rosette-shaped hieroglyph used to refer to the king in the caption abovethe sandal-bearer (Fig. 7.1)has its origins in the Near East (Smith 1992 :241–
4). Borrowing is also clear in the cylinder seal worn by the king’s servant, inthe ‘palace fac ¸ade’ motif within which the king’s name is written and in the
design of the long-necked serpopards which form the central motif on thefront of the palette (Smith 1992 :235–8). The court setting and the need to
administer the palace and extensive territories also seem to have been theimpetus for the development of writing, ﬁrst attested in Egypt in the labellingof commodities (Postgate et al. 1995 ); here again some inﬂuence from the
Near East is possible, although the hieroglyphic signs and writing system
that emerged in Egypt are very different from those in use in Mesopotamia.
The court thus emerges alongside kingship as part of an ongoing and
slow process of social stratiﬁcation. In a period of more rapid change some-time before uniﬁcation, the court took signiﬁcant steps to differentiate itselffurther from the rest of society, presumably as part of a process of self-legitimisation. This involved distinguishing the king (as the ﬁgurehead andfocus of both court and country) from his followers through associationwith the gods, providing distinct regalia and a separate cemetery. The court
itself proﬁted through its association with the king. In the First Dynastythe king was buried surrounded by members of his household (Wilkinson1999 :235–7) although the practice of retainer sacriﬁce did not last long in
Egypt. Alongside this, attempts were also made to distinguish courtiers from
the majority of the population; early evidence includes the use of titles andelaborate tombs and burial arrangements.
King and court
Itis the absolute nature of a ruler’s position which distinguishes his associates
and establishment as a court. In the ancient world this position was often,although not always, associated with very considerable power and authority.The behaviour and interaction of individuals at court belong to a broaderspectrum of behaviour around leaders, but it is the unchallenged authorityof the position of ruler (rather than the individual incumbent) that leadstothe concentration of resources, the ritualisation of the ruler’s sphere and
the invocation of divine legitimisation so characteristic of court life.

Court and palace in ancient Egypt 279
Egyptian sources of relevance to the court are dominated by the ﬁgure
of the king whose complex nature encapsulated both human and divine
attributes (Posener 1960 ;R e d ford 1995 ;S i l v e rman 1995 ;Gundlach 1998 ).
The Egyptian king wielded very signiﬁcant power and, although his com-mands were executed by others in his name, there can be little doubt astothe extent of his personal power or of public perception of the origin
of that authority (Lorton 1991 :306). The king was also closely associated
with the gods and played an essential role in the welfare of Egypt by medi-
ating between its people and the gods. Time was recorded back throughlists of ancestral kings to the reigns of spirits and before them the gods(Redford 1986 :xix–xx), providing divine precedent and ancestry for the
ruling king. Theoretically it was the king who performed all temple ritual
for the gods, although most of the time priests stood in for him. Althoughmany commentators have argued that the extent of the king’s divinity (asit was perceived by his subjects) waned over time, Quirke ( 1999 :63–4) has
pointed out that the paucity of surviving records makes any accurate assess-ment of change over time virtually impossible. It is, however, clear thatwhile the king was considered a divine being in his own right, he (or she)was also believed to be human and capable of error and weakness (Posener1960 ).
10
The divinity of the king differentiated him (and at some periods also
members of his immediate family) from the courtiers and household thatserved him. This differentiation seems likely to have been particularly usefulwhen ruling families changed as, once established, it provided clear distinc-tion between the king and the group from whom he emerged as leader.
11
On a day-to-day basis this difference was presumably marked through rit-ualised behaviour around the king and through royal regalia, dress andsetting. Rituals of kingship often followed ancient precedent, although tra-ditional formulation often masks considerable innovation. The court itselfmust have provided a signiﬁcant degree of continuity over long periods of
time, enforcing smooth transitions of power between one monarch and thenext and working to legitimise and empower the kingship on which thecourtiers depended for their own position and authority.
10Although the literary texts which provide most of the evidence for this at least in part serve
propagandistic purposes. For example, the Midd le Kingdom stories which emphasise the
amoral characters of a number of Old Kingdom kings were probably intended to emphasisepositive aspects of the characters and interests of the contemporary kings in whose courts they
were comp osed. Silverman ( 1995 :esp. 50) comments on the relationship between secondary
literature and the primary evidence on which the various interpretations are based.
11Tuthmosis III and Horemheb both claim to have been chosen to rule through oracles of the
gods.

280 kate spence
The structure of the Egyptian court
The existence of a royal court in Egypt is accepted throughout the discipline,
and the courtly context of much of the material studied by Egyptologistsis widely acknowledged. ‘Court’ and ‘courtiers’ are referred to frequentlyin secondary literature but the court as an institution has rarely formedan explicit focus of study (although see, for example, Quirke 1990 ;L o r t o n
1991 ;Coulon 2002 ;Gundlach 2004 ;Raedler 2004 ;forthcoming). Egypto-
logical dictionaries seem not to contain entries on the court or courtiers,although the seven-volume Lexikon der ¨Agyptologie has short entries under
‘H¨oﬂichkeit und Etikette’ ( L¨AII: 1229), ‘H ¨ofrang’ ( L¨AII: 1237) and ‘Hof-
zeremo niell’ ( L¨AII: 1237–8). However, there is a vast body of literature
on related subjects that are of relevance to the topic;
12these include his-
torical and prosopographical studies, and analyses of the art and literature
produced within a court setting.
On close inspection, the Egyptian court proves a surprisingly elusive
topic for research. The major problems seem to lie in deﬁning the court
and its membership and in establishing exactly how it was understood bythe Egyptians themselves. Our textual sources are limited and inexplicitand our understanding of related lexicography is at present insufﬁcient.
13
The closest equivalent to our terms ‘court’ and ‘courtiers’ is found in theEgypt ian term shenyt which derives from the verb sheni ‘toencircle’ and in
English translations is most commonly rendered as ‘entourage’ although inGerman we ﬁnd ‘Hofstaat’ or ‘Hoﬂeute’ (Erman and Grapow 1930 :511–12;
Faulkner 1962 :267–8). It is a collective term, most usually written with a
human determinative, that seems to refer to a small group of high-status
individuals close to the king rather than the whole royal household or even allof the ofﬁcials present at the palace at any time.
14The semeru ‘companions’
or ‘friends’ seem to have been an even more limited group of high-statusofﬁcials. There is also a title rekh nesu ,‘royalacquaintance’ . However, as
the opening statement of Akhenaten shows, those around the king withpotential inﬂuence were frequently referred to simply as seru,‘ofﬁcials’ or
‘functionaries’ , making it difﬁcult to distinguish between those serving the
12The bibliography presented here is far from exhaustive and where possible cites accessible
sources in English.
13Forexample, there are at least ﬁve terms for the royal palace which have been the subject of
several studies (e.g. Goelet 1985 ,1986 )but the signiﬁcance of each is still debated. Lorton
(1991 )has a limited discussion with references.
14‘…the members of the king’s entourage are at their exact places and the Thirty [the royal
council] are at their customary positions’ (Murnane 1995 :240).

Court and palace in ancient Egypt 281
king directly and other functionaries (see Quirke 1990 :43); the highest
administrators often seem to have been designated ‘royal scribes’ , but also
bore titles designating a particular area of administration (Helck 1958 :277–
8). This blurring of categories perhaps links all in authority to king andpalace but makes it hard to deﬁne the court, particularly as the use of themore speciﬁc terms seems to be rather ﬂuid, at least in the Middle Kingdom(Quirke 1990 :51–7). Quirke ( 1990 :65) also notes that the composition of
the inner palace personnel at this period seems very variable, being madeup of those high-ranking ofﬁcials who happened to be present.
Akhenaten’s classiﬁcation of those who might have inﬂuence over him
is extremely interesting but unfortunately appears to be unique (Murnaneand van Siclen 1993 :60). The list comprises: the queen and ‘any ofﬁcial in
mypresence – whether (they are) ofﬁcials “of favour” or ofﬁcials “of the
outside” , or the chamberlains, or any people who are in the entire land’(Murnane and van Siclen 1993 :40). Ranked above the chamberlains, who
are unquestionably palace ofﬁcials, are two classes of ofﬁcial: ‘favoured’and ‘outsiders’ . Does this mark the distinction between the ‘companions’or ‘entourage’ allowed access to the inner palace and those serving in theouter parts of the palace? Are those serving in high administrative ofﬁcesoutside the palace the ‘outsiders’ or do they fall with their minions under thecategory ‘any people who are in the entire land’? This latter term is unlikelytomean absolutely any Egyptian, given the royal context and the fact that
the phrase is one of a listed subset of ofﬁcials.
Inhis study of the courts at Vienna and Versailles, Duindam ( 2003 )stresses
the importance of considering the wider household of the king rather thanjust high-status courtiers. There can be no doubting the high status of royalwomen in Egypt, but personal attendants without signiﬁcant administrative
titles may also have had exceptional access to the king and therefore have hadthe potential to wield considerable inﬂuence. Unfortunately the evidence forroyalhouseholds in New Kingdom Egypt is considerably sparser than that
for courtiers, although in the Old Kingdom ﬁne tombs were provided forfavoured palace staff such as guards, musicians and low-ranking ofﬁcials,attesting to their favour with the king (Roth 1995 :40–3; Kanawati 2003 :14–
24).
15Anumber of terms for those working at the palace are found, including
aproblematical title, khenty-she ,recently translated as either ‘palace atten-
dant’ (Roth 1995 :40–3) or ‘guard’ (Kanawati 2003 :14–19), and another,
aq,translated ‘ordinary entrant’ (Quirke 1990 :36–7); Lorton argues that
15Itmay be signiﬁcant that in the Old Kingdom the court seems to have been dominated by the
royalfamily and perhaps to have been more entrenched than at later periods.

282 kate spence
the term per-nesu or ‘king’s house’ encompasses those who worked in it (i.e.
the household) as well as a physical place (Lorton 1991 :304). Servants are
shown working in palaces at Amarna but, unsurprisingly, are not listed by
Akhenaten as possible inﬂuences in his boundary stelae.
Itis worth noting that, although the Egyptian court was dominated at
most periods by prominent wealthy families, there is no hereditary nobil-ity in a strict sense. Appointments were made by the king, and those fromhumble backgrounds are known to have achieved high ofﬁce occasionally,
at least in the New Kingdom (see below). It may be, therefore, that par-ticularly favoured attendants (if they existed) could be promoted to palaceor administrative ofﬁces, thus rendering the distinction between courtiersand very inﬂuential members of the ‘household’ less stark than it may havebeen in medieval or early modern European courts. Also, as some ofﬁcialpositions involving apparently menial personal service for the king (such as
‘cup-bearers’ who looked after the king’s comfort) were known to provide
privileged access, they were highly sought after (cf. Brosius and Spawforthin this volume): many cup-bearers had close links to other insiders suchas royal nurses or tutors, or had been brought up in the palace themselves.Manyseem to have gone on to high ofﬁce (Murnane 1998 :216). It thus
seems likely that many of the king’s personal attendants were drawn fromthe families of the ofﬁcials associated with the inner palace, reducing thepotential inﬂuence of outsiders.
PapyrusBoulaq 18 provides an account of palace income and expenditure
for a period of twelve days in the late Middle Kingdom (Quirke 1990 ). The
date of the document (Thirteenth Dynasty), the fact that it does not refer tothe principal residence, and the short period it covers render it hard to assessthe broader applicability of some of the conclusions drawn from it, but itremains our only document detailing the constitution of the royal household
(albeit in a presumably limited form as it was constituted on a royal visit).Atleast some of the royal family were present: a royal wife is mentioned
along with a son, three daughters and a sister of the king. Although the kingis not mentioned in the text, Quirke ( 1990 :120) argues that he was present
during at least part of the period under discussion. Ofﬁcials are presentedin hierarchical ‘ranking blocks’ within the lists (Quirke 1990 :73). There are
nurses, magicians, musicians, a cosmetician and a hairdresser listed and also
menial serving staff and guards. Interestingly, the families of some ofﬁcialsalso seem to have been present (Quirke 1990 :90–4), but there are a num-
ber of conspicuous absences of high administrative ofﬁcials: Quirke pointsout that neither the treasurer nor any priests were present (Quirke 1990 :
51, 81).

Court and palace in ancient Egypt 283
On the basis of this document, Quirke establishes that the personnel of
the palace was administratively divided into three groups. Some, such as the
royalfamily, families of ofﬁcials, nurses and musicians were associated with
the inner palace ( kap)(Quirke 1990 :36–8, 87–97), while the outer palace
(khenty )was a sector which Quirke suggests dealt with state affairs and was
headed by the vizier (Quirke 1990 :36–8, 72–86). Finally a class of ‘ordinary
entrants’ or menial serving staff is documented (Quirke 1990 :36–7). Also
prominent in the text is an audience area of the palace where banquets wereheld.
The Egyptians thus seem to have focused on an inner circle around the
king, made up of small groups of favoured ofﬁcials with their families, andother advisors.
16The numbers involved appear to be relatively small. All of
these individuals seem to have held speciﬁc courtly titles and, given theirrole in the council chamber, were presumably involved in decision-making
although many may have been absent from the palace on ofﬁcial duties at anyone time. The outer palace also served as an important administrative centrewith its own ofﬁcials, who presumably had links with inner palace ofﬁcials
although they were usually classed separately. Links between the palace andthe provinces and other institutions were presumably maintained throughsenior ofﬁcials who held titles showing close ties to the king and courtalongside those indicating roles in institutions or the provinces. As a resultof this there is considerable blurring between the designation of courtiersand administrative ofﬁcials (cf. van Ess in this volume on the – in some wayscomparable – structure of the Han Chinese court).
Ofﬁcal documentary sources tend to fall into two groups: royal texts asso-
ciated with activities of the king, in which he is represented as an absolutepower ruling and acting in isolation, and texts associated with adminis-tration and ofﬁcialdom which suggest that the country was governed by a
hierarchical administration under one or more viziers but make little ref-erence to the king, apparently for reasons of decorum (Quirke 1990 :120;
1999). The court is the link between these two representations of power andauthority, as is clear from the courtly context and titles of those with highadministrative positions, and repeated statements that these individuals car-ried out the king’s wishes. However, the mechanisms of decision-making and
the balance and exercise of power are extremely difﬁcult to isolate (Quirke1990 :50–7; Lorton 1991 ).
16Notealso the list of ofﬁcials meeting King Amenhotep III at the gate of the palace for his jubilee
celebrations in year 30: ‘the ofﬁcials, the king’s friends, the chamberlain, the men of the
gateway, the king’s acquaintances, the crew of the bark, the castellans, and the king’s dignitaries’
(Murnane 1998 :217).

284 kate spence
The courtier
The majority of high ofﬁcials in Egypt seem to have derived from established
elite society (Murnane 1998 :212). Powerful, wealthy families are attested
in burials and inscriptions from provincial centres as well as the royal resi-dences. Some of these families were bound to the ruler through marriagealliances, and family members frequently held important administrativetitles; however, such titles were not hereditary (although at some periodsthey were monopolised by single families as is the case with the provincialrulers (nomarchs) of the late Old Kingdom to early Middle Kingdom). T ech-
nically therefore, there is no ‘hereditary nobility’ in Egypt. However, whenTutankhamun restored the Amun cult a few years after Akhenaten’s death, he
dedicated a stela in which it is stated that ‘he installed lay priests and higherclergy from among the children of the ofﬁcials of their cities, each one beingthe “son-of-a-man” whose name was known’ (Murnane 1995 :213). The
desire to bequeath one’s ofﬁce (and the wealth, rank and privileges accruingto it) to o ne’s eldest son is a common theme in Egyptian texts and there are
examples of minor ofﬁcial positions being sold, suggesting that possession
was, at some periods, considered to be a hereditary right (e.g. Parkinson1991 :110–11).
Appointments of the highest functionaries (i.e. those who would have
visited the palace and played a role at court) were made by the king. Although
relatives were occasionally appointed to vacated posts, this was unusual
during the mid to late Eighteenth Dynasty and the king’s favour wouldhave been essential to ensuring such continuity (Murnane 1998 :213–14).
17
All ofﬁcial posts brought material reward, often in the form of proceedsfrom land set aside to support the post but also sometimes in paymentsfrom the treasury. Proximity to the king brought with it the opportunityfor a courtier to distinguish himself, and favour was richly rewarded andexpressed materially. Kings could award promotions (which brought with
them wealth) or supply material gifts such as elaborate tombs and burialequipment, jewellery, luxury goods or even captives of war; these gifts aresometimes detailed in the private tombs of the recipients. Such rewards havesigniﬁcant communicative value and were actively sought by courtiers: themajority of the decorated private tombs at Amarna contain a scene showingthe king rewarding the tomb owner; these scenes and their architecturalsetting will be discussed in more detail below.
17Atcertain periods ofﬁces were more closely tied to single families; several examples are found
among the priesthood in the Nineteenth Dynasty (Kitchen 1982 :46–7, 170–1).

Court and palace in ancient Egypt 285
Those from more humble backgrounds were not precluded from rising
through the ranks and gaining status at court, however, as there is occasional
evidence for those from obscure backgrounds rising to positions of consid-
erable power. Several examples derive from Amarna itself. The ‘fanbeareron the King’s right hand’ , May, was granted a large and impressive tombat Amarna by Akhenaten (Davies 1908a :1–5, pls. I–V). He held signiﬁcant
ofﬁces, including ‘the king’s scribe, scribe of recruits, steward of “PacifyingAten” , steward of Waenre in Heliopolis, overseer of the cattle of the House of
Rein Heliopolis, [overseer of] all [the works projects] of the king, the general
of the Lord of the Two Lands’ (Murnane 1995 :145). He also held ranking
titles such as ‘sole companion’ (Davies 1908a :4–5). Despite his wealth and
status during the reign of Akhenaten, May writes: ‘I was a poor man on bothmyfather’s and my mother’s side – but the ruler built me up, he caused me
todevelop, he fed me by means of his Ka
18when I was without property . . .
he caused me to mingle with ofﬁcials and courtiers when I was the leastof underlings’ (Murnane 1995 :145). The treasurer Suty also states that his
background was humble (Murnane 1995 :186), while the general Ramose
claims to be an ‘ofﬁcial of the ruler’s making’ (Murnane 1995 :183).
The case of Senenmut, a prominent ofﬁcial and courtier during the co-
regency of Hatshepsut (a female king) and Tuthmosis III, is a particularly
revealing example dating to the earlier Eighteenth Dynasty. Senenmut’s
father had held a very lowly title and had originally been interred in a pitburial wrapped only in a shroud, until his body was exhumed and reburiedin a more elaborate tomb at Thebes by Senenmut when he became wealthyand inﬂuential (Dorman 1988 :165–9).
19Senenmut was accorded extraor-
dinary privileges during Hatshepsut’s reign: his image was inscribed behindthe doors on many of the door jambs in her mortuary temple, and theburial chamber of his lower tomb lay beneath the forecourt of the tem-ple, although the entrance was outside it. He had two tombs at Thebesand large numbers of statues portraying him have survived, suggestingverysigniﬁcant wealth. Senenmut seems to have fallen from favour dur-
ing the sole reign of Tuthmosis III and damage to his Theban tombs and hisimages in Hatshepsut’s temple suggests that he had made signiﬁcant enemies(Dorman 1988 ).
This case is revealing for a number of reasons. It demonstrates that sig-
niﬁcant changes in social status were possible in Egypt but are likely to haveowed much to the personal intervention of the king: ‘Y ou have a powerful
18The kacan here be roughly translated as ‘divine essence’ .
19When the body was excavated, soil from the original burial was found adhering to the shroud.

286 kate spence
ofﬁce, given you by the king.’20Whe ther such promotions resulted from
talent, loyalty or simply royal favour cannot be established with certainty;
autobiographical texts usually claim al lasfactors. There is no doubt, how-
ever, that privileges accrued through service at court could raise an ofﬁcial to
aposition of wealth and prestige equivalent or superior to that of an estab-
lished member of elite society, as the examples of May and Senenmut show.Interestingly, the case of Senenmut may also demonstrate limitations to the
king’s power when it came to appointments and promotions as, despite theextraordinary privileges accorded to him, he never achieved the highest-
ranking positions within government: his wealth, authority and status must
have derived instead from his position at court and his relationship to theking. This in turn suggests that our understanding of Egyptian government,which is almost entirely based on prosopographical studies of the adminis-trative hierarchy, may not always be an accurate reﬂection of inﬂuence and
decision-making processes.
Here the role of favourites amongst royal women and members of the
household must also have been signiﬁcant, although this is often difﬁcult totrace in the archaeological record. It is interesting to note here that Akhen-
aten’s cup-bearer Parennefer, who is given no administrative or ranking titlesin his Amarna tomb inscriptions, was nonetheless accorded a ﬁne decoratedtomb complete with a royal reward scene (Davies 1908b :1–6, pls. II–X). As
an ofﬁcial in charge of the personal comfort of the ruler a cup-bearer is likelytohave had particularly privileged access to the king (Murnane 1998 :216).
Interestingly, the same Parennefer is known from a tomb at Thebes and
appears to have held a priestly title at Karnak before the move to Amarna(Redford 1984 :60). His inﬂuence thus perhaps reached beyond his rank.
The individuals making up the royal court and their backgrounds varied
considerably from reign to reign and even within a single reign, as the subtle
balance of power shifted. Kings promoted or retained people whom theytrusted. Some kings maintained the services of those who had served under
their predecessors and some promoted ‘new men’ such as Senenmut or Maywho were entirely reliant on the king for their wealth and prestige. Some-times those brought up and educated with or under the king achieved highofﬁce (Bryan 1991 :261–3), although the title ‘child of the inner palace’ is not
found in any of the private tombs at Amarna. Sometimes particular fam-ilies were well represented amongst high ofﬁcials, suggesting considerable
20Lichtheim 1976 :171. From a New Kingdom text exhorting the reader to become a scribe.
Although the text is undoubtedly biased, the importance of literacy in advancing social status
through employment in administration seems to be supported by prosopographical evidence.

Court and palace in ancient Egypt 287
inﬂuence, and royal relatives such as the God’s Father Ay, who later became
king, might also be prominent at court.21
The composition of Akhenaten’s court is interesting as it shows remark-
ably little continuity with that of his predecessor (Murnane 1998 :214). His
father’s vizier, Ramose, survived into Akhenaten’s reign and is representedin his tomb being rewarded by both rulers, but in separate depictions inthe different artistic styles of the two reigns. His nephew, Ipy, was stewardof Memphis, in charge of the treasury, granary and workshops attached tothe royal domain, but then went on to become steward and ‘overseer ofthe inner palace of pharaoh in Akhetaten’ (Murnane 1998 :213–14). A man
named Aper-el (thought to be a name of Semitic origin) served as vizierunder Amenhotep III and is also described as a ‘child of the inner palace’(Zivie 1990 :151–2, 156–7). He bore the title ‘First servant of the Aten’ which
is attested only under Akhenaten, so it seems likely that he also survived thetransition. His son Huy, who was buried in the same tomb, bore military
titles, and seems to have served under Akhenaten until at least year 10 (Zivie1990 :159–66). We therefore have evidence for a few prominent individuals
and families who seem to have survived the transition of power, but otherold retainers are difﬁcult to identify.
Following the move to Amarna, courtiers are known primarily from their
inscribed tombs (Davies 1903 to1908b ), although the houses of some
prominent individuals such as the vizier Nakht can be identiﬁed in thecity itself (Peet and Woolley 1923 :5–9, pls. III–V).
22The majority of these
individuals served the king, although two are closely associated with royalwome n: Huya held the ofﬁces of steward, treasurer and overseer of the harem
under Akhenaten’s mother, Queen Tiye (Davies 1905b :1–25, pls. I–XXV;
Murnane 1995 :130–41), while Meryre II has titles associating him with
Nefertiti as well as the king (Davies 1905a :33–45, pls. XXVIII–XLVII;
Murnane 1995 :162–5). Almost all of the individuals represented in the
tombs have titles and epithets which link them closely with the king, although
there are a few exceptions such as the Chief of Police, Mahu (Davies 1906 :
12–18, pls. XIV–XXIX, XL–XLII; Murnane 1995 :147–51) and the standard-
bearer Suty (Davies 1906 :25, pls. XXXVIII–XXXIX; Murnane 1995 :185–6).
Those with priestly titles such as Meryre I (Davies 1903 ;Murnane 1995 :
21Aywas probably a close relative of Queen Tiye. His wife had also been Nefertiti’s wet-nurse
(Redford 1984 :150–1; van Dijk 2000 :292). For his tomb at Amarna see Davies ( 1908b :16–24,
pls. XXII–XXXIV, XXXVI–XLIV).
22Ab r i e fi n t r o d u c t i o nt os ome of the Amarna courtiers is given by Redford ( 1984 :149–53).
Up-to-date translations of the texts from the tombs and other sources are provided by Murnane
(1995 :107–204).

