Tips - Ritualised Friendship and The Greek City PDF

PUBLISHED IY THE PRESS SYNDICATE OF THE UNIVERSITY OF CAMBRIDOE
The Pill Building. Trumpington Street. Cambridge. United Kingdom

CAMBRIDOI! UNIVI!RSITY PRI!SS
The Edinburgh Building. Cambridge C02 2RU. UK
~West 20th Street. New York NY 10011-1$211. USA
477 Williamstown Road, Port Melbourne. VIC 3207, Australia
Ruiz de Alarc6n 13.28014 Madrid. Spain
Dock House, The Waterfront. Cape Town 8001. South Africa
http://www.cambridge.org
o Cambridge UniYenity Press 1987

This book is in copyright. Subject to statutory exception
and 10 the provisions of relevanl collective I~nsing qreemenls.
no reproduction of any part may take place withoul
the wrillen pcrmiuion or Cambridge University Press. ""0
Q)
~
Firsl published 1987 ..c
First paperback edition 2002 en
L
A (ala/ogut rtcordfor Ihis book is aWJi/ab/r from 1M Brilish Library >-

0.
o
Library of Congr,n Cala/op/llg III Pub/iral/oll do,a U
Herman, Gabriel.
Ritualised friendship and the Greek city.
BibiiOiraphy.
Includes index.
I. Greece - Social life and customs.
2. Friendship - Greece. I. Tille.
DF78.H47 1986 302.3'09495 86-4211
ISBN 0521 32541 2 hardback

ISBN 0 521 52210 2 paperback
CONTENTS
List of tables page IX

List of figures IX
Preface XI
Abbreviations XIII
1 INTRODUCTION 1
2 RITUALISED FRIENDSHIP
1 Definition 10
2 The Evidence 13 ro
~
3 Features Shared With Kinship 16 (1)
+-J
4 Features Shared With Friendship 29 ro
5 The Cross-Cultural Perspective 31 z:
""0
6 Social Status 34 (1)
+-J
..c
3 ENCOUNTER AND INITIATION en
~
1 The Etiquette 41 >-

0..
2 Preliminaries 44 0
U
3 Supplication 54
4 The Initiation Ritual 58
5 Continuity 69
. THE CIRCULATION OF RESOURCES

1 The Pattern of Exchanges 73
2 The Good Gift and the Bad 75
3 Natural Products 82
4 Valuables 88
5 Troops 97
6 Estates 106
S OBLIGATIONS: HEROIC AND CIVIC

1 'What Alkibiades Did and Suffered' 116
2 The Nature of the Obligation 118
3 Services 128
4 Xenia and Proxen;a 130
VII
VIII Contents
5 Followers, Factions, and the City 142

6 Treason and Patriotism 156
CONCLUSION 162
APPENDIX A: Xenoi. Idioxenoi. and Doryxenoi 166

APPENDIX B: Xenia, Proxen;a. and Citizenship 176
APPENDIX C: Networks during the Peloponnesian War 179
Bibliography 185
Indexes 196
co
~
(1)
+-J
co
z:
""0
(1)
+-J
..c
en
~
>-
0..
o
U
TABLES
1 Supply of Natural Products 85

2 Cyrus' xenoi 99
3 Grants of Landed Estates 109
FIGURES
1 The xenos as a foster-father

1a Attic red-figure vase. Paris, Louvre G186. Woodcut: JHS I (1880)
138 fig. 4, reproduced by permission of the Syndics of Cambridge
University Library 24
1b Attic black-figure vasco London, British Museum 8620. Drawing:
P.V.C. Braur, Centaurs in Ancient Art (Berlin, 1912) fig. 25 24
Ic Attic black-figure vase. Naples, Musco Nazionale, Collezione
Santan-gelo 160. Photo: Fotografia Foglia 25
Id Attic red-figure vasco Paris, Louvre G3. Photo: Chuzevillc 25
2 Ritual services 27
Attic red-figure vase. Palermo, Musco Nazionale V 702 (1503).
Photo: Hirmer Fotoarchiv
3 Social status
3a Antiochos of Commagene shaking hands with Apollo. Gaziantep
Museum. Photo: Jorg Wagner 37
3b Antiochos of Commagene shaking hands with Herakles. Eskikale. 37
Photo: British Institute of Archaeology, Ankara
4 A non-Hellenic institution? 51
Relief from the throne base of Shalmaneser III. Nimrud. M.E.L.
Mallowan, Nimrud and its Remains Vol 2, p. 444. Photo: Courtesy
of Lady Mallowan
5 The ritual handshake 52
Attic black-figure vase. London, British Museum 8226. Photo:
Courtesy of the Trustees
6 Communal symbo/on 53
Right hand of bronze, bearing an inscription. E. Babelon, CAtalogue
des bronzes antiques de fa bibliotheque nationale (Paris, 1895) no.
1065. Photo: Bibliotheque nationale
IX
x List of Illustrations
7 Consubstanriality 55
Scythian gold object. Leningrad, Hermitage. Photo: Hermitage
8 Terracotta symbola from Athens
8a-c Terracotta plaques. Athens, Agora Museum. Photo: American
School of Classical Studies at Athens: Agora Excavations 62
8d Terracotta plaque. Athens, Agora Museum. Drawing: IG 12 916 62
9 Ivory symbolon 64
Ivory object. Palermo, Museo nazionale. Photo: Museum
10 Persian symb%n 67
Persian pot. London, British Museum. Photo: Counesy of the
Trustees
11 Proxenia and xenia 133
Athenian decree in honour of Oiniades. Athens, Epigraphical
Museum 6796. Photo: TAP, Athens
12 The city as xenos 134
Relief. Athens, Acropolis Museum 2994. Photo: TAP, Athens
13 The demos mourns a proxenos 136
Gravestone from the Kerameikos cemetery. Photo: DAI, Athens
14 The proxenos as a foster-child 137
Relief. Athens, Acropolis Museum 2994. Photo: TAP, Athens
15 Xenia and treason
Two ostraka from Athens
I5a A. E. Raubitschek, Charites, Festschrift Lang/ott (Bonn, 1957),
240 158
15b G.A. Stamires and E. Vanderpool, H~speria 19 (1950) 390 158
PREFACE
This book owes its origin to a sudden realisation that a ubiquitous and
pervasive aspect of Greek civic life had somehow been left out of modern
accounts - an aspect which I have come to call 'ritualised friendship'. But
how was one to grasp this elusive phenomenon? The clue came from a book
which was not, strictly speaking, about the Greek city: Sir Moses Finley's
The World of Odysseus. I am very grateful to Sir Moses for having encour-
aged me to pursue this idea and for having supplied, as the supervisor of my
doctoral dissenation, unequalled inspiration and criticism. To those famil-
iar with his work, my indebtedness to him must be evident from every page
of this book. Moreover, on one issue - the 'historicity' of Homeric society-
my findings proved to be so consistent with his views that I have decided,
after serious consideration, not to refer to rival interpretations. I also owe a
special debt of gratitude to three friends, Dr Peter Garnsey, Professor
Richard Saller and Professor Brent Shaw, who in numerous helpful discus-
sions over sustained periods exercised a crucial influence on my thinking.
My warm thanks go to the examiners of the dissertation, Dr Paul Cartledge
and Professor Frank Walbank, for saving me from several errors and for
making many valuable suggestions. Thanks are also due to teachers, col-
leagues and friends, both from home and abroad, for comments and reac-
tions which made it easier for me to reconsider certain issues: Professor
Moshe Amit, Professor David Asheri, Professor David Cohen, Professor
Arnaldo Momigliano, Dr Simon Price, Professor Israel Shatzman, Professor
Anthony Snodgrass and Dr Dorothy Thompson. As usual, none of them is
to be blamed for my errors. For help in improving the text, which was far
beyond editorial, I am greatly indebted to Jane Kenrick. For help in operat-
ing Phoenix, I would like to thank Dr Miri Rubin and Dr Platon Tinios.
Last, but not the least, I should express my deep appreciation to my wife
Ora for her patience during four-odd years which for her, I know, were not
the easiest.
My studies in Cambridge have been made possible by generous grants
from the Hebrew University, Jerusalem (the Philip and Muriel Berman
Fellowship) and Darwin College, Cambridge, of which I became a member.
I also had the honour of being elected to a Leo Baeck Scholarship, an
Aylwin Cotton Fellowship, and a Humanitarian Trust grant, awards which
in difficult times made it possible to continue the research. Finally, I would
XI
XII Preface
like to thank the Faculty of Classics, Cambridge, for a grant to prepare the
typeSCript for press.
}trusalmt G.H.
September 1985
1
INTRODUCTION
Two mighty heroes of the Homeric Iliad, Diomedes and Glaukos, were
about to engage in fierce combat when they suddenly came to recognise that.
their grandfathers were bound by xenia. Xenia (xen;e, xe;ne;e or xe;n;e in
various dialects) was the Greek term for a social institution which, in the
absence of a familiar parallel, historians today render by the awkward
neologism 'guest-friendship'. Diomedes was pleasantly surprised at the re-
velation, drove his spear into the earth, and spoke to his former rival in a
friendly tone, saying, among other things:
...Therefore I am your friend and host (xeinos phi/os) in the heart of Argos;
you are mine in lykia, when I come to your country. ro
~
let us avoid each other's spears, even in close fighting. (1)
There are plenty of Trojans and famed companions in battle for me +-J
ro
to kill, whom the gods send me, or those I run down with my swift feet, z:
many Achaians for you to slaughter, if you can do it. ""0
But let us exchange our armour, so that these others may know (1)
+-J
how we claim to be guests and friends from the days of our fathers (xeinoi ..c
patroior). I en
~
Later on, in the Classical world of the cities, diametrically opposed views >-
0..
could be pronounced on precisely the same issue. In the course of a Spartan o
campaign in Asia Minor in 394 B.C., the Persian satrap Pharnabalos U
reproached King Agesilaos of Sparta for having ravaged his private estates.
He complained that Agesilaos had breached the duties of friendship. For he,
Pharnabazos, had been a friend and ally of Sparta, had provided her with
money, and had fought on the side of Agesilaos against a common enemy.
Agesilaos should have repaid him with favours instead of wronging him.
The thirty Spartiatae who accompanied Agesilaos were filled with shame at
hearing this rebuke. Agesilaos, however, by a stroke of diplomacy dis-
claimed all personal responsibility and pleaded force ma;eure:
I think you know, Phamabazos, that in the Greek states, also, men become xenoi of
one another. But these men, when their states come to war, fight with their father-
lands even against their xenoi, and if it so happens, sometimes even kill one another.
And so we today, being at war with your king, are constrained to regard all that is
his as hostile; as for yourself, however, we should prize it above everything to
become friends (Phi/ot) of yours. 2
I Iliad 6.224ff., rranslared by R. Lattimore, The Iliad of Homn' (1951).
2 Xenophon, HellnriC4 4.1.34-5; unless stared orherwise, all translations a~ adapted from
the loeb serino
1
2 Introduct;on
Only very rarely do ancient records offer such a point of vantage from
which to observe perceptions of social obligations in worlds as different yet
as intimately interconnected as the world of the epic poet and the world of
the Greek city. Crudely, the contrast can be stated thus: Diomedes and
Glaukos could exercise the rights and duties of guest-friendship freely.
Agesilaos and Pharnabazos could not; they had to take into consideration a
third, intrusive and disturbing factor, the polis.
A more refined explanation of this contrast will introduce us to a set of
problems inherent in the civic system from its very beginning and yet almost
totally ignored in modem historical research.
That the obligations of guest-friendship should be set above all other
obligations was for the epic poet a part of the natural order of things. It was
this assumption which allowed him to construct the situation that appears
implausible to us: Diomedes and Glaukos would be friends, but their
associates could continue fighting; they would avoid each other in battle, ro
~
but might kill each other's fellow-warriors. There was nothing strange or (1)
+-J
immoral about this. For the poet, adherence to the code of guest-friendship ro
was a supreme manifestation of the hero's free exercise of prowess. There z:
""0
was, in his world, neither overlord to demand feudal allegiance, nor com- (1)
+-J
munal group to claim social responsibility. The hero, the supreme pinnacle ..c
of a small social pyramid, was under no involuntary obligation to anyone; en
~
the guest-friendships he contracted were his own private affair. >-

0..
But the community tamed the hero, and transformed him into a citizen. o
U
The citizen was an entirely new creation - a social type subjected to
compulsory regulations. It was this power of the community to restrict
individual action that provided Agesilaos with his excuse. Agesilaos invoked
the communal principle, the doctrine that obligations towards the commun-
ity should override all other obligations. Civic obligations had come to take
priority even over guest-friendship: xeno; who were citizens had to fight on
behalf of their cities even at the risk of killing each other. This was an
obligation of a totally different, hitherto unknown order. For, unlike the
obligations of guest-friendship, which arose only from morality, civic
obligations were legally enforceable. And guest-friends who were citizens
could sometimes find themselves accused of treason, brought to trial,
banished or executed. The one-time hero thus became trapped in a severe
conflict - a conflict which could only be resolved by means of painful
compromises.
We are dealing not with isolated and peripheral exceptions, but with
patterns recurring at the very heart of the civic system, conflicts from which
ideas flowed and actions evolved. Both Agesilaos' argument, claiming prior-
Introduction 3
ity for the civic principle, and Phamabazos' bitter reproach, drawing its
justification from a code of personal loyalty, find echoes in numerous
examples from the world of the Greek polis itself. A few well-known cases
will serve to demonstrate this proposition and pave the way for the formula-
tion of the main arguments of this study.
Perikles in 431 B.C. foresaw the Spartan invasion of Attica and thought it
necessary to reassure the Athenian assembly that his personal bond of xenia
with King Archidamos of Sparta would not be harmful to Athenian in-
terests. He anticipated that the Spartan king might ravage most of Attica but
spare his own estates, either as a favour to a xenos or as an act of malice
intended to stir up prejudice against himself among the Athenian demos. To
avoid such embarrassment, and a possible public scandal, which in turn
could have jeopardised his position of leadership, Perikles converted his
private estates into public property. 3 The other side of the coin - the
dilemma which confronted King Archidamos - has not left such clear traces
in the historical record. But, as we shall see (Section 5.5), it could not have
been very different from the dilemma of his Athenian partner, even if his
answer to it was somewhat different.
Indeed, patriotism and guest-friendship would sometimes appear as anti- -0
(1)
+-J
thetical principles structuring the dialectics of political rivalries. Demos- ..c
CJ)
thenes, for example, while priding himself on having preferred the common
~
interest of Greece to the gifts and xenia of Philip of Macedon, accused >-
0..
Aeschines of having put his (that is, Aeschines') xenia and philia with Philip o
above the fate of the city." Two competing moral systems were involved: U
one archaic and pre-political, and the other stemming from the polis struc-
ture. It was by no means clear which one would exercise the stronger appeal
on the minds of the citizens. Aeschines alleged - or perhaps insinuated -
that Demosthenes was guilty of an impious crime: in the name of the city,
Demosthenes had arrested, tortured and put to death one of his own xenoi,
a man whose only crime was to have come to Athens to purchase goods for
the queen of Macedonia. Demosthenes retorted by saying that as a matter of
fact the man was a Macedonian spy; by executing him, he had merely "held
the city's salt as more important than the table of his own xenos". If we are
to believe Aeschines, the citizens and foreigners in the assembly raised a cry
of protest at this remark: the duties of guest-friendship were deemed more
, Thucydidcs 2.13.
• Dmlosrhencs 18.109 (0,. Ih~ Crown) and 19.248 (De Fatsa ugatio,.e); d. Dmlosthmcs
18.284.
4 Introduction
binding than those of citizenship. 5 The law of Athens would seem in this
respect distinct from the morals of its inhabitants.
Is this a safe inference? Certainty is impossible: it cannot be ruled out that
Aeschines, when commenting on the protest of the crowd, was lying. But
there are other signs pointing to the same conclusion. Much still remains
obscure about the fifth~entury treaty between the small Locrian states of
Oeanthea and Chaleon concerning the procedures by which the citizens of
Chaleon were to be judged in Oeanthea. 6 But one section of the inscription
is sufficiently clear to allow a compelling interpretation. It provides that if
the officials (xenodikai) trying the suit of a foreigner (xenos) from Chaleon
disagree, the foreigner can select a new body of men from among the
Oeantheans. Only one restriction is imposed on his freedom of choice: he
may choose neither a proxenos of his own community nor a guest-friend
(widioxenos) of his own to act as judges in his trial.? These men were
excluded since, clearly, they were believed to be prejudiced in the foreigner's ro
~
favour on account of their special relationships with him. And such (1)
+-J
favouritism was incompatible with the principles which governed commun- ro
al life. Outside the city, or before the city arose, it was one of the most z:
sacrosanct duties of a xenos to succour his partner in distress or misfortune. ""0
(1)
But now this duty clashed with the communal principle of justice, and the +-J
..c
community had to assert its precedence lest its essential principles be sub- en
~
verted. The archaic morality of guest-friendship could not be reconciled >-
0..
with communal justice. o
The same conflict between old and new, individual and community, U
morality and law, guest-friendship and citizenship, would manifest itself in
different guises in different situations. The Athenian Xenophon was invited
by a guest-friend, the Boeotian Proxenos, to join the forces of Cyrus the
younger and to become Cyrus' friend. But Socrates warned Xenophon that
by becoming a friend of Cyrus, Xenophon might bring upon himself accusa-
tions of betrayal from his fellow-citizens. For Cyrus was known to have
given the Spartans aid against Athens. Xenophon's dilemma could only be
resolved by consulting the oracle at Delphi. In the end, of course, Xenophon
joined Cyrus, and it was through this friendship that he became involved in
, Aeschines 3.22"·5 (Again't CU,iphorr) and Demo5thma 19.189·90 (D~ fa/sa Lqa·
t;Off~).
, Too (1933) no. 3.. ; for 1M larnr exrensive discussion of the inscription, sec Bravo (1980)
89Off.
, For the term widioxmo., sec below p.ll n.3. The conMxion between proxmo. and
(wid;o)xmcu is explained in Section S....
Introduction 5
the adventures described in the Anabasis. But it was also because of this
friendship that he was later exiled from his native Athens. 8
Even worse was the fate of Ismenias, the leader of the anti-Spartan
faction in The~s. In 382 B.C., the Spartan Phoibidas, acting in concert with
Ismenias' political rivals, effected a coup d'etat in Thebes and set up a
government friendly to Sparta. But Phoibidas' action proved the cause of
embarrassment for his home government. For one thing, the coup was an
improvised move, carried out without authorisation from the Spartan po/is.
For another, the forceful occupation of another state constituted a violation
of the King's Peace of 386 B.C. of which Sparta herself was a signatory and
according to which all Greek states, small and great, were to be auto-
nomous. The occupation, however, proved of great practical value to Spar-
ta, and these considerations had somehow to be brushed aside. A justifica-
tion was therefore concocted post-factum. The Spartan assembly first let
itself be persuaded by Ismenias' Theban opponent, Leontiades, that Thebes ro
~
had been hostile to Sparta in the past, and that Ismenias was mainly (1)
+-J
responsible for this. This made it morally possible for the Lacedaemonians ro
not to withdraw their garrison from Thebes and to bring Ismenias to trial. z:
They set up a jury composed of three Lacedaemonian judges and one from ""0
(1)
+-J
each of the allied states. The charges brought against Ismenias were as ..c
follows: "that he co-operated with the barbarians; that he had become a en
~
xenos of the Persian to the detriment of Greece; that he had received a share >-
0..
of the money which came from the King; and that he and Androkleides were o
chiefly responsible for all the trouble and disorder in Greece". 9 The defence U
made by Ismenias failed to persuade the court; he was pronounced guilty,
and put to death.
Once again, the contrast with the Diomedes-Glaukos encounter is reveal-
ing. In the circumstances of that world, a similar incident would have been
inconceivable. Even if we allow for a certain amount of idealisation and
poetic distortion, we must concede that none of the factors involved in
Ismenias' execution is present in the Homeric poems. Not only are they
missing, but all we know of Homeric society would militate against any
assumption of their existence. Any argument for the implausibility of the
Diomedes-Glaukos episode would therefore have to demonstrate the exis-
• Xmophon. Anabasis J.1.4H.; d. Diogenes Laertius 2.58 (= Anthologia Palalina 7.98),

quoting what appears to be a Corinthian funerary ~igram: "Albeit the countrymen of
Kraunus and Kekrops condem~ thee, Xmophon. to exile on account of thy friend (philos)
Cyrus. y~ hospitable Corinth wtlcomed thtt' For different interpr~ations of these pas·
sages in modem research. sec below p. 15. n.14.
9 Xenophon. Hel/mic4 5.2.35. Whether or not Ismmias was guilry is irrdnoant to tht
present argument.
6 Introduction
tenee in that world of a concept of communal interest; of a notion that the
community is more important than the individual; of a principle of indi-
vidual responsibility and accountability; of an agency capable of making
and enforcing communal resolutions; and finally, of moral criteria by which
guest-friendship can be judged to be 'bad'. In the world of the polis, such
factors were actively involved in the shaping of events; in the world of the
epic poet, they were totally absent. Homeric heroes seemed so completely
assured of their dominance, and Homeric communities so completely para-
lysed in their helplessness, that effectual criticism of the former by the latter
was totally out of the question.
The transition from one world to the other was effected by a radical
change in social organisation and the concentration of power. In brief, the
rulers of the Homeric age were supplanted by self-regulating communities.
But this change did not amount to a wholesale transformation. Both inside
the city and even more outside it, older social groupings and archaic ideals ro
~
maintained themselves alongside the new ones with remarkable tenacity. (1)
+-J
Our incidents form only a very partial reflection of this heritage. The ro
evidence from the age of the cities is in fact replete with similar cases z=
""0
appearing under different guises. To explain the entire process we must (1)
+-J
formulate a theory which, by means of one simple assumption, will make ...c
sense of the hundreds of apparently disparate instances. The theory I pro- en
~
pose is this. >-

0..
When during the eighth and seventh centuries B.C. the contours of the o
U
city-state were gradually drawn, the ancient world was criss-crossed with an
extensive network of personal alliances linking together all sorts of apoliti-
cal bodies (households. tribes, bands, etc.). The city framework superim-
posed itself upon this existing network - superimposed itself upon it, yet did
not dissolve it. And when the city finally became established as the domi-
nant form of organisation, dense webs of guest-friendship still stretched
beyond its bounds. Overtly or covertly, guest-friendship continued to act as
a powerful bond between citizens of different cities and between citizens
and members of various apolitical bodies. And by this persistence in the age
of the cities, it became involved in actively shaping the value system of the
polis and in formulating some of its most basic concepts and patterns of
action.
It is the aim of this study to explore the various facets, ramifications and
consequences of the co-existence of the political communities and the
archaic network of personal alliances. Several topics are brought into dis-
cussion in order to make this possible. Each follows logically from the one
before, and draws on evidence from the period between Homer and the
Introduction 7
gradual conquest of the Greek world by the Romans. Some of these topics
are so richly documented, and the documentation itself is sometimes so
repetitive, that I see no point in citing all the evidence: my generalisations
rest therefore on examples which I assume to be typical. Other topics are
overlooked in modern research, and, if necessary for the argument, I pro-
duce a collection of the relevant material (see mainly the Appendices). In
some rare cases I exceed the chronological boundaries defined above and
refer to examples from the Roman world: there are no good grounds to
believe that guest-friendship underwent significant changes even in this
period. Below I outline the main topics.
With one notable exception,1O what has traditionally been labelled in
modern scholarship as 'guest-friendship' is beset with misapprehensions. I
argue that xenia can be located within the wider category of social relations
known to anthropologists as 'ritualised personal relations'. It will then
appear that details which in the absence of a suitable model have bttn
divorced from each other (for example, the ritual of initiation, the naming
of one's son after one's xenos, the exchange of resources across political
boundaries) are in fact different manifestations of the same underlying
phenomenon. What emerges at the end of this enquiry is a social institution
with clear boundaries, well-defined rules, and a remarkable degree of inter-
nal cohesion - an institution which appears at first sight queer and implausi-
ble but ceases to be so once we view it in the light of similar institutions
from other cultures. Indeed, to contemporaries this institution seemed self-
evident even to the extent that they employed it as a metaphor for express-
ing their perceptions of the world. One of the side issues pursued through-
out this study is the logic underlying this symbolism (see mainly the Fig-
ures).
I then explore the interaction between the city and guest-friendship at the
level of ideas. Some of the most basic concepts of the city-state emerged out
of the superimposition of the communal structure upon the pre-state net-
works of guest-friendship. For example, when seen from the perspective of
the community, gift-exchange with an outsider - the essential characteristic
of guest-friendship - could appear as bribery. The antithetical notion of
abstinence from accepting gifts became the mark of the ideal citizen. To turn
this negation of heroic virtues into a term of praise and to offer communal
interest as a new standard of individual morality was probably one of the
most significant victories of the community over the hero. The hero, how-
ever, remained in some respects untamed. The external ties of guest-
friendship acted as a repository of heroic values, and disillusioned aristoc-
10 Finley (19n) looff.
8 Introduction
rats could always trust a friend abroad for assistance in their political
struggles and refuge in defeat. Such behaviour, of course, appeared from the
communal point of view as 'treachery'. Unlike its modem counterpart,
however, the ancient concept of treachery was backed by a 'positive' ethos.
The acts for which Alkibiades became known as the archetypal traitor in
history were in fact consistent with the archaic, pre-state ideals of guest-
friendship. Paradoxically, too, it was this archaic institution which provided
the city with a model of its relationships with the outside world. The
concepts, outlook and symbols of guest-friendship were transferred from
the personal level to the level of the whole community and, invested with
new meanings, they provided the city with a framework for interacting with
foreign individuals and communities.
Guest-friendship served as a device for the promotion of the material and
political interests of the elites engaged in it. Individuals integrated into
politically separated communities exchanged substantial amounts of wealth ro
~
(1)
and performed significant services for each other. Once we reveal the +-J
dynamics of these networks, it becomes apparent that the elites of the ro
ancient world were not confined to the boundaries of their immediate
z:
""0
communities (whether this be a city, a tribe, a petty kingdom or a Persian (1)
+-J
satrapal court). On the contrary, they participated at one and the same time ..c
en
both in these networks and in their immediate communities. Power, prestige ~
and resources that could be acquired through one system could readily be >-
0..
transferred to the other, and at times the horizontal ties of solidarity which o
U
linked together the elites of separate communities were stronger than the
vertical ties which bound them to the inferiors within their own communi-
ties.
Outside the city, guest-friendship functioned as a major device in the
formation of the ruling circles of the great territorial empires. The Persian
syngeneis, Macedonian hetairoi, and Hellenistic philo; were all composed of
a narrow core of kin and a wide periphery of guest-friends drawn from an
astonishing variety of localities. I explore the process of their recruitment,
the structure of the new groups to which they gave rise, and the networks of
agents through which they exercised their influence on the Greek city. I
finally argue that the attacks of these groups on the Greek city were
prompted by and large by the desire to satisfy the personal interests of the
guest-friends involved in these groups.
To be able to identify guest-friendship in its manifold manifestations, I
propose to employ a rigorous definition. This definition will have to meet
three requirements: first, it must encompass all the variants of the archaic
bond - variants of which xenia is but one example; secondly, it must
Introduction 9
separate xenia from other, seemingly similar institutions, some of the qual-
ities of which it shares; and, lastly, it must enable us to make controlled
comparisons with similar institutions in other cultures. It is mainly in order
to meet these requirements that it becomes necessary to introduce the
concept of ritualised friendship.
ro
~
(1)
+-J
ro
z:
""0
(1)
+-J
..c
en
~
>-
0..
o
U
2
RITUALISED FRIENDSHIP
2.1 DEFINITION
For analytical purposes ritualised friendship is here defined as a bond of
solidarity manifesting itself in an exchange of goods and services between
individuals originating from separate social units. This definition encompas-
ses the most distinctive features of the institution and supplies criteria for
postulating its existence even if it is not named explicitly in the evidence.
Ritualised friendship will accordingly be assumed to exist either between
people styled by one or several of the characteristic designations which are
described in detail below, or between people originating from separate
social units engaged in an exchange relationship of some duration. Excluded
are relationships between strangers that involve payments for goods and
services - as, for example, those between merchants and their customers, or
-0
mercenary soldiers and their employers. People trading specific goods and (1)
+-J
services for payments would hardly classify their relationship as one of ..c
friendship. I Transactions of ritualised friendship were supposed to be car- en
L
ried out in a non-mercantile spirit; from the point of view of the partners, >-
0..
they appeared not as an end in itself but as a means for creating a moral o
U
obligation.
These salient features, as well as some more specific attributes of ritual-
ised friendship, will gradually be elucidated and further specified. But first,
to gain an initial understanding of the institution, we start with an examina-
tion of the terminology applied to the people involved in it. Three groups of
terms can be distinguished: (a) xenos, idioxenos and doryxenos; (b) phi/os,
hetairos, epitedeios, anankaios and oikeios; (c) syngenes and euergetes. The
semantic range of these words is sometimes extremely wide; 'xenos', for
instance, in addition to 'guest-friend', can mean 'foreigner', 'stranger',
'guest', 'host' and 'mercenary soldier,.2 In this study only the functioning of
I For an emphatic assertion of this principle, see Dem05thenes 18.51-2 (On the Crown): "Is
Aachines Alexander'. hirding (l"O6(1)l~. or Alexander's xmos"? and Xmophon. Agni"''"
.....: "For had he (i.e. Agesila05) been in the habit of Idling favours (tl yciQ bWAtl la~
x6(K~ or raking payment for his benefactions (JUa6oU tUfqyhfl), no one would have felt
that he owed him anything."
1 See BoIkestein (1939) 87-8 and Whikhead (19n) 10-1. For its (uncmain) t"tymology, see
Benveniste (1969) voU, 87-101; H. Frisk, G,~chiscMI nymologi.che. Vlorterbuch vol.2
(Heidelberg, 1970), I.V. ~fi~; Gauthier (1973); Hihbrunner (1978); and Taillardat (1982).
10
1 Definition 11
these words as interpersonal designations will be considered. The procedure

is justified by the fact that the participants manifestly classified the different
meanings of the same words in different mental compartments. Mostly,
linguistic usage allowed for no ambiguity between, say, xenos as 'guest-
friend' and xenos as 'stranger'. Once we are familiar with the implicit codes
and rules of ritualised friendship, it will become apparent why this should
have been so.
Thus delimited, the words listed above imply special relations of
friendship, trust, loyalty, reciprocity, and mutual aid between the people
they characterise. However, only xenos and its rare derivatives, idioxenos
and doryxenos,J refer invariably to individuals originating from different
social units. This fact is highlighted by the curious propensity of ancient
writers to indicate the provenance of the actors. It is as if they considered
their provenance inseparable from their personalities. Thus, it appears that
a xenos, whether he came from a citY, tribe, ethnos, or some other social ro
~
unit, always had a group identity distinct from that of his partner. In other (1)
+-J
words, each individual in a xenos-dyad was an outsider with respect to his ro
partner's group. In the extant sources, no two people with the same group z:
identity are ever referred to as xenoi (Appendix A).4 To call an insider- ""0
(1)
friend a xenos seems to have amounted to a contradiction in terms. 5 +-J
..c
In prehistoric times, it was probably the family group which defined this en
~
conceptual line of demarcation between 'inside' and 'outside'. In the world >-
of the epos, it was the locality or the group of subjects. In the Greek world 0..
o
of the cities, it was the city boundary. Within Athenian territory, friends U
who came from different demes, tr;ttyes, or tribes, would call each other
1 No significant difference in uuge can be d~ected between xmos and idioxmos. The
prdix of idioxmos places ~rhaps a special emphasis on the private nature of tht" bond
distinguishing one's own xmoi from the xmoi (or prOXnfOl) of the group to which one
belonged (d. Welles (1934) 334 and Serve (1926) 62ff.). The word doryxmos appears mainly
in tragedy, and this fact, together with contradictory explanations given by kxicographers and
in scholia, suggests that the precise meaning of the term (xmoi acquired in war, spared enemit"S
turned into xmoi?) had been forgonen. For a collection of the noidence, see E. Fraenkel,
Aeschylus, Agammrrrorr (Oxford, 1950) vol.lI, 395. Hereafter, I shall assume that whatnoer I
say about xmos also holds good for idioxmos and doryxmos.
• This generalisation emt"rgn from the examination of all the pasS2gt"S listed in Appendix A
in which the panners are styled xmoi, idioxmoi or doryxmoi, and their social origin is
tra~ble. Even the fictitious cases conform to this general pattern.
J This fact is not recognised in modem literature, and the term xmos is so often mistrans-
lated that it might be impractical to argue in each case for the interpretation adopted here.
Instead of 'guest-friend', we may find 'guest', 'host', and 'stranger'. Most frequently, however,
x~"'os is rendered as 'friend', thus obscuring the distinction that the Greeks drew between
friends who were aliens and friends who were fellow-citizens.
12 Ritualised Friendship
philoi, hetairoi, or epitedeioi, never xenoi. 6 This fits in with the gloss of the
word by a lexicographer from late antiquity. A xenos, says Hesychius, is a
friend (Philos) from abroad.? Indeed, the idea of strangeness combined with
friendship is built into the assumptions of the actors. Diomedes' glee when
he says UI am your xeinos philos in the heart of Argos; you (i.e. Glaukos) are
mine in Lykia...", and erito's suggestion to Socrates, if you wish to go to
U •••
Thessaly, I have there xenoi who will make much of you and protect you, so
that no one in Thessaly shall annoy you",8 are typical of the whole institu-
tion.
A further regularity must be remarked upon in this context. The insider-
outsider dichotomy with respect to the partners' own social units was the
only one deemed necessary for a friendship to qualify as xenia. This dicho-
tomy might have coincided with or cut across other wider, real or imaginary
group divisions. A quick glance at the provenance of the xenoi listed in
Appendix A will suffice to show that xenia relationships could exist be-
tween members of different Greek cities; between members of Greek cities
and members of Greek ethne (for example, Macedonians, Epirotes); be-
tween Greeks and non-Greeks (for example, Persians, Lydians, Egyptians,
-0
Phoenicians, Romans);9 and finally, between different non-Greeks. Thus, I (1)
+-J
see no way of pinpointing significant differences between cases in which the ..c
CJ)
bond was confined within the Greek world, it extended beyond it, or it was
~
located entirely outside it. Xenia, in all these cases, seems to have followed >-
0..
an identical pattern. Furthermore, the non-Greek partners seem to have o
shown as profound an understanding of this pattern as the Greek. There are U
thus no good reasons why xenia should be regarded as an essentially Greek
institution. 10
By contrast with the xenos words, the terms included in categories (b)
and (c) above can relate to friends of the same or of different social origin.
The wider semantic range of these words stems from differences in emph-
asis. More precisely, when a speaker wished to stress the rights and obliga-
tions of ritualised friendship, the more formal xenos words would be used.
, For example. Dmlosthmn 50.21 (Againsl Po/ycks): Apollodoros of Oineis and Apollo-
doros of Leontis; Aeschines 2.73 (0" Ih~ Embauy): Chares and Kqlhi50phon, each from a
different trilt)ts of Pandionis. When. in 43 I B.C.• the population of Attica came to Athens, they
are said nnucydides 2.17.1) to have ~ provided with dlelter by friends (Philo,) and relatives
(oilteio,). not guest-friends (xmo,).
, Hesychius, S.v. ~d'Y<><;: 6 cVro;~ ~o..<><;; d. s.v. ~tv<x: ...ot bt 10U~ MO ~f:V{a~ +O-ou~.
and S.v. ~~: 6 bn;f:Vou~.
• Iliad 6.224-5; Plato, Crilo 4Sc.
• For an interesting mockrn parallel- Moslems and Jews becoming partnen to Christians in
godgarenthood relations - ~ Hammd (1968) 88.
I As argued. for example, by Gauthier (1973). I know of no study which might throw light
on the institution from a non-Greek point of view.
1 Definition 13
When the emphasis was on the sentiments that were supposed to prevail
between xenoi, words from categories (b) and (c) seem to have been prefer-
red. For our purposes, people with the same group identity who are styled
by anyone of the terms included in categories (b) and (c) do not qualify as
ritualised friends. On the other hand, an Athenian and a Spartan, for
example, or a Milesian and a Persian, do qualify, even if in the surviving
documentation they are not styled by any of these terms (provided, of
course, they meet the other requirements of the definition).
Four of the words included in categories (b) and (c) assumed a highly
technical meaning in specific historical contexts. The followers of the
Macedonian and Hellenistic rulers came to be known, respectively, as
hetairoi and philo;. In a very similar fashion, syngenes and euergetes, the
Greek renditions of Persian words, II became honorary titles bestowed upon
classes of dignitaries in the Persian court. Both in this context and in some
Hellenistic courts where the term survived, syngenes was applied to people ro
unrelated by blood to the rulers, in fact to complete strangers. l l Borrowed
~
(1)
+-J
from the ordinary routines of living, these terms were consciously turned ro
into titles signifying the status of the title holder at court by reference to his z:
degree of amity with the king. These ruling circles, as we shall see, were to a ""0
(1)
+-J
large extent made up of ritualised friends. ..c
en
~
>-
0..
2.2 THE EVIDENCE o
U
Judged by the standards of comparable ancient institutions (as, for instance,
marriage), ritualised friendship is copiously documented. Irrespective of the
literary genres in which they are cast, most of the sources have one impor-
tant feature in common: their writers were typical products of the fully
fledged city-state. Consciously or subconsciously, they employed the norms
and values of the civic ~stem in order to represent their universes. The
world of ritualised friendSh1p is thus recreated through the eyes of people
embedded in the rival civic system. This may in part explain why the
processes by which powerful princes gave up hostilities, exchanged precious
objects, and committed themselves to mutual allegiance, exercised such a
fascination for them. Indeed, numerous self-contained digressions are given
over to these events: the exotic, combined with the archaic, provides an
excellent literary theme. But there is more to it than that. For the accounts of
II Sec Wicschofer (1980) for the linguistic problem (the nlng~la; of the kin!lt. (allt"d
orosarrga; in the Persian language. according to Herodotus 8.35) and for a list of the Persian
rulers' philo;, xmoi, and ~ung~lai.
12 Cf. Atkinson (1952) 214 and Gauger (19n).
ritualised friendship are linked to some of the most decisive stages of Greek
history: the Ionian revolt, the Persian invasion, the imperialism of Athens
and Sparta, the rise of Macedon, the Hellenistic monarchies, and, finally,
Rome. It was the search for the sources of these great events that led
historians back to the moment at which the alliances were formed.
Small wonder, therefore, that the accounts of ritualised friendship were
reinterpreted in the light of these events. Actions which had been fortuitous
when performed seemed thus to acquire a new meaning in view of their
outcome; and unrelated events appeared as links in a larger scheme. Mira-
culous reversals of personal fortune were intertwined with these stories, and
served as yet another source of fascination. They tell us how enemies
became friends, the poor grew rich, and nonentities assumed high status.
The need for explanation was so acute that if the evidence was lacking it
seemed necessary to invent it. Hence the occurrence of fictitious (or quasi-
fictitious) accounts like those of Syloson, Darius and the cloak. 13 For all
their fantastic elements, however, these stories have important evidential
value: they show that, according to the writers' expectations, co-operation
between these powerful men could only have taken place if there existed
""0
between them previous relations of trust. And ritualised friendship was the (1)
+-J
only alternative to marriage for generating trust between aliens. ..c
Of all the civic writers, Xenophon merits special attention. First, en
L
Xenophon was a participant-observer: he concluded several alliances of >-

0..
ritualised friendship in person. Secondly, throughout his lifetime he ac- o
U
quired an intimate knowledge of both systems which form the subject of this
study. Born, reared and educated in the most refined of civic environments,
he left Athens at the instigation of a xenos, the Boeotian Proxenos. Most of
the rest of his life he spent away from his native city, under the protective
care of two formidable foreign friends: briefly (40211 B.C.) under Cyrus of
Persia, and for about thirty-five years under Agesilaos of Sparta. Thus it
came about that the clash between the norms of conduct inherited from the
cityless past and those imposed by the city reproduced itself in Xenophon's
own fortunes. Xenophon was expelled from Athens on account of his
friendship with Cyrus; he was recruited by Agesilaos thanks to that
friendship; and he fought in the battle of Coronea side by side with Agesi-
tJ Herodotus 3.139 fl.• d. below p.""

2 The Evidence 15
laos against his own native City.14 No doubt, he had many of the experi-
ences which he attributes to other historical figures in his writings.
In none of the literary accounts is the lengthy process through which
ritualised friendship was concluded and sustained over time reproduced in
its entirety, nor are its various stages evenly documented. Polybius devotes a
great deal of space to the encounter between Neon and Antigonos, but has
only a few words on the exchanges between Brachylles and the Macedonian
royal house. Herodotus similarly expatiates on the story of the cloak but
only refers briefly to Syloson as a tributary vassal of Darius. Nor are all the
aspects and elements of the interaction that took place in reality mentioned.
Their presence or absence in each story depends to a large extent upon the
aims, biases, literary tastes, and moods of the writers. For instance, it is
almost beyond doubt that Neon and Antigonos Doson became xeno;; yet
Polybius does not think this detail worth mentioning. Xenophon, on the
other hand, can write at length of the ritual which turned Agesilaos and ro
~
Pharnabazos' son into xeno;.15 We gain a clearer perception of the relation (1)
+-J
between what is recorded and what in fact happened if we compare our ro
evidence with a phenomenon which is essentially similar but which is z:
""0
documented from a different perspective. Greek states used to confer hon- (1)
+-J
ours and privileges upon people who benefited them in one way or another. ..c
The list of these honours and privileges, as well as a short description of the en
~
benefactions, were inscribed on stone and set up in public places. Relying on >-
0..
such evidence, M.B. Walbank was able to assemble the whole possible range o
of elements recurring in the transactions between the city and its benefactors U
in all the fifth-century Athenian decrees (elements such as proxenia, euerge-
s;a, ges enktes;s, etc.) and then note those elements that are present in each
individual decree. 16 In studying xenia, we cannot follow a similar proce-
dure. What we have are not the texts of agreements enclosed, like proxeny
decrees, within a stereotypical framework, but subjective, impressionistic
accounts of a lengthy process of human interaction. Indeed, only very rarely
14 Cf. above, p.... A different view 50metimn argued for in modem research - that
Xmophon was banished for reasons other than his friendship with Cyrus, or that at Coronea
he did not fight against his native city - seems to me ultimatel)' to stem from a reluctance to
believe that such events were likely to happen. Concerning Coronea, for example, P.J. Rahn (in
G.S. Shrimpton and D.J. McCargar (eds.), ClauiaJl Contributions. Studi~s in Honour of
Malcolm Francis McGr~gor (New York, 1981) 103-20) ignores Xenophon's constant preoc-
cupation with loyalty and patriotism, and don not take into account his obvious tendency to
underplay his own part in the battle. Rahn then suggests that Xenophon ob~rved tht" battlt" as
a by-stander but did not take an active part in it.
IS Polybius 20.5-6; Xenophon, H~lImica 4.1..19, sec Section 3.4,pp.S8ff.
16 M.B. Walbank (1978), especially the catalogues at 9-23.
were the actual terms of ritualised friendship set down in writing. I? As the
evidence stands, we simply have no way of telling what the relation is
between the elements of ritualised friendship that were used in actuality and
those which seemed to the writers worth recording. Tabulation, according-
Iy, is out of the question.
Apart .from such self-contained stories, we can catch brief glimpses of
particular aspects of the process from scattered episodes where ritualised
friendship plays a secondary role, and from general statements concerning
the rights and duties of xenoi. 18 Such details, if fitted cumulatively into a
suitable framework, can yield important information.
To contemporaries, ritualised friendship appeared as an institution
marked by its own distinctive features. To be sure, only a few of these
features were sui generis. With respect to ideology, linguistic usage and
actual practice, ritualised friendship overlapped to a great extent with two
tangential categories of social relations, kinship and friendship. Thus, an ro
~
account of the similarities and differences which existed between these (1)
+-J
institutions can offer important insights into the nature of the institution. ro
The features shared with kinship are considered first. z=
""0
(1)
+-J
2.3 FEATURES SHARED WITH KINSHIP ...c
en
~
The Greeks seem to have believed that, apart from relatives by blood and >-
0..
marriage, the family was also related through its head to a special category o
of people who were more than 'friends' yet somewhat less than 'kin'. These U
were the xenoi abroad, and though it must have been perfectly clear that
they were not kin in the proper sense of the term, the relationship in some
respects resembled kinship. The most important feature xenia shared with
kinship was the assumption of perpetuity: once the rites establishing the
relationships were completed, the bond was believed to persist in latent
form even if the partners did not interact with one another. This assumption
17 ~ for exampk. the symbolorr from Sicily. recording a bond of xmu, between a
Phoenician and a Greek and his descendanrs (/G XIV.279. d. Section 3.4 and Fig. 9. p.64); the
so-called 'Hannibal's covenant' (Polybius 7.9) recording a "sworn treaty of frimdship and
goodwill; we shall be friends (phi/oIlS). kinsmen (oiknollS). and brother'S (ad~/phOfU). on the
following conditions....•. As suggested by Bickerman (1952) 8. the strange, ungrammatical
assembly of Greek words might hide an unknown Punic idiom. For some Roman parallels. ~
Badian (1958) I t-12 and 154-5. For written records of ritualised friendship from the ancient
Near East, see Munn·Rankin (1956). No similar written records of HdJrew bn;Ih have
survived (Bickerman (1950) 26).
II Like. for example, Menandtt's one·liMr (Mcmos';choi. ed. A. Meineke. Nagmmta
Comico",," G,tI~co",," (1841) vol. IV, 357, line 615), "00 honour to your xmos and you will
acquire a good philos". or Stobaeus' maxim (3.9 p.J58) "A xmos is a just man differing not
only from a citizen but also from kinsmen".
3 Features Shared With Kinship 17
had two practical manifestations. First, the bond could be renewed or

reactivated after the lapse of many years, a variety of symbolic objects
serving as a reminder. Secondly, the bond did not expire with the death of
the partners themselves but outlived them and passed on, apparently in the
male line, to their descendants. As the case of Glaukos and Diomedes
shows, here, too, a symbolic reactivation of the bond was necessary. (The
mechanism of this feature of ritualised friendship is described in detail in
Section 3.5.)
The partners involved in ritualised friendship - and through them, their
families - were presumed to be bound by mutual affection. This affection,
however, imitated the outward manifestations but not the inward spirit of
kinship: ritualised friends were not supposed to love each other, but to
behave as if they did. Nowhere does this emerge more clearly than in the
case of Polykrates of Samos and Amasis of Egypt, two men who had never
met. When the Egyptian ruler heard about the inescapable good fortune of
his Samian xenos, he renounced the xenia so that "when some great and
terrible mishap overtook Polykrates, he himself might not have to grieve his
heart for a xenos".19 Genuine feelings could no doubt result from such
-0
simulation. For example, Pindar's description of himself and of Thorax, his (1)
+-J
Thessalian xenos, as "loving and beloved" has a palpable ring of sincerity.IO ..c
CJ)
But, as revealed in a casual remark in Isocrates' Aegineticus, affection was ~
not indispensable for a relationship to endure: >-

0..
Thrasyllos ... had inherited nothing from his parents; but having b«ome the xnros o
U
of Polemainetos, the soothsayer, he became so intimate (oike;os) with him that
Polemainetos at his death left him his books on divination and gave him a portion of
the property which is now in qurstion. 21
The underlying assumption here is that xenia need not be accompanied by
affection: the external forms of xenia would not in themselves have been
enough for Polemainetos to bequeath the property to Thrasyllos. For that to
happen, the further element of real emotional involvement was necessary.
This conventional display of emotions crops up in the sources in a fashion
somewhat unexpected. When the Egyptian raid of the Cretan ex-nobleman
(in fact, Odysseus in disguise) failed and he was in danger of losing his life,
the Egyptian king, whose territory he had invaded, warded off the attackers,
set him on his own chariot, and "took him weeping to his home". 22 When
Klearchos, the Spartan mercenary commander, was prevented by his sol-
19 Ht"rodotus 3.43.
20 Pindar, Pythian 10.66.
21 lsocratn 19.5.
22 Odys~ 14.279; d. Section 3.3 ~Iow.
diers from fulfilling his obligations to his xenos, Cyrus, he stood before
them and "wept for a long while".23 It is hard to avoid the conclusion that
the sentiments of these people were structured differently from ours.
Whether genuine or feigned, the affection displayed was that of kinsmen.
This is made clear above all by the counesy terms of reference by which
ritualised friends would address each other. In pan at least, these terms were
drawn from the kinship terminology. King Seuthes, for example, promised
Xenophon - even before seeing him face-to-face - that if he should bring the
army to him, he would not only treat him in all ways as a friend (phi/os) and
a brother (adelphos), but would even give him estates. 24 The pretence was
that xeno; were like kinsmen, and their behaviour was therefore expected to
imitate kinship roles. "Xeinos'\ said Telemachos to Mentes, the guest-
friend of his father (in fact, the goddess Athena in disguise) "you speak these
things with kindly thought, as a father to his son tt. 2S Pindar, the lyric poet,
extolled his guest-friend, Hiero of Syracuse, for being adored like a father by
his own xmo;.26 Polybius seems to have tried to give vent to his true
sentiments when he related that he himself and the young Scipio Aemilianus
"came to feel a mutual affection which could truly be compared to that of
father and son, or of kinsmen of the same blood".21It is, in the light of this,
not difficult to understand why 'father', 'brother', and 'grandfather' should
have become one of the reciprocal appellations used within the entourages
of Persian and Hellenistic rulers. 28
As for the terms of affection in general, it should be noted that their range
was wider than that of their counterpans in modem societies, and at the
same time they were less specific to anyone panicular type of relationship.
In the Homeric world, in Adkins' words, "every relationship from sexual
passion to guest-friendship, relationships whose differences we should
emphasise much more than their resemblance, is denoted by phi/otes".29
This does not seem to have changed significantly in later times, and even in
the classical age kinship, and what we translate as 'friendship', were consis-
tently regarded as variations of the same underlying principle. Xenophon,
for example, could identify the strongest phi/ia; as those which unite parents
to children, children to parents, wives to husbands, hetairo; to heta;ro;.30
1.1Xenophon, A"abas;s 1.3.2. 2S Odyssey 1.301-8.
14Xenophon. A"abas;s 1.2.25. 16 Pindar, Pythiarr 3.11.
l ' Polybius 31.25.1. There is no explicit reference to "m;a between the two.
11 Patn: 1 Maccabees 11.3 t, Josephus, ]twish A"tiq"itws 13.126, 12.148i adelphos:
OC IS 138i pappos: Xenophon. Cyropudi4 1.4.26. For the ancient Near Eastm1 paraJlel- the
employment of 'fraternity', 'paternity', 'sonship' between allied rulen - see Munn-Rankin
(1956) 16ff. 'Fraternity' does not Ittm to have been consistently used here to cknote rela-
tionships between tquals. nor 'sonship' and 'paternity' for relationships between untquilis.
n Adkins (1963) 36. The examples he cites are Odyssey t 0.336 and Iliad 3.354.
.10 Xenophon, H;ero 3.7.
For Aristotle, the untranslatable terms hetairia, oikeiotes, and syngeneia
were various forms (eide) of philia,31 and the remark he made elsewhere,
that "xenia is the firmest of all philia i" ,Jl fits in with this general recogni-
tion. This, of course, does not mean that friends or guest-friends thought of
themselves, or were thought of, as kinsmen. What it does show is that the
perceptual boundaries drawn between the different categories of positive
relationships were not as rigidly delineated as they are in modem societies.
The principles of kinship and friendship could be denoted by partially
overlapping - rather than mutually exclusive - classificatory terms. The
system of ideas was such that no contradiction was involved in calling real
kin by words used most commonly for friendship, or conversely, in calling a
friend by words derived from kinship.
Another feature in common with kinship was the custom of naming. As a
Christian father would name a son after his godfather, so a xenos would
name a son after his own xenos. Three variations of the practice stand out.
First, a man could call a son by the actual name of the xenos (whether at
binh or at some later stage is unclear). It is this custom which accounts for
the names of Alkibiades, son of Kleinias, of Athens, and of Endios, son of
Alkibiades, of Sparta. 33 The succession of names through two generations is
represented in the diagram.
Kleiniu. - - - - - - - - - - - JCetto; - - - - - - - - - - - - - - Alkibiades.

son of Alkibiades by inheritance son of Endios
: [Athens] : [Sparta]
I I
I I
Alkibiades. - - - - - - - - - - JC~"o; - - - - - - - - - - - - -- Endios.
son of Kleinias by inheritance son of Alkibiades
[Athens] [Sparta)
Mostly, it is difficult to trace such successions because of lack of explicit

evidence such as this. But names of second-generation xenoi are evident in
many cases in which the alliance involved a Greek and a non-Greek. For
example, the Greek Theokles named a son after the Egyptian pharaoh,
Psammetichos; Gorgos, brother of the Corinthian tyrant Periander, also
named a son Psammetichos; an Athenian (Alkmeonid?) named a son after
the Lydian king Croesus; and the Odrysean king Kersobleptes named a son
JIAristotle. Rhetoric 2.1381b 29. d. F.thica Nicomachea 1161b 12-17.

J2Aristotle. Mapa Moralia 2.12111 46. The doubts as to the genuineness of this work do
not impair its evidential value.
H Thucydides 8.6.3. where it is expres...ty stated that Alltibiades is a Spartan name. Cf.
Wallace (1970) 197 n.2 for the ancestry of the relationship.
10Iaos. 34 Within the Greek world itself, the practice must have been even
more widespread. Thucydides believed that it accounted for the spread of
the Hellenic name as a result of which the whole COU'l,ry came to be known
as Hellas:
Before the time of Hellen, the son of Deukalion, the name did not exist at all, and
different parts were known by the names of different tribes... After Hellen and his
sons had grown powerful in Phthiotis and had been invited as allies into other states,
these states separately and because of the connections (homilia,) with the family of
Hellen began to be called 'Hellenic' .lS
Ritualised friendship was thus responsible for the first hellenisation in
history.
This practice of naming might also account for homonyms found in
different cities. In the case of Alkibiades and Endios, the third generation
Athenian Alkibiades and the second generation Spartan Alkibiades can be
differentiated only by the name of their cities and by their patronymics. 36
But if naming was reciprocal, that is, if both panners named a son after the
other, then even the patronymics were identical. Such a possibility must
always be taken into consideration when dealing with homonyms. Thus
Liches, son of Arkesileos, appearing in the archon list from Thasos from the
year 398 B.C., need not be, as has recently been suggested, identical with the
Spanan Lichas, son of Arkesilaos, known from Thucydides. 31 The two
might be xenoi from alternative generations (see diagram). Parallels of such
homonyms - if not homo-patronymics - may be found in Plutarch and
Arketilaos - - - - - - - - - - x~"o; - - - - - - - - - - - - - - lich~

: [Spana) ~ [Thasos)
I I
I I
lichas, - - - - - - - - - - - - x~o; - - - - - - - - - - - - - - Arkesikos,
son of Arkesilaos by inheritance son of Liehes
: [Spana) : [Thasos)
I I
I I
Arkesilaos, - - - - - - - - - - - xmo; - - - - - - - - - - - - - -- Liches,
son of lichas by inheritance son of Arkesileos
[Spana) [Thasos)
Pummerichos: Meigs and Lewis (1969) no.7, d. ~Iow Section ".5 t Croesus: the
J4
1t0f41OSfrom AnavyS10S with the inscription ~Iow (LH. Jeffery in the A,,"ual of the British
School at Athens 57 (1962) 1.....) and with Herodorus 6.125 (where the chronology is impossi-
ble) for the possible origin of the name; GorgO': FGrH 90 F 59 (Nikolaos of Damascus);
lolaos: Syll.-' 195.
lS Thucydides 1.3.2, translated by Rex Wamer.
.~ The (Spanan?) Alkibiadn who made a dedication to Apollo in Delphi is likdy to be an
anc:ntor of Endios (SEG 27.135).
)1 Pouilloux and Salviat (1983).
J Features Shared With Kinship 21
Homer: Archias the Athenian and Archias the Theban are explicitly defined
as xeno;; and the two Homeric Aiantes - Aias son of Telamon, from
Salamis and Aias, son of Oileus, from Locris - were xeno; too, if judged by
their mutual conduct. 38 A series of further examples (to be treated in Section
5.4 and Appendix B) will show that the custom was far more widespread
than is normally thought.
The second variation was the naming of a son not by the name of the
xenos himself but by the name of his city or geographic area. This can best
be illustrated by the case of Thessalos, the son of the Athenian tyrant
Peisistratos, named thus, presumably, after a Thessalian xenos; by the case
of a son of the Athenian Kimon, called Lakedaimonios; and by the case of
the Spartan Libys, brother of Lysander, named thus after an ancestral
xenos, Libys, ruler of Cyrene. 39 The pattern of this variation can be elicited
from Thucydides' remark concerning Hellen cited above. For names could
circulate both ways: from a geographic area to its inhabitants and hence to
xeno; - and vice versa.
The third variation was the calling of sons by compound names bor·
rowed from the domain of ritualised friendship. Typical examples would be
names such as Xenophon, Proxenos, Philoxenos, and the Iike. 40 Though "'0
(l)
clearly rooted in ritualised friendship, and intended no doubt originally to +..J
..c:
conjure up its rights and duties, the custom in historical times must have 0'1
L
been dissociated from the institution itself. I see no way of establishing a >-
0..
.\8 Plutuch. P~/opidas 10 and Iliad 1.l101. Pouilloux and Salvi:u (1983) ignore this o
possible explanation and, assuming that liches from the Thasian archon list and the Spartan U
Lichas known from Thucydides refer to the same man (a poMibility which cannot ~ ruled out),
infer that Thucydides was :alive and wrote :after ]91 B.C. But if one lichas was a second-
generation Spartan xmos, and the other a third·generation Thasian :xmos - a possibility which
is chronologically compatible with the details - then, as pointed out by Cartledge (1984), the
conclusion is far less compelling. For objections on other grounds to the views of Pouilloux 3nd
Salviat, and for the names Lichas and Arkesilaos (or their dialectical variations) in other cities,
sec Bull. Ep;gr. (1984)..Jl4.
J9 Thess.alos: Thucydides 1.20.2. 6.55.1 and Aristotle, Th~ Athtn;an Constitut;on 17.2. d.
Herodotus 5.46.1 for a Spartan homonym; lakedaimonios: Thucydides 1.45.2. IG 12 400, d.
Plutarch, Cimon 16 and Pmcl~s 29; Libys: Diodorus Siculus 14.13.5, cf. Xenophon, H~II~""a
2.4.28. Further examples: Thucydides 3.52.5 (a Plataean p,o:unos of Sparta, called Lakon);
Thucydi<ks 4.119.2 (a Spartan called Athmaios son of Pcrikleid:a5); HerodoN5 3.55.2 (a
Spartan ailed Samios on account of an ~erg~sia - a public funeral- which, presum3bly, Ird to
a xenia ~twttn his ancntors and some Samians and which might have been the name
inherited by the Spartan admiral for the year 40211 8.C. (Xenophon, Hellen;(4 3.1.1, d.
Diodorus Siculus 14.19..~); Thucydides 5.19.24 (a Spartan called Akanthos, d. Pau~OIas 5.8.6
for an eighth·cmtury homonym); Xenophon, H~lIen;(4 1.4.2 (a Spartan called Boiorios);
Thucydidcs 8.6.8 and passim (3 Spart3n called Chalkideus); IG I 23 (3 Thespian called
Athenaios).
4() Curiously, names containing a 'xenos' component are absent from Homer. The only
exception. Polyxeinos, leader of a group of Eprians, proves this rule: he only appears in the
catalogue of ships (Iliad 2.623) and is not mentioned again in the poems (d. H. von Kamprz,
Hom~ris,be Personmnamen (Gottingen, 1982)).
correlation between these names and the standard practices of ritualised
friendship.
Finally, it should be noted that this pattern of nomenclature remained
throughout an optional and sporadic practice: people could well be xnro;
and not name a son after a partner. Yet, as we shall see, awareness of the
practice can offer imponant clues in tracing the networks of ritualised
friendship.
A funher feature which ritualised friendship shared with kinship was the
establishment of a potential link of mutual protection and help not only
between the partners themselves but also between the partners' closest
associates. Above all, xeno; were expected to show a measure of protective
concern for each other's offspring. On two occasions in the course of his
service as trierarch, Pasion's son Apollodoros was given vital assistance by
the xenoi of his father. 4 t Pausanias of Spana rescued the concubine of a
Persian magnate from the hands of Greeks engaged in slaughter and spolia-
tion on the ground of her being the daughter of Hegetorides, a xenos of his
from CoS. 42 Surely, this was a time-honoured practice, already recognisable
in its fully fledged form in the epos. Priam's son Lykaon was once captured
by Achilles and offered for sale in Lemnos. The lord of Lemnos had already
given a price for him when Eetion of Imbros, a xenos, ransomed Lykaon -
"a great price he gave", comments the poet - and set him free. 43
Sometimes, the principle of protection would be carried to extremes, and
a xenos assumed the role of a foster-father for his partner's son. In plays by
Euripides, Orestes, son of Agamemnon, was brought, following the
treacherous murder of his father, to Strophios, ruler of Phocis, a xenos of
Agamemnon's house. Strophios nurtured Orestes together with his own
son, Pylades. The two young men were xnro;, and Pylades was described as
like Orestes' brother."" This is a pattern which can be recognised in real life.
Kraugis of Megalopolis died when his son Philopoimen was still a child. But
opportunely, a xenos, Kleandros of Mantinea, took responsibility for the
rearing and the education of the orphan. 4s In somewhat later times, the
hellenised Jewish prince Antipater is said to have concluded relations of
philia and xenia with other dynasts, and in panicular with the king of the
41 Demosthenes 50.18 and 50.56 (Against Polycles).
41 Herodotus 9.76.3.
4) lli4d 2 t .42.
44 Euripides, Ekara 83 and Ore$tes 1015, where Pylades is called loa6~ civfJ(). This
fictitioul example il the only inStan~ which might indicate that one'l pla~ of origin, rather
than permanent abode, was the crucial factor determining one'l Itatus al xenos: Orestes and
Pylades were reared from early childhood in Phocil, but were xmo; in virtue of Orestes' Argive
origin.
4J Polybius 10.22 and Plutarch, Philopomlnr t. Cf. Plutarch, Aratus 3.1 for a similar story
about Aratul of Sicyon. I discuss further evidence on this topic in Sccrion S.4.
Arabs, "to whom he had entrusted his children when making war on
Aristobulus u • 46
Foster-parenthood was often romanticised - in panicular when projected
into mythology (Fig. la-d). But it must have been generally recognised that
it was primarily intended as yet another device for cementing the alliance.
For example, according to Aeschines, in 367 B.C. the Athenian general
Iphikrates was implored to assist the Macedonian royal house against a
pretender by vinue of his having been an adopted son of Amyntas, father of
Philip. Aeschines purports to reproduce verbatim the appeal which he
himself made to Philip several years thereafter:
Then ... your mother Eurydike sent for him [i.e. Iphikrates], and ... put your brother
Perdikkas into the arms of Iphikrate5. and set you upon his knees - for you were a
little boy - and said: "Amyntas, the father of these little children, when he was alive,
made you his son, and enjoyed the friendship of the city of Athens; we have a right
therefore to consider you in your private capacity as a brother (adelphos) of these
boys, and in your public capacity a friend (phi/os) to US".47
An astonishingly similar incident is preserved in Lysias· 18th oration. After
the downfall of Athens in 403 B.C., Diognetos of Athens, brother of Nikias
the general, is said to have approached Pausanias, the victorious Spartan
general, with a theatrical gesture:
...as soon as the Lacedaemonians and Pausanias had arrived at the Academy. he (i.e.
Diognetos] took the son of Nikeratos (Nikeratos was the son of Nikias) and us (i.e.
the children of Eukrates, another brother of Nikiasl. who were children. and laying
him on the knees of Pausanias, and setting us by his side. he told Pausanias and the
others present the tale of our sufferings and the fate that had befallen us and called
on Pausanias to succour (~9ijoaL) us in vinue of our bonds of philia and Junia.
and to do vengeance upon those who maltreated us. 48
These children were apparently not brought up in Pausanias· household, but
the fact that gestures reminiscent of the duties of foster-parenthood were
employed indicates how deeply ingrained an element of the institution
foster-parenthood was.
Sometimes, variations of the practice are encountered. Logbasis of Selge
was a synethes (i.e. a co-dwelling intimate) and xenos of the Hellenistic
prince Antiochos Hierax, and acted as a foster-father not to Antiochos' own
child but to the daughter of Mithridates of Pontus, whom Antiochos held
hostage. Once again, the practice is painted in the brightest of colours:
logbasis, according to Polybius, brought up the young lady as his own
46 JoscphU5.J~h Antiquities 14.122. d. The J~h War 1.181.

