Eunuchs in Antiquity and Beyond

EUNUCHS IN ANTIQUITY AND BEYOND
Frontispiece. Coin ofEumenes I showing the head of his uncle Philetairos, the
first Attalid dynast of Pergamum and reportedly a eunuch (3rd century Be).
Photo: British Museum.
EUNUCHS
IN ANTIQUITY AND BEYOND
Editor
Shaun Tougher
Contributors
Ra' anan Abusch, Ruth Bardel, Vern L. Bullough,
Niels Gaul, Shelley Hales, J .L. Lightfoot,
Lloyd Llewellyn-Jones, Margaret Mullett,
Georges Sideris, Walter Stevenson, Shaun Tougher,
Shih-shan Henry Tsai, Richard Witt
The Classical Press of Wales

and
Duckworth
First published in 2002 by
Gerald Duckworth & Co. Ltd.
61 Frith Street, London WID 3JL
(sole distributor outside N. America)
and
The Classical Press ofWales
Distributor in the United States of America:

The David Brown Book Co.
PO Box 511, Oakville, CT 06779
Tel: (860) 945-9329
Fax: (860) 945-9468
Originated and prepared for press at

The Classical Press of Wales
15 Rosehill Terrace, Swansea SAl 6JN
Tel: 01792 458397
Fax: 01792 464067
© 2002 The contributors
All rights reserved. No part of this publication may be reproduced, stored in

a retrieval system, or transmitted, in any form or by any means, electronic, mechanical,
photocopying, recording or otherwise, without the prior permission ofthe publisher.
ISBN 0 7156 3129 2
A catalogue record for this book is available from the British Library
Typeset by Ernest Buckley, Clunton, Shropshire

Printed and bound in the UK by Gomer Press, Llandysul, Ceredigion, Wales
CONTENTS
Page
Introduction Shaun Tougher Vll
Acknowledgements lX
1. Eunuchs in history and society 1

Vern L. Bullough (University of Southern California)
2. Eunuchs and the royal harem in Achaemenid Persia
(559-331 BC) 19
Lloyd Llewellyn-jones (Open University, U.K.)
3. Eunuchizing Agamemnon: Clytemnestra, Agamemnon and
maschalismos 51
Ruth Bardel (Oxford)
4. Sacred eunuchism in the cult of the Syrian goddess 71
JL. Lightfoot (All Souls College, Oxford)
5. Looking for eunuchs: the galli and Attis in Roman art 87

Shelley Hales (Cardiff and Bristol Universities)
6. Eunuchs and gender transformation: Philo's exegesis of the
Joseph narrative 103
Ra'anan Abusch (Princeton University)
7. Eunuchs and early Christianity 123
Walter Stevenson (University of Richmond)
8. In or out? Origins of court eunuchs 143
Shaun Tougher (Cardiff University, University ofWales)
9. 'Eunuchs of light'. Power, imperial ceremonial and positive
representations of eunuchs in Byzantium (4th-12th centuries) 161
Georges Sideris (College de France, Paris)
10. Theophylact ofOchrid's In Defence ofEunuchs 177
Margaret Mullett (Queen's University, Belfast)
v
Contents
11. Eunuchs in the late Byzantine empire, c. 1250-1400 199

Niels Gaul (University of Bonn)
12. Eunuch power in imperial China 221
Shih-shan Henry Tsai (University of Arkansas)
13. The other castrati 235
Richard Witt (Foundation of the Hellenic World, Athens)
Index 261
INTRODUCTION
Shaun Tougher
This introduction will not seek to provide a general overview of eunuch

history, the times and places in which eunuchs are found and the roles
that they played. This ground is ably covered by the contribution ofVern
Bullough. Rather, I wish to indicate the purpose of the conference from
which this book descended, as well as to reflect on the history and state
of the study of eunuchs.
All but one of the papers in this volume derive from a conference
entitled' "Neither Woman nor Man": Eunuchs in Antiquity and Beyond'
held in Cardiff from 26 to 28 July 1999. 1 The topic reflected the notable
flowering of interest in eunuch history in the last decade of the twentieth
century, but the conference also had a specific aim. Much of the study of
eunuchs in the 1990s was compartmentalized within individual cultures,
such as Beard on the Roman galli, 2 Ringrose on Byzantine eunuchs, 3 Tsai
on Chinese eunuchs in the Ming dynasty, 4 and Nanda on the hijras of
India. 5 These works also focused on particular issues of concern: traditional
political history in the case ofTsai, 6 cultural identity in the case of Beard,
and gender in the case of both Ringrose and Nanda. There was some
work that did take a broader view of eunuch history, but this could suffer
from a lack of expert knowledge.? Thus the conference was designed to
bring together scholars who were researching eunuchs in differing times
and cultures, as well as from differing points of view and with differing
approaches. It was hoped that this meeting would lead to the sharing of
information, views and experiences which would benefit and enrich the
study of the common subject-matter.
Those who attended the conference were happy to embrace this objec-
tive. It was remarked upon how unusual (as well as rewarding) it was to
be able to discuss one's work with those who shared one's enthusiasm. It
must be said that eunuch studies have been slow to develop. There seems
to have been a tendency to avoid the subject, perhaps due to distaste
or embarrassment. 8 In the field of ancient history, for instance, eunuchs
tended to surface in the study of religion rather than in their own right;
one thinks of the work ofNock9 and Vermaseren. 10 In the case of political
history, Badian's study of the eunuch Bagoas (the favourite of Alexander
Vll
Introduction
the Great, and sensitively immortalised in Mary Renault's The Persian Boy
(1972) ), speaks volumes; whilst Badian argues against Tarn in favour of
the existence of Bagoas, his interest is squarely focused on the question of
sources rather than on the eunuch: the article has after all the sub-heading
'A study in method' Y A breakthrough seemed to come with the classic
work of Keith Hopkins on the court eunuchs of the later Roman empire. 12
Taking a sociological approach, Hopkins investigated the reasons for the
use of eunuchs by later Roman emperors, as well as the nature of the court
eunuchs' power. For once, eunuchs were taken seriously. There followed
Orlando Patterson's own attempt to understand the usefulness of court
eunuchs, 13 as well as Peter Guyot's impressive survey of eunuchs as slaves
and freedmen throughout Graeco-Roman antiquity up until AD 395. 14
The eunuchs of antiquity (notably those of the later Roman period) also
shared in the general upswing of eunuch studies in the 1990s. 15 Despite
this, there still remains a certain tendency to overlook the subject generally.
Especially disturbing is that even two new 'guides' to late antiquity fail to
address eunuchs at all. 16 Thus it is hoped that this volume will assist in
making such omissions unlikely in future.
The conference was intended to be diverse in periods, cultures and
approaches, and the papers in the final volume span from the earliest
recorded instances of castration and eunuchs to the present day. Focuses
of interest vary, and range across politics, religion, music, literature, art,
and of course gender. Specific periods and cultures covered are the Persian
empire under the Achaemenids, classical Greece, Republican and imperial
Rome, the later Roman empire, the Byzantine empire, and in;tperial China.
Other periods and cultures are covered generally by those chapters which
take either a broader or a comparative approach. Specific contributions
on Islamic eunuchs, the medieval west, Skoptsi, hijras and the famous
singing castrati would have been welcome, but these topics are fortunately
well served by other work. 17
The volume commences with a broad introduction by Bullough, before
concentrating mainly on Greek evidence for eunuchs across the Persian,
Greek, Roman and Byzantine worlds. Llewellyn-Jones studies the eunuch
in the context of Persian 'harem' society. Bardel considers classical Greek
attitudes to castration, and mutilation generally. Lightfoot and Hales
explore the issue of eunuchs in the cult of mother goddesses (the Syrian
goddess and Cybele respectively), the former through literary evidence,
the latter through visual evidence. Abusch examines the treatment of the
eunuch in the thought of Philo of Alexandria, whilst Stevenson addresses
the question of the existence of eunuchs in early Christianity. Tougher,
investigating the ethnic origin of court eunuchs across a broad chronological
Vlll
Introduction
and cultural spectrum, focuses especially on the later Roman and Byzantine
empires. There follow three further chapters on the Byzantine empire,
Sideris and Mullett foreground positive treatments of eunuchs, whilst
Gaul turns the spotlight on the neglected eunuchs of the later Byzantine
empire. We then end with two chapters which highlight the cross-cultural
approach. In an overview of the history of the Chinese court eunuch, Tsai
emphasizes the hostility of the traditional accounts of Chinese history
towards eunuchs (which provides striking parallels with Greek prejudices),
and asserts the need for a revisionist history, whilst Witt (drawing on
evidence from ancient China to medieval Byzantium) makes the case for
the important role of eunuchs in music beyond the famous example of the
Italian castrati of the sixteenth to the nineteenth centuries.
It can be reported that the growth of eunuch studies shows no sign of
abating. Recent conferences have included sessions on eunuchs, 18 whilst
further work has continued to be published from 1999 onwards,l 9 and
more is promised. 20 As for the present volume, it is hoped that it is a
worthy reflection and justification of the burgeoning interest in eunuch
history, and also that it will assist and inspire future work.
Notes
1 The exception is Ruth Bardel's 'Eunuchizing Agamemnon: Clytemnestra,
Agamemnon and maschalismos'; this was a paper delivered at a day-school on

eunuchs also held in Cardiff, on 16 May 1998.
2 Beard 1994.
3 Ringrose 1994 and 1996.

4 Tsai 1996.
5 Nanda 1990. A second edition was published in 1999.
6 Note the more social interests of]ay 1993.
7 For instance Cheney 1995. See also my remarks on Grayson 1995 inch. 8.
8 There are of course exceptions, such as Millant 1908; Browe 1936. One
thinks especially of Byzantine studies with the work of Dunlap 1924 and Guilland
1943 (and also his subsequent studies).
9 Nock 1925.
10 e.g. Vermaseren 1977.
11 Badian 1958.
12 Hopkins 1963 and 1978, 172-96.
13 Patterson 1982, 314-31.
14 Guyot 1980.
15 e.g. Boulhol and Cochelin 1992; Schlinkert 1994; Scholten 1995; Stevenson
1995; Long 1996.

16 Bowersock, Brown and Grabar (eds.) 1999; Maas 2000.
17 On Islamic eunuchs see Ayalon 1999; Marmon 1995; Peirce 1993. On
lX
Introduction
the medieval west see Kuefler 1996. On the Skoptsi see Engelstein 1999. On
the hijras see Nanda 1990 and 1999; Jaffrey 1997. On the castrati see Barbier
1996; Rosselli 1988.
18 e.g. in 2001 there were panels on eunuchs at the 36th International Congress
on Medieval Studies in Kalamazoo and at the 20th International Congress of

Byzantine Studies in Paris.
19 e.g. on Cybele and the galli (Roller 1999); on late antiquity (Kuefler 2001;
Sideris 2000; Tougher 1999a); on Byzantium (Ringrose 1999; Tougher 1999b);

on Islamic eunuchs (Ayalon 1999); on castration in the medieval west (Murray
1999); and on the Skoptsi (Engelstein 1999). Recent general studies are Scholz
2001 (an English translation of the German original) and the more individualistic
Taylor 2000.
20 One looks forward in particular to Kathryn Ringrose's monograph on
Byzantine eunuchs.
Bibliography
Ayalon, D.
1999 Eunuchs, Caliphs and Sultans: A study in power relatiomhips, Jerusalem.
Badian, E.
1958 'The eunuch Bagoas: a study in method', CQ 8, 144-57.
Barbier, P.
1996 The World of the Castrati: The history of an extraordinary operatic
phenomenon, London.
Beard, M.
1994 'The Roman and the foreign: the cult of the "Great Mother" in imperial
Rome', in N. Thomas and C. Humphreys (eds.) Shamanism, History
and the State, Ann Arbor, 164-90. '
Boulhol, P. and Cochelin, I.
1992 'La rehabilitation de 1' eunuque dans l'hagiographie antique (IV•-VI
siecles)', Studi diAntichita Christiana 48,49-76.
Bowersock, G.W, Brown P. and Grabar, 0. (eds.)
1999 Late Antiquity: A guide to the postclassical world, Cambridge, Mass.,
and London.
Browe, P.
1936 Zur Geschichte der Entmannung: Eine religiom- und rechtsgeschichtliche
Studie, Breslau.
Cheney, V.T.
1995 A BriefHistory of Castration, Fort Walton Beach, Florida.
Dunlap, J .E.
1924 'The office of the grand chamberlain in the later Roman and Byzantine
empires', in Two Studies in Later Roman and Byzantine Administration,
New York and London, 161-234. '
Engelstein, L.
1999 Castration and the Heavenly Kingdom: A Russian folktale, Ithaca and
London.
X
Introduction
Grayson, A.K.
1995 'Eunuchs in power. Their role in the Assyrian bureaucracy', in M. Diet-
rich and 0. Loretz (eds.) Festschrift for Wolfram Freiherrn von Soden
(Alter Orient und Altes Testament 240), Neukirchen, 85-98.
Guilland, R.
1943 'Les eunuques dans I' empire byzantin', REB 1, 197-238.
Guyot, P.
1980 Eunuchen als Sklaven und Freigelassene in der griechisch-romischen Antike,
Stuttgart.
Hopkins, K.
1963 'Eunuchs in politics in the later Roman empire', Proceedings of the
Cambridge Philological Society 189, 62-80.
1978 Conquerors and Slaves, Cambridge.
Jaffrey, Z.
1997 The Invisibles: A tale ofthe eunuchs ofIndia, London.
Jay,J.W
1993 'Another side of Chinese eunuch history: castration, marriage, adoption,
and burial', Canadian journal ofHistory 28, 459-78.
Kuefler, M.
1996 'Castration and eunuchism in the Middle Ages', in VL. Bullough and
].A. Brundage (eds.} Handbook of Medieval Sexuality, New York and
London, 279-306.
2001 The Manly Eunuch: Masculinity, gender ambiguity and Christian ideology
in Late Antiquity, Chicago and London.
Long, J.
1996 Claudian's In Eutropium. Or, how, when, and why to slander a eunuch,
Chapel Hill and London.
Maas, M.
2000 Readings in Late Antiquity: A sourcebook, London and New York.
Marmon, S.
1995 Eunuchs and Sacred Boundaries in Islamic Society, New York and Oxford.
Millant, R.
1908 Les eunuques atravers les ages, Paris.
Murray,].
1999 'Mystical castration: some reflections on Peter Abelard, Hugh of
Lincoln and sexual control', in J. Murray (ed.) Conflicted Identities
and Multiple Masculinities: Men in the Medieval West, New York and
London, 73-91.
Narida, S.
1990 Neither Man nor Woman: The Hijras ofIndia, Belmont.
1999 Neither Man nor Woman: The Hijras ofIndia, 2nd edn, Belmont.
Nock,A.D.
1925 'Eunuchs in ancient religion', Archiv for Religionswissenschaft 23,
25-33.
Patterson, 0.
1982 Slavery and Social Death. A comparative study, Cambridge, Mass., and
London.
Xl
Introduction
Peirce, L.P.
1993 The Imperial Harem: Women and sovereignty in the Ottoman Empire,
New York and Oxford.
Ringrose, K.M.
1994 'Living in the shadows: eunuchs and gender in Byzantium', in G. Herdt
(ed.) Third Sex, Third Gender. Beyond sexual dimorphism in culture and
history, New York, 85-109 and 507-18.
1996 'Eunuchs as cultural mediators', ByzF 23, 75-93.
1999 'Passing the test of sanctity: denial of sexuality and involuntary castra-
tion', in L. James (ed.) Desire and Denial in Byzantium, Aldershot and
Brookfield, 123-137.
Roller, L.E.
1999 In Search of God the Mother: The cult ofAnatolian Cybele, Berkeley.
Rosselli, J.
1988 'The castrati as a professional group and a social phenomenon,
1550-1850', Acta Musicologica 60, 143-79.
Schlinkert, D.
1994 'Der Hofeunuch in der Spatantike: Ein gefahrlicher AuGenseiter?',
Hermes 122, 342-59.
Scholten, H.
1995 Der Eunuch in Kaisernahe. Zur politischen und sozialen Bedeutung des
praepositus sacri cubiculi im 4. und 5. ]ahrhundert n. Chr. (Prismata
V), Frankfurt am Main etc.
Scholz, P. 0.
200 1 Eunuchs and Castrati: A cultural history, Princeton.
Sideris, G.
2000 'La comedie des castrats. Ammien Marcellin et les eunuques, entre
eunucophobie et admiration', Revue beige de philologie et d'histoire 78,
681-717.
Stevenson, W
1995 'The rise of eunuchs in Greco-Roman antiquity', journal ofthe History
ofSexuality 5, 495-511.
Taylor, G.
2000 Castration: An abbreviated history of western manhood, New York and
London.
Tougher, S.
1999a 'Ammianus and the eunuchs', in J.W Drijvers and D. Hunt (eds.) The
Late Roman World and its Historian. Interpreting Ammianus Marcellinus,
London and New York.
1999b 'Images of effeminate men: the case of Byzantine eunuchs', in D.M.
Hadley (ed.) Masculinity in Medieval Europe, London and New York,
89-100.
Tsai, S.-s. H.
1996 The Eunuchs in the Ming Dynasty, Albany, N.Y.
Vermaseren, M.J.
1977 Cybele and Attis. The myth and the cult, London.
xu
ACKNOWLEDGEMENTS
It is a great pleasure to be able to thank those who made a significant

contribution to this publication and to the conference from which it
descended, held in Cardiff in 1999.
Pride of place goes to Shelley Hales and Lloyd Llewellyn-Jones, who
provided indispensable assistance and support throughout the preparation
for, and execution of, the conference. They were also co-organizers of the
preliminary day-school on eunuchs, held in Cardiff in the previous year.
Anton Powell deserves special mention too. He attended the original
day-school and enthusiastically urged the subsequent conference, the
proceedings of which he then offered to publish. His unstinting encourage-
ment and advice have been highly prized.
I am grateful to all those who attended the conference, and would
especially like to record the contribution of those who gave papers but who
do not appear in the volume: Mark Brustman; Deirdre Good; Dominic
Montserrat; Leo Prado; Kathryn Ringrose; David G. Smith; and David
Woods. Some of these contributors have already found outlets for their
work, whilst others are due to publish theirs in future.
I would also like to thank the staff of Aberdare Hall (the congenial
location of the conference) for their cheerful assistance and efficiency.
In the final stages of preparing the book for publication, the skills of the
typesetter Ernest Buckley have been greatly appreciated.
Lastly, I would like to dedicate this book to Raymond Davis, whose
teaching in Ancient History at the Queen's University of Belfast fostered
my fascination with the later Roman empire. ·
Shaun Tougher
Xlll
1
EUNUCHS IN HISTORY AND SOCIETY
Vern L. Bullough
Introduction
The date of the first appearance of eunuchs has been lost to history, but
castration of males whether animal or human is an old practice dating
from the time when humanity began to herd animals. Since most animals
tend to flock together and produce roughly the same number of male and
female offspring, and one male was all that was actually needed to service
dozens of females, castration to early herders served as the ideal solution
to the fighting and competition that resulted from male animals seeking
to mate with as many females as possible. Some individuals even became
specialists at it, and in the early Vedic record oflndia, the term vadhryasva
(literally, 'he who castrated horses') appears.
There are many methods for castrating animals and it is probably safest
to do so shortly after the testicles have descended in the newborn male.
Usually a string or horse-hair is tied tightly around the scrotum which
causes the testicles to turn black and drop off in about three weeks' time.
There are other ways. I remember that as a child when we castrated the
infant lambs, some of the men did so by holding the young lamb up to
their mouth and biting the testicles off. But testicles could also be crushed,
shattered, struck, edged, cut out, methods usually reserved for the larger
and older animals. The mortality rate is higher when adult males are
castrated. I was present at one castration of a horse by the local amateurs
when the horse died from bleeding to death. Some species of animals
were castrated so frequently that they developed special names, such as
gelding or ox. In fact, in parts of the United States, restaurant menus
list something called 'Rocky Mountain Oysters', i.e. testicles from the
castrated cattle.
In humans, originally, the castration process was probably quite brutal.
A slash of a sword could cut off the penis and testicles, but because this
is such a vascular area, the mortality rate of such an action would be
extremely high. Still, such drastic and painful methods could be a symbolic
way of killing an enemy. Some might even have survived to become
1
Vern L. Bullough
eunuch servants and courtiers. When eunuchs came to be treasured for

such services, their value increased and undoubtedly steps were taken
to make the process less dangerous. Sometimes only the testicles were .
removed, what we now label as castration, and this was much less life-
threatening, particularly if done while young. Total castration, removing
both the penis (penectomy) and testicles, had a higher mortality rate. We
have nineteenth-century accounts of this procedure in China where total
castration was the norm, and eunuch-maker was a speciality occupation.
The operation was expensive, was often done when the person was either
prepubescent or adolescent, and theoretically was what might be called
elective, that is the patient volunteered to have it done or family pressures
encouraged him to have it done. In preparation for the surgery, the
patient's abdomen and upper thighs were tightly bound with strings or
bandage leaving the penis and scrotum exposed. These were then washed
three times in hot pepper water while the patient sat in a semi-reclining
position on a heated, couch-like piece of furniture known in Chinese as
the k'ang. The eunuch maker repeatedly questioned the patient whether he
really wanted to go through with the surgery and if he continued to agree,
he was held firmly by assistants, and then, with one sweep of a razor-sharp
sickle-shaped knife, the penis and scrotum were cut off The urethra was
plugged and blocked off, and the wound was covered by paper soaked in
cold water after which the patient was bound tightly. The assistants then
had to walk the patient around for two or three hours before allowing him
to lie down. He was forbidden to take fluids for three days, and usually
towards the end of the period, the patient suffered almos~ as much from
thirst as from the after-effects of the operation. Mter the three days, the
urethra plug was removed and if the urine gushed out, the operation was
regarded as a success. If no urine appeared, the prognosis'was that the man
would soon die an agonizing death. G. Carter Stent, a nineteenth-century
observer, reported that over several years he saw only one patient who had
died, a 30-year-old man. 1 Mter castration, the eunuch's genitals were put
in a container where they were pickled, after which they were returned to
him for safekeeping. fu he advanced up the ranks of the palace hierarchy
or civil service, he had to show his preserved genitals. They were also
buried with him at death so that he could be restored to masculinity
in any future life.
Among the hijras in India, an Indian male cult, asceticism and renuncia-
tion of sexual desire are proven by total castration or emasculation in an
operation called nirvan, a term for the state of mind where the individual
is liberated from the finite human consciousness and is approaching the
dawn of higher consciousness. Hindu scriptures call this experience the
2
Eunuchs in history and society
second birth, or the opening of the eye of wisdom. The operation is

performed by a hijra called a dai ma (literally midwife). 2 The candidate
is bathed and seated on a small stool with the penis and testicles hanging
down and held from the back by the dai ma's assistant. The penis and
scrotal sac are tightly tied with a string in order to facilitate a clean cut.
The client attempts to get into a trance by looking at a picture of the god
Bahuchara and constantly repeating her name, Mata, Mata, Mata. The dai
ma makes two quick opposite diagonal cuts and the organs are separated
from the body. A small stick is inserted into the urethra to keep it open.
There is considerable blood and apparently nothing is done to prevent it
since it is believed to be the male element flowing out. The severed organs
are buried under a living tree. No stitches are utilized but the wound is
treated through repeated applications of hot sesame oil. The hour after
the operation is considered critical, a struggle between Bahuchara Mata,
who gives life, and her sister, Chamundeswari, who takes it. If the believer
survives for forty days, the facial hair, which has been left uncut, is pulled
out by tweezers. Turmeric is applied to the face and the body washed
off. The believer then assumes the clothing and identity of a woman.
The main occupation is to attend marriage and birth ceremonies where
the believers sing and dance. At birthing ceremonies they also examine
the baby, and often take with them any child with ambiguous genitalia,
removing such a stigmatized child from its family who themselves would
be stigmatized if they kept such an infant. Many of these children are then
recruited into the hijra sect. Many of the hijras also serve as prostitutes.
Probably most eunuchs do not have both their testicles and penis
removed but only their testicles. In Arabic, the term Khasi designated
a eunuch who only had his testicles removed while a madjbub had both
his testicles and his penis cut off.3 In the first case, often the testicles were
cut out and the scrotum cauterized by a red hot blade. An alternative was
to tie a cord around the testicles, causing them to bulge out and then
they were crushed. Tying the scrotal sac tightly above the testicles with
a cord in order to cut off circulation was also done. The mortality rate of
a simple castration was much lower than cutting off a penis which caused
major and sometimes uncontrolled bleeding, a greater chance of infection,
and often lifelong urinary problems as well, since it was more difficult to
control the dribbling. In fact, some commentators have said one could
identifY the eunuch quarters by the smell of stale urine. In some cases
of total castration, the individual was buried in sand up to the waist for
several days, again to curtail bleeding. Uniformly neither fluids nor foods
were taken for several days. Death resulted either from infection, bleeding
to death, or because of the scarring and closing of the urethra.
3
Vern L. Bullough
One of the difficulties with simple removal of the testicles was that it did
not always prevent arousal, depending upon the age at castration and the
methods used. If testicles were removed after puberty, the eunuch could
well have an erection since, although he would be sterile, he continued to
receive testosterone through the adrenal glands. If he had been castrated
by crushing the testicles rather than removing them, even at a young age,
it was still possible to have some testosterone from the testicles, depending
on the damage done to them. There might even have been a rare pregnancy
in some such cases, but the semen would have an extremely low sperm
count, making pregnancy extremely unlikely.
The appearance of eunuchs also depended in part upon the age at
castration. If castrated young, they retained the larynx of a boy, but
developed the lungs and chest of a man which gave the voice a rather
unique quality, especially in those who had good singing voices. Most did
not, and Chinese actors playing the roles of eunuchs often spoke in an
artificial, half-crying voice and groan like men in pain. Beardlessness was
another aspect of eunuchism, as was weakened muscular structure and
the fat distribution of a female.
Some have argued that in harems at least, eunuchs were not castrated to
prevent them from having sex with women, but rather to make certain the
pregnancy of the harem woman resulted from the seed of the master rather
than that of the servant or slave. Thus whether an erection was or was not
possible was of little concern, and total castration was unusual.
There were other forms of castration, some which seem to have been
particularly painful and life-threatening. A tortuous form of castration
involving splitting the penis is discussed in the Hindu Arthavaveda. 4
As far as females are concerned, although there is a rather bald-faced
statement by Athenaeus (jl. AD 200) quoting Xanthus, that Adramyttes,
King of the Lydians, castrated women and used female eunuchs, 5 it is
not clear what is meant. The modern English translations of the passage
usually say the women were sterilized. This still implies an ovariectomy
since there was early recognition of the existence of 'female testicles' but
whether surgical knowledge was up to successful performance of such
operations is doubtful to me. This is because one of the earliest mentions
of an ovariectomy in a medical work dates only from the end of the
seventeenth century when allegedly a Dutch sow-gelder had successfully
effected the removal of both ovaries from his own daughter in order
to prevent her from 'gadding' about at night. 6 Whether this is medical
folklore or reality is unclear. It was not until the end 'of the eighteenth
century that the first successful ovariectomy is recorded, by a surgeon,
the American Ephraim McDowell, and this was for the quite different
4
purpose of removing a tumour on the ovary? It was not until the twentieth
century that ovariectomy or oophorectomy was done for purposes of
sterilization but even this operation was rare, with tubal ligation being
used as an alternative.
There is an interesting sidebar which could be mentioned here, and that
is the castration and genital mutilation which took place at the end of the
nineteenth century in countries such as the U.S.A. in order to avoid the
dangers of masturbation, something I have written about elsewhere. 8 Most
of it was done in an institutional setting for the mentally handicapped
or mentally ill, but it was also occasionally done to others for 'severe'
masturbatory practices and for homosexual activity.
How frequent or infrequent was the practice of castrating males? We
know that it is recorded in ancient Mesopotamia, 9 as well as China and
India. Ammianus Marcellinus attributed the origin of eunuchism to the
legendary Semiramis who supposedly lived in the eighth century BC. 10
One of the difficulties, however, in identifYing eunuchs is the sources
themselves. A good illustration of this is the Christian Bible. For example,
the Latin vulgate version of the Bible (Genesis 37.6) reports that Potiphar,
the Egyptian official to whom the Midianites had sold Joseph, was
a eunuch. 11 A recent English translation of this calls Potiphar a courtier
of the pharaoh, 12 while the New Revised Standard Version identifies him
simply as an official of the pharaoh. 13 He was perhaps all three of these
things, but the important issues for us are that although he was identified
as a eunuch in some sources other sources obscure this, and thus some
commentators question it. 14 Also, often, in the immediate past, there was
a reluctance to deal with the topic and so euphemisms which might not
be understandable to the current generation of readers were used. We
know, for example, that many male slaves were castrated but it is not
always clear which. 15
Reasons for castration

Why were men castrated? Several reasons can be advanced: control and
domination, punishment, political reasons, need for special qualities or
abilities, religious requirements, sexual or erotic reasons, and medical
or health reasons.
Some ancient writers emphasized that eunuchs were easier to control.
For example, the Greek writer Xenophon in his biography of Cyrus the
Great of Persia wrote that Cyrus justified his policy of using eunuchs
after observing animals:
for instance, vicious horses, when gelded, stop biting and prancing about, to
be sure, but are none the less fit for service in war; and bulls, when castrated,
5
Vern L. Bullough
lose somewhat of their high spirit and unruliness but are not deprived of their
strength or capacity for work. And in the same way dogs, when castrated,
stop running away from their masters, but are no less useful for watching or
hunting. And men, too, in the same way, become gentler when deprived of
this desire, but not less careful of that which is entrusted to them. 16
Inevitably, many slaves were castrated since a castrated slave was more
expensive, in part because it was believed they were easier to control, and
there was no particular reason for the males not to be castrated unless there
were other tasks, such as fighting in the arena, which gave non-castrated
males more value. Although a high death rate was associated with the
procedure, the added profit was sufficient to make castration a regular
feature of the slave trade. There were dangers in this, however; Herodotos,
for example, tells us that Panionios castrated and traded slaves, and sold
the victims of his knife in Ephesus and Sardis, 'where they were much
esteemed because of their honesty and fidelity in every way' . 17 One of his
victims, Hermotimos, became the chief eunuch of the Persian king Xerxes
(fifth century BC). 18 In that capacity, he exacted his revenge on Panionios
by instigating a family-wide castration in which he and his four sons were
forced to castrate each other.
But castration was also used for punishment, and penises were collected
as trophies. Pharaoh Merneptah in the XIXth dynasty, for example,
memorialized that he collected a total of 6359 uncircumcised penises
after the defeat of the invading Libyan army as well as the penises of the
children of the chief, brothers of the priest, and others. 19 Whether any
of those so castrated survived is unknown, but probably the castration
took place after they had already been killed. The Assyrian' laws provided
castration, or making a person into a eunuch, as suitable punishment for
some crime with sexual overtones. 2° Castration, in fact, 'continued to be
used as punishment for certain sex crimes up to modern times and some
American states still list it as a possibility as do many other countries.
In the United States in recent years there have been several movements
to castrate, either literally or chemically, individuals involved in sex crimes,
particularly those involving adults with children. One of the early advocates
of castration for 'deviant sex acts' was Judge Lawrence Neil Turrentine
in San Diego County who, beginning in 1938, offered probation to
sex offenders who underwent castration. The practice was continued by
some of his successors. A California legislative report found that of sixty
convicted individuals who had been castrated in San Diego County, not
one committed a further sex offence.z' What was defined as a sex crime
in the 1930s and 1940s, however, was not necessarily defined as one in
the 1980s and 1990s, and many of those castrated in those years were
6
homosexuals engaged in sex with consenting partners. In recent years,

coinciding with a redefinition of sex crimes, actual physical castration
has increasingly been replaced by the alternative of chemical castration,
i.e. giving convicted male sex criminals Depo Provera or other female
hormones. How effective physical castration is in preventing sex crimes
is debatable, in spite of public belief to the contrary. Certainly castration
removes the major sources of testosterone in males and this ultimately leads
to a loss of desire and lessens the ability to ejaculate. But whilst testosterone
seems a sufficient basis, it is not a necessary one for sexual conduct, and the
adrenal glands might well give sufficient amounts for some to engage in
penetrative sex. 22 The problem is lessened with chemical castration but the
problem there is making the person take the necessary dosages.
Individuals were also castrated because their families saw advantages
for the individual and for themselves. Perhaps this had originally been the
case in those areas where the kings had harems and employed eunuchs
as servants and attendants, but eunuchs often became more than bed
attendants since their presence in the royal chamber gave them access to
the person of the ruler. In fact the possibility of this position is emphasized
by the emergence of the term chamberlain for a governmental official.
There are many historical examples. For instance, the ancient Korean
kings were surrounded by palace eunuchs as advisors, and boys from poor
families were castrated by their families in order to enter the king's service.
In theory, the recruits were given the best education possible so that they
might be a restrained and dispassionate influence on the monarchs. Many
of the eunuchs married and gave rise to eunuch families by adopting
castrated boys. 23 We know that eunuchs were part of the royal staff,
even sometimes governmental officials, in China, India, Persia, Rome,
Byzantium and in many Islamic countries. But eunuchs were more than
palace advisors; they were also generals and admirals and in the Byzantine
empire, patriarchs of the church as well, and special envoys for the rulers. 24
In China, the emperor often used them to undercut the Confucian-trained
civil service, and the rivalry between the two groups was often intense. 25
Eunuchs were trusted because by tradition in many of these areas, no
male without testicles could rule, and thus their position depended on
their loyalty to those in charge. This did not mean that there was not
'harem politics' and factions among the eunuchs, where there were large
numbers and the influence of any one individual or group depended
on their success in cultivating those in power or with a good chance of
coming to power. Eunuchs could acquire great influence on the children
of the rulers, who were often cut off from the outside world and depended
upon the eunuchs.
7
Vern L. Bullough
Eunuchs were also prized for their special qualities. Perhaps the best
example of this is that of the castrati, so important in western musical
tradition from the sixteenth to the eighteenth centuries. 26 But the castrati
singers predated the opera. They first appeared in the church choirs
which, as they became increasingly elaborate in the later Middle Ages,
sought higher voices to create greater harmonies. Since St Paul's statement,
that women were to keep silent in churches, 27 was interpreted by the
misogynistic clergy of the day to forbid women to sing in the choir,
eunuchs became the alternative. The increasingly elaborate choruses relied
on boys, falsettists, and ultimately eunuchs or castrati. 28 Technically, the
Christian church had banned self-castration for religious reasons in the
fourth century, 29 but if others did the castration, there was no ban on
utilizing their services. Poor families whose sons had good voices (or even
did not) turned to castration in the hope that it would bring them sudden
wealth. Though the number of boys castrated in the age of opera seems
exaggerated - for example, Ida Franca's claim that in a single year in the
eighteenth century more than 2000 boys had been castrated in the Papal
States of Europe, and more than 1000 in the city of Naples alone -
the numbers at any rate were large. 30 Unfortunately, castration might
have prevented the male voice from deepening, but it did not guarantee
that everyone castrated had a good singing voice. The practice of using
castrati declined somewhat towards the end of the eighteenth century,
and countries like France never or only occasionally employed them,
nor did the Protestant areas of Europe after the time of Handel. Pope
Leo XIII in 1878 finally issued an edict prohibiting the 4se of castrati.
However the effect was more gradual than immediate, since Dominico
Mustafa, a celebrated male soprano, was director of the Papal music until
his retirement in 1895. '
Still another justification for eunuchs is religionY The Gnostics,
whose growth paralleled that of Christianity, taught that men became
like beasts when they engaged in sexual intercourse. Some of them, such
as the Gnostic leader Julius Cassian, justified eunuchism. 32 Christians in
general were probably ambivalent since the scriptural sources themselves
are ambiguous. For example, when Jesus spoke of forbidding divorce or
remarriage, the disciples wondered whether it might be better simply to
remain unmarried. He said:
All men cannot receive this saying, save them to whom it is given. For there
are some eunuchs, which are born so from their mother's VfOmb, and there
are some eunuchs which were made eunuchs of men; and there be eunuchs
which have made themselves eunuchs for the kingdom of heaven's sake. 33
8
Although the meaning of the statement is unclear, it was interpreted

by most to mean only that service to God demanded a self-imposed
continence; some, however, took the words more literally. Origen, an
important figure in early Christian history, castrated himself 34 Others
did the same, but by the fourth century such acts of self-mutilation, as
indicated earlier, were forbidden by Church Councils. 35
This did not end the matter, however, and especially among the
Orthodox Christians, where castration persisted, eunuchs were a major
force. 36 The Byzantine author Theophylact of Ochrid penned at the
beginning of the twelfth century a justification of castration, supposedly
at the request of his brother who was a eunuch. 37 He emphasizes that
eunuchs could be monks, bishops, even saints, and argues that they could
have greater purity than non-eunuch ascetics.
The religious justification of castration and eunuchism is illustrated by
the Hindu sects associated with Vaishnavism. Vaishnavism held that only
the Godhead, Krishna, was truly male, and every other creature in the
world was female and subject to the pleasures of Krishna. The believers
worshipped Radha, the favourite milkmaid of Krishna, and the goal
of their devotion was the attainment of the state of a female attendant
upon Radha. Female followers of the sect granted sexual favours freely to
anyone, since Krishna himself was believed to participate in all sexual acts
with them. Male followers dressed like women, affecting the behaviour,
movements, and habits of women, even including menstruation. Many
made themselves eunuchs, and all were supposed to permit the sexual
act on their person (playing the part of women) as an act of devotion. 38
Similarly, as indicated above, the male members of the hijra sect cross-dress
as women, live as women, and are often castrated. 39
Within modern Christianity, there was the Russian sect of Skoptsi,
or eunuchs, who regarded procreation as the greatest evil and therefore
practised emasculation. 40 They were probably the most extensive users of
castration for the purpose of purification in modern Christian history.
They originated in Russia shortly after 1700 and one of their believers is
said to have been the Czar Peter III who was ousted by his wife Catherine
the Great in 17 62, for being 'less than a man'. Mostly illiterate and
largely from the peasant class, the Skoptsi probably reached their height
under Alexander I (1801-20) who indulged them. During the reign of
Nicholas I (1825-55) they were declared to be a harmful sect and this
designation remained in effect until the Soviet Revolution. Many of them
fled to Romania. In spite of their persecution, their numbers are estimated
to have reached the 100,000 mark by 1920. During the early Soviet period,
many joined the communists because of the emphasis on the communal
9
Vern L. Bullough
ideal, and some even held positions in the Supreme Soviet Council. The
government moved against them in 1929 and imprisoned their leader D.
Lomonsov for opposing socialism and spreading 'religious superstition'.
They continued to survive, however, in Russia and Romania, by going
underground. There were various levels of castration. Those of the 'small
seal' only had their testicles removed; those of the 'great seal', whether male
or female, had all of their sex organs removed. Perhaps the majority, known
as white doves, elected spiritual castration, i.e. total abstinence. 41 Many
survived the Soviet persecution and apparently the sect still exists. 42 Many
of them, before the Soviets came to power, were illiterate, and as one, who
had learned to read, commented at his trial in 1903:
We mutilate ourselves; no one taught us anything ... I followed the word of
the Evangelist and the teaching of our saviour, Matthew 19:12 ... I understood
that I had castrated myself for the kingdom of heaven ... If castration had
not been necessary, Christ would not have spoken. 43
Still another reason for castration is simply sexual purposes. According

to the scandal-mongering Roman biographer, Suetonius, Nero in the first
century AD castrated the boy Sporus and tried to make him a woman. 44 The
Roman writer Petronius in his Satyricon emphasizes this erotic aspect:
0 faeries, 0 buggers
0 eunuchs exotic!
Corne running, come running,
ye anal erotic!
With soft little hands,

With flexible burns
Corne 0 castrati,
unnatural ones! 45
Desire for eunuchs as sexual partners appears also in China, 46 and it

is a sub-theme in much of the writing on eunuchs. Even those who had
both penis and testicles removed, retained their prostate, and massaging
it through a deep penetration of the anus can result in orgasm of sorts.
There could even be an ejaculation (without sperm).
It is also possible that many of those who became eunuchs wanted to
become more like women. In fact, Christine Jorgensen, regarded as the key
figure in the emergence of transsexualism as a diagnostic surgery, simply
had her penis and testicles removed in the first surgery and began living
as a woman, an action that gave her international publicity and ultimately
led to a new diagnostic category of transsexualism. What distinguished
Jorgensen and her successors from eunuchs of the past was the discovery
10
of hormones and the development of artificial ones. This allowed the

eunuchoid male to develop female breasts, fat distribution, and to some
extent body shape, even though castration took place after puberty.
The same was true of the females who underwent mastectomies and
ovariectomies. With administration of testosterone they developed beards
and more male-looking bodies as well as lower-sounding voices. In
Christine's case, it was only several years after the initial surgery that
she underwent reconstructive surgery to make a vagina, and this was
not particularly successful. Later, however, the surgery for males-to-
females improved, but there is still difficulty with the female-to-male
in constructing a working penis. I have hypothesised elsewhere that the
willingness of males to have their penis removed in the past might well
have been due to many of the factors that we now use for the justification
of sex change surgeryY
Probably in the twentieth century, at least in the west, the largest
number of castrations and ovariectomies were done for eugenic or medical
reasons. The word eugenics was coined by Francis Galton, the cousin of
Charles Darwin, in 1885, who defined it as an applied biological science
concerned with increasing from one generation to another the proportion
of persons with better eugenic traits, that is intellectual endowment as
well as desirable physical appearance. His disciple, and first holder of
the eugenics chair at the University of London, was the mathematician
Karl Pearson. He was convinced that environment had little to do with
the development of mental or emotional qualities and both were based
on heredity. This meant that for him most social problems could be
controlled by judicious breeding, i.e. curtailing the breeding opportunities
of the inferiors who should be sterilized. Galton did not fully agree with
Pearson, but the American Eugenics movement founded in 1905 followed
Pearson. In the United States, many in the eugenic movement believed
that the white race was superior to other races and that within the 'white
race', Nordics ranked the highest. Using the developing concepts of IQ
testing, the eugenicists in the United States and elsewhere set out through
sterilization to prevent the feebleminded, insane, epileptic, the socially
undesirable, including the poor immigrants and the poverty stricken,
habitual criminals and 'moral perverts' from reproducing themselves. 48
This often meant castration for males and ovariectomies for females even
though tubal ligation had been performed since the last decades of the
nineteenth century and vasectomies from the first part of the twentieth. 49
What eventually discredited eugenics was the extremes that the Nazis
used in Germany where Jews, gypsies, homosexuals, political dissenters,
and many others were castrated or underwent ovariectomies. Most of
11
Vern L. Bullough
the United States which had castration laws on their books proceeded to
remove them in the period following World War II, although some still
remain as they do in other areas of the world. Interestingly, deliberate
sterilization through tubal ligation or vasectomies is n·ow widely used
in the United States and elsewhere as a means of contraception. Such
operations, however, did not affect the individual's sexual desires or
hormone production, and in no way could those involved be classed
as eunuchs.
Medical castration, known as orchidectomy, i.e. the removal of the male
testicles, is still utilized in some forms of cancer, such as prostate cancer, but
increasingly chemical castration instead of surgical castration is involved.
For a time also, large numbers of women who had a hysterectomy also had
an ovariectomy which, before the development of chemical hormones,
threw them into the menopause. It was also used by some as a way of
avoiding further pregnancies.
There is one further category of eunuch, which is in the process of
changing. This, for want of a better term, might be called the intersex
children. In the past, when the medical professionals encountered a baby
or child with ambiguous genitalia, they, usually with the support of the
parents and others, tried to intervene surgically, often in the process
creating eunuchs. Perhaps the most extreme example of this was the
case in the early 1960s of the Canadian twin brothers who underwent
circumcision by electric cauterisation when they were a few months old.
Unfortunately, a short in the cauterising gun caused the penis of one to
be burned off. The anxious parents, after consulting speciali~ts, decided to
raise the child as a girl, had the scrotum and testicles surgically removed,
and the child, nearly two years old by this time, changed legally from male
to female. Technically he was a eunuch. The assumption was that nurture
would win out over nature and that the child would develop normally
as a girl. Hormonal intervention would be required in the subteens and
teens, as would surgical remodelling of the labia et al. The case was much
reported in the scientific literature, particularly by John Money. He early
on stated that the child was developing normally as a girl, 5° and after much
early publicity, the case disappeared from sight. The child, however, did
not identifY as a female and in her teens rejected the girlish role assigned to
her by her parents and the medical community. After a series of traumatic
incidents in her late teens, she was finally told what had happened to
her. She then resumed her male persona, and is now regarded as a male, 51
although without a penis or testicles. '
The case, made public in the late 1990s, has become a powerful
instrument for the newly formed Intersex Society of America. The society
12
regards its role as not only preventing sex changes to children such as hap-
pened to the Canadian twin, but as opposing drastic surgical intervention
in the case of infants and children who are born with ambiguous genitalia.
With some exceptions, they insist such decisions cannot be made without
the consent of the individual involved, since the result of such surgery is
the creation of a new kind of eunuch. Obviously some corrective surgery
can be done, but if a child has an over-enlarged clitoris or a micro penis,
the recommendation is to leave it alone unless the child or adolescent later
wants some correction. Some of the advocates for the Intersex Society
want to impose a moratorium on any kind of'corrective surgery' involving
genitalia unless and until the medical profession completes comprehensive
look-back studies and finds that the outcomes of past interventions have
been uniformly positive. They also advocate that efforts be made to undo
the effect of past deception by physicians over intersex surgery. 52
The issue is obviously complicated, but the existence of such a society
and its widespread public support, emphasizes that there is a general
unwillingness to create a new eunuch class. More surprisingly, within the
past decade, those who regard themselves as transsexuals or transgendered
persons have expressed a growing reluctance to undergo major genital
surgery (especially among the females-to-males), but instead rely more
heavily on hormones and only simple castration (removal of testes) or
ovariectomy, to bring about the desired transformation. It seems clear
that, although technically there are only two anatomical sexes, male (penis
and testes) and female (vagina, uterus and ovaries), within these two
categories there are large variations in genitalia, even missing segments, as
well as in gender behaviours and in sexual preference or identification. We
have established transsexualism as a legitimate diagnosis, and developed
treatments for individuals who earlier might have chosen to be eunuchs.
They now can change their sex with surgery, hormones and legal papers.
In conclusion, what once required surgery, and was almost overwhelm-
ingly restricted to males, can now be done with hormones, but none of
the new generation have shown an inclination to be called a eunuch. At
the same time, the functions that the eunuch once performed, can be done
as well by others. There is still legal punishment of castration for some
but even this is under attack. Any one who tried to demonstrate his or
her power by cutting off the testicles or penis of a male, or removing
the ovaries of a female or even more radical surgery, would, I think,
be subject to almost universal vilification. In our increasingly sexually
egalitarian society, no one needs to undergo surgery to take on the roles
and tasks which were once limited to one sex or another. We will still have
voluntary sterilization as a means of family planning, but the emphasis is
13
Vern L. Bullough
on 'voluntary', and neither tubal ligation nor vasectomy have an effect on

hormone production or for that matter physical appearance. Females have
the capacity to become pregnant and give birth and males to make sperm,
but in today's world it is not clear that all who act and live as women are
necessarily female nor all who identifY and act as men necessarily male. We
might well have people who for religious or other reasons continue to be
castrated or have ovariectomies, but, if so, their actions should be regarded
as a voluntary decision since classification today is strictly voluntary and
eunuchs are no longer a special class, if they ever were.
Notes
1 Stent 1877. See also Matignon 1936.
2 Nanda 1999, 26-9.
3 See Pellat 1978, 1087-92.

4 Hymns ofthe Artharva-Veda 3.9, 6.133.
5 Ath. 12.515.
6 Graham 1951, 260.
7 Graham 1951, 422-4.

8 Bullough 1995, 77.
9 Asher-Greve 1998.
10 Amm. Marc. 16.6.17.
11 On Potiphar and Joseph see also the contribution of Abusch in this volume.
12 Knox 1948.
13 Holy Bible. New Revised Standard Version, Nashville, Tenn. 1990.
14 See the discussion in Kadish 1969.

15 Consider for instance the Mameluks of Egypt (e.g. Hitti 1970, 671-82)
who were recruited from slaves and whose organization and adhesion, and the
fear of them by other men, might indicate that many of the dominant leaders
were eunuchs. Hitti does not indicate that they were castrated, only that they
had been slaves. Perhaps not all of them were castrated, but the way succession
was carried out and their alliances suggested that most probably had been. This
is, however, debatable.
16 Xen. Cyr. 7.5.62-3, tr. W. Miller, London and New York 1914, vol. 2,
289. For Greek attitudes towards castration see the contribution of Bardel in
this volume.
17 Hdt. 8.105-6.
18 For Achaemenid eunuchs see the contribution of Llewellyn-Jones in this
volume.
19 Breasted 1962, vol. 3, 588, 248.
20 Pritchard 1955, 180-97.
21 Report ofthe Subcommittee on Sex Crimes, Sacramento CA 1952,47.
22 Kemper 1990, 38-65.
23 Osgood 1951, 146.
14
24 Bullough 1976, passim. Consult 'eunuchs' or 'castration' in the index.

25 See for instance the contribution by Tsai in this volume.
26 On eunuchs as singers generally see the contribution of Witt in this
volume.
27 1 Corinthians 14.34-6.
28 Heriot 1974.
29 Brundage 1987, 83.

3° Franca 1959, 96, 100-23.
31 See also the contributions of Hales, Lightfoot and Stevenson in this volume.
32 Clem. Al. Strom. 13 (91). .
33 Matt. 9.11-12.
34 Euseb. Hist. eccl. 6.8. See also Stevenson in this volume.
35 Bailey 1959, 72 n.11.
36 See the contributions of Gaul, Sideris and Tougher in this volume.
37 On Theophylact and his text see the contribution of Mullett in this volume.
38 Bhandarkar 1965, 86-7.
39 Nanda 1999.
40 Josephson 1959, 9-11, 61.
41 Steeves 1983, 171-5.
42 Much of the literature about them in western languages is in French (Rap-
paport 1937 and 1948; Pittard 1934), but see now Engelstein 1999. See also
Tomkins 1962, 156-64. Some of his claims, such as the eunuchism of Georgi M.
Malenkov, premier of the USSR from 1953 to 1955, I am reluctant to accept.
43 This quote is taken from Cheney 1995, 182, who cites V. Soukhomline, Les
proces de la secte mystique des Skoptzy, les proces celebres de la Russie, Paris 1937, as
the source, but I have been unable to locate this book.
44 Suet. Nero 28-9.
45 Tr. Arrowsmith 1987, 4.23, 36. There are many versions of the Satyricon,
many of them censored; Arrowsmith's is one of the better ones.
46 Mitamura 1970, 64 and passim.
47 Bullough and Bullough 1993.
48 For a good overview of this see Gould 1981.
49 Interestingly, vasectomies were originally advocated as a way of increasing
male potency.
50 Money and Erhardt 1972, 46-51; Money 1994 and 1975.
51 Diamond 1997.
52 Kipnis and Diamond 1998.
Bibliography
Arrowsmith, W
1987 Satyricon, New York.
Asher-Greve, J.M.
1998 'The essential body: Mesopotamian conceptions of the gendered body',
15
vern L. Bullough
in M. Wyke (ed.) Gender and the Body in the Ancient Mediterranean,

Oxford and Malden, 8-37.
Bailey, D.S.
1959 Sexual Relation in Christian Thought, New York.
Bhandarkar, R. G.
1965 Vaisnavism, Saivism and Minor Religious Systems, Varansi.
Breasted, J.H.
1962 Ancient Records ofEgypt, 5 vols., New York.
Brundage, J.
1987 Law, Sex, and Christian Society in Medieval Europe, Chicago.
Bullough, V.L.
1976 Sexual Variance in Society and History, Chicago.
1995 Sexual Attitudes: Myths and realities, Buffalo.
Bullough, V.L. and Bullough, B.
1993 Cross Dressing, Sex, and Gender, Philadelphia.
Cheney, V.T.
1995 A BriefHistory ofCastration, Ft. Walton Beach, Fla.
Diamond, M.
1997 'Sexual identity and sexual orientation in children with traumatized or
ambiguous genitalia', journal ofSex Research 34:2, 199-211.
Engelstein, L.
London.
Franca, I.
1959 Manual ofBel Canto, New York.
Gould, S.J.
1981 · The Mismeasure ofMan, New York.
Graham, H.
1951 Eternal Eve: The history ofgynaecology and obstetrics, Ne~ York.
Heriot, A.
1974 The Castrati in Opera, New York.
Hitti, P.K.
1970 History ofthe Arabs, 1Oth edn, London.
Josephson, E.M.
1959 The Unheeded Teachings ofjesus or Christ Rejected, New York.
Kadish, G.E.
1969 'Eunuchs in ancient Egypt?', Studies in Ancient Oriental Civilization
35, 55-62.
Kemper, T.D.
1990 Social Structure and Testosterone, New Brunswick, N.J.
Kipnis, K. and Diamond, M.
1998 'Pediatric ethics and the surgical assignment of sex', journal of Clinical
Ethics 9 (4), 398-410.
Knox, R.
1948 The Vulgate, New York.
16
Matignon, J.J.
1936 La Chine hermetique: superstition, crime, et misere, Paris.
Mitamura, T.
1970 Chinese Eunuchs, tr. C.A. Pomeroy, Rutland and Tokyo.
Money,J.
1975 'Ablatio penis: normal male infant sex-reassignment as a girl', Archives
ofSex Behavior 4, 65-71.
1994 Sex Errors of the Body and Related Syndromes: A guide to counselling
children, adolescents and their families, 2nd edn, Baltimore, Md.
Money, J. and Ehrhardt, A.A.
1972 Man and Woman/Boy and Girl, Baltimore, Md.
Murray, S.O.
2000 Homosexualities, Chicago.
Nanda, S.
Osgood, C.
1951 The Koream and their Culture, New York.
Pellat, CH.
1978 'Khasi. I - in central Islamic lands', in The Encyclopedia of Islam, New
Edition, vol. 4, Leiden, 1087-92.
Pittard, E.
1934 La castration chez l'homme et les modifications morphologiques qu'elle
entrafne, Paris.
Pritchard, J .R.
1955 Ancient Near Eastern Texts, Princeton.
Rappaport, I.E
1937 La castration rituelle: l'itat mental des Skoptzy, Ph.D. thesis University
of Paris.
1948 Introduction aIa psychopathologie collective: Ia secte mystique des Skoptzy,
Paris.
Scholz, P.O.
2001 Eunuchs and the Castrati. Translated from the German by J.A. Broadwin
and S.L. Frisch, Princeton N.J.
Steeves, P.D.
1983 'Skoptsy', in J. Wieczynski (ed.) Modern Encyclopedia of Russian and
Soviet History, vol. 35, 171-5.
Stent, G.C.
1877 'Chinese eunuchs' ,journal ofthe North-China Branch ofthe RoyalAsiatic
Society 9, 143-84.
Tomkins, P.
1962 The Eunuch and the Virgin, New York.
17
2
EUNUCHS AND THE ROYAL HAREM

IN ACHAEMENID PERSIA (559-331 Bc) 1
Lloyd Llewellyn-jones
In the 1961 Hollywood B-movie Esther and the King, the youthful and
glamorous Joan Collins (an unlikely Jewish heroine), together with a bevy
of nubile beauties, is escorted into a glittering harem, hung with tasselled
drapes and chiffon curtains. The colourful set represents the women's
quarters in the palace of the Persian king at Susa. The crowd of girls are
overawed at the beauty of their new home, and they giggle in wonder at
their good fortune. The noise of female laughter rises to a crescendo when
suddenly an imperious male figure emerges from a curtained entranceway
and calls for silence. He is Hegai (Robert Buchanan), the Overseer of the
Harem, and he holds sway over the court of women. He is dressed like
a peacock in gold, blue, and green robes; a turban is coiled around his
head; his chubby fingers are covered in jewels and his heavy hand clutches
a staff of office. Hegai's face is fat and smooth and his eyes are lined with
black make-up. He commands authority. Beating his staff on the ground,
he calls for silence and views his new charges with the snobbish superiority
of a grammar school headmistress. He immediately begins instructing
the 'gals' in court etiquette. He softly waves his hands and sings out his
commands: 'Bow low. Heads bent. Eyes down ... Hold your prostration.
One. Two. Three. Now rise. Slowly, slowly.' The king enters as the women
sink back onto their knees. They are scrutinized by their royal master,
Ahasuerus, the Great King, the King of Persia (Richard Egan) as Hegai
keeps watch (Fig. 1).
Instructing the women in the manner of a woman, this eunuch of
Hollywood imagination would have been easily recognizable as such to
the likes of such ancient authors and historians as Herodotos, Euripides,
Sophocles, Xenophon, Ktesias and Plutarch who, on the whole, construct
their eunuch characters along a similar line. Hegai lives in a world inhabited
by women; as of an indiscriminate sex, neither entirely man nor woman
but eunuch, he is granted special access to the women belonging to his
19
Fig. 1. Still from Esther and the King, 1961. Courtesy of BFI Posters and Stills.
master, the Persian king, whom, the film assures us, he serves with devotion
and unfailing loyalty. Hegai is the perfect product of a harem society.
Expostulating on the perverseness of these un-gendered creatures was
a regular feature of travel writing, dramas, and romantic (or erotic) novels
of the eighteenth and nineteenth centuries. Writing in the mid-1850s,
for example, the French author Gustave Flaubert set his exotic novel
Salammbo amid the barbarian hordes of mercenaries inhabiting the city of
Carthage at the time of the First Punic War. Flaubert's Carthage is a highly
sexed world of vivid passions, and his story of the priestess Salammbo's
passion for the soldier Matho is punctuated by many colourful depictions
of 'local' life, much of it violent and erotic. Flaubert associates his sex-
kitten priestess with a group of eunuchs; he writes:
All of a sudden lights appeared on the topmost terrace of the palace, the
middle door opened and [Salammbo] ... dressed in black, appeared on the
threshold ... Behind her on each side stood two long lines of pallid men,
wearing white robes with red fringes falling right down to their feet. They
did not have beards, or hair, or eyebrows; in their hands, sparkling with
rings, they carried immense lyres and they were all singing, in shrill voices,
a hymn to the divinity of Carthage. They were the eunuch priests of the
temple ofTanit. 2
20
Eunuchs and the royal harem in Achaemenid Persia (559-331 Be)
The eunuch priests add colour to Flaubert's Orientalist vision of

the past, although Flaubert may well have had first-hand knowledge of
eunuchs, as he possibly saw some on his visit to Egypt only a few years
prior to the publication of Salammbo. 3 Certainly, eunuchs were still to be
found in the Near and Middle East in the nineteenth century, and well
into the opening decades of the twentieth century too. However, although
it was calculated that no less than 8,000 eunuchs a year were imported
into Arabia, Turkey and Egypt from other countries of the Middle East
in 1890, by the 1930s N.M. Penzer, who made a pioneering study of
eunuchs and the Ottoman harem, was only able to trace one or two
of these 'strange beings' (as he termed them) in the whole of Turkey. 4
Osbert Sitwell fared better in China during a visit between 1933 and
1934, spending an afternoon drinking tea with twenty eunuchs who had
formerly served in the Imperial Palace of the Forbidden City, but were
now seeing out their days in a Refuge For Distressed Eunuchs next-door
to the fashionable Pa Pao Shan Golf Club in Shanghai. 5
So eunuchs were not merely European fantasies conjured up out
of the pages of the Arabian Nights, but a real phenomenon of eastern
life, and they had been so for millennia. They had been found in the
homes, harems, and courts of powerful men throughout the East; eunuchs
were present at the seats of power in the late Roman Empire, Christian
Byzantium and the Ottoman Empire as well as Imperial China. 6 They
were also to be found as powerful influences at the court of Assyria in
the twelfth to seventh centuries BC and, most interesting for this study,
at the court of the Achaemenid kings of Persia in the fifth and fourth
centuries BC. Here they served in the public and private apartments of
the kings and, it would appear, they were also in service to the women
of the royal family. However, not all scholars have been eager to mention
the importance of eunuchs in the Achaemenid court system, and some
veer away from making the eunuch-harem connection; most puzzling
is Maria Brosius' methodical ostracism of eunuchs from her important
study Women in Ancient Persia? However, in his recent monumental
study Histoire de !'empire Perse, Pierre Briant devotes a whole section
of his chapter on Achaemenid court life to eunuchs and interweaves
court eunuchs throughout the main thrust of his historical narrative. 8
Moreover, Briant links the Achaemenid court eunuchs with a special
function, namely serving the court of women within the royal harem. It is
the relationship fostered between the Achaemenid royal women and their
eunuchs that is the chief focus of this chapter; it is a relationship bound
up with the problematic arena of the 'royal harem'.
21
Eunuchs and the harem: two complex terms to explain, both in

the public and academic minds. Yet in scholarly and popular imagina-
tion eunuchs and the harem are invariably linked. It has already been
demonstrated how twentieth-century film-makers have assimilated the
two and have catered for popular taste, but earlier authors and dramatists,
such as Shakespeare, have also indulged and exploited the popular eunuch-
harem fantasy; 9 scholars have not been far behind. Penzer's 1936 classic
study of the Ottoman Imperial Harem devotes much space to the eunuch-
harem connection, as does Leslie Peirce's recent and more scholarly account
of the same subject. 10
By drawing on the ancient sources, I want to try to define the role of
the harem in Persian society, and the use made of eunuchs within the
institution. Where possible use will be made of as much Persian and earlier
Near Eastern evidence as is available, but this area of investigation has
severe limitations. The Persian material is not particularly forthcoming
with information on this facet of royal life (women and servants, especially
castrated slaves, too often fade into the background in the androcentric
Persian royal texts), although earlier Assyrian documentation does throw
some remarkable light on harem organization. Many scholars believe
that many of the royal institutions already well established under the
Assyrians were readily adopted by the Achaemenid rulers of Persia who
often saw themselves as the .inheritors of Assyrian rule; eunuchism was
one such institution. 11
Much of our evidence for the Persian harem and for the use made of
eunuchs within the institution stems, however, from Greek sources. This
is a frustrating reality for almost any study of Achaemenid history, but
especially so for an investigation of court life and the role of women at
the Achaemenid court. As Maria Brosius has demonstrated, the bona
fide Persian sources (such as the Persepolis Fortification Texts and the
Persepolis Treasury Texts) tell us little about the day-to-day habits of the
royal ladies, although they do confirm the importance of royal and non-
royal women in the economic life of Persia. 12 But the texts take us little
further than that, and they certainly do not inform us of any association
between a royal woman and a eunuch.
The works of the classical Greek authors Xenophon, Ktesias, and Dinon,
on the other hand, regularly attest to the use of eunuchs by Persian women,
while later Hellenistic and Roman writers, in particular Plutarch and the
unknown author(s) of the Hebrew text and the two Greek versions of the
biblical Book ofEsther, also provide valuable information:B The emphasis
of the Greek sources is always on the power of the king at court, the nobles
who stand close to him, the women of his harem, and his eunuch slaves. It
22
is thus the royal harem that becomes the battleground of the empire, where
great men rise and fall, manipulated, however, by women and eunuchs.
There is no doubt that the ancient Greeks were intrigued, puzzled, and
fascinated by the notion of eunuchism. In fact, the Greek language had
over two-dozen words for 'eunuch' and even used the figure of the eunuch
in children's riddles and rhymes. 14 However, in regard to their particular
fascination for the court eunuch, Edith Hall explains that:
The palace eunuch of the Greeks' imagination encapsulates their systematic.
feminization of Asia; emotional, wily, subservient, luxurious, and emascu-
lated, he embodies simultaneously all the various threads in the fabric of
their orientalist discourse. 15
The Greek sources are, of course, heavily biased, and the classical Greek
evidence in particular likes to portray the Persian court as a feminized
hotbed of decadence, but in the absence of Persian information proper
these sources are all we have and need t.o be used, albeit with due care
and attention. What follows here is an overview of the role of eunuchs in
the Achaemenid royal harem that combines both Near Eastern (Assyrian,
Persian, Hebrew) evidence with that of the Greek authors.
As long as the student of Achaemenid history takes into consideration
the pitfalls of using the Greek authors, their evidence can be of enormous
value, while comparisons with other ancient societies and their customs
(both near to and further removed from the Achaemenid period) can also
be informative. Therefore, it is possible to use the examples of the gendered
structures of the Hellenistic courts of the Near East, Asia Minor, and
Egypt to support the classical Greek and Persian evidence. The royal courts
of Hellenistic Pontus, Syria and Egypt can be valuable for comparative
evidence, since many of these courts consciously carried on traditions
directly associated with Achaemenid royal practice, or else assimilated the
court practices adopted by Alexander the Great after his conquest of Persia
and his keen adoption of Achaemenid royal protocol. The Hellenistic
royal reliance upon the court practices of the Achaemenid monarchy
is certainly stressed by Inge Nielsen in her study of Hellenistic royal
palaces; she suggests that to a large extent, the early Hellenistic palaces
dotted throughout the Greek world in the third and second centuries BC
(especially those of the Seleucids) were largely based on Achaemenid
models. 16 It might be reasonable to suppose that the organization of the
palace rooms, and the functions and routines carried out within them,
also contained a semblance of Persian court lifestyle.
Of the two subjects encountered in this chapter, 'eunuch' is relatively
easy to define: basically, a eunuch is a castrated man, one often found
23
serving at the courts and palaces of kings and nobles and frequently linked
with the safeguarding and serving of women. In the artistic sources, the
Persian eunuchs of the royal ceremonial city ofPersepolis are all depicted as
slim, elegant courtiers, with no hint of the corpulence or physical strength
that is apparent in depictions of court eunuchs in earlier Assyrian art. 17
The bodies of the Persepolis eunuchs are idealized in an 'Egyptianizing'
technique actively adopted by the Persian kings and their artists. For
example, a relief carved onto a door jamb of the palace of Darius I at
Persepolis possibly shows a eunuch standing ready with a towel and
perfume bottle to minister to the needs of his sovereign. He is finely
and elegantly dressed in a long wide-sleeved robe; his hair is twisted into
tight curls and caught into a small chignon at the nape of his neck; a fillet
or metal headband crowns his head. The eunuch's face is smooth, clean-
shaven and youthful; his eyes are wide and alert and his mouth extends
into a serene smile. His presence on the door jamb confirms the idea
that the rooms beyond this doorway were given over for the private uses
of the king and his immediate family. 18 All in all, some six eunuchs are
depicted in the Persepolis reliefs, often (but not always) accompanied by
a bearded official who is no doubt meant to contrast with the eunuchs'
smooth effeminacy. 19 The eunuchs frequently carry towels, ointment jars,
parasols, and fly-whisks, all of which allude to their roles within the close
entourage of the king and, perhaps, the women of the harem. 20
The Near Eastern terminology for eunuch is difficult to determine,
however. The Akkadian term for eunuch is fa ref farri (or simply fa reft),
while the corresponding Hebrew loan word is saris, which s,imply means
'one who [stands] at the head of the king'. Eduard Meyer argued that the
Hebrew word tirshatha specifically designated a palace eunuch, derived
as it was from the New Persian verb tara!, 'to cut' .zr Certainly the term
indicates a personage of high rank, since tirshatha can also be translated as
'the one to be feared or respected', that is, 'Excellency' from the Old Persian
word tarsa. According to the Old Testament, the Jewish eunuch Nehemiah
was appointed as the 'cupbearer' (Hebrew mafqeh) of Artaxerxes J.22 This is
a customary role for eunuchs attested in the Greek sources too: Xenophon
speaks of 'cupbearers' to the Persian kings in his Cyropaedia (1.3.9) and
uses the word oivoxoot. It is possible that the Greek terms oivoxoo~ and
euvoiixo~ were somehow interchangeable.
The word 'harem' is particularly problematic, both in its ancient
terminological usage and in its wider contemporary popular conception.
Influenced by vague notions of the seraglio, we have a tendency to imagine
women as shut away inside palaces, out of the sight of men (but not
necessarily out of harm's way), having as their only link to the world
24
Eunuchs and the royal harem inAchaemenid Persia (559-331 BC)
beyond the palace walls the half-men who served them and occasionally
'serviced' them. The image of a Turkish-style harem, a secluded and closely
guarded pleasure-palace filled with scantily dad, nubile concubines idling
away their days in languid preparation for nights of sexual adventure in
a sultan's bed, has become an integral part of the West's fascination with
the mysterious East. 23 This allure finds its most vivid expression in the
decadent nineteenth-century Orientalist paintings oflngres, Gerome, and
Corman, and in popular Hollywood movies of the 1940s and 1950s. 24
Most inappropriately, it is this vision of a haven of Oriental sensuality
and secret sexual delights practised by scheming, if imprisoned, women
and their spooky eunuch companions, which has heavily influenced our
interpretation of the evidence for and against the role of the harem in
many eastern societies, including that of Achaemenid Persia. 25
In dealing with the notion of what the harem is, we must first come to
terms with its definition; we need to realize that while it can be a physical
space, an· identifiable area of a palace or house used by women and eunuchs
and privileged men, a harem can also simply refer to women grouped
together; it does not necessarily need a defining space. More than that,
however, it is a term used in a metaphysical sense to describe something
which is out of bounds. The word 'harem' has at its core the Arabic 'ha'ram'
meaning 'forbidden', or 'taboo'. By implication it means a space into
which general access is forbidden (or limited) and in which the presence
of certain individuals or certain types of behaviour are forbidden. 26 The
fact that the private quarters in a domestic residence, and by extension its
female occupants, are also referred to as a 'harem' comes from the Islamic
practice of restricting access to these quarters, especially to males unrelated
to the resident females. The word 'harem' is therefore a term of respect,
evoking religious purity and personal honour, indeed in Muslim thought,
every man and woman carries with them an inner ha'ram, a self-awareness
of not overstepping personal boundaries.
This is particularly relevant to the Ottoman concept of the harem: while
not divine himself, an Ottoman sultan, 'God's shadow on earth', created
a sacred space around his physical presence, and because the sultan lived
within the harem, the physical space itself took on an added dimension
of inviolability. In many ways an Achaemenid Persian monarch resembled
the Ottoman ruler: he too was the sole representative of God on earth,
the all-powerful Ahuramazda, and, as such, he stressed his removal from
ordinary mortals by physically confining himself within the domestic
areas of his palace, away from the gaze of his subjects except for periods
of formal public audience. Thus I would contend that the term 'harem' is
totally applicable to Achaemenid royal domestic practice.
25
However, many western scholars, seduced or repelled by the notions

of hedonism and decadence encapsulated within the term 'harem', refuse
(or at the very least find it difficult) to treat the notion of 'the harem'
seriously. It might be more profitable then to couch the word 'harem'
behind some other idiom with less negative associations; in that case,
I propose that 'domestic quarters' or (better yet) 'inner court' stand in
its place as a substitute. Even though the word 'harem' expresses more
dearly the notion of what is 'taboo' or 'forbidden', the phrase 'inner court'
is an adequate alternative, because it has connotations of that which is
hidden, or that which is not easily and routinely seen or entered into. The
Imperial Ottoman palace was divided into two halves, the public world
of the 'outer court', a public place of business, spectacle, and ceremony,
and the 'inner court', a domestic arena, private, inward-looking. In the
'outer court' the Sultan met with his ministers, generals, ambassadors,
and subjects, but in the inner court he was surrounded by his family, his
mother, wives, unmarried sisters, his children, and, of course, his eunuchs.
It is possible to see Achaemenid royal practice following the same kind
of principle: the king in his role of governor of his people, commander
of his army, and representative of Ahuramazda played out a public role
within the public halls of his palaces. In a complex series of court rituals
centred in and around the vast audience chambers and terraces of the
royal citadels, the king was on display. 27 But in his private moments the
king was a shadowy figure, removed even from his chief ministers. His
private hours were spent in the 'inner court', in his domestic realm, with
the eunuchs, children, and women of the royal harem.
However, Brosius has argued that it is difficult to trace the notion of
an 'inner court', a harem, in the ancient sources on Persian women. She
states, 'It is dear ... that there is no truth in suggestions that women lived
in seclusion and were confined to the palace.' 28 This opinion is in sharp
contrast to Richard Frye's estimation that:
the harems of the Achaemenid rulers were large, for they contained not only
the wives and concubines of the king but all the women of the family, such
as sisters, mothers and others ... We do not hear of an organized harem with
many eunuchs until.. .after Darius [1]. 29
I suggest that there was such an institution as the royal harem and
that the concept of an identifiable society of women was fundamental to
both the Persian and the Greek views of court life. It is difficult to know
how the ancient Persians referred to a harem, but Frye suggests that the
Old Persian word tachara means the private quarters of the king and his
family, and his servants. 30 Greek texts suggest that there was a specific
space within the royal palace for the women and their eunuch slaves:
26
Eunuchs and the royal harem in Achaemenid Persia {559-331 BC)
Herodotos makes it manifestly dear that the male apartments of the palace
(avopEoov, 3.77-8) are separate from those reserved for the women (3.68).
The existence of separate apartments is also made explicit in his history
as he recounts the arrival of the Greek doctor Demokedes at court: he
is escorted by a group of eunuchs into the chambers of the royal ladies
(3.130, 1tapa 1:a~ £oomou yuvatKa~)Y Herodotos also notes that before
the age of five a young prince 'lives with the women and never sees his
father' (1tapa 'tiJ<n yuvm~t, 1.136). He also notes that it was the privilege
of the king's seven councillors to be admitted into the domestic heart of the
palace as long as the king was not sharing the company of his wife (3.84.2;
3.118.2). Plutarch (Them. 26) tries to account for the Persian reluctance
to allow their women to be seen in public, arguing that:
As a rule, the barbarian peoples are excessively jealous of their wives, and the
Persians outdo all others in this respect. Not only their wives, but also the
female slaves and concubines are rigorously watched, and no strange eye is
allowed to see them. They live locked up in their rooms, and if they have to
travel, they do so in carriages hung on all sides with draperies.
The Septuagint version of Esther tells of 'tijv auA,Tjv 'tTJV yuvatKEiav
(2.11; translated in the 1851 English version fittingly as 'the court of
women') and notes, in fact, that there were two such courts (see below); at
the opening of the story, Queen Vashti (or Astin, as she is named by the
Greek author) is holding a feast for the court ladies in her apartments, while
the king and his nobles dine in another area of the palace (1.9). The term
yuvatKOOVtn~ is also used by Plutarch when he tells of the living-quarters of
the concubines of the satrap of Sardis (Them. 31.2), and numerous other
Greek writers use the notion of a court of women to cast aspersions upon
the masculinity of Persian monarchs, brought up as they were surrounded
by females and castrated effeminates. As Plato states (Lg. 695 a-b):
[Cyrus] didn't notice that women and eunuchs had given his sons the
education of a Mede, and that it had been debased by their so-called 'blessed'
status. That is why Cyrus' children turned out as children naturally do when
their teachers have never corrected them. 32
An identifiable society of women and eunuchs is a notion endorsed
by Heleen Sancisi-Weerdenburg in her analysis of Ktesias' observation of
the Persian court. She notes that Ktesias may have had an ulterior motive
for portraying the role of Persian royal women in a particular way, but
in doing so she highlights his recognition that there was an identifiable
society of court women which was served by a staff of eunuchs who often
wielded considerable powers. In effect, Ktesias was studying, recording
and, to some extent, scandalizing the goings on of a harem society. 33
27
David Lewis, following A.T. Olmstead's example, took a far more

positive view of Ktesias as a major (and reliable) source for Persian court
history, and was of the opinion that the Persian harem was a highly
influential institution and that ancient reports of its influence on monarchs
should be taken seriously. As he noted:
It is customary to make fun of Ktesias when he attributes great importance
to eunuchs and queens and to say that his point of view is bound to the
harem ... I am myself disposed to take seriously stories of the irrational
caprice and wanton cruelty of monarchs. Nothing is reported of Periander,
tyrant of Corinth, which does not find ready parallels in well-attested
information about Ali Pasha of Iannina at the beginning of the nineteenth
century, and, allowing for some differences of institutions, the Persian court
will be subject to the same kind of pressures and insecurities which have
afflicted the courts of absolute monarchs down to the time of Stalin and
perhaps beyond. 34
Lewis goes on to note that, 'It seems unlikely that Greek fascination
with the Queens of Persia is solely due to the fact that they were hidden in
harems.' 35 But the harem is often mistakenly seen by ancient and modern
authors as a woman's world - domestic and private and cut-off from
the outside - and any attempt to influence events beyond its confines is
regarded as 'meddling' in a situation in which women have no right to
participate. Indeed, the idea of 'meddling women' is often put forward as
a genuine reason for the decline of the Persian empire. 36 But we need to
emphasize that an institution like the despotic Achaemenid monarchy was,
essentially, a family affair. This does not mean that this family-orientated
system was non-political; the royal family, self-contained within the inner
domestic quarters of a palace, the 'harem', was in fact the heart of political
decision-making. 37
From later Ottoman sources we know that segregation of the sexes
created for women a society that developed its own hierarchy of authority
and the same was true, no doubt, of Achaemenid Persia. Women of
superior status, the matriarchal elders, held considerable power not
only over other women but also over the younger males within the
family, because the harem was also the setting for the private life of
some men, especially young princes. Furthermore, networking within the
harem provided women with information and sources of power useful to
their male relatives. 38 It is within this context that the eunuchs became
important, because the eunuchs' ability to enter into the private world of
the harem and operate in the outer courts of the public sphere made them
invaluable aids to kings and, more interestingly, to the sometimes powerful
royal women of the inner-court itsel£ They acted as an additional set of
28
eyes and ears and operated outside the confines of the domestic heart of
the palace, relaying official messages to and fro between the inner and
outer courts, while carrying unofficial court gossip, scandals, and secrets
along the same routes.
It would appear that only representatives of the seven leading Persian
families, as well as the king's mother and his principal wife, could appear
before the king without being summoned, while all others had to make
a formal request for an audience. Under fear of the death penalty, no one
except the king, his nearest relatives and his court eunuchs could see the
royal wives and concubines; interestingly, there is not a single female image
to be found among the numerous reliefs at Persepolis, a fact which alerts
us to the idea that the female figure, certainly that of a royal female, was not
for public view. This may well be a facet ofNear Eastern art in general, since
women are notoriously absent from Assyrian palace reliefs too. 39 Egyptian
royal iconography, on the other hand, frequently promotes the image of
the royal wife, mother, sister or daughter as central to the ideology of
divine kingship. Despite the fact that the Achaemenids routinely borrowed
many Egyptian art-techniques and themes for their own iconographic
propagandistic purposes, the prominent representation of women in public
sculpture or relief was not copied. 40 It appears that there was something
fundamental to the Persian psyche that required women to be invisible,
particularly in the public and/ or artistic sphere. The official Achaemenid
iconography of conquest and ceremony has no place for women. Plutarch
tells (Them. 26) that a wild jealousy was characteristic of the Persians,
not only in respect to their wives but also in regard to their concubines
and female slaves. Widely known is the biblical tale of the demotion of
Vashti, the Persian queen, who refused to appear in public before her royal
husband, King Ahasuerus, or Xerxes, since his commands contradicted the
law which stated that women should never appear before strangers (LXX
Esther 1.11-12). 41 The required separation of royal women from public
life is a theme taken up by Chariton in his Hellenistic novel Callirhoe:
Stateira, the wife of Artaxerxes II, is driven mad with jealousy because of
her husband's new Greek concubine, the heroine of the story. She resolves
to hold a beauty contest (a familiar literary topos in Greek writings on the
Persian court) 42 to decide who is the most beautiful woman in Asia, and
offers herself as a candidate. It is only then that she is reminded that, as the
king's wife, she cannot appear in public or show herself to others, and so
she chooses a noblewoman named Rhodogune to stand in for her in the
battle against the Greek beauty (5.3-4).
Maria Brosius has proved that Achaemenid royal women moved around
the empire with unrestricted freedom and could amass great wealth from
29
their personal estates. She sees the harem, therefore, as being an institution
concocted from the minds of Greek and later writers. Yet the sources state
that even when women travelled across the country they were hidden
within curtained carriages, and the one incident recorded by Plutarch
(Art. 5.6; c£ Mor. 173 ff.) that we have of Queen Stateira breaking with
convention as she pulls back the drapes to exhibit herself to the populace,
is a device he uses to show her popularity and to compare her affability
with the cold aloofness of her arch-enemy and mother-in-law, Parysatis. 43
Brosius fails to realize that a harem is not necessarily a definable space; its
perimeters can change, and even though a woman might travel beyond
the walls of any organized harem-complex, her sense of segregation and/or
seclusion can remain. I concur with Brosius completely that Persian royal
women could be powerful and influential figures, but I want to emphasize
that their separation from an overtly public life does not necessitate any
loss of personal or even political power.
So who made up the royal harem? Without doubt, the most influential
woman at court, and someone who would certainly have been housed near
the king, was his mother. Her influence at court was usually unrivalled;
we are told that the king frequently dined with his mother (sometimes
with his wife), while a servant (probably a eunuch) with his mouth tied,
so as not to defile the king with his breath, held a large fan above the royal
pair, to keep the flies away. Courtiers ate in another room, separated from
the king and his mother by a curtain. 44
In a policy such as that of the Achaemenids, where the empire was
considered to be the personal domain of the dynastic family, it was natural
that the important women within the royal family (and in particular the
queen mother) would assume legitimate roles of authority within, and
perhaps even without, the royal household and Brosius has stressed that
queen mothers took it upon themselves to guard the safety of the throne
and of the son who occupied it. 45 Thus, queen Parysatis, the Achaemenid
dowager queen par excellence, became entangled in the deaths of the
pretender Sogdianus who threatened Darius II's accession to the throne,
and of the eunuch Artaxares the Paphlagonian who later tried to overthrow
him. 46 Sometimes, however, the power displayed by a queen mother
stemmed from personal reasons, usually from a vendetta, and had nothing
to do with the security of the dynasty (as we shall see).
Next to the queen mother in status was the king's wife or wives, since
most Achaemenid kings were polygamous; Darius I, for example, married
at least six women who bore the title king's wife. These women would
certainly have been lodged in the harem where they undoubtedly reared
their children - princesses, princes, and future heirs - and there is also
30
evidence to suggest that the unmarried sisters of kings were housed in

