1This academic investigation into the issue of kinship is the result of several workshops entitled ‘Running risks in the family: kinship put to the test of economics and politics, from Iran to the Balkans’, that were held on February 2004, 13th (Aix-en-Provence), and May 2004, 7-8th (Istanbul). These meetings were organised by the Institut d’ethnologie méditerranéenne et comparative (IDEMEC) [Institute of Comparative and Mediterranean Ethnology] and the Institut Français d’Études Anatoliennes (IFEA) [French Institute of Anatolian Studies]. They were coordinated by Gilles De Rapper and Benoit Fliche.
2Following this meeting, this fourth thematic issue of the European Journal of Turkish Studies, edited by Benoit Fliche, is comprised of eight articles on the subject of the findings of the participants of the workshop. The introduction is two-fold : firstly, we wish to present the general framework of this issue and, secondly, place the articles in each other’s perspective.
- 1 L’Homme 154-155 (Barry 2000). The contributors to this issue sought to set out comprehensively the (...)
3Studies concerning social practices of kinship are notable by the marginal place they occupy in the field, as demonstrated by an issue of the review L’Homme published in 20001, and by the publication of Maurice Godelier’s Les métamorphoses de la parenté in 2004.
4Without surveying a debate dating back to the 1970s, an explanation for the peculiarities of the field of French anthropology can be found in Levi-Straussian structuralism. In this configuration, Bourdieu’s work on ‘parenté pratique’ [kinship in practice/ practical kinship] (1979: 273) did not always find its place, even if it had a considerable influence amongst the English and American anthropologists (Collard 2000, Schweitzer 2000: 9). The latter, however, concentrate on such themes as alliances and matrimonial strategies. Few studies adopting Bourdieu’s work as their inspiration have sought to explore how actors make use of kinship outside of the family. Consequently, 30 years after the publication of Bourdieu’s first writings on the subject, Peter P. Schweitzer’s statement still seems relevant: ‘one example of such neglect is the issue of “what kinship does” (or more precisely, “what people do with or against kinship”)’ (2000: 14).
5Are we saying that this theme has never been tackled? No, of course not. The point is that even though it has been considered in several works, it has never been faced head-on. And this constitutes a second feature of studies into social practices of kinship: they do not constitute a unified area of understanding.
6The question is always merged with larger ones (methods of social reproduction, transmission and inheritance etc.), although the latter remain focused on kinship. One studies kinship in order to explain it, be it by integrating associated areas which nevertheless remain ‘subjugated’ to the subject of ‘kinship’. The reverse way, though, – i.e. using kinship to explain other social fields – is still rarer. This shift in focus will form our starting point: rather than explaining kinship, we shall attempt to understand how social actions are organised as a result of kinship.
7Our starting point is not therefore to determine what kinship is, nor to define its functions, its representations or transformations, as demonstrated recently in the first issue of the Incidence journal based on the findings of David Schneider, directed by Wilgaux (2005). Our hypothesis is that kinship may be used to do something other than just creating relatives. Thus the matter consists in explaining the reasons, the conditions and the details of practices of kinship, so as to finally grasp the influences that these social practices have on kinship itself.
8In order to define our approach more thoroughly, we must reconsider this fragmented and ill-structured field of study by distinguishing three perspectives that have sought – more than any others – to account for the social practices of kinship. That is: the ‘reticularist’ approach, the ‘solidarist’ one – the latter being subdivided into an approach focusing on exchanges and another defined as ‘strategist’. These three stances complement each other and often stem from similar theoretical starting points. They provide as many paths to explore if one seeks answers to this falsely naïve question ‘kinship, what for ?’.
9To conclude these introductory remarks, let us stress that all this is less about an anthropology of culture, than about the conditions of possibility of an anthropology of actions. To us, the structures, representations and ideologies are mere objects of study, which as such allow to account for the actions of individuals and groups. This praxeology does not aim to rule out culture from its explanatory system. The latter, on the contrary, remains one of our principal axes of research.
10While the adjective ‘reticular’ might not be the most relevant, its use can nevertheless be justified by the place occupied by the notion of network in this collection of works, a place resting on the central idea that it makes a theory of action possible.
11Challenging traditional sociological categories, these works consider society by concentrating on relationships between individuals (Eve 2002). Repeating Mitchell’s traditional proposition, a network can be defined as a specific collection of connections between a defined group of people, with an additional property – that these connections taken as a whole can be used to interpret the social behaviour of the people concerned (Mitchell 1969: 2). As such, the reticularist approach aims to bring up to date the influences that the fabric of relationships can have on individual behaviour, as well as the tensions, asymmetries and dynamics created by other members of society. Furthermore, it aims to take social changes into account,– that is, to understand how the social structures emerge from interactions between individuals determined in part by social structures (Degenne, Forsé 1994: 15).