4' Aeschines. (On the Embtwy) 2.28. The inaccuracies in the account - Philip was at that
time hostage in Thebes - do not affect my argumt'flt.
48 I.ysias 18.10 (Property of Nicias' Brothn).
fiB. la-cl The ~ as • foster-father

Apollodol'Ol' account of the story of Pel~us and Oliron (3.13.3-6), ...hen compared
WIth ritualiled friendship as rtCOIlsuue:ted from non-mythical sources. revals that
the practices prescribed by the institution underwent only minor distortions when
projected nto mythology. I paraphrase the relevant passages. Believing a false:
accusadon, Akastos desened his friend, Peleus, while asleep on Pelion, and hid his
magic sword in cow's dung. On aminltt Peleus was caught by the centaun dwelll"l
there, and would have perished jf he had not been saved by Charon, who .Iso
ratored him his S9iord. FollOWIng this euerge$ia, Chlron ~ mcd the role of n
ideal xetOS: he provided Peleus with • due as to how to leize and marry Theda. the
sea D)'IIlph whO was constandy metamorphosing herself; he acted, a IOrt of
sponsor at their wedding on Pelion (Fig. 2, p. 27), and gave Peleus an ashen-spear a.
a present (a prodigaous weapon which he would later bequeath to h • son, Iliad
16.140ff.); finally, when, becaUIe of • disagreement over the use of malPCt Thealleft
J Features Shared With Kinship 2S
ro
·C
OJ
+oJ
ro
~
"0
OJ
+oJ
s=-
d .-o
L.
>.
Co
o
U
Peleus, Chiron received the child she bore to Peleus, fed him on the innards of wIld
animals, and named him Achilles (by an absurd etymology, Achilles could mean
"non·li,,-': the inf t had nut pu his lips to the b£e&sr). As In real life, me rela-
tionship was here started by an euergesia and, once established, involved services of
a ritual and a practical nature: giving adviet, sponsoring a wedding, giving .. present,
foster-parenthood, naming, and, throughout, boetheia. The paintings above repre-
sent different venions of the apparently solemn occasion on which the Infant wai
placed under the' centaur's tutelage (for further examples, see Brommer t 1960)
249-50). In other words, the fabulous elements of the myth rested on a scaffold that
matched the categories of rcallife. It waS probably by appeal to chis real rcfCl"ent that
the vase painten hoped to conjure up the entire myth. For the traditional view of the
practice as leducation', el. H.-I. Marrou, Histoire de /·iducation dans '·antiqHiti
(Paris, 1948) 36fl. With nne 11-12 at 51 Ii for its emulation by the city, see Fig. 14 (p.
137).
daughter and treated her with special kindness. 49 Indeed, the duties of
foster-parenthood seem to have engendered a sense of responsibility which
did not cease even if the child and the actual (or potential) foster-father lived
apan. Some of lsocrates' orations and letters seem to have been written, not,
as is usually thought, in the formal tone of a master to a pupil, but rather in
the paternal tone of a xenos conscious of his responsibilities. so
Special forms of intervention were however sought not only with regard
to rearing of children but also in other life crises. Xeno;, in brief, were
expected to perform certain ritual services for each other. Satyros of Athens,
for example, undertook at considerable cost to himself to provide the
daughters of a deceased xenos, Apollophanes of Pydna, with dowries and
give them in marriage. SI Similarly in myth, the wise centaur Chiron appears
as a sponsor at the wedding of Peleus, his xenos, and Thetis, the sea nymph
(Fig. 2). Other types of ritual service concerned death. It seems to have been
the duty of a xenos to look after the earthly remains of a dead panner and
celebrate his memory. Xenophon, for instance, made a votive offering and
dedicated it in the treasury of the Athenians at Delphi, inscribing on it his
own name and that of the deceased Proxenos. "For Proxenos was a xenos",
he explained tersely. 52 The connexion between this passage and a dedica-
tion made by a Xenophon in Attica is uncenain,SJ but a grave-stele bearing
the inscription - "This is the grave of Promathos, a man who loved his
xenos" _54 might well have been set up by a xenos abiding by his duties. The
poet Simonides of Ceos seems to have been motivated by a similar sense of
responsibility, when, along with the famous epigrams he composed over the
tomb of the Spanans who fell at Thermopylae, he set up at his own expense
an inscription for the Acamanian diviner Megistias "for the sake of
xenia".ss
.9 Polybius 5.74 wirh Walbank (1957·82) voU, 600. Funher evidence on thil custom is
disamed in S«rion S.2. For parallels in the ancient Near East and in medieval vassalage, see
Weinfeld (1970) and Marc Bloch (1961) 224ff.
so 2. Nicocks, a treatise on the duties of monarchs written to Nicodes of Cyprus from
lsocrates as a 'gift' from a friend abroad; l...mer 6, To 1M Childrm ofJason, making reference
to the orator's JCnlUJ with Jason and Polyalko (presumably jason's wife, d. n.75 below) and to
an invitation on their part to come and reside with them in Thessaly; Letter 7, To TimothtlU,
the son of Klearchos, lsocrates' JCnlOS in Heracle2 Pontica. In this context should also ~
mentioned l.ctkr 8. addressed to the rulers of Myrilene. written by boerares on behalf of one
Apor, instrUCtOr in music of the children of Aphart\l$. Aphareus was the son of Hippial of
EJil and adoptive son of lsocrates (D1onysiul of Halicamassus, ISOCJ'al~s 18).
.U Demosthenes 19.192·5 (De Falu ugfftiorre), d. Diodorus Siculus 16.55.
Jl Xenophon. Anabasis 5.3.5. Cf. allO ApoIlodorus, 8ib/iolMCff 2.S.4 for the burial of the
centaur Pholos by his XnlOS, Herakln.
SJ M.Th. Mirsos and E. Vanderpool in H~spma 19 (1950) 25·6 and 391.
J4 Pfohl (1967) no.l0.
ss Herodotus 7.228.4. Callimachus, Epigr«m 2a seems to be an epitaph written for Hera·
IdeitOl, a urlOS from Halicamassus. For the adoption of this custom by the city, sec Section S.4
and Fil. 13, p.136.
3 features Shared With Kinship 27
Fig. 1. Ritual ervices

Red figure vase, now in Palermo (CorpJU Vasorum Antiquorum pls.29-30), showing
Peleus, his would-be wife Thetis, and his benefactor, protector and friend - the wise
centaur Chiton. It is not clear whether Peleus is here presenting Thetis to Chiron or
Chlron is giving his blessing to the matrimonial union. In either a xmos, like a
ChIj,.ba,D godfather, is sho~ to be inyolvcd in the ritual events of his partner's life.
The circle of protegis could be widened to include other family members,
and even dose frieods. J~son, anxious to send hi$ wife, Medea, away, could
thus promise her that his xeno; abroad would bestow benefits on her.s6
Crito offered his friend, the condemned Socrates, the chance to flee to
Thessaly and be placed under the protection of Cnto's xenoi there. S? Lykon
of Heraklea sought in Athens the aid of his xenos, Archebiades, and of
Archebiades' phi/oi. s8 And the Parian xenoiof an anonymous Aeginetan (by
naturalisation, not birth) acted as a keeper of the fonunes of the Aeginetan·s
friends. 59 The quesnon of who one's closest associates were was apparently
open to different interpretations. It is more than likel)" that it was the
mechanical extension of this principle to the whole citizen body that gavr
rise to the institution of proxeny: a man', fellow-citizens becarne, by vinuc
of their common citizenship, the proteges of his xenos. 6O
Another feature by which ritualised friendship mimicked kinship con-
s. Euripides, M~d~iI 6U.14.
57 Plato, Crito 45c.
s. DemoKhcna 52.3 and 52.21 (Ag"inst Callip",").
J. lsocrates 19.18 (A~g;If~';cus).
60 I explore this proposition in Sccrion S.4.
Ritualised Friendship 28
cems the nature of the services that the partners were expected to render
each other. The extraordinary thing is that these services were not restricted
to a narrow range of their affairs. To be sure, normally ritualised friends
would help each other solve some of the trivial problems of daily life by, for
example, lending money or supplying practical advice. The span of the
interactions was in these cases short. But sometimes - and this seems to be
the ideal purpose of the institution - they were supposed to assume respon-
sibility for the whole range of each other's affairs. In crisis or in extreme
adversity, a ritualised friend acted as a haven of refuge for his unfonunate
partner. This was no ordinary hospitality, ancient style or modern. Besides
shelter and food, the host undertook to take care of the welfare of his
panner, material and spiritual- to feed him at his table and maintain him in
his household, as if he were a foster-child. Agesilaos is even said to have
catered for the homosexual love affairs of a xenos (the son of the Persian
Pharnabazos who was staying with him), and the Boeotian Proxenos
likened his guest-friend Cyrus to his own patris, his fatherland. 61 But the
best illustration of the spirit in which this duty was conceived is supplied by
a fragment from a lost oration by Lysias. An Athenian, speaking in the law
""0
court on behalf of his xenos' son, explained charmingly that: (l)
+..J
Kephisodotos, the defendant's father, was a xenos of mine, gentlemen, and when we ..c:
0'1
were in exile, I and any other Athenian who wished to, enjoyed his hospitality in L
Thebes, and received many favours at his hands in public and in private before >-
0..
returning to our own city. Now that these gentlemen have suffered the same o
misfortunes as we, and have come in exile to Athens, realising that I have a huge debt U
of gratitude to repay (~6.tl.V x6{>t.v), I have taken them so completely into my
family (oirt~ ol)(d~ a~o,,~ "n£6t;OJ&TJV) that no visitor without prior knowledge
would know which of us was the owner of the housc. 62
The unfortunate man lived thus in his host's care and under his protection,
and so did his kin and friends. Owing to the great demands that such a
practice put on the resources of the partners, it is hardly surprising that it
was mostly confined to the very rich: the Great King of Persia, the petty
rulers at the fringes of the Greek world, the Macedonian and Hellenistic
monarchs acquired special fame for being able to provide services of this
kind. Dozens of city decrees survive honouring individuals for having bene-
.. Xmophon, H~Jltn;ca ".1.40 and Arrabasu 3.1....

62 Lylias, fragrmnt 78 (Thalheim) - Dionyliul of Halicamanul, 1S4J~ 6. In a similar
fashion, king Archidarnos, after flmng in 222 R.C. from Spana for fear of KJromenes, found
refuge in the hou~ of a xmos, Nikagofll of Messme. Nikagoras is said to have "gladly
received him in hil houK and provided for his wants" (Polybius 5.37.2).
4 Features Shared With Friendship 29
fited a city or a special group of citizens while 'spending their time' with a
ruler. 63
Thus, in several ways rituali~d friendship mimicked aspects of kinship
relations. Yet it was not kinship in the proper sense of the term. Real kinship
is given: it follows from birth. Ritualised friendship was an acquired rela-
tionship. It thus somewhat resembled kinship through marriage, since in
both marriage and ritualised friendship a ritual act was deemed necessary to
create a bond of affinity between unrelated individuals. But at this point the
similarities stop. For, as we shall see in Section 5.3, ritualised friends
performed for each other services which lay mostly outside the sphere of
kinship obligations. Acting at one and the same time as quasi-kin and allies,
they provided what can best be described as supplement to the potential of
the family group.
2.4 FEATURES SHARED WITH FRIENDSHIP
Ritualised friendship also resembled friendship in some respects. Yet it was

not friendship in the normal Greek sense of the term. Friendship in the
Greek states bound together individuals partaking of the same social system -0
(l)
+..J
and sharing similar values. Confidence was generated through a lengthy ..c:
CJ)
process of interaction and gradually developing intimacy, and the resulting
L
relationship was sustained - or disrupted - through permanent interaction. >-
0..
A good illustration of this process may be gleaned from a pseudo- o
Demosthenic passage: U
Nikostratos, whom you see here in coun, men of the jury, was a neighbour of mine
in the country, and a man of my own age (he/ikiotes). We had long known each
other (yv~~ J1tv JIOl dXE) but after my father's death, when I went to live in the
country, where I still live, we had much more to do with one another, since we were
neighbours (geitones) and men of the same age (helikiota,). As time went on we
became very inrimate (oikeios); indeed I came to feel on such intimate terms with
him (oikeios) that he never failed to win any favour he asked of me... 64
Ritualised friendship, by contrast, was concluded between persons who
originated from different, and at times, drastically dissimilar social systems,
and who had no previous record of social intercourse. Intimacy was estab-
lished not through a lengthy interaction, but abruptly, as in marriage,
61 Many such instances will ~ mcounrered in this study. I here list only a few examples.
Persia: Thucydides 8.6.1 (KaJligeitos of Megara and Timagoras of Cyzicus living at the court of
Phamabazos). Petty rulers: lsocrarn 17.3 (Trapevticus) (Sopaios, a Greek, staying with
Saryros. king of Bosporus). For Macedonia and the Hellmisric monarchies and for the
technical term l)&.(1T~V xQ()6 nv&, 5C'C Herman (1980181).
IA (Demosthenn) 53.4 (Again" Nicostratus).
through a ritual aet. 6S Funhermore, once the friendship had been estab-
lished, the partners were expected to render each other assistance not while
living together, but while dwelling at a distance from each other.
When Aristotle suggested that friendship between aliens, xenia, was the
firmest of philiai, it was this aspect of separateness that seemed to him to be
critical. In an attempt to rationalise what probably circulated as a popular
proverb, he observed that xenoi, unlike philo;, have no common object for
which they dispute with one another. Friends, on the other hand, who are at
the same time each other's fellow-citizens, compete for superiority and
engage in violent dispute. As a result, he adds gloomily, they cease to be
friends. 66
Aristotle's observation may serve as a convenient staning point for a
more systematic elucidation of the differences between ritualised and civic
friendship. Civic friends were related to each other not merely through
friendship but through a whole array of other roles. Besides friendship, they
had rights and obligations towards each other by virtue of their common
panicipation in formal institutions and in numerous informal groups and
ties, which together made up the city's social texture. 61 In addition, there
was the role of the polis itself: fellow-citizens would be linked together by "D
(l)
+..J
abstract, intangible rights and obligations even if not bound together by ..c:
personal ties of attachment. 0'1
L
As a result of all these factors, the opponunities for friction and conflict >-
0..
were greatly increased. Hence, Aristotle's pessimistic remark about the o
termination of friendships. But this dense involvement had imponant con- U
sequences for the dynamics of civic friendship. The sheer intensity and
extent of the interactions could become instrumental in compelling the
fulfilment of obligations. People interdependent through several spheres of
activity could use the sanctions available through one type of interdepend-
ence to enforce another. And, as a last reson, they could even set in motion
the city's legal machinery. The employment of such coercive means might
seem self-evident in other complex societies. But it must be borne in mind
that in the Greek city this was a recent innovation. The polis, unlike the
inchoate groups that preceded it (or the contemporary inchoate groups
outside it), could lay down and enforce formal rules regarding personal
relations. As a result, the individual was constrained not only by the implicit
dictates of pre-political morality, but by the explicitly formulated rules of
the city.
6$ The elements of the ritual are explained in Osaptcr 3.

66 Magn4 Moralia 2.121h 46.
•, Cf. finley (1983) 82.
5 The Cross-Cultural Perspective 31
A single example will suffice to illustrate how these peculiarities of civic

friendship operated. Demosthenes· father at his death had to secure the
perpetuation of his family household until his eldest son, the future orator.
reached maturity. The old Demosthenes appointed guardians for his chil-
dren, two nephews and an old friend. He bestowed valuable gifts on them,
arranged for his widow to marry one of them, and treated them as his
friends and kin. As the orator himself explained years later, "His thought
was that, if he should unite these men to me by closer ties, they would look
after my interests the better because of this added bond of intimacy
(oikeiotesr·. 68 The strategy was designed to multiply and intensify the ties
of dependence between the guardians and the family, but, alas, it proved
inadequate. The guardians abused their powers and deprived their wards of
their property. In this predicament, however, the young Demosthenes was
able to seek redress with the aid of the city's judicial machinery. He won the
case, and the guardians were ordered to pay damages. 69 Thus, although
'friendships' could be terminated, obligations would be fulfilled.
These aspects of civic friendship contrast sharply with their counterparts
in ritualised friendship. Xeno; lived far away from each other, and this
physical separation lent the institution its most distinctive feature. Although
attempts were made to multiply, as Demosthenes· father did, the ties of
interdependence, in most cases the framework of xenia remained the only
channel through which co-operation was carried out. Unless they came
together to form a new group (as, for instance, in the case of the permanent
followers of rulers), xeno; could not bring direct social pressure to bear on
one another. Nor did they have at their disposal a formal enforcement
agency which could be brought into play if obligations were disregarded.
The communities in which they Jived were not pan of larger organisations
and were mostly at war, or at least in a state of mutual hostility. Operating
outside the framework of social order which fettered the citizen, ritualised
friends could not rely on appeals to external authority.
2.5 THE CROSS-CULTURAL PERSPECTIVE
The pattern of social relationships outlined above is not peculiar to the

ancient Greek world. Institutions displaying similar features occur in other
societies. The mere observation of similarities, however, could be as mis-
leading as it might be helpful. A comparative perspective can offer real
insights only if the features to be compared are closely defined and the units
61 Demosrhenes 27.5 (Against Aphobos I) .
•9 The fact mar rhey did nor pay docs nor affect my argummr.
of comparison are rigorously delimited. More specifically, then, the prob-
lem that presents itself is this: how, in each society, can these institutions be
isolated from others into which they shade, and, once isolated, be built into
a comparative framework so as to bring out similarities and differences?
A classificatory system has been proposed by Pitt-Rivers. He envisages a
whole spectrum of social relationships marked by a common principle, that
of 'amity'. Amity, "the moral obligation to feci - or at least to feign -
U
sentiments which commit the individual to actions of altruism marks off ,
these relationships from other, non-amiable social relationships. The sphere

thus delimited represents one side of a universal dichotomy by which, from
an egocentric point of view, one's social universe is divided into friends and
enemies. Within this amical sphere five sub-categories are identified: 'real'
kinship, adoptive kinship, ritual kinship, ritualised friendship, and
friendship unritualised (sec diagram).70
'A m iable relations'
"D
(l)
+..J
..c:
0'1
Ritual kinship = Ritualised Friendship L
friendship unrirualised >-
0..
o
The two intermediate categories, 'ritual kinship' and 'ritualised U
friendship', are particularly relevant for our purposes. The concrete rela-
tionships included in these categories share two major characteristics. First,
the panners involved are in each case predominantly (but not unexcep'
tionally) non-kin - people who come either from different social units
(villages, tribes, nations), or at least from different families. Secondly, their
relationships are conceived in terms of kinship. There are, admi!'!t:~;y,
differences in the degree to which the kinship element is manifest in panicu-
lar instances. Yet, as the diagram itself shows, the distinction between 'ritual
kinship' and 'ritualised friendship' is virtually insignificant. The fun-
damental element common to all these institutions is thus the extension of
the kinship principle to non-kin.
The virtue of Pitt-Rivers' typology is that it makes it possible to
accommodate within a single analytical framework a whole range of institu-
tions which have hitheno seemed unrelated: the different variations of
blood-brotherhood, godparenthood, and bond-friendship. It came into
10 Adapted from Pin-Riven (1973) 96. For the concept of 'amity', 5« Fortes (1969).
5 The Cross-Cultural Perspective 33
being as a reaction against earlier attempts at classification that were based

on the kind of ritual needed to enter into a relationship: baptism to become
a godfather, blood rituals to become blood-brothers, and the exchange of all
sorts of sacra to become bond-friends. 71 For Pitt-Rivers, it is not the ritual
technique but the role achieved through the enactment of ritual that is the
essential typological criterion.
These simulated roles are marked by characteristics shared with kinship.
Yet, the people described by these characteristics do not identify themselves
with their real counterparts. In spite of idealistic statements to the
contrary,72 a blood-brother does not think of himself as a real brother, nor
does a godfather regard himself as a father. The essence of the relationship is
a fiction - 'as if' the incumbents were brothers, parents and friends. This
conceptual distinction corresponds to important practical distinctions. For
the partners in the relationship cannot, for the most part, assume real
kinship roles: a blood-brother is not a substitute brother, nor is a godfather
a surrogate father. In terms of their place and function in the total network
of social relations, they are all rather special kinds of friend. In other words,
real and simulated roles are complementary, not interchangeable. Only to a
very limited extent d~ the range of their duties overlap.
I propose in this study to view xenia as another instance of such ritualised
relationships.73 The institution fits the general pattern: it existed between
non-kin - indeed, complete strangers - and, as demonstrated in Section 2.3,
':'1 The best iIIustr:uion of this principle is Tegnaeus (1952), who inclucks within a single
analytical framework a wide range of past and present societies on the sole account of their
pr.1ct1sing some sort of .1l1iance through blood ritual. An identical principle is applied by
Gudeman (1971) in a cross-cuhural study of compadralgo. Explicitly, the methodology is
upheld by Davis (1977) 226-32 and 236·8. For earlier anempts to break away from such
typologies, see Eismstadt (1956) and Pin· Rivers (1968).
12 Everywhere in popular thought ritualised relationships tend to be pr~nted as somehow
stronger and more sacred than kinship ties. The central African Aundc, for example, common-
ly '·contrast blood·brotherhood with kinship, extolling the first in comparison with the
second" (Evans-Pritchard (19.13) 399). In the tribal societies studied by Tegnaeus, it is likewise
assumed that "blood·brotherhood is something greater than blood· relationship" (Tegn.1et1s
(1952) 13 and 66). Similarly, a godparent is often thought to be more imponant than the rul
parent. The explanation Biven for this in a modem Greek village wa5 that "through the parents
the child is born into t~ world of the flesh, through the godparent it is born into the world of
the spirit" (Du Boulay (1974) 163; d. Gudcman (1971) for a study of compadra:.go as a
'spiritual' tie). This should be compared with the Homeric lines "A man needs no grut grasp of
things to understand that a xeinos and hilut~s (suppliant) is no less precious than a brother"
(Odyssry 8.545-6), and with Aristotle's explanation, based probably on a popular proverb,
that "xmia is the firmest of all phi/iai" (Mapa Mora/ia 1211a 46). For the relationship
betwttn Junia and hiketeia, ~ below, ~etion 3.3.
?J The mixture of kinship and friendship elements makes it impos~ible to fit Junia neatly
into either of Pitt-Rivers' categories of ritual kinship or ritualised friendship. I have d«ided to
call the institution 'ritualised frkndship' (rather than 'ritual kinship') because th~ kinship
dement is not implied in its Greek name, in the way that it is with godparenthood and
blood·brotherhood.
34 Ritualised friendship
it was structured by a system of ideas mimicking kinship relations. Apan
from these basic features, xenia displays characteristics that are in pan
analogous to those of other ritualised relationships, in pan without parallel
in other cultures.
The imponance of viewing the Greek institution in this broader context is
twofold. First, it shows that the conceptual distinction between kinship,
friendship and ritualised friendship exists in societies outside the one under
examination. It thereby inspires some confidence that the proposal to treat
xenia as a specific, relatively autonomous, domain of social life is not
arbitrary. Secondly, it allows us, on the basis of a coherent analytical
framework, to make controlled comparisons with ritualised relationships in
other cultures. Throughout this study I shall try to show that this view of
xenia will help to bring into focus cenain aspects of the institution which
cannot otherwise be seen.
2.6 SOCiAL STATUS
Ritualised friendship appears as an overwhelmingly upper-class institution.

-0
The people involved in it belonged to the elites of the ancient world - to a (l)
+..J
small minority renowned for their wealth and identified by high-sounding ..c:
CJ)
titles such as 'hero', 'king', 'tyrant', 'satrap', 'nobleman', 'general', 'politi-
L
cian'. People of humbler standing are significantly rare. 74 Non-free men are >-
0..
absent altogether. And women are extremely rare. There are remarkably o
few references to male-female alliances - the xenia between Hekabe, Queen U
of Troy, and Polymestor, ruler of Thrace, being the most explicit one 75 -
but there is no sign of an alliance involving two women. Jason's suggestion
'4 For example, ThemilOn in Herodotus 4. 1S4 was an mrporos; Polemainetos in lsocrates
19.5 (Aqirrn;cus) and Megistias in Herodotus 7.228.4, rrumtm; Satyrot of Athens, the xmos
of Apollophanes of Pydna (Imnosthmes 19.193-4 (De Falu ugatwne)), was a comedian. It
should be noted that these were all itinerant profeslions and exttpt for the mtporos all were
·r~ble'.
Euripides, Hecuba 710. The case of Mania of Dardanul and Phamabazos (Xenophon,
Hellmiu 3.1.11 ff.) is presumably another instance of man-woman alliance. though t~ word
xmos does not appear in the text; and so seems to be that of lsocrates and the wife of Jason of
Pheru. Polyalko (Isocrates, ultn 6.1). unainty is impossible. but, as argued by G. Mathieu
in the Rude edition of lsocratel (vol.lV (1961) 169), the circumflex of noAuoAxo~ found in
the Urbinas codex, as against the almost impossible noAUQxovc; in others. seems to favour
noAUaAxciJ, not nol~ as the nominative of the name of lsocutes' xmos. For a different
interpretation of the genealogy (Polyalkes, a brother or eldest son of Jason) and for a summary
of earlier views. see H.D. Westlake, Theswly ;n the Fourth Cnrhlry 8.C. (London, 1935) 68.
6 Social Status 35
to Medea to send her to his xenoi abroad implies that the family's connex-
ions with xeno; were in fact monopolised by the husband. 76
The lower we descend on the social ladder, the less common ritualised
friendship becomes. Crito's suggestion to Socrates that he seek refuge with
his [Crito's] xeno; in Thessaly implies that he, Socrates, did not have xeno;
there. Conversdy, Andocides' bragging "I have formed xenia; and philo-
tetes with kings, with states, and with individuals too",77 was the mark of
the aristocrat. Made in the name of Pasion, the metic-banker turned into
citizen, a similar claim probably smacked of the upstart. 78 A"ivisme was
also involved, according to Demosthenes, in Aeschines' friendship with
Philip. Now that Aeschines has become Philip's xenos, the former clerk,
"the son of Glaukothea the tambourinist", parades in the agora in the
company of other aristocrats, apes their manners, and looks down on
ordinary people. 79 The conclusion thus imposes itself that the story told by
the wretched beggar (Odysseus in disguise) about his sojourn in the court of
kings was intended as proof of his noble origins - a quality which would put
him on a level with the suitors. 8o
Only when both partners enjoyed an equally elevated social standing was
-0
the relationship in harmony with its ideal image. Polybius thus relates in a (l)
+..J
reverential tone that Philopoimen, scion of noblest Arcadians, was brought ..c:
up by Kleandros, a xenos of his father, and the noblest of Mantineans. 81 0'1
L
And, when the anonymous Siphnian aristocrat (in )socrates' Aegineticus) >-
0..
wished to impress the jury with the similarity between himself and one o
Thrasylochos, he said: "we had the same philoi and xeno;".82 Such a deeply U
ingrained element of ritualised friendship was equality, that when the semi-
hellenised ruler of Commagene, Antiochos I, wished to give publicity to his
76 Euripides, M~dea 613.

n Andocides 1.145 (0" tM M)'stnies); d. lysia.s 6.48 (Agamst A"doddes), where Ando-
cides' bragging is maliciously invnted: "But for all his wulth, and the power in his pounsion,
the xenos of kinp and tyrants ...".
71 Demosthmes 50.56 (Against Polycks). As an ex-II ave, Pasion must have lacked 'ances-
tral guest-friends'.
7. Demosthmes 19.314 and 295 (De Faha LeRatimu), and 18.284 (0" tM Crown). Cf.
19.225. where Pythokln. since his visit to Philip, is said to rum .1Side whenever he m«~ the
orator. For Pyrhokles' aristocratic pedigree and the phr3se loa PalVtOV n\&xAfl. see Davi~
(1971) no.12444.
10 Odyssey 14.19tff.
I. Polybius 10.22.1; d. Pausanias 8.49.2 and Plutarch, PhiJopomten 1. In thil context we
should perhaps mention again the Jewish prince Anripatu, whose riK to power was attributed,
among other things, to his forming "relations of philia and xenia with other dynasrs"
(Josephus, }ewnh A"tiquities 14.122).
12 lsocrates 19.10.
own divine qualities, he set up a series of monumental reliefs commemorat-
ing himself as the xenos of gods and heroes (Fig. 3a,b).8.1
However, the possession of xeno; was not only a status symbol but also,
and chiefly, an asset of real value. The worth and power of a man both
within his community and outside it was in a sense in direct proponion to
the number of xenoi attached to him. It is no mere coincidence that the great
figures of ancient history, men such as Cyrus, Alkibiades and Agesilaos,
crop up persistently in the sources as having an unusually high number of
xeno;.84 True, xenia was but one of the ways these powerful individuals
forged bonds outside their own communities. The other was marriage, and
both devices could, and indeed did, serve as political alliances. But there can
be no doubt that strategically xenia was the more advantageous. First, if
judged by the example of Athens, the freedom to marry outside the com-
munity was in the Greek states to some extent restricted. In Athens, follow-
ing Perikles' law of 45110 both parents had to be citizens if their children
were to receive citizenship; else they would be classified as bastards. 85 By
contrast, we know of no state-imposed regulations designed to discourage
the conclusion of x~n;a;. Like ritualised relationships in many cultures,
xenia was surrounded by an aura of sacrosanctity. Secondly, the number of
alliances a man could forge through marriage was confined to the number
of his marriageable close kin (children, sisters). By contrast, the limit to the
number of xenoi that he could possess was set only by the number of
individuals with whom he could interact. These are perhaps the reasons why
in some cases, where both xenia and marriage were used simultaneously,
marriage was regarded as the complementary and subordinate device. Hiero
of Sicily, for example, is said to have concluded a pact of xenia with Pyrrhus
of Epirus, reinforcing it (and not vice versa) with a marriage (epigamia)
between Hiero's son and Pyrrhus' daughter. 86
I) For the role of the handshake in concluding me alliance, see S«tion 3.3 and Fig. 5, p.52.
14 See Appendix A.
In Autocratic rulen were of course free of such inhibitions. See Gernet (1953) for marriage
alliances of tyrants and Seibert (1967) for those of Helleniltic rulen.
•• Pausanial 6. t 2.3. The textual problems involved in this passage (which do not affect my
argument) are discussed by P. leveque, PyrThol (Pans, t 957) 68Off. Thil and 101M other
examples (e.g. the ~ace treaty betw«n the lydians and Medes, reinforced by a marriage
betwem the rulen' familin (Herodotus 1. 74); 0iM'US giving his daughter to ~lIerophon (1IUu1
6. t 92); Omtn' lister Electra marrying Pylades, Orestes' xmol (Euripides, Electra 1241 and
t 340)) show that, unlike some ~do·relationships in other cultures (certain types of god-
parenthood and blood-brorherhood, d. Tegnaeus (1952) and Gudeman (1971 )), xmUJ did not
create an incest taboo preventing the creation of real kinship relations.
6 Social Status 37
a b
Fig. 3a,b Social ltatu ,

Relief steles made of basalt lava, found in Sofraz Kay, showing Antioc:hos I, King of
Commagene (66-64 B.C.), engaged in a ritual handshake (dn;os;s) with a Apollo-
HeliOi and b Her.Id~. In a series of monumental reliefs (see J. Wagner in Anlik
Welt 6 (1975) 51·9 for funher examples), Anriochos posed as the x~nos of Sods and
heroes. The reliefs could not have conveyed their meaning (that Antiochos was on a
level with his divine parmers) if the belid that xenia implied equality was not
widespread. Indeed, in the insaipuon below relief a (SEG 26. 1623), one of the
epithets of Antiochos js 'THEOS'.
Within the narrow elite group from which the large- majority of ."enoi
were drawn, one's formal status, birth, or official position was of relatively
little impon~nce. What mattered most was the po~ession of a quality which
the other needed, and that IS why, an fact, a bond of ritualised friendship did
not necessarily involve ~xaet social ~uals. Ther~ is, in this respect, a
marked Incongruity between image and reahty. In spite of possible de facto
inequalities, the terminology and ethos remained those of friendship, and
friendship, as a common proverb had it, meant equality.17 Agamemnon and
87 AriltOd~. E.thIUJ Nic0m4chea 8.IIS1bS and Politics l.1281blO.
Idomeneus were thus called hetairoi; Odysseus, ruler of Ithaka, and the
wretched beggar from Crete (in fact, Odysseus in disguise), were xeinoi
patroioi;88 also xenoi were Xerxes and Pythios, or Eteokles, King of Oaxus,
and Themison, the Therean trader. 89 Both Archidamos and Agesilaos, kings
of Spana, had as their xeno; the followers of leading politicians in Phlius. 9O
In the world of Odysseus, in a not dissimilar fashion, "Agamemnon, the
most powerful of the many rulers among the Hellenes, had as his guest-
friend in Ithaca not the king, Odysseus, but one of the non-ruling aristo-
crats, Amphimedon".91 And when Herodotus wished to point out the
special status that the Arabians enjoyed in the Persian empire, he said that,
unlike the other subdued peoples, the Arabians were xeno;. not inferior
subjects. 92 This possible incongruity between ideal and reality was skilfully
exploited by the Attic orators. Subordination. flattery, and abject servitude,
they would claim, were masquerading as friendship and equality. 93 These
ro
slanderers, however, had to resort to paraphrases to express the demeaning
practice since the language itself contained no words to designate unequal
partners. When, in the beginning of the second century B.C., formal ranks
were introduced within the ruling circle of the Ptolemaic kings, the notion of -0
(l)
gradation could only be expressed in a cumbersome terminology: from +-J
among the syngeneis; from among those equal in honour to the syngeneis; ..c
CJ)
from among the first philo;; from among those equal in honour to the first L.
philo;; from among the philoi; from among the archisomatophylakes; from >-
0..
among the diadochs. 94 In sum, the Greek language did not give rise to a pair o
U
of hierarchical status designations analogous to the Roman patronus-c/iens
or the medieval dominus-vassus. Polybius, trying to interpret for his Greek
public what the Romans would have called patroni and c/ientes, could not
find a more suitable term than philoi. 9s
II Odysuy 17.522.
I' lIiM14.266; HuodoNl 7.29 and 4.154.
90 Xenophon, Htlkrti,,, 5.3.13; d. Cyropudi4 5.1.3 for the asymmetrical xmi4 berwtcn
Abradatas of SU5a and the King of Bactria.
.. Finley (1917) 103, rderriul to Odyssey 24.114.
'1 HerodoNs 3.88.
•) Demosthenes 18.50-2 (0" tht e'O","); Aeschinn 3.66 (Agai,.,' Ctt$;Pho").
... Cf. Atkinson (1952) and Mooren (1975) 2. The ranks are cited in desccndinS order. For
the seemingly &ns yariegated nomendature of me Sdeucid kingdom, see Walbank (1984a)
71·2.
•, 'J'he tmn served Polybius to render bothme me
(theoretically) egalitarian ho$,,;IitmI and
inqalitarian patrOI'INS-clierrs bond. Only much I.trr writrn (Dionysiul of Haliamassus and
Plutarch) transliterated the Larin terms into Greek or equated patro,,; with f'rOsta'ai 'hd
clatn with ptlalai. Cf. Edlund (19n) 132.
6 Social Status 39
No status distinctions being available, the quality of the benefits ex-
changed remains the single clearest indication of the partners' relative
wealth, power, and status. 96 Using these benefits as an index, it is possible
to draw up a crude typology of relative statuses. Egalitarian types of
exchange tended to take place in cases where neither partner enjoyed an
overwhelmingly powerful position in his own social unit. Apollodo-
ros the Athenian, for example, when short of money as trierarch, borrowed
from the xeno; of his father in Tenedos. 91 It is a reasonable guess that
Apollodoros' father, Pasion, performed services of a similar nature through
his banking establishment in Athens. Even when one of the partners moved
into the household of the other, the relationship could be assumed to rest on
equality and reciprocity. However, relative status might alter in the course
of the interaction. If initially it had been a relationship of equality, in the
course of time it could have shaded off into a relationship in which one
ro
partner attained a position of strength, the other a position of weakness. In
other words, a horizontal tie linking together social equals may have been
transformed into a vertical patron-client bond. Goods then would tend to be
repaid by services, protection by loyalty, and willing co-operation turned
into coercive dependence. 98
Throughout Greek history, there were three distinct types of role in
which a man who was both a client and a ritualised friend could perform
services for his patron. (a) A client might be installed, or confirmed, by his
patron as a ruler in his own social unit. According to Herodotus, for
example, Xerxes made Theomestor, a Samian, tyrant in Samos as a reward
for having fought well on the Persian side in the battle of Salamis. 99
Theopompos relates that Philip of Macedon had established Thrasydaios,
the Thessalian, as a tyrant over his own people. loo Syennesis, King of
Cilicia, was on the other hand allowed to continue ruling Cilicia as Cyrus'
subordinate. 101 (b) A client might serve his patron in his own social unit
without concentrating superior power in his hands. Such was the case of
numerous citizens of the Greek cities acting as agents of Persian, Macedo-
nian, and Hellenistic rulers. (c) A client might be transferred into his
patron's social circle and become a member of his entourage. As such, he
might either be assigned resources securing him a certain degree of inde-
pendence, or he might be left totally deprived of unmediated access to
96 These bmdits are analysed in Chapter 4.
•7 Demosthents 50.56 (Aga;"sI Pol1cks).
,. For a similar evolution obKrvtd in the Roman hospitium. d. Badian (1958) 11-12.
" Herodotus 8.85.3.
100 FG,H 115 F209.
101 Xenophon. Arrabasis 1.2.26-7.
power and resources. Both possibilities can be exemplified by the career of
Histiaios of Miletus. Darius first sent Histiaios to build a fortress in Thrace.
But, when he was forewarned that the fortress might be used as a base
against himself, Darius summoned Histiaios to the coUrt and turned him
into a counsellor. Barred from independent access to power and resources,
Histiaios would be left entirely at his patron's mercy. 102
102 Herodotus 5.23". d. 5.35.
ro
"'0
(l)
~
..c
CJ)
L.
>-
0..
o
U
3
ENCOUNTER AND INITIATION
3.1 THE ETIQUETTE
In a picturesque and apparently fictitious tale, Herodotus relates that Sylo-

son, banished from Samos by his brother Polykrates, arrived in Egypt with
Cambyses' invading army in 525 B.C. Syloson happened to be in the market
place of Memphis wearing a purple cloak when he was approached by
Darius, as yet an undistinguished officer in Cambyses' retinue. Darius took
a fancy to the cloak and proposed to buy it. Prompted by a divine inspira-
tion, Syloson gave the cloak to Darius for nothing. Darius was delighted to
accept the generous offer. Syloson, however, regretted the loss of the pre- ro
cious object and blamed his foolish good nature for it. But years later,
hearing that Darius had succeecu:d to the throne, he went to Susa proclaim-
ing that he was one of the king's benefactors (euergetes). At first Darius
could not remember any Greek benefactor. But when Syloson recounted the "'0
(l)
story of the cloak Darius recognised him at once and made him an extraor- ~
..c
dinary offer. He said that since Syloson had made him a present when he CJ)
had as yet no power, he would now bestow on him boundless gold and L.
>-
silver. The gratitude (charis) for Syloson's act should be equal to what he 0..
o
would have owed, now that he was powerful, in return for a great favour. U
Syloson said that he did not want either gold or silver, and requested instead
that Samos might be conquered and handed over to him. This was accom-
plished, and Syloson became ruler of Samos as a tributary of Persia. t
Ritualised friends belong to different worlds, and one of the major
problems facing ancient writers was explaining how their paths came to
cross. Explanations range from fortuitous conjunctions of circumstances
through premeditation and design to supernatural guidance. Odysseus and
Iphitos met in rather pedestrian circumstances in the house of a common
friend in Messcne: Odysseus came on behalf of his father to claim a debt
that the demos owed him; Iphitos came in search of animals he had lost. 2
The encounter between Pausanias and Xerxes, on the other hand, was,
according to Thucydides, a matter of cold-blooded calculation. It was made
possible by an incident that occurred some time before the Spartao regent
lost favour with his fellow-citizens. When the Greek allied forces comman-
I Herodotus 3.139 If. In paraphrasing this and some other nories, I generally omil details
which are irrelevant to 1M underlying putern which I wish to uncover.
2 Odyssry 21. 14-23.
41
42 Encounter and Initiation
ded by Pausanias took Byzantium, certain 'friends and kinsmen' (philoi kai
syngmeis) of the Great King were captured in the city. The Great King
would undoubtedly have paid a substantial sum for their ransom. But
Pausanias surreptitiously set the captives free, pretending to the allies that
they had escaped from him. This was the first act for which the king owed
him gratitude (charis). Pausanias explained in a letter that his act was meant
to please the king, and, proposing to marry his daughter, he suggested
making Sparta and the rest of Greece subject to him. Thereupon Xerxes
replied that for the release of the captives a store of gratitude was laid up for
him for ever in his house, and accepted Pausanias' proposals. He said that
he would put at Pausanias' disposal as much money and as many soldiers as
were necessary to fulfil his promise. At this point of the story, we lose track
of the exchanges between the two men. But there are good grounds for
suspecting that Xerxes kept on heaping resources and precious gifts on
Pausanias. For how else can we account for what his fellow Spartans ro
perceived as a tactless display of wealth and an un-Greek display of bound-
less ambition?.3
To Polybius, it seemed as if the friendship between Anrigonos Doson,
King of Macedonia, and Neon, the hipparch of the Boeotian federation, ""0
(l)
~
were contrived by divine providence. Antigonos was sailing on a military ..c
mission towards a city in Eastern Locris when, owing to exceptional clima- CJ)
L.
tic circumstances - some scholars suspect a seismic disturbance - his vessels >-
0..
settled on the land. The situation was further aggravated by the appearance o
of a cavalry force led by Neon. Boeotia was in those days subservient to U
Macedonia as a result of a whole series of military blunders and political
miscalculations. But a powerful internal opposition, spurred on by the
memory of the glorious days of Leuctra, was seeking ways to throw off
foreign rule. When Neon's forces besieged the trapped vessels, the Macedo-
nians had good reason to fear the worst. But Neon curiously deflected the
course of events from its predictable conclusion. For, contrary to every-
body's expectations, he spared the Macedonians and allowed them to
continue their voyage. The other Boeotians, we are told, approved of
Neon's conduct, but his Theban countrymen were not pleased with it.
Antigonos, for his part, was very grateful to Neon. Years after the event, he
conquered Lacedaemon and nominated Neon's son, Brachylles, as epistata,
thus repaying the good deed (euergesia) that Neon had rendered him. This,
comments Polybius, was but the first step in the aggrandisement of the
fortunes of Brachylles' house. Subsequently, both Antigonos and Philip kept
J Thucydidn 1.128.30.
1 The Etiquette 43
on supporting Brachylles and furnishing him with moneyt whereby Brachyl-

les managed to crush almost all opposition and to compel everybody to side
with Macedon.'-
A uniform pattern of ideas underlies these encounter stories t a pattern
common to historians as dissimilar in other respects as Herodotus t Thucy-
dides and Polybius. Two strangers (not necessarily enemies t it must be
stressed) are first brought together. The prospect of a violent exchange
looms over the encounter. Violence is averted t however t mostly against
common expectations t by some generous gesture. This gesture t together
with certain rituals t serves as a kind of triggering event which brings about a
reversal: strangeness and enmity are replaced at a stroke by warm intimacy.
And t if we are to believe the ancient authors t the friendship that results from
this consists not only of external formalities t but of genuine sentimental
attachment and large-scale co-operation.
Whatever the reality that lay behind these stories, it is clear that the ro
authors had a stereotypical pattern in their minds which t ultimatelYt must
have been abstracted from the actual practices of the ancient world. We
must infer the existence of a stylised etiquette t made up of ceremonials, a
technicallanguage t and a whole series of ritualistic devices. In other words t ""0
(l)
~
there was in the ancient world a code of manners and a context of ideas ..c
CJ)
which governed, and by modem standards made uncommonly easy, the
L.
conclusion of alliances between enemies and strangers. It is the availability >-
0.
of this code, coupled with a certain conception of man's place within his o
own group, that explains some crucial episodes in Greek history - episodes U
which seem paradoxical and nonsensical from a modern point of view. How
is it that King Demaratos, deposed from the throne and forced to flee
Sparta, went to Persia, where Darius at once "received him royally and gave
him lands and cities"?s Herodotus recounts this as if it were one of the most
natural of things. Similarly, Thucydides, in his digressions on the fate of
Pausanias and Themistokles, takes for granted the availability of such
options. What was sensational was not so much the pact of friendship with
the national enemy as the abrupt downfall of these eminent men, who were
venerated beforehand as heroes in Greece. Nor were these exceptional
occurrences. In a detailed study, 337 Greeks have been identified as having
maintained extensive ties with Persian kings or satraps in the period before
Alexander. 6 Some of these individuals were central figures in Greek history,
quite apart from their Persian involvement. Besides those already mentioned
4 PolybiU5 20.5.6.
S Herodotus 6.70.2.
6 Hofsrett« (1978).
in our study, they include such figures as Peisistratos the tyrant, his son
Hippias, Mihiades, Kallias, Lysander, Alkibiades, Antalkidas, Iphikrates
and Evagoras of Cyprus. And the same pattern was to repeat itself in later
times probably with even greater intensity. When the kingdom of Macedo-
nia was about to engulf most of the Greek mainland, Theopompos com-
plained that Philip's most trusted aids, his hetairoi,
were men who had rushed to his side from very many quarten; some were from the
land to which he himself belonged, others were from Thcssaly, still othen from the
r~t of Greece, selected not for their supreme merit; on the contrary, nearly every
man in the Greek or barbarian world of lecherous, loathsome or ruffianly character
flocked to Macedonia and won the tide of 'Philip's hnairoi'.'
Near the end of the third century B.C., the popular philosopher Teles
could produce a seemingly paradoxical argument in favour of going into
exile. Most of these exiles, he implied, deserve no pity. "For some of them
rule cities on behalf of kings, are entrusted with nations, and receive large
gifts and contributions".8 Teles' 'kings' were the Hellenistic rulers to whom
the Greek cities had lost their independence.
In shon, throughout Greek history the possibility was always there of
creating new alliances or renewing old ones, even if this meant cutting
across boundaries of group enmity.
Two peculiar features of this code of conduct are worth pointing out: its
being constrained by moral, rather than institutional, sanctions; and its
transcending political as well as cultural barriers. It is thus idle to classify, as
modern writers sometimes do, some of the elements of this etiquette as
'Greek'. 'Persian'. 'Oriental', or even 'Roman'. Allowing for slight varia-
tions in detail, they formed part of a set of conventions stretching beyond
ethnic boundaries. It is not hard to recognise in this the vestigial remains of
those primitive conditions in which behaviour was govemed by custom, and
in which what groups there were had not yet developed particularistic rules
of conduct. Thus, what we have here is an elaborated version of the private
peace treaty. Its matrix is a system of thought which assumes that interac-
tion between strangers necessarily takes the form of untrammelled violence.
The code provides for an acceptable alternative to this. Rather than being
killed or suffering a humiliating defeat, opponents are given the honourable
option of becoming allies, subdued enemies of becoming followers.
3.2 PRELIMINARIES
In real life, encounters between would-be ritualised friends were greatly
7 FG,H 115 F224 • Arhenaeus 4.1678.
• Tele$, ed. O. Hense rrubin~ 1909; 2nd edn.) 23; my rranslation.

2 Preliminaries 45
facilitated by the extraordinary geographical mobility of the Greeks. The

reasons for being abroad varied enormously. A man could conclude a pact
of ritualised friendship while fleeing his community as an exile. But he could
also do it in a public capacity, as envoy, colonist, or military leader.
Sometimes, friendships were concluded while one of the partners travelled
abroad for pleasure, or in the pursuit of an itinerant profession (for exam-
ple, traders, soothsayers, actors). The role of the pan-Hellenic festivals
should also be noted. In his typically pious tone, Isocrates remarked that,
Having proclaimed a truce and resolved our pending quarrds, we come together in
one place, where, as we make our prayers and sacrifices in common. we are
reminded of the kinship (syng~neia) which exists among us and arc made to feel
more kindly IOwards each other for the future, reviving old x~niai and establishing
new ones.'
Many of the xeno; whose praises were sung by Pindar had no doubt been
acquired in such circumstances (Appendix A).
But war provided by far the most common context for the conclusion of
friendships. Greek authors seem to take it for granted that the commanders
of joint contingents should be xeno; to each other. Of Cyrus' Greek gener-
als, Klearchos, Proxenos, Sophainetos, Sokrates and Aristippos are explicit-
ly named Cyrus' xeno;; friendship with the rest is assumed. Isagoras of
Athens and King Kleomenes of Sparta are said to have become xeno; during
the joint siege they mounted on the Peisistratids. lO Special attention is given
to friendships with leaders of potentially hostile forces. Agesilaos and Phar-
nabazos' son became each other's xeno; when their respective armies had
just stopped fighting. II When Syennesis submitted to Cyrus, it was clear
that he thereby deserted Cyrus' enemy and his own ally, Artaxerxes. 12
Finally, the competitive gift-exchange that took place before the great
invasion of Greece between Xerxes and Pythios, the wealthy Lydian, is
portrayed as a sublimation of hostilities, the aggressive aspects of which
have not escaped Herodotus' notice: Pythios' son would soon be cut in two
as a result of an inappropriate petition that Pythios made to the king. U
The physiC21 proximity of the would-be partners was not always a
necessary condition for the establishment of friendship relations. Geog-
raphical distances could be surmounted by sending embassies, and rulers
who resorted to this device behaved exactly as if they had concluded an
9 Isocrates 4.43 (Panegyr;cus).
10 Cf. l.ewis (1977) 47, for the suggestion that the X('1f;a ~ttn Perikln and Archidamos
originated when Periklcs' famer Xanthippos 2nd Archidamos' gr2ndf:uher Lrorychldas served
as foint commanden of the Grttk net't in 479 8.C.
I Xenophon, Htllmica ".1.29ff.
n Xenophon. Anabasis] .2.23ff.
J] Herodotus 7.27-9 and 38-9.
alliance personally. Cambyses' envoys did not say that they would like to
conclude a poliriC21 or military alliance with the Ethiopians, but that their
master wished to become the Ethiopian king's phi/os and xenos. Similarly,
the Indian king had reportedly sent Cyrus (II, the Great) the following
message: III desire to be your xenos, and I am sending you money, and if you
need more, send for it". 14
Once the parties were brought within geographical proximity, the first
contacts had to be made. This could prove especially troublesome if the
groups to which they belonged were in a state of hostility. A proof of
renunciation of violence was first required, and only then could negotiations
concerning friendship start. The main elements of such a rapprochement are
illustrated by the preliminaries of the Agesilaos-Phamabalos conference in
394 B.C.:
Now there was a cemin Apollophanes of Cyzicus, who chanced to be an old uno!
(b XaAaLOiJ ~~ ciJv) of Phamabazos and at that time had become a xmas of
Agesilaos also. This man, accordingly, said to Agesilaos that he could bring Phama-
buos to a conferen~ with regard to establishing friendly relations (n£Ql ~(u~).
And when Agesilaos heard what he had to say, Apollophanes, after obtaining a truce
(spondas) and a pledge (dexiafl) brought Phamabazos with him to a place which had
been agreed upon, where Agesilaos and the thirty Spartiatae with him were ...
awaiting them... And first they [Agesilaos and Phamabazos) gave each other greeting
<iUJ.~ X~lV JlQ<)OtUlUV) then Phamabazos held out his right hand <t'ilv
6£;Lclv nQOtt£vav1~) and Agesilaos held out his to meet it. IS
What needs emphasis here is the extraordinary profusion of trust-
generating devices, each fulfilling different functions and each bearing dis-
tinct, and highly pregnant, designations. To the devices appearing in
Xenophon's story - the mediator, spanda;, dexia, and the greeting - we may
add the oath, pista (or p;ste;s), and euergesia. We may now consider these
devices in some detail.
Mediators, preferably intimates of both sides, provided the inirial reas-
surance that friendship was feasible, and took the first steps for entering
into negotiations. Apollophanes was the common xenos of Agesilaos and
Phamabalos. He communicated the parties' mutual intentions and secured
.4 Herodotus 3.21 and Xenophon. CyroP4~u, 6.2.1. Cf. Herodotus 3.39 for the xmu,
between Polykratn of Samos and Amasis of Egypt. For rM UK of xmu, as a political treary
(mainly) in rM Homme world, see finley (1917) 100. for analogous usn of the Roman
hospititmt, see Mommsen (1864) 339 n.26.1n a fashion vny similar to xmia, compattmit4s in
the Middle Ages served as a means of establishi"l political allianca. O\arlemape, for
instance, "found it impouible to conceptualise his mationship with the pope orner than as a
fHKtImt comp4tnmtatis" (Bennet (1979) S). In rM ancient Near East. "political marionships
bctwem allied rulen we~ conceived of as OM of kinship. either of ·fra~mity·, ah"tJmI,
albN""". or of father and son" (Munn-Rankin (1956) 76). Hannibal's 'covenant' with Philip V
(Polybius 7.9) probably «hoes this a~t practice (Bickerman (1952».
U Xenophon. Htllnrica 4.1.29·31.
2 Preliminaries 47
pledges of non-hostility. Agesilaos' xenia with Pharnabazos' son had prob-

ably been arranged by Phamabazos himself. Cyrus' meeting with Syennesis,
whose territory he invaded, had been arranged by the latter's wife, Epyaxa.
She won over Cyrus and received pledges from him. It had been rumoured
that Cyrus had had intimate relations with her. 16 After Cyrus' death, it was
Menon, the Thessalian, whom Klearchos sent to Ariaios in a futile attempt
to bring him over again to his side: Menon and Ariaios were xeno;. 17 The
negotiations between Pausanias and Xerxes were conducted through the
mediation of Gongylos the Eretrian. 18 Such sporadic contacts turned into
widespread practice in the Hellenistic age. Powerful 'friends' would use
their strategic position at court in order to establish connexions between
their overlords and their kin, friends, and guest-friends from the Greek
cities. 19
This act of 'bringing together', 'introducing' or 'recommending' someone
to a future friend or protector was denoted by a technical term, synistanai,
which applied in such cases with remarkable regularity. Xenophon himself
had been recruited to Cyrus' expedition through the intermediary of Cyrus'
xenos, the Boeotian Proxenos, and became Cyrus' phi/os after being 'intro-
duced' to him by Proxenos. 20 The young man from the kingdom of Bospor-
us (for whom (socrates composed his Trapez;ticus) had been introduced to
Pasion, the Athenian banker, by one Pythodoros, a Phoenician. 21 And
Eumenes is said to have sent all his brothers to Rome in 180 B.C., hoping
thereby, among other things, to recommend (systesatl them to his personal
phi/oi and xenoi and to the senate. 22
A spectacular act of euergesia, a 'good deed' or a 'favour', could at the
same time serve as a declaration of non-hostility and a first step towards the
" Xmophon, Anabasis 1.2.12 and 1.2.26.

•, Xl'11Ophon. Anabasis 2.1.5.
II Thucydides 1.128.6 and Diodoms Siculu5 11.44.2. According to Diadems Siculus 11.56,
Themisrokl"' safe refuge at (he Persian (oun had been KCUred by one lysitheides, who was at
the same rime 2 philos of Xerxes 2nd an iJioxmos of Theminokles. According to another lare
version (P1utuch, Thmristokles 26.1), Themistokles' xmos was Nikogenn of Aqae, "(he
wealthiest man in Aeolia, and well acquainted with the P~sian magnat" of the int~ior". For
the confusion con~ming the ~nonaliry of this xmos, see HofstdtC1' (1978) nos. 205 and 233.
I ' This is one of the services for which members of the philoi circle were honoured in the
ciries, d. Herman (1980/1).
20 Xenophon, Anabtnis 3.1.8 and 6.1.23, d. Diogmes laertius 2.49·50.
.ll lsocrates 17. 4 (T'aproticus), d. Demosmmes 52.4 (Agairut CaJlippus) for the 'intro-
duction' of one K~hisiacks, a resident of Skyros, to Pasion by the Athenians Aristonous and
Ardlebiaoo, the xmoi of Kephisiada.
22 Polybius 24.5.3, d. 33.18.2. See Gdzer (1969) 69 for the analogy with commmJa,~.
further examples: AC'SChinn 2.1S4 (On th~ Embassy) (Demosthenes to Arinophann of
OIynthus); boaate5t Letl" 7, by means of which lsocrates himself introducN a philos,
Autokrator, to Timotheos of Heraclea. an anc"tral xmos. I am told that the mo<km Greek
verb systeino is srill used in comparable circumstances.
establishment of relations of solidarity. 23 It could thus become a most
suitable vehicle conducive to the crystallisation of ephemeral contacts into
permanent bonds. Be it a well-timed service. the saving of one's life. or a
most valuable gift. euergesia could serve as an almost perfect substitute for
the other trust-generating devices. The essential features of the concept can
be inferred from the concrete situations to which it was applied. Its built-in
assumptions are a certain short-term sacrifice of self-interest on the part of
the benefactor; a state of helplessness on the part of the beneficiary; and the
establishment. in consequence of the interaction between these com-
plementary opposites. of a sacred bond of trust between the panners. A
sudden reversal of attitudes is presumed to take place thereby. the benefici-
ary repaying the benefaction with a display of gratitude (charis).24 For. in
this world. a benefaction - a favour accepted or imposed - is like a debt: it
must be repaid at all costs. Ancient accounts make it impossible to tell
whether this sense of duty was a reflection of actual sentiments. or an
enactment of cultural conventions.
But there can be no doubts regarding the moral duties that were thought
to be associated with euergesia. Here. unlike in Judaeo-Christian morality.
the good deed was not an end in itself. an act of uncalculated. altruistic
generosity for which perhaps heavenly. but never earthly. rewards were
expected. 25 On the contrary. euergesia was unabashedly recognised as a
secular strategy in the conduct of interpersonal relations. The benefactor
put the beneficiary in a state of indebtedness. from which state the benefici-
l \ Various forms of ru~g~sia are the regular prdudn to pam of friend.,hip. The Spanans
sent to Sardis to buy gold. Croesus, rather than selling, mack a frtt gift (dotm~) of it to them.
This ftln'8tsia led to a pact of xenia Mtween Cronus and the Spanans (Herodotus 1.69). See
Karavites (1980) for exampln of ru~g"iai in Htrodotus and Thucydidn.
l4 An interesting parallel can be found in mo<km kumstvo relationships, the Balkan
tquivakm of godpa~nthood. Tht ritualistic eriqu~ hnC' seems to be more stylistd and the
language more teehnial. The~ are specific name'S for the favoun in rdum for which kumstllO
relationships a~ ntablished: 'iumswo from misfortune', and 'iumstvo from gratitude'. If a
suppliant n granted an important favour, he rec1proatn b)' giving iumstvo in return, that is.
by offering a child for baptism to the penon who has hd~ him. A favour might consist of
saving a drowning man, or of catching a man in the act of wft and saying nothing. A soldiC'r
may give iumslvo to a man who saved his life in war, and a penon who has been seriously ill
may offer one to the doctor who cured him. There is a popular belief that offers of this kind
cannot be rdu~ (Hammd (1968) 81·2).
~. The contrast betwem our encounter storin and a similar inciMnI from the Old Tnta-
mft1t is pnhaps wonh making. David, lhe pretenckr to eM throne, did not avail hirmelf of the
opportunity to kill Saul, his deadl)' tnemy. who was persecuting him. When David's mnciful
act became known to the king. he burst into rean and, overflowing with gratituck. called David
am)' son', He said that he himself was put by God at David's mercy but David did not kill him;
that was David's 'good c:kcd' (tmld): "Not often does a man find his enemy and In him go
safely on his way". However. the reward of David - to become King of Israel- did noc come
from Saul in rKiprocation, but from God a., a gift (1 Samuel 24, d. 26). In translating Biblical
passages I follow rhe New English Bible.
2 Preliminaries 49
ary could only redeem himself by a display of submission and loyalty - and
that was all there was to it. "The recipients of gratuitous benefits are always
glad to do good services (hyperetous,) to their benefactor in return for the
kindness they have received ... ", said Xenophon of the beneficiaries of his
hero, Agesilaos. 26 One of the moral precepts attributed to Kleoboulos of
Lindus, one of the Seven Sages, was that "one ought to benefit (euergetein) a
friend, so that he may be more a friend, and to make an enemy into a
friend".27 Plato considered euergesia equivalent to kinship ties in securing
loyal supporters. He remarked that whereas Dionysius of Sicily failed to
manipulate his kin and friends from abroad, Darius succeeded therein, even
though his colleagues (koinonoi) were neither his brothers nor were they
reared by himself; they were merely his associates in crushing the Mede and
the Eunuch - his ritualised friends, in our terms. 28 Never, in the concrete
examples, is the hope of divine retribution attached to a good deed. Darius
installed Syloson as client-ruler in Samos since he owed this to him for an
almost forgotten euergesia. And Antigonos, when an appropriate opportun-
ity offered itself, repaid Neon's euergesia by nominating Neon's SOn as an
official in his own service.
After Diomedes and Glaukos had rediscovered their affiliation through
xenia, they leapt down from their chariots and "clasped each other's hands
and pledged their faith" (XEt{Xl~ t'uUliA<oV Aaf3€tTJv xai, 1tl<TtWOavto).29
Whenever potential enemies renounced fighting, the exchange of 'things"
called dexia and pista (sometimes, pisteis and once enechyra), recurred with
compulsive regularity. These exchanges were mostly accompanied by oaths
(horkiai) and libations (sponda.). The function of these devices seemed so
self-evident that ancient authors felt no need to explain it. Three generalisa-
tions emerge from the evidence which follows in detail: first, that dexia and
p;sta were in general indistinguishable in meaning but slightly differed in
their connotations; secondly, that they both served as an initial proof of
non-aggression; and, finally, that the peace process might have reached its
conclusion with their exchange, or it might have further evolved into
ritualised friendship.
A passage from Xenophon's Anabasis offers a good illustration of the
system of beliefs and practices within which these pledges operated. The
Macronians objected at first to the passage of the Greek army through their
26 Xmophon, Ag~s;laus 4.4.

27 Diogen~ Lurrius 1.91.
1I Plato, U"" 7..1323; d. Xenophon, Cyropa~dia 8.7.13, where ~~rgt!s;a is SImilarly
regarded as the b6t strategy for a mon2rch fO win faithful frimds.
29 Iliad 6.233.
so Encounter and Initiat;on
territory. The Greeks, however, reassured them that they would not do any
harm. Then,
The Macronians asked whether they would give pistis to this effect. The Greeks
replied that they were ready both to give and receive pista. Thereupon, the Macro-
nians gave the Greeks a barbarian lance and the Greeks gave them a Greek lance, for
the Macroniaru said that these were p;sta; and both sides called the gods to
wimess. JO
The whole process was given a technical name in Greek, pist;n /abe;n ka;
douna;, pista poieisthai, or p;steuestha;: to pledge one's faith. As a result of
this, the partners became each other's p;stoi, and their mutual relationship
was henceforth governed by the principle of pistis, fidelity. Ritualised
friends were by definition pisto; to each other. "Pylades", says Orestes in
Euripides' Electra, "I consider you the first. man in loyalty and loving
guest-friendship unto me" (xWtov VOJ1£too xal "O-ov
!;tvov t'tJ1O(), and
"being in the pistis of the king" was one of the standard qualities ascribed to
the philo; of the Hellenistic rulers. J • Abstract concepts, personal attributes
and concrete objects are thus conceived of as somehow sharing of a com-
mon substance. For, in the world of ritualised friendship, as in the world of
Odysseus, "every quality or state had to be translated into some specific
symbol, honour into trophy, friendship into treasure, marriage into gifts of
cattle".32 And we may add: trust into ritual objects.
Pista were usually objects of little intrinsic value, but of immense symbo-
lic significance. Being imbued with the personality of the giver, they were
thought to exercise a binding force upon the personality of the recipient.
The gods, approached through oaths and libations, were called to bear
witness that this transfusion of fidelity had indeed been effected.
The term dexia in the sense of 'pledge' was abstracted from the ritual
handshake. Indeed, the handshake was deemed so characteristic an element
of the process that ancient anists used it as a visual symbol for the whole of
the alliance (Figs. 4 and 5, d. Fig. 3a,b, p.37, and Fig. 12, p.134).J3 The
handshake has two universally accepted functions: it can serve either as a
greeting sign, or as a symbol of solidarity - a pledge against resort to arms.
Cyrus the Great at his death is said to have entrusted the provinces to his
children. With a handshake (tWv 6E;uirv !~<ov), he made his own
)0 Xmophon, Anabasis 4.8.7. It is true that the partnen are here 8J'Oupi. not individuals,
but shall let that the practice was in both cases almost identical.
Wt
.H EUripidn. Ekct1a 82, d. TaiU.rdat (1982) for further examples. for rM Hdlmistic
terminology, d. Herman (198011).
J2 finl~ (l9n) 123.
H Se-e Ntumann (1965) for funher examples of vase paintings. Some kind of mreot)'ped
gesture involving bodily contact also seems to have been practised in OrMr ritualised rela-
tionships. Christian godpamnhood~ for inslllntt, involves a ritual embrace.
2 Preliminaries S1
faithful friend, the Skythian Amorges, their philos, wishing them all good if
they abided by the pact but imprecating curses upon tbe first to strike his
friend:'" There can be no doubt that it was thi5 second funaion of non-
aggression that the anaent anists chose to illustrate. For two claspU'lg hands
mutually nullify each other's aggressive potennal. The arms must be trans-
ferred to the inept lefts~ and this serves as a further proof of pacific
intentions. It was from this function of the handshake that the symbolism
was derived. For the act of greeting itself, however, a set of altogether
different gestures was reserved. Phamabazos and Agcsilaos first gave each
other greeting (Xa£Q£LV xQO(J£inav) and only then did Phamabaws hold out
his nght hand (rliv bt;..civ xQO'tdvQVto~) and Agcsilaos hold out his to
meet it (UvtUtQOVtE'VE). Similarly, Dlomedes and Glaukos had long been
engaged in conversation when they ~prang from their horses and "clasped
Fig. 4 A non-Hellenic iOltitudon?

