1 Constructing Regional Typologies of Family Patterns

In a seminal article by David Reher in 1998, the idea was advanced that there exists a specific pattern of “strong family,” which is typically found in the southern regions of Europe. But is a further specification, one that is peculiar to the Mediterranean family, really possible? The issue of the local specificity of the Mediterranean regions has emerged frequently over the last two decades. The big risk here is to get bogged down in a never-ending fragmentation of typologies. This can, however, be avoided by grounding any taxonomy on a well-founded criterion that justifies making a cleavage between north and south in Europe, as well as between different parts of the south.Footnote 1 The kind of criterion to be adopted is still the subject of much debate.

The prevalent approach in current historiographical research on the morphology of households (Hajnal 1982, 1983; Wall et al. 1983) hinges on a structural criterion: the inverse correlation between age at marriage and size of household. Based on this correlation, a dichotomous geography of systems of family formation takes shape consisting in late marriages and nuclear households in the northwest of Europe and early marriages and more complex households in the south.

Nevertheless, in moving from a country-level to a regional analysis, Laslett’s typology becomes increasingly inaccurate and new demarcation lines appear. This is particularly true in the Mediterranean area.Footnote 2 As Viazzo (2003, p. 116) points out, “testing the Mediterranean model rapidly became one of the prime goals of research for historical demographers and family historians working on southern Europe.” By the early 1990s, most of the authoritative results were contradicting the previous generalizations and showing that underlying the “Mediterranean model” was a wide range of regional and sub-regional variations.Footnote 3 This fragmentation of local family patterns challenges Laslett’s basic argument, which empirically supports the very concept of “system of formation.” The strategy of extending the range of variables used as markers of the typology, however, proved to be ineffective, producing a further multiplication of the patterns with no increase in clarity.

If the category of “system of household formation” fails to give rise to a clear-cut and meaningful typology, this is because it does not reach the core in the identity of a family system. Focusing on the timing of family formation and on the localization rules for new couples is a strategy pragmatically induced by the narrow range of the primary demographic sources in historical research, limited to series of baptisms, weddings, and funerals.Footnote 4 A family pattern, however, cannot be reduced to the timing of the formation and mode of localization of the new couple.

In the last 10 years, a remarkable gestalt switch in the way of interpreting the past and present evolution of family patterns has occurred. Interest in the core of family systems has gradually shifted from the procedures of formation to those of keeping the family together and to the morphogenesis of the family. Attention is being focussed on the way in which a family copes with criticalities at various times during life courses and on the transition between one generation and the next. In managing these criticalities, families do in fact change over time, and by changing they survive and try to remain true to their nature as far as possible.

Two corollaries result from this crucial gestalt switch. First, evidence of the morphogenetic mechanisms must be looked for beyond the mere cohabitation relationships, within the wider and more complex circle of strong kinship and non-kinship ties (Litwak 1960; Willmott 1967; Bonvalet et al. 1999). Second, the regional fissures in the family patterns cannot be solely, and deterministically, identified on the basis of the local economic structures. The ways in which reciprocal contamination occurs in norms, practices, and symbolic frames—in short, in the cultural patterns—also need to be investigated.

Both these corollaries are taken up in part by Frédéric Le Play who, by the mid-nineteenth century, had based a typology of European patterns of family formationFootnote 5 on two coordinates, one for the localization of new couples at marriage and the other for the rules of inheritance.Footnote 6 By marking on the map of Europe the regions that he considered to be dominated by the stem-family model, Le Play delineates a compact region in the middle of the continent: a belt bounded by two lines at more or less 42° and 47° latitude north,Footnote 7 including northern Spain and northern Italy (in particular the Alpine and Pyrenean regions), some Mediterranean French provinces, some German Länder, and reaching into Slovenia through the Austrian and Friuli corridor.Footnote 8

2 Finding the Core of Le Play’s Typology

Of the two coordinates on which the stem-family category is founded —a lower degree of neolocalism and the law of majorat—the latter has generally been given priority by historians and anthropologists, who believe that the advantages (in terms of household survival) from a “curtailed” partition of inheritance (as in majorat rule) allow a system of modified patrilocalism to materialize. The mainstream view about family reproduction patterns emphasizes that succession in stem family areas is first and foremost a retirement contract: parents make very sure that the heir is tied to their well-being in old age and they do so by using their property to cement links in legal, familial, social, and cultural traditions. It is neither a romantic nor a normative option but, rather, a very practical decision.

This is the mainstream view in the debate and is very authoritative. However, we need to be more cautious in this respect, since not two but, in fact, three different issues have been connected with the concept of stem-family:

The first refers to the property transfer and succession in peasant society. Second, stem family refers to household structures and living arrangements [describing] regular coresidence of three generations at least in a phase of the family life cycle, and in a more narrow definition, the coresidence of a married heir with his parents. Third, stem family refers to a particular system of family values and behavioral norms [..], a sense of identification with the intergenerational continuity of the family and his propriety [..] connected to [..] a certain image of stability and of a paternalistic authority of the housefather over his family members (Ehmer 2009, pp. 103–104).