the royal harem. The royal wives also had the privilege of dining with the
king, and, depending on their individual personalities, they too could hold
enormous influence over himY According to Herodotos, Atossa, the wife
of Darius I, was a powerful influence at court and held the particularly
important role of her husband's confidant. 48
More problematic are the king's concubines; the fourth-century Greek
sources speak of 360 concubines- one (almost) for each day of the year.
Darius III is said to have lost 329 of his when he fled from Alexander the
Great in Syria. 49 This number may seem incredible, but, as J.M. Cook
notes, when compared with later rulers of Persia, this seems 'a niggardly
quota' (the Suren, during the Safavid period (1501-1736), for example,
had 12,000 concubines). 50 All of these women, it seems, were housed near
to the king. They certainly made up the king's entourage as he moved from
one palace to another or went on military campaign. Curtius Rufus paints
a vivid picture of Darius III's household on the move (3.3.22-3):
[Following the king] at a distance of one stade came Sisigambis, the mother
of Darius, drawn in a carriage, and in another came his wife. A troop of
women attended the royal ladies on horseback. Then came the fifteen so-
called ~mamaxae' in which rode the Icing's children, their nurses and a herd
of eunuchs (who are not held in contempt by these peoples). Next came the
carriages of the 360 royal concubines.
It seems that the number of women making up the royal harem could
be well in excess of 400 at any given time. Added to this of course were
the numbers of slave girls and waiting-women found in attendance on
the royal ladies and, on top of that, we then find the ranks of the harem
swelled by the eunuchs; while not every individual woman would have
necessarily had a eunuch slave, it is well attested that powerful wives and
Queen Mothers were served by a large retinue of eunuch staff. The Queen
Mother Parysatis, for example, is linked to at least twelve eunuchs in the
sources, (only two, Artaxares and Bagapates/Mastabates, are named) but
she may have been served by many more.
Where were these women and eunuchs housed? What is the evidence
for the physical presence of harem buildings in the Persian palaces? Early
excavators of Achaemenid sites, such as Ernst Herzfeld, believed that the
ceremonial royal palace of Persepolis presented conclusive evidence of
a harem. 51 Schmitt believed that he had found archaeological evidence
for the numerous women residing at the Persian court, and followed
Herzfeld's identification of the southeast building of Xerxes' palace as the
royal harem. 52 Similarly, A.T. Olmstead declared that:
To the west of the final treasury building, and separated from it by a street,
31
was the harem which Xerxes completed for his imperious queen, Amestris.
Surrounded by the guard rooms of the watchful eunuchs was a tier of six
apartments to house the royal ladies. Each tier consisted of a tiny hall whose
roof was upheld by only four columns, and a bedroom so minute that even
with a single occupant the atmosphere would have been stifling. 53
Such an interpretation of the archaeological evidence is now unfashion-

able; it smacks, we are told, of Orientalism on the part of the early
archaeologists. Besides, tiny rooms and cramped conditions hardly cor-
respond to the image of Persian luxury found in our Greek sources.
Therefore, more recent interpretations of the tiny rooms located near
the treasury building have led them to be classified as mere storerooms,
a surplus storage area, and nothing more. 54
However, this reinterpretation of the Persepolis site seriously underes-
timates the use of palace space in antiquity. Evidence from New Kingdom
Egypt has revealed a harem suite attached to the palace of Amenhotep
III at Malkata, near Thebes, one similar in design to the harem-palace
at Medinet el-Gurob in Middle Egypt. The 1915 excavation reports
note that:
[the] collection of robing room, bath, and wardrobe rooms closely associated
with a throne room is the typical plan of the royal apartments; and this
show[s] that this dwelling was that of a personage little inferior to the king
himself... [perhaps] a very important royal lady. 55
In fact, there are eight small harem-suites located along the central
hall (throne room) of the Malkata palace-complex, enough space to
accommodate the principal royal ladies of the Egyptian court.
The disputed rooms on the Persepolis terrace are indeed very small, but
the fact that the rooms are located in the heart ofXerxes' bl1ilding suggests
that the rooms had a function which far exceeded mere storerooms. The
imperial Ottoman harem complex at the sixteenth-centuryTopkapi palace
in Istanbul, is noted for the cramped, if beautiful, conditions in which
all members of the harem lived. It would appear that high social standing
did not necessarily mean spacious living quarters.
In her catalogue of Persian palaces, Nielsen suggests that harems can
be found at Persepolis, Pasargadai, and the two palaces built at Susa by
Darius I and Artaxerxes !.5 6 We need to bear in mind, however, that lying
around Persepolis and the other Persian palaces were enormous cities
made of stone, mud-brick and tenting. Gardens and lakes graced these
vast layouts and it is probable that if larger harems of the king did exist
they were set at a distance from the ceremonial centres of the palaces. It is
in these sites that the larger harems were no doubt located. In fact, with
32
the exception of religious festivities such as the New Year celebrations,

there is no concrete evidence to suggest that even the kings regularly
resided in the palace of Persepolis, while we have firm evidence to show
that at Susa the residential quarters of the king and his court were built
away from the ceremonial palace in the so-called Ville RoyaleY When
residing at Persepolis during the religious festivities, perhaps the king was
accompanied only by the closest members of his family and their servants
who were therefore lodged in the small (tented?) harem quarters.
Given the very nature of polygamous court societies, it is possible that
the senior queens and princesses occupied palaces that were remote from
their rivals; 58 and we know that many of them had vast estates scattered
throughout the empire which were no doubt staffed by regiments of
eunuchs and other retainers. There was no love lost between some of these
women and disputes between the royal ladies, often supported and enacted
by their eunuchs, were familiar: Parysatis, the mother of Artaxerxes II,
is credited with poisoning her daughter-in-law Stateira, to give but one
example of the domestic disputes common at the Persian court. 59
Whatever the actual living conditions for the Persian royal women
may have been, it is clear that their ladies-in-waiting and eunuchs would
also have required accommodation. Eunuchs who had achieved positions
of supreme power at court would probably have been housed in close
proximity to the royal women, although whether there was any 'House
of Eunuchs' (that is to say, official housing or apartments), as we find in
Ottoman Turkish and Imperial Chinese palaces, is impossible to say. 60
There is no evidence to suggest that harem eunuchs lived in any kind
of regal splendour (although it is difficult to imagine some of the most
powerful court eunuchs whom we hear of in the sources living in penury),
but proximity to a queen and the female court was no doubt rewarding in
itself and compensated for any discomfort that may have existed.
It is also hard to tell how many eunuchs were employed at the Persian
court or within the royal harem at any one time, but even in the fragments
of Ktesias there are at least sixty. 61 During the reign of Cyrus the Great, at
the beginning of the Achaemenid dynasty, they probably only numbered
a few hundred. According to Herodotos (3.92), early in the reign of
Darius I the Persians were starting to take castrated boys from Babylon and
Assyria as tribute: 'From Babylonia, and the rest of Assyria, were drawn
a thousand talents of silver, and five hundred boy-eunuchs.' The custom of
importing eunuchs seems to have expanded from then on.
The number of eunuchs employed by the royal family probably grew
as the size of the family itself increased during its 230 year history.
Artaxerxes I, according to Ktesias, had eighteen sons by his wives and
33
concubines, but Artaxerxes II is said to have had 118, all of whom, as

immediate members of the royal family, would probably have required
a staff of eunuchs. 62 When we take into account the number of daughters
born to the monarch, then it is clear that the royal harems would have been
packed to bursting point. Herodotos notes wryly (7. 187) that:
as for the eunuchs, female cooks and concubines, no one could estimate their
number any more than the various pack-animals and Indian dogs which
followed the army. They were far too numerous to count.
It is natural to see the royal women as having their own staff of eunuchs
and other personnel; in fact according to Hellanikos, it was queen Atossa
herself who introduced eunuchs to Persia, 63 while late antique sources
claim that it was the legendary Assyrian queen Semiramis. 64 While the two
accounts are without doubt fictitious, what is interesting is the concept
of women as the originators of the custom of castrating men and placing
them- into service. The Greeks certainly linked Asian women, especially
royal women, closely with the creation or employment of eunuchs:
a fragment from Sophocles' tragedy Troilus recounts how the Trojan queen
Hekabe ordered the castration of her son's tutor, 65 and in Euripides' play
Orestes, the Phrygian slaves who accompany Helen ofTroy are no doubt
eunuchs too. One of them tells how he would fan Helen with a rounded
feather fan while he sang an Asian song (1430-4):
In Phrygian fashion, it chanced I was swaying
Beside Queen Helen the punkha fan:
On the cheeks of Helen its plumes were playing,
Through the tresses of Helen the breeze was straying,
k I chanted a strain barbarian. 66
Edith Hall has noted that the punkha, a rounded feather fan used
to keep Persian royalty cool, was frequently shown in Achaemenid art
deployed by a eunuch; it would make sense to speculate that if Euripides
knew about punkha-fanning, he may well have known who performed
it. 67 By far the best evidence that can be used to identifY this Phrygian
slave of Helen's as a eunuch is Orestes' own statement that he is, 'neither
woman, neither found in the ranks of men' (1528).
The natural connection and affinity between women and eunuchs
found in the ancient sources is simple to explain. Most cultures that
make the association regard both women and castrated men as imperfect
creatures and incomplete human specimens. The heart of the matter lies
in the absence of testicles: Aristotle claims that male testicles act like
loom weights on the vocal chords, which they stretch, and it is because
of the taut chords that men and bulls have low voices and great courage.
34
Eunuchs and women have high voices and no courage simply because
they have no balls. 68
Of course, another facet shared by eunuchs and women is their ready
sexual availability. Persian sources tell us nothing of the intimate relation-
ships between Persian monarchs and their women, so it is no surprise that
we know nothing of a king's sexual relationships with his castrated slaves.
Nevertheless, our Greek and Roman authors often hint at eunuchs' sexual
availability for their royal masters. 69 The sexual relationship between the
eunuch Bagoas and Alexander the Great is well attested in the sources and
familiar to scholarship,7° but Curtius Rufus (6.5.22) makes it clear that
a sexual relationship with a eunuch was just another aspect of Achaemenid
royal practice readily adopted by the Macedonian conqueror: 'Bagoas
[was] an exceptionally good-looking eunuch in the very flower of his
youth. Darius [III] had had a sexual relationship with him and presently
Alexander did too.'
Less well known is the story recounted by Aelian of the sexual passion
shown by king Artaxerxes for a eunuch named Tiradates, 'the most
handsome and attractive man in Asia' (VH 12.1). In the tale, the good-
looking eunuch has died and the king is plunged into deep despair until he
is introduced to a beautiful Greek courtesan named Aspasia, the daughter
of Hermotimos of Phocaea. Immediately Artaxerxes is able to discern
Tiradates' feminine features in Aspasia's striking face and his passion is
re-ignited; but not to such an extent that he can perform sex with her.
The king sends the courtesan into his bed-chamber and begins to dress
her in the late Tiradates' clothing (cr-r6A.TJ); he is struck by the woman's
resemblance to the eunuch. And it was in the eunuch's robes that Aspasia
continued to visit the king and grant him 'consolation' for his grief: a case
for psychoanalysis if there ever was one!
Like the majority of the women they served, most eunuchs found in
the ancient sources are nameless individuals whose identities are lost
to history. Those who are named and who are given even a semblance
of individuality are the eunuchs linked to kings and noblemen and to
important royal females: to wives, mothers and daughters of kings and,
occasionally, to high-ranking noblewomen. Ktesias routinely records the
names of the most important eunuchs to serve under successive Persian
monarchs. He states, for example (Ktesias 51): 'Okhos, or Darius, was
sole ruler. Three eunuchs were influential at his court, especially Artaxares,
secondly Artobarzanes, thirdly Athoos.'
But it is difficult to know if these, and other named eunuchs who are
linked so closely to the king, had a hand in harem organization, or were
considered too powerful and kept out of the harem. The Greek text of
35
Esther notes that Xerxes' seven chief eunuchs, Haman, Bazan, Tharra,
Boraze, Zantholtha, Abtaza, and Tharaba, share the king's confidence and
his dinner, but they are instructed by the king to enter into the harem
to fetch queen Vashti (Astin) (LXX Esther 1.10). The Assyrian evidence
suggests that certain eunuchs were specifically trained for harem service
and were permitted to attend on the palace women only with the direct
approval of the king himself, and only while guards were present. The
Middle Assyrian 'harem edicts' make it plain that should a eunuch
enter the harem without permission he is guilty 'of an offence', which
suggests that harem service was strictly regulated. The claustrophobic
and oppressive atmosphere of the Assyrian palace becomes apparent in
an edict dating to the reign ofTiglath-pileser III (Middle Assyrian Palace
Decree 102-12):
Either royal eunuchs (Ja-ref-farranu) or court attendants [... ] If a woman of
the palace sings or quarrels with another of her rank, and one of the royal
eunuchs ... stands listening, he will be beaten one hundred times [and] one
of his ears will be cut off. If a woman of the palace calls to a courtier while
her hips are bare [and not covered] with a loin-cloth ... [and if] he turns to
speak with her, he will be beaten one hundred times. If a courtier wishes to
speak with a woman of the palace, he may not approach closer to her than
seven paces. If someone violates this decree and the eunuch in charge of
the palace hears of it and does not punish him, the eunuch in charge of the
palace will bear the punishment ... The [eunuch] in front of the palace will
bear the responsibility for all offences.7 1
What is interesting about this harem edict is the evidence it provides

for rank among the harem eunuchs- those who are simple palace servants
who should see and hear nothing, those who patrol the corridors on
lookout, and the 'eunuch in front of the palace', who should see and
hear everything.
According to Frye, the official recorded in the Old Persian sources
bearing the title 'tacharapati' may have been the Achaemenid equivalent
of the Assyrian 'eunuch in front of the palace' and was responsible for the
upkeep and good order of the tachara, the harem. 72 But evidence derived
from the Hellenistic book of Esther (A text) suggests that the royal harem
was actually divided into smaller units: when queen Vashti refuses to
appear before her husband, the king calls for the most beautiful women
of the kingdom to be collected together in the royal harem at Susa, where
they are placed under the supervision of the eunuch Bougaios who is
titled 'the eunuch who guarded' (A text 1. 7) and the 'guardian of women'
(B text 2.8. In the B text he is called Gai; in the Hebrew t~xt he is named
Hegai). When she arrives at the Susa seraglio, Esther finds favour with
36
Eunuchs and the royal harem in Achaemenid Persia (559-331 BC)
Bougaios who treats her as a particular favourite, bestowing jewels and

serving women on her. Under his tutelage, the new harem girls (including
Esther) are pampered and preened in a twelve-month beauty treatment, in
what B.W Jones has called 'conspicuous consumption in the extreme' .73
For six months the eunuchs pound the hopeful girls into shape with the
aid of oil of myrrh, and for the last six months they continue their beauty
regime by instructing them in the arts of perfumery and cosmetics. After
twelve months of intimate eunuch attention (which speaks volumes about
how these servants were perceived to be sexually non-threatening), the
girls are ready to be admitted to the royal presence, but only if the king
calls on them by name (as he does in Esther's case).
The Hebrew text is more specific in its details than either of the Greek
texts: once a girl has been presented to the king and he has accepted her,
a girl passes from the house of the eunuch Hegai into the house of the
eunuch Shaashgaz (the Greek A text now calls him Gai), who is styled
'guardian of the concubines' (Esther 2.14). As she ascends the ladder of
success, so her title changes to the formal recognition of her place within
the harem as a royal concubine. The Greek and Hebrew versions of Esther
make it clear that royal harem society was stratified and that there was
obviously a harem hierarchy both among the women of the inner court
and among their guardians.
What were the duties of harem eunuchs? As was commonly the case
on the outside of the inner court, harem eunuchs were frequently used
as messengers: Esther has in her service several eunuchs, one of whom,
Hachratheus (LXX Esther 2.5 f£; in Hebrew he is named Hathcath), is
found carrying messages for her between the harem and the gate of the
palace and in Callirhoe the eunuch Artaxates (who is in all probability
based on the real-life figure of Artaxares the Paphlagonian) carries messages
from the favoured concubine to the king (6.7.5-13). In the opening
book of Esther, the seven favoured eunuchs are chosen by the king to
take his message to the queen who resides within the harem precisely
because they are eunuchs and are therefore able to move with ease between
the public areas of the palace and the private inner-quarters of the royal
household.
The genderless eunuch gained the privilege of being able to occupy
simultaneously the public world of male politics and the private world of
female power and became the vital connection that the Persian monarch
could utilize in communicating with his womenfolk who were, it would
appear, valuable sources of information on court intrigue and power
struggles. As has been stressed by Keith Hopkins (in a late Roman context
that works for Persia too), the constructed fluidity of movement from outer
37
court to inner court resulted in an increase in power wielded by ambitious

eunuchs, even within the harem, and the eventual rise of the phenomenon
of eunuch 'king-makers', a type of eunuch familiarly encountered in the
works of Ktesias and his fellow Greek historians.7 4
A5 intermediaries between the king and the royal ladies, and mouth-
pieces of harem networking, eunuchs are found accompanying the women
wherever they may be. They were probably present at royal hunts when
the harem attended on the king as he shot wild lions/5 and were certainly
located at the annual royal progress through the empire when the king
and his court moved from one royal residence to another or went on
campaign.76 In fact, eunuchs are frequently found accompanying the royal
ladies as they travel in convoy behind the army when the king leads his
people to war; many eunuchs are reported in the sources as being taken
captive by Greek victors.77
The intimacy that this close association between the women and
the harem eunuchs fostered can be testified by a long list of individual
eunuchs who, the literary tradition claims, loyally served the queens and
noblewomen; many eunuchs are connected with heroic deeds, and they are
often beaten for their loyalty, or give up their lives to save their mistresses,
and even die nobly beside them. Thus, Amytis, the wife of Cyrus the
Great, is aided and abetted by the eunuch Tibetheus, who secures the
throne for her son Cambyses but is horribly beaten in the process/ 8 while
in Xenophon's beautiful story of Panthea, the Lady of Susa, emphasis is
laid on the loyalty of her three eunuchs who commit suicide over the
corpse of their beloved mistress ( Cyrop. 7.3.15):
And when the eunuchs, three in number, beheld what had occurred, they
also, standing in the spot where she had ordered them to Sta!(d, drew their
daggers and drove them into their own breasts. And now even to this day
it is said that the monument of the eunuchs is still standing; and they
say the names of the husband and wife are inscribed above in Assyrian
letters; and below it is said, there are three slabs with the inscription, 'The
Mace-Bearers'. 79
The closest parallel to this self-sacrifice is located on the battlefield of

Cunaxa, where the eunuch Artapartes kills himself when he realizes his
master, Cyrus, has died. 80 Herodotos provides details of the interesting
story of the Persian nobleman Boges and the mass-suicide by sati of his
family: he burns to death on a huge funeral pyre well-stocked with the
bodies of his wife, children and servants, probably including eunuchs. 81
Despite the close affections that could develop between eunuchs and
the women they served, one thing that is conspicuously absent in the
sources is any hint of sexual scandal within the harem itself. Ktesias,
38
for example, does not usually hold back on any sordid details of harem
intrigue, but he is completely silent on the subject of sexual shenanigans
within the harem, which suggests either that he chose to ignore them
(unlikely) or that they were not a significant feature of life in the inner-
court. Later Ottoman evidence suggests that under the watchful eye of the
Queen Mother, the harem could have the atmosphere of a convent. 82 The
distinct lack of sex-scandals in the ancient sources perhaps supports the
idea that the royal harem was perceived as a real seat of power.
The harem was a safe haven for the monarchy, a place in which royal
power was fostered and grew, but it did have its dangers: it would be
wrong to give a picture of all eunuchs contentedly serving their mistresses,
since access to the secrets of the royal family and daily service to the
king and the royal women obviously fuelled regal ambitions in some
harem eunuchs, and the fragments of Ktesias are littered with instances
of eunuchs attempting to usurp power or overthrow kings. What is
interesting, though, is the fact that in the sources no royal woman is ever
found conspiring with over-ambitious eunuchs and, in fact, the texts
indicate that in the case of treason, eunuchs and the royal women of the
harem do not mix. One story in particular supports this idea, that of
the extraordinarily powerful eunuch Artaxares who served at the court of
Darius II. Ktesias (43a) notes:
The eunuch Artaxares, being influential at court, conspired against the king,
wishing to supplant him. He asked a woman to make him a false beard and
moustache so that he could look like a man, but she betrayed him.
We are not told the identity of the woman, but it is probably to be

inferred that she resides within the inner court, because we are forced
to wonder how else a court eunuch would get to meet any other kind
of woman. But we are not told her rank, nor are we given clues as to
whether she was a servant or a royal lady. As the result of their treasonable
activities, the sources tell how Persian harem eunuchs meet with hideous
torture and die horribly; but what is interesting about these executions is
the fact that they are usually enacted at the command of the royal women,
which suggests that treasonable eunuchs who had abused their position
within the harem were accountable to their mistresses and that it was the
royal women and not the king who doled out their fates. Thus Artaxares
was killed on the direct orders of queen Parysatis: 'He was arrested, handed
over to Parysatis and executed', says Ktesias (43a).
Similarly, the powerful eunuch Petasakes was blinded, flayed alive, and
then crucified on the express commands of queen Amytis, and not of
Cyrus II (Ktesias 36b). The ability to take the life of eunuchs as powerful
39
as Petasakes and Artaxares 'the kingmaker' demonstrates the authority held

by the Achaemenid queens. Eunuchs tend to be beheaded, stoned to death
or 'condemned to execution in the trough' (Ktesias 39b), a particularly
hideous device, which is described by Plutarch in gory detail (Art. 16).
However, some eunuchs, despite loyally obeying royal commands,
also become the victims of torture and death. Thus, when the eunuch
Bagapates was sent by Artaxerxes II to decapitate the corpse of his younger,
traitorous, brother, Cyrus, after his defeat on the battlefield, the Queen
Mother, Parysatis, Hew into a rage and conspired to have the eunuch
killed because Prince Cyrus had been her favourite son. Ktesias (44a)
recounts the story:
[Bagapates] at the order of the king cut the head from Cyrus [the Younger's]
body; ... his mother played dice with the king and on winning [she] took
Bagapates according to an agreement; ... he was flayed alive by Parysatis.
The same story is expanded on by Plutarch (Art. 17), who renames

the eunuch Mastabates:
The king's eunuch, who had cut off the head and hand of Cyrus, remained
alive as a mark of Parysatis' vengeance. But he was so circumspect that he
gave no advantage against him, so she framed a trap for him. She was a very
clever woman in other ways and was an excellent player of dice, and, before
the war, had often played dice with t)le king. After the war, she also readily
joined in all games with him, played at dice with him, was his confidant in
love matters ... And so once when Artaxerxes was at leisure, and inclined
to divert himself, she challenged him to play dice with her for a thousand
darics, and let him win on purpose ... She pressed him to begin a new game
for a eunuch, to which he consented. But first they agreed that each of
them might except five of their most trusty eunuchs, and that out of the
rest of them the loser should yield up any one of them the 'winner might
choose. Upon these conditions they played ... When she had got the game,
she demanded Mastabates, who was not in the number of the five excepted.
Before the king could suspect anything, having delivered [the eunuch] to
his tormentors, she ordered them to flay him alive, to put his body on three
stakes, and to stretch his skin over the stakes.
Sancisi-Weerdenburg has seen this grotesque tale as having its origins

in Indian epic tradition, where human sacrificial victims are selected by
the throw of a dice, 83 but the story stresses the point that the position of
the eunuch was maintained purely on the good will of the king and his
women. The story of Mastabates/Bagapates perfectly illustrates that the
petty squabbles and wrangles within the royal family played out in the
royal harem could have wider, and often cruel, consequences.
40
Ever since the establishment of the Achaemenid dynasty, or at least from

the early reign of Darius the Great, all royal women of note had used
eunuchs as court functionaries, specifically assigned to work for them
in the seraglio-like palace compound in a variety of jobs: fan-bearers,
masseurs, guards and chamberlains. The effect this was to have on the
Achaemenid court and governmental system was enormous. Growing up
with eunuchs, a crown prince like Xerxes, who eventually became the
ruler of the realm, naturally preferred to be surrounded by familiar faces
in a cold-blooded world of politics and intrigue. The eunuch's relationship
with the ruler developed into one of valued companion, and who could
better help him maintain his absolute power and manage his harem than
his own childhood confidants? After all, these were the trusted servants
of his mother's court. This intimate association with their ruler and his
women became the source of eunuch influence and alleged duplicity. In
the hectic world of ~ourt administration, eunuchs provided the ruler with
a convenient way to cope with the demands of his office, and to keep him
in touch with the secluded but all too powerful women of the royal harem,
especially the omnipresent queen mothers. The influence, indeed, the
threat contained within the harem should not be lightly dismissed.
It is dear that in the Achaemenid despotic institution eunuchs were
both necessary and insignificant; the harem eunuchs of Persia depended
for their survival upon the favour of the king and, as has become apparent,
specifically the royal women. If they managed to serve these select ladies
well, and foster the intimate links that were forced upon them, then
the rewards could be great- wealth, high rank, and access to the king
himself. But with power came the risk of vice, of abrupt demotion, and of
execution. The Achaemenid royal women, with or without the permission
of the king, used frequent and severe punishment to curtail excessive
eunuch power.
Because the royal household was the fundamental unit of political and
social organization in the Achaemenid empire, power derived principally
from an individual's role within the household; in the harem network
the interests of powerful men and women combined to create a variety
of factions and it would appear that personal interplay was of the utmost
importance. A wise eunuch knew how to keep his job, and secure his
position and his life; service to the royal women could be profitable, but
the dangers of harem politics could be devastating.
Thus, when Artaxerxes II learned that his mother had flayed alive
Mastabates, his favourite eunuch, he became angry and threatened Parysatis
with exile, but, according to Plutarch (Art. 17): 'she with .. .laughter told
him, "You are a comfortable and happy man indeed, if you are so much
disturbed for the sake of an old and rascally eunuch."'
41
The king shut up and the queen mother won the day. But harem
intrigues continued to mount, and the dead eunuch became the point of
focus for the rivalry that existed between Parysatis and her daughter-in-law
Stateira, Artaxerxes' wife. Plutarch continues with his tale and points out
that Mastabates, even after death, continued to be the cause of domestic
discontent: 'Stateira opposed [Parysatis] ... and was angry, for against all
law and humanity, she had sacrificed to Cyrus' memory the king's faithful
friend and eunuch'.
Acknowledgements
This chapter has benefited from the generous advice of several individuals.
Accordingly, my thanks go out to Steven Griffiths, Nick Fisher and Anton Powell.
Especial thanks go to Shaun Tougher, for providing me with an opportunity to
work up my thoughts on Persian eunuchs and for his undergraduate course on
eunuchs at Cardiff University. He has been a patient editor and a good friend.
Notes
1 Achaemenid royal chronology runs as follows (all dates are Be):
Cyrus the Great 560/59-530
Cambyses 530-522
Bardiya (Smerdis) 522
Darius I 522-486
Xerxes I 486-465
Artaxerxes I 465-424
Xerxes II 424
Sogdianos 424-423
Darius II (Ochos) 423-405/4
Artaxerxes II (Arsakes) 405/4-359/58
Artaxerxes III (Ochos) 359/58-338/37
Artaxerxes IV (Arses) 338/37-336
Darius III 336-330
2 Flaubert 1852, trans. Krailsheimer, 24.
3 See Daguerre de Hureaux 1995, 128ff.

4 Penzer 1936.
5 Sitwell1948, 279.
6 See for instance Tougher, Gaul and Tsai, this volume.
7 Brosius 1996. Eunuchs make only fleeting appearances in the book; there is
no systematic study made of eunuchs and royal women. In fact 'eunuch' does
not appear as an entry in the index.
8 Briant 1996, 279-88.
9 Shakespeare, Antony and Cleopatra 1.5.9-18, evokes a mem:Orable scene set
within Cleopatra's palace at Alexandria. The enchantress-queen sits among a bevy

of comely maidservants, soothsayers, and eunuchs. Languishing and pining for
42
her beloved Antony, Cleopatra mocks a eunuch· named Mardian, saying, 'I take
no pleasure in aught a eunuch has. 'Tis well for thee that, being unseminared, thy
freer thoughts may not fly forth of Egypt. Hast thou affections?'
The eunuch answers, 'Yes, gracious madam.' 'Indeed?' asks the queen. 'Not
in deed, madam', Mardian explains, 'for I can do nothing but what indeed
is honest to be done: Yet I have fierce affections and think what Venus did
with Mars.'
Shakespeare's Mardian is a eunuch modelled in traditional vein.
An excellent study of the fictional harem (and the use of eunuchs within the
institution) in western literature is provided by Bernard Yeazell2000.
10 Peirce 1993. See also Gibbs and Bowen 1950, 71-89, 329-33. Chinese
sources also emphasize the eunuch-harem connection. See in particular Mitamura

1963, 74ff. The view that eunuchs were the appropriate attendants for women
is even held by the Romans: in the early third-century AD, we are told of how
Plautianus, the Praetorian Prefect of Septimius Severus, had castrated a hundred
Romans of noble birth in order that his daughter, Plautilla, might be tutored by
eunuchs skilled in music and the arts. See Dio Cass. 76.14.4-6.
11 See Grayson 1995.
12 Brosius 1996, 123 ff.
13 For the textual traditions of Esther see Day 1995 and de Troyer 1995.
Even scholars who regard Esther as a historical novel concede that its author is
intimately acquainted with Persian royal custom. See discussions in Yamauchi
1990, 237ff.
14 Plato Rep. 479b.
15 Hall1989, 157.
16 Nielsen 1994.
17 See, for example, Reade 1972.
18 See Wilber 1989, 56 and Brown 1995, 96.
19 It is rare to find a depiction of two bearded officials ministering to the king,
but one such relief does exist on an exterior door jamb of the palace of Darius I.
Perhaps this demarcates the public area of the place just as the eunuch relief seems
to denote entry into the private heart of the palace. For a dear illustration of the
bearded officials holding flywhisks and parasols see Wilber 1989 plate 13.
20 In his desire to read Persepolis as a temple dedicated to the celebration of
the New Year Festival, Fennelly 1980 has interpreted the beardless figures as
eunuch-priests in the service of 'a goddess'. As yet, however, no evidence has
come to light for any concept of eunuch priests serving within the state religion
of Persia. It is better to see the Persepolis reliefs as depicting elegantly dressed
and coiffured court eunuchs.
21 Meyer 1896, 194.
22 Neh. 1.11.
23 See Ahmed 1982.
24 See further references to Orientalist paintings in Bernard Yeazell 2000.
Hollywood epics often pick up on the Orientalist discourse: in the 1956 film
Alexander the Great, for example, a blond Richard Burton as the Macedonian
43
conqueror, defeats the Persian army and - almost as a reward - discovers the
tented seraglio of the Persian monarch, Darius III. The girls, attended by fat
effeminate eunuchs, and wearing classic Scheherazade harem-pants and jewelled
bikini-tops, are seen giggling together. Suddenly alarmed at the entrance of this
mysterious, handsome stranger, they hurriedly raise their gauzy yashmaks over
their faces, but too late; Alexander, like the cinema audience, is bewitched by the
spectacle of erotic Oriental hedonism.
25 See Dixon 1992.
26 See Marmon 1995 and Peirce 1993, 3 f£
27 See Briant 1996, 217-65.
28 Brosius 1996, 188
29 Frye 1962, 97.
3° Frye 1962, 100.

31 See also Hdt. 3.69. Herodotos also reports that Persians slept with their
wives in rotation.
32 See further Athenaeus 12, 528e-f and discussions in Sancisi-Weerdenburg
1987.
33 Sancisi-Weerdenburg 1987.
34 Lewis 1977, 21-2. Olmstead's narrative of Achaemenid history lays stress
on the nature of the harem; there is even an entry in the index under 'harem
intrigues'. See Olmstead 1948.
35 Lewis 1977, 22.
36 See in particular Olmstead 1948.
37 That, of course, is not to deny the importance of the satrapy system, or
indeed the impressive Achaemenid bureaucratic infrastructure of the Empire.

For the Persian administration see Dandamaev and Lukonin 1989, 96ff., Briant
1996, 370f£, and Wiesehofer 1996, 59f£
38 Much the same situation existed in Imperial China. See Soulliere 1992 and
Mitamura 1963.
39 One exception is a small raised-relief in the British Museutp showing King
Ashurbanipal dining al fresco with his wife Queen Ashurshurat in the gardens
of the royal palace, that is to say in the domestic quarter of the harem. For
details see Reade 1983, 65 and figs. 102 and 103. For Assyrian palace reliefs
see Russell 1991.
4° For the conception and subsequent imperial function of Persian art see
Boardman 2000 and Root 1979.

41 For a discussion of the Icing's Greek and Akkadian names see Yamauchi
1990, 187.
42 See LXX Esther 2.3; Aelian VH 12.1. See further, Albright 1974.
43 Callirhoe travels through Persia in a carriage with closed curtains. See
Chariton Cal. 5.2.9.

44 See, for example, Athenaeus 4.145-6, Plut. Art. 5.5. On w~men and feasting
see Brosius 1996, 94 f£, Briant 1996, 297 f£, and Cook 1983, 139 ff.
45 Brosius 1996, 105 f£
46 For Parysatis' influence at court see Ktesias 48-51; Dinon F 15b; Plut.
44
Art. 19. 2-3. See discussion in Brosius 1996, 100ff. David Lewis' work on the
Persepolis Fortification Tablets has led him to identifY Artaxares the Paphlagonian
as a certain Artahsaru, whose large staff of servants is visible in at least eleven of
the texts dated to the 420s BC. See Lewis 1977, 21.
47 See Brosius 1996, 105 ff.
48 Hdt. 3.134. See Brosius 1996, 106ff.
49 See Briant 1998, 292 ff.

5° Cook 1983, 136.
51 For a good summary of the early excavation of Persepolis and for excellent
reproductions of the excavation photographs see Brown 1995, 73-83. Interest-

ingly, after sleeping and working in tents for the first two Persepolis dig seasons,
Herzfeld and his team transferred their accommodation and workshops into the
harem quarter of Xerxes' place, which the architect Friedrich Krefter and the
Oriental Institute builders had recreated from the deteriorated walls, columns
and capitals of the ancient structure.
52 Schmidt 1953, 255-64.
53 Olmstead 1948, 285.
54 See Root 1979, 101 and Wilber 1989, 66.

55 White 1915, 254. See further details in Lacovara 1994. For discussions of the
Egyptian New Kingdom harem see Leblanc 1999 and Tyldesley 2000, 113ff.
56 Nielsen 1994, 194.
57 Nielsen 1994, 44-7.

58 On the nature of royal polygamy and harem rivalries see Carney 2000 and
Ogden 1999.
59 Dinan F 15b; Plut. Art. 19.2-3. See Brosius 1996, 110-11.
60 See Peirce 1993, 139.
61 See Guyot 1980 for a complete list.

62 For the size of the royal family see Cook 1983, 137.
63 Frg. 169. FGH 1, 68.
64 Arnm. Marc. 14.6.17.
65 Soph. fr. 620.
66 Trans. Vellacott 1972.
67 Hall1989, 158. See also Witt, this volume.

68 Aristotle Problems 11.16, Physiognomies 807a, GA 787b-88.
69 On the feminization of the Persian eunuch see Curti us Rufus 10.1.26.
70 See in particular Badian 1958. See also Briant 1996, 280ff.
71 For a complete translation of the harem edicts (Middle Assyrian palace
decrees) see Roth 1995, 195-209. For a discussion of the Assyrian harem see
Kuhrt 1995, 526-31.
72 Frye 1962, 100.
73 Jones 1977, 175. See further comments in Baldwin 1984 and Day 1995.
74 Hopkins 1978, 172ff.
75 For the role of royal women in the hunt see Brosius 1996, 91 ff.
76 Hdt. 3.31.1; 9.108.1; Ktesias 44. See Brosius 1996 90.
77 See, for example, Plut. Them. 16.4; Aristides 9.4.
45
78 Ktesias 36b.
79 See further Levine Gera 1993, 192ff.
80 Xen. Anab. 7.3.15.
81 Hdt. 7.107.
82 Peirce 1993.
83 Sancisi-Weerdenburg 1993.
Bibliography
Ahmed, L.
1982 'Western ethnocentrism and the perceptions of the harem', Feminist
Studies 8, 521-34.
Aillagon, J.-J. (ed.)
1995 Carthage. L'histoire, sa trace, et son echo, Paris.
Albright, W.F.
1974 'The Lachish cosmetic burner and Esther 2.12', in H.N. Bream (ed.)
A Light Unto My Path, Philadelphia, 25-32.
Badian, E.
1958 'The eunuch Bagoas. A study in method', CQ 8, 144-57.
Baldwin, ].G.
1984 Esther, London.
Barnett, R.D
1970 Assyrian Palace Reliefs in the British Museum, London.
Bernard Yeazel!, R.
2000 Harems ofthe Mind, New Haven and London.
Boardman, J.
2000 Persia and the West, London.
Briant, P.
1996 Histoire de !'empire perse, Paris.
Brosius, M.
1996 Women in Ancient Persia (559-331 Be), Oxford.
Brown, D.M. (ed.)
199 5 Persians: Masters ofempire, Alexandria, Va.
Carney, E.D.
2000 Women and Monarchy in Macedonia, Norman, Okla.
Cook,].M.
1983 The Persian Empire, London.
Curtis, J.E.
1989 Ancient Persia, London.
Curtis, J.E and Reade, J.E. (eds.)
1995 Art and Empire: Treasures from Assyria in the British Museum, London.
Daguerre de Hureaux, A.
1995 'Salammbo: entre I' Orient des romantiques et 1' orientalism fin-de-siecle',
in Aillagon (ed.) Carthage, 128-37.
46
Dandamaev, M.A. and Lukonin, V.G.

1989 The Culture and Social Institutions ofAncient Iran, Chicago.
Day, L.
1995 Three Faces ofa Queen. Characterization in the Book ofEsther, Sheffield.
De Troyer, C.
1995 'An oriental beauty-parlour: an analysis ofEsther 2.8-18 in the Hebrew,
Septuagint, and second Greek text', in A. Brenner (ed.) A Feminist
Companion to Esther, judith and Susanna, Sheffield, 47-70.
Dixon, S.
1992 'The enduring theme: domineering dowagers and scheming concubines',
in Garlick, Dixon and Allen (eds.) Stereotypes of Women in Power,
209-25.
Durand, J.-M. (ed.)
1987 La femme dans le proche-orient antique, Paris.
Fennelly, J.M.
1980 'The Persepolis ritual', Biblical Archaeologist, summer edition (no
number), 135-62.
Frye, R.N.
1962 The Heritage ofPersia, London.
Garlick, B., Dixon, S. and Allen, P. (eds.)
c 1992 Stereotypes ofWomen in Power: Historical perspectives and revisionist
views, New York.
Gibbs, H.A.R and Bowen, H.
1950 Islamic Society and the ~st, vol. 1.1, London, New York and Toronto.
Grayson, A.K.
1995 'Eunuchs in power. Their role in the Assyrian bureaucracy', in M. Diet-
rich and 0. Loretz (eds.) Festschrift for Wolfram Freiherrn von Soden
(Alter Orient und Altes Testament 240), Neukirchen, 85-98.
Guyot, P.
1980 Eunuchen als Sklaven und Freigelassene in der griechisch-romischenAntike,
Stuttgart.
Hall, E.
1989 Inventing the Barbarian: Greek self-definition through tragedy, Oxford.
Hopkins, K.
Jones, B.W.
1977 'Two misconceptions re the Book of Esther', Catholic Biblical Quarterly
39, 175-81.
Kuhrt, A.
1995 The Ancient Near East, c. 3000-330 BC, 2 vols., London.
Lacovara, P.
1994 'In the realm of the sun king: Malkata, palace-city of Amenhotep III',
in D.C. Forbes (ed.) Amarna Letters 3. Essays on Ancient Egypt ca.
1390-1310 BC, San Francisco, 6-19.
Leblanc, C.
1999 Nefertari, l'aimee de Mout, Paris.
47
Levine Gera, D.
1993 Xenophon's Cyropaedia, Oxford.
Lewis, D.M.
1977 Sparta and Persia. Lectures Delivered at the University of Cincinnati,
Autumn 1976, in Memory ofDonald B. Bradeen, Leiden.
Marmon, S.
1995 Eunuchs and Sacred Boundaries in Islamic Society, Oxford.
Meyer, E.
1896 Die Entstehung des ]udenthums, Halle.
Mitamura, T.
1963 Chinese Eunuchs. The structure ofintimate politics, Tokyo.
Nielsen, I.
1994 Hellenistic Palaces: Tradition and renewal, Aarhus.
Ogden, D.
1999 Polygamy, Prostitutes and Death. The Hellenistic dynasties, London.
Olmstead, A. T.
1948 History ofthe Persian Empire, Chicago and London.
Peirce, L.P.
1993 The Imperial Harem. Women and sovereignty in the Ottoman Empire,
New York and Oxford.
Pelletier-Hornby, P.
1995 'Salammbo: la Carthage de Flaubert', in Aillagon (ed.) Carthage,
138---44.
Penzer, N.M.
1936 The Harem. An account ofthe institution as it existed in the palace ofthe
Turkish Sultans with a history of the Grand Seraglio from its foundations
·to modern times, London.
Reade, J.
1972 'The neo-Assyrian court and army: the evidence from the sculptures',
Iraq 34, 87-112.
1983 Assyrian Sculpture, London.
Root, M.C.
1979 The King and Kingship in Achaemenid Art: Essays on the creation of an
iconography ofempire, Leiden.
Roth, M.T.
1995 Law Collections from Mesopotamia and Asia Minor, Atlanta.
Russell, J.M.
1991 Sennacherib's Palace Without Rival at Nineveh, Chicago.
Sancisi-Weerdenburg, H.
1987 'Decadence in the empire or decadence in the sources? From source to
synthesis: Ctesias', in H. Sancisi-Weerdenburg (ed.) Achaemenid History
I. Sources, Structures and Synthesis. Proceedings of the Groningen 1983
Achaemenid History Workshop, Leiden, 33---45.
1993 'Exit Atossa: images of women in Greek historiography of Persia', in
A. Cameron and A. Kuhrt (eds.) Images of Women in Antiquity, 2nd
edn,London,20-33.
48
Schmidt, E.F.
1953 Persepolis I, Chicago.
Sitwell, 0.
1948 Escape With Me!, London.
Soulliere, E. ,,
1992 'Imperial women in the history of the Ming dynasty (1368-1644)',
in Garlick, Dixon and Allen (eds.) Stereotypes ofWomen in Power,
121--40.
Tydesley, J.
2000 Ramesses. Egypfs Greatest Pharaoh, London.
Vellacott, P.
1972 Euripides. Orestes and other plays, London.
White, H.G.E.
1915 'Egyptian expedition', Metropolitan Museum ofArt Bulletin 10/12, 254.
Wiesehofer, J.
1996 Ancient Persia, London.
Wilber, D .N.
1989 Persepolis. The archaeology of Parsa, seat of the Persian kings, 2nd edn,
New Jersey.
Wilson, N. G.
1994 Photius. The Bibliotheca, London.
Yamauchi, E.M.
1990 Persia and the Bible, Grand Rapids.
49
3
EUNUCHIZING AGAMEMNON:
Clytemnestra, Agamemnon and maschalismos
Ruth Bardel
The human body has recently emerged as a topic of great interest to

classical scholars, 1 a subject which is also intimately connected to issues
of sexuality and gender. In ancient Greek culture the human body 'served
as an important location for self-identification and demonstration of
authority' (Bremmer 1991, 27), and its wholeness and proficiency were
crucial to the maintenance of this identity and authority. The Greek
adult male citizen was a 'whole and bounded subject' and only his body
was 'sovereign, autonomous, and inviolable' (Stewart 1997, 8), an ideal
against which barbarian, female and slave bodies were measured and
found wanting.
Eunuchs were known of in classical Athens (see Miller 1997, 213-15)
and had been represented in drama as early as 476 in Phrynichos' Phoinis-
saU Although fascinated by eunuchs, whose mutilated and emasculated
bodies threw into confusion the clearly defined gender categories marking
Greek society, the Greeks were disgusted by what they regarded as the
savage and barbaric practice of castration. Herodotos (8.1 05-6) is clearly
outraged, not only by the nefarious practice of rendering beautiful young
boys eunuchs, but also by the castration of Hermotimos, carried out by
the Chian eunuch-trader Panionios, who had bought Hermotimos as
a prisoner of war. This account is highly pertinent in that Hermotimos
(having become the chief eunuch of the Persian king Xerxes) actually
confronts his mutilator, arguing that nothing could have merited his
castration, which made him the 'nothing' ('tO J.lTJOEV) that he now was.
Hermotimos exacts his revenge on Panionios by forcing him to make
eunuchs of his four sons who were themselves forced to 'eunuchize' their
father, no doubt so that he too, may be a 'nothing'. Drawing together
a number of literary, dramatic and other sources and focusing especially
on Aeschylus' Oresteia, this chapter sets out to illustrate that mutilation in
general, and castration in particular, was not the sole preserve of barbarians
51
Ruth Bardel
and that it is closely allied to ancient Greek notions of justice, revenge,

gender and sexuality.
The Oresteia's most prominent themes are the problems presented by
Dike, Ate, and Moira. The old, and more primitive, system of justice
based on the lex talionis - an eye for eye and tooth for tooth - is replaced
in the third play of the trilogy, the Eumenides, by the institution of the
Areopagus court. The supersession of the savage, ancient and cyclical
blood-feud by the institution of public justice and the rule of law is
bound up thematically with many other issues in the Oresteia. The clear-
cut struggle between male and female, for example, provides a rubric
within which notions of justice and revenge are articulated. This chapter
addresses the vexed problem of the practice of maschalismos (lopping off
the extremities of the body) within the context of the lex talionis which
operates so forcefully within the Oresteia. That the principles of the lex
talionis are associated with Clytemnestra and the Erinyes, those hideous
female creatures, is made quite clear in the Eumenides. Apollo, Orestes'
champion, and Athena, the goddess who defers to the male in all things,
except marriage (Eum. 737), are clearly representatives of the newly
established civic laws.
In 1948 Winnington-Ingram drew attention to Clytemnestra's political
ambition and her jealousy of Agamemnon's 'status as a man'. Despite
Clytemnestra's usurpation of male power and prerogatives and her
transgression of ideal gender roles, Clytemnestra remains unequivocally
female· and Aeschylus is keen to stress her 'womanish' ways. Zeitlin's
description of Clytemnestra as a 'monstrous androgyne' who demands
and usurps male power and prerogatives (Zeitlin 1996, S9; c£ Zeitlin
1978, 156), can only be true in a non-biological sense and, I suggest,
such a characterization obscures the trilogy's construction of the clear-cut
struggle between male and female which overlays the chief moral ideas as
one of their expressions. Clytemnestra, as a woman, exploiting 'typical'
female resourcefulness to the limit, usurps and appropriates male power
and prerogatives: on his return, Clytemnestra disarms Agamemnon, breaks
down his defences, lures him into the trap she has set, murders him in
revenge for Iphigenia's sacrifice and then mutilates his body. This small,
often overlooked, detail of Clytemnestra's mutilation (maschalismos) of
Agamemnon's body after his murder is highly significant - both in terms
of the lex talionis and in the struggle between male and female. First, I shall
briefly examine examples of the lex talionis and its relationship to revenge
and mutilation. Then, I shall discuss maschalismos, its nature and purpose,
and finally offer an explanation for its major role in Clytemnestra's
vengeance exacted upon Agamemnon. 3
52
Eunuchizing Agamemnon: Clytemnestra, Agamemnon and maschalismos
Perhaps the best-known expression of the lex talionis is the biblical

'Eye for eye, tooth for tooth, hand for hand, foot for foot. Burning for
burning, wound for wound, stripe for stripe' (Exodus 21.24-5). When
the principle of talio - vengeance - is applied, the punishment is made
to fit the crime: vengeance is thus not only repetitive and retributive,
it is also mimetic. The perpetrator of the offence or crime suffers in
exactly the same manner as his victim: the punishment may not always be
literally mimetic- it can be symbolically so. For example, in Deuteronomy
(25.11-12) it is stated that:
When men strive together one with another, and the wife of the one draweth
near for to deliver her husband out of the hand of him that smiteth him, and
putteth forth her hand, and taketh him by the secrets [the genitals/testicles]:
then shalt thou cut off her hand, thine eye shall not pity her.
In this case the 'peccant limb suffers in a grim attempt to make the
punishment fit the crime' (Driver and Miles 1935, 347).
Assyrian law is slightly less severe:
If a woman has crushed a man's testicle in an affray, one of her fingers shall
be cut off; and if, although a physician has bound it up, the second testicle
is affected with it and becomes inflamed, or if she has crushed the second
testicle in the affray both her [nipples(?)] shall be torn out. 4
In this talionic system of justice the testicle and the nipple, rather than the
hand, are analogous. In cases of adultery, Assyrian law stated that:
if the woman's husband puts his wife to death, then he shall put the man to
death; (but) if he has cut off his wife's nose, he shall make the man a eunuch
and the whole of his face shall be mutilated. 5
Both of the offending parties in a proven case of adultery were treated

equally, that is, death or mutilation for both the wife and her lover.
Should the husband choose mutilation as punishment - 'in all penalties
[involving either tearing out (the breasts) or] cutting off (the nose or ears)
of a [married woman]', a priest must be informed (Driver and Miles 1935,
425). Why a priest? Driver and Miles (1935, 425) suggest that if the priest
did not perform the mutilation himself, it was at least done under his
supervision; such punishment 'if carried out unskilfully or vindictively
would have serious consequences for the victim'. There can be no doubt
that the unskilled and vindictive tearing out of breasts or cutting off
of noses would have serious consequences for the victim, but there is,
I believe, another reason for calling the priest in. If the priest attended,
or even carried out, the operation, the punishment would be ritually or
divinely sanctioned. What is significant about the last law cited is that
53
RuthBardel
the castration of the lover, making him a eunuch, and the tearing out
of the unfaithful wife's breasts are seen as corresponding punishments.
Furthermore, the punishment removes the offending body parts, in
particular the male genitalia and the female breasts, rendering the offenders
ill-equipped to repeat their offence. In the case of facial mutilation, the
adulterous lovers would not be (dare I say) particularly attractive to their
own sex, let alone to the opposite sex. The lex talionis, as illustrated by
these examples from Assyrian law, operates on a very basic, very physical
level - a kind of vindictive symmetry.
'Primitive' legal systems are not the only arena in which mutilation
takes place: the battlefield provides another context. A relief from the
temple of Medinet Habu in Egypt is one in a series depicting the military
campaigns ofRamesses III against the Libyans (c. 1180 BC) and portrays
an assemblage of captives and spoils. The bound prisoners are led in from
the right, while in front the scribes are 'busily taking a rather gruesome
count: piled before them are the genitals and hands of the slain, allowing
the enemy's losses to be computed' (Murnane 1980, 13). Such trophies,
hands and genitals, were customary and those who brought them in were
immediately rewarded with gold. The exact totals are confusing and highly
exaggerated but what is significant are the trophies themselves. 6 The hands
and genitals, hacked off the dead enemies' bodies, act as tallies for those
dead bodies: these 'trophies' are a grim synecdoche in the vocabulary
of mutilation.
The-mutilation of an enemy's dead body on the battlefield is a fear
which not only haunts the Homeric warrior but which is also, at times,
realised. In the Iliad, after Agamemnon has killed Hippol~chus, he cuts
off his hands (XEtpm; a1to ~t<j>£1 't!lTJ~m;), severs his neck with his sword
and rolls him like a stone through the crowd (II. 11.146:._7). Priam fears
that Achilles has hacked up (!ld.Eta'tt 'ta!lrov, II. 24.409) Hector's body,
trimming the trunk of its head, arms, hands and legs and chopping these
bits up into little pieces. 7 Among the things you could do to an enemy's
body is to carve up (a1tO'tE!lV£tv) his flesh and eat it raw just as Achilles
threatens to do to Hector (II. 22.346-8), as Zeus suggests Hera wishes she
could do to the Trojans (II. 4.34-6) and as Hecabe wishes she could do
to Achilles (II. 24.212-13). 8 Severing, carving or cutting up (a1tO'tE!lVCO)
the enemy's body and ingesting its parts is, these passages suggest, the
only way to assuage personal anger. Having one's corpse thrown to the
dogs or being left as food for the birds seems marginally better than being
carved up and eaten alive; both, however, point to a fragmentation and
dissolution of the body and its distinguishing features reducing, as Vernant
notes, a human to the condition of a formless thing (1991, 71).
54
To take the example ofThyestes' feast from the Oresteia: Atreus breaks
into pieces the toes, feet and fingers and hands of Thyestes' children so
as to render the food unidentifiable ('ta !lEV nooiJpTJ Kat XEprov aK:pou~
K'tEVa~ I £8pU1t't', avro8Ev ... I avopaKa~ Ka8iJ!lEVO~.I UCJTJilU o' mhrov mJ'tiK'
ayvoiQ: ... Ag. 1594-7). 9 It is clear from this account that the hands and
feet, the well articulated parts of the human body, would act as signifiers
for what the food actually was. Removal and the fragmentation, mincing
or crushing of these parts clearly disguises identity and incapacitates,
or to employ a pun, 'disarms'. 10 8pu1t'tro denotes breaking in pieces, to
break small, crush, and enfeeble. It can also mean to make effeminate or
unmanly (cf. Xen. Symp. 8.8 and Luc. Charid. 4); in Euripides' Bacchae,
Pentheus says to Dionysus, 'You are going to compel me to be broken
into pieces' (969). With a gruesome pun on the word 'tpu<j>av I 'tpuya~,
derived from 8pu1t'tro, Dionysus responds, 'And what a breaking up it will
be' (970) (cf. Kepple 1976, 107). The process of metaphorically breaking
Pentheus into little pieces, of gradually eroding his defences, of rendering
him weak and effeminate, which began early in the play, culminates in
his literal breaking into pieces by his mother and the maenads. Pentheus'
horrid death is also a kind of lex talionis, one which is intimately bound up
with his battle against divinity (Eur. Bacc. 45). Pentheus misuses his geras
and 'kingly power' ('tupavvioa, 43). In Plato's Gorgias, the punishment
for a man who is caught criminally plotting to make himself a despot
(turannidi) is racking, castration, the burning out of eyes, grievous
torments of every kind, crucifixion or burning in a coat of pitch (473b-c)
- punishments which are strikingly similar to Apollo's description of the
torture chamber in which the Erinyes belong in the Eumenides. Tragic
heroes are broken and fragmented in a very physical manner; a number
of mythical heroes are evoked by the description of the punishments
mentioned above, for example, Oedipus, Ixion, Heracles.
Outside the context of warfare, mutilation as a punishment is particu-
larly horrifying. In the Odyssey, Antinous threatens to send Irus the beggar,
who is about to fight Odysseus, to King Echetus if he is defeated: this
would be no holiday or simple enslavement, but as Redfield remarks, it
would be 'for the sadistic pleasure of that monstrous monarch' (1975,
269-70). The pernicious Echetus would cut off Irus' nose and ears with
the pitiless bronze, and will tear out/off his genitals and give them raw to
dogs to eat (Od. 18.85-7). The exposure of the genitals, the private parts,
is a particular affront to a man's aiooo~- his self-respect- having them torn
out and thrown to the dogs is a fate which really hits 'below the belt', where
it hurts or shames the most. This is Priam's worst nightmare, that dogs
would devour his exposed genitals after his death (aioro ... K'tallEVOtO) .U
55
RuthBardel
Antinous' threat to Irus becomes a grim reality for the traitor Melanthius:
Telemachus and Odysseus' servants inflict an Echetus-style punishment on
him (Od. 22.474-7) by cutting off his nostrils, ears, hands and feet, tearing
out/off his genitals and throwing them to the dogs to eat raw. Scholia
state that Echetus was a (mythical) tyrant of Sicily who discovered many
devices (!J.TJXavai) of torture or mutilation (aiKia) for the unfortunate
strangers sent to his domain. Echetus, discovering his daughter's 'intrigue'
with a lover, cut off the lover's extremities (l]Kpffi'tT)piacr£) as well as his
genitals ('ta aioo'ia <htEKO'IfEv) and maimed (7tTJpfficras), that is blinded,
his own daughter, Metope, an example of excessive paternal jealousy (Ap.
Rh. Argonautica 4.1092).
The association between mutilation, barbarians and emasculated males
is clear from Sophocles' fragmentary play, the Troilus, in which an enslaved
eunuch asserts that the queen herself (presumably Hecabe) had his testicles
cut out with a scimitar (fr. 620: crKaA!J.lJ yap opxns ~acrtA-1s EK'tE!J.voucr'
E!J.OUs). Hall states that the mutilation of Troilus' attendant may have
been 'much emphasized' in the play and suggests that it was a significant
factor in the more general feminization, not only of Asia, but also of
the barbarian body in the Greek imagination as it manifested itself in
tragedy (1989, 157). That it was the queen herself who had this mutilation
performed suggests that the barbarian political/royal body, itself perceived
as feminine, renders its entourage equally effeminate by this (barbaric)
operation. 12 Sophocles' fragmentary Po!yxena also mentions mutilation
(l]Kpffi'tT)ptacr!J.EVOt, fr. 528 Snell-Radt) which Harpocration (146, 12
Dind.) glosses as oi AU!J.atvO!J.EVOt 'tt<Jt 1tEptK01t'tO'\.l<Jt 'ta aKpq (the outrages
and lopping off of the extremities). This is quite probably a prophecy
regarding Agamemnon's murder. 13 The words aKpm'tT)ptat;m and 7tEptK01t'tffi
are used for the 'mutilation' of the Herms in 415 Bc. 14 Thucydides records
that the Herms 'had their fronts cut/knocked up all round' (7tEptEK07tT)crav
'ta 1tp6crm1ta, 6.27 .1) a phrasing which Hatzfeld describes as a 'formule
pudique' (1940, 161), for, as he and others- for example, Keuls (1985,
13) and Henderson (1987) -maintain, the phallus was a far more obvious,
vulnerable and outrageous target. Furthermore, the removal of the Herm's
phallus negates the meaning of its image (just as castration makes a man
a eunuch and changes his meaning/status?).
Although mutilation was thought in classical Greece to be a practice
more suited to barbarians than Hellenes (Hdt. 9.79), the examples
I have given from the Iliad and the Odyssey make it clear that such acts
of mutilation were not 'reserved in epic discourse for 'performance by
non-Greeks' (Hall 1989, 25-8). When, in 458 BC, Aeschylus makes the
Chorus of the Choephoroi add the detail that Clytemnestra mutilated
56
Agamemnon's corpse, he is clearly drawing on what Hall refers to as

the 'vocabulary of barbarism' (1989, 203) frequently deployed to evoke
any character's excess, transgression, subversiveness or departure from
Hellenic virtues.
At line 439 of the Choephoroi, we are told that Clytemnestra 'foully
mangled' (£11acrxaA.icr811) Agamemnon's corpse. Sophocles, like Aeschylus,
mentions maschalismos in his Electra, adding that Clytemnestra wiped the
bloody instrument of death on the head of the murdered Agamemnon
(£11acrxaA.icr81'], K<btt A.ompotcrtv Kapc;x I K1']Atoa~ £~£11a~Ev, 445-6), 15
a gesture by which Clytemnestra signals 'on your head be it', that is,
Agamemnon must accept responsibility for the consequences of his
actions. 16 Euripides, in his Electra, does not mention the mutilation of
Agamemnon's corpse, an omission which seems to coincide with his rather
more sympathetic portrayal of Clytemnestra. I have deliberately used the
Loeb translations of the word EllacrxaA.icr81'] because they highlight the
uncertainties generated by the word: what exactly did the translator mean
by 'foully mangled'? Ancient lexicographers state that the murderer cut
off the 'extremities' of his victim, tied the severed parts on a string and
then hung it around the neck of the victim, drawing the rope under the
armpits. 17 It is from this process of drawing the string under the armpits
(!lacrxaA.at) that the practice 11acrxaA.i~nv derives its name. The parts
- 116pw - of the body thus fastened are referred to as llacrxaA.i<Jila'ta. 18
Exactly what bits of the body were cut off? Both ancient and modern
commentators simply accept the vague notion of 'extremities' assuming
them to include mainly the hands and feet (for example, Lawson 1910,
435) and sometimes the nose, ears, nails, eyes, tongue and thumbs. The
genitals are rarely mentioned: given the modern connotations of 'emascu-
late' - to deprive of the properties of the male or, to castrate - this seems
odd. Neither Kittredge's 1885 article, nor ]ebb's commentary on Sophocles'
Electra, nor Rohde's appendix - the most comprehensive treatments of
the topic to date - were clear about whether or not maschalismos involved
castration. A few critics have proposed that castration was involved: the
Freudian critic Devereux (1976, 223), for example, states, 'Now, it is
inconceivable that one of the amputated "limbs" should not have been the
phallos.' Vermeule (1979, 222 n. 16) notes that the 'moria of the dead men
which get bound round the neck very probably include genitals in practice,
although literary sources are shy of mentioning it' .19
The ancient sources - Homer, Sophocles, Aeschylus -which I have used
as examples, although they may not explicitly mention castration, have
helped to determine which 'extremities' were cut off when maschalismos was
performed. It should be clear by now that the severed 'extremities' include
57
Ruth Bardel
head, arms/hands/fingers, legs/feet, toes, nose/nostrils, ears and, in some

cases, genitals: where the genitals are not explicitly mentioned the verb
used for cutting off, 'tE!-LVro, can also denote castration. The fundamental
difference between maschalismos and other forms of mutilation such as are
implicit in the verbs ano'tE!-Lvro, aKpro'tTJpu:isro and 7t£ptK07t'tro lies in the
stringing together of the severed parts and tying them under the victim's
armpits. 20 I shall return to this point later.
Apart from the explicit reference to maschalismos in the Choephoroi, an
oblique reference is made to the myth of Ouranos' castration by Kronos
in the Agamemnon (167-73). Neither Ouranos nor Kronos is mentioned
by name, they are simply 'past and gone' - is this the fate of castrated
males, never to be mentioned again? Hesiod's account (Th. 176-82) is far
more explicit: 'the hidden boy I Stretched forth his left hand; in his right
he took I The great long jagged sickle; eagerly I He harvested his father's
genitals I And threw them off behind'. Arthur (1982, 64) has shown how,
in the Theogony as in many cultural founding myths, there is a congruence
between the sexual and political spheres. Castration, in the context of
this myth, is used as a means of usurping power and it is significant
that it is Kronos' right hand which performs the operationY In the
Argonautica (4.982-92), Apollonius, despite his averred displeasure,
facetiously recounts the aition for the island of Drepane, named after the
sickle which was used to castrate Ouranos and where Jason and Medea
married- surely an Apollonian joke!
Mutilation (aKprovia), along with castration, beheading, impalement
and the gouging out of eyes, also falls into the domain of b~rbaric atroci-
ties associated with the Erinyes mentioned by Apollo in the Eumenides
(187-90). 22 By mutilating Agamemnon's corpse, Clytemnestra aligns
herself not only with the unwholesome activities of barbafian tyrants but
also with the horrific punishments inflicted by the Erinyes on their victims.
The mutilation and castration of Agamemnon's corpse by Clytemnestra is
the most palpable attribute shared between this transgressive woman and
the Erinyes. The Erinyes, themselves born of the blood of the castrated
Ouranos (Hes. Th. 183-5), emasculate and destroy the 0"7tEP!-La of youthful
vigour (Eum. 187-8), thereby ensuring that regeneration is impossible. By
practising maschalismos on Agamemnon's dead body, Clytemnestra seeks to
eliminate the prospect of vengeance, either by Agamemnon's posthumous
ghostly activity or by his begetting children. And, as if to stress this point,
she herself appears as a ghost in the Eumenides to hound on the Erinyes in
their pursuit of revenge (perhaps precisely because Agamemnon does have
a son, Orestes, to champion his cause?).
The Chorus of the Choephoroi state that by mutilating Agamemnon's
58
corpse, Clytemnestra sought to make his murder an 'intolerable' burden

on Orestes' life (440-3). Instead of dying a glorious death at Troy,
Agamemnon is treacherously slain at the hands of his own wife in his
own home - this is how Agamemnon will be remembered. 23 This is one
crucial, though neglected, motive for performing maschalismos: there are,
however, two further possibilities which can be culled from various scholia,
lexicographers and commentaries. 24 On the basis of the explanation given
by Aristophanes of Byzantium, Rohde argues that the whole procedure
of maschalismos is intended to offer the murdered man as a sacrifice to
the gods of aversion (ano--cponatot) 25 and that the llaaxat-iallma are
the anapxai - the preliminary offering - of this 'sacrificial victim'. 26 If
these raw off-cuts (oo!l08£'tOU!l£Va) could also be called, in a comparison,
maschalismata, then - Rohde argues - there is an affinity between the
two practices designed to be a preliminary apotropaic or cathartic offer-
ing intended to avert evil. The very elusiveness of the term seems to
suggest that maschalismos was a semi-ritualised procedure. If we accept
Rohde's analogy, we may add Clytemnestra's mutilation and 'offering' of
Agamemnon's severed body parts to the other motifs of the corrupted
sacrifice in the Oresteia surveyed in Zeitlin's 1965 article. Zeitlin maintains
that the motif of the corrupted sacrifice is 'almost non-existent' in the
Choephoroi and that there is an 'ascendancy of normal ritual' in the play
(1965, 484, 508). The notion of a return to normal ritual is problematized
by the 'ritual impasse' expressed by Electra and compounded by her 'ritual
ignorance/naivete' (Cho. 85-99).
Maschalismos severs (and renders useless) the generative organs of the
male body as well as the 'operative' I active parts - the hands and feet - and
was part of Clytemnestra's intention (K--ciom, 441) 27 to make the manner of
Agamemnon's death an intolerable burden on Orestes' life (Cho. 440-3). 28
Kittredge questions the notion that the angry ghost of Agamemnon
could be propitiated by sacrificing to him his own 'extremities' - it is 'too
absurd to be entertained' (1885, 156). Such npo't£A£ta need not, however,
be offered to the dead Agamemnon. Clytemnestra offered them to the
Erinyes (Eum. 106-9: cf Zeitlin 1965, 469 with n.13). It is significant
that one factor in the Erinyes' acceptance of Athena's offer is that they will
be offered the aKpo8ivta, the first fruits (or choice parts), of sacrifices on
behalf of children and marriage rites (Eum. 834-6). In this way Athena
institutionalizes, or rather rectifies, the sacrificial travesty performed by
Clytemnestra with Agamemnon's aKpOO'tllPHlO"!la'ta which were offered
to the Erinyes.
In the 'minds of the superstitious' (Rohde) maschalismos had another
purpose. Maschalismos was designed to render the murdered man weak,
59
Ruth Bardel
feeble and unable to avenge himself ('tva, <j>acri v, acr8Evi]~ yE.vol'to rcpo~ 'tO
avn'ticracr8at 'tOV <j>ovE.a, sch. Soph. El. 445). If the dead man is mutilated,
Rohde suggests, he will not be able to hold or throw the spear which in
Athens was borne before the murdered man at his funeral and was then
set up beside his grave - certainly for no other purpose than for supplying
the dead man himself with a weapon with which to take vengeance on
his own account. 29 Basically, a ghost without feet cannot walk, a ghost
without hands cannot hold a spear. Kerrigan states, 'Agamemnon has been
subjected to the fifth-century equivalent of being staked and wreathed
with garlic' (1996, 37). Murderers who wiped the il)strument of death on
their victim's heads - as Clytemnestra did in Sophocles' Electra - did so
in order to avert defilement (!-t:Ucro~) from the dead. This is similar to
sucking the blood of the murdered man three times and spitting it out
three times- as Jason does with Apsyrtus' blood, having mutilated ('tclf.LVE
8avov'to~) him. 30 Here too the aim is the catharsis of the murderer, the
expiation of an impious deed. As a concluding comment on his Appendix
on maschalismos, Rohde poses the following question, 'What "savage"
tribe ever had more primitive ideas or a more realistic symbolism than the
Greek populace - and perhaps not populace only - of classical times in
the sinister backwaters of their life into which we have here for a moment
descended?' If Rohde's question is trimmed of its notions of savagery,
crudity and depravity, leaving only the concept of realistic symbolism,
I believe that not only will Rohde's anguished quandary be answered but
also that the practice of maschalismos will reveal itself to be- in addition to
either a propitiatory sacrifice, an apotropaic gesture or a simpJe superstition
-central to the 'somatics ofDionysiac drama' (Zeitlin 1991).
For Aristotle ( GA 716a26-34), the fundamental difference between
male and female lies in their 'generative parts', that' is the female's
uterus and the male's testes and penis, what we would call 'primary sex
characteristics' or 'biological gender'. The male and female body are
further differentiated by their performative capabilities: in the pseudo-
Aristotelian Physiognomies (806b 32-5) we are told that 'Males are bigger
and stronger than females of the same kind, and their extremities are
stronger and sleeker and firmer and capable of more perfect performance
of all functions.' By 'extremities' Aristotle means arms, legs, hands and feet
-the 'instrumental parts' ('ta opyavtKa f.LEP11, Arist. PA 647b23) which are
distinguished by possessing the faculty to perform certain actions ('ta f.LEV
clVOf.LOtof.LEPll 't(\l ouvacr8ai 'tt TCOtEtV, GA 722b32). The 'extremities' are
also the 'moving parts' (KtV'll'ttKa f.LEP11) which, when differentiated into left
and right sides, privileged the right as the 'moving side' (oE~ta KtV'll'ttKa,
Problems 6.5). Zeitlin has shown how the 'dynamics of misogyny' (1978),
60
based on a series of antitheses - not dissimilar to that of Pythagoras -

arising from the polarization of male and female roles, operates within
the Oresteia.
According to Aristotle, castration31 of the male, or the removal of the
primary sex characteristic (an Aristotelian 'principle' or apx{J) results
in that male adopting the secondary sex characteristics of the female
(GA 716b): eunuchs, for example, become like women 'for they have
a woman's voice, shapelessness and the looseness of joints' (<J>rovilv 8llAUKllV
icrxouow Kat clJlOp<l>iav Kat avap8piav, Problems 10.36). The implications
of this deduction are clear: the female possesses aJ.Lop<J>iav and avap8piav
- shapelessness and jointlessness - her body is ill-defined and badly
articulated. Arid yet, tragedy presents highly articulate, in both senses of
the word, powerful, female figures who figuratively or literally emasculate
their males. The definitive description of Clytemnestra appears within the
first 15 lines of the Agamemnon when the watchman casts her as a 'man-
counselling woman' (yuvatKO~ avop6pouA.ov KEap, 11). This incorporates
the ability to plan and the efficient use of the body in bringing those plans
to execution and fulfilment.
While he was alive Agamemnon was a king, his hands wielded fate's
appointment and a staff which all men obeyed (pacrtA.Eu~ yap Ticr8',
o<J>p' E~ll~. IJ.L6ptJ.LOV Mxo~ 1tt1tAaV'tffiV I XEPOlV 1tEtcrippo'tOV 'tE PaK'tpov,
Cho. 360-2). The hands here denote power, authority and control: it is
Agamemnon's hand which holds the sceptre. In Agamemnon's absence,
however, this power has changed hands: Clytemnestra has usurped his
position, power and status and all but literally wields the sceptre. Long
before Clytemnestra mutilates Agamemnon's corpse, she has already
rendered him effeminate, broken him into little pieces and cast him
as a barbarian, a figurative emasculation dramatically enacted in the
tapestry scene. Clytemnestra's appropriation of male power - Kpci'to~ -
is portrayed by her annexation of the male political body in general and
in particular by her exploitation of that body's, literally and figuratively,
powerful right hand. Cassandra describes how Clytemnestra's hands keep
reaching forward with murderous intent (1tpo1:EivEt o£ xdp EK I XEPO~
6p£yJ.La'ta, 1110-11).32 Through the device of the net, Clytemnestra
constricts Agamemnon's movement, in particular that of his hands and
feet (1tEOa~ 'tE XEtpo'iv Kat 1tooo'iv, Cho. 981-2; c£ 998, 1000), thereby
ensuring that she gets the upper hand and then she stabs him with a two-
edged weapon (ooA.iql JlOPCJl OaJ.LEL~ <OclJ.Lap'tO~> I EK XEPO~ aJ.L<J>t'tOJlql
PEAEJlVql, Ag. 1495-6). Surely, Agamemnon's murder is enough- why
does she also mutilate him? Is it out of, as the epic sources suggest, sheer
personal anger?
61
Ruth Bardel
Bearing in mind Clytemnestra's depiction of Agamemnon as the

darling of any Chryseis at Troy (Ag. 1439) and of Cassandra 'rubbing
Agamemnon's erection (while) on the ship's benches' (vamiA.oov 8£
crEA.!lchoov icr01:pt~i]~, Ag. 1442-3), 33 castration is the perfect punishment
for a philandering husband as the Assyrian sources suggest. The focus
on Agamemnon's phallus and his 'sexploits' does rather more than reveal
Clytemnestra's familiarity with the 'scurrilous cant of sailors' (Tyrrell
1980, 45): it isolates a part of the husband's and king's body, his phallus,
which offends and outrages Clytemnestra and her status. By lopping off
Agamemnon's offending genitals she would be making a direct attack
on his status. As we have seen in the cases of Priam and Irus, castration
is the utmost affront, not only to a man's virility, but also to his n!lil·
This blatant or, as some commentators would have it, 'obscene' reference
to Agamemnon's phallus is perhaps only paralleled by Clytemnestra's
display of her breast to her son Orestes in the Choephoroi (896-8: cf.
Iliad 22.66-89), an act which is neither blatant nor obscene: 34 both
episodes isolate a part of the body which defines that body as definitively
male and female (respectively) despite that male's effeminization or that
female's masculinisation. In a thirteenth-century chronicle by Guido delle
Colonne called Historia Destructionis Troiae, Orestes actually cuts off
his mother's breasts with a sword and then hacks her to death (Kerrigan
1996, 167).
Earlier, I mentioned that the fundamental difference between maschalis-
mos and mutilation in general lay in the stringing together of the severed
parts and tying them around the neck and under the vic~im's armpits.
None of the commentators - ancient or modern- have offered a reason,
other than semantic, for tying the severed maschalismata under the victim's
armpits. A suggestion presents itself on the basis of an Aristotelian Problem
(13.8): 'Why is', he asks, 'the armpit the most unpleasant smelling region
(8ucrooMm:a'to~)' of the body? One of the reasons proposed is that the
armpit 'has neither movement nor exercise' (cbdvT]'tO~ Kat ayu11vacr'to~).
Unless they feature in a horror film, severed hands and feet and genitals
also have 'neither movement nor exercise' and by tying them on a string
under the armpit a symbolic ineffectuality is proposed - one which
corresponds to the initial incapacitation. 35 The stringing together of the
severed body parts and the tying of the rope under the armpits associate
maschalismos with the binding imagery in the Oresteia and is, I suggest,
a literal extension of Clytemnestra's initial restriction of Agamemnon's
hands and feet with the device of the net and in so doing demonstrates
her superior dexterity. Clytemnestra has carried out the lex talionis to its
extreme. She has stood over Agamemnon's corpse and uttered a vaunting
62
speech in the manner of a Homeric warrior and then she outrages her
enemy's body just as a Homeric warrior might, thereby seeking to prevent
posthumous vengeance by cutting off his hands and feet and to sever the
link between father and son by cutting off his genitals.
Mter Agamemnon's murder, Clytemnestra declares her murderous
right hand to be a just craftsman (Ag. 1404-6). Clytemnestra's masculine
dexterity and murderous skill are also implicitly contrasted with Aegisthus'
feminine idle hands: the Chorus accuse Aegisthus of not having the courage
to commit the murder with his own hand (1633-4), a cowardice whiCh
is in keeping with his characterization as an effeminate 'stay-at-home'
(Ag. 1625-7: c£ 1224-5). The dexterity of the active right hand is a male
attribute adopted by the man-counselling Clytemnestra. The phallus is
the only part of the king's and husband's body wh~ch Clytemnestra cannot
annex, usurp, or appropriate so she lops it off- along with the hands that
once held the sceptre, that sacrificed Iphigenia and the feet that conquered
Troy and trod the purple pathway. There is, it seems, every reason for
the reticence of Agamemnon's ghost in the Choephoroi: Clytemnestra
has beaten him hands down, in life as well as in death. If, as has been
argued, revenge is mimetic and reciprocal (Nussbaum 1986) and therefore
lends itself admirably to the dramatic art (Kerrigan 1996), then the
very dynamics of revenge (and of theatre?) are literally crippled by
Clytemnestra's mutilation of Agamemnon's corpse. The 'inability' of
Agamemnon's mutilated ghost to appear in the Choephoroi is rendered all
the more bitterly poignant by the unexpected appearance of Clytemnestra's
own ghost in the subsequent play, the Eumenides, to exact revenge. One
can only conjecture that the castration of the king, the mutilation of the
body politic -Agamemnon's figurative and literal dismemberment - was
more than compensated for in the satyric Proteus with its ribald humour
and inevitable reinstatement and reassertion of the phallus. 36
Notes
1 See, for example, Stewart 1997, Montserrat 1998 and Bremmer 1991.
2 On eunuchs in ancient Greek drama, see Guyot 1980, 71-2 and Hall1989,
157-8.
3 Revenge and mutilation are intimately bound up with the failure of Agam-
emnon's ghost to appear in the Choephoroe. On this issue, see Sommerstein 1980
and Bardel1999, 120-50.
4 Driver and Miles 1935, 385.
5 Driver and Miles 1935, 389, my italics.
6 Peleus cuts out (ektemno) and keeps the tongues of the animals he slew in
a hunt as evidence of his prowess (Apollodorus 3.13.3). On the defeated warrior's
63
Ruth Bardel
head as a trophy see Vermeule 1979, 107-8 (with fig. 24).

