Families in the Greco-Roman World 9781472540683, 9781472505743

Citation preview

Abbreviations 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. Habelt IG Inscriptiones Graecae I. Iasos Blümel, W. (1985), Die Inschriften von Iasos, Bonn: R. Habelt I. Ilion Frisch, P. (1975), Die Inschriften von Ilion, Bonn: R. Habelt 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) (19902003), Pompei: Pitture e Mosaici, Vols 1-11, Rome: istituto della Enciclopedia Italiana RE Paulys Realencyclopädie der classischen Altertumswissenschaft RömMitt Römische Abteilung Mitteilungen des deutschen archäologischen Instituts SE Alpers, M. and Halfmann, H. (1995), Supplementum Ephesium, Hamburg: Deutschen Forschungsgemeinschaft

Abb r ev iat io n s

SEG Chaniotis, A., Corsten, T., Stroud, R. S. and Tybout, R. A, Supplementum Epigraphicum Graecum, Leiden: Brill TAPA Transactions of the American Philological Association Thess. Mnem. Arvanitopoulos, A. S. (1909), ΘΕΣΣΑΛΙΚΑ ΜΝΗΜΕΙΑ. 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

xi

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 Mediterranean. 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 publications 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 prosopography 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 interdisciplinary 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 JulioClaudian 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).

Preface The volumes of The Family in Antiquity series bring together scholars investigating 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 interaction 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 importance 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: http://www.iaa.bham.ac.uk/news/conferences/ oikosfamilia/index.shtml 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 osteological 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 volumes. Mary Harlow (University of Birmingham), Lena Larsson Lovén (University of Gothenburg), Ray Laurence (University of Kent), Agneta Strömberg (University of Gothenburg)

1

Introduction: From Oikos to Familia: Looking Forward? 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 archaeology. 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 archaeological 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 - R om 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 WileyBlackwell 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 sociolinguistics 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 twentyfirst 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 - R om 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 discourse. 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 development 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 archaeologists, 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 - R om 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 understanding 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 representations 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 understood 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 understanding 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 methodologies 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.

Notes   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 http://www.arachne.hum.gu.se/english/ 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: http://icls.sas. ac.uk/institute/meetingslist/conferences.html 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 - R om 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.

Bibliography 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: WileyBlackwell, 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: Routledge. —(2009), Roman Passions: A History of Pleasure in Imperial Rome, London: Continuum. 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: WileyBlackwell, 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: WileyBlackwell, 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 publications.

2

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 importance 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 aftereffect 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 fellowcolleagues’ 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 - R om a n Wo r l d

demand new perspectives and reconfigurations of kinship studies and kinship theories. 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 legitimizing 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 kinshipfocused 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 relatedness, 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 - R om 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 anthropologists 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 ‘quasifamilial’ 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 - R om 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 paterfamilias, 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 benevolence.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 hierarchical 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 magistrates’ 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 interrelated: 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 - R om 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 communication 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 interrelations 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, superdomestic 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 antagonism 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 relatedness 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 - R om a n Wo r l d

Notes   * 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. 76–78.   2 Joachim Hedén’s movie was released in Sweden on 24 September 2010; see the entry on IMDB. com: http://www.imdb.com/title/tt1537770/ (Accessed 15 November 2010).   3 See the entry on IMDB.com: http://www.imdb.com/title/tt0108778/ (Accessed 15 November 2010).   4 Cf. Peletz, 1995, pp. 364–365 on the new move from biology to choice concerning the constitutions 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 categories. 20 There are ancient parallels: Aristotle and Cicero (drawing on Aristotle) suggest in their philosophical 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 literature. 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 10.23.1.5; 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. 44–50. 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 - R om 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.

Bibliography 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), 324–357. 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-SchwesterBeziehungen 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 Archaeology. 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 Press. 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 - R om 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 FrancoAllemand 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 Publishers. Patterson, C. (1998), The Family in Greek History, Cambridge, London: Harvard University Press. Peletz, M. G. (1995), ‘Kinship studies in the late twentieth-century anthropology’, 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. 610–623. —(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 Verlag. —(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: http://circle.ubc.ca/handle/2429/17421 (Accessed 15 November 2010). —(2010), ‘Lessons from the demise and renewal of kinship studies’, Rosetta, 8, 83–99, URL: http://www.rosetta.bham.ac.uk/issue8/articles/varto-kinship/ (Accessed: 15 November 2010).