Historical research and regional studies reveal two significant points. As to the rules of non-egalitarian inheritance, property seems not to be necessarily connected with impartible inheritance. Schlumbohm (2009, p. 99) stresses that “the variety and change of household structures show that there is no rigid causal link between impartible real property and stem-family household structures.” The heir is given the task of running the productive and reproductive socio-economic unit and transmitting it undivided to the next generation. The process of transmission of a Hof seems not to imply landed property but only land and “house” use. And this was true in Continental Europe as well as in northern Portugal (Duraes 2009). Moreover, a recent analysis of property transfer and succession among urban Viennese artisans in the nineteenth century shows that the “family business” or Gewerbe (craft or trade) is an alternative way in which a robust propriety may be manifested, which can induce a stem family transfer and succession: Footnote 9

Chimney sweeps and bakers show remarkably higher percentages of transfers of the Gewerbe from father to son (pursuing) the goal of keeping the business in the family [..]. The 19th century saw an increasing concern for the family among well-to-do and rich master artisans (Ehmer 2009, p. 123).

With regard to “family values” attached enclitically to the definition of stem-family (a sort of impulse to perpetuate the “home” over generations), there seems to be no evidence that propriety of any “Hof”Footnote 10 implies such a symbolic mixture. Although a “Hofidee” is strictly observable only among Hofbauern, it does not necessarily characterize all of them. In “Capitalism, Socialism, and Democracy” (1947), Schumpeter describes the change in the capitalistic spirit as precisely the historical reduction in the family’s time horizon and the collapse of the Hof values, that is the philosophy of working primarily for the next generations along the kinship descent. The very idea of the continuity of a supragenerational entity does not appear to be a constant in the stem family model. Schumpeter found it in the romantic Rhenish proto-capitalism driven by the rule of “navigare necesse est, vivere non necesse”,Footnote 11 but no longer in the following generations. A great effort is still needed, then, to contextualize where and when a Hofidee (the family as an “ultramundane” value) is closely interwoven with a Hof.

Many of these observations converge on a major theoretical crux: are we sure that landholding and non-egalitarian inheritance are the real core of the stem family? Caroline Brettell (1991) confutes this hypothesis by arguing that “inheritance practices are not determinative” in discriminating between family models. Behind a particular lifestyle, she suggests, we can glimpse both an economic calculation and a “philosophy of life”:

One does not have a three-generation stem family because property is transmitted impartibly; one has such a family because parents want at least one child to remain at home, work on the farm, and assist them as they get older. In other words, within the broad context of the law, mechanisms for transferring property are strategies pursued to solve some of the problems faced by families of the past and the present—how to secure support in old age, how to contract a marriage for a child, how to provide for all one’s children, how to maintain the social status of all members of the family. Transferring wealth is a form of economic behavior, but as with most economic behavior patterns studied by anthropologists; it has a social dimension as well (Brettell 1991, p. 354).

Underlying any social practice that gives rise to social norms, we can see either a system of costs and benefits or a system of values (a meaning-giving system) sedimented over time. Confuting the economic reductionism of the dominant perspective, Brettell foreshadows an issue that in the 1990s was not yet part of the debate: the core of a typology of family patterns has to be sought not so much in the rules of family-formation as in its strategies of “normal” functioning. A similar conclusion is reached by Ehmer:

Family values appear as personal relationships between people, but not as a connection to a house and landholding [..]. It was not until the 20th century that ways of thinking focused on “maintaining the family line” (Mitterauer 1986, p. 312) gained significance (Ehmer 2009, p. 126).

Already in the mid-nineteenth century, therefore, the intergenerational pact of the stem family no longer involved exclusively (or predominantly) the transmission of a real propriety, but rather a set of reciprocally supporting “total prestations.” Neither the residential rules nor the rules concerning the impartible inheritance seemed to be the core of the stem family model, but rather an “essentially by-product” (Elster 1979).

3 A Dual Model of Family Morphogenesis

David Reher (1998, p. 214) also agrees that “historically the strength of familial ties appears to have conditioned the way in which succession was carried out in stem-family regions”: strong ties shape the rules of inheritance, not the reverse. The strength of family ties (that is, of the type of pact established between generations belonging to the same kinship descent) leads in many aspects of family life to important functional differences between northern and southern Europe. In particular, when children leave the parental home, in northern Europe they do so as soon as they have acquired a minimum level of economic independence (and only later get married and form a family), whereas in southern Europe home-leaving tends to coincide with marriage and finding a stable job.

Nevertheless, the age schedule of family formation is not the linchpin of the strong family. The central element in this pattern of kinship relations (Bestard Camps and Contreras Hernández 1997, p. 69) “is its character of moral obligation based on reciprocity: previous conduct to a significant degree governs present conduct”:

Kinship relations create rights and duties which an individual cannot refuse. In every day practices kinship relations are perceived as having a moral obligation and everyone respects an articulated set of rights and duties, obligations, and expectations related to her or his position in the family cycle (Ibid).