Shalmancser III, King of Assyria shaking hands with me King of Babylon, probably
Marduk-zakir-shWl'i. ShalmaneKr helped the Babylonian king suppress a rnolt and
confirmed his position on the throne of Babylon. Ninth ~entUry B.C., from the
throne-base of Shajmaneser III at Nimrud (d. M.E.L Mallowan, Nimrud and its
Rnna;lfs (London, 1966), yol.!. 444ff.). The similarity between thiS and the Greek
gestures In Figs. J, 5 and 12 is striking. Note the objects held in the right hands ot Ih~
followers, presumablt intended to be exchanged a, Jifb.
34 FG,H 688 (Kresias) 1=9.
ro
0i:
Q)
.....
ro
~
"'0
.....
Q)
.c
Fig. 5 The rituJ laandshakc 0'
.a:
Attic black figure amphon, simt century B.C., repraentil18 Herakles and the ISood' >-
centaur, PholOle Pholos alone of all the centaun made Heraldee his xmOl and 0.
protected the hero against his fellow-eentaun (ApoIIodo1'Ol2.5.4). Note the contnst o
U
between the (reversed) I)'mmetry of the claspins hands and the dissimilarity of body
contoun: xmoi arc mansen who through a ritual act become IloVlnl friends'. Of all
the elements of mls rimal (a solemn declaradon, an exchange of gifts), the hand-
shake (daiwis) lends itlelf most easily 10 vitu rcpracntauon. Basically, JeJCiosu i•
• symbolic declaration of peaceful intenriona. For two clasping rights nullify each
other's agrellive potential; the arms, of nec:esairy, have to be tr sferred to the inept
left hands. Here, as a tunher sign of peaceful intentions, Heraklcs' lion skin is taken
off and carried on the end of Ihe dub. Hennes, the god of medialon, havil1l brou(!hl
about the encounter, siu down to take a rest. For further variations on the theme, see
Brommer (1960) 135-8; Neumann (1965)
each other's hands". But nowhere does the logic underlymg this symbolic
interaction receive a more dramatic illustration than in a famous scene from
Roman history. When Popdius Laenas met King Antiochos IV in Egypt,lS
IS Polyblul 29.27.
2 Prel,minar,es 53
Antiochos greeted the Roman commander orally from_ a distance and held
out his daia.. Popilius. however, did not return his own dexia, the conven-
tional ign of friendship (nj~ cl>ll(a~ mMh)l1a) as. Polybius calls i~ but
I
handed him instead a copy of the Senatus Consultum which bid the king
withd.raw from Egypt. Cle•.rly, the Roman commander was apprehensive
that a premature e:kchange of dexiai would have- committed them to a
friendly interaction. Only when, finally, Antiochos opted for ignominious
ro
.\:
OJ
4-J
ro
~
"C
OJ
4-J
.c
.-en
L-
>-
a.
o
U
Fig. 6 Communal symbolon

Right hand of bronze with two broken fingers, bearing the insaiption "S""OO/o"
with the Velaunians" (IG XIV.2432, probably second cenlUry B.C., found in tht'
south of Fran«;e). The ling wmmunity is unknown (M. Guarducci, Epigra(",
G,~ca yol.2 (Rome, 1969) 583, sugests Massalia', nor do we have any due s to I
who the Vclaunians were. It seems, however, beyond reasonable doubt that the
allaance between the contracting communines was conceived in terms of ritualised
friendship. 5""00/0,, 'Should here be undemood u something like rreat)' of
friendship': the community has borrowed the device from the context of the private
agreements of ritualised friendship.
S4 Encounter and Initiation
friendship instead of honourable, but suicidal, hostility, did Popilius and his
suite grasp him by his right hand and gr~t him warmly.
For this is a world in which it is a virtue to respttt pledges, and a shame
to violate the obligations they impose. "I received and gave the dexia" said
Cyrus of Orontas, the friend who betrayed him, meaning thereby that
Orontas pretended faithfulness but in reality allied himself to Cyrus' enemy.
And Aeschines accused Demosthenes of murdering his own xenos, a man to
whom he had given his right hand. 36
As the pista, so the dexia could be given concrete representations. Of
Kotys, the ruler of the Paphlagonians, it is said that he disobeyed the
command of the Great King, even though it was sent along with dexia.
Following Cyrus' death at Cunaxa, we are told, certain Persians kept
encouraging the Greeks, "bringing dexiai to some of them from the King
that the King would bear them no ill-will because of their campaign with
Cyrus against him or because of anythi~g else in the past":\7 Though no
hand-tokens have survived from the Achaemenid empire, it is possible to
infer that some objects (not necessarily in the shape of hands) were used in
cases where no face-to-face confrontation was possible (Fig. 6). As in the
case of the p;sta, all that was needed was the mutual agreement of the
parties to invest some object with this function. 38
Unlike these objects, however, the whole institution of pledge-giving does
not seem to have been confined to any specific society. The Arabians, says
Herodotus, perform the pistis by means of a blood ritual, the Nasamones
and the Skythians by drinking blood from the hand of the other party (Fig.
7):\9 The ritual technique may vary from place to place, but the underlying
system of beliefs is one. Herodotus knew perfectly well what the~ exotic
people were doing; the question which he tried to answer was how.
.l.3 SUPPLICATION
Another preliminary ritual n~ds special attention since it presupposes a

situation essentially different from the ones described above, yet can lead to
.ko Xmophon Anabasis 1.6.7; A~hines 3.224 (ARainsl Ctniphon); Sophocln, Phi/okl~tes
813 (Philoktetes saying Neoptolemos f~)J.f Xf'~ x{onv), d. 1398.
J1 Xenophon, Agesi/aus 3.4, and Anabasis 2.4.1; for the Roman world, set, for example,
Tacitus, Hislori~s 1.54 (the lingoncs, according to their ancient custom, sent to dw Roman
l~ons as prnmts d~XI'iU. hasp;ti; insigne), and Nepos, Dalamn 10.1.
J' Sherwin-White (1978) suggests that hand-tokens similar to those which survive from the
Sassanid pmod might have bttn in use in the Achaemenid period. 'Bronze hands' Sftm to have
been known in Rome, d. Piganiol (1959).
.\ . Herodotus 3.8, 4.172.4 and 4.70 with 1.74.6, where the word! ho,kion and horkia are
uKd apparently as synonyms of pisl;s. For a complThensive study of blood rituals, see
Tegnaeu5 (1952).
S5
Fig. 7 Consubstantiality
Golden ornament, possibly a Skythian version of • symbolon, found in a barro~
frorn Kul-Oba near Panticapaeum, fourth or third century B.C. The barrow con-
tained the body of a semi-heUeniscd Skythian ruler (d. M.1. Rosto~ff, Iranians
and Gre~ks in South Russia (Oxford, 1922) 110ff.). The object offen a splendid
visual representation of the Skythian blood ritual (horkion) described by Herodotus
(4.70): .....they rake blood from the panies to the agreement by making a little hole
or cut in the body with an awl or a knife, and pour i( mixed with wine into a great
canhenware bowl, wherein they then dip a scimitar and arrows and an axe and a
javelin; and when this is done the makers of the ho,kion tht:mSCIves, and the most
honourable of their followers, dnnk of the blood after solemn imprecations". As
observed by V. Schiltz (in Compt~s Rmdu$ d~ "Ac"dbnie d~s Inscriptions eI
B~Uu-Umu (1975) 450), the artist has found a most original way to convey the
Idea of fraternity acquired through consubstantialiry: the profiles of the two men
drinking from the same rhyron merge into eam other in such a way as to crelte the
illusion of a single large face teen from the front. Note also Herodotus' assumption
that such rituals take place between 'big men' surrounded by followm..
a similar conclusion. The. best illustration of this contrast comes from a

scene in the Odyssey. According to one of the fictitious stories that the
Cretan stranger (in fact, Odys5Cus in disguise) told the swmeherd Eumaeus.
he once landed with his trusted friends (hetairaI) in Egypt.. His friends set
out at oner wasting the fields, carrying off women and children, and slaymg
the men. However, the news had reached the city ncarby, and in no time
they found themselves surrounded by a superior force of footmen and
chariots. No one had the spirit to stand up to the enemy, and thus some of
Odysseus' friends were killed, others carried off as captives. In this predica-
ment, a sudden inspiration from Zeus made Odysseus adopt a different
strategy. Rather than fighting back, he doffed his helmet, dropped his
shield, threw away his spear, and went to the chariot of the Egyptian king,
clasping and kissing his knees. What happened then might again appear to
modern eyes as something truly unexpected. For the Egyptian king, who~
territory had been invaded, did nothing less than protect the leader of the
invaders from his own men. As Odysseus would later recount,
...he rescued me and took pity
and seated me in his chariot and took me, weeping, homeward
with him; and indeed many swept in on me with ash spears
straining to kill me, for they were all too angered, but the king
held them off from me, and honored the anger of Zeus Protector
of Strangers (Xenias), who beyond others is outraged at evil dealings.
The sequel presents an even more surprising reversal of the initial situation:
following his salvation, Odysseus could claim that
There for seven years I stayed and gathered together
much substance from the men of Egypt, for all gave to me. 40
No technical terms appear in the text to denote the social institutions
conditioning this behaviour. Yet there can be little doubt that the ritual
which saved Odysseus' life was what the Greeks called hiketeia, and that the
bond consequently forged between Odysseus and the Egyptian king was
xenia.
Supplication occurred in the Greek world in an amazing variety of
situations - to be sure, without necessarily leading to xenia. Gods suppli-
cated other gods; humans - gods; women - men; slaves - rulers. The
suppliant, unlike the xenos, could belong to the same political unit as the
person supplicated, or to a different political unit; he could be a member of
the same social class, or of a different social class. But, whatever the
variations, one common feature always underlay the suppliant's behaviour:
that he went through "a series of gestures and procedures that together
constitute total self-abasement" .41 It is precisely this that sets off hiketeia
from the preliminary rituals described in the previous section. ") come in
front of you as a hi/utes and give myself over to you as your slave (douJos)
40 Odyssey 14.279ff., translated by R. Lanimore, The Odyssey of Homer (1967).

4. Gould (1973) 94, who furnishes the ~It available collection of evidmce on hihu;a; 1ft
also Pedrick (1982). Here I single our only exampl~ which are explicitly linked with XnfUJ.
3 Supplication 57
and ally (symmachos)t" said Gobryas to Cyrus; and this should be con-
trasted with the Indian king's message to Cyrus quoted above: ... desire to
be your xenos... ".'u In the first case t one person approaches another from a
position of temporary inferiority; in the second t from a real or pretended
equality of power relations. Other differences follow therefrom. The ex-
change of dexia and pista is a ritualistic abdication of hostilities; hiketeia is
a ritualistic admission of defeat. 4 .1 To be sure t euergesia too involves a loser
- or t at least t a person badly in need of help - but the initiative for euergesia
comes from the stronger partner; in the case of hiketeia t the initiative comes
from the defeated partner; hiketeia could be described as a solicited euerge-
sta.
A later text referring to the mythical past is worth quoting in full in this
context t since it seems to point to ritual warfare as the possible origin of this
stereotypical behaviour. "What is a dOT)'xenos?" rspear-guest-friend\ liter-
ally translated}t asks Plutarch t and by way of answer he recounts a story
which in outline at least deserves serious consideration:
In days of old the Mcgarid used to be settled in village communities with the citi7.cns
divided into five groups... Although the Corinthians brought about a civil war
among them ... none the less, because of their fair-mindedness, they conducted their
warS in a civilised and a kinsmanly way (syngenikos). for no on~ did any harm at all
to the men working in the fields, and when anyone was captured, he but needed to
pay a certain specified ransom; this his captor received after they had set him free,
and did not collect it earlier; but he who took a prisoner conducted the man to his
house and, afrer sharing with him salt and food, sent him home. He, accordingly,
who brought his ransom, was highly regarded and continued thenceforward to be a
friend (phi/os) of his captor; and, as a consequence of his capture by the spear, he
was now called doryxenos. But anyone who failed to pay the ransom was held in
disrepute as dishonest and faithless (adikos, apistos), not only among his enemies,
but also among his fellow-citizens .....
Plutarch is admittedly a poor source for archaic Greek history. Neverthe-
less, the faCt remains first, that the Greeks in Classical times did usc the
word doryxenos;4S secondly, that whatever else we know of hiketeia, and,
for that matter t xen;a t fits into this pattern; and t finally, that neither the
recruitment of a defeated enemy into the social group of the victor, nor the
prescription of rigid rules of mutual behaviour between vietor and van-
quished, is unfamiliar to scholars of primitive societies. 46 Plutarch's story is
probably idealised t but it is not devoid of heuristic value.
·42 X~nophon, Cyropa~dia 4.6.2 and 6.2.1.
4' Odys~us' action in stripping off his arms in the passag~ cited above should ~ compared
With the armed figures shaking hands in Figs. 3a,h, 4, 5, 12 and with the encounter bttw«n
Diomedes and Glaukos.
.... Plutarch, MoraJia 29SB·C.
.. < Though its specific sense was probably forgotten, d. ahove, p.II n.3 .
.." Cf. Pitt-Ri"'ers (1968) 409 and Sahlins (1974) 207. Many scattered examples may also
he f(lund in P. Bohannan (ed.), LAu' tJnd Warfare (Austin and London, 1980).
We shall return later to this conceptual association between stranger,

host, suppliant and guest-friend (Section 5.2). Suffice it to conclude at the
present moment that the rituals described in the previous sections all served
to diminish an initial hostility. Strangers randomly brought together, or
'introduced' to each other by a common friend, made alternative uses of
devices such as dexia, pista, euergesia and hiketeia in order to establish
peaceful relations.
3.4 THE INITIATION RITUAL

None of the elements so far described was sufficient in itself to convert
strangers into friends. What they created was a mere temporary bond, a
relationship established for a special purpose. They could thus serve part-
ners in a variety of other relationships, not necessarily within the context of
xenia, To conclude a pact of friendship, further devices were needed. And,
the closer we come to the crucial moment at which the transition from one
state to another was effected, the more specific to ritualised friendship these
devices become. Nowhere do we find a fully rounded account of such a
ritual. In the sequence of the Agesilaos-Pharnabazos conference, however,
we probably have something which comes very close to it:
And Phamabalos mounted his horse and rode away, but his son by Parapita, who
was still in the bloom of youth, remained behind, ran up to Agesilaos and said to
him: "Agesilaos, I make you my xmos". "And I accept you". "Remember, then", he
said. And immediately he gave his javelin (palton) - it was a beautiful one - to
Agesilaos. And he, accepting it, took off and gave to the boy in return a splendid set
of trappings (phalara) which Idaios, his secretary, had round his horse's neck. Then
the boy leapt upon his horse and followed after his father,4?
The picture emerging from this might be completed by three Homeric
scenes: the initiation ritual between Odysseus and Iphitos, and two renewal
rituals - those between Diomedes and Glaukos, and Telemachos and
Mentes. 48 I shall use these accounts to reconstruct the initiation ceremony
and to decode the logic of its symbolism, drawing on whatever else we
might hear of one or another element of the ritual from other sources.
47 Xenophon, Htllttt;ca 4.1.39. If Agnilaos and PhamabalOs ~re xmo; (which we cannot
know for ttnain), this is also a renewal ritual.
4. Ody~ 21.11-42; Iliad 6.119-236; Odyssry 1.11.5ff. The laner simply mimic the main
features of the initiation process. For a renewal ritual.as s«n through the eyes of a Roman, see
Cunius 6.5.1-S (involving Anabazos,.a hOSpt5 of Philip 11, .and Alex.andn). A parallel from the
Old Testament is 1 Samutl 18.1: "So Jonath.an and D.avid made .a solemn compact (btT;Ih)
btcau~ tach lovtd tht other .as dearly as himself. And Jonathan stripped off me cloak he was
w~ring and his tunic, and gne them to David, t~her with his sword, his bow, ~nd his belt".
See also 2 Samuel 1.26, where David in his lament over Saul and Jonathan calls Jonathan 'my
brother'. h is interesting to note that the two men bt-Iongtd to different tribts; D.avid came
from Judah, Jonathan from Benjamin.
-The ritual was made up of a complex combination of symbolic elements

which were enacted in a sequence. The whole range of possible elements
included a declaration, an exchange of objects, feasting, and again, the
taking of oaths. Both the declaration and the exchange were indispensable
for the validity of the alliance. The feasting and the oaths were optional. The
whole process could be sealed by an exchange of yet another type of object,
the symbo/on.
Agesilaos' and the Persian boy's solemn pronouncements "I make you my
xenos" and "I accept you" echoed the standard initiatory declaration em-
ployed by would-be ritualised friends. The formulary nature of this declara-
'\ tion is shown by the fact that it was not essentially affected either by the
diversity of the situations in which the pronouncements were made or by the
gaps in time between the periods from which the evidence comes. Diomedes,
having rediscovered the link between his own and Glaukos' lineages, reaf-
firmed its unbroken validity: "I am your xei,ws phi/os in the heart of Argos,
you are mine in Lykia...... 49 In a similar fashion, Mentes, once he and
Telemachos had recognised the affinity inherited from their fathers, had
reiterated the initiatory formula: "xeino; patroioi do we declare ourselves to
be of old..... :~o In the underworld, Agamemnon reminded Amphimedon: "I
declare that I am your xeinos":~1 In a very similar fashion Xerxes, having
been offered lavish hospitality and most valuable gifts by Pythios the Ly-
dian, declared that " ... in return for this I give you these privileges (gerea): I
make you my xenos.".52 And, as we have seen, the same set of words could
be applied in non-face-to·face situations, when a ruler wished to contract an
alliance through the intermediary of messengers. 53
Such declarations of intention seem to be a central element in most
ritualised personal relationships. In the tribal societies studied by Tegnacus,
the partners proclaim themselvcs in the course of the blood ceremony each
other's 'brothers', 'foster-brothers', 'cousins'. The surviving treaties of
'fraternity' 'paternity' and 'love and friendship' between the pctty rulers of
the ancicnt Near East in the second half of the second millennium B.C.
incorporate what are probably written versions of such dedarations:S4
.'1 Iliad 6.224-5 .

~l) ;ElVOl
• b·olltlA.Olv 1tatQWlO\ rllx0j.l£O'dval l; QQxtl<;. Odysse,' 1.187-8.
H Odyssey 24.114.
u Hrrodotus 7.29.2.
H Abovr. pp. 4S -6.
H Tegnarus (1952); Munn-Rankin (1956) 76; Wrinfrld (1973). It may be of interrst to
note that an almost id~ntical formula may be found in the mrdirvaJ cerrmony of homaRe.
"Sire, jr devien vostre hom", said rhe would-be vusal, 3n~ thr lord would reply: "Je '..os recoif
rt pran a homr" (Ganshof (t 964) 73, quoting from thr Etabli.ssmJerru de Saint Louis 1I.19}.
Encounter and Initiation 60
No less important an element in forging the alliance was the exchange of
a highly specialised category of gifts, designated in our sources as xenia (as
distinct from xenia, the term for the relationship itself) or dora. It was as
important to give such gifts as to receive, and refusal to reciprocate was
tantamount to a declaration of hostility. Mutual acceptance of the gifts, on
the other hand, was a clear mark of the beginning of friendship. lbe epic
poet says so explicitly: in return for the formidable bow, Odysseus gave
Iphitos a sword and a spear, to be "the first token of loving guest-
friendshi p u.55 To Herodotus, the conclusion of an alliance and the ex-
change of gifts appeared as two inseparable acts: Polykrate5, having seized
the government in Samos, "concluded a pact of xenia with Amasis king of
Egypt, sending and receiving from him gifts (dora)u.56
The exchange of these gifts differs significantly from other types of
gift-exchange both within the context of ritualised friendship and outside it.
First, rather than being delayed, the counter-gift followed promptly upon
the gift. Having declared each other as xenoi, the Persian boy gave Agesilaos
his beautiful javelin; the Spartan king accepted it, and gave in return
(aneowx£v) a splendid set of trappings. Dlomedes and Glaukos exchanged
armour in token of renewal of their inherited xenia, whereas their grand-
fathers, Oineus and Bellerophon, exchanged a belt and a golden drinking
cup. 57 The poet does not specify what the gifts were that Telemachos meant
to bestow on Mentes, but he reassures us that they were "very beautiful, ...
such a gift as dear xeinoi give to xe;noin • 58
Secondly, gift and counter-gift were expected to be commensurate, not
surpassing each other in value. Indeed, the idea of counter-giving was
u ~ ~t'vooimlc; X{M>OXTl6t~ Odyssey 21.35. The frequent rendering of xmia as 'gifts
of hospitality' seems in view of this unjustified. Suidas s.y. ~tv"ov glosses me term simply as
"gift given by xnroi·'. and the con~te example'S examined below leave no doubt that xmia
symbolised the entry into (or. as we shall see. the perpetuation of) ritualised fritndship. not the
durin of hospitality. For the word Ixnrwial with a possibly similar IftlSt in the Linear B tablrts,
lee Killen (1985).
J6 Htrodotus .1.39.2 (sec Fig. 4, p.St. for the gifts held in the hands of followers in a scene
from the ancimt Neu ust). An exception to this regularity is Iliad 7.287U: ~fter fighting a
duel. Aias and Hcktor exchange a sword and a war belt - not in or<kr ro ntablish xenia but in
toktn of temporary cessation of fighting: "Itt us now give over this fighting and! hostility for
today; we shall fight again. untiV the divinity chooses between us and gives victoryl to one or
the ormr" (translated by R. Lanimore). Indeed. thereafter they joined more than once in fince
combat (for example. lI;ad 8.330ff., 13.824ff.) but mere ill no suggestion that they were bound
by xm;a or were under any oblig~tion to spare each other. Only Sophocles would later recast
the relationship as xenia and exploit the incongruity created thereby ("the gift of Hektor. me
most hated of xttfoi to me and my greatest enemy" - Aias is made to say (A;ax 817ff.). to
aggravate the traRie situation). See funher P.E. Easterling, 'lbe tragic Horner', Bulletin of the
I"stitutt of Classical Studits 31 (1984) 1-8.
$7 Xmophon, Helltn;ca 4. J.39; Iliad 6.2 t 9·20.
$I Odyss~ 1.311-13.
expressed by words implying reciprocity: what one received and what one
gave were expected to be equal. A system of thought that assumes such
equation faces a difficulty when confronted with departure from the norm,
and the poet, faced with such an instance, was bound to resort to divine
intervention: "but Zeus ... stole away the wits of Glaukos who exchanged
with Diomedes the son of Tydeus armour of gold for bronze, for nine oxen's
worth the worth of a hundred".s'J Clearly, the value of the gifts should have
been equal.
lbirdly, assuming that gifts have at once a symbolic and a use-value, the
initiatory gifts are located somewhere between gifts that are valueless but
have a high symbolic significance (as, for example, the symbola discussed
below), and those possessed of immense use-value but which are of no
special symbolic significance (as, for example, the estates given as gifts
within the context of guest-friendship, Section 4.6). The initiatory gift
served both as an object and a symbol. Odysseus could have taken to Troy
the bow he received from Iphitos, but he preferred to leave it at home "as a
remembrance of his beloved guest-friend". 60 He would carry it with him
only on his own domains. It is this bow which, by a sudden shift of emphasis
from symbolic to use-value, supplied the plot of the Odyssey with a link
without which the final denouement would have been impossible. Megcs'
thick corselet, on the other hand, was explicitly designed to be worn in war.
The hero's father, Phyleus, received it from a xenos, Euphetes of Ephyra.
Passed on from father to son, the corselet now saved Meges' Iife. 61
Goods of commensurate worth and utility were thus simultaneously
exchanged, and it is probably this feature of xenia which has prompted
some historians to regard these gifts as forerunners of foreign trade. 6l In
strictly economic terms, no douht, the circulation of goods produced in
different societies was effeCted. This, however, was carried out in ways
different from the sort of exchanges implied in 'barter' and 'trade'. For the
exchange was not an end in itself, but a means to another end. Nothing like
a supply and demand situation, nor profit-making, was envisaged. Nor were
the objects exchanged needed by the partners. They fulfilled a psychological
need, the need to translate every state or quality into a symbol. 6 .l In genuine
trade transactions the articles are devoid of such symbolism, and their
exchange marks the end of the relationship. The gift of initiation, on the
other hand, was meant to symbolise the establishment of obligations which,
ideally, would last for ever.
~9 Iliad 6.234-6.
#,{J Odyssey 21.40.
'" Iliad 15.525ft. In rhi~ ca~. as in rhe' compact bdwee'n D~vid and Jonathan, the' counter-
gift is unknown.
62 Bolk~tcin (1939) 219, with Iirer~lure. For gifts exch~ngC'd within the framework of
xt'n;a ahe'r the nrablishment of the relationship, which form far bt"rter candidares for foreign
rrade, ~ Chaprer 4.
", Cf. Finley (1977) t 23.
62 F.ncOIwter a"d I",ti,lt,cm
-- -----,
X{VPET
AI"eN
'.'
, .
.,
'.. i
'.'
......~
. ,'.' ,.
.....
a h ( d
Fig. 8a-d Terracoua symbol.. from Athens
hrms a, h, and c were found in 1950 in tht' Agora in a rubbish pit, along with broken
ponery from the late fifth century B.C. hem d (= 1(; IZ 916) came w light in IIP8 in
the excavations at the Dip)·lon.
The production of these symho[a Involvrd thrre stages. On cia)' plaques of the size
of a domino thr leurring was first painted; the plaques were thcn cur irregularly in
two; finally, they were firrd. The rrsuhing objects were such that any given half
would join only its onginal mate. Such objects were so familiar in Athrnian life that
a srrirs of complt:x philosophical ideas could be rxprrssed by refrrrnce to them. for
example, according to the humorous theory of creation auributrd to Arisrophanes, >-
D-
"Each of us is a mere symbo[mr of a man, the result of a bisection, like the flat fish, O
~)
two out of one, and each of us is prrpetually in search of his corrrsponding
symbo/on" (Plato, Symposium 1') 1d). The underlying idea is that of complemcntary
opposites: rlthrr wuhour the other is incomplete. Aristorlr quotrs Empedoclrs in
saying that "in thr male and in the female therr is as it might be a symbo[mr" (De
Generatione Animalitmr 722h 10).
It does not serm possible ro tell whether the present objects were designrd for
xtnoi or whether, as suggrsted by H.A. Thompson (in Htsperia 20 (1951) 51-2 with
pI.25c), thry reprrsent some adaptation of this original function to a similar (so far
unknown) need of thr Athrnian government. HAI.IMOS and XS YPETAION arr
deme namrs, but l.EO (b and c, reverse) nerd not be an abbreviation of thr tribal
name LEO(NTIS): as POL in a and c (obverse), l.EO could be an abbreviation of a
personal namr. The rrading of thr Iruer traa's in a (reverse) as ERE(CHTHEIS)
srrms insrcurr. In sum, until further evidence comes to light, thr comhination of
personal and deme names cannot br rxcludrd as a possiblr oyrrall rxplanation. As it
was not uncommon for Athrnians negotiating transactions with strangers to Identify
thrmsrlvrs b)' drmr names (in the bottomry loan contract in Drmosthenes .H.1O
(Against Lacritrts), for rxample, the jOint Irndrrs identif)· themsdves as Androkles
Sphettios and Nausikratrs Kaf)'stios), the objects could well be s)'mbola used for
idrntification brtwern frirnds and d("prndanrs of xtno;. The small hole at the middle
of d suggests that the oblect could be attached to a string and worn.
It will by now have become apparent that gifts of initiation were unusual.
even extraordinary, objects, each with a provenance and a history of its own
which the heroes were fond of recounting. 64 They served, in a sense, as an
expression of their position in society, for the identity of a man of the upper
classes was determined by his relation to kin and quasi-kin abroad. In the
course of time, these attributes must have become more differentiated and
specialised, giving rise to yet another type of gift, the symbo/on, the only
function of which was to prove one's identity. This evolution seems to have
taken place somewhere in the archaic age: the concept was unknown to
Homer, and in his world the special attributes of the gifts fulfilled the
function of identification. 65 In the classical language, on the other hand, the
meaning of symbo/on was so familiar that several metaphorical usages
could be derived from it. 66 Symbola were by now not unusual precious
objects, but articles of negligible intrinsic value, bearing an explicit mark of
their usc. Pieces of bone, coins, tablets or similar objects would he broken
irregularly or cut into halves, each party taking one. (Sec Fig. 8a-d for the
illustration of the principle.) When, after the lapse of some time, the xenoi
(or their descendants or dependants) met, the test of whether they were
directly or indirectly related through xenia would be whether the halves
joined exactly. Where they did, the relationship could formally be renewed
and the obligations inherent in it fulfilled. 67 A still more elaborate version of
the same idea is represented by a splendid ivory plaque from Hellenistic
Sicily (Fig. 9). One side is carved in the shape of two clasping hands, the
most conspicuolls visual symbol of the relationship. The other bears an
inscription which tells us what the use of the ohject was: to remind its owner
of the pact of xenia which Imylch, son of lmilcho, a Phoenician, and Lyson,
son of Diognctos, a Greek, concluded.
It would be a mistake, however, to assume that xeno; resorted to such
devices as a matter of regular practice. The paucity of similar objects in the
6-4 For ~x3mple. Odyssey 4.613-18, repe~ted in 15.111·19 (th~ mixing bowl, the' work of
Hephaistos, which Mentlaos reaived from Phaidimos, king of the Sidonians). Mege!l' corstlet
and Odys5a1s' bow mmtioned above had similar historie!l.
6' The token given by Proitos to Bcllerophon (Iliad 6.168) is not, as sugge~ted by Mommscn
(1864) 338, a symbolon, sinc~ Bdlerophon did not produce ir ~s a me'ans of identifIcation on
his arrival in Lykia. Only after he had bttn entertained for nine d~y~ by the l.ord of l.)'kia W3!1
h~ asked whether he bore a token from Proitos. The object was thus designed to carry ~
message unintelligible to its carrier - a message to kill him.
fob Cf. Gauthier (J 972) 65·6, who made this observation.
"7 For example, Euripides, Ht'lena 291 and rhe scholio" to Euripid~s. Af~dt!a 613: "S)'",OO-
Ion: people who entered into rcl~rionships of xenia u~d to cut a pi~ of bone in rwo and keep
one hal( th~msdyes and leave the oth~r with their partn~rs, 50 that if they or their friends or
relatives should have <X'Casion to visit them or v;u t/trsa, they might bring the half with them
and renew the xenia". See further Gauthier (1972) 52ff.
64 Encountn and Initiation
ro
·C
..,Q)
ro
~
'0
..,Q)
.£:.
--en>-
'-
c-
o
U
I M Y 1\ x 't-~~-t' 1\ x W N 0 ( raCt.,
'N 10 A" 0 c X 1\ W poe 1 € N I A N
E n 0 H CAT 0 n poe 1\ yew N raC'a;
010 rN HToYkAITWNErrONWN
Fig. , Ivory .,.boIon

Record of an alliance of x."iIl between a Phoenician and a Greek. IvOl')' plaque from
the Hellenistic period (2nd century B.C.), found in Lily cum. It was intended to
remind its owner that one "Imylch, son of Imilcho, [surnamed] .nibalos Chloros,
conduded a (pact of) xenia with LylOn, son of Diognetos, and his descendants" (IG
XIV. 279, with o. Mason in Snnitial26 (1976) 93-8). The descendants or dosat
laoaates of Imylch or Lyson could produce this .",.boI011 as a reminder or proof of
the original compact. The -bad' grammar suggests that the present text was inscribed
by the Phoenician panner.
archaeological record can in part be explained by the availability of a more
sophisticated device, unknown to the Homeric world - the letter. Apollo-
doros communicated his request to his father's xmo; in Lampsacus through
grammata, which a friend of his, in need of help, was (0 deliver them. 6K A
letter very similar to this does in fact survive, and yields a fascinating
glimpse into the operation and psychology of ritualised friendship. It was
written by Isocrates, the Athenian orator, to one Timotheos, ruler of Hera-
clea, whose father Klearchos had apparently been a pupil and a xenos of
Isocrates. The letter opens with some courtesies and flattering comments on
Timotheos' rule. But towards its end it becomes obvious that it was in fact
intended to recommend its bearer, Autokrator, to the attention of
Timotheos, and to renew, at the same time, the xenia between Timotheos'
and Isocrates' houses. 69
Such letters, no doubt, would be as good a proof of identity as a
symbo/on. But the paucity of symbola could also be attributed to the very
persistence of the original practice in classical times: in effect, any object
marked by distinctive attributes could serve as a symboJon. All that was
needed was an exchange and the parties' mutual consent to invest an object
with this function. The norm is thus probably reflected by the case of
Demos, son of Pyrilampes, who received a golden cup as a symbo/on from
the Great King. 70 As Demos needed cash for a trierarchy, he offered to give
the cup to Aristophanes as security in return for a loan of sixteen minae.
When they both got to Cyprus (Aristophanes was elected ambassador to
Evagoras, Demos was supposed to sail there as a commander of a warship),
Demos would redeem the cup with a payment of twenty minae. The redeem-
ing would be made possible by the fact that "on the strength of that
symboJon he would command plenty of goods and also money all over the
continent [i.e., Asia Minor]". What was meant by this is made clear by some
analogous passages in which a person seeks the help of another person
known to him only indirectly through networks of ritualised friendship. 7 I In
precisely the same fashion, the Great King's friends and dependants would
now help Demos financially, thanks to the special connexions which he
would be able to prove with the king.?2 The Persian court had probably
". Demosthenes 50.18 (Agai"st Polycles), d. Diogmes Laenius 8.87 for rhe letten of
introduction born by Eudoxos of Cnidus ro NC'ktab~nis of Egypt from Agnil~os of Spana.
", lsocrarrs, utter 7.13.
70 lysias 19.25 (0" the Property of Aristopha1tes).
71 For instance, Euripidn, Medea 613-4; Plaro, C,ito 45b-c; Demosrhen~ 50.18 and 56
(A~a;"st Polydes) .
.2 Th~re is a catch in this interprdation: how could Demos usc th~ symbolo1t 15 a device for
raising money al' long as it was in Aristophane!'i' possession? I do not know the ~nswer, but I
suspect some trick was implied by the speaker.
reserved and publicised a special set of objects - Xenophon's timia - the
bestowal of which conferred some of the privileges of being in the king's
favour. 7J Several splendid vases, some bearing Xerxes', others Anaxerxes',
name in Old Persian, Elamite, Akkadian and Egyptian hieroglyphs (some
found in Halicarnassus, others in Susa) clearly correspond to such descrip-
tions (Fig. 10).74
The exchange of such objects sufficed to seal the relationship. A common
meal might have been desirable, but was by no means indispensable for the
alliance to acquire validity. 7.~ Iphitos was murdered before he and Odysseus
could share a table, and Agesilaos and Pharnabalos' son, or Diomedes and
Glaukos, had no opponunity at the time to do so. Yet, on the other hand,
"the table of one's xenos" (xenike trapeza) became a metaphor for the
sacred nature of the bond; and the terms homositos, homotrapezos, and
homospondos, "one who consumes the same food", or "sits at the same
table", or "participates in the same libations" became marks of high regard
at the cOUrts of autocratic rulers. Wishing to point out the monstrosity of
Demosthenes' crime (i.e. the tonure and execution of his xenos, Anaxinos
of Oreas), Aeschines suggested that this was no ordinary homicide. It was
much worse, the killing of a man with whom Demosthenes had eaten, drunk
and poured libations at the same table. 7ft This is what turned his execution
into an asebema, a crime against the gods. For what the eating, drinking and
libations achieved was a further reinforcement of the bond. There is an
essential notion in human cultures that the possession of a common sub-
stance is the basis of a mystical bond. And, in most rites of incorporation or
communion, consubstantiality through eating and drinking is an ever-
recurring theme (see, for example, Fig. 7, p.55).n Through these rites the
gods were invoked as witnesses, and thus the feasting rituals merged with
the hospitality rituals. In consequence, the xenos-stranger, as the xenos-
guest-friend, became the protege of Zeus Xenios. We shall return to the
consequences of this identification in Section S.2.
,-' Xmophon, Anabasis t .2.27: "a horw with 3 gold-mounted bridlc, a gold necklace and
bracdds, a gold dagger and a Pcrsian robt-". In his sp«ch ro thc noblest of Persians before the
invasion, Xerxes is similarly said ro have promised rhar "whosoevcr comes with his army bnt
equipped shall receivc from me such gifu as are rcckoned ti",iotata among us" (Herodotus
7.8).
, .. Cf. R.G. Kcnt. Old Persian (Ncw Haven. Conne\.'ticut, 1953) 115, XVs and AVs. For
orher, unmarkcd, pots and utdacts which might havc migrated from place to place as gifl" wnr
from onc rulcr to another, we' Coldst'nam (1983).
" On the significance of the common mc~1 as a de-vice crearing solidarity, s« Finley (t 977)
145-6; Gould (1973) 79·80, Murray (1983a) and (1983b).
76 Aeschin~ 3.224 (Against Ctesiphon).
" Cf. Van Gmnep (1960) 29; Finley (1977) 123 ff; Pitt-Rivers (1973) 94.
'The Initiation Ritual 67
Fig. 10 Peni'.n symbolon

Vase of alabaster, panly broken, bearing Xerxes- name in Old Persian, EJamire,
Akkadian, and in Egyptian hieroglyphs. Achaemenid period. II betokened conne-
XIOf}lI "'ith the Great King. Xenophnn could Iso have had in I'I'lipd objccr such as
this when he CQmmented: ··WhO$e gifts a~ so readily recognised a!) SQme of those
which the king gives, such as bracelets, necklaces and horses with gold-stUdded
bridles? For, as everybody knows, no one there is ever allowed to have such things
except those to whom the king has given them"' (CyropaedUJ 8.2.7). These royal
symbola served both as a marie of the prestige anaching to their owner ,.nd as
serviceable proofs uf one"s being placed under the ~ing's protection. The latter
peculiarity might have conferred practical advantages in life. For exampl~ ~n the
strength of such a symbolon th~ Athenian Demos, son of Pyrilampes, hoped to
receive a loan or a gift from other proteges of the Great King who were previously
unknown to him (d. abov~ p.6S). (I am grateful to Dr Susan Sherwin-White for
drawing my anenrion to this object.)
The relative scarcity of documentation describing this ritual should by no

means be taken as indicating its lack of importance. The ceremony belonged
to one of those basic facts of existence which need not be questioned or
described. By contrast, Herodotus' accounts of the blood rituals were pro-
bably inspired by their very oddity. The medieval rites of vassalage, the
centrality of which is beyond dispute, are also relatively poorly
documented. 78 But if direct documentation is in short supply, the signifi-
cance and frequency of the ritual of xenia seem to be confirmed by the
terminology itself. As pointed out already, the term xenos was consistently
applied to people originating from different social units (Appendix A).
However, the terminology followed only indirectly from the fact of sepa-
rateness; most immediately, the term xenos resulted from the ritual act by
which the two persons bestowed the title on each other. This is made clear
by the declaration formula itself - "I make you my xenos" - and by terms
such as "making one one's phi/os and xenos" or "becoming one's xenos" -
formulae which are probably shon-hand references to initiation
ceremonies. 79
Why, then, was the ritual necessary? Students of social relations have
encountered notorious difficulties in formulating a satisfaCtOry definition of
'ritual'. There are now as many definitions as there are theories, but none
seems to have won general acceptance. Interpretations of particular in-
stances of ritual vary, therefore, with the conceptual frameworks through
which they are approached.
Unless one accepts certain assumptions as axiomatic, the criterion by
which one can tell whether one theory is better than the other is its ability to
interpret a larger number of data by making simpler assumptions. An
interpretation of the rites of initiation is therefore offered in the light of a
theory which best meets this criterion - the theory which was laid down by
Van Gennep at the beginning of the century. It must be stressed that some
interpretation of the ritual is essential. Judged by the amount of attention
devoted to it, by the refined vocabulary in which it was expressed, and the
number of metaphorical usages to which it gave rise, the ancient peoples
were doing something which, to them at least, appeared extremely signifi-
cant. What?
Before trying to answer this question it is helpful to introduce a metaphor
,. Cf. I~ Goff (1980) 240.

7'9Iliad 4.266; Herodotus 4.1 54.3, whe~ rhe term u~ is ";(l{)OA~V btl ;dVlO;
Xenophon, AMb,u;s 1.2.3 and 7.1.8, in which case Xenophon is speaking about himself;
Aeschines 3.224 (Aga;".st Cus;phorr); lsocrates 4.43 (Parrt'gyr;cu.s) and 19.5 (Aegirrtticu.s)i
Pauunias 4.4.8; Plutarch, Solon S.2.
5 Continuity 69
used by Van Gennep to describe the perception of social boundaries in
primitive mentality:
A society is similar to a house divided into rooms and corridors. The more the
society resembles ours in the form of civilisation, the thinner are its inner panltions
and the wider and more open are its doors of communication. In a semi-civilised
society, on the other hand, sections are carefully isolated and passage from one to
another must be made through ceremonies which show exrensive parallels to the
rites of territorial passage... 80
The ritual, then, could be viewed as effecting a breakthrough in the
psychological barriers of strangeness and hostility - a passage through Van
Gennep's 'partitions'.
Admittedly, this argument is to some extent circular: we infer the 'parti-
tion' mentality from the frequent practice of ritual; there does not seem to
be an independent way of ascertaining its existence. But Van Gennep's
vision receives further confirmation if we extend our framework to the
cross-cultural category of 'ritualised relationships' (above, Section 2.5); in
all these relationships, with a few exceptions, there is an initial coincidence
of 'strangeness' and 'ritual,.81 The techniques used are, in this respect,
unimportant; it is immaterial whether the ritual consists of the baptism of a
child, the mixing of blood, the chewing of a common substance, or the
exchange of some 'sacred' object. What matters is that everywhere a drastic
change in the relative position of the partners is envisaged. The partitions of
strangeness that are broken through are at once replaced by a pretence of
friendship and excessive familiarity: everywhere a new bond is being insti-
tuted through which goods and services flow and co-operation is effected.
The ceremony initiating relations of ritualised friendship can thus be viewed
as a veritable rite de passage.
3.5 CONTINUITY
One peculiar feature of the rites of initiation was, as we have seen, to bind
together not only the participating individuals, but also their respective
lineages. For ritualised friendship was thought to outlast the individual
actors and, conspicuously mimicking kinship ties, pass on to their descen-
dants. Thus, a person could die, but the role of xenos could not. Leach's
general observation that "An enormous amount of human ritual... is
focussed around this central problem of asserting continuity in defiance of
10 Van Gennep (1960) 26.
III CE. S«tion 2.5. I 5ay 'iniri~I', since riruali~d relarionships arC' rC'markably fluid, and in
the course of rime both the conceprion of barriers and the need for ritual mighr change. Cf.
Gluckman (1962) and Pin-Rivers (1968) 412-1.l.
the threat of mortality"81 probably deserves more attention than it has

generally been accorded.
In real life, this metaphysical continuity needed constant fostering, lest it
be forgotten. People would keep a record of both their inherited and their
personally acquired xeno;. The information concerning this, together with
whatever objects were originally exchanged, would pass from fathers to
sons, serving as a reminder of the initiation ceremony.8.l It is the sons'
perspective on the ritualised-friendship stemma which is reflected in the
technical term patrikos xenos. 84
Sons would sometimes reinforce this continuity by occasional exchanges
of gifts or hospitality. Nevertheless, the link could be forgotten, and then a
renewal ceremony was necessary to reaffirm the original compact. 85 The
need ritualistically to renew the bond shows that the relationship was
thought to persist independently of individual actions or wishes. The fact
that neither they themselves nor their fathers had ever seen each other
before did not prevent Diomedes and Glaukos from recognising their
mutual obligations. All that was needed was a reiteration of the rites of
initiation. In a similar case, Darius had forgotten about the bond which tied
him to Syloson; but, when reminded of it, he suddenly reassumed the role of
generous friend. 86 Antigonos Doson nominated Brachylles as an official in
his service in repayment for the good deed done to him by Brachylles'
father; the favour owed was assumed to have devolved from father to son. 87
Finally, (socrates' estrangement from his xenos, Klearchos, on account of
the cruelty of Klearchos' rule, brought the relationship to a low ebb but did
not put an end to it. After Klearchos' death, Isocrates could therefore
propose to Timotheos, Klearchos' son, to renew their former ph;lia and
xenia (avaV£OUJ,lEVOC; 'ri}v epl1£av xat ~EV£av): paraphrasing closely the
ritual declaration, Isocrates said that he would "accept [Timotheos) hack n
(at b·Moblxof1ol.).88 And this was an ever-recurring pattern. The inference
must be that the renewal ritual was intended to reinvigorate a formal tie
which had fallen into disuse hut had not been abolished.
12 Leach (t 972) 316.
11.1 The actors throughout s«m to havc th~ most vivid recollcctions of thc smallest details of
the tv~nt .
... See Appendix A. last column, for a list of inherited relationships. This hereditary e1cment
is not prCSC11f in all ritualised relationships.
• ~ ~ technical ~m is nvavroUo6al, thc spectrum of which is somctiJnn wider than 'to
renew'; if can also lignify 'to rehear~" or 'go ovcr verbally'. d. L. Robe", Htllnr;ca 1 (1940)
96 n.S. For 'renewal' within the context of ritualised friendship, sec Thucydides 1.3.1.4 (with
phi/ill); lsocratet 4.43 (Parfegyr;cus); Polybius 33.18.2; Scho/iorf to F.uripioo. Mtd~a 613
(2.17S Schwam); Cu"ius 6.5.1-5 (hosp";u,,,).
'6 Herodotus 3.140.
., Polybius 20.5.12.
'1 lsocrafes. utter 7.13.
.5 Continuity 71
The question thus arises: Was there any formal way in which this con-
tinuity could be broken? Theory must be separated from practice. In prac-
tice, the obligations of ritualised friendship could certainly be ignored. One
could leave the bond unactivated until it fell into abeyance; one could fail to
provide much-needed assistance; and one could even inflict injury on a
partner. In all these cases interaction would in practice stop. But, was there
any formal act by which the imagined continuity could be disrupted?
The available evidence does not easily lend itself to interpretation.
'The Cretan ruler Etearchos, we are told, made one Themison, a Therean
trader, his xenos with a vicious purpose in mind. He first exacted an oath
from Themison that he would perform for him whatever service he desired,
and then bid him take his own daughter and throw her into the sea.
Themison was very angry at this most improper suggestion, and dissolved
his xenia with Etearchos. In a similar fashion, Amasis, King of Egypt, is said
to havc dissolved his xenia with Polykratcs of Samos. 89 Such tcrms as thesc,
as well as analogous terms occurring within the framework of the rela-
tionships between proxenoi and the city they represent,90 and the fact that a
disruption ceremony might be necessary in cases where the termination of
the relationship involved economic settlements, seems to suggest that such
ceremonies were indeed available. I know, however, of no source which
might indicate of what they consisted. Parallels from other ritualised rela-
tionships suggest that such ceremonies tend to mirror in reverse, although
not in perfect symmetry, the original initiation ceremonies. If applied to our
case, this would mcan that the commitment to become each other's xenos
was withdrawn, and the gifts of initiation returned. However that might be,
it seems clear that the rupture ceremony did not involve a spectacular act
similar to the feudal 'throwing of the rod' (exfestucatio) or the Roman
'breaking of the tessera' (tesseram confringere), by which the whole idea of
stoppage could metaphorically be expressed. 91 And it seems certain that it
was only seldom used. If it were common, Perikles would surely have
resorted to it to disclaim his compromising involvement with King Archida-
mos. All Perikles did, however, was to remove his estates from the context
of obligations affected by xenia; technically, Perikles and Archidamos con-
tinued to be xenoi. This fits in with what we know about public attitudcs to
19 Herodotus 4.154.4 and 3.43.2, d. Plutarch, Moralia 151F; the verb is dialuein, which is
also used to C'Xprns diYor~, d. Syll.l 364, lin~ 58. For the use of dialu~in for dissolving philia,
d. Hyperides 5.21 (Agamst Dmwsthmes).
90 See, for example, Thucydides 5.43 wirh Daux (1937) concerning Alkibiades' renew~1 of
fJTOxmia with Spina.
91 Cf. Mommsen (1864) 343; Marc Bloch (1912); i.e Goff (1980).
such procedures in other ritualised relationships: in most cases the disrup-

tion of the sacred bond entails disgrace. 92
I have outlined in this chapter the symbols, techniques and devices which
served to convert strangers and enemies into friends and intimates. It
appears from this outline that only very little has changed in this respect
following the emergence of cities; the rituals used by Greek citizens show
remarkable similarities with the rituals which had been used by Homeric
heroes and non-Greek 'big men'. In other words, the outer shell of ritualised
friendship survived almost intact into the world of the city. But not so its
substance. The entire fabric of social relations was now different, and
ritualised friendship became entangled in social and economic formations
which before did not exist. Of necessity, it affected them and was affected
by them. To gain a clearer perception of these transformations, we turn to
the material aspect of the relationship.
'2 Cf. Gudeman (1971) 59; Hammel (1968) 79; Tegnaeul (1952) 74. In Hispanic America,
many ~Ie acquire more compadratgO ties than they can maint2in. As the tie cannot formally
be broken, the solution lies in maintaining superficially correct relariomhips with most, but
developing effective exchanges with only a few (Foster in Schmidt et a!.. Ns. (1977) 22).
4
THE CIRCULATION OF RESOURCES
4.1 THE PATTERN OF EXCHANGES
In one of his fiercest outbursts of civic righteousness, Demosthenes accused

the various classes in the Greek cities of unpatriotic behaviour. He said that
each class was in its own way undermining civic independence: the notables
(gnorimoi) entrusted with political leadership took bribes, pretending that
this was nothing but xenia, hetairia, and philia with Philip; the demos,
rather than punishing these notables and putting them to death, regarded
them with admiration and envy and wanted to become Philip's friends too.
Demosthenes then continued by saying that it was because of this attitude
that the Greek states had succumbed to their enemies. The Olynthians, for
example, insofar as they had refrained from the practice, had successfully
withstood attacks by Sparta.
But when some of them began to accept bribes (dorodokein, i.e. from Phillp], when
the populace (hoi pallo;) was so stupid, or, let us say, so unlucky, as to give more
credence to those persons than to patriotic speakers, when lasthenes had roofed his
house with timber sent as a present from Macedonia, and Euthykrates was keeping a
large herd of cattle for which he had paid nothing to anybody, when one man
returned home with a flock of sheep and another with a stud of horses, when the
demos, whose interests were endangered, instead of being angry and demanding that
these men be punished, stared at them, mvied them, honoured them, and thought
them fine fellows, ... then ... nothing could save them. I
This is such a slippery piece of evidence that it seems at first sight hopeless
to try to reconstruct the reality behind it. Does the orator skilfully inflate an
insignificant practice, or does he relentlessly expose an otherwise suppres-
sed, large-scale activity? No immediately obvious clue is available by which
to choose between the alternatives. Nor docs the frequent recurrence of
similar allegations in ancient sources (mainly the Attic orators)2 necessarily
validate their truth: studies of witchcraft and gossip indicate that wide-
spread charges and calumnies often mask a relatively innocent reality. Small
wonder, therefore, that no balanced view exists in modem research either of
the prevalence of bribery in the ancient Greek city or of its impact on the
I lXmosrhenes, 19.259·66 (Dt" Fa/5a Lt"gat;orrt), d. Diodorus Siculus 16.54.
2 Apart from rhe insrances referred ro in this Chaprer, s« Davies (1971) 133-5 for refer-
ences ro Demosrhcnes' dorodoltia, and Sre. Croix (1981) 609 n.58 for a lisr of pas,ages
cont~ining accusarions. For Philip 11'5 policy of bribery, s« Hammond and Griffirh (1979)
210ff. For the 'venality' of cirizrn'generals, sec Pritchett (1974) vol.lI, 126·32.
73
74 The Circulation of Resources
social and economic status of the parties involved. Views oscillate between a
more-or-less total acceptance of accusations like those pronounced by De-
mosthenes and serious doubts as to whether bribery played any special role
at all in the extension of the power of the bribe-giver (whether he be Philip
or some other powerful figure in the world of dynasts and kings).·l
In this Chapter, I endeavour to show that it is in fact possible to arrive at
a more realistic assessment of this 'external bribery'." Once we separate
value judgements from actuality, it becomes apparent that the practice was
all-pervasive. Indeed, it was one of the deeply entrenched features of the
ancient world, in relation to which the practices inaugurated by the polis
were but a frail novelty. Upper class individuals who were integrated into
politically separated communities circulated between themselves substantial
amounts of wealth and performed significant services for each other. This
circulation of wealth did not coincide with the formal channels of interac-
tion that existed between these various communities. But it did coincide to a
large extent with networks of ritualised friendship. Unravelling the pattern
of these networks will bring into perspective a feature of ancient society
which, however central, is all too often forgotten, and has never, to my
knowledge, been systematically explored. For it will then become apparent
that the elites of the ancient world were not confined within the boundaries
of their immediate groups (whether a tribe, a petty kingdom, a Persian
satrapal court, or a polis) but participated in an extensive system of alliances
outside the groups to which they belonged. This external involvement
constituted a potent factor in the definition of their aims and in the forma-
tion of their outlook. First, it affected their social and economic standing
within their own groups: for power and prestige acquired through one
system could readily be transferred to the other, and their social status
inside the group could be improved by means of resources secured from the
outside. Secondly, it enhanced their potential for geographic mobility; for,
as in the world of Odysseus, "one who had a xenos abroad had an effective
substitute for kinsmen, a protector, representative and ally. He had a refuge
if he were forced to flee his home, a storehouse on which to draw when
compelled to travel, and a source of men and armS if drawn into battlett..s
Finally, it facilitated the formation and growth of new organisations, which
} Cf., for example, Lewis (1977) 143 and Davies (1981) 66ff. who bt-lieve rhe accuutions;
CawkweJl (1963) 204, who minimisn the effter of bribt-ry. For a position dose to me one
raken in this study, see Perlman (1976); Ste. Croix (1981) 298·9; Schuller in Schuller ed. (1982)
9·17.
.. Bribes givm within the city framC'work, such as paymmrs to individual ciriuns for mrir
votes, wtrt subject to altoFther difftrtnt mechanisms and thus fall outside the scope of this
study. For rhis distinction, see Finl~' (1983) 83·4.
, Finley (1977) 102.
in the long run proved even more successful than the polis itself. The
evidence regarding these organisations has not yet been examined in a
comprehensive perspective. But even a cursory glance at the material reveals
a regularity which transcends chronological differences and ethnic bound-
aries: throughout Greek history, ritualised friendship appears as one of the
building blocks of the Persian, Macedonian, and Hellenistic ruling circles.
4.2 THE GOOD GIFT AND THE BAD
In Classical and Hellenistic times two types of word could be used to

designate the notion which we commonly render as 'bribe'. The first, and
most common, consisted of variations, derivatives and combinations of the
words for gift: doran, dorea, dorema, dorodokia, etc. The second type
consisted of words such as chremata (money), misthos (reward, wage). In
both cases, the words were subordinated to verbs expressive of actions such
.-
as giving, receiving, accepting, persuading. 6
What is remarkable about these words is their ambiguity. For they signify
at one and the same time the concept of bribe and the (to us) logically
oppo~d concepts of 'gift't 'money' and 'reward'. In other words, there was
in the Greek language no vocabulary of bribery distinct from that of
gift-exchange itself; the same set of words served to denote both practices. C"I
I-
The question therefore arises: How could the doron-gift be distinguished >-
0-
from the doron-bribe? How could the preferred meaning be distinguished in o
use? The answer, as illustrated by a piece of rhetoric from Hyperides' speech U
against Demosthenes, seems to be that the choice was in each case dictated
by the context:
You give full permission, gentlemen of the jury, to the orators and generals to reap
rewards. It is not the laws which grant them these privileges but your tolerance and
generosity. But on one point you insist: your interests must be furthered, not
opposed, with the money they receive. Now Demosthenes and Demadcs, from actual
decree'S passed in the city and from proxeniai, have each received, I believe, more
than sixty talents, quite apan from Persian funds (basilika) and the money sent from
Alexander. If neither of these sources suffices for them, and they have now accepted
dora which threaten the city's life itself (literally: dora against the body of the city],
can we doubt our right to punish them?'
As here, it is invariably the principle of community which serves as a
point of reference; the 'gift' shades off into the 'bribe' when contrasted with
the notion of communal interest. A few outstanding examples, drawn from
6 An analy!is of the words for 'giEr' in Homeric and Classical Greek may be' found in
~nvmiste (1951) and (1968), vol.2. 65·79. As these studies take the etymologiC'S of the' words
as [heir point of depanurr, the' concept of bribt- eludes their grasp.
7 Hyperidn 5.25 (Against Dcnwslht'rft's).
a variety of situations, will suffice to show that this usage was indeed the
norm and not the exception. And the~ could be supplemented by the
examples given in the Thesaurus Graecae Linguae, s.v. doron, dorea, doro-
do/(ia, which follow the same pattern.
In the fifth and fourth centuries, the Attic orators had frequently tried to
win the sympathy of democratic juries by alleging against their opponents:
this man has accepted dorea against you. 8 At the beginning of the third
century B.C., the citizens of Chersonese had taken a collective oath not to
accept dorea to the disadvantage of the city and the citizens. 9 Even the word
dorodokia - the only one of the gift words unknown to Homer - meant
basically "acceptance of dora". 10 To transform it into "the acceptance of a
bribe", it had to be qualified. For example, we are told that King Leotychi-
das led a Lacedaemonian army to Thessaly. But, when he might have
subdued all the country, he took a great bribe (tbO>{X>b6x'lOE Q{rfUQLOV
noll6v) and withdrew from Thessaly. For this betrayal of communal in-
terests, he was tried, banished from Sparta, and his house destroyed. I I Even
.-
if this clash with public interest is not explicitly stated, its very existence is
implied by the context denoted by the words. Demosthenes alleged, for
example, that Kallias, son of Hipponikos, "was said to have taken dora on
embassy". 12 These dora were bribes since, as the orator him~lf explained
later, a man entering into the service of the community was expected to be C"I
I-
adorodoketos, that is, to abstain from gift-exchange. In sum, it was virtually >-
0-
impossible to make reference to the concept of bribery in the Greek lan- o
guage without implying the notion of communal interest. Autocratic rulers U
could only receive gifts, not bribes. For all their dislike of tyranny, Herodo-
tuS and Aristotle could not conceive of the monies that Peisistratos, the
Athenian tyrant, received from 'friends' abroad as bribes. IJ Nor did De-
mosthenes ever accuse Philip of Macedon of receiving bribes. Gift-exchange
presupposed autocratic rule in much the same way as bribery presupposed a
political community. Nowhere is this more evident than in the oath Sworn
by the Athenian nine archons "not to accept dora on account of their
office", believed to have originated in the days of Akastos, the first archon
of the new regime which usurped the functions of the Codrean basileia. 14
• DiMrchus 1.40 (Against Dmtosthnr~s); 2.1 and 25 (Agai"st Aristogiton); Hyperidcs 6.10
(M",~ral Sp~e,h); Dnnades, 0" the Twtlvt Years 21.
• SyU.) 360.
10 C£., for example. Xtnophon. Anabasis 7.6.17.
II Herodotus 6.72, d. 6.82 for similar accusations against Kleomenes. and Thucydidts
2.21.1 for accusarions againsr Pleisroanax. son of Pauuniu.
u Demosthmn 19.273 (D~ Falsa Legatio"e)
It Herodotus 1.61; Aristotle. Th~ Athnrian Co"stitution 15.2-3.
14 Aristotle. Th~ Athnr;an Co"stitutio" 3.3 and 55.5.
In itself, all this might seem insignificant, since a sense of opposition to
communal interest is also implicit in the modern concepts. The modern
concepts, however, contain a further important ingredient - the notion that
bribe-taking is condemnable in itself; that it is an absolute wrong, a sin,
even if it ;s not opposed to the communal welfare. This sense of wrong
derives from ideologies, ethical or religious, which originate outSide the
political sphere. In the ancient world, by contrast, it seems that there were
no ideological spheres outside the political with which bribery could come
into conflict. Two considerations lend some validity to this argument.
Take first the moral connotations of the words. In a study of corruption
in modern states, one author has complained that there are no 'neutral'
words to express the idea since moral disapproval is automatically implied
in all the available terms: 'bribe' can only be a 'bad' word. IS In the Greek
language, by contrast, not only were there no neutral words for bribery, but
such terms as were available carried agreeable overtones. The predominant
.-
attitude is reflected in Telemachos' suggestion to Mentes: "But I beg of you,
wait a little longer, eager though you may be to depart; stay to bathe and
enjoy good cheer and be given a gift (doron) to take aboard triumphantly,
something precious and very beautiful 10 remember me by, a true affection-
ate parting gift from me, such as dear xeinoi give to xeino;". 16 The value of
the gift was unabashedly recognised as an index of the attachment between C"I
\.....
giver and recipient, When, for example, Xenophon visited the Thraclan king >-
0-
Seuthes, a Greek living at the court tried to encourage him to bestow a gift o
on the king thus: "For I am quite sure," he said, "that the greater the gifts U
you bestow upon this man, the greater the favours that you will receive at
his hands". l7 Outside the communal sphere, then, the gift is implicitly
'good',
Secondly, consider the word adorodoketos, customarily - but mis-
leadingly - translated as 'incorruptible'. If it had meant 'incorruptible' in the
modern sense, how could it have been possible to praise CIVic office-holders
for having discharged their duties in an adorodoketos fashion? IS Such
characterisations in honorific decrees would have been ludicrous, offensive,
and out of keeping with a document intended to confer glory. Likewise,
• S Stirling (1968) 49.

•" Odysst')· 1.309-13; d. Grierson (t 959) for a similar attitude to gift-giVing in the mC'dieval
dark ages.
17 XC'nophon, Anabasis 73.20 .
•1 For C'xampk, IG II 2 lB, line 12; J. Crampa. l.ab,aunJa: Th~ Crull Insc"ptions
(Lund-Stockholm t 969- 72) no. 4.1.
how could Thucydides extol Perikles for being adorotatos (a synonym of
adorodoketos), and contrast him with individuals who, in order to satisfy
their own private ambitions and greed, promoted policies detrimental to the
state?J9 This could hardly have been a bland term denoting a tnon-sin' or
'non-offence'. Quite the contrary, adorodoketos embodied a positive virtue,
a veritable prescript for ideal behaviour. The full import of the term
becomes clear only in the context of the battle of ideologies between heroic
and civic virtues. In the process of remodelling the hero to the standards of
communal life, the polis had to replace the ideal of personal fidelity, of
which the gift was the clearest mark, with the counter-ideal of obedience to
communal rules. Adorodoketos thus came to denote a feature of the ideal
citizen - a virtue manifesting itself in a willingness to show more regard to
communal rules than to personal obligations. To turn this negation of
heroic virtue into a term of praise and offer communal interest as a new
standard of individual morality was probably one of the most significant .-
victories of the community over the hero. Thus, while lavish gift-givings had
been the mark of heroes and kings,20 abstinence from gift-exchange became
the mark of the good citizen.
The modern and ancient notions of bribery are therefore only very
partially comparable categories. They refer to different fragments of reality;
they are stored in different mental compartments; and they have different C"I
I-
overtones and reverberations. For in the Greek city, bribery was a recent >-
0-
'invention" conceptually only very slightly separate from archaic gift- o
exchange. A brief contrast between the position of the people engaged in U
gift-exchange in Homeric and civic societies may elucidate this point.
In Homeric society, gift-exchange waS the chief method of organising the
supply of goods and structuring social relations. Gifts flew in all directions,
but of particular importance were gifts exchanged between the heads of
different noble households. For one thing, it was this exchange that enabled
the circulation of wealth across household lines, introducing indispensable
extraneous commodities (metals, for example) into the otherwise closed
., Thucydides 2.65.8-9; d. 2.6O.S, where Perikln identifies himKIf IS a patriot (phi/opolis)

and lupnior to the influence of money. for Xmocratn' refusal of a gift from Antipater. see
Di08tnes Lamius 4.8. Of Socrates, the latter author writes (2.25): "He showtd his contempt
for Archelaos of Macedon and Scopal of Cranon and Eurylochos of l...an,u by refusing to
acr:J't their presents or to go to their court".
2 Cf. Xmophon's o~rvation "lavish gift-giving (polydoria) among the kings continues
even to this day" (CyropMdia 8.2.7), and Thucydides' complaint, evidently using the civic
practice as the norm, that with the Odrylian kings "it was not possible to accomplish anything
without giving gifts" (Thucydidn 2.97.2).
economy of the oikos. 21 For another, "the whole of what we call foreign
relations and diplomacy, in their peacdul manifestations, was conducted by
gift-exchange".12 These vital transactions were supported by a moral code
which portrayed the practice in an exceedingly favourable light. The central
message of the code was simple: a man's gift-giving capacity, or his willing-
ness to abide by the obligations imposed by the gift, was a measure of his
moral quality. Immoral behaviour consisted of not giving, or of breaching
the obligations emanating from the acceptance of a gift. Deserting an ally
for the sake of a greater gift might have given the notion a pejorative
shading,23 but such an evaluation was a reflection of the point of view of a
deserted partner - of a social equal, not an inferior or subordinate. No
criticism emanating from the lower levels of the social pyramid was
conceivable. 24
The polis, having inherited this mentality from the stateless past, had to
reconcile it with one of its most central, though unverbalised, doctrines.
Briefly, this doctrine held that the polis was an organic whole, the different
.-
members of which were strictly interdependent. Private actions were now
everybody's business, and had to be judged good or bad not by standards set
by powerful lords but by utility to the polis as a whole. Even justice came to
be defined as "that which is to the common advantage".25 The many thus
claimed priority over the individual, and deemed themselves justified in C"I
I-
coercing him into compliance with their norms. On these assumptions, a >-
0-
gift-exchange with an outsider quickly came to be viewed as a mark of o
misguided solidarity. For outsiders were by definition enemies, and a nexus U
with an enemy might become a threat to the whole community. Hence the
easy association between bribery and treason: ..... whenever, men of Athens,
21 Odysseus, for instance, is said to have amassed in the court of Pheidon, king of Thespro-
tians, so much bronze, gold and iron as to suHi~ to sustain ten whole gtonerations after him
(Odyssey 14.324-5). Cf. Odyssey 4.129-30. 14.285-6 and the complaint of Ishi-Adad to
Isme-Dagan in a letter from the Mari archive: "And you send me this miserable amount of
Ie-ad! ... What is lacking in your house that a brother cannot gram the wish of. brother?" (G.
Dossin, Correspo"d~nu de lasmah-Addu, yol. V of Archives Royales d~ Mar; (Paris, 1952),
no.20).
22 Finley (19n) 66.
23 Cf., for example. the remark ThemislOkles made to the Corinthian general "Nay. you of
all men will not desert us; for I will give you greater dora than the king of Medes would send
you for deserting your allies" (Herodotus 8.5).
24 Hesiod's complaints against the "gift-<kvouring kings" (Worts and Days 263-4) might
already be the m3rk of the emergent public spirit. Indeed, accountability was one of the marks
distinguishing the citizen from the autocrat. for example. when the legitimacy of the Samian
tyrant Maiandrios was challenged, Herodotus makes a nobleman ~y: "You (Mai3ndrios) are
nor worthy to reign o\'er us bC'ing a low-born knave and rascal. S« to it rathC'r that you give an
account of the moneys that you have handled" (Herodotus 3.142).
H Aristotle, EthiCiI Nicomachea 1160a4.
you see any man taking bribes, you may be sure that he is also a traitor",
says Demosthenes,26 and Dinarchus makes the same point even more force-
fully: " ... don't you realise that to take bribes in order to betray the city's
interests is one of the greatest crimes causing the most irreparable harm to
the cities?"27
There was thus an ideological conflict focussed around the practice of
gift-exchange, between the old and the new, and also between the polis and
the a-political world outside it. For there can be little doubt that the polis
was the exception, and the outside world the norm. There, nothing had
changed. Persian potentates, for example, seem never to have grasped that
the Greek city did not obey the rules of gift-exchange. Repeatedly, they seem
to send monies to leading Greek politicians in the name of 'friendship"
being apparently unaware that the noble practice had there become an
offence punishable by death. 28
This association between the exchange of goods and services and
friendship is important for analytical purposes. For goods can also be
.-
exchanged outside the context of friendship, and the two types of exchange
are mutually opposed. Crudely, the distinction is this. Outside the context
of friendship - in trading relationships, for example - the exchange is a
short-term, self-liquidating transaction. Once the benefits are obtained, the
social relationship is terminated. The transaction does not create moral C"I
I-
involvement. By contrast, within the framework of amiable relations >-
0-
(kinship, friendship, ritualised friendship), exchanges have a long-term ex- o
pectancy. Gifts beg counter-gifts, and fulfil at one and the same time a U
number of purposes: they repay past services, incur new obligations, and act
as continuous reminders of the validity of the bond. Non-reciprocation is in
this context frequently interpreted as a relapse into hostility. The signifi-
cance of the distinction for our purposes is this. The simple trading of
benefits was certainly practised in the ancient world, and may have been
important. But we hear remarkably little of it. What we hear of repeatedly
are the exchanges between partners who were morally involved. The vast
majority of attested cases of gift-exchange were conducted within the
framework of relationships that, ostensibly at least, were both warm and
expected to last.
It may now be possible to define what type of evidence might indicate the
extent and frequency of the circulation of wealth within networks of ritual-
ised friendship. Two guiding principles can be set out. First, the distinction
~mosthenes 19.268, d. 258 (Dt Falsa Legaticmt).
26
Dinarchus 2.7 (Against A,istogitorr).
21
11 The famous caKS are that of Arthmios of Zelea, Kallias of Athens and Timokrares of
Rhodes, for which see Hofstetttr (1978) nos. 53, 168 and 326.
between 'gift' and 'bribe' is irrelevant for our purposes since the 'good' and
the 'bad' gift are the same practice seen from different moral standpoints.
For an assessment of the scale of circulation of wealth, the evidence relative
to both bribery and gift-exchange will have to be considered. Secondly, only
traditional exchanges will be taken into account. There must be, in the
sources, either an assumption of amicable relationships, or a chain of giving
and counter-giving: the simple trading of benefits docs not qualify.
The goods which changed hands (or were temporarily pooled) under the
name of 'gifts' can be reduced to four major categories: natural products,
valuables (in the form of money or precious objects), troops and estates. The
non-material services which are less easily quantifiable can profitably be
classified under three headings: ritual services, private services and services
carried out within the context of political institutions. These will be consi-
dered in Chapter 5. It should be conceded that such a classification is to
some extent artificial. First, the networks of ritualised friendship combined .-
within anyone relationship goods and services of different quality. It was
the potential for effecting these combinations that rendered the exchanges
so rewarding and the relationships so efficacious: one party would be able
to provide what the other needed and, moreovcr, when he needed it.
Secondly, the classification covcrs most, but not all the goods and services
C"I
exchanged. One important item which cannot be accommodated within I-
either category is that of slaves, and it might briefly be considered here. It is >-
0-
hard to form an estimate of the extent of the practice, but the following o
U
incident, recounted by Demosthenes, together with similar accusations
levelled against Philokrates, might indicate that it was not uncommon.
He fAeschines] related an incident which, he said, had filled him with deep indigna-
tion. On his journey home Ii. e. from Philip] he met Atrestidas fan Arcadian]
travelling from Philip's court with some thirt)· women and children in his train. He
was astonished, and inquired of one of the travellers who the man and his throng of
followers were; and when he was told that they were Olynthian captives whom
Atrestidas was bringing away with him as a gift (dorea) from Philip, he thought it
was a terrible business, and burst into tears.2'J
What probably outraged Aeschines was that the enslaved people were
Greek citizens, not that slaves were given as a gift.
Despite these difficulties, however, classification is necessary both in
order to highlight the economic level at which the transactions were con-
ducted and as a preparatory step towards the exploration of the other
themes of this book.
1.9 ~mosthenrs 19.305-6, d. 309 (De Fa/sa Legatione) and Diogenes Larrtius 2.53 for the
present of captivr slavrs sem to Xrnophon by Phylopidas thr Spartan.
4.3 NATURAL PRODUCTS
The information available on the circulation of natural products along

networks of ritualised friendship is scanty and deliberately distorted. The
direct evidence consists of a few statements made in the law courts by
people trying to slander opponents and improve their own public image.
Timber and grain figure most prominently. Andocides, for example, plead-
ing after a period of exile for his reinstatement as a citizen of democratic
Athens, boasted that during the oligarchic coup of 411 B.C.
• at once proceeded to supply your [democratic] forces in Samos with oar-spars ...
since Archelaos was my ancestral xmos and allowed me to cut and export as many
as I wished ... 30
In the same speech, Andocides also made vague hints about having
secured for the city a supply of grain from Cyprus. 31 In a speech by
Dinarchus, Demosthenes was accused of having received a thousand
medimni of wheat a year from the tyrants of Pontus in return for bronze .-
statues which he set up in their honour in the agora. 32 Other examples come
from the Demosthenic corpus. Demosthenes himself alleged that Philokrates
and Aeschines had received, among other things, timber and grain from
Philip, and accused the Olynthian notables of having accepted timber and
livestock from the Macedonian monarch. 33 Finally, Timotheos, the Athe-
C"I
nian general, is said to have received a gift of timber from Amyntas, king of I-
Macedonia. 34 >-
0-
The question is this: Can we form a more adequate estimate of the scale o
U
and frequency of such transactions? In other words, were these exceptional
occurrences, or regular practices? Furthermore, is the magnitude of re-
sources mentioned in these passages a wild exaggeration, or some reflection
of reality?3s
What we sec in these passages is, in my view, merely the tip of an ice~rg.
In communities professing egalitarian ideologies, people are reluctant to
give publicity to benefits they exclusively enjoy - in particular when scarce
products are involved. Indeed, the statements cited above were not volun-
tarily made but arose in situations of conflict between individual recipients
)0 Andocides 2.11 (On his Return), d. Mriggs and l.ewis (1969) no.91.
,) IAndocidn 2.20.
.u Dinarchu~ t.43 (Against Dmrosthents).
n Demosthenes 19.114. 145 and 259·66 (De Falsa ugationt).
)4 lnmosthenrs 49.26 and passim (Against Timoth~s).
H A thousand medimni of wheat would have sufficed to feed 200 ptOple for an entire y~ar.
There is no way to estimate the amount of timber Anchines rt«ived, but it is worth pointing
out that one man was therrby enabled to provide for the needs of an entire army by mNns of
,~sou,us controlled by a singlt outsidn-friend.
3 Natural Produc.ts 83
and the community: some were accused of receiving these benefits, Others
boasted about sharing them out, but no one would freely admit to being a
recipient. Fear and desire to conform combined to make people underplay
their foreign connexions. Theophrastos' picture of one of his characters may
have been typical of the prevailing atmosphere:
Moreover, he (the Pretentious Man] may well say that he has no less than three
letters from Antipater [one of Alexander's successors] requesting his attendance
upon him in Macedonia, and albeit he is offered free export of timber he has refused
to go; he will not lay himself open to denunciation. 36
This tendency may indeed account in large part for the sparse literary
documentation. But there are important considerations which might shed a
different light on the phenomenon.
It is no mere coincidence that Macedonia is often associated with gifts of
timber, and the Pontic regions with gifts of grain. For it can hardly be
doubted that rulers rich in certain resources employed their surpluses in
exchange for what they themselves needed and in order to win over friends
and supporters. When the young man from the kingdom of Bosporus
(Isocrates' client in the Trapezitic.us) set out to see the world, he was
provided by his father with two ships of grain presumably intended both as
a primitive form of money and as a means for winning over friends. l7
Personal relations of dependence are most likely to flourish in situations in
which each party can provide benefits highly rewarding to the other at a low
cost to himself. It thus seems reasonable to assume that the surpluses of
autocratic rulers, coupled with the chronic shortages of grain and other
natural products in the Greek cities, provided an ideal context for the
creation and consolidation of such ties. Hints which might lend support to
this reasoning can be elicited from civic decrees.
Philippides the Athenian, a lengthy inscription informs US,38 secured
through intervention with King Lysimachos a gift of grain for famine-
stricken Athens amounting to 10,000 Attic medimni. The inscription, unlike
the literary passages, docs not say that Philippides had privately received
grain from Lysimachos. But one needs to credit Philippides, a part-time
courtier and part-time agent of Lysimachos,J9 with an unusual degree of
self-restraint in order to believe that he did not receive a share of the scarce
product. Inscriptions of this kind in fact shed light on the last phase of the
process - the sharing out of resources - referred to in the literary passages.
.16 l1trophrastos, Characttrs 23 ....

