Anda di halaman 1dari 213

Families in the Greco-Roman World

The Family in Antiquity

Titles in this series -

I: Families in the Greco-Roman World – edited by Ray Laurence and Agneta


II: Families in the Roman and Late Antique Roman World – edited by Mary
Harlow and Lena Larsson Lovén
Families in the
Greco-Roman World

Edited by Ray Laurence and Agneta Strömberg

Continuum International Publishing Group
The Tower Building 80 Maiden Lane
11 York Road Suite 704
London SE1 7NX New York, NY 10038

© Ray Laurence and Agneta Strömberg, with the Contributors 2012

All rights reserved. No part of this publication may be reproduced or transmitted

in any form or by any means, electronic or mechanical, including photocopying,
recording or any information storage or retrieval system, without prior permission
from the publishers.

First published 2012

British Library Cataloguing-in-Publication Data

A catalogue record for this book is available from the British Library.

ISBN: 978-1-4411-3838-5

Library of Congress Cataloging-in-Publication Data

A catalog record for this book is available from the Library of Congress.

Typeset by Fakenham Prepress Solutions, Fakenham, Norfolk NR21 8NN


List of illustrations and tables vii

Abbreviations ix
Notes on Contributors xiii
Preface xv
1 Introduction: From Oikos to Familia: Looking Forward? 1
Ray Laurence
2 Beyond Oikos and Domus: Modern Kinship Studies and the Ancient
Family 10
Ann-Cathrin Harders
3 Taking a Wider View: Greco-Roman Families and Political
Demography Theory 27
Saskia Hin
4 More than Just Gender: The Classical Oikos as a Site of
Intersectionality 48
Birgitta L. Sjöberg
5 Tracing the Oikos in Pre-Classical Corinth: The Perspective of
Iconography 60
Amalia Avramidou
6 The Language of the Oikos and the Language of Power in the
Seleucid Kingdom 84
Omar Coloru
7 Inheritance, Priesthoods and Succession in Classical Athens: the
Hierophantai of the Eumolpidai 95
William S. Bubelis
8 Women in the Hellenistic Family: The Evidence of Funerary Epigrams 106
Agnieszka Kotlińska-Toma
9 Family Relationships in Late Bronze Age, Iron Age and Early
Roman Veneto (Italy): Preliminary Considerations on the Basis of
Osteological Analysis and Epigraphy 121
Elisa Perego
vi C o n t e n t s

10 Gender, Household Structure and Slavery: Re-Interpreting the

Aristocratic Columbaria of Early Imperial Rome 143
Lindsay Penner
11 Houses, Painting, Family Emotion 159
Natalie Kampen
12 Afterword: The Future of the Ancient Greek Family 179
Mark Golden
Index 193
List of illustrations and tables

5.1a and b Corinthian pyxis from a burial near Corinth, c.550 BC.
Decoration scene, procession, banquet 61
5.2 Corinthian bottle, from Corinth, 600–580 BC. Procession of
women towards a trapeza or altar 62
5.3a and 3b Corinthian amphoriskos from Corinth, c.750–550 BC.
Sacrificial procession 63
5.4 Corinthian column krater from Caere, c.540 BC. Arrival
of the wedding couple to their new home 64
5.5 Corinthian amphoriskos from Vulci, second quarter of the
sixth century BC. Belly: Frauenfest-scene; horsemen.
Shoulder: Banquet scene  64
5.6 Corinthian oinochoe, early sixth century BC. Man, woman
and two children escape a city-under-siege 66
5.7 Pyxis lid by the Dodwell Painter, 590 BC. Youth among
women and a man; hunting scene 67
9.1 Map of Veneto 123
9.2 Reconstruction of Venetic urn with bones and grave goods 125
9.3 Reconstruction of Venetic pit grave 126
10.1 Database 152
10.2 Characteristics by gender 152
10.3 Status distribution 153
10.4 Status distribution by gender 153
10.5 Gender distribution among the columbaria 154
10.6 Status distribution among the columbaria 154
10.7 Occupational categories 155
10.8 Job distribution among the columbaria 156
10.9 Female occupational categories 156
10.10 Overall differences among the columbaria 157
11.1 View into Ala 4 facing the image of the Wooden Horse,
House of the Menander, Pompeii 163
11.2 Detail of the Wooden Horse 164
11.3 Detail of Priam between Helen and Menelaus and
Cassandra and Ajax  164
viii L i st o f i l lu st r at io n s a n d ta b l e s

11.4 Detail of the death of Laocoön and his sons  165

11.5 The Prophecy of Cassandra with Priam looking on, from
Casa della Grata Metallica  166

5.1 Catalogue of Vases (non-exhaustive) 70
5.2 Catalogue of other material (non-exhaustive) 75
5.3 Shape and Subject Relation 77
9.1 Geographic distribution of the sample (urns) 129
9.2 Chronological distribution of the sample (urns) 129
9.3 Age of individuals 130
9.4 Geographic distribution of the sample (epitaphs) 132

Abbreviations for Volume 1

Books and Journals
AAntHung Acta Archaeologica Academiae Scientiarum Hungaricae
AE L’Année Epigraphique
AJA American Journal of Archaeology
AM Mitteilungen des Deutschen Archäologischen Instituts,
Athenische Abteilung
AncSoc Ancient Society
AncW Ancient World
Anth. Pal. Anthologia Palatina
ARV Beazley, J. D. (1963), Attic Red-figure Vase-painters, (2nd
edn), Oxford: Clarendon Press
BMCR Bryn Mawr Classical Review
CAT Clairmont, C. W. (1993), Classical Attic Tombstones,
Kilchberg: Acanthus
CIL Corpus Inscriptionum Latinarum
CJ The Classical Journal
Clair Clairmont C. W. (1970), Gravestone and Epigram,
Mainz: Philipp von Zabern
CQ Classical Quarterly
CR Classical Review
CVA Corpus Vasorum Antiquorum
DNP Cancik, H., Schneider, H. and Landfester, M. (1956–),
Der Neue Pauly, Verlag J.B. Metzler
EEThess Epistemonike epeterida tes philosophikes Scholes tou
Aristoteleiou Panepistemiou Thessalonikes, Tmema
philosophias, Thessalonika
Ephesos Merkelbach, R. and Nollé, J. (1980), Die Inschriften von
Ephesos, Bonn: R. Habelt
FGH Müller, C. (1841–70), Fragmenta Historicum Graecorum
GRBS Greek, Roman and Byzantine Studies
GV Peek, W. (1955), Griechische Vers-Inschriften Bd.1,
Grab-Epigramme, Berlin: Academie-Verlag
x Abb r ev iat io n s

HZ Historische Zeitschrift
IC Guarducci, M. (1942), Inscriptiones Creticae, Roma:
Libreria dello Stato
I. Erythrae Engelmann, H. and Merkelbach, R. (1972–1973), Die
Inschriften von Erythrae und Klazomenai, Bonn: R.
IG Inscriptiones Graecae
I. Iasos Blümel, W. (1985), Die Inschriften von Iasos, Bonn: R.
I. Ilion Frisch, P. (1975), Die Inschriften von Ilion, Bonn: R.
I. Labraunda Crampa, J. (1969–1972), The Inscriptions from
Labraunda, Lund: P.Åström
JHS Journal of Hellenic Studies
JRS Journal of Roman Studies
LGPN Fraser, P. M. and Matthews, E. (eds) (1987), A Greek
Lexicon of Personal Names, Oxford University Press
LIMC Lexicon Iconographicum Mythologiae Classicae
LMA Lexikon des Mittelalters
MedArch Mediterranean Archaeology: The Australian and
New Zealand Journal for the Archaeology of the
Mediterranean World
MNN Museo Nazionale Archeologico Naples
MuZ Pfuhl, E. (1969), Malerei und Zeichnung der Griechen,
(reprint), Rome: L’Erma di Bretschneider
NC Payne, H. (1931), Necrocorinthia. A Study of Corinthian
Art in the Archaic Period, Oxford: Clarendon Press
OGIS Dittenberger, W. (1903-1905), Orientis Graeci
Inscriptiones Selectae, 2 Vols, Leipzig: S. Hirzel
Perachora Payne, H. and Dunbabin, T. J. (1962), Perachora: The
Sanctuaries of Hera Akraia and Limenia; Excavations
of the British School of Archaeology at Athens,
1930-1933, 2 Vols, Oxford: Clarendon Press
P. Giss. Grieschische Papyri im Muzeum zu Giessen
PPM Pugliese Carratelli, G. and Baldassare, I. (eds) (1990-
2003), Pompei: Pitture e Mosaici, Vols 1-11, Rome:
istituto della Enciclopedia Italiana
RE Paulys Realencyclopädie der classischen
RömMitt Römische Abteilung Mitteilungen des deutschen archäolo-
gischen Instituts
SE Alpers, M. and Halfmann, H. (1995),
Supplementum Ephesium, Hamburg: Deutschen
Abb r ev iat io n s xi

SEG Chaniotis, A., Corsten, T., Stroud, R. S. and Tybout, R.

A, Supplementum Epigraphicum Graecum, Leiden:
TAPA Transactions of the American Philological Association
Thess. Mnem. Arvanitopoulos, A. S. (1909), ΘΕΣΣΑΛΙΚΑ
Περιγραφη των εν τω αθανασακειω μουσειω Βολου γραπτων
στηλων των Παγασων, 2 Vols: Athens
ZPE Zeitschrift für Papyrologie und Epigraphik

Ancient Sources
Aesch., Ag. Aeschylus, Agamemnon
Andoc. Andocides
A.P. Anthologia Palatina
App., Syr. Appian, Syriaca
Arist., Pol. Aristotle, Politica
Arr., Epict. diss. Arrian, Epicteti dissertations
Ath. Athenaeus
Callim. Delos Callimachus, Hymn to Delos
Cass. Dio Cassius Dio
Cato, frg. Cato, Fragmente
Cic., Inv. rhet. Cicero, De inventione rhetorica
Cic., Off. Cicero, De officiis
Cic., Planc. Cicero, Pro Plancio
Dem. Demosthenes
Dig. Digesta
Diog. Laert. Diogenes Laertius
Fest. Festus
Gell., NA Aulus Gellius, Noctes Atticae
Hdt. Herodotus
Hes., Erga Hesiod, Erga
Hyper., Epit Hyperides, Epitaphos
Inst. Iust. Institutiones Iustiniani
Isoc. Isocrates
Lys. Lysias
Men., Sam. Menander, Samia
Nic. Dam. Nicolaus Damascenus
Ov., Fast. Ovid, Fasti
Plin., Ep. Pliny (the Younger), Epistolae
Plin., HN Pliny (the Elder), Naturalis historiae
Plut., An seni Plutarch, An seni respublica gerenda sit
Plut., Mor. Plutarch, Moralia
Plut., Vit. Cat. Mai. Plutarch, Vitae Parallelae Cato Maior
xii Abb r ev iat io n s

Polyb. Polybius
Suet., Aug. Suetonius, Divus Augustus
Suet., Claud. Suetonius, Divus Claudius
Notes on Contributors

Amalia Avramidou is a visiting lecturer at the University of Crete in

Rethymno, Greece. Her publications include a monograph entitled The
Codrus Painter: Reception and Iconography of Attic Vases under Pericles
(Wisconsin University Press, 2011) and many articles on Greek and Etruscan
art and archaeology. She has excavated in several sites in Greece and
specializes in Greek pottery, iconography, and cross-cultural relations in the

William S. Bubelis is an Assistant Professor of Classics at Washington

University in St. Louis. His work focuses upon the economics of ancient Greek
society, particularly where matters of religion are concerned. Recent publica-
tions of relevance to ancient social history include “The Sacred Triremes and
their Tamiai at Athens” Historia 59 (4) (2010), 385-411.

Omar Coloru is a researcher (ATER) at the Collège de France, Paris, «Chaire

d’histoire et civilisation du monde achéménide et de l’Empire d’Alexandre».
He received his PhD from the University of Pisa and the University of Paris I.
His research interests include Hellenistic history, Greek epigraphy, the Seleucid
empire, the Greco-bactrian kingdom, and the cultural exchanges between
Greece and the East. He is the author of the monograph Da Alessandro a
Menandro. Il regno Greco di Battriana (2009).

Mark Golden is professor of Classics at the University of Winnipeg, where

he has taught since 1982. He has written or edited seven books, most recently
Greek Sport and Social Status and A Cultural History of Sexuality in the Classical
World (edited with Peter Toohey).

Ann-Cathrin Harders is an Assistant Professor of Ancient History at the

University of Bielefeld. Her main research interests are focused on family
and kinship in Greece and Rome and the wider social history of the Roman
Republic and Hellenistic History. She has published a monograph on Roman
kinship, Suavissima Soror (2008). She has also published on Roman prosopog-
raphy and the Roman family. Her current research project examines “King and
queen. Power and Gender in Hellenistic monarchies”.
xiv N o t e s o n C o n t r i bu t o r s

Saskia Hin is a postdoctoral research fellow at the Max Planck Institute for
Demographic Research, where she works on Greco-Roman demography. Her
primary interest is in social and economic history, in demography, and in inter-
disciplinary approaches to these subjects. She holds degrees in Classics, History,
and Political Sciences. Previously, she has published work on the ephebeia in
the Hellenistic world, on fertility behaviour and on the Roman practice of
census taking. She is the author of The Demography of Roman Italy: Population
Dynamics in an Ancient Conquest Society, 201 BCE – CE 14 (forthcoming,
Cambridge University Press).

Natalie Boymel Kampen is Professor Emerita of Women’s Studies and Art

History, Barnard College, Columbia University. She writes on issues of gender
and social status in ancient art and on the art of the Roman provinces. Her most
recent monograph is Family Fictions in Roman Art (2009).

Agnieszka Kotlińska-Toma is Assistant Professor in the Institute of Classical,

Mediterranean and Oriental Studies at the University of Wrocław. Her research
interests cover Hellenistic literature, especially scenic genres and epigram. Her
monograph on Hellenistic tragedy (originally printed in Polish by Wydawnictwo
Uniwersytetu Wrocławskiego in 2006 with the title Tragedia hellenistyczna) will
be shortly translated into English.

Birgitta L. Sjöberg is a lecturer and researcher in Classical Archaeology and

Ancient History at Uppsala University. Her research interest is focused on
aspects of gender within the concept of the Greek oikos with a special focus
on long-term changes, for example, in the transformation of the oikos from
the Early Iron Age to the Classical period. She also has a special interest in the
economic dynamics and interaction of the Aegean in the Late Bronze Age and
the transformation associated with the early Iron Age.

Lindsay Penner is a PhD candidate in the Department of Greek and Roman

Studies at the University of Calgary. Her primary areas of interest are Roman
social history and Latin epigraphy. Her PhD dissertation focuses on the
structure and everyday functioning of the households of the extended Julio-
Claudian dynasty.

Elisa Perego is completing a PhD at the Institute of Archaeology, University

College London. Her doctoral thesis focuses on the construction of personhood
in late prehistoric and early Roman Veneto. Her other research interests include
gender, alcohol consumption, ritual, funerary archaeology and Romanisation.
Her most recent publication is the chapter ‘Engendered Actions. Agency and
Ritual in Pre-Roman Veneto’ in A. Chaniotis Ritual Dynamics in the Ancient
Mediterranean: Agency, Gender, Emotion, Representation (Stuttgart: Steiner Verlag).

The volumes of The Family in Antiquity series bring together scholars inves-
tigating the ancient family across the spectrum of classical ancient history
– from ancient Greece, to the Hellenistic Mediterranean, to the Roman Empire
through to early Christian Europe. All the writers have an interest in what could
be described as the nature of the oikos in Greece and the familia of the Roman
Empire and early Medieval Europe. The terms oikos and familia as used in
antiquity are difficult to define and any translation into a European language is
quickly found to be wanting and inadequate. Moreover, the cultural production
and reproduction of these linguistic terms occurred in very different societies
across the wide chronological span of antiquity – over 1,600 years (from c.800
BC through to c.800 AD). This diversity of usage and cultural context was one
of the attractions of these terms for an investigation that was first considered
in 2008 at a meeting to discuss collaboration of research between the present
editors of the volumes. Our intention was to extend the discussion of the family
in antiquity both in terms of chronological range, but also in terms of the inter-
action of academics working on different periods of antiquity. After all, we can
all observe in our university libraries the range of publications on the Roman
Family and then find an absence of a similar scale of writing on the family in
Archaic, Classical or Hellenistic Greece; or, for that matter, books on the family
in Late Antiquity. To address this issue, we decided to hold a conference entitled
From Oikos to the Familia: Framing the Discipline for the twenty-first century
with the proviso that we would include all who wished to deliver a paper on this
subject. We met in Gothenburg over 3 days in November 2009 with 60 speakers
from across the globe, arranged, for the most part, in three parallel sessions.
Subsequently, as organizers, we were faced with the difficult choice of which
papers to select for inclusion in the volumes. There were many good papers
that simply could not be fitted in and here we wish to acknowledge the impor-
tance of these unpublished papers that were delivered at the conference. All the
papers given in Gothenburg have shaped how authors in these volumes have
thought about their shared subject – the ancient family. All the abstracts from
the conference can be found at:
In addition, we wish to express our profound gratitude to both Mark Golden
and Natalie Kampen in providing a lead in opening the conference and setting
an agenda that was explicitly self-critical and questioning of received wisdom.
xvi P r e fac e

Both have kindly agreed to comment on the papers published in these volumes
in final chapters to each volume and to share with us their views of the future
of the subject.
The volumes of The Family in Antiquity do not seek to definitively define
oikos or familia: instead they contain different perspectives to those found
published previously, either in terms of subject matter (for example osteo-
logical analysis from the Veneto and Roman Britain), or methodologies and
perspectives drawn from outside the classical disciplines (for example in the
study of demography and kinship). One of our principle aims in these volumes
is to include a sense of the excitement and vibrancy of the ideas expressed at
the conference in Gothenburg as participants met, often for the first time, and
discussed new understandings and new thoughts about a common interest in
the family in antiquity.
These volumes are not the end of this project, but one of its outputs. To
enable younger researchers undertaking doctoral research to interact, students
from the Universities of Birmingham and Gothenburg set up a website to
enable the discussion of the life course in antiquity. It is important for this next
generation of academics to be able to interact and develop contacts in ways that
were unimaginable when we ourselves were students; and, given the current
uncertainty of the future in the present economic climate, there is a need for
us to ensure that we do not lose a generation of researchers (as occurred, for
example, in the UK in the 1980s). With this in mind, an initiative has been set
up to promote the study of the family in antiquity in Swedish Higher Education
that will involve the participation of other European scholars.
The production of these volumes has consistently reminded us of the joy,
professionalism, and enthusiasm that the participants brought with them to
Gothenburg in 2009. These qualities carried this project forward to publication.
We hope readers will see these qualities in the written versions produced
by the authors in a timely fashion for each volume. Here, we need to thank
Céline Murphy for her invaluable help in editing both volumes – she has been
meticulous in her work, we the editors take responsibility for any errors. Finally,
we must acknowledge that none of this would have been possible without the
support of the following sponsors of the conference:

Professor Göran Malmstedt and the Department of Historical Studies, University

of Gothenburg for hosting the conference
The Swedish Research Council
The Swedish Foundation for the Cooperation in Research and Higher Education
The Institute of Archaeology and Antiquity and the College of Arts and Law,
University of Birmingham
The Classical Association (UK)
The Institute of Classical Studies, University of London
Stiftelsen Harald och Tonny Hagendahls Minnesfond
The Wenner Gren Foundation
P r e fac e xvii

Berg Publishers (UK)

The Society for the Study of Childhood in the Past
The University of Kent, Faculty of Humanities (for funding the invaluable
Céline Murphy)
Michael Greenwood at Continuum
The editors recognize the contribution that all the above have made to both
Mary Harlow (University of Birmingham), Lena Larsson Lovén (University
of Gothenburg), Ray Laurence (University of Kent), Agneta Strömberg
(University of Gothenburg)

Introduction: From Oikos to Familia: Looking

Ray Laurence

The study of the family or families in ancient societies has been one of the most
productive areas of research in the fields of social history and social archae-
ology. The bibliography has continued to steadily grow and recognition of
the centrality of the subject matter to the study of antiquity can be seen in the
recently published Wiley-Blackwell A Companion to Families in the Greek and
Roman Worlds edited by Beryl Rawson (2011). Her commentary on the state of
the subject and the nature of current work that makes for interesting reading,
after all for many (especially in the Anglophone world) she built up the subject
of the Roman family through the holding of conferences and a determination to
situate family in its rightful place as a central institution for understanding the
cultures of antiquity. The news of her death in late 2010 caused the editors of
this volume and those of volume II to reflect on her contribution to the subject
and to the development of the discipline of Roman history – something that is
impossible to encapsulate in words – but we all felt a sense of loss. She traced
the academic pedigree of the subject back to the 1960s but could connect recent
work from the first decade of the twenty-first century back to work from a half
century earlier.1 She had lived through a period of academic change, just as the
editors of these volumes grew up as undergraduates and then postgraduates
reading the papers delivered at the Roman Family conferences organized by
Beryl Rawson. Her publications also endeavoured to make clear how the study
of the family was progressing both in the 2011 volume and in an earlier paper
from 2003 – again looking forward to the further development of the study of
the family.
If we are to look back to the history of the study of the family in antiquity,
we can point to a trend by which the discipline expanded from existing
work undertaken primarily by legal experts (the Wiley-Blackwell companion
contains four contributions by such scholars, about 10 per cent of the volume)
to include a greater range of evidence. Nowhere is this clearer than in the use of
archaeological evidence for dwelling in the Classical past. Not surprisingly, the
Wiley-Blackwell companion opens with this subject matter, archaeology plays
its part not least to demonstrate the presence of women and children in Roman
forts; reading the chapters (over almost 200 pages: 35 per cent of the volume),
what characterizes these studies of the household is a shift away from archaeo-
logical evidence towards textual evidence for household formations. There is
2 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

quite simply no archaeological theory for the construction of the family, even
though there is much written on gender and on age (including childhood).
In Rawson’s final volume, we see that the study of the family has effectively
colonized a wider range of evidence that can include house-plans, evidence of
artefacts, and also at the same time expanded the range of sources to include
Egyptian papyri, families of freed slaves and foreign residents, military families,
and then the Christian families of Late Antiquity. As Rawson observes, far more
of the subject matter is focused on families of the Roman period than on what
we may call families of the ‘Greek period.’2 This reflects the uneven development
of the discipline, something that we wished to address in holding the conference
in Gothenburg in November 2009. Even so, we did not end up with a volume
on the Greek Family and a volume on the Roman Family. We see a transition
or a variation to the institution that today we call the ‘family’ in Europe in the
twenty-first century. This institution can also include histories of migration and
family formation, as well as, an evolving tradition of what a family is (with or
without the legal institution of marriage being involved). There can never be
and never was, a single defined version of the family, now or in antiquity – as
even the most casual reading of the Digest of Justinian reveals. Indeed as Sabine
Huebner points out, looking from Egyptian census evidence across to evidence
from Tang China and then Medieval and Early Modern Italy: the nuclear
family while prominent in most of these societies (43–55 per cent of examples)
is matched by a significant proportion of extended and multiple families
(10–17 per cent and 18–22 per cent respectively of all examples).3 It is only in
northern Europe in the late 16th and early seventeenth century that she locates
a domination of the nuclear family (78–84 per cent of examples). This contrasts
with the evidence of funerary inscriptions that leads Huebner to find that in
this medium of commemoration the relations we associate with the nuclear
family appear predominant and in line with epigraphic patterns from the Latin
west, but quite dissimilar to the pattern found in Asia Minor in antiquity.4 The
disjuncture between census data and commemorative epigraphy points to the
difficulty of reconstructing the family: a frustration that is common to other
forms of evidence from antiquity. Families are not neat and tidy, they are often
messy and complicated and vary within a culture. The evidence we have, like
that from Egypt, may indicate that even with a presence of multiple or extended
families, the biological close kin are those that are commemorated. The impulse
to reconstruct or recreate households and family structures is a strong one, but
should not be the only impulse in the study of the family (see discussion below).
Just as the history of the family has colonized types of evidence, it has also
colonized new periods: the most intriguing is perhaps the Hellenistic period, in
which as Coloru points out in this volume: the language of kingship developed
a language of family formation, something that we also find in the Wiley-
Blackwell Companion.5 Unlike in the study of the family at Rome or in Greece,
the Royal Families are subjects for analysis and can be included alongside the
study of soldier’s families in the Roman Empire or other people of a much lower
I n t r o d u c t io n : F r om Oikos t o Familia : L o o k i n g F o rwa r d ? 3

status. The relationship of how royalty articulated and/or how people discussed
the families of the elite reflects and shapes forms of marriage. Suetonius might
only be able to find two examples of marriage between uncles and their nieces,
after Claudius’ marriage to Agrippina; but nevertheless as the biographer notes
– those marriages did exist.6 There is much that might be learnt from the socio-
linguistics of royal families in all societies in antiquity, and not just that of the
Hellenistic rulers.
The contrast between this type of work on the Hellenistic family that focuses
on the public face of royalty and the stress in the study of the Roman family
on the domestic or household setting has implications for both sides of this
academic equation.7 In the development of the study of the Roman family,
we can trace an increase in the interest in the domestic or the ‘private’ over
work on the family and public life. This coincides with how female and male
practitioners tend to play on the study of the everyday or routine activities of
modern life. Female academics have been shown to study to a greater degree
activities associated with ‘the home’; whereas male academics have chosen to
turn their attention to ‘the street.’8 This crude dichotomy of interest is relevant
to the study of the family in antiquity, since it points to a historiographical
fact that the study of the Roman Family was pioneered by women and was
informed by feminist thinking from the 1960s and 1970s and lies at the heart
of ARACHNE – The Nordic Network for Women’s History and Gender Studies in
Antiquity.9 The Marriage in Antiquity conference held at the Swedish Institute
in Rome, organized by ARACHNE, featured very few male academics.10 The
greater involvement of men in the study of gender is the subject of discussion
at a one-day workshop in London at the time of writing.11 To a degree, this
image of female and male academic foci is a stereotype and is perhaps one
that should be broken. Saskia Hin, in this volume, seeks to link the social and
demographic structure of the family to that of the state. Equally Ann-Cathrin
Harders, also in this volume, relates family and kin to a wider set of social
relations and thus expands the range of discussion beyond that of the domestic
setting. What perhaps also needs to be considered is the mobility of families
(for example that of Agrippina and Germanicus), in for example the Roman
Empire, we might equally consider such roles of the elite as signs of the creation
of ‘transnational families.’12 These are first steps in a new direction that relates
the family in antiquity to other structures of ancient societies. The approach is
also reflected in numerous papers in our own volume. What we have now in
connection with the study of the family is an understanding of its nature and
variation, something that we could not have seen 10 or 20 years ago, let alone
30 to 50 years ago.
Being in this position, with decades of work behind us, it would appear
that we might shift our perspective on how we study the family and move
out of our comfort zone and consider alternative perspectives with a view to
producing a rather different reading of the family in antiquity for the twenty-
first century. I wish to set out some perspectives, mostly drawn from James
4 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

White’s and David Klein’s book Family Theories (London, 2008), that could be
said to be theoretical in origins but that have immense practical value for the
study of the family. Frequently, the socialization of children is a familiar topic
for those who have read the literature.13 It is a process we find in discussion of
childhood in antiquity that explains many actions. However, what we do not
find in the literature alongside the socialization of children in its corollary the
stabilization of adult personalities through incorporation and anchoring within
a family.14 The relevance of the latter is particularly relevant to the discussion
of freed slaves,15 but this needs to be seen within a broader understanding of
‘stabilization’ in Roman society that might include a discussion of the context
of childhood. The adjustment to these modes of discussion to consider parental
roles in families seen in the early twenty-first century are well covered in the
existing literature on fathers and daughters, the Roman mother and so on. This
obviously does not cause the discussion to shift far from our current mode of
Looking across what has been written on the family and how students in
seminars in the UK and Sweden talk about the ‘family,’ we can easily identify
how ancient historians have utilized a framework of rational choice based on
the legal powers of fathers and absence of legal power on the part of other family
members. This causes discussion to become utilitarian and seeks to explain
options to imagined family members. The choices made are rationalized with
reference to law (particularly Roman law), the seemingly less-than-rational
familial choices made by Augustus for his family members are often explained
as being unrepresentative or politically motivated in order to rationalize them.
Such explanatory frameworks converge with an approach in the social sciences
that is implicitly underpinning work on the family in antiquity.16
There has been a surprising neglect of the role of language in the creation
of symbolic meanings of the family and the language associated with it. For
example, the names of items of clothing associated with a Roman bride was a
subject that was (and is today) investigated by ancient ethnologists to establish
the meaning of the words to reveal ideas of the origins of the clothing.17 We can
see these items as symbols and the focus of explanation, but there is a wider set
of meanings to be discussed in relationship to the daily lives of family members
and their life course. The family is a structure replete with symbols that act as
reminders of past events, both personal and related to a wider social world. Yet,
the symbolic meaning and deployment of familial terms has only partially been
investigated. What we see in literary texts is a representation of the family, but
what we do not find in the current scholarship is the analysis of the deployment
of words associated with the family. The same could be said of legal texts, we
find symbols of the family in these rather than the family as a lived experience.
Obviously, the commemoration of the dead in inscriptions leads us into the
discussion of the symbols of familial relations. Yet, surprisingly little work has
yet to address how the family is represented. The exception, of course, is in the
visual representation of the family.18 However, we need to be aware that the
I n t r o d u c t io n : F r om Oikos t o Familia : L o o k i n g F o rwa r d ? 5

representation or picture associated with a familial space could shift in meaning

according to the temporal position of viewers within the life course of a family
– a point made by Natalie Kampen in this volume.
The life course approach to the family has been implicit in the study of
marriage and family formation, but has become explicit through the devel-
opment of Richard Saller’s (1994) demographic simulation of the Roman
Family.19 Mary Harlow and Ray Laurence have pulled into the study of the
family individual concepts associated with the exponential increase of interest
in the life course and ageing within social theory in the late twentieth and early
twenty-first centuries.20 What is only partially approached is a conception of
the life course of families that can plot out the potential of families to become
socially expansive or to contract, and thus to engage with wider historical
trends such as the non-biological replacement of the senatorial elite established
by Keith Hopkins and Graham Burton.21 We have seen the introduction of age
alongside gender as an organizing social principle for the individual, but have
yet to define the development of families over the temporal period of more than
a human life span.22
Much of the above will be familiar to ancient historians and Classical archae-
ologists, but where family theory diverges more dramatically from the study
of the family in antiquity is in discussion of the sociological phenomenology
of families.23 Practitioners of this strand of thinking look at how assumptions
and stereotypes of society construct the everyday life of the family. What is
highlighted is that the existence or inclusion in a particular family ensures that
everyday life is different to that experienced by society in general. This ensures
that families create particular, or of their own, explanations – these are private
and contrast with explanations that are public. It is an approach that could be
applied to discussion of specific houses or particular ethnic groups,24 but also
could be applied to the material realm to suggest that the phenomenology of a
house or tomb could embody particular familial phenomenologies. This causes
each family to become a particular cultural form and to have its own context.
The realization of this aspect of the family in cultures causes the possibility of
generalization to become far more problematic than before, and returns us to
the study of the single house and to see within a house the inscription of the
culture of its inhabitants and their predecessors.
Looking at how the practice of the study of the families of antiquity is
constructed, we find that there is a common practice of building cases based
on evidence (mostly drawn from literary and legal texts). These are deployed to
create a narrative that ‘makes sense’ or is rational. Yet, in so doing, as ancient
historians, we are perhaps missing the role of the deployment of familial terms:
father, mother, son, daughter, wife, husband, boy, girl, virgin, youth and so
on. There is a sense that these words found in texts have been placed there by
authors, who were fully aware of the power of language and the creation of
a sense of meaning that could play on emotions or could make a text simply
much more interesting. Tim Parkin has shown how the schemes to account for
6 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

the ages of man in antiquity should be seen as literary games that are clever in
their construction as representations of the human life span.25 We are perhaps
looking at the beginning of a new historiographical trend to seek a better under-
standing the role of familial language by ancient authors for the construction of
narratives that represent people and actions from the past.26
Allied to this observation with reference to texts, we might wish to suggest that
lives lived within families should be seen as associated with a system of repre-
sentations that normalize gender roles, for example through the production of
imagery associated with a male and female couple. There is, in short, a style or
aesthetics to the ancient family and to its everyday existence.27 Certainly, nearly
all literary texts engage with an aesthetically pleasing ideal of how to live a life –
something that most, it must be said, were seen not to be capable of achieving.
What we may also observe is that through objects many people remember or
even have memories jogged, when suffering from dementia, of their life at an
earlier point in time. The objects recovered by archaeologists constitute an
aesthetics of the family, not just by presence or absence, but by their shape and
form and, above all, their familiarity. These objects create a sense or aesthetics of
living that may be integrated with the human senses in the consumption of food
and drink especially.28 Just as other human senses are engaged in the production
of the family through sex for biological reproduction, but also through touch
and smell in the creation of memory. These senses and the association of objects
with the past cause us to consider the very concept of time in families. The life
span of a human being can be reported as a narrative, but at the same time is
lived at any moment with respect to past moments in the life course and an
imagination of the future based on the observation of the lives of those older
than oneself. However, it is also recognized that the narrative of life is actually
‘redrafted and replotted’ as self-identity is re-thought and the situation of the
everyday is subject to major change (for example becoming a parent or death
of a family member) that inevitably causes the view of past existence to alter or
can resist change and become set.29 Observing such a phenomenon in antiquity
is virtually impossible, but the presence of the phenomenon needs to be under-
stood and it is essential for us to accept that Cicero in later life is a quite different
person to Cicero in his thirties. Moreover, his own past self had become the
subject of a self-nostalgia associated with the suppression of Catiline in his
consulship, some 20 years prior to his death in 43 BC.
The discussion in this introduction has shifted back and forth between texts
and archaeological material, just as Rawson’s A Companion to Families in the
Greek and Roman Worlds does. It is this interplay of two disciplines that shifts
discussion from royalty and the elite in texts to the objects found in houses
used by the ordinary people (slave, dependent, and/or master) in a household
that makes the study of the family in antiquity a vibrant and varied discursive
practice. We have opportunities to discuss representations, but also instances
of how we recover the everyday. The latter has become an area much theorized
in recent years, as the ordinary or unremarkable has become an area of study
I n t r o d u c t io n : F r om Oikos t o Familia : L o o k i n g F o rwa r d ? 7

and seen to have been constituted by the objects that are encountered and are
utilized.30 This shifts us, not back to the study of everyday life in the manner of
Jerome Carcopino, but towards a new theorized understanding of the everyday
that resists generalization and points to the family as not just central to culture,
but central to the production of an aesthetics of living in antiquity. In connecting
these realms of discourse in this way, in the future, we may see a different under-
standing develop of the family. In short, the current study of the everyday or
the ordinariness of the ancient family engaged with current thinking in cultural
theory has the potential to shift the subject towards new lines of enquiry.31 In the
meantime, the papers in this volume draw in new evidence and new method-
ologies to demonstrate that the study of the family in antiquity is not ‘done and
dusted’, but instead is still in its phase of young adulthood and seeking to define
its identity through difference to what has gone before in its academic life course.

  1 Rawson, 2011b.
  2 Rawson, 2011b, pp. 8–9.
  3 Huebner, 2011, p. 79.
  4 Huebner, 2011, pp. 87–88.
  5 E.g. Ogden, 2011.
 6 Suet. Claud., l. 26.
  7 E.g. Parkin, 2011, p. 284.
  8 Laurence, 2007, features an assertion of the street as a subject of study at the expense of the
domestic and reflects this trend.
  9 See Rawson, 2011b, pp. 8–10; for Arachne see
10 Published now by Larsson Lovén and Strömberg, 2010.
11 Entitled ‘Gender in the University Classics Curriculum’, a key question was posed: ‘Can more
male students be encouraged to do “gender” (or “women”)?’. See for programme:
12 A subject addressed in part by Noy, 2011; but see Goulbourne et al., 2010.
13 E.g. Morgan, 2011.
14 White and Klein, 2008, p. 48.
15 See Mouritsen, 2011, for an approach to slave families.
16 White and Klein, 2008, pp. 65–91.
17 Hersh, 2010, provides the evidence and some discussion, but other approaches to this evidence
remain open.
18 Cohen, 2011; Huskinson, 2011 and Tulloch, 2011, provide examples.
19 The progression from implicit to explicit can be seen in a comparison of the following:
Treggiari, 1991; Saller, 1994 and Parkin, 2011.
20 Harlow and Laurence, 2002.
21 Hopkins, 1983.
22 White and Klein, 2008, pp. 121–149.
23 Ibid.,pp. 212–215.
24 E.g. Noy, 2011.
25 Parkin, 2010.
26 Treggiari, 1999, shows the way to understanding symbols and their association with emotion.
See also in this volume the linking of imagery to emotion. However, the deployment of familial
terms in texts awaits study.
8 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

27 Highmore, 2011, pp. 9–12 and elsewhere in arguing for the conjunction of the study of
aesthetics with that of the ordinary lives of people.
28 Laurence, 2009, for discussion.
29 Highmore, 2011, p. 93.
30 There is much bibliography on this subject. Highmore, 2002, captures the development of the
discipline, but is somewhat updated by Highmore, 2011. Jacobsen, 2009, contains essays on
various approaches to the study of the everyday.
31 Highmore, 2002, provides an overview of a means to theorize the everyday.

Cohen, A. (2011), ‘Picturing the Greeks’, in Rawson, B. (2011a), A Companion to
Families in the Greek and Roman Worlds, Oxford: Wiley-Blackwell, pp. 465–487.
Goulbourne, H., Reynolds, T., Solomos, J. and Zontini, E. (2010), Transnational
Families: Ethnicities, identities and Social Capital, London: Routledge.
Harlow, M. and Laurence, R. (2002), Growing Up and Growing Old in Ancient
Rome – A Life Course Approach, London: Routledge.
Hersch, K. (2010), The Roman Wedding: Ritual and Meaning in Antiquity,
Cambridge: Cambridge University Press.
Highmore, B. (2002), Everyday Life and Cultural Theory: an Introduction,
London: Routledge.
—(2011), Ordinary Lives: Studies in the Everyday, London: Routledge.
Hopkins, K. (1983), Death and Renewal, Sociological Studies in Roman History,
Vol. 2, Cambridge: Cambridge University Press.
Huebner, S. R. (2011), ‘Household Composition in the Ancient Mediterranean
– What do we really know?’, in Rawson, B. (2011a), A Companion to Families
in the Greek and Roman Worlds, Oxford: Wiley-Blackwell, pp. 73–91.
Huskinson, J. (2011), ‘Picturing the Roman Family’, in Rawson, B. (2011a),
A Companion to Families in the Greek and Roman Worlds, Oxford: Wiley-
Blackwell, pp. 521–541.
Jacobsen, M. H. (2009), Encountering the Everyday: An Introduction to the
Sociologies of the Unnoticed, Basingstoke: Palgrave Macmillan.
Larsson Lovén, L. and Strömberg, A. (2010), Ancient Marriage in Myth and
Reality, Newcastle-upon-Tyne: Cambridge Scholars Publishers.
Laurence, R. (2007), Roman Pompeii: Space and Society, (2nd edn), London:
—(2009), Roman Passions: A History of Pleasure in Imperial Rome, London:
Morgan, T. (2011), ‘Ethos: The Socialization of Children in Education and
Beyond’, in Rawson, B. (2011b), A Companion to Families in the Greek and
Roman Worlds, Oxford: Wiley-Blackwell, pp. 504–520.
Mouritsen, H. (2011), ‘The Families of Roman Slaves and Freedmen’, in Rawson,
B. (2011a), A Companion to Families in the Greek and Roman Worlds, Oxford:
Wiley-Blackwell, pp. 129–144.
I n t r o d u c t io n : F r om Oikos t o Familia : L o o k i n g F o rwa r d ? 9

Noy, D. (2011), ‘Foreign Families in Roman Italy’, in Rawson, (2011a), A

Companion to Families in the Greek and Roman Worlds, Oxford: Wiley-
Blackwell, pp. 145–160.
Ogden, D. (2011), ‘The Royal Families of Argead Macedon and the Hellenistic
World’, in Rawson, B. (2011a), A Companion to Families in the Greek and
Roman Worlds, Oxford: Wiley-Blackwell, pp. 92–107.
Parkin, T. (2010), ‘The Life Cycle’, in Harlow, M. and Laurence, R. (eds), A
Cultural History of Childhood and Family, Vol. 1: Antiquity, Oxford: Berg.
—(2011), ‘The Roman Life Course and the Family’, in Rawson, B. (2011a),
A Companion to Families in the Greek and Roman Worlds, Oxford: Wiley-
Blackwell, pp. 276–290.
Rawson, B. (2011a), A Companion to Families in the Greek and Roman Worlds,
Oxford: Wiley-Blackwell.
—(2011b), ‘Introduction: Families in the Greek and Roman Worlds’, in Rawson,
B. (2011a), A Companion to Families in the Greek and Roman Worlds, Oxford:
Wiley-Blackwell, pp. 1–12.
Saller, R. P. (1994), Patriarchy, Property and Death in the Roman Family,
Cambridge: Cambridge University Press.
Treggiari, S. (1991), Roman Marriage, Oxford: Oxford University Press.
—(1999), ‘The Upper-Class House as a Symbol and Focus of Emotion in Cicero’,
JRA, 12, 33–56.
Tulloch, J. H. (2011), ‘Devotional Visuality in Family Funerary Monuments in
the Roman World’, in Rawson, B. (2011a), A Companion to Families in the
Greek and Roman Worlds, Oxford: Wiley-Blackwell, pp. 542–563.
White, J. M. and Klein, D. M. (2008), Family Theories, (3rd edn), London: Sage

Beyond Oikos and Domus: Modern Kinship Studies

and the Ancient Family*
Ann-Cathrin Harders

Michael Corleone: Fredo, you’re my older brother, and I love you.

But don’t ever take sides with anyone against the Family again. Ever.
(The Godfather: Part I)

The End of Kinship?

Conflict, love, and betrayal within the family form but a few of the classic
themes of Hollywood cinema. Yet not every protagonist has such strong views
on family closeness like Al Pacino’s Michael Corleone. Francis Ford Coppola’s
The Godfather saga, based on Mario Puzo’s novels, introduced a new concept of
family to the wider American audience: the Sicilian Family with capital F which
combined business interests with family values – which had fatal consequences.
In the second part of The Godfather, Michael holds true to his word and has
his traitorous elder brother and partner Fredo killed, a deed he later regrets the
most.1 There just seems to be no escape from family ties and obligations – at
least until now. ‘What’s a little family between friends?’ is the smart tagline of
the independent film Framily released in fall 2010. Young filmmaker Joachim
Hedén’s second work deals with the lives of six 30-something friends who decide
to reject family relations and obligations, sticking to friends instead. They also
form a residential community in an effort to solve their relationship troubles.2
Hedén focuses on a new social concept which emerged in the United States at
the beginning of the third millennium and recently has also found its way into
the United Kingdom. Although Hedén’s protagonists seem to experience that
‘framily’ might not be an easy solution to the forming of a supportive network,
earlier TV-productions based on this concept, like the international acclaimed
NBC series Friends (1994–2004), hint at its success and the changing views on
family and friends especially within modern urban life.3 Whereas the family is
perceived as a group of persons based on given fact and to whom individuals
are inextricably and permanently linked by pure coincidence of nature, good
friends are chosen. They are the persons most trusted, they are called for
support and are spent most time with so that it stood to reason not if, but when
friends would be the new family, a.k.a. the ‘framily’.4
B eyo n d oikos a n d domus : Mo d e r n K i n sh i p S t u d i e s 11

Yet, although the ‘framily’ distinguished itself clearly from the family in ways
of empathy and choice, it is interesting to notice that the concept itself relies
heavily on modern Euro-American notions of the traditional family, of its roles
and its social and emotional functions – starting with the denomination itself:
a blend of ‘friends’ and ‘family’. Furthermore, while the ‘framily’ denounces the
idea of the family as the core unit of social networks in which an individual is
embedded and integrated into society, it also at the same time strengthens and
confirms western assumptions on the given nature of family relations.
In his classic study on American Kinship: A Cultural Account, first published
in 1968, David M. Schneider challenged these traditional views on family and
kinship. Focusing on American kinship, Schneider pointed out that the impor-
tance given to ‘natural’ aspects of kinship – consanguinity and procreation – has
to be seen as a Euro-American peculiarity which cannot be transferred to other
societies and cultures. He argued that sexual reproduction has to be understood
as a core symbol of kinship in the American system which was defined by two
dominant orders, that of nature, or substance, and that of law, or code.5 The
sexual union of two unrelated partners in marriage and the resulting children
from this union provided the symbolic link between these two orders, the social
and the natural. According to Schneider, the idiom of nature is crucial to the
American understanding of kinship: kinship was and still is considered within
this cultural-symbolic framework as the one institution most closely linked to
the natural; while kinship in its different cultural manifestations is supposed
‘to modify nature, it cannot transcend it’.6 As kinship is thus seen as ‘the after-
effect of the natural facts of procreation’, a pre-social relationship untainted by
cultural impact and a universal given category, it may claim a privileged and
special status among human social relations.7
As Schneider placed nature and biology within the symbolic framework
of a certain society and indicated an ethnocentric bias concerning his fellow-
colleagues’ work on kinship, he stripped anthropologists of their favorite
and main topic. Since ‘the Doctrine of the Genealogical Unity of Mankind’,
as Schneider put it, had been stripped of its ‘natural’ basis, kinship could no
longer be analyzed and compared as a universal institution and as a given fact,
and so the end of kinship studies was declared.8 Yet, recently new perspectives
on kinship are discussed: they dismiss the idea of a cross culturally applicable
definition of family and kinship and try to accommodate both Schneider’s
challenge and differing cultural understandings of kinship since genealogy still
plays an important part when it comes to social relations – although kinship
is no longer seen as a cultural elaboration of biological facts. The shifting and
flexible understandings of kinship in the modern age are quite obvious due to
new technologies and globalization. Modern anthropologists are confronted
with new reproductive technologies, genetic research and genetic screening,
virtual worlds and identities, gay and lesbian families, international adoption
and donation of blood, organs, and semen. These legal, medical, technical, and
social possibilities challenge traditional views on ‘blood ties’ and family and
12 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

demand new perspectives and reconfigurations of kinship studies and kinship

Modern anthropology has responded to this modern world and has been
very productive in exploring these new understandings of kinship. It has
widened the frame of topics to social constellations which were not considered
as kin before.9 Though the ancient world lacked these technologies and the
medical possibilities attached to them, the reconfiguration of kinship studies
is, as I will argue in the following, not only productive within the ethnological
and anthropological field concerning the modern age, but can also shed new
light and broaden the perspective on the study of family and kinship groups
in ancient Greco-Roman societies. Also in antiquity, kinship had not been a
‘natural’ fact, but a culturally defined relation, and modern theories challenge
ancient historians to contemplate once more the formation and the character of
the social group they are studying when they focus on oikos and domus and the
social relations attached to it.

Family – oikos – domus – familia: What Are the

Objects of Family and Kinship Studies?
David Schneider forced modern anthropologists to reflect on their own cultural
bias with reference to their own studies of family and kinship, and to accept
that absolutely nothing is either universally given or absolute when it comes to
human social organizations. The impact of ‘nature’ on human relations was thus
challenged and placed within the cultural framework of one specific society.
However, being truthful to Schneider’s critique and declaring the end of kinship
would rob anthropologists of their main topic of investigation, though, making
some amends to Schneider’s thesis, traditionalists, such as Robert Parkin,
declare the existence of kinship as a ‘privileged cultural order over biological
universals and continuous human reproduction through birth’.10 Also Maurice
Godelier refers to marriage and birth as fundamental of kinship relations.11
Sarah Franklin, Sylvia Yanagisako, and Jane Collier on the other hand have
taken up Schneider’s gauntlet and have discussed reconfigurations of kinship
theory in which kinship is accepted as a contested analytic category; they argue
for a more flexible and open approach in contrast to biological approaches to
kinship. The impact of Gender Studies has further strengthened this view of
kinship since the relation between the natural and the social is discussed in a
similar way.12 The ‘naturalization’ of kinship and gender is seen as a means to
taking ‘difference’ for granted and treating it as a pre-social fact’13 thus legiti-
mizing hierarchies between the sexes or within a kinship group.
In order to move away from an assumed analytical opposition between the
biological and the social, Janet Carsten has proposed not to speak of kinship,
but of ‘cultures of relatedness’ instead.14 This approach signals openness to
B eyo n d oikos a n d domus : Mo d e r n K i n sh i p S t u d i e s 13

indigenous idioms that do not necessarily rely on genealogical proximity or

on any other assumed definitions when it comes to explore the functioning of
domestic life and the rearing and socialization of children. Carsten recognizes
the broadness of the term and that social relations still have to be distinguished,
but she also emphasizes the analytical opportunity to grasp relations that are
– from our modern western point of view – not part of kinship relations, but
were ascribed to other social systems.15 This approach might also provide a
more dynamic understanding of what being related provides for a particular
people – in a particular culture by not focusing on what we understand as the
‘natural’ character of kinship. Carsten does not necessarily intend to abandon
biology, but she suggests to ‘subject[ . . . ] its uses in different cultures to closer
scrutiny’ and that we should be open to different ways of connectedness via e.g.
substance, feeding, living together, procreation, or emotion.16
By advocating a contextual approach, Carsten, Franklin and others advocate
a wider frame of kinship studies and ask questions about how kinship is
created and maintained beyond the Eurocentric notions of biology and nature.
Leaving universal definitions and cross cultural comparisons aside, contextual
approaches try to describe and to understand kinship relations within the
boundaries of one specific culture first via a series of questions. What are the
bodily substances that transmit relatedness? What are the practices of kinship
and in which contexts do they take place? For what and whose benefits are
kinship relations cultivated and performed?17 By focusing on the cultural
meaning of kinship, modern anthropology has also challenged the traditional
relegation of kinship studies to specific types or domains of society, namely
primitive, rural, and small-scale societies which are described as kinship-
focused since kinship is classified as the major institution.
This localization of kinship studies had led to a specific ‘narrative of kinship’
already developed by the founding father of kinship studies, Lewis Henry Morgan.
Though kinship studies have changed since Morgan, his fundamental evolutionary
approach still dominates anthropological as well as historical analysis. Kinship
is thereby held to be the major institution of primitive, tribal societies which
dominates other social structures. Social progress and the development of the state
are seen in opposition to kinship relations; evolution is therefore only possible by
overcoming kinship bonds so kinship-focused societies turn into state-focused
societies. Anthropologists as well as historians assumed an antagonism between
state organization and kinship, and formal institutions such as governments had
to take supremacy. Especially the nineteenth century in Europe was seen as the
climax of this development; the demise of kinship structures was paralleled with
the rise of the nuclear family, which was placed in a non-political domestic sphere,
and the rise of the state. This view has been challenged recently and the nineteenth
century is rather seen as a ‘kinship hot’ historical period; kinship relations are
thereby retrieved from the ‘private’ sphere and seen as an important social factor.18
In order to study the ideas people had and expressed about cultures of relat-
edness, one usually looks at genealogy which plays an important part, but is
14 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

not an exclusive category to describe relations. In Greek and Roman societies,

an idiom can be grasped in which relatedness expressed in genealogical terms
is reflected and understood as a special social system. It is distinguished from
relations like friendship and patronage and focuses primarily on genealogical
proximity, but does not exclusively rely on descent and marriage. Ancient
historians are coping with a variety of concepts like anchisteia and syngeneia,
propinquitas and philia, as well as agnates and cognates to name but a few. Oikos
and domus involve space as a category to define groups, but the concept of
household groups cannot be seen as static and as easy to grasp as the walls that
define the house as a building.19 Usually modern historians attempt to organize
and to hierarchize the ancient concepts starting with the nuclear family as the
core unit.20 Yet, neither in Greek nor in Latin does a term exist that describes
solely and exclusively the triad of father, mother and child. Bearing Schneider’s
critique in mind, I propose to take the ancient concepts as they are and to put
the nuclear family which is so important for our understanding of kinship to
one side.
It is a truism that human beings are born into this world and that they are not
able to live on their own for a long time, but are dependent on others. Important
tasks such as feeding and nurturing and the whole process of gendering and
developing a social station and a place within the respective societies takes
place within a social framework very often, but not necessarily, described as the
kinship group. In Classical Athens and Republican Rome this is taking place
for the most time within a certain space: oikos and domus.21 In order to cope
with the different personal composition of a genealogically defined family and
a spatially defined household group, historians, archaeologists as well as anthro-
pologists have tried to separate both groups. This artificial differentiation,
however, often does not reflect social practice. A lot of these activities, which
can be classified as domestic tasks, were dealt with by persons who were not
related to each other in genealogical terms, for instance wet nurses and other
slaves. Especially vernae, home-born slaves in Roman society, who did not only
serve economic needs, but also were tied by emotional functions and ‘quasi-
familial’ bonds are elusive members of the household and the family.22 We also
encounter persons who were only for a certain time part of the household, such
as foster children or teachers.
Furthermore, persons usually ascribed to the family did not reside in the
same house: daughters moved to another household by entering marriage, but
were still considered a part of the family. In the case of adoption, usually adult
males became part of the agnatic family unit legally, but this did not necessarily
mean that they had to live with their adoptive fathers under the same roof and
that they cut all bonds to their native family.23 Stepmothers and stepfathers
might have been added to the household group – as were children from former
unions as well as orphaned nieces and nephews, and so we confront enormous
ancient patchwork families plus a retinue of slaves, servants, teachers and so
forth. Although ancient sources do not provide the information that we wish
B eyo n d oikos a n d domus : Mo d e r n K i n sh i p S t u d i e s 15

for, scholars are confronted with the need to exactly describe the composition,
organization, activities and the cultural meaning of oikos and domus, since
there is not necessarily a distinction between spatial and genealogical proximity
in social practice. Often the nuclear family is taken for granted as being the
concept most ancient families were modeled on in contrast to the concept of
the extended family. Dale B. Martin has spoken against this dichotomy and has
warned us not to apply modern concepts in an analysis of ancient societies. In
his discussion of the thesis of Richard P. Saller and Brent D. Shaw who place
the nuclear family at the heart of Roman society,24 Martin proposes to assume
a far greater variation in domestic personal relations and to rather imagine ‘a
nuclear centre with a spectrum of relations’.25 Kinship relations are not fixed but
manifest themselves in everyday action and thus generate a group over time;
hence, a diachronic analysis of families might be the answer. One concept that
embraces the fluidity and the dynamics of the household over time has been
the life-course approach as it has been presented by Mary Harlow and Ray
Laurence on the Roman household; a similar project concerning the Athenian
oikos has not been undertaken yet.26 As oikos and domus cannot be considered
as static concepts and at the same time transcend the concept of family, I
suggest widening the perspective to how households are interrelated and focus
on super-domestic relations beyond oikos and domus.

Kinship Happens
The study of family and kinship has often been separated by medievalists and
modern historians,27 as well as by ancient historians: whereas the concept of
‘family’ has been studied with a focus on domestic matters such as marriage,
systems of inheritance and the father’s authority and as an object of legal
definition, ancient kinship studies have been less institutionalized and not
so much focused on the domestic domain.28 Kinship serves as a system to
classify social relations by using, at least in Greece and Rome, a genealogical
framework. The differentiation between ‘natural’ kinship and ‘fictive’ kinship,
however, is misleading, since kinship is not to be taken as a given biological fact,
but as a social construct which mediates relatedness. Moreover, kinship is not
static, but, as Bernhard Jussen has put it, ‘kinship happens’.29 Therefore, what
we are able to grasp and analyse are cultural manifestations of this relationship
in social practice, discourse, and terminology;30 and we have to focus on the
settings in which kinship is happening and how it is maintained.
The manifestations and cultivation of kinship relations differ culturally as
well as socially. In Roman society, the household group contained a variety of
persons who were not related genealogically to each other. Nonetheless, relatives
were seen as a special group of persons being related by descent, marriage and/
or law respectively who shared a specific love and duty for each other: pietas.31
Yet, different social functions were connected with different principles of
16 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

forming kinship groups as agnates and cognates.32 Where Roman law privileged
the agnatio, especially concerning laws of inheritance, and relied to the agnatic
principle when it came to define the group of persons who were in the power of
the paterfamilias including adoptees and wives in manu, it also recognized the
maternal relatives, the cognatio. Agnates as well as cognates were not allowed
to intermarry within the sixth degree, from the second century on within the
fourth degree, as such a union was considered as incestuous. Moreover, this
legally defined set of non-marriageable persons was also distinguished by social
practice: first, the kinsmen whom one was not allowed to marry were referred
to by using a unique descriptive terminology. According to anthropological
standards, Latin kin terms follow the Sudanese pattern in which each possible
relationship receives a separate kin term. Concerning Roman rules of marriage,
this meant that within the kin group one could not marry whom one could
refer to with a specific term. Secondly, the members of this non-marriageable
kin group were distinguished by a daily kiss. The ius osculi provided every man
with the right to be kissed by his female relatives on the first occasion they met
every day. In contrast, kissing between married couples in public was censured.
Therefore, the kiss between relatives must be seen as an explicitly non-sexual
ritual to identify the members of a different sex whom a Roman was not allowed
to marry.33
In Roman religion, different rituals and festivals were more relevant to
particular groups: by participation, the worshippers not only demonstrated
their pietas to the gods and spirits, but also their group identity as members of
a social entity. At the centre of religious activity in the house stood the pater-
familias, but depending on the occasion, he dealt with different persons. There
were cults that the whole domestic unit participated in, such as the cult of the
Lares and the Penates. Furthermore, the genius of the paterfamilias was not only
worshipped by his children, but by his slaves, freedman and clients. Yet, some
religious activities were confined to agnate or cognate kin, such as remembering
the dead at the Parentalia which took place at the end of February. A sacrifice
was made to the agnatic divi parentum in order to guarantee their benevo-
lence.34 Cognate relations were also cultivated within Roman religion: during
the festival of the Mater Matuta Roman women were supposed to pray for
their sisters’ children thereby ritually demonstrating their concern for children
beyond their agnatio and their domus. The cognatio also met every year at the
festival of the Cara Cognatio or Caristeia to discuss family matters and settle
differences thereby constituting itself as a group.35
These are but a few examples in which different principles of kinship can
be demonstrated: agnatio and cognatio, co-residence and economic symbiosis.
They structured kinship and domestic grouping in Roman society and show
how relatedness was performed and cultivated. Instead of opting for a hierar-
chical organization of these groups, it is more rewarding to look at the different
social functions that were related to them. The Roman house, the domus,
provided the place for production, consumption as well as nurturing and
B eyo n d oikos a n d domus : Mo d e r n K i n sh i p S t u d i e s 17

rearing of children and care for the elderly. Among the aristocracy, the domus
was also the place for communication and thus an important factor for social
integration: Aristocrats met and mingled at the convivium, whereas in the
morning, the patronus received his clients at the salutatio. Thus symmetrical
and asymmetrical relations were cultivated in the domus which were of utmost
importance for the res publica Romana.36
Furthermore, Jochen Martin has argued that those aspects concerning the
agnatic familia and the power of the paterfamilias are not to be taken as ‘private’
aspects relegated to domestic life. Instead they are essential to the political and
social organization of the res publica Romana – especially the extensive powers
of the paterfamilias, his ius vitae necisque, have to be paralleled to the magis-
trates’ potestas. Yet, its impact on daily life must be considered as rather small as
the authority of the father was embedded: high mortality rates as well as social
norms set limits to the cases in which a father’s power over life and death was
supposed to be applied. A iusta causa was recognized in case a filius familias
turned against the res publica; a paterfamilias was then entitled to kill him. The
ius vitae necisque thus demonstrates how family and community were interre-
lated: the Roman res publica did not exercise a monopoly on violence, but relied
on the patres to defend its interests. The patres were therefore part of the Roman
political system and acted as magistrates in their own house.37
Nonetheless, the Roman family cannot be described solely by the concept of
the agnatic familia. As I have argued in depth elsewhere, the cultivation of large
cognate kin groups transcended the domestic domain and especially served the
needs of the Roman aristocracy in the political sphere.38 Cognatio and adfinitas
provided on a macro-level a large super-domestic network in which every
aristocrat was integrated. However, these kin relations were not used as a rather
static political instrument as in family factions ruling the Roman Senate – as
envisioned by Ronald Syme, H. H. Scullard and others.39 The constant creation
and cultivation of large kinship groups into which only social equals were
integrated by marriage or adoption served to constitute the Roman aristocracy as
a peer group and provided the necessary cohesion of a group that was constantly
challenged by the strains of aristocratic competition for fame, honor and rank.40
Moreover, due to its cohesive and integrative power, kinship provided a small,
regimentsfähige group of persons with the prerequisites to rule over the Roman
plebs urbana, Italy, and the Mediterranean – even though, or maybe especially
because it was not rooted in the political institutions, but underlying them.41
This example also demonstrates that kinship and society may not be treated as
discrete domains; although the kinship group performed important domestic
tasks, it cannot be reduced to a non-economical, non-political domain only
concerned with reproduction, nurture and emotional needs, but it also shaped
other domains.42
In Greece, the interdependence between the domestic domain and the
political domain must be seen as being quite different from the Roman
situation. The household was, as in Rome, seen as one of the most important
18 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

social groupings. But whereas in Rome the domains of family and society were
overlapping and the aristocratic domus was an important place for communi-
cation and social integration, the Greek oikos was seen as a sphere separated
from the polis. Although important social functions depended on the oikoi
– e.g. in Athens the status as citizen depended on the legitimate marriage of
the parents; the oikoi provided liturgies – the house was not recognized as a
legitimate place of social and political action.43
Whereas Cicero describes how the domus are interrelated by kinship ties and
constituted a social network which forms the origin of the state (seminarium rei
publicae), in the Greek version of Aristotle on which Cicero obviously draws,
kinship is not mentioned as an intermediate between family, oikos and polis.44
This very different appraisal of the integrative effects of kinship can be traced
back to the Archaic poet Hesiod. He describes how the oikoi are not integrated
into larger kinship groups, but propagated the stronger ties of neighbourhood as
an assembly of equal oikoi. He advises the stronger loyalty among neighbours:
‘If misfortune strikes your house, neighbours will come in their bedclothes;
kinsmen will dress up’.45 Hesiod proposes group formation by locality rather than
by genealogical proximity; although the villagers would necessarily be related to
each other due to local endogamy, Hesiod does not address and contemplate the
community as a large kinship group, but as neighbours (e.g. Erga 700). Thus in
Archaic Greece another method of integrating individuals into a social group
was developed. This stronger emphasis on the oikos, neighbourhood and villages,
as proposed by Winfried Schmitz,46 as well as the subordination of larger kinship
groups has to be seen in interrelation with the political domain. The loser of
this development was the Greek aristocracy. Whereas in Rome, the aristocracy
was strengthened as a group by kinship relations, the Greek focus on rather
small oikoi provided a more egalitarian environment and weakened aristocratic
claims on predominance since Greek aristocracy did not form a cohesive and
thus assertive group.47 In the long run, these different cultures of relatedness
also smoothed the way for a democratic system since aristocratic oikoi were not
interrelated via adfinitas and amicitia like in Rome and thus gained social and
political importance, but were treated just as any other fellow-citizen’s oikos.
The Greek aristocrats thus just missed their chances to rule conjointly over the
polis.48 The different emphasis on kinship in Greece and Rome and its interrela-
tions with the political domain thereby helps to understand and exemplify the
fundamental basis of two models of ancient political systems.