288 kate spence
151–62) and Panehesy (Davies 1905a :9–32, pls. II–XXVII; Murnane 1995 :
169–77) are particularly prominent, but royal scribes and military ofﬁcials
are also present. The decoration of all the tombs is focused on the king andhis family and the majority claim that the king favoured them. A few havetitles which suggest that their activi ties were not entirely focused within a
single institution: Pentu seems to hold ofﬁce both in the palace (as a royalscribe and the king’s physician) and in one of the Aten temples (Davies1906 :1–6; Murnane 1995 :179–82), while May holds courtly titles such as
‘fanbearer on the King’s right hand’ , priestly titles associated with the temple
of Re at Heliopolis and also military titles such as ‘general’ and ‘scribe ofrecruits’ (Davies 1908a :1–5, pls. I–V; Murnane 1995 :143–7).
Manykings of the early to mid Eighteenth Dynasty had been powerful
military leaders, and soldiers had sometimes ranked highly at court (Bryan1991 :279–93; 1998: 37–8).
23Although Akhenaten’s reign is usually pre-
sented as a time of peace and diplomacy, the military seems to have beenmore prominent at court during this period than is initially apparent. Mes-sages to the Egyptian king among the Amarna Letters suggest the importanceof the army with greetings such as ‘For my brother and his household, for hishorses and his chariots, for his magnates and his country may all go very well’(Moran 1992 :13). In the tomb of the chamberlain Tutu, the ofﬁcial is shown
addressing those assembled before the palace: ‘Pharaoh . . . [has ordered (?)]the commanders of hosts, masters of the horse, commanders [of the army]and [every commander of] troop s…e v e r ys e r v a n to fA t e nb e l o n g i n gt o
the Aten . . . all [. . .]s belonging to the House of Aten [and all] people . . .’(Murnane 1995 :194). The various army ofﬁcials are therefore given prece-
dence over priests and others. There are a substantial number of militaryofﬁcials amongst the courtiers at Amarna: Ay is described as ‘the king’s truescribe, his beloved, the God’s Father, Ay. The troop leader and master ofthe horse, the God’s Father, Ay’ (Murnane 1995 :107). Others such as the
general Ramose, the chief of police Mahu, the chief bowman and masterof the horse Nekhuempaaten, and the general Paatenemheb all had tombsat Amarna (Murnane 1995 :183–4; 147–51; 168). Although Horemheb is
known to have been an important general, his military title tends to belisted after a string of courtly titles during the reign of Tutankhamun: ‘thehereditary prince, the fan-bearer at the king’s right hand, the generalissimoHoremheb, justiﬁed’ (Murnane 1995 :229).
There were signiﬁcant numbers of foreigners at the royal court in the
Eighteenth Dynasty. Although there is no evidence to suggest that any of
23Forthe military during the New Kingdom see in particular Spalinger ( 1982 )and Gnirs ( 1996 ).

Court and palace in ancient Egypt 289
the nobles represented by the tombs at Amarna were foreign, visiting digni-
taries are well attested in the Amarna Letters and might spend a considerableperiod of time at court. Egyptian kings sometimes married the daughters offoreign rulers and it is clear from the Amarna Letters that these might arrivewith very large entourages and escorts (Moran 1992 :21, 81). Amenhotep III
is known to have had at least two Mitannian and two Babylonian wives andone from Arzawa (Schulman 1979 :183–4). They may still have been present
in one of the royal harems during Akhenaten’s reign, and Akhenaten him-self is known to have had at least two foreign wives (Schulman 1979 :185).
Ithas even been suggested that Kiya, a secondary wife of Akhenaten who
may have been the mother of Tutankhamun, should be identiﬁed with theMitannian princess Gilukhepa who had arrived in Amenhotep III’s harem
some thirty years earlier (Redford 1984 :150). Bilingual scribes must have
translated the diplomatic documents that passed between Egypt and her
neighbours in the Near East and some of the Amarna Letters are actuallylists of foreign words with their Egyptian equivalents (Moran 1992 :xv–xvi).
Itwas also common practice in the New Kingdom for children of foreign
rulers to be taken as surety and brought up and educated at the palace. These
are among the ‘children of the inner palace’ , some but not all of whom wenton to serve in ofﬁcial capacities in the Eighteenth Dynasty, although theyare not prominent during Akhenaten’s reign. Several foreigners achievedhigh ofﬁce within the government and foreign service of Amenhotep III(Murnane 1998 :202).
Female members of the royal family seem to have been exceptionally
prominent during the reign of Akhenaten. The king is frequently shownwith Nefertiti, his ‘great king’s wife’ , and with their daughters of whom the
eldest three are often depicted. Akhenaten’s mother, Tiye, was prominent inthe city with her own household (for her chief steward, Huya, see Murnane1995 :130–41) and ‘sunshade’ temple (Kemp 1995 :459–61). His secondary
wife, Kiya, had a brief period of exposure around the time that Nefertiti
disappears from the record, but Kiya’s images, names and titles were laterreplaced by those of Meritaten, Akhenaten’s eldest daughter who held the
position of ‘great king’s wife’ in the later years of the reign. There are likelytohave been other royal women and children, but they may not all have
lived at Amarna itself. There were ‘harem’ palaces elsewhere in Egypt at sitessuch as Medinet Gurob (Kemp 1978 )and less central members of the royal
family may have lived fairly permanently in these establishments. Otherwomen also played a role at court, with the title ‘royal ornament’ bestowed
on some wives of ofﬁcials. Tiyi, wife of the God’s Father Ay, bore the title‘king’s ornament’ and is also described as the ‘one whom the King’s Chief

290 kate spence
Wife favoured’ (Murnane 1995 :107); she had also been the wet-nurse of
Nefertiti and is described as ‘the great nurse who nourished the goddess’
(Murnane 1995 :109). Quirke ( 1990 :90–4) shows that the families of high
ofﬁcials formed part of the community of the inner palace in the late Middle
Kingdom; this seems likely to hold also for the New Kingdom although the
nature of that interaction within the residence city is uncertain.
The dual role of the court as the centre of government and the social
circle of the king is interesting. The presence of higher-ranking ofﬁcials atcourt presumably allowed the king to keep an eye on them and ensured
that his functionaries remained focused on his own person and the socialcompetition of life at court rather than on establishing entrenched power-
bases within institutions. The fact that some courtiers held titles associatedwith more than one branch of the administration (see above) must have
assisted here. The dual role of individuals as courtiers and as governmentalofﬁcers results in considerable blurring of the divisions between the palaceand other institutions, perhaps useful for maintaining an aura of ultimateauthority (Quirke 1990 :51–7).
The opportunity for interaction with the king would be essential if a
courtier were to gain royal favour. Most kings of the Eighteenth Dynasty
travelled extensively, moving within the country for religious ceremonies
and other royal activities, and many also ventured outside Egypt on mili-tary expeditions although there is little evidence that Akhenaten himselftravelled abroad (but see Murnane 1995 :101–3). Some courtiers and mem-
bers of the royal household moved around the country with the king (asis apparent from the Middle Kingdom Papyrus Boulaq 18) but others pre-sumably remained at the principal residences undertaking administrativeduties. Many ofﬁcials also served away from the royal palace for signiﬁcantlengths of time, undertaking duties at Memphis or in provincial govern-ment, leading mining expeditions or armies, overseeing building work andconducting diplomatic missions; the Amarna Letters suggest that envoys
might be detained at foreign courts for many years (e.g. Moran 1992 :7,13).
Successful discharge of a mission could, however, result in royal favour and
reward on the individual’s return to the palace. Raedler (forthcoming) sug-
gests that courtly titles expressing physical closeness to the king symbolicallymaintained proximity when ofﬁcials were absent from court.
The Eighteenth Dynasty court was therefore cosmopolitan and inclusive
of people from a variety of social backgrounds. Courtiers were not idle. Whilesome were employed in the personal service of the king and in running thepalace, others governed the country and ran the various institutions such asthe temples, treasury and army. Some ofﬁcials would have been permanently

Court and palace in ancient Egypt 291
present at court, while others were absent for long periods of time on ofﬁcial
or private business. Although some members of established families mayhave had sources of personal wealth and prestige in their home towns, theemphasis in mortuary architecture on the king and the relationship of thedeceased to him strongly suggests that personal interaction with the king,and thus presence at court, was an essential component in gaining wealthand social prestige.
Court politics
Access to the palace and the king seems to have been strictly controlled,
with security enforced by palace guards and their overseers. In addition,
anumber of ofﬁcials such as the ‘overseer of the audience chamber’ and
the ‘chamberlains’ controlled and regulated access to, and activities within,different parts of the palace. The inner parts of the palace had their own staff,and Ipy, ‘steward of Memphis’ , also served as ‘overseer of the inner palaceof pharaoh in Akhetaten’ (Murnane 1998 :214). Stewards administered the
king’s property and various supervisors oversaw the supplies kept in thepalace magazines (Murnane 1998 :212–17). The majority of ofﬁcials who
worked in the outer parts of the palace seem not to have been allowed access
tothe inner parts of the palace.
24
The Middle Kingdom ‘Story of Sinuhe’ gives a brief account of a visit to
the palace and a royal audience. Sinuhe was a courtier who had ﬂed Egyptfearing he would be implicated in the aftermath of the assassination of KingAmenemhet I. After many years in the Near East he returned to Egypt,
having received a royal pardon. The following extract describes Sinuhe’sentry to the palace: ‘T en men came and ten men went to usher me into thepalace. My forehead touched the ground between the sphinxes, and the royalchildren stood in the gateway to meet me. The courtiers who usher through
the forecourt set me on the way to the audience-hall. I found his majestyon the great throne in a kiosk of gold. Stretched out on my belly, I did notknow myself before him’ (Lichtheim 1973 :231).
The evidence suggests that prostration (‘kissing the ground’ in Egyptian
terminology) before the king was common practice. Amarna Letters sent
byvassal states refer in their opening lines to falling at the feet of the king:
‘Rib-Hadda says to the king, his lord, the Sun of [all countries]: I fall at
24Forthe administration of the palace in the Eighteenth Dynasty see, in particular, Lorton ( 1991 )
and Murnane ( 1998 :212–17).

Figure 7.2 The royal family ride in their chariots from palace to temple. The palac e, with a closed Window of Appearance, is
shown in the upper left corner of the depiction. From the tomb of Meryra at el-Amarna (Davies 1903: pl. X).

Figure 7.2 (cont.)

294 kate spence
the feet of my lord, [my] S[un] 7 times and 7 times’ (Moran 1992 :166).
Amarna period reliefs from Karnak show the king emerging from the palace
in his jubilee costume with the chamberlains lying prostrate before him
(Redford 1984 :119, ﬁg. 7.12). Prostration was also appropriate before gods,
and even Akhenaten himself is shown prostrate before the Aten with hisfamily on a balustrade block (Aldred 1988 :275, ﬁg. 25).
Inastela dating to the early Middle Kingdom an ofﬁcial describes himself
as ‘one who keeps the commoners distant from him [the king]’ (Fischer 1960:261). Ordinary people had virtually no access to the king and had to approachﬁgures of authority by petitioning at their gates and through intermediaries.This is well illustrated by another Middle Kingdom story, the ‘EloquentPeasant’ (Lichtheim 1973 :169–84). In this, a travelling trader has his goods
conﬁscated by a corrupt ofﬁcial. The trader petitions the high steward Rensi
who is in charge of the region in which the crime has taken place, waylayinghim as he emerges from his house and requesting that a servant be sent sothat he can make the complaint known. The trader’s petitions are so eloquentthat they are written down and presented to the king, who is pleased by them.Ultimately Rensi conﬁscates the corrupt ofﬁcial’s goods and gives them tothe trader. The ‘peasant’ in question thus comes to the notice of the kingwithout ever having entered his presence – via a courtier and his servant as
intermediaries.
Although ordinary people would not be able to approach the king, they
might catch sight of him as he journeyed between his palaces and the temples.The journey of the king between these structures, often accompanied byhis queen and sometimes by the royal daughters, is a common theme ofdepictions in courtiers’ tombs at Amarna (Fig. 7.2). The king most often
travels in a golden chariot. Sometimes the queen rides with him; on other
occasions she follows in her own chariot. The royal daughters and somehigh ofﬁcials also ride in chariots, while soldiers run along beside the groupand onlookers bow down and raise their arms in adoration of the king.Occasionally the king and queen are carried in sedan-chairs, similarly richlydecorated (Davies 1905b :pl. XIII). Royal barges serve a similar purpose on
longer riverine journeys: the barges are shown moored outside a palace inthe tomb of May (Davies 1908a :pl. V).
Inhis study of the courts of Vienna and Versailles, Duindam points to
signiﬁcant differences in the nature of court life at the two royal centres,contrasting the closed style of the Habsburg court which was small and
housed few courtiers with the openness of the French court at Versailleswhere thousands were housed and lived and shared entertainments with theroyalfamily (Duindam 2003 :308–9). It is the closed style of the Habsburg

Court and palace in ancient Egypt 295
court which seems to correlate more closely with what we know of the ancient
Egyptian court. Sites such as Amarna and Malkata, which will be discussed
in more detail below, show only limited provision of accommodation within
the palace compounds, and the number of high ofﬁcials associated with theinner palace is likely to have been limited (on this aspect of ancient courtssee also the chapters by Paterson and van Ess in this volume).
The closed style of the German courts is associated with creating distance
between the royal family and courtiers, whereas accessibility is created bythe open style (Baillie 1967 ). The existence of signiﬁcant differentiation
between king and courtiers ﬁts well with what we know of the Egyptiancourt and its role in promoting the divine aspects of the king’s nature. Even
in the Amarna period, the vast majority of our evidence for court life relatestoits ofﬁcial aspects: diplomatic activities; major royal ceremonies such as
thesed-festival (jubilee) and the Reception of Foreign Tribute in year 12
of Akhenaten’s reign; ofﬁcial visits to the temple; and reward ceremonies.Some unusual wall paintings from the palaces themselves show the royalfamily relaxing and eating (e.g. Arnold 1996 :ﬁgs.49and 108), and a few
private tomb depictions show similar images of the royal family. A depictionfrom the tomb of Huya, steward to the king’s mother, Queen Tiye, showsthe royal family eating, with the tomb owner waiting on them (Fig. 7.3). In
commissioning such a scene Huya is presumably depicting for all to see his
privileged access to the divine family.
Power, inﬂuence and intrigue
There is little documentary evidence from Egypt for the day-to-day intrigueswhich studies such as those of Elias ( 1983 )and Duindam ( 2003 )s u g gest
formed such a prominent part of court life. We have documentary evidencefor only the most serious court conspiracies: usually those culminating in theactual or attempted assassination of a ruler. A fascinating, although perhapsrather speculative study by Naguib Kanawati ( 2003 )c o llates textual evidence
for conspiracies from the Fifth Dynasty and analyses deliberate damage totombs, which Kanawati suggests was carried out in association with the
punishment of a group of palace ofﬁcials for involvement in the assassinationof King T eti. References found in the Old Kingdom autobiography of Weniat Abydos relate to another such incident in the reign of Pepi I and thecourtier’s role in investigating the plot (Lichtheim 1973 :19).
InAkhenaten’s boundary stelae we ﬁnd oblique and fragmentary ref-
erences to dissent, apparently culminating in year 5 of his reign and thus

296 kate spence
Figure 7.3 Huyain attendance on the royal family. The three principal ﬁgures are
Akhenaten’s mother, Tiye ( left), Akhenaten ( centre )and Nefertiti ( right ). The ﬁgure
raising a stick-like object is identiﬁed in the caption as Huya himself. From the tomb
of Huya at el-Amarna (Davies 1905b :pl. VI).
probably coinciding with the decision to move to Amarna. The dissent is
described as being worse than anything which had occurred in the reignsof his immediate predecessors (Murnane and van Siclen 1993 :41–2). Dis-
content and political struggles at court are thought likely to have led to the
signiﬁcant change in personnel which is apparent early in Akhenaten’s reignand the fact that a number of ‘new men’ are attested in high ofﬁce in hisreign. Beyond this we have no direct textual references to the events associ-
ated with the move to Amarna and the changes affecting religious practicesat the time.
The danger of being implicated in a court conspiracy is vividly illustrated
in the Middle Kingdom story of Sinuhe (Lichtheim 1973 :222–35). At the
beginning of the text Sinuhe is described as a servant of the royal harem in theservice of the wife of the crown-prince. Despite the apparent implicationsof his title, he was actually serving with the army under the leadership ofthe crown-prince when the reigning king, Amenemhet I, was assassinated,although only allusions are made to this event in the text.
25Sinuhe found
25The assassination is described in more detail in the ‘Instruction of King Amenemhet I’
(Lichtheim 1973 :135–9; Parkinson 1991 :48–52). According to this text the king was killed by
his own guards as he slept after supper. This text warns the future king against trusting a

Court and palace in ancient Egypt 297
himself in the wrong place at the wrong time and overheard a conversation
between one of the other royal sons and a messenger. T erriﬁed, he ﬂed: ‘Idid not plan to go to the residence. I believed there would be turmoil and Idid not expect to survive it’ (Lichtheim 1973 :224). He travelled to the Near
East and found favour and success in the service of foreign kings. We arenever told what it was that Sinuhe had overheard; his ﬂight perhaps seemsunlikely to a modern audience, given his claims of innocence. However, thepopularity of the text suggests that the idea of an innocent bystander beingimplicated in a plot was an entirely understandable premise for ﬂight to itsancient readers.
Sinuhe returns to Egypt in old age, after a lifetime spent promoting Egypt’s
interests abroad, tempted back by a letter from the king and the promise ofagood burial. He is given the title of ‘companion’ and is counted among
the ‘entourage’ . Meals are delivered to him four times a day from the palaceand he receives gifts and a fully equipped burial place. Interestingly, he isgiven‘ahouse and garden that had belonged to a courtier’ (Lichtheim 1973 :
233), which suggests that goods belonging to ofﬁcials might be conﬁscatedand reallocated by the king, a practice corroborated by other evidence.
26
Sinuhe’s early association with the harem is interesting as this seems to
have been the most prominent source of intrigue and conspiracy, for ratherobvious reasons (cf. van Ess in this volume). Egyptian kings often had manywivesofvarying rank. Ideally the eldest son by the ‘great royal wife’ succeeded
his father, but wives and their associates seem to have schemed to get theirownsons into positions of favour and ultimately onto the throne. The best-
documented case of a harem conspiracy occurred in the reign of RamessesIII ( c.1153 BC) and involved eleven harem ofﬁcials, a number of other
functionaries and six wives of ofﬁcials (de Buck 1937 ;K e m p 1989 :222;
Ver nus 2003 :108–20). The records even document an attempt to pervert
the course of justice by arranging for the investigating ofﬁcials to be seducedbybeautiful women (Vernus 2003 :119–20).
The threat posed by potential heirs (usually brothers or sons) as the foci for
court conspiracies must have been clearly apparent to reigning kings. From
the early Fifth Dynasty onwards, royal sons become difﬁcult to identify in therecord,and it seems that they were no longer given important ofﬁcial posts as
brother, friend or intimate: ‘It was one who ate my provisions that made insurrection; one to
whom I had given my arms was creating plots with them; one clad in my linen was looking atme as if needy; one anointed with my myrrh was passing water’ (Parkinson 1991 :50).
26The king thus uses both rewards and punishments in his treatment of courtiers. See also theMiddle Kingdom ‘Story of the Eloquent Peasant’ in which a corrupt ofﬁcial’s goods are
conﬁscated and bestowed on his victim (Lichtheim 1973 :169–84, esp. 182).

298 kate spence
they had been in the Fourth Dynasty (Strudwick 1985 :337–46). Presumably
this measure was intended to prevent them from building up power-bases
at court. From this date, evidence for king’s sons and brothers is very patchyuntil the Ramesside period, although there is occasional evidence for militarytraining or positions in the priesthood (Dodson 1990 ). Royal nurses and
tutors are known and some sons may have been brought up in provincialpalaces or perhaps even within the houses of trusted ofﬁcials or relativesrather than at the centre of court life. By contrast, wives and daughters are
more prominent and are sometimes represented in the company of the king.
The situation changed when a son was either designated as heir or
appointed to a co-regency with his father.
27Co-regencies were particularly
popular in the Middle Kingdom but are also known from the New Kingdom(Murnane 1977 ). As a practical way of ensuring a smooth transfer of power
the policy seems to have been successful, although the practice may seem tous difﬁcult to reconcile with Egyptian ideology which stressed the existenceof a single divine king. Akhenaten had an older brother, Thutmose, whowas designated as crown-prince but died before his father (Dodson 1990 );
while we know a little about this Thutmose, we know almost nothing aboutAkhenaten himself before he ascended the throne. It is possible that therewas a short co-regency with his father Amenhotep III, although this topicremains controversial (van Dijk 2000 :274–5). Evidence that Akhenaten was
concerned about the succession towards the end of his reign is provided
byclear evidence that he appointed a co-regent, named Smenkhkare (van
Dijk 2000 :279–81), although the identity and even the gender of this ruler
remain hotly debated topics.
Powermight also become concentrated in the hands of individuals, par-
ticularly if they acted as regent for a young king (as is the case with Horemhebduring the reign of Tutankhamun, although it was Ay who initially succeededTutankhamun when he died) or as vizier (Ramesses I was Horemheb’s vizier
and was appointed prince regent by him). In his coronation inscriptionHoremheb describes his relationship with the young king and his role in
27Ascene in Hatshepsut’s mortuary temple at Deir el-Bahari shows her investiture as royal heir
byher father, Tuthmosis I. In the scene, Tuthmosis I sits in a pavilion with Hatshepsut standing
before him as he presents her to his high ofﬁcials and courtiers (Naville 1898 :5–7,
pls. LX–LXII). Although this representation is usually interpreted as an invention of
Hatshepsut’s (post-rationalising her position as king) the basic premise of the scene must have
been credible. It should also be pointed out that an alternative explanation of the scene – thatHatshepsut was indeed chosen as heir but was prevented by factions at court from taking the
throne on the death of her father – seems to me equally likely. The fact that Tuthmosis III laterstates that he was chosen as king by an oracle of the god Amun suggests that his appointmentwas unusual and probably stage-managed to legitimise his position.

Court and palace in ancient Egypt 299
governing the country at that time in such a way that his consolidation of
power is clearly apparent:
When he [Horemheb] was called upon in the presence of the sovereign – the palace
having fallen into a rage – he opened his mouth and answered the king and made
him happy with what came out of his mouth. Unique and effective was he . . .
So he was administering the Two Lands for a period of many years. [The dues of
the Two Lands and the deliveries of Upper and Lower Egypt were] reported [to him],and the councils [came] to him, bowing at the gate of the king’s house. The chiefsof the Nine Bows, south as well as north, appealed to him, their arms outspread athis approach, as they did reverence to his face as to a god.
(Murnane 1995 :231)
However,kings could and did change the structure of ofﬁcial posts or create
new ofﬁces to manipulate the power balance. For example, at some point
in the early to mid Eighteenth Dynasty the ofﬁce of vizier was split intotwo, one governing the north and one the south of Egypt (Murnane 1998 :
201).
28That the king was able to upset the existing power balance at court
and within government is also clear from the fact that – regardless of hismotives – Akhenaten managed to disband the powerful and institution-alised priesthood of Amun, and presumably also that of Ptah at Memphis.Acomparison between Akhenaten’s changes and the power of Horemheb
during the reign of Tutankhamun only a few years after Akhenaten’s deathshows clearly that the power balance between king and courtiers was ﬂuidand prone to rapid and major shifts.
Akhenaten’s changes highlight the ability of the king to manipulate his
situation and succession to a certain extent. However, changes made in thereign of one king might rapidly become entrenched and could have far-
reaching implications. For the early Eighteenth Dynasty, this is particularly
apparent in the promotion of royal women culminating in the accessionof Hatshepsut. Ideally the Egyptian throne passed from father to eldestson by the principal wife, following the divine model in which the godHorussucceeded his father Osiris. Problems were particularly likely to occur
if the king produced no sons at all, or no sons by his ‘great royal wife’ .This situation seems to have arisen with surprising frequency in the earlyEighteenth Dynasty. The problem was solved by marriage to the wife or
daughter of the previous divine king and, despite several changes in the malelineage, the rulers of this period were grouped together in a single ‘dynasty’ .Intermsofthe succession, this placed emphasis on the transmission of
divinity through the mother and the female blood-line, with some of the
28Fordiscussion of administrative reforms in the Old Kingdom see in particular Kanawati ( 1980 ).

300 kate spence
more important queens given the title ‘God’s Wife of Amun’ (Bryan 2000 :
226–30). Princes by minor wives who became king also married half-sisters
by more senior wives.29This seems to have been widely accepted as a method
of legitimising political power and could elevate a prominent ofﬁcial andcourtier to the position of divine king.
30
Some of these women also wielded considerable political power, in addi-
tion to the inﬂuence that their proximity to the king may have offered (Bryan2000 :228); one of the Amarna Letters actually suggests that Akhenaten con-
sult his mother, Tiye, about a matter concerning international diplomacy(Moran 1992 :91). The power to appoint the holder of the ofﬁce of ‘second
prophet of Amun’ lay in the hands of the king’s wife from the beginningof the Eighteenth Dynasty (Bryan 2000 :229). Ahmose Nefertari, Hatshep-
sut and possibly also Ahhotep ruled as regents. The practice of appointingwomen (almost always mothers) to rule as regents for their sons dates back
at least as far as the Early Dynastic period, and is thought to have been con-sidered a good way of ensuring that the country was governed well in thebest interests of the child-king. However, during the early years of TuthmosisIII’s reign, Hatshepsut was elevated from regent to king. She was representedwearing the crowns and regalia of a king as a full co-regent with her step-
son. There has been much speculation on the nature of her relationship withTuthmosis III, but whatever the motivation behind her assumption of the
throne (in reality an ideological acknowledgement of her existing politicalpower – but a surprising one) there can be no doubt that she must havefound support for her actions within the court. Following this reign, royalwomen are considerably less prominent for several generations, perhaps a
deliberate attempt to prevent them amassing political power (Bryan 1998 :
40). However, in the Amarna period, royal women again become prominent,with Nefertiti appearing alongside Akhenaten with extraordinary frequency
(van Dijk 2000 :276). Both Nefertiti and Tiye had their own establishments
at Amarna with their own ofﬁcials and households, perhaps constitutingoffshoots or subsidiaries of the main royal court.
Women also ﬁgure prominently in the international diplomatic corre-
spondence with other Great Powers. Letters convey greetings between queens
29This is known to have happened in the case of Tuthmosis II. Tutankhamun and Smenkhkare
may also have married half-sisters (although questions about the parentage of these two kingsallow some room for doubt). Tuthmosis III was probably intended to marry his half-sisterNeferure (Hatshepsut’s daughter).
30Asmay have happened in the case of Tuthmosis I. However, Barry Kemp points out (personal
communication) that we know little about Tuthmosis I’s background. The ‘non-royal’
individuals who came to the throne during this dynasty could actually have been close relativesof previous kings who did not advertise these relationships for reasons of decorum.