12All of these works aim to define kinship as a social network rather than as a system, marking a divergence from structuralist and functionalist interpretations. They do not presuppose a world of norms and cultural prescriptions, as the symbolic structuralism of Levi-Strauss does for example; rather, they examine the way in which concrete relationships are being structured.
- 2 We should not confuse formal structuralists (who draw upon procedures of mathematical analysis) wit (...)
13The reticularist approach, however, is not homogenous. The formal ‘structural’ approaches based on complete networks differ from those that examine personal networks2. This distinction is major one (Degenne, Forsé 1994; Eve 2002). In one case, one has to gather all the information regarding the presence or the absence of relations between two members of a social group, whoever they may be. In the other, it is only necessary to concentrate on the Ego network, being aware that all the information regarding indirect relations remains unknown.
14Kinship has been dealt with along these two lines. Following Fortes (1949), numerous anthropologists attempted to make use of mathematical tools such as graphs (Augustins 1998), in particular to understand how the various exchanges within kin groups (property, estates, matrimonial exchanges, etc.) are being structured (Fortes 1949, Foster and Seidman 1981, Schweizer and White 1998). This approach is important as it implies a theory of action in the arena of kinship. The idea here is that kinship can reduce itself to a set of mathematical relations, to a graph, social life being ‘the periodic actualisation and legitimisation of networks – transient coalitions – using methods of legitimisation which are more or less credible, and therefore more or less effective’ (Augustins 1998: 18).
15Without this constituting a formal criticism, we must highlight that excessive mathematical formalisation can often lead to killing ‘flies with dynamite’ (Boissevain 1979: 393). It also supposes to have at one’s disposal a complete network in which to work, something which is rarely the case in practice.
16The other approach adopts a more ‘metaphorical’ use of the term ‘social network’ even if it also uses some mathematical formalisations. Represented by the Manchester School, it distinguishes itself from the structuralist approach by the variety of material it uses (personal networks). As Michael Eve (2002) underlines, this difference is of momentous range. Rather than considering, as structuralists do, systems of relationships and norms autonomous of one another, the Manchester School have tried to show how these systems intertwine, while anchoring their analysis at the level of the individual. The personal network of each individual extends to several relational and normative groups (the workplace, the family etc.), letting room for contradictions between different normative systems and extending the possibilities for action. This leads Mitchell to distinguish the institutional analysis, (the analysis of regular relationships in an established group, like a network of parental units,) and the analysis of the ‘network’ (the personal network). No doubt the desire to take into account the multiplicity of social arenas can be explained by the areas that the Manchester School have been dealing with. As Michael Eve recalls (2002: 194), Gluckman and his followers were sensitive to the problems of multiple adherences to different social systems, because they worked on African societies, which at the time were undergoing a period of marked transition due to urbanisation and also, one may suggest, by an autonomisation of social fields.
17This perspective leads to a major breakthrough as regards the perception of kinship. It is no longer a matter of regarding kinship as a closed system (the structuralist approach) in an attempt to explain it; rather it is about observing kinship as a network amongst many others. Rather than seeking to explain kinship as such, the Manchester School prefer to make the explanation part of a larger reticularist analysis: kinship becomes a social network.
18This is particularly evident in Jeremy Boissevain’s Friends of Friends (1974). For Boissevain, kinship is part of a network resembling others. This network presents two major characteristics: a high density and a high multiplexity of links. Yet nothing distinguishes these links from the others that can connect the individual (Boissevain 1974: 84). In other words, Boissevain denies any uniqueness to kinship. According to him, it is a reticular space like any other, though it might be more restricting and more normative than those of friendship or work. In this perspective, some authors have underlined the fact that relations with family members are ‘semi-elective’ as they can present a certain affinity, though they are also somewhat obligatory and operate within an established context. On the other hand, the amicable or amorous relationship will be described as purely affinitive (Degenne, Forsé 1994: 40).
- 3 The greater part of these works contested Parsons’ thesis according to which industrialisation isol (...)
19The solidarist approach refers to works focused on family solidarities, and which in order to do so, place at the centre of their analyses the way individuals use their own relatives. In France, this question has been dealt with by historians and sociologists more than by ethnologists as Florence Weber notes (2002: 73; 2005), whereas English and American anthropologists developed a strong interest in it3.
20Notwithstanding the disparity of these works, we can distinguish two major trends: the first tendency – which is more to be observed among historians – develops an analysis of the family’s strategy of social reproduction.
