House as Symbol

As Anna Boozer remarks:

Roman domestic spaces provide an ideal nucleus for exploring identities and memories because the Roman house served as a vessel for the cultural identity and memory practices of its inhabitants. Although Roman houses differed architecturally to varying degrees across the Empire, they retained a similar cultural role.¹

Boozer’s summary is a prelude to her discussions of house B1 at Trimithis with its suite of wall paintings, including scenes of dining, a personification of the Polis, and an Odyssean scene. Although seeing them as “Roman,” Boozer sets the paintings within the cultural complexity of a postconquest society, in which Greek, Roman, and Egyptian elements were all present.² Boozer’s use of the ethnic marker is subtle, pointing to the microhistorical concern with immediate contexts that will here be adopted. In the same way, the cities of Egypt in this period were Roman, not because they were identical with cities of other regions (they exhibit strong local and regional cultural markers), but because they were constructed within an imperial culture which, in one way or another, they reflect at a local and immediate level.³

As Jen Baird notes, much of the historiography and archaeology of the Hellenistic and Roman East has focused on the ethnic labeling of cultures.⁴ This obsession is in part historiographical, connected to the hegemonic narratives of Hellenization and Romanization. But cultural markers can also be shorthand for a variety of other social, political, and economic characteristics.⁵ Colonial discourses have tended to regard social behaviors as dependent on ethnicity and thus to reflect cultural values with deep histories. Such a separation between the colonized and the colonizer is, of course, highly tendentious and ideological: it suits colonial discourses to emphasize cultural differences so as to maintain political and economic inequalities and justify imperial endeavors. Although one might be able to imagine a colonial society in which there were cultural-ethnic groups who experienced entirely differently the everyday, it seems more likely, certainly in ancient societies, that colonial divisions were highly porous, with multiple shared experiences and repeated traversals of cultural difference, as Boozer’s examples would seem to show.⁶ Ethnic indicators may have been important as references for identity and consequently have influenced behavior, but the significance of such influences, especially when considered alongside the other constraints operative on the household community (economic, social, cultural, and political), must be open to doubt.⁷ Rather than take ethnic markers as a symbolic manifestation of socioeconomic practices of production and reproduction, I see those practices as the context in which those markers appear, as will be explored in the final paragraph.

Such an approach diverges from anthropological-phenomenological methodologies, such as Pierre Bourdieu’s 1960 (1979) influential rendition of the Kabyle house. His interpretation uncovers such rich layers in the traditional Kabyle house that we are encouraged to “read” it as a microcosm of the Kabyle cosmos. Bourdieu was likely influenced by Bachelard’s 1958 La Poétique de l’Espace and stressed the phenomenological over the sociohistorical context. As critics have shown, by 1960, the traditional Kabyle house was an anachronism that functioned as a symbol of national identity.⁸ In a complex and multifocal society, it was no longer, if it ever had been, the structuring structure, the Kabyle world in microcosm. Instead, the house was invested in a nostalgia which was counterpoised to modernity.

The poetics of such descriptions feed into our preferences as anthropologists and archaeologists for the visual.⁹ Yet, the house was not primarily a symbolic space: it was a functional space, existing within a social dynamic that cannot be disengaged from social production and reproduction.¹⁰ Any symbolic value of the house was generated in the everyday experience of the inhabiting community, and since that everyday was varied and engaged in wider contexts, the habitus could not be an unchanging essence.¹¹ Bourdieu’s poetics played into the view of cultures as “deep structures,” embedded in historical time.¹² Reading the domestic as the microsite of the cultural macrocosm makes it a primary location of acculturation and the cultural origo of the nation. In modernity, it operates in relation to changes in the nature of public space, seen as susceptible to capitalistic or imperial transformations, so as to create a dichotomy between public and private. Consequently, there often follows a discursive focus on women as guardians of a cultural tradition that needs to be secured within the domestic sphere. The dichotomous readings of public and private therefore can reinforce a particular and conservative gender politics.¹³ Any modification in domestic space then can be seen as having increased significance as symbolic of transformation in the “deep structures” of a society.

Such attitudes seem very different from those visible in the documentation from Romano-Egyptian rural communities, but we can see similar attitudes in Philo’s discussion of the “Jewish house.” In Special Laws 3.173–75, Philo argues for the exclusion of women from markets, courts, crowds, and their limitation to the domestic sphere. Men could not confine themselves, but in a hostile public environment, the safety of the house and the confinement of women therein becomes a way of preserving Jewish traditional values.¹⁴ Philo’s example seems necessarily and consciously exceptional and points us to a norm in which the functioning of the domestic was freighted differently with symbolic value and in which the boundaries of public and private were more permeable.

The House as a Location of Production

In rural Egypt, a house provided the technology required for survival. This involved shelter; food storage, preparation, and consumption; and community. The primary productive space within the domestic complex appears to have been the aule (the courtyard).¹⁵

Spatially, the house and aule were distinct, and an aule could be sold separately from its neighboring house. Yet, judging from the excavations in Karanis, every house required access to such a yard. The court might be peripheral and bordering another house and with a separate entrance.¹⁶ In the tiny houses excavated at Tebtynis the courts were in a corner (Maison 5300) or at the back (Maisons 4200, 5200, 6300).¹⁷ In her discussion of courtyards uncovered in the recent archaeological investigations at Karanis, Simpson (chapter 2, this volume) shows that courtyards were often transitional places, open to the street to the extent that they seem to have operated as passageways between residences. Many were not, it seems, gated. As Davoli (chapter 1, this volume) also shows, in the Fayum villages, the differences between street, alley, and court appear to have been negotiable, and if these areas were fully or partially roofed, within a labyrinthine village structure, distinctions between public and private were not clearly marked or, as seems likely, experienced.

The equipment and detritus of everyday production have been found in the courtyards. They were equipped with bins and storage facilities. Cooking took place in the aule. Some had ovens, perhaps for the baking of bread, though bread baking may have progressively moved from a domestic to a more commercial provision (Simpson, chapter 2, this volume). Small mills have been found, both rotary and Theban. It looks very much as though the preparation of food was a communal activity, and since the spaces of the aule were not rendered fully private and court spaces could be shared, the community performing this everyday task was not necessarily confined to a particular house. Such activities were performed alongside other productive economic activities. There were animal pens and feeding troughs in many courts. Some houses had dovecotes.¹⁸ Weaving equipment was found in the yards at Tebtynis.¹⁹ It is evident that courts were of central importance to domestic production.

