PART I
Women in a
Patronage Culture
CHAPTER ONE
The Family
SOMETIME BETWEEN 1046 and 1090, a woman named HayfÄâ bt. SulaymÄn submitted a petition to an unnamed Jewish official in Fustat.1 She had arrived in the city alone and friendless, âa feeble strangerâ with a two-year-old son in tow, desperately seeking the runaway husband who had deserted her in Jaffa. Rejected by her own relatives (âGod knows what I suffered from them and their wordsâ), and finding no help from a brother-in-law in the north Egyptian town of MalÄ«j, she told the official that he was her last hope: âI have no one to whom I can turn with my complaint, except God and your honor.â She asked him to write on her behalf to a Jewish court in Palestine, where she believed her husband had now returned, to pressure him to either return to her or divorce herâan outcome that would at least permit her to remarry.
This story was not unusual. Geniza documents mention abandoned wives so often that the topic easily deserves a book of its own.2 Constant mobility meant that many men spent more time apart from their wives than with them. Men traveled for many reasons: business, religious pilgrimage, simple wander-lust, or to evade the poll tax, poverty, debt, and the financial burden of family itself. âI was bankrupted in Alexandria (by debts owed to) Muslims,â a man writes from Fustat in an early twelfth-century charity petition; âMy children and old mother were starving. I could not bear to sit and watch them in that state, so I fled.â3 For the women that they left behind, the hierarchy of options available to HayfÄâ was typical too. Divorce was common, but a Jewish woman could not divorce her husband in his absence.4 In the meantime, a woman left without adequate means of support could seek help in two arenas: first and ideally, among her birth (and sometimes marital) relatives, or if this failed, from the government, writ small or largeâeither the class of administrative and legal authorities who ran Jewish courts and synagogues in the Fatimid and Ayyubid empires, or the Fatimid and then Ayyubid states themselves.
The rest of this book attempts partly to trace how Geniza women developed resources, or failed to, as they entered adulthood: to identify the factors in womenâs early lives that landed some in hopeless situations like the one HayfÄâ faced, and that provided others with a range of social, legal, and economic options that she lacked. This chapter and the next introduce the two institutions most visible in her letter, whose particular structure and character are central to much of what will follow: the family and the Jewish communityâmost importantly, since so many of the documents that I use are legal papers produced by the Rabbanite court of Fustat, the rabbinic legal system.
Patriliny and Its Limits
First, the family. Family ties loom large throughout the Geniza corpus. Geniza letters consistently assume that parents, children, and siblings will remain emotionally attached to one another and materially and socially support each other, and each otherâs children, throughout their lives. âI have no one but God and you,â a widow old enough to have a daughter on the verge of marriage writes to her brother from Tripoli in the 1060s,
May the Creator not deny me the sight of you! All this year, O my brother, I looked for a letter to come from you . . . when they told me you were ill, I went out of my mind. I swore not to eat in the daytime nor to change my clothes, nor to enter the bathhouseâneither I nor my daughterâuntil I received a letter from you.5
This intense affective ideal, which far outweighed the love that husbands and wives were expected to feel for each other, may be the most striking aspect of the family relationships described in the Geniza. Passages like this led Goitein to describe Geniza families as cohesive extended patrilineal clansââthe house of the fatherâ bound firmly together by âthe bonds of blood.â6
But stories about women like HayfÄâ complicate any attempt to draw a unified picture from these statements of affection. When men died or abandoned their wives and children, what âhouse of the fatherâ did they leave behind? How constant were the âbonds of blood,â and did they matter to all relatives in the same ways, or in equal measure? Much of our evidence suggests that for many people, the answers to these questions were messy and far from obvious. Closer attention to gender complicates matters too. Because women as well as men maintained significant relationships with their own birth relatives, the patrilineal family must have been porous even as an abstract idea: although the formal genealogies preserved in the Geniza exclude women, maternal kinship ties mattered a great deal in social practice.7
This chapter seeks to make sense of this counterevidence, by examining how kinship did and did not work to bind Geniza people together, focusing first on demography and then on ideology. The first part of the chapter examines Geniza families on the ground and concludes from the most systematic data available that both households and immediate family groups were more variable and fluid than has been supposed. The second part zooms out to examine Geniza Jewsâ ideas about kinship more broadly, among extended as well as immediate relatives. It presents a different way of understanding Geniza familiesâas fluid networks bound together by relationships among individual relatives rather than as predictably structured solidarity groups. I will argue both that Geniza Jews considered family ties centrally important, and that they were not attached to any particular form of household or family structure as ideal. It was the kinship bonds between individual relatives that mattered, not the shape of the broader family group that these bonds producedâa distinction that affected women in particular in crucial ways.