7 Vernant 1991, 71 with n. 37 whose footnote to this passage states, 'We will
pass over the problems of maschalismos, for which one should consult E. Rohde,
Psyche ... these problems occupy another level of analysis which will be the subject
of a future study.' I have been unable to find this 'future study'.
8 Hecabe accuses Polymestor of rending asunder and cutting Polydorus' limbs
with the iron sword (Eur. Hec. 716-17. C£ 1076-8). Schlesier (1989, 117-19)
proposes a ritual sparagmos in this context, a notion disputed by Zeitlin (1997,
182 with n. 29) and Mossman (1994, 168). It is doubtful that maschalismos is
implied: see further Bardel1999, 205-6.
9 In an article (The Guardian, 8/9197) by Gordon Burn called 'Secret Lives,
Secret Deaths' on the recent case of the West murders at Cromwell Street,
Gloucester it was reported that one of the Wests' victims' bodies, found after 21
years, had certain bones missing, especially from the feet and hands and several
fingers and toes. The kneecaps were also missing. Apparently, West had the
space to bury the body intact without mutilating it. The mutilation was, Burn
suggested, 'done for compulsive and obsessive reasons'.
10 Despite this, poetic logic permits the 6vci.pcov f.!Op<j>rof.Lacnv (Ag. 1218)
of the slaughtered children to hold their half-eaten entrails in their hands

(Ag. 1219-22).
11 This is in explicit contrast to the young man's kalos thanatos: the dead youth
is still beautiful despite the fact that his body is mangled by the sharp bronze (If.
22.71-6). See also Tyrtaeus (fr. 10.24-5) where it is a foul thing for the old to die
before the young, 'breathing forth his [the older man's] stout soul in the dust with
his genitals all bloody in his hands'. See further Richardson 1993 on this passage,
Kirk 1985 on Iliad 2.261-4 and Janko 1992 on Iliad 13.568.
12 Another fragment of the same play (fr. 623) mentions something being 'full
of maschalismata' (1tA:r1pTJ f.LaaxaA.wf.L<hcov) - that is, full of body bits (f.Lopw)

that were lopped of£
13 On Sophocles' fragmentary Polyxena, see Bardel1999, 232~1.
14 Lys. AgainstAndocides, 14.41-2; Thuc. 6.27.1; Plut. Nic. 13.2 andAle. 18.3,
19, 20.5. Like Henderson (1987), Hatzfeld (1940, 161) cites Aristophanes'
Lysistrata (tmv Epf.LOKomorov (the herm-defacers/castrators), 1093-4) as evidence
that the phalloi of the Herms were cut off. The Lysistrata was produced at least
four years after the event and its reference to the mutilation of the Herms is, in
Henderson's view, a clever way to motivate the covering up of phalloi because
serious matters are to be discussed. At 1093-4 the Spartan ambassadors - with
huge erections (because of the sex strike) take off their coats and the Chorus leader
says, 'I wouldn't do that if I were you. The herm-choppers might catch sight of
you.' See also the unidentified male who castrated himself- with a stone - on
the altar of the 12 gods (Plut. Nicias 13.2) and a fragment from Phrynichus the
comic poet where a character addresses either Hermes himself or an effigy saying,
'Dearest Hermes, don't you fall too and get yourself castrated' tfr. 58).
15 See further Rohde 1925, Appendix II; Kittredge 1885, 155-6; ]ebb on Soph.
El. 445-6 and excursus p. 211.
64
16 Resonances of this are found in the Odyssey: Penelope tells the 'bold and
shameless' handmaid that her outrageous conduct is clear and that with her 'own
head she will wipe out its stain' (19.92). See Rohde's Appendix II for further
references. On this gesture in relation to Apsyrtus' murder, see Bremmer 1997.
17 So Photios, the Suda, Hesychius and later scholars such as Rohde 1925,
Appendix II on 11acrxaA.tcr116<;.
18 As Rohde 1925, Appendix II, notes, verbs ending in -ii~etv which are derived
from the names of parts of the body, can denote according to the circumstances
the utmost variety of actions done to or with the part of the body concerned: it
would, therefore, be feasible to assume that maschalismos simply meant to tear
the arm from the shoulder at the arrripits- as Benndorf 1895, 132 suggested.
Benndorf cites a sculptured relief on which the gods appear to be tearing out the
right arm of their defeated enemy as an example of maschalismos. In response to
Benndorf, Rohde asks whether or not the Greeks would have really attributed
this much-execrated practice of cowardly murderers to the gods. One example
from Homer seems to suggest that the gods might be capable of just such a thing:
Ares in the Iliad (5.842-4) is caught red-handed- literally and figuratively-
by Athena, stripping a mortal's corpse of its armour. Given Ares' mindless and
outrageous battle tactics, which Athena is sent to check, what might he have
done to the corpse of 'huge Periphas' had Athena not intervened? Besides, the
very physical injuries the gods inflict on each other - castrations, blindings etc.
- speak for themselves. A late-archaic engraved gem is interpreted by Richter
1968, no. 742 as showing Athena in combat with the giant Enkelados: 'She has
wrenched off his arm and is running away.' On the motif of Greeks maltreating
Trojan women and children, see the panels of the Mykonos Pithos discussed
by Vermeule 1979, 114-15.
19 Vermeule 1979, 222 n. 16, observes that similar practices have been reported
in Algeria and Vietnam: elsewhere she notes that 'placing a man's severed genitals
in his mouth, as practised in the Algerian and Vietnam wars, is not a Greek
method of humiliation, unless it enters into the treatment of the body in
maschalismos' (234 n. 18). C£ Shay 1994 and Trittle 1997.
20 Vermeule 1979, 236 n. 30, notes a similar practice from Assyria: 'Esarhaddon
caught Abdimilkutte king of Kundi and Sizu like a bird in the mountains and
cut off his head: "(Then) I hung the heads of Sanduarri and of Abdimilkutte
around the necks of their nobles/chief officials to demonstrate to the population
the power of Ashur, my lord, and paraded (thus) through the wide main street
ofNineveh with singers (playing on) sammu-harps," ANEP 291'. Heracles cuts
out (apotemn6n) the ears, noses and hands of Erginus' heralds, ties the body
parts around their necks and sends this as a tribute to Erginus (Apollodorus
2.4.11).
21 See West 1966 on Th. 179-80 and Vlahogiannis 1998, 22 on the superiority
of the right hand. Vlahogiannis also notes that slaves may be mutilated but that
the 'integrity of the [Athenian] citizen body had to be kept intact' (24).
22 See Rose 1958 on 185ff. and Sommerstein 1989, on 186-90; the Greeks
associated aKprovia with Persian/barbaric practice (Hdt. 3.69.5, 3.118.2, 3.154-5,
65
Ruth Bardel
9.112.1; Xen. Anab. 1.9.13) see Hall1989, 25-8 and 204-5.

23 In one Etruscan fresco depicting Agamemnon in the Underworld (along
with the ghost of Teiresias), Agamemnon is bandaged: 'Autour de la poitrine,

il porte- comme d'autres heros morts de mort violente (ainsi Ajax et Patrocle)
- un bandage sur la blessure martelle que lui ont faite ses assassins' (Knoepfler
1993, 85 with fig. 68).
24 Kittredge 1885, 151-9, lists and analyses all of these sources; Rohde 1925,
Appendix II.
25 See Parker 1983, 220 with n. 71 for a list of such theoi apotropaioi, which
can include the dead and Apollo. On maschalismos, its etymology and ritual
see Parker 1984.
26 See Rohde for a more detailed analogy between ~acrxaA.icr~ata and ro~o
eet01)~eva. In their commentary on the Odyssey, Russo and Galiano 1992, on

22.474-7 also draw an analogy between ~acrxaA.tcr~o~ and the ritual ro~oeeteiv,
adding that the throwing of a man's genitals to the dogs to be eaten parallels
what was done with the animal's entrails during sacrifice. Such practices were
apotropaic, 'designed to ward off the posthumous vengeance of the victim'.
27 The word Kticrat is an interesting one to use in this context. As Rose 1958,
on Cho. 483-4 notes, 'one can Ktisetv a religious institution (here [Cho. 484]
something in the nature of hero-cult, or at least the tendance of a venerated
family ghost) as well as a city or the like'. Rather than instigating a hero-cult,
Clytemnestra lops it off at source.
28 Rose 1958, on 441-2 points to Orestes' own lack of resources and the
impotence of his father's ghost as being consequent to Orestes' exile and

Agamemnon's mutilation.
29 Such an interpretation would resonate with the alleged Australian aboriginal
practice of cutting the thumb off the right hand of a dead enemy in order that he
-or his ghost- may no longer be able to hold a spear (Rohde). Hughes 1991,
Appendix A, 194-8, cites the practice of maschalismos as a possible explanation
for the cut marks on bones found in graves. ,
30 Ap. Rh. Argonautica 4.447-80: the scholia glosses this with aKprotl]ptasnv.
31 For castration, Aristotle uses the verbs EK:te~vro (GA 716b; Problems 10.36)
and 1tT]p6ro (Problems 4.3, 4.26, 10.8).

32 The benign or malign intervention of a known or unknown agency can be
expressed as 'seeing someone's hand in the matter'. Zeus' influence, the stroke of
his hand, can be traced in Troy's defeat (Ag. 367-8: compare 663; Cho. 394-5).
Orestes' own hand is guided by Justice and the gods (Cho. 948; cf. 435-7) as
he exacts vengeance for his father's death. Agamemnon who endured much in
a woman's cause loses his life at the hands of a woman (Ag. 1453-4) who claims
to be the instrument of the 8ai~rov (Ag. 1500-4).
33 See Young 1964, 15; Koniaris 1980, 42-4; and Tyrrell 1980, 44-6. Might
we also see, in relation to Agamemnon's extra-marital activity, another instance of

his wasting, along with the tapestries, of the house's goods? '
34 Vermeule 1979, 236 n. 30 suggests that the moria, including the genitals,
hung around the dead man's neck would be 'natural aKpro'tl]ptacr~ata toii veKpoii
66
(Hesychius), corresponding to the mutilated breasts of the female victims as in the

tale ofMasistes' wife' in Herodotos (9.112). As Segal points out, 'Clytaemnestra's
display of the breast to Orestes builds upon earlier inversions or negations of the
life-sustaining functions of the house' (1985, 18).
35 Prometheus is chained to the rock by the will of Zeus and the hand of
Hephaestus (Aesc. PV 619) with a band that passes over his chest and under
his armpits (J.LacrxaA.tcr'tfipw;, 70-1) - an outrage (aiKciac;, 179). This binding
takes place at the edges of the earth and is of especial grief to Prometheus for,
as Hephaestus says, he can neither hear a man's <1Jrov1,v nor see a J!Op$1,v ~po'trov
(21-2). Thus the binding simultaneously incapacitates and isolates Prometheus
from the community of mortal men. In Herodotos (1.215) the golden belts worn
by the Amazons are called MacrxaA.tcr'tfipac;. In Strabo (11.8.6), however, the same
term is used of the golden harnesses which passed round the horses' shoulders
and were fastened to the yoke by the lepadnon. One reason for this dual use of the
word may be suggested by the symbolic harnessing of the wild Amazons.
36 Hall 1989, 37 states: 'satyr drama offers the insurance of a reaffirmed sense
of unindividuated masculinity, based in libidinal awareness, in order to protect

against the painful "feminine" emotions which tragedy has unleashed'.
Bibliography
Arthur, M.
1982 'Cultural strategies in Hesiod's Theogony: law, family, society', Arethusa
15,63-82.
Bardel, R.
1999 Casting Shadows on the Greek Stage, unpublished doctoral thesis,
Oxford.
Benndorf, 0.
1895 Das Monument von Adamklissi, Vienna.
Bremmer,].
1991 'Walking, standing and sitting', in J. Bremmer and H. Roodenburg
(eds.) A Cultural History of Gesture, Oxford, 15-35.
1997 'Why did Medea kill her brother Apsyrtus?', in J.l. Clauss and S.l.
Johnston (eds.) Essays on Medea in Myth, Literature, Philosophy and
Art, Princeton.
Devereux, G.
1976 Dreams in Greek Tragedy, Oxford.
Driver, G.R. and Miles, ].C. (eds.)
193 5 The Assyrian Laws, Oxford.
Gager,J.G.
1992 Curse Tablets and Binding Spells from the Ancient World, Oxford.
Garvie, A.
1986 Aeschylus Choephoroi, Oxford.
Guyot, P.
Stuttgart.
67
Ruth Bardel
Hall, E.
1989 Inventing the Barbarian, Oxford.
1998 'Ithyphallic males behaving badly; or, satyr drama as gendered tragic
ending', in M. Wyke (ed.) Parchments of Gender: Deciphering the bodies
ofantiquity, Oxford.
Hartog, F.
1998 The Mirror ofHerodotus: The representation of the other in the writing of
history, Berkeley and London.
Hatzfeld, J.
1940 Alcibiade: etude sur l'histoire d'Athenes a la fin du Ve siecle, Paris.
Henderson, J.
1987 Aristophanes. Lysistrata, Oxford.
Hughes, D.D.
1991 Human Sacrifice in Ancient Greece, London and New York.
Janko, R.
1992 The Iliad: A Commentary. Vol IV (Books 13-16), Cambridge.
Kepple, L.R.
1976 'The broken victim: Euripides' Bacchae 969-70', HSCP 80, 107-9.
Kerrigan, J.
1996 Revenge Tragedy: Aeschylus to Armageddon, Oxford.
Keuls, E.C.
1985 The Reign ofthe Phallus, Berkeley.
Kirk, G.S.
1985 The Iliad: A Commentary, Cambridge.
Kittredge, G.L.
1885 'Arm-pitting among the Greeks', A]P 6, 151-9.
Knoepfler, D.
1993 Les Imagiers de l'Orestie. Catalogue d'une exposition creee alf musee d'art et
d'histoire de Neuchatel. Novembre 1991-Fevrier 1992, Kilchberg.
Koniaris, G.L.
1980 'An obscene word in Aeschylus (I)', A]P 101, 42-4.
Lawson, J.C.
1910 Modern Greek Folklore and Ancient Greek Religion: A study in survivals,
Cambridge.
Miller, M.C.
1997 Athens and Persia in the Fifth Century BC, Cambridge.
Montserrat, D. (ed.)
1998 Changing Bodies, Changing Meanings. Studies on the human body in
antiquity, London.
Mossman,].
1994 Wild justice, Oxford.
Murnane, W].
1980 United with Eternity: A concise guide to the monuments,ofMedinet Habu,
Chicago and Cairo.
Nussbaum, M.C.
1986 The Fragility of Goodness, Cambridge.
68
Parker, R.
1983 Miasma, Oxford.
1984 'A note on cj>ovo~, 8ucria and
f.LacrxaA.tcrf.Lo~', LCM9.9 (Nov.), 138.
Rabinowitz, N.S.
1981 'From force to persuasion: Aeschylus' Oresteia as cosmogonic myth',
Ramus 10, 159-91.
Redfield, ] .M.
1975 Nature and Culture in the Iliad, Chicago.
Richardson, N.
1993 The Iliad: A commentary, Vol. VI (Books 21-24), Cambridge.
Richter, G.M.A.
1968 The Engraved Gems of the Greeks, Etruscans and Romans. Part I,
London.
Rohde, E.
1894 Psyche: Seelencult und Unsterblichkeitsglaube der Griechen, Tubingen.
1925 Psyche, trans. WB. Hillis, London.
Rose, H.].
1958 A Commentary on the Surviving Plays ofAeschylus, Vol II, Amsterdam.
Russo,]., Fernandez-Galiano, M. and Heubeck, A. (eds.)
1992 A Commentary on Homer's Odyssey, Vol III (Books XVII-XXIV), Oxford.
Schlesier, R.
1989 'Die Bakchen des Hades: Dionysische Aspekte von Euripides' Hekabe',
Metis 3, 111-35.
Segal, C.
1985 'Tragedy, corporeality and the texture oflanguage: matricide in the three
Electra plays', CW79.1, 7-23.
Shay,].
1994 Achilles in Vietnam: Combat trauma and the undoing of character, New
York.
Sommerstein, A.
1980 'Notes on the Oresteia', BICS27, 63-75.
1989 Aeschylus' Eumenides, Cambridge.
Stewart, A.
1997 Art, Desire and the Body in Ancient Greece, Cambridge.
Trittle, L.A.
1997 'Hector's body: mutilation of the dead in ancient Greece and Vietnam',
The Ancient History Bulletin 11.4, 123-36.
Tyrrell, WM.
1980 'An obscene word in Aeschylus (II)', A]P 101, 44-6.
Vermeule, E.
1979 Aspects ofDeath in Early Greek Art and Poetry, Berkeley.
Vernant, J.-P.
1991 'A "beautiful death" and the disfigured corpse in Homeric epic', in
F.l. Zeitlin (ed.) Mortals and Immortals: Collected Essays, Princeton,
50-74.
69
Ruth Bardel
Vlahogiannis, N.
1998 'Disabling bodies', in Montserrat (ed.) Changing Bodies, Changing
Meanings, 13-36.
West, M.L.
1966 Hesiod's Theogony, Oxford.
Winnington-Ingram, R.P.
1948 'Clytemnestra and the vote of Athena,]HS 68, 130-47.
Young, D.C.
1964 'Gentler medicines in the Agamemnon', CQ 14, 1-23.
Zeitlin, F.I.
1965 'The motif of the corrupted sacrifice in Aeschylus' Oresteia', TAPA
96, 463-508.
1978 'The dynamics of misogyny: myth and mythmaking in the Oresteia',
Arethusa 11, 149-81. Repr. Zeitlin 1996, 87-119.
1991 'Euripides' Hekabe and the somatics of Dionysiac drama, Ramus 20,
53-94. Repr. as 'The body's revenge: Dionysos and tragic action in
Euripides' Hekabe', in Zeitlin 1996, 172-216.
1996 Playing the Other, Chicago and London.
70
4
SACRED EUNUCHISM IN THE CULT OF

THE SYRIAN GODDESS
].L. Lightfoot
The Dea Syria, or LUpia 8£6<;, is the name given to the north Syrian
goddess Atargatis by her classical worshippers. She began as the city-
goddess of Hierapolis (the Greek name of the Aramaic town of Manbig,
or Bambyke), though from the Hellenistic period her cult was carried
across the Mediterranean to the Aegean islands, mainland Greece and
Macedonia, Egypt, and finally, under the empire, to the Danube provinces,
(possibly) Gaul, and Britain. Of course, she was long established in Syria
itself, where her cult was disseminated, brought into contact with related
cults, and fractured into a thousand minutely differentiated local forms. 1
Although it is never possible to say exactly what a deity - any deity -
'really is', we can at least say the following about Atargatis' character.
Her iconography shows her enthroned in state between lions, which she
had inherited from earlier Syro-Anatolian goddesses and developed in
tandem with Cybele, and she wore a mural crown when represented as
a city-protecting deity. 2 She was partnered by Hadad, the west Semitic
thunder god, though she had greatly eclipsed him in importance. She was
especially associated with springs of water; fish were sacred to her, and
various ceremonies in her cult involved processions to and from water. In
short, she was a beneficent, protective deity, a goddess who could double
as the Tux11 IIoA.£coc;, or protecting Fortune of the City. She had militant
characteristics, but also nurturing ones; the combination is common
throughout Greece, Anatolia, and the Near East. She is addressed, like
many Anatolian and Levantine deities, as 'mistress' - Mcrnowa, K:upta -
and the sources sometimes present her relationship to her worshippers as
that of a mistress towards slaves, 3 imperious yet magnanimous, powerful
but merciful.
Among these 'slaves' by far the most distinctive group are the goddess'
self-castrated eunuch devotees. They are known mostly from Graeco-
Roman sources, literary and epigraphical, though they are also mentioned
71
JL. Lightfoot
in a Syriac text. For details we are wholly dependent on the former, which
portray them as attached to her service, but distinct from the college of
priests in her temple. They are depicted as ecstatics, associated with loud,
hypnotic music, auloi, cymbals, and tympana, which work them into an
anaesthetic frenzy. To the amazement and terror of onlookers they wound
themselves with knives, lash themselves and each other, and new devotees,
caught up in the general hysteria, perform the ultimate in consecration to
the goddess: the act of self-castration. Thus irrevocably tied to the service
of their divine mistress, they dress as women and live by begging, touring
towns and countryside and soliciting money and goods from the public
by their awe-inspiring displays.
The background to these practices is almost irrecoverahly murky. Earlier
literature often contains the claim that sacred eunuchism is typically
'Semitic': Graillot said so explicitly, and Robertson Smith, who did more
than anyone else to form the very concept of'Semitic' religious institutions,
regarded the sacramental shedding of one's own blood as one such ethnic
practice. 4 This view also points to the presence of effeminate or transvestite
priests in the high civilizations of Mesopotamia, where cultic personnel
of ambiguous gender are associated with lnanna/lshtar in Uruk, 5 and
to various sorts of gender inversion that classical sources connect with
Syro-Palestine. 6 But precise parallels are hard to find, or evanescent.
With the Mesopotamian material one is often unsure whether or not any
given category of priest was a eunuch (let alone a self-castrate), while the
nature of the historical connections (if any) with northern Syria remain
unelucidated. With the west Semitic material one is usually, dealing, not
with eunuchism itself, but with categories - such as 'sacred prostitution'
- that are supposed to be related to it, but bring formidable difficulties of
their own.? In these areas one encounters a powerful divin~ figure, a great
goddess with the power to change human gender; 8 but a similar figure is
connected with the eunuchs oflndia, 9 and she seems too vague or generic
to be useful in establishing precise historical relationships.
Another context for sacred eunuchs is Anatolia, where they are variously
attested. Not all are called galli, the word that will become standard in
Graeco-Roman sources for religious eunuchs and eventually for eunuchs
in general, nor are all associated with self-castration. They range from
the Euvouxot who, along with other Oll!J.Ocrtot (public servants), tend the
gardens in the sanctuary of Hecate in Lagina, 10 to the dignified figure
of the Megabyzus at Ephesus, who held office along with a maiden.U
The most celebrated case is the cult of Cybele, where Classical sources
repeatedly connect her eunuch galli 12 with the Phrygian 'temple state'
of Pessinus, though far and away the bulk of the evidence comes from
72
Sacred eunuchism in the cult ofthe Syrian goddess
her western, imperial, diaspora. Cybele will figure prominently in what

follows. With all these cults the main difficulty is that none of the evidence
emerges before the advent of Greek and Roman literature and epigraphy,
in and after the Hellenistic period. The curtain rises at a late stage, and we
are left to guess what, or how much of the action, has already occurred.
Of our sources for the sacred eunuchs in the cult of Atargatis, the two
most important are literary. They complement each other. The first is the
De Dea Syria (henceforward DDS), an ethnography of Hierapolis in the
Herodotean manner, which is ascribed to the second-century Greek rhetor
and satirist, Lucian. Lucian was a native of Samosata (itself not far distant
from Hierapolis), and Aramaic may even have been his native tongue;
but in adulthood he was fully assimilated into classical literary culture.
Over a quarter of DDS is devoted to the goddess' eunuchs, who are called
yaAA.ot: it purports to give their charter-myth (§§ 19-27), and describes
their part in a Spring Festival which is allegedly the most important event
in Hierapolis' sacred calendar. The second is an episode in a novel whose
Greek original is not extant, but which survives in Apuleius' Metamorphoses
and Lucius, or the Ass (the latter again ascribed to Lucian). Its hero, in
the form of an ass, falls into the hands of a pack of mendicant eunuch
devotees, who are not called galli but priests. Whereas DDS focused on
the galli in relation to the temple, this text follows the eunuchs on their
wanderings throughout rural Macedonia; at one point the city of Beroea
is named, and there is also inscriptional evidence of Atargatis' cult there. 13
Third, to substantiate these classical sources, a Syriac text mentions the
existence of self-castration in Edessa in honour of the goddess 'Tar'atha'
(Atargatis). This is set in the time of king Abgar, who is supposed to have
ended the practice on his conversion to Christianity; but the formulation
of rules against it still in the fifth century suggests that the announcement
of its demise in the second was a little premature. 14 Fourthly, and our
only epigraphical testimony, is an inscription from Rome to which gallo
diasuriaes is a later addition. 15 But the man is a Roman citizen, which
(given that Roman law forbade human castration) militates against
his being a eunuch. To these testimonia can be added: fifth, a set of
inscriptions from the Beqaa valley (Lebanon) recording how many satchels
of alms have been collected in the course of his wanderings by a certain
Lucius, who describes himself as a oouA.o<; of the 'Syrian goddess of the
Hierapolitans' .16 A pierced hemispherical container, dedicated 't'ij Kupic;x
A'tapyan, may be the lid of a begging-bowl. 17 Neither inscription explicitly
mentions galli, and perhaps some caution is in order before assuming
(with Cumont and others) that eunuchs were indeed involved, for there
is little reason to make mendicancy their exclusive prerogative. But we
73
JL. Lightfoot
do gain some welcome context for the general scenario in Lucian and
Apuleius. Sixth and last, sources for the Sicilian slaves' revolt at the end of
the second century BC speak of their charismatic Syrian ringleader, Eunus,
from Apamea, who claimed to be inspired with dreams, prophecies, and
religious frenzy by the Syrian goddess. 18 All this behaviour is also ascribed
to the galli, whom perhaps he was emulating, but no source suggests that
he himself was castrated.
That means that our two best witnesses are also the two most 'literary',
and those hardest to take at face value. The prime difficulty about DDS
has always been taken to be the issue of its authorship. If, the argument
goes, it really is the work of the satirist Lucian, then we can no more rely
on any of its contents than we can on any of his other 'satirical' works,
such as the True Stories. In his certainly authentic works, Lucian was
a committed opponent of all religion, who draws upon established literary
and philosophical traditions to lampoon superstition and flummery. But
DDS itself is remarkably opaque, and it is difficult to say whether the belief
in Lucianic authorship is effect or in fact cause of its being read as satire.
In my forthcoming commentary I intend to argue for Lucianic authorship,
of which I a1Tl almost convinced on linguistic and other grounds; but I do
not think that the treatise must, for that reason, be satire. The difficulties in
evaluating DDS inhere rather in its genre, which is a pastiche of the fifth-
century Herodotean manner. The pastiche humorously exaggerates this
manner, sometimes approaching 'parody', but that is a word I avoid. Parody,
at any rate ancient parody, usually implies humorous transplantation into
an incongruous new context, whereas this is ethnography :ij.rmly on its
own ground. That is where the problems lie, and they are essentially those
that bedevil all ancient ethnography: the presentation of the subject-matter
as 'exotic', the failure to view the subject on its own terms (is that ever pos-
sible?), and- despite the 'exoticism' of the subject- its mediation through
a Greek (or Graeco-Roman) interpretative framework. Additional problems
here include the gentle fun at the expense of the Herodotean narrator,
whose foibles are here raised to a high degree: this is not, I maintain, 'satire'
of the cult, but it might compromise the text's evidential value.
So much to understand the literary context and positioning of the
galli in DDS. They are presented as the exotics par excellence in an exotic
cult. On 'certain days' in Hierapolis' spring festival, they gather outside
the temple, where, to the throb of drums and drone of auloi, they gash
their arms and beat each other, and onlookers are impelled to join in: the
young man 'to whom this fate has been allotted' comes into the ring, seizes
a sword which is lying at the ready, and- excises his genitals. He runs
through the streets with the severed objects, and receives female clothing
74
from the owners of whatever house he throws them into (§§50-1). Is

this horror or farce? How would we know? Do we make the best of it,
and point out (for example) that a few other sources speak of public
self-castrations, or that they parallel the use of knives (rather than flint,
the traditional substance for ritual woundings), or that a story about the
emperor Elagabalus (AD 218-222) seems to imply that he, too, threw
severed genitals - into a temple? 19 Or should we be more fundamentally
suspicious? Much of DDS can be substantiated, especially on matters of
iconography, and it does not usually engage in random fantasy. Fantasy,
especially the bizarre elaboration of the mundane, is what we find in
Lucian's True Stories; but the procedure of DDS is different. It tailors
whatever the 'reality' was to suit the ethnographical genre's expectation, if
not requirement, of marvels, 8m)f!a'ta. It is an isolated oddity, presented
without gloss or rationale, and the narrator does not even tell us in whose
honour was celebrated the festival in which this ritual took place. It fosters
the impression that the galli simply acted in a state of unthinking frenzy:
Apuleius (below) will provide a slightly different picture. And yet, despite
his refusal to interpret or explain, the narrator of course does offer an
interpretation of sorts. He speaks of auloi, tympana, inspired songs, the
cutting of arms. All this (and the language in which it is couched) orients
us towards the galli of Cybele - of whom virtually identical things are
predicated. More of this below.
The problems with our second main literary source, the novel of
Apuleius and Lucian, are of a different kind. This does not even make
ethnography's pretence of objectivity. Most of our classical sources for
religious eunuchs or galli fall into the categories of satire Quvenal, Persius,
Martial) or polemic (the Christian apologists), and this one belongs firmly
to satire. It is a wholly negative portrayal of the Syrian goddess' charlatan
'priests'. Metamorphosed into an ass, the hero is sold into the hands
of a party of eunuchs, 'those who carry the Syrian goddess around the
villages and countryside and force her to beg for alms' (Luc. Asin. 35).
He accompanies them as they tour Macedonia, dressed flamboyantly as
women, and at each stopping-place getting up to the same displays of
self-mutilation as they put on in Hierapolis, except that here they compel
their awe-struck audience to make offerings of money and provisions.
But when they are alone, they get up to the sort of vice associated with
them in the popular imagination. They kidnap strapping young villagers
and force them into acts of unspeakable degradation. Indeed, in donkey
form, Lucius fears for his own virtue. He finally escapes when they are
apprehended for stealing a golden goblet from another temple. They are
imprisoned; and Lucius is sold to a new master.
75
JL. Lightfoot
This is a document illustrating popular impressions and prejudices
about the galli. They are pathic homosexuals, extortionists, thieves, quite
without shame even when confronted with the evidence of their crimes.
The effect depends on its exploitation of familiar scenes and situations and
its reinforcement of popular prejudice; in other words, it is a collection of
topoi. We cannot take the descriptions of what the galli get up to behind
the scenes as evidence for anything very much. But the episode does
suggest that the sight of these beggars was familiar, that the audience might
at least be expected to recognize the milieu evoked by the author. Indeed,
the basic situation, in which a donkey bears divine images for a company
of mendicants on their progress through the countryside, reflects both
a well-known proverb about 'the ass that carries the mysteries', and also
a story which does the rounds in the writers of fables. 20 Here, a company
of mendicants wear out their beast of burden and then use its hide to
make tambourines, so that in death it has to withstand even stronger
blows than in life (was the author of the novel trading on his audience's
expectation that this would be a likely end to Lucius?). The sources speak
variously of 1-lTJ'tpayup'tat, n:iUot ayup'tat, and Galli Cybebes, and connect
them explicitly with Rhea and Attis.
Despite its negative presentation of the eunuchs, the novel does
contain a wealth of rich and evocative details, from which, for reasons
of space, I shall select just a couple. First, it is interesting that the self-
wounding is not presented simply in terms of anaesthetic unreflecting
frenzy, but, in the case of at least one man, precisely as a punishment
for religious disobedience and transgression, rather like the B.agellants of
early Christianity (Apul. Met. 8.27-8). Second, neither novelist calls the
eunuchs 'galli'. They usually call them 'pathics' (Kivatoot/ cinaedi), and
both use 'priests' (t£p£tc;/ sacerdotes) in a scurrilous epis~de when they
want to expose their hypocrisy. Once Apuleius also uses semiviri, Lucian
yuvatKiac;; Lucian also uses ayup-rat. But what seems to me worthy of
note is the way they speak to each other. Not only do they use a high
pitch, which is in fact characteristic of the way classical sources represent
effeminate speech. They also use feminine appellatives (Kopaata, puellae);
and they use feminine parts of speech of themselves and to each other
(Lucian Asin. 36 aau-rfj ... A-a~ouaa; Apul. Met. 8.25.4 misera; 26.1
mercata), while the narrative itself refers to them in the masculine.
They may also have assumed women's names. The subject of an epigram
by Philodemus about a devotee of the Mother of the Gods is called
Tpuyovtov, 'little dove' (Garland 3320-7 == AP 7.222),' which coheres
extremely neatly with the way the galli refer to themselves as palumbulae
in Apuleius (Met. 8.26.4). Other authors, speaking of galli in the
76
third-person, sporadically use the feminine, as Catullus does of Attis in

poem 63, or Philodemus throughout the epigram mentioned; the same
is sometimes true of references to homosexuals in classical authorsY But
it is the use of the feminine by the eunuchs themselves that catches the
attention. That would be an interesting socio-linguistic datum, provided
the representation of their speech is accurate. It might be. The hijras of
modern India are also reported to use the feminine of themselves and to
each other. 22 In the Hippocratic tract Airs, "Wtlters, Places, the Scythian
effeminates called the enarees also 'speak like women' (Hipp. Aer. 22),
but here we do not know whether the remark extends further than pitch
and vocal quality; in any case, there is no explicit indication that these
Scythians were castrated. Some of the Mesopotamian priests used the
Sumerian literary dialect Emesal, otherwise used to represent the speech of
goddesses. But this is in the context of ritual activity, and is not a datum
about their day-to-day speech.
Despite the very different genres to which our two main literary sources
belong, their differing emphases and interests, and the individual details
that emerge in the course of them, what they have overwhelmingly in
common is that their portrayal of the eunuchs is extremely dose to that
of Cybele's galli. The two can hardly be treated in isolation. DDS uses
yale/co~, the standard word for religious eunuchs (though the novel does
not). They are set in the same two basic contexts, those of (i) sacred
mendicancy, carrying round a divine statue on a donkey, and (ii) a major
spring festival, in which a day is set aside for their bloody devotions - in
Cybele's case, the notorious dies sanguinis, day of blood. 23 They have the
same techniques of inducing ecstasy, with auloi, tympana, and cymbals
(DDS §44 also mentions KPO'ta/ca, rattles, a characteristically Cybeleid
and Dionysiac instrument); their performances are described in analogous
terms (Apuleius, Met. 8.30.5, even calls 'Phrygian' the songs of the Dea
Syria's eunuchs). Both have knives and whips. 24 Both are transvestites, to
whom female appellatives are applied. Other stereotypical situations crop
up in the novel which are part of the stock-in-trade of writers on galli in
general. Lucian omits, but Apuleius includes, an episode in which they
go on tour with a universally applicable fake prophecy, and in which they
extort offerings and gifts by means of phoney warnings. Roman satirists
contain similar reports. 25
This state of affairs could be explained in a number of ways, not
necessarily mutually incompatible. The approach which has prevailed in
the scholarship hitherto has been to look to common origins. The likeliest
candidate is adjudged to have been Kubaba(t), the Hittite and neo-Hittite
city-goddess of Carchemish, 26 whose cult radiated out north and west
77
JL. Lightfoot
across Anatolia, and in whose debt both Atargatis and Cybele stood,
to varying degrees. In both cases we can point to likely iconographical
borrowings, although recent studies of Cybele have powerfully challenged
the thesis that her name itself reflects Kubaba'sY The difficulty is that
there is practically no evidence either for or against the presence of sacred
eunuchs in Kubaba's cult. It is true that the hero of DDS's aetiology
of sacred eunuchism in Hi erapolis is called 'Combabos', a name which
etymologically denotes him her servant (§§ 19-27). But the story itself has
demonstrable analogues in stories about secular eunuchs in royal courts,
and, whatever else it is, is not a i£po~ A6yo~ reflecting cultic realities in the
religion of Kubaba. Even if we concede the presence of sacred eunuchs in
the cult of Kubaba as a likelihood, this has only limited explanatory value.
First, Lynn Roller has recently argued for minimising the influence of
Kubaba on Cybele (1999, 44-53). Second, sacred eunuchs are (as already
noted) attested in many more cults in Anatolia: is Kubaba to be held
responsible for them all, or might we be dealing rather with some sort of
indigenous practice that did not simply diffuse from one single centre,
but was inherited from a prehistoric substrate, and gave rise to the very
various formulations and realizations attested in classical sources? Thirdly,
to address the particular similarity between the eunuchs of Cybele and
Atargatis, is not the appeal to distant common origins an unsatisfYing
explanation for the truly salient parallelisms, the common themes and
tropes, in classical sources?
I hope to deal at greater length with the highly complex relationship
between Cybele and Atargatis in my own edition of DDS. Certainly
Atargatis was in the debt of her north Syrian predecessors, and certainly
the influence of Kubaba was diffused throughout Anatolia, but by the
time we encounter the evidence for the cult of Atargatis, i~ and after the
Hellenistic period, it is in the process of encountering an overwhelming
tide of influence that sweeps in the opposite direction, from west to east.
That is, it is ever more in the debt of Cybele, and, to judge from the
evidence of iconography, this is not the Phrygian goddess at home in
her central Anatolian highlands, but the Cybele whose image the Greeks
constructed after they first encountered her in the cities of Ionia in the
sixth century BC. Atargatis moved closer to Cybele in many ways. Like her,
she sat enthroned between lions, or mounted on a lion, and she learned
to brandish Cybele's (Greek) attribute, the tympanum, sign of the heady
music associated with Cybele and her galli (DDS§ 15, c£ 50). 28 She seems
to have been carried abroad in company with Cybele in and after the
third century Be; classical sources were very alive to the similarities, and
reveal them housed in adjacent temples or sharing cultic personneP9
78
Tellingly, the eunuchs in Apuleius themselves call them 'sisters' (Met.

9.10.3). And our understanding of Atargatis' retinue becomes accordingly
more complicated. Whatever was diffused from northern Syria in and
after the neo-Hittite period, we need to be at least as alive to the western
influences, traditions of image-making, and perceptions that swept
eastwards in and after the Hellenistic period. We have to allow for the
impact of the west on the galli too.
Here we encounter a problem. Cybele's galli are accessible almost solely
through western sources, in and through Hellenism; there is very little
evidence for them in Anatolia, and practically none before the Hellenistic
period. 30 However, Roller identified a silver statuette (7/6th century?)
from Bayandir in south central Anatolia as possibly representing a eunuch
(since beardless) priest; if that is right, the figure's bearing and its costly
material would suggest that it was a dignified and honourable office,
nothing like the noisy vagabond hysterics beloved of the classical sources.
She is inclined to regard the association of ecstatic behaviour with the
Phrygian cult as a reflection of 'orientalising' prejudice on the part of
classical peoples about behaviour appropriate to eastern peoples. 31 That
might be one possible reconstruction of the history of sacred eunuchism
in Hierapolis itself: originally a dignified sacerdotal office, which through
and after contact with Graeco-Roman civilization became increasingly
associated with ecstatic, marginal, deviant behaviour. Crucially, the first
(and only) time we encounter the institution in Hierapolis is late - the
second century AD. (A more extreme scenario still is that it was a 'Cybeliz-
ing' western import in toto.) Yet our woeful shortage of documentation,
and complete ignorance of the prehistory, hardly permit us to prefer this to
a third scenario. That is, we might be dealing with an ancient institution
that reflects an indigenous form of popular piety. Self-castrated devotees
were never otherwise than peripheral, marginal, never part of the official
priesthood of a sanctuary, and never really written into its theology. 32
Theirs was a form of devotion that was mostly a matter of pragmatics,
whose rationale was never properly articulated (except by onlookers who
applied to it changing glosses over time), and which simply appealed to
its own precedent in order to sustain itself. But even that scenario has to
make room for an unknowable, but perhaps large, degree of historical
convergence between them and Cybele's galli, and certainly for assimilation
in their classical ethnography and representation. 33
So much is obscure. Late as Lucian's testimonium is, it does (unless we
are to write it off altogether as True History-like invention) attest sacred
eunuchs in Syro-Anatolia itsel£ far further east than anything known for
Cybele, and militates against the case that they are purely and simply
79
JL. Lightfoot
a western import. But might Hierapolis itself have confected a link with
Cybele? For that we have the testimony of a foundation myth concerning
'Rhea' (Cybele) and 'Attes' (Attis) in DDS §15, though it is obviously
adventitious, whether or not it represents local tradition (it may not).
Indeed, the presence of Attis, a divine prototype for the galli, is a significant
difference between the cults of Cybele and of Atargatis. Attis is a Graeco-
Roman accretion who is not attested in Phrygian material, nor is he, or
any corresponding figure, otherwise linked to Atargatis. 34 (Combabos, the
mythological paradigm in DDS §§19-27, is on quite a different footing.)
Does that indicate that the cult of Atargatis 'preserves' an earlier state of
affairs, an association with eunuchs before the particular, western invention
of Attis? Strikingly, the presence or absence of Attis seems to have no effect
on the literary portrayal of the galli in the two cults. 35
Whatever the history of sacred eunuchism in northern Syria, it has, by
the time we encounter it, been passed through a classical filter, whether
we mean to imply that the nature of the institution itself is altered
through contact with Hellenism, or just that it is represented with the
same biases. Have we any hope of identifying what these might be, if we
have no corrective material, nothing that allows the eunuchs to speak for
themselves, nothing that might serve as a control? Some doubts strike
quite deep. DDS and Apuleius/Lucian are far from alone in putting
all the emphasis on momentary religious ecstasy, the spectacular act of
self-castration, the blood: Catullus 63, on Attis, is but the most obvious
addition. to this category. But what no source tells us is whether their
communities included other sorts of person - born hermaphrodites
(though these would be rare) or those suffering from diseas~; those who
had chosen to live as women as a matter of personal preference; those
who had been castrated before puberty (for there were plenty around);
or perhaps even those who were not castrated at all, but had chosen to
associate themselves with the group, perhaps for economic reasons, as
professional beggars. 36 What suggests that we might think along these
lines is, first, the intrinsic improbability that the community sustained
itself only in the way Lucian describes; secondly, the fact that Pliny twice
speaks of galli castrating themselves in a way which is citra perniciem,
not dangerous; 37 and thirdly, analogies with modern communities of
religious (or quasi-religious) eunuchs. For example, the Indian hijras are
communities whose members are attracted to it for a variety of reasons,
economic, familial, physiological, sexual; more disturbingly, there are also
wide reports of coercion and kidnapping. At all events, the members do
not join as a result of momentary impulse. It seems worth raising the
possibility that our classical sources are systematically distorting the data
80
about galli to emphasize their exoticism and to de-emphasize the more

mundane aspects of their following. In a complex institution, one would
expect to find a range of behaviours, and the classical sources may well be
consistently underplaying their diversity.
Perhaps the hardest question of all is what the galli themselves thought
they were doing. How did they conceive of themselves and their religious
purpose, and how did they wish to project themselves to the rest of
society? The prior question is what sacred eunuchism per se might be
about, a question which can hardly be answered other than with regard to
the various interpretations applied to it over time. In the cult of Atargatis,
the evidence is thinner than for that of Cybele, where indications are
that over time it acquired agricultural imagery: there is talk of 'mown
corn', harvestry, reaping that is grim indeed. 38 When, in the novel, the
homosexual practices of the eunuchs are exposed, this is presented in the
light of their hypocritical claims to be puri, casti, ayvoi, which makes
sense in terms of a popular conception of their role in these terms. 39 We
do not know whether, and how, claims to purity were articulated and
circumscribed by the eunuchs themselves, though it is interesting to find
a preoccupation with purity in other contexts in the cult of Atargatis, even
where galli are not attested. In Delos, where she had a major Hellenistic
temple, she was called ayvij A<j>po<>iTIJ, ayvij 8£6~, and a little plaque
(typical of many 'oriental' sanctuaries in Delos and elsewhere) found near
the entrance documents its purity regulations. 40 There is nothing very out
of the ordinary here, save the cult's idiosyncratic preoccupation with fish-
eating. But in the imperial period, in Niha in the Lebanon, we encounter
a far stranger figure, a local celebrity called Hochmaea. A prophetess, she
practised strange forms of self-denial (abstaining from bread for twenty
years through a vow to the goddess' consort), and lived to be a hundred
years old, calling herself 'virgin of the goddess Atargatis'. Such prolonged
periods of virginity are highly unusual. 41 It seems that the cult of Atargatis
could be the focus of a number of practices of 'purity', sexual purity
among them.
DDS, on the other hand, speaks of a sort of sacred passion, an £pro~,
between the galli and women in the Holy City, which is universally
condoned (§22). Certainly, the Roman satirists indicate that the galli
had a particular appeal to women, and there are insinuations throughout
pagan and Christian writing that galli are uncontrollable in their lust for
women as well as men. But it is still uncertain (at least to me) whether
Lucian is referring to a genuine religious institution whose exact character
is unclear, or whether he is mystifYing a popular sexual stereotype in
quasi-religious terms. On the other hand, their cancelled masculinity is
81
JL. Lightfoot
a structural contrast with the phallic interpretation that Lucian puts on the
cult in Hierapolis (§§ 16, 28), and which is borne out by the dedication
of phalli in Atargatis' sanctuaries at Delos and Dura Europos (though
galli themselves are not attested in either place). 42 Were they accorded the
positive recognition of being a 'third gender'? I think not: the tendency
is rather to regard them as neither one thing nor another. 43 Then again,
when Lucian describes their castration in DDS, he speaks as if the removal
of masculinity automatically equates with femininity: 'they no longer wear
male dress, but wear women's clothes and perform women's work' (§27).
This equates with what is enacted in the cult: once a man sets aside his
masculinity, he dons female dress. But also, there is a remarkably dose
literary parallel in the Hippocratic text which deals with the effeminate
Scythians: 'they don women's dress, renouncing their masculinity; they
become women, and they work alongside the women at the same tasks'
(Hipp. Aer. 22). In both texts, it is as if masculinity negated automatically
entails female outward appearance, and even female economic activity.
What proportion of Lucian's own statement is 'fact', and what is literary
affiliation, in the tradition of fifth-century Ionian ethnography which he
pastiches, is - as always -very hard to judge.
It is difficult to sum up when so little is known, but I hope to have
collected the data for eunuchs in the cult of Atargatis and to have drawn
out some of the resulting problems. Very little is known about the
galli other than from classical sources that are hostile or in other ways
problematic, and the institution and its history are no doubt complex in
ways that we cannot now even guess. But behind the distorted picture
that classical sources present, we should be prepared to' acknowledge
influences from both east and west, both 'oriental' and 'orientalising'.
How far did Hellenism simply accord the galli of Cybele and Atargatis an
undifferentiated treatment, and how far did it actually make them alike?
We could ask of the institutions what we could ask of any individual
among them: to what extent were they really 'like that', and to what
extent made so, or made to seem so, by Graeco-Roman image-making,
prejudice, and literary tradition?
Notes
1 The only empire-wide survey available is that ofHorig 1984, 1550-75. For
the Greek evidence, see Morin 1960, and for Syria, Drijvers 1980, 76-121. For
Hierapolis itself, RAC s.v. Hierapolis (Mabbog), 27-41 (Drijvers).
2 For her iconography, see LIMC s.v. Dea Syria (Drijvers); Horig 1979.
3 For Mmtotva, KUpta, and the language of dominion applied to Atargatis,
82
see Henrichs 1976, 280-3.

4 Graillot 1912, 290-1 'Ce rite de !'eviration semble etre de provenance
semitique'; Robertson Smith 1914, 321; c£ Cumont 1910, 679.40-2.

5 For these types of priest, see the Chicago Assyrian Dictionary, s.v. kurgarru,
assinnu and kulu'u; Sladek 1974, 88-9 (kulu'u); 89-91 (assinnu); 93-5 (kurgarro);
RA v. s.v. lnanna/lstar §12, Kult- und Kultpersonal, 85-6 (Wilcke); Roscoe 1996,
213-17. For statues from Mari perhaps of eunuch temple-singers, see RA viii. s.v.
MusikA.1 §3, 468 (Kilmer) and l.B §1, 488 (Collon).
6 C£, e.g., Hdt. 1.105.4 (the goddess of Ascalon inflicts a 'female disease');
Euseb. vtt. Const. 3.55 (using the same phrase of effeminates at the sanctuary
at Aphaca, inland from Byblos).
7 Beard and Henderson 1997.
8 For their power to turn men into women, see Archi 1977; Roscoe 1996, 219.
A similar power is attributed to Aphrodite's kestos in Kyranides, 1.10.51-2, 64-8

Kaimakis, where a gem is engraved with an image of a castrated man.
9 Nanda 1999, 25-6, 33-7.
10 IK Stratonikeia, 513.34-6; c£ also 544 O"EJlVO'ta'tOV 'tfi<;Heoii euvoiixQ[v,
1101.19, 't[oii pa~8ou]xou euvouxou, who works alongside the tEpEU<; to ensure
that boys' choirs perform their duty.
11 Strab. 14.1.23 (using the imperfect: it is unclear whether this has fallen into
disuse). But note Smith 1996.

12 On whom see the contribution to this volume by Shelley Hales, with
bibliography.
13 Apul. Met. 8.24---9.10; Luc. Asin. 35---41. For the cult in Beroea, see Morin
1960, 74---7; Harig 1984, 1566.

14 Bardesanes, The Book of the Laws of Countries (ed. Drijvers) 58.20---4 (text),
59 (translation); Drijvers 1980, 76-7; for the Greek version, see Euseb. Praep.
Evang. 6.10.44.
15 CIL vi. 32462 = ILS 4280; c£ Giammarco Razzano 1982, 252.
16 Fossey 1897, 59-61 no. 68; Drijvers 1980, 119; Giammarco Razzano 1982,
253.
17 SEG 7 (1934), 801 (with references); Robert 1936, 108 no. 61; Drijvers
1980, 91-2.
18 Flor. 2.7.4-7; Diod. Sic. 34.2.5-9.
19 For public self-castrations, see Lactant. Div. Inst. 1.21.16; for the use of
knives, see Juv. 2.115-16; Mart. 9.2.14; for throwing severed genitals, see Dio
Cass. 80.11, Clem. AI. Protr. 2.15.2.
20 Henrichs 1976, 277; for the proverb ovo<; a:yolV JlUcrn'Jpta, c£ Ar. Ran. 159;
Apostolius 12.75; Suda o 382; Paus. Att. o 17 (Demon, FGrHist 327F12); for
the fable, c£ Aesop. Fab. 173 Hausrath; Babrius Fab. 141, Phaedr. Fab. 4.1,
of Galli Cybebes; Babrius re-used by Tzetz. Chi!. 13, Hist. 475, 251-67; cf.
Aesop. Fab. 193.
21 Adams 1984, 53---4 (Latin); Bain 1984,29-30 (Greek); c£ RACs.v. Effemi-
natus, 636 (Herter).

22 Jaffrey 1996, 103; Nanda 1999, 17 ('special, feminized language').
83
JL. Lightfoot
23 The calendar of 354, CIL U 260 (with commentary on 313-14); c£ Julian.
Or. 5 p. 168c-169d; Lydus Mens. 4.59-60; Macrob. Sat. 1.21.10, cf. Val. Max.
8.239-42. Note, however, that this is western, mostly Roman, evidence, and .
centres upon the figure of Attis: the Anatolian background (if any) is obscure.
24 Apul. Met. 8.28.2. For a relief found near Lanuvium showing a gallus
accompanied by a whip whose thongs are knotted with astragaloi, see Vermaseren
1977, no. 466.
25 Apul. Met. 9.8; c£ Juv. 6.517-21.
26 The standard work is still Laroche 1960; c£ Burkert 1979, 104-5.
27 Roller 1999, 46-7, 69, 124.
28 For Cybele's tympanum, see Roller 1999, 136-7, 148, 173-4, 185, 296.
29 See e.g. Cumont 1900, 2240.
3° For the emergence of this type of figure in Greek sources, see Burkert 1979,
102-3, 104; Giammarco Razzano 1982, 243-6. KUPTJPo~ first in Semonides of

Amorgos fr. 36 West (7th cent.); first mention of a castration in Plut. Nic. 13.4
(but relevance to the cult ofCybele denied by Roller 1999, 177-8); yat..:Ao~ first
in Call. fr. 761 Pf. (incert. auct.) raUm; Rhian. AP 6.173.3 = HE 3238, the
adjective yaA,A,a'io~; Arcesilaus ap. Diog. Laert. 4.43; a group of epigrams about
a yaA,A,o~, Atys (AP 6.217-20), of which the earliest perhaps is Dioscorides (AP
6.220 =HE 1539-54).
31 Roller 1999, 105 and fig. 36, 230-1, 252-4.
32 This would be a similar model to that applied by Frankfurter 1990, in his
interpretation of the ascent of pillars at Hierapolis and stylitism in northern

Syria.
33 One encounters a similar problem in the cult of Ma/Bellona, originally
from Cappadocian Comana. This is connected with self-wounding (though not

self-castration), but it is attested only in western, Latin evidence, which often
juxtaposes it with the galli. '
34 On the invention of Attis, see Roller 1999, 177-82.
35 Cumont 1910, 679.43-4 'Sicher ist, das die syrischen Galli sich von den
phrygischen kaum unterscheiden'.

36 For pre-pubertal castration, see Juv. 6.512-13 ingens I semivir; Persius 5.185
grandes Galli (their height is a giveaway). For abduction and forced emasculation
alleged of the classical galli, see Martial 3.91. For doubt that every gallus was
necessarily castrated, see also Beard 1994, 173-4; Roscoe 1996, 203.
37 Plin. NH 11.261, 35.165.
38 Julian Or. 5 p. 168d, 'tO iepov Kat a7toppTJ'tOV eepo~; Firm. Mat. Err. prof
ref. 3.1-2; c£ the fragments of a Naassene hymn ap. Heitsch (xliv) fr. 2.11-12,
calling Attis xt..,oepov otaxuv cXIlTJeevta.
39 Apul. Met. 8.29.6 sacerdotum purissimam ... castimoniam, c£ Luc. Asin. 38;
Apul. Met. 9.8.1 purissimi illi sacerdotes; for Cybele's galli, see Dioscor. HE 1541;
Babr. Fab. 141.9; Var. Eum. 135 (119); Catull. 63.17; further ~xamples of ayvo~,
castus,purus in Nock 1972, vol. i. 9. According to Ps.-Cypr. AdSenatorem 15-17,
they claimed to be casti only on the days they celebrated their rites.
40 Bruneau 1970, 470 (ayvl)), 472-3 (purity regulation= !De/2530).
84
41 IGLSvi. 2928-9; c£ Lane Fox 1986, 347-8.

42 Bruneau 1970, 473. For terracottas of the glans penis from Cybele's Palatine
sanctuary, see Vermaseren 1977, nos. 68-77; Roller 1999, 275-6 and fig. 66.
43 e.g. Val. Max. 7.7.6; Lactant. Div. lnst. 1.21.16; Prudent. Perist. 10.1071-3.
Bibliography
Adams, J.N.
1984 'Female speech in Latin comedy', Antichthon 18, 43-77.
Archi, A.
1977 'I poteri della dea lstar b.urrita-ittita', Oriens Antiquus 16, 297-311.
Bain,D.
1984 'Female speech in Menander', Antichthon 18, 24-42.
Beard, M.
1994 'The Roman and the foreign: the cult of the "Great Mother" in Imperial
Rome', in N. Thomas and C. Humphrey (eds.) Shamanism, History,
and the State, Ann Arbor, 164-90.
Beard, M. and Henderson, ] .
1997 'With this body I thee worship: sacred prostitution in antiquity', in
M. Wyke (ed.) Gender and History 9.3 Special Issue: Gender and the
Body in Mediterranean Antiquity, Oxford, 480-503.
Bruneau, P.
1970 Recherches sur les cultes de Delos a l'ipoque hellinistique et a l'ipoque
impiriale, Paris.
Burkert, W.
1979 Structure and History in Greek Mythology and Ritual, Berkeley.
Cumont, F.
1900 'Dea Syria', in REiv. 2236-43.
1910 'Gallos', in RE vii. 674-82.
Drijvers, H.J.W.
1980 Cults and Beliefi at Edessa, Leiden.
Fossey, C.
1897 'Inscriptions de Syrie', BCH21, 39-65.
Frankfurter, D.T.M.
1990 'Stylites and Phallobates: pillar religions in late antique Syria', Vig.
Christ. 44, 168-98.
Giammarco Razzano, M.C.
1982 'I "galli di Cibele" nel culto di eta ellenistica', in Ottava Miscellanea
Greca e Romana (Studi Pubblicati dall'Istituto Italiano per la storia
antica, fasc. 33), Rome, 227-66.
Graillot, H.
1912 Le Culte de Cybele, mere des dieux, aRome et dans !'empire romain, Paris.
Henrichs, A.
1976 'Despoina Kybele: Ein Beitrag zur religiosen Namenkunde', HSCP
80,253-86.
85
JL. Lightfoot
Horig, M.
1979 Dea Syria: Studien zur religiosen Tradition der Fruchtbarkeitsgottin in
Vorderasien, Kevelaer.
1984 'Dea Syria-Atargatis', ANRWii/17.3, 1536-81.
Jaffrey, Z.
1996 The Invisibles: A tale ofthe eunuchs ofIndia, New York.
Lane Fox, R.
1986 Pagans and Christians, London.
Laroche, E.
1960 'Koubaba, deesse anatolienne, et le probleme des origines de Cybele',
in 0. Eissfeldt (ed.) Elements orientaux dans !a religion grecque ancjenne,
Paris, 113-28.
Lightfoot, J .L.
(forthcoming) Lucian of Samosata: On the Syrian Goddess (Text, Translation,
Introduction and Commentary), Oxford.
Morin, P. J.
1960 'The Cult of Dea Syria in the Greek World', Diss. Ohio.
Nanda, S.
1999 Neither Man Nor W:Oman: The Hijras ofIndia, 2nd edn, Belmont.
Nock,A.D.
1972 Essays on Religion and the Ancient World, 2 vols., ed. Z. Stewart,
Oxford.
Robert, L.
1936 Collections Froehner i.: inscriptions grecques, Paris.
Robertson Smith, W
1914 Lectures on the Religion of the Semites: First Series: The Fundamental
Institutions, London.
Roller, L.E.
Roscoe, W
1996 'Priests of the goddess: gender transgression in ancient religion', History
ofReligions 35, 195-230.
Sladek, WR.
1974 'lnanna's Descent to the Netherworld', Diss. Johns Hopkins, Ann
Arbor.
Smith, ].0.
1996 'The high priests of the temple of Artemis at Ephesus', in E.N. Lane
(ed.) Cybele, Attis and Related Cults: Essays in memory ofM]. Vermaseren;
Leiden, 323-35.
Vermaseren, M.J.
1977 Corpus Cultus Cybelae Attidisque (CCCA) iii. Italia-Latium, Leiden.
86
5
LOOKING FOR EUNUCHS:

The galli and Attis in Roman art
Shelley Hales
For most Romans of the republican and early imperial periods, experience
of the eunuch came through contact with the cult of Cybele and her
consort Attis. 1 Cybele's cult, a manifestation of the Near Eastern and
Aegean worship of a mother goddess, was centred in Phrygia in Asia
Minor and was introduced to Italy in 204 BC, 2 via Rome's port of Ostia.
On her arrival, a temple was built for the goddess in the heart of Rome
on the Palatine hill,3 The cult was popular during the imperial period,
engendering much cultic activity both in and around Rome, particularly
at Ostia, and was just one of the many foreign cults operative within the
city. The mythology of Cybele's affair with Attis was canonized for the
Roman audience by Ovid at the very beginning of the first century AD. 4
In his version of the myth, Attis was Cybele's only lover but when he was
unfaithful, Cybele took revenge, driving him into a frenzy during which
he castrated himself and died. 5 Cybele then resurrected him to be her
divine consort, assured of his eternal fidelity. The cult was served by
the galli (led by the archigallus), a body of priests who purportedly (at
least in accounts of the Roman era) established their own relationship
with the goddess, as they celebrated the death of Attis, by similarly
castrating themselves. 6
The study of Cybele, Attis and her eunuch priests has had a long
history. It has inevitably involved questioning contemporary ideas of what
is Roman and exploring how to insert the alien into Roman culture_?
When Rome was still understood as the safeguard of Graeco-Roman
culture, at the hub of a vast empire which she must somehow initiate
into the ways of the centre, Cybele remained an anomaly. The difficulties
of accommodating an aniconic goddess with castrated servants into this
civilization could only be overcome by accepting a Romanized Cybele
whilst dismissing the more unsavoury parts of the cult, inevitably the ritual
castration, as something marginal and foreign. 8 As our understanding of
87
Shelley Hales
the complexities of acculturation has progressed, it has become increasingly

realized that such a division is unrealistic. The festivals of Cybele and Attis
mixed 'Roman' and 'alien' rituals. Not only the city's foreign visitors but
even her emperor himself participated in the cult.
More recently Cybele has been placed firmly in her rightful place (at
least in early twenty-first-century eyes) at the heart of Roman society. 9
In visual terms, the familiar Graeco-Roman iconography of the goddess
would seem to support the case for her integration. The famous silver
dish from Parabiago in North Italy shows the goddess wearing a turreted
crown as a sign of civilization and being pulled in her chariot by her tamed
lions. 10 She is situated at the centre of the universe, the allegories of the
sun and moon above her, the elements below her with the four seasons,
all displaying their fruits. She is accompanied in the chariot by Attis. His
occupation is indicated by the attributes of panpipes and crook, his eastern
origins denoted by his Phrygian cap, tunic and trousers. 11
Cybele could be seen everywhere in Rome. She lived on the Palatine
next door to the emperor and her image stood on the spina in the Circus
Maximus. 12 Turcan reports a cameo from Vienna which shows just to what
extent Cybele was reconciled with the Roman. It features Livia, depicted as
Cybele and Ranked by two galli. 13 It is, of course, not impossible that the
sight of Livia as emasculating mother goddess might have raised eyebrows
amongst a sceptical audience, particularly given the role in which such
an image might cast Augustus. Nevertheless, it reinforces the idea that
Rome's· first matrona and the Phrygian mother goddess were one and
the same; each drew power from the other and both safeguarded the
future of Rome .
.However, whilst Cybele might be feted, modern sch9lars have failed
to reach consensus on her galli. In this respect they are bound by the
negative reaction toward them voiced in ancient literature. This reaction
was vocalized by the parochial likes of Martial and Juvenal but also by
authors we might expect to be watching Rome with some detachment,
for example, Dionysius ofHalicarnassus. They all do their best to distance
the galli from the Roman, in terms of race, behaviour and, not least,
sexuality. 14 These opinions have caused difficulties for modern authors
trying to rehabilitate the cult of Cybele. They invite ways of thinking away
the gallus problem. Perhaps they were not her real priests but a distracting
and unwanted sideshow, not Roman but foreign. Perhaps they did not
all castrate themselves. The debate concerning the physical condition
of the priests is ongoing and, ultimately, insoluble. 15 However, it does
seem increasingly likely that the ranks of galli and archigalli featured both
castrated and whole men. By the first century there were laws in place
88
Looking for eunuchs: the galli and Attis in Roman art
forbidding Roman citizens to become eunuchs, though the effectiveness

of these laws cannot be vouched for. Carcopino tried to divide the two
categories by proposing that the archigallus, the priestly overseer of the
cult, was always a Roman citizen and so could not have been castrated. 16
If Juvenal prevents us from believing that galli were socially acceptable,
then it seems that we must compensate for his negativity by believing that
at least some were less different than others.
In 1994, Mary Beard published the most sophisticated analysis yet of the
place of Cybele and her galli within Rome. Her discussion recognized that
the expression of difference within Rome was a crucial part of mediating
Roman identity and that the galli were an important expression of that
dialogue. However, her views have not been universally acknowledged.
Sawyer, for instance, interprets the presence of Cybele as evidence of
Romans' insecurity about their own identity. In her interpretation, the
cult and, more precisely, the galli, shore up the Romans' sense of identity
by reminding them of what they are not. Simultaneously, they provide
a form of expression for those inhabitants of the city who, whether because
of class or race, cannot be absorbed into 'proper' Roman religion. 17 This
theory sees the gallus as an alternative to Roman life, presuming that
he acts beyond the realm of the Roman, rather than accepting that the
tensions between Roman and alien might actually lie within the Roman
psyche, or even within the gallus, itself. The aim of this chapter is to
consider whether the visual images of galli, and their paradigm Attis, can
shed any light on how these tensions might be resolved or exploited in
the figure of the eunuch.
The understanding of romanitas as a deliberately flexible ideal, equally
able to embrace and reject difference, is a new phenomenon in classical
scholarship. The slow reintegration of the so-called 'other' into Roman
studies is matched by a similar sluggishness within studies of ancient art
to consider the alien and the non-classical. The possible presence of the
eunuch physique in an art apparently devoted to the ideal has caused
some anxiety, an anxiety which again arises from a negativity expressed
by ancient commentators. The abhorrence of such images had a moral as
much as an aesthetic motivation; the unnatural and unreal are dangers
that should be hidden not paraded. 18 Quintilian says specifically that
eunuch priests are unfit models for artists. 19 Why, then, does Pliny the
Elder's Natural History, composed in the first century AD, record that
the late classical painter Parrhasios painted an archigallus? 2° For Jerome
Carcopino, writing in 1923, this was inconceivable; a classical painter
simply could not have painted such an unideal subject. He insisted
that the text was a corruption and in fact referred to a painting of
89
Shelley Hales
mythological pornography which later found its way into the emperor
Tiberius' bed chambers. 21 It was preferable to see Parrhasios' repertoire
including graphic depictions of straight sex than an image of someone
of dubious sexual identity.
Galli
So how do these ancient and modern aversions affect the representation of
galli? There are images of galli and archigalli in and around Rome, many
commissioned by the priests themselves. The existence of such images in
itself demonstrates the toleration towards galli, their own relative wealth
and the pride they took in their positions. But they also help to show us
how they constructed their difference.
In other ancient societies, the physical difference of eunuchs was
regularly depicted. Assyrian representations might be understood to convey
the flabbiness associated with the hormonal effects of castration, through
the double chin. The beardlessness of the eunuchs also distinguishes
them from the long-bearded Assyrian elite. The freedom to portray the
physical attributes of the eunuch condition does not simply depend on the
different tenets of Assyrian art but also the acceptance of these eunuchs
within the Assyrian social system. 22 They have a carefully defined role
in society where their subservient position renders them safe. As their
position in the system depends on their physical difference, the portrayal
of that difference acts as a counterpoint to the elite male.
The Greek and Roman audiences themselves were not entirely incapable
of imagining or appreciating other bodily forms. The art of tl,1e Hellenistic
era of the last three centuries BC shows an increased interest in the exotic
and bizarre. The hermaphrodite, a 'female' figure with male genitals,
is a typical expression of this interest. 23 This figure, al~ng with other
Hellenistic themes, would find itself most conducive to Roman, domestic
tastes, copied throughout the houses and villas of Italy, for example at
the Villa of Oplontis. 24
The Roman gallus, however, would appear to have neither of the
interdependent advantages of an art which wanted to depict, and hence
externalize, his physical difference or a clearly defined place within, or
without, the Roman social system. In general, romanitas tended towards
aiding assimilation. Literary and visual rhetoric allowed foreigners to
assume guises ofRomanness if they so chose. The adoption of the language
of veristic and then imperial portrait types by the freedmen and women
of Rome, has long since been recognized as a means by which newcomers
of disparate races became acceptably Roman. 25 These portraits defined the
Roman man and woman as demonstrable, visible types, identifiable by
their clothes, hair and attitude.
90
As a result of creating the Roman stereotype, difference was inevitably

recognizable by the way other people looked or dressed. Northern barbar-
ians wore moustaches and trousers. The galli, as an affront to certain
Roman constructions of both race and gender, were, on the surface,
similarly demarcated. Greek and Roman literature consistently reinforces
the sexual and racial difference of eunuchs by stressing how different they
look. 26 They were presented as wearing bright clothes, heavy jewellery,
make-up and sporting bleached and crimped hairY Their eunuchized
state was recognizable by this womanly paraphernalia. 28 Most of these
negative descriptions served to emphasize difference between the galli
and the literature's personae, whether angry Romans or Christians. These
galli are despicable creatures. Posing as sexually-pure priests who have
abandoned their male role, they still pleasure women. 29 They beg and
put on a frightful show, a din always preceding them. Their looks are
similarly deceptive; wigs, make-up, affectations in the way they walk -
none of these reflect the true physical condition that might be the result
of castration. Instead they are artificial devices, adopted by charlatans.
The effect in these passages is clearly to disparage the galli, to expose them
as frauds (not in the physical fact of their castration but in terms of their
duty to their goddess).
Whilst the eunuch's cross-dressing is used to such negative effect in
the ancient sources, modern writers have sought an escape route in the
effeminacy of the visual images. In the early 1930s, Guido Calza attempted
to remove the visual manifestation of eunuchs from Rome by reclassifYing
images of galli as images of priestesses who also ministered to the cult
of Cybele. He was helped by the fact that a relief survives of one such
priestess, Laberia Felicia. She is depicted holding a swag of oak leaves
and a libation bowl over an altar and wears an image of a bearded deity,
presumably Zeus, in the form of a pendant. The head is a modern
restoration which has been modelled with little regard for the beads
that hang over her shoulders. Calza used this relief as a model for his
reinterpretations, noting the similarities between the costumes of galli and
Laberia, and reclassifYing them as women. 30
Calza reproduced in a line-drawing the image of a gallus from the
Capitoline Museum (Fig. 1), at that time without a head. He wears long,
flowing robes which bear a strong similarity to feminine attire, as we might
expect from the literary descriptions of galli. However, this figure's strange
jewellery is much heavier than that of Laberia. The large pendant around
his neck is adorned with images of Attis. The central image features Cybele
between Hermes and Zeus, the pediment above, Attis with his shepherd's
crook. Two discs repeat the theme, each having a bust of Attis. A whip rests
91
Shelley Hales
over the torso's left shoulder. (Self-flagellation

was one of the more extreme rituals performed
by the galli). 31 The head wears a Phrygian cap
though this would seem an unlikely restoration
as no other extant images of galli wear this
headgear. 32
A tomb relief of another gallus, again from
the Capitoline, came in for similar treatment.
He wears similar headgear to that intimated by
the remains of the image of Laberia; over his
waved hair he wears a hood with a head-dress
bearing the image of Attis and beads that trail
down either side of his face. On the other hand,
the temple-shaped pendant and the whip are
clearly similar to the Capitoline figure. The
images of the priests show to what an extent they
had rejected an iconography that could have
integrated them into Roman society, instead
opting for the feminine and the alien. Only by
reading them as Roman women, like Laberia,
could Calza accommodate them within the
apparent demands of ancient art and the more
pressing needs of modern taste. The possibility
that deliberate gender ambiguities might exist
in the depiction of eunuch priests was thereby
avoided. 33 But the ambiguity is obvious - the
galli break gender divisions both within the cult,
Fig. 1. A gallus, Capitoline where men and women dress almost identically,
Museum. but also without the cult, where the gallus
becomes an alternative to the Roman male.
Along with the gender ambiguities, the priests and priestesses of Cybele
further demonstrate their difference by the badges and symbols which
mark their cult. But how peculiar and different are these attributes?
Their robes might be unusual by virtue of cross-dressing but they are
not, generally, anything other than Graeco-Roman. 34 Their foreignness
is implied by smaller details (the head-dress and the large pendants)
which are not necessarily confined to the galli alone. 35 Cybele shared close
affiliations with other cults, for example that of Bellona, who had a small
temple in the Sanctuary of Cybele at Ostia. Likewise dismissed as foreign,
the followers wear similar outfits, specifically a head-dress with medallions
and beads down the side. The major difference is signalled by the weapon
92
of choice of the Bellona devotees; double-headed axes as opposed to the

whips of the galli. 36 Their costumes might well imply that they are, at
least in part, constructed by a Roman view of foreignness rather than as
a reflection of their original representation in the EastY
A certain level of stereotyping of the foreign can similarly be revealed
in the Phrygian cap, misguidedly applied by the modern restorers of the
Capitoline gallus. This does not mean that either the ancient or modern
sculptors were deliberately sanitizing the cult, simply that they were
applying their own fantasies of otherness to a suitable candidate, the gallus.
Even at the most extreme manifestation of its alien nature, the cult is the
product of negotiation within the Roman imagination.
Archigalli
The images of galli do not show us eunuch bodies -like the literary texts
they rely on costume to convey the idea of eunuchness. The motif of
castration is instead hinted at by constant allusion to the image of Attis,
whose own emasculation was generally held, at least in the Graeco-Roman
mind, to be the reason for the castration rituals of the galli. 38
The sarcophagus of a priest of Cybele, found in the cemetery at Porto,
perhaps the cemetery for Ostia, reiterates the importance of making
allusions to Attis. 39 On the sides of the sarcophagus, two reliefs show the
priest in his robes, going about his business, tending the shrines of Cybele
andAttis. However, on the lid he reclines in a different guise (Fig. 2). The
figure wears heavy jewellery; rings on the hand supporting his head and
on the other arm a bracelet depicting Cybele and Attis. Calza identified
the figure as an archigallus on the basis of that bracelet, understanding
it to be the okkabos worn by priests of this rank. 40 By doing so, he was,
of course, claiming citizenship for this man, explaining away yet another
uncomfortable image. 41 Whatever his status, however, this image shows us
a new level of assimilation between the priest of Cybele and her consort.
He adopts the reclining posture that we will later see is typical of the
Fig. 2. Sarcophagus lid of an archigallus, cemetery at Porto. Source: Archivio

Fotografico della Soprintendenza Archeologica di Ostia.
93
Shelley Hales
dying Attis and wears the eastern clothes of the Phrygian shepherd. He
begins to become Attis, an identity also ascribed to the Capitoline gallus
by means of an added inscription in Greek lettering. These galli reveal
a further level of difference, one that challenges the boundary between
mortal and divine.
A second object associated with an archigallus of Ostia, Marcus Modi us
Maxximus, furthers the connection between Attis and his priests. 42 The
man's oddly-spelt citizen name and his position are proclaimed in an
inscription on the object (which seems to take the form of a modius, or
grain measure) but there are no images of him here. Instead, symbols
associated with Attis, such as his panpipes, dominate the iconography. The
identity of the donor himself is only visually realized by a cockerel,
the Latin for cockerel being gallus. 43 In this image, the Roman man we
expect from the citizen name has disappeared, becoming completely
subsumed by his post as a gallus and taking the identity of Attis. The link
between Attis and gallus is reinforced in iconography of Attis, where he is
occasionally depicted riding on a cockerel. 44 The complexity of the joke is
here deepened. Gallus is not only the name of Cybele's priests but also of
the river by which Attis was raised and later died. Substitutions like the
cock allow artists to visualize links between the mortal and divine servants
of Cybele without having recourse to the depiction of mutilation.
It would seem, then, that the costumes of the priests of Cybele
deliberately differentiate them from traditional Roman priests, not only
by asserting their own difference but, more specifically, by alluding to
the different nature of their model, Attis. We might eve)l argue, but
only if Calza was right, that the archigallus, in particular, asserted his
rank by demonstrating a more complete absorption of the shepherd. The
dichotomy between dressing up as Attis and flaunting thdir own physical
difference is exacerbated by the question of how many of these Roman
galli and archigalli were eunuchs. But even if we were to accept that we
cannot safely identifY an actual eunuch, that those who commissioned
images of themselves may have been intact Roman citizens such as Marcus
Modius Maxximus, these images still raise interesting questions about
their patrons' ideas of self-presentation. If they are not castrated, then the
stress on dressing and acting like Attis might be explicable; galli have
to assume the guise of Attis because they are blatantly not him. That
Romans could take off the costume does not necessarily alleviate the
intensity of being Attis. It does, however, draw our attention to the
fact that these are Romans apparently deliberately alien~ting themselves
from the Roman north. They are not simply there to contrast with the
Roman male because that is, in fact, what they are. For Catullus' Attis,
94
Lookingfor eunuchs: the galli andAttis in Roman art
castration means divorce from his fatherland and family. 45 But for M.
Modi us Max:x:imus, becoming Attis serves a role in Roman society worth
commemorating. Of course, it is equally possible that, underneath these
same extravagant costumes, some of these galli are indeed castrated.
But how would we know? It becomes impossible to tell exactly who is
who: women, castrates and citizen men are all brought together by their
costume, regardless of conventional boundaries.
Attis
The increasingly important role of Attis within the cult of Cybele provided
the galli with a figurehead to emulate. The insistence on depicting Attis'
costume rather than physical disability made their assimilation of the
character much easier. Attis was the ultimate in the alien - different in
terms of race, gender and his subordinate status - yet easily imitated
by wearing the right clothes. As the Parabiago plate indicates, the
iconography of Attis placed great emphasis on his dress and paraphernalia.
His shepherd's tools, trousers and cap all allude to his social status and
ethnicity, not to his physical difference.
To this extent, the popular iconographic imagining of Attis pays even
less attention to the physical difference of Attis than that of the galli.
Elements of the story, one presumes, militate against this option. Attis has
to show us why he was so beloved by Cybele, a goddess usually above the
need for male company. Given this presentation of Attis as a beautiful,
virile youth, then his 'unmanning' might come as more of a shock. How
should the archetypal self-castrate be portrayed at the crucial point of his
story, in the episode of his death? 46 The typically classical moment is the
contemplation of the event before it occurs. In a fresco from the House
of Pinarius Cerialis in Pompeii, Attis contemplates the sickle which he
will use against himself The presence of this scene in a domestic context
reflects the eagerness of Roman householders to embrace the exotic in
their homes. But the full extent of Attis' montrosity is not realized. Attis
may have the weapon ready but he remains fully dressed, safely unexposed.
Again the emphasis is so strongly on the costume that some have seen this
not as Attis himself but as an actor playing AttisY Yet another layer of
ambiguity prevents us from accessing the eunuch body.
A class of reliefs, however, does actually depict the moment of Attis'
death, always in the same location under a pine tree by the River Gallus,
his injuries always hidden. In an example from the Sanctuary of Cybele in
Ostia, Attis is again fully dressed lying in the pose adopted by the figure
on the Porto sarcophagus. An allusion to his castration is made by the
presence of testicles pictured in the space between his knees as if just fallen
95
Shelley Hales
from their rightful place. But the connection between these bodily parts
and the body itself is made very remote. 48 The alternative, as shown in
a relief from Glanum in southern Gaul, is to show him almost naked,
save his cloak and Phrygian cap, but covering up his wounds with his
hands. 49 Here we might feel one step closer to encountering a eunuch
body but, even here, Attis remains identifiable not by his deformity but
by his costume.
Even when his body is exposed, Attis' mutilation often remains non-
apparent. A popular Attis type depicts the shepherd in a slashed body suit
which billows out at the groin, revealing the crucial part of the anatomy. 5°
Almost always, the genitals are intact. Whether these images are to be
understood as Attis before his fall or perhaps as reflecting a possible
restoration after his resurrection remains unclear. What is clear is that
the majority of images of Attis avoid having to depict him in a state of
mutilation. He is the paradigm of self-castration, yet offers no visual
model of his physical difference. His physical appearance is used to remind
us that he was once a normal youth, his beardlessness an indication of his
age, rather than to define a third sex, the eunuch.
So, if the intact nature of some of the men involved might explain the
role of costume in the galli, then what might it mean for Attis? \lVhy the
accent on covering up what Attis has removed - in effect what makes
him Attis? Finally, in moving from the Ostian cemetery to the sanctuary
of Cybele and now into the shrine of Attis itself, the worshipper sees
Attis revealed (Fig. 3)_51
Fig. 3. Attis, Sanctuary of Cybele, Ostia. Vatican Museums. Photo: Fratelli Alinari.
96
This image represents the moment following that depicted in the relief
at Glanum where Attis sank down by the Gallus. Now Attis reclines by
the Gallus (here personified as a river god), apparently restored after his
ordeal. The garland of fruit and pine-cones and the vegetation in his right
hand symbolizes strength returned. The god is nearly, but crucially not
quite, naked, still dressed enough to convey his identity. His Phrygian
cap competes for space on his head with a crescent moon and a circlet of
sun-rays. 52 But here, at last, Attis' body, the body we have been promised
since we looked at the very first image of a gallus, reveals the eunuch
condition. Far from depicting mere castration, the artist has removed any
sight of male genitals. In their place he has substituted a smooth pubis, its
female characteristics complemented by the curves of the torso. A second
Attis from the same sanctuary clearly depicts the hermaphroditic nature
of the god, bereft of male genitalia. 53
Such representations are relatively rare. The reclining Attis was found
in the inner shrine of the Attideum, guarded by two Pans, ensuring that
only initiates may see his true nature. 54 The inscription below the figure
tells us that Gaius Cartilius Euplus dedicated it by the inspiration of the
goddess. This is the revelation of Attis' true nature by Cybele herself and
reveals not his reproductive power but that of the goddess. The resurrected
Attis remains a dangerous concept- even as a hermaphrodite, he is upside
down, male and female in the wrong places.
It is from this image that the galli of Ostia must have conceived their
paradigm. They subsume their identities to him to become Cybele's
servants. The robes which they wear to identifY themselves with Attis are
not there to mask the manhood of Roman citizens but to hide the truth
of Attis' emasculation. As with Attis himself, the signs of potency which
the Cybele priests exhibit, such as the sheaves of corn which sprout as
tail feathers from the cockerel of M. Modius Maxximus, do not stress
their own biological potency, but the prosperity they can bring to Rome
through the power of Cybele. The images show that their choice is just as
valid in protecting Rome as the more 'manly' job of a Ramen or pontiff.
They can likewise act as benefactors to the community, even though they
have taken a different route.
The confusion in distinguishing between gallus, archigallus and Attis
underlines the fact that the visual presentation deliberately blurs the
boundary between Roman and alien, mortal and divine. How could
anyone tell castrated and non-castrated, worshipped and worshipper
apart? The Roman citizens who presented themselves as Attis presented
themselves as challenging everything that would generally be thought of
as justifYing their Roman identity. Boundaries that the male, patrician,
97
Shelley Hales
Roman priest respected, the gallus appeared to be about to transgress.