Houses varied.²⁰ Taking the ground plan area of the houses at Karanis as an indicator, some houses were tiny, with a ground plan area of under 20m² (houses C55; C75; C5036; C5034), with the bulk of houses at under 100m², with some larger examples. The excavated Tebtynis houses were also small (C5300 is under 60m²; C6300 about 70m²; C3100 about 72m²), though the cluster of excavated houses is not necessarily typical of the village as a whole. Boozer’s figures for the Dakhla Oasis suggest a range of 99.2m² and 415m², so larger than those at Karanis. At Trimithis, House B2 is 121m² and would appear to have contained a very different household than the elaborate B1 at 225m².²¹ Considering variations in the number of stories, the balance between built and open space, and complexity of layout, we are evidently looking at a range of houses, which also seems apparent from the surviving house prices.²²

At Karanis, this range was limited: we do not have elite residences. The houses were centers of small-scale agricultural production with some differentiation in function and scale of activity within that agricultural economy, as reflected in size and decor. The relative sizes and decoration of the houses suggest a level of economic differentiation, and the presumption must be that the variety of houses was a reflection of differing status. One presumes that in a community as small as Karanis, the population of which was likely under 3,500 in the second century CE,²³ everyone must have known everyone else and all inhabitants would have been aware of microdifferentials that are for us, as outsiders, difficult to map.

The documents allow a degree of resolution unique for the Roman Mediterranean in our understanding of the house and its community. Yet the understanding that can be gleaned from those documents is often fragmentary.²⁴ Papyri found in house C123, for example, show that it belonged in successive periods to the family of Satabous and that of Gaii Iulii Sabinus and Apollinarius.²⁵ The structure consisted of two distinct but clearly related elements: a domestic structure which was typologically little different from the neighboring house and a large granary. The granary was clearly far too large to be for production by the associated household and must have provided grain storage and perhaps other administrative services for a significant number of other villagers. The granary was thus embedded within the modes of production of the village. The documents, however, tell us nothing about the operation of the granary. This is in spite of the richness of the record. For the Iulii, the documents attest to multiple economic activities, including farming and military service. They show that the house functioned as one location within a network of partially documented economic and social relations. To these activities, we ought to add the management of the granary. Consequently, any reconstruction of the economic and social activities of a residential group needs to be aware of probable evidentiary gaps and to deploy anthropological and economic models, and considerable imagination, to bridge those gaps.

Census returns show that the coresidential family was a social organization that existed in multiple varied forms.²⁶ Evidently, those forms were neither ideal nor planned: families were subject to the vagaries of fertility and mortality.²⁷ Nevertheless, families had agency. As a primary requirement, the families of the Egyptian villages had to deploy their resources to meet their subsistence requirements. The family needed to match resources, labor requirements, and organization. The complex structures attested within the census returns suggest the predominance of a strategy of avoiding the breakup of familial units on the formation of new conjugal couples or on the death of older family members. The most obvious instances of such arrangements are the frérèches. One presumes that these came into being when brothers decided not to break up a property holding on the death of a parent. Other extended and complex familial groups fulfilled similar functions, even if the processes of their formation are more difficult to reconstruct. The prevalence of complex familial groups suggests that people tended to be kept in rather than sent out to form new residential and household units.²⁸

The advantages of large familial structures are numerous, notably in the pooling of labor and other productive resources. Even if the available labor outran the landholding of the domestic unit, it could be employed in day labor or in renting or in even in diversifying income.²⁹ Keeping males in the familial unit after marriage likely allowed earlier male marriage since the setup costs for a new conjugal unit were minimized: indeed, costs were shifted primarily to the bridal side in the requirement for a dowry (though the man’s family had to provide “gifts”).³⁰ A larger unit allowed access to broader support networks, which gave the household more economic and political muscle.³¹ Multiple sources of labor equate to multiple sources of income, and that effectively is a means of spreading risk in a preindustrial society.

In contemporary impoverished environments, managing illness, mortality, and economic fluctuations become easier in larger social units that allow diversification of income sources.³² Labor is a resource that can be deployed in a variety of locales, sometimes distant from the home, as part of the livelihood strategies of familial units in order to manage risk, increase familial income, and diversify the microeconomies of such groups.³³ Consequently, rather than studying coresidential households of impoverished groups, contemporary analyses tend to focus on livelihood strategies that encompass such factors as remittances sent from family members residing at some distances from the primary familial residence.³⁴

There were not the same opportunities for diversification within Roman Egypt as there are in contemporary African communities. Yet the ample if unquantifiable evidence for the movement of young males³⁵ may reflect such strategies. We might identify regional migration in the fluctuations in registered male population by age category in the census records or in those who went “missing” in tax records.³⁶ Tax records suggest that mobility was relatively normal for members of village households.³⁷ Although it has been suggested that some of that mobility was “crisis-mobility” and, in particular, tax avoidance, records that list males of a village having a known residence elsewhere show that not all migration was covert or crisis induced.³⁸ Those moving to other villages or to the cities were presumably primarily economic migrants, and it seems likely that they maintained close ties to their home village that went beyond issues of tax liability.³⁹

Similarly, the letters that flowed between the Nilotic communities also attest the dispersal of communities. One of the characteristics of the epistolary form in Roman Egypt is the extensive list of persons to be greeted: it is evident that such lists do not make sense in the strictly familial terms in which the relations were frequently expressed (addressing people as brothers and sisters, mothers and fathers), but do attest a social need to use the mechanism of the letter to reinforce an extensive network of relationships within the home community.⁴⁰ In the Karanis documentation, the connectivity of military families, which encompassed military and civilian spheres, may have been more diversified and extensive, but was similar in kind to that of civilian families. Such networks frequently operated with what seems to have been more than one familial base.⁴¹ Strong social links were maintained between topographically distant members of a home community and meant that a unit effectively had “branches” in different locations.

A livelihood approach moves analytical focus from a particular location to the multiple locations across which a family unit might operate. Locational diversification required a more complex structuring of power relations so as to maintain a group’s coherence and renders yet more imprecise the sociological terminology we might want to apply to such groups. Our scholarly terminologies of family, household, and houseful represent complexities in residential relations, but do not necessarily map onto the various multiple strategies employed in historical societies.⁴² Although ownership of economic resources was a major source of power, it seems likely that the dependency of the individual on the family unit for their everyday needs and the networks of support that maintained the family went far beyond issues around the possession of property.⁴³

Houses and Familial Power

The operation of authority within a familial unit can be seen in archival collections such as that associated with Sakaon of Theadelphia in the Fayum. In a petition of 312, preserved in Sakaon’s papers, Aurelius Melas petitioned the authorities about Taues, a girl Sakaon had abducted.⁴⁴ Taues was the daughter of Melas’s aunt. This aunt had been married to Sakaon, and one presumes that Taues was Sakaon’s daughter. Sakaon took a new wife, Kamoution, who supposedly expelled Taues. Taues went to live with Melas. After keeping her in his house for some time, Melas married her to his son Zoilos, who was sitologos (administrator of the grain tax) of the village. Supposedly at the behest of Kamoution, Sakaon abducted Taues and demanded bridal gifts. As a consequence Melas went to the authorities (figure 7.1 for the familial relations).