Returning to Domus and Oikos

The ancient family, oikos and domus as social units and kinship groups are
important and fascinating, yet challenging topics that represent differing, yet
sometimes overlapping and fluid concepts. Especially when it comes to family
and kinship, scholars have to reflect their own family experiences and refrain
B eyo n d oikos a n d domus : Mo d e r n K i n sh i p S t u d i e s 19

from taking modern Euro-American assumptions on the natural character

of kinship for granted. Kinship is not a fact given by nature, but a cultural
construction which serves to classify individuals and create a social network
in manifold manifestations. This may be grounded in genealogical proximity,
but it does not necessarily need to be so. Kinship might be rather openly
described as a culture of relatedness, and one has to ask what practices make
people related and what does this relation do to them. This also leads to look
at household formations in a different way since these social units cannot be
described by ‘given’ and rather exclusive definitions as a group of persons living
constantly under ‘one roof ’ or sharing ‘one blood’. Both terms, the Greek oikos
and the Latin domus, refuse to specify the strategies of affiliation to this group.
As oikos and domus do not exist as a given and discrete group, super-
domestic kinship relations also do not exist within a discrete spatial entity or
social domain, apart from religion, economy and politics. The interrelation
between these domains, the way they are influencing, shaping and depending
on each other are topics yet to be explored. Within the realms of gender studies,
the dependence between the construction of gender difference within the
domestic sphere and the legal discourse on the rights of males and females has
been already discussed,49 but it also might be instructive to look at interrelations
between the economic and religious domain. It should be an aim of family and
kinship studies to break these boundaries and overcome the artificial antag-
onism of the ‘public’ domain of politics and the ‘private’ domain of the house
and the family. The equation of ‘house’ as ‘private’ as ‘non-political’ cannot be
upheld any longer.50 Furthermore, kinship relations are dynamic. Since nothing
is ‘given’ as a fact, relations have to be performed to stay intact. This also means
that kinship is not necessarily a permanent bond. The ruptures and dislocation
of kinship relations are therefore also a topic of kinship studies just as the
constitution and the cultivation of these relations.
Coming back to the statement that kinship is not given by nature, one must
also question the differentiation between ‘real’ kinship and ‘fictive’ kinship, or
kinship as a metaphor, and take social groups that define themselves as kin
for as ‘real’ as our ‘traditional’ families. It might also be of interest to analyse
how these groups, as for example, early Christians as brothers and sisters in
Christ, were coping with their pagan partners and co-residents. The analysis of
how social groups interrelate who both use the same category to create relat-
edness would further our understanding on the integrative impact of professed
genealogical closeness. We might also focus our interest on the opposite, like
residential units that are composed of persons that are not genealogically
linked – like orphans, temple communities or convents, and army barracks –
and which also do not come to mind when we speak of household groups. By
putting traditional views on family, kinship and household aside and conceiving
them as a diverse and complex culture of relatedness instead of a fact given by
nature or residence, we have to go beyond the oikos and the domus in order to
return to them and have a fresh perspective.
20 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

  * I would like to thank the editors for their highly useful comments and especially for improving
my English; I also thank Thea Fiegenbaum (Bielefeld) and Matthias Haake (Münster) for their
corrections and suggestions.
  1 On family concepts in The Godfather, see Cawalti, 1975, pp. 327–328 and Janosik, 1999, pp.
  2 Joachim Hedén’s movie was released in Sweden on 24 September 2010; see the entry on IMDB.
com: (Accessed 15 November 2010).
  3 See the entry on (Accessed 15 November
  4 Cf. Peletz, 1995, pp. 364–365 on the new move from biology to choice concerning the constitu-
tions of families.
 5 Cf. Schneider, 1980 and Schneider, 1984. On Schneider, see also Carsten, 2000, pp. 6–7;
Franklin and McKinnon, 2001, pp. 2–3; Yanagisako, 2007, pp. 33–37; Varto, 2010, p. 87.
  6 Overing, 2001, p. 8100 on Schneider.
  7 Cf. Franklin and McKinnon, 2001, p. 2.
  8 Cf. Schneider, 1984, pp. 165–177, esp. 174.
  9 Cf. the contributions in Franklin and McKinnon, 2001 and Carsten, 2000; see also Peletz, 1995.
10 Parkin, 1997, p. 3. According to Parkin, societies who are structured mainly or even solely by
kinship are of specific interest to anthropologists; on this specific ‘narrative of kinship’, see below.
11 Godelier, 1998, p. 386.
12 Cf. Godelier, Trautmann and Tjon Sie Fat, 1998, p. 4; Carsten, 2000, pp. 12–3; Overing, 2001,
pp. 8102–8103; Yanagiasko, 2007, p. 37.
13 Collier and Yanagisako, 1987, p. 29; see also Franklin and McKinnon, 2001, p. 4; Varto, 2010,
p. 88.
14 Carsten, 2000, p. 4.
15 Carsten, 2000, p. 5; cf. Varto, 2010, p. 88.
16 Carsten, 2000, pp. 33–34.
17 Cf. Yanagisako, 2007, pp. 35, 44. On the transformations of blood as the main substance
to construct kinship in Europe, see the forthcoming volume Blood and Kinship: Matter for
Metaphor from Ancient Rome to the Present, edited by C. Johnson, B. Jussen, D. W. Sabean and
S. Teuscher.
18 Cf. Overing, 2001, p. 8099; Sabean and Teuscher, 2007, pp. 4–23; Varto, 2010, pp. 92–94. On
a re-evaluation of kinship in Europe, see e.g. Lipp, 2006, and the contributions in Sabean,
Teuscher and Mathieu, 2007. See also Gestrich, 2010, pp. 68–69 in his research review on
kinship in nineteenth-century Germany. On this ‘narrative of kinship’ and the methodological
problems attached to it, see Varto 2010, pp. 84–7.
19 See Nevett, 2010, p. 5, who warns against using household and family as interchangeable
20 There are ancient parallels: Aristotle and Cicero (drawing on Aristotle) suggest in their philo-
sophical work a typology of kinship relations starting with the union of husband and wife
(Arist. Pol. 1252a; Cic. Off. 1.53–54). Modern scholarship focuses on the marriage union and
the nuclear family – the larger kinship group is usually (if at all) mentioned last; see e.g. Lacey,
1968; Dixon, 1992; Patterson, 1998; Krause, 2003; Schmitz, 2007.
21 On the Greek oikos, see Schmitz, 2007, p. 1 and Nevett, 1999, pp. 4–20 on the archaeological
approach; on the aspects of socialization in Roman society, see Harders, 2010b, with further
22 See Rawson, 2010b, on vernae. Vernae as well as free born children and foster children could all
be described by the term delicium or delicatus which demonstrates the fluidity of the concept
of the Roman family; cf. Laes, 2003. Nevett, 2010, pp. 148–149 argues for a dialogue between
archaeologists’ discussion of ‘households’ and historians’ discussion of ‘families’.
23 See, for example, the case of P. Scipio Aemilianus who was still educated by and spent most of
his time with his biological father L. Aemilius Paullus, nothing is known about the relation to his
adoptive father; see Harders, 2008, pp. 108–118. On adoption in Roman society, see Kunst, 2005.
B eyo n d oikos a n d domus : Mo d e r n K i n sh i p S t u d i e s 21

24 Saller and Shaw, 1983; see also Saller, 1994, p. 96.

25 Martin, 1996, p. 58; Cf. Rawson, 2010a, p. 610 who emphasizes the variety of Ancient family
formations; accordingly, in the Blackwell Companion to Families in the Greek and Roman
Worlds, edited by Beryl Rawson, its title stresses the pluralism of family formations. Yet, of the
contributions only the article by Jérôme Wilgaux touches the super-domestic kinship group,
the rest concentrates rather conventionally on family and household groups.
26 See Harlow and Laurence, 2002, pp. 1–3; cf. also Hareven, 2001, p. 5291, on the life-course
approach. See also the contributions in Harlow and Laurence, 2007, on age and ageing.
27 Cf. Goetz, 2009, and Jussen, 2009, on studies of family and kinship in Medieval Europe; see also
Gestrich, 2010, p. 69.
28 Cf. Martin, 1997 (with a résumé on French and German studies on the Roman family),
Harders, 2008, pp. 10–14.
29 Jussen, 1999, p. 1596 (‘Verwandtschaft ereignet sich’). See also Overing, 2001, p. 8104.
30 On the study of kinship terminology as a means to analyze the rules of a kinship system, see
Godelier, Trautmann and Tjon Sie Fat, 1998, pp. 4–5; Godelier, 1998, pp. 387–394.
31 See e.g. Cic. Planc. 80; Inv. rhet. 2.22.65. Cf. Koch, 1941, pp. 1221–1222.
32 The social functions of the Roman clan, the gens, remain obscure due to lack of sources from
Archaic Rome; during the middle and late republic, members of the gens were still recognized
as intestate heirs and had minor religious functions; on the gens in Rome and the gens as object
of modern scholarship, see Smith, 2006.
33 On Latin kin terms, see Fest., Lindsay, p. 379; Dig. 38.10.1–10; Inst. Iust. 3.6.1–12. Furthermore,
a woman addressed those males whom she was not allowed to marry as frater; see Fest.,
Lindsay, p. 379: C. Aelius Gallus frg. 24, Funaioli; cf. Bettini, 1992, pp. 157–163. Ius osculi:
Polyb. 6.11a.4 = Ath. 10.440e; Cato frg. 412 = Pliny HN 14.90.2; Gell. NA; Plut. Mor.
265D; cf. Bettini, 1992, pp. 163–170. Cato the Elder famously censured a married couple
kissing in public; see Plut. Vit. Cat. Mai. 17. 7; cf. Bettini, 1992, pp. 167–168. See Harders, 2008,
pp. 23–25; Harders, 2010a, pp. 34–37.
34 Schultz, 2006, pp. 123–131; Martin, 2009a, p. 315. On the Parentalia, see Ov. Fast. 2.533–570;
Fest., Lindsay, p. 260; cf. Scullard, 1981, pp. 74–75; Baudy, 2000. On the gens as a sacrificial unit
for the sacra gentilicia, see Smith, 2006, pp. 44–48.
35 On the Matralia see Livy 5.19.6; Ov. Fast. 6.481–482; 6.551–558; cf. Scullard, 1981, pp. 150–151;
Baudy, 1999; on the Cara Cognatio see Ov. Fast. 2.617–638; see also Scullard, 1981, pp. 74–76.
36 On the Roman domus as place of communication, see Rilinger, 1997; on the convivium, see
Stein-Hölkeskamp, 2005; on the salutatio, see Goldbeck, 2010. See also Nevett, 2010, pp.
142–143 in general.
37 Martin, 2009c, pp. 368–373 and Thomas, 1984.
38 See Harders, 2008, pp. 31–59, 318–321 and Harders, 2010a.
39 Cf. Harders, 2008, p. 51.
40 On social equality as a prerequisite to marry see Treggiari, 1991, pp. 89–94; Harders, 2008, pp.
41 Harders, 2008, pp. 314–319. On the interdependence of kinship, house and state, see Martin,
2009b, pp. 335–343.
42 On the interdependence of cultural domains in general, see Carsten, 2000, p. 13; Yanagisako,
2007, pp. 40–3.
43 See Martin, 2009b, pp. 334–345.
44 Cic. Off. 1.53–54; cf. Arist. Pol. 1252a. On Aristotle’s assessment of the place of the oikos within
the polis, see Martin, 2009b, pp. 332–334. During the Hellenistic age, the household underwent
a shift of importance which can be linked to the political change and the growing importance
of centralized monarchies. The oikoi of the elites became a ‘tool for self-promotion and political
rivalry’ (Nevett, 1999, p. 1); it therefore was rather comparable to the domus of the Roman elite
than to the oikoi of archaic peasants.
45 Hes. Erga 344–345. Cf. Cic. Off. 1.54.
46 Cf. Schmitz, 2004 and Schmitz, 2008.
47 See also Lape, 2002/3 on the standardization of family constellations in Athens under Solon
and Cleisthenes.
22 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

48 See the title of Schmitz, 2008. I am currently working on a larger project concerning the
changes in Greek kinship terminologies and kinship systems from the Archaic to Classical age.
49 See e.g. Thomas, 1992 and Martin, 2009b.
50 Cf. Harders, 2008, p. 14; Varto, 2009, p. 12; Nevett, 2010, pp. 6–7.

Baudy, D. (1999), ‘s.v. Matralia’, DNP, Vol. 7, pp. 1027–1028.
—(2000), ‘s.v. Parentalia’, DNP, Vol. 9, pp. 330–331.
Bettini, M. (1992), Familie und Verwandtschaft im Antiken Rom, Frankfurt/M.,
New York: Campus-Verlag.
Carsten, J. (2000), ‘Introduction: cultures of relatedness’, in Carsten J. ed.,
Cultures of Relatedness. New Approaches to the Study of Kinship, Cambridge:
Cambridge University Press, pp. 1–36.
Cawelti, J. G. (1975), ‘The new mythology of crime’, Boundary, 3, (2),
Collier, J. F. and Yanagisako, S. J. (1987), ‘Toward a unified analysis of
gender and kinship’, in Collier, J. F. and Yanagisako, S. J. (eds), Gender and
Kinship. Essays toward a Unified Analysis, Stanford: Stanford University
Press, pp. 14–50.
Dixon, S. (1992), The Roman Family, Baltimore, London: The John Hopkins
University Press.
Franklin, S. and McKinnon, S. (2001), ‘Introduction: relative values.
Reconfigurating kinship studies’, in Franklin, S., and McKinnon, S. (eds),
Relative Values. Reconfigurating Kinship Studies, Durham, London: Duke
University Press Books, pp. 1–25.
Gestrich, A. (2010), Geschichte der Familie im 19. und 20. Jahrhundert, München:
R. Oldenbourg Verlag.
Godelier, M. (1998), ‘Afterword: transformations and lines of evolution’,
in Godelier, M.; Trautmann, T. R. and Tjon Sie Fat, F. E. (eds),
Transformations of Kinship, Washington, London: Smithsonian Books,
pp. 386–413.
—Trautmann, T. R. and Tjon Sie Fat, F. E. (1998), ‘Introduction’, in Godelier,
M.; Trautmann, T. R. and Tjon Sie Fat, F. E. (eds), Transformations of Kinship,
Washington, London: Smithsonian Books, pp. 1–26.
Goetz, H.-W. (2009), ‘Verwandtschaft im früheren Mittelalter (I): Terminologie
und Funktionen’, in Krieger, G. ed., Verwandtschaft, Freundschaft,
Bruderschaft. Soziale Lebens- und Kommunikationsformen im Mittelalter,
Berlin: Akademie-Verlag, pp. 15–36.
Goldbeck, F. (2010), Salutationes. Die Morgenbegrüßungen in Rom in der
Republik und der Frühen Kaiserzeit, Berlin: Akademie-Verlag.
Harders, A.-C. (2008), Suavissima Soror. Untersuchungen zu den Bruder-Schwester-
Beziehungen in der Römischen Republik, München: C. H. Beck Verlag.
B eyo n d oikos a n d domus : Mo d e r n K i n sh i p S t u d i e s 23

—(2010a), ‘“Let us join our hearts!” – The role and meaning of constructing
kinship and friendship in Republican Rome’, in Mustakallio, K. and Krötzl,
C. (eds), De Amicitia – Social Networks and Relationships. Passages from
Antiquity to the Middle Ages III. Acta Instituti Romani Finlandiae, Vol. 36,
Rome: Edizioni Quasar, pp. 33–47.
—(2010b), ‘Roman patchwork families. Surrogate parenting, socialization,
and the shaping of tradition’, in Dasen, V. and Späth, T. (eds), Children,
Memory, and Family Identity in Roman Culture, Oxford: Oxford University
Press, pp. 49–72.
Hareven, T. K. (2001), ‘s.v. Family and kinship, History of ’, in Smelser, N. J. and
Balte, P. B. (eds), International Encyclopaedia of the Social and Behavioral
Sciences, Vol. 26, Amsterdam: Elsevier, pp. 5291–5298.
Harlow, M. and Laurence, R. (2002), Growing up and Growing Old in Ancient
Rome. A Life Course Approach, London: Routledge.
—(2007), Age and Ageing in the Roman Empire, Portsmouth: Journal of Roman
Janosik, M. (1999), ‘“Do You Take Sinners Here?” Family, community ritual
and the Catholic imagination in the films of Frank Sinatra’, U. S. Catholic
Historian, 17, (3), 67–92.
Jussen, B. (1999), ‘s.v. Verwandtschaft’, LMA, 8, 1596–1597.
—(2009), ‘Perspektiven der Verwandtschaftsforschung fünfundzwanzig Jahre
nach Jack Goodys “Entwicklung von Ehe und Familie in Europa”’, in Spieß,
K.-H. ed., Die Familie in der Gesellschaft des Mittelalters, Ostfildern: Jan
Thorbecke Verlag, pp. 275–324.
Koch, C. (1941), ‘s.v. pietas’, in Paulys Realencyclopädie der Classischen
Altertumswissenschaft, Stuttgart: J. B. Metzler, pp. 1221–1232.
Krause, J.-U. (2003), ‘Antike’, in Gestrich, A.; Krause, J.-U., and Mitterauer, M.
(eds), Geschichte der Familie, Stuttgart: Kröner, pp. 21–159.
Kunst, C. (2005), Römische Adoption. Zur Strategie einer Familienorganisation,
Hennef: Buchverlag Marthe Clauss.
Lacey, W. K. (1968), The Family in Classical Greece, London: Cornell University
Laes, C. (2003), ‘Desperately different? Delicia children in the Roman household’,
in Balch, D., and Osiek, L. (eds), Early Christian Families in Context. An
Interdisciplinary Dialogue, Cambridge, Grand Rapids: Wm. B. Eerdmans
Publishing Company, pp. 298–326.
La Fontaine, J. S. (2001), ‘s.v. Family, Anthropology of ’, in Smelser, N. J. and
Balte, P. B. (eds), International Encyclopaedia of the Social and Behavioral
Sciences, Vol. 26, Amsterdam: Elsevier, pp. 5307–5311.
Lape, S. (2002–3), ‘Solon and the institution of the “Democratic” family form’,
CJ, 98, 117–139.
Lipp, C. (2006), ‘Verwandtschaft – ein negiertes Element in der politischen
Kultur des 19. Jahrhunderts’, HZ, 283, 31–77.
24 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

Martin, D. B. (1996), ‘The construction of the ancient family: methodological

consideration’, JRS, 86, 40–60.
Martin, J. (1997), ‘Die französische Forschung zur römischen Familie – einige
Anmerkungen’, in Bruhns, H.; David, J.-M. and Nippel, W. (eds), Die Späte
Römische Republik – La Fin de la République Romaine. Un Débat Franco-
Allemand d’Historie et d’Historiographie, Rom: École Française de Rome,
pp. 217–220.
—(2009a), ‘Das Vaterland der Väter. Familia, Politik und cognatische
Verwandtschaft in Rom’, in Martin, J., Bedingungen menschlichen Handelns
in der Antike. Gesammelte Beiträge zur Historischen Anthropologie, ed. W.
Schmitz, Stuttgart: Franz Steiner Verlag, pp. 311–327.
—(2009b), ‘Die Bedeutung der Familie als eines Rahmens für
Geschlechterbeziehungen’, in Martin, J., Bedingungen menschlichen Handelns
in der Antike. Gesammelte Beiträge zur Historischen Anthropologie, ed. W.
Schmitz, Stuttgart: Franz Steiner Verlag, pp. 329–343.
—(2009c), ‘Familie, Verwandtschaft und Staat in der römischen Republik’, in
Martin, J., Bedingungen menschlichen Handelns in der Antike. Gesammelte
Beiträge zur Historischen Anthropologie, ed. W. Schmitz, Stuttgart: Franz
Steiner Verlag, pp. 363–374.
Nevett, L. C. (1999), House and Society in the Ancient Greek World, Cambridge:
Cambridge University Press.
—(2010), Domestic Space in Classical Antiquity, Cambridge: Cambridge
University Press.
Overing, J. (2001), ‘s.v. Kinship in anthropology’, in Smelser, N. J., and Balte,
P. B. (eds), International Encyclopaedia of the Social and Behavioral Sciences,
Vol. 26, Amsterdam: Elsevier, pp. 8098–8105.
Parkin, R. (1997), Kinship. An Introduction to Basic Concepts, Oxford: Blackwell
Patterson, C. (1998), The Family in Greek History, Cambridge, London: Harvard
University Press.
Peletz, M. G. (1995), ‘Kinship studies in the late twentieth-century anthro-
pology’, Annual Review of Anthropology, 24, 343–372.
Rawson, B. (2010a), ‘Family and society’, in Barchiesi, A. and Scheidel, W. (eds),
The Oxford Handbook of Roman Studies, Oxford: Oxford University Press, pp.
—(2010b), ‘Degrees of freedom: Vernae and Junian Latins in the Roman familia’,
in Dasen, V. and Späth, T. (eds), Children, Memory, and Family Identity in
Roman Culture, Oxford: Oxford University Press, pp. 195–221.
Rilinger, R. (1997), ‘Domus und res publica. Die politisch-soziale Bedeutung des
aristokratischen ‘Hauses’ in der späten römischen Republik’, in Winterling,
A. ed., Zwischen ‘Haus’ und ‘Stadt’. Antike Höfe im Vergleich, München: R.
Oldenbourg Verlag, pp. 73–90.
Sabean, D. W. and Teuscher, S. (2007), ‘Kinship in Europe. A new approach to a
long term development’, in Sabean, D. W.; Teuscher, S. and Mathieu, J. (eds),
B eyo n d oikos a n d domus : Mo d e r n K i n sh i p S t u d i e s 25

Kinship in Europe. Approaches to a Long-Term Development (1300–1800),

New York, Oxford: Berghahn Books, pp. 1–31.
Saller, R. P. (1984), ‘Tombstones and Roman family relations in the Principate:
civilians, soldiers and slaves’, JRS, 74, 124–156.
—(1994), Patriarchy, Property and Death in the Roman Family, Cambridge:
Cambridge University Press.
Schmitz, W. (2004), Nachbarschaft und Dorfgemeinschaft im archaischen und
klassischen Griechenland, Berlin: Akademie-Verlag.
—(2007), Haus und Familie im antiken Griechenland, München: R. Oldenbourg
—(2008), ‘Verpasste Chancen. Adel und Aristokratie im archaischen und
klassischen Griechenland’, in Beck, H.; Scholz, P. and Walter, U. (eds), Die
Macht der Wenigen. Aristokratische Herrschaftspraxis, Kommunikation und
“Edler” Lebensstil in Antike und Früher Neuzeit, München: R. Oldenbourg
Verlag, pp. 35–70.
Schneider, D. M. (1980), American Kinship. A Cultural Account, (2nd edn),
Chicago, London: University of Chicago Press.
—(1984), A critique of the Study of Kinship, (reprinted in 2007), Ann Arbor:
University of Michigan Press.
Schultz, C. E. (2006), Women’s Religious Activity in the Roman Republic, Chapel
Hill: University of North Carolina Press.
Scullard, H. H. (1981), Festivals and Ceremonies of the Roman Republic, London:
Cornell University Press.
Smith, C. J. (2006), The Roman Clan. The Gens from Ancient Ideology to Modern
Anthropology, Cambridge: Cambridge University Press.
Stein-Hölkeskamp, E. (2005), Das Römische Gastmahl. Eine Kulturgeschichte,
München: C. H. Beck Verlag.
Thomas, Y. (1984), ‘Vitae Necisque Potestas. Le père, la cité, la mort’, in Turcan,
R. ed., Du Châtiment dans la Cité. Supplices Corporels et Peine de Mort dans
le Monde Antique, Table Ronde organisée par l’École Française de Rome avec
le Concours du Centre National de la Recherche Scientifique, (Rome 9–11
Novembre 1982), Rome: Diff. de Boccard, pp. 499–548.
—(1992), ‘The division of the sexes in Roman law’, in P. Schmitt-Pantel and
Goldhammer, A. (eds), A History of Women in the West, Vol. 1: From
Ancient Goddesses to Christian Saints, Harvard: Harvard University Press,
pp. 83–138.
Treggiari, S. (1991), Roman Marriage. Iusti Coniuges from the time of Cicero to
the time of Ulpian, Oxford: Oxford University Press.
Varto, E. K. (2009), ‘Early Greek kinship’, PhD Thesis University of British
Columbia (Vancouver), URL:
(Accessed 15 November 2010).
—(2010), ‘Lessons from the demise and renewal of kinship studies’, Rosetta,
8, 83–99, URL:
(Accessed: 15 November 2010).
26 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

Yanagisako, S. J. (2001), ‘s.v. Household in Anthropology’, in Smelser, N. J. and

Balte, P. B. (eds), International Encyclopaedia of the Social and Behavioral
Sciences, Vol. 26, Amsterdam: Elsevier, pp. 6930–6934.
—(2007), ‘Bringing it all home. Kinship theory in anthropology’, in Sabean,
D. W.; Teuscher, S. and Mathieu, J. (eds), Kinship in Europe. Approaches to a
Long-Term Development (1300–1800), New York, Oxford: Berghahn Books,
pp. 32–48.

Taking a Wider View: Greco-Roman Families and

Political Demography Theory*
Saskia Hin

Introduction: Oikos, Familia and Family History

Families have their history. Both Greeks and Romans invested an amount of
effort into remembering, (re)creating, and cherishing their families that can
impress even the most fervent genealogists today. Ancestors were honoured
with recurrent rituals, ceremonies and family altars, Roman emperors empha-
sized their (imagined) blood-line to the founding father of Rome, and even
gods could not do without a family tree.
Family history as an academic discipline also has a history, and the current
volume urges us both to consider that history, and to explore ways to move
beyond it. This paper contributes to these aims by reflecting upon perspectives
previously employed in the study of the demography of oikos and familia. A
concise, selective, overview of this research (sections 2–4) shows that ancient
historians have been indebted to family historians of early modern Europe.
It also makes clear that we have not yet exhausted the potential approaches
to investigating ‘family matters’ – marriage patterns, co-residence, property
relations and the like – in the ancient world. One way to expand our insights
in the internal functioning of families is, in fact, to investigate the interaction
between families and non-kin structures in society. Putting the family back in
the larger framework of society, and employing political demography theory,
we may ask how ‘politics’ and a society’s institutions created incentives for
certain demographic behaviours, and disincentives for others, and ask how
family life was affected by the specific organizational features of the society to
which families belonged. In modern contexts, empirical research has demon-
strated that taking this wider view is a fruitful way of comparing family life in
different societies. In the second half of this paper (sections 5–8), I will explore
the potential of political demography theory for the study of the family in
ancient societies.
28 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

Reconstructing Ancient Families

In its relatively short history, demography of the ancient family, as a sub-disci-
pline, has made a great effort to capture life in a numerical format. Following
in the footsteps of modern demographers and historical demographers, ancient
historians have used a variety of sources with the aim of uncovering ‘the facts
of life’: how long a Roman might have expected to live, at what age one would –
on average – get married, how many people lived in a household and how they
were related, when people got their children and how many they were. Over the
course of the last three decades, we have gained a much better understanding of
the defining characteristics of Greek and Roman family life and of life courses in
those societies. This is not the place to consider these characteristics in detail. It
may suffice to say that, characteristically, quantification in ancient history goes
in fits and starts, but that, nevertheless, the modest accumulation of evidence
provided by twentieth-century scholarship now enables us to roughly sketch,
for example, patterns of marriage and fertility.1

Shaping Questions and Perspectives: the Influence of Early Modern Family

How did we get to where we are? Fact-finding research into ancient families
owes much to the interests of family historians of later times, in particular to
those focused on early modern Europe. This holds true especially for schol-
arship that has focused on marriage ages, on the composition of Greek, Roman,
and Egyptian households, and on their internal functioning. Ancient historians
have been, and are, keen to get a grip on the nature of the family setting(s) in
which individuals lived their daily lives: with which family members did they
share their meals, roofs and other assets, whom did they take care of and who
took care of them? Were they primarily close kin, members of the so-called
nuclear family (parents and children), or did obligations, care and responsi-
bility stretch beyond that, to include extended family such as adult siblings and
other lateral kin, and grandparents and so on? Census and tax records as well
as epigraphic evidence have been studied intensively (though not exhaustively)
in attempts to reconstruct Greco-Roman family ties and household patterns.2 In
doing so, the question to what extent ancient families lived in nuclear, extended
and multiple households respectively has raised particular interest. It has
been inspired by Anglo-Saxon research into patterns of family formation and
household structures in early modern Europe.
Scholars, studying the period before and during the process of industriali-
zation in Europe, found that distinctive patterns of marriage and of the kinship
composition of households prevailed across the continent. In north-western
Europe, marriage was late and small nuclear families consisting of parents and
children were predominant. This seemed to be in contrast with elsewhere in
Europe where one found early marriage and on average larger families. These
Ta k i n g a W i d e r V i ew: G r e c o - R om a n Fa m i l i e s 29

included a parental couple and a married child with his or her family (stem
family) or extended family members (uncles, aunts, nieces, nephews), or were
formed of two adult siblings with their families sharing a single household
(joint household). Such a pattern of correlation between geography and family
patterns was suggested first by Le Play in the mid-nineteenth century, who
identified geographic zones of different family systems.3 After its further
elaboration in the works of Hajnal and Laslett the concept of a geography of
demography was picked up widely by numerous scholars. The ‘dividing line’
between the two demographic zones was suggested to run from Trieste at
the border between Italy and Slovenia to St Petersburg in Russia and gained
a truly iconic status as an axis marking fundamentally different demographic
behaviours (both in marriage and household patterns).4 With the emphasis on
this division based on between-group variance in family studies, within-group
variance came to be downplayed and even neglected.5
The geographic distribution of household types has been classified and
interpreted as a hierarchical taxonomy. At times, explicit ideological and
developmental connotations were attached to the various types of family
systems. As ‘the fundamental form of societal organization’, family patterns
have been associated with a whole range of phenomena, from the nature of
gender relations and levels of economic development to cultural forwardness
or backwardness, even to ‘rationality’ and ‘irrationality’.6 In combination with
patterns of marriage, they have also been attributed the power to explain the
emergence of political ideologies like capitalism and communism.7
Initially, the north-western nuclear family was perceived to be the result of
industrialization,8 symbolizing ‘progress’, modernity, and growing individu-
alism. It was not until the extensive investigations performed by British scholars
in the 1970s that it was discovered that the nuclear family was in fact firmly
in place in north-western Europe well before the period of industrialization.
Along with this observation the question rose for how long ‘the west’ had
‘diverged from the rest of the world’.9 Ancient historians jumped in to fill the
void: to some extent at least, their investigations into ancient family patterns
responded to the questions proposed by early modern historians.10 Saller and
Shaw studied numerous inscriptions from Italy and the western provinces of
the Roman Empire which, as it turned out, laid emphasis upon ties between
nuclear family members. Despite the fact that they were careful not to equate
these observations to a predominance of nuclear households (as in ‘living
arrangements’), subsequent debates among ancient historians have focused on
the question whether or not their ‘nuclear family’ hypothesis was ‘right’ and/
or held true for other regions in the Greco-Roman world.11 The firm research
paradigms set by family historians of later periods thus had a considerable
impact upon research into the ancient family.
At the start of the twenty-first century, however, a growing body of research
undermines the stark ‘symbolic geographies of demographic space’ created
during the last decades of the twentieth century.12 The empirical basis for the
30 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

stark opposition between ‘east’ and ‘west’ has been found wanting: only very
few archival records had been studied for the east when Hajnal put forward
his hypothesis that came to be so widely acclaimed by subsequent schol-
arship.13 Research into the family in eastern European countries, as well as a
more thorough investigation of demographic patterns in the Mediterranean
has, first, nuanced the sharp dividing lines suggested, and, secondly, brought
variety within these regions to the fore.14 The explosion in research into family
life in the modern Mediterranean, and the variation in demographic patterns
observed over small distances in modern – but still largely agricultural – Greece
and Italy urged ancient historians on their part to move beyond the common-
alities of ancient families and to consider variation in family patterns within the
Greco-Roman world.15

Demographic Patterns as a Dependent Variable

The growing body of data and the concurrent criticism of the thesis that family
patterns broadly followed geographic zones thus created an urgency to view
family patterns as a dependent variable, as the result of – or as adaptations
to – specific circumstances. Increasingly, attention shifted to the question why
families functioned in particular and distinctive veins. A growing number of
studies into pre-modern European families started to emphasize the impor-
tance of variations in economic conditions, land inheritance systems, and
ecologic characteristics for the composition of households and the life courses
of individuals.16 Ancient historians also sharpened their conceptualization of
how these factors might have invoked differentials in demographic behaviour
between individuals and between communities. But in practice, moving beyond
the so-called ‘informed speculation’ about the causes underlying family patterns
and demographic behaviour is difficult. Data are not as fine-grained as we might
wish, and hide variation across time and space to a large extent. Yet in some
instances, evidence does allow us to see the aforementioned factors at work –
the economy, patterns of landholding and ecology/geography. Famously, the
census data from Roman Egypt shows that, as in later societies, people in the
countryside more often than urban dwellers lived in households that included
relatives other than nuclear family members.17 Changes in economic conditions
have been linked to altering demographic behaviour in the context of e.g. the
(disputed) ‘Gracchan crisis’ in late Republican Italy.18 Patterns of landholding,
ecology, and economy are thus being recognized as relevant factors that affected
demographic patterns in the Greco-Roman world.19
These three factors (landholding, ecology, and economy) are thought to
explain a large chunk of the observed variation in demographic patterns.
‘Culture’, and ‘the culture of reproduction’ (norms and values pertaining to
marriage, fertility, and gender roles) form a fourth major determinant of patterns
of demographic behaviour and family life. For the ancient world, it has been
suggested that different ‘cultures of reproduction’ might have led to differential
Ta k i n g a W i d e r V i ew: G r e c o - R om a n Fa m i l i e s 31

reproductive behaviour between the Roman elite and the population at large.
It may seem that with the identification of these four ‘causes of divergence’, we
are as close to understanding the how and why of ancient family systems as we
can get, given the state of the evidence. Indeed, for demographers, too, these
factors, which may be subsumed under the headings ‘economy’ and ‘culture’, are
the usual suspects that are under investigation as key variables.
Yet comparative research into contemporary family life points out that
variation in demographic patterns is spurred along different lines as well, not
least by differences in the institutional organization of societies. Furthermore,
the emergence of a sub-discipline branded as ‘political demography’ (which
considers both the direct and indirect effects of policy measures on demographic
patterns and behaviour) underscores the relevance of taking the embedding of
families in larger organizational structures into account. By overlooking the
question of how the family as an ‘organization’ at the micro-level might interact
with macro-level organization structures, one misses part of the dynamics that
occur between families and external factors.
A brief and selective evaluation of the ‘state of the art’ in demographic
research, in other words, suggests that ancient historians have not yet exhausted
their potential angles of investigation into the how and why of family patterns
in the Greco-Roman world. It also, once again, makes the need for comparative
and interdisciplinary research explicit. While various ways of pursuing such a
wider view are conceivable, in the remainder of this paper I shall concentrate on
one way of doing this: studying oikos and familia within a political demography
perspective. But before doing so, we first need to define more precisely what the
concept of ‘political demography’ entails.

The Political Demography Perspective: the

Interaction between Family Institutions and
‘Non-Family’ Institutions20
Coined in the early 1970s,21 the term ‘political demography’ is fondly used
by journalists and political leaders to denote practical macro-level measures
directly aimed at affecting the size, the distribution and the composition (by
age, sex, and ethnicity) of a population. There are abundant current examples
of health, fertility and migration policies instigated by states that try to affect
people’s demographic behaviour: China’s one-child policy, financial incen-
tives and compensations to childbearing, and the strengthening of ‘Fortress
Europe’ are but three prominent examples. The particular interest in state
power and state security (or: geopolitical stability) that underlies much of the
political interest in demography is far from new, and examples reflecting such
interest can be pushed back far in history. Plato’s well-known address of issues
concerning population size and population composition in his Politeia (State)
32 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

counts among the pre-Christian examples, and the topic captured the attention
of other ancient statesmen and philosophers discussing state organization.22
Whereas in Plato’s case, political demography remained at a theoretical level
or a utopian ideal, the Roman Emperor Augustus actually attempted to mould
fertility behaviour by implementing a set of rules that may truly be regarded
as the first (but largely failed) ‘family policy’ that is mirrored in contemporary
political demography targeted at the family.23
In its short history, the scientific sub-discipline of political demography
has followed suit: in concurrence with politicians and journalists, scientists
have largely been concerned with analyzing the demographic effectiveness of
historical, current, or required population policy measures – in particular in
relation to immigration and fertility issues. Beyond that, the interrelationship
between politics and demography has remained a neglected area, largely ignored
by political scientists and demographers alike. Already in their 1989 study on
demographic change in nineteenth century Italy, Kertzer and Hogan criti-
cized the narrow focus of historians’ interests in the interrelationship between
politics and demography. Historical demographers, so they observed, almost
exclusively concentrated on the effectiveness of overt attempts by governments
to intervene in reproductive processes, to the neglect of indirect, but quite
powerful, governmental influences upon demographic behaviour. Suggesting
that, in fact, ‘political change and state intervention may be key ingredients’
in demographic change, they urged historical demographers to attend to the
importance of the state in order to further our understanding of demographic
Kertzer and Hogan’s call for ‘a more ambitious and intellectually challenging
research agenda than has been the norm for demographic studies’ has not been
met with an eager response, at least not where the indirect impacts of political
measures and institutions on demographic behaviour are concerned. Almost
20 years later, echoes of the same complaint can still be heard.25 When ‘political
demography’ is considered, it is strictly limited to the impact of the state’s
economic measures and institutions upon demographic behaviour, in particular
those that are concerned directly with patterns of landholding.26 Studies of the
ancient family tend to ignore altogether the political framework within which
demographic behaviour took place. That is, beyond the direct population policy
measures of Augustus, state organizational structures shaping civic life have,
to my knowledge, not been studied for their impact upon demography in the
Greco-Roman world. Yet there is no reason why, as opposed to in later societies,
politics should have been absent as a factor shaping demographic determinants
in the context of the Greek polis-system, the Roman Empire, or the Hellenistic
Kingdoms – to name just the common triad.
In the following discussion, I explore how organizing principles in ancient
states that were meant to help establish a community, and to ensure the
maintenance of that community, indirectly affected the demographic behav-
iours of its citizens. I do so by laying out two embryonic ‘case studies’: the
Ta k i n g a W i d e r V i ew: G r e c o - R om a n Fa m i l i e s 33

first is concerned with the interrelationship between citizenship policies and

the selection of marriage partners, and compares Classical Athens and late
Republican and Imperial Rome.27 The second part of my paper investigates the
possibility of further study of the effect of housing policies on the geographical
distribution of different subsections of the population. This brief analysis by
no means covers these issues exhaustively, let alone the many dimensions in
which interaction between states and the families it contains might take place
and the forms these might take.28 However, the paper does demonstrate how
policies that were not instigated with a view to altering demographic behaviour
could have an effect on demographic outcomes. A perspective from political
demography allows us to gain deeper insights into behavioural differences
with regard to marriage, family formation, fertility, and other demographic
phenomena. Hence, above all, the current paper aims at encouraging histo-
rians of the ancient family to include ‘politics’ into multi-causal frameworks of
demographic analysis.

Citizenship Rights and the Selection of Marriage Partners

The political systems of Rome and Athens were distinctive in many respects.
The ways in which Classical Athens and the Romans in late Republican
and Imperial Rome defined and attributed the right to be a full member of
their communities were among the most fundamental differences separating
these two ancient states. Essentially, Romans defined citizenship as ‘inclusive’,
whereas Athenians employed an exclusive definition of citizenship rights.29 That
is, even when due recognition is given to subtleties, it is fair to say that Romans
were significantly more open to sharing their citizenship rights with someone
born without these than were their Classical Athenian counterparts. During
most of the Classical period, Pericles’ Citizenship Law rendered the right to
citizenship as the exclusive privilege of those who were the legitimate offspring
of a legally married couple consisting of two Athenian citizens. For others,
it was not impossible to become citizens, but their prospects of gaining this
legal status were minimal. Citizenship was granted only as a reward for special
services, provided by the individual or his hometown, to the city of Athens, and
from the 380s BC its bestowal was conditional upon the successful outcome of
two voting rounds in the popular assembly.30 In other words, during the period
in which the city-state of Athens was closer to functioning as a democracy than
any other society before, for outsiders access to the in-group of citizens with full
rights was dependent upon support by a democratic majority among the voting
citizen members. These stringent preconditions to citizenship contrast sharply
with the relative ease of gaining citizenship in ancient Rome. At an early phase
in its history, Rome started to conclude treaties with defeated neighbouring
Italian populations which gave part of them intermarriage rights with Roman
citizens (ius conubii), and turned the children of a marriage between a Roman
man and a foreign woman into Roman citizens. Furthermore, freed slaves, if
34 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

manumitted legally, would gain citizenship rights, and their offspring would
have been born as Roman citizens.31 After the Social War at the start of the
first century BC (91–88 BC) virtually all free inhabitants of Italy gained Roman
citizenship. Outside of Italy too, defeated enemies, citizens of allied cities, and
‘deserving individuals’ were frequently given citizenship, especially from the
time of Caesar onwards.32 Romans prided themselves on their tradition of
integrating ‘others’ into their growing Empire.33 Finally, under the Emperor
Caracalla in AD 212 nearly all inhabitants of the vastly expanded Roman
territory gained Roman citizenship.34 In the Roman context, decisions over
extending group membership were not made by the group, but rather by its
(non-democratic) leaders. In making their decision, they acted in their own
strategic interests, if we may believe the comments of Cassius Dio on Caracalla’s
motivations. According to him, the fact that the Emperor needed more income
from taxation was the reason why he chose to extend citizenship rights to more
inhabitants of the Roman Empire.35
At a very different level of social organization, marriage patterns in Athens
and Rome displayed quite divergent characteristics. The key question to
consider is: whether these two phenomena were fully independent of each
other, or whether there was an interrelation between the specific policies of
distributing citizenship rights on the one hand, and marriage behaviours in each
state on the other. In my view, the nature of the differences in marriage patterns
between the two societies suggests that their respective citizenship policies
affected the patterns of marriage that are observed and implies that the diver-
gences can at least partly be viewed as outcomes of state policies to citizenship.
The distinction between Rome and Athens is the possibility of intermarriage.
While there are no indications that ages at marriage or the duration of marriage
varied significantly between the two societies, it is generally held that Athenian
marriage can be characterized as endogamous, whereas at Rome exogamous
marriage was the norm.
These general characterizations are supported by various types of evidence.
In the Athenian context, epigraphic and literary evidence suggest that marriage
was seen as a way of reinforcing kinship ties, and that for this reason multiple
marriages could take place between the same families, or within a family
group.36 Kin marriage beyond that of full siblings was not prohibited, and in fact
legally enforceable for female heirs (epikleroi).37 More relevant to our question,
Athenian citizens not only practiced endogamy within their biological kin
group relatively frequently; they also married within the citizenship group. An
early attempt by Davies in the 1970s to identify marriages between Athenians
and foreign non-citizens found only three cases of civic intermarriage in the
Classical period.38 After the recent publication of Osborne and Byrne’s volume
which lists all identifiable foreigners in Athens, the task of identifying cases of
intermarriage has now been eased significantly as compared to the 1970s.39 My
investigations of this large database corroborate Davies’ findings that attesta-
tions of marriages with foreigners are rare. The number of identifiable civic
Ta k i n g a W i d e r V i ew: G r e c o - R om a n Fa m i l i e s 35

intermarriages among over 2,000 foreign immigrants for the fifth through to
the third centuries BC is small: 33 cases or 1.6 percent.40 The contrast with what
we know about Rome is striking. Here, legal circumscriptions concentrated
on the prevention of kin marriages rather than on securing kin-marriages for
female heirs.41 What about marriages between Roman citizens by birth and
immigrants? For former slaves, usually of foreign descent, and certainly not
born with citizenship rights, it was not uncommon to marry a Roman citizen.
The evidence for ex-slaves of the emperor’s household marrying Roman citizens
shows that 42 per cent of this group of ex-slaves married a Roman citizen.42
Needless to say, these slaves from the emperor’s household may be exceptional
and be seen as having added value in the marriage market – nevertheless it does
point to the integration of this group of new citizens with those descended from
One might, still, object that the reflections of intermarriage on inscrip-
tional evidence tell us more about people’s incentives or disincentives to
make their mixed marriage public than about actual intermarriage rates.
However, comparative evidence drawn from other societies broadly acknowl-
edges and affirms that actual intermarriage rates between native-born citizens
and immigrants vary substantially.43 They depend on a variety of factors, among
which should be included: socio-economic status and cultural proximity.
Recognition needs to be given to the variation in the rise and fall of the rate of
intermarriage according to changes to individual and to group characteristics.44
It is, therefore, certainly not legitimate to attribute divergences in intermarriage
practices between societies to ‘politics’ alone – or to any other single explan-
atory factor for that matter. Nevertheless, it is certainly not beyond the plausible
that part of the divergence should be explained as resulting from the particular
disincentives that the Athenian state or democracy created with reference to
intermarriage between citizens and non-citizens.
There is good evidence for the restriction of intermarriage (understood here
as a union between a citizen and a non-citizen living together as married) – and
its prohibition by Athenian law in the Classical period. This is most clearly seen
in the famous lawsuit against Neaira.45 That some people nevertheless ignored
the legal provision is also evident from this case – as well as from the inscrip-
tions mentioned above. It should be remarked though, that an impressionistic
survey of the non-citizens who succeeded in ‘conquering’ an Athenian woman
suggests that these men had something to compensate their lack of citizenship
with. Some of them were wealthy. Others came from neighbouring city-states.
For them, their greater cultural proximity may have helped, as well as, perhaps,
an initial prospect or plan of establishing the – normally virilocal – household
in the man’s city of origin, where he held citizenship. But this is speculation.
Foreign men in the inscriptional record who both came from far away and
lacked markers of wealth were less successful in marrying an Athenian citizen.
Presumably, they had nothing to compensate for their lack of citizenship rights.
Most importantly perhaps, Athenian law furthermore prevented the children
36 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

of mixed marriages from owning property. Hence parents in a mixed marriage

could not transmit property to their children. This was an issue of relevance to
most people: it is estimated that smallholder farming was dominant in Attica
and that about 75 per cent of citizens owned land.46 Intermarriage would have
caused the land to be lost and for children to be impoverished. A final factor to
consider is that disobedience of the law against intermarriage could have severe
consequences for the family group. If a lawsuit was opened, a parent responsible
for giving his child into marriage with a non-citizen could suffer disenfran-
chisement and confiscation of property. Moreover, a 1,000-drachma fine was
also imposed by the state. While the ‘love is blind’ principle might still have
driven the considerations of the individuals involved, it seems unlikely that
their parents or others responsible for the arranging of their marriage would
have been blind to their losses. It may therefore justly be concluded that the
preconditions and privileges connected to citizenship that arose from Athenian
politics, and became institutionalized in law, put up such strong disincentives
to intermarriage that we should recognize their potential to have affected
demographic behaviour.

Housing and Population Distribution

Like the choice of a marriage partner, the choices over with whom to live, or to
share a household with, and the decision over where to do so might seem to be
decisions that belong to and are taken in the private domain. If not contingent
upon an individual’s personal preferences, then they should above all be a
family matter. In reality though, the private realm does not function in isolation
of the public and political settings in which it is embedded. Hence, the state had
a role to play in shaping the nature of the households of citizens.
That politics may have had an influence in shaping the distribution of house-
holds over space – in other words, on where people located their households
– is perhaps most evident in modern contexts. Subpopulations, distinguished
by socio-economic and other criteria, tend not to be distributed evenly over the
territory of states. The causes of such distributional skews can be manifold, and
battalions of people tend to be involved in counteracting ‘geographic clustering’
when it is associated with specific problems. Governments and their policies,
at the same time, have also contributed to the process of its emergence. Within
the Greco-Roman world, perhaps the largest intentional impact of state or
leading politicians’ initiatives on the geographic distribution of subpopulations
came in the form of Greek and Roman colonization and land (re)distribution
programs. These programs targeted at land held by the state in the countryside
as well as overseas locations, and the pervasive role of politics and institutions
in directing these developments has been well recognized by ancient historians
researching the phenomena, even if recent debates have qualified the extent
to which colonization was organized.47 Also within a single city institutional
factors had an effect on distribution of population. As referred to above,
Ta k i n g a W i d e r V i ew: G r e c o - R om a n Fa m i l i e s 37

during the Classical period the right of owning and selling houses and land
(enktêsis gês kai oikias) was a privilege accorded to Athenian citizens only. As
a result, foreigners and others without citizenship rights living in Athens were
dependent on what was on offer in the rental market, with little influence on
its location.48 According to Xenophon’s analysis in the Poroi, the fact that the
right to own and sell property (enktêsis) was conditional upon citizenship was
a causal factor explaining the presence of ‘empty zones’ – vacant houses and
plots – in the city centre of Athens. If we may believe him, there was a demand
for this real estate by foreigners who could afford to buy it, but were unable to
do so because of legal prohibitions:

Consider what would happen if in addition to enjoying the blessings that are indig-
enous, we first of all looked after the interests of the foreigners residing here (metoikoi).
For in them we have one of the best sources of revenue ( . . . ) Then, since there are
many vacant houses and plots within the walls, if the polis gave the right to own and
sell property (enktêsis) to those who intend to build, who applied and seemed suitable,
as a result I think more and more worthy people would wish to live at Athens.49

This comment puts the well-known concentration of metics in the Piraeus50

in a different perspective. The geographic distribution of non-citizens over the
territory of Attica surely fitted well with the focus of their economic activities
on trade and manufacture.51 But the (implicit) emphasis on the economic logic
of metics’ ‘preference’ for the Piraeus has caused modern scholars to underplay
the importance of constraints metics might have faced over the choice of their
location. To what extent was rental housing available for them in other parts of
Attica? And how did rental accommodation get to be available at the locations
where it was available? These are questions which deserve consideration beyond
that of the context of this paper. Let it suffice here to point out another passage
by Xenophon, of which the most logical implication seems to be that the city of
Athens as a community of citizens had an active role in determining where (and
where not) immigrant renters might reside, and what their housing looked like.
The suggestion of ‘town planning’ is borne out by his description of the process
of creating new urban development. According to Xenophon, the way to go
would be for the city to collect money among its citizens. With this money, they
could build lodging houses around the harbours for ship-owners,52 naukleroi,
in addition to the existing ones to attract more foreigners. The city should also
provide public lodging for visitors, and trading places for merchants. For all of
these, the city would be the owner and hence take the revenues from rental as
a source of public income.53 Of course, it might be that Xenophon proposed
a new way of organizing a process of urban development. But there is no
indication in the text that his proposal was revolutionary in this respect. In fact,
he presents his plans as a simple expansion of current practice. If these houses,
that he suggests were built by the city to rent out to non-citizens, were indeed
an extension of existing policy, Shipley could be right in his suggestion that the
38 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

so-called Typenhäuser found in the Piraeus might be rented by metics from

their landlords.54 At the very least, what little indications literature provides
on the right to own land and housing, and on the functioning of the Athenian
housing market, urges us to take the possible role of politics and institutions
into account when trying to understand spatial residence patterns.

Other Applications of Political Demography Theory in Ancient History

The previous two sections have provided two examples of how political demog-
raphy theory might be employed as a conceptual and analytical framework by
ancient historians studying families. The current case-studies have a narrow
focus in time and space. Above all, they centre on dynamics taking place in the
city of Athens, during the period characterized by a democratic political system.
In that sense, they have limited value and applicability. But this by no means
implies that a political demography perspective is of use only to the students
of the Classical Athenian family. The main purpose of the current exercise has
been to demonstrate how a theory which predicts that institutional frameworks
at a societal level of organization have an impact on the dynamics of family life
is valuable to, and of use in, the context of the ancient Mediterranean world.
Family life and various aspects of oikoi and familiae in several other contexts
within the Greco-Roman world may, I believe, fruitfully be explored from a
political demography perspective. Thinking of Rome, for example, it seems
worthwhile to consider how the patronus-cliens system, which was arguably
one of the central structures in Roman social organization and played its role in
politics, impacted upon family dynamics and family interrelationships.55
The type of analysis made here places emphasis on the role of policies and
institutions in shaping disincentives and incentives for people to act in specific
ways, and to refrain from other strategies or behaviours, even in blocking
alternatives. Inevitably, this creates the impression of a degree of determinism:
it invokes the implicit suggestion that people did not really matter as actors.
Needless to say, of course they did. As emphasized by sociologists as well as
family historians of other historical periods, interactions of families with the
rest of society are multiple and multi-directional, and families are both subjects
and agents. This holds true no less for the interactions between families and
(political) institutions. The point is to come to a balanced evaluation of both
aspects by ‘pinpointing the ability of families to act on their surrounding world
and by estimating the effects of structural constraints on family development’, a
matter, in the words of David Reher, ‘of utmost importance’.56 Ancient historians
working on the demography of the Greco-Roman world have made many a
strong and valid case for the impact of demographic structures both on politics
and on family life.57 With this article, I have tried to make a case for investi-
gating the interrelationship between families and politics from the opposite
direction, considering the impact of political structures on demography. If ever
there was the right time and place to do so it must be, I feel, in a volume that
Ta k i n g a W i d e r V i ew: G r e c o - R om a n Fa m i l i e s 39

aims to consider new frameworks and perspectives for studying the family in
antiquity, published at the start of the twenty-first century.

  * I am grateful to the participants of the Oikos/Familia conference in Gothenburg for their valuable
comments on the paper on Citizenship and Demography I presented there, as well as to Walter
Scheidel and Mikolaj Szołtysek for their valuable comments on an earlier draft of this paper.
  1 For patterns of marriage cf. e.g. Scheidel, 2007. On fertility patterns, see Bagnall and Frier,
2006, ch. 7. On seasonal patterns of birth, see Shaw, 2001.
  2 Census records: Bagnall and Frier, 2006 (Roman Egypt). Tax registers: Clarysse and Thompson
2006a and 2006b (Ptolemaic Egypt). Inscriptions: Saller and Shaw, 1984, (Roman Empire),
Martin, 1996 (Asia Minor), Edmonson, 2005 (Lusitania), and Huebner, 2011 (Egypt).
  3 Le Play, 1877–1879, in six volumes. He referred to what is now known as the nuclear family as
‘the unstable family’. On Le Play and his influence cf. Mogey, 1955, Thornton, 2005, pp. 58–61
and Szołtysek, forthcoming.
  4 Hajnal, 1965 (marriage patterns), and Hajnal, 1982 and 1983 (household patterns). Laslett,
1983. In an eighteen-page review, Berkner, 1975 provides extensive and sound methodological
criticism of Laslett and Wall, 1972, a book in which Laslett developed the methodology to study
family structure employed in his later work.
  5 See Thornton, 2005, p. 49.
  6 Gender relations: Hartman, 2004. Family systems and the emergence of capitalism: Seccombe,
1992 (see for reviews Mentzer, 1994 and Tilly, 1993). Family systems and industrialization: see
Kertzer, 1991, p. 157 for references. Todd’s 1985 research puts emphasis on family systems and
non-democratic political systems. Szołtysek and Zuber-Goldstein, 2009, lay out the impact of
ethno-cultural ideologies on the classification of households in the East. On the ideological
dimension to historical household studies in general, see Szołtysek, forthcoming and Thornton,
2005, esp. pp. 61–72 (‘Developmental-Trajectory Interpretations’) and ch. 6 and ch. 7.
  7 Todd, 1985.
  8 D. S. Smith, 1993, however, argues that ‘the myth that industrialization transformed the family
from extended to nuclear was largely a creation of those who refuted it’, p. 342. For a rebuttal
of this view, cf. Thornton, 2005, pp. 96–97.
  9 Laslett, 1977, p. 46: ‘Can it yet be said when the west began to diverge from the rest of Europe
and the rest of the world in its familial outlook and behaviour?’ In fact, the question was already
raised by Hajnal in 1965, p. 106: ‘If the European marriage pattern is unique, it is natural to ask,
“When did it arise?” ’. Following up on this question, he briefly considers marriage patterns in
the ancient (Greco-Roman) world on pages 120–122.
10 Saller and Shaw, 1984, refer explicitly to the work of Laslett and the Cambridge Group, which
elaborated on that of Hajnal, as well as to earlier work by Le Play. Saller’s 1994 book acknowl-
edges the ‘immeasurable impact’ of the Cambridge Group on the work (p. xii). Bagnall and
Frier, 2006, adopt Laslett’s household typology. Cf. also Tacoma, 2007, p. 5 and Huebner, 2011.
11 Cf. Martin, 1996, and Huebner, 2011, for criticism on Saller and Shaw’s hypothesis. For studies
aimed at expanding knowledge on household composition in other regions of the ancient
Mediterranean, see note 2.
12 See already the overview by Kertzer, 1991. For recent criticism on the lack of foundation under-
lying assumptions on the character of family systems in ‘the east’, cf. Sovič, 2008.
13 Szołtysek, forthcoming.
14 A ‘Mediterranean pattern’ was suggested by Hajnal, 1982, abridged and slightly revised in 1983,
and laid out in R. Smith, 1981 and Laslett, 1983. See Viazzo, 2003 for an overview of subsequent
research emphasizing internal variation within the pre-modern Mediterranean.
15 Cf. Pomeroy, 1997, p. 9 commenting that ‘according to the new historical model, the Roman
family looks more different from the Greek than it used to’. See also Bradley’s 1993 review of
Evans, 1991, and Kertzer and Saller, 1991.
40 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

16 On the Mediterranean in early modern Europe, cf. e.g. Peristiany, 1976, on a range of countries;
Barbagli, 1991; Oppo, 1990; Rettaroli, 1990 and Viazzo and Albera, 1990, on Italy; Hionidou,
1995, and Wagstaff, 2001, on Greece; Gruber and Pichler, 2002, on Albania (twentieth century);
Rowland, 2002, on Portugal and Reher, 1997, on Spain.
17 Bagnall and Frier, 2006, pp. 66–69.
18 See in particular Brunt, 1987, and Hopkins, 1978, explanatory frameworks. For recent analyses
of the developments during this period, cf. the articles in De Ligt and Northwood, 2008.
19 See e.g. Horden and Purcell, 2000, pp. 379–380 (ecology), and Sallares, 1991, ch. 2.6 (economy,
inheritance systems, and family structures).
20 I adopt the term ‘non-family institutions’ from Thornton, 2005, p. 94, to denote non-family
modes of social organization, which might occur at state-level or at community-level.
21 Myron Weiner was the first to use the term in his 1971 essay written for a volume on the conse-
quences of rapid population growth commissioned by the Office of the Foreign Secretary of the
US National Academy of Sciences which advises the US government on scientific and techno-
logical issues that pervade policy decisions. Weiner, 1971. See also the Preface and Introduction
to Weiner and Teitelbaum, 2001.
22 Despite its antiquity and limited coverage, Stangeland’s 1904 work on pre-Malthusian doctrines
of population (reprinted in 1966) still remains the most encompassing study on the topic for
the Greco-Roman period.
23 On the Augustan family legislation, instituted by means of the Lex Julia de maritandis ordinibus
(18 BC) and the Lex Papia Poppaea (AD 9) see briefly Dixon, 1992, p. 79. For more elaborate
considerations, see Field, 1945; Mette-Dittmann, 1991 (a detailed guide into the legislation
itself), and Treggiari, 1991, pp. 60–80. Scholars now widely agree that his policy was mainly
directed at Roman elites, and that it was largely unsuccessful in trying to raise fertility rates
among these groups. Cf. Dixon, 1992, and Frank, 1975, who cites relevant evidence. See also
Saller, 2001, on the moral background to these legislations.
24 Kertzer and Hogan, 1989, pp. 184–185.
25 See from the side of modern demography Teitelbaum, 2006, and Weiner and Teitelbaum, 2001.
Historical demographer Sovič, 2008, pp. 151–156, also points to the neglect of factors such as
differences in taxation systems and ‘differential government policies’ in general in research on
family history and urges for consideration of these factors as underlying differences in family
26 It thus comes as no surprise that historical demographers are more familiar with the narrower
term ‘political economy’ than with the more inclusive concept of ‘political demography’.
27 I focus here specifically on democratic fifth and fourth century BC Athens. Policies regarding
citizenship differed in Hellenistic times and by region. See on this phenomenon Lomas,
2000, who focuses on the different attitudes towards mixing people from different ethnic
backgrounds in the Greek cities in Southern Italy (Magna Graecia).
28 Other relevant factors may include policies regarding military recruitment systems, state laws
conditioning employment and education, taxation systems (e.g. per person or per household
unit) and the construction and quality of transportation networks. Cf. Kertzer and Hogan,
1989, and Sovič, 2008.
29 For these characterizations, see e.g. Mathisen, 2006, for a recent view on Rome and Whitehead,
1977, and Davies, 1977–1978, on Athens.
30 See M. J. Osborne, 1981–1983, Vol. 2, p. 48, and Vol. 4, pp. 161–164.
31 Sherwin-White’s 1973 volume on Roman Citizenship still remains the most encompassing
work on citizenship issues in the Roman world.
32 See Brunt 1987, 159–165, and Sherwin-White, 1973. Cicero’s Pro Balbo, esp. 20–25 and 29–31,
contains a wealth of information with regard to the conferral of citizenship on non-Italians in
the first century BC. Crawford, 2008, now suggests that seven million people outside of Italy
might have been granted Roman citizenship by 28 BC already, but see Hin, 2009, p. 192–193,
for arguments why this claim cannot be substantiated.
33 Cf. Cicero, Pro Balbo, 31, in which he says that the decision of Roman leadership ‘from the time
of Romulus’ to adopt even enemies as citizens without any doubt ‘has been the strongest basis
of our Empire, and the thing which has more than anything else augmented Roman fame’.
Ta k i n g a W i d e r V i ew: G r e c o - R om a n Fa m i l i e s 41

34 Cf. e.g. Mathisen, 2006.

35 Cassius Dio, Roman History, 78.9. A text of the Edict is preserved in P. Giss. 40, which indeed
explicates that as a result of the extension of citizenship rights everyone from then on needed
to pay taxes.
36 Cf. Cox, 1998, ch. 2 and Vérilhac and Vial, 1998, for epigraphic evidence on marriage with kin
and between members of demes (which, because of their hereditary membership, over time
increasingly parallelled kinship ties). For literary evidence on close-kin marriage between
Athenian citizens, see R. Osborne, 1985, p. 130f.
37 On epikleroi, see Schaps, 1975.
38 Davies, 1977–1978, 111.
39 Osborne and Byrne, 1996.
40 The count here depends on the identification criteria of Osborne and Byrne, who have
identified Athenian citizens as such by reliance on the presence of a demotic, and foreigners by
the presence of an ethnikon. These criteria have been applied throughout Part 1 of their volume,
which I have used to filter out intermarriages during the fifth through third centuries BC (their
Part 2, which includes a supplement as well as ‘foreigners of unknown provenance’, without
an ethnikon, have been excluded from the analysis). On the use of the ethnikon by Greeks, see
Hansen, 1996, p. 174.
41 Until the third century BC, law prescribed that cognates and agnates to the third degree should
not marry: see Plutarch, Quaestiones Romanae, 108.
42 Weaver, 1986, p. 156.
43 In modern contexts, I have found rates between 5% (Dutch natives with Turkish immigrants
in Amsterdam, 2008) and 76% (French natives with Italian immigrants in the 1970s).
Sources: Coleman 1994 and CBS Statline (
44 See Coleman, 1994.
45 Cf. e.g. Scafuro, 1994, p. 163, and Roy, 1999, pp. 4–5.
46 Finley, 1951, pp. 56–58.
47 For colonization and land reforms in the Roman Republican context, see for example
Broadhead, 2007, and Roselaar, 2010. Good entries to the debate on Greek colonization are
provided in Lomas, 2010, and Malkin, 2002.
48 The dependence and lack of empowerment of non-citizens in this realm is neatly illustrated
by Diogenes Laertius, 10.10, which describes the will of the philosopher Epicurus. Since the
people whom he intends to be the heirs of his property do not have citizenship rights, he is
unable to bequest his house to them. They are dependent on the goodwill of the heirs Epicurus
eventually appointed, Athenian citizens, in using the place for accommodation.
49 Xenophon, Poroi [Ways and Means] 2.1 and 2.6. (Translated by Bowersock, 1968).
50 On this, see Whitehead, 1986, pp. 83–84.
51 On the professions of metics, cf. e.g. Garland, 1987, pp. 68f.
52 The most obvious translation for naukleroi would be ‘ship-owners’. LSJ point out, however, that
specifically in Athens the term was also used to denote ‘one who rented and sub-let tenement
houses’, i.e. a landlord. It remains opaque, therefore, whether the city intended to house the
ship-owners in lodging houses, or whether these wealthy foreigners lived elsewhere (in better
housing?) but were given the opportunity to rent apartment blocks in which they could provide
their foreign employees accommodation in sublet once these were hired to work in Athens.
53 Xenophon, Poroi [Ways and Means] 3.6–7 and 3.12–13. I quote the relevant passages in the
translation of Bowersock, 1968: [3.6–7:] ‘Other methods of raising revenue that I have in mind
will require capital, no doubt. Nevertheless I venture to hope that the citizens would contribute
eagerly towards such objects (. . .) [3.12–13:] ‘When funds were sufficient, it would be a fine
plan to build more lodging-houses for ship-owners (naukleroi, see on this above in note 52)
near the harbours, and convenient places of exchange for merchants, also hotels to accom-
modate visitors. [13] Again, if houses and shops were put up both in the Piraeus and in the city
for retail traders, they would be an ornament to the state, and at the same time the source of
considerable revenue’.
42 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

54 Shipley, 2005, pp. 370f. He suggests that the landlords of metics might have been the otherwise
obscure prostatai or ‘immigrant sponsors’ whose formal support foreigners needed to be regis-
tered as residents in Athens.
55 I am grateful to Arjan Zuiderhoek for his suggestion to consider the Roman case.
56 Quote from Reher, 1997, p. 10.
57 To make reference to but a few examples: Saller, 1994; Tacoma, 2006, and Zuiderhoek, 2006.