21The second tendency, encountered more specifically in sociological studies of contemporary family in Europe, lays more emphasis on the methods of exchange of goods and services. The question of the family’s social reproduction is being replaced by that of the contemporary functions of kinship for the individuals that make up the family.
22These two types of work differ as regards the discipline (history and sociology), the locality, (the first category accounting more for rural situations than the second, which is more interested in urban ones), and the object studied (the mechanisms of social reproduction and inter-familial exchanges).
23Rather than restating the entire historiography of studies into the social reproduction of the family, let it be sufficient to outline them broadly. In the course of the 1970s, the works of Laslett (1972) and the Cambridge group laid the emphasis on the household and not on the kinship links, even though it was no more considered relevant to analyse the forms of solidarity that could be practised between different households. This approach was subsequently criticised, particularly by the anthropological community. Goody, if a supporter of the Laslettian project in its revision of evolutionist plan, otherwise condemned the confusion made between household and family and the ultra-objectivism which drove Laslett to remain indifferent to the contents of family relations within the household and to domestic cycles.
24It was not until the 1980s that the theme of family solidarity emerged, in the wake of Italian micro-historians (Levi 1989). By analysing family links rather than residential links, Levi’s approach brought to the fore the forms of coalition between related non-cohabiting households – the ‘fronts of kinship’, whose collective purpose was to diminish economic and demographic risks through the initiation of strategies of activity differentiation (1989: 91). Indeed, the 17th century can be described as an era of extreme demographic instability: ‘it is death which regularly breaks unions and makes the family a much less stable and protected unit that we might have imagined. To the contrary, instability characterised the family in traditional Europe because of these repeated blows from death’ (Lebrun 1986: 25).
25This demographic context, in which families quickly found themselves ‘in shreds’ (Lebrun 1986: 25), makes it necessary to adapt and adjust continually to all sorts of setbacks. That is what explains the enormous mutability characteristic of Levi’s ‘fronts of kinship’. They can shape themselves according to situations, thus allowing households, for instance, to exchange smallholdings in the light of the evolution of their demographic setup. Consequently, these solidarities appear as central, which does not in any way imply that they are ‘natural’. Far from being gratuitous and spontaneous, they often provide the occasion for the manifestation of family hierarchies (Levi 1989: 77). Lastly, not everyone can set up such a ‘front’: to function, it requires a minimum of economic and social capital (i.e. land and relationships).
26In the beginning of the 1990s, in France, another group of historians (close to micro-history as regards their topics and orientations) approached the theme of family solidarity, in an attempt to challenge Bourdieu’s scheme (Rosental 2002: 124). While following the paths initiated by Yver, Leroy Ladurie or Augustins (with a common focus on the mechanisms of social reproduction), the perspective differs considerably: rather than rules and norms, practices are the bottom line of interpretation. These historians therefore subscribe to the Bourdieusian critic of the structualist concept of social reproduction, which makes the rule an unconscious mechanism functioning beyond the personal will and initiatives of the agents. To avoid sinking into the ‘legalism’ of structuralists (Bourdieu 1972: 1105), they have chosen to deal with family practices and strategies rather than with formal rules. The change in perspective has been considerable, since the rule is no longer a given, but a construction which has to be explained.
27This attention to family strategies and solidarities has led researchers to examine the ways and means according to which a kinship network functions, asking such questions as the operations of inheritance and succession, the distribution of power within the kinship circle or even the social factors producing the ‘fronts of kinship’.
28These questions have been addressed in a particularly plain manner by Laurence Fontaine, who shows how, in a climate of uncertainty caused by high death rates, the heads of Haut Dauphiné families in the XVIIth and the XVIIIth centuries practically managed to deal with unexpected situations, and also how a logic of control and foresight of catastrophes was put in place by very specific material arrangements such as testamentary dispositions (1994: 34). Moreover, this demonstrates that these bonds of solidarity should not be considered as natural or underestimated. In effect, this solidarity often turns out to be weak and short-term, because, as a result of mortality, ‘they do not dare take the risk of betting on a differing form of recognition’ (Fontaine 1990: 1446). Emphasis here is placed on the exterior constraints to the constitution of these fronts of kinship. Other studies, however, have shown that these fronts of kinship are often not simply adaptive responses: to quote a distinction proposed by André Burguière (1986: 112), ‘strategies of cooperation’ as well as ‘conquering strategies’ drive the market economy (Rosental 2002: 134).