He was not entirely successful. Sakaon’s claims on the girl were supported, and Melas was instructed to provide gifts before the girl could be restored to Melas and Zoilos. But Sakaon took the gifts and retained the girl, supposedly because Kamoution intended the girl for her nephew, Sarmates. So Melas went to law again. Again, he had mixed success, since the ruling stressed the importance of the girl’s views in deciding where she resided and to whom she was married.

The families clashed again thirty years later.⁴⁵ In 343, Zoilos, now a deacon, complained once more of Sakaon. Zoilos had married his son, Gerontios, to Nonna, daughter of Annous. When Gerontios was dying, Sakaon, his brothers, and Annous broke into the house and abducted Nonna. There followed an assault on Melas. Pasis, another son of Zoilos, intervened and was also assaulted. Animals were carried off. It was the assaults and the theft rather than the abduction that led Zoilos to resort to law.⁴⁶

These conflicts allow us to trace familial relations across two generations. Taues was transferred from the Sakaon residence to that of Melas with a view to her marriage to Zoilos. The export of Taues to the family of Melas was a limited transference since Taues’s mother had been Melas’s aunt. Her marriage became an issue when Kamoution, Sakaon’s second wife, sought her for her nephew. Thus, Taues was to be used to reinforce preexisting social links in different branches of the family, but could not be used both to cement relations with Kamoution’s relatives and those of Melas. A decision had to be made and conflict resulted.

We see similar issues in the marriage of Nonna. Nonna’s relationship to Sakaon was, in modern terms, remote. She was, however, part of Kamoution’s family (figure 7.2). In marrying Sakaon, Kamoution had married her maternal cousin. Her relationship to Nonna, was through her father, Kaet. Kaet’s sister was Annous whose daughter was Nonna.⁴⁷ One might speculate that the marriage of Nonna accomplished the social task that the marriage of Taues and Zoilos had been intended to perform in bringing together these two families. Over the generations, there were five marriages through which the families of Sakaon, Kamoution, and Melas were united (that of Sakaon and his unnamed first wife, Sakaon and Kamoution, Taues and Zoilos, Taues and Sarmates, and Nonna and Gerontios). But these relationships also repeatedly dissolved, two by deaths and one by abduction. On the death of Gerontios, Sakaon and Kamoution were anxious to secure Nonna, and it was that desire to bring back a woman and probably her assets (which would account for the seizure of property) into the Kamoution-Sakaon household that sparked conflict.

Although the marriage of Taues and Zoilos was not brother-sister, it showed similar characteristics. Taues was made a member of the household before her marriage.⁴⁸ The other marital relationships in the archive show a strong preference for close-kin marriage. The marriage of Taues and Zoilos was one of cousins, as was that of Kamoution and Sakaon. The abducted Taues was intended for Sarmates, and although their relationship defies English familial vocabulary, the proposed marriage was endogamous to the household. Such marriages reinforced the familial community. Some instances display a strategy for social reproduction which, rather than reproducing the family in another location, maintained this particular household into the next generation. Other instances involved women joining households to which they were already tied through familial connections.

Endogamy potentially established high boundaries between the household and the external world, minimizing the export and import of familial members.⁴⁹ In circumstances of cultural endogamy, a new member of a household (perhaps especially a woman) might face difficulties negotiating relationships. But in the Sakaon archive, the two cases of the export of women were to the same household and reversed a prior conjugal export when Sakaon received his unnamed first wife. To describe such marriages as exogamous seems inappropriate. Endogamy and exogamy are modern anthropological categories, and in the relatively small village communities of Roman Egypt, one wonders at the extent to which anyone married “outside.” The original household retained a claim on the exported person, as we can see with both Taues and Nonna, and presumably that claim was mutual: a woman exported to another household retained a financial and social interest in the original household.

Although the microhistory attested in the Sakaon archive conforms in some regards to a patriarchal operation of social and familial power, and the legal structures of Romano-Egyptian society tended to reinforce patriarchal structures, particularly through the distribution of property, the character of those households seems to have been more complex than a simple manifestation of patriarchal authority.⁵⁰P. Sakaon 1, a census return of CE 310, reflects an official focus on the men, but the attested household would appear to be the product of nonpatriarchal processes. Sakaon registered his male coresidents, who included himself (aged forty-five), his son Aeil (aged sixteen), his brother Paesis (aged fifty-five), Ammonios the son of Pasesis (nineteen), another brother Aunes (aged forty-eight), Keletes son of Kaet and the brother of Kamoution (aged fifty), his son Sarmates (aged sixteen, and the intended spouse of Taues), Alupios son of Herodes (aged thirty-five and the nephew of Kamoution), and his son Heron (aged twelve). The presence of Kamoution’s family provides the context in which Sakaon acted as the violent defender of interests that were defined by this coresidential group. The axis around which these disputes turned appears, in fact, to have been Kamoution and her genealogical family. In the context of this household, we can understand Melas’s complaint in P. Sakaon 38 that Kamoution was the source of discord as more than an instance of the standard misogynistic trope of blaming the wife. Women were both the “assets” transferred and the agents of household strategies. This might contradict patriarchal predilections, but on an everyday basis, women were major contributors to the domestic economy (through labor and assets), reliant on their conjugal house-community, and committed to the longer-term success of their households.

The everyday workings of social and economic relationships necessary for social reproduction likely encouraged social agents to maintain the relationships within and between households that marriages were intended to reinforce. The point of marital strategies was to produce long-lasting social relationships, and in this sense the “gifting” of women should not necessarily be seen as a transfer of property, but of embedding a social agent within a household. The functioning of the familial relationships thus likely depended on a continuity of social intercourse between the familial units, focusing on the women as the point of communication. But such strategies carried within them a considerable potential for clashes of authority between familial units.