Court and palace in ancient Egypt 301
as well as kings, and marriage was used as a way of cementing alliances
between these states (Schulman 1979 ;M e i e r 2000 ). Foreign bri des a rrived
in style, with considerable entourages, and must have brought alien cus-toms to the Egyptian court. No Egyptian princesses were sent abroad to
marry foreign kings in reciprocal arrangements during this period, a factthat has been interpreted as showing that the Egyptians felt themselves su-perior to the other powers with whom they corresponded (Schulman 1979 :
179–80). The extreme reluctance of Egyptian kings to send their daughtersabroad actually seems more likely to relate to the fact that royal daughtersfeatured so prominently in the process of political legitimisation, raisingthe possibility that these foreign kings or their children might be viewed aslegitimate heirs in the case of the Egyptian king’s death.
31Such an interpre-
tation corresponds well with the implications of an Amarna period queen’sextraordinary request to the Hittite king: ‘My husband died. I do not have
as o n…I fy o uw ould give me one of your sons, he would become my
husband . . . We are seeking a son of our Lord for the kingship in Egypt’
(quoted in Schulman 1979 :177 n. 1). This letter was most probably sent
byAnkhsenamun, third daughter of Akhenaten and wife of Tutankhamun
(Schulman 1978 ;Giles 1997 :311–21). The son in question was dispatched
but seems to have been intercepted and murdered before he reached Egypt.Ankhsenamun disappears from the record and Ay, a relatively elderly royal
relative who bore the title ‘god’s father’ and seems to have been inﬂuen-
tial during the reigns of both Akhenaten and Tutankhamun, took over thethrone of Egypt. Although evidence is slim outside the meagre evidence forconspiracies, it seems clear that courtiers must have played an important
role in managing smooth successions and that they also formed the main
players within factions when the succession was disputed.
These extreme cases must represent the tip of the iceberg. There must have
been a continual underlying stratum of faction and petty intrigue, ﬂatteryand back-stabbing, set against the perhaps volatile temper of an extremelypowerful ruler who could destroy a courtier. Horemheb’s pride in his abilitytocalm the pharaoh (see above) reﬂects more than the personal relation-
ship of ruler and advisor: it speaks of power achieved through the abilitytodiffuse a situation that must have been dangerous for many courtiers.
One of the most common genres of Egyptian literature at all periods isthe instruction text (Lichtheim 1996 ;Coulon 2002 ). These preach restraint,
31The only known husbands of royal princesses at this period are reigning kings, as a result of
which many relationships within the royal family were incestuous. Akhenaten probably‘married’ at least three of his daughters, with his eldest, Meritaten, serving as ‘great royal wife’
in the latter part of his reign.

302 kate spence
humility, fairness and respect for elders and superiors as the way to achieve
success, happiness and a good burial (the ultimate aim of all Egyptians). The
reason for the popularity of these texts becomes clear if they are interpreted
as a guide to weathering the difﬁculties of court life: avoiding pitfalls andachieving success within a very hierarchical society. Recommendations nottoengage in ‘a mischievous errand, nor be friends with him who does it’
(Lichtheim 1976 :150) prove particularly valuable advice within the danger-
ous world of court society and intrigue.
The residence city and the architectural settings of court life
The royal court was a social conﬁguration reliant on the presence or prox-imity of the king. The dwelling place of the king formed the setting forcourt life, and spatial aspects of the palace were fundamental to conﬁguring
and expressing status within the court and to structuring the relationshipsbetween king and courtiers. The architecture was also of immense impor-tance in projecting the king’s authority and situating him physically in rela-tion to other beings. Thus palaces are often located in close proximity totemples or even incorporated into them, expressing unequivocally the close
relationship between king and god (O’Connor 1989 ;1995 ;S p e n c eforth-
coming). Palaces were surrounded by high walls, excluding the majority of
the population and promulgating an air of secrecy around activities within,
as was also the case for temples.
32The sheer scale of preserved palaces in rela-
tion to all buildings other than temples also emphasises their importance.Egyptian kings had a number of palaces around Egypt and, interestingly,
these differ considerably in architectural layout.
33Kings also moved around
the country in boats and stayed in tents when on campaigns abroad; Akhen-aten himself records that he stayed in a ‘pavilion of matting’ on an early visittothe site of his new city at Amarna (Murnane and van Siclen 1993 :100,
105). Elements of the court must have travelled with the king and must havebeen conﬁgured in these temporary settings as well as in the larger palaces.
The king’s principal place of residence at any period was also the cen-
treofadministration and government, and was referred to simply as ‘the
Residence’ . This seems to have comprised at the very least a dwelling place and
ofﬁcial palace of the king, plus temples, institutional headquarters, ancillary
32Forthe semantic relationship between the notions of secrecy and seclusion in Egypt see
Hoffmeier ( 1985 :171–98).
33Forasummary of the most important evidence for New Kingdom palaces with bibliography
see Lacovara ( 1997 :24–41, ﬁgs. 19–38, 43); O’Connor ( 1989 ,1995 ); the latter focuses on
interpreting the symbolic signﬁcance of palace architecture.

Court and palace in ancient Egypt 303
structures and residential areas. Rolf Gundlach points to the important role
played by the main state temple and royal necropolis in relation to the Resi-dence, although it is so common to ﬁnd that these were situated far from theResidence (Gundlach 2004 :229) that they can hardly have been considered
aunity. Although the roots of the Eighteenth Dynasty were at Thebes – the
location of a number of royal palaces, the state temple at Karnak and theroyalnecropolis now known as the Valley of the Kings – increasing emphasis
had been placed on Memphis during the course of the Eighteenth Dynasty:royalpalaces were sited there, while the elite cemetery at Sakkara attests to
the important administrative activities in the region.
Textual information from the early Eighteenth Dynasty places the main
royalpalace at Thebes adjacent to the temple of Karnak, close to the pylon
of the main entrance (O’Connor 1995 :272–4); no archaeological traces of
the palace have yet been found. Amenhotep III built a major palace complexat Malkata on the west bank at Thebes for the celebration of his sed-festivals
or jubilees, held in the last decade of his reign (Kemp 1989 :213–17). He
also built a small, more isolated palace structure nearby at the site of Komel-’Abd (Kemp 1977 ),
34while the main residential palace was still situated
on the east bank near Karnak (O’Connor 1989 :79). There were several
NewKingdom palaces at Memphis, including one dating to the reign of
Tuthmosis I, and we have textual evidence for ‘mooring-places of Pharaoh’ ,
small structures in which the king could stay during his journeys around thecountry; estates might be attached to these to provide provisions for the royal
retinue when they stayed in the palace (Kemp 1989 :218). Provincial palaces
are also known from textual sources, and archaeologically from remains suchas those at Medinet el-Ghurob (Kemp 1978 ). From the later New Kingdom,
palaces were attached to the mortuary temples of kings, although the roleof these structures is a matter of debate.
Akhenaten’s royal residence at el-Amarna is the only example of a resi-
dence city for which we have very substantial archaeological remains,although a study by Peter Lacovara ( 1997 )has highlighted similarities with
the surviving features of other royal palatial centres such as Deir el-Ballasand Thebes. Four major palace structures can be identiﬁed at el-Amarna(Fig. 7.4), in addition to smaller royal pavilions within temples and at other
ceremonial sites.
35The major palaces seem to occur in pairs, with a relatively
small ceremonial palace (the North Palace) situated near the large, poorly
34Although Kemp interprets the archaeological remains of the palace as a platform, the plan
visible in the casemate construction of the f oundations rather suggests that it was a complete
building with a bedroom and columned halls.
35Kemp (1989 :261–317) provides an excellent introduction to the archaeology of the site.

304 kate spence
Figure 7.4 The Central City and North City at el-Amarna showing the location of the
four principal palaces and the Aten T emples (adapted from Kemp 1989 :ﬁgs.89and 91).
preserved residential palace at the far north of the site (the North Riverside
Palace), while in the ‘Central City’ the enormous ceremonial Great Palace is
linked by a bridge to the smaller King’s House. Alongside the palaces, the twomajor state temples to the Aten dominate the Central City, whilst admin-istrative buildings and institutional headquarters cluster around the King’sHouse, clearly marking the palace as the institutional and governmental hub
of the city (Kemp 1976 ). Residential suburbs were constructed to the north

Court and palace in ancient Egypt 305
and south of the Central City. Akhenaten was probably the only king of the
NewKingdom to construct his tomb outside Thebes: he began work on a
new royal necropolis in a prominent valley to the east of his city. A seriesof diverse ceremonial settings were constructed around the outskirts of thenatural bay in which the city was sited (Kemp 1995 ). He thus reintegrated
state cult and the royal necropolis with the governmental and ceremonialaspects of kingship within a single residence city.
The area within the palace walls was the setting for most aspects of court
life, and the very different architectural settings in existence in the distinctpalaces at Amarna or Thebes suggest varied activities, although ascribingfunction to whole palaces or rooms within palaces remains problematicalgiven the nature of our evidence.
36Insome palaces, the presence of mem-
bers of the court is apparent and is structured architecturally through theprovision of housing. The small palace of Amenhotep III at Kom el-’Abd wasconstructed on a raised brick platform which was approached by a broad
ramp (Fig. 7.5).
37Within the same enclosure wall, but built at ground level
and divided from the palace itself by walls and courtyards, were the remainsof seven houses within a single courtyard. One house is larger than the rest,although its whole plan is still smaller than the major room of the palace.Four smaller houses of identical plan and siz earegrouped together, while at
the back of the court are two further houses, similar in scale to those found inthe group of four but with a slightly different arrangement of rooms whichsuggests that they were of lesser importance. The complex perhaps providesaccommodation for seven courtiers and their own families and staff, care-fully differentiated from the king’s residence and also distinguished fromeach other to reﬂect three ranking groups. T o be chosen to accompany theking in such a restricted entourage must have been a sign of great favour,and the speciﬁc allocation of houses within would presumably have been
carefully noted by contemporaries.
Residences for ofﬁcials and court staff have also been found at Malkata
(Lacovara 1997 :56–7, ﬁg. 20). Close to the main palace complex was a
group of three large houses and eleven smaller dwellings, clearly planned as
aunit and built as an integral part of the complex. Further small houses were
constructed behind the North Palace at the site, while the nearby unexcavated
‘north village’ shows signs of organic growth suggesting that it was not built
as a part of the complex. The provision of housing for essential members
36O’Connor ( 1989 )divides the palaces into governmental, ceremonial and residential palaces.
There is likely to have been signiﬁcant overlap in these functions between palaces.
37Suchraised construction appears to have been co mmon in New Kingdom palace architecture
and is found in palaces at Ezbet Helmi and Deir el-Ballas (Lacovara 1997 :ﬁg.43).

306 kate spence
Figure 7.5 The remains of the palace of Amenhotep III at Kom el-’Abd. The palace
itself has eroded down to below the level of the brick ﬂoor in many places and only the
raised foundations remain. Accommodation for the entourage is to the east (adapted
from Kemp 1977 :ﬁ g . 2).
of the court or royal household thus appears to be a feature even of thislarger palace, but the accommodation is still very limited indeed. Malkataserved as the setting for important jubilee celebrations and presumably themajority of courtiers gathered there for the occasion were expected to ﬁndtheir own accommodation, camp or travel back to their houses in the maincity on the east bank of the Nile.
AtAmarna, very little in the way of accommodation for courtiers is
found within palaces. Both the King’s House (Fig. 7.6)and the North Palace
(Fig. 7.7)contain a fairly large house within the grounds in addition to the
palatial accommodation. It seems likely that this was for the use of a palaceoverseer as, in the case of the North Palace, it is situated within the ser-
vicequarters. Other than that, only very small lodgings are found within the
palace, presumably for more humble attendants and servants, although eventhen the provision of such accommodation seems very limited.
38Houses also
38Itshould also be noted that dwellings can be very difﬁcult to distinguish from places of work or
‘ofﬁces’ as in some cases the architecture is very similar. The role of smaller units within the
palaces is therefore difﬁcult to establish with certainty.

Court and palace in ancient Egypt 307
Figure 7.6 The King’s House at el-Amarna. A bridge linked this palace with the Great
Palace on the opposite side of the royal road (Pendlebury 1951 :pl. XVI).

308 kate spence
Figure 7.7 The North Palace at el-Amarna (adapted from Stevenson Smith 1981 :
ﬁg. 304).
seem to have been provided for chief priests in proximity to their temples
and for the overseers of institutions within those compounds. The majorityof courtiers at Amarna seem to have been expected to construct their ownhouses, which they did in large numbers, creating residential suburbs tothe north and south of the royal and administrative heart of the city. Thesehouses in turn were surrounded by the smaller dwellings of dependants(Kemp 1989 :ﬁg.97). Kemp ( 1989 :314) estimates that about half the houses
at the site have been excavated and that, on the basis of size, we shouldestimate that between 120 and 240 of these belonged to ‘ofﬁcials’ . Thesewould have included those employed in the governmental and military
institutions and the priesthood; among the largest of these are presumablyfound those who served as courtiers. The vast majority of these peopleprobably also had dwellings in Thebes and/or Memphis, and many mustalso have had holdings in their home towns.
The architecture of palaces serves to differentiate the king from his people
and to structure the relationships between the king and his courtiers. All

Court and palace in ancient Egypt 309
palaces are walled and internally subdivided with varying degrees of com-
plexity. T o the majority of the population the interior of the palace wouldhave been a mystery; to be allowed access to even the outer parts of a palacewould have differentiated an individual from the majority of Egyptians and
marked him or her out as privileged. Access within the palace was furthercontrolled by doo r-keepers and courtiers, such as the overseers of the vari-
ous parts of the inner palace (Murnane 1998 :216). How far a courtier was
allowed to progress towards the inner parts of the palace would have beenclearly readable to all others as a mark of status. Horemheb’s Edict fromKarnak makes it clear that it was a question not only of access but also of the
manner of entry: ‘They enter through the portals of . . . quickly, by horse, tothe holy area, with a dog at their feet and an attendant at the rear’ (Murnane1995 :240). Presumably only the very privileged were permitted to ride into
the palace and enter quickly with servants and dogs. Again, such speciﬁcdetails of behaviour, for which evidence very rarely survives in Egypt, wouldhave been immediately readable both to eye-witnesses (i.e. other courtiersand servants) and, by report, to those excluded from the palace.
The status achieved through access to inner parts of the palace and other
royalstructures perhaps also goes some way towards explaining a couple
of the more unusual aspects of Amarna period elite art. Private tombs atAmarna often depict plans of palaces and temples. These can be difﬁcult
torelate to the archaeological remains found on the ground; the confusion
stemming partly from Egyptian drawing conventions but also unquestion-ably from a degree of inaccuracy, which was clearly not considered a problembythe artists and tomb owners. These depictions provide a temporal setting
for the royal activities depicted in scenes, but the architectural representa-tions also give the impression that the tomb owner has knowledge of thelayout of the interior of royal buildings (however mistaken this actually is)and that he therefore has unfettered access to all parts of palaces and temples.The representations could then be interpreted as an indication of status. Sec-ondly, stelae depicting intimate scenes of the royal family relaxing beneaththe rays of the Aten (e.g. Arnold 1996 :ﬁg.88) have been found as objects of
devotion in prominent locations in a few private houses at the site. Althoughthe emphasis in interpreting these objects and their contexts has been onthe royal family as the major focus of private devotion (Ikram 1989 )i ti s
perhaps also signiﬁcant that these individuals are depicting private activitiesof the royal family within their homes, again displaying prominently theirprivileged access.
Differentiation between the king and those around him is also estab-
lished architecturally within the palace building. The focus of the palace is

310 kate spence
the royal throne, raised on a dais. It is usually centrally positioned within the
throne room, and the approach to the dais is carefully structured architec-turally and often reﬂected in decoration (Kemp 1989 :ﬁg.77; Weatherhead
1992 ). Palaces usually also contain a private suite including a niched bed-
room for the king. Elsewhere I have divided the architectural ordering of
Egyptian palaces into axial, semi-axial and non-axial structures based on the
relationship between the entry and the approach to the throne room. This
distinction most probably reﬂects the nature of ceremonial activity that tookplace within them, although it should be emphasised that there is likely tobe considerable overlap (Spence 1998 :209–14; forthcoming). The bilateral
symmetry found in axial palaces throws emphasis on the enthroned kingand focuses attention on the distance between a person arriving at the gateand the king, an impression further heightened by the many liminal spaceswhich must be crossed in order to enter his presence. The North Palace(Fig. 7.7)and the Great Palace at Amarna are examples of axial palaces.
39
Non-axial palaces feature a large columned hall approached through a
smaller columned hall. Beside or near the large columned hall is a smaller
room containing a throne dais. There is no grand or formal entrance to the
interior of the palace and no axial entrance into the throne room. This is amore private setting for the king, perhaps even the ‘audience hall’ in whichhe conducted day-to-day business and consulted with his most importantofﬁcials and courtiers. The King’s House at Amarna (Fig. 7.6)and the North
Palace at Malkata are of this type. Semi-axial palaces have non-axial entrance
sequences but internal structures which relate the throne room of theking to a number of subordinate chambers; examples include the so-calledHarimPalace at Malkata and parts of the eastern wing of the Amarna Great
Palace.
The remote location of Amenhotep III’s palace at Kom el-’Abd (a non-
axial palace) suggests that it served as a retreat from the centres of court life(Kemp 1977 ), and the palace within it thus seems likely to represent the bare
minimum of an architectural setting within which the king could functionand court life could take place. The majority of smaller royal palaces in theNewKingdom are actually of a very similar size, as are the more private suites
of rooms within much larger palaces (Spence forthcoming), suggesting acommon understanding of the core features making up the king’s dwelling.
Itis interesting to note that the King’s House at Amarna (Fig. 7.6), which
appears from its architecture, lack of axiality and tortuous entrance sequencetobe the most private of royal settings, is also the place which Kemp identiﬁes
39Also the Palace of Merenptah at Memphis, treated in depth by O’Connor ( 1995 ).

Court and palace in ancient Egypt 311
as the likely centre of government, given the fact that it is surrounded by
ofﬁces and institutional buildings (Kemp 1976 ;1989 :287–8). Although this
combination seems counter-intuitive to modern architectural readings, the
role of the domestic setting as the site of royal decision-making is well attested
in court studies (Duindam 2003 :223, 234–48) and goes some way towards
explaining the prominent role and inﬂuence that might be accorded to the
royalfamily, favourites and members of the household.
Within the many royal palaces, the king, and sometimes also his family,
interacted with courtiers and members of his household. In addition to hisarchitectural setting and throne, he was distinguished from those aroundhim by his dress, crowns and other royal regalia, perhaps by his authoritativespeech and also by his posture, gestures and behaviour. Amarna tomb scenessuggest that few stood upright in the presence of the king and there are cleardegrees of bowing, grovelling and scraping the ground adopted by foreigndignitaries, courtiers and servants in the presence of the king (Figs. 7.2,7.3,
7.8–7.10). It seems likely that this is a fair representation of interaction inthe palace. The Amarna tomb scenes and other objects from the site alsoshow the royal family embracing, kissing, relaxing and eating in a mannerentirely alien to the strict decorum usually governing Egyptian art (e.g.Arnold 1996 :ﬁgs.88, 93–8, 108). Y et courtiers are not depicted acting in the
same relaxed manner and it seems that such familiar behaviour may alsohave served to differentiate the king’s interaction with his family from hisformal engagement with inferiors.
40
Court ceremonial
Ceremonial at court covers a range of activities, from occasional festi-vals of kingship such as coronations, sed-festivals (jubilees) and funerals,
41
through speciﬁc court ceremonies, such as the appointment and rewardingof ofﬁcials, to ritualised everyday activities of the king including his regular
40The only other representations of such behaviour are found in ceramic statuettes of monkeys,
which are sometimes shown driving chariots or embracing each other with a similar lack ofreserve. However, while these have been interpr eted as satirical, potentially indicating resistance
tosuch practices and disapproval of the regime (Frankfort and Pendlebury 1933 :99), it is
worthnoting that the majority of the statuettes derive from the houses of the courtiers who
proﬁted most from the king (Freed et al. 1999 :156), and also that the king himself included a
silver statuette of a monkey and her child in gifts sent to one of his counterparts in the NearEast: ‘1 (female) monkey, with its daughter on its lap, of silver’ (Moran 1992 :30).
41InTutankhamun’s tomb a group of courtiers headed by the vizier is shown dragging the king’s
mummy in its bier (Reeves 1990 :72).

312 kate spence
interaction with courtiers; I will discuss brieﬂy examples of major festivi-
ties involving the court, the ‘Window of Appearance’ reward scenes and theeveryday activi ties of the court.
Major festivals
Akhenaten celebrated a sed-festival at Karnak in the early years of his reign
before the residence was moved to Amarna (Gohary 1992 ). The timing of
his jubilee celebration was unusual: this ancient ceremony associated withthe renewal of divine kingship was usually celebrated after thirty years onthe throne and perhaps more frequently thereafter; Amenhotep III had cel-ebrated jubilees in years 30, 34 and 37 (Kemp 1989 :213–17). Much of the
evidence for Akhenaten’s sed-festival derives from a building he constructed
at East Karnak, consisting of a massive colonnaded court decorated with stat-ues of the king and relief scenes depicting episodes in the jubilee festivities;relief scenes deriving from other structures built by Akhenaten at Karnak
are also known. The king obviously features very prominently in the deco-ration but courtiers, priests, guards and members of the household are also
shown. Courtiers are repeatedly shown bowing in the presence of the king(Redford 1984 :114–15, ﬁg. 7.13) and in one scene a courtier pours a libation
(Redford 1984 :122). Servants bustle around the palaces and bearers carry
the king’s palanquin, while attendants and soldiers are also present. A scenereconstructed from fragments shows the king wearing his jubilee costume
leaving the palace (Redford 1984 :ﬁg.7.12). As he emerges, the chief priest
of the sun cult precedes him, bowing and backing away from the king andburning incense to mark the passage of the ruler. Twelve chamberlains areshown prostrate before him. Feasting seems to have been an important partof the ceremonies.
Although the gods (or in Akhenaten’s case the Aten) played an important
role in the sed-festival, courtiers, the royal household, priests and ofﬁcials
would have been the major participants and witnesses, alongside the royal
family. The importance of their role is clear from the surviving scenes fromKarnak. Comparable but very fragmentary scenes dating to the Fifth Dynasty
and the reign of Amenhotep III have been found which show that the impor-tant role played by courtiers was by no means unique to Akhenaten’s reign(Redford 1984 :125). However, one very interesting difference in the repre-
sentation of courtiers occurs between the reliefs carved in Amenhotep III’sreign and those produced for Akhenaten. In Amenhotep III’s reliefs promi-
nent courtiers and priests are named while Akhenaten’s reliefs use only titles:the courtiers are thus stripped of their personal identity in his depictions

Court and palace in ancient Egypt 313
and appear in the presence of the king only as functionaries (Redford 1984 :
131–2).
Inyear 12 of Akhenaten’s reign, when king and court were well estab-
lished at Amarna, Akhenaten held a ‘Reception of Foreign Tribute’ which
is represented in some of the private tombs at the site. The reliefs in Huya’stomb (Davies 1905b :pls. XIII–XV) show the king and queen being carried
tothe site of the festival in a palanquin accompanied by the royal daughters
and attendants, while the relief in the tomb of Meryre II (Fig. 7.8;D a v ies
1905a :pl. XXXVII) shows the royal family seated in a pavilion. A priest
censes the empty palanquin, recently vacated by the king. Courtiers fuss
near the entrance to the pavilion, and around it are lined up grandees fromall over the Egyptian empire and sphere of inﬂuence, many bearing tributeor gifts and many with their arms raised in adoration of the royal family.Although these scenes do not show the detailed episodes of the Karnak sed-
festival reliefs, the splendour of the occasion and the organisation that musthave gone into it are clear from the depictions. They also illustrate the factthat Akhenaten’s court did not exist in isolation: there was frequent inter-action between courtiers and emissaries from other states and the Egyptiancourt was on display. The potential for foreign inﬂuence from royal wives,
envoys (Egyptian and foreign) and gifts is very signiﬁcant. An Amarna Letterchides Akhenaten for making foreign emissaries stand around in the sun for
hours on end, presumably as part of one of his ceremonial reviews (Moran1992 :39).
The ‘Window of Appearance’ reward scenes
The ‘Window of Appearance’ scene in which the king rewards and sometimespromotes a courtier is one of the most common scenes found in Amarnatombs. It is also the only common type of depiction which focuses on direct
interaction between king and courtier: although courtiers appear in manyAmarna period scenes of royal activity, the attention of the king is rarely
focused on these individuals. The scene is always architecturally structuredand highly formalised. The king, often accompanied by the queen and royaldaughters, appears at a ‘window’ which is raised on a platform and frontedbyacanopy (Fig. 7.9). The king leans out of the window and tosses golden
collars to the courtier who stands in front of the window with his arms raised,
while other ofﬁcials tie the collars around his neck. T exts make it clear thatthe golden collars usually shown were accompanied by other gifts; scribesare shown making records, presumably of value given the presence of scalesin some depictions (e.g. Davies 1905: pl. XVI). The ceremony singled out a

Figure 7.8 The reception of foreign tribute in year 12. From the tomb of Meryra II at el-Amarna (Davies 1905a: pl. XXXVII).