Paraphrasing this definition, we can reformulate the intergenerational pact of time-lag reciprocation as follows: .

The strong family is thus the steady state of a process of intergenerational lagged reciprocity. Reher (1998, p. 212) found this mechanism eloquently expressed by an elderly Spanish peasant in an interview: “First the children lived off their parents, and later the parents lived off their children. That is just the way life was.”

The mechanism of reciprocal obligations among three generations studies as follows: In the centre-north of Europe, children leave home as soon as they achieve minimum economic autonomy. From then on, parental support—both cash and in kind—is sharply reduced: what they do not spend on their children can be put aside for their own old age. In the centre-south of Europe, on the contrary, parents continue to maintain their children for longer, currently into their thirties, but the support given to children will be reciprocated when their parents grow old, though not in the same proportion. What insures that this unwritten agreement is never broken is its reiteration among the generations.

In fact, in the steady state of the reciprocation process (Fig. 1), all generations offer their free support—cash or in kind—twice in their life: both to their children and to their old parents. In the meantime, all generations receive free support twice in their life, from two different generations and at two different stages of their life: when they are young from their own parents and when they grow old from their children. The reciprocity pact is therefore a chain of lagged pacts, the concurrent interlocking of three generations. Over their life’s course, a person basically receives as much as he pays out (in a static world), though this happens only from a multigenerational perspective. The zero sum is achieved because a given set of parents received a given amount of goods and services from their parents, which they ultimately provide to their own children. This evens things out in the long run. So the issue of reciprocity certainly does exist, but the balance sheet can only be drawn up when we look at reality multigenerationally.

Fig. 1
figure 1

The chain of lagged reciprocities among generations in the strong family

Full size image

Naturally, the steady-state multigenerational pact implies that the balancing act will only work in periods of relative economic, demographic, and cultural stability. Historical changes like, for example, the current re-positioning of state intervention in welfare, can alter situations of this kind and upset the intergenerational balance: Footnote 12

Even in Sweden there is now pressure for families to mobilize their resources to take care of the elderly and disabled: legislators, administrators, and experts attempt to re-invigorate and strengthen kinship ties [..]. In post-1989, Germany the legal definition of kin has been broadened to increase the number of family members who can be summoned to take care of needy kin. North and Central Europe countries are thus moving toward what is called “a more ‘Italian’ model,” where the notion of obligation among kin is strongly underlined and a large number of relatives are legally expected to provide support. The rediscovery of kinship ties in contemporary Europe primarily obeys instrumental imperatives (Viazzo 2010, pp. 284–285).

Another observation has to do, here, with the increasing longevity in western societies, which is rapidly and profoundly altering the form and function of the family:

Even in ageing societies, families are not permanently three- or even less so four-generational. From both an individual and a family life-course perspective, the presence of three and, particularly, four generations occurs in specific phases of the individual and family life course. Most individuals and families experience the three-generation phase one or two times over the life course (once when a child or a young family, once when old), and also for some length of time; a much smaller number of individuals and families experience a short four-generation phase. But at any point in time, three-generation families account for less than half of all families, while four-generation (or more) families account for only a very small percentage of all families (Saraceno 2008, p. 7).

This major change compels us to broaden the generational exchange described in Fig. 1 and include those actors Hagestad defines as “book-end generations,” that is grandchildren and grandparents:

By observing the old, the young can explore their future selves; through interactions with the young, the old review previous stages of life (Hagestad 2008, p. 25).

Finally, it may be questioned why in this analysis of the golden rule of reciprocity no difference at all is made between the child destined to inherit and the other children, in spite of our acknowledgment that the culture of the strong family stems directly from the culture of the stem family,Footnote 13 which is an extremely selective and non-egalitarian system of intergenerational relationships. This doubt would be justified if we had been concentrating on the processes of family reproduction in the past. Instead, our aim here is to explore the current family, taking into due account those elements that have persisted since the mid- nineteenth century.

From this point of view, two differences compared to the past must be stressed. First, in a southern Europe (both Continental and Mediterranean) where the total fertility rate fluctuates steadily around the 1.3 mark, the competition for an impartible inheritance is considerably weakened by the reduced number of competitors. Second, the current family model in the Continental south of western Europe clearly hinges on the mid-nineteenth century stem family, though in a version modified by the egalitarian movement sparked off by the introduction of the Napoleonic civil code, as well described by Fauve-Chamoux and Ochiai (2009, p. 12):

With the egalitarian dispositions of the Code Civil (1804) [..] for a century, from the 1780s to the 1880s, the stem-family system of strict reproductive control was seriously attacked. In many families children refused non-egalitarian transmission and obtained enough (in land, goods, or money) to establish a new house, marrying against all customary practices [..]. In this new perspective, the classic rigid stem-family model, as traced by Le Play, had to be seriously revised.