29If the contribution made by these strategic readings is considerable, some observations can nevertheless be made. First of all, as Rosental observes (2002: 136), there is a real risk of reduction – of only grasping relationships of kinship through an economical reading, eventually overshadowing the singularity of family bonds. If kinship relationships were solely based on an economic relationship, why would they be favoured as compared to other dependants? When it come to family relations, interests and emotions often become entangled (Medick and Sabean 1984).
30The other pitfall would be to develop a ‘calculated’ reading of Bourdieu’s notion of strategy. At this point, we must briefly return to the rupture that the sociologist provoked with the mainstream structuralist theory. Denouncing the tendency of ethnologists to treat all practice as the execution of a rule, Bourdieu demonstrated that marriage was the outcome of a strategy which ‘sets in motion deeply entrenched principles of a particular tradition, and is able to reproduce, more unconsciously than consciously, some or other of the typical solutions that are explicitly named by this tradition’ (1972: 1107).
31This last sentence helps reduce the equivocality introduced by the term ‘strategy’, which a priori seems to refer to the calculated and conscious choice of a rational agent. For Bourdieu, it is not a question of subject or of liberty, but simply a reintroduction of a socialised agent into the analysis – a agent, in other words, having his own habits and habitus. The habitus makes up the connection between social structures, symbolic structures and practices. Also, matrimonial strategies are not the product of compliance with rules of kinship, but consist in ‘choosing’ the best course of action according to explicit rules of play defining the value of the cards, the game and the art of playing – all of which being known by the players.
- 4 This is a criticism addressed to the work of Laurence Fontaine, who tends to remove geographical di (...)
32The statistical consistencies that the ethnologist observes are the product of individual actions guided by the same objective and incorporated constraints – necessities enshrined in the game’s structure, or partially objectivised in its rules (Bourdieu 1985: 95). The notion of strategy does not therefore correspond to the image of an autonomous individual, unrelated to structural and factual conditions; rather, the notion of strategy corresponds to a sportsman’s idea of the game, who functions within the limits of conscience and discourse (Bourdieu 1985: 94). The analysis of strategies cannot therefore dispense with the fields in which these strategies are inscribed, nor with the habitus – i.e. the ways and means of enculturation. As a matter of fact, this cultural dimension often disappears entirely in a strategist reading4. Discussions of ‘strategy’ should not overshadow the cultural dimension of social actions.
33This very question of the place of culture is again to be dealt with if one tries an approach focused on the exchanges, characteristic of the sociology of family solidarities.
34The question of family solidarities appeared in the field of French sociology in the 1970s, particularly with the works of Agnès Pitrou (1992). She was one of the first to address the question of the different forms of sociability, exchange and solidarity between different social classes. She thus revealed how far exchanges of services, goods and information are structured in different manners according to social class. Pitrou finally demonstrated the semi-self-assured role played by kinship in times of hardship in popular environments. In the 1980s, family solidarities were studied more, since the social context was marked by a withdrawal of the welfare state and an increasing risk of social exclusion. Family solidarity hence became the counterpart to the inadequacies of public solidarity (Burguière 2002: 19). Thus emerges the hypothesis of a new ‘regime of kinship’ – a hypothesis developed particularly by Mendras and Déchaux (Martin 2002: 47). Mendras puts forward the idea that a reorganisation of kinship has taken place, marked by a decline in household cohabitation. Yet this does not signal the end of the parental unit. The latter, to the contrary, still plays an important role. However it presents itself in a new configuration characterised by the establishment of an ‘intimacy at a distance’, which allows a much larger autonomy of agents without affecting the exchanges between them. Lastly, this family sociability does not occur to the detriment of more elective sociabilities such as friends or neighbours (2003: 282-284). Déchaux adopts a similar perspective but insists on the exchange logics brought about by these new configurations of kinship (Déchaux 1990). He stresses the importance of lines of filiation as opposed to the growing precariousness of couples, which is linked to the empowerment of working females. The couple is in jeopardy whilst lines of filiation become more stable. For Déchaux, kinship now emerges as ‘the place of permanence and of symbolic roots’, even if he notes the weakening of family allegiances (Déchaux 1990: 105).
- 5 Martin sees two polar opposites which permit the classification of the different conceptions of mut (...)
35More generally, these studies show that in our contemporary society, kinship provides protection against risks and economic difficulties throughout life; whether it is about job-hunting, a marital separation, or caring for the elderly. The studies also show how the various risks differ according to social strata. The expression of relational support and even the system of mutual obligations varies markedly according to the environment (Martin 2002: 56). In traditional families, be they working class or wealthy, the duty of solidarity prevails. On the other hand those families with a strong cultural base and in which networks are less centered on the family, tend to stress values of autonomy. Practices of mutual help therefore exist which do not always have the same meaning for agents according to their social denomination, type and generation5.