THE REAL SHAPE OF GENIZA FAMILIES
Much of the available Geniza evidence for family demography is anecdotal and impressionistic. But the corpus provides one means to consider family structures somewhat more systematically, through the many marriage agreements that it contains. The Jewish legal record preserved in the Geniza includes some 1,500 to 2,500 texts that document the formation of marriagesâboth formal marriage contracts (ketubbot) and a range of auxiliary legal deeds and declarations ratifying engagements, betrothals, dower payments, dowry transfers, and individual social and financial agreements between couples and their relatives.8 Many documents of both types contain three formulae that reveal specific aspects of the coupleâs family circumstances at the moment they were composed: whether the brideâs and groomâs fathers were living or deceased; whether the bride was a first-time bride, a widow, or a divorced woman; and whether she had agreed or refused to live with particular relatives immediately after marriage. These clauses illuminate three basic aspects of family and household structure: paternal mortality during a daughterâs childhood and adolescence, remarriage, and household formation at marriage.
Among the entire corpus of currently available Geniza marriage documents, I have identified 381 documents9 that contain at least one of these three formulae.10 Except for residence stipulations, present in a small subset of these documents,11 these formulae appear in a given fragment largely by chance; honorifics indicating whether an individual is living or deceased seem to appear idiosyncratically in legal documents in general, whereas clauses identifying the bride as a virgin or nonvirgin, included in all Rabbanite and Karaite ketubbot, appear in extant ketubbot fragments as an accident of preservation. These documentsâ distribution therefore likely parallels that of the overall corpus of Geniza marriage documents. They describe marriages between couples whose dowries and promised dowers place them within a broad range of socio-economic classes. Most can be readily identified as formally Rabbanite, but a small percentage are either Karaite legal documents, or they ratify âmixedâ marriages involving a Karaite spouse.12 Finally, while all exhibit paleographic and formulary features typical of the âclassicalâ Geniza period and a bit earlierâroughly between 950 and 1250âmost are only partly preserved; only a slight majority can thus be securely dated.13 About half of these come from eleventh- and twelfth-century Fustat, with much smaller concentrations from tenth- and eleventh-century Syria, tenth- to thirteenth-century rural Egypt, and eleventh- to thirteenth-century Cairo and Alexandria. (See Figure 1.1 and Table 1.1.)
What can systematic review of these formulae tell us? Like most subcorpora preserved in the Geniza, the documents in which they appear are an inherently uneven evidence base. Unlike census results gathered and preserved in order to keep tabs on the demographic makeup of a given population, Geniza marriage documents were composed for unrelated reasons and preserved as a group by chance, facts that limit their usefulness in several ways. First, they do not yield anything like a global picture of family and household demographics. They provide no usable data at all for many questions fundamental to the study of marriage and family systems, including the prevalence of polygamy; proportions of men and women remaining unmarried; fertility rates; infant mortality; either menâs or womenâs precise ages at marriage, first reproduction, and death; andâbecause the select data points that they do preserve are frozen at the moment of each coupleâs marriageâshifts in household composition over time, either significantly before or after a particular marriage was contracted. For this reason, I use the terms family and household here in the broadest possible sense, to denote a) a group of living relatives who viewed their genealogical ties as creating some form of social connection among them; and b) a group of socially connected individuals (mostly relatives, but also nonrelatives such as slaves and household dependents) living together at a given point in time in a shared space, or several contiguous ones, within a building.
Second, the documentsâ haphazard material preservation makes it impossible to detect precisely how the particular measures that they do preserve may have varied over time or between regions and social classes. As a group, Geniza marriage documents cover a relatively broad chronological and geographic range. But only a fraction are preserved in full; this is largely a corpus of fragments, many undatable and of uncertain provenance. The demographically significant formulae that they contain often appear in bits and pieces, on small scraps of paper that contain little else. Instances of each can be tracked in isolation, but they cannot easily by correlated with each other, nor fully separated into discrete subgroups representative of specific micro-populations.
FIGURE 1.1. Datable marriage documents with demographic data, 933â1250
Geniza marriage documents thus cannot be used like census records (or like the parish registers that historians of early modern Europe have mined for demographic data). What they do offer is something far more basic, but still significant: dispersed and fragmentary as they are, this corpus of documents is the closest we can come to a representative record of the Jewish popula...