4 Properties and Corollaries of Reher’s “Strong Family”

The blood pact is an extraordinarily effective welfare-providing system,Footnote 14 based on the interlinking of three devices.Footnote 15 The first is the manifest gratuitousness of the parental “loan.”Footnote 16 The second is the time lag between the display of gratuitous parental care and the moment at which children become fully aware of it.Footnote 17 The last device is the clear disparity involved in the exchange so that children will perceive themselves as being forever indebted to their parents.Footnote 18 Overturning the traditional approach, the rules of formation of the strong family are seen to stem—as corollaries—from this kind of golden rule. And they are more complex and articulated than those typically found in the traditional approach, which is based solely on archive material.

Exit from the parental home, traditionally late and contextual to marriage, is currently being brought forward because of the spread of informal cohabitation. It is, though, still steered and fuzzily put off.Footnote 19 What makes young people, who leave home so vulnerable, exposing them to the risk of not transitory poverty, is job precariousness. Here more than elsewhere, the contradiction between the drastic deregulation of the labour market, on the one hand, and the still widely persisting male breadwinner family model, on the other, generates vulnerability to poverty (Micheli and Rosina 2009). Moreover, the earlier the parental home is left, the stronger the centripetal, cohesive force of the family comes into play with the result that children experience a parting from the home that is gradual and piloted by parents. So painful is it to leave home that it requires both a web of small liminal rites and the self-deception that it is only a transient passage:

The day before I was to leave home for good, Mum filled a basket with things you always need in a home…I remember that, one Saturday, I was supposed to go and sleep for the first time in my own home. It’s what I had decided and wanted, but on the day I said ‘if I leave now, it’s the wrong time and I’ll always feel some regret.’ So I remember saying ‘Mum, I’m too tired. I’ll stay and sleep here tonight.’ She said ‘Yes, okay, stay and sleep here tonight.’ The next day, I went away calmly but convinced that I’d miss them so very much that I told my mother not to take the sheets off my bed because I’d be back. [Single W].

Leaving home to go and live on one’s own is still so laden with significance and anxiety that the separation is not only gradual but sometimes remains incomplete, producing a sort of revolving door between the two homes, a limbo with no centre of gravity:

I still feel that my home is that of my parents. To tell the truth, I don’t feel I belong in either one or the other. [Single W].

I go back to my parents’ place once a month and I’m happy there. I’m back home. Yes home, because I still don’t really know which is my home. It’s three years now, but when I think of my home, I think of the one I left. [Single W].

Two corollaries about the formation of new family units follow from these premises: an incomplete physical divide and an imperfect role discontinuity.

Physical detachment from the parental home is incomplete, thereby merely prolonging the family umbilical cord. House (place of residence) and family (locus of relationships and affections) merge into a single category, which is endowed with a deeply physical, relational, and symbolic quality. The “casa pairal”—the Catalan term for the house where the stem family has lived continuously—well represents this synthesis:

Family refers to ‘home’ as a place of residence and it expresses a lifestyle. Family and house are related through home [..]. ‘Home’ unites ‘house’ with ‘family’. However, in the cultural context of Catalonia casa symbolizes ‘family’, and there is not a clear differentiation between ‘house’ and ‘home’ [..]. In this cultural context casa [pairal] is perceived as a means of ensuring the continuity of family identity [..]. Casa is a focus of expectation and negotiation among kin because of its capacity to express continuity as a ‘family house’ (Bestard Camps and Contreras Hernández 1997, p. 72).

The life-space of young people leaving home is, therefore, conditioned by the active role that the family of origin plays in providing a new house. Thanks to growing family affluence—and increasing property ownership—the nest formation of new couples changes into a sort of superfoetation of the parental home. New couples move to a life-space that is at the same time both outside and inside the family. The persistent practices of patrilocalism turn into a new neolocalism, characterized by a limited sovereignty. In this colonized neolocalism the material benefit of the adobe, put at the new couple’s ready disposal,Footnote 20 is paid for by a sort of praedial servitude, that is the right of parents to enter, even though discreetly, the private space of their children.

My parents owned a flat quite near theirs, so they were able to help me out. They didn’t influence us, we made our own decision. [Married W].

After the parental home is left, also the role set undergoes a transformation which remains imperfect. Though a pure patrilocalist model is no longer in force, the custom of children keeping the parental bond alive after leaving home is still prevalent. The constraint of proximity to the parental home is often an essential criterion for the choice of the new habitation, and this constraint is justified by the need for parents and their offspring to remain at each other’s disposal. The blood pact remains in force also subsequent to home-leaving and is reproduced through the iteration of small daily rituals.

It’s now a ritual: when I wake up in the morning, I call (my mother) to tell her I’m okay, I’m going to the office. In the evening she’s the one to call: everything okay then, you’re eating, you’ve eaten. We call each other at least twice a day. [Single W].