Bagnall, R. S. and Frier, B. W. (2006), The Demography of Roman Egypt, (2nd
edn), Cambridge: Cambridge University Press.
Barbagli, M. (1991), ‘Three household formation systems in eighteenth- and
nineteenth-century Italy’, in Kertzer, D. I. and Saller, R. P. (eds), The Family in
Italy from Antiquity to the Present, New Haven and London: Yale University
Press, pp. 250–270.
Berkner, L. K. (1975), ‘The use and misuse of census data for the historical
analysis of family’, Journal of Interdisciplinary History, 5, (4), 721–738 (review
of Laslett, 1972).
Bowersock, G. W. (1968), Constitution of the Athenians, Cambridge MA:
Harvard University Press.
Bradley, K. (1993), ‘Writing the history of the Roman family’, Classical Philology,
88, 237–250.
Broadhead, W. (2007), ‘Colonization, land distribution and veteran settlement’,
in Erdkamp, P. ed., A Companion to the Roman Army, Malden MA: Blackwell,
pp. 148–163.
Brunt, P. A. (1987), Italian Manpower, (2nd edn), Oxford: Clarendon Press.
Clarysse, W. and Thompson, D. J. (2006a), Counting the People in Hellenistic Egypt,
Vol. 1, Population Registers (P. Count), Cambridge: Cambridge University Press.
—(2006b), Counting the People in Hellenistic Egypt, Vol. 2, Historical Studies,
Cambridge: Cambridge University Press.
Coleman, D. A. (1994), ‘Trends in fertility and intermarriage among immigrant
populations in Western Europe as measures of integration’, Journal of Biosocial
Science, 26, 107–136.
Cox, C.-A. (1998), Household Interests. Property, Marriage Strategies, and
Family Dynamics in Ancient Athens, Princeton: Princeton University
Crawford, M. (2008), ‘States waiting in the wings: Population distribution and
the end of the Roman Republic’, in de Ligt, L. and Northwood, S. J. (eds),
People, Land and Politics. Demographic Developments and the Transformation
of Roman Italy, 300 BC – AD 14, Leiden: Brill, pp. 631–643.
Davies, J. K. (1977–1978), ‘Athenian citizenship: The descent group and the
alternatives’, The Classical Journal, 73, (2), 105–121.
Dixon, S. (1992), The Roman Family, Baltimore and London: Johns Hopkins
University Press.
Ta k i n g a W i d e r V i ew: G r e c o - R om a n Fa m i l i e s 43

Edmonson, J. (2005), ‘Family relations in Roman Lusitania: Social change in a

Roman province?’, in George, M. ed., The Roman Family in the Empire: Rome,
Italy and Beyond, Oxford: Oxford University Press, pp. 183–229.
Evans, J. K. (1991), War, Women and Children in Ancient Rome, London and
New York: Routledge.
Field, J. A. Jr. (1945), ‘The purpose of the Lex Julia et Papia Poppaea’, The
Classical Journal, 40, (7), 398–416.
Finley, M. I. (1951), Studies in Land and Credit in Ancient Athens, 500–200 B C.
The Horos-Inscriptions, New Brunswick NJ: Rutgers University Press.
Frank, R. I. (1975), ‘Augustus’ legislation on marriage and children’, California
Studies in Classical Antiquity, 8, 41–52.
Garland, R. (1987), The Piraeus. From the Fifth to the First Century BC, Ithaca
NY: Cornell University Press.
Gruber, S. and Pichler, R. (2002), ‘Household structures in Albania in the 20th
century’, The History of the Family, 7, (3), 351–374.
Hajnal, J. (1965), ‘European marriage patterns in perspective’, in Glass, D. V. and
Eversley, D. E. C. (eds), Population in History, Chicago: Aldine, pp. 101–143.
—(1982), ‘Two kinds of preindustrial household formation’, Population and
Development Review, 8, (3), 449–494.
—(1983), ‘Two kinds of preindustrial household formation’, in Wall, R., Robin, J.
and Laslett, P. (eds), Family Forms in Historic Europe, Cambridge: Cambridge
University Press, pp. 65–104 (abridged and slightly revised version of Hajnal,
Hansen, M. H. (1996), ‘City-ethnics as evidence for polis identity’, in Hansen,
M. H. and Raaflaub, K. (eds), More Studies in the Ancient Greek Polis. Historia
Einzelschriften, Stuttgart: Franz Steiner Verlag, pp. 169–196.
Hartman, M. (2004), The Household and the Making of History: A Subversive
View of the Western Past, Cambridge: Cambridge University Press.
Hin, S. (2009), The Dynamics of Life. Demography and the History of Roman
Italy (201 BC – 14 AD), Leiden: Leiden University PhD Dissertation.
Hionidou, V. (1995), ‘Nuptiality patterns and household structure on the Greek
island of Mykonos, 1849–1959’, Journal of Family History, 20, (1), 67–102.
Hopkins, M. K. (1978), Conquerors and Slaves. Sociological Studies in Roman
History 1, Cambridge: Cambridge University Press.
Horden, P. and Purchell, N. (2000), The Corrupting Sea. A Study of Mediterranean
History, Malden MA: Blackwell Publishing.
Huebner, S. R. (2011), ‘Household and Family in the Roman East and West’, in
Rawson, B. ed., A Companion to Families in the Greek and Roman Worlds,
Oxford: Blackwell Publishing, pp. 73–91.
Kertzer, D. I. (1991), ‘Household history and sociological theory’, Annual
Review of Sociology, 17, 155–179.
Kertzer, D. I. and Hogan, D. P. (1989), Family, Political Economy, and Demographic
Change. The Transformation of Life in Casalecchio, Italy, 1861–1921, Madison:
The University of Wisconsin Press.
44 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

Kertzer, D. I. and Saller, R. P. (1991), The Family in Italy from Antiquity to the
Present, New Haven: Yale University Press.
Laslett, P. (1977), Family Life and Illicit Love in Earlier Generations, Cambridge:
Cambridge University Press.
—(1983), ‘Family and household as work group and kin group: Areas of
­traditional Europe compared’, in Wall, R.; Robin, J. and Laslett, P. (eds),
Family Forms in Historic Europe, Cambridge: Cambridge University Press,
pp. 513–563.
Laslett, P. and Wall, R. (eds) (1972), Household and Family in Past Time:
Comparative Studies in the Size and Structure of the Domestic Group over
the Last Three Centuries in England, France, Serbia, Japan and Colonial
North America, with Further Materials from Western Europe, Cambridge:
Cambridge University Press.
Le Play, F. (1877–1879), Les Ouvriers Européens. Tomes 1–6, Tours and Paris:
Mame et Fils, Dentu and Larcher.
De Ligt, L. and Northwood, S. (eds) (2008), People, Land and Politics.
Demographic Developments and the Transformation of Roman Italy 300 BC
– AD 14, Leiden: Brill.
Lomas, K. (2000), ‘The polis in Italy: Ethnicity, colonization and citizenship
in the Western Mediterranean’, in Brock, R. and Hodkinson, S. (eds),
Alternatives to Athens. Varieties of Political Organization and Community in
Ancient Greece, Oxford: Oxford University Press.
—(2010), ‘Beyond Magna Graecia: Greeks and non-Greeks in France, Spain,
and Italy’, in Kinzl, K. H. ed., A Companion to the Greek World, Malden MA:
Malkin, I. (2002), ‘Exploring the Concept of Foundation: A Visit to Megara
Hyblaia’, in Gorman, V. and Robinson, E. (eds), Oikistes: Studies in
Constitutions, Colonies, and Military Power in the Ancient World, Leiden:
Brill, pp. 195–225.
Martin, D. B. (1996), ‘The construction of the ancient family: Methodological
Considerations’, Journal of Roman Studies, 86, 41–60.
Mathisen, R. W. (2006), ‘Peregrini, barbari, and cives Romani: Concepts of
citizenship and the legal identity of barbarians in the later Roman Empire’,
The American Historical Review, 111, 1011–1040.
Mentzer, R. A. (1994), ‘Review of Wally Seccombe’s A millennium of family
change: Feudalism to capitalism in Northwestern Europe’, Journal of
Interdisciplinary History, 24, (4), 695–696.
Mette-Dittmann, A. (1991), Die Ehegesetze des Augustus: Eine Untersuchung im
Rahmen der Gesellschaftspolitik des Princeps, Stuttgart: Franz Steiner Verlag.
Mogey, J. M. (1955), ‘The contribution of Frédéric Le Play to family research’,
Marriage and Family Living, 17, (4), 310–315.
Oppo, A. (1990), ‘“Where there’s no woman, there’s no home”: Profile of the
agro-pastoral family in 19th century Sardinia’, Journal of Family History, 15,
(4), 483–502.
Ta k i n g a W i d e r V i ew: G r e c o - R om a n Fa m i l i e s 45

Osborne, M. J. (1981–1983), Naturalization in Athens (4 Vols), Brussels:

Koninklijke Academie voor Wetenschappen, Letteren en Schone Kunsten.
Osborne, M. J. and Byrne, S. (1996), The Foreign Residents of Athens. An Annex
to the Lexicon of Greek Personal Names: Attica, Studia Hellenistica 33, Leuven:
Peeters Publishers.
Osborne, R. (1985), Demos. The Discovery of Classical Attika, Cambridge:
Cambridge University Press.
Peristiany, J. G. ed. (1976), Mediterranean Family Structures, Cambridge:
Cambridge University Press.
Pomeroy, S. B. (1997), Families in Classical and Hellenistic Greece. Representations
and Realities, Oxford: Clarendon Press.
Reher, D. S. (1997), Perspectives on the Family in Spain, Past and Present,
Oxford: Clarendon Press.
Rettaroli, R. (1990), ‘Age at marriage in nineteenth century Italy’, Journal of
Family History, 15, (4), 409–425.
Roselaar, S. T. (2010), Public Land in the Roman Republic. A Social and Economic
History of Ager Publicus in Italy, 396–89 BC, Oxford: Oxford University Press.
Rowland, R. (2002), ‘Household and family in the Iberian Peninsula’, Portuguese
Journal of Social Science, 1, (1), 62–75.
Roy, J. (1999), ‘Polis and oikos in classical Athens’, Greece & Rome, 46, 1–18.
Sallares, R. (1991), The Ecology of the Ancient Greek World, Ithaca NY: Cornell
University Press.
Saller, R. P. (1994), Patriarchy, Property and Death in the Roman Family,
Cambridge: Cambridge University Press.
—(2001), ‘Family values in ancient Rome’, online at: http://fathom.lib.uchicago.
edu/1/777777121908/ (Accessed 10 October 2010).
Saller, R. P. and Shaw, B. (1984), ‘Tombstones and Roman family relations in the
Principate: Civilians, soldiers, and slaves’, JRS, 74, 124–156.
Scafuro, A. C. (1994), ‘Witnessing and false witnessing: Proving citizenship and
kin identity in fourth-century Athens’, in Boegehold, A. L. and Scafuro A.
C. (eds), Athenian Identity and Civic Ideology, Baltimore and London: Johns
Hopkins University Press, pp. 156–198.
Schaps, D. (1975), ‘Women in Greek inheritance law’, The Classical Quarterly,
25, (1), 53–57.
Scheidel, W. (2007), ‘Roman funerary commemoration and the age at first
marriage’, Classical Philology, 102, (4), 389–402.
Seccombe, W. (1992), A Millennium of Family Change: Feudalism to Capitalism
in Northwestern Europe, London and New York: Verso.
Shaw, B. D. (2001), ‘The seasonal birthing cycle of Roman women’, in Scheidel,
W. ed., Debating Roman Demography, Leiden: Brill, pp. 83–110.
Sherwin-White, A. N. (1973), Roman Citizenship, (2nd edn), Oxford: Oxford
University Press.
Shipley, G. (2005), ‘Little boxes on the hillside: Greek town planning,
Hippodamos, and polis ideology’, in Hansen, M. H. ed., The Imaginary Polis.
46 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

Acts of the Copenhagen Polis Centre Vol. 7, Copenhagen: Det Kongelige

Danske Videnskabernes Selskab, pp. 335–403.
Smith, D. S. (1993), ‘The curious history of theorizing about the history of the
Western nuclear family’, Social Science History, 17, (3), 325–353.
Smith, R. (1981), ‘The people of Tuscany and their families in the fifteenth
century: Medieval or Mediterranean?’, Journal of Family History, 6, 107–128.
Sovič, S. (2008), ‘European family history. Moving beyond stereotypes of “East”
and “West”’, Cultural and Social History, 5, (2), 141–163.
Stangeland, C. E. (1966), Pre-Malthusian Doctrines of Population. A Study in the
History of EconomicTheory, (2nd edn), New York: August M. Kelley.
Szołtysek, M. (forthcoming), Rethinking Eastern Europe: Family Structure and
Co-Residence in the Polish-Lithuanian Commonwealth (17th-19th Centuries),
Series: Population, Family, and Society, Bern: Peter Lang.
Szołtysek, M. and Zuber-Gold-Stein, B. (2009), ‘Historical family systems and the
great European divide: The invention of the Slavic East’, Demografia, 52, (5), 5–47.
Tacoma, L. E. (2006), Fragile Hierarchies. The Urban Elites of Third-Century
Roman Egypt, Leiden: Brill.
—(2007), ‘Migration and the Roman family in the first two centuries AD’,
Working Paper. Available online at:
(Accessed 10 October 2010).
Teitelbaum, M. S. (2006), ‘Political demography’, in Poston, D. L. and Micklin
M. (eds), Handbook of Population, New York: Springer, pp. 719–730.
Thornton, A. (2005), Reading History Sideways. The Fallacy and Enduring
Impact of the Developmental Paradigm on Family Life, Chicago and London:
The University of Chicago Press.
Tilly, C. (1993), ‘Review of Wally Seccombe’s A millennium of family change:
Feudalism to capitalism in Northwestern Europe’, The American Historical
Review, 98, (3), 833–834.
Todd, E. (1985), The Explanation of Ideology. Family Structures and Social
Systems, Oxford: Basil Blackwell Publishing.
Treggiari, S. (1991), Roman Marriage. Iusti Coniuges from the Time of Cicero to
the Time of Ulpian, Oxford: Clarendon Press.
Vérilhac, A.-M. and Vial, C. (1998), ‘Endogamie et exogamie’, in Vérilhac A.-M.
and Vial C., Le Mariage Grec du VIe Siècle avant J. C. à l’Époque d’Auguste,
Bulletin de Correspondance Hellénique, Supplement 32, Athens and Paris:
École Française d’Athènes and De Boccard, 41–124.
Viazzo, P. (2003), ‘What’s so special about the Mediterranean? Thirty years of
research on household and family in Italy’, Continuity and Change, 18, (1),
Viazzo, P. and Albera, D. (1990), ‘The peasant family in northern Italy,
1750–1930: A reassessment’, Journal of Family History, 15, (4), 461–482.
Wagstaff, M. (2001), ‘Family size in the Peloponnese (Southern Greece) in 1700’,
Journal of Family History, 26, 337–349.
Ta k i n g a W i d e r V i ew: G r e c o - R om a n Fa m i l i e s 47

Weaver, P. R. C. (1986), ‘The status of children in mixed marriages’, in Rawson,

B. ed., The Family in Ancient Rome. New Perspectives, Ithaca NY: Cornell
University Press, pp. 145–169.
Weiner, M. (1971), ‘Political demography: An inquiry into the political conse-
quences of population change’, in National Academy of Science, Office of
the Foreign Secretary, Rapid Population Growth: Consequences and Policy
Implications, Baltimore: Johns Hopkins University Press, pp. 567–617.
Weiner, M. and Jeitelbaum, M. S. (2001), Political Demography, Demographic
Engineering, New York and Oxford: Berghahn Books.
Whitehead, D. (1977), The Ideology of the Athenian Metic, Cambridge: The
Cambridge Philological Society.
—(1986), The Demes of Attica 508/7 – ca. 250 B. C.: A Political and Social Study,
Princeton NJ: Princeton University Press.
Zuiderhoek, A. J. (2006), Citizens, Elites, and Benefactors: The Politics of Public
Generosity in Roman Asia Minor, Groningen: Groningen University PhD

More than Just Gender: The Classical Oikos as a Site

of Intersectionality
Birgitta L. Sjöberg

The aim of this chapter, which forms part of a larger research project, is to
highlight the status of different members of the family in Classical antiquity,
and their interactions within the private realm of the household and outside
the oikos with society in the public realm. Underpinning the paper is the
paradigm of intersectionality, which allows us to analyse how characteristics
such as gender, class, ethnicity, sexuality and age relate to each other and
influence outcomes, including markers of difference within the Classical oikos.
The key question for the paper is how multiple discrimination in the oikos was
constructed along the lines of the social categories: gender, class, age, ethnicity
and sexuality. For this chapter, I will illustrate the usefulness of this perspective
based on intersectionality with reference to two literary sources: Works and
Days by Hesiod and Oeconomicus by Xenophon, which both discuss the same
subject, the Greek oikos.

Intersectionality as an Analytical Tool

Intersectionality as a model can be traced back to the 1960s and 1970s when
scholars studying women of colour first employed the concept. It has since
attracted interest not only among scholars devoted to radical black feminism
but also within several other fields of research. The concept was coined by
Kimberlé Crenshaw, and has subsequently often seen use in research on
subjects linked to ethnicity and women’s studies.1 The versatility of the concept
is today displayed by its use not only in gender studies but several other fields
such as education, law, sociology and theology, and it has been taken up also
by European scholars representing a number of other fields of research.2 While
intersectionality has a wide appeal and use, it might not contribute to a precise
definition. However, intersectionality’s broad applicability is also the major
advantage of the concept. We may agree that what intersectionality does, rather
M o r e t ha n J u st G e n d e r : T h e C l a s sic a l Oikos 49

than discussing social hierarchies of gender, class, ethnicity, sexuality and age as
isolated phenomena, is to allow us to investigate how they mutually construct
one another.3 Intersectionality can therefore be regarded as an analytical tool for
understanding multiple discriminations created by the intersection of different
categories present in the oikos including: gender, class, ethnicity, sexuality and
The usefulness of intesectionality as an analytical tool within archaeology
has been explored by Margaret Conkey, who discusses the intersection between
feminist archaeology and indigenous archaeology.4 Within the field of Classical
studies the concept has, so far, been rather more neglected. Yet, as argued in
a recently constructed educational platform, the Intersectionality Tool Box
(ITB),5 intersectionality is applicable to research questions related to gender
within the Classical world. Henrik Berg discusses how Athenian masculinities
were constructed and suggests that a perspective drawing on intersectionality
can lay bare the construction of male identities.6 Such encouraging remarks
provide justification enough to move forward. Thus, with a view to illus-
trating the multifaceted interaction between different groups within ancient
Greek society, in the following discussion, intersectionality will be applied for
analytical ends, focusing on the members of the ancient Greek oikos.

How to Study Intersectionality?

How can intersectionality be studied within the realm of the Classical oikos?
Especially as pointed out by Lin Foxhall there is a limitation concerning
contexts that can be explored, because the sources are limited and produced
by free, adult, male citizens.7 Or, as pointed out by Marilyn Katz, when female
related issues are discussed, there is a tendency to focus on citizen women,
something that may be explained by the fact that hardly any information
survives about women of other classes.8 The concept of intersectionality has
seen frequent and wide use in women’s studies, but as pointed out by Leslie
McCall, methodology has been rather neglected in the academic discussion
of intersectionality. She suggests that it is necessary to use a greater range of
categories to understand the intersection within and between groups.9 This is
also relevant to the study of the Classical world, and an early application of ITB
concerns the analysis of iconographical material, primarily constituted by black
and red figure Classical vases.10
The starting point for any study of intersectionality is that a category of
identity takes its meaning as a category in relation to another category. As
such, this approach highlights the intersection between groups that go beyond
gender relationships between men and women. By applying intersectionality
to various sub-categories beyond the male–female dichotomy, it transpires that
there are situations or relations where the female can also be the dominant
party. On this a priori basis, it becomes clear that by applying the paradigm of
50 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

intersectionality to sub-categories other than the usual ones associated with a

single category, for example gender or age, we are able to extend our knowledge
of how oppression and discrimination was constituted within the Classical oikos
and in the Classical world more generally.
In addition to the original focus on the role of gender, therefore, we should
avail ourselves of the possibility of identifying sub-categories in the literary
sources at our disposal. While it may serve to make ancient society stand out
as more complex than indicated by simple dichotomies, in the end we will be
rewarded with a fuller understanding on how social relations were constituted.
As a further help in this quest, one may also contemplate the use of parallel or
contrasting cases across time and space, be they ancient or modern, drawn from
the Aegean world or elsewhere.

The Classical Oikos

The Classical oikos occupies a principal position within studies of Athenian
society and the concept has for long been an issue for discussion.11 The notion is
strongly associated with values connected to social hierarchies and ideological
constructions. The oikos was not only fundamental to social organization but
also necessary for the ideological construction of the polis. According to Fisher
the family was the basic unit and the primary focus of the activities and interests
of men and women. Both prosperity and honour were intimately bound up with
the concept. The values and the importance of the honour of the family may have
even been greater than that of the polis and the laws.12 Also Lin Foxhall agrees
that the oikos was fundamental for the ancient Greek society; however, she also
demonstrates the complexity of political and private roles with every Athenian
belonging to both polis and oikos, entwined causing problems.13 The impor-
tance and centrality of the oikos is according to Cynthia Patterson displayed by
Demosthenes in his case against Neaira where the legitimate marriage is the
important factor linking the Greek family to the polis. According to Patterson
the public significance of the family should not be neglected.14 Sarah Pomeroy
defines the oikos as the basic unit of the Greek society and the polis was a
community of oikoi rather than individual members. Pomeroy also regards the
Oeconomicus, by Xenophon, as a reflexion of normative and idealized thoughts
on the Greek household.15 The distinction between oikia and oikos has in legal
contexts been observed by Douglas M. MacDowell who associates oikos with
‘property’ or ‘family and oikia with ‘house’.16 The concept of oikos is, within the
current research project, defined as an extended family unit, understood as
including also the non-citizen persons, and the property that was part of this
family unit.
The oikos was vital for the survival of the fundamental dichotomy of
Athenian society: the house as the space for the private life of the citizen was
associated with women, whereas the male members of the oikos were part of
M o r e t ha n J u st G e n d e r : T h e C l a s sic a l Oikos 51

the public, masculinized sphere.17 Gender relations and the spatial division
within the realm of the oikos are illustrated in Classical texts by authors such as
Xenophon and Lysias. Later, the Roman architect and author Vitruvius pointed
out as a specific phenomenon characterizing the Greek house was an area called
the women’s quarter, gynaeconitis. The attitude displayed by Classical and later
authors was, according to Ian Morris, no new invention after all it is possible
to find already in the work by Hesiod who in Works and Days described how
the young female member of the house stays indoors with her mother in the
inner room.18 This, essentially normative, picture of the social life within the
Greek house has been dominant within research. As pointed out by Lisa Nevett,
the use of andron and gunaikôn has been interpreted as evidence of gender
specific locations within the house. According to her analysis, the problem is
rather more complex than this and she concludes that a possible way to look
behind the stereotyped picture of Greek houses is through a more detailed
study of archaeological remains. As she goes on to argue, physical remains and
artefacts are more appropriate to use as evidence of the actual organization of
Greek households, whereas the textual evidence has misled us.19 The emerging
picture of economic stratification, as made evident in the housing pattern,
could therefore be of some significance.20 This conclusion with respect to
economic stratification displayed in architecture is of importance to our wider
study. For now, let it suffice to note that the economic status of the household
can be assumed to be of importance as we proceed to applying the paradigm of
intersectionality to the material at hand. The hypothesis that the level of inter-
sectionality within the oikos is connected to the economic status of the house
will be made subject to further inquiry. In short, we would expect that intersec-
tionality should be more pronounced within wealthier households. This means
that the size and spatial complexity of a house cause intersectionality to vary.

Intersectionality and the Classical Oikos

The earliest written source on the function of the oikos and the obligations and
status of the individual members of the family is described in the Archaic epic
text Works and Days, by Hesiod. The relations within the oikos of a peasant
living in the Boeotian countryside, described by Hesiod in this early work
created around 700 BC where he gives advice to his brother, can be contrasted,
not least with respect to issues of social status, with that contained in the
Oeconomicus (written by Xenophon in the fourth century BC). In the latter, we
follow the discussion between Socrates and his friend Critobulus, two repre-
sentatives of the upper social class, on estate management which include all
members of the oikos, irrespective of status. The text by Xenophon is regarded
as an important and didactic contribution to our understanding of several
aspects encompassing the Classical oikos. Xenophon was in exile in Sparta
when he wrote about the management of the Athenian oikos. Also, it needs to
52 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

be remembered that the Oeconomicus describes a fourth century rather than

a fifth century oikos.21 The images projected by Hesiod and Xenophon may be
idealized versions of the Greek family group and chronologically misaligned.
Even so, it can be argued that both these texts can be used as important sources
for our understanding of intersectionality and multiple levels of discrimination.
Ian Morris has suggested that the transformation of house design between
750–600 BC display hardening gender ideologies.22 In the following it is argued
that by applying the concept of intresectionaly these attitudes are displayed on
several levels already in the works by Hesiod. In his Works and Days, the inter-
section between gender, sexuality and age is made visible in the relationships
between the shy virgin to be married to a man – most probably many years her
senior. A first instance to be observed, in addition to that of age and gender, is
the creation of the virgin, ordered and controlled by the elder heavenly ruler
Zeus.23 Further, Hesiod in the Theogony wrote of the ethnicity of women.
Hesiod is very clear that Pandora is the ancestor of evil women, the scourge of
all men on earth.24 Indeed, in both epics by Hesiod, women are unambiguously
described as the source of problems affecting men and the earth in different
ways. Hesiod describes how hard man’s life is and how he should avoid poverty
and misery and that all these disasters that affected men on earth came to be
permanent after the acceptance of a young Pandora as a gift to Epithemeus.
This acceptance was equal to a radical change of life on earth, previously lived
without evils and harsh labour or cruel disasters and aging. But when Pandora
lifted the lid from the pithos she scattered these evils about and devised misery
for men and earth.25 According to Lillian E. Doherty it is remarkable that
Hesiod does not call Pandora the mother of humans but only of genos gunaikôn,
the ‘race of women’.26 Indeed, the notion of genos gunaikôn, if interpreted as
identification of women as a group not part of a society with the male as norm,
we may read another category of intersection, that is ethnicity. The use of genos
gunaikôn may be positioned due to the negative view among Greeks of the
‘other’ as barbarians, and in the case of Hesiod women are synonymous with the
‘other’. From making this observation, we may read how Archaic women were
simultaneously positioned as virgins, women, young and ethnically different,
and thus woman became the locus of the multiple intersection of gender/age/
sexuality/ethnicity. In short, the texts by Hesiod make clear the low status that
even married women probably had and the multiple dimensions of discrimi-
nation of her in Archaic society.
Traditionally, the ideal Classical family has been regarded as an entity
formed through marital and blood ties, composed of a married couple and
their own biological children and the protection of the bloodline was therefore
of major interest for the whole society.27 We may find the importance of a son
to manage the estate inherited already in the Archaic epic Works and Days and
later in Oeconomicus, in both of which the importance of children as support
for ageing parents is mentioned.28 It is interesting to note that both Hesiod and
Xenophon mention the combination of slaves and children in aiding the elderly.
M o r e t ha n J u st G e n d e r : T h e C l a s sic a l Oikos 53

Hesiod advises the peasant not to employ a maid with a child, as a nursing
mother is an obstacle.29 In the Oeconomicus, we find a description of the house
as divided into women’s and men’s quarters and an equally important lesson is
to keep slaves separated so as not to give them the possibility to breed without
the permission of their master. The control was apparently seen as necessary;
according to the didactic text, honest slaves became more loyal if they had
children, whereas bad ones became more problematic.30 In other words, here
we find an intersection constructed by gender, age and class displaying multiple
discrimination of un-free women in particular and we may conclude that this
is irrespective of whether we speak of Archaic or the later Classical society.
Both texts suggest that marriage was formed between two individuals. From
Hesiod’s Works and Days, we learn that the suitable age for a man to be married
was around 30, whereas the appropriate age for a girl was in her teens. She
should also be a virgin – giving her husband the possibility to teach her virtuous
behaviour. In Hesiod’s description the husband oversees the activities of the
oikos including his young wife.31 The advice is possible to follow also in the text
by Xenophon who writes that the wife of Ischomachus was not even 15 when
she came into the house, and ‘she had spent her previous years under careful
supervision so that she might see and hear and speak as little as possible’.32
The intersection of age, sexuality and gender is obvious and we find that both
Hesiod and Xenophon describe a setting where marriage was between two
individuals of very different age and status.
Sexuality as a subject is avoided neither in Hesiod nor Xenophon and the
moral behaviour of the wife is emphasized in both of the didactic texts. We
need not doubt that the reputation of the girl as well as her behaviour was of
importance already in Archaic society and the opinion of the neighbours is
apparently not to be neglected.33 Also in the later work, Oeconomicus, we find
similar sentiments: Ischomachos, the husband, does not quite approve of his
young wife powdering her face and using plenty of rouge. Ischomachos is quite
clear that this is not acceptable for a wife and he advises her not to spend her
time like a slave; instead she should stand before the loom and supervise the
household.34 The association of good virtues and the loom is already found in
the Archaic society, represented by Works and Days,35 and we may observe this
attitude displayed in Classical vase painting where the mistress of the house
is depicted with a loom, often surrounded by un-free women. The control of
female sexuality is fundamental not only to the oikos but to society as a whole
and it is clear that the sexuality of the honoured wife should be distinguished
from the sexuality of un-free women. The difference in views concerning the
sexuality of free and un-free women is easily seen in Xenophon, where it is
demonstrated that the slave has little choice but to submit.36 The intersection-
ality registered in these passages is composed of age, gender, sexuality but also
ethnicity and class as slave women were non-citizens and un-free.
54 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

Intersectionality and the Greek Oikos: How Do

We Proceed?
According to Sarah Pomeroy, the Oeconomicus represents the changing view
of women in the fourth century where we find better educated women taking
part in the liberal arts and professions. In other words, women began to appear
as artists, philosophers and so forth. The Oeconomicus was a story about the
education of a young girl, now married and expected to take responsibility of
the oikos. The text also indicates that within marriage the status of women seems
to be more equal and the mutual goal of husband and wife was the increase of
property.37 Pomeroy considers the Oeconomicus to mirror a view of the oikos
as site released from ‘natural hierarchy among human beings according to
gender, race, or class’.38 Xenophon expresses, according to her, a view that men,
women and slaves, ‘potentially have the same virtues, vices and talents’.39 The
importance of the Oeconomicus for our understanding of the oikos and the
world outside the estate is also emphasized by Sture Linnér in his comments to
a Swedish translation of the Oeconomicus. We find, according to Linnér, a text
that features a set of very conservative values but also a measure of radicalism.
The virtues of the wife are primarily a consequence of Ischomachos teaching
her, the importance of which rests with the fact that he will benefit. However,
the wife is treated with respect and her importance for the supervision of the
economy of the household is emphasized. An interesting comment made by
Linnér is that by extending the responsibility of his wife it follows that the
husband was able to spend more time outside the oikos, located in public life
where the important decisions were made.40 On the other hand, might it not be
the case that the extended responsibility and authority of the wife was rather
a result of the husband making public life a priority? By so doing, he simply
has to grant his wife greater leeway, towards which end his teaching her the
relevant virtues was rather a way of ensuring that his authority was maintained.
Returning to Linnér’s analysis, he also makes the observation that the wife is
not given a name and she is not regarded as an individual: instead the wife is
an anonymous person in the text. As Linnér aptly notes, the seemingly realistic
description of an equal relationship between the wife and husband of the oikos,
is instead an idealized vision of society.41
The value of these two sources rests on the possibility that the two texts
display traditions connected to the organization and survival of the family
under rather different circumstances. Put differently, the manner in which the
household operated may have changed over time as we move from Archaic to
Classical times – and that these circumstances changed (presumably slowly)
irrespective of the social status of the household. Therefore, the question of
whether we may identify different attitudes connected to wealth and status
may provisionally be answered in the positive. It is, after all, quite obvious that
both the wife and husband in Hesiod’s Archaic family were forced to work hard
M o r e t ha n J u st G e n d e r : T h e C l a s sic a l Oikos 55

in order to ensure the survival of the family. There was little or no time for
pleasures such as we find in the Oeconomicus, where the couple seems to have
had time for other duties than those associated with hard work in the fields.
Yet such conjectures, no matter how likely, cannot be established beyond
doubt. It would therefore be of more than passing interest to pursue a wider
range of literary texts. Thus, for the wider project on intersectionality currently
in progress, the texts under investigation range from Archaic lyric, although
very fragmentary, such as Semonides, Anacreon and Sappho, to Classical
works representing different literary genres. For the Archaic period Hesiod is,
of course, a major source allowing us to approach the family values of that time
in greater detail as illustrated above – and he has few peers. This observation
must not be allowed to conceal that other sources exist. For instance, despite
its normative orientation the legislation of Solon, as related by Plutarch, is of
significance to our understanding of how Archaic society worked. Valuable
Greek historical texts that relay information on status and relations from later
periods include, in addition to Xenophon, works by Thucydides and Herodotus.
Furthermore, the philosophical texts by Aristotle and Plato are immensely inter-
esting as documents of moral values predominant within the aristocratic circles
in the fourth century. Like Greek comedy, Greek orators such as Aeschines,
Demosthenes and Lysias allow ample insight into the life of individual members
of society, not only young upper-class rascals drifting on the streets of Piraeus
insulting both women and men, but also the life of poorer citizens, non-citizens
(metoikoi), slaves, young and old. The information contained in the speeches is
naturally subjective in nature and formulated with the intention to persuade.
However, the information contained therein is of considerable interest as we
may get a glimpse not only of the laws of interest in the specific case, but we
may also get closer to the persons involved and thereby gain insights into the
status of the individuals involved. Here intersectionality affords a possibility
of applying a holistic approach where the different categories, gender, age,
sexuality, class and ethnicity meet within the oikos. As such it facilitates our
attempts to identify relationships of inequality within the Classical oikos and
the Classical polis.
For now, however, suffice it to note that already the limited examples
approached in this short article has served to extend our knowledge on, and
understanding of, inequality and discrimination within the oikos. As hierar-
chical as the ancient household may have been, it cannot be reduced to a
discussion of the issue of gender alone. Instead, by considering gender in
relation to other social identities – such as class, ethnicity, sexuality and age –
and how they mutually construct one another much can be gained to enrich our
understanding of the oikos.
56 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

  1 Crenshaw, 1989; 1991 for original concept from which developed among other Hill Collins,
1998; Flores, 2000 and Ringrose, 2007.
  2 For example Lykke, 2003; Lewander, 2004; McMullin and Cairney, 2004; De los Reyes and
Mulinari, 2005; Phoenix, 2006; Shields, 2008; Roos, 2008; Sjöberg, 2008a and Valentine, 2007.
  3 Hill Collins, 1998, p. 63.
  4 Conkey, 2005.
  5 Sjöberg, 2008b.
  6 Berg, 2010.
  7 Foxhall, 1998, p. 123.
  8 Katz, 1992, pp. 79–80.
  9 McCall, 2005, pp. 1784–1785.
10 Sjöberg, 2008b.
11 Foxhall, 1989; Cartledge, Millett and Todd, 1990; Cohen, 1991; Boegehold and Scafuro, 1994;
Gardner, 1996; Schnurr-Redford, 1996; Foxhall and Lewis, 1996; Harrison, 1998; MacDowell,
1989 and Cox, 1998. See Katz, 1992 for principal research questions.
12 Fisher, 1976.
13 Foxhall, 1989, p. 22, 43.
14 Patterson, 1994, pp. 210–211.
15 Pomeroy, 1994, pp. 31, 33, 41.
16 MacDowell, 1989, pp. 10–11, 21–22.
17 Cohen, 1991, pp. 70–97 and Pomeroy, 1997, pp. 29–31.
18 Morris, 1999, pp. 307–308.
19 Nevett, 1995; 1999.
20 Nevett, 1995.
21 Pomeroy, 1994, pp. 5, 31–33.
22 Morris, 1999, pp. 306, 311.
23 Hesiod, Theogony, 570–590 and Hesiod, Works and Days, 70–85.
24 Hesiod, Theogony, 590–591.
25 Marquardt, 1982, pp. 287–288 and Lefkowitz, 1986, pp. 113–116.
26 Doherty, 2001, p. 136.
27 Ogden, 1997.
28 Hesiod, Works and Days, 376–377 and Xenophon, Oeconomicus VII, 12–13.
29 Hesiod, Works and Days, 602–604.
30 Xenophon, Oeconomicus IX, 5–6.
31 Hesiod, Works and Days, 695–700.
32 Xenophon, Oeconomicus VII, 4.
33 Hesiod, Works and Days, 695–700.
34 Xenophon, Oeconomicus X, 2–13.
35 Hesiod, Works and Days, 779.
36 Xenophon, Oeconomicus X, 13.
37 Pomeroy, 1994, p. 58
38 Ibid., p. 66.
39 Ibid.
40 Linnér, 2003, pp. 26–28.
41 Ibid.

Berg, H. (2010), Constructing Athenian Masculinities. Masculinities in Theophrastus’
Characters and Meander’s Comedies, unpublished thesis, Uppsala University.
M o r e t ha n J u st G e n d e r : T h e C l a s sic a l Oikos 57

Boegehold, A. and Scafuro, A (eds) (1994), Athenian Identity and Civic Ideology,
London: John Hopkins University Press.
Cartledge, P., Millett, P. and Todd, S. (eds) (1990), Nomos: Essays in Athenian
Law, Politics and Society, Cambridge: Cambridge University Press.
Cohen, D. (1991), Law, Sexuality, and Society. The Enforcement of Morals in
Classical Athens, Cambridge: Cambridge University Press.
Conkey, M. (2005), ‘Dwelling at the margins, action at the intersection?
Feminist and indigenous archaeologies’, Archaelogies, 1, (1), 9–59.
Cox, A. (1998), Household Interests. Property, Marriage Strategies, and Family
Dynamics in Ancient Athens, Princeton: Princeton University Press.
Crenshaw, K. (1989), ‘Demarginalising the intersection of race and sex: A
Black feminist critique of antidiscrimination doctrine, feminist theory, and
antiracist politics’, Chicago Legal Forum, 139–67.
—(1991), ‘Mapping the margins: Intersectionality, identity politics and violence
against women of color’, Stanford Law Review, 43, (6), 241–99.
De los Reyes, P. and Mulinari, D. (2005), Intersektionalitet. Kritiska Reflektioner
över (O)jämlikhetens Landskap, Malmö: Liber.
Doherty, L. E. (2001), Gender and the Interpretation of Classical Myth, London:
Fisher, N. ed. (1976), Social values in classical Athens, The Ancient World,
Flores, L. (2000), ‘Reclaiming the ‘other’ toward a Chicana feminist critical
perspective’, International Journal of Intercultural Relations, 24, 687–705.
Foxhall, L. (1989), ‘Household, gender and property in classical Athens’,
Classical Quarterly, 39, 22–44.
—(1998), ‘Pandora outbound: A feminist critique of Foucault’s History of
Sexes’, in Larmour, D. H. J., Miller, P. A. and Platter, C. (eds), Rethinking
Sexuality. Foucault and Classical Antiquity, Princeton: Princeton University
Press, pp. 122–137.
Foxhall, L. and Lewis, A. (1996), Greek Law in its Political Setting: Justification
not Justice, Oxford: Clarendon Press.
Gardner, J. (1996), ‘Aristophanes and male anxiety; the defence of the oikos’, in
McAuslan, I. and Walcot P. (eds), Women in Antiquity, Greece and Rome 3,
New York: Oxford University Press.
Harrison, A. (1998), The Law of Athens, Vols 1 and 2, (new edn), London:
Hesiod, Theogony, (translated by Björkseson, I., 2003), Theogonin och Verk och
Dagar, Stockholm: Natur och kultur.
—Works and Days, (translated by Björkseson, I., 2003), Theogonin och Verk och
Dagar, Stockholm: Natur och kultur.
Hill Collins, P. (1998), ‘It’s all in the family: Intersections of gender, race, and
nation’, Hypatia, 13, (3), 62–82.
Katz, M. (1992), ‘Ideology and “The Status of Women” in Ancient Greece’,
History and Theory, 31, (4), 70–97.
58 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

Lefkowitz, M. R. (1986), Women in Greek Myth, London: Duckworth.

Lewander, L. (2004), Polariseringens Politik. Studier av Nation och Kön, Karlstad
University Studies, 51.
Linnér, S. (2004), Translation and comments, Xenofon Ledarskap, Stockholm:
Lykke, N. (2003), ‘Intersektionalitet-ett användbart begrepp för genusforskning’,
Kvinnovetenskaplig Tidskrift, 1, (3), 47–57.
MacDowell, D. (1989), ‘The oikos in Athenian law’, Classical Quarterly, 39, (1),
Marquardt, P. (1982), ‘Hesiod’s ambiguous view of women’, Classical Philology,
77, 283–291.
McCall, L. (2005), ‘The complexity of intersectionality’, Signs: Journal of women
in culture and society, 30, (3), 1771–1799.
McMullin, J. and Cairney, J. (2004), ‘Self-esteem and the intersection of age,
class and gender’, Journal of Aging Studies, 18, 75–90.
Morris, I. (1999), ‘Archaeology and Gender Ideologies in Early Archaic Greece’,
Transactions of the American Philological Association, 129, 305–317.
Nevett, L. (1995), ‘Gender relations in the classical Greek household: The
archaeological evidence’, Annual of the British School of Athens, 90, 363–381.
—(1999), House and Society in the Ancient Greek World, Cambridge: Cambridge
University Press.
Ogden, D. (1997), ‘Rape, adultery and the protection of bloodlines in classical
Athens’, in Deacy, S. and Pierce K. F. (eds), Rape in Antiquity, London: The
Classical Press of Wales (Duckworth).
Patterson, C. (1994), ‘The case against Neaira’, in Boegehold, L. and Scafuro,
A. C. (eds), Athenian Identity and Civic Ideology, London: John Hopkins
University Press, 199–216.
Phoenix, A. (2006), ‘Intersectionality’, European Journal of Women’s Studies, 13,
(3), 87–92.
Pomeroy, S. (1994), Xenophon Oeconomicus. A Social and Historical
Commentary, Oxford: Clarendon Press.
—(1997), Families in Classical and Hellenistic Greece. Representations and
Realities, Oxford: Clarendon Press.
Ringrose, J. (2007), ‘Troubling agency and “choice”: A psychosocial analysis
of students’ negotiations of Black feminist “intersectionality” discourse in
women’s studies’, Women Studies International Forum, 30, 264–78.
Roos, L. (2008), Att studera Medeltiden, Lund: Studentlitteratur.
Schnurr-Redford, C. (1996), Frauen im Klassischen Athen. Sozilaler Raum und
Reale Bewegungsfreiheit, Berlin: Akad.-Verlag.
Shields, A. (2008), ‘Gender. An Intersectionality Perspective’, Sex Roles, 59,
Sjöberg, L. B. (2008a), Rapport för NSHU-Projekt– För Utvecklingsprojekt Startade
2006. Pedagogisk och Metodologisk Verktygslåda för Intersektionalitetsstudier inom
Antikämnet, Institutionen för arkeologi och antik historia, Uppsala universitet.
M o r e t ha n J u st G e n d e r : T h e C l a s sic a l Oikos 59

—(2008b), ITB:

Valentine, G. (2007), ‘Theorizing and researching intersectionality: A challenge
for feminist geography’, The Professional Geographer, 59, (1), 10–21.
Xenophon, Oeconomicus, translated by Pomeroy, S. (1994), Xenophon
Oeconomicus. A Social and Historical Commentary, Oxford: Clarendon Press.

Tracing the Oikos in Pre-Classical Corinth: The

Perspective of Iconography
Amalia Avramidou

Scholars have always turned to Greek vase painting for reflections of the
ancient world, because it offers the best visual evidence of mythological, ritual
and every day life scenes. Of all Greek pottery wares, Attic receives the most
attention, while Corinthian has been less thoroughly explored, perhaps because
of its miniaturesque style and preference for animal friezes. In this light, it
comes as no surprise that studies on the iconography of women and children in
pre-Classical Corinth are not as advanced as in the case of Athens.1 Admittedly,
the number of Corinthian vases with domestic scenes is small compared to
the total ceramic production. Nevertheless, when these vases are examined
within context and in connection to other works of art, they allow us to break
away from an Athenocentric model of the Greek oikos in order to form a better
understanding of the Archaic Corinthian family.
Even though not exhaustive, the material of this preliminary study gives us
a good idea of the type of objects associated with representations of women
and children in the Archaic period. It dates mainly to the sixth century BC and
consists of about 30 vases, a wooden panel from the Pitsas cave, and more than
a dozen terracotta figurines and molds (see Tables 5.1, 5.2). The majority of
Corinthian vases depicting a child date from the first half of the sixth century,
and more specifically in Amyx’s Middle Corinthian and Late Corinthian period
(595/0–570 and 570–550 BC).2 Only a handful date to either before or after
this period. The most popular shape was the pyxis and the bottle, followed by
alabastra, aryballoi, amphoriskoi, oinochoai and column kraters. Most vases
with a known provenance originate from Corinth and the nearby sanctuary of
Perachora, although a few have turned up in Etruria. Regarding the Archaic
terracotta kourotrophoi and children, they are limited in number and consist
of both handmade and hollow examples of standing and seated kourotrophoi,
some grotesque figurines associated with pregnancy, and a couple of isolated,
possibly early examples of the so-called temple-boy type.3
This paper presents the most characteristic iconographic schemata of women
and children in pre-Classical Corinthian iconography and suggests a reason for
their increased appearance during the first half of the sixth century. For organiza-
tional purposes, I have structured the material in the following groups: women with
children on their laps, kanephoroi, wedding related episodes, and boys and youths.
T r ac i n g t h e Oikos i n P r e - C l a s sic a l C o r i n t h 61

Woman with Child on Her Lap

The motif of a woman holding a child occurs on a few Corinthian figurines and
almost a third of our catalogued vases. Starting with the terracottas, the figurine
from the Anaploga well at Corinth is perhaps the earliest of the kourotrophos type,
while three more examples were found in the sanctuary of Demeter and Kore.4
Another seated kourotrophos figurine of Corinthian ware was imported to the
sanctuary of Hera at Foce del Sele (Paestum) in the first half of the sixth century.5

Fig. 5.1a + 5.1b. Corinthian pyxis from a burial near Corinth, c.550 BC. Decoration scene,
procession, banquet (Cabinet des Médailles 94 © Bibliothèque nationale de France).

Turning to vase paintings, women with children on their laps are represented
seated on a stool wearing a peplos and bands in their hair (Figures5.1a–b, 3).6
This matronly figure is probably a mortal rather than a goddess (Demeter or
Clotho), because she is drawn in the same scale as the women surrounding her,
and is lacking any attributes or inscriptions. Children assume a stiff, wooden-
like posture and are executed as miniature adults.7 With the exception of a
striding boy on a LC alabastron,8 most children are typically identified as female
from their peplos and long hair. Few variations are noted among the female
figures accompanying the mother-and-child, who may carry spindles9 and
branches, form a chorus, or simply look on as by-standers. Occasionally, the
appearance of a sphinx and a siren alludes to funerary connotations.
The motif of a woman seated on a stool surrounded by other females finds
iconographic parallels later Attic wedding imagery, such as the lebes gamikos in
New York, and depictions of women in labour, such as the pyxis in the National
Museum in Athens showing Leto exhausted by labour pains10 – a scene that
seems to confirm Aristotle’s comments regarding the appropriate posture of a
woman during childbirth, a combination between lying and sitting.11
The mother-and-child motif is usually found on pyxides and bottles, two vessels
commonly associated with the female sphere both on account of their shape, but
also of their decoration (see Table 5.3). In particular, Ines Jucker12 has demonstrated
62 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

that bottles are often decorated with the so-called Frauenfest-scenes, i.e. ceremonies
comprising of female choruses, processions and kanephoroi, and proposed a
connection of these ritual scenes with the cult of Artemis (see Figure 5.2). In
addition to her suggestion, I would emphasize the presence of the mother-with-
child in these compositions and propose a closer association with Leto, as well.13
Vases decorated with the mother-and-child motif have been discovered in
the Heraion of Perachora, in Delos and Corinth.14 It should be noted that the
first two sites are significant cult-centres of Hera and Leto, the two main protag-
onists of the birth of Apollo and Artemis. The overall impression regarding
the woman-and-child motif is that it is found almost exclusively on terracottas
and vases associated with female rituals, while there seems to be a particular
connection with the sanctuaries of Hera (e.g. Perachora, Foce del Sele).15

The Kanephoros
Let us proceed to representations of kanephoroi. The tradition of girls carrying
offering trays occurs in many Greek cities, and is attested in both the literary
sources and the visual arts.16 Holding the kanoun, the tray with sacred objects,
on top of their heads, these girls follow the procession and chorus of women
in Frauenfest-scenes, and assist in sacrifices where both males and females are
present.17 Age definition can be challenging, since their dress is similar to a
mature women’s. However, the frequent absence of an epiblema and himation in
association to their small scale may be taken as an indication of a pre-pubescent
age. Also, the short hair of some kanephoroi should not necessarily be taken as
a symbol of lower status, as it can also indicate their imminent transition from
puberty to adulthood or perhaps a pre-nuptial ceremonial hair-cutting.18

Fig. 5.2. Corinthian bottle, from Corinth,

600–580 BC. Procession of women towards a
trapeza or altar (British Museum 65.7–20.20 ©
The Trustees of the British Museum)
T r ac i n g t h e Oikos i n P r e - C l a s sic a l C o r i n t h 63

A characteristic kanephoros scene can be found on the religious procession

depicted on the EC bottle in the British Museum.19 On the head of the
procession a woman stands next to an ornate offering table holding a wreath in
her hand. Behind her and flanked by two younger girls, a kanephoros balances
on her head a large tray laden with libation equipment. More women and
young girls follow further back. Judging by their different sizes, we may assume
that this exclusive female festival was open to women of all ages, ranging from
pre-pubescent girls to young maidens and mature women. Girls appear to have
no active role in this ritual except for their participation in the procession, in
contrast to the maidens’ responsibility of the kanoun.20
Kanephoroi were an essential component of sacrificial processions, such
as the one depicted on the LC pyxis in the Cabinet des Médailles21 showing
a woman overseeing a female flute-player, a kanephoros and two rather short
and chubby males leading a goat to sacrifice (boy-slaves?). A similar episode
decorates the MC amphoriskos from Corinth,22 only this time it is possible to
recognize the scene as a family sacrifice: the head of the household leads a bull
to sacrifice, followed by a flutist, two women and three kanephoroi.23

Fig 3a + 3b. Corinthian amphoriskos from Corinth, c.575–550 BC. Sacrificial procession
(Oslo, Mus. Etn. 6909 © Museum of Cultural History, University of Oslo)

The same interpretation can be pursued for the sacrificial procession on the
Pitsas panel from the sanctuary of the Nymphs,24 comprising of a kanephoros,
youths and maidens. Instead of forcing an identification of the largest figure as
a pregnant woman, it is preferable to recognize in this figure the male leader of
the oikos and thus interpret the scene as a family sacrifice.
Provided that our identification of the kanephoros with an adolescent girl
is correct, we come to the conclusion that maidens were expected to actively
participate in rituals associated both with the world of women and also in
religious ceremonies open to both sexes and all ages. These were occasions that
gradually introduced maidens to society, and trained them for their future roles
as mature women.
64 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

Wedding-Related Scenes

Fig. 5.4. Corinthian column krater from Caere, c.540

BC. Arrival of the wedding couple to their new
home (Paris, Louvre E 637 © RMN/Apeiron Photos)

Let us now examine a few wedding-related scenes. The LC column krater in Paris
depicts a couple in their chariot, followed by a small procession and greeted by
another group of men and women.25 I would like to interpret this scene as the
arrival of the newlyweds’ chariot at the groom’s home, a composition built
upon examples of mythological weddings.26 Our interest lies not so much in
the groom and his bride, represented here as a young woman, but rather in the
snake drawn above the horses’ heads. The snake in Greek iconography may
have a number of meanings, yet these all stress its chthonic connotations. In the
case of the Paris krater, an interpretation of the snake as an icon of Zeus Ktesios
(of the Household) and Herkeios (of the Courtyard)27 may be preferable to its
identification as a bad omen or a heroic symbol because of the overall setting of
the scene: the wedding procession has reached its final destination (the groom’s
home) and the new bride is about to be received by her husband’s family: two
men, a woman and an ephebe. The latter is drawn in white (now flaking away),
perhaps indicating his tender age. A similar domestic setting is illustrated on a
contemporary krater in Berlin, depicting the departure of a warrior.28

Fig. 5.5. Corinthian amphoriskos from Vulci,

second quarter of the sixth century BC. Belly:
Frauenfest-scene; horsemen. Shoulder: Banquet
scene (Philadelphia, University Museum MS 552
© Courtesy of the Penn Museum, image n.
T r ac i n g t h e Oikos i n P r e - C l a s sic a l C o r i n t h 65

The iconographic subjects of the MC amphoriskos in Philadelphia,29 a chorus of

women on the belly and a peculiar banquet-scene on the shoulder, have been
associated in the past with the cult of Demeter, Kore and the Moirai.30 Here
they are tentatively connected to the sphere of weddings. Starting with the
shoulder-scene, a woman and a young girl stand behind a man who reclines on
an elaborate couch.31 As a welcoming gesture the banqueter extends a kantharos
towards the approaching male figure with a rod and the woman behind him. A
seated woman with a spindle frames the scene on the right. The presence of a
girl during a banquet, baffling at first glance, is explained through a comparison
of the scene to the famous mythological banquet on the Eurytos krater32 that
depicts King Eurytos and his daughter Iole receiving as guests the competing
suitors, including the hero Heracles. In this light, the Philadelphia amphoriskos
may represent a similar subject: the arrival of a suitor at the bride’s house to
claim her hand. Such an event accounts for the presence of female members
and the young bride-to-be at the banquet presided over by her father. In this
light, the chorus of women dancing on the main frieze of the amphora and a
kanephoros are construed as a preview to the wedding celebrations.
Lastly, the LC pyxis in the Cabinet des Médailles33 (Figure 5.1) depicts several
episodes including a sacrificial procession and birth-celebrations noted above,
as well as a banquet scene, in which guests have their feet washed, and a contro-
versial subject often considered to be the decoration of a statue. The protagonist
is rendered in small-scale indicative of her young age and stands on a stool
flanked by three women: one holds a branch and appears to be dressing the girl
with a patterned peplos, while another is about to place a wreath on her long
hair. The third woman is standing in the back observing the scene.
In an attempt to find a unifying iconographic theme for all the episodes
illustrated on the pyxis, I prefer to interpret this last scene as the preparation
of a bride. By emphasizing the wedding overtones in this episode, we may
recognize in the remaining scenes different stages of a woman’s life, starting
as a kanephoros of her father’s oikos, becoming a bride to her husband, and
ultimately a mother. Such an iconographic repertoire is appropriate for a
common female vessel (pyxis), and is especially suitable as a grave gift, judging
from the reported findspot of the vase.

Boys and Youths

Let us now turn to the representations of boys and youths, largely divided into
two groups: departure scenes, and boyhood and maturation episodes. From
the first group we may single out the LC column krater in Berlin34 showing a
warrior bidding farewell to two women and a man, as he is about to mount his
chariot. A blond ephebe depicted in white approaches the hoplite and touches
his chin, a tender gesture of goodbye or perhaps an ultimate plead to stay. As in
the Paris krater (Figure 5.4), the painter has included a snake, which may again
66 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

symbolize Zeus Herkeios. Under this premise one can argue that the warrior
is leaving his home and family members to join a military campaign. This
scene was probably modeled after mythological departures of heroes, the best
example of which is the Amphiaraos krater:35 in a touching family moment, the
hero is mounting his chariot, while his wife, daughters and sons of various ages
are bidding him farewell from the doorstep of their house.

Fig. 5.6. Corinthian oinochoe, early sixth century

BC. Man, woman and two children escape a
city-under-siege (Cabinet des Médailles 179 ©
Bibliothèque nationale de France)

A rather unique episode involving two children, a boy and a girl, is depicted
on a MC oinochoe in Paris.36 Hoplites, horsemen and a chariot are fighting
in front of the walls of a city, while three figures observe the battle protected
behind the fortification. Amidst the uproar a man with a sceptre and a woman,
both dressed in elaborate clothes, are escaping from the city under siege, along
with two children. As there are no inscriptions naming the figures, the scene
maintains its ambiguous character between a mythological and historical
capture of a city.
Regarding the boyhood scenes, undoubtedly the most characteristic repre-
sentations are found on the extraordinary LPC olpe, known as the Chigi vase.37
Here boys and youths are depicted in all three friezes: the famous hoplites’
scene, the parade of horsemen and chariot, and the hare-hunt scene. Jeffrey
Hurwit has convincingly demonstrated that the iconography of the Chigi vase
reflects different stages of life from puberty to manhood in ancient Corinth.38
In addition to this vase, there are two intriguing vase-paintings that can
also be associated with boyhood. The first vase, a MPC alabastron in Madrid,39
is divided into two registers, showing rows of hoplites. In the lower frieze,
squeezed between the shields of two warriors, appears the figure of a small
warrior, stepping on or out of an amphora. This miniature figure has been
interpreted as the spirit of a fellow-warrior rising from the vessel, and thus
the whole scene has been considered a funerary allegory. Representations of
ghosts in Greek art are extremely limited, and I know of no other example from
Corinthian workshops; even on later Attic vases their presence is controversial.40
Instead, I wonder whether the small scale of the figure indicates his young age
and his placement above the amphora corresponds to a ritual initiating him to
the world of hoplites. However, lacking additional evidence neither theory can
be proven.
T r ac i n g t h e Oikos i n P r e - C l a s sic a l C o r i n t h 67

Fig. 5.7. Pyxis lid by the Dodwell Painter, 590 BC.