J7 lsocratrs 17.4.
~II 5).11.' 374.
,q ct. Shear (1978) 49.
We need to digress in order to form an idea of how the different phases are
interrelated.
The process of network building, the different aspects of which we see in
the sources, involved several stages. First, a relationship of trust was estab-
lished (or inherited) by a community member and an outsider in the fashion
described in Chapter 3. Secondly, a chain of givings and counter-givings
began. These included both symbolic gifts that were mere luxuries and
articles of real use-value. 4O The goods acted as a catalyst for the consolida-
tion of the bond. For each one of the partners, being differently situated in
the social structure and commanding access to different types of resources,
was in a position to supply what the other needed. Thus, modest gifts gave
way to large-scale co-operation, and the value of the shared resources
became an expression of the degree of confidence between the two men. The
outcome was the conversion of an initially moral relationship into an
economic partnership in which both parties had a vested interest.
The more resources the two parties commanded, the easier, and more
desirable, it became to extend the partnership. This could be done either by
merging the networks of ritualised friendship in which each one was sepa-
rately involved, or by recruiting new allies and dependants. Ritualised
friendship thus became a formula for bringing together a wide variety of
people of different social origin, and provided links in elaborate chains of
horizontal and vertical integration (made up, as we shall see, of kin, friends
and ritualised friends). The more resources a man commanded, the more
extensive his network became. Small wonder, therefore, that the most
frequently attested networks radiated from the three great centres of wealth
of Greek history - the courts of the Persian, Macedonian and Hellenistic
autocrats.
It is against this background that the honorific decrees should be consi-
dered. These are resolutions voted by the assemblies of different cities in
honour of individuals who interceded with a foreign ruler (or with the
ruler's subordinates) and thereby secured the provision of vital goods for a
city. Obliquely, these documents yield a rare glimpse into the operation and
power structure of friendship networks. The fragment of the network
revealed consists usually of three, and sometimes five (or more) men: the
40 For example: two silver baths and two tripods of gold given by Polybos of Th~ (Egypt)
to the Spartan Menelaos, and "a golden distaff and a silver basket with wheels to it and with
rims finishrd off with gold" (Odyssey 4.129-32>; the fabulous horse Boukephalas, given to
Alexander by a xtnos of the family, Demararos of Corinth (Diodorus Siculus 17.76; according
to Plutarch, Altxa"der 6, the horse was offered to Philip by Philoneikos the Thessalian); the
golden plane tf'« given, reportedly, by Darius to the father of Pythios the Lydian (Herodotus
7.27); the great many presents (dorta,) Isoc'ntn admitted to having received from Nikokles,
King of the Salaminians (Isocrates 15.40 (""'idosis)).
3 Natural Products 85
mover of the decree (a citizen of the honouring city); the honorand (either a
citizen or a foreigner); the man, or group of men, who directly controlled
the resource supplied; and finally, the ruler to whom the former was
subordinated, In cases where the honorand is credited with direct access to
the ruler himself, the chain is reduced to three links. Some typical examples
are listed in Table 1. 4 •
Table 1. Supply of Natura/I'roducts through Networks of Ritualised
Friendship
Rdrr~nce Mo\'erof Honorand(s) Subordin:ur Ruler(s)

oty, dare <kcrcr from ... of rul~r
5yll.' ]07 Unkno.....n Gorgos. Alexander

( I;1S0S. MlOniull
.3.16-.123 B.C.) (lasos)
OG/S4 Unknown Thrrsippos 'Satraps' Anripalrr,
[Nrsos. [Nesos?1 Kleitos
c..llS B.C.I
1(; II! 401 Unknown ... son of Important Unkno.....n
[Athrns. Mrtrodoros MacC'donians
323-319 B.C.) (Cyzicusl
S)·II. l 367 Epicharmos Zenon Ptolemy I
IAthens. [Unknown)
286/S B.C.I
(with Shrar
(1978)64
for thr datr)
IG 11 1 655 Simonidrs T.mo ... Audolron
[Athrns, [Athrns?1
285i4 B.C.I
(with Habichr
(1979) 52 for
thr datr)
5yll.' 374 Nikeratos Philippides Lysimachos
[Athrns. [Athrns
283/2 B.C.I
(with Habicht
(1979) 99 for
thr date)
SEC 28.60 Eucharrs Kallias Ptolemy I
[Athrns. IArhrn~)
270/69 B.C.)
IG XII.8.156 Phi ... Hippomrdon Ptolrmy III
ISamorhrace. (Sparr:!j
228-225 B.C.)
4' Thr drcrrrs arr nor numrrous. but thry form parr of a much largrr body of inscriptions
which display the sam~ pattern (th:H is. securing 3 'brnrf3etion' for a cit)' b)' interceding with a
ruler), but which I~v~ rhr narurr of thr services unm~nrioned. Thrr~ are good reasons to
bdirv~ rh:!t som~ of thr5C unspc.'cific in~riprions concral thr provision of similar goods. Srr
Herman (1980/1).
There is, in these decrees, a remarkable discrepancy between acts, ideolo-
gies and interests which is by no means apparent from the general tenor of
the texts. Take first the 'benefactions' - the acts for which the honorands
were awarded the honours. Gorgos and Minnion restored to the demos of
lasos a valuable fishing ground that had (presumably) been confiscated by
Alexander's philoi. Thersippos of Nesos, the honorand from Cyzicus whose
name is lost, and Zenon, Ptolemy's general, interceded with their superior
friends for the dispatch of grain. Hippomedon of Sparta, Ptolemy's general
in Thrace, had granted the right to export grain from the territories under
his sway:42 The case of Philippides and of Kallias of Athens is unusual since
each of them secured, through intervention with their ruler-friends, not
merely the supply - or, shall we say, the non-withholding - of grain, but its
provision, as a gift, in substantial quantities (10,000 and 20,000 medimni,
respectively).
Secondly, take the explanations given in the decrees themselves of the
motives of the benefactors. Philippides is said to have "spent much money
from his own private resources and rendered his account according to the
laws, and he has never said or done anything contrary to [the interests of)
the democracy". The other great benefactor of Athens, Kallias of Sphettos,
overtly an official in Ptolemy's service, "allowed his own property to be
confiscated in the oligarchy so as to act in no way in opposition either to the
laws or to the democracy which is the democracy of all Athenians".")
Finally, consider the interests. In remarkable contrast with the motives
ascribed to the benefactors, the operations involved in the provision of these
resources can, in my view, best be explained on three assumptions. First,
that these people formed part of extensive networks of friendship alliances,
from which all reaped substantial benefits. Secondly, that the service to the
city was, from their own point of view, but one more move in this private
exchange of benefits. And finally, that the benefits which one (or several of
them) hoped to acquire in return for the service done to the city were not, as
commonly thought, the honours recorded in the inscriptions (though these
might have been important), but the winning over, from within the city
population, of followers, allies and supporters. I shall now try to offer an
explanation of the first two propositions. A demonstration of the third must
wait until Section 5.5.
The first proposition is not difficult to demonstrate. The language of the
decrees suggests either friendly relationships or relationships of subordina-
42 Cf. Gauthirr (1979).

4.\ Philippides: SylJ.3 374, linn 48-9; Kallias: Shear (1978) 6, lines 80-2.
J Natural Products 87
tion between the honorand and the ruler controlling the resources. 44 It is
thus clear that the two men had material interests in common. On the other
hand, the nature of the connexion between the mover of the decree and the
honorand is not so clear from the surviving documentation. Apart from
accusations, the truthfulness of which it is impossible to test, the evidence is
practically lacking. This must either mean that there was no connexion at all
between the honorands and the movers of decrees - that, in other words, the
Greek city managed to suppress this form of 'corruption' - or else that it
was carried into practice in such a manner as not to leave a mark on the
historical record. Certainty is impossible, but several considerations tend to
favour the latter view. In democratic Athens, putting up motions for the
gratification of private interests would have been an offence. The people
involved would therefore have strong reasons to conceal their relationships.
The case of Archedemos, in Xenophon's words, "an excellent speaker and a
man of affairs, but poor", shows how a man could act on behalf of another
man within the city'S institutional machinery.4J Crito, the Athenian aristo·
crat, was plagued with actions by lower-class blackmailers. On the advice of
Socrates, he won over Archedemos with gifts and shows of courtesy.
Archedemos came to regard erito's house as a haven and treated him with
great respect. He soon found out that Crito's blackmailers were not without
blemish, brought 'counter-suits' against them, and compelled them to with-
draw their actions and pay compensation. Crito's many friends now begged
Archedemos to become their protector (phylax). Archedemos did so happi-
ly, "and so there was peace not only for Crito but for his friends as well". As
a result of this, Archedemos became one of Crito's philoi and was respected
by Crito's other philoi. It is not stated that Archedemos put up motions in
favour of his patron's xenoi,46 but it is a reasonable guess that he could have
done so and thus camouflaged the entire network. For an honorific decree
would not reveal any connexion between such a mover and the honorand.
A partial confirmation of this comes from the negative example of
Stratokles. When, under the rule of Demetrios Poliorketes, the political
institutions of Athens were reduced to nominal status, the movers of decrees
apparently did not think it necessary to conceal their relationships with the
honorands. In the period 307-301 B.C., Stratokles of Diomeia, a member of
the richest and most notable families from his deme, proposed and carried
through about twelve motions for different friends and proteges of
44 For the analysis of thr terminology, d. Hrrman (1980/1).

4S Xrnophon, Mttmorabilia 2.9.3-8; ser C:llhoun (1913) 48-9 and Connor (t 971) H n.l for
furthrr details on rhr Archrdemos affair.
46 We know at least of Criw's Thessalian x~oi from Plato, Crito 45c.
Demetrios. 4' No such consistent, blatant instances of 'corruption' can be
observed in the inscriptions from the period in which the assembly was
sovereign.48
For a demonstration of the second proposition, we go back to the
Archelaos-Andocides case. In return for the opportune supply of timber, the
city may have granted Andocides an honorific decree similar to those
included in our list. But from such a decree it would have been impossible to
deduce what emerges from the speech Andocides delivered before the Athe-
nian assembly: that Archelaos had mobilized his own resources simply in
order to render help to a xenos; and that Andocides supplied the timber to
the Athenian army with one purpose in mind only - to pave the way for his
reinstatement as a citizen. In other words, when once, by a rare coincidence,
the motives that lay behind a benefaction become accessible, they turn out
to be totally self-regarding. The inference must be that the mutual exchange
of benefits between ritualised friends might well be reflected in the sources
as a beneficial act in favour of a city.
In the context of ritualised friendship, the possession of grain, timber,
and other natural products was in some respects advantageous. and in
others not. On the one hand. some rulers commanded more of such pro-
ducts than they could consume, even to the extent of being able to supply
the needs of entire civic populations. On the other, the sheer bulk of these
products made them difficult to transport and their specific nature pre-
scribed a relatively narrow range of uses. Homeric heroes, like the young
man from the Bosporus. were probably always handicapped by these diffi-
culties. But within a century or two, a new means of exchange was to appear
on the scene - a device which was both easy to transport and could be put to
a multiplicity of uses: money.
4.4 VALUABLES
When Xenophon made the mythical Indian king send Cyrus the message: "I
wish to be your xenos. and I am sending you money, and if you need more.
send for it", he was inspired by one of the most common practices of
47 Thr decrttS a~ lisrrd in Osbomr (1981·3) Yo!.lI. 121 n.495. For the wulth and the
liturgic record of Stratoklt'1' family, ~ Davi" (1971) no. 129.18.
• 1 A good example of what might havr been thr pattern in this period concerns Demacks
and Hyperidn. In 3.38 B.C., Demadn managed to carry through an award of proxmy to a
close associate of Philip II (/G III 240). We do not know thr background for the grant, but
when Demacks further propowd to brstow a proxeny on Euthykrates, the Olynthian comman·
dfr who in 348 B.e. betrayed Olynthul to Philip {DemO$thenes 8.40 (On tnt Cht,sontstt);
19.343 (De Falsa ugat;o"I)~ Diodoru5 Siculus 16.53), he met with the violent opposition of
Hyperidn (F,agmmt 19).
4 Valuables 89
ritualised friendship.49 In all periods of Greek history, it is taken for granted

that ritualised friends would supply each other with money. The spirit in
which the act of giving was conceived was stereotypically rigid: one ought
to give generously, lavishly, and unconditionally. It was in particular this
aspect of the etiquette which captivated the imagination of writers accus-
tomed to more modest standards. Croesus of Lydia, like a character in a
fairy-tale, is said to have made the Athenian Alkmeon a gift of as much U
gold as he could carry away at one time on his person....~() Darius, as we

have seen, offered Syloson "boundless (apleton) gold and silver". 51 Xerxes,
in his famous letter to Pausanias, said to the Spartan King: "Let neither
night nor day keep you idle in the performance of your promises to me, nor
let them be hindered for want of gold or silver to spend". 52 And finally, of
Philip of Macedon it was said that he kept offering presents to the Theban
ambassadors Ubeginning with captives and the like, and ending with gold
and silver goblets",-B
Some of these statements are clearly fictitious, others arc probably ex-
aggerated. It is, however, hard to tell whether the exaggeration was on the
part of the actor or the story teller. For the pretence of giving away
'boundless' amounts of wealth could imply one of two things. The amount
of money one person is willing to bestow on another may be a sign of the
degree of trust he places in him. This is, surely, the message conveyed in the
passages listed above. But such a promise could also have a very different
meaning: that a person, the giver, is willing to recognise the power of the
recipient over him. The gift thus becomes a mark of submission, a form of
blackmail voluntarily paid - a tax, in its embryonic form. For example,
Pythios, the wealthy Lydian, declared his intention to contribute all his
fortune - Utwo thousand talents of silver, and of gold four million Daric
staters lacking seven thousand" - for Xerxes' war against Greece. uAIi this I
49 Xrnophon, C)'ropaedia 6.2.1. The phrasr srrangdy rchocs thr corrrspondrncr of rhr
petty rulrrs of rhr a",.;rnt Near E.ast of the S(cond millrnnium B.C. A frw rxtra~"ts from onr of
the T~II·d Amarna letters, wrirten by an Ass)'rian to an E.gyptian ruler, will suffice ro show this:
"To N:lphur ...• (grut king]. king o( F.gypt. my hrorhrr. speak! Thr words of Ashur-uballit,
king of (Assyrial, gre:lt king. ),our brnthrr. May it br well wirh you, your f:lmily. and your land.
... I hayr dispafchrd ro you :IS a pe:lcr offrring :I beautiful ro)·al ch:lrior (from among rhose)
rhat I (mysdf) driy~ and two whit~ horses I am in the procr~.. o( building a new palace. Send
mr rnough gold to derorat~ it pro~r1y If yOll are seriously disposed towards frit'nd~hip.
send much gold! It is all in thr (amily (lit. "It is your famil)''')1 Writ~ mr wh:lt yOll (your~Jf)
nrrd and it will be supplird ... (A.K. Grayson. Ass)',ian Roya/lnsc,iptions (Wirsbadcn, 1972)
It
yoU. 48·9).
(0 Herodotus 6.125. Thr fair ends wirh Croesus' doubling the amounr.
q IIrrodoru!l.1. 140.
n Thm:ydidl"S 1.129.3.
H Demosrhenes 19.139 (De ""a/sa Legalione). d. [)Iodorus Siculus 16.55.
freely give to you," he said. "For myself, I have a sufficient livelihood from
my slaves and my farms".S4 When Syennesis, King of Cilicia, was forced to
desert his former ally, Artaxerxes, and go over to Cyrus' side, "he gave
Cyrus a large sum of money for his army", clearly, as a mark of the status of
protege which he assumed. 55 It was, according to Xenophon, as a result of
numerous similar incidents of self-surrender (together with treasures and
cities) that Cyrus the younger acquired a following greater than any of his
contemporaries. 56 In this respect, again, the sums involved must have been
enormous. The Odrysian king Sitalkes, reports Thucydides, exacted from
his subjects a revenue of about four hundred talents of tribute (pharos) in
coin, as well as a similar amount in giftS (dora) in gold, silver and articles for
household use. n The latter must have come from lesser chieftains or
acephalous communities who surrendered themselves to Sitalkes'
authority. 58
Passages pertinent to transfers of money between ritualised friends are so
numerous that it would be pointless to enumerate them all. w Furthermore,
examples abound in which monetary transactions are not explicitly men-
tioned but which cannot be understood unless the transfer of substantial
sums is assumed. Therefore, rather than compiling a list, I shall try to
examine in a more detailed fashion some cases which may throw light on
other aspeCts of wealth circulation.
In Aristotle's The Athenian Constitution, it is stated of Peisistratos, the
Athenian tyrant, that after his second exile from Athens,
First he settled in the r~ion of the Thermaic Gulf at the place called Rhaecelus; from
there he proceeded to the district about Pangaeum, where he got money and hired
soldiers; then he went to Eretria. It was only in the eleventh year that he tried to
recover his rule by force, with the support of many others, in particular the Thebans,
lygdamis of Naxos, and the knights (h;pp~;s) who controlled the government at
Eretria. After winning the battle of Pallenis he occupied the city, deprived the people
of their arms, and this rime secured the tyranny firmly. Also he captured Naxos and
installed lygdamis as its ruler. 60
u HrrodofUs 7.28.
U Xenophon, Anabasis 1.2.27. Xc also Herodotus 3.13 for the sdf-impoKd taxes and gifts
paid as a mark of submission to CambYKs by the people of Libya, Cyrenr, and 8arca.
H, Xenophon, Anabasis 1.9.12.
J7 Thucydides 2.97.3.
JI For gifts as tax~s in the Hom~ric world. Sr~ Finl~ (1977) 96. For obligatory giftl (dora)
as distinct from taxn (phoroi, taRa;) in the Penian empire, d. Herodotus 3.89 and 97. For the
whole issue of th~ gift elemenr in me Penian taxation system, ~ Andreades (1933) 89·108; O.
Murray in Historia 15 (1966) 142·S6~ Hornblow~r (1982) 156ff.~ and Cook (19R.l) Ch. VIII.
~9 Thr incidents of 'rxtunal bribery' referred to in Sc-crion 4.2 would fall in rhi! catrgory
roo.
15.2-3, adapted from me translarion by P.J. Rhodn,
6() Tht Arhm;an Const;tution (Har-
mondsworth. 1984). Cf. Hrrodutus 1.61.
4 Valuables 91
Admittedly, this cannot be considered as a particularly accurate account
of Peisistratos' movements. 61 But what matters for our purposes is not so
much the details themselves as the expectations and assumptions according
to which the writers organised and interpreted the pieces of information
available to them. Checked against similar accounts derived from better
sources, the story fits perfectly into the pattern of ritualised friendship
relations.
On close inspection two, by familiar standards unusual, points stand out:
the variety of sources from which Peisistratos received assistance abroad/,l
and the high sums (in money and kind) he was given. The whole pattern of
action makes sense only if we posit, on the part of the people involved in
these networks, an expectation that the momentary sacrifices they made
would in the long run be reciprocated. Indeed, it can be stated almost as a
general rule that exchanges conducted within the framework of ritualised
friendship were marked by exceedingly long-term credit. In the case of
deposed rulers or citizens aspiring to become rulers, their helpers could hope
for a return only after, and if, the persons helped had been reinstated to
positions of power in their own social unit. Peisistratos, after his return to
power, fully repaid his debt to Lygdamis by seizing Naxos and appointing
Lygdamis ruler. In the case of Hippias, the enterprise failed: the massive
operation started by Darius to restore him to his position as tyrant in Athens
was frustrated by Greek resistance at Marathon. Xerxes bestowed monies
and goods first on Demaratos and afterwards on Pausanias and Themis-
lokles, clearly hoping that one day, when the right opportunity offered
itself, they would fulfil their promise and help him subjugate Greece. But
nowhere do we find a clearer detailed illustration of this long-term mechan-
ism than in the case of Klearchos and Cyrus the younger. Klearchos came to
Cyrus after he had been exiled from Sparta and condemned to death for
civic insubordination. Cyrus made him his xenos and gave him ten thousand
darics. Upon receiving this sum, Klearchos did not, in Xenophon·s words,
"turn his thoughts to comfortable idleness", but used it to recruit an army
and make war upon the Thracians (the enemies against whom he had
beforehand fought on behalf of Sparta). He defeated them in battle and
plundered them from time to time. The Hellespontine cities who had suf-
fered at the hands of the Thracians sent him money and support for his
troops in token of their gratitude. This idyll could probably have lasted for a
61 For the diffrrences bctwrcn Hrrodotus' and Arinodr's vrrsions of the srory, scr P.J.
Rhodr~, A C,()mm~ntary on the Amtotelian Athena;on Po/iteia (Oxford 19~ 1) 207-9.
61. To l1uace, Macedonia, Thessaly. Eretri:l, and Naxos (through Lygdamis) which appear
in this p3ssagr wr should add Argos: onr of Pri!iistr3tos' wivrs was an Argivr woman
(Hrrodorus 5.94).
long time, but, when Cyrus' summons came, Klearchos at once took the
soldiers and set out, in order that, as Xenophon makes him explain, "if he
(i.e. Cyrus) had need of me, I might give him aid in return for the benefits I
had received from him".6J
Neither the sheer coincidence of events nor the prospect of sharing
mutual benefits are adequate to explain such closely co-ordinated co-
operation between people who, we must bear in mind, as strangers, were
predisposed to treat each other as enemies. To understand such sustained,
complex, and long-term efforts, the operation of a further feature of ritual-
ised friendship must be assumed: the confidence it inspired that the other
side would be morally committed to reciprocate. This confidence acted as a
kind of primitive insurance system, increasing the likelihood that engage-
ments would be kept and obligations honoured. Coupled with the fabulous
wealth of the parties, it explains why, in the examples adduced above, we
never hear of 'payments' or 'loans'. Ritualised friends heading entire social
units bestowed favours on each other relying on the probability that when
the need arose the favours would be repaid. We need to go down one or two
steps in the social ladder to encounter transactions of a different type.
In the course of the year 362 B.C., Apollodoros the Athenian, son of
Pasion, the wealthy banker, discharged one of the most exacting duties that
the community could impose on its rich members. Apollodoros had the
command of an Athenian naval vessel for an entire year and bore the cost of
the operation out of his own pocket. This specific trierarchy, as it was
called, was beset with particular misfortune. Apollodoros, however (if we
are to believe the truth of what he said), not only succeeded in extricating
himself from a series of predicaments, but even managed, at great personal
cost and considerable self-sacrifice, to accomplish an exemplary term of
service. His triumph was only marred when at the end of his term Polykles,
his replacement, would not relieve him of his command. As a result, Apollo-
doros was compelled to bear the cost of an additional four months service -
a burden for which he seems to have been unprepared. Several attempts at
raising money failed, and the sailors kept asking for payment. However, in
this emergency, Apollodoros managed to borrow from two of his father's
xeno; in Tenedos and pay the sailors their provision money. The reasons
Apollodoros gave for his success are revealing:
... for on account of my being Pasion's son, and the fact that he was connected by ties
of xt"ia with many, and was trusted (p;steutht"a;) throughout the Greek world, I
had no difficulty in borrowing money wherever I needed it.6'4
4) Xmophon, Anabasis 1.3.3, d. 1.1.9 and 2.6.4-5.

64 Drmosrhenes 50.56 (Aga;nst Polycles).
4 Valuables 93
We do not know the specific details of Apollodoros' loan transactions,

but the examination of analogous cases permits us to establish the range of
possibilities.
It seems that loans within the context of ritualised friendship tended to be
without interest and without security. The reason for this was that loans
constituted but one item in the circulation of favours and counter-favours
that tied the partners together. Repayment was not therefore subject to
strict accountability, and could be made not only in cash but also in gifts or
opportune services. Earlier in the course of the same trierarchy, Apollodoros
sent Euktemon, his pentecontarch, to Lampsacus, giving him money and
letters to the xenoi of his father, bidding them, evidently, to help Euktemon
hire soldiers. It is not difficult to see in this a partial repayment of a loan that
Pasion made the xenoi from l.ampsacus.6~ Not even in the case of Nikostra-
tos (Nikostratos claimed that he had borrowed from xenoi in order to
ransom himself from captivity, and that it was stipulated in the agreement
(syngraphai) that he should repay the loan within thirty days or be indebted
for double the amount) do the harsh conditions turn the transaction into an
economically profitable business. The clause was merely intended as a
safeguard for lenders either short of cash or having a low opinion of
Nikostratos· credit. If it had been common for xeno; to lend at interest, it is
hard to see how Nikostratos could have failed to add this detail to his
already long list of misfortunes. 66 Indeed, professional loans and loans
carried out within the context of xenia are sometimes conceived of as
antithetical notions. A good illustration of this comes from an incident the
evidential value of which derives from a statement not self-consciously
made. Speaking before the Athenian jury, the young man from the kingdom
of Bosporus expressed his indignation at having had to endure without
protest that Hippolaidas, a xenos and epitedeios, borrow from Pasion the
banker. 67 The argument derived its force from the implicit understanding of
the jury that if the young man had money, he would have made Hippola·idas
a friendly loan. In fact, this was merely an effective roundabout way of
saying that he was totally impoverished.
Loans between xenoi thus conformed in some respects to the wider
category of friendly loans, the underlying ethical principles of which were
aptly summarised in the Problems attributed to Aristotle: "Where a loan is
involved, there is no friend; for if a man is a friend he does not lend hut
~~ Drmosrhrnn 50.18.
..t> I~m()srhrnr!il 53.10 (Against Ni(ostralus). for Nlkosrraros' mrrhod of raising Ihr
monry, ~r finley (1952) 83(f. 3nd Millrn (1983) 48.
,,-:0 lsocrarrs 17.38 (T,apeziliws).
gives" .68 Professional money lenders could lend at interest to strangers:

many of Pasion's clients were non-citizens. Some citizens could occasionally
lend at interest to other citizens who were friends or relatives. 69 But the
majority of loan transactions in fourth-century Athens - and hence, presum-
ably, in the whole of the ancient world - involved interest-free loans. The
reason for this was that the partners to these transactions were tied by
kinship, friendship, or ritualised friendship, and philia was the underlying
principle of this primitive credit system. 70
In more general terms, however, the pattern of financial aid between
xenoi was subject to pressures and risks somewhat different from those
exercised on friends belonging to the same social unit. For xenoi could help
each other in ways in which philoi could not. The most significant service, in
this respect, was that of acting as keeper of one's property abroad. 71 For
fear of revolutions, invasions and punishments at home, and from a desire
to establish a base abroad, wealthy individuals must frequently have re-
sorted to this device. 1l Two cases may be considered in some detail.
Some Siphnian aristocrats living in Aegina had at the beginning of the
fourth century transferred a large portion of their pro~rty to Paros to ~
kept there by the xenoi of one of them (the speaker in Isocrates' Aegineti-
cus). The reason for this move emerges from an indirect comment: "we
thought that the island was by far the safest".1J This, however, proved to be
I-
a serious misjudgement, for a foreign invasion causing general upheaval (the >-
0..
details of which are unclear) took place precisely in Paros. When all believed o
that their property was lost, the speaker sailed to Paros by night and got his U
friends' money out at the risk of his life.'"
It, IArisrotlel, Prob/mas 29.2, 950a28 with Fmley (1952) 85.
" lsaeus 11.42·3 (On tht F..statt of Hagnias): Arschines 1.97 (Against Timarchus): De-
mosthenes 27.9 (Against Aphobos I).
70 I borrow the rypology and thr rxamples of citizens Irnding at internt from Millen
(1983), a study based on almost nine hund~ loan transactions.
71 Instances of individuals tran5frrring or keeping a share of or thr wholr of theIr property
abroad are tOO numerous for all to br lis~. It will br enough to refer to Themisrokles, who
had monies <kpositrd with friends in Argos (Thucydides 1.138.3); to the spcakrr in Lysias' 0"
th~ Propn-ry of Aristopharrts. who had to admit, contrary to his inr~est5, that Aristophanes
had property in Cyprus (Lysias 19.40); to Konon, who kept 40 talents in realisablr assets also
in Cyprus (Lysias 19. 36, d. Davies (1971) p.508). The whole plot in Euripides' Htcuba
revolves around the murder of a xenos by his host for the sake of the money hr brought for
safekttping.
71 a. Isocrates' tactful phra~logy in describing thr motives of some Greek Ira/o; Ir'agamo;
in setding in Cyprus: "they considered Evagoras' rulr less burdensome and more equitablr man
that of their own governments at home" (Isocrates 19.51 (Atgintt;cus).
7J Isocrates 19.18.
74 Although it cannot br rulrd out that he paid the Parians for thr scrvicr, it seems on
general grounds more Iikdy that mr keeping of thr property was dOM on a frirndly basis - In
fact, as a favour to thr speakrr himself.
4 Valuables 95
Another instance emerges from Isocrates' Trapeziticus. The young man

from the kingdom of Bosporus, son of a powerful official in the king's
service, came to Athens and deposited part of his money at Pasion's bank. In
the meantime, however, his father back home was suspected of plotting
against the throne and arrested. Orders were sent from the king to the
citizens of Pontus residing in Athens to bring the young man home and take
possession of his property. In this predicament, the young man appealed to
Pasion for help. 7.~ To prevent the confiscation, the two men devised an
ingenious plot. They would surrender to Satyros' agents the money whose
existence was publicly known, but would deny the existence of the funds
deposited with Pasion. Moreover, they would try to create the impression
that as a result of the surrender of the money the young man had become
totally impoverished. 76 The young man, as he later explained, proceeded to
act in accordance with the terms of their agreement:
...1, in the presence of many listeners, had denied that I possessed anything. and
everybody had seen that money waS being demanded of me and that I was acknow-
ledging that I was indebted to others aI50. 17
Pasion took advantage of the golden opportunity and after a while flatly
denied the existence of the money deposited with him. It was this denial that
led to the suit for which Isocrates composed the speech on the young man's
behalf.
For the purposes of our analysis it is immaterial that Pasion, unlike the
Parian xenoi, did not restore the property entrusted to him. Even real
kinship ties do not provide a perfect safeguard against breach of obligations.
The significant point is that the mores of ritualised friendship could be
believed (first by the young man himself, then by the Athenian jury) to
provide sufficient assurance that obligations between strangers would be
honoured even in the absence of contractual guarantees (documents, witnes-
ses, securities). The context of ritualised friendship thus makes it easier to
understand how long-term. complex economic transactions could take
place between strangers before contractual devices and enforcement agen-
cies (such as those provided by the polis) came into existence. This point
7~ Alrhough ir is nor srarrd rxplicirly rh3r rhe two w~rr xeno;, rhr phras«>log}' used ro
describe rhrir rdarionships luv~s little room for doubr: ..... for I was on such inrim3r~ rrrms
wirh him rhar I had thr grearnr confidence in him, nor only in matters of monry, bur In
rve7thing ds~ as w~II" (17.6).
7 t 7.8. This pretence' berra)'s rh~ fal"t that rh~ financial dralings betw~n rhr two mrn
followed thr pattrm of frirndly, inrrrt$r·fr~ loans, in spirr of Pasion's being a professional
bankrr.
77 17.8-9.
, • I .: , • ':,~. : : '. ~ .' • , ' : '. ' :

cannot, of course, be illustrated by contemporary evidence. 78 But the story
about the Thesprotian king Pheidon, showing the Cretan Beggar (Odysseus
in disguise) the great wealth that Odysseus, his xenos, gathered for himself
(Ubronze and gold and iron that men had long toiled over - enough to
sustain ten whole generations after him") and entrusted to Pheidon for
keeping, however exaggerated, points to the possibility of such transactions
in the Homeric age. 79 And scattered references in Classical sources to xenoi
acting as joint lenders 80 or to a xenos bequeathing his property and profes-
sional secrets (in divination) to his panner 81 might give some idea of the
range of transactions rendered possible by this bond of solidarity.
Valuables circulating between ritualised friends became thoroughly in-
tertwined with the workings of the civic system. Pasion's son, Apollodoros,
as we have seen, was able to complete his trierarchy by virtue of monies
borrowed from his father's xeno;.82 The case of Aristophanes, the son of
Nikophemos, of Athens, suggests a more intricate variation of the same
entanglement. The following episode can be reconstructed from Lysias'
19th Oration. In the year 390/89 B.C., envoys from Cyprus arrived in
Athens to ask for assistance against the Persians. The assembly voted them a
small fleet. But the envoys still lacked money to purchase arms and to hire
light infantry and personnel for the ships. It is at this juncture that Aris-
tophanes stepped in. He gave all his fortunes and all the monies he could
borrow to the city to make up the missing sums - five talents, if we are to
believe the speaker. 8J Aristophanes' motives are stated in a fashion which
suggests that there was nothing exceptional about this donation:
What man, think you, who was ambitious of glory, and was re~iving lenets from
his. father that told him he would lack for nothing in Cyprus, and had been elected
71 The evidential value of two stories which look back to the mythical past (that of Glaukos,
the righteoUs Spanan, and the Milesian (Herodotus 6.86££.), and of Polycharn of Mnsmia
and Euaiphnos of Sparta (Pausanias 4.4.4££. and Diodorus Siculus 8.7)) is impaired by a strong
moral ising tone and by the fact that they probably reflect the outlook nOf of a single age but of
successive periods, each one ~modellin8 the story according to its own i<kals. It shoul,J .~I~\J be
pointed out that the Glaukos story is among the rare cases in which the word X~"Os is so
ambiguously usN that it seems impossible to tell whether it rders to 'guest-friend· or 'stranger'.
In any case, the relationship between Glaukos and the Milesian do not imply a series of
exchanges of some duration: the Milesian entrusted his money with Glaukos on account of his
reputation for honesty, and his descendants claimed it back.
19 Odyssey 14.321-6. The terms ~ELVloaL "'b~ ~iloaL might better be rendered here as "he
(Pbeidon) made him [Odysseus I his xmos and philos"rather than "he. welcomed him and
entertained him".
80 Demosthenes 35.8-9 (Agamst IAcritus), with Ste. Croix (1974) and Millett (1983) for the
context of the bottomry loan involved.
I' (socrates 19.5 (A~gin~ticus).
82 Demosthenes 50.56 (Agai"st Po/yd~s).
83 lysias 19.43 (0" th~ Property of Ar;stopha,,~s). For Aristophancs' considerable property
and financial <kalings, sec Davies (1971) no. 5951(C).
,j . • . .; :, ~ I: :,
5 Troops 97
ambassador (prtsbtutts) and was about to sail to Evagoras, would have Idt behind
anything that he possessed, and not have rather gratified that ruler by supplying
everything he could, with a view to recover more?fl4
The details might be confused and exaggerated, but the nexus of thought is
clear: Aristophanes was, presumably, Evagoras' xenos - his father lived in
Cyprus and had property there. When it was found out that the Athenians
lacked the funds to provide the fleet with arms and soldiers, Aristophanes
ensured that he was elected ambassador to Evagoras by making this con-
siderable outlay. For this generous act he was no doubt honoured - the
speech is replete with references to his 'love of honour' - and his credit must
have riscn. However, the outlay clearly did not fall within the modern
economic category of 'conspicuous expenditure': Aristophanes was not
trading his wealth for prestige. For, as stated in the passage, "he supplied
everything he could, with a view to recovering more". Unmistakably, he
hoped to be more than fully repaid either directly with counter-gifts or
economic concessions by Evagoras, or indirectly, out of the military opera-
tions which he would conduct together with him, making use of the Athe-
nian fleet. U A bizarre juxtaposition of the private and public sector", as
Veyne would have it, but, contrary to Veyne's assumptions, the financing of
the fleet did not rely ultimately on property owned by the benefactor inside
the city but on booty wrested, with the aid of xenoi, from outside. 8~
Quantification is impossible, but some similar cases, St. and the examples
considered in Section 4.3, suggest that the practice was by no means
unusual. Briefly, there must have been a fair degrtt of coincidence between
those involved in xenia networks and those described by Davies as members
of the liturgical c1ass. 87
4.5 TROOPS
In his book on Greek mercenary soldiers, Parke made the correct observa-
tion that Cyrus' raising of the Ten Thousand, "was distinguished only by
114 Lysias 19.18.
8.S Verne (1976) passim. For a criticism of his virws, see F. Millar in TL'; 24 March (19i8)
356, and J. Andreau, P. Schmitt, A. Schnapp (1977).
86 l~mosthenes 5.8 (Orr Iht Ptact) (Neoptolrmos, an agrnt of Philip In Arhens. was
accu~d of having gonr to Macedonia "to collet.-r sums owing him there which he might spend
on liturgies hereto); Theophrastos, CharacltrS 23.5 (rhr 3ssumption th3t thr recipirnt of thr
right to uport timber from Macedonia is a great ~nda~:tor); $)'1/.) 374. lill('$ 40-51 (rhr
contributions made by Philipptdn, Lysimachos' frirnd, to public scn·icn). For the so-called
ob·horrornn donations - tM convrnrion whr~by wealthy citizms drC'ted to ma~srraciC'S
contributed to mr public snvices under thrir chargr - s« Jonr~ (1940) 167-8. Vrynr (1976)
214-16, and Srr. Croix (1981) JO.~-6.
1:- Davies (1971) Introduction.
the magnitude of the forces which he employed and the height of ambition
at which he aimed". The precedent which, according to Parke, rendered the
undertaking unremarkable was the employment throughout the fifth cen-
tury of Greek mercenaries by Persian satraps to serve as bodyguards, fulfil
garrison duties and occasionally carry through a revolt against the Great
King. 88 But, once we ignore for a moment the Greek-Persian context and
focus our attention on the structure and composition of Cyrus· army, it
becomes apparent that Parke's observation has a much wider validity. In
fact, Cyrus adhered to a trend which on a lesser scale and with minor
variations makes its appearance in all periods of Greek history. The fun-
damental pattern is accurately mirrored in mythical tales: a powerful ruler
(or nobleman) calling his guest-friend(s) to arms to carry out a task for
which his own forces are inadequate. 89 What was thereby effected was the
pooling of military power and the temporary extension of the parties'
fighting potential. A story from Herodotus might serve here as a useful
paradigm. Croesus, King of Lydia, being hard-pressed by Cyrus of Persia,
and realising the inadequacy of his own forces, called upon his friends and
allies - Amasis, King of Egypt, Labynetos, King of Babylon, and the whole
of the Spartan community - to fulfil their pledges and send troops.9O
Xenophon's Anabasis yields a detailed picture of how such merging of
forces was carried out.
The army with which Cyrus the younger hoped in 401 B.C. to overthrow
the Persian throne consisted of about thirteen thousand soldiers raised in
different parts of the Greek world. The soldiers were probably first trained
by the recruiting captains (some in actual skirmishes and looting operations)
and then brought to Sardis, whence the march was to start. The principles in
accordance with which the entire host was organised are extremely signifi-
cant for our purposes. For the contingents each commander brought with
him were not broken up, regrouped and placed under a hierarchic chain of
command: Cyrus did not bring with him an alternative skeleton of staff
officers. Instead, as Table 2 shows, they retained their recruiting comman-
ders, original composition and surprisingly unequal numbers of men. Con-
sequently, the vast host, in Roy's words, "formed not one army, but several
Parke (1933) 23.
I.
.. For example. th~ Ic-gend of the 'Scven against Thd>n' (Sophodt'S, Oedil"" iJl C%nus
1300ff.); that of Rhesus (llUu1 10A.H, 495 and the play by Euripides); or that of Pririthoos and
Thneus (PlutlrC'h, Thnft4. 30ff.).
90 H~rodotus l.n. Croesus p~umably had treaties of xenia Ira; S),,"mach,a with Amasis
and labyn~tos in the same way as he had with Sparta (1.69). Cf. Euripides, Rhesus 335. where
Rhesus, rul~r of Thrac~, is said to have come morc as a xenos than a J)""machos, and th~
aphorism, quoted by Herodotus, (hat "annin are th~ most proper gift (doran) in Penia"
(9.109).
5 Troops 99
separate armies",91 and the captains were not so much officers as semi-
independent chieftains. The link which held the whole army together was
the relationship between the captains and Cyrus.
Table 2. Cyrus' xcnoi
Commanckr Relationship Contingents Remarks
Xenias not mentiontd 4000 hoplitts Xenias deserted at Myriandus

(Arcadia)
Klearchos xenos I 000 hoplir~ The ~hasts were Thracians;
(Sparta) 800pehasts rhe archers wert Cr("fans~ the
200 archers cavalry were Thracian
40ca\'alry
Proxl"tlo~ xmos 1.500 hophtes
(Boeotia] 500 gymnetae
Menon not mentioned ] 000 hoplitn The pelrasN we-re Dolopt"S,
[TMssaly) .500 pehasts Aunians and Ol,'nthians
Sophainctos xenos 1000 hoplirC'S
ISrymphalus)
Agias not mentiontd 1000 hoplir~
(Arcadia)
Sokratcs xmos 500hoplit~
(Achaea)
Chcirisophos not menrioned 700 hoplircs Cheinsophos jomcd Cyrus '"
(SP3rt2] Cilieia
Pasion nor menrioned .'00 hoplircs Pasion dCSC'rted at Myriandus
lMC'gara) .100 pdtasts
Sosis not mentioned 300 hoplites
[Syracu~)
Aristippos Aristippos ~nt Menon In his

[Thnsaly) stead
Klcanor not mentiontd Unknown
(Arcadia)
Pythagoras not mentioned 35 ships
[Sparta)
Tamos not menrioned Over 25 ships For T amos, S« Gomme et al.
(Egypt) (1981) voJ. V, 74
Not only were the captains probably all Cyrus' xeno; - we may legiti-
mately assume that those not designated so in the Anabasis were xeno; too -
but they had become his xeno; prior to the start of the march. We encounter
Xenias from Parrhasia (Arcadia) in Cyrus' service as early as 405 B.C.
acting as the commander of the three hundred Greek hoplites who accom-
panied him to Susa. Xenias must have been the beneficiary of some sort of
ql Roy (1967) 292.

estate in Tral1es (Caria), where he kept his wife and children. To Sardis, he
came at the call of Cyrus, bringing with him 4000 hoplites from the Ionian
cities, where he had lately served as the supreme commander of the
garrison. 92
We have already traced in a different context the exchanges between
Cyrus and the commander of the second largest contingent, the Spartan
exile Klearchos. 93 It need only be added here that Klearchos was the only
Greek to whom Cyrus had confided the real aim of the campaign. He was
also the only one allowed to attend the trial of Orontas. As a result of the
numerous favours shown him by Cyrus, in the course of time Klearchos
became a sort of first among equals.
The other prominent Spartan, Cheirisophos, one of the 'equals' in his
state, arrived with his 700 hoplites on board 35 ships led by yet another
countryman, Pythagoras. It may be that this force was sent as part of the
official aid Sparta gave Cyrus to oust his brother Artaxerxes. Yet it has to be
pointed out that Cheirisophos is said to have responded to Cyrus' private
summons (J.l£tan£r.utto~ uno KUQOU)94 precisely as did some other of
Cyrus' xeno;; and that these hoplites were not Spartan citizens but probably
Peloponnesian mercenaries. 95 Thus, it is by no means clear whether the two
Spartiates seized the opportunity of the official support for Cyrus to come
privately to his aid, or, vice versa, secured official aid as a favour to a xenos.
We are left uninformed about the precise details of the relationship
between Cyrus and the Theban Proxenos. But both the fact that, once called
upon, he came immediately bringing with him two thousand soldiers, and
his reference to Cyrus, his xenos, as a sort of nourishing fatherland
(patr;s) , 96 indicate that exchanges of considerable favours took place
between the two men even before the campaign.
Pasion, the Megarian, and Sokrates, the Achaean, had both assisted
Cyrus in the siege he mounted on Miletus as a diversion just before the start
of the march. Pasion, like Xenias, must have had an estate in Tulles. Sosis
the Syracusan is otherwise unknown, and so are Sophainetos of Stymphalus
and Kleanor of Arcadia. But the exchanges between Cyrus and his Thessa-
Han xenos, Aristippos, assume, interestingly enough, the pattern of reciproc-
ity observed in connexion with the Peisistratids. Aristippos, hard pressed by
his opponents at home, came to Cyrus and asked him for two thousand
mercenaries and three months' pay. Cyrus gave him, most generously, about
ql X~nophon. A"abasis 1.1.2; 1.4.8; 1.2.3~ 1.2.1.
" Section 4.4.
".. X~nophon. Anabasis 1.4.3.
'IS Roy (1967) 300; Cartledge (1979) 272.
96 Xenophon, Anabasis 1.2.3; 3.1.4.
, • I .: , • ,.,~, : : '. ~ .',.; ,
5 Troops 101
four thousand mercenaries and six months' pay.91 When Cyrus thought the
time had come to begin the march, he sent word to Aristippos to effect a
reconciliation at home and send him the army which he had. Aristippos, for
reasons unknown, did not come personally but sent Menon in his stead with
one thousand hoplites and five hundred peltasts only. We hear of no
expression of dissatisfaction on the part of Cyrus for this seemingly unequal
. .
reciprocation.
What thus emerges with sufficient clarity from these scattered remarks is,
first, that the history of the relationships between these men and Cyrus
followed the pattern of encounter and initiation described in Chapter 3. The
fact that most came from different social units suggests that, if the rela-
tionship was not inherited, the pact of xenia resulted either from an euerge-
s;a or, more likely, from an Cintroduction' by a common xenos (d. Section
3.2, pp.47ff.). Secondly, the bond was so strong, and the prospects of profit
so good, that when the call to arms reached them they at once abandoned
what they. had been engaged in and responded promptly - without even
knowing, it must be stressed, the real aim of the expedition. The attempt to
overthrow the Persian throne appears therefore as just another episode in
the course of the exchange of benefits between xeno;.
This basic practice of merging forces commanded or raised by xeno;
crops up with slight variations in all periods of Greek history. Some out-
standing examples will be considered here.
For the Homeric period, it is true, "we are left in rather complete
darkness about the way the Achaean army was put together". 98 But this
refers mainly to the recruitment of insiders. Agamemnon, when calling upon
his guest-friend, the nobleman Amphimedon, to help him urge Odysseus to
go to Ilion, was in fact following a common practice of recruitment among
outsiders. Amphimcdon, who did not himself command sufficient numbers
of soldiers, provided an avenue of approach to the more formidable Odys-
seus with whom Agamemnon had, so far, no formal ties. 99 Such a procedure
will suffice to explain one aspect of the composition of the Achaean and
Trojan armies - made up, as they were, of contingents headed by captains
bound together by ~rsonal ties of attachment.
The earliest non-literary evidence comes from the year 593/2 B.C. The
relevant lines of the inscription, scratched on the colossal statue of Rameses
II in Nubia, are as follows:
When King Psammetichos came to Elephantine, tho~ who sailed with Psamme-
9" Cf. Roy (1967) 298 for the textual problem! involved in this passage.
" Finley (1977) 102-3.
99 Odyssty 24.115ff.
,j' ..
. ,, :,
.' .,:
102 The C;rculation of Resources
richos son of Theokles wrote this; and they came above Kerkis as far as the river
allowed; and Potasimto had command of those of foreign speech and Amasis of the
Egyptians. 100
Psammetichos son of Theokles cannot but be the son of a Greek who

became a xenos of Psammetichos I, and, in conformity with the custom of
name exchange, called his son after the Egyptian ruler. Psammetichos II, the
present Pharaoh, now availed himself of the services of his hereditary xenos
in order to increase the levy for his Nubian expedition. The Greek Psamme-
tichos had brought with him his own ritualised friends - the individuals
who had signed their names on the statue below: a man from Teos, two
from lalysos, another from Colophon, and two of unknown ethnic origin.
In all probability, these men had each commanded a small force and
regarded themselves as peers. IOI Thus it becomes fairly certain that these
Greek men were not, as commonly thought, 'mercenaries in Egyptian ser-
vice', but aristocrats called upon to fulfil their duties as xeno; - a demand
which in view of the prospects of plunder must have appeared by no means
unattractive.
The same underlying principle can be detected in Darius' Skythian ex-
pedition. The Greek generals who served as heads of various contingents
were the tyrants of the Greek cities in Asia Minor: Miltiades the Athenian,
now tyrant of Chersonesc; Histiaios, tyrant of Miletus; Daphnis, tyrant of
Abydus; Hippoklos, tyrant of Lampsacus; Herophantos, tyrant of Parium;
Metrodoros, tyrant of Proconnesus; Aristagoras, tyrant of Cyzicus; Ariston,
tyrant of Byzantium; Strattis, tyrant of Chios; Aiakes, tyrant of Samos;
Laodamas, tyrant of Phocaea; and finally, Aristagoras, tyrant of Cyme. 102
The difference with respect to the former cases is that the voluntary bond of
xenia is here reinforced by formalised relationships of subordination and
superordination: the one-time xeno; had become Darius' clients and, at the
same time, the rulers of the Greek cities.
A final, eloquent example is provided by the commanders of the forces
under Ptolemy IV Philopator at the battle of Raphia (218 B.C.):
Eurylochos of Magnesia commanded a body of about thrtt thousand men known as
100 Translartd by Meigs and Uwis (1969) no. 7. For 593/2 B.C., ratMr than 591, as the
date of the inscription, SC'e J.D. Ray in ProcuJings of tht CAmbriJgt Philological Soc.tty
n.s.28 (1982) 85.
101 1M description of their position with respect to Psammetichos - "those who sailed with
Psammetichos son of lMoklts" - contrasts sharply with the phrase used for the rdationships
betw~ the other leaders and their forces: "Potasimto led those of foreign speech and Amasis
the Egyptians". The use of the abbreviated hypocoristic name in the sixth signature ("Pa(m)bis
of Colophon with Psammata", instead of the formal "PsamrMr1chos"), poin" to relationships
of dose familiarity between the two men, d. A. Ikrnard and O. Masson in RtvW des il"dts
grtcq"es 70 (1957) 20.
102 Herodotus 4.1.38. S« funner Hofstetter (1978).
, • I .: , • ,.,~, : : '. ~ .',.; ,
5 Troops 103
the Royal Guard; Sokrates the Boeotian had under him two thousand peltasts;
Phoxidas the Achaean, Ptolemy the son of Thrascas, and Andromachos of Aspendus
exercised together in one body the phalanx and the Greek mercenaries, the phalanx
twenty-five thousand strong ~ing under the command of Andromachos and
Ptolemy, and the mercenaries, num~ring eight thousand, under that of Phoxidas.
Polykrates (of Argos) undertook the training of the cavalry of the guard, about seven
hundred strong, and the Libyan and native Egyptian ho"e; all of whom, numbering
about three thousand, were under his command. It was Echekrates the Thessalian
who trained most admirably the cavalry from Greece and all the mercenary cavalry,
and thus rendered most signal service in the battle itself, and Knopias of Allaria too
was second to none in the attention he paid to the force under him composed of
three thousand Cretans, one thousand being Neocretans whom he placed under the
command of Philon of Cnossus. They also armed in the Macedonian fashion three
thousand Libyans under the command of Ammonios of 8arce. The total native
Egyptian force consisted of about twenty thousand heavily armed men, and was
commanded by Sosihios, and they had also collected a force of Thracians and Gauls,
about four thousand of them from among settlers in Egypt and their descendants,
and two thousand lately raised elsewhere. These were commanded by Dionysios the
Thracian. 10 .1
This morley group, in other words, consisted of ritualised friends who,
like Darius' client-rulers, had been absorbed into Ptolemy's permanent
entourage - the philo; circle - and of some ritualised friends from abroad
responding to the call to arms. To the former category belonged, for
example, Sosibios and Ptolemy son of Thraseas, a man whom we shall soon
encounter as a recipient of a landed estate in Antiochos' service; to the
latter, Andromachos of Aspendos and Polykrates of Argos, two men who,
in Polybius' words,
had recently arrived from Greece and in whom the spirit of Hellenic martial ardour
and fertility of resource was still fresh, while at the same time they were disting-
uished by their origin and by their wealth, and Polykrates more especially by the
antiquity of his family and the reputation as an athlete of his father Mnasiades. 104
Andromachos and Polykrates were thus not homeless mercenary comman-

ders, but res~etable citizens of their cities.
The inevitable question arises: Why were these generals strangers of such
diverse origin?
The answer seems to be that these men were possessed of a quality which
could not be found in plenty within any single social unit: the ability to
recruit and command armies. It was this attribute, combined with their
wealth and power, which made them such invaluable assets to their allies.
Not only could they lead their own countrymen, but they were capable of
10) Polybius 5.64. For further details on the generals, ~ Walbank (1957-82) YoU, 589-92.
104 Polybius 5.64.5-7.
. ,, :,
.' .,:
, • I .: .,: :~, : : '. ~ .',.; ,

commanding troops raised outside the social groups to which they be-
longed. Klearchos of Sparta, it will be remembered, recruited Thracians and
Cretans. Menon the Thessalian brought with him Dolopes, Aeanians and
Olynthians. In the whole of Cyrus' army, more than half were Arcadian and
Achaean hoplites, but there were only three Arcadian captains and one
Achaean, the rest being drawn from almost every part of the Greek
world. los Furthermore, men of this stamp were capable not only of recruit-
ing and commanding professional soldiers, but of mobilising the manpower
and resources of entire social units. The story of Histiaios of Miletus and
Darius might sound naive and exaggerated, but it in fact reflects the logic of
a normal practice. Darius gave Histiaios an area in Thrace in return for an
euerges;a. Histiaios started fortifying the place at once. Darius' general,
Megabyzos, then made a mild criticism of his royal master's careless gener-
osity:
Sire, ... you have given a clever and cunning Greek a city to build in Thrace, where
are forests in plenty for ship-building, and much wood for oars, and mines of silver,
and much people both Greek and foreign dwelling around, who when they have a
champion (prostates) to lead them will do all his behests by day or by night. 106
Heroes, prostatai, stratego;, as they were called, these army leaders seem
to have been drawn almost exclusively from a narrow aristocracy of war-
riors defined by upbringing, training, a certain aristocratic style, a fighting
ethos, and, most important, wealth.
A further variation of the pattern may be observed in cases where a
citizen availed himself of the services of a xenos abroad in order to promote
public interests at home - a practice which, to judge by the number of
scattered references, was not uncommon. We find, for example, the Spartan
general Brasidas in 424 B.C. commanding 1500 Thracian mercenaries. IO? In
428 B.C., the Athenian Asopios is said to have raised an army from all the
Acarnanians. 108 Cyrus' veterans were recruited by the Spartan Thibron, a
general who, in 399 B.C., "gathered troops from the Greek cities in the
mainland". 109 These scraps of information do not tell us precisely how
such forces were put together, nor is it claimed here that it could only have
been done with the aid of ::teno;. What can be suggested is that these are the
typical situations in which xeno; would be of help. They could supplement a
101 Cf. Parke (1933) 28.