26

Fa m i l i e s i n t h e G r e c o - R om 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.

3

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 emphasized 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 demonstrated 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 - R om a n Wo r l d

Reconstructing Ancient Families In its relatively short history, demography of the ancient family, as a sub-discipline, 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 History 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 scholarship 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 responsibility 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 industrialization 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 individualism. 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 - R om 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 scholarship.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 commonalities 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 importance 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 incentives 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 - R om 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 criticized 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 processes.24 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 behaviours 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 historians 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 - R om 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 divergences 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 attestations 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 citizens. One might, still, object that the reflections of intermarriage on inscriptional 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 acknowledges 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 explanatory 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 inscriptions 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 - R om 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 disenfranchisement 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 households 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 indigenous, 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 - R om 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 demography 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 investigating 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.

Notes   * 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 acknowledges 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 underlying 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 - R om 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 consequences 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 technological 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 systems. 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 (http://statline.cbs.nl/StatWeb/publication/?VW= T&DM=SLNL&PA=71888NED&D1=a&D2=1–2&D3=1–7&D4=0&D5=l&HD=100108– 1016&HDR=G4,G3,T&STB=G1,G2). 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 accommodate 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 - R om 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 registered 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.

Bibliography 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 Press. 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, 1982). 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 - R om 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: Blackwell. 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 - R om 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: http://www.let.leidenuniv.nl/pdf/ geschiedenis//tacoma_migration_and_the_roman_family__19–1-07.pdf. (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), 111–137. 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 consequences 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 Dissertation.

4

More than Just Gender: The Classical Oikos as a Site of Intersectionality Birgitta L. Sjöberg

Introduction 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 age. 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 illustrating 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 - R om 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 importance 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 intersectionality 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 intersectionality 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 representatives 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 - R om 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 intersection 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 discrimination 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 intersectionality 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 - R om 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 interesting 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 hierarchical 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 - R om a n Wo r l d

Notes   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.

Bibliography 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: Duckworth. Fisher, N. ed. (1976), Social values in classical Athens, The Ancient World, 630. 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: Duckworth. 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 - R om 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: Wahlström. 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), 10–21. 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, 301–11. 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: http://www.arkeologi.uu.se/Forskning/Projektarkiv/ Intersectionality/ 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.

5

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 organizational 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, woodenlike 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 - R om 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-withchild 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 protagonists 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 - R om 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. 150319)

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 controversial 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 - R om 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 representations 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 - R om 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 depictions 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 ‘boys-club’.

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 significance 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 represented 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 - R om a n Wo r l d

Table 5.1: Catalogue of Vases (non-exhaustive) Shape

Date

Findspot

Decoration – comments

Select bibliography

Alabastron

LPC/EC?

?

Small figure amidst warriors; young hoplite or ghost

CVA Madrid 1, IIIc, pl. 3.10a–b (32646)

Alabastron

EC/MC

Corinth Potters Quarter (aryballos deposit)

Young girl with flower

Corinth XV ii p. 104, no. 491, pls. 24, 95 (K 72)

Alabastron

MC/LC

?

MuZ 213–4, Two women 991, pl. 40. 172 seated on stool back to back, one (Berlin 4285) holding a child that is pacing, a third woman is standing with a spindle in hand, siren

Aryballos

MPC

Thebes

Athena, Aphrodite?, Dioskouroi? And a small figure (child?, xoanon?); sphinx in the back; three dogs on handles

CVA Oxford 2, III c, pl. 1.5, 36, 51 (G 146); Amyx 1988, II 619, no. 3, n. 7; Schefold 1966, 41, fig. 10 (Rescue of Helen); Arafat 2004, 32–4, fig. 2.7

Aryballos

MC

Corinth, Apollo temple

Chorus of 7 boys, one Pyrrhias is jumping in air in front of the flutist Polyterpos

Lorber 1979, 35–6, no 39, pl. 8 (Corinth C-54-1); Boardman 2007, 265, fig. 294; 158–163 Amyx 1988, II 560–1, no.17

Corinth

Procession of man, women, young kanephoroi

CVA Norway 1, pl. 4 (Oslo Mus Etn 6909); Amyx 1988 II 658, no. 4

Amphoriskos MC

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 tomb 5

Procession of women, young kanephoroi

CVA Philadelphia 2, pls. 24–6 (University 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 on stool with child in arms flanked by sphinxes; siren

Perachora II, 225, no. 2216, pl. 73

Pyxis

LC?