Figure 7.9 Akhenaten and Nefertiti at a Window of Appearance rewarding the ofﬁcial Parennefer who stands with his arms upraised
(in the second register up) wearing the gold necklaces he has been thrown by the king. The palace is represented behind the window
(Davies 1908b: pl. IV).

316 kate spence
courtier, bestowing both wealth and prestige. The frequency of its depiction
in the tombs at Amarna suggests that it was the highlight of an ofﬁcial’s
career.
The architectural setting of the scen eisparticularly interesting as it illus-
trates the care taken to differentiate king from courtier in representations
of direct interaction. The tomb scenes situate the windows within palaces,although in one instance the depiction appears to show a window in a royalrest-house within a temple complex (Davies 1903 :pls. XXV–XXVI) and a
window is also shown on a royal barge (Davies 1908a :pl. V). The majority
of representations appear to locate the windows between the outer courtof the palace and the inner parts of the building, on the axis of the palace,where this can be established (see Fig. 7.2,top left).
42Presence at the cer-
emony was thus restricted to those allowed access to the palace, temple orroyalbarge but participants were not allowed to enter the inner rooms.
Instead the king emerged from the mysterious sanctity of the inner parts
of his residence to interact formally with the courtiers. The window, lyingon axis with the throne room, provides a physical barrier between king andcourtier through which, crucially, neither can pass, stressing the ontolog-
ical difference between humans and the divine king. Other aspects of thearchitecture also stress this difference: the king was raised above the level ofthe assembled crowd while the window was decorated with potent imagesof royal authority. At Amarna, where a split lintel results in the windowbeing more of a balcony than a true window,
43the external canopy would
have thrown the fac ¸ade of the window and the courtier being rewarded into
shadow, whilst bright sun would have ﬂooded down, both highlighting andrendering shadowy and indistinct the ﬁgure of the king in the window. This
42Egyptian drawing conventions, although highly logical, can be difﬁcult to read. The Amarna
representations of palaces are governed by sche matised plans but include elements in section
and elevation. The actual location of the Windows of Appearance on the ground at Amarna
remains controversial. I would locate the principal windows in the North Palace (Spence 1999 )
and perhaps also the Great Palace. I interpret them primarily as part of axial or ceremonialpalace structures. However note that Kemp ( 1976 )locates windows in the King’s House and
Small Aten T emple. It is clear that they may also have formed elements of other structures from
which the king made an appearance such as the royal barge.
43Although in the later preserved example at Medinet Habu this feature is indeed a window(Kemp 1989 :211–13, ﬁg. 73), the Amarna examples actually seem to be balconies as they have a
broken lintel and were probably unroofed. The split lintel is a common feature of religiousarchitecture at Amarna. In traditional Egyptian architecture the lintels of doorways throughwhich the king passed on his way through a temple were decorated with a winged sun-disk,associating his directed movement with that of the sun and all its attendant cosmologicalimagery. At Amarna, the lintels of doorways are instead divided so that the direct contactbetween the actual rays of the sun and the king remained unbroken as he moved through thebuilding.

Court and palace in ancient Egypt 317
relationship between the sun-disk and the king is graphically depicted in
the reliefs as the hands in which the sun’s rays terminate reach down and
touch the king.
However,many of these same courtiers were probably allowed access
tothe inner palace in other circumstances and presumably interacted less
formally with the king during the daily routine of life in the city (for whichsee below). The ‘Window of Appearance’ scene was popular for a number ofreasons. First, it formalised the relationship between king and courtier and
crystallised it in a manner which allowed this relationship to be depicted; thiswas particularly important because each scene shows the king distinguishingone individual. Second, both the ceremony itself and the rewards distributedhad important communicative value in stressing the status of the individualand his favour with the king. The rewards are likely to have had signiﬁcantvalue (gold is prominent amongst the gifts) but luxury items are also found:among the rewards thrown to Ay is a pair of red gloves which he is shownwearing as he emerges from the ceremony to be greeted outside the palace
bymembers of his entourage (Fig. 7.10; Davies 1908a :pls. XXIX–XXX).
44
The Amarna tomb scenes feature only the highest ofﬁcials and courtiers
who had tomb chapels in which such scenes could be recorded. It is thereforedifﬁcult at Amarna to establish the frequency of such ceremonies or whetherthey were used to reward those other than top ofﬁcials. Horemheb’s edictfrom Karnak refers to the king’s use of the Window of Appearance everytendaystowatch a parade of the palace guards, during which he called to
ofﬁcers by name and tossed them rewards additional to their basic rations(Murnane 1995 :239–40). As part of the ceremony each man was seated
(presumably in the courtyard outside the window) and served with food.
This suggests that at least in the post-Amarna period the window was usedregularly for formalised interactions bet ween the king and broader sections
of the royal household and other ofﬁcials.
Royal favour expressed materially had signiﬁcant communicative value,
which was emphasised still further by the ritual nature of these rewardscenes in which access to the palace, interaction with the king, individualpreferment and material enrichment were combined in one representation.It seems t oh a v eb e e nc o n s i d e r e d acceptable to beg signs of favour from the
king (often in the form of burial equipment) and these were occasionallyrecorded. An Old Kingdom tomb inscription describes asking for a favour
and receiving more than was requested: ‘When I begged of the majesty of mylord that there be brought for me a sarcophagus of white stone from Tura,
44Apair of intricately woven gloves was found in the tomb of Tutankhamun (Reeves 1990 :156–7).

318 kate spence
his majesty had a royal seal-bearer cross over with a company of sailors
under his command, to bring me this sarcophagus from Tura. It came withhim in a great barge of the court, together with its lid, a doorway, lintel, twodoorjambs and a libation-table. Never before had the like been done for anyservant’ (Lichtheim 1973 :19).
Daily activity within the palace
Although we have little information on day-to-day activity within the palacethere are clear indications that the king’s routine was formalised and thatbehaviour within the palace was carefully structured. ‘The Duties of theVizier’ is an important text dating to at least as early as the beginning of
the Eighteenth Dynasty and outlines the principal duties of the king’s chiefminister (van den Boorn 1988 ;L o r t o n 1991 ). Every morning the vizier
received reports on the ‘affairs of the two lands (i.e. Egypt)’ in his house. He
then proceeded to the palace; his entry is described as follows:
Heshall enter in the direction of the pr-c3(Great House) facing the treasurer, as he
takes his position at the northern ﬂagstaff, and the vizier shall then move in from the
east in the doorway of the great double gate. Then the treasurer shall come to meethim and report to him, saying, ‘All your affairs are sound and prosperous, and everyresponsible functionary has reported to me, saying, “All your affairs are sound and
prosperous”’ . Then the vizier shall report to the treasurer, saying, ‘All your affairs are[sound] and pros[perous], and the closing of all the enclosures on time and theiropening on time have been reported to me by every responsible functionary’ . Thenafter each of the two of[ﬁcials] has reported to the other, the vizier shall send toopen every doorway of the pr-nsw (palace).
(Lorton 1991 :296)
This shows the extent to which the meeting of the treasurer and the vizier
was ritualised. Following the opening of the palace, the vizier entered andwent to greet the king. The text stresses the ceremonial aspect of this activity
rather than the nature of the meeting, but the fact that information on
the state of affairs was passed on to the king can be inferred (Lorton 1991 :
307). Old Kingdom evidence points to the importance of the royal lev ´ee
with very high-status courtiers bearing titles associated with the ‘House
of Morning’ involved with the king’s toilet (Kees 1914 ;Blackman 1918 :
esp. 149–2). There is little evidence for similar activity from later periods,although the ‘great chamberlain in the Great House’ Amenhotep who servedunder Amenhotep III seems to have been involved in the ceremonial dressingof the king (Murnane 1998 :216 n. 173). Akhenaten’s chamberlain, Tutu,
left an account of his behaviour in the palace and morning meetings with

Court and palace in ancient Egypt 319
the king, although the stress here is on the king giving orders rather than
receiving information. ‘My voice is not loud in the king’s house. I do not
swagger in the palac e…Ia c to n l ya c c o r ding to what he decrees as my
charg e…f o re v e r ym o r ning he rises early to instruct me.’ He also reveals
that he does not accept bribes and he does not think bad thoughts in thecompany of the favourites (Murnane 1995 :192).
Conducting temple rituals was an impor tant aspect of royal activity and
many of the tomb scenes from Amarna depict the king travelling betweenthe palace and temple in a gilded palanquin (Davies 1905: pl. XIII) or chariot(Fig. 7.2;D a v ies1903 :pl. X). These become signiﬁcant processions: the royal
family is often accompanied by attendants, and guards line the route. Longerjourneys were presumably taken in the royal barges depicted in the tomb ofMay (Davies 1908a :pl. V). Preparing for such journeys and accompanying
the king to the temple or further aﬁeld presumably formed a major part ofcourt life.
The process of decision-making must also have formed a prominent part
of life in the palace.
45Lorton points to texts which ‘conﬁrm the very real
connection, in the eyes of the ancient Egyptians, between the physical palace
and the royal throne within it, and the king’s exercise of his functions ofgovernance’ (Lorton 1991 :306); he provides a particularly clear exposition of
the relationship between the king and the vizier in terms of decision-making.AMiddle Kingdom stela describes its owner as ‘preceding the courtiers who
draw nigh to the palace; knower of [the utterance (?)] of difﬁcult (?) words(?) on the day the courtiers speak; who reports to the king in solitude;who is near of place on the day of foregathering; one to whom the kingdivulges his speech to act as a gateway for it’ (Fischer 1960 :261). This points
tothe ritualised activity masking the actual exercise of decision-making:
the ranking of courtiers approaching the palace, the difﬁculties inherent inadvising the king within the setting of the council chamber, and the courtiercharged with broadcasting the king’s decision following a private meeting.
The chamberlain Tutu’s tomb inscription, quoted above, suggests a similarsituation at Amarna (Murnane 1995 :192).
K¨onigsnovelle texts refer to the king sitting in his ‘audience hall’ consult-
ing with his advisors. Horemheb’s Karnak Edict speciﬁes that individualswerearranged according to role and rank within the audience hall: ‘I have
arranged the protocol of the inner palace, the custom of the inner quartersof the king’s companions. I have set my house t o…,t h e y…t h eh e r alds
45Goelet ( 1986 )also touches on the important governmental role of the palace although his focus
is on ceremonial activity.

320 kate spence
of the council chamber, according to their rank, while “spreading holiness”
throughout the entire house, and while the members of the king’s entourageare at their exact places and the Thirty are at their customary positions’(Murnane 1995 :240). The ritualised nature of activity within the palace is
clear here, as is the importance attached to protocol.
46Reports might also
be made within such a context (Lorton 1991 :306) and ambassadors from
important states in the Near East would presumably also have been receivedwithin the palace complex, presumably again in the audience hall.
Feasting was clearly of great importance within the palace.
47Refere nces in
the Amarna Letters suggest that foreign rulers expected invitations to dineat the palace (Moran 1992 :7)although presumably such invitations were
rarely, if ever, taken up in person. Evidence for feasting or the provision
of food for the court is signiﬁcant (e.g. Murnane 1995 :237). Horemheb’s
Karnak Edict seems to suggest that the guard feasted outside in a courtyard
while the king was present at the Window of Appearance (Murnane 1995 :
239). Papyrus Boulaq 18 (late Middle Kingdom) lists ofﬁcials feasting atthe palace, although Quirke suggests that the royal family was not actu-ally present (Quirke 1990 :42–3). It thus seems likely that the royal family
actually dined separately at feasts: either in a separate room or somewhereelse altogether (cf. Brosius in this volume on similar Achaemenid practices).Amarna reliefs from Karnak depict the king and queen feasting in the main
room of the palace (Redford 1984 :pl. 16), with food piled in front of them,
waited on by servants. Reliefs from the tomb of Huya at Amarna showAkhenaten and Nefertiti dining with the king’s mother, Tiye, and the royaldaughters (Fig. 7.3;D a v ies1905b :pls. IV–VII). Courtiers are in attendance
and Huya himself directs the people waiting on the king. However, none ofthe courtiers is shown eating in the presence of the royal family. Troops ofmusicians play beneath the main scene.
External appearance was clearly considered of great importance amongst
courtiers, and changing styles of dress, hair and jewellery can be traced
from surviving private statues and images. Wearing clean, white linen wasconsidered a mark of status in Egypt at all periods: presumably having clean
clothes indicated that one did not engage in manual activities. However,cloth was a valuable commodity and the amount of cloth required to createaparticular garment seems to have been indicative of status: thus the Amarna
reliefs contrast the loin-cloths and short kilts of soldiers and servants with
the three-quarter-length kilts of higher-status ofﬁcials (see Fig. 7.10;for
46Compare arrangements for imperial audiences in Han China: van Ess in this volume.
47Asit was in Egyptian society in general: scenes of feasting are prominent in Eighteenth Dynasty
tombs.

Figure 7.10 Reward scene from the tomb of Ay. Ay and his wife catch the rewards thrown from the Window of Appearance by the
royalfamily. Outside the palace (upper right), Ay is greeted by his supporters (Davies 1908b: pls. XXIX–XXX).

322 kate spence
clothing see Vogelsang-Eastwood 1993 ). The highest-ranking men wear a
three-quarter-length tunic over which a kilt is tied with a billowing fold of
cloth falling at the front.48Quality of cloth was also important and cloths
weregraded according to the ﬁneness of the weave (Vogelsang-Eastwood
2000 :285). High-status clothes in the New Kingdom are also often depicted
as semi-transparent, presumably an indication of the ﬁneness and delicacyof the cloth. Coloured and embroidered clothes are known from the tombof Tutankhamun and occasional archaeological ﬁnds (Vogelsang-Eastwood2000: 278–81), although coloured garments are rarely depicted in tombscenes, except on foreigners. Sandals seem frequently to have been removedas a mark of respect in the king’s presence: few are shown wearing sandals inthe king’s presence in private tomb scenes from Amarna, while Horemheb’sKarnak Edict seems speciﬁcally to refer to wearing sandals in the council
chamber as a special privilege accorded to the highest ofﬁcials. The Middle
Kingdom story of Sinuhe refers to ‘the choice perfume of the king and his
favourite courtiers’ (Lichtheim 1973 :233) and cosmetics and perfumes are
also commonly found amongst burial goods.
The life of the Amarna courtier beyond the palace
The king and the palace thus seem to have been the major focus of eliteactivity both at Amarna and at other periods of Egyptian history. The extenttowhich ofﬁcials were involved in palace life must have varied, with some
based in the palace itself as court ofﬁcials while others played important rolesas priests or military ofﬁcials and as such were presumably based at templesor in the institutional structures found in the Central City at Amarna. Thesemen were presumably also close to the king (as is certainly suggested bymany of their titles) and thus also played a role at court. The possession ofatomb was considered a mark of great status; Kemp ( 1989 :314–15) points
out that there are too few tombs at Amarna to correspond with the numberof ofﬁcials, and suggests that a tomb-plot was allocated as a favour by theking. This appears to be conﬁrmed by a statement on a small votive stelafrom the tomb of Any: ‘We (?) have seen the good things which the goodruler has done to his Scribe of the Altar. He has ordered for him a goodly
burial in Akhetaten’ (Davies 1908a :10). Other textual references from the
48Only as a result of the introduction of the vertical loom in the Eighteenth Dynasty did it
become possible to weave cloths of the width to allow such differentiation in dress. Old andMiddle Kingdom clothes are depicted as tighter and more restrictive.

Court and palace in ancient Egypt 323
Amarna tombs conﬁrm that the king provides burials: ‘May you grant me
the good funeral which is your Ka’s to give, in the tomb in which you decreed
for me to rest’ (Murnane 1995 :144; see also 178, 183). The Amarna tomb
scenes are very focused on royal activity as the source of status, prestige andmaterial wealth. The Window of Appearance scenes indicate time after timethe importance of royal favour, and courtiers must have vied continually forpositions of inﬂuence despite the paucity of evidence for such behaviour atthis period.
All courtiers and ofﬁcials seem to have constructed mansions at Amarna,
although a few of the most important of them also had ‘ofﬁcial’ residences(found adjacent to temples and within institutional buildings and palaces).Ofﬁcials were able to express their status and wealth through the size of theirhouses and by means of particular status symbols within them (Crocker1985 ;T i e tze1985 ;1986 ;Shaw 1992 ). A few houses show evidence of sig-
niﬁcant enlargement when the circumstances of the house owner improved(Spence 2004 :146–8). Hospitality also seems to have been an important
part of life amongst the elite: feasting with music and dancers is frequentlydepicted on Theban tomb wall paintings and the scenes show both menand women gathering together. Such social activities were presumably alsoprominent at Amarna. The royal family was sometimes represented in thesehouses within shrines as objects of worship.
There are certain parallels between palace architecture and elite houses
at Amarna in terms of spatial layout. This is likely to reﬂect parallels in thehierarchical nature of social relations structured by architecture:
49although
the individual courtier was clearly subordinate at court and existed withinastrict ranking system relative to his peers, within his own home he was
master. He sat to receive guests in the principal hall of his house on a chairplaced on a brick dais. A separate dwelling constructed within the grounds ofhis house was most probably for a deputy or chief attendant: this subordinaterole was in fact a privileged position which might even involve access to the
palace (Murnane 1995 :240). Outside the enclosure wall of the courtier’s
house dependants constructed small dwellings close to the source of theirownincome and status (Kemp 1989 :294, ﬁg. 98). The dual position of the
49Lacovara ( 1997 :57–60) argues that Amarna houses are not typical of Egyptian domestic
architecture but are a conscious borrowing from palace design. However, clear parallels for the
layout of the Amarna houses can be found in the Middle Kingdom planned town of Lahun(Arnold 1989 )and in New Kingdom houses at sites such as Malkata and Amara West,
suggesting that the explanation is more complex than this. The difﬁculty in assessing therelationship between Amarna houses and structures elsewhere has been exacerbated by
difﬁculties in interpreting Egyptian architectural drawings (Spence 2004 ).

324 kate spence
courtier within the hierarchy at court and in his own milieu is well illustrated
byAy’sreward scene (Fig. 7.10;Davies 1908: pls. XXIX–XXX). In the main
part of the scene Ay is rewarded by the king, the Queen and three princesses
at the Window of Appearance. Ay and his wife are shown slightly bent (thereare at least four distinct depths of bow visible in this one scene, apparentlyrelating at least in part to rank) with their arms raised in adoration and to
catch the rewards thrown to them. However, as soon as Ay moves beyond thepalace gates he is shown standing upright. Although those waiting outsidedo not bow to him they raise their arms in jubilation, a few kneel before him(and the trays of rewards) and one kisses his feet. The focus of the attentionof onlookers shifts from the king inside the palace to Ay himself outside it.Asimilar sequence of events is shown in the tomb of Tutu (Davies 1908b :
pls. XIX–XX): here supporters of Tutu kiss the ground before him.
This illustrates the deeply hierarchical nature of Egyptian society and
probably goes some way towards explaining the embeddedness of the socialstructure of the court within that society (cf. the similar observations ofBrosius in this volume, on Achaemenid Persia). As discussed above, the
relationship between king and courtier is expressed through service and
subordination on the part of the courtier and the giving of favour in theform of rewards and titles (along with the accompanying income) on thepart of the king. Manifestations of service and favour within this king–courtier relationship dominate the historical record. However, if one looks
at a broader range of – admittedly rather limited – evidence, it is clear thatthe hierarchy stretches far beyond this. The king himself serves the godsthrough temple ritual and by building and ﬁghting on behalf of the gods.Akhenaten himself is shown prostrate before the Aten, subordinating him-self and his family to the gods, and even as far back as the Pyramid T extsthe king makes himself useful to the gods through service in some of thetexts: for example in rowing Re to the west (Faulkner 1969 :158).
50The rela-
tionship between king and gods is usually expressed in royal texts as ﬁlial(i.e. subordinate but related). The subordination of courtiers and ofﬁcialstothe king has been discussed at length here, and the subordination of Ay’s
entourage has also been mentioned. In a fascinating study of servant ﬁg-urines placed in Old Kingdom elite tombs, Ann Macy Roth has pointed outthat the majority of these ﬁgures bear the names of close family members ofthe deceased courtier, and she interprets the relevance of these as allowingsubordinates to participate in the afterlife of the courtier through provisionof menial service (Roth 2002 ).
50Inmany other utterances in the Pyramid T exts, however, the gods are said to serve the king.

Court and palace in ancient Egypt 325
Conversely, the king is given favour by the gods for his service, expressed
through long life, prosperity and military victories. Members of the elite
would presumably reward their subordinates and dependants ﬁnancially
and provide them with titles and privileges. One particular favour involveddepicting the dependant within the tomb chapel of the courtier or ofﬁcial,thus allowing him to participate in the offering cult for the tomb owner. Thisis particularly prominent in Middle Kingdom tombs such as those at BeniHasan (e.g. Parkinson 1991 :78–81). However, in the small tomb of Any at
Amarna were found six small votive stelae (Davies 1908a :9–11, pls. XXI–
XXIII; Murnane 1995 :124–5). Five of these show the individual dedicating
the stela offering to or making recitations for Any. The sixth shows Any beingdriven by ‘The charioteer of the Royal Scribe Any’ , showing that courtiersthemselves must have given titles (or at least named positions) to theirsubordinates. All the dedicatees have titles, even if only ‘servant’ , although itis not clear that this title was bestowed by Any himself in all cases. Each musthave had the means to commission such a stela, however lowly a title suchas ‘servant’ may appear to modern eyes.
51These jobs presumably brought
with them income and prestige within the household. As we move down the
social hierarchy, evidence becomes very thin, but it is clear from differentialwages paid in kind (primarily bread and beer) that some individuals wouldhave been capable of supporting more dependants than others. This is alsoclear in the division of rations within a single household seen in the lettersof the Middle Kingdom farmer Hekanakht (Parkinson 1991 :101–7; Allen
2002 ).
The symbiotic relationship between the servant and the served thus
seems to stretch from the top to the bottom of Egyptian society. Serviceof some description ensured a position within this broad hierarchy andstatus was presumably reﬂected to some degree in the number of an indi-vidual’s dependants. Those with ofﬁcial titles were linked into the admin-
istrative hierarchy which covered all regions of Egypt (and its empire) andall branches of the administration. The highest ofﬁcials were then linked tothe court and brought into the orbit of the king himself. It is the embed-dedness of the court as the link between the king and the hierarchicalstructure of the administration and society as a whole which must haveensured its longevity despite numerous changes of kings, dynasties and courtpersonnel.
51The six stelae vary in the quality of the carving although all appear to have been carved by
professional artists. Interestingly, the stela dedicated by Any’s brother May is one of the poorerexamples (Davies 1908a :pl. XXIII).

326 kate spence
Conclusions
Despite the patchy nature of the evidence, aspects of the Egyptian court can
be traced from before the beginning of the historical period, through the lateEighteenth Dynasty and beyond, into the Hellenistic court of the Ptolemies,
attesting to its extraordinary longevity as an institution. The existence of aking and, by extension, his court never seems to have been challenged inover 3,000 years of history, a fact that illustrates the extent to which king
and court were embedded within the Egyptian world-view and social andadministrative structures, but also how successfully the court managed theprocess of self-legitimisation, primarily through the ﬁgure of the king.
There is no doubt that the Egyptian king was clearly distinguished from
both courtiers and the population as a whole by his divine status, despitethe fact that he might have some decidedly human attributes. However, theevidence suggests that he was not just a divine ﬁgurehead for the country,
but actively ruled it through his courtiers/administrators. There is evidencefrom several periods in Egyptian history of active reform of the administra-tion – and here we might include Akhenaten’s disbanding of the powerfulpriesthood of Amun, showing the extent to which some kings were abletoshift the balance of power between themselves and their ofﬁcials. While
Elias ( 1983 )suggests that kings in the early modern French court controlled
their courtiers by playing them off against non-noble administrators (butsee the Introduction; also Paterson in this volume), the Egyptian king’scourtiers also doubled up as administrators. Thus important ofﬁcials were
kept focused on life and competition at court rather than their institutional
power, and the king could keep his courtiers occupied and accountable.
Itis easy, however, to fall into the snare set by the Egyptian historical
documents which encourage us to vie wthe king as all-powerful and acting
in isolation. Courtiers clearly did have power, as the troubled accession ofHatshepsut as co-regent with Tuthmosis III suggests. Although there was
technically no hereditary nobility and those from humble backgrounds did
occasionally rise to positions of power, there can be no doubt that much ofthe time positions of power close to the king were monopolised by a smallnumber of powerful families, sometimes also linked to the royal family by
marriage. It was at times of succession of a new king that courtiers were attheir most powerful, as is shown by the few conspiracies that we know of,but also by the royal practice of appointing co-regents to ensure that controlwas established before the death of the reigning king. There are a signiﬁcantnumber of instances in which a prominent courtier managed to take over

Court and palace in ancient Egypt 327
the throne at the death of the king: the processes of legitimisation rapidly
established him as a divine ruler and the court continued with its business.Wesee only the extremes, and search for the ‘power behind the throne’ in
the case of child-kings or female rulers, but how many of the other kingswho form the subject of our histories were strong individuals and how manyweremanipulated by their advisors? We are unlikely ever to know, but it is
worthoccasionally re-examining our assumptions. Ultimately, the dynasties
who monopolised the ofﬁce of kingship changed many times, and it is thecourt and not kingship itself that underpins the stability of the Egyptian
social system.
Anystudy of the Egyptian court rests on generations of excellent work
byEgyptologists in many branches of the discipline: literature, history and
social history, prosopography, and art, architectural history and archaeology.Focusing on the court, however problematical it may prove as a topic, allows
reﬂect ion on the nature and context of our evidence and the interrelation-
ships between interpretations derived from the various sources. Instructiontexts, for example, take on added signiﬁcance in the light of the world of
faction, competition and intrigue highlighted by studies of court society(e.g. Elias 1983 ), despite the virtual absence of information on this aspect of
Egyptian court society from the primary sources (Coulon 2002 ). The same
might be said of domestic architecture. Comparative studies of courts proveinvaluable in reminding us just how much evidence we are missing in Egyptand in suggesting how we might go about reassembling the surviving piecesof the jigsaw.
Although the Amarna period and its aftermath tend to be treated with
great caution by historians as a result of the upheavals surrounding Akhen-
aten’s religious changes, the wide range of sources it yields provide richmaterial for examining the court. Courts tend to be conﬁgured and con-ceived spatially, and the scope for examining ar chitectural settings is widely
acknowledged in court studies (e.g. Baillie 1967 ;Elias 1983 ;Duindam 2003 ).
Amarna provides an excellent opportunity for exploring such conﬁgurations
and settings through the archaeological record, while there is enough com-parative evidence in the form of palace architecture and art that we canrest assured that conclusions drawn from this period are not irrelevant to
broader trends within the Egyptian court. The Window of Appearance, forexample, which was developed as an essential component of court life in
the Amarna period, remained an important element of palace architectureat least for the remainder of the New Kingdom. There remains considerablescope for study in this area.