Youth among women and a man; hunting scene
(Staatliche Antikensammlungen, Munich 327 ©
Staatliche Antikensammlungen, Munich)

The second vase is a MC pyxis-lid attributed to the Dodwell Painter41 and is

decorated with three, seemingly unrelated subjects: two sphinxes flanking a bird,
four hunters against a boar, and a group of two women, a boy and a man. Most
figures bear inscriptions. The young Dorimachos is depicted naked trying to
escape from a figure that holds him by the hair and prevents him from running
towards the direction of the boar hunt. Because of the unbelted peplos and a nearby
inscription that reads Alka, this figure is thought to be female.42 A second woman
called Sakis and another figure, with a kerykeion named Agamemnon, flank the
scene. Despite the name, he is probably a messenger, bringing the news regarding
the boar hunt to the group and not the king of the Achaeans. It is unclear whether
this scene should be interpreted in connection to the hunting episode and it is
equally challenging to find a reason why a woman would be holding a boy by
his hair. However, if she is merely touching his head, a gesture perhaps similar
to touching one’s chin, she could be pleading with the boy not to join the other
hunters and risk his life. An interpretation of her gesture as protective rather than
threatening is also supported by the etymology of her name.
Last but not least, we will examine the most characteristic example of youths
dancing. The MC aryballos found near the temple of Apollo at Corinth depicts
a chorus of seven youths dancing to the music of a flautist named Polyterpos.43
The remarkable lead-dancer performing a high-vault in mid-air is followed by a
chorus of six youths aligned behind him. A snake-like inscription in hexameter
names the lead dancer Pyrrhias and informs us that the ‘olpa’ belongs to him.
Scholars have argued that the vase was either given to Pyrrhias as a prize for
his victory at a dance contest or, judging from the findpot of the aryballos,44
Pyrrhias may have dedicated this vase to Apollo to show his gratitude for
his victory. According to Theocritus (2.156), an olpe was a Dorian oil flask,
equivalent to the Athenian lekythos, often held by youths during their training.
In this light, iconography, usage and ownership of the vase are associated with
the world of youths, as was the final recipient of the aryballos, the youth-par
excellence, Apollo.
68 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

The study of these vase-paintings along with previous depictions of boys in

processions and sacrifices helps us to reconstruct the different stages of their life
and maturation. Boys and youths are illustrated in domestic scenes as principal
members of the household; they are shown participating in religious activities,
and preparing for adulthood. It should be noted that the majority of depic-
tions of boys and ephebes is found on sympotic vessels, e.g. kraters, oinochoe,
olpe, which are usually connected with male gatherings and their decoration
may have served an instructive purpose for the newly initiated members of the

The Representation of Women and Children in

Sixth-Century Corinth
To conclude this overview of representations of women and children in Archaic
Corinth, I would like to offer some thoughts on their increased occurrence
during the first half of the sixth century and place them within a larger historical
frame. To this end, I wish to begin by pointing out some information from the
literary sources concerning the tyrant of Corinth Periander (c.625–585 BC). His
crimes against family-members are notorious: he killed his wife Melissa, he was
on bad terms with all his sons and even forced his youngest one, Lycophron,
into exile, not to mention the murder of his second bride-to-be, Rhadine, and
her cousin.45 All these incidents are indicative of his estranged family-relations
and a rather perverse notion of one’s oikos.
Not surprisingly Periander followed the same harsh approach towards oikoi
and domestic traditions not only with his own family-members but also in
his implementation of larger social policies.46 His attack on female luxury and
his measures against laziness and accumulation of wealth aimed towards the
decrease of power of other clans and to a distinct change from previous norms.
The most famous incident of Periander’s anti-oikos policy was his ruthless blow
against Corcyra’s population, namely his decision to send 300 young boys from
the island to be castrated in Lydia.47 Such an austere measure targeted the heart
of Corcyrean oikoi and aimed to root out any potential for revolt.
Acknowledging the issues involved the degree of accuracy and bias of our
ancient sources,48 it appears that Periander’s actions against his relatives reflect
a lack of appreciation of his oikos and particularly its female members. At the
same time, his regulations demonstrate a strict policy against any excess that
could potentially strengthen an enemy-clan or his own family-members, and
thus compromise his reign. It is safe to say that under his rule the power of the
individual greatly overshadowed the collective force of the oikos.
With these observations in mind let us return to our material. Since most of
the vases and terracottas examined here date to after the fall of the Corinthian
tyrants in c.580 BC,49 one is led to assume that this new iconographic interest
T r ac i n g t h e Oikos i n P r e - C l a s sic a l C o r i n t h 69

in women and children reflects important social changes and perhaps, new,
pro-oikos policies. This hypothesis is supported by contemporary burial and cult
practices in Corinth, which indicate a revived concern with one’s household.
In particular, even though grave-stones and plot-walls are no longer used in
the organization of the sixth-century necropolis at the North Cemetery, older
geometric burials (family plots) are respected and left untouched. We also
observe a distinct pattern of arranging graves of the same family in clusters,
and in one case, marking a family plot with a remarkable feature: a monumental
platform made of poros-stones covering four graves. In general, children in the
sixth and fifth centuries were buried in groups near the graves of the family,
although in some cases there are clusters of graves comprising solely of infants.50
Equally enlightening for our hypothesis is the study of small hero-shrines
in the city of Corinth, which were built either on top of earlier houses (LC) or
on top of graves (G), such as the Erosa Shrine in the Potters Quarter and the
Heroön of the Crossroads in the Agora, respectively.51 Most of these shrines
date to the sixth century, and their direct connections with family-residences
and ancestral burials reflect a lasting appreciation of one’s oikos and the signifi-
cance of domestic cults. This practice is an alternative to the model of public,
all-inclusive ceremonies, as it indicates a more private type of rituals with
groups revering the graves of ancestors, and families or clans venerating heroes
or gods once worshipped in a former house.
Another feature that supports our idea for a renewed interest in oikos is the
occurrence of multiple dining rooms in the sanctuary of Demeter and Kore at
Acrocorinth in the Archaic period. These facilities have been associated with
cult activities and ritual banquets of individual oikoi.52
Based on these observations, I would suggest that this shift of interest
towards the preservation of one’s oikos is the result of the oligarchy established
in Corinth after Periander’s death. The strict regulations of a tyrant with little
respect to his own family, let alone to rival Corinthian clans, are followed by
a government that favors the growth of individual oikoi, and progressively,
led to different constructions of oikos-imagery. Within this new political and
social frame, the increased presence of women and children on Corinthian
vases should be understood not merely as a new iconographic trend, but as a
reflection of larger changes in the social structure of Archaic Corinth.53
Usually female choruses, kourotrophic figures, kanephoroi and youths repre-
sented in Archaic Corinthian art are analyzed separately with no common links
between them, focusing predominantly on religious connotations and rites of
transition. By studying these iconographic subjects collectively in this paper, I
have been able to demonstrate their significance as a thematically unified group,
which along with archaeological data and literary sources help us to recreate the
Corinthian family.
70 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

Table 5.1: Catalogue of Vases (non-exhaustive)

Shape Date Findspot Decoration Select
– comments bibliography
Alabastron LPC/EC? ? Small figure CVA Madrid 1,
amidst warriors; IIIc, pl. 3.10a–b
young hoplite or (32646)
Alabastron EC/MC Corinth Young girl with Corinth XV ii p.
Potters Quarter flower 104, no. 491, pls.
(aryballos 24, 95 (K 72)
Alabastron MC/LC ? Two women MuZ 213–4,
seated on stool 991, pl. 40. 172
back to back, one (Berlin 4285)
holding a child
that is pacing, a
third woman is
standing with a
spindle in hand,
Aryballos MPC Thebes Athena, CVA Oxford 2,
Aphrodite?, III c, pl. 1.5, 36,
Dioskouroi? 51 (G 146); Amyx
And a small 1988, II 619, no.
figure (child?, 3, n. 7; Schefold
xoanon?); sphinx 1966, 41, fig.
in the back; three 10 (Rescue of
dogs on handles Helen); Arafat
2004, 32–4, fig.
Aryballos MC Corinth, Apollo Chorus of Lorber 1979,
temple 7 boys, one 35–6, no 39,
Pyrrhias is pl. 8 (Corinth
jumping in air C-54-1);
in front of the Boardman 2007,
flutist Polyterpos 265, fig. 294;
158–163 Amyx
1988, II 560–1,
Amphoriskos MC Corinth Procession CVA Norway 1,
of man, pl. 4 (Oslo Mus
women, young Etn 6909); Amyx
kanephoroi 1988 II 658,
no. 4
T r ac i n g t h e Oikos i n P r e - C l a s sic a l C o r i n t h 71

Amphoriskos EC/MC Vulci, Chamber Procession of CVA

tomb 5 women, young Philadelphia
kanephoroi 2, pls. 24–6
Museum MS
552); Amyx
1988, II 311–2
(A2), 494; Jucker
1963, 53–4, no.
33, pl. 23.2
Amphoriskos MC/LC Perachora Woman seated Perachora II,
on stool with 225, no. 2216,
child in arms pl. 73
flanked by
sphinxes; siren
Pyxis LC? Grave between Procession, CVA
Corinth and woman with Bibliothèque
Sikyon child among Nationale
females, banquet (Cabinet de
and washing of Medailles) 1,
feet, dressing 14–5, pl. 17 (94);
scene Amyx 1988 I
229, no. 1(by
the Skating
Painter); NC
878; Callipolitis-
Feytmans 1970,
51–2, no. 3, fig. 3
Pyxis MC ? Procession CVA Munich 3,
of women, 40–2, figs. 6–9,
dancing, young pl. 144.5–6; 145.
kanephoroi, 1–2; Amyx 1988,
spinners, I 229 (Munich
children and 7741)
woman with
Pyxis MC ? One frieze on Leningrad 2961;
without lid and another Amyx 1988, I
handles on body: both 229, possibly
with chains of by the Skating
women with Painter
wreaths, flutists,
kanephoroi, one
woman with
child on her lap
72 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

Pyxis MC Corinth Potters Woman holding Corinth XV ii p.

Quarter (from a child, seated 141, no. 718, pl.
inside north long woman with 103, (KP 2476)
building) spindle
Pyxis MC Perachora Dancing Perachora II,
women, girl on 171–2, no. 1783,
a throne?, girl pl. 77
behind a second
throne, woman
with spindle?,
procession of
women with
Pyxis Early MC Perachora Dancing women, Perachora II,
kanephoros 212–3, no. 2066,
pl. 77
Pyxis lid EC/MC Corinth Boar hunt scene, Lorber 1979,
woman pulling 45–6, no. 52, pl.
child, 2 more 14 (Munich 327)
figures, sphinxes by the Dodwell
and bird. Painter; MuZ pl.
Inscriptions: 45.192; Amyx
Agamemnon, 1988, II 205, no.
Dorimachos, 1, pl. 86
Alka, Sakis
Bottle MC Delos Dancing women, Jucker 1963,
kanephoroi, pl. 17.1, 3, 5, 6
woman with (Beziers IN 22);
child Amyx 1988 I
230 (no image)
by the Beziers
Bottle MC ? Dancing women, Jucker 1963,
kanephoroi, pl. 17.2, 4
woman with (Montpelier
child M. Fabre 403;
127 (SA 197));
Amyx 1988, I
230, pl. 98.2;
by the Beziers
T r ac i n g t h e Oikos i n P r e - C l a s sic a l C o r i n t h 73

Bottle MC ? Dancing women, Jucker 1963, pl.

kanephoros 20.1 (Walters
48.193); Amyx
1988, I 229, no.3,
pl. 98.1 by the
Skating Painter
Bottle EC Corinth Processions, Jucker 1963,
girls, kanephoros pl. 20. 2 (BM
NC 1070;
Amyx 1988, I
230, pl. 99.2
by the London
Bottle ?/ MC Corinth Woman combing Corinth VII.2, pl.
Oinochoe a girl’s hair? 29.164, 43.164
(CP 2485)
Oinochoe EC/MC ? Capture of city, CVA
mythological?, Bibliothèque
man, woman Nationale
and 2 children (Cabinet de
escape Medailles) 1, 11,
pl. 11.9-11 and
12 (179)
Olpe MC ? Dancing women, Jucker 1963, pl.
young girl and 23. 3 (Pr Coll;
kanephoros once on display
at the Cleveland
Museum of Art)
Olpe (Chigi LPC Tomb at Monte Boy-flutist, boys Amyx 1988, I
vase) Aguzzo, near hunting 31, no. 3; Hurwit
Cezena 2002 (Rome,
Villa Giulia
Plate MC Perachora 3 women Perachora II,
dancing (head 194, no. 1951, pl.
bands and long 77 (fr. D 115)
peploi), dog,
74 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

Column Early LC ? Departure Lorber 1979,

krater of warrior, 64–6, no. 99,
young boy in pl. 24 (Berlin
white, blond 1959.1) by the
hair, nameless Hippolytos
touches the chin Painter
of the warrior
who looks
back towards
2 women
and a man;
chariot with
another warrior
completes the
Column LC II Caere Departure scene: Lorber 1979,
krater chariot at the 81–2, no. 127,
center of with a pl. 39 (Louvre
wedding couple; E 637); Amyx
one man and 1988, II 575, no.
two women 71
in front of it
two couples
behind; more
figures dispersed
e.g. a boy named:
pais?), snake
Column LC Caere Departure of Lorber 1979,
krater Amphiaraos 78–9, no. 122,
pl. 37 (once in
Berlin F1655);
MuZ 210–11, pl.
Column LC Caere Banquet scene Lorber 1979,
krater with Eurytios, 23–5, no. 23, pl.
Iole and the 5 (Louvre E 635);
suitors Amyx 1988 II,
558–9, no. 12
EPC 725/20–690 BC; MPC I 690–670 BC; MPC II 670–650 BC; LPC 650–625 BC;
Transitional 625–605 BC; EC 605–580 BC; MC 580–560/555 BC; LC I 560/55 – 535 BC; LC II
535–500 BC
T r ac i n g t h e Oikos i n P r e - C l a s sic a l C o r i n t h 75

Table 5.2: Catalogue of other material (non-exhaustive)

Object Date Findspot Decoration Select
– comments bibliography
1. Wooden LC Pitsas Pinax A: Lorber 1979,
painted procession 93–4, no. 154,
panels towards pl. 45,46 (Athens
an altar. NM Inv A
Inscriptions: 16464/65/66/67);
part of the Orlandos, A.K.
painters’ in EAA VI 1965,
signature: [---] 201, fig. 225–6;
korinthios, Larson 2001,
dedication to 232–233; Walter-
the nymphs, Karydi 1986, 27,
and names fig. 1. Dilllon
of the three 2002, 228–9
2. Votive Relief Late 5th or Corinth Family- Corinth IX,
early 4th c. BC sacrifice, 128–9, no. 267
procession (199)
including a
young boy
3. TC figurine Archaic Corinth, Handmade; Corinth XVIII.4,
Sanctuary of seated woman 70–1 (13558);
Demeter and with necklace, Langdon
Kore on her lap via email,
another publication
seated female, forthcoming
and on the lap
of that one
of swaddled
4. TC figurine Archaic Corinth, Imported; Corinth XVIII.4,
Sanctuary of partially 70–1 (MF
Demeter and preserved 71-54); Langdon
Kore bust of via email,
woman in publication
peplos, head forthcoming
of child
against her
76 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

5. TC figurine Archaic Corinth, Handmade, (MF 10324)

Sanctuary of small; Langdon
Demeter and standing via email,
Kore figure possibly publication
holding a forthcoming
6. TC figurine Archaic Corinth Monkey with Corinth XV.2,
swaddled 178, no. 3, pl. 39
baby; (KT 30-3)
parody of
7. TC figurine EC/MC Corinth, Handmade; Corinth VII.2, pl.
Anaploga woman and 85, An 347 (MF
Well child? 11645)
8. TC figurine 5th c. BC Corinth Woman Corinth XII,
holding a 34, pl. 10. 140
girl on her (1587)
9. TC figurine 5th c. BC Corinth Woman with Corinth XII,
child 34, pl. 10. 141
10. TC figurine Late 6th early Corinth, trial Grotesque Corinth XII, 35,
5th c. BC excavation at female figure, pl. 10.153 (3456)
Penteskoufia imitating
11. TC figurine 5th/4th c. BC? Corinth Seated boy Corinth XII,
38, pl. 13. 191
12. TC figurine Archaic Corinth, Temple-boy* Hadjisteliou-
Circular Price 1969, 96,
South Shrine pl. 21.7
13. TC mold Archaic (?) Corinth, Temple boy Hadjisteliou-
Potters Price 1969, 96,
Quarter pl. 21.8; Corinth
XV.1, 105–6,
no. 61, pl. 40;
Corinth XVIII.4,
69 (KH 60)
14. TC figurines Archaic (?) Perachora Temple boy; Hadjisteliou-
3 examples Price 1969, 96, n.
total, but not 20 and 99–100
clear if all
T r ac i n g t h e Oikos i n P r e - C l a s sic a l C o r i n t h 77

15. TC figurine Second q. of Demeter Temple boy Hadjisteliou-

5th c. BC and and Kore with pilos Price1969, 99,
later Sanctuary: and clock, 105
from votive ‘shepherd
deposit boy’
with female
16. TC figurine First half of Paestum, Seated Miller
6th c. BC Heraion at kourotrophos Ammerman
Foce del Sele with child; (a 2007, 135, fig.
second one 7.5
preserves only
the infant)**
*Ca 20 TC Temple-boys and molds have been found but only one is considered Archaic, a
couple date in the second half of fifth (re-using the Archaic mold) while most in the fourth
century BC. They were dispersed all over Corinth: the Potters Quarter, TC Factory, SE
Building, Asklepeion, the North Cemetery and Circular S Shrine (Price (1969), BSA, 64, 96–7,
pls. 21, 7–11 and 22, 17, 21; Corinth XII, pp. 115–6, Corinth XVIII.4, pp. 68–9)
** Two dozen, Corinthian, female figurines were found in sanctuaries around Paestum. At
least six kourotrophoi of local workshops at Foce del Sele date in the first half of the fifth
century BC. The type will be further developed in the late fifth century to what is known as
the Pestana type
Table 5.3: Shape and Subject Relation
Subject Toddlers/ Boys and Kanephoroi Other scenes
young children youths with girls
Alabastron 1 1
Aryballos 1
Amphoriskos 2
Pyxis 5 1 4 2
Bottle 2 5
Oinochoe 1 1
Olpe 2 1
Krater 3 1
Plate 1
Panel 1 1
Total 8 10 14 4
*Note: Sample taken based on twenty-eight vases and a panel. More than one kanephoros or
boy/youth may occur in a single vase painting.
78 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

The following abbreviations are employed for this article alone:

Corinth VII.2: Amyx, D. A. (1975), Archaic Corinthian Pottery and the Anaploga
Well, Princeton: ASCSA.
Corinth XII: Davidson, G. R. (1987), The Minor Objects, (reprint), Princeton:
Corinth XIII: Blegen C. W. (1964), The North Cemetery, Princeton: ASCSA.
Corinth XV.2: Stillwell, A. N. (1948), The Potters’ Quarter – The Terracottas,
Princeton: ASCSA.
Corinth XVIII.4: Merker, G. S. (2000), The Sanctuary of Demeter and Kore:
Terracota figurines of the Classical, Hellenistic, and Roman Periods, Princeton:
Corinth XX: Williams, C. K. and Bookidis, N. (2003), Corinth, the Centenary
1896–1996, Princeton: ASCSA.
EC: Early Corinthian
MC: Middle Corinthian²
LC: Late Corinthian
PC: Proto-Corinthian
MPC: Middle Proto-Corinthian
G: Geometric

  1 See for example, Crelier, 2008; Oakley and Neils, 2003.
  2 Amyx 1988, Vol. 2, pp. 397–434, esp. 428–429.
  3 On the type, see Hadjisteliou-Price, 1969.
  4 Here Cat. nos. II.3, 4, 5, 7. For the Anaploga figurine see Corinth VII.2, pl. 85, An 347 (MF
11645). For the figurines from the sanctuary of Demeter and Kore I would like to thank Susan
Langdon for sharing with me her information. Also, see preliminary mention in Corinth
XVIII.4, 70–71 (MF 71–54), (13558). Even in areas clearly associated with female cult, e.g.
the Kokkinovrysi sanctuary, there are many female figurines, but none holding a child or
indicating pregnancy: Kopestonsky, 2009, pp. 113–120.
 5 Miller Ammerman 2007, p. 135, fig. 7.5. Here Cat. n. II.16. It is interesting to note that
according to the study of Olsen (1998, esp. pp. 384–388, fig. 1) even in the Mycenaean period
kourotrophoi figurines are rare in Corinthia compared to the Argolid. On the question of
Mycenaean terracottas and children, see Gates, 1992.
  6 See Cat. nos. I.3, 8, 9, 10, 11, 12, 16, 17.
  7 On the iconographical stereotypes for children and their association with dwarfs in medical
texts, see Dasen, 2008.
  8 Cat. n. I.3.
  9 On the association between women and weaving, see Nixon, 1999. Regarding the identification
of these scenes with rituals in honour of Artemis, see Jucker, 1963 and here nos. 12, 13.
10 Attic red-figure pyxis from Eretria, Athens Nat. Mus. 1635; LIMC VI Leto, 6.
11 Aristotle, The History of Animals, Book 7. See also the late fifth century relief showing a woman
having just given birth: NY 24.97.92; Mitropoulou, 1977, n. 66; van Straten, 1981, p. 100, fig. 43.
12 Jucker, 1963; Amyx, 1988, Vol. 1, pp. 228–230; Vol. 2, pp. 653–657. It might be worth
comparing the chains of female dancers on Clazomenian sarcophagi and Phikellura vases (e.g.
MuZ pls. 32–33, nos. 142–144) to those of the Corinthian Frauenfest-scenes.
T r ac i n g t h e Oikos i n P r e - C l a s sic a l C o r i n t h 79

13 Support for associating Artemis with childbirth comes not only from the sanctuary of Braurona
in Attica (Antoniou, 1990; Giuman, 1999; Gentili and Perusino, 2002) but also from the so
called Echinos Relief found in Thessaly, dating around 350–300 BC (Morizot, 2004; Cole, 1998;
Dillon, 2002, pp. 231–233, fig. 7.4). Alternatively, see Merker’s comments (Corinth XVIII.4,
68–73, esp. 71, n. 333) for the possibility of associating the Frauenfest-scenes with Demeter,
once the material from the sanctuary of Demeter and Kore in Corinth becomes known, a view
originally expressed by Callipolitis-Feytmans, 1970. A connection with Dionysos is favoured
by Pemberton, 2000, based on her comparative study of Fraunfest-scenes with those of padded
dancers with some emphasis on vases from Perachora. Shäffer (1997, esp. pp. 91–98) argues
against an association of the Frauenfest scenes with a single deity, because the vases originate
from various sites, ranging from sanctuaries to cemeteries, while Dillon, 2002, pp. 128–129
suggests a mixed target (i.e. many deities) depending on the inclusion of children, women with
spindles, women dancing and padded dancers. With regard to Leto and childbirth, Theocritus
18.50–51 mentions she was invoked for blessing of children.
14 Perachora: Cat. nos. I.8, 13; Delos: Cat. n. I.16; Corinth: two pyxides, one from the Potters’
Quarter (Cat. n. 12) and another from a grave outside the city (Cat. n. 9). Unfortunately there
is not enough evidence to associate the former with a shrine from the ceramic production area,
while for the latter one may only hypothesize a connection with a female grave on account of
the vase’s shape and decoration.
15 See also, Dillon, 2002, p. 14.
16 On kanephoros, see Dillon, 2002, pp. 37–41 with more bibliography.
17 Cat. nos. I 6, 7, 10, 11, 14, 16, 17, 18, 19, 22, 24, and Cat. n. II.1 (the Pitsas Panel).
18 Hdt. 4.34–35 and Callim. Delos, 269–269 mention that on Delos girls cut off a lock of their
hair and wound it around a spindle and laid it on the memorial which was inside the sanctuary
of Artemis, while ceremonial cutting of hair and a dedication of a lock to Artemis prior to
marriage is reported in Anth Pal. 6.276 and Archilochos 326 (Dillon, 2002, p. 215). For cutting
hair in case of mourning, see Larson, 1995, pp. 73, 120, n. 24 with more references. See also,
Dillon, p. 352, n. 31; Dowden, 1989, pp. 2–3; Oakley and Sinos, 1993, pp. 15–21.
19 Cat. n. I.19.
20 A similar scene, although with more music and dancing and no offering table, is depicted on
the two registers of a MC bottle in Montpellier (Cat. n. I.17).
21 Cat. n. I.9.
22 Cat. n. I.6.
23 On family religion see recently Faraone, 2008 and Boedeker, 2008.
24 The Pitsas panels were discovered in the homonymous cave near Sikyon in Corinthia and date
around 540/530 BC. Based on inscriptions, iconography and contextual finds they have been
associated with the cult of the Nymphs. The best-preserved one is Panel A, here Cat. n. II.1. For
more bibliography, see Lorber, 1979, pl. 46; Larson, 2001, pp. 232–233; Walter-Karydi, 1986, p.
27, fig. 1. Dillon, 2002, pp. 228–289 argues in favour of identifying the last figure on the right as
a pregnant woman (an unlikely possibility), in an attempt to highlight even further the female
character of the sanctuary of the Nymphs.
25 Cat. n. I.26.
26 E.g. the composition of Paris and Helen on the column krater from Italy by the Detroit Painter
(New York 27.116): Amyx, 1988, Vol.1, p. 196, pl. 79.1a-c; NC 1187.
27 On Zeus Ktesios and Zeus Herkeios, see Isaeus 8.16; Aesch. Ag., 1036–1039; Faraone, 2008, pp.
212–217 and Boedeker, 2008, pp. 228–236.
28 See here, Cat. n. I.25. On the ceremonies, see Boedeker, 2008, pp. 240–241.
29 Cat. n. I.7.
30 Callipolitis-Feytmans, 1970.
31 The banqueter with a kantharos may be compared to a similar figure on the plate in Athens
by the Skating Painter (Nat. Mus. 951): Amyx, 1988, pp. 229, pl. 97; NC 1030. However, the
scene on the plate in Athens has a totally different atmosphere including padded dancers (one
with deformed foot, others ithyphallic), women in company of dancers holding wreaths and
drinking vessels.
32 Cat. n. I.28.
80 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

33 Cat. n. I.9.
34 Cat. n. I.25.
35 Cat. n. I.27.
36 Cat. n. I.21.
37 Cat. n. I.23.
38 For the most recent treatment of the vase and bibliography, see Hurwit, 2002.
39 Cat. n. I.1.
40 E.g., when Odysseus descends to the Underworld he meets the ghost of Elpenor, a scene
portrayed on the Attic red-figure pelike attributed to the Lycaon Painter and dated around
440 BC (Boston, MFA 34.79; ARV 1045.2) and the idealized depictions of the dead frequently
decorating Attic white lekythoi; for more information of the latter, see recently Oakley, 2005.
41 Cat. n. I.15.
42 The etymology of the name Alka may give us some direction towards deciphering this scene.
Alka: valour, courage, fight: Slater, 1969, s.v. alka.
43 Cat. n. I.5.
44 Regarding the identification of the temple with Apollo, see Bookidis and Stroud, 2004.
45 Salmon, 1984, pp. 197–205. On Melissa: Hdt. 3.50.1; 5.92.η 23; Diog. Laert. 1.94; Nic. Dam.
FGH 90 F58.2; On Rhadine (second wife): Strabo 347; On Lycaphron: Diog. Laert. 1–2.
46 On Periander’s harsh measures: Arist. Pol. 1313 a 35–37; Nic. Dam. FGH 90 F 58.1; Hdt.
5.92.1–4; Salmon, 1984, pp. 197–202 with commentary.
47 Hdt. 3.48.2–53; Nic. Dam. FGH 90 F59-2-4; Diog. Laert. 1.94–95; Salmon, 1984, pp. 224–226.
48 On the accuracy and bias of the sources, see the discussion in Salmon, 1984, pp. 197–205 and
cf. Anderson, 2005.
49 Salmon, 1984, pp. 229–230.
50 On the North Cemetery, see Corinth XIII, 65–69: About 600 BC activity begun to increase in
an extraordinary way and sixth-century burials are almost double compared to PC ones and
fifth-century ones. From the fourth century the number of burials decreases significantly, while
there are limited burials from the Roman period. Sixth-century graves are mainly concentrated
in narrow strips and continue to be used in the fifth century BC with an extension in the
fourth century. In the Classical period, the area is imagined as a large rectangle cut diagonally
in the south-west corner, covering an area of c.5000 m2. As for the platform, it covered four
burials carefully aligned and close in date and kind of offerings, while the top of the platform
received four stelae. This monument offers the only certain evidence for a family burial plot
in Post-Geometric periods. On the other hand, the arrangement of burials shows clusters of
graves throughout the cemetery even though not as regular as in the Geometric era. In some
cases we can observe a sequence of graves for almost a century. Many clusters include infant
groups and are in circular arrangement. Notable is cluster n. 155 with unusual offerings. In the
sixth and fifth centuries, children were buried usually in groups placed near the graves of the
family, e.g. nos. 332 and 174.
51 The Heroön of the Crossroads in the Agora (late seventh/early sixth century BC, with renova-
tions in the Archaic period) was erected over a Geometric burial, possibly commemorating the
deceased as an ancient ancestor. Also the Stele Shrine in the area of the later Roman Forum
(second quarter of the sixth century) was built on top of a former storeroom of a private MC/
LC house, while the Erosa-Shrine in the Potters’ Quarter was constructed over the ruins of
a fifth century house, and the Terracotta Factory shrine set on top a fifth century potter’s
residence and atelier. Lastly, the Underground Shrine, probably associated with a nearby
cemetery, was set up in the late sixth /early fifth century. Especially in the case of the shrines
erected in the Potters’ Quarter, Williams (1981) argues that they were up-kept by the surviving
members of a family for nearly a generation until their cult faded away.
52 Corinth XVII.3; Dillon, 2002, p. 126.
53 Similar conclusion, although for an earlier period, reaches Langdon, 2007, p. 191, who claims
that Geometric art played a significant role not only in shaping children but also in ‘[bringing]
young communities to social maturity, to the threshold of the polis’.
T r ac i n g t h e Oikos i n P r e - C l a s sic a l C o r i n t h 81

Amyx, D. A. (1988), Corinthian Vase-Painting of the Archaic Period, 3 Vols,
Berkeley: University of California Press.
Anderson, G. (2005), ‘Before turannoi were tyrants: rethinking a chapter of
early Greek history’, Classical Antiquity, 24, 173–222.
Antoniou, A. (1990), Brauron. Symvole sten Historia tou Hierou tes Brauronias
Artemidos, Athens: En Athinais Arxaiologike Hetaireia.
Arafat, K. W. (2004), ‘Sculpture gone to pot’, in Keay, S. and Moser, S. (eds), Greek
Art in View: Essays in Honour of Brian Sparkes, Oxford: Oxbow, pp. 24–38.
Boardman, J. (2007), The History of Greek Vases, (reprint), London: Thames &
Bodel, J. and Olyan, S. M. (eds) (2008), Household and Family Religion in
Antiquity, Oxford: Blackwell.
Boedeker, D. (2008), ‘Family matters: domestic religion in Classical Greece’,
in Bodel, J. and Olyan, S. M., (2008), Household and Family Religion in
Antiquity, Oxford: Blackwell, pp. 229–247.
Bookidis, N. and Stroud, R. S. (2004), ‘Apollo and the Archaic temple at
Corinth’, Hesperia, 73, 410–426.
Callipolitis-Feytmans, D. (1970), ‘Déméter, Coré et les Moires sur des vases
corinthiens’, Bulletin de Correspondance Hellénique, 94, 45–95.
Cohen, A. and Rutter, J. B. (2007), Constructions of Childhood in Ancient Greece
and Italy, Hesperia, Supplement 41, Athens: ASCSA.
Cole, S. G. (1998), ‘Domesticating Artemis’, in Blundell, S. and Williamson, M.
(eds), The Sacred and the Feminine in Ancient Greece, London: Routledge,
pp. 27–43.
Crelier, M.-C. (2008), Kinder in Athen in gesellschaftlichen Wandel des 5.
Jahrhunderts v. Chr. Ein archäologische Annäherung, Remshalden: Bernhard
Albert Greiner.
Dasen, V. (2008), ‘“All children are dwarfs”. Medical discourse and iconography
of children’s bodies’, Oxford Journal of Archaeology, 27, 49–62.
Dillon, M. (2003), Girls and Women in Classical Greek Religion, London:
Dowden, K. (1989), Death and the Maiden: Girls’ Initiation Rites in Greek
Mythology, London: Routledge.
Faraone, C. A. (2008), ‘Household religion in ancient Greece’, in Bodel, J. and
Olyan, S. M., (2008), Household and Family Religion in Antiquity, Oxford:
Blackwell, pp. 210–228.
Gates, C. (1992), ‘Art for children in Mycenaean Greece’, in Laffineur, R.
and Crowley, J. L. (eds), EIKON. Aegean Bronze Age Iconography: Shaping
a Methodology. Proceedings of the 4th International Aegean Conference,
Aegaeum 8, Liège: Université de Liège, pp. 161–171.
Gentili, B. and Perusino, F. (eds), (2002), Le Orse di Brauron, Un Rituale di
Iniziazione Femminile nel Santuario di Artemide, Pisa: ETS.
82 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

Giuman, M. (1999), La Dea, la Vergine, il Sangue, Milano: Longanesi.

Hadjisteliou-Price, T. (1969), ‘The type of the crouching child and the temple
boys’, Annual of the British School at Athens, 64, 95–111.
Hurwit, J. M. (2002), ‘Reading the Chigi vase’, Hesperia, 71, 1–22.
Jucker, I. (1963), ‘Frauenfest in Korinth’, Antike Kunst, 6, 47–61.
Kopestonsky, T. B. (2009), Kokkinovrysi: A Classical Shrine to the Nymphs
at Corinth, Dissertation, University of Buffalo, State University of New
Langdon, S. (2007), ‘The awkward age: art and maturation in early Greece’, in
Cohen, A. and Rutter, J. B., (2007), Constructions of Childhood in Ancient
Greece and Italy, Hesperia, Supplement 41, Athens: ASCSA, pp. 173–191.
Larson, J. (2001), Greek Nymphs. Myth, Cult, Lore, Oxford: University Press.
Lorber, F. (1979), Inschriften auf korinthischen Vasen: archäologisch-epigra-
phische Untersuchungen zur korinthischen Vasenmalerei im 7. und 6. Jh. v.
Ch., Berlin: Mann.
Miller Ammerman, R. (2007), ‘Children at risk: votive terracottas and the
welfare of infants at Paestum’, in Cohen, A. and Rutter, J. B., (2007),
Constructions of Childhood in Ancient Greece and Italy, Hesperia, Supplement
41, Athens: ASCSA, pp. 131–151.
Mitropoulou, E. (1977), Corpus I: Attic Votive Reliefs of the 6th and 5th Centuries
BC, Athens: Pyli Editions.
Morizot, Y. (2004), ‘Offrandes à Artémis pour une naissance. Autour du relief
d’Achinos’, in Dasen, V. ed., Naissance et Petite Enfance dans l’Antiquité. Actes
du Colloque de Fribourg, 28 novembre – 1er décembre 2001, Orbis Biblicus et
Orientalis 203, Fribourg: Academic Press Fribourg, pp. 159–170.
Nixon, L. (1999), ‘Women, children, and weaving’, in Betancourt, P.,
Karageorghis, V., Laffineur, R. and Niemeier, W.-D. (eds), Meletemata.
Studies in Aegean Archaeology presented to M. H. Wiener as he enters his 65th
year, Aegaeum 20, Vol. 2, Liège: Université de Liège, pp. 561–756.
Oakley, J. H. and Sinos, R. H. (1993), The Wedding in Ancient Athens, Madison:
Wisconsin University Press.
—(2003), Coming 

 Past, New Haven: Yale University Press.
—(2005), Picturing Death in Classical Athens. The Evidence of the White Lekythoi,
Cambridge: The Cambridge Studies in Classical Art and Iconography.
Olsen, B. A. (1998), ‘Women, children, and the family in the Late Aegean
Bronze Age: Differences in Minoan and Mycenaean construction of gender’,
World Archaeology, 29, 380–392.
Pemberton, E. (2000), ‘Wine, women and song: gender roles in Corinthian cult’,
Kernos, 13, 85–106.
Sabetai, V. (2009), ‘The poetics of maidenhood: visual constructs of womanhood
in vase-painting’, in Schmidt, S. and Oakley J. (eds), Hermeneutik der Bilder.
Beiträge zur Ikonographie und Interpretation griechischer Vasenmalerei,
Munich: C. H. Beck, pp. 103–114.
T r ac i n g t h e Oikos i n P r e - C l a s sic a l C o r i n t h 83

Salmon, J. B. (1984), Wealthy Corinth. A History of the City to 338 BC, Oxford:
Clarendon Press.
Schäfer, A. (1997), Unterhaltung beim Griechischen Symposion, Mainz: P. V.
Schefold, K. (1966), Myth and Legend in Early Greek Art, London: Thames and
Slater, J. (1969), Lexicon to Pindar, Berlin: De Gruyter.
Van Straten, F. T. (1981), ‘Gifts for the gods’, in Vernel, H. S., Faith, Hope and
Worship: Aspects of Religious Mentality in the Ancient World, E. J. Brill: Leiden.
Walter-Karydi, E. (1986), Prinzipien der Archaischen Farbgebung, Saarbrücken:
Archeological Institute of the Saarland University.
Williams, C. K. (1981), ‘The city of Corinth and its domestic religion’, Hesperia,
50, 408–421.

The Language of the Oikos and the Language of

Power in the Seleucid Kingdom
Omar Coloru

The kingdoms born from the division of the empire of Alexander the Great have
been appropriately defined as ‘personal monarchies’,1 because according to the
Hellenistic principle of the spear-won land they were created by the bravery of
a single man, who embodied both military and political skills similar to those
of Alexander and his father Philip of Macedon. The founders of the Hellenistic
Kingdoms had to deal with the problem of creating a dynasty in order to inspire
a feeling of loyalty among the members of the court, the high officers, the new
aristocracy of the kingdom, and their subjects as well. Therefore, a united
family was the necessary requirement to ensure the stability of the kingdom
and a long reign for its ruler. Generally speaking, each Hellenistic monarchy
was image conscious and developed a public image of the royal family to be
widely disseminated to their subjects. The Seleucids are not excluded from the
framework above, even if they have been presented negatively owing to a hostile
tradition based on historians such as Phylarchus, for example, and in particular
In this paper I do not wish to take a look at the affairs, the murders, and
other stories that we find in many passages of the ancient writers,3 but rather I
draw attention to the use of words and concepts related to the sphere of familial
relations in official documents such as decrees, letters written by the kings or
other members of the royal family to the town councils, and the use of royal
epithets. I will also take into consideration, even if to a small extent, literary
works to compare them with the evidence of the documentary sources. As
Rostovtzeff pointed out in his analysis of the economy of the Hellenistic world,
the kingdom was indeed the oikos of the king and ‘it should be noted that many
titles of the king’s officials in the sphere of finance and economics were terms
borrowed from Greek private law and Greek public and private economy’.4
Following this principle, the royal family mirrors the structure of the standard
Greek family unit: it is composed of a husband (the king), a wife (the queen),
the children (the princes and princesses), and a household (the kingdom).
Worship of the founder and the ancestors of the family group shifts from the
private to the public sphere and includes the cult of the living king.
T h e L a n g uag e o f t h e Oikos a n d t h e L a n g uag e o f P ow e r 85

Taking the cue from the title of the conference in Gothenburg, Oikos to Familia,
I begin this paper by analysing the use of the word oikos in documents related to
the Seleucid dynasty. According to the common definition, oikos is understood
as equivalent to the household unit composed of husband, wife, children and
property.5 This term embraces such a wide variety of nuances that it is hardly
translatable into our concept of ‘family’. To my knowledge, oikos occurs only once
in the documents related to the Seleucid administration, that is to say a letter
written by Queen Laodike III to the council of Iasos in 192 BC.6 In this inscription
Laodike praises the council and the people of Iasos for their loyalty towards her
husband Antiochus III and towards the oikos of the Seleucids. To this document
we should add those mentioning the word oikia which occurs in: (1) a decree
from Ilion for Antiochus I recording the goodwill of the citizens towards the
king’s father Seleucos I and the whole royal house, the basilikē oikia;7 (2) a letter of
Antiochus II granting the town of Erythrae autonomy and tax-exemption for its
favourable policy towards the Seleucids.8 The question is: which meaning should
we assign to the terms oikos and oikia in this particular context?
The rarity of the use of the word oikos in this type of document is probably
due to a change in the administrative language and ideology rather than to
mere coincidence. It is plausible that oikos may have been deprived of part of
its original meaning, in particular that concerning the household management
or, more generally speaking, property. As a matter of fact, the Hellenistic
chancelleries would employ the word pragmata to designate at the same time
the business, the personal interests and properties of the king, as if the state
was a personal possession of the monarch. The ideology of the Hellenistic
kingdoms predates that of the absolutist European monarchies between the
seventeenth and eighteenth centuries. In fact, that found in the Hellenistic king
is an unshared kingship, and even when the king associates one of his sons to
the government, he nevertheless holds supremacy. The king legitimizes the
assumption of the title of basileus by a military victory, which also ratifies his
right to enjoy full ownership of the royal land because it was won by the spear,
or, in the case of a successor, because that land had been conquered by his
ancestors and consequently had become part of his inheritance. The kingdom
being a personal property, it follows that the king is the chief administrator who
manages an intense programme of audiences and official correspondence which
at times appears to be quite trying.9
As stated above, the inscription from Ilion in honour of Antiochos I refers
to the basilikē oikia. According to Classical Greek law, oikia is defined as the
house, the physical building where the family unit lives, and is part of the oikos,
but can also be translated as ‘family’. The formula employed by queen Laodike,10
appears to be a parallel form of those expressions where oikia is used: εἰς τὴν
ἡμετέραν οἰκίαν, ‘towards our house’;11 πρὸς τὴν οἰκίαν ἡμῶν, ‘towards our
house’.12 Other more explicit expressions are τὴμ πᾶσαν βασιλικὴν οἰκίαν, ‘the
86 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

whole royal house’,13 and if we want to take an example from another dynasty,
τὴν Πτολεμαϊκὴν οἰκίαν, ‘the Ptolemaic house’.14 The Ilion inscription provides
us with the key to understand the proper meaning of oikia. Our text mentions
both the terms oikia and pragmata,15 so that we are led to believe that these
words might have two distinct meanings. The first one, in union with a word
related to kinship,16 should be understood as synonym of ‘dynasty’, while the
second one seems only to refer to the administrative system of the kingdom.
As for oikos, we find a greater number of parallels in documents concerning
Roman rule in Asia Minor where this term is used as synonym of domus. In
fact, domus is attested with the same sense and use we have already found in
Greek, i.e. family, gens.17 A few examples will suffice here to illustrate the range
of usage and its convergence with examples found under the Seleucids. Oikos
is attested in a long inscription (second century AD) displaying the copies of
several letters written to the elders of the council of Ephesos for the bestowal and
confirmation of privileges in favour of the town. In particular, two letters sent
by Germanicus and one by the emperor Tiberius serve to our purpose.18 The
letter from Tiberius is above all significant because of the following expression:
‘πρὸς τε [τὸν ἐμòν πα]-/τέρα καὶ τòν οἶκον ἡμῶν’, ‘(the decree) on behalf of my
father and our house’.19 Once again, one can notice the association of oikos with
a word related to kinship as already found in the inscription from Ilion. In this
case, oikos clearly stands for the Julio-Claudian dynasty rather than a generic
reference to the Roman Empire, and through this formula Tiberius aims to
stress the special ties of friendship between the town and his family. A similar
expression can be noted in a letter of the emperor Caracalla to Ephesos: [c. 4
εὐσεβῶς διακειμένους πρὸς τòν] ἡ.μ.ε.τ.έρον οἶ.κον, ‘as you are piously inclined
towards our house’.20 Judging from the evidence, it seems that the meanings of
oikia and oikos are then overlapping. In my opinion, this fact strengthens the
point that from Hellenistic times onwards, in the framework of royal ideology
and its vocabulary, oikos takes on a more restricted meaning centred on dynasty,
while the term pragmata assumes of itself the whole range of meanings related
to the administration.
The employment of a vocabulary referring to kinship and affection among
family members is a customary phenomenon in the court etiquette of
Hellenistic dynasties, and forms a strong element inside the strategy of
consensus. Among the Seleucids, we come across several cases of this type. In
a letter to Sardis, Laodike III thanks the town for the decision taken by the
council of performing rituals to Zeus Genethlios for her safety and that of her
husband and her children.21 The word employed by the queen to designate the
children is paidion, a term more informal than the expected teknon, used, on
the contrary, by the council of Iasos when addressing the queen.22 Therefore,
paidion conveys a feeling of maternal love and gives the letter a familial and
intimate dimension. Because of its affective connotation this word is often
found in funerary inscriptions for deceased children and in votive offerings
for the safety of the members of a family. Among the Seleucids this use is also
T h e L a n g uag e o f t h e Oikos a n d t h e L a n g uag e o f P ow e r 87

attested in two dedications: the first one comes from Eastern Kilikia and is
made by Themison, nephew of Antiochus III, who dedicates an altar to Zeus
Kasios for the safety of the royal family.23 The second is an inscription from
Ptolemais in which a friend of King Antiochus VII prays to Zeus Soter for the
safety of the king, his wife, Cleopatra Thea Eueteria, and their children.24 In the
official communications with the towns of their kingdom, the Seleucids used
a particular register which aimed to create an interaction between the king
and his interlocutors:25 the language of euergetism with its set of formulaic
expressions made possible for the king to affirm his authority towards the
subjects in a subtle and diplomatic way by ‘shifting the focus of discourse
from power to benefaction’.26 Likewise the cities and their representatives used
the same language ‘to interact with the rulers, introducing petitions in such
way as to apply moral pressure on the rulers and channel reactions along
the pre-scripted lines of euergetism’.27 This kind of communication does not
entail an authoritative relationship for the king, nor a passive one as far as
the subjects are concerned, but it is a real dialogue which takes place between
the two interlocutors. In this connection Capdetrey has appropriately defined
the Seleucid rule as ‘autorité dialogique’.28 Although the vocabulary of the
documents, as I demonstrate above, falls within this general pattern, it must
be pointed out that the word paidion has a more restricted use reserved to the
royal family members and a small group of individuals who gravitated around
the court. The available evidence confirms how the family vocabulary had
become popular among the officers of the kingdom, who used it as a special
communication code in order to emphasize their privileged position and
closeness to the royal family.

The Use of Familial Terminology to Define

Dynastic Relations
It seems that the word πατήρ, ‘father’, was used as a deferent and at the same
time denoted a form of address towards older people who took on the role
of preceptors or had been guide to the young king/prince. This metaphorical
use of kinship terminology had been very common in Classical Greece and
continued to survive in the Hellenistic period.29 Flavius Josephus quotes two
letters of Seleucid kings in which such a title is employed: the first one was
written by Antiochus III, who addresses to his chief minister and long-date
friend Zeuxis by calling him ‘father’;30 likewise Demetrius II applies the same
formula to Lasthenes,31 the Cretan mercenary chief who helped him to regain
the kingdom.32 Being the only testimony at our disposal, it is difficult to say
whether it should be interpreted as a customary court title or not, even if
scholars suspect that the use of this expression is too early for Antiochus’
times, because of its isolation.33 However, as far as I am concerned, there is
88 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

no difficulty in believing that such a terminology was already in use during

the reign of Antiochus, especially if we bear in mind that it had already been
employed at an earlier date in the Greek world.34 Besides, this terminology was
common to other Hellenistic dynasties such as the Antigonids of Macedonia: to
give an example, in his correspondence with the town of Mylasa in Caria Philip
V refers to his uncle Antigonos Doson as ‘father’.35 Even in this case πατήρ is
employed in acknowledgment of Antigonos’ role of tutor of the young Philip.
To this form of address one should associate the term adelphos, ‘brother’,
that is attested by several sources both for the Seleucids and the Ptolemies: this
is the case of Lysias, prime minister and tutor of Antiochus V,36 of Jonathan
Maccabaeus, high priest of Jerusalem, who is called ‘brother’ from Alexander
Balas and Demetrius II.37 Antiochus VIII (or IX) applies the same appellation to
Ptolemy X. Alexander I and an honorary decree for Zenas,38 son of Zenophanes
and high priest of the temple of Zeus Olbios at Diokaisareia in Kilikia, describes
this individual as ‘brother’ of the kings Philip I and Philip II Barypous.39 I agree
with Muccioli when he says that patēr and adelphos are appellations halfway
between the official sphere and the honorary one,40 and I would be inclined to
consider their employment as equivalent as that of the term paidion, that is to
say a way to lay emphasis on the special and privileged relationship between the
king (or other members of the royal family) and the officials of the kingdom.
Several royal titles are related to specific degrees of relationship between the
family members: father–son, mother–son, brother–brother, and brother–sister.
It is worth noting that in the Seleucid dynasty they officially begin to be in use
only from Seleucus IV, while in other dynasties, like the Ptolemies, they already
appear in the second generation. Father–son relations are attested by two royal
titles, i.e. Philopator and Eupator that in a certain way are the expression of
two different levels of relationship between king and prince. On the one hand,
Philopator emphasizes the special affection and loyalty of the son for his father,
on the other Eupator (which can be translated as ‘whose father is good’) is an
original creation of the Seleucids,41 or, to be precise, of Lysias, prime minister
of Antiochus IV and tutor of his young son Antiochus V. Lysias attributed
this official title to Antiochos V when his father died. Since there was another
rightful claimant to the throne, i.e. Demetrius I nephew of Antiochus IV and
first in the line of succession, Lysias created Eupator in order to stress the
legitimacy of the cadet branch of the family.
The father figure occupies an important place in royal ideology, especially
that of the Seleucid court where the prince was associated with his father in
the administration of the kingdom through the institution of the coregency. An
honorary decree from Didyma for the young Antiochus I praises the prince for
having followed the example of his father, Seleucus I, in honouring the temple
of Apollo by building a monumental portico:

Since Antiochus, the eldest son of king Seleucus, has never ceased to show the people of
Miletus much good will and support; and since now seeing the great interest his father
T h e L a n g uag e o f t h e Oikos a n d t h e L a n g uag e o f P ow e r 89

also shows to the sanctuary of Didyma, judging it good to follow the policy of his father,
he announces that he will build a portico (translated by Sherwin-White).42

The same Seleucus I was hailed by the army as the best father ever43 when he
decided to give his own young wife Stratonike in marriage to his son Antiochus
who had fallen desperately in love with her and had been about to die for love.44
The title Philometor designates the affection of the son towards his mother,
but it is possible to detect two different levels in meaning: the most basic and
immediate one is usually thought to emphasize the mutual affection between
mother and son in the framework of the propaganda on family unity. The
second aspect embedded in this title occurs during an extraordinary political
situation, when a queen is called to take on the control of the kingdom. That
is just the case of Seleucus V and Antiochus VIII Grypos, sons of Demetrius
II and Cleopatra Thea: actually, their mother reigned alone after the death of
her husband from 125 to 121 BC.45 Both Seleucus and Antiochus bore the
epithet Philometor, which in this circumstance does not refer only to feelings of
affection but also of loyalty and subordination, Cleopatra being the sole ruler. It
is no coincidence that Seleucus had been executed by order of his mother when
he claimed the throne for himself; similarly Antiochus, who was associated with
power a short time after, would have undergone the same fate, if he had not
made his mother drink the poison that she had prepared to murder him.
As for the title Philadelphos, it does not differ from the general cliché that
we find in other Hellenistic dynasties.46 Without doubt one of the most inter-
esting issues in the relationships between a king and his queen is the habit
of calling one another ‘brother’ and ‘sister’. In official documents they both
make their first appearance during the reign of Antiochus III. However, Livy
refers to Apame,47 wife of the founder of the Seleucid empire, as the sister of
the king: we should not take this passage literally, because Apame was in no
way related to Seleucus, but rather she was the daughter of Spitamenes, one of
the leaders of the Bactrian resistance against the Macedonian army during the
eastern campaign of Alexander the Great. Livy is likely to have literally trans-
lated the word ‘sister’ that he had found in his Greek sources, maybe because
he was unfamiliar with the technical language of the Hellenistic chancelleries,
and he did not realize that in this specific context ‘sister’ was a court title. Far
from being exclusively an oriental tradition, the habit of husband and wife
calling one another ‘brother’ and ‘sister’ is well attested in Classical Greece and
does not imply an endogamic union.48 It is then probable that the use of the
title of adelphos and adelphē was already in use at the Seleucid court from its
Although marriage between full siblings was generally prohibited, it was
accepted in royal families as the exception to the rule in order to preserve power
inside the same family unit and to avoid external interference. The Seleucids
had recourse to this practice later, when Antiochus III gave his daughter
Laodice in marriage to her own brother Antiochus. Several hypotheses have
90 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

been put forward to provide a possible explanation for the adoption of this
practice by the Seleucids, from Achaemenid to Egyptian influences.49 Within
the melting pot of Hellenistic culture, Oriental traditions may have provided
precedents and it is certain that in some situations they were employed in the
(self-)representation of the monarch and his family, but at the same time we
cannot leave out the Greco-Macedonian background both of the dynasty and
of the dominant ethno-class. As Ogden points out, such an event had already
occurred in the Argead dynasty with the marriage between Amyntas, son of
Archelaus, and his sister.50 This issue can also be analyzed from a religious
point of view. The marriage between Zeus, king of the gods, with his sister
Hera could provide a good parallel for this custom, given that the royal couple
was also considered the human projection of the divine one. Besides, a list of
priests from Seleukeia Pieria records the existence of a cult for Seleucus I as
Zeus Nikatoros,51 and may offer support to this hypothesis. This cult could be
placed in relation to the iconography adopted by Seleucus at an early stage of his
coinage and defined ‘Alexandrine type’ by Houghton and Lorber.52 In fact, on
the reverse of these issues one can notice the representation of Zeus Nikephoros
enthroned and holding a Nike and a sceptre. The iconography seems to attest
an intermediate phase between the will of perpetuation of the memory of
Alexander the Great (whose ancestor was Zeus via the mythical Temenus and
Herakles) and the making of Apollo’s cult as a vital element in the foundation
of the Seleucid dynasty.
The negative judgement of Polybius53 on the Hellenistic basileis and their
dynasties in comparison with the virtues of the Roman Republic exerted a strong
influence upon the next generations of historians. An exception to this view was
represented by the Attalids of Pergamon, who established their political propa-
ganda of family harmony and affection, a subject that became almost proverbial.
Philip V of Macedonia continuously exhorted his sons Perseus and Demetrius to
harmony by showing them the example of the Attalids whose familial concord
was the real secret of the prosperity of that kingdom.54 The peculiar social and
cultural environment of Egypt gave birth to the cult of the Theoi Adelphoi ‘the
Sibling Gods’, to celebrate Ptolemy II Philadelphos and his sister-wife Arsinoe.
One of the most evident signs of the popularity of the family language is offered
by the (re)foundation of a huge number of cities named after a close relative,55
e.g. Apameia, Laodikeia and Antioch (Syria), Nysa (Palestine) Stratonikeia
(Caria), Arsinoe (Egypt, Greece, Kilikia, Pamphylia), Berenike (Egypt, Epirus),
or after abstract concepts such as Philadelphia (Phrygia, Palestine, Egypt). To
sum up, the language of the family, although not as well documented as in the
case of the kings of Pergamon, was then widely spread among the Hellenistic
dynasties and was developed in several ways. In the eyes of the ancient histo-
rians the case of the Seleucid dynasty was considered as a negative model of
family, where plots, hate, treachery, greed and, sometimes, madness prevailed.
During the 248 years of its existence the Seleucid dynasty faced several periods
of domestic troubles that led to conflicts and caused a progressive weakening
T h e L a n g uag e o f t h e Oikos a n d t h e L a n g uag e o f P ow e r 91

of the kingdom. Nevertheless the extant evidence shows that in this particular
matter the Seleucids were in line with the policy followed by other Hellenistic
dynasties: the focus on affection among the members of the royal family was
always emphasized as an important issue in order to convey messages of unity
and legitimacy from the court towards its subjects.

  1 Virgilio, 2003, pp. 32, 70.
  2 On the negative judgement of Polybius about the Hellenistic monarchies see Virgilio, 2003, pp.
22, 193–197.
  3 Ogden, 1999, pp. 117–170.
  4 Rostovtzeff, 1941, p. 269.
  5 See Lacey, 1968, pp. 15–32; Pomeroy, 1997, pp. 17–21; Cox, 1998; Patterson, 1998; Nevett, 1999,
pp. 4–20; Ferrucci, 2006, pp. 186–192.
 6 I. Iasos 4, l.26: ‘τον οἶκον ἡμῶν’, ‘our house’.
 7 I. Ilion 32, ll.46–47: ‘εὔνοιαν ἣν ἔχων εἴς τε τὸν πα]-/τέρα αὐτοῦ βασιλέα Σ[έλευκον καὶ τὴμ
πᾶσαν βασιλικὴν οἰκίαν’: ‘the goodwill that they (the citizens of Ilion) had towards his father,
the King Seleucus, and all the royal house’.
 8 I. Erytrae 37, l.7: ‘εἰς τὴν ἡμετέραν οἰκίαν’, ‘towards our house’.
  9 See the famous sentence attributed to Seleucus I (Plut., An seni, 790a-b): ‘If only the people
knew how tiring it was to write and read so many letters, they would not pick up a diadem
which had been thrown away’.
10 I. Iasos 4, l. 26: ‘τον οἶκον ἡμῶν’, ‘our house’.
11 OGIS 223, I. Erythrai 37, l. 7.
12 Dunst et al., 1968, l. 10.
13 I. Ilion 32, l. 47.
14 IC III IV 9*, l. 97.
15 Oikia: ll. 46–47: ‘[εἴς τε τὸν πα]-/τέρα αὐτοῦ βασιλέα Σ[έλευκον καὶ τὴμ πᾶσαν βασιλικὴν
οἰκίαν’, ‘(the goodwill) towards his father, the king Seleucus, and all the royal house’. Pragmata:
ll. 6, 7, 14, 24.
16 I. e. πατήρ, ‘father’.
17 See the list of examples provided by De Ruggiero, 1961, s.v. domus, pp. 2059–2060. Apparently
the terms oikos and oikia are not discussed in Mason, 1974.
18 Germanicus, SE 210* 4, l. 31: ‘πρὸς σύμπαντα τον οἶκον ἡμῶν’, ‘(the goodwill) towards our
whole house’; SE 210* 5, l. 42: ‘[εὐ]σεβῶς πρὸς τον οἶκον ἡμῶν δια[κεῖσθαι] ’, ‘that (you) are in
a pious disposition towards our house’ and the emperor Tiberius, SE 210* 3, l. 21.
19 ll. 21–22.
20 I. Ephesos 2026, l. 9.
21 SEG XXXIX 1284, ll. 13–14: ‘ὑπὲρ τῆς τοῦ ἀδελφοῦ/ ἡμῶν βασιλέως Ἀντιόχου καὶ τῆς
ἡμετέρας καὶ τῶν παιδίων vac./σωτηρίας’, ‘on behalf of the safety of our brother, the king
Antiochus, of us and of our children’.
22 Iasos 4, l. 62: ‘βασιλίσσης Λαοδίκης καὶ τῶν τέκνων αὐτῶν’, ‘(on behalf of the king Antiochus),
the queen Laodike and their children’.
23 ll. 1–4: ‘ Ὑπὲρ βασιλέως μεγάλου/ Ἀντιόχου καὶ Ἀντιόχου/ τοῦ ὑιοῦ καὶ βασιλίσσης/ Λαοδίκης
καὶ τῶν παιδίων κτλ’, ‘on behalf of the great king Antiochus, his son Antiochus, the queen
Laodike and the children’. See Sayar, 2001, pp. 227–234.
24 SEG XIX 904, l. 4: ‘[Κλεοπά]τ[ρ]ας Θεᾶς {ε. ..ς} Εὐετηρία.ς [καὶ] τ.ῶν. π.α.ι.δίων’, ‘ of Cleopatra
Thea Eueteria and the children’. A couple of letters survived in the space where the name of the
dedicant is expected, making it impossible to trace back his identity.
25 Ma, 2002, pp. 179–242.
26 Ibid., pp. 200–201.
92 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

27 Ibid., p. 238.
28 Capdetrey, 2007, p. 236.
29 Pomeroy, 1998, p. 18.
30 AJ 12.148.2. On Zeuxis see Savalli-Lestrade, 1998, n. 39.
31 On the career of Lasthenes see the catalogue by Savalli-Lestrade, 1998, n. 80.
32 AJ 13.127, but see also I Macc. 11, 32.
33 Bikerman, 1938, pp. 43–46, thought that in the Seleucid kingdom the title of patēr was given to
people who had also the title of syngenes, but against this view and for a discussion of the title
of ‘father’ see Gauger, 1977, pp. 3–4, 32, 329–334; Savalli-Lestrade, 1998, pp. 36–38; Muccioli,
2000, pp. 252–257.
34 Pomeroy, 1998, p. 18.
35 I. Labraunda III, n. 5 (ll. 6–7, 48), 7 (l. 12).
36 II Macc. 11.22.
37 I Macc. 10.18 and I Macc. 11.30–32.
38 OGIS 257, l. 2.
39 SEG XXVI 1460, ll. 5–6. See also Verilhac et al., 1974, pp. 237–242.
40 Muccioli, 2000, pp. 256–257.
41 Muccioli, 1996, pp. 21–35.
42 I. Didyma, 7, ll. 2–9: ἐπειδὴ Ἀντίοχος ὁ πρεσβύτατο[ς] / τοῦ βασιλέως Σελεύκου <ὑιὸς>
πρότερόν τε πολ]λὴν/ [ε]ὔνοιαν καὶ προθυμίαν παρεχόμενος δι[ετέλει]/ [πε]ρὶ τον δῆμον τον
Μ[ι]λησίων καὶ νῦν ὀρ[ῶν τòν]/ [π]ατέρα τον αὐτο[ῦ τ]ὴν πᾶσαν σπουδὴ[ν ποιούμε]­/[νο]ν
περὶ τον ἱερὸν [τò ἐ]ν Διδύμοις καλῶς ἔχε<ι>ν ε[ἶναι] /[ ὑπ]ολαμβάν[ων ἐπ]ακολουθεῖν τῆι
τοῦ πατ[ρòς προ]-/[αιρ]έσε[ι ἐ]π[αγγ]έλ[λε]ται στοὰν οἰκοδο[μήσειν κτλ.
43 App., Syr., 327.
44 Breebaart, 1967, pp. 154–164.
45 On the social and political role of the Hellenistic queens see Bielman-Sánchez, 2003, pp. 41–61;
Savalli-Lestrade, 2003, pp. 59–76.
46 Muccioli, 1994, pp. 402–422; Muccioli, 1995, pp. 41–55.
47 38.13.5.
48 Pomeroy, 1997, p. 17–18.
49 Ogden, 1999, pp. 119, 124–126; Bielman-Sánchez, 2003, p. 50.
50 Ogden, 1999, p. 125.
51 OGIS 245, ll. 10–11. Second century BC. Zenobios son of Zeno, priest of Seleucus Zeus
Nikator, Antiochos (I) Theos, Seleucos (II) Callinicos, Seleucus (III) Soter, Antiochus and
Antiochus (III) the Great.
52 Houghton et al., 2002. See for example Sardes, 3–5; Tarsus 10, Antioch on the Orontes, 13;
Seleucia in Pieria, 27–31; Laodicea by the sea, 36–37; Carrhae, 41–46; Babylon, 82–83, 87.
53 Fifth book.
54 Polyb., 23.11.
55 For a general overview see Cohen, 1995; Cohen, 2006.