106 Herodotus S.23. This is again a permanent topos. Polybius. for example, said that the
Palestinian cities of Skythopolis and Philoteria could supply armies with .11 their necessities OUf
of their ,hor" (5.70.5).
107 Thucydides 5.6.4.
101 Thucydides 3.7....
109 Xcnophon, HtUm;C4 3.1.5.
,j' ..
. ,, :,
.' .,:
5 Troops 105
citizen-general's forces with forces of their own; they could supply food,
payment to the troops, information, connexions, guidance, and so on. A
passagc from Thucydides gives what could bc a magnificd version of the
pattern. In 413 B.C., the Athenian generals Demosthenes and Eurymedon
were hcading for Sicily with an army which they gathered from Corcyra and
the mainland. They crossed the Ionian gulf and put in at the lapygian islands
called the Choerades,
and took on board 150 lapygian javelin-throwers of the Messapian tribe. Then, after
they had renewed an old friendship (phi/ia) with the local ruler. Artas, who had
provided the javelin-throwers, they went on and reach~d Metapontum in Italy. I 10
Insofar as the practice was not thought to endanger communal welfare, it

was not only tolerated but even enthusiastically approved. A good example
comes from the career of the Athenian general of noble binh, Konon. In
(socrates' words, no sooner had Konon and Evagoras, ruler of Cyprus, met
one another than "they esteemed each other more highly than those who
before had been their intimate friends (oikeious)". III After the battle of
Aigospotamoi, Konon, the only Athenian general who was on alert, found
refuge with Evagoras. Each of them separately and the twO of them together
bestowed great benefits on Athens, to the extent that Evagoras was given
Athenian citizenship. When the Spanans were ravaging Asia Minor, Eva-
goras and Konon advised the Persian king how to wage war against them
most efficiently. The Persians took their advice, and the Spartans were
defeated in the naval battle off Cnidus (394 B.C.). Following this defeat,
continues (socrates, the Spartans lost their supremacy, the Greeks regained
thcir freedom, and Athcns recovercd some of its old-time glory. He then
adds:
And although all this was accomplished with Konon as commander, yet Evagoras
both made the outcome possible and furnished the greater pan of the army. In
gratitude we [the Athenians] honoured them with the highest of honours and set up
their statues where stands the image of Zeus the Saviour, near to it and to one
another, a memorial both of the magnitude of their benefactions and of their mutual
friendship (tile; epL)Jae; tile; 1tQOe; Oll~AOU; ). III
As in the ca~ of the other goods and services circulating between ritua-
lised friends, here, too, each system was at times indispensable for the
proper functioning of the other.
110 Thucydidcs 7.33.4. I ~t out in a forthcoming paper the arguments against the prevailing
yi~ that the phi/ia here rden to a trut)· ~tween Artas and Athens.
III (socrates 19.53 (Atgi"el;cus).
III (socrates 9.57 (Evagoras).
, • I .: .,: :~, : : '. ~ .',.; ,

4.6 ESTATES
Before the final engagement with Anaxerxes' forces, Cyrus the younger
exhorted the leaders of the Greek contingents, saying, among other things:
Well, genrlemen, my father's realm extends towards the south to a region where men
cannot dwell by reason of the heat, and to the north to a region where they cannot
dwell by reason of the cold; and all that lies between these limits my brother's philoi
rule as satraps. Now, if we win the victory, we must put our philoi in control of these
provinces. I fear, therefore, not that I shall not have enough to give to each of my
philoi, if success attends us, but that I shall not have enough philoi to give to. 1l3
Even if we doubt, as some Greeks did, the sincerity of Cyrus' promise, we
can be sure that in spirit it was impeccable. Ever since the appearance on the
scene of the great territorial empires, it had always been the policy of
victorious rulers to assign newly conquered territories to dose friends (or,
alternatively, to make the heads of vanquished units friends). The number of
princely kinsmen being inadequate for the task, and centrally controlled
bureaucracies and armies being as yet non-existent, what other means were
there to attain control over an ever-growing number of subjects, and over
ever-expanding territories? The Macedonian and Hellenistic empires did
not differ in this respect from the Persian. Philip II gave to his 800 hetairo;
from abroad so much land that, according to a good contemporary source,
they derived more income from it than the richest 10,000 Greeks. 114 Alex-
ander the Great, once in Asia Minor, declared that uall the land around is
mine", I 15 and set about distributing it to his closest friends. His successors
followed suit: in the Hellenistic age, being a phi/os of a king was almost
synonymous with being in possession of an estate on the king's land. Nor
was the practice restricted to successful conquerors. The scattered docu-
mentation reveals that lesser rulers, too (men of the stamp of Seuthes of
Thrace, Satyros of Bosporus, Evagoras of Cyprus, Dionysius of Sicily), were
in the habit of granting estates to friends from abroad. 116 It seems therefore
1U Xmophon, A,.abasis 1.7.7; for a similar policy attributed to Cyrus the Gmt and
Artaxerxes I, see Xenopoon, Cyropaedia 8.6.5 (..... to many of his philoi he gave (fOOmt)
houses and servants in the various states which he had subdu~d. And even to this day those
properties, some in one land, 50fM in another, continue in the possession of the descendants of
those who then received them, while the owners themselves reside at court"), and Diodorus
Siculus 11.71.2 (..... with respect to the satraps then in office, those who were hostile to him he
dismissed and from his philo, he chose such as were competent and gave the sarrapies to
them").
114 FG,H 115 (Theopompos) F 2258.
IU OGIS 1.
116 Seurhes: Xenophon, AMbasis 7.3.2 (Alkibiades had a fortrns in the Chersonese prob-
ably given by Seuthes, Xenophon, HellmiC4 1.5.17; 2.1.25 and lysias 14.26 (Aga,nst Aid-
billtks»; Satyros: lsocrates 17. 3 (T,a~VI'CU$) (Sopaios ruler of extensive territory); Evagoras:
Andocides 1.4 (QIr the Mystmes); Dionysius: Diodorus Siculus 14.7.... In this context we
,j' ..
. ,, :,
.' .,:
6 Estates 107
that we are dealing not with a policy adopted by a particular ruler to solve a
particular problem but with a feature inherent in the very structure of these
groups. The policy thus po~s the following questions: Why did autocratic
rulers need to incorporate foreigners into the circle of their supporters? Why
did they have to effect this incorporation through grants of landed estates?
And, finally, what was the role of ritualised friendship in all this? To be able
to answer these questions, a brief survey of the practice in the Persian,
Macedonian and Hellenistic empi res is necessary.
It must be stressed at the outset that the documentation is sparse and
extremely uneven. Apart from a few generalised statements of the kind cited
above, the evidence consists of casual remarks in literary sources on the one
hand, and short-sighted, albeit magnified, references to specific grants in
inscriptions and papyri on the other. It is frequently impossible to tell why a
grant was made, in what ways it was supposed to serve the grantee, and
what aims the grantor hoped to achieve thereby. Nevertheless, allowing for
a certain margin of error, it may be possible to establish several general
categories. Having done so, it will become possible to test the evidence
against comparable material from feudal Europe.
First, it is necessary to draw a distinction between extensive areas given to
a friend to rule as a 'province' and the usually smaller territorial units given
him as a source of economic support. From the point of view of the paMners
both could appear as 'gifts' given by friend to friend, and both could stretch
over vast territories. 117 But the provinces quickly became instruments of , .
imperial administration and hence acquired qualities which overshadowed '-..-/'
their original character as gifts. Here, the analysis is restricted to estates

given as a source of income.
Only very rarely can we infer why a grant was made. Histiaios of Milerus
received Myrcinus in the Edonian land in return for the euerges;a of preserv-
ing the Danube bridge during Darius' Skythian campaign. 118 We do not
know why Phylakos the Samian was given the land, but we are told that he
was among the Great King's euergetai. 119 Demaratos the Lacedaemonian
117 Mania's principality (Xenophon, Hell~ica 3. I. .off.) s«ms to have been an extended
fid (Cook (1983) 178). According to Diodorus Siculus 19.75.1, Asandros, ruler of Carla, bcin~
h:ard pressed in war, came to terms with Antigonos by agreeing to transfer to him all his
soldiers, to leave the Greek cities autonomous, and to hold as a dorea the satrapy that he had
formnly held, remaining a steadfast friend (bebaios philos) of Antigonos.
III Ht"t'odotus 5.12, 4.139.
119 Ht"t'odotus 8.85.
should also mention the lycian dynast Gergis boasting in a verse inscription from the end of
the fifth century of having taken many acropoleis and of having given them to his syrrgm,is as a
part of his kingdom (Meigs and lewis (1968) no.93 and J. Bousqu~ in Comptes Rendus de
I'Acadlmie des Inscriptions et Btllts-uttrts (1975) 139).
r~ceived Teuthrania and Halisarna from X~rxes as a gift (doran) for partici-
pating with him in his ~xpedition against Greece, and Gongylos of Eretria
Iik~wis~ r~ceived four cities as a gift (doran) for having been the only
Er~trian to side with the Medes. 120 Lykon th~ Athenian was given "cities
and territories" for having betrayed to the Great King the rebellious satrap
Pissouthnes, his form~r master. III In an anecdote pres~rved in Athenaeus
but attributed to Duris of Samos, Alexander is said to have given Pnytago-
ras, among oth~r pr~sents (doreas), a fortr~ss which he himself asked for. III
In two other cases, much vaguer r~asons are given. Cyrus the Great is said to
have granted sev~n cities to Pytharchos of Cyzicus because h~ was a
philos.l 23 And Antiochus I of Syria ~xplained in a public letter why he had
presented one Aristodikides of Assus with a hug~ domain in these terms:
"because he, as our phi/os, has furnished us his services with all good-will
and enthusiasm".I24
What emerges from this survey is a strong link b~twe~n ritualised
friendship and the grant of estates: th~ partners are consistently portrayed
as 'friends' engaged in an amicable exchange of 'favours' (dora, euergesiai,
charites are th~ recurring words). Hence, by the logic of ritualised
fri~ndship, the granting of a land~d estate fdl within th~ larg~r category of
favour. This hdps to clarify the ~volution of th~ word dorea - which
became the technical name for a c~rtain type of landed estate, best known
from Hellenistic Egypt. Originally signifying any type of goods or s~rvices
given to establish peaceful rdations with a strang~r, th~ word came to mean
one specific type, the landed estat~. Finally, in the cont~xt of the centralised
administrations of the Ptolemaic ~mpire, dorea came to denote the status of
th~ land itself. It thus became applicable to a whole category of domains on
the king's land, a category singled out for specific treatm~nt on account of
the amicable relationship between the king and the grant~e. 125
In spite of the friendly tone of the explanations mentioned above, it
would be a mistake to accord th~m too much cred~nc~. Ritualised
friendship could serve as an explanation of motiv~s precisely because it
suggested disint~rest~dness in the form of gifts given freely, without ex-
110 Xenophon, Htlltn;ca 3.1.6; he w:as banished from Eretria on that account.
121 fCrH 688 (Ktesias)F 15 p:ara. 52.
111 Athmaeus 4.167c = FCrH 76 F 4 (Duris). The remarkable Biblical parallel is I Samuel
27.5-6: "David s:aid to AchiKh King of G:ath, 'If I stand w~1I in your opinion, grant me a place
in one of your country towns where I may settle. Why should I remain in the royal cities of your
mainty?' Achisch grante'd him Zikl:ag on that day......
12' fCrH 72 (Agathokles of Cyzicus) F6.
124 Welles (1934) no. 11.
I2J The eviden« is sp«1fic to Egypt, d. Rostovtuff (1922) 43, but there are good reasons
to Mlieve that a simil:ar process took pla« in the other empires.
6 Estates 109
pectation of return. ll6 The courteous language of ritualised friendship

could conveniently be harnessed to disguise real motivations. The nature of
the ancient evidence is such that real motivations are seldom allowed to
shine through.
We must now penetrate beyond the word and observe the concrete
variations that it served to veil. Using the density of settlements and the
degree of cultivation as an index, it is possible to distinguish three, not
mutually exclusive, types of domain: cities, estates, and what can be labelled
'uncultivated areas'. In Table 3, an attempt has been made to apply this
classification to all the cases of doreai known in some detail from the
Persian, Macedonian, and Hellenistic empires.
Table 3. Grants of l.a"ded Estates 127
(a) The PersIan Empire, 560-330 B.C.
Reference Gr.mror Rc-cipiC'nr
FGrIl471 F6 Cyru\ I Pyrharcho\ 7 C:UIt"\

ICyzicus) --..)
Hdt.6.70; l);JrJU\ I Dcmar3ros Uno,peclflecJ; ::v

.'-J
Xen.llell. 3.1.6 ISparral I Clr>'
Hdr.6.41 Danu\ I MerlOchos, son of 2 Cltlt'~
Mihiadcos
I Arhc-ns) >-
Q..
Hdr. 5.11; 2.1--4; 124 D.ariusl Hisriaios I esr.are; o
IMllc-rusl I l'lr)'
U
Hdr.5.37.1 Darius I KOt'S I CIt>·
IMyrilenel
Thuc.1.08 Xerxes I Themisroklcos .l clues
IArhensl
1241 This is a fearure which encour:aged rh~ making of puns on dorta;. In :a comIC fragmenr
from rhe rhird cenrury S.c. (PhOt'1licides, in T. Kock, Com,corum Att;corum Fragmtnta
(uipzig, 1888), vol.lll, no ..133), a man (presumably a wuuld·bco l,h,/m) 1'\ uld rn h.:IVt'
pr~tended to a courtnan that h~ would receiv~ a dorea from a king. Bur rh~n she complaJnrd:
"Now b«au~ of rhis dorea of which I am r.pc:aking rhis scoundrel h:ad me a whole year for
norhing (dorean»".
127 The rable draws on the following works. Pnsia: Sre. Croix (1972) Uff, (1981) 151ff;
Kienau (1973) 257ff.; lewis (1977) 54ff.; Brianr (1982) 291ff.; Hornblower (1982) 143ff.;
Cook (1983) 167ff. Macedonia: Hammond and Griffith (1979) 66Off.; Kienasr (197.') 257ff.
He~/t1fist;c monarchies: Rosrovttdf (1922) :and (1941) sec Index, S.\·. Dorta;; C. Prtaux,
L 'F.conom;t royale des LAgides (Brus~ls, 1939) 20ff.; Wipszycska (1961); Kre1s'ilg (1977) and
(1978); Worrte (1978); Funck (1978); Sre. Croix (1981) 119ff.; Davlcos (1984). Egyprian
dorea-hol&rs :are only Iisred if rhe receipr of a landed ~rare can readily M inferrc:-d from rhe
pap)·ri. For:a more complere lisr, s« rr vol.lV, 10061-10108. For a unique gr:anr in Homer-
seven ciricos offered by Agamemnon ro Achilles (Iliad 9.149-56 ;;; 9.291-8) - sec M.1. Finley,
'Humer and Mycen:ae: property :and renure' in Finley (1981) 2n-.ll.
, • I .: , • ,.,~, : : '. ~ .',.; ,
Reference Grantor Recipienr Typ~
Xen.lltli. 3.1.5; Xerxes I Gongylos 4-5 cities

Anab. 7.8.8, 17; (Eretria]
Thuc.l.128
Hdt.8.85 Xerxes I Phylakos 1 city(?)
(Samos]
Hdt. 8.136.5.21 Xerxes Amyntas 'of I city
Asia'
Dem.23.141-2 Ariobarzanes Philiskos 1 city
[Abydusl
FGrH 688 F15 (52) Darius II lykon Ciries and
lAlhnls) eSlafes
Arr. AMb. 1.17.8; Darius III? Memnon Estates and
Dem. 23.154,157; (Rhod~] cities
Polyaen.4.3.5;
Strabo J .3.5
(b) Macedonia to 336 B.C.
Reference Grantor Recipimt Type
Hdt.5.94 Amyntas Hippias 1 city

[Athens)
Syll..J 332 Philip II Polemokrates 2 estates
[Elimiotis?]
Syll. J 332 Philip II Koinos 1 estate
(Elimioris?l
lXm. 19.314; d. Philip II Aeschines Estat~in
18.41 (Athens] Boeotia
(Dem.] 7.32 Philip II Alexander 3 cilies
(of Epirus?1
(c) Th~ Hellenistic monarchies
Reference Grantor Recipienr Type
Syll.) 302 Krateuas Aristomenes Estat~

[Unknown]
SylU 332 Alexand~ III Ptolemy 1 ~tate
(Unknown]
Syll.J 332 Cassander Perdikkas 4 (Slates
[Elimiotis?)
Welles (1934). Antiochos I Aristodik ides Unspecified;
nos. 10-13 (Assos] I eslate
SMdis 7 (I). no.l Anrigonos I{?) Mntsimachos testate
[Unknownl
S4rdis 7 (1), no.l Anriochos III Pytheosand 1 estate
Adrastos
[Unknown]
. ,, :,
.' .,:
6 Estates 111
OGlSH Ptolemy III Ptoltmy~on I CIt)'

of lysimachos
IUnknownl
Fischer (1979) Antiochos III Ptolemy son 1 ~rate
ofThra~:n
(Unknown]
PP IV.l0064 Ptolemy II Apoll on i<><; 2 estatt"!i
ICuial
PPIV.IOO91 Ptolemy II, Nikanor I csr;lte
III orlV? (Unknown)
PP IV, 10087 Ptolemy III K:allixeno!> I estate
(Unknown)
PPIV.10107 Ptolemy IV Chry~rmos I esr.are
IUnknown)
PPIV,10100 Ptolemy III Sosiblos I estate
:and IV IAlexandrial
PPIV.l0097 Ptolemy III Senthcus 1estate :c
IEg)'ptl -
8J
~
PPIV.l0061 Ptolemy III Agarhokl~ I ~tate :1:,
andlV (Unknown) <-:-
L.
PPIV.l0068 Ptolemy II Aristol~os 1est:atc --..)
or III? IMacedonian)
~
::v
rPIV.l00n Ptolemy IV Dikaiarchos I costate
IUnknown) VI
~
PP IV.10087a Unknown Kom:anos I estate
IUnknownl
>--
Q..
-8
U
Briefly, the terms 'city' and 'cities' refer to populated areas converging
upon one or several urban ~ttlements. These areas had an economic and
social life and a history of their own prior to becoming a gift. Once they
were convened into dorea;, the grantee became the city's ruler. The term
archon, however, which is sometimes used in this context, should not be
misconstrued. The recipient of a dorea became not a civic magistrate but a
virtual despot. The territory, with all its population and wealth, was given
him as a source of income to suppon himself and his followers, to raise
troops, and, most important, to pay taxes to the overlord. Among the
recipients of doreai from Persian satraps and kings there may have been
some who exercised their rule benevolently. But, from the middle of the
third century B.C. comes a document which yields a gloomy picture of the
system.
This is an inscription preserving an exchange of letters between Ptolemy
II Philadelphus and Telmessus in Lycia, a city just recently incorporated into
the Egyptian ruler's external domains. It is possible to infer from it the
moves that preceded the correspondence with a fair degree of certainty. A
royal phi/os had probably seiz.ed the city with the intention of petitioning
the king to grant him the city as a dorea. J28 The city officials, anxious to
prevent this, dispatched an embassy who successfully interceded with the
king: Ptolemy had graciously exempted the community from being made a
dorea. In the covering letter, the community resolved to praise the monarch
and set up an altar in his honour, to the effect that Uno one may ask for the
city of T e1messus, or the villages or anything belonging to the territory of
Telmessus to be given as a dorea from any king or queen or dynast
whatsoever under any pretext". 129 The exemption, however, did not last
for long. In a Telmessian decree from the times of the next Ptolemy (III
Euergetes), a certain Ptolemy, son of Lysimachos, was hailed as benefactor
for remitting an unpleasantly long and varied series of taxes. In all probabil-
ity, Ptolemy, son of Lysimachos, ruled Telmessos as a recipient of a dorea
from Ptolemy Euergetes. 130
The cities given away in this way must have been numerous. The state-
ment that Eumenes of Cardia gave the cities of Cappadocia to his philoi may
not carry much evidential value, as it is preserved in Plutarch alone. 131 But
the scale of the practice emerges with unmistakable clarity from the peace
treaty imposed by the Romans on Antiochos III in 188 B.C. Polybius
records that it had been stipulated that "if any of the cities which Antiochos
has to give up have bttn given by him to others, he shall withdraw from
these also the garrisons and the men in possession of them".1.l2 It is thus
likely that the practice was widespread.
An 'estate,1J3 was an artificially created economic unit comprising culti-
vated territories as well as a subject population living in villages. In the
Hellenistic age, some of these estates were taken over from the Persians,
others were no doubt newly created.' A grantee could hold several territories
at once. Mnesimachos' dorea in Sardis consisted of villages, hamlets and
121 A similar attempt is recorded in an inscription from Samos dated 246-243 8.C. Some of
the most notable philo; of Antiochos I had appropriat~d lands ~Ionging to Samian cirizms.
Thanks to the intercession of Boulagoras, an eminent Samian cirizcn, with Antiochos, the
property was restored (SEG 1.366). Such restorations must have been exceprional: Boulagoras
is praised, among oth~r things, because, since the restoration of the lands, "none of the
subordinates of Antiochos ever tri~d to lay claim to what belonJed to the citizens".
u, Worrle (1978) 202. I am ht:avily indebted to this anicle for the collection of the
Hellenistic cyiden~.
130 OGIS 55. It cannot be ruled out that lysimachos was a xmas of Ptolmay of Egypt and
named his son after him. For the discussion of the identity of Ptolemy son of lysimachos, see
Rostovtzeff (1922) 4S n.50 and (1941) vol.l, 336.
131 Plutarch, Eu",mts 10. I J2 Polybius 21.43.18.
1U For the technical name ous;a, which is sometim~s used in this regard, see Kreissig (1977)
10 and (J978) 40ff.
.j' ..
. ,, :,
.' .,:
6 Estates 113
non-contiguous pieces of land scattered over several units of royal

administration. 1.14 The Egyptian doreai, attested through the papyri from a
different angle, do not seem to have departed from this pattern.
What I call 'uncultivated areas' refer to territories lacking cultivation but
not populations. Clearly, such territories were given to a grantee with the
intention that he convert them into estates. A good example is a plot of
2000 plethra of land given by Antiochos I to a certain Aristodikides of
Assus. 135 This land, as observed by Kreissig, was no ousia, but just 'land
between'. To endow it with a minimal degree of administrative unity (and,
presumably, facilitate the recruitment of labourers), Aristodikides would
have to join it to the territory of a Greek city.
To form some idea of the ties of dependence that dorea; created between
the grantor and grantee, it is necessary to find the answer to the following
questions: What kind of hold did the recipient acquire over the domains?
What obligations did the possession of the dorea impose? And finally, by
what rules were these obligations regulated? The question of hold may be
considered first.
The most important document in this respect is the inscription recording
the confirmation of four estates in the ownership of a certain Perdikkas, son
of Koinos. 136 Two of these domains were originally given by Philip II to
Perdikkas' grandfather, Polemokrates of Elimiotis. The background was
conquest: newly acquired territories were presumably distributed by lot >-
Q..
between the heta;ro;. Philip's grant carried full ownership: Polemokrates o
and his descendants were empowered to own, give away and sell the estates. U
A third estate was given by Philip to Koinos on the same conditions. A
fourth estate was given by Alexander the Great to one Ptolemaios. Perdik-
kas son of Koinos inherited the three estates granted by Philip to his father
and grandfather and bought for money the estate granted by Alexander to
Ptolemaios. In the inscription, King Cassander confirmed that all four
estates belonged to Perdikkas son of Koinos on the original conditions:
Perdikkas was free to own, give away or sell the estates. So, unmistakably,
dorea here carries full ownership.
No similar evidence is available from the Persian empire, but we know of
at least two doreai which were in the grantees' possession for at least twO
generations. We find the sons and widow of Gongylos the Eretrian ruling
the five cities Gongylos received from Xerxes more than ninety years after
1\4 Cf. Kreissig (1977) 10.

IJS Welle~ (1934) nos. 10-13.
I)" Syll. ., .H2. Cf. R.M. Errington in Journal of Hellmic Studitos 94 (1974) B; Hammond
:and Griffith (1979) .167.
the grant was made. 137 In the same period, the descendants of Demaratos
ruled the two cities that Demaratos was given by Darius. 138 It thus seems
possible that either the estates were given in full ownership or that by some
coincidence they were not reclaimed by the four (in the case of Demaratos,
five) successive Persian kings during this period. It does not seem possible to
arrive at a clear-cut conclusion on the basis of such evidence alone. But the
fact that two of Gongylos' cities, Myrina and Gryneion, appear among the
tribute-paying members of the Delian League, points again to the possibility
of holding a dorea without being formally attached to the grantor. U9 And
that, again, suggests full ownership.
Undeniably, a powerful ruler, Persian, Macedonian or Hellenistic, could
revoke a dorea at whim whatever the precise terms of the grant. Yet the
theoretical question of grant in full ownership or grant in temporary posses-
sion represents a significant difference for the psychology of the rela-
tionship, and hence for the practical exercise of the ruler's powers. The very
limited evidence examined so far suggests that in this period full ownership
was the rule. The question is, was this custom discontinued in the Hellenis-
tic age, as has been claimed ever since Rostovtzeff? 140
The answer seems to be that in the Hellenistic world there was no
uniform practice. Kings basically regarded their whole empire, including
doreai given away, as their personal property. In practice, however, some
doreai were given away on certain stipulated conditions, and others in full
ownership. The former case is exemplified by a clause in the Mnesimachos
inscription foreseeing a possible reclamation of a part of the dorea by the
king; 141 the latter by the dorea of Ptolemy son of Thraseas near Skythopolis
in southern Syria, which included both villages given in possession (egktesei)
and villages given in full ownership (eis to patrikon).142 Clearly, there was
no unified practice.
The second question, what obligations did the possession or ownership of
the dorea impose, may quickly be answered. It involved the payment of
137 Xenophon. Artabasis 7.8.8; 8.17.

u, Xenophon, Htllmic.a 3.1.6.
u, O. Ste. Croix (1972) 39.
lotO Rosrovtzeff (1922) 48, who saw dore,,; as "not hereditary but personal holdings, usually
aMOCiated with the high position occupied by the holder in the military or civil administration
of the kingdom.... Cf. Kreissig (19n) 10: Uland given err aphtm goes into the possession, but
not the ownership, of th~ receiver".
141 "If the king on account of Mnnimachos shall take away from Anemis the villages, the
allotments (Itleror1 or any other pan of fa hypolt.timmtl...". Cf. Atkinson (1972) 61, who
regards the contract as a "sale of possessio of royal land, special conditions being insened in
the contract itself to provide for the contingency of resumption of the land by the King".
•42 landau (1966) II.lVa, lines 22-3, with Fischer (1979) for emendations and mort recent
bibliography.
.j' ..
. ,, :,
.' .,:
6 Estates 115
taxes to the grantor but it did not create obligations of service, military or
other. To be sure, the grantee, when called upon, did come with his men to
assist the king in fighting. And it frequently happened that he went on
embassy for the king or promoted the king's interests in his own city. But
these services were not performed in return for the dorea. The doreai were a
means of providing the recipient with maintenance - which, no doubt,
indirectly facilitated the performance of services. They even made it possible
for the grantor to become an absentee landlord, entrust the management of
a dorea to a manager, move to the court, and become a member of the
central administration. But nowhere is the performance of services for the
king made a condition of the reception of the do rea. The duties of the
recipient were left vague and unspecified. They were neither the duties of a
tenant nor those of an administrator. Predominantly, they remained the
duties of friendship. So, in order to proceed further, and see how the
interplay of goods and services, personal relations and administrative rules
structured these societies, we must find oUt what these duties consisted of.
5
OBLIGATIONS: HEROIC AND CIVIC
5.1 'WHAT AlKIBI ADES DID AND SUFFERED'

It is not hard to see why Alkibiades should have entered history as the
archetypal traitor of all time. Until the expedition to Sicilyt he seemingly
deployed all his talents and all his wealth in the service of his native cityt
Athens. But when t in 415 B.C. t he was implicated in a supposed plot against
democracy and was recalled from SicilYt he fled to Sparta, Athens' bitter
enemy, and made recommendations about how best to inflict injury on
Athens (by sending an expedition to Sicily and establishing a post at Decelea
in Attica). When Sparta had taken full advantage of this valuable advice,
more betrayals of Athenian interests followed. In 412 B.C., Alkibiades went .-
on a mission to Ionia on Sparta's behalf and persuaded many cities allied
with Athens to revolt. This seriously impaired the war effort since it not
only reduced Athens' fighting potential but also the volume of tribute which
financed it. Then Alkibiades effected yet anothert seemingly impossible
volte-face. Realising that he had become suspect to the Peloponnesians on C"I
I-
account of a personal rivalry, he withdrew to the court of the Persian satrap
>-
Tissaphernes t became Tissaphernes' adviser t and advised him how to inflict 0-
o
damage on the Peloponnesians (namely, by cutting down the rate of pay of U
their navy). FinallYt by virtue of his good connexions t Alkibiades was not
only accepted back into his native Athens t but was even elected to the
prestigious office of strategos.
Not that extraordinary acts of treason are unknown in history. But there
is something altogether unique about treason in the Greek city t both as an
idea and a practice. For Alkibiades was merely emulating a pattern which,
to a lesser extent t was present in the careers of men such as Hippias,
Demaratos t Themistokles, Pausanias, and t as we shall see t in scores of less
impressive careers in Greek history. To judge by modern standards, what is
unusual about these men is the remarkable ease with which they could ally
themselves with the enemies of their immediate groups; the astonishingly
high positions they could secure whenever they changed sides; the devotion
they showed to the party they were allied to at anyone time; and their
effecting all these tergiversations not only without the slightest compunction
but with an apparently unwavering belief in their Own sense of justice.
A few typical excerpts from some of the most widely used manuals of
116
1 'What Alkibiades Did and Suffered' 117
Greek history will show that this pattern of action is only poorly under-
stood. In Bury's History of Greece, first published in 1900, Alkibiades was
introduced as "a young man of high birth, brilliant intelligence, and no
morality". As his schemings gradually came to be revealed, new epithets
emerged, "profane and unstable" and "renegade" being the most significant
ones for our purposes.· To Laistner, Alkibiades' morality became under-
standable only in contrast with that of Kimon: " ... whereas in Kimon an iron
sense of duty and profound love of his country were the guiding influences
of his public life, in Alkibiades self-interest was throughout his career the
overmastering passion. His youthful extravagances and arrogant disregard
of every convention were condoned as high spirits by a public fascinated
with his personal beauty... .,2 Published in 1936, this might at first sight
appear a belated return to the naivete of the previous century. But very
much the same evaluations are encountered again and again in works
published since. According to the history of Greece by Burn, which
appeared in 1966, Alkibiades "was rich by inheritance, but his extravagance
would have strained any fortune. He set out to aggrandise himself, and the
morality of the means was no object. An aristocrat, he heartily despised the
people, though naturally he flattered them in public. His natural party
would have been the conservatives; but they mistrusted him. Many respect-
able people disapproved of his morals"·) This picture is by no means
confined to the English-speaking world. E. Will, for instance, also takes
something like Laistner's 'overmastering passion' to have been the main-
spring of Alkibiades' actions: "Divinement beau, prodigieusement intelli-
gent, d'une lucidite politique qui I'apparente a Themistocle, Alcibiade ctait
toutefois rebelle a cet esprit de sou mission a la tradition et aux lois qui etait
encore Ie ciment de la cite et faisait la dignite du citoyen. On a doute qu'il flit
un produit de I'enseigncment des sophistes, mais, s'il est certain que I'indivi-
dualisme et I'ambition sans scrupules se passent d'un tel enseignement, il
apparait evident que I"esprit nouveau' avait souffle sur Alcibiade.,,4
No attempt will be made in this study to vindicate Alkibiades: moral
judgements often preclude historical understanding. But it will be argued
that characterisations such as these reflect the normative assumptions of the
modern, narional centralised state; and that they constitute a misrepresenta-
tion of the manner in which ideas and actions were interrelated in antiquity.
, The epithets a~ unchanged in rhe subsequenr edirions of rhe b()ok published, since 1951,
by R. Meiggs.
1 M.L.W. laismer, A History of the Crnlt World {rom 479 to 323 8.C. (london, 1936)
118.
1 A.R. Burn, Tht Pelican History of C,etce (Harmondsworth, 1966) 282.
4 Ed. Will, u mondt grtc tt I' orient (Paris, 1972) voU, 342.
118 Obligations: Heroic and Civic
Once we reconstruct the system of ideas of ritualised friendship, it will
become apparent that Alkibiades was operating consistently with values
which were not only traditional but, within cenain social spheres, even
highly respectable. Indeed, from the perspective of the people participating
in these spheres, it was civic morality which appeared new and disreputable.
5.2 THE NATURE OF THE OBLIGATION
Every single utterance related to xenia is permeated with an intense sense of

duty. And these duties are conceived of in astonishingly uniform terms: it is
as if everyone recognised how a xenos should behave in each specific
situation. Clearly, this recognition flowed from an ethical code implicitly
shared by all members of society. For we witness no attempts at formulating
or systematising it; at no stage in Greek history did ritualised friendship
undergo anything like the incorporation of the code of godparenthood into .-
church legislation, or the crystallisation of the feudal bond into the 'contract
of vassalage'. 5 The intricate transformation of moral principles into explicit
regulations never advanced beyond the occasional drafting of a treaty, or
the coining of an aphorism. This state of documentation is highly significant
for our attempt to reconstruct the obligations of ritualised friendship. For
what we are left with is not a written code or an inventory of duties, but C"I
I-
actions in the name of xenia and comments made on these actions. >-
0-
Three stories will serve as an introduction to the matter. Since the o
references to the duties are incidentally made, and since similar conceptions U
emerge from scores of other stories, there are no good reasons to doubt that
what we have here was the nonn, not the exception.
The first example is intertwined with the Ionian revolt of 499 B.C.
Megabates, the Persian general charged with the seizure of Naxos, dis-
covered that a watch was absent on a Myndian ship under his command. As
punishment for this negligence, Megabates bade his guards find Skylax, the
captain of the ship, and bind him in such a fashion that his head was outside
the oar-hole of the ship and his body inside. When word was brought to
Aristagoras of Milerus about this, he at once came to the aid of Skylax, his
xenos, and pleaded with Megabates for his release. When Megabates
proved impervious, Aristagoras released Skylax himself. This overt chal-
lenge to his authority so enraged Megabates that he forewarned the Naxians
of the impending invasion, thereby frustrating the whole operation. It was,
according to Herodotus, for fear that on account of this failure he would be
$ Marc Bloch (1961) 219; Ganshof (1964) 80; Mintz and Wolff in Schmidr er al. (eds.)
(1977); Goody (1983) Chapter 9.
deprived of his lordship of Miletus that Aristagoras began to plan the revolt
against the Persians. 6
The second incident is told us by Thucydides. In 424 B.C., the Spartan
general Brasidas set out for Thrace with seventeen hundred Peloponnesian
hoplites. Upon his arrival at Heraclea in Trachis t Brasidas sent a request to
his xenoi in Pharsalus to escort him and his army through Thessaly.' The
xeno; came at once and the march proceeded. Thucydides' comments on the
problem of crossing foreign territory are important for an understanding of
what followed. In the first placet says Thucydides, it was not easy to traverse
Thessaly without an escort t especially with an army. SecondlYt crossing the
territory of neighbours without their consent was generally looked upon
with suspicion by all the Greeks. ThirdlYt the commoners (plethos) in
Thessaly had always been on good terms with Athens (and therefore inimic-
al to Sparta). Thucydides notes that if Thessaly had been under democratic
rule (;sonom;a) and not in the hands of a few powerful men (dynasteia)t
Brasidas would have been unable to go forward. Even SOt Brasidas was
confronted by some Thessalians from the democratic faction. They tried to
prevent him from going further t asserting that t in proceeding against the
consent of the whole community, Brasidas was violating accepted conven-
tions (adikein). At this point, the xenoi escorting Brasidas intervened, plead-
ing for understanding. They said that if the inhabitants of the country were
unwilling t they would not conduct the march any further; it was merely that
Brasidas had unexpectedly arrived in their land and they were escorting him
since they were his xeno;. The opposing Thessalians apparently accepted the
invocation of this moral duty as partial justification of the transgression and
allowed the march to continue.
Our last incident t recounted by Xenophon, took place during Cyrus'
march against his brother Artaxerxes. When it became apparent to the
Greek soldiers that they had been tricked into an attempt to overthrow the
Persian throne t they threatened to discontinue the march. Klearchos wept
for a long while before the soldiers and then spoke as follows:
Fellow soldiers, do not wonder that J am distressed at the present situation. For
Cyrus became my xenos and not only honoured me, an exile from my fatherland, in
various ways, but gave me ten thousand darics. And I, receiving this money, did not
lay it up for my own personal use ... but I proceeded to expend it on you. First, I
wmt to war with the Thracians ... Then, when Cyru5' summons C2me, I took you
with me and ~t out, in order that, if he had need of me, I might give him aid in
return (or the benefits I received from him. But you now do not wish to continue the
6 Herodotus 5.33-5.
7 The word Thucydide1 U!e5 is ep;t~d~;o; in 4.78.1. xtno; in 4.78.4. That Brasidas (like
Ptrdikkas in 4.132.2) had XnlO; in Thnsaly, Thucydid~ assumes that his readen would take
for granted.
. ,, :,
.' .,:
march with me; so it seems that I must either betray you and continue to make use of
Cyrus' friendship (philia), or prove false with him and remain with you ....
A conspicuous common feature of all these episodes is the invocation of
xenia in order to advocate a certain course of action. The invocation was in
each case provoked by a rival premise: obedience to a higher authority in
the case of Aristagoras; infringement of territorial sovereignty in the C2se of
Brasidasj and an instinctive sense of self-preservation on the part of the
mutinous soldiers. It is precisely the gravity of these premises that throws
into relief the intensity of the sense of duty inherent in xenia. The implica-
tion is that this sense of duty, which has been ignored, or at best misinter-
preted in the modern literature, was held by contemporaries as high in the
scale of values as these rival considerations.
The conception of xenia as an institution is another characteristic emer-
ging from these incidents. There existed in the ancient Greek world a form
of behaviour appropriate to xenia in much the same way as there was one .-
appropriate to kinship and friendship. The idea was that through the rites of
initiation the personalities of the actors and their roles as xeno; became
inseparably fused; that. as a result, they became bound by mutual obliga-
tions; and that actions, attitudes and modes of conduct towards each other
flowed from these roles. It was in the name of xenia that Aristagoras
C"I
released Skylax, or that the Thessalians escorted Brasidas. The incident with I-
Klearchos shows that the ideas embodied in the institution might well >-
0-
diverge from the ideas entertained by its incumbent. For we may doubt the o
U
truth of what Klearchos said. Transparently, he thought it to be to his own
personal advantage to continue the march. To the soldiers, however, he
addressed himself not as a self-seeking personality but as the victim of social
necessity: they had to continue the march since he was bound by his role.
The fact that the soldiers yielded indicates the power of this emotional
appeal.
A society which has elaborated the conception of such an institution will
differ from one in which it is absent. The institution provided a channel of
interaction for individuals who by descent and primary attachment be-
longed to different groups and who could not otherwise have co-operated
'naturally'. Historians in antiquity seem to have attached great importance
to this fact. Without Aristagoras' rescue of Skylax, Naxos would probably
have been taken. Without the assistance of the Thessalian xenoi, Brasidas'
march probably could not have been carried out. Without the option of the
emotional appeal to the rights and duties of xenia, Klearchos might have
been unable to continue the march. Whether an abstract idea or a concrete
• Xmophon, Anabasis 1.2.3-5.
relationship, xenia could have a powerful effect on the general course of

events.
The ethos of xenia revolved around tht twin poles of idealism and
instrumentality. It called, on the one hand, for unconditional loyalty, hones-
ty and devotion even at the cost of self-sacrifice. Satyros of Athens, for
instance, rather than choosing a precious gift, asked Philip of Macedon for
the daughters of a deceased xenos, Apollophanes of Pydna. "It will bring me
no gain", he was made to say, "for I shall provide them with dowries and
give them in marriage; and I shall not permit them to suffer any treatment
unworthy of myself and their father". The sptech is said to have been met
by such outburst of applause from the other guests that Philip granted the
request, even though Apollophanes was one of the murderers of his own
brother. 9 On the other hand, expectations of reciprocity - whether immedi-
ate or delayed, whether in goods or in services - were built into almost every
single utterance or gesture connected with the institution. "For Cyrus be-
came my xenos, and not only honoured me ... but gave me ten thousand
darics...", Xenophon made Klearchos say, and the debt Klearchos owed
Cyrus for the money was inseparable from the debt he owed him as a xenos:
he repaid both by putting himself at Cyrus' service.
In this world favours accepted generated a strong sense of indebtedness,
and had to be repaid with even more pressing urgency than monetary debts.
There was thus a constant oscillation between giving and receiving, helping
and being hdptd - an oscillation that made for the perpetuation of the
bond. Where continuous exchanges were lacking, the relationship was likely
to fade: (socrates, for example, in his letter to Timotheos, felt it necessary to
apologise for asking Timotheos a favour even though he never made a
request to Timotheos' father. lo The most valued forms of repayment were
those which aimed at the benefactor's most pressing needs. The Greeks had
a special word for this, boetheia. "You will best serve your friends (phi/oi) if
you do not wait for them to ask your help, but go of your own accord at the
crucial moment to lend some aid (tv loie; xalQOiC; ~'l9fJC;)", advised Iso-
crates to Demonikos, II and an incident from Herodotus may show how
such acts of boetheia were carried out in practice. In the midst of a war
between the Lydian king Alyattes and Thrasyboulos, the tyrant of Miletus, a
third ruler, Periander, tyrant of Corinth, learnt what reply the oracle from
Delphi had given to Alyattes. Thereupon he dispatched a messenger to
9 Demosth~nes J9.J95 (D~ Fa/sa uRatio"~).

I (l 1s00~rat~, Ut,~, 7.12.
II lsocrat~ 1.25 (To Demonicus).
Thrasyboulos, "so that his xenos should be forewarned and make his plans
accordingly". 12
Saving or sparing a xenos, ransoming him from captivity, providing him
with shelter and food were the most frequent manifestations of boetheiai. lJ
It is customary to attribute these acts to a sentiment of warm humanity. But,
I suggest, the picture is more complex than that. In all societies, ordinary
favours can be repaid with counter-favours of commensurate magnitude.
Not so favours of extraordinary magnitude; these create what has been
called an 'unrepayable debt':
An un repayable debt is one given in time of extreme need at great cost. Saving a life
creates an un repayable debt ... Acknowledgment of the existence of such a debt
means that the indebted ally cannot break the alliance, and must always be ready to
respond to a call for aid, unless he can repay with a favour that is itself of such
'unrepayable' magnitude that it cancels out the initial debt. I.
An ambiguity thus springs to mind whenever we read of spared enemies.
Did the generous act arise from general considerations of humanity, or was
.-
it part of a cunning strategy designed to enlarge - or strengthen - the circle
of loyal followers? Following the capture of Thebes in 335 B.C., for exam-
ple, Alexander's troops are said to have razed the city to the ground and
enslaved the Theban survivors (including women and children) making an
exception for priestesses and those who were the xeno; of Alexander and C"I
I-
Philip and proxeno; of the Maceclonians. IS No definite answer is possible, >-
0-
but in interpreting motivations on such occasions there should be an aware- o
ness of the alternatives. U
Whether altruistic or cynically calculating, xenia exercised a constraining
effect on behaviour. To public opinion, it often appeared irresistible. Thucy-
dides, for example, reports a popular belief according to which Themis-
tokles put an end to his life when he realised that it was impossible for him
to fulfil his promise to the Great King. 16 Another indic2tion comes from a
11 Herodotus 1.20-22.
lJ Aristotle seeml again to have echoed a popular view when he dncribed the bestowal of
favours and assistance <xaq£oao6m Kat ~L) on friends (philo.). JUnt-friends (XnlO,),
and comrades (hetairai) al the gnatnt of plealures (Politics 2.1263b6). His funher comment,
that "a condition of this il the private ownership of property", is another indication of the
upper-class bias of the institution.
14 land~ in Schmidt d at (eds.) (19n) xvii. For a dncription based on first-hand observa-
tion, 1ft Douglas (1960). Some ancient examples: Histiaios and Koes sparing Darius' bridge
(Herodotus 9.97); Xenagoras of Halicamassus saving the life of Maliltes. son of Darius
(Herodotus 9.107); Neon sparing the trapped Antigonos (Polybius 20.5-6).
U Arrian, Anabasis 1.9.9 and Plutarch, AI~xQ1fd" 10.11. Cf. Uvy 1.1.1 for Aeneas and
Antenor being spared lI~tUJti iur~ hospi'i;.
16 Thucydides 1. 138.4. d. Herodotus 5.35.1 for Aristagoral' inability to fulfil hi5 promite
to Artaphrenes.
legendary story. Herodotus tells us that Themison, the trader from Thera,
was not aware when he made a pact of xenia with Etearchos, ruler of
Oaxus, that he had undertaken to drown Etearchos' daughter. Torn be-
tween his duty as a xenos and general considerations of humanity, Themi-
son enacted a mock-drowning of the young lady. The bizarre scene is clearly
intended as a kind of expiation ceremony for the obligation which Themi-
son was forced to ignore. 1 , A more realistic example comes from Polybius.
The Spartan king Archidamos, fleeing from the other king of Sparta,
Kleomenes, took refuge with a hereditary xenos, Nikagoras of Messene.
When Kleomenes held out an offer of reconciliation, Nikagoras acted as a
mediator. However, at the meeting between the two rulers, Kleomenes
treacherously put Archidamos to death, sparing Nikagoras. Polybius' de-
scription of Nikagoras' psychology is revealing: "To the outside world,
Nikagoras pretended to be grateful to Kleomenes for having spared his life,
but in his heart he bitterly resented what had occurred, for it looked as if he
.-
had been the cause of the king's death". When the right opportunity offered
itself, Nikagoras absolved himself by cunningly implicating Kleomenes in a
plot which ended in his death. 18
To what extent in general people yielded to such constraints seems
impossible to tell. Men at all times seck to reconcile self-interest with social
norms, preferably disguising the former by the latter. But the picture be- C"I
I-
comes somewhat clearer if we focus our attention on violations: what >-
0-
happened when a xenos ignored his duties? o
About the only surviving poem of Timokreon of Rhodes contains bitter U
accusations against lbemistokles the Athenian. The precise context escapes
us, but it is clear that Themistokles accepted money not to restore Timo-
kreon from exile to his native Rhodes even though Timokreon had been his
xenos. In other words, Themistokles preferred self-interest to duty and
withdrew his support precisely when it was most needed. The enraged
Timokreon called Themistokles a liar, cheat and traitor (pseustan, adikQ1"
prodotan) and imprecated curses upon him. 19 The Trojan Hektor was
seemingly guilty of a far lesser offence - he abandoned his xenos, the Lykian
Sarpedon, to the Achaeans - yet was still liable for rebuke. The Lykian
Glaukos impugned his heroic prowess in retribution - "How, then, 0
I? Herodotus 4.tH.
18 Polybius .5.37-8.
19 Timokreon, fragment 1; d. Adkins (1972a) 94.
hard-hearted, shall you save a worse man in all your company... ?" - and
threatened to withdraw the Lykian army from Troy.20
There were other ways, too, in which the need to conform with the duties
of xenia was impressed on men's minds and dominated their consciousness.
When Aeschines charged Demosthenes with the murder of his own xenos,
he called the deed an asebema, an impious crime, appealing thereby to a
whole array of mystical beliefs. 21 The horror inherent in xenoktonia - a
loaded word, the connotations of which can never fully be rendered by the
phrase 'murder of a guest-friend' - can be illustrated by the crimes with
which it was associated. Drawing a favourable contrast betWeen the polis
and the states of old, Isocrates asked:
Did there not abound in them [i.e. the states of old) murder of brothers and fathers
and xt1loi; matricide and incest and begetting of children by sons with their mothers;
feasting of a father on the flesh of his own sons... 22
Hesiod's Inferno was only slightly less terrifying. Doing wrong to a xenos is
here grouped together with cheating and stealing for gain's sake, lying with
one's brother's wife, offending against orphans, and abusing one's father.
Zeus himself is said to punish such criminals. 21 "D
(1)
The breach of obligations was a sin not only against the contracting party +-oJ
himself but also against the gods. When Herakles murdered his xenos, ..c
Q)
Iphitos, in his own house, the poet branded him as "damnable" (schetlios), I-
not paying due respect to the gods and the rules of hospitality. In consequ- >-
Q.
ence of this murder Herakles was afflicted with a disease which could only o
U
be cured by working off his guilt as a slave in the service of Omphale, queen
of the Lydians. 24 A similar offence was committed by Polymestor, ruler of
Thrace, in the Euripidean play Hecuba. Polymestor betrayed the role of
foster-father and killed Polydoros, the son of the Trojan Priam and Hecuba,
for the gold he carried with him. Hecuba in her grief is made to appeal to the
duties of xenia (nou bLxa ;tvorv;), and refer to the murder as a sacrilegious
deed. 2S In Diodorus' history, Agathokles is said to have suffered a terrible
lO Iliad 17. 140ff.; Themiltokles and Hektor broke the minimalisr ckmand not only of xnria
but perhaps of all amiable relationships in simple cultures: that people involved in 'positive'
relationships should help each o~r as much as possible and never injure each othn. On the
'norm of reciprocity', Itt GouldMr in Schmidt tt al. (eds.) (19n) 28·42.
II Anchines 3.224 (Against Q~s;pho").
u lsocrates S. 122 (To Philip).
U Works ad Days 32()'4, d. Aristophanes, Frogs 147ff. for a parody of this.
l4 Odys~ 21.26·8; Sophocles, Trach;"ia~ 68·72, 248·54, 274-6 (with Kholia); Apollo·
doros 2.6.2·3; Diodorus Siculus 4.31.5·8. Cf. Finley (1981) 116·17 for the implications of the
myth in a different context.
Ij Euripides, H~cuba 710ff.
punishment for the murder of his own phi/os and xenos, Ophellas of
Cyrene:
...for in the same month and in the same day on which he murdered Ophelia!' and
took his anny, he caused the death of his own sons and lost his own army. And what
is most peculiar of all, the god (ho theos) like a good lawgiver (nomothe'~s) exacted
a double punishment from him; for when he had unjustly slain one friend, he was
deprived of two sons, those who had 1>«n with Ophcllas laying violent hands upon
the young men. 26
In fact, the central plot of the Iliad - the abduction of Helen - revolves
around a similar sin. "You defiled me, wretched dogs·', said Menelaos to
the Trojans, "and your heart knew no fear at all of the hard anger of Zeus
loud-thundering, the xenos' god, who some day will utterly sack your steep
city".27 The abduction was a sin in itself and constituted, no doubt, an
offence against hospitality, too, but it was also - and, I suggest, chiefly - a
breach of guest-friendship. That is why, at the first encounter with Alexan-
dros, Menelaos addressed himself to the supreme deity as a betrayed xenos:
.-
Zeus, lord, grant me to punish the man who first did me injury,
brilliant Alexandros, and beat him down under my hands' srrength
that anyone of the men to come may shudder to think of
doing evil to a kindly host (xeinodokon), who has given him friendship.211
Xenophon's Anabasis furnishes an even more realistic account of such a C"I
I-
betrayal. Tissaphemes first made Klearchos his ritualised friend (homo- >-
trapezos). This gave Klearchos sufficient aSSurance to bring all the Greek 0-
o
generals to a conference with the Persian satrap. But then Tissaphernes U
treacherously murdered them all. The comment later made on this by a
Greek participant is revealing: he Tissaphernes did not even reverence
I i •••
Zeus xeniosj instead, having become Klearchos' homotrapezos, he made

that very act the means of deceiving and destroying the generals".29 When
the same Tissaphernes deceived Agcsilaos, the Spartan king claimed that by
violating the oath Tissaphernes had made the gods enemies of his side and
allies of the Greeks..l O
Glaukos' threat to withdraw the Lykian contingent and Menelaos' actual
revenge were typical. Apart from the sanction of public opinion and apart
26 Oiodorus Siculu5 20.70.3-". Nore rhar Agarhokles' sons are envisaged ro be in Ophellas'
enrourage.
1.7 Iliad U.623.5.
2. Iliad 3.351-4, d. Plurarch. Moralia 970. Zeus 35 rhe proreetor of strangers, guest·
friends. and supplianrs figures in dOlen! of passages; for example, Thcognis 143-4; Odyssry
9.266-7; 14.56; 14.414; Iliad 13.624; Plato, lAws 72ge ff., 953r, 950b; Charonda~ In
Stobaeus 2 p.IS!; Anchylus. F.umm;d~s 269-15; see furrher Gould (191.\).
2«1 Xenophon, A"abasis 3.2.4, d. 2.5.27.
}() Xenophon, Hellnr;cQ 3.4.11.
from an internalised sense of duty there was no agency outside the
framework of the relationship capable of enforcing obligations. The part-
ners themselves provided the sanctions. A man's whole moral personality
was in this respect at stake. Being left in the lurch was interpreted as an
affront to honour, and if one party ignored his obligations the other was not
only freed from all obligations but saw it as his own duty to punish the
offender. On the occasion of the soldiers' mutiny (in the course of the march
of the Ten Thousand),Jl Klearchos confessed to the soldiers both his shame
for having let Cyrus down and his fear that Cyrus might have seized and
punished him. But then, he employed Cyrus' desire to punish as an argu-
ment against the mutiny itself: And remember that while this Cyrus is a
U
valuable friend (ax;os ... phi/os) when he is your friend, he is a most

dangerous foe (echthros) when he is your enemy (polemios); furthermore,
he has an armament - infantry and cavalry and fleet - which we all alike see
and know about". 32 In most cases this desire for revenge coincides admir-
ably with calculations of expediency. In other words, it is by no means clear
whether it is the reason for, or a justification of, an action. Only myth
elevates the duty of revenge to a more idealised plane: Proteus, king of
Egypt, had to punish the Trojan Alexandros on account of a wrongful act
that Alexandros committed not against himself, but against his own xenos,
the Spartan Menelaos.JJ It was the entire institution of xenia, not merely his
own private interests, that Proteus was thus made to defend.
Not always nor everywhere, to be sure, was it possible to inflict such
punishments. But there were Other devices at hand by which xenoi could
exert pressure on one another. When, during the march of the Ten
Thousand, Xenias and Pasion deserted, Cyrus, contrary to common ex-
pectations, declared that he would not pursue and maltreat them. But he
reminded the generals that the wives and children of the deserters were held
under guard in Tralles. This veiled allusion to the families of the generals
themselves (presumably also maintained as 'guests' on Cyrus' various
estates) served no doubt as a tactful warning against further desertions.
Indeed, the courts of rulers in antiquity swarmed with the children of lesser
rulers,34 and this, I suggest, was another adaptation to ulterior purposes of
the foster-parenthood implied in xenia (Section 2.3). The whole institution,
JI Xenophon, ANabasis 1.3fE.

n Xmophon, AtUJbasis 1.3.12.
J) Herodotus 2.114-15.
34 In the Macedonian and Hellenistic coUI't'St this custOm became formalised and the sons of
nobles brought up with the royal princn (Syrttrapha'1 came to be known as Royal Pages
(basiliio; fHJiJ~s). Cf. Serve (1926) J 7ff.; Kienast (1973) 28ff.; Walbank (1984b) 227.
.j . • . _; :, ~ I: ;,
as reflected in the sources, is thus Janus-faced. Adrastos, son of Gordias, of

Phrygia, was banished by his father on account of a fratricide but taken into
the household of Croesus the Lydian by virtue of a friendship between the
two families. He lived there a princely life until his tragic killing of Croesus'
son. 3.~ Not so charitable, however, was the treatment of the son of Kerso-
bleptes, King of the Odrysai. Kersobleptes was first defeated by Philip, then
forced to surrender parts of his realm to other states and to hand over his
son to Philip. The Athenian Aeschines, who years later visited Philip's court,
defined the boy's status as 'hostage' /ft but there are good grounds to believe
that in a different context the boy would have appeared as a foster-son: for
we know of two sons of Kersobleptes by the names of lolaos and Poseido-
nios - names received, presumably, from Macedonian foster-parents.-"
Thus, the simulation of familiarity, rituals, sacred roles, divine sanctions,
gifts and services as well as blackmail and extortion were all involved in
generating obligations and sustaining the institution of xenia.
The binding character of these obligations cannot be illustrated by refer-
ence to any familiar modern parallel. But the analogy of other ritualised
relationships is suggestive. For example, an Andalusian motto says: "You
can refuse your compadre nothing; all that is yours is his".38 "Great is the
weight of the bond-friendship in this land", say people in Tikopia, one of
the Solomon islands, and the anthropologist who has described the institu-
tion explains: "The obligation to honour the bond is of a moral order, and a
betrayal by a saa bond-friend is condemned as wrong by folk who hear of
it,,:19 Among the central African Azande, "Open failure to fulfil the obliga-
tions of the pact brings upon a man not only magical retribution but also
public censure. He becomes an object of contempt to his neighbours and a
shame to his kinsmen". 40 Regarding Mexican compadrazgo, it has been
remarked that
H Herodotus 1..14-45. The same impression is conveyed by Xenophon's expressl)' rrop-

agandist stattment that" All the sons of the noblest Persians are brought up (paid~uo"tai) at
the king's court. There one may learn discretion and sdf-control (sophrosyn~) in full measure,
and nothing that is base can be either heard or seen" (A"abas;s 1.9.3.).
.\6 Aeschines 2.81-2 (0" the Embassy). For further examples, see Aymard (1954) and Amit
(1970) 143ff.
" Syll.·' 195. The two other sons of Kersobleptes had Thracia n names, Medistas and Tert'S.
For an lolaos serving as a bodyguard of Alexander IV, see IG 11 2 561. with SEG 26.86 and
Bull.Epigr. (1978) 145. Cf. also Demosthenes 19.230 (D~ Fa/sa ugalione), where Phrynon is
accused of having !lent his son to Philip even before he reached adulthood. This, of course. does
not mean that children could not be employed as hostages, like those sent by Peisistraros (0
lygdamis (Herodotus 1.64).
.18 Pin-Rivers (1973) 102.
.1 9 Firth (1936) 263.
40 Evans-Pritchard (1933) 387.
this connexion obliges the compadres and comad,es to hospitality and honesty and
all sons of good offices towards one another; and it is wonderful how conscientious-
ly this obligation is kept to, even by people who have no conscience at all for the rest
of the world. A man who will cheat his own father or his own mother will keep faith
with his compadre. 41
These parallels prove neither that xenia was immune to breaches (which,
as we have seen, it was not), nor even that in these societies there was a
correspondence between acts and ideal statements. What they show, in my
view, is that strikingly similar conceptions are present in other societies; and
that mostly, these conceptions belong to a framework of relationships
created by means of a ritual act between persons not related by blood. If we
see the norm of loyalty created by xenia as a particular variation within this
spectrum, then at least some of the difficulties concerning Alkibiades will
disappear.
.-
5.3 SERVICES
It will already have been observed that actions performed in the name of
xenia were not limited to any particular sphere of human activity: ritualised
friendship did not have a 'medium' of its own. It manifested itself partially
in any of the conventional categories of 'warfare', 'diplomacy', or 'trade', C"I
but was not fully contained in any of them. It embraced a range of co- I-
>-
operative acts as wide as one could possibly find in any human society: 0-
o
xeno; could be found providing each other assistance - and, it should be U
noted, substantial assistance - in solving family affairs; in avenging personal
grievances; in lending money; in offering shelter, refuge or asylum; in
ransoming each Other from captivity; in achieving political power; in sub-
verting governments; and overthrowing empires. The reason for this was
that ritualised friendship acted both as a substitute for and above all a
complement to kinship roles. The range (if not the intensity) of the ex-
changes carried out between xeno; was thus wider than that between
kinsmen or close intimates.
Apparently, there was no limit to what a xenos could do for another. Yet
one type of human activity is conspicuously absent from the list: low-class
activities most immediately concerned with getting a livelihood. To be sure,
xeno; did provide each other with grain and some other necessities of life.
But they were not dependent for their survival on such provisions. Ritual-
ised friendship was concerned with extracting spoils for people situated high
above the subsistence level. The services they dispensed belonged to those
41 Minn and Wolf in Schmidt et al. (eds.) (19n) 1.
3 Services 129
upper-class pursuits to which, to use Veblen's language, "a degree of hon-

Our auaches".42 The broad classificatory terms by which they arc denoted
show this: euergesia, charis, boetheia - concepts which, in marked contrast
with the universally despised banausic activities, were treated with automa-
tic approval and admiration. 'Services' within the context of ritualised
friendship should thus be understood as acts within the limits of what was
deemed honourable and consistent with the aristocratic self-image.
Even within these limits, however, the list comprises a variety not easily
amenable to classification. For one thing, all types of service could be found
within the framework of one single relationship. For another, even a service
performed within a clearly defined sphere impinged on other spheres.
Nevertheless, the services flowing between ritualiscd friends can be grouped
into ritual services, services restricted mainly to the private domain, and
those which mainly affected the public (d. Section 4.2, p.Rl). These distinc-
tions have the double advantage of introducing some order into this be-
wildering variety and enabling us to see how the evolution of the civic
.-
system was related to nctworks of ritualised friendship. In what follows, I
make only a shon reference to services already mentioned in other contexts
but describe in some detail services not encountered so far.
Ritual services. Into this category fall (a) services mimicking kinship
relations: foster-parenthood, the seuing up of funerary stones, the provision C"I
I-
of dowries, the bequeathing of property (Section 2.3), revenge (Section S.2); >-
0..
(b) bestowal of symbolic gifts (Section 4.2); (c) the composition of epitaphs, o
poems and of other literary pieces designed to express one's emotions U
towards one's xenos and spread his renown (Section 2.3). Pindar points to
an explicit link between xenia, love, and the duty to sing a xenos' praise. 4 .1 It
should not thus come as a surprise that a considerable portion of his poems
were written in honour of xenoi; nor could it be a matter of coincidence that
the Agesilaus and Cyropaedia were expressly written for the glorification of
the two great xenoi of Xenophon's life, Agesilaos of Sparta and Cyrus of
Persia.....
Private services. Here belong chiefly: (a) acts of opportune intervention
(boetheia) such as ransoming from captivity, providing shelter and mainte-
nance, giving hospitality; (b) lending money or acting as a keeper of one's
property; (c) giving expert assistance. The ability to provide the service
stems in this last case from a skill or profession that a person had acquired-
42 Vehlen (1970) 21-2.

4.' Pindar. N~m~an 7.90-2.
44 For a striking parallel in the COntext of bond-friendship in a simple tribe. see Firth (1936)
269((.
physicians, writers and artists loom large in this category - rather than from
his possessions or his advantageous position in the social structure.
Political services. Through these services, civic institutions and networks
of ritualised friendship became thoroughly intertwined. An examination of
this aspect of ritualised friendship yields a surprising picture. The elites of
the ancient world, as we have seen, were engaged in a massive work of
co-operation and exchange across political boundaries. The items they
traded, however, included not only private goods and private services but
also the interests of their entire communities: as it will be demonstrated in
Section 5.5, civic affairs were frequently subordinated to the private in-
terests of ritualised friends. In other words, to contemporaries these net-
works often appeared as organisations of supply for the individuals in-
volved in them rather than for the cities to which the individuals (or one of
them) belonged. Thus, the horizontal ties which linked the elites across
political boundaries appear at times stronger than the vertical ties which .-
bound them to their inferiors within their own communities. To substanti-
ate this claim, a reassessment of some key concepts is necessary. In what
follows, I hope to show that when ancient writers wrote of proxeno;,
epitedeioi and philo; in other cities, they meant something different from
what is customarily thought. Proxeno; may be considered first.
C"I
I-
>-
5,4 XENIA and PROXENIA 0-
o
U
The proxeny is by far the most copiously documented political institution of
antiquity. It is attested by thousands of inscriptions on stone or lead stem-
ming from a geographic area roughly coterminous with the Greek world
and covering a temporal span from the seventh century B.C. to the second
century A.D. Proxeny decrees are remarkable documents, for which it is
hard to find parallels in other cultures. Irrespective of the time or place of
publication, they display a series of features in common.
The grantor of a proxeny decree is always a community, mostly a city but
sometimes a civic subdivision or cult association. The grantee, the recipient
of the title of proxenos (and of some other titles and privileges which go
with it), is always a stranger, an outsider. This general rule does not allow
for exceptions. In all known proxeny decrees, the grantee originated invari-
ably from communities distinct from those which granted the title. 45 In
4' Cf. Marek (1984), me catalogues at 8·118. The dmri#toPf of the recipients is never
omitted from the decrees.
4 Xenia and Proxenia 131
other words, no citizen could ever become a proxenos of his own commun-
ity. This distinction between 'inside" and 'outside' persisted even after the
grant of proxeny. For the grantees were expected to further the interrsts of
the granting community either in their own communities or in a place which
(in relation to the granting community) constituted 'abroad'. An explana-
tory note to a fourth-century oration illustrates this well: "Proxenoi are
those who in their own fatherlands look after [the affairs of] other cities". 46
Concrete examples show that, as with xenia, this regularity did not allow
for exceptions.
Some proxeny decrees are extremely short and uninformative, others are
long and contain a host of minute details. But whatever their length, they
are set out within an extremely rigid stylistic framework: with rare excep-
tions, epigraphists have no difficulty in classifying an inscription as 'pro-
xeny'. Most decrees contain an introductory clause identifying the commun-
al agency issuing the document; a middle clause naming the recipient and
describing the acts by virtue of which he was given the honour; and a final
clause listing the honours and privileges which he would be given in return.
A good example of a typical decree (Fig. 11) would be this inscription from
Athens:
Gods
Resolved by the Boule and the Peop-
le, Antiochis held the prytany, Euk-
leides was Secretary, Hierokl-
es presided, Euktemon was Archon [=40817 B.C.].
Dieitrephes made the motion: Since
he is a good man, Oiniades of Pal-
aiskiathos, to the city
of the Athenians, and zealous to d-
o whatever good he can, and is of b-
enefit to [any) Athenian who comes
to Skiathos, commendation shall be given
him and he shall be recorded
as proxeno$ and euergetes of the Athe-
nians, and his descendants [as well); a-
nd that he be not harmed shall be the
responsibility both of the Boule, whichever i-
s in office, and of the gener-
als and the Athenian governor in Ski-
athos, whoever he may be on each occasion. This d-
ecree shall be inscribed by the s-
ecretary of the Boule on a stel-
46 Scholion to Aeschines 3.138 (Against Ct~s;Pho,,). See Marek (1984) 393 for other
lexicographical definitions.
e of marble and shall be set up on the A-
cropolis. He shall also be invited fo-
r xenia to the Prytaneion for t-
omorrow.·'
Proxenia was not known to Homer; only xenia was. And xenia, as
demonstrated in this study, was an essentially private institution. The ear-
liest document containing the term proxenos (more precisely, proxenwos),
dates from the last quarter of the seventh century B.C. It is a funerary
monument erected by the people of Corcyra for one Menekrates of Dean-
thea (LocriS).48 In the verse epitaph inscribed on the stone, the Corcyraeans
called Menekrates "a dear proxenos of the demos" - thereby marking a new
departure. Its significance lay in the representation of a collectivity of people
as an individual person wielding authority and conducting international
diplomatic activity (ef. Fig. 12 with Fig. 5, p.52, for the appropriation of the
visual symbolism of xenia by the city). Once established, this pattern was to
persist until the disappearance of the institution some eight centuries there-
.-
after: the proxeny will remain throughout this period an agreement between
a community of people personified as a single individual, and a 'reaP
individual outsider.
Revolutions, as opposed to invasions, are seldom capable of bringing
about a total rupture with the past. Earlier forms are mostly remodelled, C"I
I-
renamed, and re-adapted to suit new circumstances. The institution of >-
0-
proxeny was not, I suggest, exceptional in this respect. The earlier institu- o
tion was xenia; the new circumstances were the self-conscious communities U
of the city-state; and the result was proxenia. In other words, proxenia was
a communal invention using as a model xenia. 49 Someone must have hit
upon the idea of supplying the newly constituted communities with a device
similar to that which served the lords at whose expense the community had
gained power. We can even catch a glimpse of an intermediary stage
between the xenos-xenos and the city-proxenos stage which might have
facilitated the invention: powerful individuals, preferably rulers, would
47 IG I' 110, adapted (excludang lines 26·31) from the tran,lanon in C.W. Fornara, Archaic
Tim~1 to the End of tilt Pdopo"ntsia" War, 2nd edn. (Cambridge, 1983) no. 160. The format
of the translation imitat~ as clowly as possible the format of the original; hence the unconven-
tional hyphenation.
411 Meiggs and Lewis (1969) no.4.
49 Marek (1984) 387 challens" the traditional view (for a summary of which, see Gschnit·
zer (1973) 632) and denies any ronMxion betwttn proxeny and what previous scholars called
Gastrecht on the grounds that 'hospitality' nlcompassn a range of dutin tOO narrow to have
given rise to the all-purpose proxeny. But if we repla« 'hospitality' with 'rituali~ friendship',
then thne difficulties disappear.
Fig. 11 PrtUCerl~ and :.rem.

Athenian decree from the year 408n B.C. granting the ritle of proxmos and
euerget~ to Ointades of Palaisldathos (lG IJ 110 = M.B. Walbank (1978) no.87).
Proxeny decrees provide a good evidence of how proxntia evolved from xnt,a:
the m:ual declaration of xenia (". make you my .lreJ'los") is somewhat modified ("he
shall be rec:orded as pro~eno$ and ~rgete$ ~f the Athenians") but tM- underlying
belief IIllt if cre~ttS a pseudo-kinship relationship remains uncharaged (proxnria,
like xenia, is thought to pass on to the descendants) Funher, the pre-state, ritual
declaration is incorporated within a new formulary framework - eloquent evidence
of the Clry's political insritutions - and derives irs binding force not from a mystical
uttcran~ but from an explicit resolution of the demos.
ro
"C
..,
(I)
ro
~
"tJ
(I)
~
.s:::.
0)
'C
>-
C-
O
U
fiB- 12 The dty .. XaIOI

Marble relief above Athenian inscnprlon recordlnlme treacy of friendship between
Athens and Samoa in me year 405 B.C. (lG I J 127). The alliance is visualised as a
paa of xenia sealed wlm a rirual handshake between me patton goddesses of me
two tiUeI, Amena and Hera.
, • I .: , • ,.,~, : : '. ~ .',.' ,
sometimes be made philoi, xenoi and ~uergetai of enrire communities. 50 The

polis carried this custom one step further by formalising it as proxenia and
by integrating it into its political procedures.
That this was, indeed, the evolutionary process followed is shown by the
analogies between xenia and proxenia: the points of resemblance are simply
too many and too manifold to be fortuitous. Cities, like xenoi, performed
ritual services for their proxenoi: a city commemorating its deceased pro-
xenos with a tombstone and a verse-epitaph (Fig. 13) mimicked a
Xenophon or Simonides doing iust the same for their own xeno; (Section
2.3, p.26). Cities, like xenoi, envisaged the establishment of a quasi-kinship
relationship that devolved automatically upon the descendants and was
theoretically unbreakable. 5 I Cities, like xenoi, imagined themselves to be
protectors and foster-parents of their own partners and of their partners·
offspring (see Fig. 14, and d. with Figs. 1a-d, pp.24-5).Sl Proxen;a also
imitates xenia in being an all-purpose relationship: a proxenos was a
general assistant who could be called upon in case of need and whose duties
were unspecialised: above all, he was an euergetes, a benefactor. 53 Finally, the
proxeny decree itself is an elaborate, written version of the ritual of initiation
between xeno;. For what are the privileges conceived of as gifts (doreal1
given by a city to a proxenos, if not a communal version of private gift-
exchange? What is the meaning of the underlying sequence of the key
concepts (present explicitly or implicitly in hundreds of decrees) - that
erte"gl's;a cngcndcrs :I wish to confer :I cha,.is equivalcnt to the eue"gett!mQ -
JO For ~xamplt. Odyssry 9.18 (Ody~us, xmos of rhe pt,aucian~); Pindar, Pythian 9.82
(Amphirryon, xmos of the Spartans); Herodotus 1.22 (xenia and sym",achia ~ecn AlyatttS
and the Mile5ians); Herodotus 6.34 (Miltiades d~ a good deed for the Dolonci and is invit~d
to becom~ th~ir tyrant); H~rodotus 1.67 (th~ agathoergoi of Spana). d. Diodorus Siculus
11.18.1·2; H~rodotus 3.88 (Darius. xenos of the Arabians); Herodotus 1.69 (Cronus. xmos
01 the Spartans); Herodotus 8.120 (Xerx~s condudes a xenia with th~ city of Abd~ra);
Herodotus 7.116 (Xerxes, xtnos of the Aanthian~); Herodotus .1.55 (Archias. xmos of
Samians); Herodotus 5.63,90,91 and Aristotle, The Athenian Constitution 29.4 (th~ P~isistra
tida~ of Athens. xeno; of th~ Spanans); Thucydides 1.]]6 (Th~mistokles. euergelts of Cor-
cyra); Xenophon, HelleniC4 6.5.4 (Ag~~ilaos. patriltos phi/os of the Manrin~ans); X~nophon.
Cyropaedia 3.3.4 (Cyrus. euergetes of the Armenians).
H Daux (1937); Gschnitl~r (1973) 687; Ma~k (1984) 161. The rechnical t~rm for h~redit
ary proxeny (patrios proxenia. for ~xample. Syll. J 503) imitates patrilte xenia. Kallias' speech
in Spana (Xenophon. Hellenica 6.3.4) provides a good example of the beli~f in this continuity:
"Men of l.aceda~mon. as regards the position I hold as your proxmos. I am not the only
member of our family who has held it, but my father's father received it from his father and
handed it on to his descmdants".
H Examples abound of decrees stipulating protective clauses and threatening to punish
those who cau~ harm to a proxmos: IG I) 19; 27; 65; 156; 162; 164. In IG (12 111 (36312
B.C.). Athens stipulated that the newly constituted government of Ceo! exccure the man
responsible for the death of an Athenian proxmos.
H Euergetes as an honorary tirle. like proxmos. was also to some extent incompatible with
citizmship: in Gr«ce at large (but not in Asia Minor) it rended to be bestowed on foreigners
only; see Moretti (1977) and Vqne (1976) 192.
-; :, ~ ;,
Fig. 13. The dftDOI moums a pro1CettOS

Athenian grave monument for Pythagoras of Sclymbria (IG 1 :J 1154 - M.B.
Walbank (1978) no. 9, ~ 460-450 B.C.).
The rihlal obligation of %mol to bury and commemorate a dead partner (Section
2.3) was taken over by the city and applied to its relationshipl with proXerlO~ The
verse epitaph inscribed on the rectangular baR of the stele tells us that Pythagoras of
Sclymbria was buned in Athcnl at public CXpeDlC' because of has own and his
ancestors' excellence (a,~u) as proxmoi.
if not that the private etiquette of reciprocity had been transferred to the
communal level? And finally, what is the meal (xbtra) to which the city
invites its proxeno; if not an institutionalised 'fenion of the feasting which
sealed guest-friendships?s"
... In chis 1Upec"''t roo, the distinction between citizens and fompen wu meticulously
maintained~ hononnds awarded citizenship were inYited to dftprtorr: prourroi and non-cilizen
honorands, to xhri4, d. Henry (1983) 262·75.
fig. 14 The proxcnfM as 1& fOlte~-ehild

Relief above an Athenian draee (c. 430-415 B.C.) recording a grant of proxeny to
Proxenid~ son of ProxcnOl, of Cnidus (lG I,) 91 - M.B. Walbank fl978l no.64).
The female figures represent the patron-goddesses of Cnidus and Athens, Ancmis on
the left and Athena on the right. The honorand Proxenidcs appears as a foster-child
placed by has real parent under the tutelage of the foster-parent. A comparison with
Fag. la-d, p.24-S, will show Ihe samilarilics Wllh xmia.
Apan from these superficial similarities, however, [here are mor~ pro-
found points of intersection between proxen;a and xenia.
There were only two ways in which a foreigner could become a proxenos:
he could mherit the title, or he could be recommended to it by an inda~idual.
To be sure, the extut sources convey the Impression of a third way, which
can be illustrated by the Case of Nymphodoros of Abdera. Thucydides tells
us that the Athenians made Nymphodoro$ their proxenos since he was a
confidant and father-an-law of Sitalkes, King of Thracians, whom the Athe-
nians wished to bnng into an alliance. This, however, should not mislead us
into believing that the mitiative for the award of proxeny came from the
Athenians as a whole. What happened in practice was, presumably, that an
Athenian friend of Nymphodoros supplied the officials in power with the
necessary information about Nymphodoros' connexions. Upon realising the
advantages that might accrue to Athens from such a move, the officials
indeed set in motion the political machinery for awarding the proxeny, and
only this aspect of the affair is recorded by Thucydides. In brief, for a
stranger to become a proxenos, he needed a mediator who enjoyed the
citizens' confidence and who could recommend him to those in power. 55 If
he inherited the title, he had to be recommended too: proxenia, like xenia,
lasted for ever theoretically, but had always to be renewed in practice. 56
The most natural candidate to make such recommendations was a man
who was at one and the same time a citizen and a xenos. I propose to defend
this view by showing that a series of considerations, each of which is in itself
inconclusive, in fact fall together to form a coherent picture. But first a
methodological note is appropriate. It seems to me misleading to approach .-
the problem through proxeny decrees alone: their capacity to convey in·
formation is limited by their stereotypical framework; unlike the literary
sources, they cannot tell us about the bargaining and manipulations that
preceded a grant. Like marriage certificates today, they reduce a complex of
human affairs to a banal formula. To gain a three·dimensional view of the
process, other kinds of evidence must be brought into the discussion and C"I
I-
sometimes given more weight than the decrees themselves, even if in quan· >-
0-
titative terms they are less impressive. o
The first argument involves a logical exercise. It is true that we have no U
detailed information about the 'communal revolution'. But we know this
much: that the pre·political world was interlaced with a network of xenia
alliances binding the elites of different communities together; that the
community constituted itself as a final authority by reducing the power of
these elites and by incorporating them into the citizen body; that the
community devised a new institution by which to regulate its relationships
with the outside world, the proxeny. The question then arises: How was
proxenia related to xenia? There are two possible alternatives: either the
community established a formal, impersonal and official tie with individuals
JS Nymphodoros' mediation: Thucydides 2.29. For the peculiarity of Sparta in this rnpect
- che appoinrmenr of Spamn citizens as proxmo; of foreign communities by the Spartan kings
themselves (Herodotus 6.57.2) - see Mosley (1971 b).
J6 There is a mealins differma between the renewal procedures in the two contexts: xmUl
was renewed by means of a ritual aet (d. Section 3.4); proxmi4 by means of a public
resolution. Cf. Marek (1984) 161 for examples of rmewal clauses in proxeny decrees; Daux
(t 937) for Alkibiadcs' proxmy chat was renounced but remained valid nonethdm (Thucy-
elides 5.~J .2).
4 Xenia and Proxenia 1.19
abroad, or it simply harnessed the existing private ties of xenia for its own
interests. The first hypothesis would mean that the administrative machin-
ery of the Greek state somehow prefigured the rational and impersonal
bureaucracies of the modern age, and has to be discarded at once. Even in
historical times, Greek states displayed a basic mistrust of ties which were
formal, temporary and official: time and again, they included in their
embassies people who already stood in some warm and enduring rela-
tionship with the persons to be approached. 57 It is unlikely that things were
organised differently at a time when the city, the citizenship and the whole
notion of community were still in their embryonic form. On the other hand,
what could be simpler than to use existing private connexions for public
purposes? The model that presents itself, therefore, is this: in order to
extend their power abroad, communities first conducted affairs with a
foreigner through the intermediary of one who was both a citizen at home
and a xenos of the foreigner; next, they made this foreigner into a proxenos,
bringing him into relations of mutual obligation and dependence with the
.-
community itself, independently of his private bonds of xenia. Thus it came
about that networks of proxen;a; roughly coincided with the networks of
xenia; to which they owed their origin. If the appointment of proxeno; were
independent of xenia networks, it would be hard to explain why there are
sometimes several proxeno; of the same city in another city..~8 The multiple C"I
I-
proxenies must have corresponded to the multiple ties of xenia between the >-
0-
citizens of the respective cities. o
The next consideration is linguistic. The argument is that to the users of U
the language it simply seemed self-evident that proxen;a implied xenia. A
passage from the 15th oration in the Demosthenic corpus can illustrate the
whole pattern. In order to gain an understanding of the usage, a short
outline of the events leading up to the incident is appropriate. The revolt of
Rhodes against Athens in the early 60s of the fourth century constituted a
serious blow for Athens, all the more so since at the time of the revolt
Rhodes was under democratic rule. The rebellious democracy, however, did
not survive for long. In 355 B.C., cenain oligarchs, actively supported by
Mausolus, the powerful potentate of Caria, overthrew the regime and set up
an oligarchy. The coup was followed by the most commonplace practice of
Greek politics, the banishment of defeated rivals. Two years after these
events, in 353 B.C., we encounter the banished democrats in Athens. The
51 Cf. Perlman (1958) 187; Humphreys (1978b) 100-1.