36These studies on the modes of solidarity clearly make it obvious that it is necessary to reintroduce cultural dimensions into the analysis. In effect, the ideology – familial in this case – plays a fundamental role as it directs the ways and means of mutual help. There lies the first stumbling block to the ‘reticularist’ and ‘strategist’ approaches; because of the ideologies and cultural representations which guide to a large degree the ways and means of these practices, the analysis of kinship, (and more generally social networks,) cannot be achieved without taking culture into account. The cultural elements, independent of their formal structure, precede and determine it – in particular the way in which it functions.
- 6 The omission is often intentional, as is the case with Jean-Pierre Dedieu and Zacarias Moutoukias: (...)
37Now, curiously, it seems that this ideological aspect has been largely overshadowed. Everything occurs as if the researcher was concentrating on the structure of the links whilst forgetting about the mental environment which created them6. Consequently, we follow the suggestion of José Maria Imizcoz Beunza advocating a distrust of decontextualised analyses of social networks: ‘Some have defined the social network as a collection of agents linked by a series of relations endowed with certain properties. Importing such concepts may be useful, but in doing so we run the risk of phrasing the question uniquely in terms of individuality – from a perspective marked by contemporary individualism – without taking into account the specificities of the Ancien Régime. The particularity of an historian’s work consists in taking into account these specificities. What were the characteristics which distinguished these links from the type we see in contemporary society?’ (1998: 41). José Maria Imizcoz Beunza thus reminds us that these links were not the product of a free and revocable adhesion, that they were far from being constructed by the agents themselves, since they came first and imposed themselves on the agents. During the Ancien Régime, these links implied a large amount of dependence: ‘the individual was subject to his own community, to the rules which governed the house, the duty to reciprocate, the obligations of kinship, of neighbourly relations, and to those imposed by affiliation to a trade, a brotherhood, a parish, or to the many other societies in which he took part. […] Each link was governed by its own rules, which determined their internal workings. We can consider these rules as the values of the groups to which the agents are linked. They are a matter of explaining the very working of the group, its customs, and at the same time as its practice, the accumulated life-long experience of its members’ (1998: 41-46).
38This remark prompts us to define the notion of kinship with a vantage point differing from what the approach developed by Boissevain would suggest: relatives do not form a network like the others. There are major differences between their networks and networks of elective relationships. Alain Testart wisely reminds us that kinship comes before the relationships that people foster with a multitude of individuals classified in different categories such as ‘mothers’, ‘fathers’, ‘brothers’. This requires a classification of a collection of individuals recognised as relatives. That is, a taxonomy – a perfect amalgamation of differentiated categories. This classification sets in motion positions which are predetermined and organised according to a specific logic, independently of the people which can occupy them. These individuals take their place within a system and by virtue of this position, they have a function.
39This system precedes these agents and accounts for the interrelations that occur by virtue of the place that they occupy. Moreover, it has its own coherence which transcends the individuals and binds them once they enter into it. The very fact of appearing in the system of kinship as the son of a man obliges him at the same time to see himself as the ‘nephew’ of the man’s brother. Kinship therefore constitutes a system. We can choose to join the system, but we cannot change it. Distinguished from kinship, friendship does not constitute a system: it does not put differentiated positions and does not necessitate an organisation nor a classification. Thus, a network can be distinguished from a system (Testart 1999: 40-41).
40In this system, several elements play a structural role, such as terminology – this amalgam of some twenty words which designate the blood relations and the affinity that an individual maintains with others. This terminology enables individuals to position themselves vis-à-vis other individuals amongst the kinship relations which form their society; to imagine themselves, to make known to others their place within the collection of particular social relationships and to imagine the place of others within this group (Godelier 2004: 210). However, terminology does not have a direct causal effect on nor does it correspond structurally with any mode of production, economic or political. Terminology says nothing of political, religious or economic relationships which dominate the society in question (Godelier 2004: 200). This reminder of the structural role of kinship terminology does not require us to stumble into the juridicism lambasted by Bourdieu, something one should avoid by recalling that the behaviour of individuals is not to be inferred from the terminology used, but that the latter nevertheless comprises, for the agent, a frame through which he or she perceives family relations.
41The kinship system cannot only be reduced into a terminological system. It is also formed by economical relationships (transmission of property and exploitative relationships between elders and youngers), political (domination of elders and youngers) – affective, and not just symbolic, which are themselves structured by the rules of naming, the rules of residence and the rules of family representation, as Bernard Vernier illustrates in his book on Karpathos (1996).