Bielman-Sánchez, A. (2003),‘Régner au féminin. Réflexions sur les reines attalides
et séleucides’, in F. Prost, ed., L’Orient Méditerranéen de la Mort d’Alexandre
aux Campagnes de Pompée. Cité et Royaumes à l’Epoque Hellénistique. Actes
du Colloque de la SOPHAU, Rennes: Presses Universitaires de Rennes, pp.
Bikerman, E. (1938), Institutions des Séleucides, Paris: P. Geuthner.
Breebaart, A. B. (1967), ‘King Seleucus I, Antiochus, and Stratonice’, Mnemosyne,
20, (2), 154–164.
T h e L a n g uag e o f t h e Oikos a n d t h e L a n g uag e o f P ow e r 93

Capdetrey, L. (2007), Le Pouvoir Séleucide. Territoire, Administration,

Finances d’un Royaume Hellénistique (312–129 avant J.-C.), Rennes: Presses
Universitaires de Rennes.
Cohen, G. M. (1995), The Hellenistic Settlement in Europe, the Islands and Asia
Minor, Berkeley, Los Angeles, Oxford: University of California Press.
—(2006), The Hellenistic Settlement in Syria, the Red Sea Basin, and North
Africa, Berkeley, Los Angeles, London: University of California Press.
De Ruggiero, E. (1961), Dizionario Epigrafico di Antichità Romane, Vol. 2, Pt. 3,
Roma: L. Pasqualucci.
Dunst, G. and Merkelbach, R. (1968), ‘Zu dem neuen epigraphischen Dokument
aus Teos’, ZPE, 3, 170–174.
Ferrucci, S. (2006), ‘L’“oikos” nel diritto attico. Pubblico, privato e individuale
nella democrazia ateniese classica’, Dike, 9, 183–210.
Gauger, J.-D. (1977), Beiträge zur jüdischen Apologetik. Untersuchungen zur
Authentizität von Urkunden bei Flavius Josephus und im I. Makkabäerbuch,
Köln-Bonn: P. Hanstein.
Houghton, A. and Lorber, C. (2002), Seleucid Coins. A Comprehensive Catalogue,
Pt 1: Seleucus I-Antiochus III, Lancaster, PA-New York: The American
Numismatic Society and Classical Numismatic Group.
Lacey, W. K. (1968), The Family in Classical Greece, London: Thames and Hudson.
Ma, J. (2002), Antiochos III and the Cities of Western Asia Minor, Oxford: Oxford
University Press.
Mason, H. J. (1974), Greek Terms for Roman Institutions. A Lexicon and
Analysis, Toronto: Hakkert.
Muccioli, F. (1994), ‘Considerazioni generali sull’epiteto “Φιλάδελφος” nelle
dinastie ellenistiche e sulla sua applicazione nella titolatura degli ultimi
Seleucidi’, Historia, 43, (4), 402–422.
—(1995), ‘Gli epiteti di Demetrio II re di Siria’, Simblos, 1, 41–56.
—(1996), ‘Εὐπάτωρ nella titolatura ellenistica’, Historia, 45, (1), 21–35.
—(2000), ‘Crisi e trasformazione del regno seleucide tra il II e il I secolo AC:
titolatura, ruolo e competenze dei Suggeneis’, in Mooren, L. ed., Politics,
Administration and Society in the Hellenistic and Roman world, Proceedings
of the International Colloquium, Bertinoro 19–24 July 1997, Leuven: Peeters,
pp. 251–274.
—(2001), ‘La scelta delle titolature dei Seleucidi: il ruolo dei “φιλοι” e delle
classi dirigenti cittadine’, Simblos, 3, 295–318.
Nevett, C. L. (1999), House and Society in the Ancient Greek World, Cambridge:
Cambridge University Press.
Ogden, D. (1999), Polygamy, Prostitutes and Death. The Hellenistic Dynasties,
London: Duckworth/ Classical Press of Wales.
Patterson, C. B. (1998), The Family in Greek History, Cambridge, MA: Harvard
University Press.
Pomeroy, S. B. (1997), Families in Classical and Hellenistic Greece: Representations
and Realities, New York: Oxford University Press.
94 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

Rostovtzeff, M. (1941), The Social and Economic History of the Hellenistic World,
Vol. 1, Oxford: The Clarendon Press.
Savalli-Lestrade, I. (1998), Les Philoi Royaux dans l’Asie Hellénistique, Genève:
—(2003), ‘La place des reines à la cour et dans le royaume à l’époque hellé-
nistique’, in Frei-Stolba, R., Bielman, A., and Bianchi, O. (eds), Les femmes
Antiques entre Sphère Privée et Sphère Publique, Berne: P. Lang, pp. 59–76.
Sayar, M. H. (2001), ‘Von Kilikien bis Thrakien : neue Erkenntnisse zur Politik
Antiochos III. Zwischen 197–195 v. Chr. Anhand von zwei neugefundenen
Inschriften’, in Bresson, A. and Descat, R. (eds), Les Cités d’Asie Mineure
Occidentale au IIe Siècle AC, Paris-Bordeaux: Editions Ausonius, pp. 227–234.
Vérilhac, A. M. and Dagron, G. (1974), ‘Une nouvelle inscription du temple de
Zeus à Diocésarée Uzuncaburç (Cilicie)’, REA, 76, 237–242.
Virgilio, B. (2003), Lancia, Diadema e Porpora. Il Re e la Regalità Ellenistica,
Studi Ellenistici XIV, Pisa-Roma: Giardini Editori e Stampatori.

Inheritance, Priesthoods and Succession in Classical

Athens: the Hierophantai of the Eumolpidai
William S. Bubelis

For all of the exacting attention that scholars have long paid to the legal,
economic, demographic, and social dimensions of the oikos in Classical Athens,
still much remains unclear and unsatisfactory. A particular problem concerns
the degree to which an ancestral priesthood might always have been ensconced
within a given oikos and thus attached to a given patriline connecting a priest
to his direct biological ancestors. The very primacy of the oikos itself, however,
and its close connection to the nuclear family, should not be taken for granted.
This assumption has guided much of the work on the matter of succession
among ancestral Athenian priesthoods, so much so as to hinder an accurate
and reasonable understanding of the evidence at our disposal. Indeed, the
appeal of the assumption has proven strong enough that few scholars have
seen that inheritance of a priesthood might be a distinct institution in its own
right, markedly different from that surrounding the oikos and the normal trans-
mission of rights, property, and obligations.1
Any inquiry into the familial dimension of priesthoods ought therefore to
yield a stimulating trove of new perspectives and information, especially those
which challenge the underlying assumptions that have driven so much prior
The methodological challenges at hand nevertheless arise for understandable
reasons. Lacking any detailed account of priestly succession in ancient literature,
scholars have long sought to exploit the prosopographic data, drawn mainly
from inscriptions and a few literary texts, in order to determine the principles
involved. While that evidence is fragmentary and, especially in epigraphic texts,
opaque, the occurrence of patrilineal farther – son family relationship patterns
through time have long been unquestioned. Notably, such transmission would
also be hard to distinguish from the devolution of non-priestly property, rights,
and obligations more or less bound up with the oikos of a deceased priest.2 (That
priests might, through force or by their own will, abdicate their office while
obviously remaining in possession of their oikos has not yet entered the debate,
and I will return to this question later.) It is curious that although scholars have
known for some time that priestly succession among women could not have
conformed to any oikos-centred model, none seem to have questioned whether
the same might also have been true among some of their male counterparts.3
96 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

Indeed, the evidence for exclusively father-son inheritance is rather weak, for
while such patrilineal transmission might appear to be the rule for some genê
over long periods of time, those lines only seem to terminate when an external
individual assumes the priesthood.
Much more problematic are those instances where an extra-patrilineal
individual seems merely to interrupt such continuity, allowing father-son lines
to resume only after the tenure of some other person. Uncomfortable with these
patterns, a wide range of scholars has suggested that while certain genê did
employ such patrilineal inheritance (e.g. the Eteoboutadai) others must have
used allotment (e.g. the Kerykes) or a kind of priestly election or allotment
(e.g. the Eumolpidai) as the preferred or customary way of choosing their
priests.4 If it were true that the priesthood was restricted by patrilineal lines of
succession, any disjunction between the known pattern of male priests and the
conventions of inheritance would point to a fundamentally different system,
and thus require one of the proposed solutions or further suggestions. In short,
for many scholars the choice has seemed to be one of two mutually exclusive
options: priesthoods devolved along normal lines of inheritance and thus were
tied to the oikos, or the priesthood and oikos were very much separate from one
The evidence for the Eumolpidai and their most important priest, however,
the hierophantês, well illustrates how a priesthood and oikos might be entangled
with one another and yet remain distinct entities. Moreover, the critical patterns
are well enough attested that there is some hope for this particular genos to
serve as a model with which to explore wider problems involving the study of
the Athenian family in the Classical period. From the late fifth century through
to the end of the fourth century, there appear to be too many hierophantai over
almost any period of time for the office to have passed from father to son.5
Following K. Clinton’s (1974) prosopography (clarified by Lambert and Blok
[2009]), a brief list of seven identifiable hierophantai from this period along
with their precise or approximate dates renders the issue clear:6

• Theodoros (415–08)
• Arkhias (379)
• Lakrateides (in office from before 353, to at least 350/49)
• Hierokleides son of Teisamenos of Paiania (middle of the fourth century;
possibly before Lakrateides)
• Biottos (336/5–333/2)
• Eurymedon (323)
• Eurykleides (317–07)

In what is still the most thorough assessment of the hierophantai, Clinton,

following Foucart (1900) and Martha (1882), argues that given the fact of so
many hierophantai in such a short period of time the genos must have elected
the priests from some or another pool of candidates.7 The positive argument
I n h e r i ta n c e , P r i e st ho o d s a n d Su c c e s sio n 97

for election as the method actually used depends entirely upon a letter from
the Roman emperor Marcus Aurelius to the Athenians about membership in
the genê. It affirms, ever so briefly, some form of selection (arkhairesia), but we
do not know for what kind of priestly duties this method was invoked or how it
was supposed to work.8 Moreover, since the evidence speaks to contemporary
institutions and does not even pretend to have any basis in that earlier period, it
has little relevance for Classical Athens. More recently, Lambert and Blok have
subscribed to the hypothesis of allotment (following Aleshire, 1994), on very
general considerations such as the use of allotment for most magistracies of the
contemporary Athenian democracy, but with no direct bearing on the manner
in which families or larger cult groups might have operated.9 Consequently it
would seem that the pattern of known hierophantai has been regarded so utterly
at odds with the biological principle underlying patrilineal inheritance that
speculations dismissing the oikos have seemed especially inviting. Alternative
explanations, however, that do maintain the primacy of biological descent in
some form might still provide a key by which to unlock this particular pattern,
and potentially those of other Athenian priesthoods without invoking institu-
tions that are, at best, poorly attested or inappropriate.
Underlying the hypotheses of election and allotment is the notion that the
genos typically had the capacity to choose successors, or even to approve of
those selected by the priest’s closest kin. In point of fact, we simply do not
know how the priestly family related to its genos, much less how extensive a
genos might be, and thus by what degree of affinity its members might have
been related to the priest. In no case do we possess evidence that a genos ever
exercised authority over such matters as priestly succession.10 Without positive
evidence for any one particular method, however, it is difficult to disprove the
election hypothesis: since whatever degree of biological affinity there was may
simply reflect the political prominence of a single family, which in some genera-
tions might have resulted in a father being succeeded in the priesthood by a
son, and in other instances, by a brother or nephew and so on. The prosopo-
graphical evidence, however, in no way requires a method such as election or
sortition to explain the visible patterns. Since in no case but that of Hierokleides
was Clinton (or Lambert and Blok) able to establish both the patronymic and
demotic, it stands to reason that there is no positive evidence to show that these
hierophantai were not related to one another. Therefore, we should hesitate to
attribute much power to the genos without first exhausting the possibility that
the succession principle did in fact involve biological descent and affinity.
This principle is no less valid in relation to the ancient literary testimony
suggesting perhaps that some hierophantai were unmarried or sexually
abstinent. If applicable to the Classical Athenian priests (as Foucart maintained),
such evidence would surely nullify, or at least complicate, any claim that
the priesthood devolved according to inheritance, and situate election and
allotment in a more favourable light. In discussing how the people of Phlious
in Arcadia conducted their Mysteries, Pausanias (2.14) states that the local
98 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

hierophantai did not hold the priesthood for life but were elected at each
quadrennial celebration. They could even take a wife if they so wished, and
that these matters (as a whole) were the only points of difference from how the
rites were observed at Eleusis.11 While Pausanias should put to rest any further
doubt about elections at Eleusis, at least in the Roman period, his particular
phrasing with regard to wives should not be taken as evidence for the incompat-
ibility of the office with heritable descent. Pausanias does not speak of sexual
purity, and certainly not a lifetime of it, such as would obviate the production
of legitimate male heirs. Instead, he uses the standard language of taking a wife
in marriage (lambanein gynaika).12 In other words, these other hierophantai
could undertake a marriage while priests, not that they were merely allowed to
retain the wives that they already had. The contrast with the Eleusinian hiero-
phantai, then, seems to be that at a minimum the Athenian priests could not
undertake a new marriage, not that they were required to be celibate altogether,
and certainly not that they were required to maintain a lifetime of sexual purity,
such that would prevent them from having legitimate male children of their
own.13 If there was no priestly celibacy, in sum, we are compelled to look again
to close biological connections between hierophantai.
My alternative hypothesis is that the office of hierophantês was usually trans-
mitted through a single patriline, but with one critical difference from direct,
or rather exclusive father-to-son succession. The evidence best fits a model in
which the oldest available male held the office passing it upon his death to the
next oldest male still alive within his generation. Among a group of brothers,
the priesthood would devolve to the next oldest in turn until the last brother
had deceased, at which time the priesthood would move to the eldest son of
that eldest brother, and so on until all his (that is, the eldest son’s) brothers had
a chance to hold the office. In other words, transmission of priesthood from a
father to son would allow the office to pass on to the next generation but would
ultimately be delayed until all of the priest’s brothers held it in turn. What might
occur if the eldest brother among a rank of siblings had no sons or living sons is
unknown, but the sons of the next oldest brother (and thus the cousins of those
closest to priesthood) would on principle be the next closest males available to
assume the office.
This hypothesis thus marks a significant break from previous approaches
whose attention to patrilineal succession and the oikos fully eclipsed consid-
eration of alternative modes of transmission, and certainly the potential role
that brothers might play in the transfer of the priesthood in a given generation.
In many cases, by virtue of the living conditions prevalent in ancient society,
some prospective priests would have pre-deceased their tenure as hierophantês,
whether brothers or sons of priests, thus creating the impression of familial
discontinuity that prevails in our prosopography. If a young and healthy man
assumed the priesthood, then it is quite possible that he would hold it for many
decades; upon his death, the office would then transfer to his (living) brothers,
all of whom would be relatively aged, and all the more so as each brother died
I n h e r i ta n c e , P r i e st ho o d s a n d Su c c e s sio n 99

and passed the priesthood to his next oldest sibling. The death, and transfer,
of any office among already aged men would inevitably result in rapid transfer
of the office, thereby easily substantiating the (mistaken) impression in our
evidence that the hierophantai could not possibly have inherited the office.
The hypothesis of fraternal transfer would also meet the objection that not all
direct male ascendants or descendants of a hierophantês held office. A case in
point is that of one hierophantês from the early fifth century, Zakoros, who
probably did not pass the priesthood on to his son Diokles, and thus not to
his grandson, whose name is unknown to us but who cites his grandfather’s
priesthood in a speech of the late fifth century.14 Of course, it is possible that
Zakoros was a maternal grandfather, thus rendering the issue moot. Yet Blok
and Lambert note that Diokles might well have had a brother who did take up
the priesthood.15 On my hypothesis, that supposition alone would require the
brother to be older than Diokles, for it would have been that brother’s own male
children who would be eligible for another round of fraternal transfers, rather
than Diokles’ own son.
The hypothesis of fraternal transfer might be difficult to test but one case that
offers some support involves the hierophantai Hierokleides and Lakrateides,
both of whom served in this capacity in the middle of the fourth century.
Lakrateides appears to have acquired his office some time before 353 BC, the
date of a speech of Isaios (7) in which this priest figures. Given that as hiero-
phantês Lakrateides was of marriageable age (that is, about 30) in 394/3 BC,
when Isaios (7.9) mentions that he accompanied the Athenians on the way to
what would become the battle of Corinth, he should have been born no later
than c.423/2 BC.16 Thus, it would be almost impossible that Lakrateides could
have received the priesthood from his father unless his father was close to 100
years of age.17 While an Arkhias held the office in 379 BC, and could conceivably
be Lakrateides’ father, our lack of knowledge about Arkhias’ family connections
or even his deme prevents certainty. Yet, as I hope to show, even Arkhias’ tenure
as hierophantês may still conform to the model offered here, and will return to
discuss him shortly. Lakrateides is attested in this office as late as 350/49 BC,
close in time when a Hierokleides of Paiania, son of Teisamenos, also held the
office.18 The difficulty that Hierokleides poses is that our sole evidence for him,
an honorific decree of the Eleusinians, cannot yet be dated to anything more
precise than the middle of the fourth century.19 Thus, we do not know whether
Hierokleides preceded or succeeded Lakrateides, and by how much time.
Nevertheless, there is some reason to think that the two men could have
been brothers. In the first instance, it is clear that Hierokleides was likely to
have also been of significant age at the time when he held office. As Clinton
suggested, Hierokleides’ father Teisamenos may be identical with a homon-
ymous tamias of Athena attested in 414/3 BC.20 Although we cannot be sure
when Hierokleides was born, a fifth century date seems likely, and would be in
keeping with Lakrateides’ probable birth year c.423/2 BC. Much more positive
is the evidence that both Lakrateides and Hierokleides hailed from the same
100 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

deme: Paiania. While the patronym and demotic of Lakrateides as hierophantês

remains unknown, one Lakrateides from Paiania did dedicate a statue (whose
base can only be dated to the fourth century) that was perhaps situated origi-
nally in the City Eleusinion.21 A dedication there would comport well with this
Lakrateides being identified as the very hierophantês. Because the dedicator
and Hierokleides obviously hailed from the same deme, Paiania, the critical
question is whether there is sufficient probability for the two Lakrateides to
be identical, and secondly, whether Lakrateides and Hierokleides could be
biologically related to one another.22 Lakrateides’ name is so rare before the
Late Hellenistic period that we may, with all due caution, identify the men
as one and the same.23 A further step is required. If both mid-century hiero-
phantai did hail from the same deme and were roughly of the same age, then
it is conceivable that Lakrateides and Hierokleides were indeed both sons of
Teisamenos of Paiania, and thus brothers.
As noted earlier, hierophantai like Arkhias and the others who come later
in the fourth century have long posed a problem because their mere existence
seems to obviate any transmittance of the office through familial lines. Since
we do not know Arkhias’ patronymic or demotic, however, we cannot also
rule out that he too was a son of Teisamenos, or even a (younger) brother
of Teisamenos serving then at an advanced age. The eventual transfer of the
office to Biottos, about whom nothing else is known, likewise poses only a
minor quandary since Biottos could easily have been a son of Hierokleides or
Lakrateides.24 The more substantive problem, and indeed to some it might seem
to be an impenetrable one, is that Teisamenos would have been alive when one
Theodoros was hierophantês. But again, the difficulty arises from assuming that
father-son transmission of a lifetime office is the only possible means by which
the Eumolpidai could have managed this office. We do not know, and need not
assume, that Teisamenos held the position for very long, if at all, and we do not
know that he and Theodoros were not brothers, or even that they were uncle and
nephew. After all, the hypothesis of fraternal transfer from eldest to youngest,
and then back to the eldest son of the eldest brother, would sensibly cover all
these situations. The death, and transfer, of any office among already aged men
would inevitably result in rapid transfer of the office until a younger or healthier
occupant might hold it again for a lengthy period of time. Such circumstances
would also accord well with the fact that another hierophantês, Eurymedon,
occupied the office in 323 BC, whose familial relationship likewise cannot be
ruled out because his demotic and patronymic are unknown. Yet as early as
317 BC, one Eurykleides might also have served as hierophantês, if he is not in
fact a figment on the part of Diogenes Laertius (2.101) who mentions him as a
humorless priest in the reign of Demetrios of Phaleron. Since both Eurykleides
and Eurymedon are alleged by Diogenes to have laid charges of impiety against
one or another philosopher, and possess strikingly similar names, there is good
ground to think that these men are simply one and the same, although which
name would be the historically correct one cannot now be discovered.25
I n h e r i ta n c e , P r i e st ho o d s a n d Su c c e s sio n 101

A final intriguing problem remains. With each generation there is always

the possibility of creating more collateral lines, each with a potential claim
upon the priesthood, unless the degree of affinity (ankhisteia tou genous) was
perhaps more sharply curtailed compared to that usually employed for familial
inheritance in the oikos. But once a particular family line was extinguished for
lack of available adult males, the gap would have to be filled somehow. One
possible means of doing that is adoption, in which case a hierophantês who
had no younger brothers or nephews to whom he would naturally pass his
office might adopt a man from within the genos and perhaps already a close
collateral relative, surely with legitimate male children of his own.26 Our lack of
knowledge as to the makeup and organization of any genos, however, only begs
the question: what degree of affinity bound the other members of the genos to
the priest and that person’s oikos or patriline? Although this problem seems to
defy easy solution, the inevitability of low life expectancy and the possibility
that some priests might abdicate or be found unfit for office might suggest
that the genos, or at least those with the closest connection to the patriline was
involved when that line was threatened by extinction. But it would be make
good sense that those closest to the patriline would have the greatest say, or
themselves be candidates should that line actually go extinct. Such matters did
arise as a matter of course, and one case of abdication is known, involving the
Eteoboutadai in the late fourth century, when the priest of Poseidon-Erekhtheus
abdicated in favour of his younger brother.27 Curiously, the other branch of the
Eteoboutadai, who supplied the priestess of Athena Polias, also used a type of
fraternal transfer. The priestess was simply the eldest daughter of the eldest
son in a given generation; upon her death (or abdication) the office reverted
to her niece, the eldest daughter of her eldest brother.28 Such cases, real and
hypothetical alike, raise the possibility that the very system of fraternal transfer
arose (to safeguard priesthood) within the immediate family of the priest until
it could be safely transmitted to the next available descendant.
While the model that I have proposed is utterly contingent upon the
available evidence and will surely require revision, or even rejection, with
the discovery of new evidence, it does nonetheless provide a new path for
exploring the relationship between oikos and priesthood. Many fundamental
questions remain, but what the evidence concerning the hierophantai suggests
is that priestly succession might well have operated according to rules and
principles that simultaneously stood apart from the oikos but respected it as
integral to the identity of family and its priesthood. That brothers might have
played such a key role in succession as I have suggested is hardly certain, but
it is more likely that they, rather than the genos, constituted the critical factor
behind the observable pattern of hierophantai. For brothers to have possessed
such a position in (a matter hitherto regarded as restricted to fathers and sons)
raises still more intriguing questions about the potential for sibling conflict and
companionship affected or shaped by familial obligations. In sum, the oikos was
never far removed from priestly succession, for the oikos of long-dead priests,
102 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

all in principle extending that of the genos’ ultimate ancestor, still notionally
held together a string of descendants, all with their individual oikoi. In this
manner, at least one of the major priestly families of Classical Athens, the
Eumolpidai, seems to have been even more bound to the idea of the oikos, even
when in practice they operated a parallel system of succession.

  1 Even in large surveys of Attic family law and inheritance, the matter of sacral inheritance has
received very little attention: e.g. MacDowell, 1978, pp. 84–108; Wurm, 1972; Harrison, 1968,
pp. 82–162; Lacey, 1968, pp. 125–50; Broadbent, 1968, pp. 113–239; Savage, 1907, pp. 94–96,
102–06, 128–133; Lipsius, 1905–1915, pp. 339–58, 463–588.
  2 On inheritance and the Attic oikos, see especially MacDowell, 1989.
 3 Upon the death or incapacity of the Eteoboutad priestess of Athena Polias, for example, the
preferred candidate for the office was evidently the eldest daughter of the priestess’ eldest brother,
for which see Turner, 1983, p. 249; Lewis, 1997; Aleshire, 1994, pp. 332–333; followed by Connelly,
2007, p. 47. Barring intermarriage between collateral lines (or incest), a priestess’ daughter was
evidently not in line to become a priestess herself. Given that the priesthood was held for life, and
one priestess served for fully 64 years, it is not difficult to see why the office could skip an entire
generation in some cases, but it is unclear what might occur if, for example, the priestess’ eldest
niece by her eldest brother predeceased her but left a daughter of her own. In the Hellenistic
period, one priestess immediately preceded her own niece as priestess, thus fulfilling the primary
rule of descent, but the next known priestess within the immediate family is none other than
the latter priestess’ great-grand-daughter, thus seeming to violate the implicit rule against matri-
lineal succession in this priesthood, for which see Turner, 1983, p. 254. By preferring the eldest
daughter of the priestesses’ eldest brother, in effect the priesthood reverted to a single unique
patriline every time it was open. This principle of succession thus preserved the office within a
patrilineal succession but clearly diverging from normal practice since the eldest brother became
the patrilineal point of reference, not the father of the priestess and her brothers. Clearly, female
priesthoods in Athens did not entirely adhere to the oikos as the exclusive conduit through which
property and obligations had to be transferred, though prospective priestesses might certainly
have had to do so in many other respects or periods of their life. One feature of this pattern is
suggestive with respect to male priesthoods, namely, that when death, removal, or abdication of a
priestess could on some occasions require a cousin of a deceased priestess to assume the office, the
line of descent would also give the appearance that the priesthood had left one particular family
line, only to reappear several generations later, should the various collateral lines intermarry.
  4 Earlier treatments in support of inheritance (usually for specific cases, often the Eteoboutadai):
Stengel, 1920, p. 44; Hignett, 1952, pp. 64–65; Davies, 1971, pp. 172–173; Kearns, 1989, p. 71 n.
37 (unaware perhaps of Clinton, 1974, pp. 50–58 and 67); Pomeroy, 1997, pp. 149–151. Turner,
1983, pp. 249–260, offers the most extensive account of succession in female priesthoods (that
of the Eteoboutad priestess of Athena Polias), arguing for the transfer of the priesthood by
inheritance through the eldest daughter of a priestess’ eldest brother.
  5 Clinton 1974, p. 17. In their entry for Hierokleides (n. 16), Osborne and Byrne, 1994 (hereafter
abbreviated as LGPN) imply that he may be the father of Teisamenos II, but provide no further
  6 For precise details on the evidence for each hierophantês if not discussed below, see Clinton,
1974, nos. 3–9.
  7 Ibid., p. 45; Foucart, 1900, 25; Martha, 1882, pp. 35–39.
  8 The text in question is an inscribed letter of Marcus Aurelius to the Athenians concerning
disputes over religious privileges between Athenians. For text and analysis of the letter (and the
key phrase, at l. 11) see Oliver, 1970, with Greek text at pp. 3–9. I hope to treat this letter and
the position of the genê in Roman Athens in another venue.
I n h e r i ta n c e , P r i e st ho o d s a n d Su c c e s sio n 103

  9 Lambert and Blok, 2009, pp. 96–99, 115–116; Aleshire, 1994, p. 330. I hope to treat their
general claims, and to examine one or two other Athenian genê, in greater detail elsewhere.
10 On the Attic genos much has been written, but see the seminal work of Bourriot, 1976.
11 Foucart, 1900,pp. 25–28, takes Pausanias to mean that Athenian hierophantai could not have
wives, and since they held the priesthood for life, they would, on his reading of Arrian and
Pausanias together, pp. 25–28, be unable ever to take wives.
12 Examples of this phrase are easily forthcoming in Attic literature, for example: Isoc. 9.50; Dem.
34.8; Men. Samia 577. Pausanias uses the phrase several other times, each with this meaning:
4.19.6; 7.3.2; 9.34.9.
13 The claim of sexual purity is often advanced further on the basis of Arrian, (Arr. Epict. Diss.
3.21.16) who aims to disabuse others of attempting to lecture without proper preparation.
Since Foucart, 1900, 24–25, some scholars (e.g. Parker, 1983, 86–95 (periods of sexual absti-
nence); see also Stengel, 1920, pp. 31–48; Ziehen, 1913) have taken Arrian to mean that the
hierophantês was to remain sexually pure, although the term can apply more broadly to other
forms of purity or purification (such as hagneia) as Arrian himself suggests earlier. Even on the
narrow reading the term does not require us to infer that the man had to be sexually pure before
his assumption of the priesthood, as Foucart assumed.
14 Lys. 6.54.
15 Lambert and Blok, 2009, p. 115 n. 108, with Todd, 2007, p. 474.
16 See Clinton, 1974, p. 17, for the basic calculations. For the age of marriage for Classical
Athenian men, I use the figure of thirty that is often thought to be approximately correct, for
which see Sallares, 1991, pp. 148–149, and Lacey, 1968, pp. 106–107.
17 Lakrateides: Clinton, 1974, pp. 17–18 n. 4.
18 The date depends upon Didymos’ commentary (13.52; 14.42) to Demosthenes’ On Organization
19 IG II2 1188.
20 The evidence for Teisamenos is in IG I3 309, l. 2 (and restored at 333, l. 2; 355, l. 2; 371, l. 2).
LGPN does not identify the men as one and the same but provides two distinct lemmata: nos.
21 For the dedication and its brief text, see Meritt, et al., 1957, p. 216, n. 66. Note also that the
inscription has enough letters (c. 9) within which to squeeze the patronymic of ‘Teisamenou’
if spelled out fully. The editors dated the statue base only to the fourth century, and associate it
with the City Eleusinion only upon the basis, it seems, of general probability. Miles, 1998, does
not discuss the dedication in her recent treatment of the City Eleusinion.
22 Clinton, 1974, p. 18, does note the dedication and raised the possibility of identification, but
makes no mention of a possible link with Hierokleides. Lambert and Blok, 2009, p. 114, n. 104,
are uncertain about the identification and do not note the shared deme.
23 LGPN is able to cite only three other men (nos. 1–3) of this name from before the mid-fourth
century; the next attested (n. 8) is Late Hellenistic.
24 Biottos (LGPN n. 1 = Clinton, 1974, p. 20, n. 6) is attested only in the inventory of the Eleusinian
epistatai that covered the years 336–333/2 BC (IG II2 1544, l. 35), about which it is not possible
to be more precise concerning the terminus ante quem of Biottos’ tenure. Hierokledes is known
to have had a son, Teisamenos (II), who is attested at about the same time: IG II2 1496, l. 60.
25 Eurymedon: Clinton, 1974, p. 21, n. 7; LGPN, n. 3. Eurykleides: Clinton, 1974, pp. 21–22, n. 8.
LGPN, n. 4. Lambert and Blok, 2009, p. 114, n. 105, also raise the possibility of conflation.
26 Adoption might also explain a troublesome feature of the Hellenistic evidence, namely, that the
priesthood appears to switch between families of different demes. The first sign of change comes
with the late third century, when in less than a century three men from three different demes
held the office of hierophantês. Although these were probably not closely collateral branches, or
else they would have hailed from the same deme, we thus must consider that the genos did have
a stronger role in this period than the evidence otherwise would suggest. In any case, as known
from fourth-century Athens mainly, adoption would require that the man adopted change his
deme to that of the adoptive father, for which see Whitehead, 1986, pp. 68, 103.
27 Among the Eteoboutadai, the priesthood of Poseidon-Erekhtheus appears to have been
passed from father to eldest son, but this could actually be transferred by some ritual act
104 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

with legal force. According to Plut. Mor. 843E-F, in the late fourth century, the eldest son of
Lykourgos inherited the priesthood but transferred it permanently to a younger brother, whose
descendants thereafter maintained the office; in fact, the younger brother now took – legally
it appears – the position of eldest son for ritual purposes. In neither case, however, did the
transfer at all affect the oikos of either brother.
28 See n. 3 above.

Aleshire, S. B. (1994), ‘The demos and the priests: the selection of sacred officials
at Athens from Cleisthenes to Augustus’, in Osborne, R. and Hornblower, S.
(eds), Ritual, Finance, and Politics: Athenian Democratic Accounts Presented
to David Lewis, Oxford: Oxford University Press, pp. 325–337.
Bourriot, F. (1976), Recherches sur la Nature du Genos, Étude d’Histoire Sociale
Athènienne – Periodes Archaïque et Classique, Vols 1 and 2, Lille: Université
de Lille III.
Broadbent, M. (1968), Studies in Greek Genealogy, EJ Brill: Leiden.
Clinton, K. (1974), The Sacred Officials of the Eleusinian Mysteries, Philadelphia:
The American Philosophical Society.
Connelly, J. B. (2007), Portrait of a Priestess. Women and Ritual in Ancient
Greece, Princeton: Princeton University Press.
Davies, J. K. (1971), Athenian Propertied Families, 600–300 BC, Oxford: Oxford
University Press.
Foucart, P. (1900), ‘Les grands mystères d’Éleusis’, Mémoires de l’Académie des
Inscriptions et Belles-Lettres, 37, (1).
Hansen, M. H. (1991), The Athenian Democracy in the Age of Demosthenes. Structure,
Principles and Ideology, (translated Crook, J. A.), Oxford: Oxford University Press.
Harrison, A. R. W. (1968), The Law of Athens, Vol. 1, The Family and Property,
Oxford: Oxford University Press.
Hignett, C. (1952), A History of the Athenian Constitution to the End of the Fifth
Century BC, Oxford: Oxford University Press.
Kearns, E. (1989), ‘The Heroes of Attica’, Bulletin of the Institute of Classical
Studies, Supplement 57, Institute of Classical Studies: London.
Lacey, W. K. (1968), The Family in Classical Greece, Ithaca, NY: Cornell
University Press.
Lambert, S. D. and Blok, J. H. (2009), ‘The appointment of priests in attic gene’,
ZPE, 169, 95–121.
Lewis, D. (1997), ‘Who was Lysistrata?’, in Rhodes, P. J. ed., Selected Papers in
Greek and Near Eastern History, Cambridge: Cambridge University Press, pp.
Lipsius, J. H. (1905–1915), Das Attische Recht und Rechtsverfahren, Vols 1, 2 and
3, Leipzig: O. R. Reisland.
MacDowell, D. M. (1978), The Law in Classical Athens, Ithaca, NY: Cornell
University Press.
I n h e r i ta n c e , P r i e st ho o d s a n d Su c c e s sio n 105

—(1989), ‘The oikos in athenian law’, CQ, 39, 10–21.

Martha, J. (1882), Les Sacerdoces Athéniens, Paris: E. Thorin.
Meritt, B. D., Woodhead, A. G. and Stamires, G. A. (1957), ‘Greek Inscriptions’,
Hesperia, 26, 198–270, pls. 50–63.
Merker, I. L. (1986), ‘Habron the son of Lykourgos of Boutadai’, AncW, 14,
Miles, M. M. (1998), The City Eleusinion, The Athenian Agora, Vol. 31,
Princeton: American School of Classical Studies.
Oliver, J. H. (1970), ‘Marcus Aurelius. Aspects of Civic and Cultural Policy in
the East’, Hesperia, Supplement 13, Athens: ASCSA.
Osborne, M. J. and Byrne, S. G. (eds) (1994), Lexicon of Greek Personal Names,
Vol. 2, Attica, Oxford: Oxford University Press.
Parker, R. (1983), Miasma. Pollution and Purification in Early Greek Religion,
Oxford University Press: New York.
Pomeroy, S. B. (1997), Families in Classical and Hellenistic Greece. Representations
and Realities, Oxford: Oxford University Press.
Sallares, R. (1991), The Ecology of the Ancient Greek World, Ithaca, NY: Cornell
University Press.
Savage, C. A. (1907), The Athenian Family. A Sociological and Legal Study Based
Chiefly on the Works of the Attic Orators, Baltimore: The Lord Baltimore
Stengel, P. (1920), Die Griechischen Kultusaltertümer, München: C. H. Beck.
Thompson, W. E. (1970), ‘Some attic kinship terms’, Glotta, 48, 75–81.
—(1976), De Hagniae Hereditate: an Athenian Inheritance Case, Leiden: Brill.
Todd, S. C. (2007), A Commentary on Lysias, Speeches 1–11, Oxford: Oxford
University Press.
Toepffer, J. (1889), Attische Genealogie, Berlin: Weidmann.
Turner, J. A. (1983), Hiereiai: The Acquisition of Feminine Priesthoods, PhD
dissertation, University of California, Santa Barbara, CA.
Whitehead, D. (1986), The Demes of Attica, 508/7– ca. 250 BC: A Political and
Social Study, Princeton: Princeton University Press.
Wurm, M. (1972), ‘Apokeryxis, Abdicatio und Exheredatio’, Münchener Beiträge
zur Papyrusforschung und antiken Rechtsgeschichte, 60.
Ziehen, L. (1913), ‘Hiereis’, RE, 8, (2), cols.1411–1424.

Women in the Hellenistic Family: The Evidence of

Funerary Epigrams
Agnieszka Kotlińska-Toma

Funerary inscriptions are a recognized source of social and historical knowledge

on Greek society.1 In the last three decades, scholars have explored in great
detail this special kind of testimonial. Inscriptions have been used to support
a number of key areas including demographic research and mortality, as well
as social and economic aspects of everyday life and death. Moreover, scholars
have also developed a better understanding of family relationships from the
study of funerary epigrams.2 The verse inscriptions, which have more detailed
information of family members, allow us to develop an understanding of the
individual oikoi (households). In spite of the fact that there are limitations to the
genre (such as the formulaic patterns of presenting the deceased), the inscrip-
tions might be underpinned by genuine expressions of grief. In this paper, I
wish to investigate some of the well established topoi and metaphors used by
the writers to present the deceased women. The focus will be on epitymbia
(epitaphs) from the post-Classical era with a view to establishing the represen-
tation of sentiments associated with domesticity, privacy and intimacy.
Unlike men, women were only rarely presented as members of the polis in
funerary epigrams.3 Instead, texts dealing with women placed them within
the limited confines of their own households. Most scholars investigating
the representation of women in the family base their arguments on authentic
funerary epigrams (i.e. those epitymbia inscribed on stone) and omit others
that appear in literary works, which are gathered mainly in the seventh book
of Palatine Anthology. Reasons for this omission tend to be justified on the
grounds that these literary works do not represent a real phenomenon, but are
instead poetic creations. This division between the inscriptions and the other
epitymbia is mistaken. Both types of verse epigrams were poetic creations and
their means of preservation (in form of a book or on a stone) is entirely irrel-
evant. The argument that the epigrams in inscriptions were ‘real’ because they
were written for a specific person is not valid. After all, most of the texts were
written well in advance, and the name of a person was only incorporated to
the metre and carved on the stone at a later time. Moreover, some epigrams in
the Anthology were copied from stone inscriptions.4 Many epigrams by famous
poets have been written for contemporary people, as it was a custom to order
a verse epigram and to bid farewell to a friend by writing one. Therefore, to
Wom e n i n t h e H e l l e n i st ic Fa m i ly 107

understand the topoi all of these epigrams require analysis, because there was
a strong influence crossing over from more literary works to grave stele and
vice-versa.5 The existence and evolution of individual topoi is not limited to
literary creations, but needs investigation via imagery as well as texts and, in
consequence, I wish to examine the epigrams on stone in the context of the
funerary monuments with which they were associated.6 The correlation between
funerary monuments like stelae or lekythoi and epitaphs is absolutely natural
and unavoidable, as both represented the same way of depicting life and death,
and came under similar influences resulting at times in the common combi-
nation of metaphors and topoi.7 Ian Morris and Graham Oliver have suggested
that all evidence needs to be gathered and analysed in a wider context, while
taking into consideration interdisciplinary approaches to the study of death in
antiquity.8 Not every aspect of death may be considered in this short paper, but
I wish to focus on the possibilities for the interpretation of the metaphors used
in the epigrams for deceased women.
The epigrams written for dead women, and verses in which women are
presented as grieving over their family members, provide a set of similes,
metaphors and topoi, all of which can be grouped according to deceased
woman’s age and family status. One of the most interesting aspects is the status
of a daughter in the family. The burials of children have been the subject of
careful study and can be seen to be indicative of their social status.9 Much
emphasis has been placed on female infanticide and the abandonment of baby
girls in Greek society, as well as on their legal incapacity. These features are now
well known, but the emphasis on them has cast a veil over the other aspects of
a young girl’s status.

The Deaths of Young Women

In terms of epigrams, there is a large number written about the deaths of
young girls in which a very strong bond between the parents and the deceased
is emphasized. The prevalent motif in this group of epigrams is, of course,
mors immatura, which has already been analysed in detail by many excellent
scholars.10 Examining the father–daughter relationship in epigrams, one can
easily notice a very personal tone in the texts. In AP 7.643 by Crinagoras, a
child named Hymnis is presented as the one who was always bringing joy to
the whole family (ἐράσμιον αἰὲν ἄθυρμα οἰκογενὲς). In the funerary metaphors,
the common representation for little girls was in the pose of throwing their
arms around the parent’s neck. The one to popularize the motif in the funerary
epigram was Anyte11, found as early as the fourth century BC as an established
element of children’s epigrams. In AP 7.647 by Simias,12 a girl in tears embraces
her mother. Such a scene is present on some funerary stelae of the Hellenistic
period.13 It can be seen on the painted stele of Isidoros from Ibrahimieh
necropolis in Alexandria.14 Two girls raise their hands to embrace a man,
108 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

most probably their father. He touches the hand of the first one (which seems
to be older as she is a bit taller). An adolescent girl, Krito, and her mother,
Timarista, are represented on the stele CAT 2.125.15 The atmosphere of the
relief is intimate as the women are about to embrace. Only several stelae clearly
represent a mother and a daughter, for example the Rodilla and Aristylla stele.16
The non-verse epitaph, IG I 2, 1058, GV 327, under the relief depicting an older
woman sitting with a girl (standing and holding a bird in her hand) confirms
that it was the funerary monument of a deceased child.
Most of the inscribed and literary epigrams for women commemorate
unmarried adolescent girls.17 It is a visible pattern that a girl who died
unmarried was memorialized as an unfulfilled wife. The epigrams have been
examined by many scholars and constitute a model example of the inves-
tigation of this topos.18 However, the epigrams referring to young women,
married or unmarried, share the same set of metaphors and it is very often
difficult to decide whether the deceased was still a maiden or not.19 The estab-
lished topoi include the presentation of the marriage ceremony turning into a
funeral (Hymeneus turning into Hades, the marriage bed turning into a grave,
the wedding music switching to lament, the usage of wedding torches for the
funeral, and the presentation of the deceased as married to Hades or as a new
Persephone). Here, I wish to concentrate on the less recognizable topos present
in many of these epigrams – the grief of the mother, which correlates with
both wedding and funerary rituals. When a young woman dies, the grief of the
mother is represented to a greater extent in the epigrams. It was the mother
who organized the wedding and was to light the torches leading her daughter
to her future husband’s house.20 The wedding was to be the point at which the
mother–daughter relationship culminated, and both women were presented in
the epigrams as having lost this moment in that relationship and the chance to
experience the fulfilment of that relationship via the daughter’s wedding day.
The mother has to prepare the funeral instead of the wedding and the funeral
substitutes for the wedding day by becoming a celebration of regret at the loss
of the opportunity to see her daughter getting married. This can be seen most
clearly in a poem by Anyte (AP 7.649) in which a funerary monument replaces
the bedchamber with the blessed marriage-bed and a fine wedding (Ἀντί τοι
εὐλεχέος θαλάμου σεμνῶν θ’ ὑμεναίων μάτηρ). This example can be matched
by an inscriptional epitymbion on Syme:21 ‘Οὔθ’ ὑπο μητρός /Χειρῶν ἡ μελέη
νυμφίδιον θάλαμον/ἤλυθον οὐδὲ γάμου περρικαλλέος ὕμνον ἄκουσα’, ‘nor at
my mother’s hands, did I, unhappy me, go into the bridal /Chamber nor did I
hear the hymn at a beautiful wedding’, (translated by Bruss, 2005, p. 57).
From the fourth century BC onwards, the number of literary and epigraphic
epigrams referring to young wives and unmarried girls increased and was a
characteristic feature of the Hellenistic representation of the family in funerary
epigrams. This increase in topoi associated with epigrams can be correlated
with the changes in iconographic representations on the funerary monuments
(especially on loutrophoras – grave-markers for unmarried or newly wedded
Wom e n i n t h e H e l l e n i st ic Fa m i ly 109

persons). It is rather difficult to identify the mother–daughter relationship

in visual representations on artefacts, because although the older woman is
present with the bride, it does not follow that the older woman was the mother
rather than another relative of the younger woman. However, in texts, it was
usually the mother who accompanied the bride during the wedding. Wedding
scenes do appear on funerary urns for example the Hellenistic one from
Centuripe.22 The vase must have been made especially for funerary purposes
and its decoration alludes to the most common motif of marriage to death (the
deceased must have been a girl or a newly wedded woman). Its distinguishing
characteristic is the deep sadness on the faces of the bride and her companions,
and the fear in the eyes of her mother.

Childbirth and Death

The death of women in labour is found in numerous epigrams, leaving newly
wedded husbands bereaved and often also with motherless infants. The cause
of death is always clearly stated. Birth was a special female areteia comparable
to that of the warrior on a battle field, both woman and man could die heroi-
cally, and it is heroism via childbirth that we find represented on the grave
monuments and in epigrams dedicated to women.23 In the Hellenistic period,
the coexistence of life and death in a single act or moment was a focus for
intellectual investigation. The young woman was often presented as caught in
between these two violent forces, seen most clearly in the realism (or even the
naturalism) of mature Hellenism dating to the fourth–second centuries BC. As
a result, it is not uncommon to find scenes of women in labour on the funerary
stelae of this period. A very good example of the naturalistic representation
of woman dying in childbirth is a Hellenistic painted stele from Ibrahimieh
necropolis in Alexandria.24 It has a close relationship to the sculptured stelae of
the Plangon type.25 A rather later example of this type, the stele of Hediste from
Demetrias-Pagasai (c.200 BC),26 includes an epigram:

λυπρὸν ἐφ’ Ἡδίστηι Μοῖραι τότε νῆμα ἀπ’ ἀτράκτων

κλῶσαν, ὅτε ὠδῖνος νύμφη ἀπηντίασεν·
σχετλίη. οὐ γὰρ ἔμελλε τὸ νήπιον ἀνκαλιεῖσθαι,
μαστῶι τε ἀρδεύσειν χεῖλος ἑοῖο βρέφους.
ἓν γὰρ ἐσεῖδε φάος καὶ ἀπήγαγεν εἰς ἕνα τύμβον
τοὺς δισσούς, ἀκρίτως τοῖσδε μολοῦσα, Τύχη

A painful thread for Hediste did the Fates weave from their
spindles when,
as a young wife she came to the throes of childbirth.
Ah wretched one! For it was not fated that she should cradle
the infant in
110 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

her arms, nor moisten the lips of her new-born child at her breast.
One light looks upon both and Fortune has brought both to a
single tomb,
making no distinction when she came upon them
(translated by Pollitt 1986, p. 4)

In these scenes and the epigram, a choice has been made in terms of repre-
sentation – because there were other ways of commemorating those who died
in labour. However, choices were made and an emphasis was placed on the
deceased’s loss of the experience of rearing the child who, in the final epigram
example, also dies to unify mother and infant in ‘a single tomb’ – just as in
pregnancy, the unborn child was united in the body of its mother.
It is a common feature of Hellenistic aesthetics to present strange, unusual
or even incredible things with a focus on the development of the paradoxical.
This is seen most clearly in the giving of life by mothers to their children, via
birth, and subsequently in having their own life taken away. A peculiar epigram
by Antipater of Thessalonica exemplifies this phenomenon.27 The paradox of
the situation is a clear focus at two separate points in the epigram: once when
the dying mother delivers living offspring (μητέρος ἐκ νεκρῆς ζωὸς γόνος),
and subsequently when the same god takes life from the mother and gives it
to the child (εἷς ἄρα δαίμων/τῆς μὲν ἀπὸ ζωὴν εἵλετο, τοῖς δ’ ἔπορεν). This
phenomenon can also be found in epigraphic epigrams referring to mothers
dying in childbirth.28

Infant Twins
Epigrams also represent another phenomenon, the survival of a single newborn
twin while the other dies alongside her/his mother. In the literary epigram, the
subject was first undertaken by Heraclitus.29 One of the children was left with
the father to support him in his old age, while the second was kept in death
by the mother as a memento of the husband (δισσὰ δ’ ὁμοῦ τίκτουσα τὸ μὲν
λίπον ἀνδρὶ ποδηγὸν/γήρως, ἓν δ’ ἀπάγω μναμόσυνον πόσιος). This was to
become a theme that was expanded into a more elaborate poem by Antipater
of Sidon.30 The differentiation of the child who died with his mother from the
child who survived appears in the visual representation of the dead. If the
mother is seated and another woman (most probably a servant or wet-nurse)
or a father is holding the child, we may assume that the baby survived the death
of his/her mother in childbirth. In contrast, if the child also died, s/he was
positioned in the arms of his/her mother, which can be seen on many stelae.
Such an interpretation is confirmed by the famous stele of Ampharete where
the dead grandmother keeps the dead grandchild.31 Antipater’s epigram can
also be compared to a stele from the Louvre Museum which portrays a mother
keeping one child while a nurse takes the other one.32 The scene demonstrates
Wom e n i n t h e H e l l e n i st ic Fa m i ly 111

a situation in which one of the twins dies with the mother during labour. Both
the Antipater’s epigram and the stele from the Louvre present the same paradox
of twins split between the living and dead members of the oikos.

The Husband, His Dead Wife and Her Father

It would be wrong to assume that the mother, who died during or soon after
labour, was always commemorated by her husband. In fact, very often it is
parents who set up memorials to these young mothers. Young mothers are
mostly memorialized by their parents,33 with a particular prominence of her
father grieving the loss of his daughter. Most of these epigrams mention at least
one of the woman’s parents, usually the father, but there are some cases when
both parents are named. It might be an indication that the parents participated
in the costs or paid for the monument and the epigram. Nevertheless, it is also
a clear indication of the strong affiliation of daughters to their biological kin.
The emotional bond between the newly wedded woman and her father’s oikos
was not yet weakened by a long separation, like in AP 7.166. In fact, the young
woman was not yet integrated in her husband’s family. One of the inscriptional
epigrams presents the woman as ‘οὔτε γυνὴ ( . . . ) οὔτε τι κούρη’ which means
that she was not yet considered as a real woman and wife because she had not
given birth.34 A girl in between her father and her husband is presented in the
epigrams by Julian. AP 7.600 and 601 show a 16-year-old wife mourned by both
(σοὶ γενέτης, σοὶ πικρὰ πόσις κατὰ δάκρυα λείβει).

The Deaths of Siblings

Only a few epigrams let us enter the world of sisterly relations, but they are
revealing.35 Epigram AP 7.662 presents a 7-year-old girl who died following
the death of her brother. The cause of death according to the author, Leonidas
of Tarentum, was the bereavement of a 20-month-old sibling: ποθέουσα τὸν
εἰκοσάμηνον ἀδελφὸν. In the epigram by Callimachus,36 Basilo committed
suicide from grief on the day her brother died. Posidippus, in his partially
preserved poem on Myrtis,37 presents two older brothers burying their young
sister. The close relationship between siblings is suggested on the well-known
stele of Mnesagora and Nikochares on which a girl hands in a toy or a bird-pet to
her baby brother.38 The picture is complemented by epigram IG II/III2 12147, GV
95 from which we know that the monument was erected by the grieving parents.
The stele of Chairestrate and Lysandros is of the same iconographic type.39 The
affection between brother and sister is more visible when we compare the scenes
with the funerary reliefs for mothers with children.40 In fact the older sisters
were represented in exactly the same format as mothers, perhaps indicative of
their role as older women and ‘substitute’ mothers in the oikos.
112 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

The Husband and his Dead Wife

A good wife is commemorated and remembered via the presentation of her
domestic virtues. Conventional inscriptions would normally consist of arete and
sophrosyne, and a traditional statement that she was better than other women41
or more famous because of her virtues. From the end of the Archaic period,
women’s funerary epigrams and stelae became ever more elaborate. A whole
range of female virtues were shown by using some commonly recognized signs
and symbols. On existing stelae, sirens usually symbolize the underworld and
loutrophora signify marriage.42 These signs were universally recognized to such
an extent that exchanging them with others – more suitable for a man, became
a riddle in two epigrams by Antipater of Sidon,43 where reins symbolize being in
charge of the household (one of the most important duties of wives), a muzzle
symbolizes the virtue of not being talkative, animals such as a cock and an owl
symbolize the stay for a night, while wool-making and a bitch represent taking
care of children.44 The muzzle and reins would be a typical sign for a horseman
and a cock is very often represented on funerary reliefs for young men. Such
an elaborate set of metaphors is in contradiction to the usual or expected ones,
which are also mentioned in epigram AP 7.424: such as a spindle and a loom.
The funerary iconography often portrays seated women with kalanthoi or
epinetra, which symbolize their diligence.45 The less conventional approach was
to show a wife as possessing two personalities represented by two goddesses.
To her husband, she acted like Aphrodite while to strangers she acted like
Athena Pallas.46 Yet greater precision can be found in AP 7.337, in which the
anonymous author enumerates a wife’s virtues which were also associated as
virtues of famous men: εὐγένεια, ἀρετή, ἤθεα, σωφροσύνη, (nobility of birth,
moral character, virtue and prudence).
Mentioning physical beauty in a very few banal words or omitting it
altogether is another characteristic of funerary epigrams.47 This description
of the decent wife was both obvious and irrelevant. Her inner beauty would
normally be more important than physical beauty. Epigram AP 7.695 refers to
the relief carved on a funerary monument presenting the portrait of Kassia.
Even though the passer-by can see her beauty clearly, the poem points out
her virtues as more admirable than her appearance. A popular motif was the
juxtaposition of gold and other ornaments and garments with the inner beauty
of a wife.48 The epigrams were of special interest to the scholars who compared
them with the most popular way of presenting women on funerary monuments
– a seated figure which either looks into the jewellery box (or keeps a piece of
jewellery) or receives it from her servant. These scenes have usually been inter-
preted as representations of well-to-do women. The most famous stele of this
iconographic type is the funerary monument for Hegeso.49 The scene is usually
read as the adorning of the lady of the house.50 The jewels are seen as symbols
of the woman’s personal wealth, or are suggested as representations of her
dowry. In my opinion, the juxtaposition of inner virtues with the fact that the
Wom e n i n t h e H e l l e n i st ic Fa m i ly 113

women seem to be focused on the ornaments is not surprising, especially if we

read the scenes as scenes of farewell to earthly life symbolized by the removal
of the ornaments – and a parting glance to the jewellery – rather than scenes
of adornment. My interpretation can be supported by other representations of
this type, where the accompanying woman (a servant or a daughter) takes the
jewellery box from the lady, or even takes the box away.51 If this is the case, the
representation of these artefacts does not contradict the well-established topoi
in Greek poetry.
The topoi of arete and sophrosyne associated with wives are in stark
contrast to the epigrams commemorating lovers and hetaerae. In the case of
the latter, both their beauty and entertainment skills are emphasized.52 Four
literary polished epigrams for deceased wives are of great significance here.
An interesting example of praising women was made via a comparison with
a mythological heroine. Alcestis, who devoted her life to save that of her
husband, could be a better mythological example of the ‘good’ wife. Epigram
AP 7.691, which seems to have been written for a specific person, presents a
deceased named Callicratia who is praising herself as the new Alcestis, who
died for her husband, loving him more than her own life and that of her
children. Another reference to the myth can be traced in an epigram GV1392
from Thrace in which a husband swears not to take another woman to his bed.
Another mythological paradigm for a good wife can be located in Penelope,
as both her faithfulness and diligence were the main virtues of wives.53 This
phenomenon is also found in inscriptional epigrams. The deceased appears to
be as virtuous as both Alcestis and Penelope in inscription IG XII 7.494 from
The most common way of presenting a married couple on funerary
monuments is via the so-called dexiosis where the husband and wife hold hands
in a gesture of greeting or farewell.54 This can be also traced to verse epitaphs.
One shows a woman whose husband is travelling and whose hand she therefore
cannot hold at the time of death.55 The epigram also presents another very
important aspect of family relationships: peaceful death takes place only after a
farewell to the family.56
The expressions of attachment in marriage can be found in epigrams. A wife
having died soon after her husband because she was unable to live without him
is represented as sharing his tomb after her death.57 This presents unification in
death, which is another common topos for a married couple. The topos of simul-
taneous death could easily be interpreted as a pure fiction had it not been for
many examples, especially of very old couples. In epigram AP 7.475, a girl dies
three months after losing her newly wedded husband, and the cause of death,
according to the epigrammatist, is the sorrow of the soul (οὐλομένῃ ψυχῆς
δύσφρονι τηκεδόνι).58 The epigram is also interesting because the young couple
are not commemorated by the typical metaphors associated with a wedding-
funeral formula (although the husband is called nymphios). Instead, it presents
a picture of a lamenting woman following the beloved one to the gate of Hades.
114 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

The Bereaved Mother

Another set of topoi is reserved for mothers who lost their children. Children
make a woman fulfilled but in having them, her social status changes – she
becomes a personification of the family, and is central to its constitution.
Without her, the family would not have existed in the first place. The funeral
epigrams and monuments often pay tribute to this fact.59 It is deeply rooted in
the tradition of naming the father in Greek epigrams – not only by the form
of patronimics – but also via the family’s genealogy that was disrupted by the
death of the child. However, there are epigrams written about dead children, in
which the name of the father is omitted, and instead mourning mothers provide
the parental focus. In a way, they use these bereaved mothers as an element
to illustrate a universal truth: no one can love and despair to a greater degree
than a mother.60 The death of a child is an inversion of natural law by which
a mother should precede her child in death.61 However, there is yet another
cultural assumption hidden in these texts. Mothers in Greece prepared corpses
for burial with a preference given to biological mothers in this role.62 In this
context, the body of the child belonged to its mother. As she is the one nursing
and caring for infant hygiene, she is also the one predestined to perform the
rites of preparing the corpse for burial. Women and girls were also by tradition
responsible for sanitation and caring for the sick, and were familiar with the
procedure of cleaning the body. In many ways, there was an expectation that
children would predecease their mother and be mourned by her. An obser-
vation confirmed in cenotaph epigrams express the misfortune of the dead
predeceasing their parents who are not mourned by a mother wailing over their
body.63 The ritual performed by the mother during the funeral64 is also visible in
epigram AP 7.468 by Meleager in which a mother dresses her son as an offering
to Hades, and is presented as the principle mourner and the person who breast-
fed and loved her dead son.
Songs sung in mourning appear in epigrams that reveal the relationship of
living women to the dead. In epigram AP 7.467 by Antipater of Sydon, a mother’s
song of lament is used as an element of the poem: ‘Τοῦτό τοι, Ἀρτεμίδωρε, τεῷ
ἐπὶ σάματι μάτηρ/ἴαχε δωδεκέτη σὸν γοόωσα μόρον’ (‘This is how for you,
Artemidoros, your mother lamented over thy grave, grieving for your death at 12
years of age’). After these introductory verses, Antipater paraphrases a ritual song
called goos which was a spontaneous lament commemorating the dead performed
by female relatives. In goos, kinswomen enumerated the virtues of the deceased,
evoking them from the dead. As a lament stressing the personal bereavement they
felt, it in fact consisted of a highly emotional wail. The nature of goos is presented
in another poem by Antipater via the repetition in three consecutive verses of
the word ὤλεθ (gone, perished). Alexiou also noticed a literary echo of goos, in
epigram AP 7.486 of Anyte in which Kleino invokes her daughter from the dead.65
Especially interesting is the anonymous epigram AP 7.334 from Cyzicus, in
which the dead son appreciates the mother’s efforts in raising him more than
Wom e n i n t h e H e l l e n i st ic Fa m i ly 115

his father’s. As it seems to be based directly on a text from an inscription, it

would appear that the epigram was ordered, or maybe even written (it presents
detailed information on the boy’s life), by the mother or a close relative, and
therefore the mother’s loss is stressed here. The ultimate way for a mother to
experience sorrow is death in mourning.66

Verses presenting grandmothers are rather rare. First of all, we should notice
that grandmother does not necessarily mean an old woman since, in getting
married at 15, a Greek wife could have had grandchildren at the age of 30. In
the Greek Anthology we can find a few epigrams on women who were remem-
bered for giving birth to the greatest number of children. Epigram AP 7.484
presents a mother of 10 children who was also a grandmother. Others, such as
seen in epigrams AP 7.224 and 743, are found to have given birth to 29 children.
Long-living woman are represented and recognized for their role in the household
or in raising children. That is why epigram AP 7.726 by Leonidas of Tarentum,
praises an 80-year-old woman as keeping poverty at bay. Another older woman
named Nikarete67 is remembered as diligent and frugal. Being a grandmother was
a sign of the fulfilment in life, a typical example being IG II/III2 3.2 11 998. Two
recently discovered epigrams by Posidippus are devoted to older women. The first,
AB 47 concerning Onasagoratis, not only stresses that she lived 100 years, but also
indicates that she witnessed the birth of three generations of descendants. The
second, AB 59, was devoted to Menestrate, an 80-year-old woman, seen to have
been blessed by the gods with two generations of children. In both cases, being
a fortunate old woman meant witnessing the fecundity of children and grand-
children in the family. It is very hard to find matching examples of these long
lived women in funerary stelae, perhaps because the representations of women
on funerary monuments were usually ageless. The already mentioned stele of
Ampharete showing a grandmother who holds her dead grandson is one of the
most heartbreaking monuments of antiquity. Nevertheless, the portrait does not
differentiate her from those who were mothers in terms of composition with
others found in the Classical period. It would appear that the stele was cut and
then adapted to include an inscription referring to a grandmother and her grand-
child. Women in old age become more visible on funerary stelae in the Hellenistic
period. Even so, it is very often difficult to distinguish the representation of an
old nurse from a grandmother. In such cases, only the epitymbion is a reliable
source of information. An interesting inscribed stele is CAT 1.934, in which the
epigram reveals that the deceased woman, Chairestrate, was an august priestess,
beloved wife and grandmother.68 The relief shows Chairestrate as a seated and
older woman with a temple key in one hand. In front of her stands a young girl
(probably her granddaughter) with a large tympanon which might suggest the
inheritance of the post by the female members of the family.
116 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

The Hellenistic period was a time of profound change in the social, economical
and political spheres. Depopulation and the fragmentation of family ties as well
as increasingly common elementary education among women and political activ-
ities of Hellenistic queens caused a noticeable change in mentality and perception
of female role models. In addition, interests in individuals rose in Hellenistic art,
literature and philosophy, therefore rendering funerary epigrams more personal
and more detailed in terms of information regarding the deceased. In comparison
with the brief Classical Attic funerary inscriptions which represent feminine
stereotypes, the Hellenistic epigrams are noted for their variety and elaborateness
of topoi. Very often, epitaphs present personal emotions and individual and
subjective perspectives on death. Funerary epigrams concerning the dead female
members of a family create a picture of the deceased in the form of the paragons
of womanhood, praising them respectively as wives, mothers, grandmothers or
daughters and sisters. The inner rhetoric of the epitaphs, specific formulae and
the well-established metaphors build the concept of a family based on female
virtues, skills and care, all of which are in contradiction to the common view
underestimating the woman’s role as a creator and keeper of the oikos. The overall
pattern found in the examples presents a woman fulfilling her obligation as wife
and mother, or in the case of girls waiting to be engaged in family responsibilities.
These funerary epigrams provide us with a glimpse into the daily affairs of the
women inside the oikos, but most of all, demonstrate the ethos of women inside
the Greek family. The problems with the correlation of the source material and
reality are familiar to every scholar concerned with family studies. It is difficult
to establish weather the topoi mirror everyday life or represent wishful thinking.
Above all, the epigrams must praise the deceased and topoi should construct
idealizing norms for domestic life. Finally, it is necessary to return to the issue
which was raised in the first section of this article. No distinction in using topoi
exists between inscriptional and literary epigrams as both are poetic creations
of the same genre. Of course, in many cases the manuscript tradition preserved
much more elaborate epigrams than texts found on stones, yet one should
remember that a collection of poems like Palatine Anthology was a selection
par excellence. Most of the inscriptional epitaphs were written in advance and
adjusted to the particular person after the purchase of a funerary stone. We
should nevertheless remember that family members made conscious decisions to
choose the proper epigram and funerary monument as a tribute to the deceased.