However, the gallus served a goddess at the heart of Roman religion,
a religion fostered by the emperors. 55 That galli chose to have themselves
commemorated in this guise, and simultaneously paraded their citizen
names, implies that these people thought they could be both gallus and
Roman. But how could anyone, on sight, distinguish between the Roman
and the non-Romangalli? Just as the costumes worn by Cybele's ministers
negated the gap between the male and female, similarly they elide the
distance between Roman and foreign, castrated and not-castrated. They
create an ambiguity that can allow the cult to involve and evolve. Beard's
work on the gallus concludes that the cult parades the differences between
Roman and non-Roman as a means of dealing with 'cultural uncertainty'
and competing claims of what it is to be Roman. 56 It is also the case that
this is not a distinction that need be played out between Romans and
foreigners, it is a juxtaposition which is within and has to be faced by
every player in the empire, citizen or not. From the imperial family at the
heart of Rome to the provincials of the western empire, the spread of the
cult of Cybele would appear to confirm the integral role of the cult, galli
and all, within Roman identity. 57
These images, then, cannot help us with a positivist history of eunuchs.

They do not tell us whether these men are eunuchs or not and they do
not tell us whether they are Romans or not. The iconography of the gallus
deliberately masks these aspects of identity in the pursuit of otherness.
In the same way, the conventions of veristic portraiture opened up the
Roman 'look' to people from all races and statuses.
The galli have remained hidden behind the fa<_;:ade of prejudices
propagated by an ancient and modern rhetoric which tri'es to construct
what is Roman according to a limited social and geographic definition.
These constructions hide the true variety of the Roman experience.
Images of galli reflect attempts to visualize difference and the tensions
and contradictions within Roman identity. The difficulty of portraying
the galli as anything other than Roman bodies in fancy dress reflects an
inability of the Roman audience either wholly to alienate or rehabilitate
the gallus or archigallus. Whilst the Assyrian eunuch is given his own body
type as a reflection of his compartmentalized status, the Roman gallus
remains on the threshold because he is neither the opposite of the Roman
nor the same as the Roman but a part of the Roman. The eunuch was
never exposed and made monstrous like the hermaphrodite who, by its
exposure, was safely defined as apart and distant. The reality of the gallus,
his proximity to and inclusion within Roman life, meant that he could
98
Lookingfor eunuchs: the galli andAttis in Roman art
not be fully ostracized. M. Modius Maxximus reduces his identity to

a series of symbols and signposts and so avoids having to negotiate his
physical appearance. Attis, too, appears in the domestic and public context
as a male body in a Phrygian outfit. It is only in the confines of the
Attis sanctuary that a different body can be revealed. Only here was the
monstrosity of Attis and his irreversible difference on show.
Notes
1 For a general introduction to the cult see Vermaseren 1977. Roller 1999
offers a more recent alternative. Hepding 1903 represents the comparable study
of Attis.
2 See Livy 25.1.6. Also Ov. Fast. 4.247-348; Herodian 1.11.3-5. Gruen 1990,
5-33 discusses the arrival.

3 Vermaseren 1977, 41-3.
4 Ov. Fast. 4.221-44.
5 There are many variations of the myth. Paus. 17.9-12 offers two alternative
versions. Hepding 1903, 98-122 offers a myriad of stories.

6 Vermaseren 1977, 96-101.
7 There is an extensive bibliography on the role of the galli in Roman society.
Beard 1994 is easily the best. Sawyer 1996 and Roller 1997 continue to see
the gallus as wholly alien.
8 Sawyer 1996, 123 suggests that the incorporation ofCybele into the Roman
pantheon may have been something of a 'Trojan horse'. That is to say that the
Romans do not realize what they are letting themselves in for. Roller 1997,
549, makes the difference even more explicit- galli in Asia can be respected by
Romans while galli in Rome must be abhorred.
9 Beard 1994, 180-1.
10 Vermaseren 1977, 71-87, deals with the iconography of Cybele. For this
plate see 72-4. The dating of the Parabiago plate is uncertain. The debate is
summarized by Elsner 1998, 209.
11 The iconography of Attis is discussed in Vermaseren 1966.
12 Vermaseren 1977, 41-51, provides a brief tour of Cybele's presence in
the city.
13 Lambrechts 1952, 251 ff. See also Turcan 1996, 43. Other images linking
emperors or their wives to the cult can be found in Calza 1946, 201-3. Roller
1999, 313, mentions, for example, an image (in the Getty) ofCybele surmounted
by a portrait head of Livia.
14 Dion. Hal. 2.19.4-5.
15 Beard 1994, 174, discusses this issue and refers the reader to Sanders 1972.
16 Turcan 1996, 347 n. 75, suggests that okkabos might not only mean bracelet
but necklace as well. In that case, in fact, might all our images be of archigalli?
17 The cult remains apart from traditional religion, only of importance to
outsiders within Rome. Sawyer 1996, 124-5.
99
Shelley Hales
18 See Vitr. De arch. 7.5.3-4 and Hor. Ars P. 11.1-5. For a discussion see
Elsner 1995, 51-8.
19 Quintilian is talking about the eunuchs who served the cult of Artemis. Smith
1996, however, doubts that these priests were eunuchs: see esp. 332-5.
20 Plin. NH35.70.
21 Carcopino 1923, 267-99. Turcan 1996, 49, is also inclined towards this
explanation. For Tiberius as art collector see Suet. Tib. 43.

22 Reade 1998 provides an introduction to the iconography of such reliefs.
23 The issue of hermaphroditism is clearly an important theme for the myth of
the castrated Attis. For a discussion of the role of the bisexual being in ancient
myth see Delcourt 1958.
24 See DeCaro 1987.
25 The principal advocate of this approach has been Kleiner 1977.
26 Popular Roman perceptions of eunuchs can be found in Juv. 6.511-21
and Mart. 5.41

27 See Dion. Hal. 2.19.3-5; Varro Sat. Men. 120.4; Juv. 6.516; Polyb. 21.37.4-7;
Hdt. 4.76; August. De civ. D. 7.26.

28 This cross-dressing element is discussed in this volume by Lightfoot. Roller
1999, 320-5, draws a parallel berween the galli and the hijras of India who
similarly adopt women's clothes.
29 Mart. 3.81.
°
3 Calza 1932, 227-31.
31 Prud. Cath. 10.1076 ff. states that flagellation is a ritual of the archigallus.
32 See, for example, Vermaseren 1977, 99.
33 Catull. 63 provides a literary counterpart in trying to construct the gender
ambiguities of the castrated male. For a brief discussion of this poem see Roller
1997, 551-2.
34 In Varro's Eumenides, the hero is transformed into a gallus simply by putting
on women's clothes: see Roller 1999, 308.
35 Roller 1999, 295, believes that these, at least, are attested in Asia Minor.
36 The relief of a priest of Bellona is pictured and discussed 'by Elsner 1998,
208-9.
37 In her contribution to this volume Lightfoot considers the possibilities of
western influence moulding the cult of Cybele.
38 Roller 1999, 237 ff., feels that this is a sign of the Romanization of the cult as
a western audience tried to make sense of the alien rituals connected to Cybele.
In that case it might be suggested that the very appearance of Attis on the galli is
a sign of a western construction of an eastern cult.
39 The discovery and identification of this sarcophagus is the subject of Calza
1932.
4° Calza 1932, 233.
41 It is interesting to note that Roller's brief discussion of the Capitoline gallus
promotes this figure to the rank of archigallus as well: Roller 1999, 295.
42 Published in Calza 1946, 215-16. I would like to thank Mary Beard for
letting me see a paper on 'Images of castration, or Attis at Ostia'.
100
43 Mart. 13.63 and 13.64 make use of this joke.

44 LIMC2.2, Attis 294, p. 35.
45 Catull. 63.58-67.
46 For a discussion of such images see Vermaseren 1966, 31-8.
47 See, for example, Vermaseren 1966, 56.
48 Published in Calza 1946, 223--4.
49 Vermaseren 1966, 35.

50 Vermaseren 1977, 142.
51 Published in Calza 1946, 216-17.
52 These symbols are often interpreted as markers of Attis' syncretic identity.
See Turcan 1996, pp. 67-8.

53 Published in Calza 1946, 221-2, billed as a hermaphrodite. It is published
explicitly as an Attis in LIMC 2.2, p. 24.

54 A description of the Pans can be found in Calza 1946, 208-10.
55 For instance, the imperial involvement of Augustus is discussed in Turcan
1996, 43-4. The extensive interest shown by Claudius is mentioned passim,

particularly by Carcopino 1923. It is perhaps no surprise to find that Elagabalus
was a fan: SHA, Heliogab. 7.1-3.
56 Beard 1994, 183-7.
57 Turcan 1996, 58-65.
Bibliography
Beard, M.
1994 'The Roman and the foreign: the cult of the "Great Mother" in imperial
Rome', inN. Thomas and C. Humphrey (eds.) Shamanism, History and
the State, Ann Arbor, 164-90.
Calza, G.
1932 'Una figura-ritratto di archigallo', Historia 6, 221-37.
1946 'Il santuario della Magna Mater a Ostia', Mem. Pont. Ace. Rom. 6,
183-205.
Calza, R.
1946 'Sculture rinvenute nel santuario', Mem. Pont. Ace. Rom. 6, 207-27.
Carcopino,].
1923 'Attidea II. Galles et archigalles', Melanges d'archaeologie et d'histoire
40, 237-324.
DeCaro, S.
1987 'The sculptures of the villa ofPoppaea at Oplontis: a preliminary report'
in E. MacDougall (ed.) Ancient Roman Villa Gardens, Washington
D.C., 77-133.
Delcourt, M.
1958 Hermaphrodite. Mythes et rites de !a bisexualite dans l'antiquite classique,
Paris.
Elsner,].
1995 Art and the Roman Viewer, Cambridge.
101
Shelley Hales
1998 Imperial Rome and Christian Triumph, Oxford.

Gruen, E.
1990 Studies in Greek Culture and Roman Policy, Leiden.
Hepding, H.
1903 Attis, seine Mythen und sein Kult, Gieszen.
Kleiner, D.E.E.
1977 Roman Group Portraiture: The fonerary reliefi of the Late Republic and
Early Empire, New York.
Lambrechts, P.
1952 'Livie-Cybele', La Nouvelle Clio4, 251ff
Reade, J.
1998 Assyrian Sculpture, 2nd edn, London.
Roller, L.
1997 'The ideology of the eunuch priest', Gender and History 9.3, 542-59.
Sanders, G .M.
1972 'Gallos' in Reallexiconfor Antike und Christentum, vol. 8, Stuttgart.
Sawyer, D.
1996 Women and Religion in the First Christian Centuries, London and New
York.
Smith, ].0.
1996 'The cult of Artemis at Ephesus', in E.N. Lane (ed.) Cybele, Attis and
Related Cults: Essays in memory ofM.j Vermaseren, Leiden, 323-35.
Turcan, R.
1996 The Cults ofthe Roman Empire, Oxford.
Vermaseren, M.J.
1966 The Legend ofAttis in Greek and Roman Art, Leiden.
1977 Cybele and Attis. The myth and the cult, London.
102
6
EUNUCHS AND GENDER TRANSFORMATION:

Philo's exegesis of the Joseph narrative
Ra 'a nan Abusch
Introduction
The reflections on eunuchism and castration contained in the voluminous
writings of Philo of Alexandria (c. 30 BC-AD 45), especially in his several
exegetical treatments of the Joseph narrative in Genesis, provide a critical
key for examining Philo's conception of the categories of male and
female. 1 The foremost practitioner of biblical interpretation among the
Greek-speaking Jews of Alexandria and our primary source for Jewish
philosophical thought in the Hellenistic period, Philo sought to integrate
the assumptions of scriptural revelation and the traditions and techniques
of philosophical dialectic. The novel forms of cultural accommodation
pioneered by Philo point the way to pagan and Christian philosophical
thought oflate antiquity, in which the innovations forged in the Hellenized
Near East would come to seem commonplace. Like these later thinkers,
such as Plotinus and Origen, Philo was vexed by the complexity of
situating the human body within his philosophical programme. 2 It was
with the aim of clarifYing the place and function of the human body and
the biblical commandments to which it is subject that Philo explores the
figure of the eunuch.
The aim of this chapter is to investigate the relationship between Philo's
use of gender categories and his attitudes toward eunuchism, castration,
and circumcision by exploring his allegorical exegesis of the Joseph narra-
tive. 3 I argue that his various, and sometimes contradictory, treatments
of castration have important implications for Philo's conception of the
commandment of male circumcision. 4 The centrality of circumcision for
the creation of visible Jewish difference in antiquity is unquestionable. 5
The act and sign of circumcision did not, however, function in a social
and cultural vacuum. Circumcision was not simply an anomalous Jewish
peculiarity, but, to the Greek and Roman elites of the early empire,
Jewish circumcision belonged to a larger category we might best term
103
Ra 'anan Abusch
'genital mutilation'. Circumcision was, at least in part, viewed by the

surrounding culture through attitudes shaped by the emergent social
reality of castration and the increasing prominence of eunuchs in political
and social life.
As a Jewish man living in a Hellenistic world among both non-
circumcised gentiles and circumcised Jews, Philo experienced circumcision
as a source of tension. Alien to his surrounding culture but so important
to his own, circumcision served as a locus of hesitation about his own
tradition and his own conceptions of gendered sexuality. Philo's ambivalent
reading of the complex figure of Joseph and the eunuchs he encounters
in Egypt illuminates in important ways his explicit defense of the practice
of circumcision.
Gender transformation and the spiritual life

For Philo, the figure of the eunuch serves as a fertile cultural signifier.
Philo betrays an awareness that castration both reaffirms and endangers
his rhetoric of male supremacy. The ambiguous figure of the eunuch
represents the instability inherent in Philo's 'anthropology', in which
corporeality and sense-perception are fused to intellect and soul within the
compound human being. This conception of human existence is mapped
out in Philo's writings in terms of a stark, though permeable, gender
hierarchy. The malleable male body of the eunuch thus provides Philo
with a dynamic image with which to explore the process of human growth
and transformation so crucial to his philosophical programme. 6
For progress is indeed nothing less than the giving up of the female gender
by changing into the male, since the female gender is material, passive,
corporeal and sense-perceptible, while the male is active; ratiqnal, and more
akin to mind and thought? ·
Gender instability is crucial for Philo's conception of spiritual progress

that he understands as the ability to control and ultimately to transcend
human embodiment through pure intellection. 8 In Philo's blueprint for
human nature, all people, male as well as female, are bisected into the
complementary facets of mind and body, components that are in turn
conceptualized in gendered terms. All bodies, both those of men and those
of women, are gendered female in their imperfection.
This language of progress through gender transformation is echoed
in Philo's discussions of female virginity. 9 For Philo, virginity is radically
opposed to the embodied existence of physically mature women for whom
the stages of menses, defloration and parturition signifY their proper
social function and location. 10 The readoption by Philo's female biblical
104
Eunuchs and gender transformation: Philo's exegesis ofthe joseph narrative
characters of their former virgin state after menopause simultaneously

removes them from the community of women and gives them access to
the male deity as recipients of his divine seed.
The union of human beings that is made for the procreation of children
turns virgins into women. But when God begins to consort with the soul,
He makes what before was a woman into a virgin again, for he takes away
the degenerate and emasculate passions which made it womanish and plants
instead the native growth of unpolluted virtues. Thus God will not talk with
Sarah till she has ceased from all that is after the manner of women (Gen.
18.11), and is ranked once more a pure virgin. 11
The purity represented by the female virgin seemingly represents a rejection

of corporeal desire and experience, but in fact encodes Philo's gendered
language of spiritual procreation in which God is imagined as the potent
male who actively sows his seed in the receptive, feminized human soul. 12
Here Philo's language contains a tension between a rejection of sexuality
and the reinscription of gender categories on a cosmic level. Like the two
halves of a medieval diptych, male and female merge as complementary
images into an integral whole, whose components, when differentiated,
serve as the paradigms not only for human sexual existence, but more
importantly for the actors in the drama of the spiritual life ..
It is by means of this permeability of gender categories that Philo
fashions his language of human religious experience. The human subject
in Philo's writings is almost invariably a male subject. This male subject
must navigate a double transformation in order to adopt a feminized
and passive relationship to the divine, while still safeguarding his male
identity. The embodied male subject sheds his female corporeal aspect,
here conceived of as the 'emasculating passions', while, at the same time,
imagining his soul as a receptive and passive virgin impregnated by God's
divine seed. All the while, he must safeguard his social and physical reality
as a man. I want to argue that the gender indeterminacy of the eunuch,
his status as 'neither male nor female', o1h' appEv OU't£ 8f\A.u, a common
phrase echoed in Philo, represents an alternative strategy for this crisis
of embodiment.
Eunuchs and castration in Philo

Philo include castrates among a lengthy list of different types of male
gender-bending and consistently censures them as transgressors of divinely
ordained gender categories. In one colourful and strident passage, Philo
writes:
In former days the very mention of (pederasty) was a great disgrace, but now
it is a matter of boasting not only to the active partner but to the passive
105
Ra 'a nan Abusch
partners, who habituate themselves to endure the disease of effemination

(v6crov 9itA.Etav). They let both body and soul run to waste, and leave no
ember of their male nature to smoulder. Mark how conspicuously they braid
and adorn their hair, and how they scrub and paint their faces with cosmetics
and pigments and the like. In fact, the transformation of the male nature
to the female is practiced by them as an art and does not raise a blush ...
Certainly you may see these hybrids of man and woman continually strutting
about through the thick of the market, heading the processions at the feasts,
appointed to serve as unholy ministers of holy things, leading the mysteries
and initiations and celebrating the rites of Demeter. Those of them who,
by way of heightening still further their youthful beauty, have desired to
be completely changed into women and gone on to mutilate their genital
organs, are clad in purple like signal benefactors of their native lands ... , each
of them a curse and a pollution of his country. 13
Philo's language here is sweeping and inclusive. He associates the passive

male partner with the ritual attendant of Demeter, 14 who represents the
goddess in the public space of the city. He is attuned to the minute details
of the body language and self-presentation that these 'effeminate' figures
share, believing the ruined state of their souls to be displayed on the
outside. The 'mutilation' of the male genitals represents the most extreme
manifestation of this degeneracy. Yet, even this remarkable and coherent
condemnation betrays undercurrents of ambivalence. Philo seems to
recognize the allure and sheer physical beauty of these effeminates, even
if their powerful sexuality is dangerous and illicit. It is precisely this
juxtaposition of sexuality and the negation of reproductive capacity
that so infuriates Philo. Castration generates an ambival~nt product,
desexed, yet unlawfully sexual, non-reproductive, yet associated with
fertility cults
Philo offers a variety of metaphors for the eunuch scattered throughout
his writings. This imagery fuses the ritual vocabulary of biblical cult
with the rich philosophical language associated with correct perception
and reason. The eunuch is variously like an infertile field 15 or the urban
multitude, 16 in each case unable to generate wisdom, ayovo~ cro<jlia~.
He employs traditional philosophical vocabulary to link the eunuch's
genital mutilation to the deformation of the capacity to use the organs of
perception to pierce through the veil of appearances. The eunuch is like
a man with a cataract, who is 'under the dominion of appearances and
does not perceive what is truly excellent' .17 Like Lucian's exploration of the
fate of a eunuch philosopher whose reproductive organs are considered
prerequisites for attaining a chair of philosophy, 18 Philo attests to the fact
that a man's reproductive capacity was brought to bear on his capacity to
participate in philosophical debate and speculation
106
In many of the passages in which he voices his condemnation of

eunuchs along with other effeminate men, Philo brings the same proof
text, Deuteronomy 23.1: 'He whose genitals are crushed (8/caoia~) and
he whose testicles have been cut out (a1tOK£KOjljl£Vo~) shall not enter into
the assembly of the Lord.' Here, as in Leviticus 21.18-24, the 8/caoia~,
the man with crushed testicles, and the a1tOK£KO!l!l£Vo~, the man whose
testicles have been cut out, 19 are barred from the communal cultic service
along with the child of the prostitute20 and certain classes of foreigners. 21
It is the crushed testicles of the 8/caoia~ that characterize those who
'destroy the perfect Forms', and thus adulterate the unified image of
god. Likewise, the a1tOK£KOjljlEVO~ (or in parallel texts the a1tOK01tOlV) is
likened to the atheist, one who 'denies the very existence of god'. 22 His
severed genitals embody this absence. Philo's imagery thus brings the
ritual system governing the boundaries of the cultic community to bear
on philosophical activity. The Hebrew Bible is concerned with a wide
range of physical deformities, which disqualify one from participation in
sacrificial cult; physical integrity is more central than ethical character
per se. In contrast, Philo explicitly links embodied states to ethical and
philosophical disposition. The more abstract notion of properly gendered
male identity is, for Philo, a prerequisite for participation in both ritual
community and the philosophical life.
Philo's exegesis of the Joseph narrative

Philo's use of the figure of the eunuch in his allegorical writing highlights
these tensions. The vast majority of occurrences of the term Euvouxo~ and
of the related term c:maorov in the Philonic corpus are in fact found in
Philo's exegesis of the Joseph narrative in his several separate treatments
of this themeY Picking up on the use of these terms by the Septuagint
in its description of Joseph's first master Potiphar, 24 Philo creates a vivid
presentation of the events that take place in Potiphar's house. In so doing,
Philo develops Joseph into an ethical and spiritual model.
The story recounted in Genesis chapters 37 to 48 is familiar. Joseph is
the favourite son of Jacob who, having provoked the jealousy and ill-will
of )his brothers, is taken down to Egypt by traders and there sold into
slavery. Mter several notable adventures, including his initial period of
employment in the house of a high government official named Potiphar,
whom the Septuagint labels an Euvouxo~ rather than the ambiguous
0"'.,0 of the Hebrew Bible, Joseph moves his way through the administra-
tive bureaucracy, finally securing the most powerful position in the
Egyptian government after the Pharaoh. And, through a series of dramatic
circumstances, Joseph's newly acquired position of power enables him
107
Ra 'a nan Abusch
to ensure for himself and his family material survival in the face of
dwindling resources.
The gaps in this story are as impossible to resolve as they are numerous.
At the heart of any understanding of this narrative is the hermeneutic
puzzle of how we should read the character of Joseph; is he a model for
self-restraint and piety in the face of temptation or perhaps a vain, self-
promoting pretty-boy? 25 This question vexed the Rabbis as well. Both
possibilities are found side-by-side in Genesis Rabba, a compilation of
Rabbinic exegetical traditions on the book of Genesis. In an extended
treatment of this passage found in Genesis Rabba 87.3, the Rabbis
present Joseph as an ethical model, but also as a tattle-tale who acts with
condescension towards his brothers and as a dandy who calls attention
to his beauty and flair. 26 They suggest that Joseph has been purchased
by Potiphar to serve as a catamite and that Potiphar's castration for his
transgressive sexual preference is a punishment from God.U The Midrash
here even suggests that Joseph 'may be compared to a man who is sitting
in the marketplace penciling his eyes, fixing his hair, and prancing about
(,~i'V~ l"l.,l'1Q, ,,V~~ ii'r10, ,"~~V~ fUQWQQ ).' 28 The Midrash explains
that God, infuriated by Joseph's vanity, sends the wife of Joseph's master
to test his self-restraint. 'If you are a real man,' God says in the Midrash,
'here is a she-bear, come, wrestle with it.' This description of]oseph's body
movements, mannerisms and personal grooming habits go hand in hand
with the characterization of Joseph as Potiphar's potential catamite. In
fact, a careful reading of this description of Joseph's distinctive habitus
suggests that the Rabbinic composers of this Midrash were \J.S attuned as
their Hellenized counterpart to the connection between body language
and ethical character. 29 Like Philo, the Rabbis were sensitive to the clues
encoded in the biblical narrative. 30 1
Certainly Joseph's beauty and self-presentation make' him suspect.

This characterization, coupled with the nature of Joseph's career - first
as a household slave and then as a powerful figure within the Egyptian
bureaucracy -lends credence to my suggestion that his is the classic career
of a eunuch. Like many eunuchs, Joseph begins his career as a young slave.
Joseph, like many young slaves, especially the more attractive ones, 31 would
have fetched a far higher price for the nomadic traders on the Egyptian
slave market if he had been castrated before the sale or by the traders en
route. Like many eunuchs in the ancient world, Joseph is foreign born and
from beyond the boundaries of the ruling empire. And, like them, once
his social and family ties are severed, he becomes entirely dependent
on the support and goodwill of his employers. 32 The jobs he is given
both at the beginning of his career and at its culmination are the highly
108
specialized jobs of household administration and government planning

and policy. A rigorous education would have been absolutely essential
for a Canaanite shepherd-boy to learn and master the intricacies of the
Egyptian governmental bureaucracy. Of course, this reading goes against
the grain of the explicit biblical narrative in which Joseph ultimately
marries and has children. 53 Nevertheless, the question of how to read his
character and his identity as a man lies at the heart of any interpretation
of this narrative. As we will see below, Philo himself grasped the logic of
this portrait of Joseph as a 'literary' eunuch.
Dependent on the Greek translation of the Bible, Philo emphasizes
the fact that Joseph encounters a number of eunuchs during his time in
Egypt. 34 Philo describes Joseph's first master, Potiphar, as:
a eunuch gelded of the soul's generating organs, a vagrant from the men's
quarters, an exile from the women's, a thing neither male nor female, unable
either to shed or receive seed, twofold yet neuter, base counterfeit of the
human coin, cut off from the immortality which, through the succession
of children and children's children, is kept alive for ever, roped off from the
holy assembly and congregation. 35
Philo's allegorical reading of Joseph's encounter with his master's wife
similarly emphasizes her husband's eunuchism. He writes:
For the eunuch and chief cook, in truth the mind itself, deals not in the
simple pleasures but in excessive ones also; it (this type of mind) deserves
the title of eunuch as one who is incapable of begetting wisdom (ayovo~
cro<)lia~). 36
He goes on to elaborate that Potiphar's wife represents Pleasure in its

most extreme form, because she is not only a woman, but, as the wife of
a eunuch, depends on a man whose business it is to provide the excessive
pleasures of the appetites.
This same language is echoed in Philo's allegorization of the baker and
wine-steward whom Joseph later encounters in prison:
Why is it that not a single one of these offices is entrusted to a real man or
woman? Is it not because nature has trained men to sow the germs of life
and women to receive them, and the mating of these two is the cause of
generation and of the permanence of the All.
These eunuchs are instead to be likened to:
a soul which is impotent and barren, a soul which has been made so by
emasculation (E.~~:.uvouxtcrJ.l.EVll~). For such a soul is neither able to drop the
truly masculine seeds of virtue ('ta ap~:.'tfic; app~:.va ro~ a/cTJ8roc; crn£pJla'ta) nor
yet to receive and foster what is so dropped, but like a stony field is only
109
Ra 'anan Abusch
capable of blighting the successive growths, which were meant to live. (He
is like) a craftsman whose work it is to produce pleasure and can produce no
fruit of wisdom. He is neither male nor female (ou'tE &ppT)v ffiv OU'tE Si)A.Eta),
for he is incapable of either giving or receiving seed (mr£p1-1ma). None
such does Moses permit to enter the congregation of the Lord, for
what use can he find in listening to holy words when the knife has cut
away the power of faith and the store of the truth. 37
In this passage, Philo explicitly compares the stewards at the Egyptian court
to those who, by separating themselves from the cycles of reproduction,
have removed themselves from the community of Israel. These passages
employ a set of stock images that travel from one context to another. For
Philo, the eunuch combines physicality, passion, and pleasure with a lack
of reproductive capacity. Eunuchs are a perversion of Philo's conception of
human nature, an embodied nullification of his cosmic economy.
Joseph as eunuch in Philonic exegesis

Negative symbolic valence
Considering the complexity of the character of Joseph and the potency of
the figure of the eunuch within Philo's allegorical idiom, it is perhaps no
surprise that Philo plays with the notion that Joseph himself is a eunuch.
Comparing Joseph unfavorably to Abraham's wisdom and Noah's nobility,
Philo writes that Joseph here represents:
the mind which loves the body and the passions and has been sold into
slavery to the chief-cook of our compound nature, Pleasure. Castrated of all
the male reproductive organs of the soul, and living in indigence of noble
practices, he is unable to receive the divine message, debarred from the holy
congregation in which the talk and study is always of virtue. 38,
Philo articulates his doubts about Joseph as an ethical model using the very
same language and imagery he has otherwise reserved for other castrated
characters from the biblical Joseph narrative. At the conclusion of this
passage, Philo prays on his own behalf that he be unlike this castrated
Joseph figure: 'My soul, if you are snared by the hook of passion, endure
to become prisoner rather than a prison keeper.' Embedded in a cultural
context in which the social reality of eunuchism and castration was always
close at hand, Philo too recognizes the logic of applying this image to Joseph
in order to condemn what he sees as Joseph's less noble characteristics.
Positive symbolic valence

Even more provocative, however, is Philo's portrayal of Joseph as
a eunuch, not in negative hermeneutic play on the complexities ofJoseph's
110
characterization, but instead with the aim of depicting Joseph as a paragon

of self-control and abstinence. This piece of exegesis comes in one of the
passages I have cited above, LegumAllegoriae 3.236, in which Philo depicts
Potiphar's wife as the figure of corrupting Pleasure, wed to a eunuch
who serves 'none other than Pharaoh, destroyer (lit. scatterer) of noble
things'. Yet, Philo interrupts the flow of this invective in mid-stream
with a 180-degree turn in argument:nion, offering a radically different
understanding of eunuchism.
According to another account (Ka-t' aA.Aov A.Oyov) it would be noblest
(aptawv 39 ) to become a eunuch (to c:uvouxov yc.v£a8m), if (in this way) our
soul should be able to escape wickedness and unlearn passion. So Joseph too,
the self-controlling character (6 E:yKpariJ<; tp67to<;), when pleasure says to him
'Sleep with me and, being human, indulge in human passions and enjoy the
delights that come in life's course', refuses to comply with her. 40
The phrase used here, Kat' aA.A.ov A.Oyov, might alternatively be translated
'According to an alternative interpretation'. While this phrase may merely
mark the contrast between the two interpretations Philo provides, it
may indicate that he is here transmitting a received exegetical tradition.
Whatever the best translation of this phrase may be, it is certain that the
abrupt change in discourse indicates that this 'alternate' reading stood out
for Philo as much as it does for the modern reader. It is no longer Potiphar
who is the eunuch, but Joseph. And the eunuch, far from representing
emasculate and emasculating passion, signifies the transcendence of the
physical and sexual self.
In light of Philo's frequent condemnations of eunuchs and castration,
this interpretative tradition is novel and surprising. Yet, I believe that this
passage articulates an undercurrent in Philo's approach to the mutually
constituting problems of gender and embodiment. In fact, in a passage
dealing precisely with the struggle to overcome human desire, Philo
relates:
To my thinking, those who are not utterly ignorant would choose to be
blind rather than see unfitting things, and to be deprived of hearing rather
than listen to harmful words, and to have their tongues cut out to save them
from uttering anything that should not be said ... Certain wise men, they tell
us, while being tortured on the wheel to induce them to reveal secrets have
bitten off their tongue, and so contrived a worse torture for their torturers,
who found themselves unable to obtain the information they wanted. It
is better to be made into a eunuch than to rage after sexual intercourse
(E:~c.uvouxta8Tjvm yc. !llJV Ufl£tvov f17tpo<; auvouaia<; A.unav). 41
How are we to understand such texts in which castration serves as a trope

for the spiritual perfection of the wise?
111
Ra 'a nan Abusch
Castration as spiritual progress in Philo

In his writings, Philo consistently uses the same language of excising to
describe castration as well as to describe the process of separating soul from
body. In these contexts, the words EK'tEjlVOJ, anoK01t'tOJ, and anO'tEjlVOJ,
often occurring in pairs or interchangeably, signifY not the 'excision' or
'crushing' of the male genitals, but the rejection of physicality. 42
More significantly, the semantic field that undergirds Philo's description
of both castration and spiritual progress is identical to his discussion of
the allegorical meaning of circumcision.
Circumcision assimilates the circumcised penis to the heart. For as both are
framed to serve for procreation, thought being generated by the spirit force
in the heart, living creatures by the reproductive organ ... Thus the legislators
thought to punish the organ of sexual intercourse, making circumcision
the figure of the excision of excessive and superfluous pleasure (m::pt'tOJlTJV
nEpt't'tll\; EK'tOJlTJV Kat n:A.Eova~ou<JTJ\; iJoovil\;). 43
Philo's intentional play on the words 1tept'tOjlTJV (circumcision) and

1tept't'til~ EK'tOjlTJV iJ8ovfJ~ (excision of superfluous pleasure) emphasizes the
ethical symbolism of the act of circumcision. For Philo, circumcision is
not merely an empty commandment, but represents the profound spiritual
truth that the male individual must, through struggle, learn to overcome
the body by cutting out the passions from the heart. And, unlike many
radical allegorizers in his own community, Philo recognized the need to
enact such commandments physically as well as in spiritual terms. 44
In one sense, both circumcision and castration define the boundaries
of community, one a prerequisite for inclusion, the othe,r a mark of
exclusion. 45 Within Philo's Platonizing framework, however, castration,
similar to circumcision, provides an apt metaphor for spiritual progress.
Philo cannot ignore the symbolic power of castration. Far from imagining
the eunuch as a barren field, we might imagine him as a properly pruned
vine that is to bear wholly virtuous fruit. In many respects then I believe
that Philo betrays an awareness that all circumcised Jewish men have in
some respects undergone an alteration to their reproductive organs as
a ritual of sanctification to ensure their inclusion in a sacred community.
Every circumcised Jewish man is, as it were, a sacred castrate.
Literal and allegorical reading strategies: Philo's legacy

This parallelism between circumcision and castration, of course, raises
the question of literal as opposed to allegorical understandings of such
radical acts as castration and other forms of self-mutilation. I believe that
what we find here are early signs of a massive shift between the world of
sacrificial cult, in which wholeness of body, just as proper genealogy, is
112
a prerequisite for approaching the divine, and a world in which impurity

is understood as a state inherent in the human condition. We are thus not
far from a world in which the rationale for Origen's alleged self-castration
would have been clearly understood as a spiritual lesson inscribed in the
body. Indeed, although Origen attenuated in his writings the message of
this lesson enacted in his youth, 46 he is still grappling with the significance
of castration for the problem of embodiment in his later works. In his
commentary on Matthew, likely penned in his old age, 47 Origen uses
Philo's approval of castration cited above in his justification of Jesus' call
to self-castration at Matthew 19.12:
And Philo, who enjoys a high reputation among intelligent people for many
subjects discussed in his treatises on the Law of Moses, says in the book
entitled On that the worse is accustomed to attack the better that 'it is better to
be made into a eunuch than to rage after sexual intercourse'. 48
Although elsewhere in this commentary Origen repudiates castration, the

inclusion of Philo's unusual endorsement of the practice attests to the
durability of the logic that sustained it as well as the controversial power it
continued to impose on the philosophical imagination. 49 Within Origen's
radical call to self-transformation, this act makes perfect sense.
In many respects, the eunuch resembles the primordial human being
in Philo's reading of the Genesis creation story, the rcpffi1:o~ avepomo~.
Like the eunuch, this figure, created in the divine image (1:ou 1m1:a 'tTJV
dKC>Va 8wu), is 'neither male nor female' (oth' app£v o{h£ 8ft/-;u). 50 It
refuses the consequences of gender differentiation which first established
the irreparable barrier between the human and the divine. The fall of the
primordial human into an embodied, engendered existence is inverted
by the eunuch's repudiation of his human condition. Yet, unlike this
primordial human, Philo's gender hierarchy guarantees that the figure
of the eunuch must always also represent the dangerous slippage back
into the passive and emasculating experience of sensual perception and
passion. In his often contradictory reflections on the ethical and spiritual
valence of the figure of the eunuch, Philo prefigures intellectual debates
concerning the role of self-mutilation within early Christianity. The figure
of Joseph represents an important interpretative locus for Philo's seminal
articulation of the increasingly vital desire among his contemporaries to
mould and transform the gendered and embodied self
Acknowledgements
I owe a great deal of gratitude to mentors at Princeton and in Amsterdam. Jan
Maarten Bremer and Ross Kraemer have added much keen insight. I want to
113
Ra 'anan Abusch
express special appreciation to David Runia for having lent a virtual stranger
much time and creative energy. And, as always, Jan Willem van Henten and
Athalya Brenner have provided a special combination of ideal working conditions
and generous mentoring. I dedicate this chapter to my mother, Susan Abusch,
who has battled through so many transformations of her body and her self
Notes
1 For the texts ofPhilo's writings extant in Greek, I have used Colson 1929-65,
unless otherwise noted. For Quaestiones in Genesin and Quaestiones in Exodum,

I have used Marcus 1953.
2 For the impact of Philo's thought on Origen's conception of castration, see
Caner 1997 and Stevenson in this volume.

3 For the classic discussion of Philo's use of gender categories, see Baer 1970.
Baer's remarks are surprisingly close to feminist readings of Philo's use of gender
categories, despite his 'ideological' naivete. 'It is precisely Philo's depreciation
of woman that permits him to use her as a symbol of sense-perception, and,
on the other hand, his castigation of female sense-perception and the natural
world which leads in turn to a further devaluation of woman' (p.40). However,
where Baer perceives a fundamental difference berween Philo's use of male and
female within the created sphere and within the sphere of the undifferentiated
and uncreated Original Man, npiiho~ av8prono~, Mattila 1996 rightly argues that
Philo's gender categories pervade his anthropology and ontology at every level,
thereby shaping his conception of the divine as well. .
4 For discussion of Philo's attitude towards and conception of circumcision see
Borgen 1982, Hecht 1984 and Collins 1985.

5 I provide here a very partial list of only the most important discussions of
circumcision as a mark of difference: Barclay 1998; Borgen 1982; Boyarin 1992;

Collins 1985; Cohen 1993. Tacitus himself already states this rationale explicitly:
'They (the Jews) instituted circumcision of the genitalia so t~at they could be
recognized by their difference' (Hist. 5.5.1-2).
6 This characterization of Philo's 'gender gradient' as dynamic and permeable
is intended as a corrective to Mattila's emphasis on its 'pervasiveness and rigidity'

(1996, 129). Mattila's systematizing treatment of Philo's conception of gender
leads her to formulate an almost algebraic model that does not take seriously the
dialectic berween Philo's penchant for static abstractions and his commitment to
a transformational anthropology.
7 Qptaest. in Ex. 18.
8 Foucault 1986, 64-5, has noted the importance of self-control for the
cultivation of the male subject in late antiquity. 'Enkrateia is characterized more

by an active form of self-mastery, which enables one to resist or struggle, and
to achieve domination in the area of desires and pleasures ... Located at the axis of
struggle, resistance, and combat, it is self-control, tension, continence; enkrateia
rules over pleasures and desires, but has to struggle to maintain control.' As many
critics have noted, Foucault does not pay sufficient attention to the gender-specific
114
aspect of this ethos of enkrateia. In Philo, spiritual and material self-cultivation

through the practice of physical restraint coupled with philosophical activity is
consistently formulated in gendered language.
9 See Baer 1970, 51-5, on Philo's spiritualization of virginity and its relation-
ship to the divine impregnation of the soul. See also Som. 2.273, Deus Imm. 138,
and Spec. Leg. 1.129 for the comparison between virgins and widows.
10 Recent discussions of Philo's representations of women, both biblical figures
and contemporaries, have been superficial and unsatisfying (Wegner 1982 and
1991; Sly 1990). Kraemer 1989 provides a richer account of the function of
gender within Philo's idealized portrait of the female proto-monastic Therapeutae
in De Vita Contemplativa.
11 Cher. 50. In Praem. Poen. 158-60, Philo writes: 'For when the soul is
"many," full that is of passions and vices with her children, pleasures, desires,
folly, incontinence, injustice, gathered around her, she is feeble and sick and
dangerously near to death. But when she has become barren and ceases to produce
these children or indeed has cast them out bodily she is transformed into a pure
virgin.' C£ Quaest. in Gen. 4.117-19, 242; Quaest. in Ex. 2.3.
12 See the comprehensive table at Baer 1970, 58-61 for sources relating to the
divine impregnation of the soul. Adjectives such as 'untrod' (c'i~a-ro<;), 'simple'

(anaM<;), 'young' (v£o<;), and the noun 'blossom' (c'iv8o<;), normally used to
describe the virgin (nap8£vo<;), are in this context juxtaposed to Philo's procreative
language; see especially Som. 1.199 where the invisible seed (aopa-ro<; crnopa) of
divine speech (Myot) is said to impregnate 'the still young and simple souls' (ai
En VEal Kat anaA.a\, 'lfUxai). C£ Cher. 46; Migr. Abr. 34-5; Mut. Nom. 134;
Abr. 100-1; Deus Imm. 136-7; Praem. Poen. 159-60; Poster. C. 134-6; Spec.
Leg. 2.29-31; Leg. All. 3.40; Det. Pot. Ins. 60, 147; Ebr. 30 for similar uses of
0"1tep~a and crnopa. For the use of agricultural imagery in Graeco-Roman ethical
literature, see duBois 1988 and King 1994.
13 Spec. Leg. 3.40--42. The notion of the 'female disease' is contained in a wide
variety of passages in the Philonic corpus, e.g. Vit. Cont. 60 and Spec. Leg.
1.325 and 3.38--40. See Szesnat 1998, 97-107, for an excellent discussion of
this theme.
14 The reference here to Demeter is inexplicable. Philo, or a later tradent, no
doubt conflated this native Greek deity with one of several mother goddesses
whose attendants were castrated priests, such as the galli of the Anatolian goddess
Cybele. On the relationship between goddesses and their castrated attendants,
see the standard treatment by Vermaseren 1977 and, for a recent reassessment,
Sodergard 1993. See also Lightfoot and Hales in this volume.
15 Ebr. 211.
16 ]os. 58-66.
17 ]os. 58-9.
18 Eunuch 10, Demonax 12. See Gleason 1990 and 1995 for brilliant discussion
of this material. She analyses the function of rhetorical training and performance
for the cultivation of authority and the concomitant construction of masculinity
among Roman sub-elites, arguing that a rigorous semiotics of gesture, tonality,
115
Ra 'anan Abusch
and style by which men scrutinized each other undergirded their strategies of
competition and alliance.
19 This distinction between 'crushing' and 'cutting' perhaps echoes the medical
procedures of compression and excision described by Paul of Aegina in his Epitome

ofMedicine, V1.68 (cited in Tougher 1997, 175).
20 LXX translates Hebrew ,~QQ (meaning uncertain, probably 'bastard') as 6
eK n:opvll<; ('son of a prostitute').

21 The Ammonite and Moabite of the following verse, Deuteronomy 23.4
22 Spec. Leg. 2.344.
23 The relevant texts are found in De Somniis 2, De ]osepho, and finally in
a more general work called Legum Allegoriae 3.

24 HB Genesis 37.36 and 39.1 both read l''!V,~ C"',C. LXX translates tqi
<maoovtt «<>aparo and 6 euvouxo<; ci>aparo respectively. The Hebrew C"',C has
two possible, and not necessarily incompatible, meanings in the present context,
'eunuch' and/or 'officer'. In only some biblical texts does C"', C carry the
particular meaning of 'eunuch' (e.g., Isaiah 56.3-4; Esther 2.3, 14-15; 4.4-5).
The Akkadian origin of the word is sha~reshi, 'he who is chief'. There is still no
scholarly consensus about the exact meaning of this term. A large number of
artistic representations of the unbearded sha-reshi exist in which they are depicted
alongside the sha-ziqni, the bearded ones; these two classes of court officials are
frequently used together as a comprehensive term for Assyrian governmental
officials, but what exactly these words signifY has received insufficient attention.
See Grayson 1995 for the latest synthetic treatment of the question. LXX makes
this identification unambiguous.
25 For a collection of such traditions see Kugel1990 and Neihoff 1992.
26 For similar traditions, see BT Sotah 12b, 36b; Genesis Rabba 83.6, 86.3-6,
91.6. Levinson 1997, 269, reads these traditions through the lens of Graeco-
Roman theatrical culture, demonstrating that 'through a cross-coding of the
gender and cultural codes in the Joseph narrative, the hegemonic discourse of
the theater is exploited to denigrate the dominant foreign cvlture as a form of
deviant identity'. Although sympathetic to Levinson's focus on ethnicity, this
chapter instead emphasizes the way Philo's ethical and philosophical categories are
mapped as gender transgression and transformation. It still remains to contrast
Rabbinic and Philonic exegesis of the Joseph narrative in terms of their (differing?)
relationships to the surrounding Graeco-Roman culture.
27 In an attempt to explain the apparent identification of Potiphar ~~"'Q,~)
the 'eunuch' with Potiphera (V,~ "''Q,~ ), the priest of On of Genesis 41.45,
the father of Aseneth and eventual father-in-law of Joseph, the Rabbis suggest
in Genesis &tbba 87.3 that Potiphar had acquired Joseph in order to use him for
sexual purposes and was castrated as a result. The punishment of castration is
derived from a word play on Potiphar's name, whose last element V,~ means
'to destroy', and hence 'to castrate'.
28 Genesis Rabba 87.3. Translation mine.
29 Bourdieu 1977, 74, defines habitus as 'systems of durable, transposable
dispositions, structured structures predisposed to function as structuring
116
structures, that is, as principles of the generation and structuring of practices

and representations'.
30 Along with Philo, other Hellenistic retellings of the Joseph narrative pick up
on many of these same clues in the biblical text. In the Testament ofjoseph, as in
Philo, Joseph is portrayed as the enkratic man (tou £yKpatou<;) par excellence,
an ethical model of temperance and chastity (Hollander 1981). His obsessive
concern for his chastity transforms his sexual potential into a powerful and
attractive religious force. Similarly, in the proto-Romance joseph and Aseneth,
Joseph's combination of beauty and aloofness lend him power and authority.
Aseneth even mistakes the angel who comes to her for Joseph, perhaps alluding
to Joseph's youthful and indeterminate beauty.
31 Joseph is the only man in the Bible described as beautiful and comely,
M~,Q M~~, ,~l'1-M~~ (Genesis 39.6). These epithets may merely echo Joseph's
maternity through Rachel, ·since the phrase is only otherwise used to describe
her (Genesis 29.17). Yet, later narrative and exegetical treatments of Joseph's
character and behaviour, Jewish, Christian, as well as Islamic, make his beauty
central to their interpretations of the narrative.
32 This description of the court eunuch draws heavily on Hopkins 1978, where
for the first time a sociological analysis is applied to the function of the eunuch
at the late Roman imperial court.
33 Genesis 41.50-52.
34 Jubilees chs. 34 and 39 likewise emphasizes that, like Potiphar, the two
stewards whom Joseph encounters in prison are eunuchs.

35 Som. 2.184.
36 Leg. All. 3.236. Translation mine. Mut. Nom. 173, a passage likewise linking
Potiphar's eunuchism and his role as chief-cook, argues: 'Potiphar is eunuch
and chief-cook; eunuch because he has scant store of excellence and has lost
by mutilation the soul's organs of generation, unable further to sow and beget
anything that tends to discipline; cook because in cook-like fashion he slaughters
living beings, chops and divides them, piece by piece, limb by limb, and moves in
a chaos oflifeless carcasses, immaterial rather than material.'
37 Ebr. 211-12. C£ Ebr. 220.
38 Deus lmm. 111.
39 aptcrtOV can also mean 'finest' or 'best'. Regardless of what the best translation
is, this word is especially emphatic in this context, where it possesses both
a normative and a philosophical meaning.
40 Leg. AIL 3.236-7.
41 Det. Pot. Ins. 176. Origen cites this very passage in his justification of Jesus'
call to self-castration at Matthew 19 .12. See below for discussion.
42 See Som. 2.64 where Philo writes: 'For just as we find on trees, to the
great damage of genuine growth, superfluities which the husbandmen purge

(Ka8atpoucrt) and cut away (a1totE.Ilvoum) to provide for their necessities, so the
true and simple life has for its parasite the life of falsity and vanity, for which
no husbandman has hitherto been found to excise (a1tEKO'IfE) the mischievous
overgrowth, root and all.' C£ Mut. Nom. 173; Leg. All. 3.8; Som. 25; Ebr. 69.
117
Ra 'a nan Abusch
43 Spec. Leg. 1.9. Later in Spec. Leg. 1.303 Philo adds: 'But the law says that some
are uncircumcised in the heart. Circumcise the hardness of your heart ... prune
away from the ruling mind the superfluous overgrowths sown and raised by the
immoderate appetites of the passions and planted by folly, the evil husbandman
of the soul.' Likewise, at Mig. Abr. 92, he writes: 'It is true that receiving
circumcision does indeed portray the excision of pleasure and all passions, and
the putting away of the impious conceit, under which the mind supposed that it
was capable of begetting by its own power: but let us not on this account repeal
the law laid down for circumcision. Why, we shall be ignoring the sanctity of
the Temple and a thousand other things, if we are going to pay heed to nothing
except what is shown us by the inner meaning of things. Nay, we should look on
all these outward observances as resembling the body and their inner meaning
as resembling the soul.'
44 Spec. Leg. 1.1-10.
45 Wolfson 1987, 196, likewise highlights the role of circumcision as a pre-
requisite for inclusion in the religious community and, therefore, for access
to religious experience in Rabbinic and mystical Judaism. He finds the most
powerful expression of this 'nexus between circumcision and the appearance of
God' in a text from Numbers Rabba 12.10. At the core of this complex exegetical
passage lies the statement: 'The "Daughters of Zion" are those [males] who were
distinguished by circumcision, for if they were uncircumcised, they would not
have been able to look upon the [divine] presence.'
46 Hanson 1966, 82, argues in favour of the historicity of Origen's self-
castration: 'In view of this evidence of self-castration as a known and on the

whole approved custom in the Christian Church of Origen's time, it seems
captious to doubt that he did perform this act in his youth, even if in his old
age he decided that it was not a permissible one.' Brown 1988 likewise supports
the historicity of this report.
47 Crouzel1989, 43.
48 Commentarium in Evangelium Matthaei 15.3 citing Det., Pot. Ins. 176. See
n.41 above. Translation from Runia 1993, 160. ·

49 Stevenson, in his contribution to this volume, argues that, even here in his
interpretation of Matthew 19 .12, Origen's affirmation of the practice of castration

is embedded in a rhetorical method that aims to 'sublimate' the reader's desire for
self-mutilation into the realm of language. Whether or not this interpretation is
valid, Origen's embodied state as a castrate in conjunction with his radical call
to self-transformation continued to exert a hold on the Christian imagination.
His willingness to cite Philo approvingly in the context of Jesus' call to self-
castration reaffirms the vitality of this practice. This view of Origen is indebted
to Brown (1988, 169) who writes: 'What Origen may have sought, at the time,
was something more deeply unsettling. The eunuch was notorious (and repulsive
to many) because he had dared to shift the massive boundary between the
sexes. He had opted out of being male ... Deprived of the standard credential
of a philosopher in late antique circles - a flowing beard - Origen would have
appeared in public with a smooth face, like a woman or like a boy frozen
118
into a state of prepubertal innocence. He was a walking lesson in the basic

indeterminacy of the body.' See also Kuefler 2001 for synthetic analysis of
Christian discourse concerning the figure of the eunuch.
50 Op. Mund. 134.
Bibliography
Baer, R.A.
1970 Philo's Use ofthe Categories ofMale and Female, Leiden.
Barclay, ].M.G.
1998 'Paul and Philo on circumcision: Romans 2.25-9 in social and cultural
context', NTS 44, 536-56.
Bourdieu, P.
1977 Outline ofa Theory ofPractice, trans. R. Nice, Cambridge.
Boyarin, D.
1992 '"This we know to be the carnal Israel": circumcision and the erotic life
of God and Israel', Criticallnquiry 18, 474-505.
1995 'Homotopia: the feminized Jewish male and the lives of women in Late
Antiquity', Diffirences7.2, 41-81.
Brown, P.
1988 'I beseech you: be transformed', in The Body and Society: Men, women,
and sexual renunciation in early Christianity, New York, 160-77.
Caner, D.
1997 'The practice and prohibition of self-castration in early Christianity',
VC51, 396-415.
Cohen, S.J.D.
1993 '"Those who say they are Jews and are not": how do you know a Jew
in antiquity when you see one?', in S.J.D. Cohen and E. Frerichs (eds.)
Diasporas in Antiquity, Atlanta, 1-45.
Collins, J.J.
1985 'A symbol of otherness: circumcision and salvation in the first century',
in]. Neusner and E. Frerichs (eds.) 'To See Ourselves as Others See Us~·
Christiam, jews and 'Others' in Late Antiquity, Chico, 163-86.
Colson, F.H. (trans.)
1929-1965 Philo, Loeb Classical Library, 10 vols., Cambridge, Mass.
Crouzel, H.
1989 Origen, trans. A.S. Worrell, San Francisco.
duBois, P.
1988 Sowing the Body: Psychoanalysis and ancient representations of women,
Chicago.
Foucault, M.
1986 The Use ofPleasure, trans. R. Hurley, New York.
Gleason, M.W
1990 'The semiotics of gender: physiognomy and self-fashioning in the
second century C.E.', in D. Halperin,]. Winkler, and F. Zeitlin (eds.)
Before Sexuality, Princeton, 389-415.
119
Ra 'anan Abusch
1995 Making Men: Sophists and self-presentation in ancient Rome, Princeton.

Grayson, A.K.
1995 'Eunuchs in power: their role in the Assyrian bureaucracy', in M. Dietrich
and 0. Loretz (eds.) Vom Alten Orient zum Alten Testament: Festschrift
for Wolfram Freiherrn von Soden zum 85. Geburtstag, Neukirchen-Vluyn,
85-97.
Hanson, R.P.C.
1966 'A note on Origen's self-mutilation', VC20, 81-2.
Hecht, R.D.
1984 'The exegetical context of Philo's interpretation of circumcision', in F.E.
Greenspahn, E. Hilgert and B.L. Mack (eds.) Nourished With Peace:
Studies in Hellenistic judaism in memory of Samuel Sandmel, Chico,
51-79.
Hollander, H.
1981 joseph as an Ethical Model in the Testaments of the Twelve Patriarchs,
Lei den.
Hopkins, K.
King, H.
1994 'Sowing the field: Greek and Roman sexology', in R. Porter and M. Teich
(ed.) Sexual Knowledge, Sexual Science: The history ofattitudes to sexuality,
Cambridge, 29-46.
Kraemer, R.
1989 'Monastic Jewish women in Graeco-Roman Egypt: Philo Judaeus on
theTherapeutrides', Signs 14,342-70.
Kuefler, M.
2001 The Manly Eunuch: Masculinity, gender ambiguity, and Christian ideology
in Late Antiquity, Chicago.
Kugel, L.
1990 In Potiphar's House, New York.
Levinson, J.
1997 'An-other woman: Joseph and Potiphar's wife. Staging the body and
politics',jQR 87, 269-301.
Marcus, R. (trans.)
1953 Philo Supplements, Loeb Classical Library, 2 vols., Cambridge, Mass.
Mattila, S.L.
1996 'Wisdom, sense perception, nature, and Philo's gender gradient', HTR
89.2, 103-29.
Meeks, WA.
1974 'The image of androgyne: some uses of a symbol in earliest Christianity',
History ofReligions 13, 165-208.
Neihoff, M.
1992 The Figure ofjoseph in Post-Biblical jewish Literature, Leiden.
Ringrose, K.M.
(ed.) Third Sex, Third Gender: Beyond sexual dimorphism in culture and
120

Runia, D.
1993 Philo in Early Christian Literature, Minneapolis.
Satlow, M.
1995 Tasting the Dish: Rabbinic rhetorics ofsexuality, Atlanta.
Sly, D.
1990 Philo's Perception ofWomen, Atlanta.
Si:idergard, J.P.
1993 'The ritualized bodies of Cybele's galli and the methodological problem
of the plurality of explanations', in T. Ahlback The Problem of Ritual:
Papers read at the symposium on Religious Rites held at Abo, Finland,
13th-16th ofAugust 1991, Stockholm, 167-93.
Szesnat, H.
1998 '"Pretty boys" in Philo's De Vita Contemplativa', The Studia Philonica
Annual10, 87-107.
Tougher, S.
1997 'Byzantine eunuchs: an overview, with special reference to their creation
and origin', in L. James (ed.) Women, Men, and Eunuchs: Gender in
Byzantium, New York, 168-84.
Vermaseren, M.J.
1977 Cybele and Attis: The myth and the cult, London.
Wegner, J.R.
1982 'The image of women in Philo', SBL Seminar Papers 21, 551-63.
1991 'Philo's portrayal of women: Hebraic or Hellenic', in A.J. Levine (ed.)
Women Like This: New perspectives on jewish women in the Greco-Roman
World, Atlanta, 41-66.
Wolfson, E.R.
1987 'Circumcision, vision of God, and textual interpretation: from Midrashic
trope to mystical symbol', History ofReligion 27, 189-215.
121
7
EUNUCHS AND EARLY CHRISTIANITY
Walter Stevenson
Introduction
Modern students of early Christianity have tended to avoid the topic
of ancient castration. 1 For instance, while reading through the Penguin
translation of Eusebius' Ecclesiastical History, one comes upon an early
Father named Melito the Eunuch. The editor's footnote states that 'eunuch'
was just an early Christian way of saying 'celibate'. 2 Of course we might
expect someone translating for a popular audience to brush off the topic,
but the discomfort with eunuchs lives also at more detailed levels of
scholarship. If one were to glance quickly at the most recent edition of
Melito the Eunuch, one would learn in the introduction that, 'we may
infer. .. he (Melito) was ascetic, since "eunuch'' at this period usually meant
"celibate"' - this without any further evidence given. 3 Perler's edition
states the same opinion citing a list of passages, which I will show do not
fully support his point. 4 Only Blank's edition is open to the possibility
that Melito was a castrated man. 5 And yet, in spite of a wide array of
celibate Fathers in the second and third centuries, there are none except
Melito called the 'eunuch'. Though scholars may well have good reason
to avoid a subject they find distasteful, it may still be worth the time to
consider the evidence.
In this chapter I will try to show that, in fact, eunuchs were prominent
in the early development of Christianity, and in particular that the
castrated state of the third-century Church Father Origen is significant
in understanding both the social and intellectual origins of Christianity.
Though Origen's emasculation leads to a fruitful and interesting discussion
of castration as sexual denial in the second and third centuries, this
chapter will turn instead to investigate a fundamental paradox in Christian
Alexandria. 6 It seems that Christians were practicing castration, while
officially discouraging it. Alexandrian authors were subtly allegorizing the
prevalent Judea-Christian texts on eunuchs, while believers were acting on
remarkably literal interpretations of these texts. And the veritable founder
123
Walter Stevenson
of Christian theology, Origen, was arguing against castration though he

had apparently emasculated himsel£
I will begin with a discussion of Perler's assertion on the meaning of
'eunuch', then provide a survey of the historical evidence for Christian
castration, treating Origen, his Alexandrian context, and Judea-Christian
sacred texts. Once I have outlined the nature of Judea-Christianity's
paradoxical reaction to eunuchs, I will attempt to consider this paradox
from rhetorical and historical perspectives. Finally I will tentatively argue
that Origen, through his life and his writings, steered early Christianity
onto a new course for discussing and understanding castration.
Euvouxo~ in early Christian language

It would be tendentious to argue that Euvouxo~ never means 'celibate'
in the first few centuries of our era. But it can be shown that the term
'eunuch' held a far more complex range of meanings than 'celibate' in
early Christian texts. Both Greek and Latin have explicit words to describe
a person who is unmarried (or refraining from sexual activity): aya~o~ and
caelebs seem most common. Both Greek and Latin also have a variety of
words for 'eunuch' varying from Euvouxo~ to crnaorov to the more explicit
8A-aoia~, 8A-t~ia~ anoKEKO~~E.vo~, and castratus. To complicate the issue
further, there clearly are a variety of 'eunuchs' living at the time of our
early Christian authors. I would break these down into those who are
born in a variety of conditions without strong masculine characteristics,
those who had 'moderate' destruction of the gonads (8A,aoia~ and 8A,t~ia~)
and those who had radical surgery (anoKEKO~~E.vo~ and ca,stratus). Given
the complexity of both terminology and culture, it would be simplistic to
decide that Christian authors from Matthew to Methodius, whenever they
used any word for eunuch, meant precisely 'celibate'. '
Inspection of Perler's list of citations will show how complex the terms
and issues are. First Matthew 19.12: almost everyone will agree that
Matthew did not intend a literal interpretation of these words:
Eicrt v yap euvouxot Ot nvec; EK KOtA.iac; J.!TJ'tpoc; eyevvitSTJcrav o1hroc;, Kat eicrt v
euvouxot Ot 'ttvee; euvouxicr9Tjcrav U1t0 trov av9pcimrov' Kat Eicrt v euvouxot
Ot ttvec; EUVOUXt<JaV eautouc; bt<l 'tllV ~acrtA.etaV tffiv oupavrov.
There are eunuchs born so from their mother's womb, there are eunuchs
made so by human agency and there are eunuchs who have made themselves
so for the sake of the kingdom of Heaven.
But it is not clear that the author intended us to substitute 'celibate' or

'incapable of marriage' every time we see t::uvouxot.? If Matthew wanted
'celibate' he could have written aya~o~, as Paul did in Corinthians 7.32, or
124
Eunuchs and early Christianity
7tap9£vo~ , as in Apocalypse 14.4, or a7t07tap9£v£ucra~ 8 or something more

specific. ayaJ.LO~ is very precise, cX1t01tap9EV£'UO"a~ implies heroic ascesis,
and 7tap9£vo~ would draw up images of humble, maidenly purity, while
Euvouxo~ certainly could not do the same for Matthew's contemporary
readers. 9 All these terms can roughly mean 'celibate', but, as Origen's
apparent misunderstanding shows, the author of Matthew's Gospel made
a great mistake writing £Uvouxo~ if he really meant 'celibate'.
Justin Martyr at 1.15 of his first Apology does seem to be reading
Matthew 19.12 figuratively, though we will see that a few pages later he
applauds a young Christian for trying to have his gonads excised. 10 If this
young man's petition had been successful, there is no doubt he would
have been called Euvouxo~.
Athenagoras interestingly juxtaposes the word 7tap9Evia with the word
Euvouxia, 'Ei 8£ -co £v 7tap9£viq Kat £v Euvouxiq J.lEtvm J.l&A.Aov 7tapim:'Tlcrt
'CeQ 9£cQ ... ', 'if remaining in maidenly purity and in the state of a eunuch
offers more to God ... '. 11 Though the author may be padding his text here
with tautology, it is just as likely that two distinct states were intended.
This passage does not exclusively support the meaning 'celibate'.
Since Eusebius and Methodius wrote significantly later, they are less
forceful witnesses to Perler's assertion. Nevertheless Eusebius concedes that
Origen castrated himself, even if he insists that Origen made a mistake
in interpreting Matthew's usage of £Uvouxo~. If £Uvouxo~ only meant
'celibate', then both Eusebius and Origen are showing ignorance in
discussing the various possible interpretations of the word. Methodius, in
a Platonizing dialogue on chastity, has one of the characters make reference
to Matthew 19 .12, saying that Christ was discussing a7to7tap9£v£ucrav-r£~
when he discussed the different types of eunuchs. 12 Methodius' character
felt the need to gloss £Uvouxo~ with a7to7tap9£v£ucra~, and this implies
that the meaning of £Uvouxo~ as 'celibate' was not immediately obvious
to Methodius or his readers.
Which brings us to Tertullian. Perler is right that Tertullian used spado
to mean 'celibate' with reference to Matthew 19.12 in his De Baptismo 6.
But this usage is not consistent throughout his work. In fact, Tertullian uses
the words castratus, eunuchus, and spado often, and sometimes apparently
interchangeably. For instance, his rhetorical penchant leads him to vary
terms apparently for euphony, as when he states the following in his de
Monogamia 3:
... etiamsi totam et solidam virginitatem sive continentiam Paracletus hodie
determinasset, ut ne unis quidem nuptiis fervorem carnis despumare per-
mitteret, sic quoque nihil novi induere videretur; ipso Domino spadonibus
aperiente regna caelorum, ut et ipso spadone; ad quem spectans etApostolus,
propterea et ipse castratus, continentiam mavult.
125
"Walter Stevenson
... even if the Holy Spirit had prescribed complete and genuine virginity
or continence so that He would not allow the fervor of the flesh to froth
over even in a single marriage, even so nothing surprising would seem
to be introduced. The Lord himself opened the Kingdom of Heaven to
eunuchs [spadonibus], as he himself was a eunuch [spado]. And the Apostle,
looking at his example, moreover himself a eunuch [castratus], preferred
continence.
Here not only is Christ a eunuch, which clearly implies a celibate, but
Matthew becomes a castratus. It appears that Tertullian, having used the
word spado twice already, just wanted to get an effect with a different
word that would both imply celibacy and also make reference to a reading
of Matthew 19.12. It would be absurd to argue from this example that
Tertullian thought Matthew was a surgically altered eunuch.
But there are other passages in which Tertullian seems to cross over to
a more nuanced usage, for instance, in his Adversus Marcionem 1.1. After
a blistering attack on Marcion and his native Pontus, the fiery Mrican
states that a bestial home gives rise to a bestial man: 'Jam et bestiis illius
barbariae importunior Marcion. Quis enim tam castrator carnis castor, quam
qui nuptias abstulit?' ('Now Marcion is even more troublesome than the
beasts of his barbarous homeland. Who is such a self-castrating beaver as
one who rejected marriage?') Here the myth of beavers as self-castrators is
applied to Marcion, who clearly denigrated marriage. 13 If it was good that
Matthew was a castratus, why is it bad now that Marcion wants to make
other people castrati? While we were to understand above that Matthew
was not a literal, surgically altered castratus, this beaver must be seen as
a literal castrator for the denigrating image to work on Marcion. Are
we to equate Matthew with the 'castrated' followers of Marcion? Upon
consideration it is no surprise to find that Tertullian is not using his terms
very consistently.
Here is another example. While, as we have seen above, Christ is the
great spado, elsewhere God himself is the anti-spado: 'Filius spadonis esse
non possum, maxime qui patrem habeam eundem, quem et omnia' ('I cannot
be the son of a eunuch, especially since I have the same Father as all
Creation does.') Tertullian makes no pretence of being consistent or even
careful with these terms. And he is not alone among early Christian
authors.
At this point it should be clear that the term cuvouxoc; is not so simply
understood as the tradition implies. The variety of meanings placed on the
language of eunuchs and castration undermines any effort to substitute
a simple term like 'celibate' every time someone is called a eunuch. A more
nuanced study is required.
126
The problem of early Christian eunuchs

Though the sketchy sources of second-century Christianity will not
support any broad generalizations on Christian eunuchs, it is possible
to start from one specific and very significant example. Origen demands
our attention due to his enormous influence on the development of
Christianity. His relatively well documented life can provide us with focus
on this topic, though it also presents us with a fundamental interpretive
problem: his writings strongly forbid castration while he apparently
emasculated himself. This quandary has led to a variety of explanations and
even inspired some to doubt that his castration is historical. 14 I would like
to start by arguing that Origen's self-emasculation is historical, and fits well
into our understanding of Christian ethics in second-century Alexandria.
Our major source for Origen's castration is Eusebius' Ecclesiastical History.
Eusebius' account of the life of Origen is well known as more panegyric
than biography, and so it seems strange that the historian would spend
several pages explaining away, as he calls it, the 'immature and youthful
action' (npayj.Hl n 1tE1tpaK'tat <j>pEVO~ j.lEV chEAOU~ Kat VEaVtKll~), if it
never happened. A brief list of parallels will show that self-castration by
a young and ardent Christian would not have been out of keeping with
the times in which Origen lived.
Evidence for castration among early Christians

~ustin lkfartyr
The most telling example is a testimony from Justin Martyr (Apology
1.29) who in the mid-second century was arguing to the Roman emperor,
Antoninus Pius, that Christians are sexually chaste and either marry to
have children, or, refraining from marriage, suppress their sexual urge
(£vEKpa't£UOj.l£8a). Justin wrote:
Indeed one of our Christians, in order to convince you ('pagans') that

licentious intercourse is not part of our religion, requested a permit from the
prefect Felix allowing him to have a doctor remove his gonads. 15
This ardent believer needed a permit since, beginning with the emperor
Domitian's broad legislation against castration in the late first century, and
continuing with a variety of additional restraints throughout the second
century, Roman law forbade emasculation within the boundaries of its
empire. Of course we are told that this young man was refused, but it
is clear Justin not only believes that Christian desire for self-castration
is noble, but also that this philosophical emperor would readily believe
the account, understand the context, and perhaps even applaud such
heroism.
127
"Walter Stevenson
Gnostic 'Encratites'
It is normally assumed that the group of early heretics called Encratites
used strong figurative language to dissuade believers from marrying.
The stance of this conjectured group is reconstructed by scholars from
the attacks of orthodox Fathers (as well as the less-than-fully-orthodox
Tertullian's passages cited above). But only Clement extensively quotes
these teachers, and their language does not clearly imply that believers
should not physically emasculate themselves. What influence these
Alexandrian teachers would have had on the adolescent Origen is a matter
of conjecture, but the possibility that they would favor literal self-castration
should at least be considered. For instance, Julius Cassian published a work
called IlEpt Euvouxim;, or On Eunuchism. In it he taught that marriage
should be avoided apparently through self-castration:
Let no one say that since we have appendages such that some of us are
designed female and some male, some for receiving and some for implanting,
God designed us to be this way. For if this were God's plan, towards which
we are striving, he would not have blessed eunuchs: the prophet (Isaiah)
would not have said 'they are not fruitless trees' taking an analogy from
the tree for the man who by deliberate choice emasculates himself from
ideas of this sort. 16
Julius Cassian here has created an interesting anti-natural law positing

that the nature of our created bodies should not outweigh the words of
Isaiah who praises eunuchs. 17 His emphasis on the physical appendages
and their uselessness is particularly strange and interesting. Though this
passage does not provide conclusive proof that Julius encouraged physical
self-castration, it leans as much in this direction as towards encouraging
figurative celibacy.
Clement, who has preserved the quotation from Juli~s Cassian, com-
ments that this statement puts Julius clearly in the tradition of the school
of Valentin us, perhaps the most popular of second-century Alexandrian
Christian teachers. In Alexandria itself we know that Valentinus had
a large following, and Clement sums up this school's opinion in this
way:ts
How could someone not rationally give credit to the Saviour, if he trans-
formed us, and freed us from error, and from the commingling of our parts
and appendages and genitals? 19
Once again there is a fixation on the appendages ('ta IJ-Opta) and a very
unfigurative description of human sexuality. If this is just encouragement
to live a virginal life, it is worded in a very strange way.
A similar line of thought is added by Basilides, another of the very
128
influential second-century Alexandrian teachers. Basilides states:

Those who have castrated themselves (oi Euvouxicravn::~ E.amou~) on
account of the kingdom of God choose this most rational way to avoid
the difficulties of marriage, fearing the trouble of procuring necessities for
their family. 20
Though, in fact, the followers of Basilides were known for their liberal
sexual lifestyle, 21 their teachings on castration seem to have stuck closely to
those ofValentinus and Julius Cassian. And although all of these teachers
are traditionally understood to be figuratively promoting celibacy, the
quotations themselves use the language of castration. It is unlikely that
these influential second-century teachers made an explicit priority of
promoting castration, but their influence should not be underestimated.
In an intellectual climate where the body is described with 'useless'
appendages, and where its functions are negated, it is easy to imagine
some turning to surgical solutions.
Castrated monks
In addition to Gnostic testimony, there is evidence of whole monastic
orders requiring castration for members. 22 Epiphanius, in his catalogue
of heresies composed in the fourth century, describes one of these, the
Valesians:
I often hear of Valens. He was an Arab, and though I don't know anything
definite, I have considerable suspicion that his sect still lives on to this day
in Bakathi in the region of Philadelphia on the other side of the Jordan
river. .. They were members of the Church until that time when their
insanity came to its fullness and they were excommunicated. For they are
all castrated ... 23
It is remarkable that Epiphanius, who shows his dear rejection of Christian

castration, concedes that this monastic order was accepted as part of the
Church for some time, though the members were all castrated. One might
question Epiphanius as a source, but it would seem strange for a Christian
apologist deliberately to create stories of legitimate castrated orders in the
fourth-century Church. Rather it is more probable that such orders did
exist not only in the early-fourth century, but as Epiphanius concedes,
into his own time. Indeed it is possible that castrated orders existed right
throughout the history of eastern Christianity. 24
Christian tradition
Another body of evidence arguing for the historicity of Origen's castration
is Christian tradition. The Church began to legislate against eunuchs
129
"Walter Stevenson
from the beginnings of canon law, and the implied need for this frequent
legislation itself argues that castration was common. 25 Furthermore
theologians of the fourth and fifth centuries were quick to label castrated
Christians as heretics, with Origen the most celebrated example. But
in spite of the later theological movement to stamp out the practice of
castration, the record tells us of several monastic orders whose initiates
were castrated, and a variety of individual eunuchs stretching all the way
back to the first Ethiopian Christian, the courtier who appears in the Acts
of the Apostles. 26 In short, church history shows us that Christian eunuchs
were well known in the second and third century.
Origen's sources
Finally Origen himself in his own commentary on the Gospel of Matthew
reveals several other possible sources that may have influenced him towards
castration. The widely influential second-century text, the Sentences of
Sextus, stated: 'Cast aside every part of the body that does not lead you
to self control; for it is better to live continently without some body
part than destructively with it', as well as: 'You may see that people have
saved the health of the rest of their body by chopping off and casting
away parts. How much better to do this for continence?' 27 He even
found a citation from Philo to support castration: 'it's better to castrate
yourself than struggle with illicit sexual intercourse' .28 And so when we
consider that castration was discussed and esteemed among Alexandrian
Christians, that Eusebius concedes Origen's castration as the only blot
on his otherwise stainless life, and that Church history; suggests the
commonness of castrated Christians, it seems that the evidence points to
the historicity of Origen's self castration.
Alexandrian rejection of castration

Philo
And so it is surprising how consistently Alexandrian writers, who later
gained orthodox status, rejected the practice of castration. Alexandria,
a great centre of scriptural study in the first two centuries after the birth
of Christ, fostered one of the most influential foundations of Christian
exegesis. And it is clear that the most influential Alexandrian exegetes from
Philo to Origen all agreed that castration was unholy. Though we have
just seen Origen quoting Philo in support of castration, Philo repeatedly
argues elsewhere that eunuchs are symbols of atheism, are 'cut off' from
immortality, and in this strong statement from his Special Laws, are rightly
barred from the synagogue. He is discussing specifically Deuteronomy 23
130
which forbids eunuchs to enter the assembly:

Thus, knowing that in assemblies there are not a few worthless persons
who steal their way in and remain unobserved in the large numbers which
surround them, it guards against this danger by precluding all the unworthy
from entering the holy congregation. It begins with the men who belie
their sex and are affected with effemination, who debase the currency of
nature and violate it by assuming the passions and the outward form of
licel).tious women. For it expels those whose generative organs are fractured
or mutilated, who husband the flower of their youthful bloom, lest it should
quickly wither, and restamp the masculine cast into a feminine form. 29
Though this description raises many difficulties in understanding ancient

eunuchs, the message is unambiguous: eunuchs are a desecration and
should not even enterthe synagogue. 30
Clement
The early Alexandrian Christians, who show themselves to be Philo's
disciples in method and content, not only agree with him but quote
him. Clement of Alexandria, teaching a generation before Origen, in an
evangelical work addressed to pagans, states:
Don't you see Moses, the veritable prophet of truth, ordering that those
with mutilated genitals (8A.a8ia<; and anoKEKOf.Lf.LEVo<;) cannot enter the
assembly? And also the son of a whore? He is showing allegorically through
the above the type of the atheist, the one cut lacking procreative power,
the other divine strengthY
Here and elsewhere Clement, like Philo, disparages eunuchs.
Origen
Origen follows in this tradition as well. It is interesting that Origen does
not treat the topic of eunuchs nearly as frequently as Philo and Clement,
often avoiding opportunities in his commentaries; for example, in his
extant homilies on Leviticus he wholly ignores the strictures against
castration. But when he comes to the key passage of the New Testament on
eunuchs, the passage that inspired him to emasculate himself as a youth,
he, like Clement, turns to Philo. In beginning an historical survey of
eunuchs in the Jewish tradition, he starts with the castrated servants of
Pharaoh in Genesis 40.1, and calls them 'those unable to give birth to any
good' (ayovot nav-ro~ KaA.oii). 32 Origen, following the lead of Clement
and Philo before him, disparages castration.
It seems, then, that Origen had castrated himself, like other ardent
Christians in Alexandria, and did so under the influence of current beliefs.
131
WTalter Stevenson
But it is also dear that the most influential scriptural exegetes of Alexandria
strongly disapproved of castration, following their reading of Scripture.
Therefore it will be useful to look more closely first at the Judea-Christian
tradition, and then its Alexandrian hermeneutics.
Judea-Christian tradition on castration

The earliest Hebrew legislation against eunuchs comes from Leviticus
(21.18-20, 22.24) where castration causes ritual impurity most importantly
in the priests, but also in their sacrificial animals:
None of your descendants, for all time, may come forward to offer the food
of his God if he has any infirmity, for none may come forward if he has an
infirmity, be he blind or lame, disfigured or deformed, or with an injured
foot or arm, a hunchback, someone with rickets or ophthalmia or the scab
or running sores, or a eunuch ...
You will not offer Yahweh an animal if its testicles have been bruised,
crushed, torn or cut off. 33
The strength of this taboo that extends all the way to sacrificial animals
is striking, and indicates that the early Hebrew people rejected any form
of emasculation. Deuteronomy 23, as we have seen, repeats the strong
prohibition against castrated men participating in religious ritual:
A man whose testicles have been crushed or whose male member has been
cut off must not be admitted to the assembly ofYahweh.
The Torah is unambiguous in legislating against any form of castration.