Grave between Corinth and Sikyon

Procession, woman with child among females, banquet and washing of feet, dressing scene

CVA Bibliothèque Nationale (Cabinet de Medailles) 1, 14–5, pl. 17 (94); Amyx 1988 I 229, no. 1(by the Skating Painter); NC 878; CallipolitisFeytmans 1970, 51–2, no. 3, fig. 3

Pyxis

MC

?

Procession of women, dancing, young kanephoroi, spinners, children and woman with baby

CVA Munich 3, 40–2, figs. 6–9, pl. 144.5–6; 145. 1–2; Amyx 1988, I 229 (Munich 7741)

Pyxis without handles

MC

?

One frieze on lid and another on body: both with chains of women with wreaths, flutists, kanephoroi, one woman with child on her lap

Leningrad 2961; Amyx 1988, I 229, possibly by the Skating Painter

72

Fa m i l i e s i n t h e G r e c o - R om a n Wo r l d

Pyxis

MC

Corinth Potters Quarter (from inside north long building)

Woman holding a child, seated woman with spindle

Corinth XV ii p. 141, no. 718, pl. 103, (KP 2476)

Pyxis

MC

Perachora

Dancing women, girl on a throne?, girl behind a second throne, woman with spindle?, procession of women with wreaths

Perachora II, 171–2, no. 1783, pl. 77

Pyxis

Early MC

Perachora

Dancing women, Perachora II, kanephoros 212–3, no. 2066, pl. 77

Pyxis lid

EC/MC

Corinth

Boar hunt scene, woman pulling child, 2 more figures, sphinxes and bird. Inscriptions: Agamemnon, Dorimachos, Alka, Sakis

Lorber 1979, 45–6, no. 52, pl. 14 (Munich 327) by the Dodwell Painter; MuZ pl. 45.192; Amyx 1988, II 205, no. 1, pl. 86

Bottle

MC

Delos

Dancing women, kanephoroi, woman with child

Jucker 1963, pl. 17.1, 3, 5, 6 (Beziers IN 22); Amyx 1988 I 230 (no image) by the Beziers Frauenfest Painter

Bottle

MC

?

Dancing women, kanephoroi, woman with child

Jucker 1963, pl. 17.2, 4 (Montpelier M. Fabre 403; 127 (SA 197)); Amyx 1988, I 230, pl. 98.2; by the Beziers Frauenfest Painter

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, girls, kanephoros

Bottle ?/ Oinochoe

MC

Corinth

Woman combing Corinth VII.2, pl. a girl’s hair? 29.164, 43.164 (CP 2485)

Oinochoe

EC/MC

?

Capture of city, mythological?, man, woman and 2 children escape

Olpe

MC

?

Dancing women, Jucker 1963, pl. 23. 3 (Pr Coll; young girl and once on display kanephoros at the Cleveland Museum of Art)

Olpe (Chigi vase)

LPC

Tomb at Monte Aguzzo, near Cezena

Boy-flutist, boys hunting

Amyx 1988, I 31, no. 3; Hurwit 2002 (Rome, Villa Giulia 22679)

Plate

MC

Perachora

3 women dancing (head bands and long peploi), dog, kanephoros

Perachora II, 194, no. 1951, pl. 77 (fr. D 115)

Jucker 1963, pl. 20. 2 (BM 65.7-20.20); NC 1070; Amyx 1988, I 230, pl. 99.2 by the London Frauenfest Painter

CVA Bibliothèque Nationale (Cabinet de Medailles) 1, 11, pl. 11.9-11 and 12 (179)

74

Fa m i l i e s i n t h e G r e c o - R om a n Wo r l d

Column krater

Early LC

?