42His book the shaping of the unconscious by social relationships demonstrates that the effective family exchanges correlate with economic and symbolic exchanges within the family exchanges are generated socially, can also serve practical uses and regular conduct (1996: 16). This structuration of individual emotions legitimation of relationships of domination and exploitation, transform objectively economic services into badges of devotion and affection. To understand this social structuration of feelings, it is not sufficient to bear in mind terminology, it is also necessary to explain economic exchanges the movement of property at the moment of inheritance and succession, and the symbolic exchanges at work through the rules of naming and likeness which enables the classification of members of the family (1996, 1999).
43Kinship does not therefore constitute a network like any other. It is the product of representations and ideologies communicated by the socialisation of the family. It is a system which shapes economic, symbolic and emotional exchanges between its members, and which is also moulded by reproductive relations and forces that the members foster between each other. Kinship is therefore a dynamic system.
44We shall be pursuing a double argument. Firstly, more than a network, kinship constitutes a system in which the cultural element cannot be overlooked. The second argument is that kinship attempts to serve other purposes than merely ‘kinship’. These social functions are not limited to those described by Godelier, that is to conceive, to raise, to educate, to recognise a child, to transfer property etc. or to facilitate such strategies of exclusion or inclusion as described by Peter P. Schweitzer (Godelier 2004: chap VI; Schweitzer 2000: chap VIII). Kinship is also about going out hunting with one’s brother-in-law, to quote a well-known example. In other words, kinship involves many things other than mere kinship, such as doing business, engaging in politics, globetrotting, being at war etc.
45To study these practices, we have decided to consider chiefly societies characterised by an autonomisation of social fields. This autonomisation is not accompanied by a decline in familial structures, which continue to be one of the major instruments of social reproduction. As a matter of fact, kinship in these societies is a privileged instrument used in order to invest in different social fields, or even to pass from one to another. It is therefore a central resource for the members of a society as well as a setting for action. Thus we are concerned with societies that might be defined as ‘familialist’, yet in which the social reproduction system is not theoretically founded on the family.
- 7 This term is used from Iran, through Turkey, to the Balkans. Sometimes presented as ‘more than brot (...)
46We have chosen to focus on a geographical region stretching from Iran to the Balkans: i.e., on countries endowed with meaningful kinship systems and, furthermore, having in common the use of the Turkish term ‘bacanak’ to indicate the husband of the sister of the ‘woman of ego’7. The comparative perspective is not limited by this main area of investigation: Bernard Vernier's article is concerned principally with Nigeria and Claire Autant-Dorier's on Turks living in France (Vernier 2006; Autant-Dorier 2006). Indeed, this area is more arbitrary than scientifically determined. The unity of this thematic issue is not based on geography, but on theoretical options.
47We must state first of all that this anthropology of action is inspired, sometimes voluntarily, by the sociology of collective actions. Indeed, this sociology provides numerous theoretical tools that we would regret leaving aside, like the concept of ‘frame’, of alignment and dissonance of frames, of mobilisation, etc.
48Our purpose is to understand the conditions of the mobilisation of kinship, then to examine the way fronts of kinship are structured, and finally to analyse how kinship reacts to its social uses.
49Not all relatives are equal and the agents do not resort to just anyone when they decide to engage in an act of kinship. Firstly, we must extricate the conditions of mobilisation of kinship by asking ourselves why an individual calls upon a relative rather than on someone else. Why choosing links of kinship rather than other kinds of social ties? In what situations and for what types of activities does one turn to one’s relatives instead of other acquaintances (friends, clients, professionals)? As such, what are the determining factors in the use of kinship?
50The first reason explaining why the societies under scrutiny favour kinship relationships for social action, is that the family is a space of safety and confidence. This is demonstrated by Christian Bromberger (2005) in his study of Ghilan [in Northwest Iran]. Despite profound demographic and social transformations, the family remains the preferred social space for going into business, finding jobs and trading. More particularly, the ‘agnats’ remain the privileged partners for trading and taking economic risks. Indeed, even if maternal relatives, friends and those from the same hamlet, can be used in migratory situations, ‘Only “agnatic” links stand the test of time’ (2005: 139). This is because they offer a reassuring quality as well as a more important means of control. ‘Daily, all agents’ behaviour passed through this screen: childish insolence, bad manners (not indicating in time one’s congratulations for fiancés…), advantages – imaginary or real – that someone or other would have taken from one’s relatives… The price of agnatic solidarity is one of constant surveillance, kindled by allies’ (Bromberger 2005: 139-140).