Such clashes are at the heart of the petition of Dionysia (P.Oxy. 2.237) against her father. It is a complex case that began with a dispute between Dionysia and her father Chaeremon over land that was contracted to her as part of her marriage settlement. Chaeremon had mortgaged this land and got into difficulties with the repayment. Dionysia demanded the land. Chaeremon responded by demanding return of all assets that he had bestowed on her in her marriage contract. This failed (unsurprisingly), but Chaeremon’s next move was to attempt to dissolve the marriage (which would entail the property returning to him). Dionysia successfully resisted this, arguing that there was no provision in law for a father to extract an adult daughter from a contracted marriage against her will and that this was an attempt to pervert the prior legal decision. Dionysia quoted a similar case that came before the Prefect Flavius Titianus resulting from a Sempronius abducting his daughter from her husband, Antonius. Sempronius alleged that Antonius had accused him of incest (presumably with the daughter) and that he was within his rights in dissolving the marriage. Antonius responded by claiming that a father had no rights over the dowry or the woman whom he had given in marriage, a position the prefect supported.

Even if they failed to break up their daughters’ marriages, these fathers, like Sakaon, felt that they had rights over their daughters that extended into other households. Such claims were not necessarily patriarchal (the household of Sakaon was also the household of Kamoution), but the requirement to control social and economic resources and preserve the integrity of the familial unit on which all depended necessitated adherence to familial hierarchies, the offense to which led to conflict in these cases.

The adventures of the women of the Sakaon-Kamoution family suggest the need for nuance in understanding the social and symbolic significance of the house. The perceived need to abduct the women and the use of violence in so doing attests to the importance of location in controlling resources and protecting individuals. Sakaon’s seizure of the women gave him control over the key “assets” in the dispute with Melas. Within the physical space of the house, the household was able to exercise authority over and protection of its members.

The everyday productive activities by which the household maintained itself likely reinforced the household’s dense social bonds and further integrated the women. The fact that familial authority extended beyond the boundaries of a particular house did not render those boundaries insignificant. The house remained the location of everyday productive activity and intense social interaction. As the scene of production and reproduction of the family, it must have carried powerful symbolic value within the village, providing identity for its residents. Nevertheless, it was but the primary spatial location of a family’s social and economic activity.

Houses as Property

The Sakaon archive suggests a desire to integrate the household and maintain the house as a place of control. That social pressure was in some tension with inheritance practices that fragmented property ownership. In the Sakaon case, even though the family had various properties in the metropolis and the village, the residence of Kamoution and her family was in the busy village house of Sakaon.⁵¹ In other instances, a house might be divided into multiple, sometimes very small fractions.⁵² Sales were sometimes used to consolidate property within the family.⁵³ Other sales were between neighbors who might also have been relatives.⁵⁴ In some instances, jointly inherited property was sold, perhaps to simplify the family’s property holdings.⁵⁵

House sale documents show that families often owned multiple houses and clustered those houses together in a way that offset any property divisions. To take two examples, P.Mich. 5.277 documents the sale of a house in Tebtynis in which the seller co-owned a neighboring house with the buyer and another neighbor was the vendor’s brother. In P.Mich. 5.276 five siblings sold a seventh share and a half share of a house and aule, which they had inherited from their father, to someone who looks from his patronymic to have been a relative. Two of the siblings owned a neighboring house and aule. The very fact that we have sales documents for transactions within extended families suggests that fragmented ownership could cause tensions, but they also point us away from the house as an insular unit, representing the family.

We can see the familial management of a cluster of houses in a small archive from Oxyrhynchus. The weaver Tryphon bought a half share of a house from his cousin Pnepheros. The house had been divided between Thamounis, Tryphon’s mother, and her unnamed sister. In the next generation, the half share passed to the sister’s son, Pnepheros. Thamounis was living, but it is a safe assumption that Tryphon expected to inherit her share, so that the house would, at the point of her death, once more be united. Neighboring lots were owned by Thamounis and Tausiris, sister of Pnepheros.⁵⁶ The original arrangement, then, seems to have been a plot of three houses, owned by a single family, which became tenurially divided between siblings and cousins and which the familial unit cooperated to maintain control over.

A further example comes in a census declaration (P.Oxy. 12.1547) from early second-century Oxyrhynchus from a family of stone cutters. Petosiris, son of Dionusios, grandson of Petosiris, registered a half of a house. Of the remaining half share, two-thirds (one-third of the total house) belonged to his wife Tetoeus daughter of Thoonis, one-third inherited from her father, one further third from her brother Petosiris. The final sixth of the house was owned by Papontos, a third sibling. Thirteen people were registered in the declaration, including a son of the late Petosiris, brother of Tetoeus. It seems clear that a single house belonging to a Petosiris was inherited by Dionusios and Thoonis. It was then passed onto the next generation when one half share was inherited by Petosiris and the other by Tetoeus, Petosiris, and Papontos. The marriage of the cousins Thoonis and Tetoeus brought some tenurial consolidation, which was furthered by the death of the second Petosiris. But the entire family were coresident and all the men were stone cutters. It seems probable that they operated together commercially and as a single domestic unit.

Such tenurial divisions were probably not reflective of living arrangements. Husselman reports some interior doors with bolts and bars, though predictably most lockable doors were to the street. She has also has plates showing bricked-up doorways.⁵⁷ Small houses, with single points of entry, single staircases, and seemingly one route of access round the house, could not easily have been separated into multiple units. The same considerations applied in the tiny Tebtynis houses.⁵⁸ There may have been tensions, but houses required the occupants to use the communal space of the aule, and it seems likely that such communal living was a social and economic necessity.

The fundamental question I seek to answer is, What are we looking at when we look at a rural house in Roman Egypt? The first part of my answer begins with theory. I argue that anthropological views of the house have embedded within them certain features of a colonial viewpoint. The house is seen in contrast to public space. It is valued primarily as a culture-producing machine. As the scene of family life, the house is portrayed as a deep social structure, the defender and inculcator of tradition. Such views make the house a site of nostalgia and cultural resistance, marginal to the “real history” of cultural and economic development that was taking place in the public spaces. By contrast, I argue that the Romano-Egyptian village house operated not as a point of symbolic resistance, but as the primary location of everyday production and consumption, by which the familial unit survived from day to day, and of social and biological reproduction, by which the familial unit survived from generation to generation.

Our historical analyses are attracted to the relative clarity of visible spaces. When we look at a house, we are tempted to believe that we are looking at the familial unit. It is an expectation that works quite well for modern bourgeois society. In the villages of Egypt, there was a link between the space on the ground (the house) and the familial unit, but that association was imprecise. The house operated as a central point, but there was fuzziness in the relationship of spatial and social boundaries. The meaningful familial unit extended beyond the coresidential group. The economic requirements to generate livelihoods carried individuals to distant places, but they remained members of a topographically concentrated community. Women transferred from one residence to another, but retained ties to the original familial unit.