Both the essential criterion and the corollaries of the strong family make it evident that Reher’s contribution is an important step forward in the search for an effective typology of family patterns in Europe. He does not stop on the threshold of the family’s abode and simply count the number of people going in or coming out; he in fact enters the black box of the mechanisms of reciprocal support between generations.

5 What Characterizes the Mediterranean Family Pattern?

Reher observes that in Europe “it is not difficult to identify areas where families and family ties are relatively strong and others where they are relatively weak”:

The dividing line, in some ways, is actually much simpler, with the central and northern part of Europe (Scandinavia,Footnote 21 the British Isles, the Low Countries, much of Germany and Austria), together with North American society, being characterized by relatively weak family links, and the Mediterranean regionFootnote 22 by strong family ties. (Reher 1998, p. 203).

Reher suggests, in sum, that the strong family is an anthropological pattern covering the whole of Europe below the 47th degree of latitude north. Nevertheless, as historiographical and anthropological research has advanced, some specificities of the Mediterranean regions that are not reducible to a sort of indistinct Pangea have once again emerged. The differentials in the fertility trends over the last few decades (Micheli 2000) confirm this cleavage.

Hionidou (1995) affirmed that Greece is part of an area—including Italy, Spain, and Portugal—marked by a “different vein,” which she calls strictly “southern European Mediterranean.” Nonetheless, the scientific awareness of a Mediterranean specificity already existed before the 1990s, stemming from a suggestion by R.M. Smith (1981) that the cultural trait peculiar to this area lay in female honour. Even earlier, in the 1960s, Arensberg had observed that

Rather than Spanish as such, the Andalusians are better classified with, for instance, Sicilians and others, more of whose ways they share, as Mediterranean; while the Gallegos and the Basques are better classified with others to the north as people of the Atlantic fringe (Arensberg 1963, p. 79).

As Viazzo (2003) comments:

Arensberg was expressing a growing conviction, also voiced by other anthropologists in the same years, that the communities of the Mediterranean possessed more similarities between different countries than the tenets of modern nationalism would have us believe and, at the same time, that they could differ significantly from other communities within their National frontiers (Viazzo 2003, p. 118).

Two distinct patterns of strong family divide the regions in the south of Europe. We need, therefore, to shift from a country to a sub-country level of analysis. It is true that the utility of a country-level analysis.

follows from the (usual but not universal) coincidence of culture and language with national boundaries, and also with the common influences on demographic behavior of uniform national systems of taxation and of family welfare arrangements (Coleman 2002, p. 324).

Nevertheless, between the north and south in both Spain and Italy the organization of the family exhibits differences which clash with this presumed homogeneity. The cleavage between the two regions of the south-European family runs sometimes within and at others across the borders of the nation-states. As Reher writes:

The specific boundaries of different family systems are often not crystal clear and there is much sub-regional difference. For example, in some respects Ireland does not fit well into northern European family patterns, there are indications that northern and southern France often walk divergent paths, and the southern fringes of Spain, Italy, or Portugal often show distinct characteristics from the northern parts of those same countries. Within individual societies, there is also much room for heterogeneity affecting families and family life. This multiplicity of forms and behavior, however, does not negate the existence of more general regularities affecting large areas of Europe (Reher 1998, p. 203).

This is why we draw demarcation lines both within and across national borders and plot a map of macro areas, in which southern Europe is split into two parts. Taking as a basis the same corpus of in-depth interviews used in the previous section,Footnote 23 let us try to delineate the communalities and specificities of the Mediterranean model.

The golden rule of lagged reciprocity between generations unites both the patterns of strong family: previous behavior determines current behavior—“I know I’ll receive because I’ve given.” However, whereas in the Continental south of Europe the pact of reciprocal insurance is a blood pact, that is it develops along lines of kinship descent binding two generations to one another, in the Mediterranean south of Europe two significant variants can be detected. First, the new partnership contract sets an insurmountable limit to the pervasiveness,Footnote 24 virtually boundless, of the moral obligation toward parents.Footnote 25 Second, the blood pact spreads beyond the parent–child dyad to involve a larger circle of kin. Family agency includes a dense network of ties.Footnote 26

[a] The Mediterranean regions seem to be characterized both by a clear-cut physical divide and a plain role discontinuity. Whereas in the Continental south a strong proximity of children to the parental home is the rule in the tradition of the stem family model, in the Mediterranean area physical separation from the parents’ home involves a clean break with the new couple staking out their own clearly defined life-space. As Le Play (1871) already observed, the Mediterranean regions seem to be marked by a neolocalist pattern of family formation. Silverman (1968), for example, traced a boundary separating the “deep south” (of Italy), where the nuclear family is the rule, from the culture of a “central Italy,” where the adhesiveness of the family stems from the force of attraction of the sharecropping system. Similarly Pitt-Rivers, in his ethnographic study of an Andalusian town, emphasized the great significance of the physical divide for new couples:

People feel very strongly that every family should possess its own house, and to marry without a separate home is regarded as a make-shift arrangement (..). The economic advantages which might accrue from forming a larger family unit are offset by the desire to be free of the tensions which make family life impossible where there is more than one family in the house.Footnote 27 « Cada uno en su casa » (each one in his own house), that is the only way to live peacefully. « Casada casa quiere » (housewife wants house), the saying rubs in the point. (Pitt-Rivers 1971 [1954], p. 99).