Departure of warrior, young boy in white, blond hair, nameless touches the chin of the warrior who looks back towards 2 women and a man; chariot with another warrior completes the scene

Lorber 1979, 64–6, no. 99, pl. 24 (Berlin 1959.1) by the Hippolytos Painter

Column krater

LC II

Caere

Departure scene: chariot at the center of with a wedding couple; one man and two women in front of it (inscriptions), two couples behind; more figures dispersed e.g. a boy named: HOLOIUS? (o pais?), snake

Lorber 1979, 81–2, no. 127, pl. 39 (Louvre E 637); Amyx 1988, II 575, no. 71

Column krater

LC

Caere

Departure of Amphiaraos

Lorber 1979, 78–9, no. 122, pl. 37 (once in Berlin F1655); MuZ 210–11, pl. 42.179

Column krater

LC

Caere

Banquet scene with Eurytios, Iole and the suitors

Lorber 1979, 23–5, no. 23, pl. 5 (Louvre E 635); 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 – comments

Select bibliography

1.

Wooden painted panels

LC

Pitsas

Pinax A: procession towards an altar. Inscriptions: part of the painters’ signature: [---] korinthios, dedication to the nymphs, and names of the three women: Euthydika, Eukolis, Ethelonche

Lorber 1979, 93–4, no. 154, pl. 45,46 (Athens NM Inv A 16464/65/66/67); Orlandos, A.K. in EAA VI 1965, 201, fig. 225–6; Larson 2001, 232–233; WalterKarydi 1986, 27, fig. 1. Dilllon 2002, 228–9

2.

Votive Relief

Late 5th or Corinth early 4th c. BC

Familysacrifice, procession including a young boy

Corinth IX, 128–9, no. 267 (199)

3.

TC figurine

Archaic

Corinth, Sanctuary of Demeter and Kore

Handmade; seated woman with necklace, on her lap another seated female, and on the lap of that one indication of swaddled child

Corinth XVIII.4, 70–1 (13558); Langdon via email, publication forthcoming

4.

TC figurine

Archaic

Corinth, Sanctuary of Demeter and Kore

Imported; partially preserved bust of woman in peplos, head of child against her breast

Corinth XVIII.4, 70–1 (MF 71-54); Langdon via email, publication forthcoming

76

Fa m i l i e s i n t h e G r e c o - R om a n Wo r l d

5.

TC figurine

Archaic

Corinth, Sanctuary of Demeter and Kore

Handmade, small; standing figure possibly holding a child

(MF 10324) Langdon via email, publication forthcoming

6.

TC figurine

Archaic

Corinth

Monkey with swaddled baby; parody of kourotrophos

Corinth XV.2, 178, no. 3, pl. 39 (KT 30-3)

7.

TC figurine

EC/MC

Corinth, Anaploga Well

Handmade; woman and child?

Corinth VII.2, pl. 85, An 347 (MF 11645)

8.

TC figurine

5th c. BC

Corinth

Woman holding a girl on her shoulder

Corinth XII, 34, pl. 10. 140 (1587)

9.

TC figurine

5th c. BC

Corinth

Woman with child

Corinth XII, 34, pl. 10. 141 (3457)

10. TC figurine

Late 6th early 5th c. BC

Corinth, trial excavation at Penteskoufia

Grotesque female figure, imitating pregnancy?

Corinth XII, 35, pl. 10.153 (3456)

11. TC figurine

5th/4th c. BC?

Corinth

Seated boy

Corinth XII, 38, pl. 13. 191 (1409)

12. TC figurine

Archaic

Corinth, Circular South Shrine

Temple-boy*

HadjisteliouPrice 1969, 96, pl. 21.7

13. TC mold

Archaic (?)

