Moving beyond impressionist accounts derived from literary sources, demographic study of the ancient family now focuses on quantifiable features such as marriage age and household structure. In general, and in keeping with later Mediterranean practice, early marriage for women and late marriage for men appears to have been common among Greeks and Romans. Like other elites in history, Roman aristocrats entered unions at unusually young ages, in the early to mid-teens for women and the late teens for men (Lelis et al. 2003: 103-25). Non-elite customs can only be assessed indirectly, by measuring shifts in commemorative patterns in epitaphs: thus, the age at which spouses replaced parents as commemorators for young adults is taken to reflect the age of first marriage. This method implies a substantial gap between a mean female marriage age of around age 20 and male marriage around age 30 in the Western half of the empire (Shaw 1987; Saller 1994: 25-41). However, as the available evidence is largely limited to urban environments and the first few centuries ad, we are left wondering about marriage practice in the countryside - where men may have married earlier (as they did in late medieval Tuscany, for example) - and about conditions in republican Italy. The latter is particularly vexing because our understanding of the social impact of Roman mass conscription of young men critically depends on the average age of first marriage (Rosenstein 2004): if recruits had already acquired spouses and children, their absence might have been more disruptive than in the event of delayed marriage. As it is, the current model of late male marriage papers over big gaps in our knowledge but is simply the best we have got (Scheidel 2007b). By comparison, the Roman Egyptian census returns indicate slightly less delay, with first marriage in the late teens for women and from the early twenties onward for men (Bagnall and Frier 1994: 111-18).
The same census documents allow us to determine the mix of nuclear and complex households in that province: well over half of all recorded individuals belonged to extended-family or multi-couple households (Bagnall and Frier 1994: 57-74). While the Greek evidence is meager, Roman conditions are once again inferred from funerary epigraphy: since most deceased free civilians were commemorated by members of the nuclear family (spouses or children), single-couple households are thought to have been common (Saller and Shaw 1984). Urban bias, however, raises the possibility that rural households may have been more complex, as they were in Roman Egypt. Complex households are likewise known from other Eastern provinces of the Roman empire (D. Martin 1996; Sadurska and Bounni 1994). Household composition matters because it is associated with the degree of autonomy of married couples - who may strike out on their own (“neolocality”) or remain embedded in extended families - as well as economic performance. Extended families provide better safeguards against risk but also are also conducive to higher fertility that may lower living standards, whereas neolocality would make it harder for widows and orphans to cope.
These broader consequences of ancient marriage and household patterns have gradually begun to attract attention among historians (see fig. 13.2). The later men married, the more likely their wives were to be widowed and their children to grow up fatherless (Krause 1994-95). In Roman society, paternal mortality severely constricted the actual scope of patria potestas, a father’s (fairly) absolute authority over his household (Saller 1994: 114-32). Divorce, generally easier to come by than in
Publisher's Note:
Permission to reproduce this image online was not granted by the copyright holder. Readers are kindly requested to refer to the printed version of this chapter.
13.2 Approximate proportion of Roman men at age x with at least one living relative in a given category. (Source: Saller 1994: 52)
The recent past, would have added further to the instability of families. All in all, we end up with a picture that has much in common with modern conditions of fluidity and hybrid reconfiguration: step-parenting and adoption of relatives were common, creating complex arrangements that can only be documented for elite circles (K. Bradley 1991: 125-76) but would likely have occurred across all classes. Stereotypical ideologies of patriarchy were hard to reconcile with demographic realities.
Meanwhile, what is arguably the single most striking feature of Greco-Roman marriage has failed to raise any curiosity at all - the fact that Greeks (after Homer’s heroes) and Romans were strictly (serially) monogamous regardless of their socioeconomic status, just like modern Westerners but unlike most other early civilizations. While our own experience might tempt us to take this for granted, we must ask how this principle came to be so firmly established even among (customarily polygynous) elites - the egalitarian ethos of the city-state is a plausible candidate -, how it coexisted with de facto polygyny facilitated by sexual congress with chattel slaves (Scheidel 2009), and how it became entrenched in Christian doctrine that survived the fall of the Roman state and ensured its survival and spread in later European (and subsequently world) history. In this strangely neglected area, ancient history has a vital contribution to make to our understanding of the global evolution of marriage.
World History