51The security character of kinship is not stable in time. It can quickly disappear because of important social change. Gilles Dorronsoro shows for example that the clan was a social unit of solidarity in a Kurdish village in Southeast Anatolia until the 1980s (Dorronsoro 2006). Along with rural-urban migration and the beginning of armed clashes, the clan is no longer an institution protecting its members. It quickly loses its social function.
52However, we must note that as is the case for familial solidarities, the reassuring quality of kinship cannot exist without a relatively constraining reticular configuration and above all, a familialist ideological frame which allows these constraints to be justified.
53The familial frame must not be underestimated. Indeed, it explains certain political and economic practices, as Michel Bauer has demonstrated in an article on French chief executives (1991). Bauer advances the proposition that business is often a familial reality before it becomes an economic reality. This is particularly remarkable when problems of transmission occur: then the pater familias gets the better of the homo economicus, and familial logic appears to be at the heart of economic logic (1991: 23). Familialist ideology hangs over the general politics of enterprise. It determines agreements and transfers on the market of enterprises. The presence of the family, though more discrete, appears to be all-determining as concerns the structure of capital in each business: rarely is the business’ capital open to those outside the family, at least while the business remains of a modest size. In the end, familial realities hang in an important and determining manner over the daily functioning and/or the development of the firm. There is strong vividness of dynastic dreams, therefore of familial ideology.
54The question of ideological frames in which actions take place therefore has to be raised, bearing in mind the perspective that has been revealed by studies on collective actions – i.e. that we deal with collections of beliefs orientating and legitimising those actions (Snow 2001: 28).
55Let us make a preliminary remark. Like the security character, ideological frames can change with time. As far as family is concerned, these changes can be very quick as Gilles Dorronsoro demonstrates (Dorronsoro 2006). Moreover, these ideological frames concerned with the family that dominate the social context of action and define whether or not it is legitimate to make use of kinship are produced by several agents: the state, family, or even individuals. We must therefore distinguish ideological frames promulgated by the state, the family, and individuals.
56First, the state plays an essential role in constructing these ideological frames, as is demonstrated here by the contributions of Galia Valtchinova regarding Bulgaria and Gilles de Rapper on Albania (Valtchinova 2006; De Rapper 2006).
57Gilles De Rapper’s paper deals with ‘biography’. During the communist regime in Albania (1944-1991) everyone was evaluated and classified through his or her ‘biography’ [biografi], which was a judgement, in political terms, on one’s personality, acts and familial background. The ‘biography’ was largely responsible for the authorities’ attitude towards the individual. Gilles De Rapper tries to understand what the word ‘biography’ meant for the people themselves. It appears that ‘biography’ is often synonymous with ‘family or lineage background’. Such an observation leads to explain ‘biography’ as the conjunction of a political concept, the ‘class struggle’ as it was understood by both the authorities and the people, and the family or lineage [fis] as an institution of Albanian society. We see here how the state performs this ideological frame (De Rapper 2006).
58Galia Valtchinova demonstrates this in her case study on the uses of kinship in transborder exchange and ‘trade’ in one of the numerous ambiguous border areas in the Balkans. Drawing on fieldwork carried out between 1997 and 2000 in the neighbourhood of Trun, the study focuses on the local practices and images of trans-border contact and exchange between kin. Galia Valtchinova asks how cross-border ties of kinship are made meaningful and used by local people during socialism and in the first post-socialist decade. More specifically, she asks how and if ‘kinship’ and the ‘family’ has influenced the black market trade and smuggling activities which have been part of everyday occupational activities for most of the local people in the nineties. She shows that because of its moralisation and its politicisation by the Bulgarian State, kinship should not be used as a network to trade. The nationalist and socialist ideological frames shape the uses of the kinship network: it is shameful to trade with family because, although trading is shameful, kin relationship would become irrelevant (Valtchinova 2006).
59The ideological frames promoted by the state use many channels to diffuse like media (Sirman 2004). The media has played a role in defining the ideological frames on the family. Berna Ekal, in her article on the gün (rotating savings and credit associations) observes the influence of television in the urban women's elaboration of the mother-in-law's role (Ekal 2006).
60However, one should not believe that it is the only way in which these frames are produced. Taking the example of Karpathos, Vernier shows how an ideological frame, which often naturalises relationships of domination, is produced on a family level on a day-to-day basis (Vernier 1996). In the same way, Berna Ekal shows how the role of mother-in-law is constructed in everyday interaction. She analyses well the importance of the everyday context. Gün [day], as a specific form of rotating savings and credit associations in urban Turkey, is a distinct ground for women where middle-class values and norms are performed. In this context, discussions on being a kaynana [mother-in-law] help to consider the ways in which the notion of family is central to the perception of women. To oppose the role of kaynanas ‘in the past’ is women’s way of claiming to comply with what they perceive to be the ‘modern’ way of forming a family and, hence, of being ‘modern’. Ekal argues that proper kinship roles shape the relation of people to the people other than their kin (Ekal 2006).