  1 See the introduction to Peek, 1960; Sourvinou-Inwood, 1996, pp. 140–279.
  2 Excellent study on depicting family members in the inscribed epigram is Tsagalis, 2008, pp.
183–198, on women see Strömberg, 2008; Burton, 2003.
Wom e n i n t h e H e l l e n i st ic Fa m i ly 117

  3 Comp. Breuer, 1995, pp. 30–100. On the high ratio of family oriented epitaphs for woman see
Tsagalis, 2008, p. 184.
  4 That is indicated by a phrase ex lithou or giving the place of origin.
  5 Excellent study of the topic in Bettenworth, 2007.
  6 See discussion on the genre in Bruss, 2005, pp. 2–18; Kazazis, 1996–1997, 9–17; and Dignas,
2003, pp. 177–178 with further bibliography. Useful in examining the artefacts: Sojc, 2005.
  7 Although not many existing artefacts and epigrams can be collated, comp. Clair, pp. 17–18,
Day, 1989, 21.
  8 Oliver, 2000, pp. 4–6; Morris, 1992, p. 202.
  9 Houby-Nielsen, 2000, with further bibliography. Recent study on the Greek funerary art for
children Schlegelmilch, 2009, pp. 27–68.
10 On the death of aoroi: Garland, 2001, pp. 76–86; Griessmair, 1966; Vérilhac, 1982; Kazazis,
1996–1997; all with further bibliography.
11 AP 7.646 in which Erato is embracing her father.
12 The authorship is uncertain see Gow and Page, 1965, p. 516.
13 A parallel representation is a son stretching his arms to embrace his mother (like on the stele
Athens NM 767).
14 Funerary slab with a soldier and two girls [Greek; from the Soldiers’ Tomb, Ibrahimieh necropolis,
Alexandria, excavated 1884] (04.17.4). In (2000), Heilbrunn Timeline of Art History, New York:
The Metropolitan Museum of Art,
15 Rhodes Archaeological Museum 13638.
16 Athens NM 766, CAT 2.051, Clair, n. 27.
17 Humphreys, 1980, p. 93.
18 Comp. Lattimore, 1942, pp. 192–195, on the special topic of death before marriage in Greek
tragedy and other literary genres see Rehm, 1994; Seaford, 1987 and Szepessy, 1972.
19 Anyte (7. 486 and 649), Erinna (7. 710 and712), Antipater of Sidon (7.711), Meleager (7.182
and 476), Mnasalkas (7. 488), Parmenion (7.183), Ps-Sappho (7.489), Theodoridas (7. 528),
Philippus (7.186), Antipater of Thessalonice (7.185), Antonius Thallus (7.188) Julian, Prefect
of Egypt (7.600), Paulus Silentiarius (7.604).
20 See Phoen. 344–346.
21 Athens NM 766, CAT 2.051, Clair, n .27. CV 947. The whole poem and commentary see Bruss, 2005,
p. 56f; Schlegelmilch, 2009, p. 75f, (who thinks that Syme was a child) with the picture of Syme’s stele.
22 Illustration available on the Museum’s internet page:
5278&collID=13&dd1=13. Accession number 53.11.5.
23 See Vedder, 1988.
24 Funerary stele with a woman in childbirth [Greek; from the Soldiers’ Tomb, Ibrahimieh
necropolis, Alexandria, excavated 1884] (04.17.1). In (2000), Heilbrunn Timeline of Art
History, New York: The Metropolitan Museum of Art, –
25 320 BC, Athens NM 749. This, and many other examples of funerary monuments representing
women dying in childbirth see Demand, 1994, pp. 156–165.
26 Thess. Mnem., 97, pp. 1–5, discussed by Pollitt, 1986, pp. 4–5, 194.
27 AP 7.168, evidently literary.
28 E.g. CEG 604, ‘εἰς φῶς παῖδ› ἀνάγουσα βίου φάος ἤν[υσας αὐτή]’, ‘bringing a child to light
(Cleagora) lost the light of her own life’.
29 AP 7.465.
30 AP 7.464.
31 Kerameikos Museum P695, CAT 1.660, Clair n. 23.
32 CAT 2.810.
33 E.g. GV 1507=SEG VII 369.
34 IG IX, 2 638. On transmission to womanhood through childbirth see Dillon, 2002, p. 215.
35 On the young siblings in the funerary art Lewis, 2002, p. 182, (with examples of sibling burials
in ancient Greece).
36 AP 7.517.
118 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

37 AB 54.
38 Athens, National Museum 3845, CAT 1.610, Clair n. 22.
39 CAT 1. 575.
40 CAT 1.662 and 1.867.
41 E.g. CEG 493, CEG 546, analysed by Tsagalis, 2008, p. 178f.
42 On the types of sirens and other symbolic representations see Kurtz and Boardman, 1971, p.
43 AP 7.424 and 425.
44 Similar riddle-epigram preserved on the stele of Menophila, see Gutzwiller, 1998, p. 265f.
45 CAT 1.309. It can be also any other object used in weaving, like shuttle, wool on the spool,
spindle on the stele from Cyrene (comp.: Breuer, 1995, pp. 90–94).
46 AP 7.599.
47 On the rate of ‘beautiful’ in the Greek epigram see Breuer, 1995, Tables II and IV, s.v. schönheit.
48 E.g. οὐχὶ πέπλους, οὐ χρυσὸν ἐθαύμασεν ἐμ βίωι ἥδε IG II/ III 23,2, 11162 = GV 1810.
49 Athens, National Museum 3624, CAT 2.150.
50 For the discussion see Clairmont (CAT, Vol. 2, pp. 96–97), who rejects the theory of adorning
Hegeso as a bride-to-be of Hades. Both jewels and a servant are the markers of Hegeso’s status
and wealth. See Leader, 1997, p. 689.
51 As on the stele of Kalliarista in the Museum of Rhodes, CAT 2.335b, Clair n. 32, epigram GV
893, and CAT 2.431 and 2.750.
52 E.g. the series of literary epigrams from AP 7.217 to 222.
53 AP 7.557 by Cyrus the Poet.
54 On dexiosis in the funerary art see Pemberton, 1989; Davies, 1985 with the further bibliog-
raphy. Naturally all the members of the family can be presented shaking hands.
55 AP 7.735 by Damagetus.
56 Compare the lament of Andromache in Iliad 743.4 and some remarks on this passage in Holst-
Warhaft, 1992, pp. 112–113.
57 AP 7.378.
58 Detailed analysis see Gow and Page, 1965, pp. 275–276 and Bruss, 2005, p. 77.
59 The oldest funerary stele of woman and child has been found in Anavysos (Athens, NM4472)
and comes from the sixth century (Jeffery, 1962, 145–146, n. 2; Richter, 1961, pp. 42–43, figs.
151–53, n. 59).
60 AP 7.467.
61 AP 7.261.
62 Havelock, 1981, p. 112. Shapiro, 1991, p. 635, points out that women were better suited to deal
with the pollution of death as they have been experiencing the pollution of giving birth. On
pollution caused by death see Parker, 1983, pp. 33–35.
63 AP 7.404.
64 On the role of the women in the funeral rites see Garland, 2001, pp. 21–37.
65 Alexiou, 2002, p. 109. On goos see also Reiner, 1938, pp. 8–61.
66 AP 7.608 and 644.
67 IG II/III2 12254, GV 328: ἐργ[ά]τις ὀσα γυνὴ φειδωλός τε ἐνθάδε κεῖμαι. Νικαρέτη
68 Pireus Museum 3626, Clair n. 26; IG II/III2 6288, GV 421.

Alexiou M. (2002), The Ritual Lament in Greek Tradition, (2nd edn, revised by
Yatromanolakis, D. and Roilos, P.), Lanham: Rowman and Littlefield.
Bettenworth A. (2007), ‘The mutual influence of inscribed and literary epigram’,
in Bing, P. and Bruss, J. S. (eds), Brill’s Companion to Hellenistic Epigram
Down to Philipp, Leiden: Brill, pp. 69–93.
Wom e n i n t h e H e l l e n i st ic Fa m i ly 119

Blundell, S. and Williamson, M. (1998), The Sacred and the Feminine in Ancient
Greece, London, New York: Routledge.
Breuer, C. (1995), Reliefs un Epigramme Giechischer Privatgrabmäler, Köln: Böhlau.
Bruss, J. S. (2005), Hidden Presences. Monuments, Gravesites, and Corpses in
Greek Funerary Epigram, Leuven, Paris, Dudley MA: Peeters.
Burton, D. (2003), ‘Public memorials, private virtues: Women on classical
Athenian grave monuments’, Mortality, 8, 20–35.
Davies, G. (1985), ‘The significance of the handshake motif in classical funerary
art’, AJA, 89, 627–640.
Day, J. W. (1989), ‘Rituals in stone: Early Greek grave epigrams and monuments’,
JHS, 109, 16–28.
Demand, N. H. (1994), Birth, Death, and Motherhood in Classical Greece,
Baltimore: Johns Hopkins University Press.
Dignas, B. (2003), ‘Posidippus and the mysteries: Epitymbia read by the ancient
historian’, in Acosta-Hughes, B.; Kosmetatou, E. and Baumbach, M. (eds), Labored
in Papyrus Leaves. Perspectives on an Epigram Collection Attributed to Posidippus
(P. Mil. Vogl. VIII 309), Cambridge: Harvard University Press, pp. 177–186.
Dillon, M. (2002), ‘Girls and Women in Classical Greek Religion’, London, New
York: Routledge.
Garland, R. (2001), The Greek Way of Death (2nd edn), Ithaca New York:
Cornell University Press.
Gow, A. S. F. and Page, D. L. (1965), The Greek Anthology. Hellenistic Epigrams,
Vol. 2 Commentary and Indexes, Cambridge: Cambridge University Press.
Griessmair, E. (1966), Das Motiv der Mors Immatura in den Griechischen
Metrischen Grabinschriften, Innsbruck: Wagner.
Gutzwiller, K. J. (1998), Poetic Garlands: Hellenistic Epigrams in Context,
Berkeley and Los Angeles: University of California Press.
Havelock, C. M. (1981), ‘Mourners on Greek vases: Remarks on the social
history of women’, in Hyatt, S. L. ed., The Greek Vase, Latham New York, pp.
Holst-Warhaft, G. (1992), Dangerous Voices: Women’s Laments and Greek
Literature, London, New York: Routledge.
Houby-Nielsen, S. (2000), ‘Child burials in ancient Athens’, in Sofaer Derevenski, J.
ed., Children and Material Culture, London, New York, Routledge, pp. 151–166.
Humphreys, S. C. (1980), ‘Family tombs and tomb cult in ancient Athens:
Tradition or traditionalism?’, JHS, 100, 96–126.
Jeffery, L. H. (1962), ‘The inscribed gravestones of archaic Attica’, BSA, 57,
Kazazis, N. J. (1996–1997), ‘Mors immatura as mature art: from the subliterary
to the literary ancient Greek funerary epigram’, EEThess, 6, 9–30.
Kurtz, D. and Boardman, J. (1971), Greek Burial Customs, London: Thames &
Lattimore, R. (1942), Themes in Greek and Latin Epitaphs, Urbana: University
of Illinois Press.
120 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

Leader, R. E. (1997), ‘In death not divided: Gender, family, and state on classical
Athenian grave stelae’, AJA, 101, 683–699.
Lewis, S. (2002), The Athenian Woman: An Iconographic Handbook, London,
New York: Routledge.
Morris, I. (1992), Death-Ritual and Social Structure in Classical Antiquity,
Cambridge: Cambridge University Press.
Oliver, G. J. (2000), The Epigraphy of Death: Studies in the History and Society of
Greece and Rome, Liverpool: Liverpool University Press.
Parker, R. (1983), Miasma: Pollution and Purification in Early Greek Religion,
Oxford: Oxford University Press.
Peek, W. (1960), Griechische Grabgedichte, Berlin: Akademie-Verlag.
Pemberton, E. G. (1989), ‘The dexiosis on Attic gravestones’, MediArch, 2, 45–50.
Pollitt, J. J. (1986), Art in the Hellenistic Age, Cambridge: Cambridge University
Rehm, R. (1994), Marriage to Death: The Conflation of Wedding and Funeral
Rituals in Greek Tragedy, Princeton, NJ: Princeton University Press.
Reiner, E. (1938), Die Rituelle Totenklage der Griechen, Stuttgart-Berlin: Kohlhammer.
Richter, G. M. A. (1961), Archaic Gravestones of Attica, Cambridge, Mass.:
Harvard University Press.
Schlegelmilch, S. (2009), Bürger, Gott und Götterschützling: Kinderbilder der
Hellenistischen Kunst, Berlin, New York: Walter de Gruyter.
Seaford, R. (1987), ‘The tragic wedding’, JHS, 107, 106–130.
Shapiro, H. A. (1991), ‘The iconography of mourning in Athenian art’, AJA, 95,
Sojc, N. (2005), Trauer auf Attischen Grabreliefs, Berlin: Reimer.
Sourvinou-Inwood, C. (1996), ‘Reading’ Greek Death: To the End of the Classical
Period, Oxford: Oxford University Press.
Strömberg, A. (2003), ‘Private in life-public in death. The presence of women
on Attic classical funerary monuments’, in Larsson L. and Strömberg A. (eds),
Gender, Cult, and Culture in the Ancient World from Mycenae to Byzantium:
Proceedings of the Second Nordic Symposium on Gender and Women’s History
in Antiquity, Helsinki, 20–22 October 2000, Sävedalen: Paul Åströms förlag,
pp. 28–37.
Szepessy, T. (1972), ‘The story of the girl who died on the day of her wedding’,
AAntHung, 20, 341–357.
Tsagalis, C. C. (2008), Inscribing Sorrow: Fourth-Century Attic Funerary
Epigrams, Berlin, New York: Walter de Gruyter.
Vedder, U. (1988), ‘Frauentod-Kriegertod im spiegel der attischen Grabkunst
des vierten Jahrhunderts v. Chr’., AM, 103, 161–191.
Vérilhac, A. M. (1982), Paides Aoroi. Poésie Funéraire. Texte Critique et
Commentaire des Epigrammes, Athènes, Vols 1 and 2, Athenai: Grapheion
Dimosieumaton tes Akademias Athenon.
Vermeule, E. (1981), Aspects of Death in Early Greek Art and Poetry, Berkeley-Los
Angeles-London: University of California Press.

Family Relationships in Late Bronze Age, Iron

Age and Early Roman Veneto (Italy): Preliminary
Considerations on the Basis of Osteological Analysis
and Epigraphy
Elisa Perego

The aim of this paper is to investigate the structure of the family in Late
Bronze Age, Iron Age and Early Roman Veneto (Italy, c.1050 BC – AD 25)
through the analysis of selected features of the local funerary record, namely
epitaphs and the treatment reserved to cremated dead bodies. In the attempt
to reconstruct some aspects of Venetic kinship structure, my primary focus
is on the widespread custom of placing the bones belonging to different
cremated individuals in a single cinerary vessel (bone mingling). This
practice, which spread in Veneto from at least the Late Bronze Age to the
Roman period, and is also known in other Italian regions, is interpreted as
the extreme attempt to reunite in death individuals who must have been
tied together by special bonds also in life. In this paper analysis of a sample
of more than 400 urns from different Venetic localities is used to shed light
on the frequency and modalities of bone mingling practices and to identify
the social criteria which motivated the burial of a person alone or in an urn
with other people. By analysing the overall pattern of sex and age distri-
bution of the sample, I suggest that bone mingling practices were adopted
for the burial of close relatives, probably in most cases members of nuclear
families. The scrutiny of late Venetic and early Latin epitaphs inscribed
on cinerary urns dating between the third century BC and the early first
century AD further reinforces this hypothesis. However, the existence of
possible alternatives in the disposal of the dead might be suggested by a
number of epitaphs.
122 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

Veneto in the First Millennium BC
Today, Veneto is an administrative region located in north-eastern Italy between
the Alps, Lake Garda, the Adriatic Sea and the Po River (Figure 9.1). According
to mainstream Italian scholarship, during the first millennium BC this area was
inhabited by an Indo-European population, the Veneti of the Greco-Roman
tradition.1 An Indo-European language slightly similar to Latin – the so-called
‘Venetic’ – is attested on several hundred inscriptions spread across the region.2
The observed variability in settlement pattern, population density and material
culture may have been motivated by landscape diversity, interregional contact,
external cultural influences and independent development of central versus
peripheral areas. Moreover, the presence of ‘ethnic minorities’ (e.g. Celts) in
some areas of Veneto from at least the fifth century BC is strongly suggested
by both archaeological and epigraphic evidence.3 The main Iron Age Venetic
settlements were Este and Padua in central Veneto, Altino on the Adriatic Sea,
Oderzo and Concordia in eastern Veneto, and Treviso and Montebelluna in the
Piave Valley.4 Oppeano and Gazzo Veronese, which flourished in the Verona
countryside between the tenth and the fifth centuries BC, declined after the area
was presumably taken by the Etruscans and then by the Celts.5
The transition between the Late Bronze Age and the Early Iron Age
(c.1050–900 BC) in this region was characterized by important social realign-
ments accompanied by a partial shift in the settlement pattern. Over the
following two centuries, a progressive increase in social complexity and stratifi-
cation is found in both funerary and settlement evidence.6 The transformation
of Venetic settlements from loose aggregations of huts to more structured proto-
urban centres started at the latest in the eighth century BC. At Este and Padua,
the appearance of a few graves of exceptional wealth seems to demonstrate the
emergence of an elite able to acquire exotic and luxurious goods not available
to others. A dense network of interregional contacts with Bologna, central Italy,
Lombardy, Continental Europe, and – later – the Adriatic emporia of Spina
and Adria was established over the entire Iron Age.7 The earliest evidence of
formal cult activity in sanctuaries dates between the late eighth and the sixth
centuries BC.8 The main Venetic centres possibly developed an urban layout
as early as the sixth century BC. Padua may have developed as a prominent
centre in the region from the fifth century BC.9 An increasing intervention of
the Romans in local affairs dates at least to the second century BC. The process
of Romanization seems to have been mainly gradual and peaceful. The Veneti
are presented as allies of Rome by Greco-Roman historical sources, and to date
there is no clear archaeological evidence of extensive warfare and destruction in
the region.10However, major social and political developments which ultimately
led to the disappearance of the local language, independence and lifestyle took
place in the last two centuries of the first millennium BC.11 The ius Latii and
Fa m i ly R e l at io n sh i p s 123

Roman citizenship were granted to northern Italy, including Veneto, in 89 and

49 BC, respectively.12

Fig. 9.1. Map of Veneto with the main sites mentioned in the paper. 1: Valle di Cadore;
2: Belluno; 3: Concordia; 4: Oderzo; 5: Covolo; 6: Montebelluna; 7: Ponzano Veneto; 8;
Treviso; 9: Altino; 10: Padua; 11: A Petrarca; 12: Este; 13: Montagnana (Saletto); 14: Villa
Bartolomea (Lovara); 15: Oppeano; 16: Gazzo Veronese (drawn by author).

Venetic Funerary Evidence

Several thousand graves, dating between c.1050 BC and AD 25, have been
excavated in the region. Regular excavations began in the late nineteenth
century and have continued to the present day. Many grave assemblages dug up
124 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

in the past have been lost or were never properly published. The vast majority
of the tombs excavated in the last 20 years with exceptional scientific accuracy
remain unpublished. Our knowledge of Venetic funerary rituals is mainly based
on a dataset of more than one thousand graves from different localities, which
have been excavated and published over the last 130 years.13 Cremation was the
main funerary rite practised between the Late Bronze Age and the Early Roman
period. Inhumation was adopted less frequently and possibly for the deposition
of people of a low status.14 Inhumation graves are not considered in this paper,
owing to the poor documentation of this burial type. When cremation was
adopted, the bones of the dead were selectively collected from the pyre, washed
and placed in a bronze or ceramic urn. Urns were generally covered with an
upside-down bowl or a lid, and then placed in a tomb either in the form of a
container or as a burial into the ground (Figures 9.2 and 9.3). Graveyards were
always located outside the settlements, although sometimes infants were buried
as inhumations under the floors of houses.
The structure of cremation graves varies, according to the locality, the rank
of the dead and the availability of raw material. At Este, the most common
tomb container was a box of stone slabs of variable size, the so-called cassetta.
At Padua, it was more common to place the urn(s) in a wooden container or
in a large pot (dolium). Everywhere, grave goods including ornaments, tools,
vessels and sometimes weapons were placed in the tomb with the dead. The
composition and wealth of the grave assemblage varied according to the rank,
age, gender and social or kin affiliation of the deceased. Some graves were
accompanied by a tombstone. Multiple graves were not uncommon. These
may entail the deposition of more than one urn in a tomb container, and/or
the burial of more than one person in a single urn. Recent excavations have
proved that tomb containers were often reopened to deposit additional bones
of other individuals.15 Clusters of tombs probably belonging to relatives or
social affiliates were placed near each other and sometimes were covered with
a collective burial mound. The dimension, shape, structure and spatial organi-
zation of collective tumuli may vary according to the locality, historical phase
and social standing of the owners. For example, a hierarchical disposition of
the tombs is evident in the largest mounds built at Este between the sixth and
the fifth centuries BC, with the erection of the most prestigious cassette in the
middle of the tumulus, while small cassette, simple pit graves and inhumations
were located on the tumulus edge, in clear subordinate positions.16

Literacy and Writing

The extent of literacy in Iron Age Veneto is unknown. The number of inscrip-
tions recovered up until now is not large (less than a thousand over a period
of six centuries) but this may be a product of preservation – obviously texts
written on perishable items cannot be recovered. The bulk of the corpus
comprises a few hundred poorly preserved inscriptions consisting of single
Fa m i ly R e l at io n sh i p s 125

Fig. 9.2. Reconstruction of Venetic urn with bones and grave goods. 1: ceramic urn; 2;
ceramic urn lid; 3: fragment of bronze bracelet; 4: fragment of bronze plaque; 5: amber,
glass and bronze pendants and beads (drawn by author; 1 and 2 adapted after Salzani 2008).
126 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

Fig. 9.3. Reconstruction of Venetic pit

grave. 1: burial earth bound; 2: pyre
debris; 3: ceramic urn with bones and
grave goods (fibula and bracelets); 4:
dining set (drawn by author).

or repeated letters. The remaining data include about 250 texts generally
consisting of no more than 10 words. Only one inscription recently found at
Este is significantly longer (around a hundred words, half of which are now
lost).17 Writing was probably adopted from the Etruscans in the first half of
the sixth century. However, the nature of its introduction into Veneto is still
uncertain. A Greek-modified alphabet was used to write the local language up
to the late second or first centuries BC, when Venetic started to be gradually
be supplanted by Latin. Inscriptions written in Latin still included Venetic
personal names down to the first century AD.18 Venetic texts are found on a
wide range of stone, ceramic and metal artefacts, including gravestones, urns,
urn lids, dining and drinking vessels, and bronze votive offerings. The context
of use for Iron Age Venetic writing seems to have been limited mainly to the
ritual sphere.19 The vast majority (90 per cent) of the inscriptions published to
date have been found in sanctuaries and cemeteries, where they were offered
to the deity as votive dedications, or to the dead as epitaphs.20 With very few
exceptions, the texts were formulaic, and usually record theonimis (i.e. divine
appellatives) and people’s personal names. Verbs and nouns – including terms
to indicate kinship relations – are rare.
Regardless of their brevity, Iron Age and early Latin inscriptions provide
some information on Venetic kinship relations. Venetic society seems to have
been mainly patrilineal. Standard Venetic onomastic formulas were composed
of a given name accompanied by a second name in the form of an adjective
ending in –io-/(i)ko- in the masculine form and in –ia- in the feminine form.
These forms are usually interpreted as patronymics (e.g. Uko Ennonios, Aviro
Broijokos, Ebfa Baitonia).21 However, several exceptions are known to this rule.
People were sometimes designated by their given name or second name only
(e.g. Lessa, Broijokos). Adjectives in –iaio- (masculine) and –iaia- (feminine)
are also attested and have been interpreted as metronimics (e.g. Osts Katusiaios,
Nerka Trostiaia).22 Male given names were sometimes followed by two adjec-
tives in –io- (e.g. Urkl(e)i Egetorioi Akutiioi). The meaning of the second form
in –io- is unclear, although it might have derived from the name of a close
Fa m i ly R e l at io n sh i p s 127

relative other than the father (e.g. grandfather), or it may have been used to
sanction anomalous ‘judicial’ or personal situations in the context of Iron Age
Venetic society.23 Women’s onomastic formulas were often composed of a given
name followed by an adjective ending in –(i)na- or –kna- (e.g. Kanta Makna,
Fougonta Lemetorna). The name in –na- was sometimes accompanied by a
supposed patronymic in –ia- (e.g. Fougonta Fugisonia Brigdina). Until recently,
female names in –na- were interpreted as gamonymics deriving from the name
of a woman’s husband. This hypothesis has been undermined by recent osteo-
logical analysis, which indicates that adjectives in –na- were also given to infant
girls.24 Male names in –(k)nos- are also sporadically attested, either as the first or
the second adjective in the onomastic formula (e.g. Moldo Boiknos, Aios Paplons
Fremaistknos). Their function is still unexplained, although there has been some
speculation they might have derived from ‘Celtic’ onomastic forms.25 Finally,
there is some evidence that some Venetic second (or third) names may have
been used as gentilicia (family or clan names), as in the cases of the Pannarii
from Altino and the Andetici from Padua and Belluno.26 Venetic onomastic
formulas started to be gradually supplanted by Latin onomastic forms from
about the first century BC.

Venetic Kinship and Family Structure

Osteological Evidence
Osteological analysis of cremated human remains has become an important
feature of Venetic archaeology in the last 20 years.27 When cremated bones are
preserved, important information can be gained on the funerary treatment
undergone by the corpse, the sex and age of the deceased, and the minimum
number of individuals placed in the urn. The osteological dataset analysed in
this paper includes a minimum of 523 dead individuals buried in 413 urns
excavated in 8 different Venetic localities, namely Este, Padua, Gazzo Veronese,
Montebelluna, Altino, Saletto di Montagnana near Este, Ponzano Veneto in
the Treviso province, and Lovara in the Veronese.28 In this paper, the evidence
available is analysed through SPSS on the basis of five variables, i.e. the age and
sex of the deceased, the number of individuals buried in a single urn, the place
of burial, and the chronology of the deposition. This preliminary survey does
not take into consideration other important variables such as the burial type,
the precise location of the tomb in the cemetery, the total number of urns placed
in a single tomb container, and the grave assemblage buried with the deceased.
These issues are widely discussed in my doctoral thesis (in preparation) and will
be included in future publications.
The geographic distribution of the sample is not uniform. The vast majority
of the urns included in my dataset are from Este (235) and Gazzo Veronese
(115). Twenty-three urns are from Montebelluna and 28 from Padua. The data
128 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

from Lovara, Altino, Ponzano Veneto and Saletto are not statistically significant
(11 urns in total, of which 5 are from Altino). The precise provenence of one
urn is unknown (Figure 9.1; Table 9.1). A dating has been possible for 351
urns out of 413. In 62 cases, a chronology is not available owing to a poor
excavation or publication record. The dataset includes 84 urns dating between
the early tenth and the late ninth century, 75 between the early eighth and the
late seventh, and 148 between the late seventh and the late fourth, with a peak
of occurrences in the sixth (86). Only 44 urns date between the late fourth
century BC and the Early Roman period (Table 9.2).29 The scarcity of data for
this phase is probably due to poor archaeological evidence. As more recent
tombs are nearer to the soil surface, they have often been destroyed and were
not recorded prior to their destruction. It must be noted that all the urns dating
between the early tenth and the late ninth centuries BC but four are from
Gazzo Veronese, while the Padua sample includes only graves dating from the
sixth century BC onwards.
The sex of the deceased was identified for 247 individuals out of 523. Sexing
is always a complicated procedure with cremated remains, and is obviously
impossible for immature individuals and badly preserved and/or incomplete
adult bodies.30 If we exclude the Late Bronze Age/Early Iron Age Gazzo sample,
where men almost equal women (30 to 38), women correspond to around 70
per cent of the sexed sample (125 individuals) while men represent only 30 per
cent (54 individuals). An explanation for such an uneven distribution is difficult.
Although the possibility of significant errors in the osteological analysis cannot
be entirely discounted, it must be noted that (a) the Gazzo sample shows that the
diagnostic parameters adopted by Venetian anthropologists are not positively
biased towards the identification of women, and (b) an overwhelming female
presence, especially at Este, is confirmed by my preliminary archaeological
analysis of the grave assemblages. An alternative explanation is that during the
Iron Age men were more likely to die away from home (e.g. in war) and to be
buried where they died. A more suggestive hypothesis – which requires further
investigation – is that certain categories of men may have been intentionally
excluded from the burial ground or given alternative forms of burial.
Age was determined for 477 individuals out of 515. Children between birth
and 13 years are 184, probably with a very low percentage of infants between
birth and 12 months.31 Adults between 21 and 40 years are 233. Adolescents
(aged between 14 and 20 years) and mature people between 41 and 60 years
are 31 and 33 respectively (Table 9.3). This evidence suggests that Venetic
cremations were not representative of a standard funerary population for a
pre-industrial society, as both adult men and very young children are under-
represented. The complete lack of senile people is also notable, although it is
unclear whether this is due to imprecise osteological analysis, a very short life
expectancy or the exclusion of these people from the formal burial ground. It
must be kept in mind that inhumation graves are not discussed in the present
Fa m i ly R e l at io n sh i p s 129


200 Este
100 Other


Number of urns



100 8th–7th
80 Late 4th – AD25

60 ND



Number of urns
Table 9.1 Geographic distribution of the sample (urns) and Table 9.2: Chronological
distribution of the sample (urns)

An estimation of the minimum number of individuals placed in each single

cinerary vessel has been possible for 411 urns out of 413. Overall, 334 urns
(81 per cent) contained one individual, while 77 urns (18 per cent) contained
at least two individuals. Of the latter, 73 urns contained not fewer than two
individuals (17 per cent), three urns contained at least three individuals (1 per
cent) and one urn contained a minimum of four individuals (0.2 per cent).
Urns containing two or more individuals have been found at four out of the
eight settlements considered in the sample (Este, Padua, Gazzo Veronese
130 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d


100 ND


Number of individuals
Table 9.3 Age of individuals
and Montebelluna). There is no evidence of bone mingling at Altino, Saletto,
Ponzano and Lovara but this is probably due to the scarcity of data from
these settlements. In particular, bone mingling is surely attested at Early
Roman Altino, as I discuss later on, and it is therefore possible that this ritual
was performed at this site also during the Iron Age. The percentage of urns
containing more than one individual vis-à-vis single urns varied at different
burial sites, ranging from the Benvenuti cemetery at Este, where 16 urns out
of 56 were multiple (30 per cent), to the Colombara cemetery at Gazzo, where
only 5 urns out 48 contained more than one dead (12 per cent). Concerning
the chronology of the finds, bone mingling practices are attested over the
entire period considered, from the Late Bronze Age to the Early Roman phase.
The persistence of bone mingling practices in Early Roman Veneto is further
confirmed by the recent excavation of the Roman cemeteries of Altino.32
According to preliminary osteological analysis on a sample of 345 cremations
dating c.25 BC – AD 50, four urns contained at least two individuals each
(1 per cent).
Overall, 405 urns out of 413 contained at least one individual for whom age
was verified. The age distribution of the sample shows that 116 urns out of 405
contained one child buried alone, 161 urns contained one adult, 24 contained a
young individual and 27 a mature person. When taking into account the urns
containing more than one dead individual (77 out of 405), several age combi-
nations are possible. The most common case entailed the deposition of one or
two children in the same urn with one or two adults (39 occurrences). Children
were also buried with other children (ten cases) and more rarely with young or
mature individuals (one and five cases, respectively). Adults were buried with
other adults (with no children) in 12 cases. More rarely they were associated with
Fa m i ly R e l at io n sh i p s 131

young individuals (four cases) and never with mature people. In four cases, one
adult was placed in the urn with an individual whose age is unknown. Seldom
young people were buried together (one case) or with a mature individual (one
case). If we exclude the four cremations for which the age of the second deceased
person is uncertain, 55 multiple urns out of 77 contained at least one child.
Almost half of the children included in my sample were buried in the same urn
with another person. The impression, therefore, is that bone mingling practices
were often adopted for the burial of infants. This hypothesis is further confirmed
by findings from the recently excavated Via Umberto I cemetery at Padua. Here,
more than a half of the numerous multiple cremations analysed so far contained
the remains of children buried with adults.33
Overall, 237 urns out of 413 contained at least one individual whose sex
was identified. Almost half (114 out of 235) contained a single woman, while
62 contained a single man (about 26 per cent). If we consider the urns which
included more than one individual (58 out of 235), 12 contained a woman
and man, 34 a woman and an individual of uncertain sex and 12 a man and
an individual of uncertain sex. In the majority of the cases, the individual for
whom sexing has been impossible is a child. Children, therefore, were often
buried together either with a woman or with a man. In two cases, a child was
placed in the urn with both a woman and a man.
The overall pattern of sex and age distribution of the sample suggests that
bone mingling practices were adopted for the burial of close relatives, probably
the members of nuclear families. First, this is indicated by the complete absence
of both urns containing the remains of same sex adults and urns containing
more than two adults. The deposition of a woman and a man together would
suggest the burial of a couple. The common interment of adult (or mature)
individuals with children should indicate the deposition of parents (or even
grandparents) with their sons/daughters. Children buried together may have
been siblings. The burial of a complete nuclear family is probable in those
two cases in which the remains of a man, a woman and a child were mingled
together. The sex/age distribution of the small Roman Altino sample confirms
this pattern, as two multiple urns out of four contained a couple of adults
each, while the other two contained a mature woman and a child, and an adult
woman with an infant, respectively.34

Epigraphic Evidence
When osteological analysis is lacking, epigraphic sources are an important tool
to recognize occurrences of bone mingling. The structure of Venetic and early
Latin epitaphs was usually very simple. In the vast majority of the cases, the
epitaph consisted of the plain name of the dead, generally in the nominative
or dative case (e.g. Fogotna). More articulated inscriptions were rare, although
sometimes the name of the dedicator of the epitaph was mentioned with the
name of the deceased (e.g. Ivanta Soccina Pusioni ma, i.e. Ivanta Soccina to
132 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

Pusione). ‘Speaking’ inscriptions are also sporadically attested (e.g. Plutevei

Panarioi ego, i.e. I am [the urn] for Plutev- Panarios). Early Venetic sepulchral
epitaphs c.550–300 BC were usually inscribed on tombstones set up outside the
grave and generally found sporadic far from their original location. Therefore,
these texts are useless to establish the number of individuals buried in a single
urn, even when they mention more than one individual. Late Venetic and
Early Roman epitaphs were usually inscribed on the body or the lid of cinerary
vessels buried in the grave. Multiple epitaphs on the same vessel, often clearly
inscribed by different people in different moments, as well as the presence of a
single epitaph mentioning more than one person, strongly suggest that the urn
was used for the deposition of two or more individuals.
Overall, my epigraphic dataset includes 106 inscribed cinerary vessels and
urn lids from different Venetic localities. More precisely, 81 urns are from
Este,35 15 from Montebelluna,36 4 from Covolo (Treviso),37 2 from Altino,38 2
from Valle di Cadore in northern Veneto,39 and 1 each from Arquà Petrarca,
near Este, and Belluno (Figure 9.1; Table 9.4).40
Several other inscriptions from Este, Arquà Petrarca and Altino have been
found in a tomb, inscribed on a few pottery fragments, some ceramic containers
and a bronze cup reused as a tripod.41 Owing to poor archaeological evidence,
it is unclear whether these inscriptions were epitaphs recording the name of
the dead buried in the urn, or property markers inscribed on culinary vessels
offered to the deceased. Therefore, they are excluded from the sample. The
inscriptions included in the dataset mainly date between the third century
BC and the late first century BC, although there are some possible earlier
and later examples. The vast majority are written in Latin, although many



70 Este
50 Altino

40 Valle



Number of epitaphs
Table 9.4 Geographic distribution of the sample (epitaphs)
Fa m i ly R e l at io n sh i p s 133

of them still record Venetic personal names. The most recent pure Venetic
epitaph whose chronology is secure, Este inscription 81, dates to the early
first century BC.42 One of the latest occurrences of the Venetic language in the
funerary context is Treviso inscription 4, which dates to the Augustan period.43
Despite the use of the Latin alphabet, the declension and the onomastic forms
attested in this epitaph are probably still Venetic.44 Almost all the inscribed
urns mentioned in this paper were excavated between the end of the eight-
eenth century and the first half of the twentieth. The context of discovery is
generally poorly documented, and osteological analysis has been possible
only in five cases.45 On the contrary, almost all the graves excavated in the
last 20 years and recently published do not contain inscriptions. Therefore, a
direct comparison between osteological and epigraphic evidence is generally
Eight epitaphs out of 106 are poorly preserved, incomplete, or difficult to
translate.46 Another ambiguous occurrence is Latin Este inscription 3, which
refers to two female individuals. The first woman, Septuma Sex f(ilia) T. Rutili
uxor (Septuma, daughter of Sex(tus), wife of T. Rutilius), is mentioned in the
nominative case, while the second name, Aemiliae, is probably in the dative
case. Here, it is unclear whether Septuma and Aemilia were buried together, or
the former was the dedicator of Aemilia’s epitaph.47 Of the rest, 88 epitaphs out
of 97 mention only one dead individual,48 while nine document the presence of
at least two people in the cinerary vessel.49 Seven multiple urns out of nine date
from the first century BC onwards and confirm the persistence of bone mingling
practices in Early Roman Veneto. If we consider the several inscriptions from
Este, Arquà Petrarca and Altino whose function as epitaphs is uncertain, only
one, Este inscription 121, may have mentioned two individuals. Unfortunately,
the meaning of this text is unclear.50 Overall, the epigraphic evidence presented
above suggests that bone mingling practices were adopted for about 10 per
cent of the burials included in my epigraphic dataset. This proportion is lower
than the percentage (18 per cent) found in my osteological sample. A possible
explanation is that the adoption of bone mingling practices declined in Late
Iron Age and Early Roman Veneto, as also suggested by Tirelli’s Altino sample.
It is also probable that some of the epitaphs included in my dataset referred to
non-Venetic people who settled in Veneto in the Early Roman period and did
not practice bone mingling at all. However, we cannot exclude some of these
inscriptions that did not mention all the individuals buried in the urn, especially
in the case of children, who may have been sometimes denied post-mortem
Closer inspection of the inscriptions sheds some light on the kinship ties
which may have existed between the people buried together. In three cases,
the epitaphs clearly corroborate my previous analysis, and demonstrate that
the individuals buried together were close relatives. The best example is
offered by Este inscription 111.52 This epitaph states that the urn was given by
a woman called Fugenia to her parents Fougonta and Egtor– Lamusio(s), who
134 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

were therefore buried together as a couple.53 It is noteworthy that Fugenia and

her mother had very similar given names, both deriving from the well-known
Venetic onomastic stem F(o)ug-. As I discuss later, the adoption of similar
personal names by the members of a same family was probably an intentional
social strategy to underline strong kinship relations.54 Another notable occur-
rence is Latino-Venetic Belluno inscription 1.55 This epitaph was inscribed on
a bronze vase probably used as an urn, although the original grave assemblage
is now lost. According to the most probable translation of the text, the vessel
was a ‘property’ of – or was selected as an urn by – a man called Enonio(s), to
be used by himself and two other male individuals called Onte and Appios.56 All
the three men share a common second name, possibly a gentilicium (Andetici),
and therefore were relatives. Although it is impossible to define their kinship
relationship more precisely, it is probable that at least Enonio(s) was an adult
man, as he was the dedicator of the epitaph and the owner of the vessel. Onte
and Appios may have been his sons or younger relatives, a hypothesis entailed
by their somewhat subordinate position vis-à-vis Enonio(s) in the text. A third
epitaph, Latin Treviso inscription 1 from Montebelluna, testifies that two male
individuals, L(ucius) N(e)ppiaqus L(uci) f(ilius) (Lucius Neppiaqus, son of
Lucius), and T(itus) Neppiacus L(uci) f(ilius) (Titus Neppiacus, son of Lucius),
were buried together in the cinerary urn. The onomastic formulas suggest they
were relatives, and possibly a father and his son, or siblings.57 Interestingly, two
other inscribed urns found in the same locality, and probably in the same tomb
of Treviso inscription 1, may have belonged to Lucius’ and Titus’ relatives. The
two epitaphs in fact refer to a man called L(ucius) Neppiacus son of Sex(tus), and
to a second man called L(ucius) Neppiacus, respectively.58
The evidence offered by Venetic Este inscription 80, dating to the second
century BC, is more ambiguous.59 This epitaph was inscribed on a cinerary
vessel found in Benvenuti tomb 123, a wealthy multiple cassetta grave excavated
at Este in 1879.60 The cassetta appears to have been reopened several times
between the third and the first centuries BC to allocate a total of ten urns,
four of them inscribed with an epitaph. Personal names from the well-known
onomastic stem Frem- recur on three epitaphs out of four (Frema, Fremaistna
and Fr), probably to remark the kinship ties existing between the individuals
buried in the tomb.61 In the case of Este inscription 80, the short epitaph
indicates that the cinerary vessel contained the remains of at least two people,
a female individual called Kanta and a male person called Fougont-.62 The two
given names are followed by a shortened second name, Fr, clearly deriving
from the Frem- root. It is unclear whether the latter referred to both Kanta and
Fougont- or to Fougont- only. Interestingly, a tiny urn also found in the tomb
and contemporaneous to the previous one, mentions a dead individual, possibly
a child, called Vant- Fougontios (i.e. son of Fougont-), who may have been a son
of the two.63
In other cases, the existence of a kinship bond between the individuals
buried together in the urn is suggested by the use of very similar given names
Fa m i ly R e l at io n sh i p s 135

for the two people. This might be the case of Latino-Venetic Este inscription
107.64 According to the epitaph, the urn contained the remains of two female
individuals, Vanta and Ivanteia Fremastina, who share very similar names
deriving from the common onomastic root (I)vant-.65 A similar occurrence is
probably Este inscription 106, which also mentions two female dead buried
together, Fo[-i]a Ostina and Fougonta Toticina.66 Although Ostina’s given name
is incomplete, it probably derived from the F(o)ug- onomastic stem, similarly
to Toticina’s. If we maintain that bone mingling practices were adopted for
the burial of close relatives, it is notable – although not unexplainable – that
the same –na- adjective was not necessarily shared by kin affiliates. A similar
instance is offered by Este inscription 127, on a bronze vase used as a cinerary
vessel found in Benvenuti tomb 123/1879, and dating to the first phase of use
of this grave (c.275 BC). According to the epitaph, two female individuals with
different second names, Frema Boialna and Rebetonia Votina, were buried
together.67 Marinetti still maintains that at least in some cases the –na- adjective
may have indicated a woman’s marital status. Frema and Rebetonia, therefore,
may have been sisters, sisters-in-law, or even a mother and her daughter,
obviously married with two different men (*Boialos and *Votos, respectively).
This hypothesis, however, would imply the deposition of two adult women
together, an occurrence which is absent in my osteological sample.
The other three inscriptions (Este XX, XXII and XXIX) which mention
the placing of two people together in the urn, do not shed much light on the
relationships existing between these individuals. Este inscription XXII suggests
the deposition of a woman and a man together. The two may have been a
wife and her husband, although other explanations are obviously possible.
According to Pellegrini and Prosdocimi, the woman, Iva Aidria Vol. f(ilia) (Iva
Aidria, daughter of Vol.), was possibly Venetic.68 Her praenomen (given name)
may have been an abbreviation of Venetic I(u)vanta, while her patronymic
may have derived from another common Venetic name, Voltiomnos. Indeed,
the very same use of a female praenomen, which was not common in Latin
onomastic formulas but was the norm in Iron Age Veneto, further reinforces
this hypothesis.69 The man’s onomastic formula (M. Rutili L. f.) (M. Rutilius, son
of Lucius) does not offer much insight into his relation to Iva. Este inscription
XXIX mentions two female individuals. One, Canta Paphia C.[ . . . ] uxor
(Canta Paphia, wife of [ . . . ]), is designated with a longer onomastic formula,
while the second, Moltisa, has only one name.70 Apart from Este inscriptions
107 and III discussed above, another epitaph which mentions two female
individuals, the first with an expanded onomastic formula and the second
with a shorter one, is Este inscription XX (Konia Cn. f. Libona or Libonia and
Qu[-]rta C. f.) (Konia, daughter of Cn(eus) Libona/Libonia; Qu[-]rta, daughter
of C(aius)).71 It is notable that in two of these peculiar female epitaphs (Este
inscriptions III and XXIX), the person granted an expanded onomastic
formula was an adult woman and a wife (i.e. Septuma Sex f(ilia) T. Rutili
uxor and Canta Paphia C.[ . . . ] uxor). This might have been the case of Este
136 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

inscriptions 107 and XX as well, if the –na- adjective was a gamonymic (Libona
and Fremastina). If we take into account my previous osteological dataset, it is
possible that Vanta, Moltisa, Aemilia and Qu[-]rta were girls buried with their
mothers. They may have received a shorter onomastic formula either because
children were sometimes denied a complete name vis-à-vis adults or because
their lineage was already evident after their parents’ names. In this regard, it
is noteworthy the great variability attested in the structure of these possible
mother/daughter epitaphs. For example, whereas Qu[-]rta was granted both a
given name (deriving from a numeral) and a patronymic, Moltisa’s onomastic
formula is composed of a single element (probably a given name). The case of
Este inscription III is even more intriguing. Here, Septuma is designated with a
numeral used as a praenomen, a patronymic (Sex f) and a form indicative of her
marital status (Rutili uxor). Interestingly, she is not given a gentilicium, which
is the only component of Aemilia’s onomastic formula. A possible explanation
is that Septuma’s kin affiliation was made explicit through her association
with Aemilia’s name, a hypothesis which would further confirm their close

This paper has brought together two different lines of enquiry – osteological analysis
of cremated human remains and epigraphy – to shed some light on the structure
of the family and kinship relations in the Italian region of Veneto between the Late
Bronze Age and the Early Roman period. My analysis focused on a selected, and
very characteristic, trait of Venetic funerary rituals, that is bone mingling. The
sex/age distribution of the osteological sample discussed in this paper strongly
suggests that the burial of two or more people together in the urn was reserved to
close relatives and probably to the members of nuclear families. This burial pattern
seems to have been maintained for over a millennium, regardless the huge social
and political developments that Venetic society must have undergone over time.
The hypothesis that bone mingling was adopted for the deposition of close relatives
is generally strengthened by the epigraphic evidence presented, although some
inscriptions are not incompatible with alternative forms of disposing of the dead.
The evidence discussed in this paper is obviously not exempt from pitfalls.
These include the uneven geographical distribution of both samples, the partial
lack of chronological correspondence between the epigraphic and the osteo-
logical datasets, and the impossibility to perform bone analysis in the case of
many inscribed urns. Moreover, owing to the brevity of the publication, it has
been impossible to discuss in more detail the archaeological and historical
context of the finds. Still, this work provides interesting insights into Venetic
kinship structure. Overall, it is possible to suggest that, in the late prehistoric
period in Veneto, an extraordinary social and ritual importance was given to
the bond established between parents and their children, as well as between
Fa m i ly R e l at io n sh i p s 137

siblings and spouses. Bone mingling practices were still adopted during the
period of Romanization and may have become an important means of cultural

  1 Capuis, 2009a.
  2 Marinetti, 1992; 1999; 2001; 2008a; 2008b; Pellegrini and Prosdocimi, 1967 and Prosdocimi, 1988.
  3 Capuis, 2009b; De Marinis, 1999; Marinetti, 2001 and Prosdocimi, 1988.
  4 Capuis, 2009a.
  5 De Marinis, 1999 and Capuis, 2009b, 182.
  6 Capuis, 2009a.
  7 Capuis, 2009a and Capuis and Chieco Bianchi, 1992.
  8 Ruta Serafini, 2002.
  9 Maggiani, 2002.
10 Capuis, 2009b and Marinetti, 2008a.
11 Cuscito, 2009.
12 Chieco Bianchi and Calzavara Capuis, 2006, pp. 301–319.
13 A full dataset with a comprehensive bibliography is in my PhD thesis (in preparation).
Publication on Venetic funerary rituals is vast (e.g. Bianchin Citton et al., 1998; Capuis,
2009a; Capuis and Chieco Bianchi, 1992; Chieco Bianchi and Calzavara Capuis, 1985; 2006;
Gambacurta and Ruta Serafini, 1998; Ruta Serafini, 1990 and Salzani et al., 2000).
14 A full discussion of Venetic inhumation rites is in my PhD thesis (in preparation). This
analysis proves that although Venetic inhumation burials were indeed – on average – less
wealthy than cremation graves, a great variability in tomb structure, location of the tomb in
the cemetery, and composition of the grave assemblage is attested for both cremations and
15 Gambacurta and Ruta Serafini, 1998.
16 Balista and Ruta Serafini, 1998.
17 Marinetti, 2008b.
18 Unfortunately, many Venetic and early Latin inscriptions from Veneto have been randomly
found, or were excavated in the past and poorly documented. Therefore, a precise chronology
of these finds is often lacking. The most recent Venetic inscriptions known so far were inscribed
on votive offerings found at the Romano-Venetic sanctuary of Auronzo di Cadore in northern
Veneto, and seem to date not earlier than the first century AD (Marinetti, 2008a).
19 Whitehouse and Wilkins, 2006.
20 It should be noted, however, that several dozens of inscriptions have been recently
excavated in settlement contexts (e.g. Marinetti, 2008b). This evidence suggests that
Venetic writing may have been more widespread than previously acknowledged, even in
non-ritual contexts.
21 Patronymics in –io- were common in southern and central Veneto, while the form in –(i)ko- is
mainly attested in Cadore (northern Veneto). The feminine form in –(i)ka- is rare or absent, as
inscriptions with female names are very rare in northern Veneto (Marinetti, 2001).
22 Second and third names in –iako- and –iaka-, possibly the equivalent to ‘southern’ metro-
nymics in –iaio- and –iaia-, are attested both in northern Veneto and at Este (e.g. Molo
Arbonkos Ostiako[-], Fugia Fougontiaka). In one case, a second name in –iako- was accom-
panied by the word libertos (freedman; Eskaiva libertos Arspetijakos), leading to the speculation
that the –iako-/–iaka- adjective may have indicated the freedman status (Marinetti, 2001;
Pellegrini and Prosdocimi, 1967, p. 254).
23 Prosdocimi, 1988, p. 381.
24 This is the case of a small cinerary urn found in late Venetic – early Roman Benvenuti tomb
123/1879 (Este). This urn, which is inscribed with a –na- adjective in the nominative case
138 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

(Fremaistna) proved to contain the remains of an infant aged from birth up to seven (Chieco
Bianchi and Calzavara Capuis, 2006, p. 289).
25 Marinetti, 1996, p. 99.
26 Marinetti, 2008a, p. 156.
27 For example Drusini et al., 1998; 2001.
28 A full dataset with a complete bibliography is in my PhD thesis (in preparation).
29 This chronological categorization roughly corresponds to the four ‘periods’ of historical
development of the Este culture established by Prosdocimi, 1882, and still adopted (with
adjustments) by mainstream Italian scholarship.
30 Drusini et al., 1998.
31 The exact number of very young infants included in my sample is unknown because (a) infant
bones, being very fragile, are more difficult to identify and (b) a precise determination of the
age of cremated individuals is generally impossible.
32 Tirelli, 2001, p. 248.
33 Gamba and Tuzzato, 2008, p. 65. The Via Umberto I cemetery was excavated in 2002–2003 and
is largely unpublished. The tombs excavated so far include 520 cremations, 169 inhumations,
and 3 burials of horses (Gamba and Tuzzato, 2008, p. 59). Unfortunately, the precise number
of cremations analysed to date by Venetian anthropologists is not specified in Gamba and
Tuzzato, 2008.
34 Tirelli, 2001, p. 248.
35 (For readers not familiar with numbering systems of Venetic epigraphy, throughout these
endnotes, Arabic numerals indicate Venetic epitaphs; whereas those in Roman numerals
indicate Latino-Venetic and Latin inscriptions). Este inscriptions 76, 77, 79, 80, 81, 82, 83, 88,
89, 90, 91, 92, 93, 95, 98, 99, 103bis, 104, 105, 106, 107, 108, 109, 110, 111, 113, 127, 128, I, II,
XLVIII, LIII, LIV, LV, LVI, LX LXVIII (Chieco Bianchi, 1987; Chieco Bianchi and Calzavara
Capuis, 2006, pp. 302–304; Marinetti, 1992; Pellegrini and Prosdocimi, 1967 and Prosdocimi,
36 Treviso inscriptions 1, 3, 5, I, II, III, V, VI, VIII, IX, X, XI, XII, XIII (Pellegrini and Prosdocimi,
1967). Another inscription (Pusilae) is mentioned in Prosdocimi, 1988, p. 260.
37 Treviso inscriptions 2, 4, IV, VII (Pellegrini and Prosdocimi, 1967).
38 Treviso inscriptions 11, 12 (Marinetti, 1999)
39 Cadore inscriptions 74 and XV (Pellegrini and Prosdocimi, 1967; Prodocimi, 1988, pp.
40 Este inscription LXIX and Belluno inscription 1 (Pellegrini and Prosdocimi, 1967 and
Prosdocimi, 1988).
41 These include Este inscriptions 78, 84, 85, 86, 87, 94, 96, 97, 100, 101, 102, 103, 112, 114, 121,
123, 124, 126, 133, 134, XI, XVI, XLIX, L, LI, LII, LVII, LVIII, LIX, LXI, LXII, LXIII, LXIV,
and other eleven smaller ceramic fragments with no classification number (Marinetti, 1992;
Pellegrini and Prosdocimi, 1967,Prosdocimi, 1988). A few other possible epitaphs from Altino
have been published in Marinetti, 1999 and Tombolani, 1987.
42 The inscription records a male Venetic name in the nominative case, namely Lemetor Urkleiars
(Chieco Bianchi and Calzavara Capuis, 2006, pp. 313–314).
43 Bandelli, 1999, p. 660.
44 The epitaph reads Fema Matricai, that is Fema to Matrica.
45 Este inscriptions 95 (urn found in Este Benvenuti tomb 123/1879), Este inscription 128 (bronze
vase containing a ceramic urn found in Este Ricovero tomb 23), Este inscription LXVIII
(ceramic urn found in Este Ricovero tomb 123b), and Treviso inscriptions 11 and 12 (urn lids
found in Altino Fornasotti tomb 1). The first two inscriptions mention a single dead individual
each (Fremaistna and Nerka Trostiaia, respectively), a datum confirmed by the osteological
analysis (a birth–7-year-old child, and adult, sex unknown). The presence of a single person in
the urn seems to be confirmed also in the case of Treviso inscriptions 11 and 12 (i.e. Plutevei
Panarioi ego: I (am the urn) for Plutev- Panarios; P[----Pan]narioi e[go]: I (am the urn) for
Fa m i ly R e l at io n sh i p s 139

P[--- Pan]narios). The third epitaph from Este, Este inscription LXVIII, possibly mentions one
individual only, but the urn may have contained the remains of two people.
46 Este inscriptions 89, 93, 103bis, 113, XVIII, and Treviso inscriptions 2, 3, XI.
47 Pellegrini and Prosdocimi, 1967, pp. 236–238.
48 Este inscriptions 76, 77, 79, 81, 82, 83, 88, 90, 91, 92, 95, 98, 99, 104, 105, 108, 109, 110, 128, I, II,
LX, LXVIII, LXIX,Treviso inscriptions 1, 4, 5, 11, 12, II, III, IV, V, VI, VII, VIII, IX, X, XI, XII,
Cadore inscription XV. In the case of Este inscriptions 108, XVII and XLIV (on three different
cinerary urns), the name of the deceased is repeated twice. Although it is probable that these
double inscriptions mentioned the same person, we cannot exclude a priori that two people
with the same name were buried together. In the case of Este inscription LXVII the situation is
less clear. Chieco Bianchi (1987, pp. 195–197) indicates that the urn may have contained the
remains of an adult male and a child, while Balista et al. (1988, 280) mention the single burial
of a man, with at least one bone element possibly referring to a second deceased. The epitaph
is a Latin onomastic formula mentioning a male person (C. Coponi Scirpae). It is noteworthy,
however, that a shortened version of a similar or identical name (C Cop) is also inscribed on
the urn. Therefore, the second epitaph may refer to the adult man again, or mention a second
individual, perhaps a child with the same of his ‘father’.
49 Este inscriptions 80, 106, 107, 127, XX, XXII, XXIX, Belluno inscription 1 and Treviso
inscription I.
50 Prosdocimi, 1988, pp. 253–255.
51 Chieco Bianchi, 1987, pp. 195–197.
52 Pellegrini and Prosdocimi, 1967, pp. 230–231; Prosdocimi, 1988, p. 260.
53 Fougontai Egtorei filia Fugenia Lamusioi, which translates to: to Fougonta and Egtor- Lamusio(s),
(their) daughter Fugenia. The chronology of this urn is problematic. The epitaph is written in
a pre-Augustan version of the Latin alphabet but the grave goods found inside include objects
dating to the first century AD. A possible explanation is that the vessel was re-used several
times. However, as the urn was excavated more than one century ago, an incorrect description
of its content is not unlikely
54 Chieco Bianchi and Calzavara Capuis, 2006, p. 294.
55 Pellegrini and Prosdocimi, 1967, pp. 451–452; Prosdocimi, 1988, pp. 307, 382.
56 Enoni Ontei Appioi sselboisselboi Andeticobos ecupetaris /piis., i.e. Enonios’ ecupetaris. To Onte,
Appios and himself, Andetici. Ecupetaris and similar forms are recurrent terms on Venetic and
early Latin epitaphs from Este, Padua, Belluno and Altino. They probably translate to ‘urn’ or
‘tomb’, with a reference to the (high) status of the deceased, but their precise meaning is unclear
(Prosdocimi, 1988, pp. 296–299).
57 Pellegrini and Prosdocimi, 1967, p. 419.
58 Pellegrini and Prosdocimi, 1967, p. 420. The name ‘Neppiacus’ is probably a pre-Latin form
used as a gentilicium (Pellegrini and Prosdocimi, 1967, p. 408).
59 Pellegrini and Prosdocimi, 1967, pp. 198–199; Prosdocimi, 1988, p. 256.
60 Chieco Bianchi and Calzavara Capuis, 2006, pp. 276–294.
61 It is notable that a broken gravestone recovered from a violated grave located nearby, Benvenuti
tomb 120, mentions a man called Frematos, possibly a relative of the people buried in Benvenuti
tomb 123/1879 (Chieco Bianchi and Calzavara Capuis, 2006, pp. 264–265, 294).
62 Another inscription, written by a different person, has been found on the same vessel (Kan).
It may represent a shortened version of Kanta’s name, who perhaps was the first to die, or
a reference to a third dead. Unfortunately, the bones from this vessel have been lost and no
osteological analysis has been possible.
63 It is perhaps noteworthy that at least two people with names deriving from the same onomastic
stem of Fougont- Fr and Vant- Fougontios from Benvenuti tomb 123/1879 were buried in the nearby
tomb of the Titini (Fogotna and Fugsia Titinia). The most ancient epitaph in the tomb of the Titini
is Este inscription 81 (Lemetor Urkleiars), which I mentioned before. The Urkl- onomastic root is
rare at Este and appears in only five inscriptions dating between the fifth century BC and the early
140 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

Roman period. Significantly, second names deriving from this stem were associated with given
names from the F(o)ug- onomastic root in two inscriptions out of five (Este inscription 13: Fougont-
Urkleios; Este inscription 47: Fugia Urkleina. Chieco Bianchi and Calzavara Capuis, 2006, p. 317).
64 Pellegrini and Prosdocimi, 1967, pp. 226–227.
65 Ivanteia may have been a patronymic in –ia-. The two names were inscribed by different
people. The second inscription was added later and partially covers the first text.
66 Pellegrini and Prosdocimi, 1967, pp. 224–226.
67 This is the most probable translation. Another possibility is that the epitaph was given to a first
woman called Frema Boialna Rebetonia and to a second female individual designated with the
–na- adjective only (Votina) (Marinetti, 1992, pp. 150–152, 157–159).
68 Pellegrini and Prosdocimi, 1967, p. 255.
69 Bandelli, 1999, p. 659.
70 Canta’s inscription is damaged. Another possible reading is Canta Paphia C. f(ilia) Eni
uxor (Canta Paphia, daughter of C(aius), wife of En-) (Pellegrini and Prosdocimi, 1967, pp.
261–263). Canta is a Venetic name (Kanta).
71 Pellegrini and Prosdocimi, 1967, pp. 251–253.
72 Prosdocimi (1988, p. 260) suggests that in this case the two may have been sisters, and not a
mother and her daughter (who should have had different gentilicia).
73 This paper partially derives from research carried out during my PhD at UCL. I want to thank
my supervisor, Dr Corinna Riva, for her support over these years. I also thank Ray Laurence
and anyone who commented on an early draft of this work. Finally, I acknowledge the Institute
of Archaeology, UCL, and the Classical Association for providing financial support to join the
‘Oikos/Familia’ Conference in Sweden.