The administrative process of registering a coresidential group tells us very little about the dynamic process by which that group operated.⁵⁹ For the household, there was no clear rule as to who belonged and who did not. The endogamous practices of the Romano-Egyptians knitted families and households into complex structures that show little affinity with, for instance, the simpler patriarchal structures of the Roman family.⁶⁰ When the extended family lives next door; when endogamous practices are used strategically to maintain the integrity of localized property holding; when the export of women to other households (if it happens) does not break their tie to the natal household; when such transfers are repeated interchanges between the same familial groups over generations; when men might be expected to work in distant locations so as to diversify and maximize the livelihood of the familial unit; when the boundaries between public and private, between courtyard, house, and street, are ill-defined; when familial units use house sales to consolidate and rationalize property holding within topographically concentrated familial blocks, then the boundaries of house, family, and community lose clarity and the family appears more communal than domestic. A house might represent a concentration of the hierarchically organized power of the household, dominated by its leading figures, but the membership, interests and boundaries of that social unit extended beyond any particular architectural structure.

The house was the scene of the economic, social, and biological processes by which the familial unit was reproduced day to day, year to year, and generation to generation. In circumstances of poverty, close control over familial resources (people and property) was essential. The functional importance of familial units accounts for their social power. Any individual (male or female) required a familial unit.⁶¹ The familial unit was productive of livelihoods and social relationships, established social status, and defended individuals in the competitive environment of the village. The power structure of the familial unit looked inward and outward. On the inside, it managed social relationships and took the decisions, working within economic, demographic, and cultural constraints to give shape to the household. On the outside, the unit was competitive with other similar units and necessarily had to engage with political and administrative structures. The house was a topographical center for the functioning of a familial unit, but it was not a boundary that limited and contained those activities.

My argument has tended to downplay the house as a signifier of cultural identity. Although the macronarratives of acculturation and political identity place considerable emphasis on ethnic markers, it is only if we assume that such markers were signifiers of modes of social reproduction that we would associate their adoption with fundamental economic and social change. The basic challenge that faced the inhabitants of these rural communities was not whether they were Roman, but how to manage everyday issues of production and reproduction and how to protect and advance the interests of their familial units. In such a context, it becomes easier to understand why new ethnic markers might be adopted in local competitions for status and power. These houses were Roman since they sat within the spatial and cultural frames of the Roman Empire, but there is no reason to think that the adoption of non-Egyptian artifacts had much effect on the everyday practices of village society. First and foremost, the rural houses of the Fayum were machines for living and nodal points in the social, economic, and cultural production and reproduction of the familial units.

Notes

1Boozer 2010, 138.

2See also Boozer 2011; Boozer 2014.

3In Alston 2002a, I avoided the issue of Romanization, since it is not a question that can usefully be asked of the complex documentation.

4Baird 2014, 32–33.

5This issue has an almost endless bibliography, but see, indicatively, Mattingly 1997, 2013, especially 43–72; Keay and Terrenato 2001; Cooper and Webster 1996.

6A characteristic of the colonial situation is ambivalence, a being in-between state in which the categories of the colonial never quite fit. See, for example, Bhabha 1990; 2012; Said 1999; Chakrabarty 2000. For a discussion of ambivalence in a Roman text and related bibliography, see Alston 2018.

7In Alston 2022, I emphasize the complexity of individual and group agency in the production of houses.

8See Silverstein 2004; Herzfeld 1987, 8.

9For the visual in archaeology, see the essays in Edwards, Gosden, and Phillips 2006. Mitchell (1991, 19–31) argues that the act of observation is wrapped up in the power dynamics of the colonial situation.

10Lefebvre (1991, passim, but especially 7, 27, 33, 94, 14) argued that an emphasis on poetics of space (codes of interpretation) blinds us to processes of production and the lived experience of space.

11For the notion of the everyday, see Highmore 2002. Whereas notions of the everyday can become a structural anthropology of advanced cultures, emphasizing system above agency, the everyday also contains within it an element of contingency and praxis, as expressed in De Certeau 1988.

12See Smith 2004, 16–19, on the continuous process of national reinvention, the development of a uniform public culture, and the historical roots on which the nation feeds. Augé (1995) argues that spaces of (inter)national consumption are seen as empty spaces or non-spaces in contrast to the spaces of deeper meaning and cultural resonance.

13See the summary in Levine 2007; for the link between gender and cultural value, see Wilson 2003, 1–12; Chakrabarty 1997.

14In Quaestiones in Genesim 4.15 the house is a metaphor for the soul, which is divided between male (reasonable, rational) and the female (emotional) parts. There is a similar passage in the Legum Allegoriae 4.238–40. In In Flaccum 86–91, the Roman raid on Jewish housing leads to an explanation of the confinement of women to the inner quarters. For the Jewish house in Alexandria, see Alston 1997, 2002b. For a recent discussion of the ethnic conflict, see Atkinson 2006.

15See Husselman 1979, 49–54.

16Good examples from Karanis in Husselman 1979 at House C45 (plan. 29), House C51 (plan 35), House C56 (plan 38), House C62 (plan 40), House C57 (plan 42).

17Hadji-Minaglou 2007.

18Husselman 1953.

19Hadji-Minaglou 2007, 47–68.

20Egyptian villages contrast with the more uniform housing at places like Olynthus or Priene. See Robinson and Graham 1938; Hoepfner and Schwandner 1994. For critical discussions see Morgan 2010; Nevett 1999, 74–78.

21See Boozer 2012. Some elements are, of course similar: B2 has a clustered plan around room R7 which is not very different from that for B1, around room 2 for example. Yet rooms 1/11/14 provide “luxury” space, which is not evidently part of the design of other houses.

22Alston 2002a, 63–67.

23Alston 1995, 121.

24Van Minnen 1994.

25Claytor 2013; Husselman 1952; Husselman 1963.

26Bagnall and Frier 1994, 53–74; see also Huebner 2013 for the dynamics of family.

27Pudsey (2011) shows that the attested structures within the census returns are not the products of ideological preferences but of particular familial events. The approach undermines the macrohistorical approaches, such as Goody 1983.

28Bagnall and Frier 1994, 57–67.

29For land rentals, see Rowlandson 1996.

30Marriage provided an influx of wealth and labor. For marriage gifts, see P. Sakaon 38. For the relationship of familial structure to early male marriage, see Huebner 2013, 48–50.

31Sen (1982) sees a lack of social capital as a root cause of extreme poverty.

32For modern parallels, see Sen 1994; and Bledsoe 1994. Ismail (2006) argues that the operation of extensive social networks was essential for poorer households. It is only with the development of a middle class that large family-based networks have ceased to be crucial, a change that has resulted in shifts in family structure and changes in age of first marriage: see Singerman 2007; Engelen and Puschmann 2011.

33Kaag 2004; de Haan 2007.

34For a contemporary ethnographic account of the importance of remittances, see Ghosh 1998.