328 kate spence
The Egyptian evidence also offers a fascinating opportunity to examine
the origins of the court as it emerged alongside kingship as part of a complex
process of social stratiﬁcation that can be observed in the archaeologicalrecord.Duindam ( 2003 :318) writes that ‘it makes no sense to use “court”
as a bland equivalent for group dynamics around leaders’ . This may betrue, but the origins of the court in such behaviour are clear, as are the
various culturally determined strat egies for legitimising existing political
power through processes of differentiation – whether that power was gainedrapidly through conquest or built up gradually over centuries. This perhaps
explains both the frequency with which courts are encountered around the
world across the centuries, and the fact that they prove so difﬁcult to model.
Itis the processes that can be modelled, while the courts themselves remain
diverse although with frequent parallels. In a long-term view, the Egyptianevidence also illustrates the extent to which court society was continually
evolving in unexpected directions as the actors sought to cope with the ﬂuid
nature of power relations between king and courtier and to ﬁnd new ways ofexpressing the differentiation between king and courtier and courtier and
subject, so essential to maintaining the status quo.

Bibliography
Abka’i-Khavari, M. 2000. DasBild des K¨ onigs in der Sasanidenzeit .Munich
Adamson, J. 1999a. The making of the ancien-r ´egime court 1500–1700. In Adamson
1999b: 7–41
(ed.) 1999b. The Princely Courts of Europe: Ritual, Politics and Culture under the
Ancien R ´egime 1500–1700 .L o n d o n
Aldred, C. 1988. Akhenaten: King of Egypt .L o n d o n
Alf¨oldi, A. 1952. AConﬂict of Ideas in the Late Roman Empire .O xford
1962. A coin portrait of Julian Apostata. AJA 66: 403–5
1970. Diemonarchische Repr¨ asentation im r¨ omischen Kaiserreiche ,Darmstadt
(reprint of: (i) Die Ausgestaltung des monarchischen Zeremoniells im r ¨omi-
schen Kaiserhofe. MDAI(R) 49 (1934): 3–188; (ii) Insignien und Tracht der
r¨omischen Kaiser. MDAI(R) 50 (1935): 3–158)
Allen, J. P . 2002. The Hekanakht papyri .N e wY o r k
Altheim, F. 1962. Geschichte der Hunnen ,B dV .B e r l i n
Altheim, F. and Stiehl, R. 1954. Ein asiatischer Staat: Feudalismus unter den Sasaniden
und ihren Nachbarn .Wiesbaden
1957. Finanzgeschichte der Sp¨ atantike .Frankfurt
Andronicos, M. 1984. Verg ina.Athens
Anson, E. M. 1985. Macedonia’s alleged constitutionalism. Classical Journal 80:
303–16
Antoine, M. 1989. Louis XV .P a r is
Apostolid `es, J.-M. 1981. Leroi-machine: spectacle et politique au temps de Louis XIV .
Paris
Archibald, Z. 1999. Macedonian administration. CRn.s. 49: 163–5
Arnold, D. 1996. The Royal Women of Amarna: Images of Beauty from Ancient Egypt.
New York
Arnold, F. 1989. A study of Egyptian domestic buildings. VariaAegyptiaca 5: 75–93
Asch,R.G.2003. Nobilities in Transition 1550–1700: Courtiers and Rebels in Britain
and Europe .L o n d o n
AschR. G. and Birke, A. M. (eds.) 1991. Princes, Patronage and the Nobility: The
Court at the Beginning of the Modern Age .O xford
Atkinson, J. E. 1994. ACommentary on Q. Curtius Rufus’ Historiae Alexandri Magni
Books 5 to 7,2 .Amsterdam
Aujolat, N. 1983. Eus ´ebie, H ´el`ene et Julien I–II. Byzantion 53: 78–109 and 421–52

330 Bibliography
Avery, W. 1940. The adoratio purpurae and the importance of the imperial purple
in the 4th century. MAAR 17: 66–79
Badian, E. 1958. Alexander the Great and the unity of mankind. Historia 7: 425–44
1981. The deiﬁcation of Alexander the Great. In H. J. Dell (ed.), Ancient M acedo-
nian Studies in Honor of C. F. Edson ,28–71. Thessaloniki
1985. Alexander in Iran. In I. Gershevitch (ed.), The Cambridge History of Iran ,
vol. II: The Median and Achaemenian Periods ,420–501. Cambridge
1996. Alexander the Great between two thrones and heaven: variations on an old
theme. In A. Small (ed.), Subject and Ruler: The Cult of the Ruling Power in
Antiquity, JRA Suppl. 17: 11–26
2000. Conspiracies. In Bosworth and Baynham 2000: 50–95
Baer, K. 1960. Rank and Title in the Old Kingdom: The Structure of the Egyptian
Administration in the Fifth and Sixth Dynasties. Chicago
Baillie, H. M. 1967. Etiquette and the planning of state apartments in baroque
palaces. Archaeologia 101: 167–99
Baines, J. 1990. Restricted Knowledge, Hierarchy, and Decorum: Modern Percep-
tions and Ancient Institutions. Journal of the American Research Centre in Egypt
27: 1–23
1995. Origins of Egyptian kingship. In D. O’Connor and D. Silverman (eds.),
AncientEgyptian Kingship ,95–156. Leiden
1998. The dawn of the Amarna age. In D. O’Connor and E. H. Cline (eds.),
Amenhotep III: Perspectives on His Reign ,271–312. Ann Arbor, MI
Balsdon, J. P . V . D. 1962. Roman Women .L o n d o n
1969. Lifeand Leisure in Ancient Rome .L o n d o n
Bang, M. 1921. Die Freunde und Begleiter der Kaiser. In Friendl ¨ander 1921: 56–76
Bard, K. 1994. From Farm ers to Pharaohs: Mortuary Evidence for the Rise of Complex
Society in Egypt .Shefﬁeld
Bardill, J. 1997. The Palace of Lausus. AJA 101: 67–95
1999. The Great Palace of the Byzantine emperors. JRA 12: 217–28
Barnes, T. 1981. Constantine and Eusebius. London
1982. The New Empire of Diocletian and Constantine .Cambridge, MA and London
1989. Emperors on the move. JRA 2: 247–61
Bartsch, S. 1994. Actors in the Audience: Theatricality and Doublespeak from Nero to
Hadrian .Cambridge, MA
Baynes, N. 1955. Byzantine Studies .L o n d o n
Beik, W. 1985. Abso lutism and Society in Seventeenth Century France .Cambridge
Bek, L. 1983. Quaestiones Convivales :the idea of the Triclinium and the staging of
convivial ceremony from Rome to Byzantium. Analecta Romana 12: 81–107
Berger, P . 1981. The Insignia of the Notitia Dignitatum. New Y ork
Berve, H. 1926. DasAlexanderreich auf prosopographischer Grundlage ,2vols. Munich
Bickerman, E. 1963. A propos d’un passage de Chares de Mytilene. PdP 18: 241–55
Bielenstein, H. 1954. The restoration of the Han Dynasty, part I. Bulletin of the
Museum of Far Eastern Antiquities 26: 1–209

Bibliography 331
1976. Lo-yang in Later Han times. Bulletin of the Museum of Far Eastern Antiquities
48: 1–144
1979. The restoration of the Han Dynasty, part IV . Bulletin of the Museum of Far
Eastern Antiquities 51: 1–300
1980. The Bureaucracy of Han Times .Cambridge
1986a. The institutions of Later Han. In Twitchett and Loewe 1986: 491–5191986b. The Later Han Dynasty. In Twitchett and Loewe 1986: 279–87
Bird,H.(trans. and comm.) 1993. Eutropius: Brev iarium. Liverpool
(trans. and comm.) 1994. Sextus Aurelius Victor :DeCaesaribus. Liverpool
Blackman, A. M. 1918. The House of the Morning. Journal of Egyptian Archaeology
5: 148–65
Blockley, R. 1983. Fragmentary Classicizing Historians of the Later Roman Empire .
Liverpool
Boardman, J. 2000. Persia and the West .L o n d o n
Borchardt, L. 1913. DasGrabdenkmal des K¨ onigs S’a
3Hu-Re’. Band II:DieWand-
bilder .Leipzig
Borza, E. 1983. The symposium at Alexander’s court. AncientMacedonia 3: 45–55.
Thessaloniki
1999. Before Alexander: Constructing Early Macedonia. Publications of the Asso-
ciation of Ancient Historians 6. Claremont, CA
Bosworth, A. B. 1980. Alexander and the Iranians. Journal of Hellenic Studies 100:
1–21
1988. Alexander the Great: Conquest and Empire. Cambridge
1995. AHistorical Commentary on Arrian’s History of Alexander ,vol. II: Commen-
tary on Books IV–V .O xford
1996. Alexander and the East: The Tragedy of Triumph .O x f ord
2000. A tale of two empires: Hern ´an Cort ´es and Alexander the Great. In Bosworth
and Baynham 2000: 23–49
Bosworth, A. B. and Baynham, E. J. (eds.) 2000. Alexander the Great in Fact and
Fiction.O xford
Boucharlat, R. 1997. Susa under Achaemenid rule. In J. Curtis (ed.), Mesopotamia
and Iran in the Persian Period: Conquest and Imperialism 539–331 BC ,54–67.
London
2001a. Pasargades 2001. Rapport sur les travaux de la mission conjointe soumis
`al’Organisation du Patrimoine culturel de l’Iran, Direction de la recherche,
www.achemenet.com/recherche/sites/pasargades/pasargades.htm
2001b. The palace and the royal Achaemenid city. Two case studies – Pasargadae
and Susa. In I. Nielsen (ed.), The Royal Palace Institutions in the First Millennium
BC,113–24. Aarhus
Bourriau, J. 2000. The Second Intermediate Period (c. 1650–1550 BC). In Shaw
2000: 185–217
Bowersock, G. 1996. The vanishing paradigm of the Fall of Rome. Bulletin of the
American Academy of Arts and Sciences 49: 29–43

332 Bibliography
2000. Selected Papers on Late Antiquity .B a r i
Bowersock, G., Brown, P . and Grabar, O. 1999. Late Antiquity: A Guide to the Post-
classical World .L o n d o n
Bremmer, J. 1978. An imperial palace guard in heaven: date of the Vision of
Dorotheus. ZPE 75: 82–8
Brent, A. 1999. The Imperial Cult and the Development of Church Order .S u p plement
toVCTexts and Studies 45. Leiden, Boston and Cologne
Briant, P . 1985. Les Iraniens d’Asie mineure apr `es la chute de l’Empire ach ´em´enide.
Apropos de l’inscription d’Amyzon. DHA 11: 167–95
1989a. Histoire et id ´eologie: les Grecs et la ‘d ´ecadence’ perse. In M.-M. Mactoux
and E. Geny (eds.), M´elanges Pierre L´ evˆeque ,vol. II, 33–47. Besanc ¸on
1989b. Table du roi, tribut et redistribution chez les Ach ´em´enides. In P . Briant
and C. Herrenschmidt (eds.), Letribut dans l’empire perse ,35–44. Paris and
Louvain
2001. Bulletin d’histoire ach´ em´enide ,vol. II. Paris
2002 (French original 1996). From Cyrus to Alexander: A History of the Persian
Empire .Winona Lake, MN
2003. Darius dans l’ombre d’Alexandre .P a r is
Brilliant, R. 1963. Gesture and Rank in Roman Art .N e wH a ven (=MCAAS 14)
Brooker, G. 1977. The Civic World of Early Renaissance Florence .P rinceton
Brosius, M. 2000. The Persian Empire from Cyrus II to Artaxerxes I .L o n d o n
2002. Women in Ancient Persia 559–331 BC .O xford
2003. Alexander and the Persians. In J. Roisman (ed.), Alexander the Great ,169–93.
Leiden
2006. Investiture I. Achaemenid times. EncIr XIII: 181–7
Brown, P . 1971. The World of Late Antiquity. London
1992. Power and Persuasion in Late Antiquity .Madison, WI
1995. Review of T. F. Mathews, The Clash of Gods (1993) .A r t Bulletin 77: 449–
502
1997. The world of late antiquity revisited: a report. SO72: 5–32
1998. Christianization and religious conﬂict. In A. Cameron and P . Garnsey (eds.),
The Cambridge Ancient History ,vol. XIII, 632–64. Cambridge
Brunt, P . A. 1976–83. ArrianHistory of Alexander andIndica, 2 vols. Loeb Classical
Library. Cambridge, MA
1988. The emperor’s choice of amici .InP.Kneissl and V . Losemann (eds.),
Alte Geschichte und Wissenschaftsgeschichte: Festschrift f¨ ur Karl Christ zum 65
Geburtstag ,39–56. Darmstadt
Bryan, B. 1991. The Reign of Tuthmosis IV .Baltimore
1998. Antecedents to Amenhotep III. In D. O’Connor & E. H. Cline (eds.), Amen-
hotep III. Perspectives on His Reign :27–62. Ann Arbor, MI
2000. The Eighteenth Dynasty before the Amarna period (c.1550–1352 BC). In
Shaw 2000: 218–71
Buck, D. 1988. The reign of Arcadius in Eunapius’ Histories. Byzantion 68: 15–46

Bibliography 333
Burkert, W. 1985 (German original 1977). Greek Re ligion .O xford
Butz, R. and Dannenberg, L.-A. 2004. Theoriebildungen des Hofes. In Butz et al.
2004: 1–41
Butz, R., Hirschbiegel, J. and Willoweit, D. (eds.) 2004. Hofund Theorie:
Ann¨aherungen an ein historisches Ph¨ anomen .Cologne, Weimar and Vienna
Cameron, A[lan]. 1968. Gratian’s repudiation of the pontiﬁcal robe. JRS58: 96–102
1970. Claudian: Poetry and Politics at the Court of Honorius .O x f ord
1982. The empress and the poet: paganism and politics at the court of Theo-
dosius II. YClS27: 217–90
1985. Polyonomy in the Late Roman aristocracy. JRS75: 164–82
2000. The poet, the bishop and the harlot. GRBS 41: 175–88
Cameron, A[veril]. (ed.) 1969/70. Agathias on the Sassanians. DOP 23/24: 67–183
1976. Corippus .InLaudem Justini Augusti Minoris. London
1987. The construction of court ritual: the Byzantine Book of Ceremonies. In
Cannadine and Price 1987: 106–36
Cannadine, D. and Price, S. 1987. Rituals of Royalty: Power and Ceremonial in Tra-
ditional Societies .Cambridge
Carney, E. D. 2000a. Artiﬁce and Alexander history. In Bosworth and Baynham
2000: 263–85
2000b. Women and Monarchy in Macedonia .Norman, OK
Carrata-Thomas, F. 1995. Il problema degli eteri nella monarchia di Alessandro Magno .
Turin
Carter, E. and Stolper, M. W. 1984. Elam: Surveys of Political History and Archaeology.
Berkeley
Cereti, C. G. 1997. Primary sources for the history of Inner and Outer Iran in the
Sasanian period. In Archivum Eurasiae Medii Aevi 9: 17–71
2001. La letteratura pahlavi .Milano
Chavannes, E. 1895–1905. Les m´ emoires historiques de Se-ma Ts’ien ,5vols. Paris
Cheng Ch’iao 1987. T’ung-chih .Peking
Chilver, G. E. F. 1979. AHistorical Commentary on Tacitus’ Histories I and II .O xford
Ch’¨uT’ung-tsu 1972. Han Social Structure .Seattle
Coarelli, F. 1978. Il ‘grande donario’ di Attalo I. In IGalli e l’Italia 231–55. Rome
Cohen, R. and Westbrook, R. 2000. Amarna Diplomacy: The Beginnings of Interna-
tional Relations. Baltimore
Costa, E. 1972. The ofﬁce of castrensis sacri palatii in the 4th century. Byzantion 42:
358–87
Cotton, H. 1984. The concept of indulgentia under Trajan. Chiron 14: 245–66
Coulon, L. 2002. Cour, courtisans et mod `eles ´educatifs au moyen empire. Egypte
Afrique & Orient 26: 9–20
Couvreur, S. 1899. Li Ki ou M´ emoires sur les biens´ eances et les c´ er´emonies ,2vols. Ho
Kien Fou
Crocker, P . 1985. Status symbols in the architecture of el-Amarna. Journal of Egyptian
Archaeology 71: 52–62

334 Bibliography
Crook, J. A. 1955. Consilium Principis: Imperial Councils and Counsellors from Augus-
tus to Diocletian .Cambridge
1996. Augustus: power, authority, achievement. In A. K. Bowman et al. (eds.),
The Cambridge Ancient History ,2nd edn, vol. X, 113–46. Cambridge
Da Vinha, M. 2004. Lesvalets de chambre de Louis XIV .P a r is
Dagron, G. 1974. Naissance d’une capitale: Constantinople et ses institutions .P a r is
1996. Empereur et prˆ etre: ´ etude sur le ‘c´ esaro-papisme’ .P a r is
Dalby, A. 1996. Siren Feasts: A History of Food and Gastronomy in Greece .L o n d o n
and New Y ork
Dandamayev, M. A. 2002. Courts and courtiers. In EncIr 6: 356–9
D’Arms, J. 1990. The Roman convivium and the idea of equality. In O. Murray (ed.),
Symp otica ,308–20. Oxford
Daryaee, T. 1995. National history or Keyanid history? The nature of Sasanid Zoroas-
trian historiography. Iranian Studies 28: 129–141
2001/02. Memory and history. The reconstruction of the past in late antique
Persia. Nˆame-ye Irˆ an-e Bˆ astˆan1/2: 1–14
2002. Mind, body, and the cosmos: chess and backgammon in ancient Persia.
Iranian Studies 35: 281–312
Daskalopoulos, M. 1993. Divinit ´es et cultes en Mac ´edoine dans l’antiquit ´e. Unpub-
lished PhD thesis, University of T ours
Davidson, J. 1997. Courtesans and Fishcakes .L o n d o n
Davies, N. de G. 1903. The Rock Tombs of El Amarna ,P a r tI :The Tomb of Meryra .
EEF Memoir 13. London
1905a. The Rock Tombs of El Amarna ,P a r tII:The Tombs of Panehesy and Meryra
II.EEF Memoir 14. London
1905b. The Rock Tombs of El Amarna ,Part III: The Tombs of Huya and Ahmes .
EEF Memoir 15. London
1906. The Rock Tombs of El Amarna ,P a r tI V : Tombs of Penthu, Mahu, and others .
EEF Memoir 16. London
1908a. The Rock Tombs of El Amarna ,P a r tV :Smaller Tombs and Boundary Stelae .
EEF Memoir 17. London
1908b. The Rock Tombs of El Amarna ,P a r tVI:Tombs of Parennefer, Tutu, and A¨ y.
EEF Memoir 18. London
de Blois, F. 1990. Burz¯ oy’s Voyage to India and the Origin of the Book of Kalilah wa
Dimna .L o n d o n
de Buck, A. 1937. The Judicial Papyrus of Turin. Journal of Egyptian Archaeology 23:
152–64
de Crespigny, R. 1975. The harem of Emperor Huan: a study of court politics in
Later Han. Papers on Far Eastern History 12: 1–42
Deichmann, F. 1989. Ravenna: Haupstadt des sp¨ atantiken Abendlandes 2.3, Stuttgart
Delmaire, R. 1995. Lesinstitutions du Bas-Empire romain de Constantin ` aJustinien .
I.Lesinstitutions civiles palatine .P a r is
1997. Les usurpateurs du Bas-Empire et le recrutement des fonctionnaires. In
Paschoud and Szidat 1997: 111–27

Bibliography 335
Demandt, A. 1989. Der Hof. In Demandt, Die Sp¨ atantike ,vol. II.1.b, 231–44. Munich
Dickmann, J. A. 1999. Domus frequentata: Anspruchsvolle Wohnen in pompeianischen
Stadthaus .Munich
Dodson, A. 1990. Crown Prince Dhutmose and the royal sons of the Eighteenth
Dynasty. Journal of Egyptian Archaeology 76: 87–96
Dorman, P . 1988. The Monuments of Senenmut. London
Drake, H. A. 2000. Constantine and the Bishops ,Baltimore
Dubs, H. 1938, 1944, 1955. The History of the Former Han Dynasty ,3vols. Baltimore
Duindam, J. 1995. Myths of Power: Norbert Elias and the Early Modern European
Cour t.Amsterdam
2003. Vienna and Versailles: The Courts of Europe’s Dynastic Rivals, 1559–1780 .
Cambridge
2004. Norbert Elias and the history of the court: old questions, new perspectives.
InButz et al. 2004: 91–104
Dumser, E. A. (ed.) 2002. Mapping Augustan Rome. Journal of Roman Archaeology
Supplementary Series 50, Portsmouth, RI
Dunbabin, K. M. 1991. Triclinium and stibadium. In W. J. Slater (ed.), Dining in a
Classical Context ,121–48. Ann Arbor, MI
Duval, N. 1997. Les r ´esidences imp ´eriales. In Paschoud and Szidat 1997: 127–
55
Dvornik, F. 1955. The emperor Julian’s ‘reactionary’ views on kingship. In
F.Weitzmann (ed.), Late Classical and Medieval Studies in Honour of A. M.
Friend,71–81. Princeton
Eck, W. 2000. The emperor and his advisers. In A. K. Bowman, P . Garnsey and
D.Rathbone (eds.), The Cambridge Ancient History 2nd edn, vol. XI, 195–213.
Cambridge
2002. Imperial administration and epigraphy: in defence of prosopography. In
A. K. Bowman, H. M. Cotton, M. Goodman and S. Price (eds.), Representa-
tions of Empire: Rome and the Mediterranean World .Proceedings of the British
Academy 114, 131–52. London
Eck, W., Caballos, A. and Fern `andez, F. 1996. Das se natus consultum de Cn. Pisone
patre .Vestigia 48. Munich
Edwards, C. 2003. Incorporating the alien: the art of conquest. In C. Edwards and
G. Woolf (eds.), Rome the Cosmopolis ,44–70. Cambridge
Eichert, O. 1967 (1893). Vollst¨andiges W¨ orterbuch zu dem Geschichtswerke des Quin-
tus Curtius Rufus .Hildesheim
Elias, N. 1983. The Court Society .O xford
Elliott, J. H. and Brokliss, B. (eds.) 1999. The World of the Favourite .N e wH a venand
London
Elster, J. 1993. Political Psychology .Cambridge
Enderlein, V . and Sundermann, W. (eds.) 1988. Sch¯ahn¯ame Das persische K¨ onigs-
buch. Miniaturen und Texte der Berliner Handschrift von 1605 .Leipzig and
Weimar
Engler, F. K. 1980. DerGoldherr besteigt den weißen Tiger .Z u r ich