Balista, C. and Ruta Serafini, A. (1998), ‘La necropoli della Casa di Ricovero.
Storia della ricerca’, in Bianchin Citton, E., Gambacurta, G. and Ruta Serafini,
A. (eds), ‘Presso l’Adige Ridente’. Recenti Rivenimenti Archeologici da Este a
Montagnana, Padova: ADLE, pp. 17–28.
Balista, C., Ruta Serafini, A., Drusini, A. and Rippa Bonati, M. (1998), ‘I
resti umani cremati della necropoli Ricovero di Este: metodi di studio e
prospettive’, Quaderni di Archeologia del Veneto, 4, 267–285.
Bandelli, E. (1999), ‘La romanizzazione attraverso l’epigrafia: il Veneto e il
modello etrusco’, in Paoletti, O. ed., Protostoria e Storia del ‘Venetorum
Angulus’. Atti del XX Convegno di Studi Etruschi ed Italici. Portogruaro –
Quarto d’Altino – Este – Adria 16–19 Ottobre 1996, Pisa: Istituti Editoriali e
Poligrafici Internazionali, pp. 652–664.
Bianchin Citton, E., Gambacurta, G. and Ruta Serafini, A. (eds) (1998), ‘Presso l’Adige
Ridente’. Recenti Rivenimenti Archeologici da Este a Montagnana, Padova: ADLE.
Capuis, L. (1992), ‘Este preromana. Vita e cultura’, in Tosi, G. ed., Este Antica. Dalla
Preistoria all’Età Romana, Padova: Zielo, pp. 40–108
—(2009a), I Veneti. Società e Cultura di un Popolo dell’Italia Preromana, Milano:
—(2009b), ‘La romanizzazione del Venetorum Angulus’, in Cuscito, G. ed.,
Aspetti e Problemi della Romanizzazione. Venetia, Histria e Arco Alpino
Orientale, Trieste: Editreg Sas, pp. 179–205.
Fa m i ly R e l at io n sh i p s 141

Chieco Bianchi, A. M. (1985), Le Necropoli di Casa di Ricovero, Casa Muletti

Prosdocimi, Casa Alfonsi, Roma: G. Bretschneider Editore.
—(1987), ‘Dati preliminari su alcune tombe di III secolo da Este’, in Vitali, D.
ed., Celti ed Etruschi nell’Italia Centro-Settentrionale dal V Secolo a.C alla
Romanizzazione. Atti del Colloquio Internazionale. Bologna 12–14 Aprile
1985, Imola: Bologna University Press, pp. 191–236.
—(2006), La Necropoli di Villa Benvenuti, Roma: G. Bretschneider Editore.
Cuscito, G. ed. (2009), Aspetti e Problemi della Romanizzazione. Venetia, Histria
e Arco Alpino Orientale, Trieste: Editreg Sas.
De Marinis, R. C. (1999), ‘Il confine occidentale del mondo proto-veneto
– paleo-veneto dal Bronzo finale alle invasioni galliche del 388 a.C’., in
Paoletti, O. ed., Protostoria e Storia del ‘Venetorum Angulus’. Atti del XX
Convegno di Studi Etruschi ed Italici. Portogruaro – Quarto d’Altino – Este –
Adria 16–19 Ottobre 1996, Pisa: Istituti Editoriali e Poligrafici Internazionali,
pp. 511–564.
Drusini, A., Crivellaro, F. and Carrara, N. (1998), ‘Studio antropologico degli
incinerati’, in Bianchin Citton, E., Gambacurta, G. and Ruta Serafini, A.
(eds), ‘Presso l’Adige Ridente’. Recenti Rivenimenti Archeologici da Este a
Montagnana, Padova: ADLE, pp. 36–47.
—(2001), ‘Studio antropologico dei resti ossei cremati di Colombara (Vr)’,
Padusa, 37, 137–147.
Gamba, M. and Tuzzato, S. (2008), ‘La necropoli di Via Umberto I e l’area
funeraria meridionale di Padova’, in I Veneti Antichi: Novita’ e Aggiornamenti,
Sommacampagna: Cierre Edizioni, pp. 59–77.
Gambacurta, G. and Ruta Serafini, A. (1998), ‘Etre reliés dans la mort: deux
exemples du rituel funéraire de l’Age du Fer de Padoue et d’Este’, European
Journal of Archaeology, 1, (1), 91–113.
Malnati, L. (2002), ‘Monumenti and stele in pietra preromani in Veneto’,
in AKEO. I tempi della Scrittura. Veneti Antichi: Alfabeti e Documenti,
Montebelluna: Museo di Storia Naturale e Archeologia, pp. 127–138
Marinetti, A. (1992), ‘Lingua e epigrafia di Este preromana’, in Tosi, G. ed., Este
Antica. Dalla Preistoria all’Età Romana, Padova: Zielo, pp. 127–172.
—(1996), ‘Iscrizione venetica. Ritrovamento occasionale’, in Malnati, L.,
Croce della Villa, P. and Di Filippo Ballestrazzi, E. (eds), La Protostoria tra
Sile e Tagliamento. Antiche Genti tra Veneto e Friuli, Roma: Esedra Editrice,
pp. 99–100.
—(1999), ‘Gli apporti epigrafici da Altino preromana’, in Tirelli M., and Cresci
Marrone, G. (eds), Vigilia di Romanizzazione. Altino e il Veneto Orientale tra
II e I secolo a.C., Roma: Quasar, pp. 75–95.
—(2001), ‘Il Venetico di Lagole’, in Fogolari, G.and Gambacurta, G. (eds),
Materiali Veneti e Romani del Santuario di Lagole di Calalzo al Museo di Pieve
di Cadore, Roma: Giorgio Bretschneider Editore, pp. 59–73.
—(2008a), ‘Aspetti della romanizzazione linguistica nella Cisalpina orientale’, in
Urso, G. ed., Patria Diversis Gentibus Una? Unità Politica e Identità Etniche
142 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

nell’Italia Antica. Atti del Convegno Internazionale, Cividale del Friuli, 20–22
Settembre 2007, Pisa: ETS, pp. 147–169.
—(2008b), ‘Venetica I’, Quaderni di Archeologia del Veneto, 24, 189–193.
Pellegrini, G. B. and Prosdocimi, A. L. (1967), La Lingua Venetica, Vol. 1,
Padova-Firenze: Zielo.
Prosdocimi, A. L. (1882), ‘Notizie delle necropoli euganee di Este’, Notizie dagli
Scavi, 3–35.
—(1988), La lingua’, in Fogolari, G. and Prosdocimi, A. L. (eds), I Veneti
Antichi: Lingua e Cultura, Padova: Editoriale Programma, pp. 223–440.
Ruta Serafini, A. ed. (1990), La Necropoli Paleoveneta di Via Tiepolo a Padova.
Un Intervento Archeologico nella Città, Este: Zielo.
—(1998), ‘Il rituale funerario: nuovi spunti metodologici’, in Bianchin Citton,
E., Gambacurta, G. and Ruta Serafini, A. (eds), ‘Presso l’Adige Ridente’. Recenti
Rivenimenti Archeologici da Este a Montagnana, Padova: ADLE, pp. 75–98.
—(2002), Este Preromana: Una Città e i Suoi Santuari, Treviso: Canova.
Salzani, L., Drusini, A. and Malnati, L. (2000), ‘Orfeo in Veneto. La tomba 13
della necropoli di Lovara (Villa Bartolomea)’, Quaderni di Archeologia del
Veneto, 16, 138–148.
Tirelli, M. (2001), ‘. . .ut. . .largius rosae et esc[a]e. . .poneretur. I rituali funerari
ad Altinum tra offerte durevoli e deperibili’, in Heinzelmann, M., Ortalli,
J., Fasold, P. and Witteyer, M. (eds), Culto dei Morti e Costumi Funerari
Romani. Roma, Italia Settentrionale e Province Nord-occidentali dalla Tarda
Repubblica all’Età Imperiale. Convegno Internazionale Roma 1–3 Aprile 1998,
Wiesbaden: Dr. Ludwing Reichert Verlag, pp. 243–256.
Tombolani, M. (1987). ‘Materiali tipo La Tene da Altino (Venezia)’, in Vitali,
D. ed., Celti ed Etruschi nell’Italia Centro-Settentrionale dal V Secolo a.C alla
Romanizzazione. Atti del Colloquio Internazionale. Bologna 12–14 Aprile
1985, Imola: Bologna University Press, pp. 171–178.
Whitehouse, R. and Wilkins, J. (2006), ‘Veneti and Etruscans: issues of
language, literacy and learning’, in Herring, R., Lemos, I., Lo Schiavo, E.,
Vagnetti, L., Whitehouse, R. and Wilkins, J. (eds), Across Frontiers: Etruscans,
Greeks, Phoenicians and Cypriots. Studies in Honour of David Ridgway and
Francesca Romana Serra Ridgway, London: Accordia Research Institute,
UCL, pp. 531–548.

Gender, Household Structure and Slavery:

Re-Interpreting the Aristocratic Columbaria of Early
Imperial Rome
Lindsay Penner

The funerary epitaphs of the Early Imperial aristocratic columbaria around

Rome are an invaluable source for the lives of the slaves and freed slaves of
those large households. More specifically, they can reveal how gender roles
influenced economic production and occupational tasks, additional marital,
familial, patronage, or slavery bonds between members of a larger household,
and the overall social structure of the household’s members. Did female slaves
and freed slaves function primarily in a familial role, as wives and mothers to
the labouring male slaves and freed slaves, or did they also possess vital occupa-
tional and even administrative roles within the aristocratic household?
The five columbaria discussed here can all be firmly dated to the late first
century BC and the early first century AD based on references to the owners
of the households mentioned in each columbarium. The Monumentum Liviae,1
the Monumentum Filiorum Drusi, and the Monumentum Marcellae2 all belong
to close relations of the Emperor Augustus (c.27 BC – AD 14). The first was
associated with the household of Augustus’ wife Livia, the second with the
households of Livia’s son Drusus and his immediate family,3 and the third with
the household of Augustus’ niece Marcella Minor, daughter of his sister Octavia.
I have included two additional contemporary columbaria, the Monumentum
Statiliorum,4 and the Monumentum Volusiorum.5 The former was associated
with the household of T. Statilius Taurus (cos. suff. 34 BC, cos. ord. 26 BC) and
his descendants up to his great-granddaughter Statilia Messalina, third wife
of the emperor Nero (c.54 AD – 68 AD), while the latter was associated with
the household of L. Volusius Saturninus (cos. 3 AD) and his three children.
All five columbaria lay along major roads just outside the city of Rome:6 the
Monumentum Liviae and the Monumentum Volusiorum lay along the Via Appia,
the Monumentum Drusi and the Monumentum Marcellae lay between the Via
Appia and the Via Latina, and the Monumentum Statiliorum lay at the inter-
section of the Via Labicana and the Via Praenestina. As all five columbaria are
roughly contemporary, in use from the late first century BC to the middle of the
first century AD, they provide an excellent comparison of large households in
the early Roman Empire.
144 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

The sixth volume of the Corpus Inscriptionum Latinarum (CIL) contains only
inscriptions from Rome, including those inscriptions known to have originated
in the five largest columbaria of the first century AD, as discussed above.7 These
columbaria comprise approximately 1,500 inscriptions, almost all brief epitaphs
no more than a few lines in length. Owing to their brevity, there is little value in
analysing these inscriptions individually; studied en masse, however, they reveal
the commemorative trends, occupations, and social roles of the population of
each columbarium and thus, of the associated households.
However, the inherent difficulty of drawing conclusions from such diverse,
brief epitaphs has frequently led to uncritical usage of epigraphic material.
Within individual epitaphs, we can assume no more information than is actually
provided: a Greek cognomen, for example, is not necessarily an indication of
freed status. Furthermore, in order to make any generalizations about Roman
society or family structure, inscriptions must not be studied on an individual
basis, but as a collection relating to a single theme, and in as large a collection
as is possible or feasible. Finally, owing to the vast number of inscriptions, any
statistical analysis demands careful methodology, thoroughly detailed in the
study itself, in order to understand how the inscriptions themselves were inter-
preted and analysed in order to draw the eventual conclusions.
In order to permit a detailed statistical analysis, I entered each legible
name into a database (Figure 10.1), along with associated demographic and
relationship information when that information was provided, including
gender, legal status, age at death, occupation, and social relationships such as
marriage, children, patronage, and slavery. After I entered all legible names into
the database, I transferred that information into Predictive Analytics Software
(PASW), version 17.0,8 permitting both the calculation of percentages for the
various categories and comparison through statistical analyses. Through the use
of analyses borrowed from the social sciences, such as χ-squared and analysis
of variance (ANOVA), it is possible to determine whether even minute differ-
ences between groups are due to chance alone or whether there is an actual,
statistically significant difference. For example, are there more women in
Marcella’s columbarium than in Livia’s columbarium? The difference between
the two percentages is statistically significant only when the probability of its
being due to random chance – through the preservation of a slightly varying
body of evidence, for example – is less than some predetermined threshold.9
Throughout this paper, any differences mentioned are indeed statistically signif-
icant. One caveat: there is not sufficient time or space to discuss each and every
analysis performed, and thus these results focus on the particularly interesting
or relevant differences.
G e n d e r , Hou se ho l d S t ru c t u r e a n d Sl av e ry 145

Gender in the Households of Imperial Rome

The initial purpose of the present study was to examine gender roles within
large households as reflected in the columbaria, primarily by comparing the
information provided by men and women, and the relationships considered
vital enough to record in an epitaph (Figure 10.2). As with CIL 6 as a whole,
the sample included approximately two-thirds men (67 per cent) and one-third
women (33 per cent). But what about their roles? Treggiari, in her study on the
Volusii, stated that female slaves and freed slaves had only two responsibilities
within large households: marriage and childbearing.10 At first glance, these
results would seem to bear that out: women were indeed more likely to have
spouses and children, while men were more likely to record their occupation, an
administrative role within a collegium, and relationships beyond marriage and
children, to other family members, or to their own patrons, slaves, and freed
slaves. It is safe to say, therefore, that female members of large households were
more frequently commemorated in familial roles than were men, but it does
not necessarily follow that these were their sole functions within the household.
Such a disparity between men and women reveals more about epigraphic habit
than it does about the existence of actual relationships. It reveals instead the
primary role, the role in the household through which commemoration was
expected to occur, while masking the potential for multiple simultaneous roles
within a household structure, for both men and women. Indeed the ‘large’
household seems to have consisted of smaller interconnected social units,
whether these comprised a married couple and children, extended relatives
and other close connections, workers affiliated through similar occupations, or
patrons, slaves, and freed slaves.11
Epitaphs do not always record full legal status, with indications of ownership
for a slave, patronage for a freed slave, or filiation for a freeborn citizen. When
these indications appear, status classification is simple. However, in order to avoid
skewing the results, I remain extremely conservative in assigning status where
there is no explicit indication beyond the name itself. Those with a tria nomina are
free, but it is impossible to determine whether they are freeborn or manumitted,
making them ‘uncertain free’. Although those providing only a single cognomen
are probably slaves, free citizens do occasionally omit their nomina owing to space
limitations or financial constraints, and thus those with only a cognomen comprise
the ‘single name’ group.12 For example, freed slaves of well-known historical
figures frequently omit their tria nomina, providing only the name of their patron,
as their nomen can be inferred from a statement such as ‘Liviae libertus’.13 Overall,
the individuals in the columbaria are largely free (60 per cent), whether freeborn
or manumitted (Figure 10.3), revealing that freed slaves and even their freeborn
children retained vital roles within large households, and in fact constituting a
larger proportion of the household population than did slaves (no more than
40 per cent). Affiliation with a particular household, even a large one with only
minimal contact with the owner, did not end with manumission.
146 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

However, there are considerable status differences between men and women
(Figure 10.4). Women are significantly over-represented in the higher status
groups, while men are over-represented in the lower status groups: more
women are freeborn, uncertain free, or freed, while more men are Imperial
freedmen or slaves, slaves, or provide only a single name. This is probably
another artefact of the epigraphic habit.14 Overall, women are less likely to be
commemorated in an epitaph, as illustrated by the skewed sex ratio in CIL 6,
whereas the Roman population’s sex ratio was likely closer to equal. As a result,
women who do receive an epitaph are more likely self-selected from higher
status groups, where financial and social factors permit and encourage the
expense of an epitaph. Furthermore, gender differences among Imperial slaves
and freed slaves reflect the composition of the Imperial household, as Weaver
and Boulvert have demonstrated.15 The columbaria in question date almost
exclusively to the Julio-Claudian dynasty, and without a formalized civil service,
much of the necessary administrative work was done within the household,
jobs which were filled by men, not by women.

Differing Households, Differing Experiences of

Although I anticipated finding major differences between the genders, I
also expected that those differences would remain consistent across the five
columbaria, but the results showed otherwise. Upon closer inspection, the
columbaria in question did not have uniform populations with identical or
even similar demographic characteristics. Household structure, occupations,
relationships, and even epigraphic habits were by no means uniform across the
five columbaria.
Even something as simple as the sex ratio within the household varied
between the columbaria (Figure 10.5). Statistically, there were more males in
the households of Livia (70 per cent male, 30 per cent female) and the Statilii
(69 per cent male, 31 per cent female), but more females in the households of
Marcella (63 per cent male, 37 per cent female) and the Volusii (59 per cent
male, 41 per cent female), as compared to the entire sample (67 per cent male,
33 per cent female). The reason for the difference is unclear, but combined
with the other differences within households which I will outline in the rest
of this paper, it suggests that there are major differences in overall household
structure and commemorative habits between the five columbaria, a prelim-
inary indication that even contemporary households of comparable size and
status varied enormously, depending on economic specialities and emphasis on
status and relationships.
Status differences between the five columbaria reflect the obvious differences
in their households of origin, as Imperial freed slave or Imperial slave status
G e n d e r , Hou se ho l d S t ru c t u r e a n d Sl av e ry 147

requires current or former ownership by a member of the Imperial dynasty.

The three columbaria closely affiliated with the Julio-Claudian dynasty, which
belonged to Augustus’ wife Livia, her son Drusus, and Augustus’ niece Marcella,
contained more Imperial freed slaves and slaves than did the columbaria of
the Statilii and the Volusii (Figure 10.6). Beyond that obvious difference, it is
notable that Marcella’s household included considerably more free individuals
(76 per cent), especially the freeborn (2 per cent), while the household of the
Statilii had a greater number of slaves (20 per cent) and single names (41 per
cent). This is primarily the result of differences in epigraphic habit between the
two households, although those differences do reveal a great deal about the
internal culture of their respective households. Marcella’s household strongly
encouraged continued participation in the household’s social network after
manumission and into successive generations, whereas the internal culture of
the household of the Statilii focused on slaves, who had no connections beyond
those bonds formed within the household itself. The Statilii certainly had
numerous freed slaves and freeborn children of freed slaves, just as Marcella
did, but they seem to have largely chosen commemoration elsewhere, in
separate single grave tombs or their own household tombs, rather than within
the household’s columbarium, although they would have had access should they
so desire. As a result, members of Marcella’s household retained a stronger affili-
ation with their original household following manumission and even into the
next generation, choosing to continue their affiliation with her household past
death rather than present themselves as entirely separated from it.
Tracking individual occupations is a virtually impossible task owing to the
huge variation in job titles and the vast number of specialized occupations.
However, even highly specialized occupations fall into obvious categories of
similar occupations: medici and obstetrices are clearly both medical professions,
while pedisequi and cubicularii are both personal attendants of differing sorts.
The number of categories remains large (Figure 10.7), but such grouping permits
the application of statistical analyses. In the sample as a whole, from all five
columbaria, the distribution is logical. The largest categories represent occupa-
tions that, by their very nature, demand large numbers of workers. It takes
numerous seamstresses to produce and mend clothing for the entire household
(11 per cent), while attendants, especially pedisequi, are most valuable to the
aristocracy when displayed in large numbers (20 per cent). Conversely, highly
specialized occupations that would have required considerable training and
even apprenticeship, such as medicine (3 per cent) and finance (4 per cent),
are represented by much smaller contingents, also illustrating their relative
necessity. It is unlikely that any household, regardless of its size, would require
more than one or two doctors at any given time.16 Even massive households
with populations numbering thousands would not have needed numerous
workers fulfilling the same, highly specialized and training-intensive tasks.
However, although it may seem obvious that large households would
require a similar combination of tasks in order to function independently, job
148 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

distribution varied considerably between the five columbaria (Figure 10.8),

indicating that there was in fact no fixed proportion of household occupa-
tions. Beyond the occupations necessary for the functioning of an aristocratic
household – cooks, accountants, doormen, and so on – even large households
tended to specialize in particular occupations. Both Drusus (34 per cent) and
the Statilii (26 per cent) had large numbers of attendants, especially bodyguards
such as custodes and Germani, within their households, and the Statilii also
had a considerable number of tradesmen (13 per cent), such as carpenters and
gardeners, who could not have had enough work to occupy them through the
estates of the Statilii alone, and were likely involved with the amphitheatre built
and administered by the Statilii.17 Livia’s household had more seamstresses (12
per cent), providing additional support for Augustus’ claim that he only wore
clothes made by the women of his household,18 or in this case simply within his
wife’s household. Marcella’s household contained numerous medical workers
(12 per cent) and library staff (9 per cent),19 while the Volusii specialized in
childcare workers (11 per cent) and entertainers (11 per cent). Large house-
holds must have specialized in this way for economic purposes: note that
the occupations in question provide products and services that can easily be
sold to individuals and households beyond the immediate household. Rather
than attempting to exist as a self-sustaining entity, able to function completely
independent from other households, these Roman aristocratic households seem
to have specialized in particular fields, enabling them to invest more financial
and material resources and more labour into improving their products and
services in one or two areas, while renting or purchasing in other fields where it
was appropriate, or borrowing from the households of close family members.20
With regard to female workers in particular, the proportion of women
reporting job titles is consistent with Joshel’s study of CIL 6 as a whole (14
per cent),21 with women in the columbaria no more or less likely to record an
occupation (13 per cent). The distribution of female workers is concentrated in
certain expected areas (Figure 10.9), namely clothing (33 per cent), childcare
(14 per cent), and appearance (14 per cent), although women were certainly not
limited to those areas, as we might expect, working in medicine (3 per cent),
administration (3 per cent), and literature (3 per cent) as well. Overall, women’s
listed occupations tend to fall in the centre of the occupational hierarchy of
the Roman household, with men holding the highest posts in administration
and finances and the lowest posts in attendance and generalized household
staff. Although certain occupations were clearly male, with no female workers
dealing with food, money, or trade, female workers were not limited to tradi-
tionally ‘feminine’ occupations. In addition, they are particularly subject to
a self-selection bias. The decision of whether or not to include a job title in
an epitaph must certainly reflect a sense of pride in one’s occupation and
that occupation’s importance to one’s identity: all information included in
an epitaph, particularly brief columbaria inscriptions such as those forming
the basis for the present study, ought to be considered highly relevant and
G e n d e r , Hou se ho l d S t ru c t u r e a n d Sl av e ry 149

important to those commemorating the deceased. Although the lack of data has
skewed any results we can obtain now about the population as a whole, owing
to differences in commemorative practices and habits for men and women, the
very nature of the Roman household demanded more male workers than female
workers. Flory has convincingly argued that the majority of women within a
large household would have had husbands or permanent partners, whereas only
higher-ranking men would have been able or permitted to obtain a spouse.22
As a result, a woman may have worked in a low-status occupation, such as
scrubbing the floors, but her epitaph commemorates her in her role as wife
and mother, leaving no trace of women in low-status occupations. By contrast,
the low-status male slave, still unmarried and with no bond to the household
beyond his occupation, records that job title on his epitaph. Information on
female workers, as a result, has been skewed by the material itself, as males are
more likely to be commemorated within their occupational roles, while females
are more likely to be commemorated within their social roles.
Beyond occupational tasks, some members of these households chose or were
chosen to take on additional tasks related to the administration of the colum-
baria as officers of the household’s collegium. These are easy to identify, as they
provide titles such as decurio, immunis, and magister.23 The formalized structure
of a collegium was not strictly necessary in order to administer a columbarium,
as Drusus’ columbarium contains no record of such an organization, although
the other four columbaria do include officers of such organizations. These were
primarily male (87 per cent), although unlike the highest administrative and
financial posts of the household itself, not exclusively male. Female members of
a household were certainly capable of becoming officers of a collegium in their
own right. Officers were more likely to be Imperial freed slaves (10 per cent) or
Imperial slaves (13 per cent), most likely owing to the large collegium running
Livia’s columbarium. They record their job titles more frequently (34 per cent),
generally occupations of higher status within the household’s internal hierarchy,
such as household management (19 per cent) or administrative positions (27
per cent), rather than those of lower status, such as childcare (2 per cent) or
entertainment (2 per cent). Officers were no more or less likely to record marital
(21 per cent) or other familial relationships (9 per cent), but they were less likely
to record children (1 per cent) or patronage and slavery relationships (14 per
cent) than non-officers.24 Overall, it is clear that collegial officers held a higher
social position within their households’ social networks: regardless of legal
status, higher social position within a household tends to be correlated with
higher-status occupation and thus with an increased likelihood of marriage,
especially given the sex ratio of the Roman household.
The columbaria are probably most different in terms of their commemo-
rative habits, the tendencies of their associated inscriptions to include certain
information in their epitaphs, ranging from age at death to job to a variety of
familial and patronage relationships (Figure 10.10), while frequently omitting
much other information that we might wish to have about the dead. There is
150 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

a clear distinction between households such as Livia’s and that of the Statilii,
where there is an obvious emphasis on occupation and on relationships created
through slavery and patronage, and households such as Marcella’s and that
of the Volusii, where smaller units based around children and other familial
relationships were more prominently commemorated. The household’s culture
of slave-owning would have affected how people chose to commemorate both
themselves and others, from what relationships they emphasized to what infor-
mation – if any – they included in their epitaphs, thereby creating individualized
household identities as depicted in their variable commemorative patterns.
Both Livia’s household and that of the Statilii included more men, which
resulted in a greater focus on occupation, with job titles more frequently given,
and a tendency to develop smaller familial units comprised Imperial freed slaves
with their own slaves and freed slaves in turn, rather than through marriage and
childrearing. By contrast, Marcella’s household and that of the Volusii contained
more women, and were accompanied by an increased focus on children and
marital familial units. Interestingly, the household of the Volusii tended to
include within its structure smaller, almost nuclear familial units based around
married couples, whereas marital relationships were less frequent in Marcella’s
household. The reason for this difference is unclear. However, it is possible
that Marcella’s connections to the extended Julio-Claudian family altered the
distribution of familial units. Familial units within the Julio-Claudian house-
holds may have been spread across separate households (for example, shared
between Marcella and her mother Octavia) rather than being contained within
a single household, as they were in the household of the Volusii. However, this
is no more than conjecture. The household of Drusus is by far the smallest of
the five, which perhaps explains the absence of a collegium to administer the
columbarium. Like the household of the Volusii, commemorative habits in
Drusus’ household led its members to include more information, such as age
at death and occupation, in their epitaphs, rather than providing a name alone
with no other information, which is a fairly common practice in columbaria
inscriptions. The differences in commemorative focus between these colum-
baria, ranging from marriage to patronage to occupation, suggests that large
Roman households were not necessarily identical and that each individual
household’s internal culture and sense of identity and community affected its
members’ lives as well as their deaths. These elements would suggest that the
culture of a household created quite different social structures for slaves and
freed slaves, as well as experiences of slavery that varied considerably between
large households.
G e n d e r , Hou se ho l d S t ru c t u r e a n d Sl av e ry 151

Implications for Understanding the Roman Family

and Slavery
Based on these analyses, it becomes clear that the Roman household cannot and
should not be understood as an unchanging entity, with every example beyond
the obvious exception of the emperor’s own household being fairly similar.
Household size, structure, Imperial connections, and economic specializa-
tions influenced the composition of the household in terms of gender, status,
and necessary occupations. Furthermore, as demonstrated by the unique
characteristics of the households using these five columbaria, epigraphic habit,
family and slave relationships, and gender roles also varied from household
to household. Simply put, a comparison of these five columbaria reveals that
Roman households, even in the same time period and within an extended
family, had distinctive identities and differing cultures of slave-owning that had
a strong impact on how both men and women functioned as spouses, parents,
workers, and leaders within the aristocratic household.
More broadly, these results have significant implications for our under-
standing of slavery in the early Roman Empire, with regard to both slaves
and slave owners. Experiences of slavery varied greatly from household to
household, and were by no means as uniform as they might initially appear.
In some cases, slaves were encouraged to marry among themselves and have
children, while in others the emphasis was more on occupation. In some
households, freed slaves retained a stronger affiliation after manumission,
choosing burial within that household structure rather than outside it. Even
sex ratios varied by household, with some households having significantly more
or less women than the expected average. In addition, slave owners seem to
have developed their household staff to specialize in particular areas, trading
and borrowing workers from other households when their services became
necessary. Each household becomes, as it were, an individual unit with its
unique hierarchy, specific abilities and needs, and even its own unique culture,
different than the household next door. Pliny the Younger claims that servis res
publica quaedam et quasi civitas domus est,25 which is indeed an apt description
of the Early Imperial household.
152 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d


Yes No

ML 3926-4326 MLD 4327-4413 MM 4418-4880 MS 6213-6640 MV 7218-7394

Name Age Yes No

Gender Male Female
Status Uncertain free Slave Job Yes No
Freeborn Imperial slave Category
Freed Single name
Imperial need Unknown Collegium Yes No Family Yes No

Role Commemorated Other Marriage Yes No Slavery Yes No

Dedicator Unknown Children Yes No Name Yes No
Number Order Duplicate Yes No

Copy record Next record

Figure 10.1 Database

13 13
33.9 27.4 27.6 33.4
49.9 51.9

87 87
72.6 72.4
66.1 66.6
50.1 48.1

Job Age Collegium Marriage Children Other Patronage Whole

family or slavery sample
Figure 10.2 Characteristics by gender
G e n d e r , Hou se ho l d S t ru c t u r e a n d Sl av e ry 153

Singe name

Uncertain free

Imperial slave



Imperial freed
Figure 10.3 Status distribution

22.9 27.5 27

76.1 73
59.7 55.6

Uncertain Imperial Imperial Single Whole

Freed Slave
free freed slave name sample
Figure 10.4 Status distribution by gender
154 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

29.7 30.1 30.6 33.4


70.3 69.9 66.6
62.9 59

Livia Drusus Marcella Statilii Volusii All

Figure 10.5  Gender distribution among the columbaria

17.1 18.8 18.5
6.9 0
13.1 2.7 7.3 Single name
4.4 Imperial slave
29.3 10.8
19 0
0.3 Slave
14.6 8.8 5.1
1.9 2 Imperial freed
4.5 20.2
6 Freed
0.3 1 Freeborn
0.2 11.3

0 15.5 Uncertain free

52.3 51.7
36.2 38.8

Livia Drusus Marcella Statilii Volusii All

Figure 10.6 Status distribution among the columbaria
G e n d e r , Hou se ho l d S t ru c t u r e a n d Sl av e ry 155





Food 5%
Religion 2% Money
2% Medicine 4%
Figure 10.7 Occupational categories
156 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d







Attendance Administration Household Clothing Trade

Appearance Entertainment Childcare Artisan Money

Medicine Literature Religion Food Military

Figure 10.8  Job distribution among the columbaria





3% Artisan

Medicine 3%
3% Entertainment
Figure 10.9  Female occupational categories
G e n d e r , Hou se ho l d S t ru c t u r e a n d Sl av e ry 157

Family 10.8

Children 8.5

Marriage 16.4

Patronage/slavery 15.9

Job 30.6

Collegium 4
0 14.6

Age 17.4

0 5 10 15 20 25 30 35 40 45

All Volusii Statilii Marcella Drusus Livia

Figure 10.10 Overall differences among the columbaria

 1 CIL 6.21415. The Monumentum Liviae was excavated in 1726, and little remains today besides
the inscriptions themselves (Treggiari, 1975a, pp. 48–49).
 2 AE 1996, 00253. Both the Monumentum Marcellae and the Monumentum filiorum Drusi are no
longer extant; they were originally part of a columbarium complex between the Via Appia and
the Via Latina known since before the fifteenth century (Hasegawa, 2005, p. 22).
  3 Drusus had three children: Germanicus, Livilla, and the future emperor Claudius.
 4 CIL 6.06213. The Monumentum Statilii was first excavated in 1875 (Hasegawa, 2005, pp. 8–9).
 5 The Monumentum Volusiorum was first excavated in 1825, although no actual structure
remains (Hasegawa, 2005, p. 19).
158 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

  6 Ibid., pp. 5–6.

 7 Livia: CIL 6.03926 – CIL 6.04326; Drusus: CIL 6.04327 – CIL 6.04413; Marcella: CIL 6.04418 –
CIL 6.04880; Statilii: CIL 6.06213 – CIL 6.06640; Volusii: CIL 6.07281 – CIL 6.07394.
  8 This statistical analysis software was formerly known as the Statistical Package for the Social
  9 As is common practice, I have used a threshold, or p-value, of 0.05, or 5%.
10 Treggiari, 1975b, p. 395.
11 Flory, 1978, passim.
12 Cf. Joshel, 1992, p. 39.
13 For example, CIL 6.03927, CIL 6.04053, CIL 6.04250, CIL 6.04352, and CIL 6.04390. There are
numerous other examples, especially among Livia’s freed slaves buried within her columbarium.
14 Cf. Flory, 1978, p. 88.
15 Weaver, 1972; Boulvert, 1974.
16 These medical workers would have treated the free and slave members of the household rather
than the aristocratic owners.
17 Cass. Dio. 51.23.1. In fact, this amphitheatre provides an excellent example of the potential
for multiple job categories to be involved in a single project: the maintenance of an amphi-
theatre would demand numerous workers, including but not limited to tradesmen, custodians,
managers, and artisans.
18 Suet. Aug. 73.
19 These library staff may in fact be linked to the libraries within the Porticus Octaviae, erected by
Augustus sometime after the death of Marcellus, Marcella’s brother, in 27 BC on behalf of his
sister Octavia.
20 Cf. Treggiari, 1975a, pp. 54–57.
21 Joshel, 1992, p. 16.
22 Flory, 1978, p. 87.
23 Cf. Nielsen, 2006, p. 205.
24 See Figure 10 for these numbers for the whole sample.
25 Plin. Ep. 8.16.3.

Boulvert, G. (1974), Domestique et Fonctionnaire sous le Haut-Empire Romain,
Paris: Les Belles Lettres.
Flory, M. B. (1978), ‘Family in familia: kinship and community in slavery’,
American Journal of Ancient History, 3, 78–95.
Hasegawa, K. (2005), The Familia Urbana during the Early Empire: A Study of
Columbaria Inscriptions, (PhD diss), Nottingham University.
Joshel, S. (1992), Work, Identity, and Legal Status at Rome: A Study of the
Occupational Inscriptions, Norman: University of Oklahoma Press.
Nielsen, H. S. (2006), ‘Collegia: A new way for understanding the Roman
family’, Hephaistos, 24, 201–214.
Treggiari, S (1975a), ‘Jobs in the household of Livia’, Papers of the British School
at Rome, 43, 48–77.
—(1975b), ‘Family Life among the Staff of the Volusii’, TAPA, 105, 393–401.
Weaver, P. R. C. (1972), Familia Caesaris: A Social Study of the Emperor’s
Freedmen and Slaves, Cambridge: Cambridge University Press.

Houses, Painting, Family Emotion

Natalie Kampen

After some years of living in a house in which there were wall-paintings in
many rooms, did the Roman owners of the House of the Menander in Pompeii
begin to take the pictures for granted? Did they stroll by without noticing them?
After how many dinner parties did the guests, having reclined before the same
paintings in this same dining room so many times, cease to find anything witty
or philosophical to say about them? My paper is not, actually, about boredom
but about how paintings may stay the same while the emotions of viewers
change. It is in some ways a very art historical essay and not directly about the
image of familia or oikos. But in other ways it seems to me to add something
to the collective toolbox, because it suggests some ways to think about intimate
emotions, family emotions, in a physical as well as a historical context. The
paper is about how people experience and control deep emotions with the
aid and complicity of the objects around them. We normally think about this
kind of emotional experience in relation to the tomb and use tombstones
and epitaphs, sarcophagi and funerary statues as our evidence. I suggest here
that the house and its decoration can be used too, as they have already been
used in relation to desire and eroticism.1 However, this approach only works
if we combine the data from words: consolations, epic and poetry, letters and
epitaphs, with the experience, physical and psychic as well as aesthetic, of the
images. Moreover, it includes not just the framed pictures of Greek myths so
often the focus of Roman art history, but the background panels, dados and
pretend mouldings that accompany and comment on them.2 The experience of
the room, as well as of the decorative ensembles within its walls, frames, shapes
and helps to control emotion, as I hope this case study will show.
Along with thinking about the ways in which pictures construct viewers or
viewers read pictures, this paper explores the question of how the two processes
intersect differently depending on the emotional state of the individual viewer
at a given moment. One could, of course, extend this to the shifting population
of viewers living in and visiting the house, at different stages in their lives,
in different conditions of servitude or autonomy, and with differing experi-
ences to bring to the pictures. Rather than discussing the images as if there
160 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

were one ideal or typical viewer or as if the viewer always felt the same way
about life, or death, the paper tries to imagine him/her/them as having fluid
and volatile emotions as well as experiencing images from inside complicated
personalities. It will put Virgil’s Aeneas, weeping before the paintings of Troy,
together with the tragic father Priam as he appears on Roman wall-paintings;
the case study here concerns those who have, as was so common in the Roman
world, lost one or more people they loved and the ways in which the condition
of mourning might affect their readings of pictures.3 Further, the paper asserts
that the pictures and their rooms participate actively in constructing the viewer
or viewers as particular kinds of mourners. It tries to imagine within Roman
terms how a viewer passing by the paintings in a Pompeian house may have,
quite suddenly, experienced the sort of terrible pain of loss that struck Aeneas
as he came before Dido’s Carthaginian temple.4 The paper examines the way
emotion works in process rather than as product, and it focuses on paintings
from the House of the Menander connected with Laocoön and Cassandra and
with Priam, King of Troy, whose tragic loss of wife and children, home, nation,
life itself, is both one of the great stories of Greek epic and foundational for the
story of Rome.5
Let us begin with a famous scene of emotional change, as Virgil presents it
in the Aeneid, written late in the first century BC. The hero Aeneas has left Troy
after the Greeks destroyed the city; he has lost his homeland and many dear to
him, and, buffeted by the winds of angry gods, finds himself shipwrecked on the
shores of Carthage. He approaches the temple of Juno.

Here in this grove a strange sight met his eyes and calmed his fears for the
first time. Here, for the first time,
Aeneas dared to hope he had found some haven,
for all his hard straits, to trust in better days.6

His state of mind, then, is calm and reassured after the storm. But as he walks
around the temple, ‘amazed’ by the splendour of its decoration, he comes upon
the scenes of ‘the battles fought at Troy’ and reacts by ‘feeding his spirit on
empty, lifeless pictures’ which causes him to groan, ‘the tears rivering down his
face’.7 His emotions have changed quite suddenly, and he falls into the pictures
and his memories and associations as he stands ‘spellbound, eyes fixed on the
war alone’.8 He has moved from his present place and time, from quietly looking
at a temple, into a time of memory and pain, an elsewhere that the pictures force
upon him. It does not matter here that Aeneas has been taken in by a mere
image; instead, what is important is the abruptness with which his emotions
change when he realizes what he is looking at. The power of the paintings to
change his feelings, instantly, without his bidding, and his inability to control
his response is what matters. Virgil wants us to understand how powerful
pictures can be, despite their sophistry, and how vulnerable to that power even
the strongest man can be.
Hou se s , Pa i n t i n g , Fa m i ly E m o t io n 161

Many men of the Roman elite struggled against tears and grief, although
perhaps a tragedy of the proportions that struck Priam and Troy would have
been as appropriate a cause for tears to them as Cicero claimed the crisis of the
Republic was to him. Great men wept over great public tragedies. But the more
domestic tragedies, the deaths of those one loved, might well have caused such
men far greater struggle since the rules of class and gender precluded protracted
grief and mourning. Presumably they managed, manfully, to repress their
sorrow and move on as they were supposed to do (although not everyone did,
as our sources will reveal), but what might the effect on them or on their wives
have been of the daily sight of images of tragic familiar sorrows?
Here is the first-century AD rhetorician Quintilian grieving over the loss
of his son, ‘of whom I had such expectations and in whom I rested the sole
hope of my old age’.9 Quintilian lost his wife when she was only 19 and then
his younger son at the age of 5 and finally his second and most beloved son. He
swears to the excellence of that son, ‘by my own troubles, by the misery that my
heart knows, by those spirits of the departed who are the gods of my grief ’ and
knows that he lives only to suffer as he remembers the way ‘this delightful child
preferred me to his nurses, preferred me to the grandmother who brought him
up’.10 He has ‘reason to be grateful for the grief I suffered, a few months before,
in the death of his excellent mother, who indeed was beyond all praise’.11 His
sorrows are profound, but presumably, like so many men of the upper ranks, he
steeled himself, returned to work and to public life, even if he stayed away from
it for what his friends felt was far longer than was appropriate.12 Seneca’s advice
to a friend was clearly what one heard repeatedly. In the Epistulae Morales, he
insists that his correspondent is behaving ‘like a woman in the way you take
your son’s death . . . A son, a little child of unknown promise, is dead; a fragment
of time has been lost . . .’.13 He goes on: ‘There are countless cases of men who
have without tears buried sons in the prime of manhood – men who have
returned from the funeral pyre . . . and have straightaway busied themselves
with something else’.14 Cicero knows this and so does his correspondent Brutus,
writing to each other in the middle of the first century BC, but even in the face
of what Amanda Wilcox has characterized as the ‘stern and even contentious
rhetoric of Roman consolation’,15 men and women suffered and grieved in a
world where people died young and too often a person lost family members one
after another. How could the mourner then endure facing images of tragic loss
and suffering while walking through his or her own house? How could one bear
to dine before a picture of an untimely death?
The argument proposed here is speculative, since it depends on a modern
reading of Pompeian walls and builds on the evidence of texts that express
sentiments concerning deaths in the family. Other readings are possible.16 The
point, however, is that images that might normally be passed by and ignored, or
discussed and reacted to as sites for literary or philosophical or erotic or even
joking conversation or thought, might also become, under the right conditions,
sites for the experience of and resistance to grief.17 There is, one could argue, a
162 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

complicated dance going on between people trained to contain their grief and
images that manage emotions through their formal properties. What I mean by
managing emotions through their formal properties will emerge shortly.

Pictures and Emotion in the House of the Menander

Just off the atrium of the splendid house of the Menander in Pompeii is a small
room with red walls and pictures taken from the story of the Fall of Troy.18 The
paintings date to the second or third quarter of the first century AD, and were
in place at the least for 10 or 15 years before the eruption of Mount Vesuvius;
they may even have been visible for as much as 50 years.19 The lower wall of
this highly visible space has a black dado and the upper wall a white ground for
delicate architectural motifs. Between them the red panels flanking the Trojan
pictures are enframed with fine golden borders and enlivened by small floating
figures (Figure 11.1). Contrasting with the emphatic flatness of the red panels,
the framed Trojan pictures seem not so much hung on walls as placed before
openings with architectural devices and garlands that invent a space behind.
The room, close to the front door of the house, is open to the atrium and thus
accessible, at least visually, to clients, strangers, and guests as well as to members
of the family. This was the part of the house most open to the public, and it
shared many of its visual motifs and overall compositions with the paintings in
the atrium and tablinum as well.
Three framed images with a coordinated program take their positions at
the centre of each wall in ala 4. On the back wall is the scene of the wooden
horse with Cassandra rushing past and gazing at it in terror. Hidden from our
and the Trojans’ eyes are the Greeks who, against Cassandra’s advice, will be
brought into Troy inside the horse. Priam, her father, will lose his realm and
his family, and the prophetess already knows this (Figure 11.2)20. The left wall
shows Cassandra beset by Ajax while Helen is manhandled by her aggrieved
husband; between the two stands the helpless Priam (Figure 11.3). The Greeks
have won, and Ajax tears Cassandra from the protection of the goddess’ image
in order to violate and enslave her just as the Greek Menander recaptures his
errant wife whose face launched those thousand ships. And on the right wall
Laocoön and his sons are dying their horrific deaths. Punished by the gods, as
is Cassandra, for making prophecies of doom for Troy against the wishes of the
gods, Laocoön must not only suffer his own excruciating death but must watch
his innocent sons die by the same serpents (Figure 11.4).
Clearly, one aspect of the viewer’s experience will have been the recognition
of the fall of Troy and of its relationship both to Homer and to Virgil, as well as
its literary resonance. The pictures functioned as well to raise questions about
the fate of those who offer prophecies and of those defeated by the gods as
well as by their enemies in war. Not only do the three scenes act in harmony,
they operate as pairs too. Cassandra appears in the rape and the scene of the
Hou se s , Pa i n t i n g , Fa m i ly E m o t io n 163

Figure 11.1 View into Ala 4 facing the image of the Wooden Horse, House of the
Menander, Pompeii (© Scala/Art Resource New York)

wooden horse, prophets suffer their dreadful fates in the rape and in the death
of Laocoön, and the rape and Laocoön also raise questions about the place of
fathers in the Trojan cycle. Both Priam and Laocoön are responsible, on some
level, for the tragedy visited on their children, and their helplessness is a central
theme in the facing images. In one, the father is dying as his sons lie lifeless near
him, while in the second, the father stands, passive. On one side his daughter is
defiled and on the other his daughter-in-law who is also the adulterous cause of
war and death is treated to the violence of a wronged husband.
164 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

Figure 11.2 Detail of the Wooden Horse (© Peter Stewart)

Figure 11.3 Detail of Priam between Helen and Menelaus and Cassandra and Ajax
(© Peter Stewart)
Hou se s , Pa i n t i n g , Fa m i ly E m o t io n 165

Figure 11.4 Detail of the death of Laocoön and his sons (© Peter Stewart)

The Power of Fate and the Powerlessness of Parents

The modern viewer might expect somewhere in the repertoire of Roman
painting to find fathers helping or protecting their children, but in fact only
Aeneas, carrying his father and leading his son, can be seen in that role.21
Elsewhere in Pompeii, Agamemnon covers his face as his daughter is taken
away to be sacrificed in order to allow the Greek ships to sail to Troy, while
in other houses Admetus’ parents look on passively as the messenger explains
that someone may die in place of their son; they do nothing.22 And nowhere in
Pompeian painting is Priam himself ever an active force (Figure 11.5). In some
of the pictures he sits with his children and grandchildren, a poignant image
of family before the storm; he listens passively to Cassandra’s prophecies or he
watches inert as around him everyone suffers and dies.23
Fathers are, in short, almost never represented in Pompeian painting as
other than passive before the gods and fate or actively engaged in causing
harm to their children. An odd choice for a society whose father image, the
pater familias, included both supportive concern for wives, and children and
autocratic bullying. This was, after all, the man to whom early Roman law
gave enormous power over the members of his household, but he was also, as
demographic evidence strongly suggests, not likely to live to see his children
into adulthood. The reality of the pater familias may well have been different
from the ideological construction, but neither one permits us to imagine that
166 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

Figure 11.5  The Prophecy of Cassandra with Priam looking on, from Casa della Grata
Metallica (Marie-Lan Nguyen, Wikimedia Commons)

the paintings are reflecting attitudes and practices going on in the house they
decorate. Indeed, we could imagine at one moment that the painted passive
fathers reveal the kind of ambivalence toward the powerful father ideology that
appears in Roman comedy, while at another that these helpless men stand as
contrasts to normative fatherhood among the Roman occupants of the house.
Both are likely to be true, since the social ambivalence can be documented
Hou se s , Pa i n t i n g , Fa m i ly E m o t io n 167

just as can the displacement of that ambivalence from the Roman father onto
the father in myth or the fathers of conquered Greece. That the fathers in the
paintings are Homeric rather than Roman, mythic rather than contemporary,
permits them to become a site for multiple readings, and the changing emotions
of viewers can find purchase there for even the most contradictory feelings
about fathers and family relations.
Nowhere in the imagery of these wall-paintings is the kind of interven-
tionism of modern family relations visible. This is equally true for fathers and
for mothers, some of whom endanger or kill their children (Medea of course),
some of whom are helpless or passive (Hercules’ mother as the baby is attacked
by snakes), and only one or two of whom are saved by the gods along with their
children (Danae and Perseus).24 For modern westerners, willingly to do nothing
in the face of a child’s danger is almost inconceivable in an era when our
metaphors include battling disease and conquering epidemics or counselling
our sorrows into submission. By contrast, in ancient myths intervention is often
fatal, as the popular paintings of the fall of Icarus show.25 The fate of Priam’s
daughter and Laocoön’s sons comes, then, as no surprise, although it may well
have brought tears to the eyes of their viewer, whose loved one had died.
This is not to suggest, however, that a weeping parent would necessarily feel
guilt at the death of a child nor read Priam and Laocoön in the paintings and
stories as blameworthy by their role in the fates of their children . . . In some
sense, death was always already a fait accompli given the nature of Greek tragedy
and the Homeric and Virgilian epics. It was, after all, from those grand sources
that so much of the subject matter on the Pompeian walls came and which
formed so much of the cultural repertoire for imagery and its interpretation.
Further, there is no indication either in the consolation literature or in epitaphs
that parents believed they could intervene to change a child’s fate. Mentions of
‘cruel fate’ and loss are present, but laments for failed responsibility or inter-
vention are extremely rare. For example, although there are many inscriptions
that speak of the fates or singular fate as jealous, cruel, and wilful, cutting short
a life or snatching someone away, especially someone whose death is untimely,
there are few cases of parents being called crudelis or inpia.26 In these cases,
we must be seeing guilt about not being good enough to a child or not having
been sufficiently observant of the needs of the gods, but these do not imply the
parent’s ability to save the child.
The image of the parent who either endangers a child or fails to protect it
points to a belief in the impossibility of effective intervention as much as to
the conventional Roman belief in fate. Thus in Pompeian houses all over town,
Daedalus repeatedly watches his son plunge to his death, and the wall paintings
remove any sense of suspense. We know what must happen.27 The old debates,
following on the contention of Philippe Ariès and others about emotional
connections between parents and young children, seem to have missed this
element of intervention.28 The argument that parents in times of high infant
mortality did not invest much emotion in their young children, or that it was
168 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

only with lowered mortality rates that people began to develop affective depth,
whether in practice or language, neglects the reality not only of death rates but
of the lack of possibility for intervention and its consequences in lack of articu-
lated guilt, in the popularity of stoicism, and in the wide-spread belief in fate
and the power of the gods to determine the future.

Public Decorum, Time, and Memories of the Dead

Despite the lack of ability to intervene and to change the workings of Fate,
people mourned the loss of their loved ones as deeply as Aeneas mourned the
loss of Troy. The evidence of elite literary production as well as of non-elite
epitaphs, formulaic though they may be, articulates the pain. But the pain
abates gradually, and as time passes, it is experienced in different ways, as Pliny
the Younger explains in his letter to a friend about the death of Fundanus’
beloved daughter. Pliny, however, tells us something more about the pain and
its life-span when he speaks of Fundanus’ loss. He says:

[Fundanus] is indeed a cultivated man and a philosopher who has devoted himself
from youth to higher thought and the arts, but at the moment he rejects everything
he has so often heard and professed himself: he has cast off all his other virtues and is
wholly absorbed by his love for his child . . . If then you write anything to him in his
very natural sorrow, be careful not to offer any crude form of consolation which might
suggest reproof; be gentle and sympathetic. Passage of time will make him readier to
accept this: a raw wound shrinks from a healing hand but later permits and even seeks
help, and so the mind rejects and repels any consolation in its first pangs of grief, then
feels the need of comfort and is calmed if this is kindly offered.29

Not only time but actions themselves can soothe the pain, as Cicero hopes to do
in building a memorial shrine for his adult daughter Tullia on the grounds of a
suburban hortus or garden.30

As far as practicable in this age of culture, I shall naturally hallow her with every kind
of memorial which Greek and Latin genius can supply. Perhaps that will gall my wound.
But I consider myself as virtually bound by a vow and pledge, and the long expanse of
time after I shall cease to be is of more account to me than this little span, which yet
seems to me too long. I have tried everything and found no comfort.31

How should a father who suffers so acutely then view the helpless Priam or
the tormented Laocoön as he walks through his house? How should a mother
face such images? Will they feel as Pliny suggests people do, when in a letter
to Macrinus, he says, ‘We seek consolation in sorrow in the busts of our dead
we set up in our homes . . .’.32 An aristocratic privilege, but one might take
equal comfort from the portrait of a deceased relative even in a Pompeian
Hou se s , Pa i n t i n g , Fa m i ly E m o t io n 169

house of a freedman’s descendant.The question, then, is whether the comfort

comes as quickly as consolation literature demands. One suspects that the
insistence that mourning be brief describes a wish for public decorum
(perhaps both for purposes of class solidarity and recognizability and also to
avoid opening the wounds of one’s fellows) rather than the denial of private
emotions. It is these emotions that even elite men acknowledge when they
speak of themselves, although they seldom speak as Pliny does of others in
quite so sympathetic a way.
How much do consolation literature and letters describing loss and sorrow
really say about sorrow rather than about a decorum which is always external,
a matter of behaviour meant to shape feeling rather than to index it?33 Their
descriptions of people suffering and grieving seem vivid, even as the prescrip-
tions for behaviour remain rigidly conventional and classed. The consolations
may show gendered behaviour, since women seem to have been given a little
more leeway in emotion than men, but the prescriptions reveal that class
trumps gender when it comes to advice. Seneca’s famous consolation to Marcia,
indicating that she was still devastated by the death of her son after three years,
proposes that she too has the class obligation to control her grief. He woos her
self-control both by compliments, telling her that she is ‘as far removed from
womanish weakness of mind as from all other vices’ (1.1) and remarking, ‘That
correctness of character and self-restraint which you have maintained all your
life, you will exhibit in this matter also; for there is such a thing as moderation
even in grieving’ (3.4).34 Despite all efforts to help her recover, Seneca says that
Marcia still suffers as acutely as ever, and he suggests that she consider the two
aristocratic models of female mourning and choose between them. ‘Through
all the rest of her life Octavia (having lost her son Marcellus) set no bounds to
her tears and moans, and closed her ears to all words that offered wholesome
advice’ (2.4), whereas Livia suffered terribly on the loss of Drusus, but

as soon as she had placed him in the tomb . . . she laid away her sorrow, and grieved no
more than was respectful to Caesar or fair to Tiberius, seeing, that they were alive. And
lastly, she never ceased from proclaiming the name of her dear Drusus. She had him
pictured everywhere, in private and in public places, and it was her greatest pleasure to
talk about him and to listen to the talk of others – she lived with his memory (3.2).35

In both cases, the image of the one lost too soon is part of the discussion, and
although Octavia refuses any image of her son, Livia cherishes images of Drusus
and this acts as the sign of recovery as well as of appropriate class conduct.

Voicing Emotions: from Epitaphs to Paintings

What of non-elite mourning? Almost all the evidence is in the form of epitaphs,
most of which are terse and thoroughly decorous.36 They give a name and an age,
170 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

perhaps an adjective or a relationship. Far more unusual than the word or two
of affection are the verse epitaphs and the prose commemorations (some made
for non-elites whereas others were made for the likes of Regilla, commemorated
in a long poem by Marcellus of Side at her villa on the Via Appia). These often
speak explicitly of sorrow and detail the qualities of the deceased.37 The debates
about the reliability of epitaphs as evidence for emotions are well-known, but
I want to use the inscriptions to indicate a kind of horizon of expectation, to
echo Jauss.38 We cannot know who was sincere and who was not, nor can we
know how much of a clichéd language of mourning revealed emotions and
how much concealed. But as Keith Hopkins and others have proposed, there
is surely room for thinking about the epitaphs as indicative of typical ways to
give voice to what was both regularly expected and sometimes experienced.39
In the first century AD, the parents of an 11-year-old boy from Benevenum call
themselves most unhappy and miserable and say that their loss has condemned
them to daily weeping40 while a father says that he will always search for his
darling daughter, the 9-year-old Asiatica. ‘Sadly shall I often imagine your
face to comfort myself.’41 Some of these people had houses with pictures on
the walls, but if Pompeii and Rome (or Magdalensburg and Ephesus) are good
examples, they were more likely to have to imagine the faces of the dead than to
have their images in the house. Pompeian painting gives very little evidence of
portraiture, although perhaps lost panel paintings once offered comfort to the
bereaved.42 Unlike the sarcophagi that came into use in the second century AD
and represented some of the same myths that we find on the walls of Pompeii,
the painters never seem to insert portraits in place of the faces of Priam,
Cassandra or the other protagonists.43 The myths in the house keep the dead
abstract rather than, as the sarcophagi do, blurring the line between the actual
and the mythic dead. Surely one would have an even harder time controlling
one’s emotions if the daily confrontation with tragic stories insisted so heavily
on their direct connection with one’s own life. It must have been quite different
to go annually to the tomb to look upon Alcestis dying for her beloved husband
Admetus when both wore the faces of a real husband and wife. What emerges
here is that most people grieved for their dead in one way or another and
that decorum could not always contain and control the pain. The works of art
participate actively in allowing for the maintenance of decorum either by the
way they are physically situated in a room or a tomb or the way they represent
the myths that remind or warn one of loss.

Re-Reading Roman Pictures via Familial Emotions

The notion that grief is experienced in a variety of ways and with varying
intensities, even for young children, suggests that individual viewer responses
changed with time even if the pictures themselves remained the same and even
if they dealt through tragic myth and epic with loss and suffering. Readings of
Hou se s , Pa i n t i n g , Fa m i ly E m o t io n 171

mythological paintings have seldom taken account of this kind of temporality,

and have tended to fall into two categories, neither of which accounts adequately
for the kinds of fluid and even volatile states of feeling that I am proposing as
the experience of many people in an era of high mortality rates. Apologizing
for the oversimplification, it appears that an older form of scholarly interpre-
tation searched the mythological paintings for meanings that were unitary and
timeless.44 Whether finding a coherent iconographic meaning in one painting
or in programmes within a room, these readings operated on the premise that
the original intent of artist and patron at the time of production determined
future experience and interpretation. These were essentially static readings,
as when we interpret ala 4 of the House of the Menander simply as about the
Trojan war’s end or about the interaction of the pendant pictures. More recent
approaches, seen especially in the work of Bettina Bergmann and Katharina
Lorenz, propose both the need to imagine the experience of the viewer as
she moves through the house or moves around a room and also the need to
acknowledge that certain pictures and groupings encourage the experience
of time.45 Thus the relationship of ala 4 to the atrium with its shrine and wall
paintings, perhaps with a stemma of the family on the wall, perhaps with its
crowd of visitors hanging about, becomes visible. So too does the changing
character of the atrium and the ala, since the passing of the day brings to the
spaces different light and shadow with its different occupants and activities.
Paintings themselves can initiate the experience of time, as when we under-
stand the sequence of events at the end of the Trojan war. Thus Bettina Bergmann
describes the way paintings instruct in and evoke memory while at the same
time some pictures and groups may depend for their meanings on a sense of
temporality. In describing pictures in the house of Jason where heroines such
as Medea, Phaedra and Helen, are in the act of making desperate decisions,
Bergmann allows us to experience the moment coming alive as the room evokes
an empathic experience of time itself.46 I think she is right about this, but I see the
possibilities for a still more active engagement with the time of viewing. Similarly,
when Timothy O’Sullivan describes the experience of an educated viewer walking
past the Odyssey frieze in the house on the Esquiline in Rome and points to the
relationship of the viewer’s own physical movement with the changing scenes
of the paintings, he takes us further into a sense of time.47 But neither of these
studies takes us far enough into the perplexing nature of viewing enacted by a
mutable individual, one who lives with ever-changing feelings. The multiple and
passing moments of reception for the people who live in and visit a house shift
their experience of the images, the rooms, the house itself.
To incorporate into the analysis not only the mythological pictures but the
effect of the rest of the room and of the surrounding rooms in the house is to
experience more fully the temporal and emotional context. At the House of the
Menander, the wall panels in red, the illusions of open space in white, and the
black dado, in short, the majority of the room’s surfaces, remind us of the ability
of colour and line to counteract through visual pleasure the pain of the myths
172 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

in their relatively contained spaces. So too, the garlands and delicate floating
figures with their celebratory and idyllic quality restrain the harshness of the
stories and restore emotional balance. Only when the room is experienced
in multiple time frames does all this become clear. The walls are seen by the
moving viewer and the scanning eye, by the viewer throughout the day and
throughout the years, as his or her life changes.
Not all paintings evoke powerful feelings of loss or grief, and in fact the issue
is less what the painting is doing than what the viewer is doing with it. Let us
end with a rather silly text about viewing and emotions in order to make this
point clear. In the Satyricon of Petronius, written perhaps for the entertainment
of the court of Nero in the mid-first century AD, there is a section where the
narrator, Encolpius, having been abandoned by his lover, finds himself in an art
gallery looking at famous old paintings.