As for role discontinuity, we should bear in mind Edward Banfield’s (1958) conclusions. When exploring the rules of relationships within a community in Mediterranean Italy, Banfield stressed that “amoral familism” is not to be identified with an individually selfish pattern of behavior but rather with a variant of altruism, in which the social circle of reference is that of the nuclear households. In the narratives of interviewees living in the Mediterranean regions, there still emerges today a tendency to maintain a clear-cut role discontinuity, stemming from the major rule of being loyal to the new family as the “true and real” family.Footnote 28 It is a role discontinuity that does not depend on a generational discontinuity because of the long-term modernization process; on the contrary, it fits in perfectly with the rules handed down from parents to children, which exclude any scenario of revolving door.Footnote 29

[b] What distinguishes the two Mediterranean models has to be sought not so much in a different mode or timing of new-couple formationFootnote 30 as, rather, in a system of attractive and repulsive forces, aimed at holding the different family units together in a web of kinship solidarities.Footnote 31 Consequently, the device of time-lag reciprocation concerns not simply the parent–child dyad but, more precisely, the relationship between family unit and kinship. Whereas in the Continental south of Europe the family of origin is the real centre of gravity for the entire kinship system, and its adhesiveness curbs children’s natural propensity toward leaving home, in the Mediterranean area the family of origin looses its “centripetal force,” and the intergenerational pact involves a real “kinship federation” among family units, which interweave with one another through both blood ties and a shrewd strategy of alliances.

Undoubtedly, the two regions of southern Europe do have one feature in common: the crucial role performed by family support in reducing the vulnerability of young people when leaving home. However, what distinguishes the Mediterranean regions from those of the Continental south is the fact that such support is not given by the family of origin, but rather by the whole kinship federation, and in this extended version support ends up assuming a “monopolistic” role in social regulation (Micheli and Rosina 2009).

In conclusion, while the two souths of Europe have in common a regular functioning of the strong family, what distinguishes them is a different code of home-keeping, the aim of which is to insure that the trajectory of the new family units follows in the wake of the parental unit and the kinship system is reproduced from one generation to the next. Whereas in the Continental south the pact of solidarity—namely, a web of strong ties rooted in the intergenerational pact—gravitates solely to the “stem” family, in the Mediterranean south by contrast it gravitates toward a federative pact among family units, a complex system of counterbalances on which they all hinge. What distinguishes the two southern kinship regimes, therefore, has not so much to do with the moment and the way a new couple (like the formation of a new planet) become detached, as with their continuous efforts over time to trace equilibrium orbits by maintaining the correct distances between the different family units (as if they were different planets) of an integrated system.Footnote 32

6 Gift and Asabiyyah: Anthropological Roots of the Strong Family

A boundary line runs through the regions of southern Europe, separating those lying on the mountainous backbone in the Continental south from the others facing the Mediterranean rim. What are the roots of this cleavage? Reher himself makes a suggestion, incurring a precious slip of the pen. When disaggregating Europe into two models, strictly circumscribed by national borders, he looks for the historical roots of this bipolar geography in a bifurcation of the processes of cultural sedimentation: southward Roman culture and law; northward the Germanic counterparts. But, surprisingly, what Reher says sketches a Europe divided not so much in two as in three distinct local stratifications:

In the northern part of the continent, Christianized forms of familial organization ended up meshing gradually with existing Germanic legal and social traditions based, among other things, on the importance of the tribe, the individual, and the relevant social position of women. In southern Europe the influence of the Germanic tribes was much more superficial and short-lived. Besides, from the early eighth century on there were a series of Muslim incursions, strongest in Spain and in the Balkan Peninsula but also present in southern Italy, which tended to bring back, once again, oriental family structures, so central to Islamic societies, that are based on the overriding importance of kin ties. (Reher 1998, p. 214).

There emerge from Reher’s account, therefore, three focal points of cultural diffusion. In the north and centre-north of Europe, Germanic social organization and civic law form the base on which the Roman equivalents were grafted. In the Continental south, Roman culture prevailed, though with many pre-Roman contaminations. In the Mediterranean south a new cultural stratification, of Islamic origin, spread from the eighth century onwards for a few centuries.Footnote 33

In this culture the role of incubator for young generations is not played by the blood family but rather by kinship heritage, which is neither economic nor human capital but rather relational capital: Reher emphasizes the “overriding importance of kin ties.” However, the term “kin ties” does not make explicit a meaning that has been overlooked until now: the Islamic strong ties capital radiates out from the parental home, but is not limited to blood ties in the strict sense. Islamic strong ties capital has in fact wider and more fluid borders than the Continental strong family. We can understand this better if we look at a seminal text on Western culture by an historian of the decline of the Islamic Empire, which confirms the specific cast of actors that play a role in the kinship regime of the Mediterranean regions.