336 Bibliography
Erman, A. and Grapow, H. 1930. W¨orterbuch der ¨Agyptischen Sprache ,vol. IV . Leipzig
Errington, M. 1996. Assembly, Macedonian. In OCD3192–3
Faulkner, R. O. 1962. AConcise Dictionary of Middle Egyptian .O x f ord
1969. The Ancient Egyptian Pyramid Texts .O xford
Favro, D. G. 1996. The Urban Image of Augustan Rome .Cambridge
Finsen, H. 1969. La r´esidence de Domitien sur le Palatin. Copenhagen
Fischer, H. G. 1960. The Inscription of ’In-it.f ,B o r no f Tﬁ.Journal of Near Eastern
Studies XIX: 258–68
Fowden, E. 1999. The Barbarian Plain: Saint Sergius between Rome and Iran. The
Transformation of the Classical Heritage 28. Berkeley
Fraisse, J. C. 1974. Philia: la notion d’amiti´ edans la philosophie antique .P a r is
Frank, R. I. 1969. Scholae Palatinae :The Palace Guards of the Late Roman Empire .
Rome
Franke, D. 1991. The career of Khnumhotep III of Beni Hasan and the so-called
“decline of the nomarchs” . In S. Quirke (ed.), Middle Kingdom Studies ,51–67.
New M alden
Frankfort, H. and Pendlebury, J. D. S. 1933. The City of Akhenaten ,P a r tII:The North
Suburb and the Desert Altars. EEF Memoir 40. London
Fredricksmeyer, E. 2000. Alexander and the kingship of Asia. In Bosworth and
Baynham 2000: 136–66
Freed, R. E., Markowitz, Y. J. and D’Auria, S. H. 1999. Pharaohs of the Sun: Akhenaten,
Nefertiti, Tutankhamun. Boston
Friedl ¨ander, L. 1921. Darstellung aus der Sittengeschichte Roms in der Zeit von
Augustus bis zum Ausgang der Antonine ,9th–10th revision, vol. IV , ed. George
Wissowa with contributions by M. Bang, F. Drexel, U. Kahrstedt and O. Wein-
reich. Leipzig (English translation: Friedl ¨ander, 1908. Roman Life and Manners
under the Early Empire ,trans. of 7th edn by J. H. Freese, A. B. Gough and L. A.
Magn us, London)
Friedrich, C. and Brzezinski, Z. 1965. Totalitarian Dictatorship and Autocracy .
Cambridge, MA
FuLo-ch’eng 1977. Hsi Han de chi ko cheng-chih chi-t’uan. In HanT’ang shih lun
chi.Taipei
Gag´e, J. 1971. Lesclasses sociales dans l’empire romain ,2nd edn. Paris
Garnsey, P . and Humfress, C. 2001. The Evolution of the Late Antique World .
Cambridge
Garnsey, P . and Saller, R. 1987. The Roman Empire: Economy, Society, and Culture .
London
Garrison, M. B. and Root, M. C. Forthcoming. Seals of the Persepolis Fortiﬁcation
Table ts ,II.Images of Human Activity .Chicago
Gascou, J. 1981. Un ´enigme ´epigraphique: S ´ev`ere Alexandre et la titulature de Giuﬁ.
AntAfr 17: 231–40
Geertz, C. 1983. Local Knowledge .N e wY o r k
Gelzer, M. 1968. Caesar, Politician and Statesman ,trans. P . Needham. Oxford

Bibliography 337
Geyssen, J. 1998. Presentations of Victory on the Theodosian obelisk base. Byzantion
68: 47–55
Gibson, S., Delaine, J. and Claridge, A. 1994. The triclinium of the Domus Flavia: a
new reconstruction. Papers of the British School at Rome 62: 67–100
Giele, E. 2006. Imperial Decision-Making and Communication in Early China: A
Study of Cai Yong’s Duduan. Opera Sinologica 20. Wiesbaden
Gignoux, P . (ed.) 1991. Lesquatre inscriptions du mage Kirdir: textes et concordances.
Cahiers de Studia Iranica 9. Paris
2006. Prol ´egom `enes pour une histoire des id ´ees de l’Iran sassanide: convergences
et divergences. In Wieseh ¨ofer and Huyse 2006: 71–81
Giles, F. J. 1997. The Amarna Age: Western Asia. Warminster
Gnirs, A. M. 1996. Milit¨ar und Gesellschaft: ein Beitrag zur Sozialgeschichte des
Neuen R eiches.Studien zur Arch ¨aologie und Geschichte Alt ¨agyptens; Bd. 17.
Heidel berg
Gnoli, G. 1989. The Idea of Iran: An Essay of Its Origin. SOR 62. Rome
1993. Iran als religi¨ oser Begriff im Mazdaismus. Rheinisch-Westf ¨alische Akademie
der Wissenschaften, Vortr ¨age G 320. Opladen
1999. Farr(ah). In Encyclopaedia Iranica IX: 312–19
Goelet, O. 1985. TwoAspects of the Royal Palace in the Egyptian Old Kingdom .A n n
Arbor, MI
1986. The term stp-s3 in the Old Kingdom. Journal of the American Research Center
in Egypt 23: 85–97
Goffman, E. 1967. The nature of deference and demeanor. In Goffman (ed.), Inter-
action Ritual: Essays on Face-to-Face Behavior ,47–95. Garden City, NY
Gohary, J. 1992. Akhenaten’s Sed-festival at Karnak .L o n d o n
Goukowsky, P . 1978. Essai sur les origines du mythe d’Alexandre (336–270 av. J.-C.) ,
vol. I. Lesorigines politiques .Nancy
Grabar, A. 1936. L’empereur dans l’art byzantin .P a r is
1967. Christian Iconography: A Study of Its Origins .P rinceton
Greatrex, G. and Bardill, J. 1996. Antiochus the praepositus :aPersian eunuch at the
court of Theodosius II. DOP 50: 171–98
Gregor, W. 1997. Interaktion, Repr ¨asentation und Herrschaft. Der K ¨onigshof im
Hellenismus. In Winterling 1997b: 27–71
Grenet, F. 2001. La geste d’Ardashir ﬁls de Pˆ abag: K¯ arn¯amag i Ardaxˇ s¯er ¯ıP ¯abag¯ an.
Paris
Grifﬁn, M. 2003. De Bene ﬁciis and Roman society. JRS93: 92–113
Grzybek, E. 1999. Le meutre et son ch ˆatiment dans la Mac ´edoine antique. In Ancient
Macedonia. Sixth International Symposium (Ancient Macedonia 6), vol. I, 463–
9. Thessaloniki
G¨uterbock, K. 1906. Byzanz und Persien in ihren diplomatisch-v¨ olkerrechtlichen
Beziehungen im Zeitalter Justinians .B e r lin
Guidi, M. and Morony, M. 1991. Mazdak. In The Encyclopaedia of Islam .N e w edn,
vol. VI: 949–52

338 Bibliography
Guilland, R. 1955. Les titres auliques reserv ´eaux eunuques. REB 13: 65
1969. Etudes de topographie de Constantinople byzantine .Amsterdam
Gundlach, R. 1998. DerPharao und sein Staat: die Grundlegung der ¨Agyptischen
K¨onigsideologie im 4. und 3. Jahrtausend. Darmstadt
2004. Zu Strukturen und Aspekten Pharaonischer Residenzen. In Butz et al. 2004:
225–47
Gutsfeld, A. 2004. Der Pr ¨atorianerpr ¨afekt und der kaiserliche Hof. In Winterling
1998: 75–102
Gyselen, R. (ed.) 1999. La science des cieux: sages, mages, astrologues. ResOrientales
12. Bures-sur-Yvette
(ed.) 2002. Charmes et sortil` eges: magie et magicians. ResOrientales 14. Bures-
sur-Yvette
Hall, E. 1989. Invent ing the Barbarian .O x f ord
Hallock, R. T. 1969. Persepolis Fortiﬁcation Tablets .Chicago
1978. Selected Fortiﬁcation texts. CahDAFI 8: 109–36
Hammond, M. 1959. The Antonine Monarchy .R o m e
Hammond, N. G. L. 1983. Three Historians of Alexander the Great .Cambridge
1989. The Macedonian State: Origins, Institutions and History .O xford
1990. Royal pages, personal pages, and boys trained in the Macedonian manner
during the period of the T emenid monarchy. Historia 39: 261–90
1991. The various guards of Philip II and Alexander III. Historia 40: 396–418
Han-shu 1962. Ed. Peking: Chung-hua shu-ch ¨u
Hardy, G. 2000. Worlds of Bronze and Bamboo ,N e wY o r k
Harries, J. 1988. The Roman imperial quaestor from Constantine to Theodosius II.
JRS78: 148–72
Hartmann, U. 2002. Geist im Exil. R ¨omische Philosophen am Hof der Sasaniden. In
M. Schuol, U. Hartmann and A. Luther (eds.), Grenz¨ uberschreitungen: Formen
des Kontakts zwischen Orient und Okzident im Altertum. OrOcc 3. Stuttgart:
123–60
Hatzopoulos, M. 1988. Undonation du roi Lysimaque .Meletemata 5. Athens and
Paris
1994. Cultes et rites de passage en Mac´ edoine .Athens
1996. Macedonian Institutions under the Kings ,vol. I: AHistorical and Epigraphic
Stud yand vol. II: Epigraphic Appendix .Meletemata 22. Athens and Paris
Heather, P . 1998a. Senators and senates. In A. Cameron and P . Garnsey (eds.), The
Cambridge Ancient History ,vol. XIII, 184–210. Cambridge
1998b. Themistius: a political philosopher. In Whitby 1998: 125–49
Heisserer, A. J. 1980. Alexander the Great and the Greeks .Norman, OK
Helck, W. 1954. Untersuchungen zu den Beamtentiteln des ¨ agyptische Alten Reiches.
Gl¨uckstadt
1958. ZurVerwaltung des Mittleren und Neuen Reiches .P r o b l eme der ¨Agyptologie
3. Leiden
1986. Politische Gegens¨ atze im alten ¨Agypten .Hildesheimer ¨Agyptologische
Beitrage 23. Hildesheim

Bibliography 339
Heltzer, M. 1994. Neh.11,24 and the provincial representative at the Persian court.
Transeuphrat` ene8: 109–19
Herman, G. 1987. Ritualised Friendship and the Greek City .Cambridge
1997. The court society of the Hellenistic Age. In P . Cartledge, P . Garnsey and
E. Gruen (eds.), Hellenistic Constructs: Essays in Culture, History, and Histori-
ography .Berkeley, CA: 199–211
Hermann, A. 1938. Die ¨agyp tische K¨ onigsnovelle .Leipziger ¨Agyptologische Studien
10. Gl ¨uckstadt
Herrmann, G. and Curtis, J. 1998. Reﬂections on the four-winged genie: a pottery
jar and an ivory panel from Nimrud. IrAnt 33: 107–33
Hirsch, S. W. 1985. The Friendship of the Barbarians: Xenophon and the Persian
Empire .Hanover, NH
Hoffmeier, J. K. 1985. Sacred in the Vocabulary of Ancient Egypt: The Term DSR, with
Special Reference to Dynasties I–XX. Orbis Biblicus et Orientalis 59. G ¨ottingen
Holum, K. 1977. Pulcheria’s Crusade AD 421–2 and the ideology of imperial victory.
GRBS 18: 153–72
1982. Theodosian Empresses .B e r k e ley, CA
Hopkins, K. 1961. Social mobility in the later Roman empire: the evidence of Auso-
nius’ , CQ11: 239–49
1963 [1978]. Eunuchs in politics in the late Roman empire, PCPS 1963 ( =Hop-
kins, Conquerors and Slaves ,Cambridge 1978: 172–96 ( =ch.IV,with revisions))
1978a. Conquerors and Slaves .Cambridge
1978b. The political power of eunuchs. In Hopkins 1978a: 172–961980. Taxes and trade in the Roman Empire, 200 BC–AD 400. JRS70: 101–25
1995/6. Rome, taxes, rents and trade. Kodai: Journal of Ancient History 6–7: 41–75
(reprinted in W. Schiedel and S. von Reden (eds.) 2002. The Ancient Economy .
Edinburgh)
Hou H an shu 1965. Ed. Chung-hua shu-ch ¨u. Peking
Hours, B. 2002. Louis XV et sa cour: le roi, l’´ etiquette et le courtesan .P a r is
Howard-Johnston, J. D. 1995. The two great powers in late antiquity: a comparison.
InA. Cameron (ed.). The Byzantine and Early Islamic Near East ,III:States,
Resources and Armies. Studies in Late Antiquity and Early Islam 1. Princeton:
157–226
Humbach, H. and Skjærvø, P . O. 1978–83. The Sassanian Inscription of Paikuli ,
3pts. Wiesbaden
Hunt, T. D. 1998. The Church as a public institution. In A. Cameron and P . Garnsey
(eds.), The Cambridge Ancient History ,vol. XIII, 238–76. Cambridge
Huyse, P . 1999. Die dre isprachige Inschrift ˇS¯abuhrs I. an der Ka‘ba-i Zarduˇ st(ˇSKZ)
(CII, pt. III, vol. 1, texts I), 2 vols. 1–2. London
2002. La revendication de territoires ach ´em´enides par les Sassanides: une r ´ealit ´e
historique? In Huyse (ed.). Iran: questions et connaissances. Actes du IVe Congr` es
europ´ een des ´ etudes iraniennes organis´ ep a r la Societas Iranologica Europaea ,I :
La p´eriode ancienne .Studia Iranica, cahier 25. Paris: 297–311
(in press). Histoire orale et ´ ecrite en Iran ancien entre m´ emoire et oubli

340 Bibliography
Ikram, S. 1989. Domestic shrines and the cult of the royal family at el-’Amarna.
Journal of Egyptian Archaeology 75: 89–101
Inwood, B. 1995. Politics and paradox in Seneca’s De Bene ﬁciis .InA.Laks and
M. Schoﬁeld (eds.), Justice and Generosity ,241–65. Cambridge
Jacobs, B. 1994. DieSatrapienverwaltung im Perserreich zur Zeit Darius’ III .
Wiesbaden
Joann`es, F. 2004. The Age of Empires: Mesopotamia in the First Millennium BC ,trans.
Antonia Nevill. Edinburgh
Johnson, D. 1977. The Medieval Chinese Oligarchy .Boulder, CO
Jones, A. H. M. 1964. The Later Roman Empire 284–602: A Social, Economic and
Administrative Survey .O x f ord
1966. The Decline of the Ancient World .L o n d o n
Jones, B. W. 1992. The Emperor Domitian .L o n d o n
Jones, C. 2002. The Great Nation: France from Louis XV to Napoleon .Harmondsworth
Jugel, U. 1976. Politische Funktion und soziale Stellung der Eunuchen zur sp¨ ateren
Hanzeit (25–220 n.Chr.) .Wiesbaden
Kanawati, N. 1977. The Egyptian Administration in the Old Kingdom: Evidence on Its
Economic Decline. Warminster
1980. Governmental Reforms in Old Kingdom Egypt. Warminster
2003. Conspiracies in the Egyptian Palace: Unis to Pepy I .L o n d o n
Kantorowicz, E. 1963. Oriens Augusti: lever du roi. DOP 17: 119–77
Kaptan, D. 1996. The Great King’s audience. In F. Blakolmer et al. (eds.), Fremde
Zeiten ,2vols., 259–71. Vienna
Kay, N. M. 1985. Martial Book XI, A Commentary. London
Kees, H. 1914. Pr-dw3tand Db3t .Recueil de Travaux 36: 1–15
Kelly,C. 1994. Later Roman bureaucracy: going through the ﬁles. In A. Bow-
man and G. Woolf (eds.), Liter acy and Power in the Ancient World ,161–76.
Cambridge
1998. Emperors, government and bureaucracy. In A. Cameron and P . Garnsey
(eds.), The Cambridge Ancient History ,vol. XIII, 138–83. Cambridge
1999. Empire building. In Bowersock, Brown and Grabar 1999: 170–95
Kemp, B. J. 1976. The Window of Appearance at El-Amarna and the structure of
this city. Journal of Egyptian Archaeology 62: 81–99
1977. A building of Amenophis III at K ˆom El-‘Abd. Journal of Egyptian Archaeology
63: 77–8
1978. The Har ˆım-Palace at Medinet el-Ghurab, Zeitschrift f¨ ur¨Agyptische Sprache
und Altertumskunde 105: 122–33
1989. AncientEgypt: Anatomy of a Civilization. London
1995. Outlying temples at el-Amarna. In Kemp (ed.), Amarna Reports VI .EES
Occasional Publications 10. London: 411–62
Kettenhofen, E. 2002. Die Einforderung der achaimenidischen T erritorien durch die
S¯as¯aniden – eine Bilanz. In S. Kurz (ed.). Festschrift I. Khalifeh-Soltani zum 65.
Geburtstag .Aachen: 49–75

Bibliography 341
Kettering, S. 1986. Patrons, Brokers, and Clients in Seventeenth Century France .
Oxford
Kienast, D. 1973. Philip II von Makedonien und das Reich der Achaimeniden. Munich
Kiilerich, B. 1993. Late Fourth Century Classicism in the Plastic Arts .O d e nse
Kirkpatrick, J. 1982. Dictatorship and Double Standards: Rationalism and Realism in
Politics .N e wY o r k
Kitchen, K. A. 1982. Pharaoh Triumphant: The Life and Times of Ramesses II .Cairo
Klodt, C. 2001. Besche idene Gr¨ osse. Die Herrschergestalt, der Kaiserpalast und die
Stadt Rom: Literarische Reﬂexionen monarchischer Selbstdarstellung .G¨ottingen
Kloft, H. 1970. Liberalitas Principis .Boehlau
Kokkinos, N. 1992. Antonia Augusta: Portrait of a Great Roman Lady .L o n d o n
Kolb, F. 2001. Herrscheridiologie in der Sp¨ atantike .B e r lin
Konstan, D. 1997a. Friendship and monarchy: Dio of Prusa’s 3rd Oration on King-
ship. SO72: 124–43
1997b. Friendship in the Classical World .Cambridge
1983. Three Imperial Capitals: Topography and Politics .L o n d o n
Krautheimer, R. 1986. Early Christian and Byzantine Architecture ,4thedn.
Harmondsworth
Krentz, P . and Wheeler, E. L. 1994. Polyaenus: Stratagems of War ,vol. I. Chicago
Kroll, W. 1931. Kuss. RESuppl. 5: 512–20
K¨uhn, W. 1987. Der Kuss des Kaisers: Plinius paneg.24.2. W¨urzburger Jahrb¨ ucher
fur die Altertumswissenschaft 13: 263–87
Kuhrt, A. 1995. The Ancient Near East 3000–300 BC ,2vols. London
2001. The palace(s) of Babylon. In I. Nielsen (ed.), The Royal Palace Institutions
in the First Millennium BC ,77–93. Aarhus
Kuran, T. 1995. Private Truths, Public Lies: The Social Consequences of Preference
Falsiﬁcation. Cambridge, MA
L¨A: W. Helck, E. Otto and W. Westendorf (eds.) 1975–92. Lexikon der ¨Agyptologie ,
vols. I–VII. Wiesbaden
Lacovara, P . 1997. The New Kingdom Royal City. London
Lane Fox, R. J. 1997. The Itinerary of Alexander: Constantius to Julian. CQ 47:
239–52
Lanfranchi, G. B., Roaf, M. and Rollinger, R. (eds.) 2003. Continuity of Empire?
Assyria, Media, Persia .P a d u a
Larsen, M. T. (ed.) (in press). Palace, King and Empire .Copenhagen
Laurence, R. and Paterson, J. J. 1999. Power and laughter: imperial dicta. Papers of
the British School at Rome 67: 183–97
Lavagne, H. 1988. Operosa antra: recherches sur la grotte ` aRome de Sylla ` aHadrien .
BEFAR 212. Rome
Lee, A. D. 1993. Information & Frontiers: Roman Foreign Relations in Late Antiquity .
Cambridge
Legge, J. 1861–72. The Chinese Classics ,5vols. Hong Kong and London
1885. Li-chi, Sacred Books of the East ,vols. XXVII and XXVIII. Oxford

342 Bibliography
Lendon, J. E. 1997. Empire of Honour: The Art of Government in the Roman World .
Oxford
Leprohon, R. J. 1995. Royal ideology and state administration in Pharaonic Egypt.
InJ.M. Sasson (ed.), Civilizations of the Ancient Near East ,vol. I, 273–87. New
Yo r k
L´evy, E. 2001. Review of Hatzopoulos 1996. Gnomon 73: 267–70
Lewis, D. M. 1977. Spar ta and Persia .Cincinnati
1987. The king’s dinner (Polyaenus IV, 3, 32). AchH ist2: 79–87
Lichtheim, M. 1973. Ancient Egypt ian Literature ,vol. I: The Old and Middle King-
doms. Berkeley
1976. Ancient Egypt ian Literature ,vol. II: The New Kingdom. Berkeley
1996. Didactic literature. In A. Loprieno (ed.), Ancient Egypt ian Literature: History
and Forms ,243–62. Leiden
Liebeschuetz, J. H. W. 1972. Antioch: City and Imperial Administration in the Late
Roman Empire .O xford
2001. The uses and abuses of the concept of ‘decline’ . In L. Lavan (ed.), Recent
Research in Late-Antique Urbanism .JRA Suppl. Ser. 42, 233–8. Portsmouth, RI
Liverani, M. 1990. Prestige and Interest: International Relations in the Near East
ca. 1600–1100 BC. Padua
Loewe, M. 1974. Crisis and Conﬂict in Han China .L o n d o n
1986a. The concept of sovereignty. In Twitchett and Loewe 1986: 726–461986b. The structure and practice of government. In Twitchett and Loewe 1986:
463–90
1992. The imperial tombs of the Former Han dynasty and their shrines. T’oung
Pao78: 302–40
Loprieno, A. 1996. The ‘King’s Novel’ . In Loprieno (ed.), Ancient Egy ptian Literature:
History and Forms ,277–95. Leiden
Lorton, D. 1991. What was the pr-nsw and who managed it? Aspects of royal adminis-
tration in ‘The Duties of the Vizier’ . Studien zur ¨Agyptischen Kultur 18: 291–316
MacCormack, S. 1981. Artand Ceremony in Late Antiquity .B e r k e ley, CA
McCormick, R. 1985. Analysing imperial ceremonies. Jahrbuch der ¨Osterreichischen
Byzantinistik 35: 1–19
2000. Emperor and court. In A. Cameron, B. Ward-Perkins and M. Whitby (eds.)
The Cambridge Ancient History ,vol. XIV, 135–61. Cambridge
MacDonald, W. L. 1982. The Architecture of the Roman Empire ,vol. I, rev. edn. New
Haven
McManners, J. 1998. Church and Society in Eighteenth Century France ,2vols. Oxford
MacMullen, R. 1962. Roman bureaucratese. Traditio 18: 364–78 ( =MacM ullen
1990: ch. 8)
1964a. Some pictures in Ammianus Marcellinus. ArtBulletin 46: 435–56 ( =Mac-
Mullen 1990: ch. 9)
1964b. Social mobility and the Theodosian Code. JRS54: 49–53

Bibliography 343
1988. Corruption and the Decline of Rome .N e wH a ven
1990. Changes in the Roman Empire: Essays in the Ordinary .P rinceton
Macuch, M. 1992. Charitable foundations, I. In the Sasanian period. In Encyclopaedia
Iranica V:380–2
1994. Die sasanidische Stiftung ‘f ¨ur die Seele’ – Vorbild f ¨ur den islamischen waqf ?
In P . Vavrou ˇsek (ed.), Iranian and Indo-European Studies: Memorial Volume of
Otakar Klima ,Prague: 163–80
Maguire, H. 1999. The good life. In Bowersock, Brown and Grabar 1999: 238–57
Mango, C. 1981. Daily life in Byzantium. J¨OB31: 337–53
Manley, B. 1996. The Penguin Historical Atlas of Ancient Egypt .Harmondsworth
Mansel, P . 2006. The globalisation of court history. The Court Historian 11, 1: 77–101
Marmon, S. 1990. Eunuchs and Sacred Boundaries in Islamic Society .O x f ord
Mathew, G. 1963. Byzantine Aesthetics .L o n d o n
Matthews, J. F. 1975. Western Aristocracies and Imperial Court AD 364–425 .O x f ord
1989. The Roman Empire of Ammianus .L o n d o n
Mathews, T. F. 1993. The Clash of Gods .P rinceton
Medrei, M. 1996. Suet. Nero 31.1: Elementi e proposte per la riconstruzione del
progetto della Domus Aurea. In C. Panella, Meta Sudans ,vol. I, 165–88. Rome
Mee, C. and Spawforth, A. 2001. Greece: An Oxford Archaeological Guide .O xford
Meier, S. A. 2000. Diplomacy and international marriages. In Cohen and Westbrook
2000: 165–73
Millar, F. 1965. Epictetus and the imperial court. JRS55: 141–8 (reprinted in H. M.
Cotton and G. M. Rogers (eds.), Rome, The Greek World, and the East ,vol. II,
105–19. Chapel Hill and London, 2004)
1967. The Roman Empire and Its Neighbours .L o n d o n
1977. The Emperor in the Roman World (31 BC–AD 337). London
Millar, F. and others. 1981. The Roman World and Its Neighbours ,2nd edn. New Y ork
Miller, M. C. 1997. Athens and Persia in the Fifth Century BC: A Study in Cultural
Recept ivity.Cambridge
Moles, J. L. 1985. The interpretation of the ‘Second Preface’ in Arrian’s Anabasis.
Journal of Hellenic Studies 105: 162–8
Mommsen, T. 1887. R¨omisches Staatsrecht .Leipzig
Monteﬁore, S. S. 2003. Stalin: The Court of the Red Tsar. London
Montserrat, D. 2000. Akhenaten: History, Fantasy and Egyptology. London
Mooren, L. 1975. The Aulic Titulature in Ptolemaic Egypt: Introduction and Proso-
pography .Brussels
1977. La hi´ erarchie de cour Ptol´ emaique: contribution ` al ’ ´etude des institutions et
des classes dirigeantes a l’´ epoque hell´ enistique .Studia Hellenistica 23. Louvain
Moran, W. L. 1992. The Amarna Letters. Baltimore
Morony,M.1989. Bah ¯ar-e Kesr ¯a. In Encyclopaedia Iranica III: 479
Morris, I. 1964. The World of the Shining Prince: Court Life in Ancient Japan .O xford
Muir, E. 1997. Ritual in Early Modern Europe .Cambridge