.u For example, Ari~ophanes, Birds 1021; Euripides, Ion 55 1, 1039 and ~ndromach~
1103; Welles (1934) no.3, para.S (of Teas in lebedos); Thucydides 3.2.3 (of Athens in
Mytilene); Thucydi<ks 3.70.1 (of Corcyra in Corinth).
very men who beforehand instigated the secession of Rhodes from Athens
now appealed to Athens for help. Not surprisingly, the response of the
Athenians was chilly. It was Demosthenes who tried to induce his country-
men to adopt a different course of action. Wishing to persuade the popular
assembly to help the Rhodians to return to their city, Demosthenes said,
among other things:
I should never have made this proposal, had I thought that it would benefit the
Rhodian democrats alone, for I am not a proxmos of the men, nor is any of them
privately my xmos."
What Demosthenes meant to say thereby was that his proposals to help
the exiles stemmed from genuine concern for the Athenians' welfare, not
from private obligations or from an expectation of benefits. The casual
nature of the remark betrays the built-in expectations that proxenia and
xenia went hand in hand. Some other examples also show that the one
followed from the other. But the clearest indication comes from rhetorical
figures. Reviling his arch-enemy Meidias, Demosthenes once said:
.-
Is there a vote on? Meidias of Anagyrous is put forward as a candidate. He acts as a
proxmos of Ploutarchos (nAo'VtciQxou ~£v2i). He knows all the secrets; the city
is too small for his aspirations. 60
The technical impossibility of a man being another man's proxmos is here
skilfully exploited to convey a package of ideas which can only be unravel- C"I
I-
led by a cumbersome paraphrase: Meidias of Athens and Ploutarchos, >-
0-
tyrant of Eretria, are known to be xeno;; 61 Meidias manipulates this xenia o
for private gain by sacrificing the city's interests (i.e. by revealing its secrets U
to Ploutarchos); to make the malpractice appear acceptable, Meidias con-
ducts his manipulations under the guise of proxenia. 62 The oxymoron
would have been pointless if proxen;a did not imply xenia.
The final considerations are founded on statistical probability. In Appen-
dix B, an attempt has been made to relate grants of proxenies and of
citizenship of the classical age to existing ties of ritualised friendship. It
seemed advisable to include grants of citizenship for two reasons. First,
because they carry the idea of proxeny one step funher: they not only
" IXmosthenes 15.15 (For th~ LiMrly oftM Rhodi4ns). a., for similar usage!, Too (1933)
no.34, lines 11-12, discussed above, p.4; FGrH 392 (Hermailaos of Chios) F6; Pau5anias
3.8.4.
60 Dm10sthenn 21.200 (Against MMias); my translation.
" It is explicitly stated in Demosthmes 21.110. For the background of the incident. see
Brunt (1969).
'2 Further example!: Aeschine!2.191 (On t"~ E",baSlY) (Demosthenes betrays the intern"
of the city to Thebes, whose proxnros he is); Dcmosthmes 18.82 (0" 1M era"",) (Anchines
aets as a ·proxenos· of ambassadon who come from Philistides of <>reus and Kleitarchos of
Eretria).
4 Xrnia and Proxenia 141
convert the grantees into ext~rnal allies but effect their actual incorporation
into thr communal group. Secondly, citizenship grants were relatively rare:
for thr approximatively 94 rpigraphicaJly attrstrd grants of proxeny in
fifth-crntury Athens, there are only 6 ~pigraphically attc:sted grants of
citizenship.b) If, indeed, xenia ties tum out to undrrpin the grants of these
much more jealously guardrd prerogatives, we might rxpect a similar
mechanism to apply to the grant of proxenies.
In identifying ties of xenia, use has been madr of the attributes of the
bond drscribed in the prrvious chapters. Thry include: (a) the explicit
mention of xenia or philia between individuals belonging to diffrrent com-
munities; (b) rderences to thr exchange of benefits betwern thr grantre and
a contemporary citizen (or between their ancestors); (c) naming (as de-
scribed in Section 2.3, p.19ff.). By using these pointers, 23 cases have been
identified in which it can be said with varying degrees of certainty that the
proxeny or citizenship had been secured through the intervention of a
citizen-xenos (Appendix B).
.-
As compared with thr hundreds of surviving decrees honouring for-
rigners, this figure might seem to argue for the insignificance of xenia ties.
Funher thought, however, suggrsts that thr contrary might be the case. To
havr a fair comparison, the analytical groups should be cardully idrntifird,
and M.B. Walbank's collrction of grants of proxc:nies from fifth-century C"I
I-
Athens may offer a useful sampling base. The 94 grants included in this >-
0..
study may be dividrd into two categories: 84 grants attested only o
epigraphically,64 and 10 attested both rpigraphically and in litrrary sources U
(or only in thr latter).65 It might at first sight seem ho~lrss to try to infer
ties of xenia from proxeny decrees alone. But even in the inscriptions
included in this first category there are three in which thr namr of the
grantre points to a possible citizen-xenos. 66 Given the facts that, first, two
people could well be xeno; without resorting to the custom of naming;
secondly, that there might be other cases of homonymity which, brcause
thry do not belong to such prominent citizens, escape our attention; and
thirdly, that in 38 out of thr 84 inscriptions included in this catrgory thr
name of the grantee is illrgible 67 - thrre is a fairly high number. In a fourth
6_' Cf. M.B. Walbank (1978) nos. 1·94 with Osborne (1981-83) nos. 1·6. some of which are
collective grants. For the whole period c. 420·140 B.C., Osborne has collected about 118
grants of Athenian citizenship; grants of proxmy for the same period come c10~ to one
thousand.
64 Nm. 3·4. 6·17.19·29.31-44.45 (but d. p.252). 46. 48·59, 61·5, 67·9, 71·89, 91-4.
65 Nos. 1.2,5, 18. 30.47. 60. 66, 70, 90.
66 No. 11. a Thespian called Athenaios; no.32. Nikias of Crete. presumably a xmos of the
Athenian general; no. 74, Thoukydidn of Pharsalus, prnumably a x~"OJ of the historian's
family.
67 Nos. 10. 13. 20, 21, 23. 26, 28, 31. 33-7, 40·2, 44, 5 I, 53·9,62.69. 71, 72, 76, 77, 79.
83. 84. 89. 92, 93. 94.
142 Obligat;ons: Hero;c and Civic
inscription from this category, it is possible to infer from the content itself
the existence of a hereditary relationship between the honorand, Apollodor-
os of Selymbria, and the proposer of the rider, Alkibiades. 68 Indeed, the
suspicion that xeno; were involved in securing the grants is confirmed if we
move to the second category - grants attested also (or only) in literary
sources. Here, there are only 4 cases out of lOin which the meagre literary
evidence does not point to a possible xenia connexion. 69 It is therefore hard
to accept Marek's conclusion that most of the proxeny grants served the
political interests of the community and were largely independent of private
interests.'o Communities no doubt perceived it to be to their own advantage
to make use of the connexions of their influential citizens; and it also seems
clear that the grantees did benefit the communities either directly - by giving
a gift or performing a valuable service for the whole community - or
indirectly - by helping out their own xenos in a situation in which he
represented communal interests.'1 There was thus a cenain element of
symbiosis between public and private interests. In general, however, we
shall see that the interests of xeno; only rarely coincided with civic interests:
structurally and inherently, they were opposed to them.
5.5 FOLLOWERS, FACTIONS AND THE CITY
We can test this claim through the evidence of Thucydides. Thucydides is an

especially valuable source for our purposes since by every conceivable
criterion the polis-system achieved in his lifetime the peak of its evolution.
The kings of the mythical age, the petty chieftains of the dark age, as well as
the tyrants of the archaic age, had all disappeared from the scene. And the
days had not yet come when Hellenistic monarchs controlled the cities'
destinies through their network of agents. Most communities in Thucy-
dides' time were independent of extraneous powers (unless subjected by
force) and expressed their collective will through "institutions which sub-
jected even the most powerful men to formal organs and rules of
authority".72 And yet, even in these circumstances, the interplay between
ritualised friendship and the civic system was far from negligible.
•1 No. 86.
" Nos. 30, 47. 60. 66. In nos. 1.2,5, 18, 70 and 90. the xnrilJ is evident from the literary
sources, 1ft Appendix B.
'0 Marek (1984) 134.
,. For example, when Byzantium admitted the winning Spanan gmeral. l.ysander. in 40S
B.C., thole Byzantines who had beforehand betrayed ~ city to Alkibiadcs went to Arhms and
were given citizenship (Xenophon. H~lInr;c4 2.1.3).
12 Finley (1979) 90, his italics.
The conflict between the community and powerful individuals forms a

theme running through Thucydides' history from beginning to end. It is
unnecessary to present all the examples here. For a pattern to emerge, it will
suffice to examine a selection. Three xenoi-dyads are singled out: Perikles of
Athens and King Archidamos of Sparta; Alkiphron of Argos and King Agis
of Spana; and, finally, Alkibiades of Athens and Endios of Sparta. We may
stan with Perikles and Archidamos.
There is, in Thucydides' history, a story of a strange incident which has
not so far received a satisfactory explanation. 73 It concerns the first invasion
of Anica by the Peloponnesians in the year 431 B.C.:
Here [in Oenoe, Attica] the Peloponnesians prepared to make an assault and sp~nt
much time with nothing to show for it. For this Archidamos was severely criticised.
Even during the period leading up to the war people had thought him weak and
friendly to Athens (Toi~ 'A9Jtvaim; t1tLnlbElo;), because he had not spoken in
favour of a full-scale war effort. 74 And in the period after mobilisation his reputation
had suffered still further because of the delay at the Isthmus, and the slow progress .-
of the march afterwards. The present hold-up in front of Oenoe seemed worst of all;
for during this time the Athenians were bringing their property into the city, and the
Peloponnesians considered that, if it had not been for the delaying tactics ofArchida-
mos, they might have made a rapid advance and found all this property still outside
the walls. Thus, during the siege of Oenoe, there was bitter feeling in the army
against Archidamos. The reason why he still held back was, they say. because he C"I
expected that the Athenians would not face the idea of letting their land be devas- I-
tated and would make some conciliatory gesture while it was still untouched. 75 >-
0..
Archidamos' dilatoriness, described by Thucydides in such striking terms,
o
U
is, in my view, related to his xenia with Perikles, and I propose to suppon
that view by showing that this and some other sections of books 1-2 simply
make no sense unless we accept it. On the other hand, if we do accept this
view, then the problematic sections fall into a pattern which is logically in
accord with similar sections in Thucydides and factually consistent with
what we know of ritualised friendship from other sources.
The obvious alternative explanation is the one implicitly accepted in
modem research: that Archidamos, being a man of intelligence and modera-
tion, was opposed to the war on practical grounds. He simply foresaw that,
given the prevailing balance of power, Sparta could not win it, and therefore
resorted to dilatory tactics.
It is true that this explanation cannot be totally discarded: Thucydides
, J Thucydides 2.18, ci~d here; this should be ~ad with the comments of Gomme (1945 -81 )
vo!.U, 67ff.
74 The rderencc is to Archidamos' conciliatory speech in Thucydides 1.80-85.
7$ Adapted from the translation by R. Warner, Thucydid~s: Hislory of the Ptlopo,.n~sian
War (Harmondsworth, 1972), my italics.
himself makes somr allusions to it. But hr makes rven clrarer allusions to
the view propounded here - in fact, he almost states it explicitly. For what is
the role of 2.13, in which the xenia between Perikles and Archidamos is
revealed? A considerable mental effon is needed to grasp the full impon of
this fact, to realise that the, to us, unusual characteristics of ritualised
friendship were for Thucydides self-evident. He took it for granted that
powerful individuals from different communities were involved in networks
of alliance; that a continuous flow of gifts, resources, favours and services
was the compulsory expression of this involvement; that the alliances were
structured by an aristocratic code of conduct and buttressed by an heroic
ethos; and, finally, that they were inherently opposed to communal in-
terests. So obvious a pan of his reality were these features of ritualised
friendship, that it sufficed to indicate them all with the aid of a few simple
words. And that is precisely what Thucydides does in 2.13.
One qualification is necessary. It might legitimately be asked, why, if this .-
was the case, Thucydides did not make an explicit refrrence to it. In my
view, the reason why he docs not say that Archidamos' tactics flowed from
xenia is that he did not have such information as would pass the test of his
critical methods: he guessed it. And thus, instead of stating the connexion,
he suggested it. By constantly presenting Archidamos' actions as seen by the
Peloponnesians - Uthey said", Uit was thought", are the recurring formula C"I
I-
- Thucydides makes it clear that neither he, the story-teller, nor Archida- >-
0-
mos, the actor, shared this view. On the other hand, the xenia itself he o
presented as an objective fact. What he signified by this fact was a general U
obligation to help one's panner - an obligation which, by the standards of
action to which Thucydides was accustomed, did not exclude the possibility
of subordinating the entire interests of the Peloponnesians to it: this possi-
bility is foreshadowed by the examples of Pausanias and Themistokles in
Book 1, and is fully realised in the case of our other two xenoi-dyads.
Archidamos and Perikles, in a similar fashion, pursued a cenain policy
which tried desperately to reconcile their private involvement with their
public responsibilities. Both were criticised by the demos for not making
satisfactory representation of communal interests. And both had in the end
to sacrifice their xenia in order to remain in power. By a refined mastery of
composition, Thucydides thus interweaves the story of the escalation of the
rivalry between Athens and Spana with thr story of the tragic friendship
between Perikles and Archidamos. If Archidamos' behaviour indeed flowed
from his cautious character, then chaptrr 2.13, in which the xenia is
revealed, would be a casual aside providing a mere anecdote, an almost
redundant piece of information. If, however, the view propounded here is
accepted, then chapter 2.13 becomes a clue to the behaviour of the two
leaders, a central motif which runs through the first two books.
The events leading to our next incident can briefly be summarised.
Following the peace of Nikias in 421 B.C., a major reshuffle of political
alliances took place. In 419 B.C., Athens entered into an alliance with Argos
and her allies, Elis and Mantinea. The quadruple alliance at once initiated a
series of expeditions against Epidaurus, a city allied with Spana. Thucy-
dides describes the Spanans' motives for taking counter-measures in ex-
ceedingly lucid terms: first, her Epidaurian allies were in distress; secondly,
some of her allies in the Peloponnesus had revolted; finally, other allies were
not well disposed towards her. Thus, for fear of more desertions, the
Spanans assembled an army which Thucydides would later describe as "the
finest Hellenic army that has ever been brought together ... a force that
looked as though it could hold its own not only against the Argive League
but against another such League in addition". This army, now, under the
leadership of Agis, son of Archidamos, marched against Argos. The two
rival armies, the Peloponnesian and the Argive, were forming up in order of
battle, the Peloponnesians securing for themselves a more advantageous
"D
position, when our remarkable incident occurred: (1)
+-oJ
The greater part of the army of the Argives and their allies, so far from realising how ..c
Q)
dangerous their position was, thought that they were going to fight on very favour-
I-
able conditions, with the Spartans cut off in their own country dose to their city. But >-
two men among the Argives - Thrasyllos, who was one of the five generals, and Q.
o
Alkiphron, who was proxmos of the Spartans - just when the armies were on the U
point of meeting, went forward, held a conference with Agis, and urged him not to
bring on a banle, saying that the Argives were prepared to submit to fair and ~ual
arbitration any complaints that Sparta had to make against them, and to make a
treaty and live in peace for the future. In saying this they spoke entirely for
themselves, with no authority from the mass of the army. Agis, too, in accepting
their proposals, acted on his own responsibility, and did not even discuss the
question with the majority. He took into his confidence only one man among the
high officers who were serving with him, and made a truce for four months, in the
course of which time the Argives were expected to carry out what they had prom-
ised. He then immediately led his army off, giving no explanation to any of his other
allies. 76
It is necessary to indicate that existing interpretations of the passage are
unsatisfactory. Gomme recognises the difficulty but leaves it unresolved. 77
Bury and Meiggs misrepresent the text: "The soldiers of both Thrasyllos
and Agis were confident of victory, but the generals were of another mind.
Agis, as well as his antagonist, considered his position precarious, and
76 Thucydides 5.59.4-60.1, adapted from me translation by R. Warner.

77 Gommc (1945-81) vol.lV, 82-3.
consequently they came to terms... "./~ But everything in the text suggests
that the Peloponnesians, including Agis, were hoping to carry a major
victory. Another modem interpreter, Kagan, reaches an explanation only at
the cost of introducing elements and logical links which are absent from
Thucydides' text: "Thrasyllos and Alkiphron were oligarchs who wanted to
avoid a breach with Sparta. They spoke to Agis on their own authority,
without consulting the people. They were encouraged to think that their
actions would be applauded by the people both because they would have
extricated the Argive army from a potentially dangerous situation and
because the blame for the failure to engage and defeat the Spartans could be
laid to the Athenians, who had failed to appear".'9 But, to point to one
difficulty alone, Thrasyllos and Alkiphron could not have been encouraged
to think that their actions would be applauded by the people, since the
greater pan of the Argive army thought that they were going to win.
The clue, I suggest, is the word proxenos. Once again, the implications of .-
this word can only be conveyed by a cumbersome paraphrase: that
Alkiphron had xeno; in Spana - Agis was, in all probability, one of them;
that he held the Spanan proxeny in consequence of this xenia; that in Argos,
he had a group of followers around him who, out of loyalty to him, were
supponing Spana and were therefore opposed both to other factions in
Argos and to the demos. Once we accept these implications of the word, the C"I
I-
difficulties in interpreting the passage are removed. What happened on the >-
0-
battlefield, in fact, was that Alkiphron and Thrasyllos, who belonged to the o
same Argive faction, perceived that the battle between Argos and Spana U
would jeopardise their interests. They therdore prevailed upon Agis, who
represented the Spartan end of the network and happened to be, at the same
time, the leader of the Peloponnesian army, to do his best to aven the banle
- explaining, probably, that such a banle would run against his own
interests, too. And, as a favour to a xenos, Agis avened the battle.
That this interpretation is not too far-fetched, can be seen in our last
incident. The pattern observed in the first two examples appears here in its
full light: Thucydides probably had in this case first-hand information of
every single move.
To set the scene, we must move slightly back in time to the complex
political combinations that followed the peace of Nikias. In Athens, it will
be remembered, the question was whether or not to conclude an alliance
with the Argives and their allies. An alliance with the Argives, the bitter
711J.8. Bury and R. Mtiggs, A History of Gruu (I.ondon, 1975), 4th edn., 289.
~ D. Kag2n. TIn P~au of Nidas and th~ Sidli"" Exp~djtion (Ithaca and London. 1981)
100, repeating an argumenr ser our originally in Cltwital Philology 4 (1962) 209·18.
enemies of Sparta, would necessarily mean the end of the recent alliance
with Sparta. The most prominent figure among those who wished to dis-
solve the alliance with Spana was Alkibiades. It seemed better to him to
have an alliance with the Argives, not least because he considered himself to
have been slighted by the Spanans. For the Spartans had negotiated the
alliance with Athens through Nikias and Laches, not himself. They had
overlooked him because of his youth, and failed to show the honour that
was due him on account of his proxeny with Spana, which he had inherited
from his ancestors. Publicly, Alkibiades did his best to discredit the Spar-
tans. He alleged that they were not trustworthy, that their only object in
concluding the treaty with the Athenians was to be able in this way to crush
the Argives and proceed then against the Athenians themselves. He sent a
personal message to the Argives, urging them to come to Athens with the
Mantineans and the Eleans as quickly as possible, and stating that he would
do everything he could to bring about an alliance. The Argives sent their .-
envoys at once. At the same time, however, a rival embassy from Spana also
arrived in Athens - an embassy which the Spartans thought would be
acceptable to the Athenians. The embassy was intended to forestall the
Athenian-Argive alliance and, at the same time, offer some concessions and
explanations for previous moves made by the Spanans. The ambassadors
C"I
spoke in front of the Athenian Council on these points, making it clear that I-
they had come with full powers (autokratores) to reach an agreement. This >-
0-
speech made Alkibiades afraid that, if repeated in front of the Assembly, it o
might persuade the demos not to enter into the alliance with Argos. He U
therefore resoned to an intrigue. We can now follow Thucydides literally:
By giving pistis to the Spartans, he persuaded them that, if they made no mention of
their full powers in the Assembly, he would give Pylos bade to them (which was one
of the objects of the embassy], saying that he himself would get the Athenians to
agree to this, just as it was now he who was opposing the idea; he would also
arrange for the other points to be settled. His plan was to drive a wedge between the
Spanans and Nikias, and also he intended by attacking them in the assembly for
having no sincerity in their intentions and for never saying the same thing twice to
bring about the alliance with Argos, Elis and Mantinea. And so indeed it happened.
When they came forward in front of the people and, in reply to a question, said just
the opposite of what they had said in the Council - namely that they had not come
with full powers - the Athenians lost all patience with them, and listened instead to
AJkibiades, who now attacked the Spartans even more bitterly than before. They
were in fact prepared to bring in the Argives and those who had come with them
immediately and to make them their allies. 8o
It is unnecessary to dwell here on the counter-measures taken by Nikias
10 Thucydide1 5.45.3·4, adapted from the translation by R. Warner.

to save the Spartan alliance. Suffice it to say that Alkibiades' intrigues
proved in the end successful, and the alliance between Athens, Argos, Elis
and Mantinea was brought about. To unravel the intrigue itself, the best
way to proceed seems to be by answering the questions that, according to
Thucydides' most authoritative commentators, bring out the difficulties in
the text: "(1) Why should the Spartans listen at all to their enemy Alkibiades
rather than to Nikias their friend? (2) How did Alkibiades persuade them
that their denial would help their cause? (3) How, after this, could Alki-
biades and Endios collaborate so closely in 413/12?,,81 The answers are as
follows.
(1) The Spanans were not merely 'Spanans', the representatives of their
states. For among them was Endios, son of Alkibiades, who was a heredit-
ary xenos of Alkibiades, son of Kleinias, the Athenian. 82 It is typical of
Thucydides that he does not state this fact at this point of his narrative. The
explanation for this is that the existence of xenia could be conveyed in a
number of roundabout ways. When Thucydides, or any other writer for that
matter, says that two persons - say, A from city A' and B from city B' -
were on friendly terms with each other, or that they themselves or their
ancestors exchanged favours, or that A was proxenos of B' or B of A', then
we can be almost sure that this presupposes a xenia network. Thucydides
himself provides the proof. For example, Endios and Alkibiades are fre-
quently presented in his history as engaged in the exchange of favours. An
explicit announcement of their xenia, however, is delayed until the last
book. 83 The implication must be that Thucydides assumed that the descrip-
tion of these exchanges sufficed to convey the idea of xenia. Another
example is furnished by Alkibiades' Argive network. In 5.43.3, Alkibiades is
said to have sent a message to Argos privately (idiai) bidding them to come
at once. He said that he himself would co-operate with them to the utmost.
Again, this exchange seemed to Thucydides perfectly adequate to make it
clear that xenia was involved. In 6.61.3, he could thus refer to "Alkibiades'
xenoi in Argos" as to a detail which had already been introduced. In sum,
question (1) poses a difficulty only because it rests on erroneous assump-
tions. The ambassadors listened not to an enemy, but to a friend. The
Spanans at home hoped that Endios, Alkibiades' xenos, would prevail on
Alkibiades and bring about the cancellation of the Athenian-Argive alliance.
In fact, the contrary happened: it was Alkibiades who prevailed on Endios
to disregard his ambassadorial duties and make it possible for himself to
bring about the Athenian-Argive alliance.
II Gomrne (1945·81) vol.JV, 51·2, followins the formulation of ~ problem by J. Hameld,
Altib;ad~ (Paris, 1940) 89ff.
11 Thucydidcs 8.6.3 .
•J Thucydidcs 8.6.3.
(2) Alkibiades' m~thod in persuading the Spanan envoys to betray their

city's interests provides the answer to the second question. Clearly, Alki-
biades did not trick them: the pistis he gave, and his unbroken friendship
with Endios, show this. Alkibiades succeeded in persuading the envoys to
deny their being autokratores because he did not deal with them merely in
the capacity of state representatives. He himself and Endios were xenoi, and
xenia was a 'whole-person' relationship: the two men's public and private
spheres of action merged inseparably into each other. The negotiations were
not guided merely by the concern to make the best possible representation
of their community's interests: there were other considerations at work, no
less important. We cannot tell in each particular case what these considera-
tions were: our sources, with an eye on the political, are too one-sided to
reveal other aspects of the relationship. However, the range of possibilities
can be guessed from the types of exchange outlined in the previous sections
of this book: xenoi were bound by affective ties, compelled by economic
interests, and constrained by standards of conduct. It is a reasonabl~ guess
that such considerations were enough to compensate Endios for this diplo-
matic fiasco. The passage thus presents difficulties only if we assume that
ancient diplomats, like their modern counterparts, were motivated by a
single-minded drive to succeed in the representation of their communities.
But if it is recognised that the fulfilment of their missions was not the sole
measure of their success; that they judged themselves by standards different
from those of the societies which they represented; and that in addition to
their membership of political communities, they were involved in a private
matrix of external pressures and obligations - then these difficulties dis-
appear.
(3) Not only did Alkibiades' manoeuvre not put an end to the friendship
with Endios, but it seems to have intensified it. There can be little doubt
that, in 415 B.C., after Alkibiades' recall from Sicily, it was Endios who
secured his admission into Sparta. And the two men continued to interact
even after Alkibiades fled Spana. When already in Persia, Alkibiades pri-
vately offered Endios his aid in organising the revolt in Ionia and thus
securing the Persian alliance for Sparta. The credit for this would go to
Endios, not to the Spanan king Agis, Alkibiades' enemy.84 And this is
precisely the pattern that can be detected in our first two incidents and that
emerges from dozens of other cases: armies would go to war, while xeno;
pursued their individualistic interests; cities would engage in diplomacy,
while xeno; conducted their own rivalries through the agency of political
institutions. Alkibiades of Athens acted in concert with Endios of Sparta
against Nikias of Athens, a man whom the Spartans believed to be friends of
114 Thuqdides 8.12.2, d. 8.17.2.
theirs and who was connected by ties of xenia with Pausanias, the Spartan
king. 85 Inter-state relations and networks of ritualised friendship thus fol-
lowed independent courses. If this fact is recognised, then the third difficulty
raised by Thucydides' commentators will also be removed.
The generalisation may thus be risked that neither the formal institutions
of the polis nor the informal factions within the polis were the only foci of
power shaping inter-state politics. There was the funher factor of upper-
class coalitions extending beyond the bounds of the city. 86 These coalitions
consisted of a leader, surrounded by a narrow circle of kinsmen and friends,
from the inside; of xenoi attached to the leader, also surrounded by kinsmen
and friends, from the outside; and of a third circle of lower-class, unrelated
people, attached to both former groups not on a kinship or quasi-kinship
principle, but in return for payment, or in full personal dependence, as
slaves. These action-sets, varying both in their overall size and in the relative
size of each of its constituent elements, form throughout Greek history a
ubiquitous phenomenon. Thucydides provides a good example of their
.-
panern of operation:
... th~e was an Athenian called Kylon, a vietor of the Olympic Games, belonging to a
noble family, and a powerful man himself. He had married the daughter of
Theagenes, a Megarian, who at that rime was the tyrant of Megara. Kylon went to
Delphi to consult the god, and the reply he received was that he was to seize the C"I
Akropolis of Athens... Theagenes gave him some troops and, summoning his own I-
friends (philol1 to join him, ... he seized the Akropolis with the intention of making >-
0-
himself tyrant... 87 o
U
In fact most, if not all, of the exchanges of goods and services considered in
this study were conducted within the framework of such groups. For one
peculiar feature of these groups was the multiplicity of uses to which they
could be put, and it is this versatility which deserves closer examination.
When Alkibiades disobeyed the summons recalling him to Athens, the
Athenians "sentenced him and those with him" (ot f,1£t'!xdvOtl) to death. 88
Alkibiades, like most xeno; figuring in this study, was always accompanied
by a close circle of followers - so close that the Athenians deemed them
indistinguishable from himself. In the sources, these followers are denoted
by a variety of unimpressive words, "those with/around/following some-
u l.ysias 18.10 (On 1M Proptrly of Nitias' Brothn).

I. The stress should be on the word ·coalitions·, 5i~, as the case of Alkibiades and Nikiu
shows. there were clearly several within a single ciry.
" Thucydidn 1.126.
IIThucydide5 6.61.7, d. 6.53.1.
body", or simply Hone's phi/oi orland hetairoi orland epitedeioi u being the
general formula. 89 But whenever we catch a glimpse of their composition,
an amazing variety is revealed. A passage from Herodotus provides a typical
example. When Polykrates, tyrant of Samos, was treacherously murdered
by Oroetes, Oroetes did not know what to do with Polykrates' hetairoi, and
his dilemma gave Herodotus an opponunity to describe their composition.
They included: Demokedes from Croton, a physician; some Samians -
evidently free men; and a funher group, designated as the xeno; and douloi
of the followers themseives. 9O
Evidently, all members of an entourage could, in case of need, act as
all-purpose attendants. There seems to have been, however, a meaningful
differentiation of tasks. Apan from the slaves, the need for whom is self-
evident, we frequently encounter physicians and soothsayers. These specia-
lists seem to enjoy almost as high a status as the leader, and their function
requires no explanation. What is less obvious is the need for citizens and
xenot.
Citizen-followers could promote their patron's interest at home. It might
not be incidental that Alkibiades' case reminded Thucydides of the tyrants
of the sixth century, who allowed the city to be governed by the laws
Hexcept that they took care to see that there was always one of their own
men in office". 91 Followers could assume political offices, put up motions
in the assembly, canvass for agreeable candidates, and generally enhance
their patron's good reputation. Alkibiades' epitedeioi were successful even
to the extent that they succeeded in having him elected strategos. And when
Alkibiades sailed in, after he had spent eight years in exile and held a long
record of treacherous dealings, his public image had improved to the extent
that
the common crowd of Piraeus and the city gathered to his ships, filled with wonder
and desiring to sec the famous Alkibiades. Some of them said that he was the best of
the citizens; that he alone was banished without just cause, but rather because he
was ploned against by those who had less power than he and spoke less well and
., For example, Thucydides 1.3.3; 5.42.1; 6.53.]; 8.65.1. Xenophon, Anaba$is 1.5.8;
Hellenica 3.1.10; 3.5.4; 5.2.36; .~.3. 13; 6.5.6·8 (numerous refercnces). l.ysias B.19 (Against
Agoratus); lsocratt$, Lett~ 2.8,13; uttv 6.]2; Demosthenes 20.59 (Against Uplines); 56.7
(Against Dion)'sodorus); (59).99 (Against Ntana); Hellenica Oxyrhynchia 12.2, 13.1 (d.
Calhoun (1913) 7 n.3 and Aurenche (1974) 15-29). Th~ followers are sometimes described in
inscriptions: M.B. Walbank (1978) no.85.line 27. IG 11 2 109; IG 11 2 226. For the Hellenistic
period, see Polybius 5.48.12; 5.63.1; 4.48.12; 13.6. As some of these laner examples show, in
the Hellenistic age the expression 01 :'CfQ£ nva Kerns to have become a paraphrase for the
person himself. I know of no serious study of this important subject.
90 Hcrodotus 3.] 25 .
91 Thucydidn 6.54.6. The expression o+Wv autci1v includes both relatives and friends,
MnCC my rendering as 'one of their own men'.
, • I .: , • ,,'~, : : '. ~ .',,: '. ' :

ordered their political doings with a view to their own private gain, whereas he was
always advancing the common weal, both by his own means and by the power of the
state. 92
Citizen-followers would also be of help by virtue of their own networks
abroad. For example, the followers of Podanemos of Phlius were the xeno;
of King Archidamos of Spana; the followers of Prokles of Phlius were the
xeno; of Archidamos' son, Agesilaos. 93 Thus, the networks of the followers
would be added to those of the leader and constitute a significant extension
of his power. Small wonder, therefore, that apan from the immediate
interests that a leader could secure by means of his close followers, he had a
permanent interest in extending it. A 'benefaction' - preferably supplied by
a xenos from abroad at a low cost (Section 4.3) - was marvellously suited to
that end. Once the demos was persuaded that the leader was a 'good man',
the number of his followers would increase and the avenues for political
ascendancy would be open.
The second element in the entourage - the xeno; - played a com-
plementary role. Polykrates took with him his hetairoi's xeno;, apparently
in order to facilitate his movements: a friend's friend was almost as valuable
an asset as the friend himself. But there was much more to it than that. For a
xenos-follower, a foreigner himself, would form the link with other factions
abroad. And this is why, in a xenia network, the crucial links were those
between fathers bound by xenia and each other's sons: a son of a xenos
would become a member of the panner's entourage. Pharnabazos' son
performed precisely this task among Agesilaos' followers, just as the Athe-
nian Phrynon's son would have done at Philip's coun. 94 And such was
generally the role of the children (and family) of xmo; in the entourage of
rulers: it is not difficult to recognise here the pattern of foster-parenthood. It
would have been in the patron's interest to feed, protect and benefit his
followers, and it seems that their promotion to citizenship in the patron's
city was one of the manifestations of this interest. Personal patronage would
thus be exercised through the vehicle of political institutions.
The Greek world was thus interlaced with a web of informal alliances
which were not congruent with the official foci of power of the city-states
themselves. Appendix C lists these extended alliance groups for the period
92 Xenophon, HellmiC4 1.4.13...., d. 1.4.19, where the group designated as ~i'edeioi

indudes a cousin, oiktrioi and philoi.
91 Xenophon, Hellmiu 5.3.13. The term ol ,uQl noMvfJ&OV includes both Podanemos
and his followen, and it is therefore impossible to tell whether the followers acquired meir
xenia wim Archidamos mrough the agency of Podancmos, or, via rJnS4, Podanemos through
one of his followers. Both options must have been open.
94 Xenophon, HelinriC4 4.1.40 and Demosthencs 19.230, 233 (De falsa u,atio"e).
'I' • . ; :, ~ I: :,
5 followers. Fact;ons, and the City 153
of the Peloponnesian War by the political units to which they belonged.

Perhaps the most significant feature of these groups to emerge from this list
is their permanence. For, in the vast majority of cases, the co-operative
interactions mentioned in this study flowed from relationships which were
either inherited or had been established earlier. Nowhere is this clearer than
in the case of Alkibiades. Only once (out of some 12 instances) did he start a
relationship with a view to co-operation; in most of the cases, he availed
himself of well-entrenched connexions (Appendix C).9S
These factions were only to a very minor extent fettered by political
ideologies. Alkibiades, for instance, was opposed to the alliance between
Argos and Sparta despite the fact that his Argive xeno; professed oligarchic
ideas (they effected an anti-democratic cOtlP).96 Pausanias, king of oligar-
chic Sparta, "was on exceptionally friendly terms with the leaders of the
popular faction in Mantinea t>.9 7 And Thrasyboulos in 404 B.C. effected a
pro-democratic coup by means of an army of retainers provided - to judge
by its composition and Thrasyboulos' connexions - by a Thracian xenos. 98
After the victory and the restoration of democracy, Thrasyboulos' retainers
were befittingly rewarded with Athenian citizenship. Factions of ritualised
friendship thus cut across the ideological division between oligarchy and
democracy. Not even the welfare of their own communities would prevent
them from pursuing their particularistic interests. A puzzling feature of
Greek inter-state politics was the betrayal of a city by friends from the
inside. It is true that sheer bribery did playa part. But repeatedly, the
'traitors' turn out to be proxeno; of the attacker's communities or xeno;,
philo;, ep;tede;o; of the commanding generals. 99
These factions, moreover, showed a remarkable capacity for expansion.
They grew by a process which can most aptly be described as 'pyramiding';
several lesser leaders, each accompanied by his own followers, and each
controlling valuable resources, became the follower of a more powerful
leader. 'The composition of the ruling circles of the Persian, Macedonian and
Hellenistic ruling circles shows that this is indeed how they were formed,
Within each circle two distinct elements stand out: a narrow core of inti-
mates (kin, friends, xeno;); and a wide periphery of xeno;. It is this
9S The relationship with Phamabazos (Xenophon. H~lIen;ca 1.3.12). AJkibiadrs' collection
of a hundred talent5 in the Ceramic gulf (Xenophon, H~lIen;ca 1.4.8) can be anributed either to
an arg,.,o/ogia or to provision by xeno;. as in t~ case of Peisisrratos (Aristotle, Th~ Athenimr
Constitution 15.2·3, d. Section 4.4, p.90).
96 Thucydides 6.61.3.
97 Xenophon, Htllenica 5.2.3.
91 Ct. Middleton (1982).
99 ct. Losada (1972).Gerolymaros (1986) reaches the same conclusion through the analY5is
of proxen~s.
periphery which is of most interest for our purposes. For, insofar as the
central leader could satisfy the needs of these xenoi in terms of protection
and resources, and insofar as he could render them assistance in subduing
their enemies, there was no limit to the ability of the periphery to expand.
First, a leader himself could have an almost unlimited number of xeno;.
Secondly, the lesser leaders brought with them both their own followers and
their own xeno;, and the process would repeat itself the lower we descend
the pyramid and the further we move away from the centre. It is this feature
which accounts for the astonishing ethnic diversity of the ruling circles: the
inclusion of foreign leaders was a necessary price of expansion. The relevant
evidence may briefly be reviewed.
The Persian circle of syngenei5 included oriental non-Persians and Greeks
both as client-rulers and as office-holders on the king's behalf. Quantitative
evidence is missing, but the scanty literary fragments point to eight landed
estates in the Persian empire held by Greeks as against 21 by non-Greeks.
This would seem to indicate a general trend 100 since it accords both with the
examples discussed elsewhere in this study - of Greeks being drawn into the
high spheres of power at the Persian court - and with the pattern displayed
in Cyrus' attack on Artaxerxes: it is a reasonable guess that, had the
expedition succeeded, Cyrus' non-Persian xenoi would have been put in
positions of power.
Macedonia under Philip and Alexander displayed a similar pattern.
Thcopompos' venomous statement concerning Philip's hetairoi (U some
were from the land to which he himself belonged, others were from Thessa-
Iy, still others from the rest of Greece ... nearly every man in the Greek or
Barbarian world of a lecherous, loathsome, or ruffianly character flocked to
Macedonia and won the title of 'Philip's hetairo;' It), 101 suggests that both
the heads of surrounding tribes and Greek citizens were absorbed by the
centre. This seems to be confirmed by casual references to particular persons
becoming hetairai or receiving gifts (timber, grain, estates) from Philip, 102
'00 Cook (1983) Fig. 12 facing p. 178. See funher Hofltetter (1918) and Wiesdt6f~ (1980)
for Greeks in Penian service.
101 FG"H 115 F 2H, d. Demosthenn' remark thlt Athenians expelled for loose morals had
been recruited to Philip's court (Demosthcnes 2.19 (~c01fd Oly.tthi«)). Ind lsocrates' remark
thlt Philip had around him the ablest of Ma~nianl as well as famous and intelligent men
from Greece ((socrates 5. t 9 (To Philip)).
101 For example, Kallias of Chalkis was named one of Philip'. hetai"oi (Aeschines 3.89
(Alamst Ctesiphcm); Aeschines himself was accused of having become Philip's xmas and
pbilos (Demosthenes 19.314 (D~ Falsa Ltgatione); Philokrates and Anchinn were given
houses. timber and grain by Philip (Dcmosthmes 18.41 (On th~ OoWN). 19.145); Xenophron.
one of me 'thirty tyrants' in Athens., who clearly acted as a Mtairos (Dcmosthena 19.196h
Demaratos of Corinth was a xmas of the Macedonian royal family turned into Iwtairos (Bave
(1926) 00.253). For I collccrion of the secondary eviMnce on Ma~nian h~t4iroi. see
Thomes (1955).
and by some prosopographical data. The people explicitly referred to as

hetaira; of Alexander in the list compiled by ~rve, which I use as a
sampling base, clearly fall into three groups: (a) 31 individuals from Mace-
donia proper; (b) 9 from neighbouring territories (Lyncestis, Elimiotis,
Orestis, Amphipolis, Eordaea); and (c) 17 from remote cities or ethne
(Larisa, Teos, Crete, Acarnania, Cyprus, Pharsalus, Corinth, Mytilene, Car-
dia, Epirus, Thrace, and Persia).103 It is not difficult to recognise in the 9
people from group (b) the chieftains of tribes subdued in the first imperialis-
tic expansion of the Macedonian kingdom. Nor can there be any doubt that
the individuals from remote areas (group (c)) were either xeno; of old,
promoted to positions of power, or the heads of subdued enemies turned
followers.
After the dismemberment of Alexander's empire, this trend seems to have
intensified. Studies of the epigraphical evidence from the Hellenistic ~riod
show conclusively that the phi/a; of the rulers were not imposed from
above, from within the narrow circle of Macedonian dynasts, but were
recruited from below, from among the elites of the Greek cities. 104 Acting
simultaneously as leaders of civic factions and officials of the central power,
these elites were able to hold together empires that might stretch over three -...)
::v
continents. lOS 4-J
In sum, throughout its history, the Greek city was torn by conflict
between upper-class factions who derived their power and resources from
foci of power which lay outside its boundaries. Networks of alliances linked
factions from several cities and radiated from the great empires located at
the fringes of the world of cities, creating a system of external friendships
that could offer rewards - wealth, fame, position - even more tempting than
those of the city itself. It was a system that had not changed significantly
since the days of the epos. What had changed was the range of services
available for exchange: when the city came into being, the services that
could be performed through political institutions came to be added to the
10J The groups are as follows: (a) Nos. ~9, 57, 83. 157. 164,269,357.380,412,445,466,
467,~80,~94,501.502.545.554.556,598.602,603.606,613,621. 782, 802,645.648,
700, 757. (b) Nos. 37 (Lyn«sris). 250 (Lyn«sris), 1~3 (Elimioris). ~39 (EJimioris), 4~6
(Orcstis), 627 (Orcstis), 780 (E1imiotis), 668 (Eordaea). (c) Nos. 101 (Amphipolis). 139
(Ph.nalol). 253 (Corinrh). 258 (Greece ?), .102 (Mytilene). 308 (Gre«e ?), 317 (Cardia), 17
(Teos), 292 (Greece ?), 350 (Thrace), 427 (lIIyria). 46~ (Mytilene), ~86 (Penia), 521 (Lari~),
544 (Cme), 5~8 (Epirus). 586 (Penia. to judge by the name), 788 (Acamania). 719 (Cyprus).
Nos. 72. 745 and 789 are unclassifiable and are therefore left out of account. In doubtful casn
I have accepted Berve's conjectures.
104 Habicht (1958); Herman (t 980/1); Walbank (198~a).
10$ Cf. Bagnall (1976). who providn a good example of how remote territories would be
controlled through networks of friends. In the third century B.C.. the 'Ptolemaic possessions'
outside Egypt included Cyrenaica, Cyprus. Phoenicia, Palestine. Code·Syria. and parts of Asia
Minor and the Greek islands.
, • I .: , • ,,'~, : : '. ~ .',,: '. ' :
156 Obligat;ons: Hero;c and Civic

repertoire of 'private' services that had circulated in the pre-political world.
By the same token, however, there emerged a class of men who, unlike the
aristocracy, the rich, and their clients, were free of such ties of dependence
and uninvolved in the alliance system. This was the demos, and it is to the
point of view of the demos that we now turn.
5.6 TREASON AND PATRIOTISM
The betrayal for private gain of fatherland (patris) and fellow-eitizens

(andres), wrote Pausanias in the second century of our era, had been
endemic in Greece since the earliest days. The concrete examples with which
he illustrates this general proposition are highly relevant to our subject:
In the reign of Darius, the son of Hystaspes, the king of Persia, the cause of the
lonians was ruined because all the Samian captains excrpt eleven ~rayed the Ionian
fleet. After reducing Ionia, the Persians enslaved Eretria also, the most famous
citizens turning traitors - Philagros, the son of Kyneas, and Euphorbos, the son of
Alkimachos. When Xerxes invaded Greece, Thessaly was betrayed by Aleuades, and
Thebes by Attaginos and Tim~enidas, who were the foremost citizens of Thebes.
After the Peloponnesian war, Xenias of Elis attempted to betray Elis to the lake-
daimonians under Agis, and the so-called xeno; of Lysander at no time relaxed their
efforts to hand over their countries to him. In the reign of Philip, son of Amyntas,
Lakedaimon is the only Greek city to be found that was not betrayed; the other cities
in Greece were ruined more by treachery than they had previously been by
PIague.... 106
The remarkable thing about Pausanias' diatribe is the incongruity be-
tween the elevated status of his actors - captains, leading citizens, xeno; -
and their ponrayal as traitors. People possessed of such laudable traits can
hardly be fitted into any familiar stereotype of 'traitors'. 'Elites' are com-
monly regarded as the most authentic representatives of their societies,
embodying traits which reside in their inferiors only to a lesser degree.
Pausanias, on the other hand, pictures the elites of the Greek cities as the
enemies of their own societies, and it is this picture that calls for an
explanation.
The very fragmentary evidence indicates that, in the early days of the
polis, the portrayal of the elites in such terms would have been inconceiv-
able. For Homer. as we have seen, it was self-evident that the hero should
remain loyal to his xeno; above al1. 107 And two centuries or so later, Pindar
106 Pausanias 7.10.1-3. The passage carries the history of Greece down to the conquest by
the Romans on similar principln.
107 a. Finley (1977) 116-17. The evidence produced by Glftnhalgh (1972) for 'patriotic
pride' in the age of Homer shrinks to insignificance when compared with some of the genuinely
patriotic utterances from the Classical age cited below.
'I' • . ; :, ~ I: :,
could still pamt a favourable picture of Thorax of Larisa, a man who
harboured Xerxes and opened the way for Mardonius' invasion of
Greece. lOR The conclusion must therefore be drawn that the critical outlook
which Pausanias expressed and clearly accepted without question had been
extricated, by means of a gradual and painstaking process, from converst'
premises. An anecdote incorporated into Herodotus' history might point to
the first germs of the emerging alternative. Mardonius, having invaded
Athens, sent a messenger to the Athenian Council in Salamis offering
friendship. Lykidas, one of the councillors, declared that it seemed to him
best to receive the offer and lay it before the Assembly. "This was the
opinion which he declared", comments Herodotus, "either because he
received money from Mardonius, or because the plan pleased him". But the
Athenian councillors were so wroth that they stoned Lykidas to death. 109
Only dim traces are visible of the process by which this crude form of
communal solidarity became refined and gathered ideological weight. But
we can sec enough to conclude that it did so in opposition to upper-class
individuals united by xenia. Throughout Greek history, it was the commun-
ity that arrayed itself against the one-time hero, against xenoi plotting its
plundering, subjection and exploitation.
In the first instance, the community defended itself b}' punishing them. In
Athens, ostracism was perhaps the best-known instrument of punishment.
The people who were ostracised at the beginning of the fifth century were of
high social status, of noble birth, and some of them, like Themistokles, are
known from literary sources for their involvement with the Persians. llf} It
seems thus likely that even if in later times ostracism became a tool of
political rivalry inside Athens,lll its original purpose was to restrict the
ability of upper-class citizens to draw on foreign allies in order to subvert
the people's rule. "Kallixenos traitor", says one potsherd from the early
fifth century, unmistakably echoing an angry cry of the Demos; "Aristeides,
brother of the Persian general Datis", says another (Fig. 15a,b), and there
can be little doubt that what was envisaged here were Persian magnates
acting in concert with Athenian xeno; to the detriment of the Athenian
collectivity.
The Peloponnesian War was punctuated by events which fit a similar
pattern. The xenoi of rival forces who preferred their private alliances to
communal duties were criticised, banished or executed. Perikles gave his
11111 Pindar, Pythian 10 and Herodotus 9.1. It should be noted, however. thar the poem
preceded the inva~ion.
1M Herod()tu~ 9.5.
110 Meiggs and Lewis (1969) 40ff. give a usdul catalogue.
III Cf. Finley (1973h) 26.
158 ObligQtions: H~o;c and Civic
APIITI---J
TO~~ ---
AAEAf ---
ro
'-
(lJ
......
• b
ro
:E
fiB. lS.,b Xmi. aad tmuon '0
OStrlka from Athens from the earl, fifth century bearing me inlcriPbons • "Aria· Q)
......
teides, brothet of Dati,", and b "Kallixenos traitor--. (For the reading, d. A.Eo .c
RaubitlCbek in Charit", Nstsebrift Ltmglo~ (Bonn 19S7) 240; G.A. Stamires and .-0'
'-
E. Vanderpool in H~spma 19 (1950) 390; and Meigs and Lewis (1969) 42). As ~
'brother- is a convenient synonym of 'unos' - Pylades, for instance, in plays by a.
o
Eunpldes js styled both as Orestes' xmos and brother (Eur. E.l. 83 and Or. 1015)-it U
is evident that the idea of treason is here expressed through the" metaphor of
ritualised friendsillp.
estate to the demos as proof of his loyalty, but his patriotism and abstinence
from bribery were nonetheless impugned. 112 ArchidamOi himself, as we
have seen, was severely criticised, and his case reminded the Athenians of
Pleistoanax, son of Paosanias, who fourteen years before the .ar had been
banished from Spana on account of a similar, apparently inexplicable,
withdrawal from Attica. 1 U The Spartans refrained from their original idea
of imposing a fine on Agis (the king who failed to give battle at Argos) and
of razing his bOUie, but they enacted a law preventing him from leading an
army OUt of the city without the consent of ten Spanan advisers elected
by the city. I 14 The Argives made an angry demonstration against those who
made the truce without consulting the many (Plnhos): Thrasyllol eteaped
IU In Thucydides 2.60, he iI made to answer such chUJel.
tl) Thucydidcs 2.21.1. He wu luapcacd of "bcqJ penuadcd 1¥itb the aid of 1IIODC1-•
.... Thucydidcs 5.63.4.
being stoned by the skin of his teeth, and his property was confiscated. And
Themistokles, the Athenian, after his death in Asia Minor, had to be buried
in Attica by his relations, without the Athenians' knowledge - "for it was
not lawful to bury him there, as he had been banished for treason". 11~
There are good reasons to believe that it was with similar situations in mind
that the Athenian law of treason was formulated: "if anyone shall be traitor
of the po/is,... he shall be tried before a popular jury (dikasterion), and if he
be convicted, he shall not be buried in Attica, and his property shall be
confiscated". 1 16
But there were subtler means at hand by which to curb aristocratic
power. An intriguing process comes into play whenever a group subjected to
self-imposed regulations enters the scene. "After the rise of cities", observed
a student of primitive societies, "men became something different from
what they had been before".117 The case for ancient Greece can be illus-
trated by the example of Alkibiades. This inveterate aristocrat, whose own
behaviour was clearly shaped by the ideal of the Homeric hero, could at the
same time declare himself a "lover of his city" and crave desperately for
recognition by his fellow Athenians. The same ambivalence runs through
the careers of most of the other major figures of Greek history. They tried
persistently to have the best of both worlds: like Alkibiades, they engaged in
intrigues by means of xenia connexions; but, when required, they posed as
loyal citizens. I 18 Xenia and patriotism pulled in two opposing directions,
and, in practice, xenia could be preferred to patriotism. But the fact remains
that patriotism would not be left unrecognised.
For the community set new ideals of behaviour and generated hitherto
unknown criteria for passing moral judgements. It devised new points of
reference for interpreting the world, and imposed a way of thinking which
took the city as its point of departure. Most important, it prompted its
members to remodel their own personalities to meet these standards. What
set Agesilaos and Alkibiadcs apart from the pre-state heroes was that they
yielded - however panially - to this drive. What distinguished Perikles and
Archidamos from Diomedes and Glaukos was their repudiation of xenia -
II ~ Thucydides 1.138.6.
116 Xenophon, Hellenica 1.7.22.
117 Redfield (1953) v.
1111 For example, when Xenophon's patron, Agesilaos, received a letter offering xenia and
phi/ia with the Great King of Persia, he is said to have told the messenger: "Tell his majesty,
that there is no need for him to send me private Imers, but, if he gives proof of friendship for
Sparta, and goodwill towards Greece, I on my part will be his friend with all my heart. But if he
is found plotting against them, let him not hope to have a friend in me, however many letters I
may receive" (Xcnophon, Agesi/aus 8.3). The whole C'Ssay may be viewed as an unsuccessful
attempt to iron out the inconsistency ~een A~ilao~' image as a loyal xenos and an
obedient servant of his city.
however nominal this repudiation may have been - in order to conform to

the ideals of the city. Loyalty to the demos, the country, the polis -
originally a mere response to threats to the whole community - was gra-
dually superseded by an ideological gloss which took on a life of its own.
"The most valuable thing in man's life", says Demosthenes, accusing Aes-
chines of treason and xenia with Philip, "is supporting the policy of the
many, and having the same friends and the same enemies as your fatherland
(patris)".119 This was, then, the premise that had to be accepted, that had to
become self-evident, in order to make it possible to discover that the line
separating traitors from patriots was, in fact, almost identical with the line
separating the upper classes from the lower:
Who then are the people who commit these monstrous crim~ [i.e. embezzling public
money, fomenting political troubles, overthrowing governments, etc.) ? Persons who
fancy themselv~ important enough to be called xe"o; and philoi of Philip, men
itching for military commands and eager for military distinction, men who claim
superiority over the common people (hoi polloi). At Megara the other day was not
Perillos tried before the Three Hundred on a charge of visiting Philip? And did not
Ptoedoros, the first man in all Megara for wealth, birth and reputation, come
forward and beg him off, and send him back to Philip? The sequel was that one of
the pair returned with an alien army at his back, while the other was hatching the
plot at home. Indeed, there is no danger, no danger whatsoner, that requires more
anxious vigilance than allowing any man to become stronger than the people. 120
We must return to the incidents analysed in the previous sections in order to
realise that, despite its explicit bias, Demosthenes' rhetoric is not unduly
removed from reality. The upper classes of the Greek city were involved in a
network of alliances across community lines; the lower classes were con-
fined within their laterally insulated communities; and the upper classes did
display more solidarity with those of their kind outside their communities
than they did with the lower classes inside them. The portrayal of the
'foremost of citizens' as traitors is thus central to an ideology that was
propagated by the demos (or on their behalf) to protect themselves - and the
community as a whole - from external, upper-class coalitions.
Alkibiades was, indeed, accused of serious crimes. But the speeches
portraying him in an unpleasant light were written with one clear purpose in
mind - to win over the support of the demos. Thus they say what the demos
would like them to say. When it suited a speaker to attack Alkibiades, he
accused him of treason, of aiming at tyranny, of sexual depravity, and,
119 Demosthenn 18.284 and 280 (0" tbe Crown). a~rly, the orator was using a diche.
el., for very similar phrases, lsocrates 16.41 (Ccmct17tmg th~ Ttam of HorKS); Plato, Laws
12.9S5b-c; and Polybius 1.14.
110 Dtmosthenes J9.295 (Dt Falsa ugatio"t), d. 18.109ff. (On the Crown).
6 TreaS01I and Patriotism 161
generally, of not acting in conformity with civic norms. When a speaker

wanted to defend Alkibiades, he played down Alkibiades' adventures
abroad and presented him as a loyal citizen. 121 However, in the extant
sources, another point of view persisted. Thucydides more than once makes
it clear that Alkibiades acted from personal motives of self-interest. Not
even once, however, does he condemn him on that account; never does he
call him by any of the derogatory terms used above - let alone 'traitor'.
Indeed, he even makes him maintain that what clearly was treason was not
treason. 121 It is therefore likely that from the moral standpoint of Thucy-
dides the identification of xenia with treason was by no means obvious.
In the Greek world of the cities, then, unlike in the modern world, the
notions 'treason' and 'patriotism' had overtones of class conflict. Consi-
derations of nationalism, on the other hand - considerations which lend to
the modern concepts their most distinctive colouring - were almost entirely
absent. The Greeks who co-operated with the Persian kings a century before
Demosthenes, and who would co-operate with the Hellenistic rulers not
many years later, came from the same high aristocracy as those who
co-operated with Philip of Macedon. As to Alkibiades - true, he betrayed
his inferiors in his native city, but he remained loyal to his upper-class
friends abroad; he violated the principles of action imposed by the demos,
but remained faithful to the heroic principles of ritualised friendship.
121 This tvidcnce is summarised in ~ag('r (1967).