In the Mouqaddima,Footnote 34 Ibn Khaldun proposed a sort of theory of collective action, hinging it upon the concept of “Asabiyyah,” an abstract noun derived from “asabah” meaning male sibs of a common lineage (Gabrieli 1930). Varyingly translated as “esprit de corps” or group solidarity, this term has a particular semantic depth: it is true that “Asabiyyah” is based upon blood bonds resulting from “nasab” (genealogy, blood descent), but it widens out to include as naturally belonging components some non-blood reciprocity ties, such as alliances (hilf) or patronages (wala):

Feeling that joins us to a blood kinship is a natural fact. Everybody loves his own parents and family. [..] The same happens with people who either place themselves under our protection or made an alliance with us. As the words of the Prophet say: « The sense of belonging to a family is what leads us to help anyone else of the family » . Kinship just consists of that blood force which leads to a reciprocal support, in case of need (Ibn Khaldun 1985, pp. 154–155, my italics).

These words veer abruptly away from our traditional idea of “strong tie.” Ibn Khaldun does not assert that the drive to reciprocate results from the blood tie; on the contrary, kinship itself is defined as stemming from the establishment of a reciprocity pact. Hence, a blood tie does not descend only and necessarily from a kinship relationship (what common sense calls “blood tie”) but can descend also from a preliminary and more general practice of reciprocation.

What, then, are the communalities between the two southern family patterns and what are their specificities? Marcel Mauss’ “Essay on the Gift’” (1923–1924) also helps us to understand them better. The gift, emblem of a secondary sociability not so far from the “sabiyyah” concept, is every instance of property or service given with no guarantee of return aiming to create, re-create or keep alive a social tie (Godbout and Caillé 2001). The modern nature of the gift is often evoked by comparing it to time banking. Nevertheless, what clearer example is there than the perpetual free loan inscribed in the intergenerational blood pact? The blood pact is the most misunderstood among the “total prestations” because of a sort of “telescopic” fallacy which makes us overlook what is immediately apparent.

As in the intergenerational blood pact, the gift implies reciprocation,Footnote 35 which primes a loop of “expectations of expectations” (Elster 1979), and this interiorized loop in turn triggers the senses of duty, guilt, and shame. Moreover, again as in the intergenerational blood pact, the longitudinal dimension of time is fundamental to the gift. When splitting up the “total prestation” into three distinct obligations—giving, receiving, reciprocating—Mauss (1923–1924, p. 93) stresses that “the intrinsic nature of the gift is, whatever the society, to create a time-lag constraint. Whichever the reciprocation, time is needed.” What makes the gift incommensurable with exchange is therefore a mix of three elements: gratuitousness, compulsoriness and, chiefly, the time lag before the obligation is fulfilled.

When studying the logic of action in blood donors, Titmuss (1979) overturned the syntactical sequence of the economic rule “do ut des” by using a non-conditional future in the subordinate clause: I give, so that you will give. The same overturned syntactical structure is adopted by Bestard Camps and Contreras Hernández (1997) to define the blood tie proper of the Catalan “pairalismo”: “I know I’ll receive [non-conditional future] because I’ve given [past tense].”

Since time-lagged reciprocity is always a trap, in which the imperative nature of the gift preliminarily implies something “already received,” the blood pact is thus the last link in an intergenerational chain resistant to breakage. It is a vicious circle no different in its dynamics from a blood feud between two families or clans.

All the kinship regimes—weak, Continental strong or Mediterranean strong family—share the no-limit delay on reciprocation, though with a varying importance in the role played by the blood pact. Nevertheless we can find further suggestions that are useful to split up the two southern family patterns by identifying the different casts of actors contracting the pact. In the last part of the “Essay on the Gift’,” two distinct modes of total prestation are briefly sketched out by Mauss. As to southern Europe—Reher’s Roman Europe—focusing on the “nexum,” a form of contract in Roman law, he says:

Things are not the inert entities they are considered to be by both Justinian law and our juridical systems. First of all, they are part of the family: the Roman ‘familia’ includes not only people but also the ‘res’ [..]. Going back to ancient times, the word ‘familia’ denotes more and more the ‘res’ that are part of it. The best etymology of the word ‘familia’ goes back to the Sanskrit dhaman, that is house. Moreover, things belong to two types. Roman law distinguished between ‘familia’ and ‘pecunia’, the things of the house [..] and the cattle, grazing in the countryside away from the farm buildings. And it distinguished also between ‘res mancipi’ [the things of value, buildings and children included, that can only be alienated by following the formula of mancipatio] and ‘res nec mancipi’.. (Mauss 1923–24, pp. 131–132, my italics).