344 Bibliography
M¨uller, C. 1979. Untersuchungen zum Erdaltar she im China der Chou- und Han-Zeit .
Munich
Murnane, W. J. 1977. AncientEgyptian Coregencies. Studies in Ancient Oriental
Civilization 40. Chicago
1995. Texts fro mthe Amarna Period in Egypt .Society of Biblical Literature: Writ-
ings from the Ancient World 5. Atlanta, GA
1998. The organization of government under Amenhotep III. In D. O’Connor
and E. H. Cline (eds.), Amenhotep III: Perspectives on His Reign ,173–221. Ann
Arbor, MI
Murnane, W. J. and van Siclen, C. C. 1993. The Boundary Stelae of Akhenaten. London
Murray,O.1996. Hellenistic royal symposia. In P . Bilde et al. (eds.), Aspects of
Hellenistic Kingship ,15–27. Aarhus
(ed.) 1990. Symp otica .O x f ord
Naville, E. 1898. The Temple of Deir el Bahari. Part III:End o fN o r thern Half and
Southern Half of the Middle Platform .L o n d o n
Nielsen, I. 1994. Hellenistic Palaces: Tradition and Renewal .Studies in Hellenistic
Civilization 5. Aarhus
(ed.) 2001. The Royal Palace Institutions in the First Millennium BC: Regional
Development and Cultural Exchange between East and West ,Athens
Nienhauser, W. 1994–2002. The Grand Scribe’s Records ,vols. I, II and VII. Bloom-
ington, IN
Nixon, C. E. V . and Rogers, B. S. 1994. InPraise of the Later Roman Emperors: The
Panegyri Latini. Berkeley, CA
Nock, A. D. 1947. The emperor’s divine comes, JRS37: 102–17 ( repr. in Nock 1972:
653–75)
1972. Essays on Religion and the Ancient World ,e d . Z. Stewart. Oxford
Noethlichs, K. 1991. Hofbeamter. RAC 15: 1111–58
1998. Strukturen und Funktionen des sp ¨atantikes Kaiserhofes. In Winterling 1988:
13–50
North, D. C. 1981. Structure and Change in Economic History .N e wY o r k
O’Connor, D. 1989. City and palace in New Kingdom Egypt. Cahiers de Recherches de
l’Institut de Papyrologie et Egyptologie de l’Universit´ edeLille: Etudes sur l’Egypte
etle Soudan anciens 11: 73–87
1995. Beloved of Maat, the Horizon of Re: the royal palace in New Kingdom Egypt.
InD.O’Connor and D. Silverman (eds.), AncientEgyptian Kingship ,263–300.
Proble me der ¨Agyptologie 9. Leiden
Olson, M. 1993. Dictatorship, democracy and development. American Political
Science Review 87: 567–75
Palagia, O. 2000. Hephaestion’s pyre and the Royal Hunt of Alexander. In Bosworth
and Baynham 2000: 167–206.
Panaino, A. 1999. La novella degli scacchi e della tavola reale. Un’ antica fonte orien-
tale sui due giochi da tavola pi` udiffusi nel mondo eurasiatico tra Tardoantico e
Medioevo e sulla loro simbologia militare e astrologica. Testo pahlavi, traduzione

Bibliography 345
ecommento al Wiz¯ ariˇsn ¯ıˇcatrang ud nihiˇ sn ¯ın ¯ew-ardaxˇ s¯ır ‘La spiegazione degli
scacchi e la disposizione della tavola reale’ .Milan
2001. Greci e Iranici: confronto e conﬂitti. In Settis, S. (ed.), IG r e c i ,vol. III:
IG r e c io l t r el aG r e c i a .Turin: 79–136
2003. The ba γ¯an of the Fratarakas: gods or ‘divine’ kings? In C. G. Cereti,
M. Maggi. and E. Provasi (eds.), Religious Themes and Texts of Pre-Islamic Iran
and Central Asia: Studies in Honour of Prof. Gherardo Gnoli on the Occasion ofHis65th Birthday on 6th December 2002. Beitr ¨age zur Iranistik 24. Wiesbaden:
265–88
Pandermalis, D. 1997. Dion.Athens
1999. Dion. Hˆ eAnaskaphˆ e.Athens
Pani, M. 2003. La corte dei Cesari: fra Augusto e Nerone .R o m e
Parkinson, R. B. 1991. Voices from Ancient Egypt: An Anthology of Middle Kingdom
Writ ings. London
Paschoud, F. and Szidat, J. (eds.) 1997. Usurpationen in der Sp¨ atantike. Historia
Einzelschriften 111. Stuttgart
Paterson, J. 2004. Autocracy and political economy. Mediterraneo Antico 7.2: 571–89
Patlagean, E. 1992. De la chasse et du souverain, DOP 46: 257–63
Pearson, L. 1960. The Lost Historians of Alexander the Great .American Philological
Monogr aphs 20. New Y ork
Peck, E. H. 1992. Clothing, IV . In Encyclopaedia Iranica V:739–52
Peet, T. E and Woolley, C. L. 1923. The City of Akhenaten ,P a r tI :Excavations of 1921
and 1922 at el-’Amarneh .EES Memoir 38. London
Pendlebury, J. D. S. 1951. The City of Akhenaten ,Part III: The Central City and the
Ofﬁcial Quarters. EES Memoir 44. London
Perron, Y. 1990. D’Alexandre `aN´eron: le motif de la tente d’apparat. La salle 29 de
la Domus Aurea. In J. M. Croisille (ed.), Neronia IV: Alejandro Magno, modelo
de los emperadores romanos ,211–29. Collection Latomus 209. Brussels
Pharr, C. (trans.) 1952. The Theodosian Code .N e wY o r k
Posener, G. 1960. Dela divinit´ ed uP haraon .Cahiers de la Soci ´et´eAsiatique 15. Paris
Postgate, N., Wang, T. and Wilkinson, T. 1995. The evidence for early writing:
utilitarian or ceremonial? Antiquity 69: 459–80
Potter, D. S. 2004. The Roman Empire at Bay AD 180–395 .L o n d o n
Potts, D. T. 1999. The Archaeology of Elam. Cambridge
Price, M. J. 1991. The Coinage in the Name of Alexander the Great and Philip Arrhi-
daeus ,2vols. Zurich and London
Price, S. 1984. Rituals and Power: The Roman Imperial Cult in Asia Minor .Cambridge
Quirke, S. 1990. The Administration of Egypt in the Late Middle Kingdom: The Hieratic
Documents .N e wM alden
1999. Visible and invisible: the King in the administrative papyri of the late Middle
Kingdom. In R. Gundlach and W. Seipel (eds.), Das fr¨ uhe ¨agyp tische K¨ onigtum:
Akten des 2. Symposiums zur ¨ agyptischen K¨ onigsideologie in Wien ,63–71.
Wiesbaden

346 Bibliography
Raedler, C. 2004. Die Wesire Ramses’ II.– Netzwerke der Macht. In R. Gundlach and
A. Klug (eds.), Das ¨agyp tische K¨ onigtum im Spannungsfeld zwischen Innen- und
Außenpolitik im 2. Jahrtausend v. Chr .Wiesbaden
Forthcoming. Rank and favour at the early Ramesside court. In R. Gundlach and
J.Taylor (eds.), Egyptian Royal Residences: Structure and Function
Rance, P . 2000. Simulacra pugnae: the literary and historical tradition of mock
battles in the Roman and early Byzantine army, GRBS 41: 223–75
Rawski, E. S. 1998. The Last Emperors: A Social History of Qing Imperial Institutions .
Berkeley, CA
Rawson, E. 1975. Caesar’s heritage: Hellenistic kings and their Roman equals. JRS
65: 148–59
1985. Intellectual Life in the Late Roman Republic .L o n d o n
Redford, D. 1984. Akhenaten: The Heretic King .P rinceton
1986. Pharaonic King-Lists, Annals and Day-Books: A Contribution to the Egyptian
Sense of History. Mississauga
1995. The concept of kingship during the Eighteenth Dynasty. In D. O’Connor
and D. Silverman (eds.), AncientEgyptian Kingship ,157–84. Probleme der
¨Agyptologie 9. Leiden
Reeves, N. 1990. The Complete Tutankhamun .L o n d o n
Richter-Bernburg, L. 1998. On the diffusion of medical knowledge in Persian court
culture during the fourth and ﬁfth centuries A.H. In Z. Vezel et al. (eds.), La
science dans le monde iranien ` al ’ ´epoque islamique .Tehran: 219–33
1999. Iran’s contribution to medicine and veterinary science in Islam AH 100–
900/AD 700–1500. In J. A. C. Greppin et al. (eds.), The Diffusion of Greco-Roman
Medicine into the Middle East and the Caucasus .Delmar: 139–67
2000. Medicine, pharmacology and veterinary science in Islamic eastern Iran and
Central Asia. In C. E. Bosworth and M. S. Asimov (eds.), History of Civilizations
of Cent ral A sia,vol. IV: The Age of Achievement: AD 750 to the End of the Fifteenth
Cent ury,p t2 , The Achievements .P a r is
Ringrose, K. 2003. The Perfect Servant: Eunuchs and the Social Construction of Gender
in Byzantium. Chicago
Rist, J. 1967. Plotinus: The Road to Reality .Cambridge
Roddaz, J.-M. 1984. Marcus Agrippa .R o m e
Rogers, R. S. 1947. Roman emperors as heirs and legatees. Transactions of the Amer-
ican Philological Association 78: 140–58
Roller, M. B. 2001. Constructing Autocracy: Aristocrats and Emperors in Julio-
Claudian Rome .P rinceton
Rollinger, R. 2003. The western expansion of the ‘Median’ empire: a re-examination.
InLanfranchi, Roaf and Rollinger 2003: 289–319
Root, M. C. 1979. The King and Kingship in Achaemenid Art .Acta Iranica 3. Leiden
Roth, A. M. 1991. Egyptian Phyles in the Old Kingdom. The Evolution of a Sys-
tem of Social Organization .Studies in Ancient Oriental Civilization, 48.
Chicago

Bibliography 347
1993. Social change in the Fourth Dynasty: the spatial organisation of pyramids,
tombs, and cemeteries, Journal of the American Research Center in Egypt 30:
33–55
1995. ACemetery of Palace Attendants, including G 2084–2099 G 2230 +2231 and
G2240. Boston
2002. The meaning of menial labour: ‘servant statues’ in Old Kingdom Serdabs,
Journal of the American Research Center in Egypt 39: 103–21
Rouech ´e, C. 1984. Acclamations in the late Roman empire: new evidence from
Aphrodisias. JRS74: 181–99
Royo,M.1999. Domus Imperatoriae: topographie, formation, et imaginaire des palais
imp´eriaux du Palatin .R o m e
Rubin, Z. 1995. The reforms of Khusro Anushirwan. In A. Cameron (ed.), The
Byzantine and Early Islamic Near East ,vol. III: States, Resources and Armies .
Studies in Late Antiquity and Early Islam 1. Princeton: 227–97
1998. The Roman Empire in the ResGestae Divi Saporis –the Mediterranean world
in S¯as¯anian propaganda. In E. Dabrowa (ed.), Ancient Iran and the Mediter-
ranean World .E l e c t r u m2 .K r a k ´ow: 177–85
2000. The Sasanid monarchy. In A. Cameron, B. Ward-Perkins and M. Whitby
(eds.), The Cambridge Ancient History ,2nd edn, vol. XIV . Cambridge: 638–61
2002. ResGestae Divi Saporis :Greek and Middle Iranian in a document of Sasan-
ian anti-Roman propaganda. In J. N. Adams, M. Janse and S. Swain (eds.),Bilingualism in Ancient Society: Language Contact and the Written Text .O xford:
267–97
Sako, L. 1986. Le rˆole de la hi´ erarchie syriaque orientale dans les rapports diplomatiques
entrelaPerse et Byzance aux Ve–VIIe si` ecles.P a r is
Saller, R. 1980. Anecdotes as evidence for Roman imperial history. Greece and Rome
ser. 2, 27: 69–83
1982. Personal Patronage under the Early Empire .Cambridge
1989. Patronage and friendship in early Imperial Rome: drawing the distinction.
InA. Wallace-Hadrill (ed.), Patronage in Ancient Society ,49–62. London
Salzman, M. 1990. On Roman Time: The Codex Calendar of 354 .B e r k e ley
Sancisi-Weerdenburg, H. 1980. Y auna en Persai: Grieken in een ander perspectief,
PhD thesis, Leiden University
1985. The death of Cyrus: Xenophon’s Cyropaedia as a source for Iranian history.
InPapers in Honour of Mary Boyce ,459–71. Acta Iranica 25. Leiden
1987. Decadence in the empire or decadence in the sources? From source to
synthesis: Ctesias. AchH ist1: 33–45
1988. Was there ever a Median empire? AchH ist3: 197–212
1989. Gifts in the Persian empire. In P . Briant and C. Herrenschmidt (eds.), Les
tributs dans l’empire perse. Actes de la Table Ronde de Paris 12–13 d´ ecembre 1986 ,
129–146. Paris
1993. Persian food: stereotypes and political identity. In J. Wilkins, D. Harvey and
M. Dobson (eds.), Food in Antiquity ,286–302. Exeter

348 Bibliography
San-fu huang-t’u chiao-cheng 1980. Hsi-an
Savalli-Lestrade, I. 1998. Lesphiloi royales dans l’Asie hell´ enistique .H a u t e sE t u d e s
du Monde Gr ´eco-Romain 25. Geneva
Scharf, R. 1994. Comites und comitiva primi ordinis .M a y e n c e and Stuttgart
Scheer, T. S. 2000. DieGottheit und ihr Bild .Zetemata 105. Munich
Schlinkert, D. 1994. Der Hofeunuch in der Sp ¨atantike: ein gefahrlicher Aussenseiter?,
Hermes122: 342–59
1996. Vom Haus zum Hof: Aspekte h ¨oﬁscher Herrschaft in der Sp ¨atantike, Klio
78: 454–82
1998. Dem Kaiser folgen. Kaiser, Senatsadel und h ¨oﬁsche Funktionselite ( comites
consistoriani )von der ‘T etrarchie’ Diokletians bis zum Ende der konstantini-
schen Dynastie. In Winterling 1998: 133–60
2002. Between emperor, court and senatorial order: the codiﬁcation of the Codex
Theodosianus, Ancient Society 32: 282–94
Schmidt, E. 1953. Persepolis ,vol. I: Structures, Reliefs, Inscriptions .Chicago
1957. Persepolis ,vol. II: Contents of the Treasury and Other Discoveries .Chicago
1970. Persepolis ,vol. III: The Royal Tombs and Other Monuments .Chicago
Schmitt, E. 1953, 1957. Persepolis ,vols. I–II. OIP 68.69. Chicago
Schneider, R. M. 2006. Orientalism in late antiquity: the oriental other in imperial
and Christian imagery. In Wieseh ¨ofer and Huyse 2006: 241–78
Scholten, H. 1995. Der Eun uch in Kaisern¨ ahe: Zur politischen und sozialen Bedeu-
tung des praepositus sacri cubiculi im 4. und 5. Jahrhundert n.Chr. Prismata 5.
Frankfurt
1998. Der oberste Hofeunuch. Die politische Efﬁcienz eines gesellschaftlich
Diskriminierten. In Winterling 1998: 51–74
Schulman, A. R. 1978. ‘Ankhsenamun, Nofretity, and the Amka affair. Journal of the
American Research Center in Egypt 15: 43–8
1979. Diplomatic marriage in the Egyptian New Kingdom. Journal of Near Eastern
Studies 38: 177–93
Scott, J. 1990. Domination and the Arts of Resistance: Hidden Transcripts.
Princeton
Seeck, O. (ed.) 1962. NotitiaDignitatum .Frankfurt (repr. of original 1876 edn)
Shahbazi, A. S. 2001. Early Sasanians’ claim to Achaemenid heritage. Nˆame-ye Irˆ an-e
Bˆastˆan1.1: 61–73
Shaw, I. 1992. Ideal homes in Ancient Egypt: the archaeology of social aspiration.
Cambridge Archaeological Journal 2.2: 147–66
(ed.) 2000. The Oxford History of Ancient Egypt. Oxford
Shepherd, M. S. 1967. Liturgical expressions of the Constantinian triumph. DOP
21: 57–78
Sherwin-White, S. and Kuhrt, A. 1993. From Samarkand to Sardis: A New Approach
to the Seleucid Empire .L o n d o n
Shih-chi 1959. Ed. Peking, chapter 16/759
Shih-san ching chu-shu (The Thirteen Classics) .1980. Peking

Bibliography 349
Silverman, D. 1995. The nature of Egyptian kingship. In D. O’Connor and
D.Silverman (eds.), AncientEgyptian Kingship ,49–92. Probleme der
¨Agyptologie 9. Leiden
Sivan, H. 1996. Was Theodosiu sIau s u r p e r ? Klio78: 198–211
Smith, H. S. 1992. The making of Egypt: a review of the inﬂuence of Susa and
Sumer on Upper Egypt and Lower Nubia in the 4th millennium B.C. In
R. Friedman and B. Adams (eds.), The Followers of Horus: Studies Dedicated to
Michael Allen Hoffman ,235–46. Egyptian Studies Association Publication 2.
Oxbow Monographs 20. Oxford
Smith, R. B. E. 1995. Julian’s Gods: Religion and Philosophy in the Thought and Action
ofJulian the Apostate. London and New Y ork
1999. T elling tales: Ammianus’ account of the Persian expedition of Julian. In J. W.
Drijvers and E. D. Hunt (eds.), The Roman World and Its Historian: Interpreting
Ammianus ,89–104. London and New Y ork
Smith, R. R. R. 1985. Roman portraits: honours, empresses and late emperors. JRS
75: 209–21
1997. The public image of Licinius I: portrait sculpture and imperial ideology in
the early 4th century. JRS87: 170–302
2002. The statue monument of Oecumenius: a new portrait of a late antique
governor from Aphrodisias. JRS92: 134–56
Spalinger, A. 1982. Aspects of the Military Documents of the Ancient Egyptians .Yale
Near Eastern Researches 9. New Haven
Spawforth, A. 2006a. The Complete Greek Temples .L o n d o n
2006b. ‘Macedonian times’: Hellenistic memories in the provinces of the Roman
near east. In D. Konstan and S. Said (eds.), Greeks on Greekness: Viewing the
Greek Past under the Roman Empire .Proceedings of the Cambridge Philological
Society supplementary vol. 29, 1–26. Cambridge
Spence, K. E. 1998. Orientation in Ancient Egyptian royal architecture. Unpublished
PhD dissertation, University of Cambridge
1999. The North Palace at Amarna. EA15: 14–17
2004. The three-dimensional form of the Amarna house, JEA 90: 123–52
Forthcoming. The palaces of el-Amarna: an architectural analysis. In R. Gundlach
and J. Taylor (eds.), Egyptian Royal Residences: Structure and Function
Spencer, D. 2002. The Roman Alexander: Reading a Cultural Myth .E x e t e r
Spycket, A. 1980. Women in Persian art. In D. Schmandt-Besserat (ed.), Ancient
Persia ,43–5 (with pls. 21–6). Malibu
St Clair, A. 1996. Imperial virtue: four late antique statuettes. DOP 50: 147–62
Starr, C. G. Jr. 1949. Epictetus and the Tyrant. Classical Philology 45: 20–9
Stausberg, M. 2002. DieReligion Zarathushtras ,B dI .S t u ttgart
Ste Croix, G. de 1954. Suffragium: from vote to patronage, British Journal of Sociology
5: 33–48
Steinby, E. M. 1993–2000. Lexicon Topographiae Urbis Romae ,vols. I–VI. Rome
Stern, H. 1954. Remarks on the adoration under Diocletian, JWCI 17: 184–9

350 Bibliography
Stevenson Smith, W. 1981. The Art and Architecture of Ancient Egypt, re v. W. K.
Simpson. Harmondsworth
Straub, J. A. 1967. Constantine as Koinos Episkopos: tradition and innovation in
the representation of the ﬁrst Christian emperor’s majesty, DOP 21: 57–78
Strommenger, E. 1962. F¨unf Jahrtausende Mesopotamien: Die Kunst von den
Anf¨angen um 5000 v. Chr. bis zu Alexander dem Grossen .Munich (Eng. trans.
London 1964)
Stronach, D. 1987. Archaeology ii. Median and Achaemenid. EncIr 2: 288–96
1997. Anshan and Parsa: early Achaemenid history, art and architecture on the
Iranian plateau. In J. Curtis (ed.), Mesopotamia and Iran in the Persian Period:
Conquest and Imperialism 539–331 BC ,35–53. London
Strong, R. 2002. Feast: A History of Grand Eating .L o n d o n
Strudwick, N. 1985. The Administration of Egypt in the Old Kingdom. London
Summers, G. D. 2000. The Median empire reconsidered. A view from Kerkenes Da ˘g.
AnSt 50: 55–73
Sundermann, W. 1977. Mazdak und die mazdakitischen Volksaufst ¨ande. Altertum
23: 245–9
1981. Mitteliranische manich¨ aische Texte kirchengeschichtlichen Inhalts. Schriften
zur Geschichte und Kultur des Alten Orients, Berliner Turfantexte 11. Berlin
1983. Review of Humbach/Skjærvø 1978–1983. Kratylos28: 82–9
Swift, L. and Oliver, J. H. 1962. Constantius II on Flavius Philippus, AJP83: 247–64
Syme, R. 1939. The Roman Revolution .O xford
1986. The Augustan Aristocracy .O x f ord
Talbert, R. J. A. 1984. The Senate of Imperial Rome .P rinceton
Tamm, B. 1963. Auditorium und Palatium .Stockholm
Tanabe, K. 1983. Iconography of the Royal-Hunt Bas-Reliefs at Taq-i Bustan. Orient
(T okyo) 19: 103–16
Tarn,W.1948. Alexander the Great ,2vols. Oxford
Therasse, J. 1976. Quintus Curtius Rufus .I n d e xV e rbor um. Relev´ es lexicaux et gram-
maticaux .Hildesheim and New Y ork
Thompson, D. L. 1981. The meetings of the Roman senate on the Palatine. AJA 85:
335–9
Tietze, C. 1985. Amarna: Analyse der Wohnh ¨auser und soziale Struktur der Stadt-
bewohner. Zeitschrift f¨ ur¨Agyptische Sprache und Altertumskunde 112: 48–84
1986. Amarna (T eil II): Analyse der ¨okonomischen Beziehungen der Stadt-
bewohnner. Zeitschrift f¨ ur¨Agyptische Sprache und Altertumskunde 113: 55–78
Tjan Tjoe-som 1949, 1952. PoHuT’ung: The Comprehensive Discussions in the White
Tiger H all,2vols. Leiden
Tod, M. N. 1948. ASelection of Greek Inscriptions ,vol. II: From 403 to 323 BC .O xford
Tomei, M. A. 2000. Le case di Augusto sul Palatino. MDAI(R) 107: 7–36
Toner, J. P . 1995. Leisure and Ancient Rome .O x f ord
Torelli, M., 1983. Gli spettacoli conviviali di et `aclassica: documenti archeologici su
possibili fatti genetici e sviluppi. In Spettacoli conviviali dell’ antichit` aclassica
alle corte italiane del’ 400: Atti del VII convegno di studio ,51–61. Viterbo

Bibliography 351
Tougher, S. (ed.) 2002. Eunuchs in Antiquity and Beyond .L o n d o n
Trigger, B. G. et al. 1983. Ancient Egy pt: A Social History. Cambridge
Tuplin, C. 1987. The administration of the Achaemenid empire. In I. Carradice (ed.),
Coinage and Administration in the Athenian and Persian Empires ,109–66. BAR
Inter national Series 34. London
Turcan, R. 1987. Vivre ` al ac o u rd e sC ´ esars .P a r is
Twitchett, D. and Loewe, M. 1986. The Cambridge History of China ,vol. I. Cambridge
Ullmann, M. 1978. Islamic Medicine .Edinburgh
Unvala, J. M. 1921. The Pahlavi Text King Husrav and His Boy. Published with Its
Translation, Transcription and Copious Notes .P a r is
Vallat, F. 1998. Elam i. The history of Elam. EncIr 8: 301–3
VandeMieroop, M. 1997. The Ancient Mesopotamian City. Oxford
2003. AHistory of the Ancient Near East ca.3000–323 BC .O xford
van den Boorn, G. P . F. 1988. The Duties of the Vizier: Civil Administration in the
Early New Kingdom. London
van Dijk, J. 2000. The Amarna Period and the later New Kingdom (c.1352–1069
BC). In I. Shaw (ed.), The Oxford History of Ancient Egypt ,272–313. Oxford
van Ess, H. 2004. Die Aufzeichnungen des Historiographen, die erste Standard-
geschichte Chinas, und das Aufkommen des b ¨urokratisch organisierten Zen-
tralstaates. In M. Hose (ed.), Große Texte Alter Kulturen: Literarische Reise von
Gizeh nach Rom ,87–110. Darmstadt
Verlet, P . 1985. Le chˆ ateau de Versailles ,2nd edn. Paris
Vernus, P . 2003. Affairs and Scandals in Ancient Egypt ,trans. D. Lorton. Ithaca, NY
Veyne, P . 1976. Lepain et le cirque: sociologie d’une pluralisme politique .P a r is
1981. Client `ele et corruption au service de l’ ´etat: la v ´enalit ´edes ofﬁces dans le
Bas-Empire. Annales ESC 36: 339–60
1990. Bread and Circuses: Historical Sociology and Political Pluralism .Trans. Brian
Pearce. London
Vogelsang-Eastwood, G. 1993. Pharaonic Egyptian Clothing .L e iden
2000. T extiles. In P . Nicholson and I. Shaw (eds.), AncientEgyptian Materials and
Technology ,268–98. Cambridge
V¨olcker-Janssen, W. 1993. Kunst und Gesellschaft an den H¨ ofen Alexanders d. Gr. und
seiner Nachfolger .Quellen und Forschungen zur antiken Welt 15. Munich
vonGall, H. 1990. DasReiterkampfbild in der iranischen und iranisch beeinﬂußten
Kunst parthischer und sasanidischer Zeit. Te h e r aner Forschungen 6. Berlin
Wallace-Hadrill, A. 1982. Civilis Princeps: between citizen and king. JRS72: 32–48
1983. Suetonius .L o n d o n
(ed.) 1989. Patronage in Ancient Society .L o n d o n
1994. Houses and Society in Pompeii and Herculaneum. Princeton
1996. The imperial court. In A. K. Bowman and others (eds.), Cambridge Ancient
History ,2nd edn, vol. X, 283–308. Cambridge
Wang Zhongshu 1982. HanCivilization ,trans. K. C. Chang and others. New Haven
and London
Ward-Perkins, J. B. 1981. Roman Imperial Architecture .Harmondsworth