122 Thucydides 6.92.4, d. 4. t 14.3, where the same argument is attributed to Brasidas.
Polybius would later spell out a very similar viewpoint. see Polybius 18.13- J" with Walhank
(1957·82) vol. II, 564·70.
CONCLUSION
I have attempted in this study to abstract for the purposes of analysis a

category of social relationships and values which can be identified both in
the pre-political world and in the Classical and Hellenistic worlds of the
Greek cities. 'Ritualised friendship' forms a connecting thread, as it were,
between these worlds, and serves as an analytical tool to throw into perspec-
tive cenain aspects of ancient society.
Perhaps the most important conclusion that emerges from this way of
looking at the ancient evidence is the persistence, in the world of the cities,
of the horizontal cleavage between the upper and the lower classes - which,
though perhaps not as deep as that observed in the world of Odysseus, l was
certainly deeper than is commonly held in modem research. A minority of
people at the top of the social pyramid strove to differentiate themselves
from those below them, and employed a variety of means to that effect.
Apart from a separate code of manners, fictive genealogies, the cultivation
of leadership qualities, and certain occupations, it was through their parti-
cipation in a social sphere outside the city that they distanced themselves
from their inferiors. The main prerequisites for entering this sphere were
wealth, power and status; the main avenues of access, guest-friendship and
marriage.
The ancient world was thus united at its highest social level by a web of
complex alliances. However, this was a superficial unity. Neither the Greek
world as a whole, nor any individual polis, came close to anything which
might be termed a "nation'. Neither the Greek world - since, despite some
claims to unity on the basis of cultural homogeneity, the upper-class
alliances did not respect either cultural or ethnic boundaries: the Persian.
Macedonian, and Hellenistic 'enemies' formed an integral part of the Greek
network. Nor the individual polis - since the most significant social division
of the ancient world, that between the upper and lower classes, was not
confined within its bounds. Having discarded the model of the nation state
for the social structure of the ancient Greek world, we are left with a type of
agrarian society which has been described by Gellner. In the characteristic
agrarian state, according to him, "the ruling class forms a small minority of
the population, rigidly separate from the great majority of direct agricultu-
I Finley (1977) 1H.
162
Conclusion 163
ral producers, or peasants". Below this horizontally stratified minority,