It is therefore remarkable that we find the following text in the so-called
trito-Isaiah (56.3), probably written down at about the same time as
Leviticus and Deuteronomy:
No foreigner adhering to Yahweh should say, 'Yahweh will utterly exclude
me from his people.' No eunuch should say, 'look, I am a dried-up tree'. For
Yahweh says this: to the eunuchs who observe my Sabbaths and choose to do
my good pleasure and cling to my covenant, I shall give them in my house
and within my walls a monument and a name better than sons and daughters;
I shall give them an everlasting name that will never be effaced.
This text appears deliberately to undercut the traditional ancient laws

recorded in Leviticus and Deuteronomy. We have already seen that it
lends itself well to allegory, but taken literally or allegorically the eunuch
is tolerated and perhaps even privileged.
The book of Wisdom, composed by a pious Jew pmbably in Greek
around 100 BC, goes one step further than Isaiah (Wis. 3.13):
Blessed the eunuch whose hand commits no crime, and who harbours no
132
resentment against the Lord: a special favour will be granted to him for his
loyalty, a most desirable portion in the temple of the Lord.
It is tempting to include this work as the beginning of the Alexandrian

tradition. The general thought and style of the work is Hellenized like
that of the Alexandrian exegetes who follow. And in it eunuchs have won
a special place in the Jewish community. 34
So it is no surprise that we have the following quotation of Jesus in
the Gospel of Matthew (19.12) composed about a century and a half
after the Book ofWisdom:
There are eunuchs born so from their mother's womb, there are eunuchs
made so by human agency and there are eunuchs who have made themselves
so for the sake of the kingdom of heaven. Let anyone accept this who can.
Here the eunuch is the symbol of the most gifted members of the
congregation. In summary, from the time of Levitican legislation the
tradition has apparently made a complete reversal on eunuchs. In the
Torah they are forbidden and the procedure of castration is even unclean
for animals. The post-exilic prophetic tradition undercuts this stance,
and by the time of Hellenized Judaism eunuchs have become figures of
prestige and examples of virtue.
A modern critic would tend to understand an historical development
when reading these passages. The rigid taboo of Leviticus and Deuter-
onomy gave way to acceptance in Isaiah and Wisdom which in turn rose to
an exalted status in the Gospel of Matthew. But our early Alexandrian crit-
ics did not see a development. Rather they felt challenged by a discrepancy
between Jewish law and the prophetic tradition. It was unacceptable, as
they understood it, that Isaiah and Jesus would have flatly denied what
they considered Moses' legislation against eunuchs. Starting from faith
these critics had to find a critical strategy that would harmonize Scripture's
conflicts. This strategy was allegory.
Alexandrian hermeneutics
How allegory became central for interpreting Jewish Scripture is a well
discussed question outside the scope of this chapter. Suffice it to say that
by the time of Philo, allegorical interpretation had become central to
Judea-Christian exegetes in Alexandria. An appropriate example can be
drawn from Origen's influential commentary on the Song ofSongs. 35 The
text itself simply relates the story of a dark-skinned lover, disparaged by
other women, who wins the love of the king. But Origen asserts that
on a spiritual level the lover represents the new Christian Church made
up of all the races of the ancient world, the other women represent the
133
"Walter Stevenson
Jews, and the beloved represents Christ, who loves all these women, but
who chooses to marry the dark-complexioned church. 36 Origen insists
that both the literal and spiritual meanings are valid, but implies that the
spiritual or allegorical is privileged.
Recent works on Alexandrian hermeneutics have varied greatly. For
instance, David Dawson has seen strong Hellenic influence and has argued
that figures like Clement were close to Stoic 'professors' in their strategies
and goalsY Henri Crouzel, however, the most authoritative biographer of
Origen, argues that the sort of allegorical reading Origen practices appears
in the letters of Paul as well as Revelation. 38 While avoiding the intriguing
issue of Greek or Hebrew influence on Alexandrian hermeneutics, I would
call attention to Origen's rhetorical techniques and strategies. After all,
ancient rhetoricians taught that allegory was merely one of many rhetorical
techniques. 39
Origen as rhetor
Origen can easily be seen as similar to other professional teachers of his
time. We know that he was connected to famous ancient figures like
his own 'pagan' teacher Ammonius under whom Origen studied with
Plotinus; that is, we know that Origen was very much a part of the world
of the professional sophist/philosopher. 40 In this form of analysis Origen,
like professional teachers in antiquity or beyond, would have sought to
carve out 'market share' with well placed attacks on competitors who
would be cast as heretics and pagans in these polemical battles. This is
an attractive view for us considering how strong the Greek traditions of
rhetoric were in Alexandrian culture, and especially in the works of Philo,
Clement and Origen. 41
Two aspects of Origen's interpretation of Matthew' 19.12 suggest
a fundamentally rhetorical purpose: that Origen seems to presume that
(Matthew's) Jesus would follow the rules of rhetoric; and that Origen
shows some strong rhetorical tendencies himself
The foundation of Origen's discussion of this passage appears to be that
Jesus was speaking figuratively at Matthew 19 .12. He describes two groups
as his opposition. First he discusses those who take the whole passage
as literal - those who argue that there are eunuchs who are born that
way from their mothers, eunuchs who were castrated by other humans,
and those who physically castrate themselves for the kingdom of heaven.
And second, he attacks the majority who interpret the first two categories
literally and the third as figurative. Origen proceeds to argue that the first
group is na'ive and susceptible to embarrassing attack even from the most
tolerant pagans. He applies almost his whole discussion to the second
134
group, asserting that they are misunderstanding Jesus' use of figures:

'I want to defend the sequence (aKoA.ou9ia) of the three types of eunuchs
and approve the figurative nature ('tp07tOA.oyia) of the third ... ' 42 The
technical rhetorical terms OKOAOU9ta and 'tp01t0AOyta seem tO imply that
Jesus employed a rhetorical strategy, and that interpreting the passage
literally is an affront to Jesus' skill.
Origen unguardedly reveals his own rhetorical strategies. Two examples
from Origen's argument may help to illustrate this point. He calls upon
Jesus' warning against greeting passers-by on the road as found in Luke
10.4: 'greet no one on the way'. What would happen if all Christians read
this passage literally and stopped greeting anyone on the road? Origen
responds that Christians would be seen as fools (!lrop6<;). When directly
discussing those who castrate themselves, he responds even more strongly
that these Christian eunuchs would bring shame (aicrxuvT]) not only on
other Christians, but on humanity in general, even for those pagans who
are accustomed to tolerate almost any aberrant behaviour. 43 Origen is not
using his allegory here to privilege a divine truth, but rather to mould
a Christianity that is not repulsive to non-Christians. This process of
crafting a message, not based on 'truth', but rather on the expectations
of the pagan audience, seems to support the thesis that Origen spoke as
a traditional Hellenistic rhetor.
Origen himself uses language figuratively in his discussion of Matthew
19.12. He seems deliberately to string together flamboyant puns. They
vary from the standard rhetorical paradoxes like 'they cut themselves out
of an unchecked lust (epro<;) for chastity' to bold ones like, 'eunuchs who
father pains (yevvav 7t6vou<;) for themselves as well as fathering dissection
of their bodies', to several games on the Greek verb 'to cut', 'tE!lVEtv, 'the
A.Oyo<; that leads to castration is the most cutting ('t!lT]'ttKffi'ta't6<;) while the
A.Oyo<; that empowers the soul to win cuts out desire (eK'tE!lVEtv)', and, 'if
someone takes up the livirtg and powerful and cutting ('tO!lol'tepo<;) A.Oyo<;
"beyond a dagger double cutting (oicr'tO!lo<;)" he could cut out (EK'tE!lVOt)
desire from the soul without touching the body' .44 He. ends his diatribe
by stating that it is not intended to be a schoolboy exercise in rhetoric
that dispatches all the points of his adversaries. But we can easily become
confused by his playful language, and see a fundamentally rhetorical
purpose in his work.
Origen's arguments are not at all aimed against the practice of castration.
In fact, he dismissed the literalists with a few lines and then proceeds to
argue against his second group of interpreters, those who read the first two
ofJesus' eunuchs literally, but the last category figuratively. By reading the
last group figuratively these exegetes are in fundamental agreement with
135
"Walter Stevenson
Origen that Jesus did not intend his followers to castrate themselves. Their
mistake, and the mistake that Origen spends most of his discussion of
the passage on, is this: that they are undermining Jesus' rhetorical intent,
and starting down the slippery slope to literalism. It is difficult to see
Origen's charges as anything but 'academic'. It seems to be his strategy to
expose the incompetence of his competition (in this case they are probably
Gnostic heretics who must be disparaged). This competitive strategy adds
another rhetorical stroke to our understanding of Origen's writing.
And so the overall picture of Origen as rhetor/sophist would look
something like this: Origen projects his own rhetorical world view by
describing Jesus' intentions in Hellenistic rhetorical terms; his arguments
are meant to appeal to the predilections of a pagan audience rather than
to some privileged Christian truth; he himself uses many rhetorical tropes
that seemingly undercut the seriousness of his point; he uses the oldest of
tropes, the recusatio that he is not stooping to petty rhetorical polemics;
and his motivation seems less theological or moral and more polemical.
Starting from a close reading of his exposition, we seem justified in
concluding that Origen writes more as a rhetorician than a theologian.
Origen as evangelist
But there is a problem with this argument. Modern scholars of Origen
have dated his commentary on Matthew to near the end of his life. 45 By
this time Origen was undoubtedly one of the most famous and prestigious
intellectuals of his period. He even achieved the honour of invitation
from the emperor's mother, Julia Mamaea, to her salon. 4~ By old age,
it is difficult to imagine him having any serious rhetorical competition
for students or prestige. Even the slights to his reputation apparently
originating from his old bishop in Alexandria, Demetri'us, would have
been lost, rhetorically at least, in the mists of time. In short, though
Origen shows techniques and strategies similar to his pagan peers, his
fundamental motivation seems different.
To find this motivation it is only fair to look at his own statements
of exegetical purpose. At the beginning of his discussion of Matthew
19.12 Origen states this:
Before I argue what seems to me the true interpretation of this passage, I will
describe the two leading interpretations. And I will refute these as much as
possible so that, while guarding against any errors, we will be able to live
better if we follow the correct teachingY
It is very likely that Origen is refuting not only the still attractive teachings
of the Valentinian school, but also refuting his own surgical example. One
136
could easily suspect that, in a character such as Origen's, his self-criticism

could inspire more heated rhetoric than even the hated Valentinians.
Nevertheless his stated goal is clearly to improve the moral quality of
his audience's life.
Perhaps more instructive is this general statement of exegetical purpose
taken from his seventh homily on Leviticus:
I am not trying to explain Scripture but rather to build up the Church,
though I think that a sagacious listener can find in these discussions of mine
clear paths of interpretation. 48
Here we see that he self-consciously considered his role to be rhetorical.

But his rhetoric was working to further the institution for which he
had often risked his life. His evangelical method, like the Apostle Paul
before him, drew heavily on Graeco-Roman culture specifically to attract
Graeco-Romans to Christianity. It would seem to be overly sceptical
to dismiss Origen's statements of purpose without at least considering
them. In fact, these statements do not seem to contradict the portrait
of Origen as rhetor.
So in summary, two sides of Origen are clear from this analysis: first,
that he grew up and was educated in a highly rhetoricized culture; and
second, that he was self-consciously using his rhetorical skills to build
up the Christian Church.
Origen's teaching pioneers Christian sublimation?

The confused language surrounding asexual men in our early Christian
authors shows the impossibility of making a simple judgment. The
disparities in early Christian preaching and practice showed the Church
struggling with a fundamental paradox. A survey of Christian Scripture
showed significant conflict, at least as interpreted in the first few centuries
of our era. And, of course, Origen personally embodied this conflict
by castrating himself and then preaching against self-castration. But all
this conflict can lead to a tentative conclusion on the role of Origen in
our understanding of eunuchs in early Christianity. Origen seems to have
regretted his own self-castration, and desired to steer the early Church away
from his own example. But he chose not to participate in the polarized
traditions of Judea-Christianity on castration. He neither attempted to
legislate against the practice as did Hebrew and later Christian law, nor
did he tolerate eunuchs as did Isaiah and the Book of Wisdom. Instead
he followed a strategy that became known as essential to Christianity. He
wanted to direct people away from physical to verbal expression of their
desires. Just as he redirected the lush and seductive physical £proc; (desire)
137
"Walter Stevenson
of the Song of Solomon into the realm of language and contemplation

with his allegorizing exegesis, so here he seems to transfer the prevalent
desire for castration into the realm of language. The puns on cutting,
so common in his interpretation, could lead a young ascetic bent on
self-mutilation away from the physical act and into the realm of language.
Presumably such young men could dull their ascetic frenzy by expressing
it in vivid language and poetic meditation.
If this tentative conclusion can be taken seriously, it points to an aspect
of Christianity that has become increasingly evident in the subsequent
ruminations of monastics (Apothegmata, et al), Lives of the Saints and on
into the practice of the modern confessional. Christianity; more than any
other religion, displaced what it saw as dangerous physical desires from
the body into discourse, into long and often verbally explicit confessions
of illicit desire. It may seem ironic that Origen felt the need to sublimate
his fellow Christians' desire for castration. But it is possible that Origen's
unique combination ofJudeo-Christian traditions with Hellenistic rhetoric
fostered one of Christianity's most salient features. In this way, Origen's
teaching on castration would have to be seen as seminal.
Notes
1 The term 'eunuch' is a broad one certainly extending far beyond 'castrated
man'. I will discuss the variety of meanings in early Fathers below, but will assume
for this p_aper that the vast majority of eunuchs in antiquity were castrated. An
exhaustive treatment of the terms can be found in Maas 1925. The classic article
on ancient eunuchs is Hopkins 1963 (also Hopkins 1978, 172-'-96). See also
Tougher 1997, and see Stevenson 1995 for a broad treatment of eunuchs in the
second and third centuries.
2 Williamson 1964, 231.
3 Hall1979, xi.
4 Perler 1966 lists Matt. 19.12; Justin, I Apol 15.4; Athenagoras, Supplicatio
33.1; Euseb. Hist. eccl 6.8.2; Methodius, Symposium 1.1.10; Ps.-Clement, De

Virginitate 2.1; and Tert. Ad Uxorem 1.6.
5 Blank 1963, 18: 'Ob sich die Bezeichnung "Meliton der Eunuch", die
Polykrates von Ephesus (Eusebius KG 4.24.5) gebraucht, an Mt. 19.12 anschliesst,

also nur "iibertragen'' zu gelten hat und nur die Ehelosigkeit Melitons bezeichnen
soll, oder ob Meliton vor seiner Bekehrung zum Christentum Bezeihungen zu den
Mysterienreligionen (etwa den Gallen der phrygischen "Grossen Mutter") hatte,
was an sich nicht unwahrscheinlich ist, oder ob er, ahnlich wie nach ihm Origenes,
Mt. 19.12 wortlich nahm, lasst sich nicht mehr mit Sicherheit entscheiden. Doch
diirfte die Annahme, dass Meliton Eunuche im physischen Sinne war, naher
liegen, da sich der Zuname auf diese Weise am besten erklart'.
6 See Brown 1988, esp. chap. 8 in which Origen's castration is discussed.
138
See also Scholz 1997, 143-72, for a discussion of sexual renunciation in early
Christianity.
7 See the Revised English Bible, ad foe. 'For while some are incapable of
marriage because they were born so, or were made so by men, there are others
who have renounced marriage for the sake of the kingdom of heaven. Let those
accept who can.' This sort of narrowly interpretive translation seems at the least
to obscure the original text.
8 Methodius, Symposium 1.1.
9 See Clem. AI. Paed.3.4.24: 'There are many eunuchs who are pimps, and
since it's thought that they cannot be lovers, they can help out those seeking
a love affair without any suspicion'.
10 See Justin, I Apol. 1.29.
11 Athenagoras, Leg. pro. Christ. 33.3; the translation is mine.
12 Methodius, Symposium 1.2.
13 See Juv. 12.34, where Juvenal implies that beavers chew off their own
gonads.
14 This tradition began in antiquity with Epiph. Panarion 64.3.
15 Justin, I Apol. 1.29.
16 Clem. AI. Strom. 3.13, the translation is mine, with the last phrase clearly
stated as E:amov euvouxil~ov-ca.

0 0 0
17 Isaiah 56. See discussion below.

18 See Tert. Adv. Valent. 1.1.
19 Clem. AI. Strom. 3.13. See also 3.7 in which Valentinus is quoted as teaching
that Christ was so continent (E:yKpaTI]<;) that, though he ate food, it never was
digested. It seems that the physical nature of humanity was not sacred to him.
°
2 Clem. AI. Strom. 3.1.
21 Clem. AI. Strom. 3.7.
22 See also Epiph. Adv. haeres. 3.2.1095 for another group in the east.
23 Epiph. Adv. haeres. 2.1.
24 The tradition had well known followers in modern Tsarist Russia, the Skoptsi,
some of whose monasteries were known to be very popular. See Engelstein 1999,
and Rapaport 1943.
25 See Canon 1, for instance, of the Council of Nicaea: 'If any one in sickness
has been subjected by physicians to a surgical operation, or if he has been castrated

by barbarians, let him remain among the clergy; but, if any one in sound health
has castrated himself, it behoves that such an one, if [already] enrolled among
the clergy, should cease [from his ministry], and that from henceforth no such
person should be promoted. But, as it is evident that this is said of those who
wilfully do the thing and presume to castrate themselves, so if any have been
made eunuchs by barbarians, or by their masters, and should otherwise be found
worthy, such men the Canon admits to the clergy'.
26 8.26-40. See lrenaeus, Adv. haer. 3.20, where this eunuch is called the
founder of Ethiopian Christianity, and also Scholz 1997, 143-5.

27 Commentarium in Evangelium Matthaei 15.3, Sententiae Sexti 13.
28 Det. Pot. Ins. 176. See in this volume Abusch's fascinating discussion of this
passage and Philo's complex thought on eunuchs.
139
Walter Stevenson
29 Spec. Leg. 1.325.

30 Abusch argues that Philo took an ambiguous attitude towards eunuchs,
and rightly shows several passages that seem to approve of eunuchs. Though
his argument is convincing, it seems to work at a level of abstraction that does
not undermine the fundamental social interdiction demonstrated throughout
Philo's work.
31 Clem. AI. Protr. 2.16. See also Philo, Mig. Abr. 69, which appears to be
Clement's source here.

32 Commentarium in Evangelium Matthaei 15.5. Philo, in his treatise De
Ebrietate 220, discusses at length the Pharaoh's eunuchs calling them (cro<j>ia~
iiyovot) 'those unable to give birth to wisdom'. See Abusch's penetrating discussion
of this passage in this volume.
33 These translations are taken from the New Jerusalem Bible (New York,
1985).
34 See Abusch who points out some very promising paths of interpretation for
a development in Jewish views of eunuchism.

35 Origenis in Canticum Canticorum 2.45b-46.
36 See De Lange 1976 for a discussion of Origen's anti-Semitism.

37 Dawson 1992.
38 Crouzel1989.
39 Though allegory was held in disregard generally, see Quint. lnst. 5.11.21,
6.3.69, and 8.11.14. The influence of Greek teaching is obvious here since
Quintilian is merely transliterating the Greek word.
40 See Crouzel1956.
41 Buell1997 develops the term 'self-authorization' for Clement's position. This
notion can be easily transferred to Clement's follower Origen.

42 Commentarium in Evangelium Matthaei 15.4.
43 Commentarium in Evangelium Matthaei 15.1-2.

44 See Abusch who briefly discusses Philo's punning.
45 See Crouzel 1989, 43: 'A work of Origen's old age, the, Commentary on
Matthew is on the whole less mystical and more pastoral than the Commentary
on john.'
46 If we can trust Euseb. Hist. eccL 4.21.3-4.
47 Commentarium in Evangelium Matthaei 15.1.
48 Homiliae in Leviticum 7.1.
Bibliography
Amir, Y.
1983 Die hellenistische Gestalt des ]udentums bei Phi/on von Alexandrien,
Neukirchen-Vluyn.
Bardy, G.
1931 Origene, Paris.
Blank, J.
1963 Meliton von Sardis: Vom Passa, Freiburg im Breisgau.
140
Brown, P.
1988 Body and Society, New York.
Buell, O.K.
1997 'Producing descent/dissent: Clement of Alexandria's use of filial meta-
phors as intra-Christian polemic', Harvard Theological Review 90,
89-104.
Cadiou, R.
1935 Lajeunesse, Paris.
Caspary, G.E.
1979 Politics and Exegesis: Origen and the two swords, Berkeley.
Clark, E.
1992 The Origenist Controversy: The origin of an Early Christian debate,
Princeton.
Crouzel, H.
1956 'Origene et Plotin eleves d'Ammonios Saccas', Bulletin de litterature
ecclesiastique 57, 193-214.
1989 Origen, San Francisco.
Dawson, D.
1992 Allegorical Readers and Cultural Revision in Ancient Alexandria, Berkeley.
De Lange, N.R.M.
1976 Origen and the jews, Cambridge.
Engelstein, L.
London.
Hall, S.G.
1979 Melito ofSardis on Pascha and Fragments, Oxford.
Hopkins, K.
Maas, E.
1925 'Eunuchos und verwandtes', Rheinisches Museum fur Philologie 74,
332-76.
Nautin, P.
1977 Orig{me: sa vie et son oeuvre, Paris.
Nikolaus, W
1997 Praeparatio Evangelica: Studien zur Umwelt, Exegese und Hermeneutik
des Neuen Testaments, Tiibingen.
Perler, 0.
1966 Meliton de Sardes: sur la P!tque et fragments, Paris.
Rapaport, I.
1948 Introduction a la psychopathologie collective: la secte mystique des Skoptsy,
Paris.
Runia, D.T.
1993 Philo in Early Christian Literature, Minneapolis.
141
-willter Stevenson
Scholz, P.O.
1997 Der Entmannte Eros, Dusseldorf.
2001 Eunuchs and Castrati: A cultural history, Princeton.
Stevenson, W
1995 'The rise of eunuchs in Greco-Roman antiquity', journal of the History
ofSexuality 5, 49 5-511.
Strecker, G.
1967 'Math. 19.12 und die alten Christen', in Walter Bauer: Aufsiitze und
Kleine Schriften, Tubingen.
Torjesen, K.J.
1986 Hermeneutical Procedure and Theological Method in Origen's Exegesis,
Berlin.
Tougher, S.
and origin', in L. James (ed.) Women, Men, and Eunuchs: Gender in
Byzantium, New York, 168-84.
Van den Hoek, A.
1995 Christian Controversy in Alexandria: Clement's Polemic against the
Basilideans and Valentinians, New York.
1998 'The "catechetical" school of early Christian Alexandria and its Philonic
heritage', Harvard Theological Review 90, 59-87.
Williamson, G.A. (trans.)
1964 Eusebius: The history ofthe Church from Christ to Constantine, London.
142
8
IN OR OUT? ORIGINS OF COURT EUNUCHS
Shaun Tougher
The study of eunuchs in history has tended to focus on their place and
function at royal and imperial courts, despite the existence- as this volume
seeks to demonstrate - of other interesting and fertile topics of study, such
as gender, religion, and art. Thus it is with some reservation that I address
the more traditional political topic here. But the court role of eunuchs in
history is important, and one that our historiographical sources certainly
report, given their major theme of politics. My starting point is the court
of the later Roman empire (c. fourth to sixth centuries AD), the institution
responsible for the increased presence and visibility of eunuchs in Roman
society and in our sources. The role of the eunuch in the late Roman
court system, in which eunuchs could rise to the heady political and
social heights offered by the post of grand chamberlain (praepositus
sacri cubiculi), has been analysed in a classic article by Keith Hopkins,
seeking an answer as to why eunuchs became so valued an instrument
of the late Roman emperors. 1 Part of the answer for Hopkins was the
outsider element of eunuchs; the fact that they were imported from non-
Roman territory made them outsiders in Roman society, forcing them into
a dependent mutually-vital relationship with their master the emperor. 2
This judgement has recently been emphasized again by Schlinkert. 3 In this
chapter I wish to explore whether other court systems where eunuchs had
a role shared this factor of the eunuch as ethnic outsider. If this is indeed
a constant feature, this would enable Hopkins' analysis to inform more
generally the study of eunuchs in history. I wish to consider in particular
whether the foreign eunuch is a feature of the court of the continuation
of the Roman empire in the east, which we call the Byzantine empire (c.
seventh to fifteenth centuries AD). I will suggest that the factor of the
court eunuch as an ethnic outsider is not an essential constant, and
certainly not in the case of Byzantium, where homegrown eunuchs seem
to have become the norm. This fact may have reduced the efficacy of their
role in the Byzantine political system, resulting in the disappearance of
the phenomenon of the powerful court eunuch. Ultimately, the general
143
Shaun Tougher
question 'Why eunuchs?' still remains open, but the answer is surely
dependent on the physical nature of eunuchs, rather than on where they
came from.
The later Roman empire

The established view, then, is that for the late Roman court eunuchs were
to be literal outsiders, drawn from territory beyond the empire. This view
is apparently well supported by late antique Roman imperial legislation.
Constantine I (306-37) ruled that eunuchs were not to be created within
the Roman empire, whilst Leo I (457-74) explicitly identifies the tolerated
source of eunuchs, the importation of castrated barbarians. 4 General
ancient comments can also provide support. Aelius Donatus, agrammaticus
active in Rome c. AD 353, remarks in his commentary on Terence's
play The Eunuch that in the East most eunuchs came from Armenia. 5
In the sixth century Procopius alleges that in the time of the emperor
Justinian I (527-65) the majority of the eunuchs to be found at the
imperial court in Constantinople had their origins in Abasgia, on the
eastern shore of the Black Sea, a region subject to the Lazi; the two kings of
the Abasgians operated a significant trade in castrated subjects. 6
If we look at specific late antique court eunuchs whose origins are
known, the view finds broad confirmation, as Scholten in her study on
grand chamberlains in the fourth and fifth centuries has shown. Scholten
does acknowledge, however, that caution is needed.? We are entirely
dependent on the information which our sources provide, and they are
certainly not full. We are not always told where individual eunuchs
come from; indeed individuals may not even be explicitly 'identified as
eunuchs. Unfortunately names, where we have them, are not that helpful
for deducing ethnic origin; for, as Scholten points out, most were picked
for the eunuchs after they became slaves. Scholten can provide only eight
examples of grand chamberlains for whom we have information on origin
or social status. 8 These include: Eutherius, an Armenian; 9 Eutropius, an
Armenian; 10 Mamas, an Armenian; 11 and Antiochus, a Persian. 12 Looking
beyond Scholten's chronological boundary into the sixth century we find
another good example, Narses the Armenian. 13 From the information we
have, Scholten estimates that for the fourth and fifth centuries the great
majority of chamberlains were drawn from Armenia or Persia. 14
The comparative approach

My enquiry is, then, to what extent is this outsider__~k_f!l:~l}_u:.:h::tr~<,:!erisric
of oth_<:t_.C:.<?!:I:£t systems where eunuchs are employed, and thus to what
extent does H~:pidns' analysis apply to other courts. This brings us to the
144
In or out? Origins ofcourt eunuchs
field of comparative history, one which I believe offers great potential

fertility for the study of eunuchs in history, and was prominent in the
conference which led to this volume. An example of a previous attempt
in this field is Grayson's fascinating quest to throw light on the difficult
subject of eunuchs at the court of ancient Assyria. 15 Grayson's endeavour,
however, is not always successful; errors arise when he discusses those
cultures in which he is not an expert. For example, he asserts that all
eunuchs in the Chinese system were Chinese; he implies that all eunuchs
in the Ottoman system came from Mrica; he overstates when he says of
Byzantium that noble families 'often had one of their members castrated
to enter state service and attain high positions'; and he seem~ to believe
that Byzantine eunuchs could marry. 16 Grayson also misses useful points
of comparison; just as Assyrian terminology seems to distinguish two
groups of court officials, the bearded and eunuchs, 17 so definitely does
a ninth-century Byzantine court text. 18 Thus, whilst the comparative
approach has the potential to. reward, it also has its dangers; not being an
expert in other fields can lead one to make, or accept, factual errors, which
in turn can lead to faulty deductions. Lack of knowledge of other fields
can also prevent the formation of valid and illuminating comparisons.
Despite these risks this is surely a valid and promising approach, worth
attempting. I shall consider several courts where eunuchs also found a role,
and try to identifY their ethnic origin.
Ancient Assyria
Grayson, at his most compelling, advocates that there were indeed court
eunuchs in ancient Assyria. Most of these eunuchs, he argues, derived
from foreign captives and tribute, but looking to the examples of China
and Byzantium, he suggests that eunuchs who reached the highest offices
could be from the Assyrian nobility itsel£ 19
Achaemenid Persia (c. 550-330 Be)

There is no doubt that eunuchs had a role here. 20 The classic statement on
the Achaemenid court's use of eunuchs is found in Xenophon's Cyropaedia,
in the Greek's presentation of the Persian Cyrus' own reasoning: eunuchs,
haying no families of their own, would focus their devotion and care on
their master; eunuchs have a natural need for a master as they are rejected
by all others; they are not cowards and weaklings; and they evince the
utmost fidelity. 21 Grayson says that one source of the court eunuchs was
the provinces, citing Herodotos who tells us that as well as one thousand
talents of silver, 500 boy-eunuchs were drawn from Babylonia and the rest
of Assyria as the tribute of the satrapy. 22 Grayson states that it is, however,
145
Shaun Tougher
unclear if Persians themselves were made eunuchs, though he thinks it is

possible on analogy with Byzantium. 23 A fuller account of the evidence
available is presented by Guyot; not only were eunuchs drawn from
Babylonia, 24 but there were also Greeks, one a Paphlagonian. 25 Herodotos
records the castration of Greek boys after the Ionian revolt was crushed by
the Persians, 26 and he also presents us with a Greek eunuch with his own
story, the Pedasian Hermotimos, chief eunuch of Xerxes (486-465)_27
As part of the story of Hermotimos Herodotos details the workings of
the Chian slave-trader Panionios who castrated boys and sold them at
Ephesus and Sardis. The case of the Paphlagonian eunuch appears not in
Herodotos, but in the Persica of Ktesias of Knidos, a Greek doctor who
worked at the Persian court, c. 405-397. 28 The eunuch is Artoxares, court
eunuch of Artaxerxes I (465-424/3), 29 then chief eunuch of Darius II
(423-405), whom he supposedly sought to supplant as king, with the
aid of a false beard and moustache. Despite Ktesias' predilection for
mentioning eunuchs, Artoxares is the only one whose origin he records. 30
Hellenistic kingdoms
With the conquest of the Achaemenid empire by Alexander the Great
and the subsequent establishment of the Hellenistic kingdoms, the
Macedonian royal courts opened to the influence of the eastern model.
The old kingdom of Macedonia itself was not affected it seems, but there
is evidence for eunuchs at the Ptolemaic and Seleucid courts, as Grayson
notes, though a fuller picture is provided by the survey of Guyot. 31 Under
the Ptolemies we encounter such diverse figures as Aristonikos (active
under Ptolemy V) 32 and Eulaios (active under Ptolemy VI), 33 whilst under
the Seleucids we meet Krateros (under Antiochus VII). 34 Concerning the
ethnic origin of such Hellenistic eunuchs, Guyot states 'that in general
those who served the Graeco-Macedonian elite of the Hellenistic courts
were barbarians. He argues that Eulaios was an imported slave, and
although Aristonikos and Krateros identifY themselves in inscriptions
respectively as the son of Aristonikos, from Alexandria, and the son of
Krateros, from Antioch, he sees these as fictitious claims to ease their
assimilation to the elite. 35
Beyond the Seleucid and Ptolemaic kingdoms, a most striking case
is that of the kingdom of Pontus under Mithridates VI (120-63 BC),
that infamous opponent of Rome and member of a dynasty of Persian
descent. 36 Eunuchs appear often in Appian's account of Mithridates and
his court. 37 We meet Dionysius a naval officer, Tryphon a would-be
fortress-holder, special agents and general attendants on the king and the
royal women. Yet on the question of origin we are frustrated. There is no
146
direct statement about origin and, as Guyot notes, while all but one have
Greek names, this reveals nothing about their place of origin, as slaves here
too were given new names. 38 It can be noted however that Appian records
that Mithridates' son punished the pro-Roman city of Amisus in Pontus
by selling the inhabitants into slavery and castrating the boys. 39
One Hellenistic eunuch whose origin we do appear to know is the
famous Philetairos (Frontispiece), a Paphlagonian. 40 Strabo in particular
details the history of this first Attalid dynast of Pergamum. 41 The story
runs that he became a eunuch by accident, when as a child he was crushed
in a crowd. Thus Philetairos appears to be a special case, for he was not
deliberately created for service. However this depends on whether one
believes the story. It could be malicious slander, or one could argue that
it was devised to cover up a deliberate castration; Guyot prefers the latter
interpretation. 42
Medieval Islam
The court eunuchs in this context, Grayson says, are again foreigners. 43
Ayalon, the major contributor to this field so far, 44 commented:
the overwhelming majority of the eunuchs ... had been brought over from
outside the borders of the Muslim lands. They had thus been usually
transplanted from a different civilization and different religion ... to a Muslim
environment. 45
Ayalon contrasts this with other civilizations where eunuchs could be

drawn from the local population and have 'very strong local roots', citing
in particular Byzantium and China; he remarks 'Nepotistic attitudes of
eunuchs towards their relatives were quite common in Byzantium and
China'. Ayalon states that for the eunuch within the Islamic system 'there
was nothing to divert his exclusive devotion from his patron'. 46 Given the
diversity of races able to be exploited as eunuchs in the Medieval Islamic
world we find both so-called 'black' and 'white' eunuchsY
Ottoman empire
On this medieval and modern Islamic empire, Grayson remarks: 'the
eunuchs in Ottoman Turkey were all foreigners, coming from Mrica,
including Ethiopia'. 48 Certainly it is true that Mrica was key in the supply
of 'black' 49 eunuchs 'via Egypt, present-day Libya, the Red Sea, Arabia,
and the Persian Gulf'. For the study of this supply Toledano is notable
for his exploitation of the Turkish archival source of the Register of the
Biographies of the Imperial African Eunuchs, yet even with this resource
he notes that it is 'difficult to determine the exact places of origin of the
147
Shaun Tougher
African eunuchs'. 50 The Ottoman empire, however, also consumed white

slaves, notably from the Caucasus, though Penzer adds that there were in
addition Hungarians, Slavonians and GermansY A good example comes
from the sixteenth century, that of the Hungarian rebel Gazanfer. 52 He
was a convert to Islam, and castrated himself to remain in the service
of prince Selim when he became sultan in 1566.53 He became the chief
white eunuch and head of the privy council, and served the son and
grandson of Selim. Of specific regions of the Caucasus, Penzer identifies
Armenia, Georgia and Circassia, and notes that when the 'Circassian
slave-dealers had been driven from their country in 1864 they merely
carried their ... trade nearer Constantinople, in Rumelia, and on the Asiatic
coast at Brusa' .54 Overall, Toledano is content with the view that the
eunuchs were a foreign element in society and 'unlike other slaves, in most
cases ... did not form alternative family ties by marriage'. 55
China
China is perhaps the most famous imperial court that employed eunuchs. 56
On the ethnic origin of those used, Grayson declares 'there is no evidence
that any of the eunuchs were foreigners'. 57 This however is authoritatively
contradicted by Tsai; there was indeed a domestic supply, but the 'non-
Chinese peoples who lived in China's peripheral regions' were also
affected. 58 Under the Ming emperors most foreign eunuchs were supplied
from Annam and Korea. Tsai argues:
There were compelling reasons for placing eunuchs from outside races and
tribes in positions of trust, because one of the major concerns of the emperor
was the security of the imperial line. The best way to preserve it and keep
court secrets was to use foreign-born eunuchs. 59
Byzantium
Coming at last to Byzantium, we do find the classic scenario of outsider as
court eunuch. 60 In the late seventh century we meet Stephen the Persian,
who was treasurer (sakellarios) for the emperor Justinian II during his
first reign (685-95). 61 On the fall of Justinian Stephen's fate was to be
dragged bound by the feet to the market of the Bull and burned to death. 62
A ninth-century example is Damianos the Slav, who was chief eunuch
(parakoimomenos) of the emperor Michael III (842-67). 63 In the early
tenth century we also encounter Samonas the Arab. 64 Samonas (perhaps
in origin a prisoner of war) was a domestic servant in an, elite household
in Constantinople. He entered imperial service when he informed the
emperor Leo VI (886-912) of a plot against his life; he then rose to
148
become the emperor's chief eunuch. In the eleventh century the emperor
Constantine IX Mono machos ( 1042-59) appointed as grand hetaireiarch
the eunuch Constantine the Saracen. 65 Eunuchs could also find their way
to Byzantium from the west; in 949 the Italian ambassador Liudprand of
Cremona presented four penis-less eunuchs, known as Carzimasians, to
the emperor Constantine VII (913-59) as a gift; these eunuchs were items
usually sold by the merchants ofVerdun in Moslem Spain. 66
But for Byzantium more emphatic, and in apparent contrast to the
practice of the later Roman empire, is the internal production of eunuchs. 67
Evidence for this phenomenon is particularly pronounced for the tenth
and eleventh centuries, reflected in general comments and specific cases.
The tenth-century Arab writer Masudi matter-of-facdy notes that the
Byzantines, like the Chinese, did practise the castrating of their own
children. 68 This finds corroboration, for instance, in the story of the
background of Constantine the Paphlagonian, chief eunuch in the early
tenth century to the emperor Leo VI and then to the empress Zoe
Karbonopsina (fourth wife of Leo VI, regent for Constantine VII). 69
The story runs that Constantine was castrated by his father for the sake
of a career in Constantinople, and our source comments that this was
a custom amongst farmers in that region. 70 There are other good examples
of Byzantine eunuchs, even if we do not always necessarily know their
exact origin. 71 It is striking that several other Byzantine eunuchs whose
origins are known happen to be Paphlagonians, as Paul Magdalino has
recently highlighted. 72 Magdalino cites seven cases between the dates
906-1042, the most famous being John the Orphanotrophos.73 This
court eunuch even managed to secure the throne for his own brother,
Michael IV (1034-41), and then also for his nephew, Michael V (1041-2).
Strikingly John was not the only eunuch of the family; so were his brothers
Constantine and George.
In or out?
From this overview of the ethnic origin of eunuchs employed at a range
of royal and imperial courts the following observations can be made. As
noted at the start, the evidence provided by our sources is a major issue.
How full is the picture available? We have seen that the origin of eunuchs
is not always recorded. Despite this it is very dear in certain cases that it
is outsiders who are required; the Ottoman empire is a particularly good
example. At the same time, however, we have seen that domestic supply
exists. As Grayson noted, this is especially true for Byzantium and China,
and he is thus moved by these comparative examples to deduce that it
was an element of the Assyrian and Persian courts. Whether we accept
149
Shaun Tougher
these deductions or not, the picture, despite the supposed advantages

of foreign eunuchs,74 remains mixed. This is surely most striking in the
case of Rome and Byzantium, for the former was the intimate precursor
of the latter.
How can we account for this contrast between late Roman and
Byzantine court eunuchs? Magdalino, considering the specific case of
the Paphlagonian production of eunuchs, raises the issue of supply. 75 He
points out that under Justinian, according to Procopius, the creation of
eunuchs in Abasgia and thus their importation into the Roman empire,
was piously ended. 76 It is interesting to find Justinian in his own legislation
subsequently attempting to stamp out both the castrating of Romans and
castration by Romans, which he specifically contrasts with the termination
of the practice by the barbarians.77 Could this then be the context in
which domestic supply (specifically of Paphlagonians for Magdalino78 )
supplanted foreign supply?
It is a tempting hypothesis but we need to acknowledge that Justinian's
concern about the creation of Roman eunuchs was not new, as earlier
Roman imperial legislation makes quite dear.7 9 Constantine's decree that
eunuchs were not to be created within the Roman empire was not merely
an ideological statement, for he lays down penalties; the perpetrator was
to be subject to capital punishment and both the slave who had been
castrated and the place of castration were to be confiscated. Leo I desires
the importation of castrated barbarians in the context of railing against
the ownership of Roman eunuchs (whether castrated within or beyond
the empire). Ayalon already shrewdly drew attention to the implications
of these decrees, contrasting the repetitive Roman legisl~tion against
castration with the Islamic world where 'similar edicts are practically
unknown'. 80 The late antique legislation reveals that the 'emperors were
hoping to stop the practice of creating and selling Roman eunuchs, rather
than that they did stop it; the ideal of foreign eunuchs dearly did not
entirely match the reality. 81
It is instructive also to return to the late antique grammaticus Aelius
Donatus; he reports that most eunuchs in the east came from Armenia,
but that those in the west came from Gaul. 82 A Gallic eunuch is met
in Ambrose. 83 Wiedemann notes with alarm the contradiction these
references reveal between imperial law and practice, and seems to find relief
in the proposition that Celts beyond Roman jurisdiction (i.e. Ireland) are
meant. But perhaps it is not necessary to explain away the contradiction.
Despite these qualifications it still seems to hold true tha't Byzantium has
seen a shift. Whether this is due to the question of supply is debatable, 84
but it is dearly another sign of transition from ancient Rome to medieval
150
Byzantium. 85 A knock-on effect of this shift may have been the undermin-
ing of the political value of court eunuchs in Byzantium and the eventual
disappearance of the politically powerful court eunuch, the last truly
notable example being Nikephoritzes in the second half of the eleventh
century under the emperor Michael VII Doukas (1071-8). 86 This has
puzzled Byzantinists, and explanations have ranged from the westernisation
of Byzantine high society to the change in the nature of imperial rule
with the advent of aristocratic rule by the Komnenoi. 87 But perhaps the
Byzantines simply came to realize that homegrown eunuchs no longer
yielded the advantages that their non-Roman counterparts had once
provided. The isolation and devotion of the court eunuch had been lost;
eunuchs such as John the Orphanotrophos had their own agendas. It is
interesting to note Shepard's exploration of the use of foreigners in his
service by Alexios I Komnenos (1081-1118); perhaps this emperor was
striving to find an alternative to the failed eunuch system. 88
But maybe such an interpretation overvalues the belief that the ethnic
outsider status was such an essential and valuable commodity where court
eunuchs were concerned. We have encountered more than one culture
where a domestic supply of eunuchs existed. Was it such an advantage to
have foreigners, as historians such as Ayalon and Tsai think? If so, why did
some courts not follow this strategy rigorously? In the case of Byzantium,
a foreigner like Samonas could have more care for his own circumstances
than those of the emperor. 89 Ktesias records striking details concerning
Izabates, one of the leading eunuchs ofCambyses II (530-522). 90 Izabates
was not entirely isolated in society, for he had a cousin Kombaphis, also
a eunuch. This Kombaphis was said to have great influence with the
Egyptian king Amyrtaios, and engineered the loss of Egypt to the Persian
Cambyses, in order to become governor of the country. It appears that
eunuchs could act, or be imagined acting, contrary to the wishes of their
master and for their own benefit. Thus the assumption that eunuchs who
were ethnic outsiders were intrinsically advantageous to their masters
should be questioned. 91 Of course it must be remembered that eunuchs
could be outsiders for other reasons beyond their ethnic origin. As will
have been clear from this chapter, many eunuchs who worked at royal and
imperial courts were of slave origin. Low social origin is also a recurring
factor. But more fundamentally we need to realize that the essential fact
about eunuchs who served at courts is that they were eunuchs. Certainly
this was Patterson's response to Hopkins' analysis in his book Slavery
and Social Death. 92 Patterson found Hopkins' answer to 'Why eunuchs?'
unconvincing. Notably exploiting evidence from Byzantium and China,
he preferred to argue that what is crucial about court eunuchs is their
151
Shaun Tougher
symbolism, not where they come from: the ultimate ruler requires the
ultimate slave; and they are a mediating symbol between the human and
the divine, the subject and the ruler. Whether we agree with this answer
or not, the point that the solution lies with the condition of the eunuch
as a castrated man is surely true. 93 The comparative approach (focusing on
the aspect of the ethnic origin of court eunuchs), through the divergences
found across time and cultures, tends to reinforce this conclusion.
Notes
1 Hopkins 1963, which was the basis ofHopkins 1978, 172-96.
2 Hopkins 1963, 75 and n. 3; 1978, 189 and n.45.
' 3 1 Schlinkert 1994. This idea is echoed by Ayalon 1979a on Islamic eunuchs
and Tsai 1996 on Chinese eunuchs: see below for further comment.
4 These laws were preserved by Justinian, Cod. lust. IV.XLII. For Roman
legislation concerning castration from Domitian to Constantine see Guyot 1980,

45-51. For comment on Leo's legislation see Guilland 1943, 199.
5 See Guyot 1980, 31; Wiedemann 1986, 139; Scholten 1995, 28 and
n.124.
6 Procop. tVtlrs 8.3.12-21.
7 See Scholten 1995, esp. 28-33.
8 As Scholten 1995 stresses, these eunuchs were mostly of slave status; they
were to be bought within the empire as commodities. As such they were sundered
from their society of origin, forced into total dependence upon their buyers and
owners, enhancing their outsider status.
9 Jones, Martindale and Morris 1971, 314-15; also Dunlap 1924, 270-1;
Guyot 1980, 201-2.

10 Martindale 1980, 440-4; also Dunlap 1924, 272-84; Guyot 1980, 31.
11 Martindale 1980, 704-5. It seems that Mamas was not a ty,pical product of
the slave trade in eunuchs; he was castrated for medical reasons, and had family
contacts at the monastery of St Theodosius near Jerusalem.
12 Martindale 1980, 101-2; also Greatrex and Bardill1996.
13 Martindale, 1992 B, 912-28. See also Dunlap 1924, 284-99; Fauber 1990.
14 Hunt 1996, 569, states that the imperial eunuchs at the late Roman court
were 'usually freed slaves from Armenia or Persia'.

15 Grayson 1995. The main problem is one of terminology, but there is also
the question of the reading of visual images.
16 Indeed while the emperor Leo VI, Nov. 26 (Noailles and Dain 1944, 100-5)
does permit eunuchs to adopt, in Nov. 98 he still insists on the ban on marriage,
since the purpose of marriage in Christian society was the production of children
(Noailles and Dain 1944, 320-7).
17 Grayson 1995, 92.
18 The so-called Kletorologion of Philotheos: see Tougher 1997a, 171.
19 Grayson 1995, 95. Xen. Cyr. 5.2.28, 5.3.8, 8.4.2, certainly includes the
152
story of the noble eunuch Gadatas (taken as an Assyrian by, e.g., Gera 1993, esp.
203--4), though his castration was a punishment inflicted by the Assyrian king
rather than being due to a deliberate decision to become a court eunuch. Notably
Gadatas did eventually acquire a court role, but for the Persian king Cyrus, whose
chief mace-bearer he became. See Guyot 1980, 204.
20 See the contribution of Llewellyn-Jones in this volume.
21 Xen. Cyr. 7.5.59-65. On this text and eunuchs in Ktesias and Herodotos
also, see Gera 1993.

22 Hdt. 3.92.
23 Grayson 1995, 89.

24 Not only Herodotos' reference but perhaps also Bagistanes: Guyot 1980,
188.
25 Guyot 1980, 31.
26 Hdt. 6.32.
27 Hdt. 8.104-6. Hermotimos was a victim of war, as Gera 1993, 218,
comments.
28 On Ktesias and the nature of his text see Sancisi-Weerdenburg 1987. The
text was in part preserved by Photios who, as a ninth-century Byzantine, would

have been familiar with the phenomenon of court eunuchs due to their place
within his own society.
29 Though exiled to Armenia for his defence of Megabyzos.
30 In connection with the ethnic origin of Artoxares it is interesting to note that
Lewis 1977, 20-1, has suggested that the portrayal of the Athenian demagogue
Cleon by Aristophanes in his Knights as the Paphlagonian slave owes something
to this real-life powerful Paphlagonian eunuch, rather than being explicable
simply as a pun on 1ta<!>A.<ism (paphlazo), meaning 'to bluster'. Dover 1972,
89, notes that the 'representation of [Kleon] as a Paphlagonian is designed to
suggest, in accordance with the common forms of political antagonism, that
he is not of true Athenian origin', as well as making us think of 'bluster'. It
seems that Artoxares also served as the model for the character of Artaxates, the
most powerful and trusted of the Persian king Artaxerxes' eunuchs in Chariton's
Callirhoe: see Goold (ed.) 1995, 11.
31 Grayson 1995, 88; Guyot 1980, 92-120.
32 A rare eunuch for he is praised: Polyb. 22.22. One wonders why Ammianus
did not note this example of a good eunuch in his quest to find an example from
history prior to that paragon Eutherius: Amm. Marc. 16.7.
33 Eulaios is cast in the more typical role as an effeminate corrupter: Diod. Sic.
30.15. See also Polyb. 28.20-1.

34 He held the rank 'of the first friends'; he was chief physician, and grand
chamberlain of the queen.

35 Guyot 1980, 103. This may seem rather bold, but he argues that we can see
the same phenomenon in the case of Philetairos, for whom see below.
36 See for example McGing 1986, Hind 1994.
37 And note also Ammianus' story about Menophilos a eunuch ofMithridates:
Amm. Marc. 16.7.9.
153
Shaun Tougher
38 Guyot 1980, 99.

39 App. Bell. Civ. 2.91.
40 See Guyot 1980, 219-20.
41 Strabo 13.4.1.
42 The most recent discussion is Ogden 1999, 199-201.
43 Grayson 1995, 87.
44 And now Ayalon 1999. But see also Marmon 1995.
45 Ayalon 1979a, esp. 69, repr. 1988, III. See also Ayalon 1977, repr. 1979b,
III.
46 Ayalon 1979a, 70.
47 Ayalon 1979a, 72-3.
48 Grayson 1995, 86. See also Penzer 1936, 125-51.
49 But see the qualification ofToledano 1984.

50 Toledano 1984.
51 Penzer 1936, 119-23. See also Gibb and Bowen 1950, 76-82.
52 From Peirce 1993, 12, 179.
53 His brother Jafer also had himself castrated, but did not survive the
operation.
54 Penzer 1936, 122.
55 Toledano 1984, 381.
56 See for instance the contribution ofTsai in this volume.
57 Grayson 1995, 86.
58 Tsai 1996, 14-17. See also Mitamura 1970, 52-65. Jay 1993, stresses that
eunuchs could have family life after castration, thus undermining the view that it
was loyalty to the emperor through lack of any other social ties that made eunuchs
so desirable. She does not seem to consider the issue of origin.
59 Tsai 1996, 16.
60 AI-Jahiz (d. 868-9) in his Animals and his Refutation of the Christians,
accuses the Byzantines of kidnapping Muslim children and castrating them

later: see Pellat 1969, esp. 88, and Allouche 1939. I thank Al-An;lin Abou-Seada
for providing me with this information. On Byzantine eunuchs see also the
contributions of Gaul, Mullett and Sideris in this volume.
61 Kazhdan (ed.) 1991, vol. 3, 1955. He features in the histories ofNikephoros
and Theophanes.
62 Nikeph. 40.37-41: Mango 1990, 96-9. Also Theoph. 369: Mango and Scott
1997, 515.
63 Guilland 1943, 220. Theoph. Cont. 5.16.
64 Kazhdan (ed.) 1991, vol. 3, 1835-6; Janin 1935; Jenkins 1948, repr. 1970, X;
Ryden 1984; Tougher 1997b.

65 Guilland 1943, 211. Sky/.: Thurn (ed.) 1973, 438.63-5.
66 Antapodosis 6.6.
67 For a more detailed consideration of this transformation see Tougher
forthcoming. '
68 Lunde and Stone 1989, 345; Ayalon 1979a, 75 and n.4.
69 See Tougher 1997b, 200-1; Magdalino 1998, 143-4.
154
70 See Tougher 1997a, 178-9; Magdalino 1998, 148.

71 See Guilland 1943: Niketas under Nikephciros II Phokas; Niketas ofpisidia;
Eustathios Kyminianos, under Alexios I. There is also Stephen the Pergamene,
under Constantine IX: SkyL: Thurn (ed.) 1973, 423.42; 428.82, 85, 94; 429.29;
430.32.
72 Magdalino 1998. See also Tougher 1997a, 178-80.
73 Amongst the other seven are Constantine Gongylios, Joseph Bringas, and
the uncle of Symeon the New Theologian, if not Symeon himself. On John
see Kazhdan (ed.) 1991, vol. 2, 1070; Janin 1931. Angold 1991, esp. 22 and
n. 48 usefully draws attention to a connection between John and the Komnenoi:
'Isaac I Comnenus granted an annual sum of 24 nomismata to the monastery of
Theotokos Dekapolitissa for a candle to burn in memory above the grave of]ohn
the Orphanotrophos.' It seems odd that Michael Psellos does not make more
of John's condition as a eunuch.
74 See below for further reflection on this issue.
75 Magdalino 1998, 149-50.
76 Procop. "!Vttrs 8.3.12-21. The palace eunuch Euphratas, himself an Abasgian,
was sent out to relay Justinian's command.

77 Nov. 142. The emperor also insists that those who have been castrated within
the empire are to be freed.

78 Though eunuchs from Paphlagonia have been met previously, e.g. the
Achaemenid eunuch Artoxares and the Hellenistic eunuch Philetairos.

79 Preserved by Justinian, Cod. lust. IV.XLII (see also Dig. XLVII, 8, 4.2,
for Hadrian's legislation on castration). For later legislation on castration see

e.g. Leo VI, Nov. 60 (Noailles and Dain 1944, 222-7). Leo still disapproves
of the practice, though thinks it wrong that the practisers should be castrated
themselves. Instead goods are to be confiscated, they are to be exiled, and the
victim if a slave is to be freed. He however says that if the castrated person was
free they are responsible for themselves. Leo also recognizes that often castration
is necessary on the grounds of health.
80 Ayalon 1979a, 70.
81 Whether Roman eunuchs served at the Roman court is a moot point.
82 Guyot 1980, 31; Scholten 1995, 28 and n. 124.
83 Wiedemann 1986.
84 And note also that Byzantium can still acquire foreign eunuchs, as pointed
out above.
85 See Tougher forthcoming.
86 For Nikephoritzes see Kazhdan (ed.) 1991, vol. 3, 1475; Angold 1997,
121-4; Lemerle 1977, 300-2. Nikephoritzes' family had their origins in the
theme of Boukellarion.
87 For the westernisation theory see Guilland 1943, 234. For the Komnenian
solution see Kazhdan and Wharton Epstein 1985, 69-70. See also the discussion
in Gaul's contribution to this volume. Gaul stresses that eunuchs are at least still
to be found at the late Byzantine court.
88 Shepard 1996.
155
Shaun Tougher
89 But see Jenkins 1948 for a loyal explanation of Samonas' flight.

90 Note also for example that Euphratas the Abasgian eunuch of Justinian was
survived by a nephew: Procop. SH29.13.
91 Woods 1998 now even questions Ammianus' image ofEutherius' devotion to
Julian; he was devoted rather to another master, Constantius II.

92 Patterson 1982, 317-31.
93 For a recent attempt to answer this question for Byzantium see Ringrose
1996.
Bibliography
Allouche, I.S.
1939 'Un traite de polemique de christiano-musulmane au IX• siecle', Hesperis
26, 123-55.
Angold, M.
1991 'The Byzantine state on the eve of the battle of Manzikert', ByzF 16,
9-34.
1997 The Byzantine Empire, I 025-1204: A political history, 2nd edn, London
and New York.
Ayalon, D.
1977 'The eunuchs in the Mamluk Sultanate', in Studies in Memory ofGaston
W0it, Jerusalem, 267-95.
1979a 'On the eunuchs in Islam', jerusalem Studies in Arabic and Islam 1,
67-124.
1979b The Mamluk Military Society, London.
1988 Outsiders in the Lands ofIslam: Mamluks, Mongols and eunuchs, London.
1999 Eunuchs, Caliphs and Sultam: A study in power relationships, Jerusalem.
Dover, K.J.
1972 Aristophanic Comedy, Berkeley and Los Angeles.
Dunlap, J .E.
1924 'The office of the grand chamberlain in the later Roman and Byzantine
empires', in Two Studies in Later Roman and Byzantine Administration,
New York and London, 161-324.
Fauber, L.H.
1990 Narses: Hammer ofthe Goths, Gloucester and New York.
Gera, D.L.
1993 Xenophon's Cyropaedia: Style, genre, and literary technique, Oxford.
Gibb, H.A.R. and Bowen, H.
1950 Islamic Society and the W0st, vol. 1, London, New York and Toronto.
Goold, G.P. (ed.)
1995 Chariton, Callirhoe, Cambridge, Mass., and London.
Grayson, A.K.
1995 'Eunuchs in power. Their role in the Assyrian' bureaucracy', in
M. Dietrich and 0. Loretz (eds.) Festschrift for Wolfram Freiherrn von
Soden (Alter Orient und Altes Testament 240), Neukirchen, 85-98.
156
Greatrex, G. and Bardill, J.

1996 1\ntiochus the praepositus: a Persian eunuch at the court ofTheodosius
II', DOP50, 171-97.
Guilland, R.
1943 'Les eunuques dans l'empire byzamin', REB 1, 197-238.
Guyot, P.
Stuttgart.
Hind,J.G.F
1994 'Mithridates', in ].A. Crook, A. Lintott and E. Rawson (eds.) The
Cambridge Ancient History, vol. IX, 2nd edn, Cambridge, 129-64.
Hopkins, K.
Hunt, E.D.
1996 'Eunuchs: secular', in S. Hornblower and A. Spawforth (eds.) Oxford
Classical Dictionary, 3rd edn, New York and Oxford, 569.
Janin, R.
1931 'Un ministre byzantin. Jean l'orphanotrophe (XI< siecle)', EO 30,
431---43.
1935 'Un arabe ministre aByzance: Samonas (IX<-Xe)', EO 34, 307-18.
Jay, J.W.
1993 1\nother side of Chinese eunuch history: castration, marriage, adoption,
and burial', Canadian journal ofHistory 28, 459-78.
Jenkins, R.J.H.
1948 'The flight ofSamonas', Speculum 23,217-35.
1970 Studies on Byzantine History ofthe 9th and I Oth Centuries, London.
Jones, A.H.M., Martindale, J.R. and Morris,].
1971 Prosopography of the Later Roman Empire, vol. 1, AD 260-395,
London.
Kazhdan, A. (ed.)
1991 Oxford Dictionary ofByzantium, 3 vols, New York and Oxford.
Kazhdan, A.P. and Wharton Epstein, A.
1985 Change in Byzantine Culture in the Eleventh and Twelfth Centuries,
Berkeley, Los Angeles and London.
Lemerle, P.
1977 Cinq etudes sur le XJe siecle Byzantin, Paris.
1986 Byzantine Humanism, tr. H. Lindsay and A. Moffatt, Canberra.
Lewis, D.M.
1977 Sparta and Persia, Leiden.
Lunde, P. and Stone, C.
1989 Masudi: The meadows ofgold. The Abbasids, London.
McGing, B.C.
1986 The Foreign Policy ofMithridates VI Eupator King ofPontus, Leiden.
157
Shaun Tougher
Magdalino, P.
1998 'Paphlagonians in Byzantine high society', H BusavnviJ MtKpa Acria
(6o~-12o~ at.), Athens, 141-50.
Mango, C.
1990 Nikephoros of Constantinople Short History, Washington D.C.
Mango, C. and Scott, R.
1997 Chronicle ofTheophanes Confessor, Oxford.
Marmon, S.
1995 Eunuchs and Sacred Boundaries in Islamic Society, New York and
Oxford.
Martindale, J.R.
Cambridge.
1992 Prosopography of the Later Roman Empire, vol. 3, AD 527-641, 2 vols.,
Cambridge.
Mitamura, T.
Noailles, P. and Dain, A.
1944 Les novelles de Leon VIle sage, Paris.
Ogden, D.
1999 Polygamy, Prostitutes and Death: the Hellenistic dynasties, London.
Patterson, 0.
1982 Slavery and Social Death. A comparative study, Cambridge, Mass., and
London.
Peirce, L.P.
1993 The Imperial Harem, New York and Oxford.
Pellat, C.
1969 The Life and "U!Orks ofjahiz. Translations ofselected texts, tr. ,D .M. Hawke,
London.
Penzer, N.M.
1936 The Harem, London.
Ringrose, K.M.
(ed.) Third Sex, Third Gender. Beyond sexual dimorphism in culture and
1996 'Eunuchs as cultural mediators', ByzF23, 75-93.
Ryden, L.
1984 'The portrait of the Arab Samonas in Byzantine literature', Graeco-
Arabica 3, 101-8.
Sancisi-Weerdenburg, H.
1987 'Decadence in the empire or decadence in the sources? From source
to synthesis: Ctesias', in H. Sancisi-Weerdenburg and A. Kuhrt (eds.)
Achaemenid History 1, Leiden, 33--45.
Schlinkert, D.
1994 'Der Hofeunuch in der Spatantike: Ein gefahrlicher AuSenseiter?',
Hermes 122, 342-59.
158
Scholten, H.
1995 Der Eunuch in Kaisernahe (Prismata V), Frankfurt am Main, Berlin,
Berne, New York, Paris and Vienna.
Shepard,].
1996 '"Father" or "scorpion''? Style and substance in Alexios's diplomacy', in
M. Mullett and D. Smythe (eds.) Alexios I Komnenos, vol. 1, Belfast,
68-132.
Thurn, I. (ed.)
1973 Joannis Scylitzae: Synopsis Historiarum, Berlin and New York.
Toledano, E.R. .
1984 'The imperial eunuchs of Istanbul: from Mrica to the heart of Islam',
Middle Eastern Studies 20, 379-90.
Tougher, S.
1997a 'Byzantine eunuchs: an overview, with special reference to their creation
and origin', in L. James (ed.) Women, Men and Eunuchs. Gender in
Byzantium, London and New York, 168-84.
1997b The Reign ofLeo VI (886-912), Leiden, New York and Koln.
Forthcoming 'From Rome to Byzantium: the transformation of the court
eunuch'.
Tsai, S-s. H.
1996 The Eunuchs in the Ming Dynasty, Albany.
Wiedemann, T.
1986 'An early Irish eunuch?', LCM 11.8 (Oct.), 139-40.
Woods, D.
1998 ~mianus and Eutherius', Acta Classica 41, 105-17.
159
9
'EUNUCHS OF LIGHT' 1
Power, Imperial Ceremonial and Positive Representations
of Eunuchs in Byzantium (4th-12th centuries AD)
Georges Sideris
The study of Byzantine eunuchs has been dominated by a negative image

of these men, reflecting the 'eunuchophobic' point of view found in
some Byzantine writers. Eunuchs are often described in Byzantine sources
as creatures who are both physically and morally deformed. 2 Byzantine
writers are especially harsh when they speak about imperial eunuch
servants, the cubicularii (chamberlains), and their chief, the praepositus
sacri cubiculi (grand chamberlain).
Such reproaches are particularly sharp during the fourth century, where
we find them expressed by both pagan and Christian authors. Among the
pagans, the criticisms of Eusebius the grand chamberlain of the emperor
Constantius II (337-61) by the historianAmmianus Marcellinus, are very
famous. Nevertheless Ammianus is not systematically opposed to eunuchs,
and a number of his criticisms can be read simply as literary play. 3 The
poet Claudian's attacks, however, are severe. At the end of the fourth
century he wrote two fearsome invectives against Eutropius, the grand
chamberlain of the emperor Arcadius (395-408). 4 Among the Christians,
the bishops Athanasius of Alexandria, Basil of Caesarea and Cyril of
Alexandria attacked eunuchs vividly, the latter two both drawing up a list
of their vices. 5 But here the opposition to eunuchs is linked with religious
questions. In the case of Athanasius and Basil of Caesarea, eunuchs are
Arians, opposed to orthodox faith, and they encouraged the emperors
Constantius II and Valens (364-78) in this direction. Cyril's criticisms
were more hypocritical, because he tried to buy the eunuchs' religious and
political support with many significant gifts. 6 Several centuries later, in
his early twelfth-century treatise, In Defence of Eunuchs, the archbishop
of Ochrid and Bulgaria, Theophylact, was to list and refute the charges
usually brought against castrated servants? Theophylact says that he wrote
this treatise because his brother, a eunuch himself, was irritated by the
attacks against eunuchs. 8
161
George Sideris
When we read these attacks, we can see that the charges are the same
in the fourth as in the twelth century. 9 It would be tedious to list these
alleged traits which, in fact, are stereotypes. In general authors described
eunuchs as unjust, evil-minded, greedy, easily corrupted, egotistical,
jealous, ~owardly, deceitful and servile. Sometimes, they denounce their
indiscriminate sexual depravity, asserting that they have relations with
either men or women. 10 The fact that eunuchs were often slaves or
former slaves feeds the negative images, for their lowly situation is seen as
increasing their natural servility and their vices. 11
Often, these attacks are linked to a conception of eunuchs derived
from physiognomical treatises and the medico-philosophical tradition.
Physiognomy claimed to interpret the traits of the soul from physical
characteristics. The bodily corruption engendered by castration involved
the soul's corruption. As a result of losing their virility, eunuchs lost all
the qualities which made the masculine character: temperance, courage,
fidelity, sense of justice, moral sense, altruism. In fact, since eunuchs are
androgynous, their traits are close to the characteristics assigned to women.
Nevertheless, even if eunuchs are no longer men, they do not become
women, so they do not have the advantage of womanly qualities; feminine
aspects mean, in the case of eunuchs, vice and weakness. 12
Nevertheless, this 'eunuchophobia' is just a political tendency. We find,
throughout Byzantine history, a stream of thought favourable to eunuchs,
and particularly to imperial chamberlains. I will present here testimonies
by different authors, from the fourth to the twelfth centuries, which paint
a positive picture of Byzantine eunuchs.
Praise of Eutherius and Mardonius

In the fourth century, more precisely in the 380s, Ammiaims Marcellinus
included in his history an encomium, in a literary style, of the eunuch
Eutherius. This eunuch was the grand chamberlain ofJulian, when he was
in Gaul serving as the Caesar of the emperor Constantius II. Ammianus
knew Eutherius, whom he had met at Rome, and the eunuch was an
historical source for him. Ammianus praises his 'extraordinary powers of
memory', so we can assume that he was directly informed by Eutherius
himself of the main events of his life. In fact, in his work (including
a long digression), Ammianus gives us a short but precise biography of
the eunuch. 13 Eutherius was born in Armenia, where he was captured as
a child by hostile tribesmen. He was castrated and sold to some Roman
traders who took him to Constantine's palace, where he. lived, was well
educated and 'gradually gave evidence of virtuous living and intelligence'. 14
He served under the emperor Constans in the western part of the Roman
162
'Eunuchs oflight:· ... Positive representations ofeunuchs in Byzantium
empire. Later he became grand chamberlain to Caesar Julian in Gaul,

whom he was willing to criticize. Ammianus stresses his virtues and
qualities:
conspicuous for his remarkable keenness in devising and solving difficult
and knotty problems, he had extraordinary powers of memory; he was eager
to do kindnesses and full of sound counsel. .. In unrolling many records of
the past, to see to which of the eunuchs of old I ought to compare him,
I could find none. 15
In fact, Ammianus here describes Eutherius as having the qualities usually
ascribed to males and not to eunuchs, thereby implying that Eutherius
was a real male.
When Julian became emperor himself he dismissed Constantius'
eunuchs. He hated in particular the grand chamberlain Eusebius, and
set up a tribunal at Chalcedon which tried Constantius' supporters and
condemned the eunuch to death. 16 Nevertheless Julian praised the culture
and the moral standards of his tutor, a Scythian eunuch, Mardonius, who \
had formerly been the teacher of his mother BasilinaY
Eunuchs and sanctity in Byzantium

Byzantine Christians could read in the Bible veritable panegyrics of
eunuchs. Some passages were particularly important for the Church
Fathers. They interpreted the term 'eunuch' in both the literal and the
metaphorical sense. 18 Notable in the Old Testament is Isaiah 56.3-4:
neither let the eunuch say, Behold, I am a dry tree. For thus saith the Lord
unto the eunuchs who keep my sabbaths, and choose the things that please
me, and take hold of my covenant; Even unto them will I give in mine house
and within my walls a place and a name better than of sons and of daughters:
I will give them an everlasting name, that shall not be cut off.
And in the New Testament Matthew 19.12:
For there are some eunuchs, which were so born from their mother's womb:
and there are some eunuchs which were made eunuchs of men: and there be
eunuchs, which have made themselves eunuchs for the kingdom of heaven's
sake. He that is able to receive it, let him receive it.
In Acts 8.26-40, Philip baptizes the Ethiopian eunuch of Candaces,
queen of Ethiopia. 19
Above all, the imperial eunuchs' fame is based on the example of
Diocletian's eunuchs. These chamberlains count among the Church's first
martyrs during the Great Persecution of 303. 20 Byzantine hagiography
praises the courage of many eunuchs who chose death rather than apostasy.
163
George Sideris
In fact only some of these eunuch saints were real eunuchs. Nevertheless,
the most important thing is that Byzantines worshipped them as eunuch
saints. In the twelfth century, Theophylact names Indes, Yakinthos and.
Proteus. 21
Several eunuch chamberlains are praised for their faith, their piety
and their charity. We have quotations from both Orthodox and non-
Chalcedonian writers. Thus, Mark the Deacon, in his Life of Porphyry,
bishop of Gaza in the late fourth to early fifth centuries, describes the
eunuch Amantios, castrensis (steward of the imperial household) of the
empress Eudoxia, as someone truly irreproachable who did not cease to
distribute alms, to fast, to give shelter to numerous guests and to contribute
to pious works. 22 In 419-20, the bishop Palladius ofHelenopolis dedicated
his Lausiac History to the grand chamberlain Lausus, whom he qualified as
a 'very faithful servant of Christ'. He praised him for, among other things,
his generosity, his alms, his virtue, and his love of knowledge. 23
In his Lives ofthe Eastern Saints, John of Ephesus describes the lives of
the chamberlains Mishael and Theodore. They were both monophysites,
like John of Ephesus himself Mishael was grand chamberlain, probably
under the emperor Anastasius I (491-518). According to John, Mishael
was 'a Christian and merciful and ascetic man'. John speaks of Mishael as
a monophysite saint and martyr, a victim of persecution:
and (he) was moreover perfect in all spiritual things, insomuch that he
underwent exile for the sake of the truth of the right faith, that he might
not communicate with the synod of Chalcedon, insomuch that he spent
a considerable number of years in the exile.
At last Mishael was recalled to Constantinople. He completed his time
in the palace and retired. He gave his wealth to the poor, then 'lived an
ordinary and poor life, down to extreme old age, and thus departed from
the world bearing great and noble testimony'. 24
The chamberlain Theodore served first under Mishael and eventually
retired with great wealth after serving as castrensis under Justinian I (527-65).
According to John of Ephesus, Theodore 'began eagerly and joyfully to
scatter that gold, and to sow all his wealth in good soils which are the
hands· and bellies of the poor'. He distributed all his property, freed his
slaves, giving them gifts, sold his silver and his valuable clothes. He lived
like a poor man in a villa called Serna, where Mishael was buried, and
there he ended his life. Theodore had a brother, whose name was John,
who was also a chamberlain and imitated his ascetic life. The Life of the
Blessed Theodore calls him 'the blessed Theodore ... who, while he was in
the body, practised a heavenly and divine mode of life'. John of Ephesus
164
praises his ascetic way of life, his charity and his monophysite faith: 'his
soul was thus intoxicated with the fervour of divine love, and no worldly
property was regarded by him'. In fact Theodore is a real monophysite
eunuch saint. 25
In his chronicle, Theophanes Confessor shows how the iconoclast
emperor Leo IV (775-80) persecuted some chamberlains in 780, because
they worshipped holy icons. He paraded them in chains through the Mese,
the main street of Constantinople, then he locked them up. One of them,
Theophanes, who was formerly chamberlain and parakoimomenos (lord
of the imperial bedchamber), died. Theophanes Confessor considers him
a saint: 'Whereupon the said Theophanes died, thus becoming a confessor
and winning the crown of martyrdom.' 26
Many patriarchs were eunuchs. During the first iconoclasm there were
two eunuch patriarchs, Germanos I (715-30) (who was castrated when
his father, the patrician Justinian, was put to death under Constantine IV
because he had plotted against the emperor Constans II) and the iconoclast
patriarch Niketas I (766-80)_27 Methodios I (843-7) was perhaps also
a eunuch. 28 The eunuch patriarch Ignatios (847-58 and 867-77) was
the son of the emperor Michael I. During the tenth century there were
three eunuch patriarchs: Stephen II (925-7), Theophylact (933-56) (son
of the emperor Romanos I), and Polyeuktos (956-70). 29 There were
many eunuch bishops in Byzantium. For instance Theophylact cites the
archbishop ofThessalonike, and the bishops of Pydna, Petra, and Edessa
in Bulgaria. 30
Physical and spiritual beauty of eunuchs

The handsome eunuch, whose physical beauty reflects the beauty of his
soul, is a common subject in writings from the early Byzantine period.
Leontios Scholastikos, in a famous epigram which is a veritable miniature
panegyric, praises the beautiful heart and pretty face of Kallinikos, as well
as the sweetness of his voice:
Thou conquerest in beauty of soul as much as in beauty of face, for thou
possessest everything that is worthy of thy name, and ever in the bed-chamber,
sending the emperor to sleep, thou dost sow all gentleness in his eats. 31
Kallinikos was a very powerful grand chamberlain of the emperor Justinian.
He was one of the men who supported Justin II's accession to the throne
after Justinian's death in 565. 32 In Justin's reign, the panegyrist Corippus
says of Kallinikos:
you too (since you have lived thus in faithfulness and will continue to live so)
shall together with your rulers gain everlasting fame, glory and name. 33
165
George Sideris
This echoes Isaiah 56, and Corippus clearly links eunuch imperial court
service and celestial service. Furthermore Corippus gives this description
of the eunuch Narses, Justin's sword-bearer:
He was in gold all over, yet modest in dress and appearance, and pleasing
for his upright ways, venerable for his virtue, brilliant, careful, watchful
night and day for the rulers of the world, shining with glorious light: as the
morning star, glittering in the clear sky, outdoes the silvery constellations with
its golden rays and announces the coming of day with its clear flame. 34
Corippus here associates light and the eunuch's beauty. This association,
which is a topos in the case of an emperor, is astonishing here, concerning
a eunuch. Moreover, it is announced in an official panegyric, which means
that this theme was officially recognized as early as Justin's reign. We can
see here that the ideology of the panegyric sets out, within the context
of imperial service, themes which associate on the one hand Byzantine
court eunuchs and the kingdom of God, and on the other hand Byzantine
court eunuchs and light.
In the tenth-century Book ofCeremonies of Constantine VII (913-59),
imperial ceremonies, where eunuchs play an important role, are conceived
as an image on earth of the celestial kingdom, and they show the empire's
beauty. 35 The Peloponnesian patroness of emperor Basil I (867-86),
Danelis, brought a hundred graceful eunuchs to Constantinople for the
needs of the imperial court. 36
Eunuchs and angels 37

The vision of the luminous beauty of eunuchs, associated· with purity and
the soul's beauty, finds an important development in relation to angels.
Perhaps at the beginning of the 640s, Leontios, bishpp of Neapolis in
Cyprus, wrote a biography of John the Almsgiver, patriarch of Alexandria
(61 0-19). In the. course of his life John leaves Alexandria, which is
threatened by the Persian armies. He is called to Constantinople by the
emperor. During his journey, a eunuch appears to him:
When continuing their course they had reached Rhodes, the Saint whom God
had called saw with his waking eyes a eunuch in gleaming (E.~acrtpantovta)
apparel, a golden sceptre in his right hand, standing by him and saying:
'Come, I beg you, the King of Kings is asking for you!' 38
This passage is particularly interesting. We see that seventh-century

hagiography likens imperial eunuchs to angels. The theme of light, which
is used in accounts of imperial ceremonies to describe imperial eunuchs,
was also used to describe angels. We have here a complex connection.
The text uses two common themes to link eunuchs and angels. Eunuchs
166
'Eunuchs oflight~· ... Positive representations ofeunuchs in Byzantium
and angels are both royal messengers and they both introduce dignitaries
to the sovereign (the Byzantine emperor, Basileus, on earth and God in
heaven). In fact, the angel with his golden sceptre plays here the role of an
imperial ostiarios, with his golden stick. 39 God sends one of his messengers
to call John to him. Leontios explicitly draws a parallel between the
imperial eunuch and the angel when he says:
Without delay he [John the Almsgiver] forthwith sent for the patrician,
Nicetas, and said to him with many tears: 'You, my master, called me to
go to our earthly king, but the heavenly King has anticipated you and has
summoned to himself my humbleness.' He then related to him the vision
which he had just seen of the eunuch, or rather of the angel. 40
Leontios reinforces this association between eunuchs, light and angels

when he speaks about different miracles which occur after John's death:
For on the same day that this blessed man took his departure from this
life to go to his Lord, one of those who have practised the angelic way of
living and follow the monastic discipline, an admirable and virtuous man,
Sabinus by name, living in Alexandria, fell, as it were, into an ecstasy and
saw John, honoured of God, come out of his own palace with all the clergy,
bearing candles and going to the king, as, said he, a eunuch chamberlain
had summoned him. 41
The angel is described as a real eunuch chamberlain.

The so-called Oneirocriticon ofAchmet, a work that interprets dreams,
dated from the ninth to the eleventh centuries, indicates that if one
dreams of a gracious and tall eunuch, the eunuch should be considered
an angel, because a eunuch is pure, resembles an angel, and does not
feel carnal desire. 42
In the Narratio de Sancta Sophia (an anonymous text that perhaps
dates to the second half of the ninth century and brings together often
legendary tales about the erection of the Great Church of Constantinople)
an angel appears on two occasions in the guise of a handsome eunuch
(cuvouxo~ ... ropaio~ 'tql £ton) in dazzling or white dress, resembling an
imperial court messenger. The luminous white refers both to the dress of
angels in the Bible and to the clothing worn by eunuchs at court. The
court eunuchs appear here as veritable physical and earthly images of
the celestial angels. 43
The long version of a hagiographic text from the tenth century, entitled
The Vision ofthe Monk Kosmas, tells the story ofKosmas, formerly a servant
(koitonites) at the imperial palace under the emperor Alexander (912-13),
now a monk in the imperial monastery of the Theotokos tou Eusebiou
near the Sangarios River in the theme of the Optimates in Asia Minor.
167
George Sideris
In a vision, which probably occurred on 3 June 933, Kosmas travels

to celestial Jerusalem. 44 He reaches the palace of the Lord and there
on a terrace he finds luminous eunuchs. The heavenly palace is here
modelled after the palace complex in Constantinople, while the imperial
ceremonies are transposed into the celestial world. Eunuchs are thus
equated with angels, and we do not really know which one is the model
for the other. 45
Theophylact of Ochrid, in his treatise in favour of eunuchs, described
a eunuch as the most gracious of men, defended the pure eunuchs who
served the virtuous empresses, and said of certain of them that they were
portraits of the Logos. 46
The patrician, praipositos and sakellarios Leo

Thus, the idea of the beauty of eunuchs certainly existed in Byzantium,
and was not confined to literary sources. A miniature illustrates the relation
between the themes of the physical and spiritual beauty of eunuchs,
imperial ceremonial and the power of eunuchs. It dates probably to
940 and is placed at the beginning of a Bible, the Codex Reginensis
Graecus 1 B, fol. 2v. The Bible was originally in two volumes, and was
commissioned by the eunuch Leo the sakellarios (imperial treasurer) for the
monastery of Saint Nicholas, founded by his brother, the protospatharios
Constantine. It is a dedicatory miniature, a full page illumination, with
verses composed by LeoY It is in the antique style and belongs to
the so-called Macedonian Renaissance. 48 The miniature represents Leo
as praipositos, patrician and sakellarios, offering the B\ble that he had
commissioned to the Virgin Mary, Mother of God. Christ, through the
Virgin, accepts this gift. 49 Leo draws a parallel between the gift of this
Bible (the purpose of which is to receive the favour of Saint Nicholas, the
Virgin Mary and Christ) and the gift of their soul that the monks have
made to God. 5° Leo is represented in proskynesis, and is probably shown in
his court dress, consisting of a red cloak, trimmed in gold, under which
we recognize his white clothing. 51 The Virgin is dressed in a long cloak
and a maphorion, a cape, in purple, the colour of imperial majesty. The Virgin
and Christ are nimbed in gold. There is a rotunda behind the Virgin,
who stands on a pedestal. The composition is constructed along a diagonal
starting from Leo's right foot, passing through the Bible in the eunuch's
hands and then the Virgin's right hand and ending at the hand of Christ,
thus linking the eunuch to the Virgin and to Christ. 52
The text inside, like that outside the frame of the image, serves several
purposes. In the first place, it describes what is happening: the gift of the
Bible by a powerful court eunuch, and its acceptance. But, in addition,
168
the epigram that surrounds the image, makes dear its spiritual meaning:
Leo's gift of the Bible is to redeem his sins. This miniature illustrates the
eunuch's piety. 53 The eunuch's face, beardless and grey-haired, is turned
toward the Virgin with an expression and a movement of the eyes that
suggest profound contemplation. The beauty of the clothing and the
gestures, and the serenity of the eunuch's peaceful face, express the state
of grace, both physical and spiritual, in which the eunuch finds himself. 54
The piety and beauty of the eunuch were major themes related to the
power of eunuchs at the imperial court. These are the themes portrayed
in the miniature by the sakellarios Leo, which ultimately proves to be
an image of power. 55
In conclusion, as well as negative criticism of eunuchs, there also existed
in Byzantium, as studies have confirmed in recent years, a body of positive
representations of eunuchs. In particular, eulogies and favourable speeches
demonstrate the existence of a veritable eunuchophilia in Byzantium. But
it is power, and the way that it is portrayed, that appears central. Imperial
eunuchs in the emperor's service benefited from the effects of power
generated by their position in the court hierarchy and their participation
in imperial ceremonies. Further, they were themselves in a position to
elaborate portrayals of power favourable to them. The elegy of Kallinikos
in the sixth century and the miniature of the sakellarios Leo in the tenth
century are expressions and dazzling representations of this system.
Notes
1 This phrase ('<Protoetoe'ic; euvouxot') is taken from The Vision of the Monk
Kosmas: see Angelidi 1983, 861.88.