35As surveyed in Braunert 1964; Adams 2016.

36Bagnall and Frier (1994, 91–94) show an underrepresentation of young adult males, precisely the population who might be “exported” for economic reasons.

37As Adams (2016, 258) concludes: “Migration was an everyday feature of Roman Egypt, a society that, by any historical standard, was a mobile one.”

38Bell 1938 argues that “going-up” reflected a crisis situation, relying heavily on the Nemesion archive for evidence. Yet “those who have gone up” may have been seeking work elsewhere: see SB 4.7462; P. Ryl. Gr. 4.595. For a more nuanced approach to the Nemesion archive, see Hanson 1979; 1988; 1989.

39In SB 16.12632, Nemesion registers those away from the village in Hiera, Ptolemais Nea, Ptolemais Euergetis, Babylon, and in Alexandria and associated districts. SB 14.11481 is a list of persons resident in Alexandria. The fragmentary P.XV Congr. 14 lists men in Ptolemais Neas, the villages of the Boukoloi(?), the villages of Haryotos, in Syron and Ptolemais Hormou, in the metropolis (Ptolemais Eueregetis), and five in Magdolon. These men were not in hiding.

40See Alston 2005 for discussion and further references.

41Adams 1977; Pighi 1964; Alston 1995, 127–38; Alston 1999.

42See Hammel 1984 for a succinct discussion of the problems. For variation within a single society, see Hammel 1980.

43This continues the argument of Alston 2005, which suggests that the documentation is skewed toward direct ascendants and descendants by the legal processes of inheritance and property registration, but that the social workings of family encompassed a wide grouping, including matrilineal relatives. Looking at familial economics primarily through issues around the ownership of real estate might better suit modern economies than the premodern.

44P. Sakaon 38.

45These were two of the more prominent families in the village of Theadelphia. See P. Sakaon 52 for Pennis son of Sakaon and Zoilos serving as komarchs (village leaders) in 326.

46P. Sakaon 48.

47The family can be reconstructed through P. Sakaon 37; 40; 59; 94.

48See Huebner 2007 for a suggestion that a considerable proportion of sibling marriages fell into this pattern. For sim-pua marriages in China as a parallel instance, see Wolf and Gates 1998; Wolf 1995.

49Huebner 2013, 141–61.

50Alston 2005.

51P. Sakaon 1; 59; 60.

52For small fractions in Fayumic documents, see, for example, P.Petaus 10; 11; CPR 1.223; P.Lond. 2.334 (p.224); P.Mich. 10.583; P. Hamb. 1.15; P. Gen. 1.44.

53See likely examples, SB 1.5117; BGU 15.2476; P.Stras. 4.208; P. Ryl. 2.161; P. Mich. 5.253; P.Mich. 5.285–86; CPR 1 223 (probably), P.Ryl. 2.160(b); P. Ryl. 2.160(d).

54P. Mich. 10.583; SB 14.11895; SB 1.5246; SB 1.5247; P. Vind. Tand. 25; P.Mich. 5.241; P.Stras. 4.602; PSI 8.908 (though the neighboring property also seems to have been sold); PSI 8.910; BGU 1.350; P. Mich. 6.428; P.Ryl 2.162; P. Petaus 14; P.Ryl. 2.160(d).

55PSI 8.909; P.Ryl. 2.289.

56P.Oxy. 1.99. For the archive, see Biscottini 1966; Brewster 1927.

57Husselman 1979, 41–44 (doorways and doors), Pl. 45–50 (bolts and bars), Pls. 57, 58 (blocked doorways).

58Hadji-Minaglou 2007.

59As argued by Wilk and Netting 1984. See also Cooper and Donald 1995 for the prosopographic methodologies required to go from census returns to kin relations.

60Although in law and in practice, the pater exercised considerable authority in the household (Shaw 1987), lived experience was undoubtedly more varied than legal theory (Saller 1986; 1987; 1994).

61Huebner 2019 looks for the single person in the pre-Christian Roman Egyptian census records and identifies a surprisingly high percentage of single adults over the age of fifteen (just under 42 percent). Her definition of single relates to those persons who were not obviously in a conjugal relationship. They were not, however, solitary and tended to cohabit with children or other close relatives. Boozer 2019 looks for singles in the archaeological record. She finds evidence of singles in particular residential structures, military sites, field houses, and artisanal workshops. These examples would seem to confirm the argument in the second section of this chapter that individuals might spend some considerable time away from a particular familial unit, but that does not mean that those individuals did not belong to that familial unit.

Works Cited

Adams, C. 2016. “Migration in Roman Egypt: Problems and Possibilities.” In Migration and Mobility in the Early Roman Empire (Studies in Global Social History/Studies in Global Migration History), edited by L. de Ligt and L.E. Tacoma, 264–84. Leiden: Brill.

Adams, J.N. 1977. The Vulgar Latin of the Letters of Claudius Terentianus (P.Mich.VIII, 467–72). University of Manchester. Faculty of Arts, No. 23. Manchester: Manchester University Press.

Alston, R. 1995. Soldier and Society in Roman Egypt: A Social History. London: Routledge.

Alston, R. 1997. “Philo’s In Flaccum: Ethnicity and Social Space in Roman Alexandria.” Greece & Rome 44: 165–75.

Alston, R. 1999. “The Ties that Bind: Soldiers and Societies.” In The Roman Army as a Community (Journal of Roman Archaeology Suppl. 34), edited by A. Goldsworthy and I. Haynes, 175–95. Portsmouth, RI: Journal of Roman Archaeology.

Alston, R. 2002a. The City in Roman and Byzantine Egypt. London: Routledge.

Alston, R. 2002b. “Reading Augustan Alexandria.” Ancient West and East 1: 141–61.

Alston, R. 2005. “Searching for the Romano-Egyptian family.” In The Roman Family in the Roman Empire: Rome, Italy, and Beyond, edited by M. George, 129–57. Oxford: Oxford University Press.

Alston, R. 2018. “The Utopian City in Tacitus’ Agricola.” In The Production of Space in Latin Literature, edited by W. Fitzgerald and E. Spentzou, 235–59. Oxford: Oxford University Press.

Alston, R. 2022. “Spaces of Desire: Houses, Households, and Social Reproduction in the Roman World.” In Housing in the Ancient Mediterranean World: Material and Textual Approaches, edited by J. Baird and A. Pudsey, 442–69. Cambridge: Cambridge University Press.

Atkinson, J. 2006. “Ethnic Cleansing in Roman Alexandria in 38.” Acta Classica 49, 31–54.

Augé, M. 1995. Non-places: Introduction to an Anthropology of Supermodernity. London: Verso.