Corinth, Potters Quarter

Temple boy

HadjisteliouPrice 1969, 96, 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; 3 examples total, but not clear if all Archaic

HadjisteliouPrice 1969, 96, n. 20 and 99–100

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 5th c. BC and later

Demeter and Kore Sanctuary: from votive deposit with female figurines

Temple boy with pilos and clock, ‘shepherd boy’

HadjisteliouPrice1969, 99, 105

16. TC figurine

First half of 6th c. BC

Paestum, Heraion at Foce del Sele

Seated kourotrophos with child; (a second one preserves only the infant)**

Miller Ammerman 2007, 135, fig. 7.5

*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 young children youths

Kanephoroi

Other scenes with girls

Shape Alabastron

1

Aryballos

1 1

Amphoriskos

2

Pyxis

5

Bottle

2

1

2

5

Oinochoe

1

Olpe

2

Krater

3

Plate

1 1 1 1

Panel Total

4

8

1

1

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 - R om 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: ASCSA. 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: ASCSA. Corinth XX: Williams, C. K. and Bookidis, N. (2003), Corinth, the Centenary 1896–1996, Princeton: ASCSA. Dates: EC: Early Corinthian MC: Middle Corinthian² LC: Late Corinthian PC: Proto-Corinthian MPC: Middle Proto-Corinthian G: Geometric

Notes   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 - R om 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 renovations 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

Bibliography 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 & Hudson. 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: Routledge. 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 - R om 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 York. 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-epigraphische 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 
of 
Age
 in
 Ancient
 Greece: 
Images
 of 
Childhood
 from 
the 
Classical
 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. Zabern. Schefold, K. (1966), Myth and Legend in Early Greek Art, London: Thames and Hudson. 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.

6

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 Polybius.2 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

Oikos/Oikia 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 - R om 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 - R om 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 interesting 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 translated 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 foundation. 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 - R om 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 propaganda 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 historians 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.

Notes   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 - R om 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 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 ἔχεν 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.

Bibliography 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. 41–61. 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 - R om 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: Droz. —(2003), ‘La place des reines à la cour et dans le royaume à l’époque hellénistique’, 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.

7

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 transmission 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 research. 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 - R om 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 another. 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 institutions 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 generations 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 prosopographical 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 - R om 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 incompatibility 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 hierophantai, 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 transmitted 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 consideration 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 hierophantês 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 homonymous 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 - R om 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 originally 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 hierophantai 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 - R om 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.

Notes   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 matrilineal 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 support.   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 abstinence); 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 (13). 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. 18–19. 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 - R om 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.

Bibliography 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. 187–195. 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, 41–50. 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 Press. 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.

8

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 inscriptions 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 representation 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 irrelevant. 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 combination 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 - R om 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 investigation 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 established 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 heroically, 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 - R om 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 representation – 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 - R om 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 interpreted 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 Amorgos. 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 simultaneous 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 weddingfuneral 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 - R om 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 observation 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 breastfed 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

Grandmothers 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 remembered 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 grandchildren 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 grandchild. 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 - R om a n Wo r l d

Conclusions 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 activities 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.

Notes   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, http://www.metmuseum.org/toah/works-of-art/04.17.4 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: http://www.metmuseum.org/Works_of_ Art/collection_database/greek_and_roman_art/terracotta_vase//objectview.aspx?OID=13001 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, –http://www.metmuseum.org/toah/ works-of-art/04.17.1 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 - R om 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. 135. 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 bibliography. 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 HolstWarhaft, 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.

Bibliography 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. 103–118. 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, 115–153. 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 & Hudson. 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 - R om 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 Press. 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, 629–656. 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.

9

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

Introduction 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 distribution 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 - R om a n Wo r l d

Background 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 realignments accompanied by a partial shift in the settlement pattern. Over the following two centuries, a progressive increase in social complexity and stratification is found in both funerary and settlement evidence.6 The transformation of Venetic settlements from loose aggregations of huts to more structured protourban 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 - R om 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 organization 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 inscriptions 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 - R om 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 adjectives 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 osteological 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 - R om 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 underrepresented. 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 article.