2 These criticisms are often topoi, inherited from the classical tradition: see
Ringrose 1999, 124.

3 See Sideris 2000.
4 See Long 1996. For the text see Hall (ed.) 1985, 143-89.
5 Basil: To Simplicia, a heretic, in Courtonne (ed.), letter 115, 19-20; Cyril:
Sermo steliteuticus adversus eunuchos, PG 77, col. 1105, 1108, 1109; Athanasius:
Historia arianorum ad monachos, PG 25, ch. 38, col. 737.
6 See Brown 1992, 16-17.
7 For Theophylact and his treatise see the contribution of Mullett in this
volume.
8 Gautier (ed.) 1980, 289, 1.1-6.
9 But see the comments of Mullett in this volume.
10 See for example Amm. Marc. 16.7.78, 18.5.4; Theophylact, in Gautier
(ed.) 1980,293, ll.9-17. For comment see Boulhol and Cochelin 1992, 55-6;
Ringrose 1994, 93-5 and 102-3. On eunuchs and homosexuality see Claudian,
169
George Sideris
In Eutropium, I, 65-77; Cyril, Sermo steliteuticus adversus eunuchos, PG 77, col.

1108 C; Theophylact, in Gautier (ed.) 1980, 295, 11.14-16. For comment see
Tougher 1999b, 94; Long 1996, 122-3.
11 Claudian In Eutropium, I, 181-6; Basil, To Simplicia, a heretic, in Courtonne
(ed.), p. 20,11.17-23. For comment see Hopkins 1978, 189.

12 See on these aspects: Treatise on Physiognomy, in Andre (ed.) 1981, 3-7, 40,
78, 98-128, 51-6, 83, 110-12 and 124-37; Adamantii Physiognomonica, B, 3,

in Foerster (ed.) 1994, 351-2; Amm. Marc. 16.7.7-8, 18.4.4-5; Theophylact,
in Gautier (ed.) 1980, 293, 11. 15-20 and 295, 11. 1-6. For comment see Evans
1969, 12-13; Ringrose 1994, 94; Long 1996, 121-34. On eunuchs as effeminate
see also Tougher 1999b; Guyot 1980, 38-42.
13 Amm. Marc. 16.7.2-8, 20.8.19, and 20.9.1-4. See also Jones, Martindale
and Morris 1971, 314-15; Thompson 1947, 20 and 80; Woods 1998, 106;
Tougher 1999a, 65-8.
14 Amm. Marc. 16.7.5.
15 Amm. Marc. 16.7.5-8, tr. Rolfe (ed.) 1982, 229-31.
16 Amm. Marc. 22.3.12. See also Matthews 1989, 92-3, 275.
17 Julian. Mis., in Lacombrade (ed.) 1964, 351b-352c, 175-6. On Mardonius
see also Jones, Martindale and Morris 1971, 558; Browning 1975, 36-7.
18 See for instance the contribution of Stevenson in this volume.
19 On these aspects see Boulhol and Cochelin 1992, 60-2; Ringrose 1999,
127.
20 Lactant. De mort. pers. 11.3, 14.2-4 and 15.2. Chamberlains Dorotheos,
Gorgonios, Petros: Euseb. Hist. eccl. 8.1.4 and 6.1-5. See also de Gaiffier 1957,
18-26; Jones, Martindale and Morris 1971, 270 (Dorotheos 2), 398 (Gorgonios
I); Guyot 1980, 194-5 (Dorotheos), 206 (Gorgonios), and 218 (Petros).
21 In Gautier (ed.) 1980, 327, 11.8-9. See also de Gaiffier 1957, 36-46; Boulhol
and Cochelin 1992, 62-74.

22 In Gregoire and Kugener (eds.) 1930, 36.15-16, 31; 37.20-21, 32; 52.3-14,
43. See also Martindale 1979, 66. ,

23 In Butler (ed.) 1967, Prologue, 10,11.9-17, 11, 1.4 and 23-5, 12,11.7-16,
(citation 11, 1. 23). On Lausus see also Martindale 1979, 660; Scholten 1995,
230-1.
24 John of Ephesus, Lives of the Eastern Saints 57, tr. Brooks 1925, 200-1. See
also Martindale 1992, 892-3.

25 See John of Ephesus, Lives ofthe Eastern Saints 3.57, tr. Brooks 1925, 200-6
(citations, 200, 202 and 203). See also Martindale 1992, 1244-5.
26 Theoph., in de Boor (ed.) 1883, 453, 11.10-20, tr. Mango and Scott 1997,
625.
27 On Germanos' career and on Niketas see Auzepy 1999, 289-92; Guilland
1943, 202 (= Guilland 1967, 168).

28 See Ringrose 1999, 126 and 133; Guilland 1943, 202-3 (= Guilland 1967,
168). '
29 On these and later eunuch patriarchs see Guilland 1943, 203 (= Guilland
1967, 168-9).
170
30 In Gautier (ed.) 1980, 297, ll.4-5.

31 Anth. Graec. 16.33, tr. Paton 1960, 179. Leontios plays on Kallinikos' name,
compounded of 'beauty' and 'victory'.
32 See Cameron (ed.) 1976, notes 65-78, 131-2; Martindale 1992, 260-1.
33 Corippus, In laudem Iustini 1.86-8, tr. Cameron (ed.) 1976, 88-9.
34 Corippus, In laudem Iustini 3.224-30, tr. Cameron (ed.) 1976, 106. On
Narses see Martindale 1992, 930-1.

35 De. cer., Vogt (ed.) 1967, for instance preface and ch. 1, 1-28.
36 Theoph. Cont. in Bekker (ed.) 1838, 318. See also Tougher 1997, 168;
Ringrose 1994, 95.

37 On the subject of eunuchs, angels and light see also Ringrose 1996, 86-9;
Mango 1980, 151-5.

38 Life ofjohn of Cyprus 52.34-9, in Festugiere with Ryden (eds.) 1974,403.
English translation from Dawes and Baynes 1948, 255.

39 On the ostiarios see Guilland 1955, 79-84 (= Guilland 1967, 286-9).
40 Life ofjohn of Cyprus 52.39-46, in Festugiere with Ryden (eds.) 1974, 403.

41 Life ofjohn of Cyprus 60.1-10, in Festugiere with Ryden (eds.) 1974, 408.

42 Achmetis Oneirocriticon, in Drexl (ed.) 1925, 10, 6, ll. 7-11; tr. Oberhelman
1991, 89. For the author and the date see Mavroudi 1998, vol. 1, 16-50 and
126-8; Oberhelman 1991, 11-13.
43 In Preger (ed.) 1901, 10, 86-7, and 11, 88-90. See also Dagron 1984,
22, 50, 192, 200-2, 230-4, 268-9, and nn. 84-98. On the association of the
terrestrial court with the heavenly court see Maguire 1997.
44 Angelidi 1983, ll. 1-5, 20-9 and 72-164, 79-80, 82-6, and see the introduc-
tion 74-6. On koitonites see Oikonomides 1972, 301 and 305. On celestial
Jerusalem see Angelidi 1982, 207-15.
45 Angelidi 1983, 86-7, ll.185-208, and see the commentary, 97 n.19 and
20.
46 In Gautier (ed.) 1980, 295, 11.23-4, and 319, 11.10-13. See also the
contribution of Mullett in this volume.

47 For a reproduction of the image see Mathews 1997, 89.
48 Dufrenne and Canart 1988, 11-12, 65-6; Mathews 1977, 96-8; Spatharakis
1976, 7 and 13; Brubaker 1999, 397.

49 Mathews 1977, 98-9.
50 See Mathews 1997, 88-90.
51 See Dufrenne and Canart 1988, 21; Spatharakis 1976, 9; Cutler 1995, 160
(repr. Cutler 2000); Ringrose 1996, 78.

52 Dufrenne and Canart 1988, 20-1; Spatharakis 1976, 8.
53 See Spatharakis 1976, 9-10; Dufrenne and Canart 1988, 20; Olster 1994,
419-20. On manuscripts, portraits, patronage and art in the middle Byzantine

period see also Mullett 1997, 227-8.
54 Dufrenne and Canart 1988, 21-2; Mathews 1997, 90.
55 On the possible access of Leo to imperial scriptoria in the production of his
171
George Sideris
Bible see Mathews 1977, 94. On the image of imperial power in the 9th-11th
centuries and imperial piety see ]olivet-Levy 1987, 441-70.
Bibliography
Andre, J. (ed.)
1981 Anonyme Latin, Traite de Physiognomonie, Paris.
Angelidi, c.
1982 'Remarques sur la description de la Jerusalem celeste', jOB 32-3,
207-15.
1983 'La version longue de la vision du moine Cosmas', Analecta Bollandiana
101, 73-99.
Auzepy, M.-F.
1999 L'hagiographie et l'iconoclasme byzantin. Le cas de la Vie d'Etienne le
Jeune, Aldershot.
Bekker, I. (ed.)
1838 Theophanes Continuatus, Bonn.
Boulhol, P. and Cochelin, I.
1992 'La rehabilitation de l'eunuque dans l'hagiographie antique (IV•-VI<
siecles)', Studi diAntichita Christiana 48, 49-76.
Brooks, E.W.
1925 'John of Ephesus. Lives of the Eastern Saints', Patrologia Orienta/is 19,
153-282.
Brown, P.
1992 Power and Persuasion in Late Antiquity. Towards a Christian Empire,
Madison.
Browning, R.
1975 The Emperor julian, London.
Brubaker, L.
1999 Vision and Meaning in Ninth- Century Byzantium. Irrzage as exegesis in the
Homilies of Gregory ofNazianzus, Cambridge. .
Butler, C. (ed.)
1967 The Lausiac History ofPalladius. A critical discussion together with notes
on Early Egyptian Monachism II, Hildesheim.
Cameron, Averil (ed.)
1976 Flavius Cresconius Corippus. In laudem Iustini Augusti minoris Libri
IV, London.
Courtonne, Y. (ed.)
1961 Saint Basile. Lettres, vol. 2, Paris.
Cutler, A.
1995 'The pathos of distance: Byzantium in the gaze of Renaissance Europe
and modern scholarship', in C. Farago (ed.) Reframing the Renaissance:
Visual culture in Europe and Latin America, 1450:---1650, London and
New Haven, 127-63.
2000 Byzantium, Italy and the North. Papers on cultural relations, London.
172
Dagron, G.
1984 Constantinople imaginaire. Etudes sur le recueil des Patria, Paris.
Dawes, E. and Baynes, N.H.
1948 Three Byzantine Saints, Oxford.
De Boor, C. (ed.)
1883 Theophanis Chronographia, vol. 1, Leipzig.
De Gaiffier, B.
1957 'Palatins et eunuques dans quelques documents hagiographiques',
Analecta Bollandiana 75, 17-46.
Drexl, F. (ed.)
1925 Achmetis Oneirocriticon, Leipzig.
Dufrenne, S. and Canan, P.
1988 Die Bibel des Patricius Leo. Codex Reginensis Graecus IB, Zurich.
Evans, E.C.
1969 'Physiognomies in the ancient world', Transactions of the American
Philosophical Society n.s. 59, 5, 1-101.
Festugiere, A.J., with Ryden, L. (eds.)
1974 Uontios de Neapolis. Vie de Symeon le Fou et Vie de Jean de Chypre,
Paris.
Foerster, R. (ed.)
1994 Scriptores Physiognomonici Graeci et Latini, vol. 1, Stuttgart and Leipzig.
Gautier, P. (ed.)
1980 Theophylacte d'Achrida, 1, Discours, traites, poesies, Thessalonike.
Gregoire, H. and Kugener, M.-A. (eds.)
1930 Marc le Diacre.Vie de Porphyre eveque de Gaza, Paris.
Guilland, R.
1955 'Etudes de titulature byzantine. Les titres auliques reserves aux eunuques',
REB 13, 50-84.
1967 Recherches sur les institutions byzantines, vol. 1, Berlin and Amsterdam.
Guyot, P.
Stuttgart.
Hall, J.B. (ed.)
1985 Claudii Claudiani Carmina, Leipzig.
Hopkins, K.
]olivet-Levy, C.
1987 'L'image du pouvoir dans l'art byzantin a l'epoque de la dynastie
macedonienne (867-1056)', B 57, 441-70.
Jones, A.H.M, Martindale, ].R., and Morris,].
1971 Prosopography ofthe Later Roman Empire, vol. 1, AD 260-395, London.
Lacombrade, C. (ed.)
1964 L'Empereur julien. CEuvres completes, vol. 2.2, Paris.
Long,].
1996 Claudian's In Eutropium Or, how, when, and why to slander a eunuch,
Chapel Hill and London.
173
George Sideris
Maguire, H.
1997 'The heavenly court', in H. Maguire (ed.) Byzantine Court Culture from
829 to 1204, Washington D.C., 247-58.
Mango, C.
1980 Byzantium. The Empire ofthe New Rome, London.
Mango, C. and Scott, R.
1997 The Chronicle of Theophanes Confessor. Byzantine and Near Eastern
History AD 284-813, Oxford.
Martindale, J.R.
Cambridge.
1992 Prosopography of the Later Roman Empire, vol. 3, AD 527-641, 2 vols.,
Cambridge.
Mathews, T.F.
1977 'The epigrams of Leo sacellarios and an exegetical approach to the
miniatures of Vat. Reg. Gr. I', Orientalia Christiana Periodica 43,
94-133.
1997 '42. Bible of Leo sakellarios', in H. C. Evans and W.D. Wixom (eds.)
The Glory of Byzantium. Art and Culture of the Middle Byzantine Era
AD 843-1261, New York, 88-90.
Matthews, ].
1989 The Roman Empire ofAmmianus, London.
Mavroudi, M.V.
1998 The So-called Oneirocriticon of Achmet: A Byzantine book on dream
interpretation and its Arabic sources, Thesis Harvard University, Cam-
bridge, Mass.
Mullett, M.
1997 Theophylact of Ochrid: Reading the letters of a Byzarztine Archbishop,
Aldershot and Brookfield.
Oberhelman, S.M.
1991 The Oneirocriticon ofAchmet. A Medieval Greek and Arabic treatise on
the interpretation ofdreams, Lubbock, Texas.
Oikonomides, N.
1972 Les listes de preseance byzantines des IX' et X' siecles, Paris.
Olster, D.
1994 'Byzantine hermeneutics after iconoclasm: word and image in the Leo
Bible', B 64, 419-58.
Paton, W.R. (ed.)
1960 The Greek Anthology, vol. 5, London and Cambridge, Mass.
Preger, T. (ed.)
1901 Scriptores Originum Constantinopolitanarum, vol. 1, Leipzig.
Ringrose, K.M.
(ed.) Third sex, Third Gender. Beyond sexual dimorphism in culture and
174
tion', in L. James (ed.) Desire and Denial in Byzantium, Aldershot and
Brookfield, 123-37.
Rolfe, J.C. (ed.)
1982 Ammianus Marcellinus, vol. 1, Cambridge, Mass. and London.
Scholten, H.
1995 Der Eunuch in Kaiserniihe (Prismata V), Frankfurt am Main, etc.
Sideris, G.
2000 'La comedie des castrats. Ammien Marcellin et les eunuques, entre
eunucophobie et admiration', Revue belge de philologie et d'histoire 78,
681-717.
Spatharakis, I.
1976 The Portrait in Byzantine Illuminated Manuscripts, Leiden.
Thompson, E.A.
1947 The Historical Work ofAmmianus Marcellinus, Cambridge.
Tougher, S.
1999a 'Ammianus and the eunuchs', in J.W. Drijvers and D. Hunt (eds.) The
Late Roman World and its Historian. Interpreting Ammianus Marcellinus,
London and New York, 64-73.
1999b 'Images of effeminate men: the case of Byzantine eunuchs', in D.M.
89-100.
Vogt, A. (ed.)
1967 Constantin VII Porphyrogenete. Le Livre des ceremonies, vol. 1, Paris.
Woods, D.
1998 'Ammianus and Eutherius', Acta Classica 41, 105-17.
175
10
THEOPHYLACT OF OCHRID'S
IN DEFENCE OF EUNUCHS
Margaret Mullett
Theophylact of Ochrid's In Defence ofEunuchs 1 has achieved iconic status2

in the study of Byzantine eunuchs. It is a text written for a eunuch which
presents a vigorous defence of the state of eunuchism. In this chapter,
instead of engaging with the text in its place in eunuch history, or in the
Byzantine gender system, I shall look at it as the product of a particular
author at a particular time for a particular reader, and so examine the
primary reception of this text.
Theophylact of Ochrid and Demetrios Hephaistos

Theophylact Hephaistos was born in Euboia around the middle of the
eleventh century, and educated in Constantinople in the 1070s, where
he taught in a school and then served as maistor ton rhetoron (master
of rhetors) during the 1080s. He was appointed to the autocephalous
archbishopric of Ochrid, fifth of the Byzantine archbishops, shortly after
his last speech as maistor, in 1088-9. He appears to have been still in his
see in 1108 towards the end of the second Norman War, and a pair of
poems on Symeon the New Theologian is dated to 1125/26.3 His writings
are numerous: best-selling commentaries on the Gospels, the Epistles, the
Minor Prophets, the Psalms, the Acts of the Apostles, as well as saints' lives,
speeches to the emperor, discourses on the liturgy, on Latins, on eunuchs,
fifteen poems and a rich letter-collection of 130 letters. 4 This collection
has been studied to cast light on Byzantine feudalism, taxation, the rural
peasantry, the management of the archdiocese of Ochrid, the use of the
classics in Byzantium, the medieval history of Macedonia and Bulgaria,
patterns of heresy and Greek prose rhythm, 5 and- very latterly- Byzantine
letter-collections. This recent study 6 establishes him as manipulating
a complex network of 127 persons, with strong links into the Komnenian
imperial family and into the intellectual life of the capital, which he exploits
for the benefit of his archdiocese and his clients and friends.
177
Margaret Mullett
Theophylact is a well-known figure? His brother Demetrios is far more

shadowy, known only from Theophylact's works: the letters, two poems
and two works commissioned by him, our text and a work on liturgical
questions. 8 There must have been one other sibling of Theophylact
and Demetrios: George Tornikes was Theophylact's nephew. 9 Not all
commentators are certain that the eunuch brother who commissioned our
text is Demetrios, but it is most convincing and economical to assume
a single brother, Demetrios. 10 This eunuch brother was addressed in
two letters and mentioned in fourteen more, from the very beginning
of the correspondence to very nearly the latest letters which survive. He
was, according to the letters and poems, dear to Alexios I Komnenos
(1081-1118), a client of the ex-basilissa Maria of Alania, a pupil of
Theodore Smyrnaios and possibly of Niketas ho tou Serron, known to the
bishop of Kitros, Nicholas Kallikles, to Niketas the imperial doctor and
to Gregory Kamateros, all close associates of Theophylact. He had the
entree to court (though probably not a member of the clergy of the Great
Church), carried nine ofTheophylact's letters, and his death, probably in
1107, is a central event of the letter-collection. It is dealt with in two letters
as well as in two poems, and prepared for by three letters about a brother's
illness. He died - apparently of consumption - despite Theophylact's
efforts to treat him with oxysacchari; he may be the only recorded casualty
of Byzantine letter-exchange for he bore letters across the freezing Vardar
during a winter which was fraught for Theophylact with accusations of
calumny, and when he needed to mobilize all his friends in support.ll
Theophylact's reactions show the centrality of the relationship as well
as of the event: '
The brother on whom my breath depended, who was refilly everything to
me, who would throw himself in the path of fire and swords so that I could
live relaxed and free from pain
who will contain the violent attacks of the praktors (taxcollectors)? Who will
close the frog-mouths of the sekretika (financial bureaux)? Who will be the
friend of wise judges? Who will be respected by senators for the worth of his
character? To whom shall I reveal the pain of my suffering, now that I no
longer have a doctor for my torment?
I long for an ocean of tears to mourn my brother, who shone before me in
the road fated to darkness. I stumble along the dark way because of his death,
who has suffered extinction. My heart in every fibre is gripped by Niobe's
grief If God were to shape me into stone I would again bewail my suffering.
Time is no healer; suffering lasts and stays fresh. 12 '
This portrait is compatible with the eunuch Theophylact describes as
178
Theophylact of Ochrid's In Defence of Eunuchs
a model of a life of decorum, and his text as the child of a very wise man,
a friend of purity, a eunuch remarkable for his wisdom. 13
The text and Theophylact's other works

Let us look now at the text and its structure.
Theophylact of Ochrid, Logos (In Defence of Eunuchs)

Iambic protheoria: Theophylact, his brother and the audience.
Protheoria: Theophylact, his brother, the gift, the audience.
Setting: Theophylact is in Thessalonike when the emperor is there and hears
two people discussing eunuchs, one criticizing the other, a eunuch, for having
his nephew castrated.
Speech of the non-eunuch: Castration
• is in opposition to the Creator
• against the law of Moses
• against the canons of the Apostles
• against the Fathers
• against Justinian's legislation
• ruins the character of the child
• exposes him to dangerous passions
• in the palace will make him effeminate
• in the theatre will certainly debauch him and in church ruin the liturgy.
Transition: He pauses, gathering thoughts; the eunuch smiles gently (he is the
most charming and cultured of men); his criticisms are of foreign eunuchs not
the friendly Thessalonike variety.
Speech of the eunuch: Castration
• is not in opposition to the Creator
• is not against the law of Moses
• if by loving families in aid of piety is not against the canons or heretical
• has contemporary justification; Justinian was unable to conform to his own
legislation
• is condemned by Photian synod for a reason
• does not ruin the character of all, and not all ruined characters are eunuchs
• in the palace: may expose child to pious empresses
• in the church: many eunuchs live in priestly purity
• in the theatre: eunuchs are outnumbered by non-eunuchs
• in church music: support of David, Ephrem, John Chrysostom.
The Fathers criticized people other than eunuchs.
There are eunuch saints, patriarchs, bishops, priests, monks.
Proof of purity is in its achievement.
179
Margaret Mullett
Conclusion: the eunuch asks if he should go on: no, says the other or I shall
become a eunuch myself... He is not saying that chastity is impossible without
castration, just that the chastity of eunuchs is praiseworthy.
They get up, embrace and kiss; the eunuch takes his (listening) nephew in his
arms and kisses him, pleased with the debate; they part.
Setting: Theophylact does not ask them where they are from (not Thessalonike)
lest he is drawn into the debate. This is what Theophylact has brought his
brother from Thessalonike, not without difficulty, though his memory still works
even in old age.
It is described as a logos in both manuscripts 14 (the title In Defence of

Eunuchs is not found) and it is preceded by two protheoriai (prefaces), the
first one in iambics. Dieter Simon (1994) thought it was an encomium,
and toyed with the idea that it was a progymnasma, a school exercise. 15
It is certainly a strange kind of dialogue. Structurally it has two shells,
the commissioning of the work by Demetrios and the conversation in the
streets of Thessalonike overheard by Theophylact. I have elsewhere seen
the setting as characteristic of the revival of fiction at the time, but this is
not crucial to our present analysis. 16 So we have introductory material, the
speech of the non-eunuch, a transition, the speech of the eunuch, a false
ending in which he offers what I take to be a joke: 'Shall I go on?' he says,
'No,' says the other 'or I'll have myself castrated!' The eunuch resumes his
conclusion, the pair say their farewells and Theophylact brings his present
home to Demetrios. So it is not a dialogue, nor a conversation, more
like a pair of speeches in a disputation, like those Alexia$ used to arrange
against the Armenians or the Latins. 17 But compared with the fragment
of a speech against an Armenian 18 the tone is completely different, much
less aggressive and polemic. The heart of the text is the two speeches, the
first, which is much shorter in which criticisms of eunuchs are laid out,
and the second, the longer one, in which they are refuted. The structure
of the refutation keeps to the structure of the criticisms, but then adds
additional material in a spectacular conclusion.
The speech of the non-eunuch sets out 19 reasons why the eunuch should
not have castrated his nephew. It is against various authorities (sources of
tradition in Orthodoxy) and will ruin the character of the child, opening
him to dangerous passions like cupidity, avarice, ambition, jealousy,
irascibility (a short-list from an ascetic handbook perhaps); contact with
women in the palace will give him feebleness of spirit and vigour of lust,
languour, softness and luxury; contact with other eunuchs in the theatre
will predispose him to indecent talk and encourage him to 'play the man'
in licentiousness. Compared with Basil the Great's list of eunuch vices,2°
180
Theophylact ofOchrid's In Defence of Eunuchs
the non-eunuch does not include greed for food, fickleness or meanness;
other Byzantine stereotypes he neglects are lack of control, lack of courage,
weakness and incapacity for natural affection. Nor does he emphasize
aspects of gender, as did Basil ('neither feminine nor masculine') or Masudi
on eunuch sex ('they are women with men and men with women'). 21 But
then his speech was incomplete when the eunuch began to reply.
The speech of the eunuch is much longer. 22 He responds with vigour,
and in different ways to each criticism. To the charge that eunuchism is
against nature, he offers a comparison with asceticism; to Deuteronomy
23.2 he offers Isaiah 56.3-5; 23 on the canons he claims both that all
depends on the purpose of castration (good in childhood for chastity or
bad in adolescence for contraception) and that a historical perspective is
necessary: castrates are no longer particularly susceptible to heresy, for
example, and now eunuchs are really useful to the state. To the charge that
it is against Justinianic legislation he retorts that Justinian couldn't keep
his own law, and in view of Narses it was just as well; he pre-empts any
use of the Photian synod by pointing out the bad blood between Photios
and Ignatios; the alleged vices, he says, are not worth refuting, and in any
case they are less bad than the special vices of non-eunuchs; eunuchs in
the palace can be influenced by good empresses; if you criticize eunuchs
in the church you are running down Christianity itself; in the theatre,
he concedes that he could add to the complaints - but refuses to accept
that this condemns all eunuchs. In the liturgy eunuchs are not solely
responsible for church music; and he then launches into general responses
both negative and positive, before the sparkling conclusion. In this text
Theophylact is using a whole panoply of argument to delight and console
his brother.
There are constant reminiscences of other works by Theophylact, in
particular the logos on the errors of the Latins written for an ex-pupil. 24
The stress on oikonomia, the emphasis on reading the spirit as well as
the letter of the law, the dismissal of traditional arguments - all except
the filioque (the phrase added in the west to the creed and disputed by
the east) in the one text, or his refusal to credit all the charges of eunuch
vices in the other - show a similar perceptive tolerance, an openness
to difference which he still combines with a sense that he is a stickler
for the canons. 25 In others of his works, however, when not working to
commission or involved in his arguments, and not writing to a eunuch,
he is not so favourable towards eunuchs. In particular his commentary
on the key biblical passage Matthew 19.12 takes a traditional patristic
metaphorical interpretation which excludes actual cutting, 26 and says
explicitly that to become a eunuch for the kingdom of heaven does not
181
Margaret Mullett
mean cutting yourself (for that is accursed) but to be continent. In his

letters he occasionally lets slip pejorative views of eunuchs with whom he
has dealings, or even standard stereotypes. In letter G31 we meet a eunuch
praktor he nicknames Scylla and Charybdis; in G66 at the end of a letter
in praise of the deacon, priest and cantor Gregory whom he would like to
keep in Ochrid if he could get permission from his hegoumenos (abbot),
Theophylact adds that he would hate 'to be deprived of such a man, and,
which is rare among eunuchs, of good character and most candid'. 27 And
his poem 13 is a robust invective, which catalogues a eunuch's crimes in
terms which outdo the non-eunuch's speech in our text: the eunuch is
debauched, promiscuous, a friend of prostitutes, a deflowerer of virgins,
randier than a goat; he is both Priapus and Pan. 28 But his point is not that
all eunuchs are monsters of this kind as a result of the 'change in their
nature' but that this particular eunuch is a monster among eunuchs, who
have a natural gift of purity. 29 We are reminded of G76 where Theophylact
talks of taking iambic revenge against an enemy who had previously
inspired only laments, and we suspect special pleading, perhaps even like
Basil the Great's notorious letter 115 to Simplicia with its list of eunuch
vices. Theophylact maintains throughout an open mind as to the potential
of any human being; he is not however immune from the stereotypes of
his day, or the temptation to use them in pique or anger.
The text, the world and its critics

Theophylact's logos is a remarkable piece, for all the reasons I have
mentioned, and it deserves to reach a wider audience. It has not attracted
a great deal of interest, although the two modern critical editions of Gautier
and Spadaro appeared within a year of one another, and Byzantinists have
taken some time to see its importance in eunuch history.
Its autobiographical status, its dating, and the specific details contained
in it have attracted the most interest, including the light it sheds on the
remarkable career of the eunuch Symeon the Sanctified, who was grand
droungarios of the Watch (a high legal official) at Constantinople under
Nikephoros III Botaneiates (1 078-81), carried out a mission for Alexios
during the revolt of Basilakes, got permission to take the monastic schema
and moved to Mt. Athos with three beardless members of his household,
where he became second founder and hegoumenos of Xenophontos.
Because of the arrogance of his three associates he was thrown off the
mountain by the protos (head of the Holy Mountain) but was reinstated
by Alexios, who sent Symeon Senachereim to force him to come back,
and the protos Paul to reinstate him. He held the unique status of the
only permitted eunuch on Mt. Athos. What our text adds is that he
182
also founded a monastery of eunuchs in Thessalonike as well as ruling

strictly on Athas. 30
D. Simon's study sets Theophylact firmly in a legal context with a view
to discovering the Byzantine concept of norms. He makes two interesting
observations: firstly, that Theophylact may have been trained as rhetor
and lawyer but that he did not have all the texts and further is not very
good at using them; secondly, that Theophylact concentrates much more
on the legal proscriptions than on the insults and vices, where he might
have expected the rhetor to triumphY Only Kathryn Ringrose and Shaun
Tougher have instead set this text in the context of eunuch history; both
are convinced that it is a crucial text, allowing us (should we wish) to
see it as reflecting a major change in Byzantine attitudes to eunuchs by
the twelfth century. 32
Rather than echo this view I shall make two more cautious points. One
relates to the human context of the work, the other to its polemic content
and intellectual context.
The personal is the political

The first point is that this is above all a personal document. Simon realized
this when he was trying to fit it into the genre of progymnasma. It has
a clear commission history, like others of Theophylact's works, 33 but is
rooted here in the most important relationship Theophylact recorded.
Although the text has no manuscript title, it does have two prooimia or
protheoriai, and it explains the prehistory ofDemetrios' request. Demetrios
was annoyed at criticism of eunuchs, and was particularly incensed at the
lumping together of good and bad: what was needed was a judgement '1m't'
avopa', case by case. 34 Theophylact presents himself (in both protheoriai)
as writing the piece originally for Demetrios, but converting himself by
his own eloquence and now offering it to help all and to persuade those
opposed to his viewpoint. It is a doron and a charis for Demetrios and
a paraklesis, a gift, a grace and a consolation. This marks it out clearly
from the text otherwise most comparable, the Discourse against the Latins
commissioned by his old pupil Nicholas, a deacon and kanstresios of
Hagia Sophia, the future bishop of Malesova. These descriptions make
it more comparable to letters, which are regularly described in Byzantine
collections as gifts, as endowed with grace and as consolations. 35 This
reading underlines not only the generic freedom of the text which we
noted earlier, but also its clear interactive status: it is not a treatise to be
read at any point in the future; it is a text written for someone, now.
Secondly, the structure of the text is also very personal. We have seen
that the double shell of Theophylact and Demetrios on the outside
183
Margaret Mullett
and, inside that, the argument of the eunuch and his interlocutor in
Thessalonike, surrounds not a dialogue but a disputation in which each
makes one speech divided by a transition. The argument is personalized:
the point of it is to criticize the eunuch for castrating his little nephew, who
is present throughout, listening keenly and is then kissed over and over
again at the end. The protagonists dispute on a basis of cordial respect:
theirs is the body-language of collegiality, if not greater intimacy, and they
address one another as 9aUJ.HXO"tE, XPTJO"'t6ta'tE av8prov ('wonderful', 'most
excellent of men'). 36 The relationship between the uncle and nephew is
clearly meant to be an example (for he is after all a living refutation of
criticisms of eunuchs), of an idealized picture of responsible parenthood.
In this, in contrast to the young bloods who have themselves cut in order
to make themselves more attractive to the lighter kind of woman, devoted
parents are seen to support their children's choice of eunuchism in zeal
for purity and cleanliness. Eunuchs are seen in a double and very loving
affective literary shell.
Thirdly, this is a very local document. It is set in Thessalonike when
the emperor was there. The balance of probability is that this imperial
presence is not that of the first Norman War in the 1080s, but of the
second, in the 11 OOs. If this is so, Theophylact brought back his agogimon
to Demetrios probably in his see of Ochrid or in his estate at Ekklesiai on
the Vardar, where Demetrios probably died in 1107. 37 The protagonists
are not from the city, but other people in the text are: the metropolitan
of the city, Theodoulos, is revealed as a eunuch, as is Theophylact's other
correspondent the bishop of Kitros. Theophylact wro~e one letter to
Theodoulos (G72) as a colleague and friend, with no trace of the old
rivalries of the sees of Ochrid and Thessalonike after the conquest of
Bulgaria discernible in the sigillia ofBasil II (976-1025). The relationship
with the bishop of Kitros seems much more intimate, a close friend
and confidant to whom Theophylact wrote four letters over the period
1097-1107. They sent gifts to one another, incense and cinnamon, and in
letter G52 Theophylact shared with the other the problems of the passing
through of the First Crusade; while in G 113 and G 121 he confided the
story of Demetrios' illness and death. Also revealed as eunuchs (though
contemporaries could in any case presumably tell) are the bishops of Petra,
a suffragan ofThessalonike, and Edessa, a suffragan ofTheophylact. 38
Finally, another local hero who is mentioned in the text is Symeon the
Sanctified. He appears as the culmination of the recital of holy eunuch-
saints, deacons, priests and bishops: '
There are eunuch monks also, like Symeon, the charming, the graceful, the
prudent, whom we have seen in this town, and who led a community of
184
monks on the holy mountain with all strictness, and who has founded here
(Thessalonike) a monastery of eunuchs. 39
The story is not told; simply Symeon is sketched in to add to the local
colour and the strength of the case.
Given these personal and local features, can this text survive at all as
a document of the Zeitgeist, or merely as evidence for a narrower society:
that remarkable homosocial continuum represented by the complex emo-
tions and relationships ofTheophylact's network with their characteristic
concerns and discourse? 40
Ringrose's case for a developing acceptance of eunuchs in the state and
in some areas of the church is built on her need to find a context for
Theophylact's text. 41 There is, however, a sense in Theophylact that what
he is writing is not going to be accepted by all, and, just as Simon noted
that Theophylact is not fully briefed on the legal casework, so Theophylact
seems rather unaware of the better hagiographical backing for a developing
tolerance of eunuchs in Byzantine society. His saints, apart from Symeon
the Sanctified, are all martyrs of the period of the tetrarchy in the late third
and fourth centuries, and they are largely popularised in the Metaphrastic
menologion, a hugely popular text in aristocratic circles of the eleventh
century. He fails, however, to name names, for example of the patriarchal
eunuch-saints or Symeon the New Theologian, or indeed anyone more
recent or relevant. 42 Perhaps Theophylact's is what it appears to be,
a highly individual account, not written because of a gathering support for
eunuchs in the empire- or contrariwise because of a sudden blow to their
political importance and influence through the advent of Komnenian
family government. 43 Paul Magdalino has suggested recently that eunuchs
may have become less high-profile under the Komnenoi, but that this
does not necessarily mean that they became less useful. The fact that they
appear to make a come-back at the end of the twelfth century under the
Angeloi (a highly suitably named dynasty) would in itself suggest that
they had not yet disappeared from view. Even some of the old sources of
supply of eunuchs might still continue, though continuing reference to
Paphlagonians in the sources should probably not be taken to mean that
there is continuing attestation ofPaphlagonian eunuchs. 44
What should be borne in mind is the noticeable personalization of
all politics and politicization of all personal relationships at the time of
Alexios Komnenos. No emperor knew better how to exploit relationships
of kin, fictive kinship, patronage, or liege-homage. 45 And Theophylact
saw everything, even his own identity, through his network of personal
relationships which he manipulated to stave off the threats of the demosion
(the fisc), to protect his parishioners, to maintain and develop friendships.
185
Margaret Mullett
So for Theophylact to couch a logos in an interactive and affective shell

simply indicates the importance he placed upon it, its status as gift, its own
interactive nature. It means that he was predisposed by his relationship
with Demetrios to take the part of the eunuch, and was - or claims to
have been - convinced by his own advocacy. It does not mean that we can
assume from the existence of this work that there was a current debate
over the place of eunuchs in society, still less a reassignment of gender
roles to include what Ringrose has called 'the ultimate socially constructed
gender category'. 46
Gender-bending, competing angels and nocturnal emission

My second point is that Theophylact's text seems to be less about gender
or even about masculinity than about the relationship between eunuchism
and asceticism, about different routes to agneia and sophrosyne, and
katharotes (chastity, purity, inner cleanliness). And here we may well be
touching on a more topical issue. It is of course entirely characteristic of
Theophylact to avoid the obvious, or to support the losing side, or to
refuse to categorize and to look for the good in all sides. But the text does
deal, in the crucial first and last sections of the eunuch's speech, with the
additional fact that the eunuch is disputing with an ascetic.
Haven't you caused your body to waste away with fasting and neglect of
washing and all the forms of spiritual askesis? Haven't you made yourself
thin rather than in good shape, sallow rather than fresh-complexioned, weak
rather than robust? 47
Alternative models for the third gender are proposed and opposed, and the
result is to cast doubt on traditional asceticism. Recent work by western
medievalists has suggested that clergy were excluded from key masculine
roles such as war and child-begetting, and may have been regarded as
a separate gender. 48 That this possibility is present even in a society with
eunuchs may be seen clearly from Theophylact's outrageous suggestion
that ascetics act equally against nature, but do not have the courage of
their convictions to excise their sexual organs.
You have modified the nature of your sexual organs, since the formation
of sperm is the natural function of testicles which were created for the use
of procreation - but you deplore the production of sperm because of your
love for excellent virginity. If you, being in perfect possession of reason, had
decided not to use your sexual organs for the purpose for which they were
created and then at the right time decided to have them refUoved, you would
not have been worthy of criticism. 49
We can say, can't we, that this transformation of your body has transgressed
186
the laws of the Creator, since he created you to be in good health and you
have transformed yourself into the opposite?
Theophylact's overall point is both issue-dodging, and fundamentally

radical. He refuses to accept the structural arguments, insisting on the
case-by-case approach. That this is not gendered in terms of 'we are all
men' but that there is a genuine anthropology is seen in the section on 'at
the court' where pious empresses we have known make an entrance. There
are good eunuchs and bad eunuchs, good bearded men and bad bearded
men, good empresses and bad empresses. And he saves up the existence of
the bad to make his best case for the existence of choice. 5° The importance
of this is in the context of the patristic view which arose in response to
Origen's bold stroke in castrating himself. 51 St Basil said that 'eunuchs
are doomed to/damned by the knife', and that 'they are chaste without
reward thanks to the knife'. 52 To excise your sexual organs is as Ringrose
has put it, to cheat, to opt out of the core activity of asceticism which is
the agon with demons, the fight against the demons of porneia which is
the sine qua non for the Byzantine holy man. 53 This means that eunuchs
got no credit for chastity, purity, or celibacy, because they did not have to
struggle for it. On the other hand, failures in chastity, purity or celibacy
were seen as par for the course, the natural debauched state of the eunuch.
Theophylact triumphantly seizes on this inconsistency. It is the very
existence of the bad, the debauched, the impure eunuch, which shows up
the achievement of the celibate, the chaste, the saintly. This sophrosyne
(a virtue in the encomium tradition open to men and women, but seldom
to eunuchs 54 ) was chosen by eunuchs no less than by the other genders,
and, says Theophylact, they deserve credit for their achievement. 55
This attitude is not unprecedented, in that not all texts follow the
Basilian line. For example in the seventh century John Moschos' Pratum
spirituale seems on occasion to run counter to the patristic line. The story
of the Cilician priest who had difficulty baptizing beautiful converts
and was miraculously castrated does provide a way out for the future,
a model for the patriarch Methodios among others. This is clarified by the
story of the woman who found that her eyes were tempting men to lust,
and blinded herself with a shuttle: she might also be thought to 'cheat',
though vicariously. 56 But the questioning of the Basilian line is presented
in Theophylact in terms of a comparison between the ascetic life and the
life of a eunuch. Both of course are angelic lives, in the sense that eunuchs
are regularly compared to the angels at the court of heaven, who are in
a similar way 'neither male nor female.' 57
The tendency of the middle Byzantine period, however, seemed to be to
privilege the monks as livers of the angelic life and wearers of the angelic
187
Margaret Mullett
schema. A monastery is where one would expect to find angels at this date.
Theophylact's text shows an ascetic clearly contrasted with a eunuch, and
in the nature of the polemic if the eunuch wins, the ascetic has to lose,
not just the argument, but the primacy of status. If the eunuch is a living
refutation of calumnies against eunuchs, the ascetic shows the limitations
of his way of life; it is no less than eunuchism a life against nature, in
which the body is destroyed for the sake of the spiritual exercises which
allow the approach to God. A pagan, Theophylact says, would be justified
in criticising eunuchs for unnatural practices- but you ascetics? 58
The topicality of this text is remarkable, but again a personal angle
obtrudes. When Theophylact received letter G37 from Symeon the
hegoumenos of the monastery at Anaplous, which told of the death of
the ex-hegoumenos, it made him properly happy - and drew the ribaldry
of his kyklos, well aware of how much he had suffered himself from
scandal caused by the actions of monks. 59 However, other letters show
him struggling to support monastic communities and individuals in all
parts of his archdiocese (as he does a eunuch community in letter G96). 60
And Theophylact's reaction to the death of Symeon's predecessor seems
perfectly proper in view of his anxiety for the fortunes of the monastery
under an abbot who had been a hesychast rather than an administrator.
There is no reason to believe that Theophylact was deeply opposed to
monks. But from the last section of the eunuch's speech emerges a sense
of the limitations of asceticism with a view to chastity, purity and inner
cleanliness: sophrosyne, agneia and katharotes. This, if it does date from
the first decade of the twelfth century, is an early sign ,of the crisis of
confidence in holy men which was to manifest itself in the twelfth century
and lead both to what Beck has called 'hagiographic disappointment'
and to the curious disappearance of the holy man as 'defined by Peter
Brown. 61 Fifty years later, criticisms of extreme asceticism were more
commonplace: Tzetzes satirizes the wandering monks in the streets of
Constantinople, dripping with chains, thrusting their (fake) sores under
the noses of passers-by, Ptochoprodromos reveals the hypocrisies and
class structures of successful monasteries, and Eustathios ofThessalonike
called for reform. 62 Back at the time ofTheophylact, Alexios Komnenos,
like marty emperors before him, forged a close alliance between empire
and the holy: under his mother's influence he took a holy man along
in his baggage on campaign, he cultivated and patronized many holy
men (though, as Armstrong has shown, at the cheapest possible rate and
preferably at his womenfolk's expense); 63 his great reconquest of Anatolia
was marked by the convergence of two important holy men, Meletios from
Myoupolis, present at the profectio bellica by vision, and Christodoulos
188
from Patmos. He took his whole family out on daytrips to his nearest holy
man, St Cyril Phileotes, and he acquired the reputation of an emperor
who was the friend of monks, philomonachos, who treated them with
impeccable consideration and proper body-language, and who was able
on his deathbed to draw on his investment: 'All hermits living in caves or
on mountains or leading their lives in solitude elsewhere were urged to
make long supplications for the emperor.' 64 In contrast Alexios' son John
II (1118-43) assisted his wife to found one very high-profile monastic
establishment in Constantinople, whilst his grandson Manuel I (1143-80)
attempted a monastic experiment at Kataskepe, a kind of All Souls of
monastic superstars, protected by generous grants from the need to exploit
their own estates. 65 There seems to be a need to clean up at least the
wilder fringes of monasticism: there still are stylites on the squares of
Constantinople, but tolerance for the fools (saloi)and migrants (xenoi)
and grazers (boskoi) of a previous age was running short. The biographer
of St Cyril Phileotes was careful to defend him against potential criticism
for lack of stability, for asset-stripping and heavy economic activity, and for
extreme ascetic practices: he begins with heavy flagellation and chains, but
comes off chains and is commended for modesty by his spiritual adviser.
The life frequently commends moderate ambition and records realistically
partial achievement: Cyril is shown as damaged and wounded by his
battles with porneia, which he combats with a red-hot coal, unthinkable in
earlier saints' lives, where the holy struggle terribly but win. At the heart
of the life is a debate with an abbot who accuses Cyril of excess: he wins
by not disputing the charge, but accepting it with humility. A nephew is
advised not to set his sights too high, but he is anxious to beat Cyril in
fasting, and eventually gives up fasting altogether. When people go wrong
in this life they go badly wrong, taking to the hills as bandits or smeared
by association with a Black Mass. And it all looks back nostalgically to
the golden days of the emperor Alexios, the friend of monks. But there is
also a sense that there is a limited pool of talent, that saints are not to be
expected these days, that a moderate approach is the safest. 66
This suspicion of extreme asceticism may account for the success of the
spiritual texts of the monastery of the Theotokos Evergetis under its modest
first founder Paul. 67 The foundation charter of this monastery had been
adopted by the time ofTheophylact's logos into two Constantinopolitan
typika, one of which, Phoberou, also incorporated large passages of a letter
of Paul Helladikos. 68 The Evergetis spiritual anthology had what looks like
a recension for pious laypeople, and was hugely influential in orthodox
monasticism. Their line on fasting, Krausmi.iller has shown, was lax
compared with much contemporary practice. 69 Thus, writing at the time
189
Margaret Mullett
of the last flowering of the traditional alliance between emperor and holy
man, Theophylact echoes the modest aims of Evergetis and prefigures
the revulsion against the punk monks70 of the capital in the mid-twelfth .
century.
He also pulls out of the hat a last great illusion. He finds a way in which
eunuchs have a clear advantage over non-eunuch ascetics, and he makes
this his last major point. It is devastating.
I think that freedom from the flow of semen stands above all for the man
who is a lover of cleanliness, and who does not suffer being polluted by
involuntary and natural occurrences. Enjoying this freedom, we avoid the
scruples of conscience which you suffer even if reason persuades you not
to classifY them as pollutions and not to think that the elevation of the
bulging member is anything, this elevation which the member boasts of
on the grounds that it is the upright member. You will not deny that your
conscience is troubled, and particularly when you are persuaded by the
words of Basil the Great.7 1
Theophylact clearly believes (contrary to most of his contemporaries)

that eunuchs have neither orgasms nor erections. He then presses his
advantage, picks up the idea of willing chastity and the reward of virtue,
and concludes that 'dearly our purity proceeds from a choice which
works along with this condition of our bodies, and in consequence of
this it is rewarded'. A last reminder of the inconsistency which gave him
this leeway:
I myself am a witness, having shown in the matters in which you said many
eunuchs are immoral, that the chaste are pure by choice.
Then the joke, and back to scholarly qualification and almost patronizing
condescension to the stars of the ascetic life:
I am not laying down a law from what I have said about eunuchs that no-one
can be chaste otherwise, for it is possible with many struggles and heightened
abstinence - which are found too seldom in comparison with the crowds of
people who practise priestly celibacy. 72
And the opponent is finally tarred with the brush of intolerance, and
worse, love of the flesh and envy (phthonos). This may say something
about the relations between monks and eunuchs at the time. Two other
sources offer intertextual enlightenment. We can see from the Diegesis
merike, a late twelfth-century account of scandals on Mt Athos in the
reign of Alexios Komnenos, that the existence of bpys and beardless
ones in monasteries was hardly less shocking than the presence of Vlach
shepherdesses. 73 When the typikon ofPhoberou adopted the letter of Paul
Helladikos, in the context of concern about porneia in monasteries, both
190
male and female, it repeated a long story told by the eunuch Eutropius,
a secretary of Juliana Anicia, who entered a monastery but was tempted
by the beauty of his godson who came on visits. Even after his departure
the Devil tempted him with visions of the boy so that his 'member began
to burn with fire and rise in insurrection against me and to exude drops
of filth so that my thighs were befouled with unclean moisture'; God
rescued him and freed him from the temptation which was destroying
him. 74 We can see from these texts that there was a problem for some on
holy mountains in a fear of the seductive powers of boys and eunuchs
in monasteries, hardly less than those of women and female animals:
philosarkia perhaps.
But Theophylact's marshalling of nocturnal emission (or the lack of
it) as the strongest argument for the reinstatement of eunuchs in the
Byzantine gender system of sophrosyne has wider implications than the
strange death of the Byzantine holy man, or the particular hang-ups of
the garden of the Panagia. If patristic scholars and western medievalists
are right to see discussion of nocturnal flux as a focus for debate about
the boundaries of the Christian community, here we see a strong bid for
inclusivity from a position of strength for eunuchs, and, despite ascetic
techniques for control of wet dreams, a serious question-mark placed
beside the place of the ascetic.7 5 But then of course, this argument may
again simply be part of the transactional content ofTheophylact's network,
of crucial significance to his brother, his friends and himself, but bending
no gender-system of Byzantium. All depends on how we read this text.
Notes
1 Theophylact ofOchrid, Logos, in Gautier (ed.) 1980, 287-331; Spadaro (ed.)
1981. The title is not original.
2 See for example Cheney 1995, ch. 7, 99-125, where the author includes
a rough translation of the logos. (I am grateful to Vern Bullough for sending me

this chapter.) The two basic readings of the text are by Ringrose 1999, and
Simon 1994. Ringrose sees it as a positive text which is the culmination of
a changed attitude to eunuchs discernible from the period of iconoclasm in
the eighth and ninth centuries onwards; Simon's is a critical commentary on
Theophylact's handling of normative text. I should like to convey rather more
my impression that this is an attractive, energetic and personal piece of rhetoric,
fully characteristic ofTheophylact's oeuvre.
3 On Theophylact see Mullett 1997.
4 For the commentaries and saints' lives see MPG 123-6; for the discourses,
speeches and poems see Gautier (ed.) 1980; for the letters see Gautier (ed.)
1986.
191
Margaret Mullett
5 See Nikolaev 1951; Xanalatos 1937; Papayanni 1989; Harvey 1993. Praechter
1892; Panov 1971; Obolensky 1948; Katicic 1957.

6 Mullett 1997.
7 He made it into the ninth edition of The Encyclopaedia Britannica 1888,
vol. 23, 277.

8 On Demetrios see Gautier (ed.) 1980, 15-22, and 1986, 59-60; Mullett
1997, 173-7.
9 George Tornikes, in Darrouzes (ed.) 1970, ep. 6, 118.14-16; ep. 10,
128.9-10.
10 See the argument in Mullett 1997, 174.
11 Theophylact, in Gautier (ed.) 1986, ep. 110, To the Doctor of the Emperor
Kyr Niketas, 530-2; ep. 121, To the Bishop ofKitros, 558; ep. 122, To the Bishop
ofDebra, 561.
12 Theophylact, in Gautier (ed.) 1986, ep. 121, 559; Gautier (ed.) 1980, poem
15, 377; poem 14, 369.

13 Logos, iambic protheoria, in Gautier (ed.) 1980, 19.
14 Laurentianus gr. 59-12, fol. 222-8; Vindob.theol.gr. 43, fol. 122v-32r.

15 Simon 1994, 6.
16 Mullett 1997, 76-7; 280.
17 On for example the visit of Peter Grossolano in 1112, or the disputation with
Armenians at Philippopolis in 1114, see Grumel1933; Alexiad 14.8.9.

18 Theophylact, in Gautier (ed.) 1986, ep. G135, From a Letter to Tibanios
the Armenian, 595-7.

19 Gautier (ed.) 1980, 293-5.
20 Basil, ep. 115, To Simplicia. On the negative image of eunuchs in Byzantine
thought see also the contribution of Sideris in this volume.

21 Masudi, tr. Lunde and Stone 1989, 345-6.
22 Gautier (ed.) 1980, 297-329.
23 On these biblical passages see for instance Stevenson in this volume.
24 On the Error ofthe Latins, in Gautier (ed.) 1980, 247-85,.

25 C£ Theophylact, in Gautier (ed.) 1986, ep. G66, To the Chartophylax, 365-6;
G84, To Niketas ho tou Chalkedonos, 441.13; G108, To Makrembolites, 527.3.

26 Theophylact, In Matt. 19:12, MPG 123, 352-3. Ringrose inexplicably hails
this reading as a great leap forward.

27 Theophylact, in Gautier (ed.) 1986, ep. G31, To Kamateros, 233-5; ep. G66,
To the Chartophylax Kyr Nikolaos, 367.40.

28 Theophylact, in Gautier (ed.) 1980, poem 13, 367-9.
29 Theophylact, in Gautier (ed.) 1986, ep. G76, To Mermentoulos, 405; c£
Basil., ep. 115, To Simplicia.

30 Theophylact, in Gautier (ed.) 1980, 329. On Symeon see Petit 1903, 18-29;
Papachryssanthou 1975, 132, no 19; Morris 1995, esp. 279-80.

31 Simon 1994, 25-7.
32 Ringrose 1994 and 1996; Tougher 1997 and 1999. S'ee now also Sideris
in this volume.
33 e.g. Theophylact, On the Error of the Latins, in Gautier (ed.) 1980, 247-85:
192
identification of the pupil through the lemma and a note ofDemetrios Chomatenos,
105; on Maria of Alania as Theophylact's literary patroness see Mullett 1984.
34 Theophylact, in Gautier (ed.) 1980, 297.3 and 299.28.
35 See Mullett 1997, 14, 32-4; and 1981.
36 Theophylact, in Gautier (ed.) 1980, 297.9-10.
37 On the second Norman War see Chalandon 1900; McQueen 1986; Shepard
1996.
38 Gautier (ed.) 1980, 297. On the bishop of Kitros, see Mullett 1997, 184.
Theophylact, Gautier (ed.) 1986, ep. G72, 387; ep. G2, 303-5; ep. Gll, 539;
ep. 121, 559.
39 Gautier (ed.) 1980, 329.
40 On Theophylact's network, see Mullett 1997, 163-222; for the placing of
twelfth-century letter-networks on a homosocial continuum, in the sense of

Kosofsky Sedgwick 1985, see Mullett 1999b.
41 Ringrose 1999, 130-7.
42 Simon 1994, 26-7; on later saintly support see Ringrose 1999, 132-4.
43 See for example various papers in Mullett and Smythe (eds.) 1996; the
fundamental studies remain Stiernon 1961, 1963, 1964, 1965, 1966.

44 Contra Magdalino 1998.
45 Mullett 1999a. On fictive kinship see Macrides 1987; on liege homage see
Pryor 1984; Shepard 1996.

46 Ringrose 1999, 123.
47 Gautier (ed.) 1980, 299.
48 Swanson 1999; Cullum 1999.
49 Gautier (ed.) 1980, 299.
5o Gautier (ed.) 1980, 319.

51 On Origen see for instance Stevenson in this volume.
52 Basil, ep. 115, To Simplicia.
53 Ringrose 1999, 126-30.

54 I am grateful to Shaun Tougher who points out that lgnatios the Younger in
PG 105, 501C does possess sophrosyne.

55 On sophrosyne see Menander Rhetor in Russell and Wilson, 90 and Hill
1999, 182-3.
56 John Moschos, Pratum spirituale, ch. §3, MPG, 87.3, 2853-6; ch. 60, MPG,
87.3, 2912-13. On the patriarch Methodios see Krausmiiller 1999.

57 On the angelic life see Morris 1995, 32; on representations of angels see
Peers 2001. On eunuchs and angels see also the contribution of Sideris in this
volume.
5B Gautier (ed.) 1980, 299.
59 Gautier (ed.) 1986, ep. G37, 253-7.
60 Gautier (ed.) 1986, ep. G96, 491.1.
61 Beck 1959; the model of the holy man proposed in Brown 1971 lay
behind much of Hackel (ed.) 1981, including Magdalino 1981. See now
Howard-Johnston and Hayward (eds.) 2000.
62 John Tzetzes, ep. 104, To Demetrios Gobinos, in Leone (ed.) 1972, 150-2;
193
Margaret Mullett
Ptochoprodromos, III, in Hesseling and Pernot (eds.) 1910, 48-71; Eustathios of

Thessalonike, On Hypocrisy, in Tafel (ed.) 1832, 88-98. See Angold 1994.
63 Morris 1995, ch. 10, 267-95; Angold 1995, 45-72; Armstrong 1996.
64 Nicholas of Methone, Life ofSt Meletios, ch. 27, in Vasilievskii (ed.) 1986,
27.21-28.10; Christodoulos, Testament and Codicil, in Miklosich and Muller

(eds.) 1890, 82, 86; Nicholas Kataskepenos, Bios kai politeia kai merike thaumaton
diegesis tou hosiou patros hemon Kyrillou tou Phileotou, in Sargologos (ed.) 1964;
Alexiad 15.11.9.
65 For the typikon of the Pantokrator monastery see Gautier (ed.) 1974; on
Kataskepe see Magdalino 1993, 298-9.

66 For Cyril's unequal battle with porneia, see ch. 43, in Sargologos (ed.) 1964,
206; for the pivotal debate with the abbot, ch. 29, in Sargologos (ed.) 1964,
127-36; for withdrawal from chains, ch. 16.3, in Sargologos (ed.) 1964, 89-90;
for the nephew, ch. 52, in Sargologos (ed.) 1964, 245-8; for the Black Mass, ch.
53, in Sargologos (ed.) 1964, 249-55.
67 See Mullett arid Kirby (eds.) 1994 and 1997.
68 Tj;pikon of Kecharitomene, in Gautier (ed.) 1985; typikon of Phoberou, in
Papadopoulos-Kerameus (ed.) 1913, 1-88. For the use of Paul Helladikos' letter,
quoted by Dominic Montserrat at the conference, see Thomas and Hero 2000,
vol. 3, 873-4 and Jordan 2000.
69 Krausmiiller 1996.
70 A phrase coined by Michael Angold.
71 Gautier (ed.) 1980, 329.9-15.
72 Gautier (ed.) 1980, 329.25-31.2.

73 Diegesis Merike, in Meyer (ed.) 1894, 163-84, at 170-1: 'Again at another
time c.ertain persons reminded the emperor Lord Alexios saying, "Our holy
Master, even if the monks of the Holy Mountain do not have sheep or flocks, yet
they do have boys and beardless youths." The thrice-blessed and·famous emperor
whose judgement is imperial muzzled them in the words of the Gospel, saying,
"They have Moses and the prophets." They added and answere4 back the emperor
saying, "Holy Master, about beardless youths, the prophets 'and every word of
the Righteous One were relaxed so that 'on account of boys and their scandals
we cannot impose rule in the monasteries because there are more boys than
gerontes.' " And the emperor said, "What ought we to do about the boys? Herod
is dead and I (171) am not turning into a child-slayer, and all the more if you
realize that their mothers are going to come to us lamenting."' This complaint
does not go away, and the emperor is forced to threaten nose-slittings and other
fearsome punishments for monks who continue to complain about the eunuchs
and beardless boys.
74 Tj;pikon Phoberou, ch. 57, in Papadopoulos-Kerameus (ed.) 1913, 8, tr.
D. Montserrat. Note however that John ofPhoberou, following Paul Helladikos,

believes that eunuchs are able to feel desire, have sex and ejaculate; it is simply
that they produce sterile seed. '
75 Brakke 1995; see the application to the medieval west in Leyser 1999. For
an anthropological perspective see Stewart 2002.
194
Bibliography
Angold, M.J.
1994 'Monastic satire and the Evergetine monastic tradition in the twelfth
century', in Mullett and Kirby (eds.) The Theotokos Evergetis, 86-102.
1995 Church and Society in Byzantium under the Komnenoi 1081-1261,
Cambridge.
Armstrong, P.
1996 ~exios I, holy men and monasteries', in Mullett and Smythe (eds.)
Alexios I Komnenos, 219-31.
Beck, H.G.
1959 Kirche und theologische Literatur im byzantinischen Reich, Munich.
Brakke, D.
1995 'The problematization of nocturnal emissions in early Christian Egypt,
Syria and Gaul',Journal ofEarly Christian Studies 3.4, 419-60.
Brown, P.
1971 'The rise and function of the holy man in late antiquity', ]RS 71,
80-101.
Chalandon, F.
1900 Les Comnene, I, Essai sur le regne d'Alexis I Comnene, Paris.
Cheney, V.T.
1995 A BriefHistory of Castration, Fort Walton Beach, Florida.
Cullum, P.H.
1999 'Clergy, masculinity and transgression in late medieval England', in
D.M. Hadley (ed.) Masculinity in Medieval Europe, London and New
York, 178-96.
Darrouzes, ]. (ed.)
1970 Georges et Demetrios Tornikes, Lettres et discours, Paris.
Gautier, P. (ed.)
1974 'Le typikon du Christ Sauveur Pantokrator', REB 32, 83-111.
1980 Theophylacte d'Achrida, I, Discours, traites, poesies, Thessalonike.
1985 'Le typikon de la Theotokos Kecharitomene', REB 43, 5-165.
1986 Theophylacte d'Achrida, II, Lettres, Thessalonike.
Grumel, V.
1933 'Autour du voyage de Pierre Grossolano, archeveque de Milan a Con-
stantinople en 1112', EO 32.
Hackel, S. (ed.)
1981 The Byzantine Saint, Birmingham.
Harvey, A.
1993 'The land and taxation in the reign of Alexios I Komnenos; the evidence
ofTheophylact ofOchrid', REB 51, 139-54.
Hesseling, D.C. and Pernot, H. (eds.)
1910 Poemes prodromiques en grec vulgaire, Amsterdam.
Hill, B.
1999 Imperial Women in Byzantium, 1025-1204: Power, patronage and
ideology, London and New York.
195
Margaret Mullett
Howard-Johnston, J. and Hayward, P.A. (eds.)

2000 The Cult of Saints in Late Antiquity and the Middle Ages: Essays on the
contribution ofPeter Brown, Oxford.
Jordan, R.H.
2000 'John of Phoberou: a voice crying in the wilderness', in D.C. Smythe
(ed.) Strangers to Themselves: The Byzantine outsider, Aldershot, 61-73.
Katicic, R.
1957 'Die akzentuierte Prosarhythmus bei Theophylakt von Achrida', Ziva
antika 7, 66-84.
Kosofsky Sedgwick, E.
1985 Between Men: English literature and male homosocial desire, New York.
Krausmiiller, D.
1996 'The Athonite monastic tradition during the eleventh and early twelfth
centuries', in A. Bryer and M. Cunningham (eds.) Mount Athos and
Byzantine Monasticism, Aldershot, 57-66.
1999 'Divine sex: patriarch Methodios's concept of virginity', in L. James (ed.)
Desire and Denial in Byzantium, Aldershot, 57-65.
Leone, P.A. (ed.)
1972 Joannis Tzetzae epistulae, Leipzig.
Leyser, C.
1999 'Masculinity in flux: nocturnal emission and the limits of celibacy in
the early Middle Ages', in D.M. Hadley (ed.) Masculinity in Medieval
Europe, London and New York, 103-20.
Lunde, P. and Stone, C.
1989 The Meadows of Gold. The Abbasids, London.
McQueen, W.B.
1986 'The Normans and Byzantium, 1071-1112', B 56, 427-77.
Macrides, R.
1987 'The Byzantine godfather', BMGS 11, 139-62.
Magdalino, P.
1981 'The Byzantine holy man in the twelfth century', in Hackel (ed.) The
Byzantine Saint, 51-66.
1993 The Empire ofManuel! Komnenos, 1143-1180, Cambridge.
1998 'Paphlagonians in Byzantine high society', H Bul,;avnvr'l Mucpa Acria
(6o~-12o~ at.), Athens, 141-50.
Meyer, P. (ed.)
1894 Haupturkunden for die Geschichte der Athos Kloster, Leipzig.
Morris, R.
1995 Monks and Laymen in Byzantium 843-1118, Cambridge.
Miklosich, F. and MUller, J. (eds.)
1890 Acta et diplomata graeca medii aevi sacra et profana, vol. 6, Vienna.
Mullett, M.
1981 'The classical tradition in the Byzantine letter', if/- M. Mullett and R.
Scott (eds.) Byzantium and the Classical Tradition, Birmingham, 75-93.
1984 'The "disgrace" of the ex-basilissa Maria', BS 45, 202-10.
1997 Theophylact of Ochrid. Reading the letters of a Byzantine Archbishop,
Aldershot.
196
1999a 'Friendship in Byzantium: genre, topos and network', in J. Haseldine

(ed.) Friendship in Medieval Europe, Stroud, 166-84.
1999b 'From Byzantium, with love', in L. James (ed.) Desire and Denial in
Byzantium, Aldershot, 3-22.
Mullett, M. and Kirby, A. (eds.)
1994 The Theotokos Evergetis and Eleventh-Century Monasticism, Belfast.
1997 Work and Worship at the Theotokos Evergetis, Belfast.
Mullett, M. and Smythe, D. (eds.)
1996 Alexios I Komnenos, vol. 1, Belfast.
Nikolaev, B.A.
1951 Feodalni otnoshenie v pokorenata ot Bizantiia bulgariia otrazeni v pismata
na Teofilakt Okhridski Archiepiskop Bulgarski, Sofia.
Obolensky, D.
1948 The Bogomils. A study in Balkan neomanichaeism, Cambridge.
Panov, B.
1971 Teofilakt Ohridski kako izvor za srednovekovnata istorija na Makedonskiot
narod, Skopje.
Papachryssanthou, D.
1975 Actes du Protaton, Paris.
Papadopoulos-Kerameus, A.l. (ed.)
1913 Noctes Petropolitanae, Petrograd.
Papayanni, E.S.
1989 'Phorologikes plerophories apo epistoles tou megalou Basileiou (329/31
-379) kai tou Theophylaktou (1050/55-1125/6)', in H kathimerine zoe
sto Byzantio, Athens, 391-407.
Peers, G.
2001 Subtle Bodies: Representing angels in Byzantium, Berkeley and Los
Angeles.
Petit, L.
1903 Actes de Xenophon, St Petersburg.
Praechter, K.
1892 'Antike Quellen des Theophylaktos von Bulgarien', BZ 1, 399-414.
Pryor, J.H.
1984 'The oaths of the leaders of the First Crusade to emperor Alexius
I Comnenus: fealty, homage-pistis, douleia', Parergon n.s. 2, 111-51.
Ringrose, K.M.
1999 'Passing the test of sanctity: denial of sexuality and involuntary castration',
in L. James (ed.) Desire and Denial in Byzantium, Aldershot, 123-37.
Russell, D.A. and Wilson N.G.
1981 Menander Rhetor, Oxford.
Sargologos, E. (ed.)
1964 La vie de saint Cyri!le le Phi!eote, moine byzantin (f111 0), Brussels.
197
Margaret Mullett
Shepard, J.
1996 ' "Father" or "scorpion"? Style and substance in Alexios's diplomacy', in
Mullett and Smythe (eds.) Alexios I Komnenos, 68-132.
Simon, D.
1994 'Lobpreis des Eunuchen', Schriften des historischen Kollegs, Vortrage 24,
5-27.
Spadaro, M.D. (ed.)
1981 'Un inedito di Teofilatto di Acrida sull' eunuchia', RSBS 1, 4-38.
Stewart, C.
2002 'Erotic dreams and nightmares from antiquity to the present', journal of
the Royal Anthropological Institute, n.s. 8, 279-309.
Stiernon, L.
1961 'Notes de prosopographie et de titulature byzantines 1', REB 19,
273-83.
178-98.
184-98.
222-43.
1966 'Notes de prosopographie et de titulature byzantines 5, REB 24,
89-96.
Swanson, R.N.
1999 'Angels incarnate: clergy and masculinity from Gregorian reform to the
Reformation', in D.M. Hadley (ed.) Masculinity in Medieval Europe,
London, 160-77.
Tafel, T.L.F. (ed.)
1832 Eustathii Opuscula, Frankfurt-am-Main.
Thomas, J. and Hero, A. C.
2000 Byzantine Monastic Foundation Documents, 5 vols, Washington D.C.
Tougher, S.
and origin', in L. James (ed.) Women, Men and Eunuchs: Gender in
1999 'Images of effeminate men: the case of Byzantine eunuchs', in D.M.
89-100.
Vasilievskii, V. (ed.)
1986 'Bioi Meletou tou neou', Pravoslavnyi Palestinskii Sbornik 6.2.
Xanalatos, D.
1937 Betrage zur Wirtschafts- und Sozialsgeschichte Makedoniens im Mittelalter,
hauptsachlich auf Grund der Briefe Theophylakts von Achrida, Munich.
198
11
EUNUCHS IN THE LATE BYZANTINE EMPIRE

c. 1250-1400
Niels Gaul
In spite of the increasing interest in the function and fate of eunuchs*

in Byzantine society, 1 our perception of the role of eunuchs during the
last centuries of Byzantium still relies largely on Rodolphe Guilland's few
examples which he lists to support his observation that ' ... it is probable
that their influence on politics and government was less important than
earlier?
Guilland's argument is twofold. On the one hand, he argued that under
the Palaiologoi 3 Byzantine society became more and more 'westernized';
the Byzantine aristocracy accepted Latin customs and, accordingly, Latin
prejudices. In this context of western values castration eventually turned
out to be a fault rather than an advantage: the eunuch came to be
considered as 'an inferior being' .4 On the other hand, Guilland emphasized
that the emperors of the house of Palaiologos had much less reason to
fear being overthrown than their predecessors of the ninth to eleventh
centuries, as by now dynastic thinking was securely established. 5 In short,
there was no further need to employ eunuchs as a 'lubricant' to neutralize
over-ambitious aristocrats. 6
There is, in fact, little evidence to challenge Guilland's general verdict:
under the Palaiologoi, eunuchs were less 'important' than in earlier
centuries. However, the recent completion of Erich Trapp's prosopography
of the Palaiologan period provides more concrete data for our examina-
tion? Among the c. 30,000 Byzantines registered in the prosopography
we encounter about fifteen individuals unmistakenly known to have
been eunuchs (see Appendix for an overview). Certainly there were more
eunuchs altogether, at least at the imperial court, as can be concluded from
phrases like 'one of the [emperor's] eunuchs' (n~ ·nov Euvouxoov) or 'and
many other eunuchs' (Kat 1toA.A.rov E:-c£poov Euvouxoov) 8 - although we are
not given any information about their specific titles or functions.
*The term 'eunuch', in spite of its broad meaning in late antiquity and thereafter, here
refers only to males who were castrated on purpose, not to eunuchs 'by nature'.
199
Niels Gaul
Here I shall merely attempt to offer some adjustments, especially to the

first part of Guilland's argument; that we are less likely to find eunuchs in
a distinguished position does not necessarily imply that the institution of
eunuchism itself became despicable to the Byzantines. On the contrary,
considering the fundamental changes the Byzantine governmental system
had undergone between the days of the Macedonian dynasty and the
Palaiologan period, the decline of political eunuchism must not strike us
as particularly surprising. The Komnenian invention of 'extended family
government'; 9 the Laskarid 'day-to-day household government' evolving
during the days of the Nicaean exile; 10 the Palaiologan government, heir
of the Nicaean court on a higher scale; 11 finally, the changing function and
increasing importance of aristocratic women from the eleventh century
onwards, 12 all this no longer discouraged the members of the imperial
family and ambitious high aristocracy from taking part in political affairs,
but, on the contrary, encouraged them to share in the exercise of imperial
power. Under these circumstances it seems only natural that those high-
ranking offices and dignities once reserved for eunuchs were by now, if still
in existence, granted to close relatives or oh:etot 13 of the emperor.
This chapter will briefly consider the origins of late Byzantine eunuchs,
discuss their place in society and at the imperial court and, finally, glance
at fourteenth-century attitudes towards eunuchs.
Origin and place in society

The information that can be gained about the origin of eunuchs of the
Palaiologan period is well-nigh nil. 14 The absence of family names in the
case of most of our samples, and the presence of otherwise unattested
surnames like Eonopolites (APP., i) and Kallikrenites (APP., ii and iv),
may indicate a rather low social background for Palai~logan eunuchs.
The appearance of two Kallikrenitai in two successive generations is
a puzzling phenomenon, and certainly it would be tempting to assume
that the uncle's fortune made a nephew succeed in his steps. At least the
latter, John (APP., ii), seems to have accumulated considerable wealth:
he was granted possession of the Vassos monastery in Constantinople
under the condition that he would finance its restoration. 15 The only
eunuchs from otherwise well-attested families are George Pepagomenos
(APP., ix) 16 and John (APP., xii), the uncle of the well-known polyhistor
Nikephoros Gregoras. Both of them achieved fairly high-ranking offices
in the ecclesiastical hierarchy.
With very few exceptions, late Byzantine eunuchs seem to have occupied
lower ranks in the patriarchal or imperial household. This is confirmed
first and foremost by those court officials whom we encounter among our
200
Eunuchs in the late Byzantine empire c.l250-1400
samples and in historiographical sources (c£ .APP.). None of them- except

for the J.LEya~ 8ou~ (of Trebizond, APP., viii),l? whom Pseudo-Kodinos
placed seventh in the eighty positions of his list of precedence -was of
a particularly high rank. 18 We find further evidence in a Trapezuntine
horoscope, compiled for the year 1336 by an anonymous writer. 19 Among
the addressees, the author placed eunuchs in a middle rank and dose to
the noble women, observing the following order: emperors; grandees;
grammarians and notaries; prelates and clerics; court officials and com-
manders of the army; abbots (yepov'tE~) and eunuchs; noble women;
businessmen and merchants; envoys; actors; simple folk and people
of the marketplace. For the moment, he says, the eunuchs shall face
opposition. But after a while they will experience better treatment, well-
being (KaA.rocruv'll), fellowship (oiKEirocrt~, presumably with the emperor)
and assignments from the imperial court (evepyEtat EK 'tffiv ~acrtA.tKffiv
auA.ffiv). 20 In September, obstacles and humiliation are supposed to occur;
in November, both eunuchs and women are said to feel grief and sorrow; in
January, however, a glorious time is foretold. 21 Finally, there are a few lines
in a thirteenth- or fourteenth-century vernacular romance, Kallimachos
and Chrysorrhoe, which might suggest that eunuchs were still organized
in a strict hierarchy and obeyed the same strict ceremonial as in earlier
centuries: 'They walked in, they bowed down to earth, they left at the
same moment, as was their custom, in an order according to their ranks
(Ka'ta 't'l)v 'tU~tv 'tOU'tffiV). ' 22
This obvious presence of eunuchs at the fourteenth-century imperial
court is silently contradicted by their almost complete absence from
the only detailed treatise on court ceremonial which has come down to
us from the Palaiologan period, commonly known under the name of
Pseudo-Kodinos. Pseudo-Kodinos makes, unlike Philotheos in the ninth-
century Kletorologion, 23 no special mention of offices, dignities, or dress
reserved for eunuchs, thus implying a rather open-minded attitude towards
them at court. Obviously, the intention and the need to distinguish them
from their fellow-courtiers who were 'normal' men had vanished. Thus
we can only speculate whether some of the minor offices in his list of
precedence, particularly some of those closely connected with personal
attendence on the emperor or the empress, were still traditionally held
by eunuchs. 24
The case ofStjohn bishop ofHerakleia 25

Only in the- certainly not representative- case of StJohn (1249-1328)
do we learn the reasons for the child's castration. They are described
by his nephew Nikephoros Gregoras in a work written in praise of this
201
Niels Gaul
saintly eunuch-uncle and teacher, the Life of Stjohn of Pontoherakleia

(BHG 2188): 26
As, however, a terrible disease befell the new-born babe immediately after
his birth, demanding to take away either his genitals or his soul, the parents,
after having compared the one with the other and carefully measured the
outcome, decided on the less arduous way for the child, so that since he
would, after all, be henceforth no good passing his life at home, he might be
of use for the women's quarters of the imperial palace. 27
As far as methods of castration are concerned, Gregoras later tells us that
his uncle was castrated by means of 'an iron' (crUiT\poc;). 28 We are not told,
though, which parts of John's genitals were affected by the operation,
nor in what way. It is worth comparing this account with some earlier
anecdotes, as, for example, related in the Lives of the na-rpixtoc; Niketas
(BHG Auct. 1342b), of Nikephoros, bishop of Miletos (BHG 1338),
and of Constantine, son of St Metrios (BHG 2272), none of whom
were affected by any kind of disease. However, in Gregoras' view, John's
disease seems to be the only - or perhaps the most - acceptable excuse
for the castration. 29 Surely one must not forget that John sprang from
a rather well-off family, and that his parents' decision was, allegedly, not
influenced by financial needs. But both Niketas' and Nikephoros' parents
were moderately wealthy. 30 Gregoras' eagerness to ascribe John's castration
to the latter's disease may well indicate that for a noble familf 1 it was
no longer, at least not officially, acceptable to have a eunuch reckoned
among its members, although by the twelfth century, temporarily at least,
the purposeful castration of boys had become socially acceptable. 32 This
apparent change might be due to western influences - or simply derive
from the fact that castration was no longer a prerequisi~e for acquiring
influence at court. Those middle-ranking offices at court not reserved
for the emperor's family and his oiK£tOt were by now, it seems, open to
castrated and 'normal' men alike. On these grounds, it is impossible to
decide if late Byzantine eunuchs were merely a result of bad fortune, or if
they continued to be produced for the demands of the court.
Later on, in a chapter dedicated to the physical effects of John's castra-
tion,33 Gregotas' attitude clearly suggests that, in his opinion, his uncle,
although a eunuch, was as much a proper man as any man. For him,
John's castration means in no way that John was, from his childhood
days, destined to live a life of celibacy. 34 On the contrary, the boy is
especially likened to his non-castrated peers at the imperial court, that
is, most probably, to the numerous Nicaean pages called apxov'tonouA.ot
and nmMnouA.a. 35 He had to face carnal desire as any other saintly male
202
Eunuchs in the late Byzantine empire c. 1250-1400
youth, and had to overcome it successfully (actually, he was the only one
who succeeded in subduing it). Another, quite important, conclusion
that seems possible from this passage is that eunuchs were no longer
brought up exclusively 'in the shadows' of the women's quarters but in
a more relaxed atmosphere, and had considerably closer contact with
the world beyond. 36
On the other hand, prejudices against eunuchs seem to be concealed
under a rather thin surface; and Gregoras, albeit grudgingly, obviously feels
obliged to defend his uncle against the charge of effeminacy, against those
who doubt that eunuchs feel 'heterosexual' desire. He cleverly does so by
turning the argument commonly used against 'effeminate eunuchs' to his
uncle's advantage: if eunuchs are more womanlike, yuvcnKroOEcr't£pov, 37
it is even more difficult for them to achieve saintliness; 38 accordingly his
uncle deserves even higher praise.
Despite Gregoras' attitude, the social network surrounding the castrated
holders of offices at court and in church, namely George Pepagomenos
and Michael Kallikrenites, does not allow us to conclude that eunuchs in
distinguished positions were generally perceived as outsiders. Both received
flattering letters and poems, from the (notorious) epistolographer Michael
Gavras and the court poet Manuel Philes respectively39 - a profile likely to
be shown also by non-castrated officials. In his poems, however, Manuel
Philes could not refrain from hinting repeatedly at George Pepagomenos'
eunuchism. In the last line of a poem praising the foundation of a church
of St George, the personified church itself makes a pun on the donor's
physical condition, putting d)vou~, 'well-minded', instead of the expected
£uvouxo~, and employing the rather rare term 'tO!J.ta<; ('eunuch'). 40 Another
poem41 addresses Pepagomenos in his function as !J.Eya<; EKKA1')crtapx1')~;
the writer obviously seeks permission for two aged people to get married
(vv. 9-10). 42 The addressee is referred to as an 'unmarried man' (literally:
'virginal man', nap8£vo~, v. 4) and as 'avoiding yourself any wedded union'
(<j>£uyrov 0£ 'tijv crus£u~tv) (v. 3); his inability to procreate is paraphrased
euphemistically (v. 6): 'you who do not intend to produce offspring'
('t£K£tv ouMv 8£A.rov). Again the poem culminates in a pun, this time
on the receiver's surname (vv. 7-8): although Pepagomenos bears the name
of the bitter cold (Pepagomenos derives from the verb nayoro/-ffivro, 'to
freeze'), he kindles the flame that suits weddings. It is certainly not without
irony that a eunuch was in charge of marital law at the church of Hagia
Sophia, and we may imagine Philes choosing his words rather carefully.
Still, the mere fact that the poet alluded to Pepagomenos' eunuchism twice
allows us to conclude that the latter felt quite comfortable with it.
203
Niels Gaul
Eunuchs at the imperial courts