Bachelard, G. 1958. La Poétique de l’Espace. Paris: Presses Universitaires de France.

Bagnall, R.S., and B.W. Frier. 1994. The Demography of Roman Egypt. Cambridge Studies in Population, Economy and Society in Past Time 3. Cambridge: Cambridge University Press.

Baird, J.A. 2014. The Inner Lives of Ancient Houses: An Archaeology of Dura-Europos. Oxford: Oxford University Press.

Bell, H.I. 1938. “The Economic Crisis in Egypt under Nero.” Journal of Roman Studies 28: 1–8.

Bhabha, H.K. 1990. “DissemiNation: Time, Narrative, and the Origins of the Modern Nation.” In Nation and Narration, edited by H.K. Bhabha, 291–322. London: Routledge.

Bhabha, H.K. 2012. The Location of Culture. London: Routledge.

Biscottini, M.V. 1966. “L’archivio di Tryphon, tessitore di Oxyrhynchos.” Aegyptus 46: 60–90; 186–292.

Bledsoe, C.H. 1994. “ ‘Children Are Like Young Bamboo Trees’: Potentiality and Reproduction in Sub-Saharan Africa.” In Population, Economic Development, and the Environment edited by K.L. Kiessling and H. Landberg, 105–38. Oxford: Oxford University Press.

Boozer, A.L. 2010. “Memory and Microhistory of an Empire: Domestic Contexts in Roman Amheida, Egypt.” In Archaeology and Memory, edited by D. Borić, 138–57. Oxford: Oxbow Books.

Boozer, A.L. 2011. “Forgetting to Remember in the Dakleh Oasis, Egypt.” In Cultural Memory and Identity in Ancient Societies, edited by M. Bommas, 109–26. London: Continuum.

Boozer, A.L. 2012. “Globalizing Mediterranean Identities: The Overlapping Spheres of Egyptian, Greek and Roman Worlds at Trimithis.” Journal of Mediterranean Archaeology 25 (2): 219–42.

Boozer, A.L. 2014. Amheida II: A Late Romano-Egyptian House in the Dakhla Oasis: Amheida House B2. New York: New York Library. http://dlib.nyu.edu/awdl/isaw/amheida-ii-house-b2/.

Boozer, A.L. 2019. “Looking for Singles in the Archaeological Record of Roman Egypt.” In The Single Life in the Roman and Later Roman World, edited by S.R. Huebner and C. Laes, 57–84. Cambridge: Cambridge University Press.

Bourdieu, P. 1979. “The Kabyle House or the World Reversed.” In Algeria 1960, translated by R. Nice, 133–53. Cambridge: Cambridge University Press. Reprinted in P. Bourdieu, The Logic of Practice, 271–83. Cambridge: Polity Press, 1990.

Braunert, H. 1964. Die Binnenwanderung: Studien zur Sozialgeschichte Ägyptens in der Ptolemäer und Kaiserzeit. Bonn: L. Röhrscheid.

Brewster, E.H. 1927. “A Weaver of Oxyrhynchus: Sketch of a Humble Life in Roman Egypt.” Transactions of the American Philological Association (1927): 132–54.

Chakrabarty, D. 1997. “The Difference—Deferral of a Colonial Modernity: Public Debates on Domesticity in British Bengal.” In Tensions of Empire: Colonial Cultures in a Bourgeois World, edited by F. Cooper and A.L. Stoler, 373–405. Berkeley: University of California Press.

Chakrabarty, D. 2000. Provincializing Europe: Postcolonial Thought and Historical Difference. Princeton: Princeton University Press.

Claytor, W. 2013. “Penthemeros Certificates from the Granary C123, Karanis.” Bulletin of the American Society of Papyrologists 50: 49–75.

Cooper, D., and M. Donald. 1995. “Households and ‘Hidden’ Kin in Early-Nineteenth-Century England: Four Case Studies in Suburban Exeter, 1821–61.” Continuity and Change 10 (2): 257–78.

Cooper, N.J., and J. Webster, eds, 1996. Roman Imperialism: Post-Colonial Perspectives: Proceedings of a Symposium Held at Leicester University in November 1994. Leicester Archaeology Monographs 3. Leicester: University of Leicester.

De Certeau, M. 1988. The Practice of Everyday Life. Berkeley: University of California Press.

de Haan, L. 2007. “Studies in African Livelihoods: Current Issues and Future Prospects.” In African Alternatives, edited by P. Chabal, U. Engel, and L. de Haan, 59–72. Leiden: Brill.

Edwards, E., C. Gosden, and R.B. Phillips, eds, 2006.Sensible Objects: Colonialism, Museums, and Material Culture. Oxford: Berg.

Engelen, T., and P. Puschmann. 2011. “How Unique is the Western European Marriage Pattern? A Comparison of Nuptiality in Historical Europe and the Contemporary Arab World.” History of the Family 16 (4): 387–400.

Ghosh, A. 1998. In an Antique Land. London: Granta.

Goody, J. 1983. The Development of the Family and Marriage in Europe. Cambridge: Cambridge University Press.

Hadji-Minaglou, G. 2007. Tebtynis IV: Les habitations à l’Est du temple de Soknebtynis. Cairo: Institut français d’archéologie orientale.

Hammel, E.A. 1984. “On the *** of Studying Household Form and Function.” In Households: Comparative and Historical Studies of the Domestic Group, edited by R. McC. Netting, R.W. Wilk, and E.J. Arnold, 29–43. Berkeley: University of California Press.

Hammel, E.A. 1990. “Household Structure in Fourteenth-Century Macedonia.” Journal of Family History 5: 242–73.

Hanson, A.E. 1979. “Documents from Philadelphia Drawn from the Census Register’.” In Actes du XVe Congrès international de papyrologie, [Bruxelles--Louvain, 29 août-3 septembre 1977], edited by J. Bingen and G. Nachtergael, 60–62. Brussels: Fondation égyptologique Reine Élisabeth.

Hanson, A.E. 1988. “The Keeping of Records at Philadelphia in the Julio-Claudian Period and the ‘Economic Crisis under Nero.’ ” In Proceedings of the XVIII International Congress of Papyrology: Athens, 25–31 May 1986, edited by B.G. Mandilaris, 261–77. Athens: Greek Papyrological Society.

Hanson, A.E. 1989. “Village Officials at Philadelphia: A Model of Romanization in the Julio-Claudian Period.” In Egitto e storia antica dall’ellenismo all’età araba: Bilancio di un confronto: Atti del colloquio internazionale, Bologna, 31 agosto-2 settembre 1987, edited by L. Criscuolo and G. Geraci, 429–40. Bologna: CLUEB.