129

Fa m i ly R e l at io n sh i p s 250

Este

200

Gazzo Padua

150

Montebelluna Altino Other

100

50

0 Number of urns 160 140 120

10th–9th 8th–7th

100

6th–4th 80

Late 4th – AD25

60

ND

40 20 0 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 - R om a n Wo r l d 250

200 0–13 14–20

150

21–40 41–60 ND

100

50

0 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 combinations 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 - R om 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 90 80 Este

70

Montebelluna

60

Covolo

50

Altino

40

Valle Other

30 20 10 0 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 eighteenth 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 impossible. 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 recognition.51 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 - R om 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 occurrence 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 - R om 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 relationship.72

Conclusion 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 osteological 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 resistance.73

Notes   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 inhumations. 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’ metronymics 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 accompanied 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 - R om 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, III, IV, V, VI, VII, VIII, IX, X, XII, XIII, XIV, XV, XVII, XVIII, XIX, XX, XXI, XXII, XXIII, XXIV, XXV, XXVI, XXVII, XXVIII, XXIX, XXX, XXXI, XXXII, XXXIII, XXXIV, XXXV, XXXVI, XXXVII, XXXVIII, XXXIX, XL, XLbis, XLI, XLII, XLIII, XLIV, XLV, XLVI, XLVII, 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, 1988). 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. 308–309). 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, IV, V, VI, VII, VIII, IX, X, XII, XIII, XIV, XV, XVII, XVIII, XIX, XXI, XXIII, XXIV, XXV, XXVI, XXVII, XXVIII, XXX, XXXI, XXXII, XXXIII, XXXIV, XXXV, XXXVI, XXXVII, XXXVIII, XXXIX, XL, XLbis, XLI, XLII, XLIII, XLIV, XLV, XLVI, XLVII, XLVIII, LIII, LIV, LV, LVI, 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

64 65 66 67 68 69 70 71 72 73

Fa m i l i e s i n t h e G r e c o - R om 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: FougontUrkleios; Este inscription 47: Fugia Urkleina. Chieco Bianchi and Calzavara Capuis, 2006, p. 317). Pellegrini and Prosdocimi, 1967, pp. 226–227. 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. Pellegrini and Prosdocimi, 1967, pp. 224–226. 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). Pellegrini and Prosdocimi, 1967, p. 255. Bandelli, 1999, p. 659. 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). Pellegrini and Prosdocimi, 1967, pp. 251–253. 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). 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.

Bibliography 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: Longanesi. —(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 - R om 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.

10

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 occupational 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 intersection 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 - R om a n Wo r l d

Methodology 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 interpreted 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 differences 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 significant. 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 - R om 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 Slavery 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 preliminary 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 affiliation 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 occupations 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 - R om 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 occupations. 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 households 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 traditionally ‘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 columbaria 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 commemorative 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 - R om 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 information – 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 households 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 columbaria, 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 specializations 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 understanding 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 - R om a n Wo r l d