352 Bibliography
Wataghin-Cantino, C. 1966. La Domus Augustana: personalit` aep r oblemi dell’
architettura Flavia .T u r in
Watson, B. 1961. Records of the Grand Historian of China ,2vols. New Y ork
1974. Courtier and Commoner in Han China .N e wY o r k
Weatherhead, F. 1992. Painted pavements in the Great Palace at Amarna, Journal of
Egyptian Archaeology 78: 179–94
Weaver, P . R. C. 1972. Familia Caesaris: A Social Study of the Emperor’s Freedmen
and Slaves .Cambridge
Weber, G. 1997. Interaktion, Repr ¨asentation und Herrschaft. Der K ¨onigshof im
Hellenismus. In Winterling 1997b: 27–72
Weber, U. Prosopographie des fr¨ uhen Sasanidenreiches (www.uni-kiel.de/klassalt/
projekte/sasaniden/index.html)
Weinberg, J. 1999. The international elite of the Achaemenid empire. Reality and
ﬁction. ZA W 111: 583–608
Weitzmann, K. (ed.) 1979. The Age of Spirituality .P rinceton
Whitby, M. 1987. On the omission of a ceremony in mid-sixth century Constantin-
ople. Historia 36: 462–88
(ed.) 1998. The Propaganda of Power: The Role of Panegyric in Late Antiquity.
Mnemosyne, Bibliotheca Classica Batava. Supplementum 183. Leiden
Wieseh ¨ofer, J. 1980. Die ‘Freunde’ und Wohlt ¨ater des Grossk ¨onigs. StIr 9: 17–21
1994. Die‘dunklen Jahrhunderte’ der Persis .Zetemata 90. Munich
2001. Ancien tP e r sia,2nd edn, rev. London
2005a. R ¯um as enemy of Iran. In E. Gruen (ed.), Cultural Borrowings and Ethnic
Appropriations in Antiquity .OrOcc 8. Stuttgart: 105–20
2005b. Iraniens, Grecs et Romains .P a r is
2006. ‘Randkultur’ oder ‘Nabel der Welt’? Das Sasanidenreich und der Westen.
Anmerkungen eines Althistorikers. In Wieseh ¨ofer and Huyse 2006: 9–28
2007a. Chusro I. und das Sasanidenreich. Der K ¨onig der K ¨onige ‘mit der
unsterblichen Seele’ . In M. Meier (ed.), Sie schufen Europa .Munich: 195–215
2007b. Narseh, Diokletian, Manich ¨aer und Christen. In A. Mustafa, J. Tubach and
G. S. Vashalomidze (eds.), Inkulturation des Christentums im Sasanidenreich .
Wiesbaden: 161–9
2007c. From Achaemenid imperial order to Sasanian diplomacy: war, peace and
reconciliation in pre-Islamic Iran. In K. Raaﬂaub (ed.), War and Peace in the
Ancien tW o r ld.Oxford: 121–40
Inpress (a). Household, Residence and Court: The Persian King and His Palace
Wieseh ¨ofer, J. and Huyse, P . (eds.) 2006. ¯Er¯an ud An¯ er¯an: Beitr¨ age zu den Beziehungen
zwischen Ost und West in sasanidischer Zeit. OrOcc 13. Stuttgart
Wilkes, J. 1986. Diocletian’s Palace, Split: Residence of a Retired Emperor .Shefﬁeld
(repr. with corrections Oxford, 1993)
Wilkinson, T. A. H. 1999. Early Dynastic Egypt. London
2000. Political uniﬁcation: towards a reconstruction, Mitteilungen des Deutschen
Arch¨ aologischen Instituts, Abteilung Kairo 56: 377–95

Bibliography 353
Winter, E. 1988. Legitimit ¨at als Herrschaftsprinzip: Kaiser und ‘K ¨onig der K ¨onige’
im wechselseitigen Verkehr. In H.-J. Drexhage and J. S ¨unskes (eds.), Migr atio
etCommutatio: Studien zur Alten Geschichte und deren Nachleben. Th. Pek´ ary
zum 60. Geburtstag am 13. September 1989 dargebracht von Freunden, Kollegenund Sch¨ ulern St. Katharinen: 72–92
Winter, E. and Dignas, B. 2001. Rom und das Perserreich: Zwei Weltm¨ achte zwischen
Konfrontation und Koexistenz .B e r lin
Winter, I. 1986. The king and the cup. Iconography of the royal presentation scene
on Ur III seals. In M. Kelly-Buccellati (ed.), Insight through Images: Studies in
Honour of Edith Porada ,253–68 and pls. 63–6. Malibu, CA
Winterling, A. 1985. Der Hof de rK u r f ¨ ursten von K¨ oln 1688–1794. Eine Fallstudie zur
Bedeutung ‘absolutistischer’ Hofhaltung .B o n n
1997a. Vergleichended Perspektiven. In Winterling 1997b: 151–70(ed.) 1997b. Zwischen ‘Haus’ und ‘Staat’: antike H¨ ofe imV e r g l e i c h .Historische
Zeitschrift Beiheft 23. Berlin
(ed.) 1998. Comitatus: Beitr¨ age zur Erforschung des sp¨ atantiken Kaiserhofes .B e r lin
1999. Aula Caesaris: Studien zur Institutionalisierung des r¨ omischen Kaiserhofes in
der Zeit von Augustus bis Commodus 31 v. Chr.–192 n. Chr .Munich
2004. ‘Hof’ . Versuch einer idealtypischen Bestimmung anhand der mittelalter-
lichen und fr ¨uhneuzeitlichen Geschichte. In Butz et al. 2004: 77–90
Wintrobe, R. 1998. The Political Economy of Dictatorship .Cambridge
Yarshater, E. 1983. Iranian national history. In The Cambridge History of Iran ,
vol. III/1. Cambridge: 359–477
Y¨uYing-shih 1967. Trade and Expansion in Han China .B e r k e l e yand Los Angeles
Zanker, P . 1988. The Power of Images in the Age of Augustus. Ann Arbor
2002. Domitian’s Palace on the Palatine and the imperial image. In K. Bowman,
H. M. Cotton, M. Goodman and S. Price (eds.), Representations of Empire: Rome
and the Mediterranean World ,105–30. Proceedings of the British Academy 114.
London
Zivie, A. 1990. D´ecouverte ` aSaqqara: le vizir oubli´ e.P a r is

Index
Abyd os,277
access to ruler, 23,39–40,108–9,125,153,
250,291–5,seealso salutatio
acclamations, 211
Achaemenids, seecourts ,palaces ,Persia,
wome n
adventus ,210,226
Aegeae, 91,seealso Verg ina
Aelius Seianus, Lucius, 141–2
Aem ilius Lepidus, Marcus (consul AD 6), 124
agentes i nr e b u s ,199
Agricola,seeIulius Agricola
Agrippa, seeVipsanius Agrippa
Ahura Mazda, 36, 62, seealso Ohrmezd
Ai, Han emperor, 257,258
Akhenaten, 267–8
and ceremonial, 312–17
seealso ‘Appearance, Window of ’
Akhetaten, seeel-Amarna
Alexander III (‘the Great’), Macedonian
king, 82–120,177–8,224
Alexander Severus, 177
Alf¨oldi, Andreas, 163,176
Amarna, seeel-Amarna
Amarna Letters, 288–90
Amarna period, 267–328,seealso Akhenaten ,
Egypt
Ambrose, bishop, 203,223
Amenhotep III, 274,289,312,seealso Kom
el-’Abd
amici ,amicitia ,143–4,208,seealso friends
Ammianus Marcellinus, 162,169,201,221
‘Amun, God’s Wife of’ , 299,300
Amun, priesthood of, 299,326
analogical thinking, 260
Ankhsenamun, 301
Anshan, 21
Antioch (Syrian), 191
Antonius Musa, 143,seealso physicians
apadana ,seepalaces (Persian)
Apollo Palatinus, temple of (Rome), 152,153
‘App earance, Window of’ , 313–17
Arca dius, Roman emperor, 162Archelaus, Macedonian king, 90
aristocrats, seenobility, nobles
Assyria, Assyrians, 23–5
Aten, 250–1,324
Athenaeus, 83
Athens, Athenians, 92
atrium ,128
audience-giving, audiences, 98–9,108,233,
245–8
audience-hall, 319
Augustan History ,161–2
Augustus, 124,132,135,136,140,145,146,152
aula,127,140
aul¯e,seecourts ,terminology for
Aurelian, Roman emperor, 173
Aureli us Victor, 170
Ausonius, 185
Ay,Egyptian king, 287,288,301,see
‘Appearance, Window of ’
Babylon, Babylonia, 23
banquets, banqueting, seefeasting
barbarian, barbarians, 89
basileion ,128,188
Berve, Helmut, 83
Bessus, 101–2
Bielenstein, Hans, 233
bishops, 186
Bisitun, 27,37
bodyguards, 85,126,131,197,241,282
‘boys, ro yal’ ,91
Britannicus, 141
brokerage, brokers, 85,93,135,141,seealso
courtiers
Brown, Peter, 158,160
bureaucracy, 161,180–1,198–9,228–9,see
also Handynasty
Caesar, Julius, 131–4,151
Caligula, Gaius, 136,148,152,153
Calpurnius Piso, Gaius, conspirator, 137
Calpurnius Piso, Gnaeus (consul 7 BC), 141
Caracalla, emperor, 145,177

Index 355
Cassius Dio, 123,124,145
castration, 261,seealso eunuchs
Catulus, seeLutatius Catulus
ceremonial, ceremonies, 89,209,226–7,239,
248,311–22,seealso adventus ,
proskun ¯esis,seating ,sacriﬁces
chamberlain, 318,seealso cubicularius ,
praepositus
Chang, Later Han emperor, 237
Ch’ang-an, Chinese capital, 236–7,238–9,241
Chao, Han emperor, 255–6
Chao Fei-yen, Han empress, 256–7
ch’ao, seecourts ,terminology for
Chares, 83,94,118–20
charisma, charismatic, 64,140,175
Ch’eng, Han emperor, 237,256–7
Ch’in Shih Huang-ti (First August Emperor
of the Ch’in), 235
China, seeHandynasty
Christianity, 186–7
church and court, 186–7
observance of emperors, 220
Cicero, 133–4
civilis princeps ,civilitas ,174,205,206,208,219
Claudian, 162
Claudius Marcellus, Marcus
(consul 51 BC), 134
clementia ,133
Clodius Thrasea Paetus, Marcus, 145
clothes, clothing, seedress
comitatus ,168,188,196–209
comites ,143–4,181,seealso amici
Companions, 85–6,99,108,110
concubines, seeeunuchs ,harem, women
Confucianisation, Confucius, 237,245
consistory, 198
Constantine I, 158,172–4,180–3,186,212,
213,220
Constantine Porphyrogenitus, 170
Constantinople, 158,183–4,189,212,213,
231,seealso Great Palace
Constantius I, 175
Constantius II, 177,210,215,224
convivium ,139,149–50
Cornelius Fronto, Marcus, 146
Cornelius Lentulus Augur, Gnaeus, 144
court, courts
deﬁnitions of, 2–3,18–19,135,140–1,
167–8
late Roman images of, 226,227
origins (Egypt), 275–8
as redistributors of wealth, 135
Sasanian, 58–79terminology for, 25,87,168,196,233,280,
seealso aula
courtiers
as brokers, 135,140–1
Assyrian, 24
Egyptian, 94,284–91
recruitment of, 53–4,66–71
seealso friends
cubicularius ,142,202,seealso chamberlain
cup-bearer, 85,93,100,282,286
CyrilofAlexandria, bishop, 203
Cyrus(II) the Great, 47–9,54–5,112
Darius I, 51–2
Darius III, 93–102,112
Datianus, 185,199
diadem, 176
dining, dinners, seeconvivium ,feasting
Dio, seeCassius Dio
Dio of Prusa, 209
Diocletian, 157,161,172–4,175,176,177,
206
diptychs, 216,219
Dium, 93,96,105
divinity, of rulers, 173,279,280–3,seealso
proskun ¯esis,ruler-cult ,s a c r iﬁces,
sed-festival
doctors, seephysicians
domestication, court as instrument for, 165
Domitian, 123,126,137,148,149–50,153–4
Domus Flavia, 154,193,seealso Palatium
dress, 93,111,147,320–2
of Han emperors, 246,248,249
Duindam, Jeroen, 281,294,328
Egypt
court, origins of, 275–8
king and court, 278–9
royalresidences, 302–11
seealso Amarna period
Eighteenth Dynasty (Egypt), seeAmarna
period
Elam, 21–2
el-Amarna, 284,287,303–5,306–8,310–11
courtiers’ tombs and houses, 322–3
seealso ‘App earance, Window of ’
elephants, 110
Elias, Norbert, 4–7,17,83,121,127,133,134,
139,145,164,225,241,263,326
emperor, Chinese, seeHandynasty
empresses
late Roman, 229
seealso wome n

356 Index
Epaphroditus, alibellis ,142,143
Ephippus, 99,118
Epictetus, 124,142
equestrians, equites ,131,132,139,143,146,
179,183
eunuchs, 26,93,100,129,162,202–8,241,
seealso Handynasty
Eusebia, 212
Eusebius, 178,186
Eutropius, historian, 170
Eutrop ius,praepositus ,203,207
examination system, seeHandynasty
familia ,seefreedmen ,slaves
FanYeh, historian, 265
favourites, 208,257,286
feasting, feasts, 41–4,86,87,90–3,98,
99–101,126,186,220–2,320,seealso
convivium
feudalism, 78
FirstEmperor(China), seeCh’in Shih Huang-ti
ﬂattery, 86,87,124,130,132,136
Flavius Abinnaeus, 216
Flavius Philippus, 185
freedmen, 124,125,129,142–3,seealso
familia ,slaves, social mobility
Friedl ¨ander, Ludwig, 126
friends, 39,73,84,140,seealso amici ,comites ,
Companions, kiss
Gag´e, Jean, 127
Gaius, emperor, seeCaligula
Galba, Roman emperor, 138
Gallus, emperor, 221
gestures, seekissing ,proskun ¯esis
Gibbon, Edward, 158–60,163,
204
gift-giving, gift s,54–6,107–8,138,
150–1,284
Gobryas, 37–8
governance, 319,seealso audience ,consistory ,
feasting ,gift-giving ,Handynasty
(emperor’s role) ,intrigue
Gratian, 223
Great Palace, 193–4,222,seealso
Constantinople ,Hippodrome
Handynasty
bureaucracy, 240,259–60
emperor’s role, 245–50
eunuchs, 259–62
examination system, 243–5
harem, 252,253–9history, 234
palaces, 238–9,241
seealso hunting
HanKao-tsu, ﬁrst Han emperor, seeLiu Pan
Han-shu ,seePan ku
harem, 93,97,100,206,289–90,297,seealso
eunuchs ,Handynasty ,wome n
‘harem’ palaces, 289
Hatshepsut, 285,299,326
hazarapatiˇ s,27–9
Helicon, cubicularius ,142
Hetairoi, seeCompanions
hierarchy, 324–5,seealso rank,seating
Hippodrome, 213
homosexuality, seeAi, Han em peror
Honori us, emperor, 162
Hopkins, Keith, 164,204
Hore mheb, 298,seealso ‘App earance ,
Window of ’
HouHan-shu ,seeFan Yeh
houses, of Roman senators, 130,151
hunting, hunts, 44–5,76,111,222–4,250–1,
254–5
HuoKuang, Han regent, 255–6
Husraw I, 75–8
immortality, and Han emperors, 249
inscriptions, royal, 62,seealso Bisitun ,
Naqsh-i Rustam
intrigue, 262,295–302,327,seealso ﬂattery
investiture
of Persian king, 41
late Roman, 213,215,216
Iulius, Agricola Gnaeus, 138,148
jesters, 251
Jones, A. H. M., 163,228
Jovian, 221
jubilee, seesed-festival
Julian the Apostate, emperor, 169,178,208,
220,221,224
Justinian, 74
Karnak, 312,seealso Thebes
Kienast, Dietmar, 92
King’s Eye, 30
kings, Hellenistic, 129,130
King’s Table, seefeasting, feasts (Persian)
kiss, kissing, 110,139,143,147–8,215
Kiya,secondary wife, 289
Komel-’Abd, 305,310
K¨onigsnovelle ,267
Kuang-wu, Later Han emperor, 237

Index 357
Lactantius, 192
La Fontaine, 166
Later Han dynasty, 237,258,seealso Han
dynasty
Lausus, palace of, 195
Lent ulus, seeCornelius Lentulus
Leonnatus, 104
Libanius, 221
Licinius, emperor, 192
Ling, Later Han emperor, 261
literature, and Han emperors, 252,seealsoPan
Ku,Ssu-ma Ch’ien
Liu Pan, ﬁrst Han emperor, 236,237,240,253
Livia, empress, 141
‘Livia, House of’ , 152
Livius Drusus the Y ounger, 130
Louis XIV , 121,166,169,248
Loewe, Mi chael, 233
Lo-yang, Chinese capital, 237,239,257,
258
L¨u, Han empress, 237
Lupercalia, 131,132
Lutatius Catulus, Quintus, consul 78 BC,
151
Macedon, Macedonians, seeAlexander III ,
Persia
MacMullen, Ramsay, 163
Maecenas, 140,141,144,153–4
Malkata, 305,310
Marcus Aurelius, 125–6,146
Mark Antony, 132
‘marquess’ , Chinese noble title, 240
Mathew, Gervase, 163
May,royalfanbearer, 285
Medes, Media, 22
Memphis, 303
Milan, 189,223
Millar, Fergus, 127,173
mobility, social, 185,seealso freedmen
Mommsen, Theodor, 126
Munatia Plancina, 141
music, at Han court, 251
Naqsh-i Rustam, 27–9,41,62
Narcissus, freedman, 141,142,143
Narmer P alette, 275–6,278–9
Narses, Sasanian king, 176
Nefertiti,287,289–90,300
Nero, emp eror, 124,145,152,153
Nicomedia, 192–3
nobility, nobles, 36–9,66–71,240,282
notarii ,201NotitiaDignitatum ,170,201,202
obeisance, see proskun ¯esis,prostration
obelisk, 213,214–16
Ohrmezd, 65,seealso Ahura Mazda
Opis, 104
Ottomans, seeTurks
pages, 70, seealso ‘boys, royal ’
palaces
Hellenistic, 130
Persian, 25,46–53
late Roman, 187–96
seealso Egypt ,Great Palace ,Handynasty ,
Palatium ,Pasargadae ,Pella ,Persepolis ,
Susa
palatium ,188
Palatium, at Rome, 128,151–5,194
Pallas, freedman, 143
PanKu, historian, 258,265
Parennefer, cup-bearer, 286
Parnaka, uncle of Darius I, 29–30
Pasargadae, 41,47–9
Pella,90
Perrault, Charles, 166
Persepolis, 50–2,93,96
Persia and Persians, ancient, 17–20,56,223
inﬂuence on Macedonians, 90–2,93–101,
112
seealso Sasanians
Peucestas, 103
Philip II, Macedonian king, 86,87,90–2
physicians, 34–5,140,seealsoAntoni us Musa
Piso, seeCalpurnius Piso
Plancina, seeMunatia Plancina
Pliny the Elder, 144
Pliny the Y ounger, 124,128,209
Plotinus, 178
Plutarch, 124
Polyaenus, Macedonian writer, 95,109,
118–19
praepositi ,202–4,206,seealso eunuchs
praetorian prefects, 199–200
processions, 212,seealsoadventus ,ceremonial
proskun¯ esis,103–6,176,178
prostration, 291
Ptolemy, Macedonian Companion, 100
Pulcheria, 230
purple, adoration of the, 173,176,213,
214–16
quaestor sacri palatii ,171
Quintilian, 134

358 Index
Ramesses III, 297
ranking, ranks, social, 66–71,87,102–6,
264
Ravenna, 191
religion, seeAhura Mazda ,Akhenaten, Aten ,
Christianity ,divinity ,Confucianisation ,
Handynasty ,sacriﬁces
Rome, 95,175,188,196–209,210
Rostovtzeff, Michael, 158–60
ruler-cult, 176,seealso divinity
sacriﬁces, by Han emperor, 245
Saint-Simon, duc de, 123
Sakkara, 277,303–5
salutatio ,127,130,139,143,144,146,147,
148,153,215
Sasanian Persia, Sasanians, 175–6,223,seealso
court ,Sasanian
satrapies, satraps, 20,35–6,111
seating, 104,246,248
sed-festival, 311–13
Sejanus, seeAelius Seianus
Sempronius Gracchus, Gaius, 130
senate, senatorial order, senators, Roman, 126,
127,130,131,132,139,146,175,179,
182–5
Senenmut, Egyptian ofﬁcial, 285
Sextus of Chaeronea, 126
Shabuhr I, 72–3
Shih-chi ,seeSsu-ma Ch’ien
‘Sinuhe, Story of’ , 291,296–7
slaves, 273, seealso freedmen
Smenkhkare, 298
‘Son of Heaven’ , seeHandynasty
Split, Diocletian’s palace at, 191–225
Ssu-ma Ch’ien, Chinese historian, 252,
253–9
Statius, 126,150
succession, imperial, at Rome, 139
Sueton ius, 125,174
Susa, 21,52–3,89,96,97,111,120
Synesius, 162
table, royal, seefeasting
Tacitus, 124,137–8,139,174
tents, 33,86,87,90,93,94–7,112–20
Te t r a rc hs, T etrarchy, 158,172,189
Thebes (Egyptian), 303,seealso MalkataThemistius, 211,221
Theodosian Code, 170–1,202,212,219
Theodosius I, 208–9,213,215,217,221–2,
226,227
Theodosius, Missorium of, 217
Thrasea Paetus, seeClodius Thrasea Paetus
throne, 91
Tibe rius,125,128,138,141–2
Tiye, q ueen, 287,289,300
Trajan,148,149,154
Trier, 193
Turks, Ottoman, 96
Tutankhamun, 284,289
tyrant, as a monarchical model, 122
Tzu-fu, seeWei,empress
Ursicinus, 215
ushers, 93–4
Valentinian I, 198,215,221,224,226–7
Vergina, 105,seealso Aegeae
Versailles, 153,164–7,174,186,206
Vespasian, 144,153,154
Veyne, Paul, 164,172,186
Vienna, 127
villas, late imperial, 195
Vindolanda, 128
Vipsanius Agrippa, Marcus, 140,141,144,
152,153
Vitelli us, Aulus, emperor, 140
Vitruv ius, 130
Voltaire, 166
Wallac e-Hadrill, Andrew, 127,164
Weber, Max,134,165,242
Wei, Han empress, 254–6
Weifamily, seeWei,empress
Wen, Han e mperor, 237,249
Winterling, Aloys, 127,165,241
wome n, at court, 24–5,31–4,64,129,141,
289–90,299,seealso empresses ,Eusebia ,
Handynasty ,harem
Wu, Han emperor, 237,240,242–5,250–1,
256–7
Y¨uan Hung, historian, 266
Zanker, Paul, 154,164

Copyright Notice

© Licențiada.org respectă drepturile de proprietate intelectuală și așteaptă ca toți utilizatorii să facă același lucru. Dacă consideri că un conținut de pe site încalcă drepturile tale de autor, te rugăm să trimiți o notificare DMCA.

Acest articol: This page intentionally left blank [606314] (ID: 606314)

Dacă considerați că acest conținut vă încalcă drepturile de autor, vă rugăm să depuneți o cerere pe pagina noastră Copyright Takedown.

This page intentionally left blank [606314]

Copyright Notice

LOCUL ȘI ROLUL TRANSPORTULUI AERIAN ÎN CADRUL TRANSPORTULUI INTERNAȚIONAL [303165]

Date preluate de pe www.worldvaluessurvey.orgWVSOnline.jsp 267 [600091]

Oprea M. Andra Maria [605715]

Denumirea obiectivului de investiții Analiza economică și detalii de execuție privind contruirea unei clădiri P+1E cu funcțiunea de grădiniță cu… [621234]

ACADEMIA DE STUDII ECONOMICE A MOLDOVEI FACULTATEA CONTABILITATE CATEDRA CONTABILITATE ȘI AUDIT Sorina CAZACU CONTABILITATEA CHELTUIELILOR (în baza… [622530]

INTRODUCERE ………………………….. ………………………….. ………………………….. ……….. 5 CAPITOLUL I – ASPECTE… [623564]

Copyright Notice

Similar Posts