Herzfeld, M. 1987. Anthropology through the Looking-Glass: Critical Ethnography in the Margins of Europe. Cambridge: Cambridge University Press.

Highmore, B. 2002. “Introduction: Questioning Everyday Life.” In The Everyday Life Reader, edited by B. Highmore, 1–36. London: Routledge.

Hoepfner, W., and E.-L. Schwandner. 1994. Haus und Stadt im klassischen Griechenland. München: Deutscher Kunstverlag.

Huebner, S.R. 2007. “’Brother-Sister’ Marriage in Roman Egypt: A Curiosity of Humankind or a Widespread Family Strategy?” Journal of Roman Studies 97: 21–49.

Huebner, S.R. 2013. The Family in Roman Egypt: A Comparative Approach to Intergenerational Solidarity and Conflict. Cambridge: Cambridge University Press.

Huebner, S.R. 2019. “Single Men and Women in Pagan Society: The Case of Roman Egypt.” In The Single Life in the Roman and Later Roman World, edited by S.R. Huebner and C. Laes, 37–56. Cambridge: Cambridge University Press.

Husselman, E.M. 1952. “The Granaries of Karanis.” Transactions and Proceedings of the American Philological Society 83: 56–73.

Husselman, E.M. 1953. “The Dovecotes of Karanis.” Transactions and Proceedings of the American Philological Association 84: 81–91.

Husselman, E.M. 1963. “Texts Discussed at the Annual Meeting, 28 December 1962: Two Archives from Karanis.” Bulletin of the American Society of Papyrologists 1: 3–5.

Husselman, E.M. 1979. Karanis: Excavations of the University of Michigan in Egypt, 1928–1935: Topography and Architecture: A Summary of the Reports of the Director, Enoch E. Peterson. Kelsey Museum of Archaeology Studies 5. Ann Arbor: University of Michigan Press.

Ismail, S. 2006. Political Life in Cairo’s New Quarters: Encountering the Everyday State. Minneapolis: University of Minnesota Press.

Kaag, M. 2004. “Ways Forward in Livelihood Research.” In Globalization and Development: Themes and Concepts in Current Research, edited by D. Kalb, W. Pansters, and H. Siebers, 49–74. Dordrecht: Kluwer Academic Publishers.

Keay S.J., and N. Terrenato, eds. 2001. Italy and the West: Comparative Issues in Romanization. Oxford: Oxbow.

Lefebvre, H. 1991. The Production of Space. Translated by D. Nicholson-Smith. Oxford: Blackwell.

Levine, P. 2007. “Introduction: Why Gender and Empire?” In Gender and Empire, edited by P. Levine, 1–12. Oxford: Oxford University Press.

Mattingly, D.J., ed. 1997. Dialogues in Roman Imperialism: Power, Discourse and Discrepant Experience in the Roman Empire. Journal of Roman Archaeology Supplementary Series 23. Ann Arbor, MI: Journal of Roman Archaeology.

Mattingly, D.J. 2013. Imperialism, Power, and Identity: Experiencing the Roman Empire. Princeton: Princeton University Press.

Mitchell, T. 1991. Colonising Egypt. Berkeley: University of California Press.

Morgan, J. 2010. The Classical Greek House. Exeter: Bristol Phoenix Press.

Nevett, L.C. 1999. House and Society in the Ancient Greek World. New Studies in Archaeology. Cambridge: Cambridge University Press.

Pighi, G.B. 1964. Lettere latine d’un soldato di Traiano, P. Mich. 467–472. Bologna: Zabichelli.

Pudsey, A. 2011. “Nuptiality and the Demographic Life-Cycle of the Family in Roman Egypt.” In Demography and the Graeco-Roman World New Insights and Approaches, edited by C. Holleran and A. Pudsey. 60–98. Cambridge: Cambridge University Press.

Robinson, D.M., and J.W. Graham. 1938. The Hellenic House: A Study of the Houses Found at Olynthus, with a Detailed Account of Those Excavated in 1931 and 1934 (Excavations at Olynthus Pt. 8). Johns Hopkins University Studies in Archaeology No. 25. Baltimore: Johns Hopkins University.

Rowlandson, J. 1996. Landowners and Tenants in Roman Egypt: The Social Relations of Agriculture in the Oxyrhynchite Nome. Oxford Classical Monographs. Oxford: Clarendon Press.

Said, E.W. 1999. Out of Place: A Memoir. London: Granta Books.

Saller, R. 1986. “Patria potestas and the Stereotype of the Roman Family.” Continuity and Change 1: 7–22.

Saller, R.P. 1987. “Men’s Age at Marriage and Its Consequences for the Roman Family.” Classical Philology 82: 21–34.

Saller, R.P. 1994. Patriarchy, Property, and Death in the Roman Family. Cambridge: Cambridge University Press.

Sen, A. 1982. Poverty and Famines: An Essay on Entitlement and Deprivation. Oxford: Oxford University Press.

Sen, A. 1994. “Population and Reasoned Agency: Food, Fertility, and Economic Development.” In Population, Economic Development, and the Environment, edited by K.L. Kiessling and H. Landberg, 51–78. Oxford: Oxford University Press.

Shaw, B.D. 1987. “The Family of Late Antiquity: The Experience of Augustine.” Past and Present 115: 3–51.

Silverstein, P.A. 2004. “Of Rooting and Uprooting: Kabyle Habitus, Domesticity, and Structural Nostalgia.” Ethnography 5: 553–78.

Singerman, D. 2007. The Economic Imperatives of Marriage: Emerging Practices and Identities among Youth in the Middle East. The Middle East Youth Initiative Working Paper 6. Dubai: Wolfensohn Center for Development, Dubai School of Government. http://www.meyi.org/publications.html.

Smith, A.D. 2004. The Antiquity of Nations. Cambridge: Polity.

Van Minnen, P. 1994. “House-to-House Enquiries: An Interdisciplinary Approach to Roman Karanis.” Zeitschrift für Papyrologie und Epigraphik 100: 227–51.

Wilk, R.W., and R. McC. Netting. 1984. “Households: Changing Forms and Functions.” In Households: Comparative and Historical Studies of the Domestic Group, edited by R. McC. Netting, R.W. Wilk, and E.J. Arnold, 1–28. Berkeley: University of California Press.

Wilson, K. 2003. The Island Race: Englishness, Empire, and Gender in the Eighteenth Century. London: Routledge.

Wolf, A.P. 1995. Sexual Attraction and Childhood Association: A Chinese Brief for Edward Westermarck. Stanford: Stanford University Press.

Wolf, A.P., and H. Gates. 1998. “Modelling Chinese Marriage Regimes.” Journal of Family History 23: 90–99.