244 paola sacchi and pier paolo viazzo
nearness, a closeness of relationship which rests not only on blood connection or
physical proximity but above all on emotional attachment: the qarayib, she reports,
“would be the people upon whom one has no hesitation to call if there is a need”
(1984: 56). These are findings of considerable importance in themselves, and more
generally because they prompt some final considerations about what is often held to
be one of the long-term distinctive features of households in the Mediterranean,
namely their fundamental role in welfare provision.
Household and family as welfare agenciesOne of the central aims of the comparative research campaign launched by Laslett was
to shed light on how domestic groups had taken care of orphans, the elderly, and
other needy people. The discovery that north-western Europe displayed a marked
prevalence of neo-local households implied that many individuals were bound to find
themselves at some stage of their life cycle without familial support, in a condition of
actual or potential hardship. The existence in England of a deep-rooted state system
of poor relief led Laslett (1988) to articulate his “nuclear-hardship hypothesis” and
surmise that in England and north-western Europe help from the collectivity must
have been of the utmost importance, whereas help from the family was of little sig-
nificance since moral obligations to reside with one’s kin in order to provide support
were either absent or weak. The opposite presumably obtained where various forms
of complex household had prevailed: here one would expect the family—or, rather,
the household as a co-resident group of kin—to be the primary and almost exclusive
provider of welfare.
Evidence consistent with this prediction has indeed been retrieved from several
parts of the Mediterranean and from different periods. Since the 1980s studies of the
living arrangements of the elderly in the modern age have shown, for instance, that
in northern Spain, southern France, and central Italy old people overwhelmingly
resided in extended or multiple households, and almost no widows or widowers lived
in households without at least one of their children (Viazzo, 2003: 129–130).
Intriguing corroboration has come from Hübner’s (2007) interpretation of the
vexed question of “brother–sister marriages” in Roman Egypt. Her contention is
that these married couples, so frequently recorded in the census returns, were legal
but not biological siblings and that we are mostly dealing with uxorilocal marriages
of adopted sons to natural daughters that provided a solution to a problem which
was critically serious in a society where—she maintains—married sons and their wives
were the main care providers for elderly parents and a high proportion of aging men
and women had no living son because of the harsh demography of Roman times.
Rather than a “unique case in world ethnography,” as claimed by Hopkins (1980:
304), this seemingly disconcerting form of marriage should be seen as a family strat-
egy which served several aims, the most important one being to ensure support in
old age.
Whatever the merits of these studies, any generalization to the whole Mediterranean
area would be hazardous, for we have seen that large sectors of southern Europe were
actually dominated by nuclear households. What is more, one can doubt that living
together is the only form and channel of welfare provision other than help from a
collectivity which is only too often identified with the state rather than, say, with
neighbors or friends. Even if one accepts that in Mediterranean countries cultural