Most specificities of the Continental strong family are echoed in this quotation: the incomplete and reversible divide between the parental unit and the new couple, and that between the parents’ abode and the new one, points to a persistent pattern of total fusion of both children and their abode into the home. In Roman law, children and the abode were both res mancipi, namely precious and inalienable “things” belonging to the home.

Kinship descent assumes in this context the form of a direct relationship between owner and property. And this form is crucial for social reproduction. In the north of Europe, by contrast, the role of the kinship relationship fades into the background to the extent that other reciprocity performances involve a larger circle of actors. Here a total prestation different from the nexum was widespread in ancient times. It developed not into kinship descent but rather into a web of strong ties outside the family unit: the circle of relationships between different clans within a tribe or between different homes within a clan. Mauss presents as an example of this different device the rites of Gaben, the “night of gifts”:

On the occasion of a marriage, the guests, normally the whole village [..], bring wedding presents whose value is far greater than the wedding expenses. In some German villages, the ceremony of Gaben itself forms the bride’s dowry, which is handed over to her on the morning of the wedding day [..]. The munificence of the gifts is a pledge for the fecundity of the young couple (Mauss 1923–24, p. 154).

In the Gaben ritual a way of social life is formalized that is based not so much on kinship ties as on the relationships between individual and community. Awareness of the virtuous circle linking individual and community together is no doubt the founding formula of the north-European system of civic rules. Reher’s distinction between a weak family area and a strong family area is confirmed by Mauss. However, while in the Continental south the fusion of both house and children into the category of the home has its roots in the social practices governed by Roman law and culture, also in the Mediterranean regions many clues point to a reproduction system based on a loose-knit web of enlarged reciprocities, which in turn stem from a system of kinship alliances (the “purest kind” of total prestation, according to Mauss) between clans or family units.

In the Mediterranean regions, we find a family system whose centre of gravity does not lie on the home but rather on alliances and reciprocations. It is not a social reproduction system in which blood ties produce reciprocity but, conversely, a system where reciprocations generate blood ties. This makes it a system ruled by practices and customs more similar to the Gaben culture than to the nexum culture. Northern Europe and the Mediterranean south appear, in sum, to share a culture of enlarged reciprocities, as against a one-centred home-based culture, marked by Roman law. This leads us to suppose that in the Mediterranean regions Roman culture was cross-fertilized by the waves of other historically overlapping cultures.

7 An Old Question to be Reopened

What has been attempted here is to find empirical confirmation for the hypothesis that a “different vein” distinguishes the “strong family” patterns characterizing the Continental south of Europe from that predominant in the Mediterranean regions. A large corpus of in-depth interviews seems to support the hypothesis of a dual identity of the strong family, differently shaped in two belts in southern Europe. Finally, exploring the historical roots of this cleavage between the two southern family patterns, we have searched for some clues to their existence in two seminal texts: Ibn Kaldun’s Mouqaddima and Mauss’ “Essay on the Gift’,” with particular reference to Mauss’ categories of “total prestation” and nexum and Ibn Khaldun’s “asabiyyah” and “nasab.” The views of these two social scientists with respect to the aetiology of a reciprocity device compel us to reopen a methodological questionFootnote 36 and reassert its importance.

We cannot but take note of the fact that, as the debate now stands, the search for the roots of a social practice—such as the intergenerational pact—is firmly driven by a predominant criterion that is the existence of a previous and clear-cut economic motive. An authoritative example is in Reher, where he discovers an economic explanation for the cultural cleavage between family patterns in Southern and Northern Europe in some seventeenth-century marriage contracts:

In Catalonia and the Basque Country of Spain, stem-family areas par excellence, the obligation to co-reside en una mesa y compañía (‘at one table and in the company of’) with the parents was normally stipulated quite simply in the marriage contract (capitulaciones matrimoniales). In much of central and northern Europe, veritable retirement contracts between parents and their children were drawn up listing in great detail the rights and obligations of children and parents [..]. The history of these contracts was frequently fraught with the intergenerational strife they were designed to minimize. Contracts such as these are simply unimaginable in a southern European context (Reher 1998, p. 211).

The primacy of the rational choice approach in the formation of normative cultural behavior patterns is currently overwhelming.Footnote 37 Nonetheless, this need not be the only possible approach. Different family patterns can have their imprinting in symbolic constructions of reciprocity rules that differ from one another. In Ibn Khaldun’s “nasab” genealogy, the rational strategies of collective action derive from the establishment of affective (not necessarily “blood”) ties. And Mauss does not trace the sedimentation of a reciprocity rule back so much to economic reasons, as to cultural and symbolic devices. As Caroline Brettell (1991) has pointed out, what lies behind a lifestyle are an overlapping economic calculus and “philosophy of life.” Whichever social practice is to be explained, it is no doubt reassuring to ground it on economic causality; but this is by no means sufficient.