Figures Yes

ML 3926-4326 Name Gender Status

Role

Male

MLD 4327-4413

MM 4418-4880

MS 6213-6640

Age

Yes

No

Job

Yes

No

No

MV 7218-7394

Female

Uncertain free Freeborn Freed Imperial need

Slave Imperial slave Single name Unknown

Commemorated Dedicator

Number

Other Unknown

Order

Copy record

Category Collegium

Yes

No

Family

Yes

Marriage

Yes

No

Slavery

Yes

No No

Children

Yes

No

Name

Yes

No

Duplicate

Yes

No

Next record

Figure 10.1 Database

13

13 33.9

49.9

27.6

72.6

72.4

33.4

51.9

87

87 66.1

50.1

Job

27.4

Age

Collegium Marriage

Figure 10.2 Characteristics by gender

66.6

48.1

Children

Other family

Patronage or slavery

Whole sample



153

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

Singe name 22% Uncertain free 39% Imperial slave 7%

Slave 11% Freed 15% Freeborn 1% Imperial freed 5%

Figure 10.3 Status distribution 6.2 27.5

22.9 40.3

27

33.4

44.4

76.1 59.7

55.6

Uncertain free

Freed

Imperial freed

93.7 72.5

Imperial slave

Figure 10.4 Status distribution by gender

Slave

73 66.6

Single name

Whole sample

154

Fa m i l i e s i n t h e G r e c o - R om a n Wo r l d

29.7

30.1

70.3

69.9

62.9

Livia

Drusus

Marcella

30.6

37.1

41

69.4 59

Statilii

Volusii

33.4

66.6

All

Figure 10.5  Gender distribution among the columbaria

17.1

10.8 18.8

6.9 13.1 4.4

29.3

19

0

1.9 20.2

8.8

0

52.3

10.8 5.1

1

Livia

38.8

30.1

Drusus

22.4

Marcella

Statilii

Volusii

Figure 10.6 Status distribution among the columbaria

Imperial slave Slave

Freed Freeborn Uncertain free

51.7

0.5 36.2

Single name

Imperial freed 15.4

0.3 15.5

7.3

2

6 11.3

21.6

0 19

0.3 4.5

0.2

40.7

2.7

14.6

14.4

18.5

6.5

All



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

Administration 15%

155

Household 12%

Trade 7%

Appearance 6% Attendance 20% Entertainment 5%

Artisan 5%

Military 0% Religion 2%

Food 2% Literature 3%

Childcare 5% Medicine 3%

Figure 10.7 Occupational categories

Money 4%

156

Fa m i l i e s i n t h e G r e c o - R om a n Wo r l d

All Volusii Statilii Marcella Drusus Livia Attendance

Administration

Household

Clothing

Trade

Appearance Medicine

Entertainment

Childcare

Artisan

Money

Literature

Religion

Food

Military

Figure 10.8  Job distribution among the columbaria

Attendance 17% Clothing 33% Appearance 14%

Childcare 14% Religion 3%

Artisan 2%

Medicine 3%

Literature 3% Administration 4%

Figure 10.9  Female occupational categories

Entertainment 7%



157

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 11.5 20.5 9.6 10.8 10.5 11.3

Family

7.2 16.6 5.6

Children

8.5 5.3 4.4 18 40.5 11

Marriage

16.4 22.6 18 26.7 23.9 35

Patronage/slavery

15.9

29.3 32.1 20.7

17.6

Job

30.6

9.8 28.6 23.5 6.8 3.4 4

Collegium

5.2 14.6

0 10.9 12.2

Age

17.4

10.2

22.6 1.8 0

5 All

10 Volusii

15

20 Statilii

25 Marcella

30

35 Drusus

40

45

Livia

Figure 10.10 Overall differences among the columbaria

Notes  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 - R om 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 Sciences.   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 amphitheatre 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.

Bibliography 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.

11

Houses, Painting, Family Emotion Natalie Kampen

Introduction 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 experiences 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 - R om 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 - R om 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 - R om 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 - R om 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 interventionism 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 intervention 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 - R om 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 articulated 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 prescriptions 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 - R om 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 interpretation 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 understand 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 - R om 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

Notes   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 (translated 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 - R om 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. 117–154. 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 adquiescit (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 translation 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 monimentorum 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 adplicueris 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. 145–714. 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).

Bibliography 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 - R om 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 Press. Jauss, H.-R. (1982), Toward an Aesthetic of Reception, Minneapolis: University of Minnesota Press. Kassel, R. (1958), Untersuchungen zur griechischen und lateinischen consolationsliteratur, 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: Brill. 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 - R om 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 Series. —(2004), Mit Mythen leben: Die Bilderwelt der römischen Sarkophage Munich: Hirmer. Zevi, F. (2004), ‘Cicero and Ostia’, Journal of Roman Archaeology, Supplement 57, 15–31.

12

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 everything 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 SigismundNielsen 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 - R om 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 multifamily 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 more. 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 of 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 paidagogoi 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 distinctions, 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 - R om 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 of 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 networks. 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 establishes 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 - R om 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 of 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 contemporary 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 unconventional 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 - R om 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 marriage. 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 eighthcentury 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-developing 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 of 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.

Notes  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 - R om 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 households.   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, 2006. 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 of 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 sympathetic.’ 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 unpublished 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 forthcoming, 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

Bibliography 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, 141–70. 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 diplomatique 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 - R om 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, 398–399. —(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, 57–91. 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 GrecoRoman 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 productivity 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 of 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, 109–118. 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 inheritance 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 - R om 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: Oldenbourg. Shaw, B. D. (1992), ‘Explaining incest: brother-sister marriage in GraecoRoman 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.