I cried out as if I were in a desert, among these faces of mere painted lovers, ‘So even
the gods feel love. Jupiter in his heavenly home could find no object for his passion,
and came down on earth to sin, yet did no one any harm. The Nymph who ravished
Hylas would have restrained her passion had she believed that Hercules would come
to dispute her claim. Apollo recalled the ghost of a boy into a flower, and all the stories
tell of love’s embraces without a rival. But I have taken for my comrade a friend more
cruel than Lycurgus himself.’48

For Encolpius, all the pictures are about his own situation, his own misery, but
then he is joined by Eumolpos, who presents himself as an art expert and whose
chatter causes Encolpius’ attention to shift away from himself until Eumolpos
notices that his companion’s attention is riveted on a picture representing the
capture of Troy. Within minutes, just as Aeneas’ emotions and attention shifted,
so do those of the shallow but charming Encolpius. His attention will shift once
more when Eumolpus attracts the attention of other visitors to his commentary,
and they respond by throwing stones at him.49
The protean nature of viewing, and of emotions, comes through in this text
that is as far from epic and consolation as the first century can go. Virgil and
Petronius allow us to understand the experience of viewing both in time and in
space, as Encolpius shifts his gaze and Aeneas rounds the corner. We too come
to understand the way that images in context can play a role in managing as
well as manipulating emotions. The painted rooms of Pompeii are interacting
with their viewer so that one may find different levels of affective intensity
depending on one’s state of mind, but at the same time the painted rooms are
also performing acts of containment for that intensity. Seeing in time as well
as in space becomes a key to understanding both Roman visuality and the
experience of emotions.
Hou se s , Pa i n t i n g , Fa m i ly E m o t io n 173

  1 For example, Elsner, 2007, pp. 132–176; Clarke, 1998, passim, or Valladares, 2005, pp. 206–242.
There has been an explosion of interest especially in the erotics of viewing in relation to Roman
painting, but this has only just begun to raise phenomenological issues of the physical position
of the viewer (the most important exception to this, as will be obvious from the citations below,
is Bergmann).
 2 On the critical importance of the decorative ensembles in Roman houses, see especially
Bergmann, 2002, pp. 15–45.
  3 Saller has provided useful models for the potential loss of family members of various ages in the
Roman household; see Saller, 1994, passim but esp. 12–70. For more recent debate on the use
of demographic data for the study of Roman society, see as well Scheidel, 2001, pp. esp. 1–82.
 4 Aeneid, Book 1, 543–598.
  5 On the House of the Menander, Ling and Ling, 2005, pp. 41–46 and 73–75. The versions of the
stories in the ala are to be found in Book 2 of the Aeneid rather than in the Homeric texts, but
the pictures differ from the texts in a number of details such as the combination of Helen and
Menelaus with Cassandra’s attack by a single figure rather than a group.
 6 Aeneid, Book 1.450–452: Hoc primum in luco nova res oblata timorem leniit, hic primum
Aeneas sperare salutem ausus, et adflictis melius confidere rebus (translated by Fagles, 2006,
Book 1.543–547, p. 62). On this passage and important comparanda that reveal the extent to
which Roman writers understood that pictures from epic and tragedy evoked strong emotions
in a variety of viewers, see Bergmann, 1994, pp. 225–256, but especially 248–249 where Porcia
weeps before a painting showing the separation of Hector and Andromache as she grieves
over her own separation from Brutus. The ideas in this essay obviously owe much to those of
Bergmann’s paper on the House of the Tragic Poet.
 7 Aeneid, Book 1.464–65. Sic ait, atque animum pictura pascit inani, multa gemens, largoque
umectat flumine voltum (translated by Fagles, 2006, Book 1.562–64, pp. 62–63).
 8 Aeneid, Book 1.494–95: Haec dum Dardanio Aeneae miranda videntur, dum stupet, obtutuque
haeret defixus in uno. . .(translated by Fagles, 2006, Book 1.596–97, p. 64).
 9 6.1–14, 9–15: Illum enim de quo summa conceperam, et in quo spem unicam senectutis
reponebam. . . (translated by Russell, 2009, pp. 9–15).
10 6.10: Iuro per mala mea, per infelicem conscientiam, per illos manes, numina mei doloris. . .; 6.
8–9: ut ille mihi blandissimus me suis nutricibus, me aviae educanti, me omnibus qui sollicitare
illas aetates solent anteferret.
11 6. 9: Quapropter illi dolori quem ex matre optima atque omnem laudem supergressa paucos
ante menses ceperam gratulor.
12 6.14.
13 99.2–3, 130–131. Molliter tu fers mortem filii; quid faceres, si amicum perdidisses?
Decessit filius incertae spei, parvulus; pusillum temporis periit (translated by Gummere,
1925, pp. 130–131).
14 99.6, 132–133. Innumerabilia sunt exempla eorum, qui liberos iuvenes sine lacrimis extulerint,
qui in senatum aut in aliquod publicum officium a rogo redierint et statim aliud egerint (trans-
lated by Gummere, 1925, pp. 132–133).
15 Wilcox, 2006, p. 238.
16 For example, Baltussen, 2009, pp. 355–369, esp. 359, 361, and 363.
17 See for example, Bergmann, 1994, and, on emotional identification, Zanker, 1998, pp. 40–48,
esp. 46–47.
18 Most recently on the House of the Menander: Ling and Ling, 2003, as well as the catalogue of
the 2003 exhibition at the Antiquarium at Boscoreale, Stefani, 2003.
19 Santoro, 2005, pp. 97–123, esp. 111–112; and Ling and Ling, 2005, pp. 41–46.
20 For a listing of scenes concerned with Cassandra in Pompeian painting, see Hødske, 2007, pp.
212–216; for Laocoön, 256.
21 Book 9.13.5.
22 Agamemnon: Pompeii Regio VI.8.3.5: House of the Tragic Poet peristyle 10: MNN 9112, PPM
IV, 1993, pp. 552–553, nos 47–48; Alcestis and Admetus: VII.12.28: Casa del Balcone Pensile,
174 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

PPM VII, 1997, p. 594; V.1.18: House of the Epigrams peristyle, PPM III, 1991, pp. 550–552,
nos 22–23; VI.8.3.5: House of the Tragic Poet: MNN 9026, PPM IV, 1993, pp. 546–547, no. 36;
VI.Is. Occ.: Podere di Irace: MNN 9025; Ercolano unknown house: MNN 9027.
23 Priam: VI.3.7: ‘Accademia di Musica’ triclinium 14, PPM II, 1990, p. 277 no. 56; I.2.28: Casa
della grata metallica triclinium I: MNN 111.476, PPM II, 1990, p. 277 no. 56; VI.10.2: Casa dei
Cinque Scheletri triclinium: MNN 8999, PPM IV, 1993, p. 1040 no. 18.
24 Medea: IX.5.18: House of Jason, cubiculum 23: MNN 114321, PPM IX, 1999, p. 687, no. 23;
VI.9.6–7: House of the Dioscuri peristyle 53: MNN 8977, PPM IV, 1993, p. 975, no. 223;
Ercolano unknown house: MNN 8976; IX.8.3–6: Casa del Centenario atrium 2: PPM IX, 1999,
p. 912, no. 14. Zeus with Herculiscus, Alcmene and Amphitryon: VII.3.10 oecus I, PPM III,
1991, p. 103; Ercolano unknown: MNN 9012; VI.15.1: House of the Vettii triclinium n, PPM V,
1994, p. 527, no. 105. Danae and Perseus: VII.2.44–46: Casa dell’Orso Ferito triclinium c, PPM
VI, 1996, p. 762; III.1: MNN 9552; V.1.18: House of the Epigrams exedra O: MNN 111212, PPM
III, 1991, p. 557, no. 37; IX.2.23 (destroyed), PPM IX, 1999, p. 97.
25 V.2.10, cubiculum Q, PPM III, 1991, p. 839, no. 19; I.7.7: House of the Priest Amandus,
triclinium b, PPM I, 1991, pp. 394–395, no. 10–11; I.10.7: Casa del Fabbro, triclinium; I.9.5:
Casa dei Cubiculi floreali, aka Casa del Frutteto, triclinium 11; V.5.3: Caserma dei Gladiatori,
peristyle, PPM III, 1991, p. 1072, no. 6; Unknown house: MNN 9506; Unknown house:
London, British Museum P28.
26 For these themes, see Lattimore, 1942, passim, esp. pp. 142–158, and 183–199, as well as, more
recently, the work of Christian Laes, such as Laes, 2004, pp. 43–75, especially page 53, note
31 on inscriptions addressing unjust Fate as ‘invidia fata’; or Nielsen, 1997, pp. 169–204, with
examples 198–202. Cf. CIL VI 2.9783, p. 1113: inpia morte perit; VI.4.24659, p. 2486: mors
acerba immature abstulit; or VI.4.25063, p. 2517: Heu crudelis nimis fatum . . .
27 Daedalus and the fall of Icarus: Von Blanckenhagen, 1968, pp. 106–143.
28 Aries, 1962; Stone, 1977, 405–480 and 651–652. For a still useful discussion of the literature,
see Golden, 1988, pp. 152–163. On the use of inscriptions in the debate, see King, 2000, pp.
29 V.16.8–11: Est quidem ille eruditus et sapiens, ut qui se ab ineunte aetate altioribus studiis
artibusque dediderit; sed nunc omnia, quae audiit saepe quae dixit, aspernatur expulsisque
virtutibus aliis pietatis est totus (translated by Radice). Proinde si quas ad eum de dolore tam
iusto litteras mittes, memento adhibere solacium non quasi castigatorium et nimis forte, sed
molle et humanum. Quod ut facilius admittat, multum faciet medii temporis saptium. Ut enim
crudum adhuc vulnus medentium manus reformidat, deinde patitur atque ultro requirit, sic
recens animi dolor consolationes reicit ac refugit, mox desiderat et clementer admotis adqui-
escit (translated by Radice Pliny, Letters and Panegyricus, 1969, pp. 380–381).
30 Baltussen, 2009; and see also Zevi, 2004, pp. 15–31. Cf. Erskine, 1997, pp. 36–47. For the trans-
lation as well as a discussion of the problems connected with the memorial, see Shackleton
Bailey, 1966, pp. 404–413. His edition of the text may be found on pages 90–91: Ego, quantum
his temporibus tam eruditis fieri potuerit, profecto illam consecrabo omni genere monimen-
torum ab omnium ingeniis suptorum et Graecorum et Latinorum. Quae res forsitan sit
refricatura vulnus meum. Sed iam quasi voto quodam et promisso me teneri puto, longumque
illud tempus cum no ero magis me movet quam hoc exiguum, quod mihi tamen nimium
longum videtur. Habeo enim nihil temptatis rebus omnibus in quo acquiescam.
31 Cicero XII.18.1 (translated by Shackleton Bailey, 1966, pp. 90–91).
32 II.7.6–7,100–101: Etenim si defunctorum imagines domi positae dolorem nostrum levant,
quanto magis hae quibus in celeberrimo loco non modo species et vultus illorum, sed
honor etiam et gloria refertur! (translated by Radice Pliny, Letters and Panegyricus, 1969,
pp. 100–101.
33 Consolation literature: Kassel, 1958; Baltussen, 2011; Graver, 2002, and Wilcox, 1996,
pp. 267–287.
34 Moral Essays, 1.1, 2–3 and 14–15: Nisi te, Marcia, scirem tam longe ab infirmitate muliebris
animi quam a ceteris uitiis recessisse. . .; 3.4: Si ad hoc maximae feminae te exemplum adpli-
cueris moderatius, mitius, non eris in aerumnis nec te tormentis macerabis (translated by
Basore, 1932, pp. 2–3 and 14–15).
Hou se s , Pa i n t i n g , Fa m i ly E m o t io n 175

35 2.4, 10–11 and 12–15: Nullum finem per omne uitae suae tempus flendi gemendique fecit
nec ullas admisit uoces salutare aliquid adferentis, ne auocari quidem se passa est; and 3.2: ut
primum tamen intulit tumulo, simul et illum et dolorem suum posuit, nec plus doluit quam
aut honestum erat Caesare aut aequum saluo. Non desiit denique Drusi sui celebrare nomen,
ubique illum sibi priuatim publiceque repraesentare, libentissime de illo loqui, de illo audire:
cum memoria illius uixit, quam nemo potest retinere et frequentare qui illam tristem sibi
reddidit (translated by Basore, 1932, pp. 10–11 and 12–15).
36 Lattimore, 1942.
37 Laes, 2004.
38 Debates on the interpretation of the emotional content of epitaphs: King, 2000, Hopkins, 1983,
pp. 220–226. Jauss, 1982, esp. chap. 1.
39 Hopkins, 1983, pp. 220–226.
40 CIL IX.1973
41 CIL II.3771
42 But see Thompson, 1979, pp. 78–92 and Anderson, 1987, pp. 127–135, esp. 130.
43 Zanker, 2004, passim, but see especially the useful discussion of identification through
portraiture in mythological imagery, pp. 45–50. Another perspective on portraits of the dead
may be found in Petrucci, 1998.
44 An important representative of this form of interpretation, immensely useful at the time,
can be seen in Thompson, 1960–61, pp. 36–77. See also Brilliant, 1984, for an exploration of
interpretive strategies over a variety of media. For comments on the state of the study of Iliac
themes, see Santoro, 2005, pp. 104–106.
45 Bergmann, 1994; Lorenz, 2008 and Clarke, 1991. Earlier precedents include Drerup, 1959, pp.
46 Bergmann, 1996.
47 O’Sullivan, 2006, pp. 133–152.
48 83. 164–67 and 167–79: Inter quos etiam pictorum amantium vultus tanquam in solitudine
exclamavi: ‘Ergo amor etiam deos tangit. Iuppiter in caelo suo non invenit quod diligeret,
sed peccaturus in terris nemini tamen iniuriam fecit. Hylan Nympha praedata temperasset
amori suo, si venturum ad interdictum Herculem credidisset. Apollo pueri umbram revocavit
in florem, et omnes fabulae quoque sine aemulo habuerunt complexus. At ego in societatem
recepi hospitem Lycurgo crudeliorem (translated by Heseltine, 1919, pp. 164–67).
49 83–90 (translated by Heseltine, 1919, pp. 167–79).

Ancona, R. and Greene, E. (eds) (2005), Gendered Dynamics in Latin Love
Poetry, Baltimore: Johns Hopkins University Press.
Anderson, C. (2002), The Built Surface, Vol. 1: Architecture and the Pictorial Arts from
Antiquity to the Enlightenment, Aldershot, Hants, and Burlington, VT: Ashgate.
Anderson, M. L. (1987), ‘Portrait medallions of the Imperial Villa at Boscotrecase’,
AJA, 91, (1), 127–135.
Aries, P. (1962), Centuries of Childhood: A Social History of Family Life, New
York: Knopf.
Baltussen, H. (2009), ‘A grief observed: Cicero on remembering Tullia’, Mortality,
14, (4), 355–369.
—(2011), Acts of Consolation. Approaches to Loss and Sorrow from Sophocles to
Shakespeare. Papers from the 2007 London Colloquium, Institute of Classical
Studies, Senate House, Cambridge: Cambridge University Press.
176 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

Bergmann, B. (1994) ‘The Roman house as memory theatre’, Art Bulletin, 76,
(2), 225–256.
—(2002), ‘Playing with boundaries: painted architecture in Roman interiors’, in
Anderson, C. (2002), The Built Surface, Vol. 1: Architecture and the pictorial
arts from Antiquity to the Enlightenment, Aldershot, Hants, and Burlington,
VT: Ashgate.
Brilliant, R. (1984), Visual Narratives: Storytelling in Etruscan and Roman Art,
Ithaca, NY: Cornell University Press.
Burzacchini, G. (eds) (2005), Troia tra Realtà e Leggenda, Parma: Monte
Università Parma.
Clarke, J. R. (1991), Houses of Roman Italy, 100 BC–AD 250: Ritual, Space and
Decoration, Berkeley: University of California Press.
—(1998), Looking at Lovemaking: Constructions of Sexuality in Roman Art 100
BC – AD 250, Berkeley: University of California Press.
Docter, R. F. and Moorman, E. M. (eds) (1999), Proceedings of the XVth
International Congress of Classical Archaeology, Amsterdam, July 12–17,
1998: Classical Archaeology towards the Third Millennium: Reflections and
Perspectives, Amsterdam: Allard Pierson Series.
Drerup, H. (1959), ‘Bildraum und Realraum in der römischen Architektur’,
RömMitt, 66, 145–174.
Elsner, J. (2007), Roman Eyes, Visuality and Subjectivity in Art and Text,
Princeton, NJ: Princeton University Press.
Erskine, A. (1997), ‘Cicero and the expression of grief ’, in Braund, S. and Gill, C.
(eds), The Passions in Roman Thought and Literature, Cambridge: Cambridge
University Press.
Golden, M. (1988), ‘Did the Ancients care when their children died?’, Greece
and Rome, 35, (2), 152–163.
Graver, M. (2002), Cicero on the Emotions: Tusculan Disputations 3 and 4,
(translation with commentary), Chicago: University of Chicago Press.
Gummere, R. M. (1925), Seneca Ad Lucilium Epistulae Morales, Vol. 3, London:
Heinemann, and New York: G. P. Putnam’s Sons.
Heseltine, M. (1919), Petronius, London: William Heinemann, and New York:
G. P. Putnam’s Sons.
Hopkins, K. (1983), Death and Renewal, Cambridge: Cambridge University
Jauss, H.-R. (1982), Toward an Aesthetic of Reception, Minneapolis: University
of Minnesota Press.
Kassel, R. (1958), Untersuchungen zur griechischen und lateinischen consolations-
literatur, Munich: Beck.
King, M. (2000), ‘Commemoration of infants on Roman funerary inscriptions’,
in Oliver, G. J. ed., The Epigraphy of Death: Studies in the History and Society
of Greece and Rome, Liverpool: Liverpool University Press.
Laes, C. (2004), ‘High hopes, bitter grief: Children and their virtues in Latin
literary inscription’, in Partoens, G., Roskam, G. and Van Houdt, T. Virtutis
Hou se s , Pa i n t i n g , Fa m i ly E m o t io n 177

Imago: Studies on the Conceptualization and Transformation of an Ancient

Ideal, Louvain: Editions Peters/Société des Etudes Classiques.
Lattimore, R. (1942), Themes in Greek and Latin Epitaphs, Urbana: University
of Illinois Press.
Ling, R. and Ling, L. (2005), Insula of the Menander at Pompeii, Vol. 2: The
Decorations, New York: Oxford University Press.
Lorenz, K. (2008), Bilder machen Räume, Berlin: De Gruyter.
Nielsen, H. S. (1997), ‘Interpreting epithets in Roman epitaphs’, in Rawson, B.
and Weaver, P. (eds), The Roman Family in Italy: Status, Sentiment, Space,
Oxford: Clarendon Press.
O’Sullivan, T. (2006), ‘The mind in motion. Walking and metaphorical travel in
the Roman villa’, CP, 101, 133–152.
Partoens, G., Roskam, G. and Van Houdt, T. (eds) (2002), Virtutis Imago:
Studies on the Conceptualization and Transformation of an Ancient Ideal,
Louvain: Editions Peters/Société des Etudes Classiques.
Petrucci, A. (1998), Writing the Dead: Death and Writing Strategies in the
Western Tradition, Stanford, CA: Stanford University Press.
Pliny, Letters and Panegyricus, Vol. 1, (translated by Radice, B., 1969), Cambridge,
MA: Harvard University Press, and London: William Heinemann.
Rawson, B. and Weaver, P. (eds) (1997), The Roman Family in Italy: Status,
Sentiment, Space, Oxford: Clarendon Press.
Russell, D. A. (2001), Quintilian: The Orator’s Education, Cambridge, MA:
Harvard University Press.
Saller, R. (1994), Patriarchy, Property and Death in the Roman Family, Cambridge
and New York: Cambridge University Press.
Santoro, S. (2005), ‘I temi iliaci nella pittura pompeiana’, in Burzacchini, G. ed.,
Troia tra Realtà e Leggenda, Parma: Monte Università Parma.
Scheidel, W. (2001), Debating Roman Demography, Leiden and Boston:
Seneca, Moral Essays, Vol. 2, (translated by Basore, J. W., 1932), London:
William Heinemann, and New York: G. P. Putnam’s Sons.
Shackleton Bailey, D. R. ed. (1966), Cicero’s Letters to Atticus, Cambridge:
Cambridge University Press.
Stefani, G. (2003), Menander: La Casa del Menandro di Pompei, Milan: Electa.
Stone, L. (1977), The Family, Sex, and Marriage in England, 1500–1800, New
York: Harper and Row.
Thompson, D. L. (1979), ‘Painted portraiture at Pompeii’, in Pompeii and the
Vesuvian Landscape, Washington, D. C.: Archaeological Institute of America
and Smithsonian Institution.
Thompson, M. L. (1960–61), ‘The monumental and literary evidence for
programmatic painting in antiquity’, Marsyas, 9, 36–77.
Valladares, H. (2005), ‘The Lover as Model Viewer: Gendered Dynamics in
Propertius I.3’, in Ancona, R. and Greene, E. (eds), Gendered Dynamics in
Latin Love Poetry, Baltimore: Johns Hopkins University Press.
178 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

Virgil, The Aeneid, (translated by Fagles, R., 2006), New York: Viking Press.
Von Blanckenhagen, P. H. (1968), ‘Daedalus and Icarus on Pompeian Walls’,
RömMitt, 75, 106–143.
Wilcox, A. (1996), ‘Paternal grief and the public eye: Cicero Ad Familiares 4.6’,
Phoenix, 59, (3/4), 267–287.
Zanker, P. (1998), ‘Mythenbilder im haus’, in Docter, R. F. and Moorman, E.
M. (eds) (1999), Proceedings of the XVth International Congress of Classical
Archaeology, Amsterdam, July 12–17, 1998: Classical Archaeology towards the
Third Millennium: Reflections and Perspectives, Amsterdam: Allard Pierson
—(2004), Mit Mythen leben: Die Bilderwelt der römischen Sarkophage Munich:
Zevi, F. (2004), ‘Cicero and Ostia’, Journal of Roman Archaeology, Supplement
57, 15–31.

Afterword: The Future of the Ancient Greek Family

Mark Golden

My focus on the Greek family in particular has three justifications. First, I know
more about it. Second, the conference which inspired these remarks was held in
Sweden, home of the world’s most successful source of household furnishings.
The name of that firm – IKEA – bears an uncanny resemblance to oikia, a
common Greek word for the household and those connected to it. Could this
be mere coincidence? I think not, and I am sure Henning Mankell would agree.1
Finally, the study of the Greek family seems to me to offer special challenges but
also the chance to do new and exciting work – which maybe even work which
students of the Roman family would find worth their while for a change.2
Some 30 years ago, when Richard Saller arrived in Cambridge, he told Moses
Finley (then Professor of Ancient History) that he intended to work on the
Roman family. ‘Why?’ asked Finley. ‘The lawyers have already written every-
thing that needs to be said.’ In the decades since, Saller and others (including
Keith Bradley, Suzanne Dixon, Judith Evans-Grubbs, Beryl Rawson, Brent
Shaw, Hanne Sigismund-Nielsen) have transformed our understanding of the
Roman family and of the European family as a whole. Of special importance,
they have asserted the primacy of the nuclear triad, the emotional core of father,
mother and children which was once thought to have developed only after
Antiquity. This new paradigm has not gone uncontested – this is still an active
area of research. For example, Bradley has stressed the vulnerability of the
Roman nuclear family to death and divorce and the role played by slave nurses
and minders in the lives of its children.3 Pointing to the many epitaphs with no
identified dedicator, Sigismund-Nielsen suggests that for many of the urban
working class, mostly freed slaves, it was their workmates, organized informally
or in collegia, who filled the familial role.4 These ‘choice families’, as Sigismund-
Nielsen calls them, were what Ann-Cathrin Harders terms a ‘framily’. Dale
Martin even questions methodology developed by Saller and Shaw in their
ground-breaking study of funerary inscriptions from the Roman west: how
should we count joint or multiple commemorations on the tombstones on
which the argument for the predominance of the nuclear family is based?5
Jonathan Edmondson has in fact used the method both of Saller and Shaw and of
Martin to analyze two epigraphic samples from Lusitania and reached virtually
identical results, both pointing to the predominance of the nuclear family.6 This
is not to say that extended kin and others did not share the household with
it and play a significant role in the lives of its members. Nor is it to discount
180 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

variation. Our best evidence for household composition, from Greco-Roman

Egypt, reveals a range of household types, from individuals living alone to
conjugal, extended and multiple family dwellings, and there are differences (as
Saskia Hin reminds us in this volume) between urban areas and the countryside
too. Variation is also striking in Martin’s examination of inscriptions from Asia
Minor and the Levant, with the proportion of those outside the immediate
family as commemorators rising from one in four in Bithynia to almost three
times as many at Olympus in Lycia.7 In addition, siblings commemorate and
cousins marry more often in the east than in the western provinces. In general,
as Edmondson observes, the Greek east is characterized by a ‘broader emphasis
on maintaining links and emotional ties with more distant kin’.8
Let us turn to the Greeks then. What have students of the Greek family been
doing since Saller and the rest began their Roman revolution in family studies
(and reminded us that we should always seek a second opinion about lawyers’
advice)? When Daniel Ogden’s Greek Bastardy appeared in 1996, I reviewed it
and noted some of the other fine English-language work related to the Greek
family which had been published in the previous few years: Lene Rubinstein’s
book on Athenian adoption, Barry Strauss’ on fathers and sons at Athens,
Nancy Demand’s on birth and motherhood, the volume by Oakley and Sinos
on weddings, Patricia Watson’s work on Greek and Roman stepmothers, my
own on Athenian childhood.9 The time was right, I thought, for a new synthesis
to replace Lacey’s useful but inevitably outdated The Family in Classical Greece,
published in 1968 and a surprising survivor from that tumultuous year. If only
my views had as much influence on my own family! Soon enough there were
two such syntheses, very different, equally excellent books by Sarah Pomeroy
and Cynthia Patterson.10 I have reviewed these elsewhere and will not do so
now.11 I am more concerned with one of their unanticipated consequences: the
virtual disappearance of the Greek family.
Already in the 1970s French scholars had dispelled the old image of multi-
family groups, phratries and gene, controlling Archaic Greek communities and
standing in opposition to the new institutions and social forms of the territorial
state.12 Patterson surveyed the nineteenth-century sources for this mirage –
Bachofen, Fustel de Coulanges, Maine, Morgan, Engels – and summarized the
prevailing view: ‘polis and family developed in a relationship of creative and
productive partnership’.13 In fact, Athens’ phratries may stem (like so much else)
from Cleisthenes and most of her gene may be one hundred years younger and
The nuclear family was ready for its close-up. Instead, scholarship focused on
the household: the third edition of the Oxford Classical Dictionary went so far
as to omit the topic altogether, directing those interested in ‘family, Greek’ to
an entry on ‘household, Greek’. We still await Greek equivalents for The Roman
Mother, Brothers of Romulus, Fathers and Daughters in Roman Society, or the
five volumes of papers from the international conferences on the Roman family
originally organized by the late Beryl Rawson.14
Af t e rwo r d : T h e F u t u r e o f t h e A n c i e n t G r e e k Fa m i ly 181

Certainly there are advantages in privileging the household in this way. The
household provided the economic context for the family’s life cycle.15 The birth
of children put a strain on its resources, often met by bringing in outsiders
(slaves, wage workers), especially at peak times such as harvest. These were
then let go or sold in times of crisis – perhaps one in four crops in Attica was
inadequate – and when children grew old enough to contribute their labour.
These outsiders might affect the family’s emotional economy too. A source
of continuity in the flux caused by demographic realities, nurses and paida-
gogoi also remind us that the roles of ‘fathers’ and ‘mothers’ are complex and
culture-bound. Engendering and birthing, nurturance, training, sponsorship
can be tasks for different adults, not all of them kin.16 Slave mistresses, however
common in fact, figure in Greek drama precisely to raise questions about the
family’s boundaries. In Agamemnon, Cassandra and the king enter on a cart
meant to mirror a wedding procession, an image reinforced by references to a
bride’s veil. Is the captive, a princess after all, a threat to Clytemnestra’s place in
the family? Was the queen herself (like any wife) not once an outsider? Had she
not too brought yet another, Aegisthus – kin but hostile – into the home? In
Euripides’ Andromache and in comedies of Menander we detect contemporary
voices debating the definition of family and its relation to the household. The
structural and other similarities of the household’s subordinate members –
women, children, slaves – has proved itself to be an especially fruitful subject.17
And of course our interest in the Greek household echoes an especially clear
and convincing voice – Aristotle’s in Politics.
Aristotle nevertheless has his limitations and biases. As Giulia Sissa puts it,
Politics is ‘an essay in social anthropology rather than a record of the reality
of the family in ancient Athens’.18 Aristotle’s presentation is motivated, at least
in part, by a reaction against Plato’s dismissal of the family in The Republic.
His inclusion of slaves in the oikos is in fact eccentric.19 His account of the
motivation for Pericles’ citizenship law of 451/0 BC, which established the legal
basis of the Athenian nuclear family for generations, is internally inconsistent
and anyway improbable.20 He exaggerates the role played by disputes over
epikleroi at Phocis – family feuds – in the causes of the Third Sacred War and
(as a comparison with Thucydides shows) likely does the same in respect to
conflicts at Mytilene.
All this amounts to a plea for reinstating the Greek family on our agenda.
What should a new synthesis entail? First, it should consider ‘family relations’,
both external links (with the polis as a whole, with other kin groups, with
outsiders such as slaves) and internal ties. It should recognize that even the
smallest family was nuclear, not an indivisible atom. So when the Athenians
deposited their dependants on Salamis in the face of the Persian advance (says
Lysias), they felt pity for their children, yearning for their wives, compassion
for their parents.21 In another funeral speech, Hyperides makes other distinc-
tions, saying that his fellow citizens owe it to their brave forebears that fathers
have grown famous, mothers are looked up to in the city, sisters make worthy
182 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

marriages, and that children will find a way to the people’s good will.22 How do
those with different roles relate? (Fathers to sons? Fathers to all others?) How
do different groups interact (men to women, the old to the young)? Second,
how does the family change over time? Do relations between individuals alter
as they age? Did the Greeks (for example) share the view of Martha Cochrane
in Julian Barnes’ England, England: ‘After the age of 25, you were not allowed to
blame anything on your parents’?23 Perhaps not: the prosecutor in an Athenian
lawsuit says that a father – a man over 50 above all – should restrain and curb
his wanton sons.24 What was the effect of high mortality? The influence of the
mother’s father and brother is prominent in myth, and maternal kin have been
found to be more supportive than a father’s family in the Athenian courts.25
One explanation: wives were normally much younger than their husbands, so
more of their kin survived. Does wealth matter? The rich may have handed over
property earlier, as dowries to their daughters and as inheritances pre-mortem
to their sons, simply because they had more than enough to live on. Is this
why bridegrooms on Attic vases and suitors on Menander’s stage (these at least
members of the elite) seem so young? And how about long-term change? In
some ways it seems that even modern Greek families are a lot like ancient ones,
as in their naming practices.26 Yet we can also identify changes within relatively
short periods of time: the Peloponnesian War kept many men away from home
for long stretches or forever, and women’s role in the family probably grew to
(help) fill the gap. The effects of other secular changes are more problematic.
Alexander’s conquests reshaped the Greek world. Yet Reit van Bremen considers
the Hellenistic family to have become distinctive only a century and a half later
and Dorothy Thompson denies that there is such a thing as the Hellenistic
family at all: ‘No single model [ . . . ] can cover all the evidence’.27 In any case, few
families could have been aware of even momentous developments over time.
While Egyptian family archives extend over 11 generations, there is nothing
remotely similar from Greece, where oral tradition reached back to the time of
a great-grandfather and (judging from Andocides) was not very reliable then.28
The extent of the ‘family’ at any one time is a third issue for consideration.
In Athens, no law set out the membership of the oikos; orators were free to
shrink or expand it as their cases required. So Andocides provides a list of
his relations denounced by Diocleides. This family is made up of his father,
a brother-in-law, the father’s cousin, his son, and his brother-in-law.29 The
anchisteia, ‘those nearest’, extended to the children of cousins (and so to
members of different oikoi by any measure). This category was important for
inheritance of property and determining who was to fulfil certain obligations
(prosecuting homicides and, for women, mourning). But it was unstable,
defined for each individual and in existence for his or her lifetime only, and
never acted as a group. Moreover, it is unclear whether all forms of inheritance
followed the same rules (as William Bubelis argues in the case of priesthoods
in this volume).30 At its narrowest, the family was the nuclear triad, the father,
mother and child(ren) who ate and slept in the same dwelling, conducted
Af t e rwo r d : T h e F u t u r e o f t h e A n c i e n t G r e e k Fa m i ly 183

regular religious rites together, and were entwined in a mesh of economic

symbiosis and emotional solidarity. I will round off these remarks on the future
of the ancient Greek family with a few comments on two themes, myth and
domestic space. Neither plays a role in Patterson’s or Pomeroy’s book and both
confirm the importance of the nuclear family as opposed to more extended kin
Much more than the Romans’, the pantheon of the Greeks was structured
as a family, with Zeus, an adult male, at its head.31 But the divine family, like
that of mortal Greeks, practised partible inheritance. Zeus, whose domain is
the sky, rules over the gods both literally and owing to his superior strength,
while his brothers hold sway over the sea (Poseidon) and the underworld
(Hades). This is not a system of primo- or ultimo-geniture. In fact, Zeus is
both first- and last-born, a sign that he is special. He was the youngest son of
Cronos and Rhea, but – since Cronos swallowed all his older offspring and was
prevented from treating Zeus the same only because Rhea hid the baby away
and substituted a stone – he has effectively spent more time alive as well. His
success, won at his father’s expense, may reflect a tension within many Greek
families. The first-born son must always have been welcome, but a second posed
a problem. Certainly he was insurance against his brother’s death, a misfortune
all too common. But (if both survived) he must also be a drain on the family’s
resources (at first at any rate) and these might not be sufficient to support two
families when the time came to pass them down.
Conflict between fathers and sons did not start with Cronos and Zeus:
Uranus too had tried to prevent the birth of his children by concealment, within
the body of their mother Earth, and Cronos had deposed him (and freed his
other children) by castration. Such struggles needed to be carried to extremes
among the gods because immortal fathers do not weaken with age, let alone
leave the field free by dying. Zeus too faced the prospect that a son would
grow to be greater and overthrow him. However, he avoided that fate by letting
Peleus marry Thetis and engender Achilles. In this way as in others he estab-
lishes a new natural order, in which fathers and sons need not be enemies. As a
result, Hector (in The Iliad) may hope that the infant Astyanax will grow up to
surpass him. Unlike the gods, Greek men could not expect to live forever; their
ambitions to live on in memory required a son to succeed (in both senses of the
word). The Odyssey too offers a model of how human fathers and sons should
behave. Laertes retired as king of Ithaca in favour of his son Odysseus and
Telemachus did not attempt to supplant his father Odysseus in his absence.32
At the end of the poem, all three generations stand together to fight common
enemies in defence of their home and family.
How about the family’s women? Greek myths include instances of Zeus
giving birth from his own body. Athena (sprung from his head) is a female
figure with a warlike nature and a bent for helping male heroes gain their ends;
Dionysus (once bound up in his thigh) is a male god surrounded in myth by
female followers and with a feminine side of his own. Zeus, it seems, is powerful
184 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

enough to produce gods who have a share in both genders. He is sometimes said
to be Hephaestus’ father too. But there is another version, in which he is Hera’s
son alone – and he is the one god to be physically imperfect, suffering from
what we must vaguely call (to use the jargon of the National Hockey League)
a lower-body injury. Even in birthing, women’s special area of expertise, the
queen of the gods is not a match for her mate. Ideally (as male characters in
tragedy lament) women would not be necessary for the family’s continuation
at all. Are they even really a part of it? They do move from one to another at
marriage and return to their original family, of birth, on divorce (and Athenian
law permitted fathers to force the dissolution of a marriage until the birth of a
male heir and perhaps after). Poised as they are between families, liable to move
among them, the position of women is hard to establish securely. Greek myth
presents them with choices, all of them likely to be wrong.33 Medea abandons
her family of birth, betraying her father and murdering her brother, in order
to run off with Jason and found a new family with him. Althaea, on the other
hand, avenges her brothers by causing her son’s death. In each case – and this
is the point – women’s decisions on family loyalty are ruinous to men, and they
are blamed for their role in a social system in which they are rarely agents.
We should not forget that the families of myth were real to most Greeks
(and could conventially be presented as real by sceptics). Omar Coloru’s
contribution to this volume outlines the efforts made by Hellenistic dynasts to
present themselves as champions of family values. At the same time, sharing
the same – mythical – family also played a political role, as the basis of alliances
between city-states in the eastern Mediterranean. Such linkages continued to
count in the Greek cities of the Roman Empire, though they are much less
prominent in the Latin west.34 There are many examples (as well as others
for different purposes, such as the upgrading of a local festival).35 A recently
discovered inscription from the third century BC, new evidence for a treaty
between Athens and Cydonia on Crete, appeals to their mythical ancestors, Ion
and Cydon, both sons of Apollo.36 This family relationship, attested by other,
independent sources, is likely to be traditional rather than a fiction invented for
the purpose.37 There is therefore some likelihood that the link between these
two distant cities was inspired by their family connection in the distant past.
Before we reach our final destination, domestic space, we must take a detour
through one of the more remarkable byways of ancient family history. Those
who think of classicists as stodgy and repressed will be surprised to find many
recent pages of two of the profession’s flagship journals devoted to discussions of
brother–sister sex, marriage and reproduction. Goodbye, Mr Chips, indeed. The
sibling marriages of the Ptolemies have quite a few parallels elsewhere; Sheila
Ager’s lead article in Journal of Hellenic Studies convincingly portrayed it as a
means for monarchs to flaunt their exceptionality, elevating them to the ranks
of divine models such as Isis and Osiris and Zeus and Hera.38 That ordinary
brothers and sisters in Egypt of the Hellenistic and Roman periods shared this
custom has long been recognized. Though the consensus was briefly threatened
Af t e rwo r d : T h e F u t u r e o f t h e A n c i e n t G r e e k Fa m i ly 185

by an article by Sabine Huebner arguing that those our sources term ‘brother’
and ‘sister’ are in fact not biological siblings but participants in a pattern of
adoption and marriage widespread in the eastern Mediterranean, the evidence
of census records and other official documents, literary texts (which stress
that the Egyptian marriages were indeed thought to be unusual), and naming
practices in my view establishes the existence of full sibling marriage beyond
a doubt.39 A series of sophisticated studies by Walter Scheidel has brought this
anomalous (though not quite unique) phenomenon to the attention of social
scientists and evolutionary biologists.40 In its own way, it is as revolutionary as
the discovery of the Roman nuclear family and, like that research, underscores
the continuing relevance of the study of Greek and Rome to more contem-
porary concerns.
How are we to explain so queer a custom? There is no agreement. The
approach I find most helpful is to place the practice into the context of Greek
colonization of Egypt after the conquests of Alexander.41 Greeks who stayed
in or came to Egypt wanted to marry other Greeks, but found it hard to find
partners of a suitable age and status, all the more in the smaller and isolated
centres where most of them lived. Settlers came from many parts of the Greek
world, each boasting its own customs. Many of their home communities
allowed marriages of half-siblings, though some restricted them to children of
the same father (as in Athens), others to those of the same mother (Sparta). Far
from home as they were, the force of traditional taboos was weakened. And the
step from marrying half- to full siblings may have been easier if this taboo was
not very strong in any case: Aristotle leaves brother–sister incest off his (short)
list of the most shameful forms.42 Brother–sister marriage may represent a
strong form of both the endogamy characteristic of the ancient Greeks and the
solidarity between brothers and sisters so memorably expressed by Antigone.43
I do not think this hypothesis is controverted by the fact that brother–sister
marriages are attested among native Egyptians too. Elite mores are often taken
up by others and in fact Hellenistic Egypt provides an example: the adoption
of a kyrios after the Greek fashion, by Jewish and eventually some Egyptian
women who were not required by law to have one.44 But more certainly needs to
be said. Among much else of value, Janet Rowlandson and Ryusoke Takahashi
have foregrounded the role of the inheritance regime of Greco-Roman Egypt.
Partible as elsewhere in the Greek world, inheritance in Egypt was unconven-
tional in allotting an equal share to women. Since both fathers and mothers
thus transmitted property to both sons and daughters – often giving daughters
their share as dowry – holdings became exceptionally fragmented, a tendency
amplified by the Egyptian custom of raising all their children, daughters too.
For Rowlandson and Takahashi, it is the fragmentation of agricultural land and
the inconvenience of accounting for small lots in the registers of real property
introduced by the Romans in the late first century BC which are crucial. I
wonder if this does not slight the importance of domestic space. Awkward
though they might be, small and scattered landholdings could be worked or
186 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

rented out or sold to a neighbour without much disruption. Also liable to

the requirement that they be registered as private property, Egyptian houses
were less easy to divide, especially since they typically had just one entrance
or staircase.45 Furthermore, as Rowaldson and Takahashi point out, Egyptian
marriages (like most in the Greek world) were virilocal: sibling marriage
allowed daughters to take ownership of part of the parental home at the same
time as they continued to live within it. It was therefore less disruptive than
other endogamous arrangements, such as marrying a cousin. Maintaining
the integrity of domestic space was likely one motivation for brother-sister
This raises one last issue: just how much domestic space did the Greek
family have? I used to imagine that the typical Greek house was small, squalid
and dark, much like the hovels of the poor in Engels’ account of working-class
Manchester in the earlier nineteenth-century. Many of my colleagues, I suspect,
shared this impression. In any case, I was wrong. We now have data from some
300 houses for the period from 800 to 300 BC, from all over the Greek world:
from Acragas, Athens, Olynthus, Miletus, cities, villages, the countryside. These
appear to have been quite spacious on the whole, in the Classical period at least,
with a median ground area for roofed portions of the house of approximately
1,000 (fifth century) to 2,200 square feet (fourth century) or more.46 But such
dwellings are about ten times bigger than English tenant houses in the 1830s
and comparable to those of eighteenth-century Boston or middle-class areas
of Winnipeg today. What is more, the houses which have been excavated were
generally well made, with stone socles, walls of stone or sundried brick, tiled
roofs, earthen floors, or when they had to be waterproof, constructed from
crushed rock. In Ian Morris’ words, ‘fourth-century Greek houses were large
and quite comfortable, even by the standards of developed countries in the
early twenty-first century’.47 Now, this evidence is skewed towards the larger
end of the scale. Smaller houses, at least those less soundly built, are harder to
discern in the material record and may have not seemed worth the trouble of
excavation. In addition, some Greeks certainly lived in apartments or shared
homes. And houses grew larger over time, so that those in Hesiod’s eighth-
century Boeotian village were much more likely to match my expectations
than those in the Athens of Aristotle, 400 years later.48 Still, Greek houses were
reasonably affordable: the single-family home was not the privilege of the elite
alone.49 With all due allowance for scepticism, I am inclined to think that the
houses we have are not so unrepresentative of a significant proportion of the
stock which once existed. And one factor, the disappearance of evidence for
upper storeys, may push the size of known houses down somewhat, though we
cannot know how much area such structures covered or who lived in them. If
Greek houses were as spacious as I now think, they would fit a newly-devel-
oping consensus that the Greeks were richer than we once believed in general:
they and their animals were larger, their agricultural practices more productive,
their access to resources more equal.50
Af t e rwo r d : T h e F u t u r e o f t h e A n c i e n t G r e e k Fa m i ly 187

Might this change the way we think about the Greek family? Answering this
question forces us to enter the realm of speculation, even imaginative fiction,
and run the risk of the criticism Virginia Woolf levelled at Edwardian novelists:
‘They have given us a house in the hope that we may be able to deduce the human
beings who live there.’51 However, I am willing to hazard a guess: since the Greek
house, at least in the Classical period was a more pleasant environment than
we have usually assumed, our understanding of the relation of the public and
private spheres in the Greek polis and of the importance of family life needs
adjusting. To put it crudely: the more spacious and comfortable the home, the
more time family members (especially adult males) would spend and enjoy
there.52 Without question this bald statement is subject to a number of variables.
Even a large dwelling can become crowded if there are enough people in it.
Our best evidence for household size from the Egyptian census records reveals
that just over one-third of Greek households included more than 5 adults and
that one packed in 22.53 Cultures are notoriously unpredictable in their ideas
about privacy and personal space but a house with 22 adults would probably
feel overstuffed anywhere. What is more, the attractions of domestic space
depend on factors other than space too; on furnishings, conveniences and so
on. And the draw of the home may be counteracted by the scale and splendour
of public amenities. Nonetheless, I think we should start to conceive of the
ancient Greeks as homebodies as well as participants in one of the more vital
and vigorous public cultures we know. After all, our evidence seems to show
that in general they cooked and ate at home (rather than in the market or at
stalls in the street); they slept at home (rather than in their workplaces). There
are further implications: for example, while no one now believes that Athenian
women were confined to the home, let alone to a small part of it, it is probable
that a considerable portion of their time was spent indoors. A re-evaluation
of that environment therefore contributes to a less pessimistic picture of their
lives. I am, however, happy to leave further exploration of the relevance of myth
and the role of domestic space to the next synthesis on the Greek family, one
which I am confident my words will again conjure up.

 1 Swedes discovered Winnipeg long ago, whether or not the Vikings made the Kensington
rune-stone. Anders Hedberg and Ulf Nilsson led our Jets to the World Hockey Association
championship in the late 1970s and Thomas Steen starred for the National Hockey League
Jets before becoming a Winnipeg city councillor (where he plays right wing, I am sorry to
say). Even one of Henning Mankell’s victims has family in Winnipeg: ‘one daughter . . . lives
in Canada, Winnipeg – wherever that is’, Faceless Killers (originally published 1991), tr. S. T.
Murray, New York: Viking 1997, 32. The ARACHNE conference was my first opportunity to
return their visits and I enjoyed every moment of it. Many thanks to the organizers, Mary
Harlow, Ray Laurence, Lena Larsson Lovén, and Agneta Strömberg, for their clear thinking
and hospitality and to the Gothenburg students and others for good conversation. This version
of my keynote talk is indebted to Geof Kron, Lisa Nevett, Richard Saller and Walter Scheidel.
188 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

  2 Ann-Cathrin Harders’ paper in this collection shows that old-fashioned comparisons between
Greeks and Romans can still be as productive as consideration of more exotic cultures.
  3 Bradley, 1993.
  4 Sigismund-Nielsen, 2006.
  5 Saller and Shaw, 1984; Martin, 1996, with the brief response of Rawson, 1997.
  6 Edmondson 2005, pp. 215–217, cf. Scheidel forthcoming.
 7 Martin, 1996, 47–60. Lindsay Penner’s chapter in this volume provides another example
of careful quantitative analysis yielding an unexpected extent of variation, in Roman elite
  8 Edmondson, 2005, pp. 216–217.
  9 Review: Golden, 1996. English-language work: Rubinstein, 1993; Strauss, 1994; Demand, 1994;
Oakley and Sinos, 1993; Watson, 1994 and Golden, 1990.
10 Pomeroy, 1997 and C. Patterson, 1998.
11 Patterson: Golden 1998; Pomeroy: Golden, 1999.
12 Bourriot, 1976; Roussel, 1976.
13 C. Patterson, 1998, p. 69.
14 Dixon, 1988; Bannon, 1997; Hallett, 1984. Rawson: the most recent collection of papers arising
from these important meetings is Dasen and Späth, 2010.
15 For the life cycle of the Greek peasant household, see Gallant, 1991.
16 See in general Goody, 1999 (on West Africa.)
17 See, for example, the studies collected in Moggi and Cordiano, 1997; Joshel and Murnaghan,
1998 and Birgitta Sjöberg’s discussion of ‘intersectionality’ in this volume. Athenian orators
urged audiences to go to war to set themselves off from their slaves and wives and to emulate
their valorous ancestors: Hunt 2010, pp. 108–33.
18 Sissa, 1996, p. 194.
19 MacDowell, 1989, cf. Ferrucci, 2006, pp. 186–192 on the differences in Aristotle’s usage of oikia
from that of Xenophon and Attic legal texts and inscriptions.
20 Blok, 2009.
21 Lys. 2.34, 39.
22 Hyper. Epit. 27.
23 Julian Barnes, England, England, New York: Alfred A. Knopf, p. 22. Martha is relevant here
because she is doing graduate work on the history of the Sophists, p. 45.
24 Dem. 54.22; cf. Lys. 19.55, where a speaker, now 30 years old, claims he has never had a dispute
with his father.
25 Myth: Bremmer, 1983; courts: Humphreys, 1986.
26 Bresson, 2006.
27 Van Bremen, 2003; Thompson, 2006, p. 94.
28 Egypt: Thompson, 2006, p. 106; Athens: Thomas, 1989, pp. 95–154. Tomb-cult too extended
over only three generations in Greece: Antonaccio, 1993, p. 63.
29 Andoc. 1.47, cf. Dem. 59. 6, 12.
30 For the inheritance of sitesis, the right to dinner in the Prytaneion, see now MacDowell, 2007.
31 Much of the seminal work on myth and the Greek family and society more broadly has been
produced by scholars writing in French – Louis Gernet, Georges Dumézil, Jean-Pierre Vernant,
Nicole Loraux. See most recently Renaud and Wathelet, 2007, and, on Zeus’ position, Brulé,
32 For another approach, invoking ‘matrimonial succession’ in kingship in the heroic or Bronze
Age, see Finkelberg, 1991.
33 Visser, 1986.
34 For Rome, see now Battistoni, 2009.
35 Jones, 1999 provides a wide-ranging account; local festival: see, e.g., L. E. Patterson, 2010.
36 Papazarkadas and Thonemann, 2008.
37 This is stressed by Curty, 2005.
38 Ager, 2005, cf. now Müller, 2009.
39 Huebner, 2007, with the responses of Remijsen and Clarysse, 2008 and of Rowlandson and
Takahashi, 2009.
Af t e rwo r d : T h e F u t u r e o f t h e A n c i e n t G r e e k Fa m i ly 189

40 E.g., Scheidel, 1996, 1997.

41 See especially Shaw, 1992.
42 Arist. Pol. 2. 1262a25.
43 As Trollope observed in The Eustace Diamonds (originally published 1871), Oxford: OUP
1973 (10), ‘Brothers do not always care much for a brother’s success, but a sister is generally
44 Pomeroy, 1984, p. 121. Note here that unlike Greeks and native Egyptians, Romans and Jews
were forbidden by law to marry siblings.
45 Muhs, 2008. Though Muhs emphasizes the care Egyptians took to prevent the division of
homes, his study of Hawara in the Fayum reveals no sibling marriages.
46 See the table in Morris, 2005, p. 110. The fullest collection of data that I know, in an unpub-
lished paper by Geof Kron, concludes, ‘there is remarkably little variation in the sizes of
citizens’ houses, which seem to have varied from 2200 to 3300 square feet’ in the Classical
period (Kron, unpublished). Major collections of evidence include Robinson and Graham,
1938; Hoepfner and Schwandner, 1994 and Trümper, 1998.
47 Morris, 2005, p. 123.
48 Morris, 2004, 2005, pp. 107–125.
49 Geof Kron estimates that nearly a third of Athenian citizen families at the end of the fourth
century BC, and perhaps as many as three-quarters, could afford to buy a house: Kron forth-
coming, cf. Ober, 2010, 258.
50 See, e.g., Kron, 2005, 2008, Ober, 2010.
51 Woolf, V. (1924), Mr Bennett and Mrs Brown, London: Hogarth Press, p. 18.
52 It may be significant here that Hesiod’s Works and Days demonstrates considerable concern
about neighbours. Did his rural lifestyle combine with a smaller and less pleasant house to
induce him to spend more time out of doors than later Greeks?
53 Thompson, 2006, 102

Ager, S. L. (2005), ‘Familiarity breeds: incest and the Ptolemaic dynasty’, JHS,
125, 1–34.
Antonaccio, C. (1993), ‘The archaeology of ancestors’, in Dougherty, C. and
Kurke, L. (eds), Cultural Poetics in Ancient Greece. Cult, Performance, Politics,
Cambridge: Cambridge University Press, pp. 46–70.
Battistoni, F. (2009), ‘Rome, kinship and diplomacy’, in C. Eliers ed. Diplomats
and Diplomacy in the Roman World, Leiden and Boston: Brill, Mnemosyne,
(Supplement 304), pp. 73–97.
Blok, J. H. (2009), ‘Perikles’ citizenship law: a new perspective’, Historia, 58,
Bourriot, F. (1976), Recherches sur la Nature du Génos, Lille: Université de Lille
III/H. Champion.
Bradley, K. R. (1993), Discovering the Roman Family, New York and Oxford:
Oxford University Press.
Brulé, P. (2006), ‘La parenté selon Zeus’, in Bresson, A. et al. (eds), Parenté
et Société dans le Monde Grec de l’Antiquité à l’Age Moderne, Bordeaux: De
Boccard, pp. 97–119.
Curty, O. (2005), ‘Un usage fort controversé: la parenté dans la langue diploma-
tique de l’époque hellénistique’, AncSoc, 35, 101–17.
190 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

Dasen V. and Späth, T. (eds) (2010), Children, Memory and Family Identity in
Roman Culture, Oxford: Oxford University Press.
Demand, N. (1994), Birth, Death, and Motherhood in Classical Greece, Baltimore
and London: Johns Hopkins University Press.
Edmondson, J. (2005), ‘Family relations in Roman Lusitania: social change in a
Roman province?’, in George, M. ed., The Roman Family in the Empire: Rome,
Italy and Beyond, Oxford: Oxford University Press, pp. 183–229.
Ferrucci, S. (2006), ‘L’“oikos” nel diritto attico. Pubblico, privato e individuale
nella democrazia ateniese classica’, Dike, 9, 183–210.
Finkelberg, M. (1991), ‘Royal succession in heroic Greece’, CQ, 41, 303–16.
Golden, M. (1990), Children and Childhood in Classical Athens, Baltimore and
London: Johns Hopkins University Press.
—(1996), ‘Review of D. Ogden, Greek Bastardy’, BMCR, 7, 718–722.
—(1998), ‘Review of Patterson, The Family in Greek History’, Phoenix, 52,
—(1999), ‘Review of Pomeroy, Families in Classical and Hellenistic Greece’, CR,
49, 157–159.
Hallett, J. P. (1984), Fathers and Daughters in Roman Society, Princeton:
Princeton University Press.
Hoepfner, W. and Schwandner, E.-L. (1994), Haus und Stadt im Klassischen
Griechenland, (2nd edn), Munich: Deutscher Kunstverlag.
Huebner, S. R. (2007), ‘Brother-sister marriages in Roman Egypt: a curiosity of
humankind or a widespread family strategy?’, JRS, 97, 21–49.
Humphreys, S. C. (1986), ‘Kinship patterns in the Athenian courts’, GRBS, 27,
Hunt, P. (2010), War, Peace and Alliance in Demosthenes’ Athens, New York:
Cambridge University Press.
Jones, C. P. (1999), Kinship Diplomacy in the Ancient World, Cambridge MA:
Harvard University Press.
Joshel, S. R. and Murnaghan, S. (eds) (1998), Women and Slaves in Greco-
Roman Culture: Differential Equations, London and New York: Routledge.
Kron, G. (2005), ‘Anthropometry, physical anthropology and the reconstruction
of ancient health, nutrition and living standards’, Historia, 54, 68–83.
—(2008), ‘The much maligned peasant. Comparative perspectives on the produc-
tivity of the small farmer in classical antiquity’, in De Ligt, L. and Northwood,
S. (eds), People, Land and Politics. Demographic Developments and the
Transformation of Roman Italy, 300 BC – AD 14, Leiden: E. J. Brill, pp. 71–119.
—(forthcoming), ‘Comparative evidence and the reconstruction of the ancient
economy: Greco-Roman housing and the level and distribution of wealth
and income’, in Callatay, F. and Wilson, A. (eds), Long-term Quantification in
Ancient Mediterranean History, Oxford: Oxford University Press.
—(unpublished), ‘The use of housing evidence as a possible index of social
equality and prosperity in classical Greece and early industrial England’.
Lacey, W. K. (1968), The Family in Classical Greece, London: Thames and Hudson.
Af t e rwo r d : T h e F u t u r e o f t h e A n c i e n t G r e e k Fa m i ly 191

MacDowell, D. M. (1989), ‘The oikos in Athenian law’, CQ, 39, 10–21.

—(2007), ‘Hereditary sitesis in fourth-century Athens’, ZPE, 162, 111–113.
Martin, D. B. (1996), ‘The construction of the ancient family: methodological
considerations’, JRS, 86, 40–60.
Moggi, M. and Cordiano, G. (eds) (1997), Schiavi e Dipendenti nell’Ambito dell’
‘oikos’ e della ‘familia’. Atti del XXII Colloquio GIREA, Pontignano (Siena),
19–20 novembre 1995, Pisa: Edizioni ETS.
Morris, I. (2004), ‘Economic Growth in ancient Greece’, Journal of Institutional
and Theoretical Economics, 160, 709–742.
—(2005), ‘Archaeology, standards of living, and Greek economic history’, in
Manning, J. and Morris, I. (eds), The Ancient Economy: Evidence and Models,
Stanford: Stanford University Press, pp. 91–126.
Muhs, B. (2008), ‘Fractions of houses in Ptolemaic Hawara’, in Lippert, S.
and Schentuleit, M. (eds), Graeco-Roman Fayum – Texts and Archaeology.
Proceedings of the Third International Fayum Symposion, Freudenstadt, May
29 – June 1 2007, Wiesbaden: Harrassowitz, pp. 187–195.
Müller, S. (2009), Das Hellenistische Königspaar in der Medialen Repräsentation:
Ptolemaios II. und Arsinoe II, Berlin and New York: De Gruyter.
Oakley, J. H. and Sinos, R. H. (1993), The Wedding in Classical Athens, Madison:
University of Wisconsin Press.
Ober, J. (2010), ‘Wealthy Hellas’, TAPA, 140, 241–286.
Ogden, D. (1996), Greek Bastardy in the Classical and Hellenistic Periods,
Oxford: Oxford University Press.
Papazarkadas, N. and Thonemann, P. (2008), ‘Athens and Kydonia: Agora I
7602’, Hesperia, 77, 73–87.
Patterson, C. B. (1998), The Family in Greek History, Cambridge MA and
London: Harvard University Press.
Patterson, L. E. (2010), ‘Strabo, local myth and kinship diplomacy’, Hermes, 138,
Pomeroy, S. B. (1984), Women in Hellenistic Egypt: from Alexander to Cleopatra,
New York: Schocken Books.
—(1997), Families in Classical and Hellenistic Greece. Representations and
Realities, Oxford: Oxford University Press.
Rawson, B. (1997), ‘“The family” in the ancient Mediterranean: past, present,
future’, ZPE, 117, 294–296.
Remijsen, S. and Clarysse, W. (2008), ‘Incest or adoption? Brother-sister
marriage in Roman Egypt revisited’, JRS, 98, 53–61.
Robinson, D. M. and Graham, J. W. (1938), Excavations at Olynthus. VIII. The
Hellenic House, Baltimore: Johns Hopkins University Press.
Roussel, D. (1976), Tribu et Cité, Paris: Les Belles Lettres.
Rowlandson, J. and Takahashi, R. (2009), ‘Brother-sister marriage and inher-
itance strategies in Greco-Roman Egypt’, JRS, 99, 104–139.
Rubinstein, L. (1993), Adoption in IV. Century Athens, Copenhagen: Museum
Tusculanum Press.
192 Fa m i l i e s i n t h e G r e c o - Rom a n Wo r l d

Saller, R. P. and Shaw, B. D. (1984), ‘Tombstones and Roman family relations in

the Principate: civilians, soldiers and slaves’, JRS, 74, 124–56.
Scheidel, W. (1996), ‘Brother-sister and parent-child marriage outside royal
families in ancient Egypt and Iran: a challenge to the sociobiological view of
incest avoidance’, Ethology and Sociobiology, 17, 319–340.
—(1997), ‘Brother-sister marriage in Roman Egypt’, Journal of Biosocial Science,
29, 361–371.
—(forthcoming), ‘Epigraphy and demography: birth, marriage, family and
death’, in Davies, J. and Wilkes, J. (eds), Epigraphy and the Historical Sciences,
Oxford: British Academy/Oxford University Press.
Schmitz, W. (2007), Haus und Familie in Antiken Griechenland, Munich:
Shaw, B. D. (1992), ‘Explaining incest: brother-sister marriage in Graeco-
Roman Egypt’, Man, 27, 267–299.
Sigismund-Nielsen, H. (2006), ‘Collegia: a new way for understanding the
Roman family’, Hephaistos, 24, 201–213.
Sissa, G. (1996), ‘The family in ancient Athens (fifth-fourth century BC)’, in
Burguière, A. et al. (eds), A History of the Family 1. Distant Worlds, Ancient
Worlds, (translation by Tenison, S. H. et al.), Cambridge MA: Harvard
University Press, pp. 194–227.
Strauss, B. S. (1993), Fathers and Sons in Athens. Ideology and Society in the Era
of the Peloponnesian War, London: Routledge.
Thomas, R. (1989), Oral Tradition and Written Record in Classical Athens,
Cambridge: Cambridge University Press.
Thompson, D. (2006), ‘The Hellenistic family’, in Bugh, G. ed., The Cambridge
Companion to the Hellenistic World, Cambridge: Cambridge University Press,
pp. 93–113.
Trümper, M. (1998), Wohnen in Delos: eine Baugeschichtliche Untersuchung zum
Wandel der Wohnkultur in Hellenistischen Zeit, Rahden: Leidorf.
van Bremen, R. (2003), ‘Family structures’, in Erskine, A. ed., A Companion to
the Hellenistic World, Oxford: Blackwell, pp. 313–330.

Words in italics are in Latin or are titles history 27

Numbers in italics are figures life 30–1, 38
nuclear 2, 14–15, 29, 131, 179–80
adelphos 88–9 role 145, 170
adult 61, 128, 130–1 unit 150
aesthetics 6–7 fate 162, 165, 167–8
affection 86, 88, 89, 91, 111 father 4, 15, 17, 87–8, 96, 107, 110–11, 114, 163,
affiliation 19, 111, 145, 147, 151 165–8, 181–5
age 52–3, 62, 127, 130–1, 152, 157 funeral 108, 114
aristocracy 17–18
Aristotle 18, 181 gender 3, 48–55, 143–6, 151, 152–4, 169
Athens 14, 18, 33–4, 37, 95, 182 genealogy 13, 114
Augustus 32, 143, 147–8 genos 96–7, 101–2
genos gunaikôn 52
bone 124–37 grandmother 115–16
boyhood 65–6 grave 69, 124
brother 88–9, 98–9, 101, 111, 185–6 grief 106, 161, 170, 172 see also emotion
burial 69, 80n, 121, 124, 128, 131, 136
Hesiod 18, 52–5
childbirth 109–10 Theogony 52
children 4, 13, 52, 60, 69, 86, 98, 101, 110, 114, Works and Days 48, 51, 52–5
128, 131, 133, 136, 150, 165–7, 181 household 1, 2, 3, 14–15, 19, 28–30, 35–6,
Cicero 6, 18, 168 51, 54–5, 69, 85, 143, 145–51, 179–81,
citizenship 33–7 187
columbaria 143–51 housing 37–8
husband 53–4, 89, 111–13
daughter 101, 108–9, 185–6
death 106–7, 109–16, 161–3, 167 iconography 60, 67, 90
demography 27–9, 31–3, 38 imagery 6, 107, 167
discrimination 48–50, 52–3, 55 inheritance 15, 16, 95–7, 182, 185
domus 12, 14–19, 86 inscription 86, 95, 106, 112, 116, 124, 126, 131–6,
144, 148, 170 see also epigram
emotion 11, 116, 159–62, 167, 169–72, 181 see also intersectionality 48–55
epigram 106–8, 116 see also inscription king 84–5, 87–9
epigraphy 121, 136 kinship 11–19, 86, 121, 126, 133–4, 136
epitaph 113, 116, 121, 131–6, 144–6, 160–70,
179 language 2, 4, 6, 84, 87, 90, 122, 170
law 4, 11, 33, 35–6, 50, 85, 181–2
familia 12, 17, 31, 38 see also family loyalty 84, 88, 89, 184
family 1–7, 10, 15, 19, 28–30, 50, 85, 106, 113–14,
121, 136, 144, 151, 167, 179–87 see also familia manumission 145, 147, 151
194 I n d e x

marriage 3, 12, 16, 28, 33–6, 53, 89, 98, 108–9, 113, Rome 14, 17–18, 33–5, 38, 143, 170
145, 152, 157, 185–6 see also wedding
intermarriage 34–6 sexuality 52–3
memory 6, 160, 171, 183 sibling 34, 89, 101, 111, 131, 184–6
mortality 106, 167, 182 sister 89, 111, 181, 184, 185–6
mother 53, 61–2, 89, 108–11, 114–16, 168 slave 4, 145–7, 149–51, 152–5, 181
grandmother 115–16 social network 11, 18, 149
social relation 3, 11, 13, 50
name 54, 126–7, 131–6, 144–7, 152–3 social status 51, 107, 114
son 88–9, 95–100, 183
oikia 50, 85–6, 179 space 36, 50, 172
oikos 12, 14–15, 18–19, 31, 48–55, 60, 68–9, 85, domestic space 186–7
95–8, 101–2 succession 95–8, 101–2

painting 60, 61, 66, 159–60, 165, 167, twin 110–11

parent 107, 111, 165, 167 vase 49, 60–2, 66, 70
paterfamilias 16–17, 165 virtue 53, 54, 112–14
patriline 95, 98, 101
polis 18, 50 wedding 64–5, 108–9 see also marriage
politics 19, 27, 32, 35–6, 38 wife 53–4, 89, 98, 108, 111–13, 149
population 31–3, 36, 128, 149 woman 52, 61–5, 106–16, 149
property 36–7, 50, 85, 182, 185
Xenophon 37, 51–5
queen 89, 116 Oeconomicus 48, 50–5

relative 15–16, 30, 90, 121, 124, 131–6 youth 65–9, 77