As far as eunuchs at the imperial court are concerned, they partly seem
to have retained their well-acknowledged function as messengers (media-
tors).43 Michael VIII (1259-82) sent 'one of his eunuchs' to arrest the
11£ym; 1tPtlllltKTJptoc; Kassianos, who was charged with high treason. 44 It
was also Michael who was the last emperor to entrust a eunuch with
the command of an army, namely Andronikos Eonopolites (APP., i).
During the first civil war, Andronikos II (1282-1328) repeatedly sent
Michael Kallikrenites to his grandson, Andronikos III (1328-41). 45 In
these tense situations, it was the eunuch who twice suffered from a nervous
breakdown, not his companion Theoleptos bishop of Philadelphia. Thus
Michael - or rather Kantakuzenos' depiction of Michael46 - contributes
to and reinforces the common prejudice against the weak and effeminate
character of eunuchs. Finally, Andronikos II let the monk Neophytos
(APP., xiv) accompany Patriarch Athanasios II of Alexandria on a marriage
embassy to Cilicia on behalf of his son, Michael IX.
There also seem to be considerable hints at a special relationship between
empresses, or female members of the imperial family, and eunuchs. St
John of Herakleia started his career as a servant in the empress' chambers
and was later a novice of the empress' spiritual fatherY John Kallikrenites
(APP., ii) was an oixetoc; of the empress and accordingly rewarded by
her. Philialetes (APP., v) was given a piece of land by an otherwise
unknown 1tpro'to~ecrnapw who was the emperor's aunt. The only mention
of eunuchs in Pseudo-Kodinos' list of precedence draws a connection
between eunuchs and the empress: during the coronatiqn ceremony the
empress leans either 'on two relatives of the closest kind (cruyyevrov 'tOOV
YV'Ilcrtro'ta'trov) or, if she has none, on two eunuchs'. 48 The superlative here
employed, on the one hand, indicates that eunuchs were still considered
to be trustworthy servants, while, on the other hand, very close relatives
were understood as sexually above suspicion. This text originally referred
to the coronation of Anna of Savoy by her husband, Andronikos III, in
1327 - this Italian princess was indeed unlikely to have had 'very close
relatives' with her. We encounter another hint at political eunuchism in
the environment of the same empress, Anna of Savoy, in the third book of
John VI Kantakuzenos' history, where the usurper describes how adherents
of the empress - at the time both dowager empress and John's major
antagonist in the second civil war (1341-7)- tortured his beloved mother
Theodora to death; her immense wealth was confiscated immediately. 'So
huge was the amount of treasures discovered', writes Kantakuzenos,
that, although many of the empress' [Anna's] men shared it in the first
instance among themselves- one part was kept by [Alexios] Apokavkos, the
204
Eunuchs in the late Byzantine empire c.I250-I400
!J.Eym; 801)~, another by Uohn] Gavalas, the 1tpro'tocr£~acr'to~ .. .-and [again

another] by [Manuel] Kinnamos the !J.UO''tn.:6~ [private secretary], who was
then the 'ta!J.ia~ [treasurer] of the imperial treasuries, and by many other
eunuchs, who had great influence on the government (o'i 1tapE8uvacr'tEUov),
[so huge was the amount of the treasures that] they could nonetheless wage
war against the emperor [the writer, Kantakuzenos, himself] just by paying
for their needs from what was left. 49
This passage is certainly meant to be highly polemical towards John VI's

enemies, likening 'real men' in a rather ambiguous phrase to eunuchs, and
eunuchs to morally despicable 'real men' .5° Kantakuzenos' commentary
reminds us of the special relationship between empresses and eunuchs
so often hinted at in earlier centuries; 51 he certainly makes his point for
reasons of political propaganda. One could reasonably argue, however, that
eunuchs whose influence went far beyond the empress' privy chambers
stirred the reader's suspicion, were meant to indicate the weakness of
Anna's government, and were certainly not well appreciated. The rather
old prejudice concerning greedy, selfish and arrogant behaviour of court
eunuchs comes to the surface.SZ Thus it seems of some interest that of the
few mentions of eunuchs in the fourteenth century two are connected
with Anna of Savoy, who was the only woman regent we encounter in the
entire Palaiologan period and, as has been pointed out, came from abroad:
presumably a foreign empress, being thrown into Byzantine court society,
to maintain her rightful place had no choice but to rely on trustworthy
eunuch attendants.
In another passage in the same history, in the account of the marriage
of John Kantakuzenos' daughter Theodora to the Ottoman Emir Orhan
at Selymvria in summer 1346, 53 eunuchs figure in the princess' retinue:
Md torches were fastened all around her, which were invisibly held by
eunuchs who would bend down on their knees.' 54 Notably these eunuchs
occur at a stage when John had already crowned himself emperor at
Adrianople (21 May 1346), but not yet gained access to the imperial
palace of Constantinople (8 February 1347): 55 must we assume that he
hurried to acquire eunuchs to emphasize his own and his daughter's
imperial status? Or did eunuchs still serve in aristocratic households, and
not only at the imperial court? For the time being these questions cannot
be answered satisfactorily. 56
Eunuchs in vernacular romances

Different attitudes toward the gender of eunuchs are presented in three
of the five chivalric vernacular romances written within the boundaries of
the Byzantine empire during the thirteenth and fourteenth centuries. 57 We
205
Niels Gaul
encounter eunuch attendants in Velthandros and Chrysantza, Kallimachos

and Chrysorrhoe, and Livistros and Rhodamne. Prominent already with
the ancient novel-writers as a motif of slavery and daintiness (describing
the Persian court), 58 they appear to have been re-introduced into the
Byzantine romance by Constantine Manasses towards the end of the
twelfth century. 59 In his Aristandros and Kallithea a viper (£xtova), the
mother of venom (<j>apJlaKOI.t:r'rcrop) that once had bitten a eunuch, burst
herself when the eunuch turned out to be more poisonous than the snake:
'The viper tasted blood much more poisonous than hers, [blood] that
completely overpowered her death-bringing venom.' 60
As it is reasonable to assume that these romances circulated first and
foremost among the learned circles of the court, that they were intended
to meet a courtier's taste, and accordingly reflected to a certain, 'semi-
realistic' degree the customs and habits at court, one may with caution
attempt to exploit the mention of eunuchs in these romances for our
purposes. 61 Thus in Kallimachos and Chrysorrhoe, 62 eunuchs receive the
same bad press from which they had suffered in Constantine Manasses'
romance. At some point in the story, the king of a neighbouring country
abducts Chrysorrhoe and puts her under the guard of three evil eunuchs in
his palace. Eventually Kallimachos succeeds in finding his wife and reveal-
ing himself to her. He enters the king's service in the guise of a gardener,
while Chrysorrhoe, in order to spend her nights with Kallimachos,
craftily claims that her condition would be much improved by sleeping
unattended in the garden. Mter a while, however, the three eunuchs
discover Chrysorrhoe's ruse, and send a letter to their king who happens
to be on campaign, urging him to take some action against the girl. The
king, upon receipt of the letter, utters a tirade against them which reads
as follows (vv. 2283-9): ·
Perhaps some are angered and urge the cutting off of hair and the putting
on of black clothes, the refraining from flesh, the moaning. 63 And especially
the y£vo~ of eunuchs, this y£vo~ of double sex, or rather, of no sex at all,
is insidious and envious making up such lies as do make me angry and stir
me to scorn and wrath towards the girl, and so they free themselves from
tempting me to put them in black clothes.
Finally the king's lamentation culminates in a pun which makes use of

the homophony of the Greek word for 'well' (~::u) and the first syllable
of 'eunuch' (£1JVouxor;), adding the contradictory term KaK6t; ('bad') to
the element £U- (vv. 2291-4):
But how, again, can they [the eunuchs] dare to turn to such a lie, to
write a thing against nature itself? That now she has become a whore, my
206
gold-gleaming maiden? - They lie, the wicked eunuchs! ('PEuoovmt oi

lCUKEUVOUXOt.)
More generally, these three eunuchs seem to frighten the court: after they
have thrown the unhappy Kallimachos into prison, the frightened senior
gardener will not answer Chrysorrhoe's question about Kallimachos, for
fear of the eunuchs (vv. 2341-2). The king's cry: 'this y£voc, of double sex,
or rather, of no sex at all' is certainly to be connected with the old dilemma
of where to place eunuchs on the Byzantine ladder of gender hierarchy:
in the middle, at either end, or at both? 64 It is sufficient demonstration
that the oldest perceptions of eunuchs and their gender were still alive
in the fourteenth century.
However, the bad press eunuchs received in Kallimachos and Chrysor-
rhoe is counter-balanced by their appearance in the other two novels.
In Velthandros and Chrysantza we meet an (unnamed) court eunuch
functioning in a way quite similar to that assigned to Michael Kallikrenites
by Andronikos II during the first civil war. Velthandros' father, Rodophilos,
~a<nA£UC, of the Romans ('the Byzantines'), who once expelled his own son
from the realm, feels death approaching and sends a eunuch messenger
to bring Velthandros back. This eunuch seems to be an old and trusted
member of the imperial household; although his tide in the court hierarchy
is not specified, it is to him and not the captain (KOilTJC,) of the ship that
Velthandros directs the final account of his adventures (vv. 1224-304).
Rhodamne's eunuch page in the Livistros romance is the only eunuch
character assuming so important a role that he is introduced to the reader
by name, Vetanos. 65 This young eunuch (he is repeatedly referred to as
'little eunuch', £uvouxonou/..oc,) is praised for his exceeding beauty as
well as for a fair share in his lady's words, secrets, and ruses (49-51).
In a situation quite similar to that of Chrysorrhoe, Vetanos - although
set to guard his lady's seclusion by her father himself, Chrysos ~a<nA£UC,
of Argyrokastron - helps Rhodamne and Livistros to stay in contact,
playing a key role in the exchange of secret letters (40-1053). 66 Warning
Rhodamne against the powers of Eros in the beginning (325-38), her
longing for Livistros subsequently moves him to pity. He alerts his lady to
the fact that Livistros' letters are dedicated to her, not one of her ladies-
in-waiting (394-9). It is also he who urges her to reply to these letters
(570-4), and finally even to exchange rings with Livistros as tokens of
their mutual love (765-8). For some reason Vetanos, who is befriended by
one ofLivistros' men (58, 248, 476-81), appears to be very eager to serve
Livistros right from the beginning, addressing him first in a refined letter
(461-75, accordingly we may conclude that he was properly educated)
207
Niels Gaul
and meeting him face to face later on (989-1027, 1092-5). Livistros,

albeit of royal blood, greets Vetanos like a peer rather than an 'inferior
being', holding his hand, 'kissing him ten thousand times', embracing
him, and thus acknowledging Vetanos' outstanding role as 'mediator'
(f!EO"asrov av8pro7tO~, 10.23) of the developing relationship with Rhodamne.
Finally, he rewards him richly according to custom. In Livistros' opinion
'the whole y£vo~ of eunuchs is very fond of flattery (KoA.aKda), and
is likely to get involved in the subject of love' (1 002-3). Thus the
Euvouxo7touA.o~ assumes the more traditional role of a mediating courtier,
utterly trustworthy and amiable, as well as the more innovative function
of a helper of Love herself, transgressing not only the borders between his
lady and Livistros, but also between the two lovers and the supernatural
power of Eros.
Conclusion
Eunuchism featured at the Byzantine imperial and patriarchal courts, and
possibly in aristocratic households as well, at least up to the beginning of
the fifteenth century. Undoubtedly, however, eunuchs were less numerous
and - accordingly, one might assume - less influential than in earlier
centuries.
The reasons for this decline cannot solely be attributed to Guilland's
supposed 'westernization' of Byzantine society. This argument seems
valid only in so far as the imperial court as a whole - first under Manuel
I Komnenos, 67 and especially after 1204- increasingly resembled a western
court. The way of conducting the empire's government undeniably
changed, but this development, one would think, was independent of
an individual emperor's attitude towards eunuchism. On the contrary, in
the few cases where criticism occurs, it draws very much upon the same
prejudices against eunuchs' gender and character as were detectable in
earlier centuries, not betraying any particularly western derivation (where,
supposedly, such a systematic criticism could hardly have developed,
considering the absence of regular contact between Latins and eunuchs).
The only hint at a possible influence of western ideas, judging from
Nikephoros Gregoras' Life of Stjohn, is the fact that it was no longer
desirable for a well-off family to reckon a eunuch among its members. On
the other hand, the silence of Pseudo-Kodinos on distinctive features of
their outward appearance as well as the rather 'open-minded' upbringing
of StJohn at the imperial court ofMichael VIII may suggest that eunuchs
were much more integrated into the 'normal' court society than hitherto.
Nor does Guilland's second argument, concerning dynastic stability,
provide sufficient explanation; it does so only in so far as a desire for
208
dynastic security supported the development of a governmental system

based upon the imperial family.
Thus the most obvious and rather 'internal' reason for the decline of
eunuchism remains the fact that from the eleventh and twelfth centuries
onwards the more influential and prominent offices and duties formerly
reserved for eunuchs happened to be fulfilled partly by relatives or oh:Etot
of the emperor, partly by the women of the imperial and aristocratic
families who had strengthened their position in the household administra-
tion.68 Eunuchs remained, however, a distinctive feature of Byzantine
court society, as may be concluded from their occurrence in historio-
graphy as well as in vernacular romances. Their mostly favourable or, at
least, indifferent treatment in our sources suggests that eunuchs of the
Palaiologan period acted in accordance with what they were expected to
do: to serve sufficiently and silently in the more private quarters of the
imperial, and possibly the aristocratic, houses, and that for this kind of
service they were well respected and even appreciated.
APPENDIX
1. Eunuchs known by name in the thirteenth to early fifteenth
centuries
1.1 Eunuchs at court
1.1.1 Constantinople
[i] Andronikos Eonopolites (Avop6vn:oc; 'Hovo7tOAt't1'Jc; ('Irovo7tOAt't1'Jc;?),
PLP #6713): llEyac; opouyyapwc; in Constantinople (1286-9),
'ta'tac; 'tfjc; auA-f\c; (1280/1); recommended the later patriarch
Athanasios to Emperor Andronikos II.
[ii] john Kallikrenites ('Iroavv11c; KaUucp11Vt't1'Jc;, PLP #I 0370): otKEtoc;
of the Empress Helene Dragas, wife of Manuel II Palaiologos
(about 1400).
[iii] Karvas (Kap~ac;, PLP #11145), died mid-December 1291(?):
servant of Andronikos II in Constantinople; was murdered while
he tried to prevent Michael Komnenos Angelos, heir to the
domain ofThessaly, from fleeing.
[iv] Michael Kallikrenites (Mtxm1A- KaUtKP11Vt't1'Jc;, PLP #10371):
7tpOKa8it!levoc; "COU Komovoc; in Constantinople (1321-30/1),
1tavcr£~acnoc; cr£~acr'toc; and otKetoc; of emperor Andronikos II,
c. 1330. Sent to Andronikos III thrice during the civil war; later
the patriarch sent him on a diplomatic mission to Armenia and
Cilicia.
[v] Philialetes (<l>tAtaATt"C'llc;, PLP#29822), died before 1401; founder
209
Niels Gaul
of the Theotokos Amolyntos church in Constantinople on a piece

of land which he was given by an otherwise unspecified npwco-
~ecrnapta.
1.1.2 Trebizond
[vi]? Theodore (the) Eunuch (9t::63ropo~ Euvouxo~, proper name?, PLP
#6305): 'ta'ta~ at the imperial court (1386).
[vii]? Therianos (the) Eunuch (9eptavo~ Euvouxo~, proper name?, PLP
#6306): 'ta'ta~ at the imperial court (1381).
[viii] john ('Iroavv11~, called ~::uvouxo~ or EK'tOf.!ta~, PLP #8597), died
March 1344 in Limnia near Trebizond: f.!Eya~ oou~ ofTrebizond
(1332-March 1344), founder of the LOO'tTJP 'tOU <1>apou monastery
near Trebizond; he became a member of the senatorial faction
and kept Michael Komnenos, emperor ofTrebizond, prisoner in
Limnia; was killed when the latter was freed.
1.2 Eunuchs in the church
[ix] George Pepagomenos (r~::oopyw~ Ilt::nayrof.!EVO~, PLP#22357): f.!Eya~
EKKAll<Jtapxll~ ofHagia Sophia in Constantinople (1326/7).
[x]? Eunouchos (Euvouxo~, proper name?, PLP#6304): hymnographer
and npro'tO\jfUA'tll~, monk of the Philanthropenos monastery
either in Constantinople or in Thessalonike (fifteenth century
or earlier).
[xi] Theodosios (e~::oMcrta~, PLP #7176): monk of the Myrelaion
monastery in Constantinople (1315).
[xii] ·john ('Iroavv11~, PLP #8609): born 1249 close to the River
Parthenios/Pontos, died 1328; brother of Nikephoros Gregoras'
mother; archbishop of Pontoherakleia (1295-1328).
[xiii] joannikios ('IroavviKto~, PLP #93651): cleric (end of thirteenth/
beginning of fourteenth century).
[xiv] Neophytos (Nt::o<l>mo~, PLP#20162): monk, ambassador.
[xv]? Theodore (e~::ooropo~, PLP #7477): of Slavonic origin, called
'beardless' (possibly a eunuch?); writer (autobiography) and
ypaf.!f.!a'ttKO~ on Mt Athos (1242/3-65), expelled from Mt Athos
in 1263 but called back half a year later; a homosexual.
[xvi]? joannikios (the) Eunuch ('IroavviKto~ Euvouxo~, proper name?,
PLP #91897): tepOf.!Ovaxo~, fourteenth century.
1.3 Unspecified cases
[xvii] Xeros (811p6~, PLP#20916); mentioned 1285.
[xviii]? Eunouchos (Euvouxo~, proper name?, PLP #6303); landowner in
Melitziani/Strymon, c. 1341.
[xix] Chresimos (Xpilcrtf.!O~, PLP #30991); owner of a house in Con-
stantinople, before 1342.
210
2. Eunuchs who are 'nameless' or fictitious

2.1 Historiography and ceremonial literature
[i] George Pachymeres, Relations historiques, ed. A. Failler (Paris,
1984-2000).
[ii] John Kantakuzenos, Historiarum libri IV, ed. L. Schopen (Bonn,
1828-32).
[iii] Pseudo-Kodinos, Traite des offices, ed. J. Verpeaux (Paris, 1976).
[iv] Horoscope for the Trapezuntine Empire, 1336, ed. S. Lampros,
Neoc; 'EivlvTJvaJ.Lvru.trov 13 (1916).
2.2 Vernacular romances
[i] Kallimachos and Chrysorrhoe, ed. C. Cupane (Torino, 1995).
[ii] Velthandros and Chrysantza, ed. C. Cupane (Torino, 1995).
[iii] Livistros and Rhodamne, ed. J. A. Lambert (Amsterdam, 1935).
Acknowledgements
I am much indebted to Professors E. Trapp (Bonn) and E.M. Jeffreys (Oxford)
as well as to Mr D.W. Farrell (Oxford) and, especially, to my mother, who
read through half of the prosopographical dictionary of the Palaiologan period,
generously helping me with my hunt for late Byzantine eunuchs.
Notes
Translations from Greek into English are my own. Byzantine Greek '~' is
transcribed as 'v' rather than 'b'.
1 Tougher 1997 provides a useful introduction to the phenomenon ofByzantine
eunuchism plus a convenient survey of recent research.

2 Guilland 1943, 234 (italics mine): 'Avec la restauration de l'empire sous les
Paleologues (1261-1453), !'influence de !'Occident se fit plus grande aByzance.

La haute societe byzantine adopta bien des usages latins et aussi bien des prejuges
occidentaux. I.:eunuque fut regarde desormais comme un etre inftrieur et son
inferiorite physique devint des lors pour lui une tare plutot qu'un avantage.
Les basileis, d' autre part, mieux assures sur leur trone, n' avaient plus les memes
raisons de recourir au service des eunuques. Aussi les textes aux 14e et 1Y s.
mentionnent-ils rarement les eunuques. Ce silence des textes ne permet pas,
du reste, de conclure a la disparition des eunuques. Mais if est vraisemblable
de supposer que leur influence sur fa politique et sur le gouvernement est moins
importante que jadis.'
3 The last ruling family of Byzantium, 1259-1453.
4 Guilland 1943, 234 (cf. n. 2).
5 In my opinion, this argument fails to persuade: although by the fourteenth
century dynastic thinking may have been more common among ordinary - or
politically dissatisfied- people (cf. e.g. Macrides 1981), Komnenian as well as
211
Niels Gaul
Palaiologan emperors were certainly well aware of the danger of revolts initiated
by members of their own families. As Magdalino 1993, 190, has it: 'The threat
which the reigning emperor had formerly faced from the heads of rival families
had moved within his own family.' It should be noted that the founder of the
Palaiologan dynasty, Michael VIII, established himself on the throne by means of
usurpation, and that John V Palaiologos in his youth experienced the usurpation
ofJohn VI Kantakuzenos. Compare moreover the fate of Andronikos II and John
V Palaiologoi respectively in their later years, as well as the rivalry among the
younger siblings ofJohn VIII (see, in general, Nicol1993).
6 Hopkins 1978, 180.
7 Trapp 1976-96 [hereafter PLP]. This present study is effectively based upon
references to eunuchs that can be traced in the PLP. It has to be acknowledged that
further sources of such kind as the 1336 horoscope for the empire ofTrebizond
(Lampros 1916), which are not represented in the PLP and remain unknown
to me for the moment, might further modifY our perception of eunuchs in the
later Byzantine empire.
8 Pach. Rel. hist. 4.681.19; Kant. Hist. 2.223.22.
9 Kazhdan and Wharton Epstein 1985, 69-70; Magdalino 1993, 180ff.
and Magdalino 1996, 146-55, esp. 147 n.6 for earlier references. However,
Magdalino does not discuss the changing role of eunuchs during these years,
simply stating 'there is enough to indicate that eunuchs remained an influential
force behind the obvious power structures of the sekreta and the imperial family'
(1993, 260). See also the comments of Mullett in this volume.
10 Angold 1975, 147-81. As long as a prosopography of the Nicaean Empire
remains a desideratum of Byzantine studies, any examination of eunuchism at

the Nicaean court will be difficult to undertake.
11 Laiou 1973; Raybaud 1968, 156-251.
12 Laiou 1981, esp. 249-57, and 1982, esp. 199-201. Cf. ,most recently
Garland 1999, 159-224 (with some caution), and Hill1999.

13 On oiKetot see Verpeaux 1965.
14 Tougher 1997, 177-80, and especially in this volume, has' pointed out that
the Byzantines, disregarding their own legislation, did produce eunuchs within
the boundaries of their empire. Thus, lacking any evidence, there is no reason to
assume that any of our samples carne from abroad.
15 Miklosich and Muller II, 388-9.
16 Cf. Schreiner 1971, 156-60, for a prosopography of the Pepagornenoi
family; see also PLP #22340-71.

17 Unfortunately, our sources do not provide sufficient information to identifY
the reasons why John achieved such a high office.

18 The JlE'Ya<; 8pounapw<; ranked thirty-second (Ps.-Kod. off. 178.12-15,
179.11-15, 249.11-250.12; Guilland 1951; Kazhdan (ed.) 1991, vol. I, 663

s.v. 'droungarios tes viglas'), the npoKa9iJJ.1evo<; toii KOttmvo<; sixtieth (Ps.-Kod.
off. 176.6-14, 186.8-12), a cre~acrto<; seventieth (Ps.-Kod. off.139.30; Kazhdan
(ed.) 1991, vol. III, 1862-3 s.v. 'sebastos').
19 Lampros 1916; c£ PLP 14864.
212
20 Lampros 1916,40.13-15.
21 Lampros 1916,44.26-8, 45.8, 45.22--4.
22 Cupane (ed.) 1995, 190 (vv. 2226-7).
23 C£ Tougher 1997, 171-2.
24 This may hold true especially for the office of tata~, which we find (possibly)
no less than thrice among our group of fifteen. The exact duty performed by
a tat<l~ is unknown; the office was not introduced before the end of the twelfth
century and ranked fairly low, at the thirty-sixth position (Ps.-Kod. off 138.17).
In the thirteenth century he belonged, together with the myKEpVTJ~ and the
E1tt tfi~ tpartESTJ~, to the three major aulic functionaries appointed by Michael
VIII for his son, Andronikos II (Kazhdan (ed.) 1991, vol. iii, 2013-14 s.v.
'tatas'). Thus the office of tata~ rather suggests a, possibly honorific, service
in close proximity to the emperor than an elevated status at court. Both the
other offices were reserved exclusively for eunuchs up to the eleventh century
(Guilland 1967, 237-50).
25 On StJohn of Herakleia see generally Guilland 1926, 4-5 and 126-8, and
Laurent 1930.
26 On hagiography of the Palaiologan period in general c£ Macrides 1981,
Laiou-Thomadakis 1980 and Talbot 1991.

27 Greg. Vit. joan. 34.6-11.
28 Greg. Vit. joan. 36.3.
29 Castration for medical reasons was permitted from the ninth century
onwards, c£ novel LX of Leo VI (Noailles and Dain 1944, 222-7).

30 Vit. Niket. 1; Vit. Nikeph. 136.4.
31 Nikephoros Gregoras was 'snob' enough to despise John Apokavkos, I!E'Ya~
oou~ under Anna of Savoy, as a homo novus of obscure origin, c£ Magdalino

1984a, 61--4.
32 For a discussion of the earlier lives c£ Ringrose 1999, 130--4 and 1996,
75-93. C£ also Tougher 1997, 179-80, on the ninth to eleventh centuries.

33 Greg. Vit. joan. 34.6-37.13.
34 Theophylact, in his dialogue in defence of eunuchs (c£ Simon 1994 and the
contribution of Mullett in this volume), was the first to maintain that a castrated
man's soul is not affected by the <jlucrt~ of his body. For a detailed discussion from
a Byzantine point of view of whether or not castrated men could achieve sanctity
c£ Ringrose 1999, esp. 126ff. By the end of the fourth century eunuchs were
clearly not 'eligible' for sanctity as, according to the Fathers, their celibacy was
imposed upon them rather than being their own achievement.
35 A novelty of the Nicaean court in exile, who were trained for a career in
military or civil administration, c£ Angold 1975, 174-81.

36 Ringrose 1994, 94-7.
37 Greg. Vit. joan. 36.15.

38 Eunuchs were commonly likened to women: like the latter, they were said
to be unable to control their passions; c£ Ringrose 1994, 86-94, and Tougher

1997, 171 and 1999, 94-8. On the rather ambiguous treatment offemale saints
in Byzantium see Talbot 1996, ix-xv.
213
Niels Gaul
39 Gavras sent two letrers to Kallikrenites (epp. 48, 214) and Pepagomenos (epp.
409, 444) each; Philes dedicated one poem to Kallikrenites (anecd. 38) and two to
Pepagomenos (poem. I p. 137, 357). Compare PLP #10371 and #22357.
40 Phil. poem. I p. 137 (nr. 283.10).
41 Phil. poem. I p. 357 (nr. 191).
42 This office did not come into existence before the beginning of the fourteenth
century, c£ Darrouzes 1970, 136, 284-8; before then, the eKKA.lJcrtapxnc; was
usually in charge of the sacristy of a church or monastery (c£ Darrouzes 1970,
136; Kazhdan (ed.) 1991, vol. I, 682 s.v. 'ekklesiarches'). In Darrouzes' slightly
later list P2 the office is linked to the 11£:yac; xaptoq>uA.a~ (p. 573, 3): 'flyouv 6
IJ.Eyac; eKKA.lJcrtapxnc;'. In list R, while both offices are listed separately, the office
of IJ.Eyac; xapto(j>uA.a~ is described as follows (p. 574, 4-5): 'as the right hand of
the patriarch, he is a legal representative (eKOtKoc;) in all matters, taking care of the
exhortations (1tpotpo1tac;) and the wedding contracts (yaiJ.tKa cruvaA.A.awma)'.
Although no definition of the office of IJ.Eyac; EKKAlJcrtapxnc; is given, he may well
have been concerned with questions of canonical law too.
43 See Ringrose 1996 on the various 'mediating' functions of eunuchs in
Byzantine society.
44 Pach. Rel. hist. 4.681.19.
45 Kant. Hist. 1.64.12-65.14, 1.94.13-95.10, 1.118.16-119.7.
46 Cf. Tougher 1999, 93-100. This eunuch's behaviour seems worthy offurther
study: was Michael Kallikrenites a truly timid character, a 'typical eunuch', or

did he possibly act in accordance with a role which he was supposed to play in
accordance with the (unwritten) laws of political communication in Byzantium,
as some recent research carried out in western medieval studies (c£ Althoff 1997
with further bibliography) on the public display of body-language and emotions
may suggest? Althoff's collection of examples may alert us to the importance of
public display of body-language also in Byzantine politics.
47 On the importance of spiritual relationships in eleventh- and twelfth-century
Byzantium c£ Morris 1981, 46-50, and 1995, 90-110.

48 Ps.-Kod. off 261.3-21.
49 Kant. Hist. 2.223.16-224.1.
5° Kantakuzenos' choice of words seems to imply that some of his opponents
referred to in this passage, preferably the last ones mentioned, were eunuchs
themselves. However, both Alexios Apokavkos (PLP #1180) and Manuel Kin-
namos (PLP #11724) were certainly married; nothing seems to be known about
John Gavalas (PLP #93286).
51 Ringrose 1994, 96 and 515 n. 41.
52 Cf. Tougher 1999, 92.
53 C£ Bryer 1981.
54 Kant. Hist. 2.588.7-9.

55 Nicol1996, 75-83.
56 Nor does the limited scope of this chapter allow any discussion of the
impact this marriage may have had on the development of eunuchism in the.
Osmanh harem.
214
Eunuchs in the late Byzantine empire c.I250-1400
57 For a general overview compare Beaton 1996, 101-16, with Agapitos and
Smith 1992, 73-90. The classification 'vernacular' is somewhat misleading,
because there is general agreement that all three of these novels were produced
in, as well as for, a milieu closely related to the imperial court of Constantinople.
The Kallimachos romance, for example, may have been written by a nephew of
Emperor Michael VIII, cf. Beck 1971, 124-5, and Beaton 1996, 104 and
108-9. However, Agapitos and Smith 1992, 131, and Agapitos 1991, 15-16,
suggest some caution. While at present no final agreement on the chronological
order of these novels has been achieved (though see Agapitos 1993 and 1999,
112 n. 7), Agapitos has recently proposed - in a paper delivered at the xxe
Congres international des Etudes byzantines (August 2001) - to date Livistros
and Rhodamne as early as the Laskarid period (he considers Theodore II Laskaris
to be the author of the romance).
58 Guyot 1980, 72-7, analyses appearances of eunuchs in the ancient Greek
novel: they always appear in a Persian context, where Chariton depicts them
neutrally, while Heliodorus and lamblichus make ample use of prejudices and
characterize them as cruel. C£ also Scarcella 1996, 233f. on the pejorative
attitude displayed towards them.
59 Beaton 1996, 79. Although Manasses, writing his novel, was influenced by
ancient models (c£ Mazal 1967, 132, and above, n. 58), his dislike of eunuchs
was not limited to this genre: it is a prominent feature also of his world chronicle.
See most recently Magdalino 1997, 163, with further references. Hunger 1978,
II, 125, already acknowledged that in the twelfth-century novel eunuchs fit 'eine
auf das byzantinische Milieu der Zeit bezogene Darstellung'.
60 Mazal1967, fr. 80,6-9 (p. 184), c£ also fr. 110 (p. 192) andfr. 161 (p. 204).
See also Cupane (ed.) 1995, 194 n. 146.

61 For further examples of a fruitful sociological exploitation of the Byzantine
romance and epos (DigenisAkritis) see Pieler 1971, Magdalino 1984b and 1989,
and some essays in Beaton and Ricks 1993. A 'semi-realistic' setting seems to
hold true also for the Byzantine Achilleid's Naples version, where the use of
the title and of the insignia ofOecmcrtTJ<; (Smith 1999, esp. vv. 88, 116, 175-7,
366-78 and commentary) designates the emperor's son and heir to the throne,
as in contemporary politics. See also n. 57.
62 The Kallimachos and Velthandros romances are quoted from Cupane (ed.)
1995.
63 I understand these rather unclear verses to express the eunuchs' hope that the
king would punish Chrysorrhoe by forcing her to accept a nun's habit.

64 C£ Ringrose 1994, 86-90 and 102-9.
65 All quotations from the Codex Scaligeranus version, ed. Lambert 1935.
66 C£ Agapitos 1991, 218-22.
67 There is general agreement that under the Angeloi emperors the phenomenon
of eunuchism temporarily revived, c£ Guilland 1943, 233, and Kazhdan and

McCormick 1997, 179-80, carefully refining Guilland's statement.
68 See Tougher's contribution to this volume for an alternative or, rather,
additional suggestion.
215
Niels Gaul
Bibliography
Agapitos, P.A.
1991 Narrative Structure in the Byzantine Vernacular Romances. A Textual and
Literary Study ofKallimachos, Belthandros andLibistros, Munich.
1993 ''H xpovoA.oytKft aKoA.ouSia 'tcOV ll'U8t<J'tOPTJI!CltCOV KaA.A.i~-Laxo~,
BeA.Savopo~ Kat AWtcr'tpo~ ', in N.M. Panagiotakis (ed.) Origini della
letteratura neograeca, Venice.
1999 'Dreams and the spatial aesthetics of narrative presentation in Livistros
and Rhodamne', DOP 53, 111-28.
Agapitos, P.A. and Smith, O.L.
1992 The Study ofMedieval Greek Romance, Copenhagen.
Althoff, G.
1997 Spielregeln der Politik im Mittelalter, Darmstadt.
Angold, M.
1975 A Byzantine Government in Exile. Government and society under the
Lascarids ofNicea {1204-61), Oxford.
Angold, M. (ed.)
1984 The Byzantine Aristocracy, IX-XIII Centuries, Oxford.
Beaton, R.
1996 The Medieval Greek Romance, 2nd edn, London.
Beaton, R. and Ricks, D. (eds.)
1993 Digenis Akritas: New approaches to Byzantine heroic poetry, Aldershot.
Beck, H.-G.
1971 Geschichte der byzantinischen Volksliteratur, Munich.
Bryer, A.A.M.
1981 'Greek historians on the Turks: the case of the first Byzantine-Ottoman
marriage', in R.H.C. Davis andJ.M. Wallace-Hadrill (eds.) The Writing
ofHistory in the Middle Ages. Essays presented to Richard William Southern,
Oxford.
Cupane, C. (ed.)
1995 Romanzi cavallereschi bizantini, Torino.
Darrouzes, ].
1970 Recherches sur les 6~P~Pina de l'eglise byzantine, Paris.
Delehaye, H.
1895 'Vita Sancti Nicephori Episcopi Milesii saeculo X', Analecta Bollandiana
14, 129-66.
Failler, A. (ed.)
1984-2000 Georges Pachymires: Relations historiques, 5 vols.,Paris.
Fatouros, G. (ed.)
1973 Die Briefe des Michael Gabras, 2 vols., Vienna.
Garland, L.
1999 Byzantine Empresses. Women and power in Byzantium AD 527-1204,
London.
Guilland, R.
1926 Essai sur Nicephore Gregoras: l'homme et l'ceuvre, Paris.
216
1943 'Les eunuques dans l'empire byzantin', REB l, 197-238. Reprinted

1967, 165-97.
1951 'Etudes de titulature et de prosopographie byzantines. Les chefs de la
marine byzantine: Drongaire de la flotte, grand drongaire de la flotte,
due de la £lotte, megaduc', BZ 44, 212-40. Reprinted 1967, 535-62.
1967 Recherches sur les institutions byzantines I, Berlin and Amsterdam.
Guyot, P.
Stuttgart.
Hackel, S. (ed.)
1981 The Byzantine Saint, London.
Halkin, F.
1957 Bibliotheca Hagiographica Graeca, 3 vols., Brussels.
1969 Auctarium Bibliothecae Hagiographicae Graecae, Brussels.
Hill, B.
1999 Imperial Women in Byzantium I 025-1204. Power, patronage and ideology,
Harlow.
Hopkins, K.
Hunger, H.
1978 Die hochsprachliche profane Literatur der Byzantiner, 2 vols., Munich.
Kazhdan, A. (ed.)
1991 The Oxford Dictionary ofByzantium, 3 vols., New York, Oxford.
Kazhdan, A.P.(t) and McCormick, M.
1997 'The social world of the Byzantine court', in Maguire (ed.) Byzantine
Court Culture, 167-97.
Kazhdan, A.P. and Wharton Epstein, A.
1985 Change in Byzantine Culture in the Eleventh and Twelfth Centuries,
Berkeley, Los Angeles and London.
Laiou, A.
1973 'The Byzantine aristocracy in the Palaeologan period: a story of arrested
development', Viator 4, 131-51.
1981 'The role of women in Byzantine society', JOB 31.1, 233-60 (= xvi.
lnternationaler Byzantinistenkongress (Wien 1981), Akten Ill).
1982 'Addendum to the role of women in Byzantine society', JOB 32.1,
198-204 (= xvi. Internationaler Byzantinistenkongress (Wien 1981),
Akten III I).
Laiou-Thomadakis, A. E.
1980 'Saints and society in the late Byzantine empire', in A. E. Laiou-
Thomadakis (ed.) Charanis Studies, New Brunswick.
Lambert, J.A.
1935 Le roman de Libistros et Rhodamni, Amsterdam.
Lampros, S.P.
1916 'Tpancl;ouv'ttaKov 'QpocrK6mov 1:ou ihou~ 1336 ', N£o~ 'EA.A.T\VOJ.lVTlf.LCOV
13,33-50.
217
Niels Gaul
Laurent, V.
1930 'La personnalite de Jean d'Heraclee (1250-1328). Oncle et precepteur
de Nicephore Gregoras', 'EA.A.T]vtKa 3, 297-315.
1934 'La vie de Jean, metropolite d'Heraclee du Pont', ApxEtOV TIOV't01l
6, 3-63.
Macrides, R.
1981 'Saints and sainthood in the early Palaiologan period', in Hackel (ed.)
The Byzantine Saint, 67-87.
Magdalino, P.
1984a 'Byzantine snobbery', in Angold (ed.) The Byzantine Aristocracy, 58-78.
1984b 'The Byzantine aristocratic oikos', in Angold (ed.) The Byzantine
Aristocracy, 92-111.
1989 'Honour among Romaioi: the framework of social values in the world of
Digenis Akrites and Kekaumenos', BMGS 13, 183-218.
1993 The Empire ofManuel I Komnenos, 1143-1180, Cambridge.
1996 'Innovations in government', in M. Mullett and D. Smythe (eds.) Alexios
I Komnenos, vol. 1, Belfast, 146-66.
1997 'In search of the Byzantine courtier: Leo Choirosphaktes and Constan-
tine Manasses', in Maguire (ed.) Byzantine Court Culture, 141-65.
Maguire, H. (ed.)
1997 Byzantine Court Culture from 829 to 1204, Washington, D.C.
Martini, E. (ed.)
1900 Manuelis Philae carmina inedita, Naples.
Mazal, 0.
1967 Der Roman des Konstantinos Manasses, Vienna.
Miklosich, F. and Muller, J. (eds.)
1860-90 Acta et diplomata graeca medii aevi, Vienna.
Miller, E. (ed.)
1855-7 Manuelis Philae Carmina, 2 vols., Paris.
Miller, W.
1926 Trebizond. The last Greek empire, London.
Morris, R.
1981 'The political saint in the eleventh century', in Hackel (ed.) The
Byzantine Saint, 43-50.
1995 Monks and Laymen in Byzantium, 843-1118, Cambridge.
Nicol, D.M.
1968 The Byzantine Family ofKantakouzenos, Washington, D.C.
1993 The Last Centuries ofByzantium, 2nd edn, Cambridge.
1996 The Reluctant Emperor, Cambridge.
Noailles, P. and Dain, A.
1944 Les novelles de Leon VIle Sage, Paris.
Papachryssanthou, D.
1968 'Un confesseur du second iconoclasme: la Vie du Patri<;e Nicetas (t836)',
TM3, 309-51.
Pieler, P.E.
1971 'Recht, Gesellschaft und Staat im byzantinischen Roman der Palaiolo-
genzeit',]OB 20, 189-221
218
Eunuchs in the late Byzantine empire c.I250-I400
Raybaud, L.-P.
1968 Le gouvernement et !'administration centrale de !'empire byzantin sous les
premiers Paliologues (1258-1354), Paris.
Ringrose, K.M.
tion', in L. James (ed.) Desire and Denial in Byzantium, Aldershoi:,
123-37.
Scarcella, A.M.
1996 'Social and economic structures in the ancient novel', in G. Schmeling
The Novel in the Ancient World, Leiden, New York and Cologne.
Schopen, L. (ed.)
1828-32 Joannis Cantacuzeni eximperatoris historiarum libri IV, 3 vols.,
Bonn.
Schreiner, P.
1971 'Eine griechische Grabinschrift in Corridonia',JOB 20, 149-60.
Simon. D.
1994 Lobpreis des Eunuchen, Munich.
Smith, O.L. (t) (ed.)
1999 The Byzantine Achilleid. The Naples version, Vienna.
Talbot, A.-M.
1991 'Old wine in new bottles: the rewriting of saints' lives in the Palaeologan
period', inS. Curcic and D. Mouriki (eds.) The TwilightofByzantium,
Princeton, N.J.
1996 Holy Women ofByzantium, Washington, D.C.
Tougher, S.
1999 'Images of effeminate men: the case of Byzantine eunuchs', in D. Hadley
(ed.) Masculinity in Medieval Europe, London and New York, 89-100.
Trapp, E. (ed.)
1976-96 Prosopographisches Lexikon der Palaiologenzeit, Vienna.
Verpeaux, J.
1965 'Les oikeioi. Notes d'histoire institutionelle et sociale', REB 23, 89-99.
1976 Traite des offices, Paris.
219
12
EUNUCH POWER IN IMPERIAL CHINA
Shih-shan Henry Tsai
Most of the stories about Chinese eunuchs come from the twenty-three
standard Chinese Dynastic Histories, which were long a monopoly of
Confucian historiographers. These historical literati used their monopolistic
writings to promote the politics of the sage-kings and the imperial system.
And, if the system derailed, they felt that they were responsible for mending
it. Unfortunately, cruel political realities often defeated such idealism, and
ultimately they attributed all evils to the despised and hateful eunuchs,
when in fact the cause of the ills of their society was the very imperial
system that the Chinese intelligentsia gleefully served. Consequently,
Chinese historians rarely openly and persistently criticized the autocratic
political system and the tyranny generated by it. Instead, they singled out
the eunuchs as the scapegoats and refused to treat the group of courtiers
as a social and political complex. This chapter attempts to deviate from
orthodox Chinese ideological abettors and tries to give eunuchs a more
balanced treatment. It presents Chinese eunuchism in the context of
imperial despotism, court politics, and the eunuch institutions the emperor
created and which received their power from him.
Nobody knows when and how eunuchs were first institutionalized
in China, except that castration was frequently used as a substitute for
the death penalty. In China, palace eunuchs were called siren, meaning
'waiters in the palace chambers'. They were also called huanguan, which
was a recognized official title during the Shang dynasty (1765-1223 BC),
appearing on the Shang oracle bones: Since then, both the Chinese words
huan and guan mean officials or officialdom. According to Zhouli (Rites
of the Zhou dynasty, 1122-256 Be), the king invested one queen, three
madams, nine concubines, twenty-seven varied ranks of consorts, and
eighty-one court ladies for the duties of the Inner Court. In conjunction
with this system, the Zhou king also employed castrated men to supervise
royal chambers and guard his harem. In spite of the long-held suspicion
that Zhouli was a composition of the late Han dynasty, it provides all
221
the official ranks of eunuchs as chamberlains in the palaces. The first

identifiable Chinese eunuch in the Zhou court was Meng Zi who seemed
unable to accomplish what he wished. He poured forth his anger and
frustration in a poem which was later collected in the Book ofOdes, one of
China's six ancient classics.' However, by Confucius' time (551-479 BC),
a decidedly negative image of the eunuchs had already been established
in Chinese society. Confucius was said to have been so irked by the 'evil
example' of a prominent eunuch named Yang Qu of the state of Wei
that, after staying for only a month, he decided not to serve the duke of
Wei. The first notorious eunuch who drew the attention of the ancient
chroniclers was Pei who, according to the historian Zuo Qiuming, served
under three mutually hostile dukes during the sixth century BC. Pei was
to become the stereotype eunuch- a person notorious for his sycophancy,
ruthlessness, treachery, greed, and luxury.
During the Qin dynasty (221-207 BC), more eunuchs were employed
to serve the ambitious empire builder Qin Shihuangdi, who was believed
to have kept over 3000 concubines in his palaces, establishing a new
agency called Zhongchangshi for the sole purpose of managing the ever-
increasing number of court-eunuchs. In 210 BC, while on his routine state
inspection tour, the First Emperor of Qin died. Immediately, concerns
were raised about who should succeed to the throne, and an attendant
eunuch by the name of Zhao Gao suggested that the news of the emperor's
death be kept from the public, including the emperor's oldest son and
the Heir Apparent Fu-su, until they could move the troops safely to the
capital at Xianyang (near present-day Xian). Secretly, ~he eunuch Zhao
Gao and the emperor's youngest son Hu-hai ordered the construction of
a special coffin to slow the decomposition of the emperor's dead body
and to obscure the smell. Openly, they pitched tern for the emperor
and brought meals to him every day as if the emperor were still alive. In
the meantime, the eunuch Zhao Gao, the Prime Minister Li Si, and the
emperor's youngest son successfully plotted to kill the Heir Apparent and
to make Hu-hai the Second Emperor of the Qin dynasty. Mterwards,
Zhao Gao was promoted to head the eunuch agency Zhongchangshi,
and on his recommendation the young emperor ordered all of his father's
consorts who did not bear sons to follow the First Emperor to the grave;
so, too, the artisans who had worked on the terra cotta figures and the
tomb. Mter winning the total trust of the young monarch, Zhao Gao
felt that Prime Minister Li Si knew more than he was supposed to know
about their secrets and had now become the biggest threat to the new
court. With the young emperor on his side, Zhao Gao replaced Li Si as the
Prime Minister while callously but methodically purging Li's associates.
222
Eunuch power in imperial China
But popular discontent mounted, and in only a few months, rebellions

broke out all over the empire. During the early autumn of 206 BC, when
the rebels began to march into the valley of the Wei River, Zhao Gao
murdered the Second Emperor Hu-hai. But a month later, the army led
by Liu Bang, founder of the Han dynasty, crushed the Qin defenders.
Zhao Gao, the first really powerful eunuch in Chinese history, ultimately
became a casualty. 2 He was forever identified with usurpation, political
intrigues, and murders. The tale of Zhao Gao comes primarily from the
Han historians Sima Qian, who lived in the second century BC, and
Jia Yi (201-169 BC). Both Sima and Jia portrayed the First Emperor
of Qin as a murderer and oppressor who failed to rule with humanity
and righteousness. The grand eunuch Zhao Gao, who was the shadow
of the emperor, was then singled out to share the blame of Qin's many
evil policies.
In the succeeding dynasties, court eunuchs continued to grow in
number and influence and ultimately became an important part of China's
imperial apparatus. Although Chinese society generally viewed these
people as anathema, their status had, by the beginning of the Christian era,
been firmly established. During the Eastern Han dynasty (AD 25-220), the
eunuch agency Zhongchangshi was re-established which made it possible
for high ranking castrated courtiers to gain access to the emperor and the
empress. Six dowager empresses successively promoted Zhongchangshi
eunuchs to powerful positions and unwittingly sowed the seeds for the
dynasty's downfall- at least according to Confucian historiographers. In
135, the eunuchs were permitted to adopt sons, and their power grew
with their wealth as some of them owned large agriculturallands. 3 Early
in AD 189, Emperor Ling of the Eastern Han dynasty died at the age of
thirty-two, and because his son, the new Emperor Shao, was only thirteen
years old, the Empress Dowager - by the name of He - took over the
helm of the state. She immediately promoted her older brother He Jin,
who had earlier resisted the messianic rebels called the Yellow Turbans in
Luoyang, to be the Grand Commandant while appointing twelve grand
eunuchs to manage the Inner Court Zhongchangshi. He Jin, however,
sided with the bureaucrats and pressured his sister the Empress Dowager
to get rid of all the eunuchs, charging them with rampant corruption
and abuses of power.
Once, in a heated exchange, the chief eunuch Zhang Rang, who held
the rank of marquis, asked He Jin: 'You said that we inside the place are
corrupt. Please tell me, Sir, from the ministers on down, who is loyal and
honest?' 4 The final showdown came in September of AD 189, when Zhang
Rang's confidant, the eunuch Qu Mu, slew the Grand Commandant He
223
Jin during a court audience. He Jin's deputies in turn brought their troops
to the capital Luoyang and put to death more than 2000 eunuchs in
retribution. The chief eunuch Zhang Rang took the teenaged Emperor
Shao and the Dowager He and fled northward toward the Yellow River.
But after being surrounded by his enemies, Zhang Rang jumped into the
river and drowned himself, while his patron and protector the Dowager
He was forced to take poison. The Emperor Shao was then deposed and
succeeded by his eight-year-old half-brother, the Emperor Xian. But that
was also the end of the Eastern Han dynasty. A strong man named Cao
Cao who would hold sway of China during the next turbulent decades
was actually an adopted grandson of a court eunuch. As a consequence,
orthodox Chinese historians almost universally attribute the downfall of
most of the dynasties to the lascivious conduct and insolent power of the
eunuchs, a power derived from their intimacy with the young monarch
and the empress dowager, from access to offices inside and outside the
palace, and from a passive government beset with corruption, fear, and
inept leadership. A typical example was the last Emperor of Chen dynasty
(582-9), during the Northern and Southern Dynasties, who lived in
seclusion in his harem and seldom received his officials. He authorized
his chief eunuchs Cai Tuoer and Li Shanchang to deliberate on state
documents. And frequently his favourite concubine Zhang was sitting
on his lap when he discussed state affairs with eunuchs Cai and Li and
gave the imperial sanctions. 5 It is clear that the historian Yao Shilian, who
compiled the official history of the Chen dynasty, wanted this politically
charged cliche- the weak and lazy emperor, the sly and cynical eunuchs
- to be embedded in Chinese historiography. '
The next dynasty that was to be constantly plagued with eunuch
problems was the prosperous and culturally brilliant Tang dynasty
(618-906). During the early years of the Tang, power centred on the
emperor and three to seven chancellors. Mter the mid-eighth century,
however, the eunuchs began to undermine the integrity of the chancellery,
as many of them were appointed commanders of the army. According to
the Old Tang History, one of China's twenty-three official histories, the
Tang court generally maintained over 4600 ranked eunuchs who owned
60% of the property and land in the capital city and directly participated
in the decision-making process. 6 And because there were so many court
intrigues and bloody coup d' etats, a handful of daring eunuchs, who
sided with the winners, often received rewards and were subsequently
entrusted with more power. A case in point was Gao Lishi, who in 713
helped Emperor Ming Huang get rid of his great aunt, the Grand Princess
Taiping, effectively removing a real threat to Ming Huang's regime. The
224
death of the Grand Princess Taiping signalled not only the end of half
a century of women's domination of the Tang court, but also the rise of
the eunuch powers. Gao Lishi was ultimately invested as the Duke of Qi-
hence the beginning of the tradition in Chinese history oflavishly granting
noble titles to castrated men. All told, the Tang emperors invested one
eunuch prince, twenty-seven eunuch dukes, four eunuch marquises, two
eunuch earls, four eunuch viscounts, and seven eunuch barons?
During the decade of 745-55, Emperor Ming Huang, then in his
late sixties and early seventies, became infatuated with his legendary
imperial consort Yang Guifei who, at the prime age of thirty-something,
encouraged the ageing monarch to indulge in music, dancing, and opera.
As Emperor Ming Huang lost interest in the management of the state,
the chief eunuch Gao Lishi literally became the de facto ruler of the Tang
court. Even the great minister Li Linfu, who directed the empire from
737 until his death in 752, and the commander of the Northern Army,
An Lushan (of Sogdian and Turkish stock), sought favour with Gao
Lishi. Other high officials of the central government in Xian and the
provinces were reduced to ingratiating themselves with the favourites
of Gao to maintain their power and position. Princes and dukes alike
called him 'Master Gao', officials high and low considered him their
real boss, and Emperor Ming Huang himself affectionately called Gao
'My Commander'. Old Tang History gives us the following biographical
sketch of Gao:
Gao Lishi owned immense property and wealth, far surpassing those of
the princes and the nobles. He gave money to construct Buddha statues
and the famous Treasure-Longevity Monastery, as well as a gigantic Daoist
temple ... At the northwestern corner of the capital, he built a five-wheeled
water mill that could grind 300 bushels of wheat daily. Mter the bell at the
Treasure-Longevity Monastery was cast and was ready for use, the entire
court came to the monastery to take part in the ceremony. Courtiers took
turns to hit the bell, each hitting it from ten to twenty times. 8
In the winter of 755-6, An Lushan marched at the head of his armies

(180,000 men and 30,000 horses) toward Xian, and the ensuing military
rebellion, which lasted until 763, drastically changed the Tang defence
system. Afterwards, recruitment of soldiers took the place of conscription
until mercenaries came to dominate the army near the end of the Tang
dynasty. In the meantime, the Tang emperor appointed his grand eunuchs
as commanders of palace guards and army supervisors. Among the most
powerful eunuch commanders was Li Fuguo (jl. 756-62), who totally
dominated Emperors Suzong and Daizong. 9 And as the eunuchs' inroads
into the Tang military establishment grew unabated, they became virtual
225
kingmakers, instrumental in selecting seven of the last eight Tang emperors.

But once again, the eunuchs had to pay a heavy price for wielding such
enormous power. In 903, the warlord Zhu Quanzhong arrived in Xian
and executed 162 court eunuchs and their collaborators. He later arrested
and killed 300 more and ordered every eunuch who had been involved
in military affairs to commit suicide. With most of the eunuchs gone,
so went the Tang dynasty.
During the next fifty-four years of the so-called Five Dynasties, China
witnessed political division and chaos, but very few eunuchs rose to
prominent positions. Upon ascending to the throne, Zhao Kuanyin, the
founder of the Song dynasty and fully aware of the eunuch problems in
previous dynasties, limited his castrated servants to only fifty and also
took measures to keep them under strict control in the capital city of
Kaifeng. But his younger brother and successor Zhao Kuanyi (976-97)
increased the number of eunuchs to 280, and during the next two reigns,
the number continued to swell, from several hundreds to several thousands.
Four chapters of Song Shi [History of the Song Dynasty] - from juan 466
to juan 469 - provide brief biographies for fifty-three prominent eunuchs
during the Song dynasty. Among them, forty-three rose to prominence
during the Northern Song (960-1126) while ten became influential during
the Southern Song (1127-1279). In addition, twenty-seven were military
commanders who at one time or another fought off invaders in Song's
frontier regions. Generally speaking, during the first hundred years of the
Song dynasty, which governed then the largest nation on earth, China
enjoyed the reign of a succession of emperors who were conscientious and
able administrators, as well as the general absence of any notorious perfidy
on the part of the eunuchs. However, several eunuchs were given military
command. In the year 1103, upon the recommendation of the Prime
Minister Cai Jing, Song Emperor Huizong appointed the grand eunuch
Tong Guan, a native of Kaifeng, to lead an army 100,000 strong against
a nomadic people of Himalayan origin called the Qingtang tribe. In caring
for their herds, the Qingtang tribesmen constantly looked for water and
grass, consequently extending their forays into the lands of the settled
agricultural communities at Hezhou and Taoxi, in present-day Gansu
province. The Song emperor was delighted that the eunuch commander
Tong Guan had accomplished something significant as he crushed the
raiders within only a month. After this successful campaign, Tong Guan was
promoted to be a 'Regional Supervisor' in the Song military hierarchy and,
five years later, received an additional title as a 'Regional Commandant'.
By 1115, Tong Guan was commanding all the troops, totalling some
800,000, that were deployed along the empire's western frontiers. Two
226
years later, Tong was again promoted to head the Bureau of Military
Mfairs, or Shumiyuan. Moreover, he was also named the 'Grand Preceptor
of the Heir Apparent', an unprecedented honour for a eunuch. During the
two decades when Tong Guan took part in policy-making, he consistently
advocated a detente with Song's powerful northern neighbour, the Liao
empire - founded by a Mongol people called Qidan. But in 1125 the
Jurched, a Tungusic people from whom the Manchus later descended,
dismantled the Liao and immediately announced their intention to take
on the Song Chinese. Tong Guan was then despatched to Taiyuan, but
when he failed to stop the enemies from entering the capital city of
Kaifeng, faculty and students from the Imperial University repeatedly
requested the removal of all of the eunuchs. When the situation finally
got out of control, Emperor Huizong abdicated and fled with Tong Guan
southward to Yangzi. However, in order to satisfY the demands of the anti-
establishment elements, the new Emperor Qinzong ordered the execution
of Tong Guan in the summer of 1126. Within only a few months,
however, the Jurched captured not only the new Emperor Qinzong but
also the retired Emperor Huizong, sending both of them to Manchuria,
never to return. 10
Tong Guan was forever blamed for causing the downfall of the Northern
Song dynasty and ever since, no other Song emperors, who continued the
dynasty as Southern Song in Hangzhou, would dare to appoint eunuchs to
any militarily responsible positions. When China was under Mongol rule
(1279-1368) eunuchs were rarely active in the political or military arena.
Nevertheless, in 1348, Jia Lu was appointed a Taijian, or grand eunuch,
to manage flood control and maintain China's hydraulic infrastructure.
Taijian, a unique eunuch title in China's officialdom, was thus created and
henceforth became an important element in China's imperial history.
When Zhu Yuanzhang (known as Emperor Hongwu) founded the Ming
dynasty in 1368, he was also keenly aware of the potentially pernicious
eunuch problem and decided to limit the number of court eunuchs to
fewer than 100. Even though he was later to increase the number of palace
servants to more than 400, he reportedly also decreed that no eunuch be
permitted to learn books or to give advice on political matters. 11 Even
the few eunuchs with whom Emperor Hongwu might chat were kept
dutifully awed and never allowed to discuss politics. In 1384, he had the
following inscription engraved on an iron tablet in front of his Nanjing
palace: 'Eunuchs are forbidden to interfere with government affairs. Those
who attempt to do so will be subjected to capital punishment.' 12 Shortly
before his death he ordered that eunuchs should no longer be allowed to
wear the uniform of government officials and that their rank should not
227
exceed the fourth grade. Officials of all departments were forbidden to

communicate with eunuchs by written documents.
The commands of the founder of the dynasty were disobeyed soon aft~r
his death. When Yongle (1402-27) usurped the throne of his nephew,
Emperor Jianwen (1398-1402), he had to rely upon the help of the
eunuchs because the court ministers had remained loyal to his nephew.
After his successful coup d'etat, Emperor Yongle moved the capital to
Beijing, rewarded his eunuchs by giving them high ranks and showered
them with special favours, even putting some of them in charge of military
and diplomatic affairs. In the ensuing years, Emperor Xuande, Yongle's
grandson, established a classroom known as Neishutang, or Inner Court
School, and appointed four scholars from the Hanlin Academy to teach
some 200 to 300 young eunuchs whose average age was about ten. Some
of these educated eunuchs were to become secretaries of the emperor
and often had communication with court officials. As the eunuchs were
becoming literate, they became a dominant group to be reckoned with.
The eunuchs were engaged in espionage and internal security, military and
foreign affairs, judiciary review, tax and tribute collection, the operation
of imperial monopolies, and so on. They acted as the trusted minions of
the throne, and by extension, the state.
Not only did the Ming eunuchs secure power bases in the court, but
also they proliferated because the supply of eunuchs, primarily by means of
self-castration, exceeded the demand for them. Beginning with Emperor
Chenghua (1465-87), the Ming court was forced to take in more young,
poor castrati than it could afford. At the end of the fifteenth century, there
were approximately 10,000 eunuchs in the Forbidden C!ty and in various
Ming princely establishments. 13 In 1572, Emperor Longqing unwillingly
selected 3250 such persons and assigned them to different bureaux and
directorates of the imperial household. In 1601, Emperor Wanli picked
4500 castrated men off the streets of Beijing. He kept 3000 for his own
use and gave fifty to each of the territorial princes, twenty to every imperial
prince (his own sons), and ten to each prince of second degree (the sons
of every imperial prince). In 1620, some 20,000 such people swarmed
into the capital, begging for whatever jobs the Ming government could
give them. 14 By the end of the Ming dynasty in 1644 it was estimated
that the total number of eunuchs in China had reached 100,000 in a total
population of approximately 150 million. If each of the 100,000 eunuchs
had, on average, retained two household servants and also adopted a son
or supported a nephew, that would mean that the total eunuch personnel
the Ming government had to feed would be around 400,000. Tsai (1996)
asserts that this was a new variable, and one that should be injected into
228
all Ming historical discourse. As Lenin once said, quantity has a certain
quality of its own. And indeed, there were many skilled, intelligent, and
capable grand eunuchs who distinguished themselves as commanders of
the army, admirals, ambassadors, architects, engineers, financiers, and
superintendents of various agencies.
The Ming eunuchs were not just household servants hewing wood and
drawing water- they actually made up a third administrative hierarchy,
participating in all of the most essential matters of the dynasty. They
exercised their power in all areas of government, frequently challenging
and surpassing those of scholar-bureaucrats and the military establishment.
Among the most powerful and notorious eunuch bosses were Wang Zhen
in the 1440s, Wang Zhi in the 1470s, LiuJin in the early 1500s, and Wei
Zhongxian (jl. 1620-7) near the end of the Ming dynasty. On the other
hand, there were exemplary eunuchs who left remarkable careers behind
them. Nguyen An, a castrato sent from Vietnam as tribute to the Ming
emperor, designed and constructed the fabulous Forbidden City during
1410-22. Feng Bao, serving both as the Director of the Eastern Depot
(Ming's espionage agency) and as the Director of the Ceremonial - the
Ming emperor's general chief of staff- helped to guide the statecraft of
the Ming during the 1570s. But the Ming eunuch who is best known
to the Western World is Admiral Zheng He, who led seven maritime
expeditions, between 1405 and 1433, to some thirty states in south-east
Asia and along the Indian Ocean, reaching as far as Hormus in the Persian
Gulf and Somaliland in Mrica. Each of these pre-Columbian voyages
involved tens of thousands of government troops and employed more
than sixty-five ocean-going vessels that travelled several thousand miles.
Such phenomena ultimately inspired Levathes to write a lively account
of Ming's naval reconnaissance entitled When China Ruled the Seas. 15
The influence of the eunuchs was so pervasive that one might even call
the Ming government a government of half-men. Such a designation of
imperial China will assuredly irritate male chauvinist Chinese scholars.
But when the peasant rebels, led by Li Zicheng, entered the Forbidden
City through its back door during the late spring of 1644, and after almost
all of the officials had fled their posts, it was the eunuch commandant
Wang Chengen who stayed with the last Ming Emperor Chongzhen,
helping His Majesty to commit suicide in the Coal Hill. Afterwards, the
eunuch Wang took his own life.
During the Qing dynasty, the last of China's imperial dynasties, there
would be no more eunuch admirals or commandants, even though there
were a few influential eunuchs, such as An Dehai and Li Lianying, both of
whom served the Empress Dowager Cixi near the end of the dynasty. The
229
Manchu rulers placed severe restrictions on the court eunuchs: for example,
they were not allowed to leave the Beijing city gates; those who broke the
rule were always executed. Normally, the Qing emperor would only accept
castrated boys who were under fifteen years of age and unmarried. Before
entering palace service, these emotionally scarred and physically deformed
castrati were thoroughly examined by the palace head eunuch who made
sure that their 'treasure parts' had been permanently severed from their
bodies. Castrati who were fifteen years and older and those who had been
married were given to the princes and princesses. A Manchu prince, rank
7a, could keep as many as forty castrati under a eunuch supervisor, while an
8a rank prince was allowed to have thirty castrati plus a eunuch supervisor.
Decades later as the number of members of the imperial family grew at
an exponential rate, there was also an increasing number of palace women
who needed the service of eunuchs. Beginning with Emperor Qianlong
(1736-96), middle-aged castrati (some of whom had been previously
married) were recruited from the princely establishments to work in the
Forbidden City. And this is how Li Lianying, a handsome hairdresser in his
late twenties, was brought to serve the vivacious Empress Dowager.
A native of Hejian, some sixty miles south of Beijing, Li Lianying
became an orphan when he was still very young. Constantly struggling
just for mere subsistence, Li held all kinds of odd jobs, including as
a leathersmith and a hairdresser. Through the influence of another eunuch
named Shen Yulan from Hejian district, Li decided to make a radical
change in his life - by voluntarily having himself emasculated so that he
could enter the palace service. At the outset, he worked in various inner
court storehouses and held no office. But Li Lianying's talents were soon
discovered. When the Empress Dowager was looking for a hairdresser
to make a fashionable hairdo for her, Li was recommended. With this
stroke ofluck and because Li Lianying was very clever with his tongue, he
soon became Her Majesty's favourite eunuch. Also because the Empress
Dowager was an avid opera fan, the jack-of-all-trades Li Lianying decided
to learn how to act on the stage so that he could entertain his imperial
patron. He was gifted enough to play not only a male role, but also
a female role, and even the role of a clown. By the time Li was forty years
old, he had become Her Majesty's trusted confidant and daily companion.
His power was equal to a 1a grand councillor, and many of the high
officials, including the highly respected governor-general Li Hongzhang,
had started to ingratiate themselves with him.
In 1886, Li Lianying accompanied I-huan, the Prip.ce of Chun and also
the grandfather of Qing's last emperor Henry Pu-yi, for an inspection of
the new navy in Tianjin. In the course of the tour, a censor reprimanded
230
Li Lianying for insolence, and warned of the dangers of eunuch rule.

However, I-huan defended the accused and the censor was degraded. In
fact, several of the Qing army commanders and navy admirals, most of
whom were later to fight the Japanese during the Sino-Japanese War in
1894, became convinced that Li Lianying was their best insurance to
maintain their positions. They called Her Majesty's hairdresser 'Master
Li', who in turn exacted fees from these corrupt high military personnel. 16
Li used the graft he took to buy such luxurious items as cuckoo clocks,
and also to purchase official positions for his adopted sons, Li Fuheng, Li
Fude, Li Fuli, and Li Fuhai, in the Ministries of Personnel, War, Justice,
and Public Works respectively. During the Boxer Rebellion and on the
approach of the Allied troops toward the Chinese capital, the Empress
Dowager fled Beijing, accompanied only by a handful of eunuchs, among
them her indispensable hairdresser Li Lianying. Like so many eun'uchs
before him, Li functioned as the Dowager's agent, keeping a prying
eye on the behaviour of the emperor, the princes, and other ambitious
ministers. But unlike eunuch bosses in previous dynasties, Li Lianying was
not a power player, mainly because he held no institutionalized offices.
He generally resigned himself to managing such routine affairs as food,
entertainment, birthdays, and so on in the inner court. Unlike Zhao Gao
of the Qin, Zhang Rang of the Han or Wei Zhongxian of the Ming, Li
Lianying played very little role in the downfall of the Qing dynasty. Even
though he also became a lightning rod for blame, his life was spared when
his protector the Empress Dowager died in 1908. Nevertheless, he had
lost all of his influence and allure, and was demoted to manage the court
gardens until his death in 1911.
Li Lianying was the last influential court eunuch, but with the downfall
of the Qing dynasty, so ended the disturbing and, in the eyes of twenty-
first century westerners, an irrational element in China's imperial past. It
is evident that a more objective and in-depth study of Chinese eunuchs
is needed for the topic to break free from the traditional framework in
which it has been studied. The thematic stress of usurpation of power by
the eunuchs and the conceptual framework of historical harm and damage
done by the eunuchs should be re-examined in the light of Chinese
institutional tyranny and despotic political traditions. A closer look
at the eunuchs' social backgrounds and their ways of life is needed.
One might even want to investigate whether the eunuch system was
a necessary cement that held imperial institutions together and made
China's polygamous society work. In short, it is only fair that eunuchs be
allowed to speak for themselves and be seen as the subjects rather than the
objects of Chinese dynastic history.
231
Notes
I Only 305 poems were selected in the Book of Odes. The eunuch Meng Zi's
rankling in his heart was well expressed in a xia ya folk song style. See Guo
Shaoyu et al. (eds.) 1979, 7-10.
2 Sima Qian 1964, juan 6 on 'Basic Annals of the First Emperor of Qin',
102-5.
3 Fan Ye et al. (eds.) 1979, juan 78, 2518.
4 Fan Ye et al. (eds.) 1979, juan 78, 2534-6.
5 Yao Shilian 1980, juan 7, 132.
6 Liu Xu et al. (eds.) 1979, juan 184, 4753-5.