there is "another world, that of the laterally separated petty communities,
which generally live inward-turned lives, tied to the locality by economic
need" (see diagram). 2
'\,...~....- stratified, hori7.ontally

~~~~~~~~~~~~~~~....-;segregated layers of
military, administrative.
clerical and sometimes
commercial ruling class
laterall)' insulated
communities or
agricultural producers
It seems to me that of all the models that arc available, it is this which best
accommodates the details of the evidence. However, allowing for these
basic similarities between ancient Greek society and the societies from
which Gellner's generalisation is drawn, ancient Greece displayed features
which cannot easily be found in any type of agrarian society, and it is to the
elucidation of these peculiarities that we must finally turn.
What makes the case of ancient Greece unique is the polis itself. The polis
superimposed itself upon this archaic form of social structure and evolved,
2 Adapted from Gellner (1983) 9-10.

164 Conclusion
in a sense, out of the conflicts created thereby. Most major political moves
of Greek history - wars, invasions, revolutions - can be explained in terms
of this struggle: the d~mos, the lower-elasses, and the polis, on the one
hand; the aristocracy, the upper-classes, and coalitions outside the polis on
the other. For roughly two centuries in the Classical age the lower classes
extricated themselves from dependence on their superiors and set up an
alternative form of social organisation. The one-time heroes thus lost their
effective supremacy. Some were transformed into right- and duty-bearing
members of formally constituted political bodies, while others joined their
peers outside the city, seeking, with their help, to regain power. In the
Hellenistic age, these attempts were now crowned with success; most cities
were manipulated, if not effectively ruled, by leaders of factions supported
by rulers abroad.
The simultaneous involvement of the upper classes with the city and with
foreign rulers had important consequences, too, for the structuring of the
societies outside the city. The petty kingdoms on the fringes of the Greek
world, or the Macedonian and Hellenistic monarchies, were in some re-
spects reminiscent of the social pyramids headed by the Homeric heroes. Yet
there were important differences. First, in the Homeric world, friendship
(philia and xenia) was, apart from marriage, the only bond to create
enduring obligations between peers. In the Classical and Hellenistic worlds,
friendship indeed remained the main bond, but there were in addition
relationships of subordination and superordination which to some extent
were institutionalised. People who originally started their careers in the
Persian, Macedonian and Hellenistic ruling circles as philo;, xeno;, and
hetairai, would quickly become generals, satraps, treasurers and the like in
the royal administration: the basic obligations of friendship came to be
superseded by obligations of service.
Secondly, although these societies shared, both in terms of the circulation
of resources and the ritual of social relations, some features with European
feudalism, they were not feudal in the proper sense of the term. It will suffice
to refer back to the bestowal of landed estates in order to make this point
clear. Given the nature of the evidence, it is hard to penetrate the intricacies
of the practice, but even so it seems clear that such grants followed no
common pattern. There were important geographical and chronological
differences in the way in which the relationships between the grantors and
grantees were structured, and, furthermore, the grants never became an
indispensable element of the bond: one man could serve, support, and be
loyal to another man on the basis of friendship and exchange of benefits
Conclusion 165
alone without receiving a landed estate from him. Unlike in feudal Europe,]
the two institutions of ritualised personal relationships (there, vassalage)
and the granting of estates (there, the benefice), never merged together to
give rise to a new system of social relations.
, Cf. Ganshof (1964) 15ff.
C)
APPENDIX A
XENOI, IDIOXENOI AND DOR YXENOI
The list below roughly follows the chronological order of the sources (not
the dates of the individuals engaged in the bond). If a relationship is attested
in several sources, the later source(s) are cited together with the earliest.
Secondary sources are only included here for the sake of completeness; the
generalisations in the text are based on evidence stemming from the time
period defined in the Introduction.
This selection of sources is dictated by the terminology used in the texts;
it thus comprises only a small fraction of the instances delimited by the
definition of ritualised friendship. In all the passages cited, there is either
explicit reference to xenia, idioxenia or doryxenia (hospitium, in the Roman
sources). or, as in the case of Tydeus and Hippolochos, the existence of the
bond follows necessarily from its existence between their ancestors.
PX2 and PX t mean, respectively, xenoi through grandfathers and
fathers. (The term patrikoi xenoi covers both possibilities; hence my 'ances-
tral xenoi'.) Ancestors not styled explicitly as xenoi are only listed if they are
said to have interacted within the context of xenia. The absence of either
PX2 or PX t indicates that the bond was generated during the lifetime of the
partners themselves. (The terminology, however, is not compelling: two
people might be called simply xeno; even if they inherited the relationship.
Such might have been the case of Perikles and Archidamos.)
With the exception of the Homeric examples, the historicity of which it is
unnecessary to discuss here. fictional cases are marked F; cases in which the
borderline between imagination and reality seems unclear are marked F?
Cases in which the persons involved had, for a variety of reasons, not
formally become xenoi, but the word appears in the text (one expresses his
wish to become another man·s x~nos; an offer of xenia is rejected; the
existence of xenia is an accusation or insinuation which cannot be verified)
are marked NX.
166
Appendix A 167
Name and proven;1nce N;1me and provenance Referm~ Rem;1rk,
Oineus Bdlerophon 11.6.215

(Lycia] (Argos]
Tydeus Hippolochos II. 6.21 S PXI
[Lycia) [Argos)
Diomtd~ Gbukos II. 6.224.231 PX2
(Lycia) [Argosl
Paris Harpalion 1/.13.660-1
[Troy) [P;1phl;lgoni;1)
Euphrtes Phyltus II. 15.532
(Ephyra) [Doulichion?]
Hektor SarpWon II. 17.150
(Troy) [Lycia)
Htktor Ph3i nops l 11. 17.S84
[Troy) [Abydos)
P';2mos Et"rion 1/.21.42 :c
[Troy) IImbrosl
-
:lJ
Agamtmnon Amphimedon Od.24.114 ~
[Mycmc) [Ithaca) ~
.--
Ody~us Iphitos Od. 21.13.21 L
[Ithaca) [Oech2lial '--"
J)
Odysseus Idomtneus Od.19.191 4-J
[lthacal [Crrte)
Cl
Odysseus Beggu 2 Od.17.522 ~
[Ithaca) [Crete) :>--

Odysseus Alkinoos Od.9.18 NX :::l.
(Ithaca] (Phaeacial ,0 .
'-...-/'
Odysseus Pheidon Od.14.322

[Ithaca) [Tht1proria)
Dmetor Begg;1,l Od.17.442
[Cyprus] [Crete)
Telemachos Peisistratos Od.15.195~ PX2
(Ithaca) (Pylos)
Telem;1cho5 Mentes' Od.1.176.0I7. PX2
[Ithaca) (Taphos) 313.417
Themi~tokles Timokreon Timocr., fro I
(Athtns] (Rhodes) = Plut. Thmr;sl. 21
Pindu Thusyhoulm Pind. I. 2.48
(ThtbtsJ (Acugas]
Pindar Hiero Pind. P. 3.69
IThebts) (Syracusc)
Pindar Thorax Pind. P. 10.64
[Thebts) [Th~saly)
II.e. Apollo in disRUisc.

1 I.e.Odysscus in disgui~.
'I.e. Athena in disgUise.
168 Xcnoi, Idioxenoi, and Doryxenoi
Name and provenance Name and provenance Reference Remarks
Pindar Thearion Pind. N. 7.61

[Th~sl [Aegin~1
Herakl~ Aiakos Pind. N. 7.86 F
(Thebes! IAeginal
Omtts Strophios Pind. P. 11.34 F
[Amycl~el (Phocis)
Omtcs Pylades Pind. P. 11.16; F
[Amyc:lael [Phocis) Eur. EI. 82;
Soph.EI.16
Croesus AdraCitos Hdt. 1.43 F
[Lydial IPhrygial
Chileos Unknown ;u"o; Hdt.9.9
[Tegeal ISparul
Kleomenc.-s Isagoras Hdt. 5.70; d. Arist.
(Sparta) (Athens) All,. Pol. 20
Pausanias Hegetondes Hdt.9.76
(Sparta) [Cosl
Themison Etearc:ho5 Hdt. 4.154 F
(Thera) (Oaxus)
Camby~s Phanes Hdt.3.7
[Persia I [Halicarnassus)
Camby~ Unnamed king Hdt.3.21 F?;NX
(Persia) [Ethiopia I
Xerxes Pythios Hdt.7.29
[Penial [Lydia!
XerxeS Demaralos Hdt.7.237
(Penia) [Spart~1
Periander Thrasyboulos Hdt.l.20;

(Corinth I (Miletus) Diog. uert. 1.95
Histi:aios Unnamed xeno; Hdt.S.30
(Miletus) [Naxosl
Ari5tagoras Skylax Hdt.5.33
(Miletusl (Mindusl
Simoni<ks M~i~tias Hdt.7.228.3-4
(Ceusl (Acarnanial
Polykrates Amasis Hdt. 3.39,40
[S~mosl [Egyptl
Polykrates' followers Unnam~xmo; Hdt.3.125
(Samosl (Unknownl
Perikles Archidamos Thuc. 2.13;
[Athensl (Sparta) Plut. Per. 23.2
Alkihiades Endios Thuc.8.6.3 PXJ
(Athens) [Spartal
Alkihiades Unnamed xeno; Thuc.6.66.3;
(Athens) (Argos) 0100. 13.5.1
Brasidas Unn;lmed xeno; Thuc. 4.78.4
(Sparta) (Thessalyl
Appendix A 169
Name and provenan<:e Name and provenance Referen"c Rcm,lrks
Perdikku Unn3med x,no; Thuc.4.1.17.2

(Macedonl (ThcssalyJ
Agamemnon Strophios Aesc:hyl. A. H80 f
(Argosl IPh()(is)
House of Ag2mcmnon P)'ladcs Aeschyl. Ch. 562 ..·.I'X 1
[Argos) (Ph<Kisl
Herakles Keyx Soph. Tr. 40 r
(Thebnl [Trachisl
Aia~ Hektof Soph. Ai. 817 F
(Salamis) (Tro)·J
Heraklc.-s Eurytos Soph. Tr. 263 F
rl1lcbes) IEuryteial
House of Agamemnon Phanoteu'i Soph.F.45~ F
[Argosl [Phoc.-isl
Hektor Rhesos fur. Rht$ . . n.s f
[Troy I lThracel
lason Unnamed nno; Eur. Mt'd. 613 F
[Iolcus] IUnknownl
Priamos Polymcstor F.ur.llu.7 F
[Troy) [Thri1cel
Polydoros Polymesror F.ur.lftc. 19.82 f. PXI
(Troy I /Thrace)
Uekabe Polym~tor Eur. flee. 710 F
(Troyl (Thrace)
PolyneikC's Unnamed xl'no; fur. PI,. 402 F
[Thebes] [Unknown I
Polyneikes
[Thebes]
Theseus
[Athens)
fur. Sit"". 9.JO F
Sophoklcs Hermesileos FGrH 392 {Ion}

(Athens) (Chios)
Cnro Unnamed xena' Plat. Cr;t. 4.se. 5.ld
(Athensl IThessaly)
Menon The Great Kmg Pbt. Ail'''. 78e-<l PXI
(Lari~1 (Persia)
PI:uo Dion [Pl:al.ll.ett~,. 3.316c
(Athens) (Syracuse I
Andocidc.-s Unknown xmOl Ando4:. I. 132
(Athensl [Delphi (?) and
elsewhere (?)1
AndocidC'S Unnamed king.'i Andoc. 1.145
(AthensJ and individuals
[Unknownl
Andoeidt'S Archelaos Andoc:.2.11 PXI
[Athensl IM2(edonl
Kephisoooros Unn:1mcd xenos 1.)'5. fro 78
(Thebes) [Athensl
170 Xenoi, Idioxenoi, and Doryxenoi
Name Ind provenance Name and provenance Reference Remarks
Eunomos Dionysiu5 Lys.19.19

[Athens) (Syracuse)
Andocides Unnarmd rulers Lys.4.48
[Athens) (Unknown)
Lysias Thrasydaios P. Ox,. 1606.168;
[Athens) [Elisl Plut. Mor. 835f
Pausanias Diognetos lys.18.1O
[Sparta) (Athens)
Agcsilaos Artaxerxes II X(n. Ag~s. 8.3; NX
[Sparta) [Persia) Plut. Agts. 23.6
Agesilao5 Pharnabazos son
I Xm. Hell. 4.1.39;
(Sparta] [Persia] Plut. Ages. 13.1
Agesillos Apollophancs Xm. Hell. 4.1.29;
[Spartl) (Cyzicus) Plut. Ages. 12.1
AgC'Silaos Unnam~x~o; Xen. Age". 2.23
(Sparta] [T~eaJ
Agtosilaos Mausolos Xen. A~s. 2.2'"
[Sparta) (Caria)
Agnilaos Tachos Xm. Ages. 2.27
[Splrel' (Egypt)
AgesilaO$ Prokles and his Xm. H~II. 5..tI3
[Sparta) followers
[Phlius)
Cyrus Unnamed king Xen. Cyrop. 6.2.1 F
[Persia) [India)
Cyrus Klearchos Xen. A"ab. 1.3.3,
[PersilJ [Spartal cf.1.1.9
Cyrus Aristippos Xen. AMb. 1.1.10
[Penil) [Thessaly)
Cyrus Proxenos Xen. A"ab. 1.1. 11
[Persia) (Barotia)
Cyrus Sophainetos Xm. AMb. 1.1.11
[Persia) [Stymphalos)
Cyrus SokratC'S Xen. A"ab. 1.1.11
(Persia) (Achaeal
Xenophon Proxenos Xen. A"ab. 3.1.4,
[Athens) [Baroria) 5.3.5
Xmophon Kleandros Xm. A"ab. 6.6.35,
(Athens) [Sparta) 7.1.8
Menon Ariaios Xen. A"ab. 2.6.28,
[urisa) [Persia) 2.1.5,2.4.15
IImenias Unnamed ruler Xen. Hell. 5.2.35 NX
(Thebes) (Persia)
Apollophanes Pharnabazos Xen. Hell. 4.1.29;
[Cyzicus] (Penial Plut. Ages. 12.1
4 For tM discussion of the problems involved in this passage, see Hornblower (1982) 174-5.
Appendix A 171
Name and provenance Nam~ and provenance R~ference Remarks
Antalkidas Ariobarzancs Xen. Hell. 5.1.28

[Spana) [Persia)
Archidamos Podantmos and Xtn. Hell. 5.].1]
[Spana) his followers
[Phlius)
Ahr:adatas Unnam~kinK Xen. C)·rop. 5.1.3 F?
[Susa) [Bactria)
~m05thenc."S Anaxinos Atschin.3.224
(Athens) (Oreus)
Thrasyllos Poltmam~tos lsocr. 19.5
(Siphnos) (Unknownl
Speaker in lsocr. 19 Unknown xeno, Isocr.19.18
(Siphnos) (Paros)
Thrasyllos and Unknown xt>no, Isocr. 19.22
the Speaker (Troezen)
(Siphnos) :-c
:-
Philippos Sopaios Isocr.17.43 QJ
[Ath~s) (Bosporu'i) 4-J
:D
Philippos Sopaios' son Isocr.17.43 PXI -.
..::...
(Athens) (Bosporusl
'-'
Hippolaidas Sopa ios' Wn boer. 17.]8 QJ
4-J
[Athens?) (Bosporus)
Isokratcs Klearchos lsocr. Letter 7.13 Cl
(Athens) [Htraclt:a ] ~
bokrates
>-
Tlmmhros Isocr. I.etter 7.13 PXI. NX 2-
(Athens) [Heraclea) c'
Isokratc."S bson
0
Isoer. I.e"" 6.1
[Athens) (Pherae)
lsokrates Pol)'alko~ lsocr. I~tt" 6.1
(Athens) [Pherat)
lsokra tc."S Childrtn of lason Isoer. l.etter 6.4
[Athens) [Pherae)
Satyros Apolloph:ants Dem.19.194;
(Athen~) (Pydna) d. Diod. 16.55
Philip A('s"'hin('s l)('m.18.51.284; NX
(Macedon) (Athens) 19.314
Philip Pythoklc."S Dem.19 ..H4
(Macedon) (Athens)
Philip Philoet a!. I~m. 19.140
(Macedon) [Thehcs)
Pasion Unnamed xeno, Oem.50.18
[Athens) [Lampsacus)
Pasion Kle:an:ax Dem.50..S6
(Athens) ITen~os)
P:asion Ept'r.uos Oem.5CU6
[Athens) (Tenedosl
'for the reading Poly:alko, rat~r that Pol)·alkes. see P. ]4 n. 75
Name and provenance Name and provC1lanct Referenct Remarks
Nikostrltol Unnamed :unoi Dem.53.1-12

(Athms] [Unknown]
Aristonous lykon Dem.52.3
[Athms] [Heradea]
Archebiades lykon Oem. 52.3
[Athens) [Heraclea)
Meidi'l Plootlrchos Oem. 21.1 10
[Athens) [Eretrial
Androkln Nausikrarn [Dml.) 35.8
(Athens) (Carystus)
Alexlnckr Unnamt'd ;dioJCenoi fGrH 12,SF4 (Ch:uts);
(~ta«don) [Unknown) d. Ael. Var. Hist. 8.7
Ht'fakIeitos Kallimachos Call. Ep;gr. 2a
[Haliclrnassus} [Cyrene)
lason Iphitos Apoll. Rhod. 1.208 F
(lokus) [Phocil)
Antiochus II(?) Unn:amed idioxeno; Welles (1934) no. 17
[Syria) [Unknown)
Imylch Lyson IG XIV.279
[Phomieiln Sicily?) [Greek Sicily?)
Kraugis Kleandros Polyb. 10.22.1;
[Megalopolis) (Mantinea] P1ut. Philop. I;
Paul. 8.49.2
Philopoimrn Kleandros Polyb.10.22.1; PXI
[Megalopolis) [Mantinea) Paus. 8.49.2;
Plut. Phi/op. 1
Philopoimm Timolaos Polyb. 20.12.2; d.
[Megalopolis) [Sparta) Paus. 8.51.2;
P1ut. Ph;/op. 1
Archidamos Nikagoras Polyb.5.37.2
[Spartl] [Mn5CflC)
Araras Nikophanes Polyb.2.48.4 PXI
[Sicyon] [Mtgalopolis)
Aratos Kt>rkidas Polyb.2.48." PXI
[Sicyonl [Mtjalopolis)
Eummes Unnamed JCeno; Polyb.24.S.3
[Pergamuml [Rome]
Amlus Unnamed xeno; Polyb.33.18.2
[Pergamum) IRome)
Logbuis Antiochus Hierax Polyb.5.74.4
[~Igel [Syria)
Timothe05 lason Nepos, Timoth. 2
[Athms) [Thessaly]
Hnakles Pholos Apollod.2.5.4 F
[Thebn) [Pholoe)
Philip Unnamed JCeno; Arr. Anab. 1.9.9
[Ma«don) [Thebes)
Appendix A 173
Name and provenance Name and provenance Reference Rtmark~
Alexander Unnamedxeno; Arr. Anab. 1.9.9

(MaCt'don) [~~)
Themistokles Lysithtides Diod. t 1.56;"
(Athens) [Unknown) d. Plut. Them. 26
Belesys Leader of Arabs Diod.2.H.6
[Babylon} (Arabia)
l.ibys Aristokritos Diad. 14.1.3.5
JRuler of area [Sparta)
aroundSiwah)
l.ibys l.ysander Diod. 14.13.5 PXl
(Rulerof arta (Spart2)
around Siwah)
Lihys libys Diod. 14.13.5 PXI
[Rulerof area [Sparta)
around Siwah)
Philip Unnamed xt"noi Diod. 16. 91.5~
[Macedon) [all of Greece)
Hephaistion Ballonymos Diod. 17.47; d.
[Pella) [Tyre) Curt.4.1.19
Xenodikos Unn:amed idio;unoi Diod.20']1
(Acragas) (Gcla)
Agathokles Ophellas Diod.20.70.3
(Therm:ae Himeraeae) (Ruler of Cyrene)
Polyperchon Unnamed ;d;ox~no; Diod.20.20.2
(MaCt'don) [Unknown)
Polychares Euaiphnos Diod. 8.7; d. F
(Messenel (Spartal Paus. 4.9.1
Pherekydes Perilaos Diog. Laen. 1.116 F?
(Syros) [M~sene)
Thcmistokles Nikogenes Plut. "lMm;sI. 26; d.

JAtMns) [Aegae) Diog.l.aert. 1. 101-2
Kotys Unnamed x~nos Plut. Mor. 174d F?
[Thrace) (Unknown]
Solon Epimenidn Plut. Mor. 157d, F?
(Athens) (Crete) H8a
Solon Anacharsis Plut. Sol. 5.2; F?
[Athens) (Scythia) Diog. Laen. I. 102
Solon Chilon Plut. Mor. 151 f f?
(Athens) [Sparta)
Diokles Thain Plut. Mor. 146c F?
(Unknown) [MjJ~sl
Periandt'f Amasis Plut.Mor.152f F?
[Corinth) (Egypt)
Philip Philon Plut. Mor. ) 78c
(Macedon) (Thebes)
"For the problems concerning Lysitheides, d. Hofstetter (1978) no. 205.

, • I .: , • ",:, : : '. ~ .',,: '. ' :
Name and provenance Name and provenance Refermtt Rtmarks
PeriJcles Ktphalos Plut. Mar. 83Sc

(Athens) (Syracuse)
Demaratos Philip P1ut. A/tx. 9; d.
(Corinth] (Macedon) Dem.18.295
KJeinias Unnamed Icings Pluto Arat. 4.2
(Sicyon) [Macedon, Egypt?]
KJeinias Unnamed xmo; Plut. Arat. 3.1
(Sicyon) {Arsos)
Aratos Unnamed xmo; Pluto Arat. 3.1 PXl
(Sicyon] (Af'8os)
An axerxes Anralkidas Plut. Artax. 22.3
[penia] (Sparta)
Lysander Unnamed xmo; Plut. Lys. 8.1
(Sparta) (Miletus)
Lysander Unnamed xmo; Plut. L)'s. 13.4, 19.1;
(Sparta] (Unknown) Paus.7.10.3
Dion Synalos Plut. Dio" 25.6
[Syracuse] [urthage)
Dian K2J1ippos Plut. Mor. 1761; d.
[Syracuse) [Athms] Di01f 55ff.
PhililtOS Unnamed xmo; Plut. Dian 11.4
[Syracuse] (Adria)
Dionysius Archyta. and Plut. Diem 18.2
[Syracuse) fellow Pythagoreans
[Tarentum)
Dionysiul Plato Pluto Dimt 22.1
[Syracuse] [Athens}
Agesilaos Unnamed xmo; Plut. Mor. 212e F?
[Spana] [Asia)
Pyrrhus uius Fabriciul Pluto Pyrr. 20.1 NX,F?
[Epirus) (Rome)
Archill Archias Plut. Pelop. to
[Athens) fTheba)
Alexander Pholcion Plut. PhaG. 17.19
[Macedon) (Athens]
Hiero Pynhus Paus.6.12.3
(Syracuse) (£pi",,]
Agi. Xmias PaUl, 3.8.4; d.
[Sparta) (Elis] 7.10.3
Philip ArtabazOi Curt. 5.9.1,6.5.2; d.
(Macedon) [Penial Diod. 16.52.3
Hephaisrion Two unnamed hospites Curt. 4.1.17; d.
(PeUa] (Sidon) Plut. Mor. 340c
. ,,
" ' .' ,,:
, • I .: ::, : : '. ~ .',.: ,
Appendix A 175
Name and provtnance Name and provenance Remarks
Alkibiadn Alkibiades 7
[Athens) (Unknown)
71f Harpokration's dncription of Alkibiades of Phegous as ·A).x~L(l00u ~ is correct (a

sugg~tion which is somewhat reinforced by the homonymity), then Alkibiades of Phegous
referred to in Andocides 1.65 (0" th~ MY$t~ri~s) cannot be the cousin and fellow-exile of
Alkibiadcs arrested by Thrasyllos in 409 B.C. (Xenophon, HtllnJica 1.2.13). If, Ind only if.
Harpokration is right, Alkibiades of pt,egous is a %~"OJ of unknown origin who became an
Athenian citizen. presumlbly by virtue of the effo"s of his Athenian panner. Cf. Davies (J 973)
no. 600 VI.C.
. ,,
" ' .' .,:
APPENDIX B
XENIA, PROXENIA AND CITIZENSHIP

City granting the Name and provenance Xmosin
Reference deaee and date ofgranrtt granting city Identified by
1. Hdt. 8.136; d. W. I Athens Alexandros Peisistraatos? ex.f. belween ancestors

[before 480 B.C.) [Macedonia] Hippias? [Hdt. 5.9.4; Arin. Arb. Pol. 15.2J
2. Aeschin. 3.258-9; d. Athens Anhmios Themistoklcs? d. Wallace (1970) 202
W.2 [before 48 1/0 B.C.) [ZeJea)
3.lsocr.15.166;d. W.5 Athens Pindar Mepkles xmos;poem
[co 475 B.C.) [Thebes) (Pind. P. 7J
4./GI J 23;d. W. 11 Athens Athenaios ? nam6
[466-40 B.C.I [Thespiae] [decree itself]
n 5. Ion of Chios, FGrH 392; Athens Hmnesilaos Sophokles xmos n
0 d. W.18 (440/39 B.C.) [Chios] [rexritself] 0
"'<
~
CQ
6. Thuc. 2.29.5·,2.67.2 Athens Sadokos, son of Sitallees Nymphodoros of family ties "
'<
::!.
[431-30 B.C.] [Thrace] Abdera [Thuc. 2.29.1) CQ
....
J
~ (pro:cmos) ....
J
~
c. c.
3: 7. Thuc. 2.85.5; d. W. 32 Athens Nikias Niki~the name
3:
....
I»
~
[429-8 B.C.) [Gonyn) general? [text itself] ....
I»
~
~. 8. Thuc. 3.52.5 Spana La leon ;)
name ~
!!. [427 B.C.) [Plaraea] !!.
[teXt itself]
9. Amen. 3.108f-109a; Athens Ana Dcmosthmes or hereditary pbilos
Suidas,s.v.Anas;d. W. 70 [c. 41413 B.C.) [Messapial Eurymedon [Thuc.7.33.4J
10. Thuc. 8.92.8;cf. W. 74 Athens Thoukydides namesake in name
[41211 B.C.) [PhanaJus) Athens? [text itself]
II./G I) 118; d. W. 27 Athens Apollodoros Alkibiades ex.f.
and 86 [408nB.c.J [SelymbriaJ [decree itself]
12./GI J I17;d.W.90 Athens Archclas Andocides xmos
[407/6 B.C.) [Macedonia] [Andoc.2.11]
tJ. Thuc. 5.43.2, 6.89.2 Spana Alleibiades Endios hereditary xmos
[sincec. 550 B.C., [Athens) [ThUc. 8.6.3]
d. Davies (1971) 16)
City granting the Name and provenance Xnrosin
Refcnncc decree and d2te ofgrantft granting city Wentified by
14./G 1'113-; DmI. 12.10; Athens Evagoras Nikophemos? ex.f.

d.O.3 (401B.C.) (Salamis] Aristophann? [Lys. 19.23.36;
Konon? lsocr. 9.54. 51)
IS./G III 20. with AthalS Evagoras Konon ex.f.
Lewis and Stroud (1979) (39312 B.C.] (Salamis] [decree irsclf)
16./G III 103 -; d. O. 10 Athens Dionysios Eunomos xmos
(369/8 B.C.] (Syracu5C) [Lys.19.19]
l
17./G II 106;ef. T.13S Athens Koroibos namesake name
(36817 B.C.] (Sparta]
18./G 112 110 A~ Mmdaos Timotheos ex.f.
[36312 B.C.] (Pelagonia] (decree itself)
son's name Athmaios
(T.148]
19./G 11 2 130 Athens Lachar~ son of Charcs Chares,the name
(355/4 B.C.] (Apollonia] ~raJ? (dKrec itsdf)
20. IG 11 2 133 AthalS Philiskos Generals of ewrgesi4
[355/4 B.C.] (Sestus] 36615 B.C. (decree irsclf)
21. Dem. 21.200 Euboea? Demosthcnn Ploutarchos xmos
(bdore 350 B.C.] (Athens] (~m.21.110]
l
22.IG Jl 237-; d. 0.16 Athens Phonnion Phormion name; inherited Xnti4
(33817 B.C.] [Acamania] [Thuc. 2.80; d. O. vol. 4 p. 44]
23./G II l 405-;d. 0.21 Athens Amyntor. son of namesake? name
(33~3 B.C.] OenKtrius [PA no. 750]
[Unknown I
KEY: T... Too (i 948) vol. 2 W.· M. B. Walbank (1978) O. = Osborm (1981-83) ex.!. = exchange offavours
• • citizenship; abKncc of - = proxmy
, • I .; , • ,:::, : : '. ~ .',.: ,
APPENDIX C
NETWORKS OF RITUALISED FRIENDSHIP

DURING THE PELOPONNESIAN WAR
(432 - 404 B.C.)
The identification of the relationships has been made according to the

characteristics of the bond traced in the previous sections: terminology of
ritualised friendship (xeno;, philo;, epitede;o; etc. from different cities,
Section 2.1), terminology of proxeny (Section 5.4), or exchange of resources
or services characteristic of ritualised friendship (Chapters 4 and 5).
Proxenies are only listed from the point of view of the proxeno;, not of
the cities granting the title.
Individuals attested epigraphically have only been included if both their
name and origin are known. In doubtful cases I have accepted M.B. Wal-
bank's conjectures.
, • I .: .,: ::, : : '. ~ .',.: ,
180 Networks during the Peloponnesian War
Political unit and name Relarionship with ... Rtference Idtnrifitd by ...
Acha~1I
lykon AmENS W.50 prx.t.
Abdna
Nymphodoros Sitalkes 11me. 2.29.1 d. W. 30 r.f.r
[l1uace)
NymphodorO$ Perdikkas Thue.2.29.6 r.f.r
[Macedonia)
Nymphodoro5 All-fENS Thue.2.29.1-3 prx.r.
Acaman;a
Unnamed Demosmenes Thuc.7.57.10 rJ.r.
(ArhmsJ
Alea
Asteas All-fENS W.49 prx.r.
Andros
Antiochides ATHENS W.60 prx.r.
Phanosthenes All-fENS W.60 prx.t.
Argos
Alkiphron SPARTA Thuc.5.59.4 prx.t.
UnnarMd SPARTA Thuc. 5.76.2,5.82.3, prx.t.
5.83.2
Thrasyllos Agis Thuc.5.59.4 prx.t.
(Sparta]
Eustrophos SPARTA Thuc.5.40.3 prx.t.
Aison SPARTA Thuc. 5.40.3 prx.r.
UnnarMd AUcibiadn Thuc.6.61.3 rJ.t
[Arhms]
Athnu
Alkibiades Endios Thuc. 6.89.2, 5.45.2, r.f.t.; ex.f.
(Sparta] 8.6.3
Alkibiades Unnamed Thuc.8.17.2 r.f.t
[Mil~os)
Alkibiades Chalkideus Thuc. 8.17.2 exJ.
lSparta'
Alkibiades Unnamed Thuc. 6.29.3, d. rJ.t
lManrineaJ 6.61.5
Alkibiades Unnamed Thuc. 6.61.3; lsocr. 16.9 r.f.t.x; ex.f.
lArgos)
Alkibiades Unnamed [Andoc.] 4.30; Satyros ex.f.
(Ephesos) in Athen. 12.534d,
d. Plut. AI,. 8.6
Alkibiades SPARTA Thuc. 6.89.2, 5.43.2 prx.t.
Alkibiade-s Unnamtd Thuc. 8.14.1; d. rJ.r.
IChios} [Andoc.) 4.30
. ,,
" ' .' .,:
Appendix c: 181
Politic~1 unit and name Relationship with ... Referentt Identified by ...
Alkibiades Tissaptlcmes TImc.8.45-8 r.f.t.; e:d.

[Persia)
Alkibiades Apollodoros W.86 prx.r.; ex.f.
(Selymbria)
Alkibiades Seuthes Xen. Hell. 1.5.17; ex.f. (estate)
[TIlrace) 2.1.25; Diod. B.l0S.3ff.
Alkibiades Pharnabazos Xen. Hell. 1.3.12 d.t.
[Persia I
Alkibiades Unnamed Xen. Hell. 1.4.8 ex.f.(?)
{Ceramic gulf)
Andocides Archelaos Andoc.2.11 d.t.; ex.f.
[Macedonia)
Demosthenes Anas Thuc 7.33,4 d.t.
IMesupia)
Demosthenes Unnamed Thuc.7.57.10 d.t.
[Akarnania)
Eurymedon Anas Thuc.7.33.4 d.t.
[Messapia)
Hipponikos SPARTA Inferred from prx.t.
Xen. Hell. 6.3.4
Kallias SPARTA Xen. Symp. 8.39 prx.t.
Lakedaimonios, SPARTA Inferred from prx.t.
son of Kimon Andoc. 3.2; d.
Aeschin. 2.172
Perikles Archidamos Thuc. 2.13.2 d.t.
[Spanal
Nikias Pausanias lys.18.10 d.t.
[Sparta)
Nikia5 SYRACUSE Diod. 13.27.3 prx.t.
GmriruJ
Xanthippes ATI-fENS W.63 prx.t.
Chios
T ydeus, son of Ion Kimon TIme. 8.38.3; d. ~~.f.
[Athens) F. Jacoby in CQ
61(1947)I,n.9
CIa:ommae
Herakleidn ATHENS W.47 prx.t.
Cleonae
The fatherof ATHENS W.80 prx.t.
Echembrotos
C"idus
Proxenides ATIiENS W.64 prx.t.
Political unit and name Relationship with ... Reference Identified by ...
Co/opho"
ApoHophanes ATHENS W.39 prx.t.
Corcyra
Peithios ATHENS Thuc.3.70.3 prx.t.
Corilfth
AristeUs POTIOAEA Thuc. 1.60.3; 2.67.4 prx.t.
Unnamed CORCYRA Thuc. 3.70.1; prx.t.
Diod. 12.57.2
Gorty$
Nikias ATIiENS Thuc. 2.85.5 prx.t.
Polypos ATHENS W.91 prx.t.
Hnacl~a
Sotimos ATHENS W.46 prx.t.
LIma
Aristonous ATHENS W.29 prx.t.
Maudo"i4
Perdikkas Nymphodoros Thuc.2.29.6 d.t.
[Abdera]
Perdikkas Unnamed Thuc.4.137.2 r.f.t.
[Thesuly]
Perdikkas Nikonides Thuc.4.78.2 d.t.
[TheS$lly)
Krisan ATHENS W.24 poe.t.
Archelaos Andokides AmicK. 2.11 r.f.t.; ex.f.
[Athens)
Archela(o)s ATHENS W.90 prx.t.
Meuapi4
Artas Demosthenes and/or Thuc. 7.33.4;cf. W. 70 d.t.
Eurymedon?
[Athens)
Milno$
Unnamed Alkibiades Thuc.S.17.1 d.t.
[Athens)
Mytik"e
Dexandros ATHENS Thuc. 3.2.3; Arist. Pol. prx.t.
1304a9;d. W.38
The Pnsi4" Empirt
Phamabllos Alkibiades Xen. Hell. 1.3.12 r.f.t.
[Athens]
Appendix C: uu
Political unit :and n:ame ... Relationship ..... Ith ... Rderence identifIed "y ...
Tissaphern~ Alkibl:ad~s Thuc. 8.47.2 d .t.
(Athens!
Orchommus
Potamodoros ATHENS W.4.~ prx.r.
Eurytion ATHENS W.4.S prx.r.
Pharsa/us
Thoukydides ATHENS W.74 prx.l.
Ph/ius
Polystraros ATHENS W.48 rrx.l.
P/alaea
Naukleid~ and his Eurymachos Thuc.2.2.2 d.t.
followers (ThchesJ
~
Sdalhus
Oiniadcs ATHENS W.87 prx.t.
-
ill
~
Se/ymbria v
~
L
Apollodoros ATHENS W.86 ex.f.:prx.t.
'-'
Empedos ATHENS W.27 rrx.r. Q)
~
Sici/)'
Cl
Archonides
[kmon
ATHENS
ATHENS
W.66
W.66
prx.t.
rrx.r.
->-.
~
Sparta ,0
,-.-J
Agis Thrasyllos Thuc.5.59.4 ex .f.
(Argos!
Archidamos Perikl~ Thuc:. 2.1 ].2 r.f.l.
(AthensJ
Brasidas Unnamed Thuc. 4.78.1 d.r.
[Thes.sal)·)
Unknown bkon Thuc ..~.52.5 r. f. t.
(Platac:a!
PausaOlas Niki3s Lys.18.10 d.t.
(Athens!
Klearchos BYZANTIUM Xen. Hell. 1.1..35 prx.t.
Chalkideus Alkibiades Thuc.8.17.2 ex.f.
Taras
H~e1()(hos ERETRIA Meiggsand lewis prx.t.
(196Q) no. 82
Herakle1tos ERETRIA Syll.' 106 prx.t.
TeRea
Unnamed SPARTA Thuc.5.64.1 prx.t.
Political unit and n:ame . . . Relationship with ... Rderence identified by ...
Thasos
Sons of Apemamos ATHENS W.61 prx.t.
Sthorys ATHENS W. 78 prx.t.
Tb~lHs
Eurymach05 Naukleides and Thuc.2.2.2 d.t.
followers
IPlataeal
Thessaly
Nikonides Perdikkas Thuc.4.78.2 d.t.
[Macedonial
Unnam~ Perdikkas Thuc.4.U7.2 d.t.
IM3ccdonia I
Kallippo~ ATHENS W.65 prx.t.
Strophakos Chalkidians Thuc. 4.78.1 prx.t.
Thoukydid('S ATHENS Thuc. R.92.2 prx.t.
Thrace
Silalkes N)'mphodoros Thuc.2.29.1 r.f.t.
IAhderal
KEY: d.t... ntuali~ friendship terminology prx.t.... proxeny terminology

ex,f.... exchange of faVOUr!! r. = ritual W... M. B. Walhank (197M).
BIBLIOGRAPHY
Adcock, Sir F. and D.J. Mosley (1975) Diplomacy in Ancient Gruce. London.
Adkins, A.W.H. (1960) M~';tand R~sponsibility. A StudyofG,~~k Valu~s. Oxford.
(1963) , "Friendship" and "self-sufficiency" in Homer and Aristotle', Classical
Quarterly. n.5. 13, 33-45.
(1971) 'Homeric values and Homeric society" Journal of Hellenic Sttldi~s 91.
1-14.
(1972a) Moral Valu~s and Political R~haviour in Ancient Greece, from HomeT to
the End of the Fifth Century. London
(1972b) 'Homeric gods and the values of Homeric society', Journal of Hdlenic
Studies 92, 1-19.
Amit, M. (1970) 'Hostages in Ancient Greece' Rilfista di filo/ogia e di istruzione
classica 98, 129-47.
Andreades, A.M. (19.13) A History of Greek Public Fi"ance, translated by C.N.
Brown. Cambridge, Mass.
Andreau J., P. Schmitt and A. Schnapp (1978) 'Paul Ve)'ne et I'evergetisme" Annates
E.s.C. 33, 307-25.
Atkinson, K.M.T. (1952) 'Some ohsen'ations on Ptolemaic ranks and titles" Aeg)'p-
tus 32, 204-14.
(1972) 'A hellenistic land-conveyance" Historia 21, 45-74.
Aurenche, O. (1974) Les groupes d'Alcibiade, de l.iogoras et de Tellcros. Paris.
Aymard, A. (1954) 'Philip~ de Macedoine, otage a Thebes" R~vue des etudes
anciennes 56, 15-36. [Reprinted in Etudes d'histoire ancienne (Paris, 1967),
418-38.J
(1958) 'Us ttrangers dans les cites grecques aux temps c1assiques" Recueils de la
Socihi J~an Bodin 9, 119-39.
Badian, E. (1958) Foreign Clienttlae. Oxford.
Bagnall, R.S. (1976) The Administration of the Ptolemaic Possessions Outside
Egypt. Lciden.
Belshaw, C.S. (1965) Traditional F.xchange and Modem Markets. Englewood Cliffs,
N.j.
Bennet, M. (1979) 'Spiritual kinship and the baptismal name in traditional European
society', in L.O. Frap~1I (ed.), Principalities, Powers and Estates: Studits in
Medieval and Early Modem Gor.'ernment and Society (Adelaide) 1·13.
Benveniste, E. (1951) 'Don et echange dans Ie vocabulaire indo-euro.,een" L'Annee
socio/ogique 3e serie, II, 7-20. IReprinted in Problemes de linguistique gener-
ale, vol. I ( Paris, 1966), 315-26.
(1969) u vocabulaire des institutions indo-~uropeen"es, 2 vols. Paris.
Berve, H. (1926) Das Alexanderreich auf prosopographisch" Grundlage, 2 vols.
Munich.
185
186 Bibliography
Bickerman, E. (19]8) Institutions d~s Seleucid~s. Paris.

(1944) 'An oath of Hannibal'. Transactions and Prouedings of the American
Philological Association 75, 87- 102.
(1950) 'Couper une allian~', Archive d'histoire du droit oriental 5. [Reprinted in
Studi~s in Jewish and Christian History (Lciden, 1976) 1-32.)
(1952) 'Hannibal's covenant', American Journal of Phil%gy 73, 1-23.
Black-Michaud, J. (1975) Feuding Societies. Oxford.
Bloch, Marc (1912) 'Les formes de la rupture de I'hommage dans I'ancien droit
feodal', Nouvelle revu~ historique de droit fran~is et etranger 36, 141-77.
(1961) Feuda/ Sodety, 2 vols., translated by L.A. Manyon, first published 1940.
Chicago.
Bloch, Maurice and G.S. Guggenheim (1982) 'Compadrazgo, baptism and the
symbolism of a second binh', Man, n.s. 16, 376-86.
Blok, A. (1969) 'Variations in patronage', Soci%gisch~ Gids 16, 369-86.
(1971) 'On brigandage with special reference to peasant mobilization', Socio/o-
gische Gids 18, 208-16.
(1974) The Mafia of a Sicilian Village, 1860-1960. Oxford.
(1981) 'Rams and billy-goats: a key to the Mediterranean code of honour', Man,
n.s. 16, 427-40.
Boissevain, J. (1966) 'Patronage in Sicily', Man, n.s. 1, 18-33.
(1974) Friends of Friends. Oxford.
Boissevain, J. and j.C. Mitchell (eds.) (1973) N~twork Analysis: Studi~s in Human
Interaction. The Hague.
Bolkesrein, H. (1939) Wohlt(jtigk~it lind Arm~npfl~g~ im vorchristlich~n Alurtum.
Utrecht.
Bravo, 8. (1980) 'Sylan', Annali della Scuo/a Norma/~ Super;or~ di Pisa 10/3,
675-987.
Briant, P. (1982) Rois, tributs ~t paysans. Paris.
Brillant, M. (1877-1919) 'Xenia', in Ch. Daremberg and Edm. Saglio (eds.), Diction-
nair~ dts antiquites grecqu~s et romain~s IX, 1008-10.
Brommer, F. (1960) Vas~nlisttn lur gritehischen Heldensage, 2nd edn. Marburg.
Brunt, P.A. (1952) 'Thucydides and Alcibiades', RetJUe des itutks grtcques 65,
59-96.
(1965)' "Amicitia" in the late Roman Republic', Proceedings ofth~ Cambridge
Philological Soci~ty n.s. 2, 1-20. I Reprinted in R. Seager (ed.), The Crisn of
the Roman Republic (Cambridge, 1969) )99-218.1
(1969) 'Euboea in the time of Philip 11', Classical Quarterly n.s. 19,245-65.
Calhoun, G.M. (1913) Athenian Clubs in Politics and Litigation. Austin, Texas.
Campbell, J.K. (1964) Honour, Family and Patronage. Oxford.
Canledge, P.A. (1979) Sparta and Lakonia. A Regional History 1300-362 B.C.
london.
(1981) 'The politics of Spanan pederasty', Proceedings of the CAmbridge Philolo-
gical Soci~ty n.s. 27, 17-36.
(1982) 'Sparta and Samos: a special relationship?', Classical Quarttrly 32, 243-
65.
(1984) 'A new lease of life for Lichas son of Arkesilas'? Uv~rpool Classical
Monthly 9.7 Ouly 1984) 98-)02.
Bibliography 187
Cawkwcll, G.L. (1963) 'Demosthenes' policy after the Peace of Philocrates', Classic-
al Quarterly 13, 120-38 and 200-13.
Chroust, A.-H. (1959) 'Treason and patriotism in ancient Greece', Journal of the
Histo r)' of Ideas 15, 280-8.
Coldstream, J.N. (1983) 'Gift exchange in the eighth century B.C:, in R. Hagg (eu.),
The Greek Renaissance of the Eighth Centflry B.C.: Tradition and ImtolJation
(Stockholm) 201-7. [Proceedings of the Second International Symposium at
the Swedish Institute in Athens, 1-5 June, 1981.)
Connor, W.R. (1971) The New Politicians of Fifth-Century Athens. Princeton, N.J.
Cook, J.M. (1983) The Persian Empire. London
Cordell, D. (1979) 'Blood partnership in theory and practice: the expansion of
Muslim power in Dar AI-Kuti', Journal of African History 29, 379-84.
Corradi, G. (1929) Studi ellen;st;ci. Turin.
Daux, G. (1937) 'Alcibiadc, proxene de l.acedemone', in Milanges offerts a A.-Ai.
Desrousseaux, 117-22. Paris.
Davies, j.K. (1971) Athenian Propertied Families. Oxford.
(1981) Wealth a"d the Power of Wealth in Classical Athms. New York.
(1984) 'Cultural, social and economic features of the Hellenistic world', in F.W.
Walbank (ed.), The Cambridge Ancient History 2nd cdn., Vll.l, 257-320.
Davis, j. (1977) People of the Mediterranea". london.
Donner, H. (1961) 'Der "Freund" des Konigs', Zeitschri{t fiir die Alttestammtl;che
Wissenschaft, n.s. 32, 269-77.
Dorey, T.A. (1956) 'Aristophanes and Cleon', Greece & Rome, n.s. 3, 133-9.
Dorsinfang-Smets, A. (1958) 'us ttrangers dans la societe primitive', Recueils Je la
societe Jean Bodin 9, 59-73.
Douglas, M. (1960) 'Blood-debts and c1ientship among the Lele', Journal of the
Royal Anthropological Institute 90, 1-28.
Dover, K.J. (1974) Greek Popular Morality in the Time of Plato and Aristotle.
Oxford.
Du Boulay, J. (1974) Portrait of a Greek MOllntain Village. Oxford.
Edlund, E.M.1. (1977) 'Invisible bonds: clients and patrons through the eyes of
Polybius', Klio 59, 129-36.
Ehrenberg, V. (1937) 'When did the polis rise?', JOllrnal of Hellenic Studies 57,
147-59.
(1969) The Greek State, 2nd edn. Oxford.
Ei5C'nsradt, S.N. (1956) 'Ritualized personal relations" Man 96, 90-5.
(1963) The Political Systems of Empires. New York.
Eisenstadt, S.N. and l. Roniger. (1980) 'Patron-client relations as a model of
structuring social exchange', Comparative Studies in Society and History 22,
42-77.
(1984) Patrons, Clients and Friends. Cambridge.
Evans-Pritchard, E.E. (1933) 'Zande blood-brotherhood', Africa 6, 369-401.
Finley, Sir M.1. (19.52) Studies in LAnd and Credit in Ancient Athens, SDO-200 R.C.:
the Horos Inscriptions. New Brunswick.
(1973a) The Ancie"t F.conomy. London.
(197.3b) Democracy Ancient a"d Modern. l.ondon.
(t 975) The Use and Abuse of History. London.
188 Bibliography
(1977) The World of Odysseus, 2nd edn. London.

(1979) FArly Greect: The Bronu and Archaic Ages, 2nd edn. London.
(1981) Economy and Society in Ancient Greect, edited by B.D. Shaw and R.P.
Saller. London.
(1983) I'olitics in the Ancient World. Cambridge.
Firth, R. (1936) 'Bond-friendship in Tikopia" in Custom is King: Essays Presenud
to R.R. Mardt on his Seventieth Birthday. London.
(1972) 'Verbal and bodily rituals of greeting and parting\ in J.S. La Fontaine
(ed.), The Interpretation of Ritual; Essays in Honour of A.I. Richards. (lon·
don) 1-38.
Fischer, Th. (1979) 'Zur Seleukideninschrih von Hefzibah', Zeitschrift fur Papyrolo-
gie und Epigraphik .1.3, U 1-8.
Forrest, W.G. (1966) The Emergenu of Greek Democracy. London.
Fortes, M. (1962) 'Ritual and office in tribal society', in Gluckman (ed.), (1962)
53·88.
(1968) 'Of installation ceremonies', I'roceedings of the Royal Anthropological
Institute 98, 5-20.
(1969) 'Kinship and the axiom of amity', in Kinship and the Social Order
(London) 219-49.
Foster, G.M. (1961) 'The dyadic contract: a model for the social structure of a
Mexican ~as~nt village', America" Anthropologist 63, 117]·92. [Reprinted
in Schmidt et al. (eds.) (1977) 15-28.1
Funck, B. (1978) 'Zu den Landschenkungcn hellenistischen Konige', Klio 60, 45-55.
Ganshof, F.L (t 964) Feudalism, translated by Ph.Grierson, first published 1944.
London.
Gauger, J.-O. (1977) 'Zu einem offenen Problem des hellcnistischen Hoftitelsystems.
Ein persischer Ehrentitel syngenes?', in A. Lippold et al. (eds.), Bonner Fest-
gobe Johannes Straub, 137-58. [Btiheft 39 of Bonner Jahrbucher.l
Gauthier, Ph. (1971) 'l..es XENOI dans les textcs Atheniens', Revue des etudes
grecqlles 84, 44-79.
(1972) Symbola. l.es ttrangers et la justiu dans les cites grecqlles. Nancy.
(1973) 'Notes sur l'etranger et "hospit3Iire en Grece et a Rome', Ancient Society
4, 1-21.
(1979) 'EEArorH IITOY: Samothrace, Hippomedon et les Lagides', Historia
28, 76-89.
Gellner, E. (1962) 'Patterns of rural rebellion in Morocco during the early years of
independence', European Journal of Sociology 3, 297-.11 1. [Reprinted in
Gellner (1981) 194-206.J
(1979) 'Rulers and tribesmen', Middle Eastern Studies 15,106-13. [Reprinted in
Gellner (1981) 221-30.]
(1981) Muslim Society. Cambridge.
(1983) Nations and Nationalism. Oxford.
Gellner, E. and J. Waterbury (eds.) (1977) Patrons and Clients in Mediterranean
Societies. London.
Gelzer, M. (1969) The Roman Nobility, translated by R. Srager. Oxford.
Gennep, A. Van. (1960) The Rites of Passage, translated by M.B. Viudom and G.L.
Caffee, first published 1909. London.
Bibliography IR9
Gernet, I.. (19.38) 'us nobles dans la Grece antique', Anna/~s d'histoirt iconomique
et sociale 10, 36-43.
(1948) 'La notion mythique de la valeur en Grece'. Journal dt psycho/ogie
nONnale et pathologique 41, 415 -62. [Reprinted in 1.. Gerner, Anthropo/ogie
de la Grece antique (Paris, 1968) 93-173; English translation in R.G. Gordon
(ed.), Myth, Religion and Socitty (Cambridge, 1981) 111-46.1
(1953) 'Mariages de tyrans', in Hommages a l,ucitn Febvre (Paris) 41·53.
Gerolymatos, A. (1986) Espionage and Treason. A Study of th~ Proxenia in Political
and Military Intelligence Gathering in Classical Greece. Amsterdam[Forth-
coming. I have been able to see the typescript through the generosity of Mr
J.e. Gieben.]
Gillis, D. (1979) Collaboration with the Persians. Wiesbaden. [Historia, Einuls-
chriften 34.]
Glotz, G. (1904) I..a solidariti de la famille dans Ie droit criminel en Greet. Paris.
(1929) The Greek City and its Institutions, translated by N. Mallinson. London.
Gluckman, M. (1962) Essays on the Ritual of Social Relations. Manchester.
(1965) Politics, lAw and Ritual in Tribal Society. Oxford.
Gombrich, Sir E.H. (1982) 'Ritualized gesture and expression in an', in Tht Imagt
and the Eye, 63-77. Oxford.
Gomme, A. (1945-1981) A Historical Commentary on Thucydides, 5 vols. Oxford.
[vols. 4 and 5 with A. Andrewes and K.J. Dover]
Goody, E. (1972) , "Greeting", "begging", and the presentation of respect', in .J,S.
La Fontaine (ed.), The Interpretation of Riwal; Essays in Honour of A.I.
Richards (London) 39-71.
Goody, J. (1961) 'Religion and ritual: the definitional problem', British Journal of
Sociology 12, 142-64.
(1963) 'Feudalism in Africa?', Journal of African History 4, 1-18.
(1983) The Dtvelopment of tht Family and Marriagt in F.urope. Cambridge.
Goody, J. (ed.) (1973) The Charact~r of Kinship. Cambridge.
Gould, J. (1973) 'Hikereia', Journal of Hellenic Studies 93, 74-103.
Greenhalgh, P.A.L. (1972) 'Patriotism in the Homeric world', Historia 21, 528-37.
Grierson, P. (1959) 'Commerce in the Dark Ages: a critique of the evidence"
Transactions of th~ Royal Historical Society 9, 123-40.
Griffith, G.T. (1935) The Mercenaries of the Hellenistic World. Cambridge.
Gschnitzer, F. (1973) 'Proxenos', in Real-Encyclopiidi~ der classischen Altertums-
wissenschaft Supp!. 13, 629-7.10.
Gudeman, S. (1971) 'The Compadrazgo as a reflection of the natural and spiritual
person', Procudings of the Royal Anthropological Institute 101, 45-67.
Habicht, Chr. (1958) 'Die herrschende Gesellschaft in den hellenistischen Monar-
chien" Viertd;ahresschrift fur Sozio I- und Wirtschaftsgeschichte 45, 1-16.
(1979) Untersuchungen zur politischen Geschichte Athens im 3. Jahrhundert
v.Ghr. Munich.
Hammel, E.A. (1968) Alternative Social Structures and Ritual Relations in th~
Balkans. Englewood Cliffs, N.J.
Hammond. N.G.L. and G.T. Griffith (1979) A History of Macedonia 550-336 B.C.,
voUr. Oxford.
Hands, A.R. (1968) London. Chariti~s and Social Aid in Creect and Rome. London.
190 Bibliography
Henry, A.S. (1983) Honollrs and Privileges in Athenian Decrees. Hildesheim,

Zurich, New York.
Herman, G. (1980/1) 'The "friends" of the early hdlenistic rulers: servants or
officials?" Talanta 12/13, 103-49.
Hiltbrunner, O. (1978) 'Hosris und EENOI" in S. Sahin et al. (eds.), Studien VIr
Religion und Kultur Kleinasiens. Festschrift fur Friedrich Karl Domer tum 65.
Geburtstage. Leiden.
Hiltbrunnff et al. (1972) 'Gastfreundschaft', in Th. Klauser (ed.), Reallexilt.on fUr
Antike und Christentum VIII (Stuttgart), 1061· 123.
Hobsbawm, E.J. (1969) Bandits. London.
Hofstetter, J. (1978) Die Griechen in Pers;en. Prosopograph;e der Griechen im
persischen Reich vor Alexander. Berlin.
Homans, G.c. (1957-58) 'Social behaviour as exchange" American journal of
Sociology 63, 597-606.
Hopkins, K. (1983) Death and Renewal. Cambridge.
Hopkins, K., P. Garnsey, and C.R. Whittaker (eds.) (1983) Trade in the Ancient
Economy. London.
Hornblower, S. (1982) Mausolus. Oxford.
Humphreys, S.C. (1978a) Anthropology and the Greeks. London.
(1978b) 'Public and private interests in classical Athens" The Classical journal
20, 97-104. [Reprinted in The Family, Women and Death (London, 1983)
22-32.]
Ihering, R. von. (1887) 'Die Gastfreundschaft im Altertum', Deutsche Rundschau
51,357-97.
Jonts, A.H.M. (1940) The Greek City from Alexander to justinian. Oxford.
Kakridis, H. (1963) lA notion de I'amitie et de I'hospitalitt chet Homere. Thessalo-
OIca.
Karavites, P. (1980) , "Euergesia" in Herodotus and Thucydides" Revue interna·
t;onale des dro;ts de I'ant;quite 27, 69·79.
Kaufman, R. (1974) 'The patron·c1ient con~pt and macropolitics: prospects and
problems" Comparative Studies in Society and History 16, 287·308.
Kenna, M.E. (1976) 'The idiom of family" in Peristiany (ed.) (1976) 347-62.
Kiefer, T.M. (1968) 'Institutionalized friendship and warfare among the Tausug of
Jolo', Ethnology 7, 225·44.
Kienast, D. (1973) Philip 11 vorr Makedonien und das Reich der Achaemeniden.
Munich.
Killen, J.T. (1985) 'The linear B tablets and the Mycenaean economy" in A.M.
Davies and Y. Duhoux (eds.), Linear B: A 1984 Survey (Louvain·le-Neuve)
241·305. [Proceedings of the My~naean Colloquium of the Vllhh Congress
of the International Federation of the Societies of Classical Studies, Dublin, 27
August-l September 1984.}
Kirk, G.S. (1968) 'War and the warrior in the Homeric poems', in J.·P. Vernant
(ed.), Problmres de fa guerre en Grtce anc;enne (Paris) 93·117.
Kreissig, H. (1977) 'Landed property in the "Hellenistic" Orient" Eirene 15, 5-25.
(1978) Wirtschaft und Gesellschaft im Sefeukidenreich. Berlin.
landau, Y.H. (1966) 'A Greek inscription found near Hefzibah', Israel Exploration
journal 16, 54·70.
Bibliography 191
Lane, K. (1968) 'Kollektivhesitz lind Suarsschan: in Griechenl:lnd', N,uhrichtrn drr

Akademie der Wissenschaften in Gottingen, Phil.-Hist. Klasse (194617), 64-
75. [Reprinted in G. Gigon et al. (eds.), Kleine Schriften (Munich) 294-312.1
Lauffer, S. (1974) 'Die Liturgien in der Krisenperiode Athens', in E.e. Weiskopf
(ed.), Hellenische Poleis, voLl (Berlin) 146-59.
Le Goff, J. (1980) 'The symbolic ritual of vassalage', in Time, Work, and Culture i"
the Middle Ages, translated by A. Goldhammer (Chicago and I.ondon) 2.P-
87.
Leach, E.R. (1966) 'Ritualization in man in relation to conceptual and social
development', Philosophical Transactions of the Royal Society B, 251, 403-8.
(1972) 'The influence of cultural context on non-verbal communication in man',
in R.A. Hinde (ed.), Non- Verbal Communication (Cambridge) 314-44.
Lecrivain, Ch. (1877-1919) 'Hospitium', in Ch. Daremberg and Edm. Saglio (eds.),
Dictionnaire des antiquites grecques et romaines 111.1, 294-302.
Levy, H.L. (1963) 'The Odyssean suitors and the host-guest relationship', Transac-
tions and Proceedings of the American Philological Association 94, 145-53.
Lewis, D.M. (1977) Sparta and Persia. Leiden.
Lewis, D.M. and R.S. Stroud (1979) 'Athens honors King Euagoras of Salamis',
Hesperia 48, 180-93.
Losada, L.A. (1972a) The Fifth Column in the Peloponnesian War. leiden.
(1972b) 'Fifth columns in the Pdoponnesian war: How they worked and the
defense against them', Klio 54, 125-45.
Marek, Chr. (1984) Die Proxenie. Frankfurt.
Mauss, M. (1921) 'Une forme ancienne de contrat che7. les Thraces', Re,',tt' des
etudes grecques 34, 338-97.
(1966) The Gift, translated hy I. Cunnison, first puhlished 1925. London.
Mayer, A.C. (1966) 'The significance of quasi-groups in the study of complex ,o.
societies', in M. Banton (ed.), The Social Anthropology of Complex SOCIeties '-.-/'
(London) 97-122. [Reprinted in Schmidt et al. (eds.) (1977) 43·S4.1

McGregor, M.F. (1940) 'The pro-Persian parry at Athens from 510 to 480 B.C." in
Athenian Studies, Presented to William Scott F~guson (Cambridge, Mass.)
70-95. [Harvard Studies in Classical Philology, Suppl. voLl.]
Meiggs, R. (1949) 'A note on Athenian imperialism', Classical Rel/iew 63, 9·12.
Meiggs, R. and D. Lewis. (1969) A Selection of Greek Historical Inscriptions.
Oxford.
Middleton, D.F. (1982) 'Thrasyboulos' Thracian support" Classical Quartt"rl)1 .U,
298-303.
Millen, P. (1983) 'Maritime loans and the structure of credit in founh-centut)·
Athens', in Hopkins et al. (eds.) (1983) 36-52.
Mintz, S.F and E.R. Wolf (1950) 'An analysis of ritual co-parenthood (Compadraz-
go)', Southwestern Journal of Anthropology 6,341-54. [Reprinted in Schmidt
et al. (eds.) (1977) 1-15]
Momigliano, A. (1934) Filippo il Macedone. I-lorence.
Mommsen, Th. (1864) 'Das romische Gastrecht und die romische Clientcl', in
Romische Forschungen (Berlin) ]2 t -90.
Mooren, L. (1975) The Aulic Titulature in Ptolemaic Egypt. Introduction and
Prosopography. Brussels.
192 Bibliography
Moretti, L (1977) 'Milero, Ie sue colonie, e I'istituro dell'evergesia" Rivista di

fi/%gia et di istruvone classica lOS, 5-11.
Mosley, D.]. (l971a) 'Cimon and the Spartan proxeny', Athenaeum n.s. 49,431-2.
(1971b) 'Spartan kings and proxmy" Athmamm n.s. 49, 433-5.
Munn-Rankin, J.M. (1956) 'Diplomacy in western Asia in the early second millen-
nium B.C." Iraq 18, 68-11 O.
Murray, O. (1983a) 'The Greek symposion in history" in Tria Corda. Scritti in
onore di Arnaldo Momigliano (Como) 257-72.
(l983b) 'The symposion as social organisation', in R. Hagg (ed.), The Greek
Renaissance of the Eighth Century B.C.: Tradition and Innovation (Stock-
holm) 201-7.
Neumann, G. (1965) Gesten und Gebiirden in d~ griechischen Kunst. Berlin.
Ormerod, H.A. (1924) Piracy in the Ancient World. Liverpool.
Osborne, M.J. (1981-83) Naturalization in Athens, 4 vols. Brussels.
Palmer, L.R. (1956) 'The concept of social obligation in Indo-European" in G.
a
Redard (ed.), Hommages Max Niedermann (Brussels) 258-69. [Coli. Lato-
mus, 23.1
Parke, H.W. (1933) Greek Mercenary Soldiers. Oxford.
Pecirka, J. (1966) 'A note on Aristotle's conception of citizenship and the role of
foreigners in fourth century Athens', Eirene 6, 23-6.
P~rick, V. (1982) 'Supplication in the Iliad and the Odyssey" Tra"sactions Qnd
Proceedings of the American Philological Association 112, 125-41.
Pcristiany, J.G. (ed.) (1965) Honour and Shame. The Values of Medite"anean
Society. London.
(1968) Contributions to Medite"anean Sociology. Paris and The Hague.
(1976) Medite"anean Family Structures. Cambridge.
Perlman, S. (1958) 'A note on the political implications of proxenia in the fourth
century B.C.', Classical Quarterly. n.s. 8, 185-9J.
(1976) 'On bribing Athenian ambassadors', Greek, Roman, and Byzantine Stu-
dies 17, 223-33.
Pfohl, G. (1967) Greek Poems on Stone, 1: Epitaphs {rom the ~venth to the Fifth
unturies B.C. Leiden.
Piganiol, A. (1959) 'Fides et mains de bronze', in Melanges H. Levy-Bruhl: Droit de
I'antiqu;tt et sociologie ;uridique (Paris) 471-3.
Pitt-Rivers, J. (1965) 'Honour and social status" in Peristiany (ed.) (1965) 21-77.
(1968) 'Pseudo-kinship', in International Encyclopedia of Social Sciences VIII,
403-0.
(1971) The People of the Sie"a, 2nd edn. Chicago.
(1973) 'The kith and the kin" in Goody (ed.) (1973) 89-105.
(1976) 'Ritual kinship in rhe Mediterranean: Spain and the Balkans', in Peristiany
(ed.) (1976) 317-34.
(1977) The Fate of Shechem, or the Politics of Sex. Cambridge.
Polanyi, K. (1977) The Livelihood of Man. New York, San Francisco, London.
IEdited by H.W. Pearson.1
Poralla, P. (1913) Prosopographie d~ Lakedaimonier. Breslau.
Bibliography 193
Pouilloux, J. and F. Salviat (198.l) "Lichas, l..acedemonien, archonte a Thasos et Ie

livre VIII de Thucydide" Comptes Rendus de I'Academie des Inscriptions et
Belles-Lettres, 376-403.
Pritchen, W.K. (1971-79) The Greek State at War, 3 vols. Berkeley, Los Angeles,
London.
Pusey, N.M. (1940) "Alcibiades and to philopoli", Harvard Studies in Classical
Philology 51, 215-31.
Quiller, 8. (1981) "The dynamics of the Homeric society', Symbolae Osloenus 56,
109-55.
Radcliffe-Brown, A.R. (1952) Structure and Function in Primitive Society. London.
Redfield, R. (1953) The Primitive World and Its Transformation. Ithaca, N.Y.
Reina, R.E. (1959) 'Two patterns of friendship in a Guatemalan community',
American Anthropologist 61, 44-50.
Rhodes, P.J. (1985) What Alcibiades Did or What Happened to Him. Durham.
[lnaugurallecrure.]
Rostovtzeff, M.I. (1922) A Large Estate in Egypt in the Third Century R.C. New
York.
(1941) The Social and Economic History of the Hellenistic World, .3 vols.
Oxford.
Roy, J. (1967) "The mercenaries of Cyrus', Historia 16, 287-323.
Runciman, W.G. (1982) "Origins of states. the case of archaic Greece', Comparatil'e
Studies in Society and History 24, 351-77.
Sahlins, M.D. (1968) Tribesmen. Englewood Cliffs, N.J.
(1974) Stone Age Economics. London.
Saller, R.P. (1982) Personal Patronage under the Early Empire. Cambridge.
Schmidt, S.W. et al. (eds.) (1 977) Friends, Followers, and Factions: A Reader In
Political Clientelism. Berkeley.
Schuller, W. (ed.) (1982) Korruption im Altertum. Munich.
Scon, J.c. (1972) 'The erosion of the patron-client bonds and social change in rural
southeast Asia', Journal of Asian Studies 32, 5-37.
Seager, R. (1967) 'Alc.:ibiades and the charge of airning at tyranny', Historia 16,
6-18.
Seiben, J. (1967) Historische Beitrcige zu den dynaslischen Verbindungen in he/le"is-
tischen Zeit. Wicsbaden. ! H istoria, Einzelschriften, 101
Shaw, B.D. (1984) "Banditry in the Roman Empire', Past and Present 105, 3-52.
Shear, T.L. (1978) KallicJs of Sphnlos and the Ret'olt of Athens in 286 B.C.
Princeton, New Jerse)'. IHesperia: Supplement XVII.]
Sherwin-White, S.M. (1978) "Hand-tokens and Achaernenid practice', Ira" 16. 183.
Sinl, C. (1890) Die Gebiirden der Griechen lind Romer. Leipzig.
Smelser, N.J. (1958·59) "A comparative view of exchange systems', I:conomic
Development and Cultural Change 7, 173-82.
Snodgrass, A.M. (1974) "An historical Homeric society?'. Journal of Hellenic St"dies
94, 114-25.
(1980) Archaic Greece. Berkeley.
Sragakis, G. (1975) Studies in Homeric Society. Wiesbaden. ( HlStor;a, l-:inulschrif-
ten. 26.)
194 Bibliography
Stahl in. G. (1954) ';fvo~. ~fvl~ ..... in G. Friedrich (ed.l. Theologisches Wort~buch
zum Neuen Testament V (Stuttgart, 1954) 1-36. [English translation by G.W.
Bromiley (1967)]
Ste. Croix. G.E.M. de (1972) The Origins of the Peloponnesian War. London.
(1974) 'Ancient Greek and Roman maritime loans'. in H. Edey and B.S. Yamey
(eds.), Debits, Credits, Finance and Profits. (A collection of papers presented
to William Baxter.) (London) 41-59.
(1981) The Class Struggle in the Ancient Greelr World. London.
Stirling. P. (t 968) 'Impartiality and personal morality', in Peristiany (ed.) (1968)
49-64.
Taillardat, J. (1982) 'CZ»IAOTH1:, nl1:TI1: et Foedus'. Revue des ttudes grecques
95. 1-14.
Tegnaeus, H. (1952) Blood-Broth~s. An Ethno-Sociological Study of the Institution
of Blood-Broth~hoodwith Special Reference to Africa. Stockholm.
Thome5. F.e. (t 955) 11 problema degli eteri nella monarchia di Alessandro Magno.
Turin.
Thompson, H.A. (1953) 'Athens and the Hellenistic prince5', Proceedings of the
American Philosophical Society 97. 254-61.
Tod, M.N. (193.1-48) A Selection of Greelr Historical Inscriptions, 2 vols. Oxford.
Veblen, Th. (1970) The Theory of the Leisure Class. first published 1899. London.
Vernant. J.-P. (1980) Myth and Society in Ancient Greece. translated by J. 1.I0yd.
London.
Veyne. P. (1976) Le pain et Ie cirque. Sociologie historique d'un pluralisme poIi·
tique. Pa riSe
Walbank. F.W. (1940) Philip V of Macedon. Cambridge.
(1951) 'The problem of Greek nationality" Phoenix 5. 41-60.
(1957-82) A Historical Commentary on Polybius. 3 vols. Oxford.
(1984a) 'Monarchies and monarchic ideas'. in F. W. Walbank (ed.). The Cam-
bridge Ancient History. 2nd edn .• VII.I (Cambridge) 62-100.
(1984b) 'Antigonus and Macedonia'. in F. W. Walbank (ed.). The Cambridge
Ancient History, 2nd edn .• VII. I (Cambridge) 221-56.
Walbank. M.B. (1978) Athenian Proxenies of the Fifth Century B.C. Toronto and
Sarasota.
Wallace. M.B. (1970) 'Early Greek proxenoi'. Phoenix 24. 189-208.
Weber. M. (1947) The Theory of Social and Economic Organization. edited by T.
Parsons. New York.
Weinfeld. M. (1970) 'The covenant of grant in the Old Testament and in the ancient
Near East'. Journal of the Am~ican Oriental Society 90. 184-203.
(1973) 'Covenant terminology in the ancient Near East and its influence on the
West'. Journal of the American Oriental Society 93. 190-9.
Welle5, e.B. (1934) Royal Co"espondence in the Hellenistic Period. New Haven.
Whitehead. D. (1977) The Ideology of the Athenian Metic. Cambridge.
(1982/3) 'Sparta and the thirty tyrants'. Ancient Society 13/14. 105-30.
Wiesehofer. J. (1980) 'Die "Freunde" und "Wohltater" des Grosskonigs', Studia
Iranica 9. 7-21.
Wilhelm, A. (1942) 'Proxenie und Euergesie', in Attische Urkunden. V (Vienna)
11-86.
Bibliography 195
Wilson. B.R. (ed.) (1970) Rationality. Oxford.

Wipszycska. E. (1961) 'The bwQ£o of Apollonios the Dioiketes in the Memphite
Nome'. Klio 39. 153-90.
Wolf, E. (1969) 'Kinship, friendship and patron-c1i~m relations" in M. Damon (ed.),
The Social Anthropology of Complex Societies (london) 1-22. [Reprinted in
Schmidt er al. (eels.) (J9n) 167-78.]
Wolski, J. (1973) 'MH~I1:M01: er son imponancc en Grece a I'epoque des guerres
mediques" Histona 22, 3-15.
Worrle, M. (1978) 'Epigraphische Forschungen zur Geschichte Lykiens', Chiron 8,
201-96.
INDEX OF SOURCES
Nott: The sourcn listed in tM Apptndica are not included.
UTERARY TEXTS - GREEK
AESCHINES Magna Moralia

L Against 'r;",arcbus 2....ll..J la46 19n. ,12, .1On. ~ 33n. Z2
~ 94n.~ Politics
2.. On thl' F.mbassy 2, 1161b6 l22n. U
2Il 2.1n.~ l, 12K7b 30 .17n.81
Z1 12n.6 I Prohlt",s I
Hl 127n.,ln 29 2 950'128 94n.68
154 47n.22 Rhttor;c
ill 140n.62. 21381h19 19n.31
.t. Against Ctt.iphon ARRIAN
~ 3Kn.9.1 AnalMsis
~ 154n. lUl. U.i 122n.1.5
ill Uln.~ L11JI till (fabk:!l
224 4n. .h 54n. ~ 66n. ~ 68n. rt l24n. ATHENAEUS
21 4167b 44n.2
AESCHYLUS 4 167" 108n. ill
F.umi'nidts
269.75 125n.28 CALLIMACHUS
ANTHOLOGIA PALATINA F.pigram 2a 26n.55
7,98 ,h.S
ANDOCIDES DEMADES
1. On fflt MysltritS ()n tht Tlw/llt YtaN
~ l06n. un 21 76n.H
145 .150. U. DEMOSTHENES
2.. On his RtfU", 2.. .~cond Olynthiac
11 81n..10 1.2 154n. U!.1
2J} H2n.ll 1. Un thi' Ptau
APOLLODORUS 8 97n.~
LY 260. 52• .u (caption) 17. ()n Hd/ontSJus!
262.3 l24n.24 II 100 (Tabk J)
1,13.3·6 oM (caption) H.. ()n tht Chnsontst
ARISTOPHANES ~ 88n.48
Birds 1...L for th(" l.ibtrty of tilt Rlmdians
1021 139n.~ IS 140n.59
frogs 1.ll On thl' C,mlln
147ff. 124n.2J i.l lU1{Tablc.l). 154n. 102
ARISTOTLE ~ IOn. I. lHo. 93
T~ Athmian Constitution Kl 140n.62
.u 76n. 1..! l09fl. 3n. 4, 1600. 120
~ 76n. U. 9On. ~ 15.1n. ~ 280 16Un. ill
!ll 21n.39 284 30.4,.15n, 79. 160n. l l j
2ti USn. 50 1..i. nt fal.., ugation(" -
i5...S. 76n. ~ ill H2n.3.l
0(" Gmnationt A"imalium 139 89n,SJ
722hlO 62. (caprion). 145 82n. 33, 154n. l.!l2
ft";(a Nicomachta 189.90 4n.5
IIHhS 37n.87 1.2.2.=5 26n..~ 1
116()J4 79n.ll l..tY 340. 74
1161h12·17 19 0.31 ill 121n.9
196
Index 197
19h I 54n. 1ill. 11 44 2 47n. l1i
2H Un. 79 l..Ub 47n. 1Jt
230 127n.37,152n.94 11 71 2 IOhn.ill
23.l 152n.94 1A...Z...4 I06n. ill
248 3n.4 14. Il.~ 2.1n.~
HH 80n.26 14 19 S 2..Jn. N
259·66 73n. I,X2n. .lJ lh..iJ KHn.4H
268 80n.1.!l ~ 7)n. 1
273 76n.11 l.6.-S...S 2hn.51,X'Jn.ii
29.~ 35n. 79. 160n. 120 17 18 1·2 USn. 50
305·6 81".2,2 11..Z.6 K4n.40
309 81n.29 1975 1 107n. u:.
314 35n. 79. 110 {Table .n. IHn. lU2. 2070 3.4 IHn.2(,
343 -- - H8n.48 DlOGENES LAF.RTlUS
2!1 Against uptints liJ 49n.l.::..
il 151 n. .&2 L25 7Hn. e!
21. Against Mtidias 249-50 47n.!!!
110 140n.61 UJ Hln.29
2llli 14On. hQ U1i 5n.~
21. Against A,istoaatts .4-H i8n. ~
HU 110 (Table 11 &Jrr 65n.hX
154 ill (1'~ble :!l DlONYSIUS OF UALICARNASSUS
1lI ill (T ~ble 11 lsatus
21.. Against Aphobos I 6. 2Hn.62.
.i 3In.6H Isocrates
~ 94n.62. Ui 16n. .lli
l i Against lAcritlls
R·9 69n.H!l EURIPIl>ES
ill 61. (caprinn) Andromache
~ Against Timotheus 1103 1.J9n. 5H
2h fUn.H Hutra
.ill.. Against Po/)'t:lts 8l 50n. 11
La 22n.4I,65n.68,6Sn.7I,93n.61 H3 22n. ±L 1..i8. (cJptlonl
21 12n.6 1241 3"n.8.6
1.6 22n.41 • .15n. 78.39n.9i,65n. 71.92n. 1340 36n.8h,
!rl. 96n. 8l Huuba
12. Against Callippils 710ff. .l4n. 7S, l14n.25
J 27n.i8. Hdma
! 47n.21 291 63n.6.2
2.1 27n.iH. Ion
153. Against Nicostratusl 551 09n.5X
! 29n.!cl 1039 09n. iH.
ill 93n.6b Mtdta
i6.. Against Oionysodorus 613· 14 21n. ~ Hn. ~ 610. !l1.(SdW!iOll),
Z 1$ 1n. 82 65n. z.L 7On. lti (schu/",n)
159. Against Nt"tra) Oust~s
~ 15In.~ 1015 22n. 44. ill (captinn)
OINARCHUS Rhtsus
L Against Dnnosthnttts 335 9~n.W
4!l 76n.8
4.l 82n. .12 FGltH
2.. Against Aristogiton 72F6 (Agalhoklts of Cyzicus) 108n. ill
! 76n.8 76F4 (Duris of Samos) 108n. ill
Z 80n.21 9OF59 (Ni,ol:lu~ of Damascu\) 10n. '4
2.1 76n.!! 115 F209 (Theopompos of Chi~) .19n. l.!lli
DlOOORUS SICULUS 115F224 (Thropompos of Chlo\) 44n. -.
4,·H,~·8 124n.24 15 4 n. l.!l.J
8.Z 96n.ZH 115f2258 (Theopompo§ of Chios) 106n. 1 ~
198 Index
392F6 (H~rmnilaosof Chios) 104n.~ 6Jlbff. 96n.Z&
472f6 (Ag~thoklnof Cyzicus) 1~ (Table}) 6...ll.S 2On. Hs 89n. 10
688F9 (Kte5i~s of Cnidus) 51n. .H U 66n.11
688F 15 (Ktesias of Cnidus) 108n. 121, 110 1...1.l.:!1 38n.89,45n. 13.84n.~9Qn.ii
(T~ble 11 Lll:1 - - 59n.ll
2.l8..:..2 45n.U
HELLENICA OXYRHYNCHIA 1..1.lb 135n.50
1U 151n.89 7228,4 26n.~34n.~
ill 151n.89 8.i 79n.21
HERODOTUS ILli J.1n. U
1.20·2 122n. !b U5n. 50 8.Jli 39n. 99, 107n. 119,llQ (Table .1)
1.34·45 127n.35 8...I.lD. 135n.50
1.61 76n. lit 90n. 60 R.l.l6 UQ (Table 11
1.64 127n. .lZ 2..l 157n.1illl
1.67 l.15n.50 ti 157n.109
1.69 4Hn.2J,9Hn.~ 135n.1O 2.16.J 22n.42
1.74 J6n.Hb 2..22 22n. 42, 122n. ~
1.74.6 54n.~ UflZ 122n. ~
1.77 98n.~ 2....1ili 98n.90
2,114-15 126n..\3 HF.SIOO
II 54n. ..l2 Works a"d Days
.Lll 90n. .H 263-4 79n.M
].21 46n.~ 320-4 124n.21
.Lll. 46n. 14.60n.56 HOMER
III 17n. 19, 71n. 89 Iliad
.Lll 21 n. 39, 135n. .ro 2.623 21n.40
ll1l .18n.92, l.15n. 50 3 3S 1-4 125n.28
~ 9On.58 .l...ll.4 18n.29
ill 90n.!! U66 38n. ~ 68n. Zj
.l..1lS IS In. 90 6,119·236 58n.48
.hIll 14n. 13.41n.l Ufl8 63n.61
.1...Hfi 70n.86,89n.51 UU 36n.8.6,
.un 79n.24 6219·20 6On.11
un 54n. ~ 15. (caption) 6224ff In.l,59n.49
UJ.B 102n. JJl2 6..lli 49n.2.2-
!...1..li 107n. ill 6234·5 61n.~
~ 123n. U 7,287ft. 60n.i6.
~ 34n. ~ 38n. ~ 68n. ~ Zln. ~ 8..J..lO 60n.i6.
4.172.2 54n.39 9 149·56 109n. ill
ill 109 (Table:ll 9,291·8 100n.ill
i...U 107n.ill 104U 98n.n
.t.2..1 ill (Table 3) 10495 98n.S!
.t.2..Y 4On. 102, l04n. 106, 109 (Table:ll 11623·5 125n.27,125n.28
.w..b1 119n.6- l.l.Zill. 21n.38
.Lll 4On. 102, 122n. Ln 13824 6On.56
Ll1J. 109 (Tab~ 11 IS,525ft. 61n.61
.uu. 21n.39 16,140ff. M (caption)
i..6.3 135n.50 17 140{f 124n.20
.un 135n.1O OdysMy
.uJ I.lIn.50 I. 115H 58n.48
~ 91n. 62. llQ (Table 3) 1.187·8 59n.50
.uM -109 (T~b~}) 1.307·8 18n.25
~ USn. 50 1 309· 1 1 77n. 16.60n.58
M.1 109 (Table 3) 4,129·)2 79n.21.84n.~
6.1U 138n. 55 4613·1 K 63n.64
6.ZO 4.Jn.1.. 109 (T~ble}) R..54S·6 .lln.71.
fW1. 76n.11 2...l.8. 135n.50
6JI2 76n. U 9266·7 125n.28
Index 199
10 336 18n.29 utter 6 (To the Childrm of jason) 26n. .so.
H.16 125n.28 34n. Lt 151n. 82
14.191 H. 35n.80 utt" Z (To Timotlmls) 26n. SO. 47n. ~
J4279 17n. 22, .56n. 40 65n. ~ 70n. ~ 121n.1O
J4185·6 79n.21 utter H. (To the Ruins of My/ilmt) 26n. ill
14.321·6 96n.1!J.
J4 324-5 79n.21 JOSEPHUS
14,414 l25n.28 jtwish Antiquit~s
15 JII·19 63n.6! 12.J48 18n. .2..8.
17512 38n.88 13 126 18n.28
21.11·42 5Rn.4R 14 122 23n.46,35n.Hj
21 14·23 41n.2 l'he j~ish War
21 26·8 124n.24 LJJU 23n.46
2.Lll 60n.ll
2..1..M! 61n.60 KOCK, J., ('..<>micorum Alticorum fragmmta
2..l..n 22n.tl vol. III, no..H3 (Phomicidas) 109n. 126
24 114 38n. 91, 59". 51
24.1 Hff. 101n.99 LYSIAS
HYPER IDES !L. Against AndO€ides
1. Again$t Dnnosthenn !8 35n.21
~ 88n.48 I I Against Ago,atus
II 71n.R9 ~ 151".82-
II 75n. 7 a Against Alcibiades
!L. funeral Speech 1A.26 100n. ill
10 76n.8 1..8.. On the Property of Nicias' Brother
10 23n.48.15On.89
ISAEUS 1.2. On the Praperty of Aristophanes
11... On t~ F.staU of Hagnias 18 97n.84
4l.:.l 94n.~ 2.5 65n. ~ 6I (caption)
ISOCRATES .lb 94n. Zl
L To Dnnonic," !O. 94n. Zl
21 121 n. II !J 96n.&3
LNicodes 26n.ifi
4.. Panegyricus MENANDER
U 45n. ~ 68n. ~ 7On. a.s. Monostichoi
1. To Philip 615 16n.18
1.2 154n. Ull
ill 124n.U PAUSANIAS
~ Evagoras ~ 140n,12
11 105n. ill .... 4f( 96n.ZH
l..S.... Antidosis !.ti 68n. 79
~ 84n.40 1JL6 21n.39
l6.. Concernin, th~ Team of Horus 6....12..J 36n.M
!l 16On.lli 7 10 1-3 156n.1Jl6
11.. l'rapevticus uu. 35n.H.l
1 29n.63. 106n.~ PINDAR
~ 47n. 21, R:ln, .37 Nnnean
6 95n.25. 1.2{hl 129n.43
U 21n. & 95n. ZZ Pythian
.l! 93n.6I .ll1 18n.26
C!.. Aeginnic," 2Jl2 135n..m
i 17n.21,34n.74,68n.79,96n.81 10 157n. lD.H
~ 35n.82 liL6.6 17n.20
lH 27n. ~ ~ Zb 97n. M PLATO
U 96n.8.3 Crito
11 94n.7.1. lli 27n. 57, 65n. 71, 87n. 46
II 105n.111 Laws
utter 2.8,13 (To Philip, I) 151 n. 89 729eff. 12.5n. .2..8.
200 Index
950b 125n.28 OMipu. al CoIonu.
9He 125n.28 BOO 98n.&i
9SSb·c 16On.lli PhiiOlltltS
Leun-Z 49n.28 81.1 54n.16.
Symposium Trachi"iat
lliiJ 6..2 (c~prion) 6&Zl 124n.24
PLUTARCH 248·54 124n. .M
A/nand" 274-6 124n.24
6 84n.!£} STOBAEUS
1.0..1.1 122n.15 .1.9 p.3S8 16n. 1.8.
Aralus !.l p.W (Charondas) 125n.28
1..1 22n.!1 STRABO
Cimon Ll..5. UO(Tabk 1)
I.h. 21n.39 TElES
EumtntJ 21 44n.8
1.0 112n. ill THEOGNIS
Mora/ia 143-4 125n.28
~ 12Sn.18 THEOPHRASTOS
1llf 71n.&i Characlns
29SB·C 57n.~ 2M 83n.16
Ptlopidas 2.U 97n.86
ill 21n.38 THUCYOIDES
Pnk/t' Ll.l 2On.ll
n 21n.39 LLl 151n.89
Philopomrnr U!U 21n..19
1 22n.45.35n.8.1 u.u 21n.n
Solon LH£l:j 143n.~
U 68n.~ Lll6 15On.8.Z
Thmri"Olle. 1 118·}O 42n. 3. 47n. ~ i l l (T:able 11
2..6....1 47n.1.8 I J 29 3 8CJn.ll
ThtUUS Lll.6. 13Sn. SO
3Off. 98n.&i Lll8 94n. Z!.. 109 (Table:!h 122n. !!.
POLYAENUS ISem. i l l
~ 1..1.0 (Table 11 2...1J 3n.3. 144. 145
POLYBIUS 118 143n.73
~ 16On.lli 2....l.Z...I 12n.6
.. 48 12 151n.89 2..l..L..I 76n. II. 1S8n. ill
5...3.1.2. 28n.6..2 2....2!l 138n. .55
.t..lH 123n. 1.8 2...6.ll 78n. 19.1S8n. i l l
SAlt12 151n.89 2.65.8·9 78". I..i
U1...1 l5In.89 2.!12.l. 78n.2.0
1M. 103n. 103. 103n. 11M UZJ 9On.S1
.t.z!1.j l04n.ill6 .l.l.3 139n.58
1.1.! 26n.~ l.ZA l04n. Ul8
il 16n. 17. 460. ~ .l...S.l..1 21n.1i
Ulll 22n.45. 35n.81 .LZ!U 139n.58
116 ISln. &i ~ 119n.7
18.11·14 16In.1ll 4.114,) 161n. III
2{LH 15n. 15,43n.4,70n.87, 122n.~ 4 1192 21n.39
21 43 18 112n.lll 4 132,2 I 1em. 7
M....Ll 47n.22 ~ l04n. 1.01
lUI S2n.ll 5 1924 21n.39
31 2S I 18n.27 .ul.l ISln.~
B 182 47n. ~ 7On. 85 i..4.l 71n.89, 138n. 56
5...fi..l ~
SOPHOCLES 5 45 ).4 147n.8O
A;ax 5 594.60 J 145n.16
817ff. 6On.5.6 5..6..3.A. 1S8n. l..H
Index 201
6.53.1 l50n. 88, 151n. 89 7.2.25 18n.24
6.54.6 15ln.91 7.3.2 106n. 116
6.55.1 21n.39 7.3.20 77n.17
6.61 148, 150n.88, 153n.96 7.6.17 76n. 10
6.92.4 161n.122 7.8.8 110 (Table 3), 114n. 137
7.33.4 7On.85, 105n. 110 7.8.17 110 (Table 3), 114n. 137
8.6.3 19n.33, 148n.82, 148n.83 Cyropudia
8.6.1 29n.6.1 1.4.26 18n.28
8.6.8 21n.39 3.3.4 135n.50
8.12.2 49n.84 4.6.2 57n.42
8.17.2 49n.84 5.1.3 38n.90
8.65.1 151n.89 6.2.1 46n.14,47n.20,57n.42,89n.49
TIMOKREON 8.2.7 67 (caption), 78n. 20
Fragment 1 123n.19 8.6.5 106n.113
8.7.13 49n.28
XENOPHON 8.17 114n.137
A~silaus Htllnrica
3.4 54n.37 1.3.12 1S3n.95
4.4 10n.l,49n.26 1.4.2 21n.39
8.3 159n. 118 104.8 153n.95
A"abasis 104.13-4 152n.92
1.1.2 lOOn. 92 1.4.19 152n.92
1.1.9 92n.6.1 1.5.17 106n.1I6
1.2.1 lOOn. 92 1.7.22 H9n.116
1.2.3 68n.79, lOOn. 92, lOOn. 96 2.1.3 142n.71
1.2.3-5 12On.8 2.1.25 l06n. 116
1.2.12 47n. 16 204.28 21n.39
1.2.23 45n. 12 3.1.1f1. 2In.39,34n.75
1.2.26-7 390. 101,47n. 16, 66n. 73,9On.55 3.1.5 104n. 109, 110 (T~ble 3)
1.3.2 18n.23 3.1.6 108n. 120, 109 (T3ble 3), 114n. 138
1.3.3 92n.63 3.1.10 151n.89
1.3.12 126n.31,126n.32 3. I. 10fl. 107n.117
1.4.3 lOOn. 9.. 3.... 11 125n.30
104.8 lOOn. 92 3.5.4 15ln.89
1.5.8 ISln.89 4.1.29ft. 45n. 11,46n. IS
1.6.7 54n.36 ".1.34-5 In.2
1.7.7 106n.1I3 4.1.39 15n.15,58n.47,60n.57
1.9.3 127n.35 4.1040 28n.61,152n.94
1.9.12 9On.56 5.2.3 153n.97
2.1.5 47n.17 5.2.35 5n.9
204.1 54n.37 5.2.36 15 In. 89
2.5.27 125n.29 5.3.13 38n.90, 15In.89, 152n.93
2.604-5 92n.63 6.3.4 USn. 51
].1.4 5n.8,100n.96 6..~A 135n.50
3.1.8 47n.20 6.5.6-8 H In. 89
3.2.4 12Jn.29 Hitro
4.8.7 500.30 3.7 19n.30
5.3.5 26n.52 Mmrorahi/ia
6.1.23 47n.20 2.9.3-8 87n.45
7.1.8 68n. 79
202 Index
INSCRIPTIONS - GREEK
A"",",' of Iht Brilish School at Athms no. 138 18n.28
51 (1962) 144 IOn. 34 OSBORNE (1981-83)
nos. 1-6 141n.63
Bull. £pi"
(1978) 145 127n. 37 PFOHL (1967)
(1984) J 14 21n. 38 no. 10 26n.54
Chiron Sardis 7 (1)

8 (1918) Worrle (1978)
;;II 112n.129 no. I 110 (Table 3)
CompttS RtfUlus tk L'Audm.;t da SEC
IJlScriptiorrs 1.366 112n. 128
J.Bousquet(1975) 139 101n.116 26.86 127n.37
CRAMPA,J. (1969-72), lAbrmmda 26.1623 37 (caption)
no. 43 11n.18 27.135 2On.36
28.60 =Shear (1987) 85 (Tabk 1)
Haperia SHEAR (1978)
19 (1950) 25-6 26n.53 p.6. lines 80-2 86n.43
19(1950)390 158 (caption) syII'
20 (1951) and pl.2Sc 62 (caption) 195 2On.34, 127n.37
302 110 (Table 3)
IG 307 85 (Table l)
11 400 21n.39 332 110 (Table 3), 1I3n. 136
11 916 62 (caption) 360 76n.9
IJ 19 1350.52 3~ 71n.89
I' 23 21n.39 367 85 (Table I)
I) 27 135n.52 374 83n. 38,85 (Table I), 86n. 43, 97n. 86
135n.52 503 135n.51
" 65
" 91 137 (caprion)
" ltO 132n.47, 123 (caprion) TOO (1933)
" 127 134 (caprion) no. 34 4n.6,14On.59
"3 lS6 135n.52
1 162 135n.52 M.B. WALBANK (1978)
.' 1~
1-' 1154
135n.52
136 (caprion)
no. 9
no. 64
136 (caption)
137 (caption)
11 1 109 151n.89 no. 85 151n.89
11 1 111 135.52 no. 86 142n.68
111 223 nne 18 no. 87 133 (c_prion)
111 226 lSln.89 nos. II, 3~ 74 141n.66
11 1 240 88n.48 nos. 1,2,5, 18,30,4',60,66,70,90 141n.
11 1 401 85 (fable I) 65
111 561 127n.37 nos. 10, 13, 20, 21, 23, 26, 28, 31, 33-7,
111 655 85 (fable 1) 40-~ 44, 51, 53-9,62, 69, 71, 72, 76, 77,
XII.8.156 85 (fable I) 79,83,84,89,92-4 141n.67
XIV.279 16n.17,64(caprion) nos. 1,2,5,18,30,47,60,66,70,90 142n.
XIV.2432 53 (caption) 69
nos. 1-94 141n.63
MEiGGS AND LEWIS (1969) nos. J-4, 6-17, 19-29,31-44,45,46,48-59,
no. 4 132n.48 61-5,67-9,71-89,91-4 141n. ~
no. 7 2Op.34
no. 91 82n.30 WEtES (1934)
00.93 107n.116 no. II 108n. 124
pp.40-2 157n.l10,158(caprion) nos. 10-13 110 (Table 3), 113n. 135
no. 3, para. 5 139n.58
OGIS
no.l I06n.115 u;tschrift fi, Papyro/ogit u"d Epigraph;'
no.4 85 (Table I) 33 (1979) = Fischer (1919) III (Table 3),
no. 55 111 (Table 3), 1I2n. 130 114n. 142
Index 203
LEXICA
HESYCHIUS ~~ 12n.7
~£ivoe; 12n.7
~fvot 12n.7 SUIDAS
~mov 6On.55
OTHER ANCIENT SOURCES

CURTIUS MACCABEES
6.5.1·5 7On. 85 1.11.31 180.28
DOSSIN. G. (1952) Co"~spo,,dmu th NEPOS

ItUmah·Addu Dift4mn J0.1 5"n.37
no. 20 790.21
OLD TESTAMENT
GRAYSON. A.K. (1972) Assyria" Royal 1. S.mud 18.1 S8n... 8
'''scriptio"s 1. Samuel 24, 26 "8n.2S
vol. 1,48-9 890. 49 1. Samuel 27, 5-6 108n. 122
2. Samuel 1.26 58n.48
KENT. R.G. (1953) Old Persia"
p. 115, XVs .nd AVs 66n. 74, 67 (caption) TACITUS
Historin 1.5" 540.37
LlVY
1.1.1 122n. IS
INDEX OF PROPER NAMES

Note: The names lisred in rhe AppendiCft are nor included.
Abradaras, 38n. 90 Alexander (me Great), IOn. I, 43, 58n. "8,
Achilles, 22, 25 (caption), 1090. 127 75, 83, 84n. 40, 85, 86, 106, 108, 1 B,
Ach isch, 108n. J22 J22, IS", J55
Adkins, 18 Alexander III, 110
Adr.sros, 110, 127 Alexander IV, 127n. 37
Aenras, 122n. IS Alexander (of Epirus), 110
Aeschines, 3,4, 100.1,23,35,54,66,81,82, Alexandras (of Troy), 125, 126
110, 124, 127, 140n. 62, 154n. 102, 160 Alkibiades (of Athens), 8, 19,20,36, ...., 71n.
Agamemnon, 22,37,38,59, 101 90, 116-18, 128, B8n. 56, 142, 143,
Agarhok'es (~cipienr of tlo,~a), 111 147.51, 153, 155, 159-61
Agathokles (Sicillian tyrant), 12", 125n. 26 Alkibiacks (of Spana), 19, 20n. 36, 148
Agrnor, 26n. 50 Alkimachos, 1S 6
Agnilaos, I, 2, JOn. I, 1", 15, 28, 36, 38, Alkiphom, 1"3, 145, 146
45-7, 49, 51, 58-60, 65n. 68, 66, 125, Alkmeon,89
129, 135n. SO, 152, 159 Afyanes, 121, 135n. SO
Agias, 99 Amasis, 17, 46n. I", 60, 71, 98, 102
Agis, 143, 145, '46, 149, 156, 158 Ammonios, J03
Aiakes, 102 Amorges,51
Aias (son 01 o;leus), 21 Amphimedon, 38, 59, 101
Aias (!K)n of Tdamon), 21, 6On. S6 Amphitryon, 135n. 50
Akanthos, 21n. 39 Amyntas, 23, 82, 156
Akastos (Athenian archon), 76 Amyntas ('of Asia·), 110
Akasros (Peleus' friend), 2.. (caprion) Anaxinos, 66
Aleuades, 156 Andoci~,35,82,88
204 Index
Androkleidcs. 5 Arknila05 (example of name), 21n. 38
Androkles, 62 (caption) Arkesilaos (of Sparta), 20
Andromachos, 103 ArkesilaOl (of Thasos), 20
Antalkidas, 44 Artabazos, 58n. 48
Antmor, 122n. IS Artaphrmes, 122n. 16
Antigonos I, 107n. 117. 110 ArtIS, 105
Anrigonos Doson, IS, 42, 49, 70, 122n. 14 Artaxerxes I, 100n. 113
Anriochos I (of Syria), 108, 110, 112 Artaxerxes 11,45,66,90, 100, 106, 119, 1S4
Antiochos III, 103, 110, Ill, 112 Artemis, 114n. 141, 137 (caprion)
.Anriochos IV, 52, 53 Arthmios, 8On. 28
Anriochos Hierax, 23 Asandros. J07n. 117
Anriochos I (of Commagene), 35-7 (caption) Ashur-uballilt 89n. 49
Anripatn (Alexander's successor), 78n. 19, Asopi05, 104
83,85 Athena. 18, 134 (aprion). 137 (caprion)
Anriparn Uewish prince). 22. JSn. 81 Athenaeus., t 08
Aphareus, 26n. SO Athenai05 (of Sparta), 21 n. 39
Apollo-Hellos, 37 (caprion) At~aiOl (of Thespi$), 21n. 39, 1.. ln. 66
Apoliodoros (Athmian writer), 24, 52 Atresridas. 81
(caprion) Attaginos, 156
Apoliodoros (of Leonris), 12n. 6 Audoleon, 85
Apollodoros (01 Oineis), 12n. 6 Autokrator, 47n. 22, 65
Apollodoros (son of Pasion), 22, 39, 65, 92,
93,96 Bellerophon,36n. 86,60, 63n. 65
ApoUodoros (of Sttymbria), 142 Buve,155
ApoUonios, 111 8oiotios, 21n. 39
ApoIlophanes (of Cyzicus), 46 Boulagoras, 112n. 128
Apollophanes (of Pydna), 26, 34n. 74. 121 Brachylles, IS, 42, 43, 70
AratuSt 22n. 45 Brasidas, 104, 119, 120, 161n. 122
Archebiades, 27, 47n. 21 Bum, 117, 145
Archedemos, 87
Archelaos, 78n. 19, 82 Camby~ 46, 9011.55
Archias (of Athens), 21 Cassander, 110, 113
Archias (of Thebes), 21 Chalkideus, 21n. 39
Archils (xmas of Sarmans), 135n. SO Chares, 12n. 6
Archidarnos ((ifth-cmtury Spartan king), 3, Charlemagne, 46n. 14
38,45n. 10, 71, 143, 144, 145, 152, 158, Cheirisop~ 99, 100
159, 166 Chiron, 24, 2S (caprion), 26, 27 (caption)
Archidamos (third-century Splrtan king), Chrysermos, 111
28n. 62, 123 Crito, 12, 27, 35, 87
Ariaios,47 Croesus (lydian ruler), 19, 20n. 34, 48n. 23,
Ariobarzanes, 110 89, 98, 127, 135n. 50
Amtagora. (tyrant of Cyme), 102 Croesus (an Alkrneonid), 19
Aristagora. (tyrant of Cyzicus), 102 Cyrus II, 46, SO, 88,98. l06n. 113, 108, 109,
Arisragoras (of Milerus), 118, 119, 120, 135n.50
122n. 16 Cyrus (the younger), 4, Sn. 8, 14, 18, 28, 36.
Aristeidcs, 157, 158 (caprion) 39, 45, 47, 54, 57, 88, 91, 92, 98-101,
Arisrippo5y 45, 99, 100, 101 104-5, 106, 119-21, 126, 129, 1S4
Aristobulus, 23
Aristodikides, 108, 110, 113 Daphnis, 102
Aristollos. III Darius I, 14. IS, 40, 41, 43. 49, 70, 84n. 40,
Aristomma, 110 89, 91. 102, 103, 104, 107, 109, 114,
Ariston, 102 122n. 14, USn. SO. 156
Arislonous. 47n. 18 Dlrius II, 110
Aristophanes (of Olynthus), 47n. 22 Darius III. 110
Aristophanes (poet), 62 (caprion) Oatis, 1S7, 1S8 (caption)
Aristophanes (son of Nikophemos), 65. 94n. David, 48n. 25, 58n. 48, 61n. 61, 108n. 122
71, 96, 97 Davies, 97
Aristotle, 19, 30. 62 (caprion), 76, 90, 93 Demades, 75, 88n. 48
Index 205
Oem.rams (of Corinth), 84, 154n. 102 Gob'las,57
Dema,atos (01 Sparta), 43, 91, 107, 109, 114, Gommc,145
116 Gongylos, 47, 108, ltO, 113, 114
Demetrios Poliorketes, 87, 88 Gordias, 127
Demokedn, 151 Gorgos (brother' of Periander), 19, 20n. 34
Demoni1cos, 121 Gorgas (01Ia505), 85, 86
Demos, 65, 67 (caption)
Demosthenes (general), 105 Hannibal, 16, %n. 14
Demo$t~nes (Ofator), 3,31,35, 47n. 22,54, Ht'Jttorides, 22
62 (~prion), 66, 13·76, 80·2, 124, 139, Hekabe (or Hecuba), 34, 124
140, 160 Hekror, 6On. 56, 123, 124n. 20
Deu1calion, 20 Helen, 125
Dicitrephn, 131 Hellen, 20, 21
Dikaiarchos, I 11 Hephaistos, 63n. 64
Dinarchus, 80, 82 Hera, 134 (caption)
Diodorus, 124 HerakJritos, 26n. 55
Diognctos, 23, 63, 64 (caprion) Heraklcs, 37 (caption), 52 (caption), 124
Diomcdes, 1, 2, 5, 12, 17, 49, 51, 57n. 43, Hermes. 52 (caption)
58·61, 66, 70, 159 Herodotus, IS, 38, 39, 41, 43, 45, 54, S5
Dionysius (of Sicily), 49, 106 (caption), 60, 68, 16, 118. 121, 123, 151,
Dionysios (Thracian), J03 157
Duris, 108 Herophantos, 102
Hcsiod, J24
Echekrates, 103 Hicro I, 18
Eetion,22 Hicro 11, 36
EJectra, 36n. 86 Hierokles, 131
Empedocles, 62 (caption) Hippias (of Athens), 44, 91, 110, 116
Endios, 19, 20. 143, 148, J49 Hippias (of Elis), 26n. 50
Epicharmos, 85 Hippoklm, 102
Epyaxa.47 Hippolaidas, 93
Etearchos, 71, 123 Hippolochos, 166
Eteokles. 38 Hippomedon, 85.86
EU.Jiphnos, 960. 78 Hisriaios, 39, 40, 102. 104, 107. 109, 122n.
EuchaFn,8S 14
Eudoxos,6Sn.68 Homer, 21. 63, 76, B2, 156
Eukleides, 131 Hy~des, 75, 88n. 48
Eukrates, 23 Hyst2spe5, 156
Euktemon (archon). 131
Euktcmon (pentceontarch), 93 ldomcneus, 38
Eumaeus,55 Imilcho, 63, 64 (caprion)
EUrMnes (of Cardia), 112 Imylch, 63, 64 (caption)
Eummcs (of Pcrgamum), 47 lolaos, 20, J27
Euphttes,61 Iphik,atcs, 23, 44
Euphorbos, IS6 Iphiros, 41, 58, 60, 61, 66, 124
Euripides, 22, SO, 124, 158 Isagoras, 45
Eu'ldike, 23 Ishi·Adad, 79n. 21
Eu'llochos (of Larina). 18 Isme·Dasan, 79n. 2]
Eu'llochos (of Mapcs1a), 102 Ismmias.5
Eu'lmcdon, 105 I socra tes, 17,26,35,45,47, 65, 70, 83, 94,
Euthykratcs. 73. 88n. 48 95, lOS, 121, 124
Evagor.s,44,6S, 94n. 72,97,105.106
Jason, 26n. SO, 27, 34
Gellner. 162. 163 Jona~an,S8n.48,61m.61
GerBis, 107n. 116
Glau1cos (lykian), I, 2. 5, 12, 17, 49, 51, Kegan, 146
57n. 43, 58-61, 66, 10, 123, 125, 159 Kallias (of Chalkis), IS4n. 102
Glaukos (Spartan), 96n. 78 Kallias (son of Hipponilc.os), 44, 76, 80n. 28,
Glaukothea, 35 13Sn.51
206 Index
Kama. (of Sphmos), 85, 86 lysitheidcs,47n. 18
Kalligeitos, 29n. 63 Lyson, 63, 64 (caption)
Kallixenos (mipimt of estate), 111
Kallixenos (victim of ostracitm), 1S7, 158 Maiandrios, 79n. 24
(caption) Mania, 34n. 75, 101n. 117
KephUria~,47n. 21 Mardonius, 157
Kephisodotos, 28 Marek. 142
Kephdophon, 12n.6 Masistes, 1220. 14
K~hS, 19, 127 MlnoU, IS (caption)
Kimon, 21, 117 Mausolus, 139
KJandros. 22, 35 Medea, 27, 35
Kleanor, 99, 100 Mcdistas, 127n. 37
Klearchos (lsocrares' xmos), 26n. SO, 65, 70 Mcpbates, 118
Klcarchos (mercmary commanda'), 17, 45, MCSabyzos, 104
41, 91, 92, 99, 100, 104, 119, 120, 121, Mega, 61, 63n. 64
125, 126 Mqpsrias, 26,34n. 74
Klcinias, 19 Mcidias, 140
Kleitarchos, 140ft. 62 Meigs, 145
Kleitos, 85 Mcmnon, 110
KJcoboulos, 49 Mcnander, 16n. 18
Kleomcncs I, 45, 76n. 11 Menckrltes, 132
KIeomcncs III, 28n. 62, 123 Mmelaos, 63n. 64, 84n. 40, 125, 126
Knopias, 103 Menon, 47, 99, 101, 104
Koes, 109, 122n. 14 Mm~, 18,58,59,60,77
Koinos, 110, 113 Metiochos, to'
Komanos, 111 Metrodoros (father of honorand), 85
Konan, 94n. 71,105 Metrodoros (tyrant), 102
Kotys, 54 Milriadn, 44, 102, 109. 135n. SO
KnteUas, 110 Minnion,86
Kraugis,22 Mithrida~ (of PanNI), 23
Kmssig, 113 Mnasiades, 103
Kylon. 150 Mnesimacho., 1l0, 112, 114
Kyncas, 156
Naphur, 89n. 49
Labynetos. 98 Ncon, 15,42,49, 122n. 14
Laches, 147 Ncoprolemos, 97n. 86
Laistner, 117 Nausikrates, 62 (caprion)
Lakcdaimonios, 21 Nektabanis, 6Sn. 68
Lakon,21n.39 Nikagoras, 28n. 62, 123
Laodamas, 102 Nikanor, 11I
Lelch,69 Nikcratos (mover of decree), 85
Leontia~ S Nikcratos (son of Nikias), 23
Leocychidas, 45n. 10, 76 Nikias (of Athens), 23, 145-9, 1SOn. 86
Libys (of Cyrcne), 21 Nikias (of Crete), 141n. 66
Libys (of Spana), 21 Nikogencs, 47n. 18
Lkhas (example of name), 21n. 38 Nikoklcs, 26n. SO, 84n. 40
Lichas (of Spana), 20, 21n, 38 Nikophemos,96
Lichcs (of 111asos), 20, 21 n. 38 Nikostratos, 29, 93
Logblsis, 23 Nymphodoros, 137, 138
Lypmis, 90, 91, 127n. 37
tykaon,22 Odysseus, 17, l5, 38,41, SO, 55, 56, 57n. 43,
Lykidas, 157 58, 60, 61, 63n. 64, 66, 74, 96, 101,
Lykon (of Athens), 108, 110 J35n.50, 162
Lrkon (of Heraldea), 21 0ilcus,21
Lysander, 21,44, 142n. 71, IS6 Oincus, 36n. 86, 60
Lyaias, 23, 28, 96 Oiniades, 131, 133 (caption)
Lysimlchol (King), 83, 85, 112 Omphale, 124
Lysimachos (UNOS of Ptolemy?), 112 Ophellas, 125
Index 207
Or6tn, 22, 36n. 86,50,158 Pleistoanax, 76n. II, 158
Oroetn, lSI Ploutarchos (of Eretria), 140
Orontas, 100 Plutarch, 20, 57, 112
Pnytagoras, 108
Pambi!, 102n. 101 Podanemos, 152
Parapira, 58 Polemainetos, J7, 34n. 74
Parke, 97, 98 Polemokrates, 110, 113
Pasion (Athenian banker), 22, 35, 39, 47, 92, Polyalkes, 34n. 75
93, 94, 95, 96 Polyalko, 26n. 50, 34n. 75
Pasion (of Megara), 99, 100, 126 Polybiu~ 15, 18,23,35,38,42,43,53, 103,
Pausanias (traveller and geographer), 156, 112, 123
157 Polybos, 84n. 40
P:iusanias (Sparran genn-al), 22, 23, 41-3, 47, Polychar~ 96n. 78
89, 91, 116, 144, 158 Potydoros, 124
Pausanias (Spartan king), 150, 153 Potykles, 92
Peirithoos, 98n. 89 Polykrates (of Argos), 103
Peisistratos, 21, 44, 76, 90, 91, 127n. 37, PoIykrates (of Samas), 17,41, 46n. 14,60,
153n. 95 71, 15 I, 152
Peleus, 24, 25 (caption), 26, 27 (caprion) Polymestor, 34, 124
Perdikkas II, 119n. 7 Polyxeinos, 2 Jn. 40
Perdikkas (Philip's brother), 23 Popilius taenas, 52-4
Perdikkas (son of Koinos), 110, 113 Poseidoni05, 127
Periander, 19, 121 Porasimto, 102
Perikleidas, 21 n. 39 Pouilloux, 21n. 38
Perikles, 3, 36, 45n. 10, 71, 78, 143, 144, Priam, 22, 124
157, 159, 166 Proitos, 63n. 65
Perillos, 160 Prolcl~ 152
Phaidimos, 63n. 64 Promathos, 26
Pharnabazos, 1,2,3,34n. 75,46,47,51,58, Proteus, 126
153n.95 Proxmides, 137 (caprion)
Pharnaba~s'5On,15,28,45,47,58,66,152 Proxenos (of Boeotia), 4, 14, 26, 28, 45, 47,
~don. 79n.21.96 99, 100
Philagros, 156 Proxenos (of Cnidus), 137 (caption)
Philip II, 3, 23, 35, 39, 43, 44, 58n. 48, 73, Proxenos (example 01 name), 21
74, 76,81,82,84n.40,88n.48,89, 106, Psammara, 102n. 101
110, 113, 121, 122, 127, 152., 154, 156, Psamrnerichos (I and II, Pharaohs), 101-2
160, 161 Psammetichos (son of Gorgas), 19
Philip V, 46n. 14 Psammetichos (5On of Theokles), 19, 101-2
Philippides, 83, 85, 86, 97n. 86 Ptoedoros, 160
Philiskos, 110 Ptolemaios (recipient of estate), J 13
Philistides, 14On. 62 Ptolemy I, 85
Philokratn, 81, 82, 154n. 102 Ptolemy II, 85, 111, 112
Philon, 103 Ptolemy III, 111, 112
Philoneikn, 84n. 40 Ptol~y IV, 102-3, J 11
Philopoimm, 22, 35 Ptolemy (5On of lysimach05), 111, 112
Philoxenos (example of narne), 21 Ptolemy (5On of Ptolemy), 110
Phoibidas, 5 Ptolemy (5On of Thraseas), 103, 111, 114
Pholos, 26n. 52, 52 (caption) Pylades, 22, 36n. 86, 50, 158
Phoxidas, 103 Pyrilampes, 67 (c:iption)
Ph ry non, 127n. 37, 152 Pyrrhus,36
Phryon's 5On, 152 Pythagoras (of Selymbria), 136 (caption)
Phylakos, 107, 110 Pythagoras (of Sparta), 99, 100
Phyleus,61 Pytharchos, 108, 109
Phylopidas, 81 Pytheos, 110
Pindar, 17, 18, 45, 129, 156 Pythios, 38, 45, 59, 84n. 40, 89
PiS5Outhnes, 108 Pythodoros, 47
Pitt-Riven, 32-4 Pythokles, 35n. 79
Pbro, 49, 62 (caprion)
208 Index
Rahn. 15n. 14 Thessalos, 21
R.rneses II, 101 Thetis. 24, 26, 27 (caption)
Rhesus, 98n. 89, 98n. 90 Thibron, 104
Rostovtzeff, 114 Thorax. 17. 1S7
Roy. 98 Thoukydidn (of Pharsalus), 141 n. 66
Thrasyboulos (of Arhens). 153
Salviat. 21n. 38 Thrasyboulos (of Miletus), 121, 122
Samios, 21n. 39 Thrasydaios, 39
Sa~n, 123 Thrasyllochos. 35
Satyros (of Athens). 26, 34n. 74, 121 Thrasyllos (Argive general). 145, 146, 1S8
Satyros (of Bosporus), 106 Thrasyllos (xmas of Polemainetos), 17
Saul. 48n. 25. 58n. 48 Thucydides. 21. 21,41,43. 78, 90, lOS, 119.
Scipio Aemilianus, 18 122, 137, 138. 142-8, ISO. 151. 161
Scopas, 78n. 19 Timagoras, 29n. 63
Senrbeus. tit Timegenidas, 156
Seuthes, 18. n. 106 Timokrares. 80n. 28
Shalmaneser III. 51 (caprion) Timokreon, 123
Simonides (mov~ of decree). 85 Timotheos. (Athenian general), 82
Simonides (poet), 26. 135 Timotheos (Isocrares' xenos). 47n. 22. 65.
Sitalkes, 90. 137 70. 121
Sky)ax. 118. 120 Tissaphernes, 116. 125
Socrates (Athenian philosopher), 4, 12, 27. Tydeus. 61. 166
35, 78n. 19. 87
Sokrates (Boeotian). 103
Sokrates (Cyrus' xmas), 45. 99, 100
Sopaios, 29n. 63 Van Genncp. 68-9
Sophaineros, 45. 100 Veblen. 129
Sophocles. 60n. 56 Veyne,97
Sosibios, 103. III
Sosis, 99. 100 Walbank, M.B.. IS. 133 (caption), 136
speaker in lsocrates 19, 35. 94 (caption). 141
speaker in Lysias 19, 94n. 71 Will, 117
Suatokles, 87. 88n...7
Stratris, 102 Xanthippos, 45n. 10
Strophios, 22 Xenagoras, 122n. 14
Symnesis. 39. 45. 47. 90 Xenias (of Arcadia). 99. 100. 126
S~05On, 14. 15,41.49.70.89 Xenias (of Elis). 156
Xenocrates. 78n. 14
Tamos, 99 Xenophon.... 5n. 8.14.15.18.26.46.47.
Tegnaeul, 59 49.66.67 (caprion). 77. 81n. 29. 87. 88,
Telamon,21 90. 92, 98. 119, 121, 125, 129. 135
Telemachos, 18, 58. 59, n Xmophon (example 01 name), 2 t
Too, « Xenophron. IS"n. 102
Terrs, 127n. 37 Xerxes, 38. 39, 41. 42. 45. 47. 59. 66. 67
Theagenes, ISO (aption), 89, 91, 108-10, 113, 135n. SO.
Thcmison, 34n. 74, 38. 71. 123 156, 1S7
Themistoldes, 43, 47n. 18. 79n. 23. 9t, 94n.
71. 109. 116. 117. 122. 123. 124n. 20, Youns man (from lsocrates, Trapniticus),
135n.50. 14~ 15~ 159 83, 88. 93. 9S
Theokles, 19. 102
Theomestor.39 Zenon. 85. 86
Theopompos, 39. «, 154 Zeus, 61, 124i Saviour. 105i Xmios. 56. 66.
Theophrastos. 83 125
Thenippos, 85, 86
Indn 209
INDEX OF SUBJECTS
accountability, 2, 6, 10, 79n. 24, 93 institutions and rules of, 30, 78, 138, 142,
ad~/phos, Stt 'brother' ISO, 164; given as gift, 43, 108, 111-12;
administration, 106, 114n. 140, 115, JJ9, anirude to bribny, 73-81; ideals
164 propagated by, 97, 159-60; attitude to
adorodoltetos, 76-8; U~ also bribery rai5ing troops by xmoi, 104-5; as
adorotatos, SH adorodoketos partners in proxenies, 130-42; turning
affection, 13, 17-18,48, 129, 149 xmia into proxmia, 132-42; clashing
agenu, 39, 83,95, 97n. 86, 142 with upper-cialS coalations, 138, 163
agreerMnts, 53, 54, 93, 95, 132; SH also citizenship: obigarions of, 3, 4; ideals of, 18,
documents and symbolon 79, 151, 159, 161; gr~nrs of, lOS, 141;
alliances, 5,9, 14,23, 34, 36,43,44,46, SO, incompatible with '~"trgts",', U5n. 53.
53 (caprion), 59, 60, 74, 98, 102, 134 promotion to, 140-2, 152, 177-8
(caption), 137, 144-9, 152, 156, 179-84, dassn: conflict of, 8, 119, 130, ISS, 160-4;
allies, 29,44,45,54,56-7, 74, 79, 84, 86, 90, up~ and lower, 28, 34, 73, 74, 104,
116, 141, 145, 157 128-9, 149, IS 7, 160; pursuih
altruism, 32, 48, 121 appropriate to, 129; ~e also dites, dmros
ambassadors, U~ embassies clients, 14, 38, 39, 102, 156
amity, 13, 32, 80 client rulers, IS, 39, 40, 41, 49, 10, J02, 103,
Anabas;s, s~~ Ten Thousand 1S4
ana"ltaios, 10 communities, stt city
ancestral proxt,,;a, proxmos, 133, 135, 137, compadral.go, ~t godparenthood
138, 147 comparative perspective, J 1-4
ancnrral xtnia, xmos, I, 21,59, 69-72, 82, conspicuous expmditure, 97
92, '02, 125, 138, 141, 152, 148, 166f{' consubstantiality, SS (caption), 66
corruption, n, 87, 88; ste also brihery
bank operations, 39, 93, 94, 95 courtiers, see roy~1 offici~ls
baptism, 69
benefactions, ~~ t"ngts;a dt-brs, 41,48, 121; unrepayable, 122; su also
benefih, s~t n4n~si4 loans
~rith, 16n. 17, 58n. 48 declararion, 59, 68, 11, 133 (caption)
betrayals, sn r~ason de~ttes, 75, n, 84-8, 130, 13', 135, 137
blood-brothnhood, 32, 33, 55 (caption) (caprion), 138
blood rituals, 33n. 71, 54, 55 (caprion), 59, donas, 3, 41,73, B3, 144, .46, 147, 156,
68,69 1S7, 160, 161, 164; $It also classes
boethtia, 4, 23, 25 (caprion), 48, 121-2, 129; dex;a, 46, 49, 50-4, 57, 58; ue also
Stt also opportune help handshake, pledges
bond-friendship, 32, 33, 127, 129n. 44 dex;osis, see ha"dshalt~
booty,97 documents, 16n. '7,77, 95, BO, 131, 135,
bonowing, S~t loans 137 (caption); SN also agreements,
bribery, 5, 7, 73, 73-81, 1S3, 1S8 decfftS
'brother'. 18, 22, 2.l, 33, 46n. t 4, 55 doctoni, 15 I
(aption), 58n. 48, 59, 19n. 21, 89n. 49 donations, oh honorem, 97; ue also liturgies
burnucracin, g t administration dor~tJ, 76, 108, 109, III-IS, 135; Stt also
burial, 26, 135, 136 (caption); $H also ""dn estates, gifh
epitaphs dorema, see doron
dorodo#tia, 75, 76; ut also bribery
caprives, 57; ~t also slaves dOrmt, dora, 60, 15, 16, 77, 90, 108; Stt also
charis, 28, 41, 42, 48, 91, 108, 129, 135 ""der estates, gifh
chieftains, 99, 142, 155 doryxlnos, 10, 11,57,166
chremaUl, In money dowries, 26, 121, J29
city: in conflict with ritualised friendship,
1-1, 14.4, 144, 153, 163, 16~; qafitarian ethos, ~t ""tln ritualised
enforcement agencies of, 5-6, 30, 95; as a friendship
self-conscious community, 6, 132, 142, elites, 34, 104, 155, 156; ~t also classes,
157; friendship within, 29-3 I; formal social srarus
210 Index
embassi~ 45, 65, 76, 112, 115, 139, 147, (caption), 52 (caption), 58, 59, 70, 76, 78,
148 135, 144; spirit of giving of, 10, 89, 90,
encounter, 41-58, 101 95, 108-9, 109; of initiation, 24 (caption),
epitedeia, epikdeios, 10, 12, 93, 119n. 7, 41, 58, 60-3, 71; symbolic, 24, 32, S0, 58,
130, 143. lSI, 152, 153, 179 60, 66, 67 (caption), 69, 84, 129;
epitaphs, 5n. 8, 26, 129, 132, 136 (caption); counter-gifts, 41, 60, 61 n. 61, 80, 97;
see also burial cities given as, 43, 108, 112; of real
estates, 18, 100, 103, 106-lS, 154; given as use-value, 59, 61, 73, 75, n, 81-J 1S; and
gifts, 61, 104, 107-8, 111-12; reason of tra~ 61, 80, 95; of hospitality, 70; as
grant, 107; documentarion of, 107; types bribes, 73-81; in Homeric society, 78-9;
of, 107, 109, 112-13; in Penian. long-term effect of, 80, 91-2; estates
Macedonian and Hellenistic empires, granted as, 81, 107 i slaves as, 81; natural
109-111; ownenhip of, 113-14; produet5 as, 81-8; given to cities, 83, 86,
revocation of, 114; and feudal benefices, 142; troops supplied as, 84, 105; money
164-5 i S« dlso dare" .15, 88-92; valuables as, 88-97; as taXeI,
etiquette, 41-4 89, 90; as loans, 92-3; as civic privileges,
ftltrglSia: communal, lS, 84-6, 105, 135, 135; ut also dorta
152; private, 24-5 (caption), 42, 46, 57, godparenthood, 12n. 9, 19,27 (caprion), 32,
58, 101. J04, 107, 108, 112, 129; as 33, 72n. 92, 118, 127, 128
stamng mechanism, 25, 47-9, 101
eungeta: private, 10, 107, 112; Persian
'good deed" s~~ ~~sia
grain, 1ft u"tIn natural products
court tide, 13, 41, 107; communal, 131, grammata, 1ft letters
135 'grandfather', 18
exchange, see gifts gratitude, 1ft cham
expertise, 96, 103-4, 129-30, 151 GCftks in Penian service, 43, 98, 154, 157,
161
factions,S, 142-56, 164; composition of, purlns. 46, 49, 54
150-2, 154-5; and political ideologies, guest-friend(s), sn XePfOl, ritualised
153; expansion of, 153-4 friendship
family, see kinship guest-friendship, see ritualised friendship
'father', 18, 59
favoun, 3, 4, 28, 100, 121, 122, 148; see also handshake, 37 (caption), 46, 50-3, 63, 64
(caption), 134 (caprion); su a/so d~x;a,
,har"
feasting, 57, 59, 66, 136 gesrures
feudalism, 2, 38, 59, 107. 118, 164-5 hand-tokens, 54
fictitious kinship, ue NIUIer kinship Hellenistic ruins, su .,,,dn ruling circles
fidelity, see ""der rirualised friendship, norm hetairia, hetairas, 10, 12, 18, 19, 38, 55,
of loyalty of l22n. 13, lSI, 152; Macedonian court
followers, 44, 86, 90, 111, 122, 146, 148-56 title, 8, 13, 44, 73, 106. 113, 154, ISS,
foreigneR, 3, 4; see also mangen 164
'foster-brother', 59 hero: me raming of, 2, 7, 78, 138, 157, 159,
foster-parenthood, 22-6, 35, 124, 125, 126, 160, 164; Homeric, 1, 2, 6, 34, 72, 88,
127, 129, 135, 137 (caption), 152 104, 156, 159, 164
fraternity, see 'brother' hilteuia, biltetes, sn supplication
friends, friendship: in general, 4, 11, 12, 14, homonymity. 20-1, 141, 177-8
42, 43, 56, 76, 80, 83, 84, 93, l15, 120, homot,.ap~%as, 125; see also
127, 144, 149, 150, lS3, 160, 161; consubstantiality, feasting
unritualised, 16, 29-31, 32-4, 58; of horltio", hm-Itia, see oaths
friends, 27, 47, 65, 152; wimin the city, hospitality, 10, 28, 59, 6On. 55, 66, 70, 122,
29-3 I; of communities, 143, 148, 149 124, 125, 126, 128, 129, 132
hospitium, 38n. 95, 39, 46n. 14, 54n. 37,
122n. IS, 166
gesrures, 24 (caption), 25 (caption), 27 hostages, 23, 28, 100, 126-7
(caprion), 37 (caption), 43. 5On. 33, 5 I
(caption), 52 (caption). 53 (caption), 55
.
idioxePfos, 4, 10, II, 166
(caprion), 56, 57n. 43, 60n. 55, 6On. 56, initiation, see .",tIn rites
64 (caption), 134 (caption), 137 (caption) interest (in moneylending), 93, 94
gi~ exchange of, 3. 7, 13, 42, 45, 50, 51 interests: communal, 3, 75, 79, 86, 139, 140,
Index 211
•
142, 152, 153; private, 48, 86, 123, 126, Pdoponnesian War, 119, 143·53, 157,
142, 149, 152, 153, 161; a king's, 115; 179-84
communal subordinated (0 private, 130 phi/ia, 3, 19,23,30,46, 73, lOS, 120, 141
introduction, 47, 101 phi/os: private, 1, 10, 12, 18,23,38,46,47,
Ionian revoh, 14, 118·19 SO, 51, 56, 57, 87, 106, 107, 108, 112,
122n. 13, ISO, 153, 164, 179; Hetlenisric
kinship: ral, 16·29, 19, 31, 32, 34, 49, 80, court title, 8, 13, 38, SO, 103, 106, 108,
84, 128, ISO, ISJ; pseudo- , 16·29,32-3, 109n. 126. 112, 155; of communities, 23,
46n. 14,63,69, 128, 133 (caption), 135; 126, 135; Persian court ririe, 42, 106,
adoptive, 32; riwal, 32 108; see also u"der ruling circles
Ittm.sh!O, 48n. 24 phi/aleS, see affection
pistlJ, pistis, 46, 49·50, 54, 57, 58, 147, 149;
letten, 26, 47, 65, 89n. 49, 93, II~ 121, see also pledges
159n. 118 pledges, 46, 47, 98; see also dexia, pista,
libations, 49, S0, 66 symbalorr
liturgies, 65, 83, 86, 92, 96-7, 116 plunder, 97, 102
livestock, su rmdn natural products precious objects, 13,58·67,81, 89
loans, 28, 39,65,67 (caption), 92-4, 96, 121, preliminaries, 44-58
128, 129 polis, set city
loyalty, lee under ritualiKd friendship p~rty; pub'ic, 3, bequeathed to xmos, J7,
96, 129; confiscation of, 95, 159; estates
marriage, 13, 14, 16, 24 (caption), 26, 27 as, 112-14; see a/so estiltes, natural
(caprion), 29, 31, 36, 42, SO, 91n. 62, products, safekeeping
138, 1S0, 162, 164 proxenia, see proxtntt-s
meals, see feasting proxmies,4, 15,75, 122, 130-42, 145, 1S3;
mediators, 46-7, 52 (aption), 101, 123, 138 relation to xmia, 27, 71, 135, 139, 146;
~runaries: soldi~, J0, J7-18, 98, 100·4, inheritance of, 135, 137, J47;
commanden, 17,45,98-101, 103 recommendations to, 137-41, 177·8;
metals, 79, 89, 90, 96 peculiarity of Spartil, 138n. 55; renewal
money, 42, 46,65,75,76,83,86,88·92,95, of, 138; implying xmia, 139-40, 146;
101, 113, 119, 121 multiple, 139; fifth-emrury Athenian,
141-2, 180-8
naming, 7, 19-22, t 02, ."1
narion, J62-3; ue also nationalism ransom. 22, 42, 57, 93, 122, J28, 129
nationalism, 117, 161, 162 rroprocity, 41, 60, 61, 80, 84, 90, 91, 92, 93,
natural produm, 82·8i grain. 82·6, 154; 100. 101, 108·9, 121. 124, 136
livestock, 82; timber, 82·3, 88, 154 recommendations: to citizenship, 141, 177-8;
netWorks, 6, 7, 8. 22, 74, 80, 81. 82, 84·6, to proxmia, 137-41, 177·8; to xenia,
129, 138. 139, 144, 146, 148, 150·2, ISS, 46·7,65
160. 162, 179·84 recruirment. 97·105; see also troops
rmewal: of xmia, J7,45, 58, 60, 63, 65, 70;
oaths. 49, S0, 54, 55, 59, 71, 76, 125 of proxmia, 138, Jtt also u"dn rites
obligations: civic, 2·8; of ritualised resources, 7, 73·115, 15J; su also lI"dn
friendship, 2·8, to, 18.54,61,120·8, estates. Bifts, money, troops
140, 144, created by grant of estate!l, revenge, 23, 125, 128, 129, 135n. 52
113·15, 164·5; of proxen;a, 139 right hand~ see dtxia
oiltnos, o;lce;otes, 10, 19, 28, 29, 31, 105 rirC'S: of initiarion, 7, 29, 30, 33, 58·69, 120,
opportune help. 28; su a/so btHlhna 133; of passage, 68·9; see also renewal
ostracism. 157-8 ritual kinship, 31·4, 59
outsiders, set strangers ritual objects, 49-50, 53-5 (caption)
ritual warfare, 57
pappos, see 'grandfather' rirualised friendship: in conflict with
parernity, stt 'father' communal principle, 1·7, 143·4, 153,
patri#tt proxe"ia, ste anctSrral proxnria 157-9; morn of, 1,2,118·28; violations
patriltos xenos, see ancestra' xenia of, I, 51, 54, 66, 71, 123-7; obligations
patriotism, 15n. 14, 73, 78n. 19. 156-61 of. 2·8, 10, 18, 54, 61, 120·8; norm of
parronage (Roman), 38 loyalty of, 3, 13, 39, 49, 50, 121, 128,
patron-client bond (analyrical term), 39 156, 16l; evidence for, 7, 13·16, 41·3,
212 I"dex
68, 118-20; definition of, 8, 9, 10-13, supplication, 33n. 72, 54-8
162; terminology of, 10-13, 96n. 78, 179; symOOlorr, 53 (caprion), 55, 59, 61, 62
'stronger than kinship', 33n. 72; (caption), 67 (caption)
egaliterian ethos of, 37-9, 57, 61; social symbolism, 7, 8, 36, SO, 51, 58, 59, 60·1, 72.,
units of partners in, 11-12., 32, 166-75; 132
not specifically Greek, 12; its affinity to symmachos, ue alliances, allies
kinship, 16-29; bequeathed to syrretMs, 2J
descendants, 16-17, 69-72, 135, 141; syrtgenes. Syrtgeneia, 10, 19, 38, 45, 106n.
termination of, 17, 71·2, 135; its affinity 116, 154; Ptolmlaic COU" ride, 38;
to friendship, 29-31; in a cross-cultural Penian cou" title, 8, 13, 42, 154
penpective, 31·4; status of partnen in, syrristanai. ue inrroduction,
34-40, IS 1-61; and marriage, 36; and recommendations
dient rulers, 39-40; etiquette of, 41-4; its
role in mediation, 47; and the gods, SO, taxes, 90, 111, 112, J IS
66, 124-5; and the pooling of foren, Ten Thousand, the, 4·5, 49·50, 97·101,
97-105; and the grant of estates, 108; and 119-20, 125, 126
treason, 116-18, 128, 142-61; sanctions timber, ue "nder natural produet5
of, 126-7; services performed in the name trade, traders, 34n. 7S, 4S, 61, 71, 80, 95,
of, 128-30; responsible for proxmin, 123, 128
141, 146; and political ideologies, 1S3; treachery, see treason
and international politics, ISO; sn also ~ason,2,4,8, 76, 79,80,116-18,140,153,
'brother" 'sons' 156·61
ritualised relationships, 7, 31-4, 69 ",carin, 4, 46n. 14,53 (caption), 59, 98, 118,
royal officials, 29n. 63, 40, 42, 44, 49, 70, 134 (caprion); SN also alliances
83, 85, 86, 95, lIS, 116, 137, lS4, ISS, trierarch)', 92; see also liturgies
164 troops, 42, 81, 90, 97-105, 111, ISO
royal pages, 126n. 34, 127 truce, 46, 49, 145
ruling circles, 8, 75, 84, 109-11; Hellenistic, trust, 14, 48, 84, 92., 100; ue also ritualised
8, 38, 44, SO, 103, 142, 153, 1S5, 164; friendship, norm of loyalty of
Macedonian, 8, 153-4, 164; Persian, 8,
13n. II, 66, 103, 1S3-4 valuables, 88·97; see also loans, money,
precious objects
safekeeping, 27, 94-6, 124, 129 vengeance, see meRge
security (in moneylmding), 65, 93, 95
sentiments, see affection wealth, circulation of: 8, 7J-S, 81, 96, 104,
services, 81, 128-30; ritual, 25 (caprion), 26, 105, 144; see also ""dn gifts, city,
81, 129, 135, 136 (caption); private, 81, money, grain
129-30, 1S6; political, 81, 130, 146, 1S5, women, as partnn's in xenia, 34
1S6 widioxmos, ue idioxenos
sharing out, see liturgies
slaves, 56, 81, 150, 151 xenia: private, 60; communal, 132, 136; see
social status, 34-40, 74, 97, 103, 104 also feasting, gifts of initiation
soldiers, see troops xeineie. x~;"ie. xemos. xenia. xenos: private
'sons', 22-3, 28, 42., 48n. 25, 1S2 1-23, 24 (caprion), 26·8, 30, 31, 33·9,
soothsayers, 17, 34,45, 151 45-7, 50, 56~66, 68·71, 73, 74, n, 82,
spo"dai, 46, see also libations, truce 87, 88, 91-7, 98-102, 104, 118-29,
strangers, strangeness, 4, 10, 11, 12, 43, 58, 131-44, 146, 148-61, 164, 166, 179; of
62 (caprion), 68, 69, 72, 92., 94, 95, 103, communities, 38, 135; sa also under
108,130·1, 136n. 54,139 ritualised friendship
submission, 37·40, 44, 45, 49, 56, 89, 90, xeinoi patroio;. see anastral xntia
121, 1S5 xenoktonia, 3, 54, 66, 124, 125
CAMBRIDGE
UNIVERSITY PRESS
www.cambridge.org
ISBN 0-52'-52210-2
IIII l1li11
9 780521522106

Tips - Ritualised Friendship and The Greek City PDF

Caricato da

Informazioni sul documento

Titolo originale

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

Tips - Ritualised Friendship and The Greek City PDF

Caricato da

Copyright:

Formati disponibili

PUBLISHED IY THE PRESS SYNDICATE OF THE UNIVERSITY OF CAMBRIDOE

The Pill Building. Trumpington Street. Cambridge. United Kingdom

o Cambridge UniYenity Press 1987

A (ala/ogut rtcordfor Ihis book is aWJi/ab/r from 1M Brilish Library >-

ISBN 0521 32541 2 hardback

List of tables page IX

1 The Etiquette 41 >-

. THE CIRCULATION OF RESOURCES

S OBLIGATIONS: HEROIC AND CIVIC

5 Followers, Factions, and the City 142

APPENDIX A: Xenoi. Idioxenoi. and Doryxenoi 166

1 Supply of Natural Products 85

1 The xenos as a foster-father

the guest-friendships he contracted were his own private affair. >-

• Xmophon. Anabasis J.1.4H.; d. Diogenes Laertius 2.58 (= Anthologia Palalina 7.98),

pose is this. >-

these words as interpersonal designations will be considered. The procedure

Xenophon was a participant-observer: he concluded several alliances of >-

tJ Herodotus 3.139 fl.• d. below p.""

had two practical manifestations. First, the bond could be renewed or

not indispensable for a relationship to endure: >-

Kleiniu. - - - - - - - - - - - JCetto; - - - - - - - - - - - - - - Alkibiades.

Mostly, it is difficult to trace such successions because of lack of explicit

JIAristotle. Rhetoric 2.1381b 29. d. F.thica Nicomachea 1161b 12-17.

Arketilaos - - - - - - - - - - x~"o; - - - - - - - - - - - - - - lich~

46 JoscphU5.J~h Antiquities 14.122. d. The J~h War 1.181.

fiB. la-cl The ~ as • foster-father

Fig. 1. Ritual ervices

.. Xmophon, H~Jltn;ca ".1.40 and Arrabasu 3.1....

2.4 FEATURES SHARED WITH FRIENDSHIP

Ritualised friendship also resembled friendship in some respects. Yet it was

6$ The elements of the ritual are explained in Osaptcr 3.

A single example will suffice to illustrate how these peculiarities of civic

2.5 THE CROSS-CULTURAL PERSPECTIVE

The pattern of social relationships outlined above is not peculiar to the

these relationships from other, non-amiable social relationships. The sphere

'A m iable relations'

being as a reaction against earlier attempts at classification that were based

2.6 SOCiAL STATUS

Ritualised friendship appears as an overwhelmingly upper-class institution.

76 Euripides, M~dea 613.

Fig. 3a,b Social ltatu ,

102 Herodotus 5.23". d. 5.35.

3.1 THE ETIQUETTE

In a picturesque and apparently fictitious tale, Herodotus relates that Sylo-

on supporting Brachylles and furnishing him with moneyt whereby Brachyl-

• Tele$, ed. O. Hense rrubin~ 1909; 2nd edn.) 23; my rranslation.

facilitated by the extraordinary geographical mobility of the Greeks. The

pledges of non-hostility. Agesilaos' xenia with Pharnabazos' son had prob-

" Xmophon, Anabasis 1.2.12 and 1.2.26.

26 Xmophon, Ag~s;laus 4.4.

Fig. 4 A non-Hellenic iOltitudon?

Fig. 6 Communal symbolon

Another preliminary ritual n~ds special attention since it presupposes a

a similar conclusion. The. best illustration of this contrast comes from a

40 Odyssey 14.279ff., translated by R. Lanimore, The Odyssey of Homer (1967).

We shall return later to this conceptual association between stranger,

3.4 THE INITIATION RITUAL

-The ritual was made up of a complex combination of symbolic elements

.'1 Iliad 6.224-5 .

Fig. , Ivory .,.boIon

Fig. 10 Peni'.n symbolon

The relative scarcity of documentation describing this ritual should by no