Index Words in italics are in Latin or are titles Numbers in italics are figures adelphos 88–9 adult 61, 128, 130–1 aesthetics 6–7 affection 86, 88, 89, 91, 111 affiliation 19, 111, 145, 147, 151 age 52–3, 62, 127, 130–1, 152, 157 aristocracy 17–18 Aristotle 18, 181 Athens 14, 18, 33–4, 37, 95, 182 Augustus 32, 143, 147–8 bone 124–37 boyhood 65–6 brother 88–9, 98–9, 101, 111, 185–6 burial 69, 80n, 121, 124, 128, 131, 136 childbirth 109–10 children 4, 13, 52, 60, 69, 86, 98, 101, 110, 114, 128, 131, 133, 136, 150, 165–7, 181 Cicero 6, 18, 168 citizenship 33–7 columbaria 143–51 daughter 101, 108–9, 185–6 death 106–7, 109–16, 161–3, 167 demography 27–9, 31–3, 38 discrimination 48–50, 52–3, 55 domus 12, 14–19, 86 emotion 11, 116, 159–62, 167, 169–72, 181 see also grief epigram 106–8, 116 see also inscription epigraphy 121, 136 epitaph 113, 116, 121, 131–6, 144–6, 160–70, 179 familia 12, 17, 31, 38 see also family family 1–7, 10, 15, 19, 28–30, 50, 85, 106, 113–14, 121, 136, 144, 151, 167, 179–87 see also familia

history 27 life 30–1, 38 nuclear 2, 14–15, 29, 131, 179–80 role 145, 170 unit 150 fate 162, 165, 167–8 father 4, 15, 17, 87–8, 96, 107, 110–11, 114, 163, 165–8, 181–5 funeral 108, 114 gender 3, 48–55, 143–6, 151, 152–4, 169 genealogy 13, 114 genos 96–7, 101–2 genos gunaikôn 52 grandmother 115–16 grave 69, 124 grief 106, 161, 170, 172 see also emotion Hesiod 18, 52–5 Theogony 52 Works and Days 48, 51, 52–5 household 1, 2, 3, 14–15, 19, 28–30, 35–6, 51, 54–5, 69, 85, 143, 145–51, 179–81, 187 housing 37–8 husband 53–4, 89, 111–13 iconography 60, 67, 90 imagery 6, 107, 167 inheritance 15, 16, 95–7, 182, 185 inscription 86, 95, 106, 112, 116, 124, 126, 131–6, 144, 148, 170 see also epigram intersectionality 48–55 king 84–5, 87–9 kinship 11–19, 86, 121, 126, 133–4, 136 language 2, 4, 6, 84, 87, 90, 122, 170 law 4, 11, 33, 35–6, 50, 85, 181–2 loyalty 84, 88, 89, 184 manumission 145, 147, 151

194 I n d e x marriage 3, 12, 16, 28, 33–6, 53, 89, 98, 108–9, 113, 145, 152, 157, 185–6 see also wedding intermarriage 34–6 memory 6, 160, 171, 183 mortality 106, 167, 182 mother 53, 61–2, 89, 108–11, 114–16, 168 grandmother 115–16 name 54, 126–7, 131–6, 144–7, 152–3 oikia 50, 85–6, 179 oikos 12, 14–15, 18–19, 31, 48–55, 60, 68–9, 85, 95–8, 101–2 painting 60, 61, 66, 159–60, 165, 167, 170–2 parent 107, 111, 165, 167 paterfamilias 16–17, 165 patriline 95, 98, 101 polis 18, 50 politics 19, 27, 32, 35–6, 38 population 31–3, 36, 128, 149 property 36–7, 50, 85, 182, 185

Rome 14, 17–18, 33–5, 38, 143, 170 sexuality 52–3 sibling 34, 89, 101, 111, 131, 184–6 sister 89, 111, 181, 184, 185–6 slave 4, 145–7, 149–51, 152–5, 181 social network 11, 18, 149 social relation 3, 11, 13, 50 social status 51, 107, 114 son 88–9, 95–100, 183 space 36, 50, 172 domestic space 186–7 succession 95–8, 101–2 twin 110–11 vase 49, 60–2, 66, 70 virtue 53, 54, 112–14 wedding 64–5, 108–9 see also marriage wife 53–4, 89, 98, 108, 111–13, 149 woman 52, 61–5, 106–16, 149

queen 89, 116

Xenophon 37, 51–5 Oeconomicus 48, 50–5

relative 15–16, 30, 90, 121, 124, 131–6

youth 65–9, 77