Io Tuotuo et al. (eds.) 1979, juan 466-9, 13599 and 13658-62.
II ZhangTingyu et al. (eds.) 1979, juan 304, 7765.
I 2 ZhangTingyu et al. (eds.) 1979, juan 304, 7765.
I 3 Hongzhi Veritable Record, 70, 12th moon of 5th year, Hongzhi reign;
Zhengde Veritable Record, 11, 3rd moon of 1st year, Zhengde reign.
I 4 Wanli Veritable Record, 11, 3rd moon of 1st year; 205, lith moon of 16th
year, and 358, 4th moon of 29th year, Wanli reign.
I 5 Levathes 1994.
I 6 Hummel (ed.) 1943, 298, 385.
Bibliography
Anderson, M.M.
1990 Hidden Power: The palace eunuchs ofimperial China, Buffalo, N.Y.
Crawford, R. C. '
1966 'Eunuch power in the Ming dynasty', T'oung Pao 49, 115-48.
Fan Ye et al. (eds.)
1979 Hou Han shu [History of the Later Han Dynasty], Taiwan reprint.
Gardner, C.S.
1961 Chinese Traditional Historiography, Cambridge, Mass.
Gu Rong and Ge Jingang
1992 Wuheng weiqiang: Gudai huanguan qunti de wenhua kaocha [Fog across
the Curtain Wall: An examination of the collective culture of eunuchs
in ancient times], Xian.
Guo Shaoyu et al. (eds.)
1979 Zhongguo lidai wenlun xuan [Selected Literary Essays throughout China's
Dynasties], Shanghai.
Hummel, A.W (ed.)
1943 Eminent Chinese ofthe Ch'ing Period, Washing~on D.C.
Jurgel, U.
1976 Politische Funktion und Soziale Stellung der Eunuchen zur Spiiteren
Hanzeit (25-220 n. Chr.), Wiesbaden.
232
Kuwabara, ].
1936 'Shina no kangan' [The eunuchs m China], Toyo shi setsuen 22,
344-58.
Leng, D.
1990 Bei yange de shouhu shen: huanguan yu Zhongguo zhengzhi [Castrated
Eunuchs and Chinese Politics], Changchun.
Levathes, L.E.
1994 When China Ruled the Seas: The treasure fleet of the Dragon Throne,
1405-1433, Oxford.
Liu Xu et al. (eds.)
1979 Qju Tang shu [History of the Old Tang Dynasty], Taiwan reprint.
Millant, R.
1908 Les eunuques: atravers les ages, Paris.
Mitamura, T.
1963 Kangan: sokkin seiji no kozo [Chinese Eunuchs: The structure of intimate
politics], Tokyo.
Shimizu, T.
1932 'Jigu kangan no kenkyu' [A study of eunuch's self-castration], Shigaku
zasshi 43, 82-128.
SimaQian
1964 Shi]i [Historical Records], Taiwan reprint.
Stent, G.C.
1877 'Chinese eunuchs' ,journal ofthe North China Branch ofthe RoyalAsiatic
Society n.s. 10, 166.
Tsai, S.H.
1996 The Eunuchs in the Ming Dynasty, Albany, N.Y.
Tuotuo et al. (eds.)
1979 Song shi [History of the Song Dynasty], Taiwan reprint.
Yang Lien-sheng
1960-61 'Female rulers in Chinese history', Harvard journal ofAsiatic Studies
23,47-61.
Yao Shilian
1980 Chen Shu [Book of the Chen Dynasty], Taiwan reprint.
Zhang Tingyu et al. (eds.)
1979 Ming shi [History of the Ming Dynasty], Taiwan reprint.
233
13
THE OTHER CASTRATI
Richard Witt
Western castrati of the sixteenth to the nineteenth centuries are well-

researched. 1 Castrato singers of other times and places have vanished,
eclipsed by eunuch politicians. 2 The balance needs correcting: 3 often the
music and the political discourse were closely allied, the eunuch singer
becoming solace and confidant - what Arab scholars4 called nadfm - to
some ruler or high official, in a kind of music therapy: 5
Thou shalt present me as an eunuch to him;
It may be worth thy pains; for I can sing
And speak to him in many sorts of music
That will allow me very worth his service ... 6
The resulting sense of security, reflecting perhaps a more complex field of

male experience than the listener's own,? coloured early Near Eastern terms
for the eunuch- Akkadian sa resi, Iranian sa bestan, Syriac m'haimna ('the
trusted'), and the classical Greek 'polite' term Euvouxo~. 8
Musical aesthetics
The quality of the eunuch voice perhaps varied with the age at which
emasculation occurred. 9 In Ottoman Turkey 'amputative' emasculation,
removing the whole genital system, might be done shortly after puberty,
when the vocal cords had already enlarged and the voice had deepened.
Thus Osmin, an Ottoman harem overseer in Mozart's Die EntjUhrung
aus dem Serail, has basso profondo arias, not male soprano coloratura.
But in China, where the operation was done very early in life, the
voice characteristically remained high-pitched, landing eunuchs with the
unflattering nickname 'crows' .10
In the case of the castrati, who were produced by 'excisive' or 'partial'
emasculation, with the testicles, and no more, surgically disconnected
or removed, there was an 'optimal' age: not below seven and not above
twelve, and usually between eight and ten. 11 With the large proviso that
the testicles were excised completely without botching, the voice, instead
235
Richard Witt
of deepening, remained 'feminized in a masculine body', 12 in a treble

register 'halfway between child and woman'. The subject - if still alive -
was typically left with a high baritone or light tenor voice. He learned to
command the female register (as could a singer who was not biologically
a castrato, but a 'natural' eunuch) by projecting his voice 'artificially'
upwards 13 in folsetto, the technique where the female voice 'sits behind'
the normal voice, and which Elizabethan musicians called 'feigning' .14 In
Greece, Rome and Byzantium, the Near East, and probably classical India,
pre-pubertal castration, with the 'biological' voice pitch rising, was the
norm. 'Shriller than a eunuch's voice', ran the Roman proverbY
The truism that novelty is the spice of music applies forcefully to the
overlapping categories primo uomo, male alto, countertenor, folsettista,
and true castrato. Their novel timbre 16 and nimbleness, whether plain
or 'feigning', has always gripped or appalled listeners, among them
Schopenhauer. Indian, Greek and Arab theorists all made attempts to pin
down its acoustic qualities: few survive. 17 The school of Aristotle likened
the eunuch's speaking and singing voice to those of other groups falling
short of full male biological entelechy - prepubertal children, women,
old men, and 'people enfeebled by disease' .18 Implicitly weighed against
the 'normative' male voice, 19 the corporate sound of a tragedy chorus,
for example, it was found wanting - 'febrile', 20 lacking the stamina21 to
hold a note steady22 (one symptom of rabies being 'a trembling, dry,
eunuch-like (crJta8onorv;), hoarse voice'). 23 The favoured description was
'thin (A.EJt'to~) and shrill (6~u~)': 24 the voice of Favorinus the Sophist,
a hermaphrodite, 'sounded shrill, thin and highly strung, the tuning which
Nature has bestowed on eunuchs'. 25 Aristotelians ascribed this lack of
brio either to motor organ malfunctioning, or to comparative windpipe
narrowness, considered to raise pitch by accelerating the air jet. 26 With
greater scientific sophistication, an Arab musicologist discusses the inability
of singers with 'weak' vocal organs- presumably including castrati- to
hear the highest register of their accompanying instruments. 27 A Syrian
description of the idealized female voice of a noblewoman from Smyrna-
with eunuchs at her beck and call- has the same technical feel:
Every note is tender as tender can be; neither heavy, so that it suits with
the masculine, nor so exceedingly thin as to be acutely feminine and simply
insipid. No: her voice is like the voice of a boy not yet past puberty- stealing
on the ear deliciously, refreshingly, gently, and, when it ceases, lingering
like a faint echo. 28
This is the 'incomparable softnesse and sweetnesse' that John Evelyn found
in the castrato voice; that emasculation has always sought to conserve; 29 and
236
The other castrati
that, in eighteenth-century Naples, was iconized by dressing prepubertal

castrati as 'little angels' to sing laments over dead children. 30
Timbre, stretched compass, and artistic chutzpah aside, it was how
the castrato used his voice - his distinctive musical idioms, borrowed
or invented - that made him sought after, in classical China, India, and
the Near East as in seventeenth-century Italy. Reconstructing all of these
idioms without scores is clearly impossible, but comparative musicology
and textual analysis uncover three or four.
Specific to China were the basic Qing dynasty castrato styles: 'refined
restraint', for sentimental scenes; and 'rumbustious', for vigorous action
and crowd scenes. Since (as in Greek tragedy) female roles were sung by
biological males, the actor-singer cultivated an artificially raised register
- the distinctive 'high, short, falsetto' delivery that impresses listeners to
'Peking Opera' today. 31 It has been traced to vocalization at funerals -
the desire to become oneself a ghost (again a form of self-traumatization)
talking with ghosts in their own thin language. 32
In the ancient Near East, eunuchs were constantly connected with
dancing, cymbals, drums, piping, and ululation. 'Ululation', 'a noise made
by a combined motion of the tongue, throat, and hand vibrated rapidly
over the mouth', 33 was an onomatopoeia (like 'warbling'), derived from the
eastern Mediterranean battle-cry alald. Still commonplace throughout the
Islamic world, it expresses triumph, at weddings especially. In Zululand,
it greets the discovery that a bride is virgo intacta. Hair-raising, even at
a distance, the effect is of a ferocious, jubilant communal hollering, pulsing
rapidly on a single indeterminate high pitch. 34 The sound generates at
the base of the tongue 35 (recalling both the gutturalis 36 of medieval music
and the gorghegghi, 'continuous warbling in the throat', one of many
ornaments which Italian castrati learned). Rhythmic handclapping often
supports it. Ululation, gender-sensitive in Greek usage, 37 was the technical
term for the vocalization of Cybele's galli, 38 in a cult with a musical
core. 39 Catullus makes his eunuch soloist cue a chorus of galli into 'high-
pitched, tongue-fluttering ululation'. 40 What impressed onlookers was
less the ecstatic texts than their instrumental accompaniment, on hand-
drums, pipes, and cymbals, 41 with ancillary clapping 42 and single-note
trills. Though not confined to Cybele music or the very similar Dionysus
music, 43 ululation was characteristic of both, and of other trance or
manic behaviours. 44 Having experienced the nineteenth-century Anatolian
version- 'shrill' and sung by women- the archaeologist Layard concluded
that 'when an Arab or Kurd hears this tahlehl he almost always loses his
senses through excitement and is ready to commit any desperate act', 45 an
illuminating comment on the psychology of self-castration. 46
237
Richard Wz'tt
Eunuch singers are also implicated in a second Near Eastern vocal

idiom, tEpE:tt~Etv, 'chirruping'. Originally this meant 'twittering', like
a bird: cross-dressing cult dancers were said to 'twitter' and 'chirp'; 47 and
Italian castrati were recommended to listen to and imitate birdsong. 48
In Greek musicians' parlance, 'chirruping' came to mean something like
'doodling wordlessly at the octave'. A Byzantine pseudo-classical verb
EV'tEpE'tt~Etv was coined for the singing of castrati. 49 Elsewhere tEpni~nv
describes the 'lewd ditties' of an 'effeminate' Hellenistic male actor 'in soft
raiment' miming classic women's roles; 50 and the low-powered singing,
'turning to childish treble' ,51 of decrepit old men, a category with whom
eunuchs were lumped.
The acoustics and melodic use of 'chirruping' are further illuminated
in a late, classicizing epigram:
... When your plectrum strikes the top string at the far right (of the seven
strings constituting the octave), the bottom note at the far left vibrates
spontaneously, with a thin squeak (A£7ttov 1motpi/~oucra). What is happening
here is t£p£ttcrj.ta-resonance. 52
We need to remember that Near Eastern musical theory and practice

divided the territory across which the human voice melodized into two
zones, male and female. 53 Biological males occupied their own, paramount
vocal territory ('bass'); women and pre-adult males an ancillary territory
('treble'). For both cultural and physiological reasons, an adult male singer
woulq avoid trespassing on female territory. Similarly, if women (or boys)
had to sing the same note series as a group of adult males, they shadowed
the melody at the interval of the octave, so important in ancient music. 54
It was this gendered system that castrato tEpE'tlcrlla infringed.
'Chirruping' combined various strands. It was associated with stringed
instruments, 55 especially those regarded as 'Asiatic': the two-stringed
cSixopcSov, ancestor of the Turkish baglamas, 56 and the 'many-stringed'
Phoenician harps, 57 one of which was actually called cr1tacSt1;, 'eunuch'. 58
In describing David as 'chirruping' (£tEpE'tl~Ev) the Psalms to the ten-
stringed harp (\lfaAt1lptov), the Patriarch of Jerusalem was using the
mot juste; 59 in another context, Roman law, the \lfaAt1lptov could imply
'unmanliness'. 60 Only harps could play a melody's male and female form
simultaneously, in parallel octaves. 61 The harpist could thus shift the
melody from the low zone (male) to the high zone (female) and back
at will; a leap which the human voice too could make. 62 This shift was
traditional for Levitical cantors performing Psalm 46 in the style alamot'
variously interpreted as 'falsetto', 'singing at the octave', and 'hidden'
(i.e. 'feigned'?). 63 They were accompanied by a Palestinian relative of the
238
The other castrati
crnaot~, the 'throaty' neve!, 64 which, in the book of Amos, accompanies

effeminate 'chiefs' oflsrael singing to amuse themselves. 65 A Syrian writer
connects harp 'tEp£ncrjla with piping (Kpoujla) and 'debauched singing'. 66
Classical Arabic music had a comparable idiom, sayhat ('outcry'), where
'one part of the verse is sung in the high octave, the other part in the
low octave'. 67
'Chirruping' was an intimate act of mouth music, 'keeping the voice
inside oneself', as in tuning a lyre; 68 or in humming, 69 whistling/ 0
vocalize,7 1 and the relaxed register-shifted crooning still heard in Greece
or India, 72 and popularized in western music by Rudy Vallee73 and,
on occasion, Elvis Presley.74 Lastly, 'chirruping' was, although not as
good as 'real singing',7 5 a pleasant enough sound, suitable for harvesters
relaxing. 76
Metrically, Anatolian castrati typically worked with the tumbling
syncopations of the rapid three-bar 18-beat 'galliambic', paralleling Chinese
'martial arts' music with its clashing cymbals and violent accelerations.77
Historical distribution
Castrato singers-and-dancers, like other eunuchs, belong to the 'high
civilizations' of Far Eastern Asia and Western Asia. In China, where court
eunuchs date from at least the eighth century BC, 78 castrati sometimes
enjoyed high social status. Particularly celebrated, in the secondary Han
capital, Luoyang, were the 'Palace Attendants of the Yellow Gates', a cadre
of castrati aged from 10 to 12, who led the chu yi ('banishment of
pestilence') ceremony, surviving under the Tang, for the passing of the Old
Year.7 9 Holding twirl-drums, twelve teams of ten boys in animal costumes
sang a simple leader-and-chorus chant to ward off evil. A very similar
fertility-cum-longevity ritual took place in Korea on the Night of the
Twelfth Moon (New Year's Eve), three hundred palace eunuchs chanting
prayers and swinging blazing brands. 80
In about AD 1500, on imperial orders, the Chinese eunuch Dai Yi
taught at least one other eunuch- Huang Xian, from Guangxi province-
the country music repertoire xumen zengchuan; this presumably included
songs. Fifty years later, Huang compiled a tablature (Wugang qinpu),
one of many such, for the seven-stringed zither qin. 81 In his own words,
'this clearly shows that there was an orderly lineage among teachers and
friends': a long, solid tradition of eunuch dilettanti versed in music.
A quite different eunuch identity emerged when K'un-ch'i art theatre
became popular in the Qing dynasty (1644-1911). Professional castrati
established a monopoly of stage vocal music, and most court actors were
palace eunuchs: seven hundred or so at any given time. 82
239
Richard Witt
In Indian musical culture, 'sweet-voiced eunuchs' already feature in

the Sanskrit Natya-shdstra, dated between the second century BC and the
third century AD. 83 Barred from the main sacrifices, 84 they assisted nqt
as cantor-priests but as minor attendant vocalists. Their acquired vocal
flexibility gave them the edge in one of the two main classical types: the
florid 85 'lively' song (khayal). This ranged over two and a half octaves, far
beyond the standard male compass, contrasting in every possible way with
the reflective 'masculine' song dhrupad. 86 Its successor is perhaps today's
hijra chants to spur on candidates for emasculation. 87
The Near Eastern epicentre was always Anatolia: 'Outside of Asia
Minor the eunuch priest is rare.' 88 Though the institution of eunuchs
was conventionally associated with Nineveh and Samurramat's reign, 89
there seem to have been castrati musicians in Anatolia and Mesopotamia
several centuries earlier. 90 In lshtar's temple, the Akkadian assinnu - in
Sumerian, kurgarru, perhaps identifiable with the galli - were chanters
of dialogues and managers of ceremonial. 91 The ongoing MELAMMU
database gives promise of relevant information. It classifies Ishtar's
devotees into 'effeminates' and 'eunuch devotees' or kurgarrulgallos
(3.2.2.1 and 2), and has entries for 'self-mutilation' and 'intoxicating
music and dance' (3.2.3.4 and 5); for 'self-castration' and 'transve<s>tism'
(3.2.4.1 and 2); and above all, for 'choirs of castrated singers' (3.3.5.1),
implying that eunuch singers were subject to the same strict codes as
eunuch palace officials.
Cross-dressing musicians 92 appear in Anatolia, as in Minoan Crete, 93
Egypt, 94 and classical Greece. 95 Ur-Nashe96 is just one of many identifiable
temple singers portrayed as women in Sumerian reliefs: Though gender
ambivalence in dress did not automatically indicate eunuch status, Ishtar
could certainly be the patron of eunuchs. 97 Furthermore, galli cross-dressed
as a highlight of the spring equinox festivities at Hierapolis98 (discontinued
only in AD 502). They simulated women walking, and wore 'falsies', as did
Elagabalus (218-22), accomplished singer-and-instrumentalist and priest
of the Elagabal cult at Emesa, just south ofHierapolis. 99
In Egypt the presence of singing eunuchs is still to prove. 100 They would
not be relevant to Israel, where emasculation was repugnant, were it
not for the first three chapters of Daniel. Here the seemingly historical
merges with myth and the miraculous. Did Maccabean propagandists write
this narrative in 168-165 BC, or is it essentially contemporary with
the events it relates (605-535 Be)? Four male children of the Israeli
nobility - good-looking, wise, and 'without blemish', led by Daniel
(a name from Phoenician legend) - are recruited to Babylon's civil service
by Nebuchadnezzar's 'prince of eunuchs'. They are rinsed of Israeli
240
The other castrati
identity by being given new, Assyrian names and clothes; Daniel finds
'tender-compassion' 101 from the king's own sa resi. Were they regendered,
as well as renamed and reclothed? Jews took it for granted that the three
companions - and, presumably, Daniel himself- had been eunuchs. 102
Daniel's text twice hints at eunuch singing: in the Soliloquy of Daniel
himself (2.20-3), and in the second Miracle Tale (3.1-30) of his three
companions. The Soliloquy, accepted as canonical, comprises a blessing
(2.20-2) in psalm metre, therefore sung, and a freer verse of thanksgiv-
ing (2.23), declaimed. In the Aramaic original of the Miracle Tale,
the companions in the oven do not sing. But in the Greek Septuagint
translation, with its own dating problems, there is a triple 'apocryphal'
addition between lines 3.23 and 3.24 of the original: the 'Prayer of Azariah'
(apocr.add. 1-22), declaimed; a short prose narrative; then a long paean
of blessing (apocr.add. 26-68) by all three companions in unison, in an
antiphonal style resembling certain Psalms. (Under its Latin incipit of
Benedicite, the Song of the Three Children became part of the Gallican
Mass, 103 sung by a deacon or by young trebles, leaving a rich legacy to
European music). It must remain an open question whether there were
Jewish castrati outside or inside Israel, and whether or not the Benedicite
was a later invention.
For ancient Asia Minor, Greece, and Italy, and for the axial capitals of
Rome and Constantinople, evidence remains sparse. With one exception,
Timotheos' Persae, 104 no classical text prepares us for Euripides' coup de
theatre (lines 1369-1502) in Orestes- the climax, musically speaking,
not only of this play, but of its poet's lifelong efforts to escape from the
normative Greek male timbre and explore the singing voice of single non-
standard characters, 105 effectively inaugurating the operatic aria. 106 He is
already experimenting with this technique of monody when introducing
a 'fledgling' voice a third of the way through Alkestis. 107 By Orestes, he
has developed it beyond recognition, into a long, brilliant aria 108 (in
lieu of a messenger speech) articulated, just as in Timotheos, through
complex lyric metres. This 'showstopper' is sung by a character who
appears nowhere else in the play, and who all but outrages the Hellenic
proprieties. Established, even before his arrival, as Phrygian, he quickly
identifies himself as a minor Persian court official, a punkah-wallah. If
modern commentators are undecided whether to regard him as a eunuch or
simply a slave, Euripides himselfleft his audience small margin for doubt.
The mere mention ofPhrygia was enough to induce a knee-jerk association
with castration for some viewers. 109 Even if the new character had never
opened his mouth, his dress language gives him away. Euripides explicitly
tells us he had slip-on deerskin shoes, and an official fan - a notorious
241
Richard Witt
symbol of gender ambiguity 110 - which he displays onstage once if

not twice.U 1 Judging by the Magi in Byzantine art, not to mention
the costume-design miniatures for the eunuch Chaereas in the Vatican
Terence, 112 he would also have worn Oriental headgear, two-piece tunic
and trews (shalwar). The scholiasts' consensus is that this was a eunuch.
They point, unanswerably, to the high voice, to line 1570 and, above
all, line 1528 ('neither man nor woman'). 113 And once the Phrygian gave
tongue, his creator- within the constraints of telling a thrilling tale- did
not let the audience forget his gendering and origin. The empathy with
Helen at her female spinning, the kiss-curl, the 'womanly' cowardice, and
a passing reference to Ganymede, all act to blur gender. The adjectives
~cip~apoc;, ~£voc;, '18atoc; and <l>pu~/<l>puytoc; are piled high, and the
illusion of experiencing native Phrygian music is sustained by a series of
references to Asiatic styles.
Musically, the character was a young male professional vocalist with
a voice of extraordinary flexibility and endurance. There are two pos-
sibilities. Either he actually was a castrato: it was perfectly possible in
Alkibiades' Athens to find a eunuch for a job of work, 114 from a slave-
supplying area like Thrace, Syria, or Phrygia. (This seems to me at least
as plausible as the theory that the hard-worked actor who had just left the
stage dressed as Hellenic Elektra returned in full Phrygian fig to sing the
Barbarian.) 115 Or he could give a very good imitation of a castrato- 'not
a eunuch, but merely doing the work of a eunuch', to adapt Hippolytus 116
- just as in Rome actors playing young men shifted to a tremulous or
effeminate voice whenever his part quoted the words of an old man or
a woman. 117 A statuette found in South Russia seemingly shows an actor
personating a gallus. 118
The Orestes aria was the jewel in a celebrated score, of which a scrap,
maybe, survives. 119 Like Farinelli, the singer worked with contrasted
sections: 120 six in all, with faultlessly spaced rests to take breath. (I time
this section at 24 minutes in performance: about the same as Bri.innhilde's
closing scene in Gotterdammerung, 9 minutes longer than the eighteenth-
century castrato aria Queltusignolo che inamorato, and a display of vocal
athletics that gives the lie to the general scepticism in antiquity about
eunuchs' staying power.) The voice part lay cruelly high, 121 with a wide
dynamid 22 and thematid 23 range, calling for the utmost bravura. As in
Act 3 of Tristan, violent shifts of metre represented febrility alternating with
exhaustion- the swings of mood 124 and 'unmanly' tears 125 conventionally
ascribed to the castrato. Virtuoso effects included agonized repetition,
rhythmic tension, 126 and alliteration. Narrative and comment intertwined,
as in a Christian Passion cantata, calling for consummate 1m6Kptcnc;.
242
The other castrati
Catullus 63, the 'Attis ode' in galliambic rhythm, 127 makes an interesting
foil to the music from Orestes. Declaimed, not sung, 128 it too was designed
in aria sections, had a ferocious mood-swing from ecstasy to despair -
furioso to maestoso, in the poet's own dynamic markings 129 - and was
spiced with references to Asiatic music. 130 The illusion of actual music
was maintained by percussive play with the letters d (or t), s, and 1,
imitating respectively the pattering frame-drum, the sibilant sistrum, and
the glottal trill of ululation. Catullus went one better than Euripides in
representing the worshippers' psychic state by obsessive repetition; and
added a Roman touch of his own, a spate of elisions building up to and
away from a climax.
Catullus would have been in his early twenties at the date of the 'Bona
Dea' scandal, when the young Clodius 'lowered his virility and heightened
his voice'. 131 As the Greek biographer tells it, 132 'there were no males
present'; Clodius, cross-dressing as a professional woman harpist, was
'a beardless youth'; and one of the maids spotted the difference of his
vocal timbre from that of a biological woman. The next generation but
one saw an extended vogue for light, high voices among public speakers,
particularly actors. 133 The light, high delivery, which Cicero himself had
experimented with, to his contemporaries' annoyance, was classed as
an 'Asiatic' fashion. Adjectives chosen to pin it down were 'soft(ish)'
(mollis, mollior), 'womanish' (muliebris, ejfeminata), 'slender' (gracilis),
'broken' ifracta), and even 'running riot' (exultans), recalling descriptions
of western castrato timbre. Quintilian, fascinated as well as appalled,
wrote of'a transparent, shot-silk style' that tended to emasculate whatever
it dealt with. 134
It is from the dying days of this 'Asiatic' fashion that Juvenal's sixth
Satire dates. His savage twelve lines attacking some Roman matrons'
preferences for gelded males is famous, but it has gone unnoticed that
the subsequent six-line coda, dripping with innuendo, can and should be
read as referring to a castrato:
If music's what she fancies, where is the pop singer who'll be able to keep
his cock-pin stiff for long? She's always got an organ or two in her hands;
a swathe of jewels glitters across the sound-chest; the strings throb to the
curly plectrum her soft Hedymeles gets to work with: it's him she hugs, him
she treats herself to, lavishing kisses on his ingratiating plectrum. 135
The language here is sexually charged, especially 'curly', 'plectrum', and
'gets to work' .136 For anyone knowing Greek, 'Hedymeles' would have
evoked not only upmarket literary models - Sappho, Anacreon - but
smooth skin and caressing melodies, a good advertising ploy. 137 If the
243
Richard Witt
music-loving Hadrian discouraged eunuchs, a later Roman emperor's

'hard man', Plautianus, was having both boys and men castrated, partly to
provide his daughter with reliable instructors in, amongst other subje.cts,
'music appreciation'. 138
Meanwhile in Anatolia, for a thousand years or more, the worship of
Kubab-Cybele had been running riot from Chios and Lydia-Phrygia in
the mountainous west to Hierapolis in the flat Mesopotamian east. 139
Eventual Roman tolerance of mystery religions merely confirmed Cybele's
potency. The cult had a musical core, the passion narrative of Attis the
young castrato, and its deafening jam sessions made many converts,
despite the cultural resistance apparent in the legend of Anacharsis and
in the banning of galli from the temple at Eresos on Lesbos (and of
women intending to 'behave like galli' from its precinct). 14° Cybelism,
the 'ululating cymbal', was a major target for the Apostle Paul: 141 it had
already become 'ecumenical'. In Roman times the castration-inducing
liturgy was a highly-structured two-day event. Musically it made use of
multi-stringed instruments, pipes, hand-drum, panpipes, vocalization
at the octave, ululation, and stamping feet. Its imagery threw the high
castrato tessitura into relief by mannered references to the lion, the animal
typical of low voice pitch (~apu~pOj.!O~). Genre epigrams describe how
a eunuch priest of Cybele is trapped in open country by the Phrygian
lion, Rhea's mascot, and is delivered de ore leonis -like the eunuch Daniel
- by performing the correct magic, frantic beating of the cult drum. 142
The Cybele texts were chanted by large numbers of people in trance, led
by a rambling eunuch core (the galli) and a solo vocalist, an Attis-figure,
not to be identified with the supervisory archigallus. 143 (The AStarte cult
music at Hierapolis followed a similar hierarchical pattern.) Nero sang an
Attis-hymn, possibly of his own invention, to a stringed instrument, but
not by any stretch of the imagination could his voice have been mistaken
for a eunuch's, since it was 'short-winded and muddy' .144 The opening of
an Attis-hymn survives; 145 and with it a Phrygian dogma, almost Sufic
in tone, that by wearing 'extraordinary dress' and singing 'a song of the
greatest mysteries' to harp accompaniment, the performer renders the
hymn 'subliminally'.
Attis music reappears in a group of epigrams in the Anthologia
Palatina, 146 and a common thematic stock - a lock of hair dedicated,
a danced journey, accompanying percussion or ululation, refreshment
from the chase - seems to reflect an actual 'passion play' with music. The
earliest of them, by Rhianos, is also the most interesting: it differs from the
ruck in that the central figure is not Attis himself-herself but 'Achrylis the
girl from Phrygia'. :AxpuA.t~ looks like a variant or a mistake for :ApxuA.t~;
244
The other castrati
and there was certainly a male name ApxuA.O<;, also accented on the final
syllable. Was this 'girl from Phrygia' a Proustian Albertine, a regendered
boy -like the singing Attis, or like Euripides' singingpunkah-wallah? Was
'she' yet another castrato (or 'born eunuch') singer? 147
A wholly unexpected transcultural parallel to the Phrygian cult is the
music of the Skoptsi, the secret sect that spread across Russia, from about
1770 onwards, surviving even into the first Stalinist years. 148 Skoptsi
ritual included hymn singing, exhortations, and frenzied dancing which
culminated in spontaneous castrations. 149
It is worth asking why the castrato singer did not flourish in the
sophisticated, 'para-Byzantine' climate of the Caliphate. Certainly Islam
had, like China, eunuch (khisiyan) scholars knowledgeable about singers
and singing, as witness an anecdote of the famous Abbasid musician ibn
Jami'. 150 Incognito, he played a composition of his in the palace music
salon at Baghdad, with its partitioning veil. Hardly had he finished, than
half a dozen eunuchs emerged from behind the veil. They had been sent
to ask what the piece was and who it was by. By me, said the composer,
whereupon back they went, and out came the Head Eunuch, shouting
'Liar! It's by ibn Jami'!' Nevertheless, the 'praiseworthy voice' was firm,
pure and bleached of timbre 151 - thus excluding vox tremula - and was
not on any account to be soft and effeminate 'like that of a young married
woman', 152 so that the elegant gentleman troubadours of the Arab courts
had no time for eunuchs. Artistic taste was secondary anyway, for the
great barrier was ideological. Mohammed - intermittently - viewed all
music as 'of the Devil'; 153 he had expressly forbidden mutilation of the
human body; 154 and eunuchs were included in his formal curse upon 'men
who imitate women' .155
In the Young Rome at Constantinople, by contrast, true castrato singers
were well dug in by the end of the fourth century at the latest. The
first we hear of them is the empress Eudoxia's court favourite, Brison,
staunch opponent of Arians and close friend of John Chrysostom. 156 The
Arians had been staging weekend late-night protests with music, drawing
on a repertoire of hymns with ideological content (Arian theology)
and ideological form (Antioch-style antiphony). Brison had to organize
a competing all-night sing-in, as leader of the Homoousians' UllVCflOOL 157
In the upshot, the emperor banned public performance of the evidently
contagious Arian hymns. Profiting by experience, Chrysostom now com-
manded Brison to see to it that Brison's own singers- including castrati
like himself? 158 -were trained in antiphony, so that their Salvation Army
should steal the enemy's thunder. Antiphony may have been a Syrian
speciality, like metrical sermons; and we should also reckon with the
245
Richard W'itt
musical politics ofEdessa. This was a city with, like Nisan, a great tradition
of singing, harping, and drumming. It was home to the musical settings of
those 'bittersweet' songs of Bar Daisan which St Ephrem failed to drown
out with Christian canticles and responds sung by his own choir. 159 At its
celebrated spring festival, the old heathen tales were sung all night long by
citizen vigil-keepers. No eunuchs they (or at least not after about the year
AD 200, when with king Abgar's conversion to Christianity the penalty for
self-emasculation in Syria and Urhai was the loss of a hand). 160 However
they bared their genitals, like Attis, and followed their leader, the Dancer,
with 'singing, shouting' (i.e. ululation), and 'lewd behaviour'. 161
Were castrati singers like Brison the rule, or the exception, in the first
four centuries of Christian Byzantium? Perhaps to start with they were
perceived, like chanting nuns in Greek Orthodoxy today, as a necessary,
but controversial and uncomfortable innovation. In the long run, they
contributed creatively to the evolution of church music, 162 developed
into a guild of cantors jealously guarding its secrets; 163 and provided
an acceptable solution to the logistic problems caused by muting the
female voice in church. Moreover, of the 'philanthropic' orphanages
and monasteries founded at Constantinople, particularly during the
Macedonian dynasty (867-1056), one or two were eunuch training
centres. 164 The Lazarus monastery was reserved for eunuchs by Leo
VI, 165 himself no mean church musician, 166 and was visited officially by
emperors. A sixth-century foundation, the Orphanage of St Paul with
its choir school, was given a new lease of life by Alexios I Komnenos
(I 081-1118), 167 foreshadowing the Naples orphanages, those first
European conservatoires. 168 On the Feast of Lights its choristers sang
for the court: the music was simple, an antiphonal melody chanted
three times. 169 The teaching communities of eunuch mbnks organized at
Thessalonike by the 'sweet, witty, thrifty' abbot Symeon surely lost no
opportunity to perfect their 'kalophonic' chanting. 170 Eunuch bishops,
too, inevitably shaped musical practice when they intoned the liturgy:
notably in Thessalonike-Ochrid; 171 on Leukas and throughout tenth-
century Greece; 172 and at Constantinople itself, where Germanos I, the
first of a series of eunuch Patriarchs, 173 was not only a hymnographer of
distinction, but a practical musician who 'used melodies and canticles
(qaf.La·ta) to soften the rigours of all-night fasting'.
Conditions thus favoured the rise of a school of castrati cantors. A key
text is Theophylact of Ochrid's brief treatment of music in his In Defence
of Eunuchs. 174 'What is so criminal', asked Theophylact broadmindedly,
'about castrati in church chirping up with a brothel tune sanctified
by Christian lyrics?' This passage attests eunuch singers' experimental
246
The other castrati
fusion of plainchant and secular melodies, a key process in European

music history - contrafactum, as it came to be called - just at its very
beginning. 175
A century later, Theodore Balsamon, Patriarch of Antioch in (once
more) Syria, remarked, prior to condemning their 'florid and theatrical
chanting', 176 that all cantors ('Jfall:tm) were now castrati, and that in days
of old the regiment of cantors consisted not only of castrati, but also
those who were not 'such' .177 In other words, by the late seventh century
there were substantial numbers of castrati (made, not begotten) among
church singers in Constantinople, and by the mid-eleventh century they
had cornered the market. The beardless singers of Middle Byzantine
wall-paintings bear out this conclusion. 178
A notable musical 'happening' of early October 1147 confirms Balsa-
man's testimony. With the Second Crusade cooling its heels outside the
walls of Constantinople, Manuel I had sent a team of clerics to Louis
VII's camp, to celebrate an 'ecumenical' mass in honour of St Denys. The
western observer noted: 179
These clergy differed from our own in wording and in order of service. Yet
they made a favourable impression, so sweet was their handling of their
voices. For there was a mixture of voices - the more robust mixed with the
slender [gracilis], or in other words, castrato voices mixed with male voices
(for many of the clerics were castrati) - and this was very soothing to the
spirits of the Franks.
He added, in a faint twelfth-century echo of the wild leaping of the
galli: 'They made the onlookers jocund by the seemliness of their body
movements, and their restrained clapping of the hands and bending
of the limbs'. Was the mixed timbre of the emperor's picked choir the
sound which Ignatii of Smolensk heard in the coronation acclamations at
St Sophia in 1392 and described as 'singing of weird beauty'? 180
In the fledgling Rus state the Byzantine princess Anna of the Khazars
brought her own choir with her to her marriage in Kiev in 988 or 989. 181
There is no way of telling whether it included castrati. But one cannot
imagine Kievans taking kindly to castrato euphonica; and in the spiritual
support team later dispatched to Vladimir, the 'numerous priests, deacons,
and cantors' were all Slavs. 182 Thereafter, Russian Orthodoxy drifted
away from the Constantinopolitan rite; 183 and with the lone exception
of an enterprising castrato cantor named Manuel, who came from
Constantinople to Smolensk in the later twelfth century and founded
a singing school, 184 Byzantine castrato singing seems to have made no
headway in the vast Slav hinterland. Even in multicultural Thessalonike
of the Palaiologoi, the surprising phenomenon was not castrati but a choir
247
Richard Witt
of women as well as men, with the two sexes singing in antiphony. 185
Castrato singers are missing from the archives of Ragusa, and, still more
significantly, Venice. The castrati - or falsettisti - of the Papal choir were
almost certainly due to political contacts with Spain, where a fashion for
falsettisti dimly seems to have evolved out of the late eleventh-century
Mozarabic liturgy. 186
We have now seen castrato vocalists performing an impressively wide

range of roles in contrasting Eurasian cultures: as boon companions,
as operatic stars, as music teachers, as dilettanti, as votaries in frenzied
Anatolian cults, and as cantors in decorous Byzantine worship. Plainly it
is just not true that the castrato was common only among Italian singers
of the seventeenth-eighteenth century. The evidence is anything but
homogeneous, and I do not think it will be added to substantially. 187 But
its wide scatter in time and place indicates that, in the complex web of
human song, the voice of the eunuch was considerably more prominent,
and considerably more influential, than has yet been recognized.
Notes
1 Especially Habock 1927; Heriot 1956; Walker 1980; Barbier 1996.
2 Der Neue Pauly 1998, 4.256-8, ignores eunuch musicians.
3 Guyot 1980, 57: 'Bei den musikalischen, tanzlerischen und schauspielerischen
Darbietungen haben die Eunuchen sicher auch gesungen' (my italics).

4 Sawa 1989, 119.
5 See Moreno 1988.
6 Shakespeare, Twelfth Night 1.2.56-9.
7 For Xavier Bichet (1771-1802), 'whereas the eunuch enjoys less vital energy,
the phenomenon of life develops in him with more fullness' (Oster 1990,
#8547).
8 The 'impolite', non-Attic, term was crmHirov. At Naples the young castrati
were called di pulitezza, 'the eunuchs': Barbier 1996, 44.

9 Barbier 1996, 12.
10 See Anderson 1990, 307-10.
11 See Walker 1980, 875.
12 Barbier 1996, 241.
13 This skill reached its zenith with the 18th-century castrati, whose mean
range was something like g or a (bottom of the bass stave) to between f" and d'"
(top of the treble stave and above).
14 Babr. 103.5; c£ Hesych. s.v. t£p£'ttcrJla'ta.
15 Dio Chrys. Or. 62.6; cf. Guyot 1980, 58 n. 58.
16 'Certain notes he called his "fourth voice" - strange, sexless, superhuman,
uncanny': Emma Calve on the castrato Domenico Mustafa (1829-1912); c£
248
The other castrati
Bergeron 1996, 174-5, on Moreschi. 'His beautiful supernatural voice cannot

be compared with any woman's' for 'freshness and fullness of timbre' and 'silver
purity': Schopenhauer on Crescentini, in Heriot 1956, 119.
17 On the physiological changes defining the castrato voice, see Defaye 1983;
Barbier 1996, 16-18.

18 Arist. [Pr.] 11.34, 11.62; cf. [De audib.] 802al.
19 Though lying close to the octave d-d', well above the modern bass register, the
'normative' male range of classical Greece was papm£pa, 'deepish', 'heavyish'.

2° Cp. Arist. [Pr.] 19.37; Galen 16.608.
21 Arist. [Pr.] 11.62; c£ Quint. Inst. 11.3.91. Actually the eunuch has 'the
larynx of a youth with the chest and lungs and force of expiration of a man':
Raguenet, in Walker 1980. Farinelli's victorious duel of stamina and articulation
with a virtuoso trumpeter is famous: Bergeron 1996, 181.
22 Manniche 1991, 14, discusses possible tremolo and humming on the letter
n in Egypt.
23 Hippoc. [Ep.] 19.
24 Aent6~: Arist. [Pr.] 11.16. The Latin equivalent is tenuis (Pomponius, apud
Macr. Sat. 6.4.12; Quint. Inst. 11.3.32) or exilis (Quint. Inst. 1.11.1, 11.3.15;
Plin. NH 11.51.270), associated with the (undifferentiated) female voice, and
with elderly men. Stated positively, the singer 'uses less of the voice'. o~u~: Arist.
[Pr.] 11.34; 11.62. But at Pl. Resp. 398e the 'effeminate' harmonia is the 'slack'
(i.e. lowish in tessitura?) Auouni.
25 Philostr. VS 8.
26 Arist. [Pr.] 11.58, 11.62; [De audib.] 801a9-10.
27 al-Hasan, in Shiloah (ed.) 1972, ch. 127; c£ ch. 201.
28 Lucian, Essays in portraiture 13.
29 Evelyn, Diary, 1687, 4.547; c£ S.Wilcox to www. Vocalist, 23.9.1994: 'It's
as if they [the 'old-style' mid-20th century counter-tenors] are trying to model

themselves on boy trebles rather than on other adult singers'.
30 Barbier 1996,17,38.
31 Reinhard 1956, 44.

32 Scott 1972, 148; Strassberg 1976.
33 Layard 1854, 79. Kuipers 1999 calls it 'a widespread but poorly described
form of vocalization'.
34 Catull. 63.24.
35 New Grove Dictionary of Music, London 1980, 'Iraq' §3; c£ Catull. 63.28
(trepidantibus linguis). Phonology of'throat singing': Jansson, www-throat.

36 Walter Odington, Summa de speculatione musice (c. 1280-1320), 94: 'thus
called because it is formed by a pumping motion of the throat' (gutturalis dicitur

quia cillenti gutture formatur). The choice of the rare verb cillere ('to pump away',
in a sexual sense) must be intentional.
37 A distinction was made between male ululation (aA.aMil;etv: e.g. Eur. Cyc. 65;
Hel. 1352, mentioning drum and double pipe) and female ululation (oA.oA:ul;etV:
e.g. Hom. Od. 3.450; Aesch. Eum. 1043). The shift from a too may represent
the physiological difference between male and female voices.
249
Richard Wttt
38 e.g. Rhianos, Anth. Pal. 6.173.3. On the galli see also the contributions of
Hales and Lightfoot in this volume.
39 See below, and also Thomson 1997, 373.
4o Catull. 63.24, cueing line 28.
41 Ov. Met. 4.333.
42 e.g. Ov. Met. 4.353.
43 Turcan 1996, 30.
44 In war: Pind. Dithyrambs fr. 61.10; in worship: Ov. Met.1l.l7, and Ap.
Rhod.Argon. 3.1218.
45 Layard 1854, 79-80.
46 C£ Turcan 1996, 64.
47 Kilmer 2000.
48 Zacconi 1596, 58.

49 See below.
50 Lucian, Dancing 2.
51 Cf. Pl. Leg. 664d-70b.
52 Agathias Anth. Pal. 11.352.3-6, adopting the persona of an associate of the
music theorist Aristoxenus.

53 Classical Greece: West 1992, 252; India: Bake 1957, 214; Java: Sachs 1943,
130; China: Sachs 1943, 114-18, 137.

54 Known as 'the answering voice': avti<jlrovo~, -Etv: Pl. Leg. 812a; Arist. [Pr.]
19.18 and 19.19 (c£ 19.14).

55 C£ Landels 1999, 74; Mathiesen 1999, 272-4.
56 Euphro Adelphi fr. 1.34; guaranteed by 1tTJ1Ctt~ 8ixop8o~, So pater fr. 11 K.
57 Maas and Snyder 1989, 147-55; Michaelides 1978, 264; West 1992,
70-80.
58 Poll. Onom. 4.59.
59 Sophronios of Jerusalem Vita Cyri et Joannis 4, PG 87.3385b; c£ Psalm
144.9.
60 mollities: Cic. Har. resp. 21.
61 Maas and Snyder 1989, 150.
62 Orff Carmina Burana, 22, 19; c£ Bergeron 1996, 173, 'yodelling vocals'.
63 See e.g. Sendrey 1969, 122-3.
64 Loaned to Greek as va~A.a. C£ Sachs 1940, 115 £; West 1992, 77 with
note 134.
65 Amos 6.4-5.
66 Lucian Nigr. 15.

67 Sayha: the top note of the melody considered as an octave span.
68 Theophr. Char. 27.15; Plin. NH 11.51.267. C£ West 1992, 68.

69 Pallad. Hist. Laus. 18.
70 Anastasios of Sinai '08m6~ 12, PG 89.201b. Whistle-tone in falsetto:
S. Willoughby, www- Vocalist 1.11.1994.

71 C£ Phld. De musica 18 Neub., where instrumental music and tepettcrJ.ta are
formally contrasted with verbal performance.

72 Described by an Internet contributor as 'interlude hummings'.
250
The other castrati
73 Croon and croodle are onomatopoeic verbs. Musical crooning, usually

associated with Vallee and Bing Crosby, was originally a routine modelled by
a white actor in the 1820s on one of an elderly Negro in Louisville.
74 e.g. in King ofthe Whole Wide World.
75 ConsultArist. [Pr.] 918a30£
76 Aristoph. fr. 339; cf. Babr. 9.4.
77 In the 'refined' style, the voice copies the sostenuto and delicate embellish-
ments of a stringed instrument.

78 Anderson 1990, 16. See also the contribution ofTsai in this volume.
79 Text in the Treatise on Ritual section of the Han Hou su: Bodde 1975, 153.
80 Dolby 1976.
81 See Wellesz (ed.) 1957, 119.
82 Dolby 1976.
83 Cited by Sachs 1943, 159.

84 Mah!tbh!trata 8.45.25.
85 Sangitaratn!tkara 1.4.35.
86 Fox Strangways 1914, 286.
87 Nanda 1990; www-hijra. Hijra music has been grafted on to western pop
by the Indian rapper Devang Patel.

88 Gray in Hastings (ed.) 1912, 581.
89 e.g. Amm. Marc. 14.6.17.

90 Bienkowski and Millard (eds.) 2000, 110.
91 Leick 1994, 159-60, 168.
92 See the contribution of Lightfoot in this volume.
93 Maas and Snyder 1989, 2 and fig. 2a.
94 Manniche 1991, 88-9.

95 C£ Reinach 1926, 139.
96 Brown (ed.) 1994, 153.
97 Leick 1994, 224.
98 See Segal1970, 47, 53, 106.
99 SHA, Heliogab. 26.5, 32.8.

100 Though see Manniche 1991, 93-5 and pl. 56.
101 Non-sexual, and on its own not nearly enough to place Daniel as a eunuch
(let alone a gay).

102 See Danielou 1957, 581.
103 See Hughes (ed.) 1955, 74-6. The legacy includes: the Daniel Plays; St
Francis, Laus Solis; G. Gabrieli, In ecclesiis; Eisenstein-Prokofiev, Ivan grozhni

(Part 2, shots 319-44); Britten, The Burning Fiery Furnace.
104 See Bassett 1931; Janssen 1989.
105 Euadne (a sketch for Wagner's Brtinnhilde) in Supplices; the blinded
Polymestor in Hecuba; Kasandra at her 'black hymeneal' in Trojan Women;

the protagonist of Helen, who musically dominates its opening; the singing
Oedipus in Phoenician Women; and - if indeed by Euripides - the Muse herself
in Rhesus.
106 C£ Pian 1975, 65-8.
251
Richard Wt'tt
107 Eur. Ale. 393-415.

108 Decharme 1906, 361; c£ Bergeron 1996, 176-7, on Handel's Messiah.
109 C£ Greg. Naz. adv. Julian. 1.70.
110 Anac. 21.13; Pherekrates 64; Ter. Eun. 3.5.47.
111 Perhaps at 1385-6; undoubtedly at 1429.
112 Codex Vaticanus lat. 3864, fol. 24, 25, 33; with Ter. Eun. 609, 847, 905-7,
1015-6. See also Barton 1935, 32; Leclercq 1922, 744; West 1987, 39.
113 Dindorf (ed.) 1843, 297, 301.
114 Pl. Prt. 314c. Eunuchs were familiar enough at Athens to be part of
a children's riddle: Pl. Resp. 479c.

115 West 1987, 38.
116 Hippol. Haer. 5.4.
117 Quint. Inst. 11.3.91, objecting on principle.
118 Picard 1930, 5f£ and pl. 1.
119 Pohlmann 1970, 78-82; Landels 1999, 248; Mathiesen 1999, 70 and
120-2.
120 See Decharme 1906, 362. Farinelli favoured a straightforward A-B-A aria
structure: Barbier 1996, 94.

121 West 1992, 270 with note 44.
122 From maximum energetic movement (1416), through 'oriental' languor
(1427-30), to dead stillness (1490).

123 From the enormous canvas of land, underground, sky, sea (1373-9, 1398,
1475, 1495-8) down to the painterly detail of Helen's golden slippers against
Orestes' jackboots (1468-9).
124 1376-9/1381-4, 1395-7/1403-7, 1482-8; Decharme 1906, 362.
12 5 1031-2. C£ Diod. Sic. 17.66.
126 Decharme 1906, 362.
127 Fedeli 1977 and 1978; Johnston 1929; Oksala 1969.
128 Mart. Ep. 2.86; and all modern commentators, e.g. Quinn 1973, 284.
129 Catull. 63.42-9; rabidus (38) ... miseriter (49).
130 Elision here mimicking castration? C£ Anth. PaL 6.234.
131 Cic. Plane. 86.
132 Plut. Cic. 19; Caes. 9-10.
133 Krenkel1947.
134 Quint. Inst. 8, preface 20; c£ 12.10.12: quod procul absit.
135 Juv. 6.379-84.

136 e.g. Copa 2; Hor. Ep. 8.20.
137 Italian audiences had emotive pet names for their castrati: Barbier 1996,
83-6.
138 Cass. Dio 76.14.4-5.
139 Lucian dea Syra 15. C£ Vermaseren 1986, 5.465.
140 Hdt. 4.76; Clem. Al. Protr. 2.24.1.
141 1 Corinthians 13.1.
142 Anth. PaL 6.51, 6.94.6, 6.173, 6.217-20 and 6.237.
143 See Segal1970, 47; Turcan 1996, 43-65.
252
The other castrati
144 Cass. Dio 61.20; c£ Landels 1999, 201.

145 Hip pol. Haer. 5 .4; <j>tA.ocro<j>ouJ.!eva 5 .1.
146 Anth. Pal. 6.173 (Rhianos); 6.217-9, 220 (Dioscurides); 6.234 (Erucius).
147 See for instance Gow and Page 1965, 2.506.
148 See now Engelstein 1999.
149 C£ Tortchinov 1998.
15° abu'l Qasim Isma'll al-Jami', Aghani 6.78-80.

151 al-Hasan, in Shiloah (ed.) 1972, ch. 195.
152 al-Hasan, in Shiloah (ed.) 1972, ch. 213.
153 Farmer 1929, 26.
154 al-Buhari al-Dhaba'i, wa'l-said25; c£ Pellat 1978, 1089.
155 Hadtth of the imam Ahmad abu Huraira; Hadtth of al-Kaba'ir Lith-thahabi:,
who also condemns cross-dressing.

156 Socrates Hist. eccl. 6.8. C£ Moran 1986, 25; McKinnon 1987, 37.
157 This was the post later known as maistor: see Wellesz 1960, 208.
158 Hopfner 1938, 1.1.399, thinks so. Browe 1936, 85, says not, and Guyot
1980, 58 note 58, is inclined to agree with him.

159 See Segal 1970, 34, 37, 53.
160 [Bar Daisan], Book ofthe Laws ofthe Countries (early 3rd cent.), ch. 45.
161 See Segal1970, 106.
162 Theophylact, in Gautier (ed.) 1980, 327.3.
163 Palikarova Verdeil1953, 71; Landels 1999, 173.
164 Constantelos 1968, 88-110; Moran 1986, 24-6.
165 Guilland 1943, 204.
166 Cf. Tillyard 1928/30 and 1930/1.
167 See Moran 1986, 26.
168 See Barbier 1996, 36-7.
169 Philotheos, in Oikonomides 1972, 187.4-16.
172 Liudprand, Embassy to Constantinople 63.

173 Guilland 1943, 202-3. See also Sideris in this volume.
174 Theophylact, in Gautier (ed.) 1980, 323.7-325.8. On Theophylact and his
text see the contribution of Mullett in this volume.

175 See for example Cattin 1984, 1.127-31; Hiley in McKinnon (ed.) 1990,
128-30.
176 Cf. Heriot 1956, 10.
177 Balsamon PG 137.532. 'Not such': 'Mlj tow\hot', a very early instance of
this adjective as a euphemism for deviation from a sexual norm; in contemporary

modern Greek it implies 'adult male homosexual'.
178 Moran 1986, 26.
17 9 Odo, in Berry 1948, 4.68; Moran 1986, 1.
180 See Palikarova Verdeil1953, 34.
181 Chronicle ofNovgorod 4.81-2.
182 Or so claims our source, the suspect Chronicle ofjakim Uoachim]. Palikarova
253
Richard \Vitt
Verdeil1953, 66-8, accepts the evidence of the Chronicle here.

183 Palikarova Verdeil1953, 64-73.
184 Barbier 1996, 8.
185 Timarion 10.
186 The proof text is a Bull of Pope Sixtus V (1545-90): cf. Walker 1980.
Mozarabic liturgy: Barbier 1996, 8.

187 'Early music specialists consistently face the problem of making intelligent
inferences from many different grossly incommensurable sources, and must learn
to live with the inevitable uncertainty' (van der Velde, of Moreschi's 1902-3
gramophone recording).
Bibliography
Anderson, M.M.
1990 Hidden Power: The palace eunuchs ofImperial China, Buffalo, N.Y.
Andreis, J.
1974 Music in Croatia, Zagreb.
Bake, A.
1957 'The music oflndia', in Wellesz (ed.) The New Oxford History ofMusic,
vol.1, 214.
Barbier, P.
1996 The World ofthe Castrati, London.
Barton, L.
1935 Historic Costume for the Stage, London.
Bassett, S.E .
. 1931 'The place and the date of the first performance of the Persae of
Timotheus', Classical Philology 36, 153-65.
Beard, M. and North, J.A.
1990 Pagan Priests, London.
Bergeron, K.
1996 'The castrato as history', Cambridge Opera]ou~nal8/2, 167-84.
Berry, V.G.
1948 Odo de Deuil, de profectione Ludovici VII in Orientem, New York.
Bienkowski, P. and Millard, A. (eds.)
2000 Dictionary ofthe Near East, Philadelphia.
Bodde, D.
1975 Festivals in Classical China: New Year and other observances, Princeton.
Browe, P.
1936 Zur Geschichte der Entmannung, Breslau.
Brown, D.M. (ed.)
1994 Sumer: Cities ofEden, Time-Life.
Burkert, W
1987 Ancient Mystery Cults, Harvard.
Burney, C.
1771 The Present State ofMusic in France and Italy, London.
254
The other castrati
Campbell, M. and Created, C.

1987 The Musician's Guide to Acoustics, London.
Cattin, G.
1984 Music ofthe Middle Ages, Cambridge.
Constantelos, D.].
1968 Byzantine Philanthropy, Rutgers.
Courtney, E.
1980 A Commentary on the Satires ofjuvenal, London.
Dalley, S.
1989 Myths from Mesopotamia, Oxford.
Dalton, O.M.
1901 Catalogue ofEarly Christian Antiquities, London.
Danielou, J.
1957 'Daniel', Reallexikon for Antike and Christentum 3.575-85, Stuttgart.
De Groot, J.J.M.
1892-1910 The Religious Systems of China, Leiden.
Decharme, P.
1906. Euripides and the Spirit ofhis Dramas, New York.
Defaye, Ph.
1983 Les Castrats, aspects phoniatriques, Unpublished dissertation, Limoges
University.
Dindorf, G. (ed.)
1843 Scholia Graeca in Euripidis Tragoedias, Oxford.
Dolby, W.
1976 A History of Chinese Drama, New York.
Engelstein, L.
London.
Farmer, H.G.
1929 History ofArabian Music, London.
Fedeli, P.
1977 'Dal furor divino al rimpianto del passato', Giornale Italiano di Filologia
29,40-9.
1978 'Struttura e stil del monologo dell'Attis', Rivista Filologica e d'Istruzione
Classica 186, 31-52.
Ferguson, J.
1985 Catullus, Lawrence, Kans.
Fox Strangways, A.
1914 Music ofHindostan, Oxford.
Friedlander, L.
1895 juvenalis Saturarum Libri V, Leipzig.
Gautier, P. (ed.)
1980 Thiophylacte d'Achrida, I, Discours, traites, poesies, Thessalonike.
Gow, A.S.F. and Page, D.L.
1965 The Greek Anthology, Cambridge.
255
Richard Witt
Granet, E.
1926 Danses et legendes, Paris.
Guilland, R.
1944 'Fonctions et dignites des eunuques', REB 2, 185-225.
Guyot, P.
Stuttgart.
Habock, F.
1927 Die Kastraten unde ihre Gesangkunst, Stuttgart.
Harich-Schneider, E.
1973 Japanese Music, Oxford.
Hastings, F. (ed.)
1908 Encyclopedia ofReligion and Ethics, vol. 2, Edinburgh.
1912 Encyclopedia ofReligion and Ethics, vol. 5, Edinburgh.
Heriot, A.
1956 The Castrati in Opera, London.
Hopfner, Th.
1938 Das sexual/eben der Griechen und Romer, Prague.
Hughes, D.A. (ed.)
1955 The New Oxford History ofMusic, vol. 2, London.
Jaffrey, Z.
1996 The lnvisibles: A tale ofthe eunuchs ofIndia, New York.
Janssen, T.H.
1989 Timotheus' Persae: A commentary, Amsterdam.
Johnston, M.
1929 '"Jazz" in Catullus', Classical Weekly 23, 24.
Kazhdan, A. (ed.)
1991 The Oxford Dictionary ofByzantium, 3 vols., New York and Oxford.
Kilmer, A.D.
2000 'Music and dance in ancient western Asia', in Sasson (ed.) Civilizations
ofthe Ancient Near East 4.2611.
Kluckhohn,C.
1979 'Myths and rituals: a general theory', in WA. Lessa and Z.E. Vogt (eds.)
Readings in Comparative Religion, 4th edn, New York.
Krenke!, WA.
1947 'Hypothermia in ancient Rome', American journal of Philology 68,
394-5.
Kuipers, J.
1999 'Ululations from the Weyewa highlands (Sumba): simultaneity, audience
response and models of co-operation', Ethnomusicology 4313.
Landels, J.G.
1999 Music in Ancient Greece and Rome, London:
Layard, A.H.
1854 A Popular Account ofDiscoveries at Nineveh, New York.
256
The other castrati
Leclercq, H.
1920 'Daniel', in F. Cabrol and H. Leclercq (eds.) Dictionnaire d'Archeologie
chrt!tienne, vol. 4.1, Paris.
1922, 'Eunuques', in F. Cabrol and H. Leclercq (eds.) Dictionnaire d'Archeologie
chretienne, vol. 5.1, Paris.
Leick, G.
1994 Sex and Eroticism in Mesopotamian Literature, London and New York.
Maas, M. and Snyder, ].Mel.
1989 Stringed Instruments ofAncient Greece, Yale.
McGee, T.J.
1998 The Sound of Medieval Song: Ornamental and vocal style according to
the Treatises, Oxford.
McKinnon, J.
1987 Music in Early Christian Literature, Cambridge.
McKinnon,]. (ed.)
1990 Man and Music: Antiquity and the Middle Ages, Basingstoke and
London.
Mann, S.E.
1984 An Indo-European Comparative Dictionary, London.
Manniche, L.
1978 'Symbolic blindness', Chronique d'Egypte 53, 13-21.
1991 Musicians in Ancient Egypt, London.
Mathiesen, T.J.
1999 Apollo's Lyre, Nebraska.
MELAMMU
1999 Database Project [DOC Version 2.1], Intellectual Heritage ofAssyria and
Babylonia in East and West, ed. S. Parpola, Helsinki University, State
Archive of Assyria Centre of Excellence.
Michaelides, S.
1978 The Music ofAncient Greece, London.
Mitamura, T.
Moran, N.K.
1986 Singers in Late Byzantine and Slavonic Painting, London.
Moreno,].
1988 'The music therapist: creative arts therapist and contemporary shaman',
The Arts in Psychotherapy 15, 271-80.
Moses, P.
1960 'The psychology of the castrato voice', Folia phoniatrica 12.
Nanda, S.
Oikonomides, N.
1972 Les listes de preseance byzantines des IX' et X' siecles. Introduction, texte,
traduction et commentaire, Paris.
Oksala, P.
1960 'Das Geschlecht des Attis bei Catull', Arctos 6, 91-6.
257
Richard Witt
Oster, P.
1990 Dictionnaire des citations franfaises, Paris.
Palikarova Verdeil, R.
19 53 La musique byzantine chez les bulgares et les russes (du IX' au XIV' siecle),
Copenhagen and Boston.
Pelikan, E.
1876 Geschlechtlich-medizinische Untersuchungen uber das Skoptzentum in
Russland, Giessen.
Pellat, Ch.
1978 'Khasi 1', Encyclopedia ofIslam, 4th edn, Leiden.
Pian, R.C.
1975 'Aria structure patterns in the Peking opera', in ].1. Crump and W.P.
Malm (eds.) Chinese and japanese Music-Dramas, Ann Arbor, 65-8.
Picard, Ch.
1930 'Les apprets de 1'ordination du Galle d' apres une terre-cuite d'Odessa',
Revue de l'Histoire des Religions 102, 5-12.
Pighi, G.B.
1932 '11 canto d'Attis', Rivista musicale italiano 39, 34-40.
Pinnock, F.
2000 'Erotic art in the ancient near east', in Sasson (ed.) Civilizations of the
Ancient Near East, 4.2530.
Pohlmann, E.
1970 Denkmaler altgriechischer Musik, NUremberg.
Quinn, K.
1973 Catullus, London.
Reinach, Th.
1926 La musique grecque, Paris.
Reinhard, K.
1956 Chinesische Musik, Eisenach and Kassel.
Roller, L.E.
Sachs, C.
1940 History ofMusical Instruments, New York.
1943 The Rise ofMusic in the Ancient World, East and West, New York.
Sadie, S. (ed.)
1980 New Grove Dictionary ofMusic and Musicians, London.
Sasson, J.M. (ed.)
2000 Civilizations ofthe Ancient Near East, New York.
Sawa, G.D.
1989 Music Performance Practice in the Early 54.bbasid Era, Toronto.
Scholz, P.O.
2001 Eunuchs and Castrati: A cultural history, Princeton.
Scott A. C.
1972 The Theatre in Asia, London.
Segal, J.B.
1970 Edessa: 'The Blessed City', Oxford.
258
The other castrati
Sendrey, A.
1969 Music in Ancient Israel, New York.
Shiloah, A. (ed.)
1972 La perfection des connaissances musicales, Paris.
Strassberg, R.
1976 'The singing techniques of K'un-ch'i and their musical notation',
Chinoperl6, 45-81.
Streeck, U. (ed.)
1993 Das Fremde in der Psychoanalyse. Erkundungen uber das 'Andere' in Seele,
Korper und Kultur, Miinchen.
Sundberg, J.
1977 'The acoustics of the singing voice', Scientific American (March),
82-91.
Tatnell, R.S.
1965 'Falsetto practice: a brief survey', The Consort 22, 31-5.
Taylor, G.
2000 Castration: An abbreviated history of western manhood, New York and
London.
Thomson, D.F.S.
1997 Catullus: A commentary, Toronto.
Tillyard, H.J.W
1928/30 'Em9tva Avo:cr'ta<H!LO:: the morning hymns of the emperor Leo VI',
Annual ofthe British School at Athens 30, 86-108
1930/31 'Em9tva Avo:cr'tacrt!LO:: the morning hymns of the emperor Leo VI',
Annual ofthe British School at Athens 31, 115-47
Tortchinov, E.A.
1998 'The mysteries of the "suffering gods"', lnternational]ournal ofTransper-
sonal Studies 17/2, 149-59.
Tosi, P.F.
1742 Observations sur le chant figure, Paris.
Turcan, R.
1996 The Cults ofthe Roman Empire, Oxford
Vermaseren, M.J. (ed.).
1977-1989 Corpus Cultus Cybelae Attidisque, 7 vols., Leiden.
Walker, T.
1980 'Castrato', in Sadie (ed.) New Grove Dictionary ofMusic and Musicians,
London.
Wellesz, E. (ed.)
1957 The New Oxford History ofMusic, vol. 1, London.
1960 Byzantine Music and Hymnography, 2nd edn, Oxford.
West, M.L.
1987 Euripides: Orestes, Warminster.
1992 Ancient Greek Music, Oxford.
Wiesehofer, ].
1994 Das antike Persien, Munich and Zurich.
259
Richard "W'itt
Wille, G.
1967 Musica Romana, Amsterdam.
Willink, C.W
1986 Euripides: Orestes, Oxford.
Wu Zhao (ed.)
1981 Qinqu jicheng, vol. 1, Beijing.
Wynn, V.T.
1970 'Measurements of small variations in "absolute" pitch', journal of
Physiology 220, 627-37.
Zacconi, L.
1596 Prattica di musica, Venice.
Electronic sources (Internet websites)

www-byzantium = www. tiac.net/ users/wgenest/ ars/ source/society/eunuch.h tml
www-daniel = www.bible.org/ docs/ or/books/dan/ dan3.htm
www-dilbook = www.ltg.ed.ac. ukl ~chrisbr/ dilbookl node 19 .html
www-hijra = www.bme.freeq.com/culture/970101/hijra.html
www-infibulation = www.bme.freeq.com/culture/981115/slave.html.
www-medieval = www.medieval.org/emfaq/misc ('Answers to topics/Voice names')
www-throat = www.jyu.fi/ ~sjanssonlthroat.htm 'A miraculous method of singing:
On throat singing ofSouth Siberia'
www-yodel = www.tfmpage.com/forum/14208.20.13.07.html
www-Vocalist = www. vocalist.org and www.lists.oulu.fi/vocalist
260
INDEX
Abasgia 144, 150 Apuleius, Metamorphoses ch. 4 passim

Abgar ofEdessa 73, 246 Arcadius 161
Achaemenid dynasty ch. 2 passim, Aristonikos 146
145-6 Aristotle 34, 60-1
Adramyrtes, King of the Lydians 4 School of 236
Aelian 35 Armenia 144, 148, 150, 162
Aelius Donatus 144, 150 Artapartes 3 8
Aeschylus, Oresteia ch. 3 passim Artaxares the Paphlagonian 30-1, 35,
Mrica 145, 147, 229 37,39-40,146
Agamemnon, in Greek drama ch. 3 Artaxates 37
passim Artaxerxes I 24, 32-3, 146
Ahasuerus, King of Persia 19, 29 Artaxerxes II 29, 33-4, 40-1
Ahuramazda 25, 26 Arthavaveda 4
Alexander the Great 23, 31, 35, 146 Artobarzanes 35
Alexander I, Byzantine Emperor 167 Artoxares: see Artaxares
Alexander I, Czar 9 asceticism 186-91
Alexandria ch. 7 passim, 146, 166-7 Aspasia, Greek courtesan 35
Amantios 164 Assyria 22, 33, 145
Amenhotep III, Pharaoh 32 Atargatis/Dea Syria, Cult of ch. 4
Amestris 32 passim
Amisus 147 Athanasius of Alexandria 161
Ammianus Marcellinus 5, 161-3 Athenaeus 4
Ammonius 134 Athenagoras 125
Amyrtaios 151 Athens 51, 60, 242
Amyris, wife of Cyrus the Great 38-9 Athoos 35
An Dehai 229 Athos, Mt. 182-3, 190
An Lushan 225 Atossa, wife ofDarius 31, 34
Anastasi us I 164 Attis 76-7, 80, ch. 5 passim, 243-6
Anatolia 72, 78-9, 188, 240, 244
Angeloi, dynasty 185 Babylon/Babylonia 33, 145-6, 240
angels 165-6,187-8,237 Bagapates/Mastabates 31, 40-2
Anna of the Khazars 247 Baghdad 245
Anna of Savoy 204-5 Bagoas, favourite of Alexander the
Annam 148 Great 35
Anthologia Palatina 244-5 Bahuchara Mata 3
Antioch 146, 247 Balsamon, Theodore 247
Antiochus the Persian 144 Basil I 166
Antiochus VII 146 Basil II 184
Antoninus Pius 127 Basil ofCaesarea 161, 180-2, 187,
Appian 146-7 190
261
Index
Basilides 128-9 Carthage 20

Basilina, mother of] ulian 163 Carzimasians 149
Bayandir 79 Cassian, Julius 8, 128-9
beard, false 3 9, 146 castrati 8, 13 passim
beavers 126 castration ch. 1 passim, ch. 3 passim,
begging 72-3, 75-7, 80, 91 ch. 6 passim, ch. 7 passim
Beijing 228, 230, 231 chemical 6-7, 12
Bellona, cult of 92-3 effects 4, 34-5, 61, 90, 106-7,
Beqaa valley 73 110, 111, 162, 180-1, 187,202-3,
Beroea 73 235
Bible: methods 1-5, 107, 124, 147, 202,
Acts 8.26--40 130, 163 235
Amos 239 pre-pubertal 80, 181, 235-6
Daniel 240-1 positive views of 110-112, 128,
Deuteronomy 23.1 107, 130-2; 130, 132
23.2 181; 25.11-12 53 reasons for 5-14, 53-4, 58, 62, 81,
Esther 22, 27, 29, 35-7 93,108,127,145,148,181,202,
Exodus 21.24-5 53 221,230,244
Genesis 37.6 5; 37--48 107; self- 8-9, 71-5, 79-80, 88, 95-6,
40.1 131 106, 112-13, ch. 7 passim, 148,
Isaiah 56.3-5 128, 132, 163, 166, 187,228,230,237,240,246
181 see also 'legislation'
Leviticus 21.18-24 107, 131-2; Catherine the Great 9
22.24 132 Catullus 77, 80, 94-5, 237, 243
Matthew 19.12 8-10, 113, 124-6, Chaereas 242
133-6, 163, 181-2 Chamundeswari 3
Psalms 238 Chariton, Callirhoe 29, 37
Wisdom 3.13 132-3 Chen Dynasty 224
Bishop of Edessa, eunuch 184 Chenghua 228
Bishop of Kitros, eunuch 184 China 5, 7, 10, 21, 149, 151, 226-7,
Bishop of Petra, eunuch 184 235,237,239
Boges 38 Choir, Papal 248
Book of Odes 222 Chongzhen 229
Bougaios 36-7 Christodoulos ofPatmos 188-9
Boxer Rebellion 231 Cicero 243
Brison 245-6 Circassia 148
Buchanan, Robert 19 circumcision 103-4, 112
Byzantium 7, 21, 149-51, 168-9, Cixi, Empress Dowager 229-30
177, 191, 199, 246 Claudian 161
Clement 128, 131, 134
CaiJing 226 Clodius 243
Cai Tuoer 224 Clytemnestra, in Greek drama ch. 3
Cambyses 38 passim
Cambyses II 151 Collins, Joan 19
Candaces, Queen of Ethiopia 163 Combabos 78, 80
Cao Cao 224 Confucius 222
Carchemish 77 Constans 162
262
Index
Constans II 165 Domitian 127

Constantine I 144, 150, 162 Dura Europos 82
Constantine IV 165
Constantine VII 149, 166 Echetus, King 55-6
Constantine IX Monomachos 149 Edessa 73, 246
Constantine the Paphlagonian 149, Egan, Richard 19
202 Egypt 21, 54, 104, 107, 147, 151,
Constantine the Saracen 149 240
Constantinople 144, 148-9, 164-8, ejaculation 10, 190
177,182,188-9,200,205,245-7 Elagabalus 75, 240
Constantius II 161-3 Emesa 240
Corippus 165-6 Enarees, Scythian effeminates 77, 82
Cormon 25 Eonopolites, Andronikos 204
Crete 240 Epiphanius 129
Cubicularii ch. 9 passim erections 190-1
Praepositus Sacri Cubiculi 143-4, desire 202-3
161 Erinyes 52, 55, 58-9
Cybele, cult of 71-3, 75, 77-8, 80-1, Esther 36-37
ch. 5 passim, 237, 244 Esther and the King 19, 20
Cyril of Alexandria 161 Eudoxia 164, 245
Cyrus the Great 5-6, 27, 33, 38, 145 Eugenics 11-12
Cyrus II 39 Eulaios 146
Cyrus, brother of Artaxerxes II 38, 40, eunuchs:
42 adoption 7, 223, 228, 231
and angels 166-8, 187-8, 237
Dai Yi 239 Assyrian 21-2, 36, 53-4, 90, 98,
Daizong 225 145, 149
Damianos the Slav 148 Byzantine 7, 9, 21, 143, 145,
Daniel 240-1, 244 147, 148-51, ch. 9 passim, ch.lO
Danielis 166 passim, ch.ll passim, 245-7
Darius I 24, 29-33, 41 Chinese 2, 7, 21, 33, 145, 147-9,
Darius II 30, 39, 146 151, ch. 12 passim, 235, 239
Darius III 31, 35 and Christianity 8, 8-9, 113, ch. 7
Darwin, Charles 11 passim, 161, 163-5, ch. 10 passim,
David 238 201-3,245-8
Day of Blood 77 court 7, ch.2 passim, 78, 108-9,
Delos 81-2 ch. 8 passim, ch. 9 passim, ch. 11
Demeter 106 passim, ch. 12 passim, 239
Demetrios, eunuch brother ofTheo- Egyptian 104, 109-10, 131, 151
phylact of Ochrid 178-80, 183-4, ethnic origin 108, ch. 8 passim,
186, 191 185,200
Demetrius, Bishop of Alexandria 136 female 4
Demokedes 27 gender 20, 37, 51,76-7, 82,91-2,
Depo Provera 7 96, ch.6 passim, 162, 181, 186-7,
Diegesis Merike 190 191,202-3,205-7,238,240-2
Diodetian 163 Hellenistic 146-7
Dionysius 146 Indian 72, 240 (see also Hijras)
263
Index
Islamic 7, 21, 147, 149, 245 Evergetis, monastery 189-90

Jewish 240-1
(knowledge of) at Athens 51, 242 fan/fanning 34, 41, 241-2
Korean 7, 148, 239 Farinelli 242
Late Roman 21, 143--4, 150-2, Favorinus 236
161-6,245-6 Feng Bao 229
marriage 7, 145, 148, 203, 230 Five Dynasties 226
messengers 37, 146, 166-7, 201, Flaubert, Gustave 20-1
204,207-8 Salammbo 20-1
military 7, 146, 204, 225-7, 229 Forbidden City 21, 228-30
monks 129-30, 183-5, 191, 204, France 8
246 Fu-su 222
Ottoman 21-2, 26, 33, 145,
147-9,235 Gai: see Hegai
Persian 5-6, 19--42, 145-6, 149, Gaius Cartilius Eupulus 97
206,241 Galli ch. 4 passim, ch. 5 passim, 237,
positive views of 110-12, 128, 240,242,244,247
132-3, ch. 9 passim, ch.lO passim Archigallus ch. 5 passim, 244
religious 7, 8-10, 20-1, ch.4 Galton, Francis 11
passim, ch. 5 passim, 106, ch. 7 Ganymede 242
passim, 163-5, ch. 10 passim, Gao Lishi 224-5
201-3,240,244-6 invested Duke of Qi 225
as scapegoats 221, 224 Gaul 71, 96, 150, 163
sexual activity (or lack of) 10, 35, Gavras, Michael 203
38-9, 76,81,91, 162,180-2, Gazanfer 148
190-1,204,222 Genesis Rabba 108
singing 8, ch.13 passim Georgia 148
stereotypes 23, 76, 161-2, 180-1, Germanos I, Patriarch of Constantino-
203-7,222,224,242 ple 165, 246
terminology 3, 23--4, 72, 77, 107, Germany 11
123,124-6,145,221,235 Gerome 25
and women ch. 2 passim, 61, 82, Glanum 96-7
91-2, 146, 162--4, 168, 178, Gnostics 8, 128-9
180-1,187,201-8,223,230-1 Gregoras, Nikephoros 200-3, 208
visual evidence 24, 34, 79, ch. 5
passim, 168-9,247 Hachratheus/Hathcath 37
Eunus of Apamea 74 Hadad 71
Euripides, Orestes 34, 241-2 Hadrian 244
Eusebius, Ecclesiastical History 123, Han Dynasty 221,223, 239
125, 127, 130 Eastern 223
Eusebius, Grand Chamberlain 161, Handel 8
163 harem/s 4, 7, ch. 2 passim, 235
Eutherius 144, 162 Assyrian 36
Eutropius 144, 161 Chinese ch.12 passim
Eutropius, secretary ofJuliana Egyptian 32
Anicia 191 Ottoman 25-6, 28, 32, 39, 235
Evelyn, John 236 Persian ch. 2 passim
264
Index
He, Empress Dowager 223-4 John ofHerakleia, St 200-4, 208

HeJin 223-4 John Moschos 187
Hecate, temple of at Lagina 72 John the Orphanotrophos 149, 151
Hegai 19-20, 36-7 eunuch brothers of 149
Hejian 230 Jorgensen, Christine 10
Helen ofTroy 34, 242 Joseph 5, ch. 6 passim
Hellanikos 34 Julia Mamaea 136
Hermaphrodite/s 80, 90, 97-8, 236 Julian 162, 163
Hermotimos 6, 51, 146 Juliana Ani cia 191
Herms, mutilation of 56 Justin II 165-6
Herodotos 6, 27, 51, 145-6 Justin Martyr 125, 127
Hesiod, Theogony 58 Justinian I 144, 150, 164-5, 181
Hierapolis 71, 73, 78-80, 82, 240, Justinian II 148
244 Juvenal 243-4
Hijras 2-3, 9, 77, 80, 240
Hochmaea 81 Kaifeng 226-7
Homer: Kallikrenites, John 200, 204
Iliad 54 Kallikrenites, Michael 203-4, 207
Odyssey 55-6 Kallinikos 165-6, 169
homosexuality/homosexuals 5-7, 11, Kantakuzene, Theodora, daughter of
76-7,81,105-6,108,131,203 John VI 205
Hongwu 227-8 Kantakuzene, Theodora, mother of
Hormus 229 John VI 204
Hu-hai 222-3 Kantakuzenos, John VI 204-5
Huang Xian 239 Kiev 247
Huizong 226-7 Kombaphis 151
Komnenos,Alexiosl 151,178,180,
1-huan, Prince ofChun 230-1 182,185,188-90,246
Ibn Jfuni' 245 Komnenos, John II 189
Ignatii of Smolensk 247 Komnenos,Manuell 189,208,247
Ignatios I, Patriarch of Komnenian Dynasty 151, 177, 185,
Constantinople 165, 181 200
lnanna: see Ishtar Korea 239
India 5, 7, 72, 77, 239 Krateros 146
lngres 25 Krishna 9
intersex children 12-13 Kronos, castration of Ouranos 58
Intersex Society of America 12-13 Ktesias 27-8,33, 35,38-9, 146, 151
Ionian revolt 146 Kubaba(t) 77-8
Ireland 150
Ishtar 72, 240 Laberia Felicia, Priestess of Cybele
lzabates 151 91-2
Laskarid Dynasty 200
Jia Lu 227 Lausus 164
Jia Yi 223 Lazarus, monastery 246
Jianwen 228 legislation, on castration and eunuchs:
John, Chamberlain 164 Assyrian 6, 53-4
John of Ephesus 164-5 Biblical 107, 110, 130-3, 181
265
Index
Church 123, 129-30, 181 Mardonius 163

Islamic 245 Maria of Alania 178
Lex Talionis 52-5, 62 Mark the Deacon 164
Papal 8 Maschalismos ch. 3 passim
Roman 73,88-9,127,144,150, masturbation 5
181,238,244 Masudi 149, 181
Syrian 246 McDowell, Ephraim 4
Lenin 229 Medinet Habu, temple of 54
Leo, the Sakellarios 168-9 Megabyzus, the, at Ephesus 72
Leo Bible 168-9 Meletios ofMyoupolis 188
Leo I 144, 150 Melito 123
Leo IV 165 Mengi Zi 222
Leo VI 148-9, 246 Merneptah, Pharaoh 6
Leo XIII, Pope 8 Mesopotamia 5, 72, 240
Leontios, Bishop of Neapolis 166-7 Methodius 125
Leontios Scholastikos 165 Methodios I, Patriarch of Constantino-
Leukas 246 ple 165, 187
LiFuguo 225 Michael I 165
Li Hongzhang 230 Michael III 148
Li Lianying 229-31 Michael IV 149
adopted sons 231 Michael V 149
Li Linfu 225 Michael VII Doukas 151
LiShanchang 224 Ming Dynasty 148, 227-9
Li Si 222 Ming Huang 224-5
Li Zicheng 229 Mishael 164
Ling 223. Mithridates VI 146-7
Liu Bang 223 Mohammed 245
LiuJin 229 Money, John 12
Liudprand of Cremona 149 Mozart 235
Livia 88 music 72, 74-5, 77, 181, 225, ch.13
Lomonsov, D. 10 passim
Longqing 228 Mustafa, Dominicb 8
Louis VII 247 mutilation ch. 3 passim; 74, 77, 96,
Lucian 106, 236 and n. 28, 239 103-4, 106, 187
and n.66
De Syria Dea ch. 4 passim Nanjing 227
Lucius, or the Ass ch. 4 passim Naples 8, 237, 246
Lucius, slave of' Syrian goddess of the Narratio de Sancta Sophia 167
Hierapolitans' 73 Narses 144, 181
Luoyang 224,239 Narses, sword-bearer of Justin II 166
Nazis 11
Mamas 144 Nebuchadnezzar 240
M;~,nasses, Constantine 206 Nehemiah 24
Manuel, castrato cantor 247 Neishutang 228 ,
Marcion 126 Neophytos, eunuch monk 204
Marcus Modius Maxximus, Nero 10, 244
archigallus 94, 97, 99 Nguyen An 229
266
Index
Nicholas I, Czar 9 Petroni us, Satyricon 10

Niha (Lebanon) 81 phalli, dedicated at sanctuaries 82
Nikephoritzes 151 Phileotes, Cyril 189
Nikephoros, Bishop of Miletos 202 Philes, Manuel 203
Nikephoros III Botaneiates 182 Philetairos 147
Niketas I, Patriarch of Philialetes 204
Constantinople 165 Philip, Apostle 163
Niketas, St, eunuch 202 Philo of Alexandria ch. 6 passim,
Nineveh 240 130-1, 133--4
Philodemus 76-7
Odo of Deuil 247 and n. 179 Philotheos, Kletorologion 145 and
Oneirocriticon ofAchmet 167 n. 18, 201
orchidectomy 12 Phoberou, monastery 189-91
orgasm 10, 190-1 Photios 181
Orhan, Ottoman Emir 205 Phrygia 241-2, 244-5
Origen 9, 103, 113, ch. 7 passim, 187 Phrynichos 51
orphanages 246 Plato 27
Orphanage of St Paul 246 Plautianus 244
Ostia 87, 92-7 Plotinus 103, 134
Ouranos, castrated by Kronos 58 Plutarch 27, 29
ovariectomy 4-5, 11-14 Polyeuktos, Patriarch of Constantino-
Ovid 87 ple 165
Pompeii 95
Palaiologan Dynasty ch. 11 passim, Pontus 126, 147
247 Kingdom of 23, 146-7
Palaiologos, Andronikos II 204, 207 Potiphar 5, 107-9, 111
Palaiologos, Andronikos III 204 Presley, Elvis 239
Palaiologos, Michael VIII 204, 208 Procopius 144, 150
Palaiologos, Michael IX 204 Pseudo-Kodinos 201, 204, 208
Palladius 164 Ptolemaic court 146
Panionios 6, 51, 146 Ptolemy V 146
Panthea, Lady of Susa 38 Ptolemy VI 146
Paphlagonia 149-50, 185 Pu-yi, Henry 230
Parabiago dish 88, 95
Parysatis 30-1, 33, 39, 40-2 Qianlong 230
Pasargadai 32 Qin Dynasty 222-3, 239
Paul, St 8, 137, 244 Qin Shihuangdi 222
Paul Helladikos 189, 190 Qing Dynasty 229-31, 237
Pearson, Karl 11 Qinzong 227
Pei 222 QuMu 223
Pepagomenos, George 200, 203 Quintilian 89, 243
Pergamum 147
Persepolis 22, 24, 29, 31-3 Radha 9
Persia 7, 144 Ragusa 248
Pessinus 72 Ramesses III 54
Petasakes 39--40 Rhea 76, 80, 244
Peter III, Czar 9 Rhianos 244-5
267
Index
Rhodogune 29 Susa 32-3, 36

Romance, vernacular, Byzantine 201, Suzong 225
205-8 Symeon the New Theologian 185
Aristandros and Kallithea 206 Symeon the Sanctified 182-5, 246
Kallimachos and Chrysorrhoe 201,
206-7 Taiping, Grand Princess 224-5
Livistros and Rhodamne 206-8 Tang Dynasty 224-6, 239
Velthandros and Chrysantza 206-7 Old Tang History 224-5
Romania 9-10 Terence 144, 242
Romanos I 165 Tertullian 125-6, 128
Rome 7, 73, ch. 5 passim, 162 Theodore 164
Russia 9-10, 245, 247 Theodoulos, Metropolitan ofThessalo-
nike 184
Samonas 148-9, 151 Theophanes, Chamberlain 165
Samurramat: see Semiramis Theophanes Confessor 165
Seleucids 23, 146 Theophylact, Patriarch of Constanti-
Selim 148 nople 165
Semiramis 5, 34, 240 Theophylact of Ochrid 9, 161,
Sentences ofSextus 130 164-5, 168, ch.lO passim, 246-7
Shaashgaz 37 Thessalonike 180, 183-4, 246-8
Shakespeare 22 and 42-3 n. 9, 235 Tianjin 230
and n. 6 Tibetheus 3 8
Shang Dynasty 221 Tiglath-pileser III 36
Shao 223-4 Timotheos 241
Shen Yulan 230 Tiridates, beloved of Artaxerxes 35
Sima Qian 223 Tong Guan 226-7
Sisigambis 31 Topkapi palace 32
Sitwell, Osbert 21 transgender 13
Skoptsi 9-10, 245 transsexualism 10-11; 13, 80
slaves/slavery 6, 22, 51, 107-8, transvestism 72, 75, 77, 82, 91-2,
147-51,162,206,241-2 240,243
Smolensk 247 Trough, the 40
Sogdianus 30 Tryphon 146
Song Dynasty 226-7 Turrentine, Lawrence Neil, Judge 6
SongShi 226
Northern Song 226-7 Ur-Nashe, temple singer 240
Southern Song 226-7 Uruk 72
SongofSongs 133-4, 137-8
Sophocles, Troilus 34, 56 Vaishnavism 9
Spain 66, 149, 248 Valens 161
Sporus 10 Valentinians 136-7
Stateira 29-30, 33, 42 Valentinus 128-9
Stephen the Persian 148 Valesians 129
Stephen II, Patriarch of Constantinople Vallee, Rudy 239 ,
165 vasectomy 11-12, 14
Strabo 147 Vashti, Persian Queen 27, 29, 36
Suren, Safavid 31 Venice 248
268
Index
Verdun 149 Xian 222, 225-6
Vietnam 229 Xian, Emperor of Eastern Han
viper 206 Dynasty 224
Vision ofthe Monk Kosmas 167-8 Xianyang 222
Xuande 228
Wang Chengen 229
Wang Zhen 229 Yang Guifei, consort of Ming
WangZhi 229 Huang 225
Wanli 228 Yao Shilian 224
Wei: YongQu 222
duke of 222 Yongle 228
state of 222
Wei Zhongxian 229,231 Zhang, concubine 224
women: Zhang Rang 223--4, 231
Biblical 104-5 Zhao Gao 222-3, 231
Byzantine aristocratic 200, 204-5, Zhao Kuanyi 226
209 Zhao Kuanyin 226
Chinese ch. 12 passim Zheng He, Admiral 229
Clytemnestra ch. 3 passim Zhongchangshi, eunuch agency 222-3
Persian ch. 2 passim Zhou Dynasty 221-2
Zhu Quanzhong, Warlord 226
Xenophon, Cyropaedia 5-6, 24, 38, Zoe Karbonopsina 149
145 Zuo Qiuming 222
Xerxes 6, 31-2,41, 51, 146
269

Eunuchs in Antiquity and Beyond

Uploaded by

Copyright:

Available Formats

You might also like

Eunuchs in Antiquity and Beyond

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Eunuchs in Antiquity and Beyond

Uploaded by

Copyright:

Available Formats

EUNUCHS IN ANTIQUITY AND BEYOND

The Classical Press of Wales

Distributor in the United States of America:

Originated and prepared for press at

© 2002 The contributors

All rights reserved. No part of this publication may be reproduced, stored in

ISBN 0 7156 3129 2

Typeset by Ernest Buckley, Clunton, Shropshire

1. Eunuchs in history and society 1

5. Looking for eunuchs: the galli and Attis in Roman art 87

11. Eunuchs in the late Byzantine empire, c. 1250-1400 199

This introduction will not seek to provide a general overview of eunuch

Agamemnon and maschalismos'; this was a paper delivered at a day-school on

3 Ringrose 1994 and 1996.

12 Hopkins 1963 and 1978, 172-96.

13 Patterson 1982, 314-31.

1995; Long 1996.

on Medieval Studies in Kalamazoo and at the 20th International Congress of

Sideris 2000; Tougher 1999a); on Byzantium (Ringrose 1999; Tougher 1999b);

It is a great pleasure to be able to thank those who made a significant

EUNUCHS IN HISTORY AND SOCIETY

eunuch servants and courtiers. When eunuchs came to be treasured for

second birth, or the opening of the eye of wisdom. The operation is

Reasons for castration

homosexuals engaged in sex with consenting partners. In recent years,

Although the meaning of the statement is unclear, it was interpreted

Still another reason for castration is simply sexual purposes. According

With soft little hands,

Desire for eunuchs as sexual partners appears also in China, 46 and it

of hormones and the development of artificial ones. This allowed the

on 'voluntary', and neither tubal ligation nor vasectomy have an effect on

2 Nanda 1999, 26-9.

3 See Pellat 1978, 1087-92.

7 Graham 1951, 422-4.

13 Holy Bible. New Revised Standard Version, Nashville, Tenn. 1990.

14 See the discussion in Kadish 1969.

21 Report ofthe Subcommittee on Sex Crimes, Sacramento CA 1952,47.

22 Kemper 1990, 38-65.

23 Osgood 1951, 146.

24 Bullough 1976, passim. Consult 'eunuchs' or 'castration' in the index.

29 Brundage 1987, 83.

in M. Wyke (ed.) Gender and the Body in the Ancient Mediterranean,

EUNUCHS AND THE ROYAL HAREM

The eunuch priests add colour to Flaubert's Orientalist vision of

Eunuchs and the harem: two complex terms to explain, both in

However, many western scholars, seduced or repelled by the notions

David Lewis, following A.T. Olmstead's example, took a far more

evidence to suggest that the unmarried sisters of kings were housed in

Such an interpretation of the archaeological evidence is now unfashion-

the exception of religious festivities such as the New Year celebrations,

concubines, but Artaxerxes II is said to have had 118, all of whom, as

What is interesting about this harem edict is the evidence it provides

Bougaios who treats her as a particular favourite, bestowing jewels and

court to inner court resulted in an increase in power wielded by ambitious

The closest parallel to this self-sacrifice is located on the battlefield of

We are not told the identity of the woman, but it is probably to be

as Petasakes and Artaxares 'the kingmaker' demonstrates the authority held

The same story is expanded on by Plutarch (Art. 17), who renames