61This paper is closely linked to that of David Behar who shows how these frames are incorporated or held by the agents (Behar 2006). He examines specific patterns of adaptation and transmission within families belonging to Turkish high bourgeoisie. He presents the various levels and criteria of heir’s affiliation to the lineage and shows that internalisation of such schemes of perception and evaluation is directly related to the process of individualisation. Their activation requires compromise with the plurality of situations encountered during the heir’s socialisation, between parental expectations and personal ambitions and between justification of oneself regarding significant others and justification of oneself regarding oneself. Behar shows how the achievements of previous generations are interpreted as part of a continuing renegotiation of heir’s identification to the role he perceives as assigned to him.
62The ideological frames of kinship are therefore plural. They are not produced by the same agents and they do not necessarily present the same perspective (see Dorronsoro 2006; Behar 2006). Indeed the perspectives produced can come into direct competition, or even cause a confrontation. This confrontation can be exacerbated when it concerns access to various resources, as Gilles Dorronsoro demonstrates (Dorronsoro 2006).
63These different kinds of frame lead us to pose a first question, well-known by the sociologists of mobilisation: that of the alignment and the dissonance of these different ideological frames. Galia Valtchinova’s study perfectly illustrates this problem (Valtchinova 2006). Through a study of cross-border petrol trade between 1994 and 1999, Galia Valtchinova explains why Bulgarian citizens preferred not to make use of their relationships on the other side of the border to trade. This refusal can be explained by a moralisation of kinship by the state, which considered such collusion as a hindrance on the move towards democracy. To use one’s relatives to traffic contraband would be frowned upon. This is something we can reformulate as a change in the order of symbolic capital: to have recourse to one’s relatives would signify an important loss of symbolic capital within the family. Hence, people would dispense with kinship and use weaker, less constraining kinds of links, with which it was possible to bargain without scruples. However, this example also demonstrates how important the alignment of ideological frames is. Here, be it at the individual, family or state level, one frame dominates: that which denounces the use of family ties.
64Still, there are numerous instances where there is dissonance between different frames, for example between familial and state frames. Using kinship can be stigmatised as nepotism for example. Other times, it is between the familial and the individual frames that there is dissonance. In Turkey, it is not rare that an individual leaves the family group, following one’s ascent into the civil service. Hence he is perceived by the rest of the group as a provider of services, even if he himself refuses to take on this role, having adopted the ideological frame of the state, whether by strategy or by internalization following a secondary socialising (college, university etc.). Thus, there are dissonances between individual and familial frames, which can lead to misunderstandings and conflicts: the members of the family group do not understand why ‘he who succeeded’ does not help them more willingly.
65This is clearly demonstrated in Gilles Dorronsoro’s article (Dorronsoro 2006). He presents how new frames produced in the course of the processes of urban migration and politisation came into competition with the clan. Generally, people engaged in the PKK are former urban migrants who left the city with their own experiences, keeping a distance with the clanic ideology. In this context, agents could engage in the PKK, breaking with this former ideological frame. Affiliation to a clanic or family-friendly ideological frame is therefore not natural: we see here the importance of primary socialisation and especially secondary socialisation, that sometimes brings primary socialisation into question (on this point, see Behar 2006 and Fliche forthcoming 2007).
66Thus, the dissonance and the alignment of the frames constitute a condition for the use of kinship. The question is to know according to which social logics, the alignment or the dissonance comes about.
67Describing the configuration of frames is not enough. It is also necessary to retrace the individual’s horizon of expectation. Does one have a right to expect something from a cousin, a brother or a sister? In other words, to whom does one have a right and what does one have a right to ask? It is vital that we capture the horizon of expectation of the individual. While we may agree with Jean-Pierre Dedieu and Zacarias Moutoukias when they suggest, ‘that the stating of a link between several individuals does not in itself authorise any inference as regards its consequences’ (1998: 9), we must however add that it should allow inferences as regards expected consequences. These expectations exist in behaviour and in the elaboration of strategies (Autant-Dorier 2006; De Rapper 2006; Fliche, Massicard 2006; Valtchinova 2006).
68At this point the importance of the aforementioned systems of kinship appear manifest. Let us here consider the example given by Bernard Vernier of the Island of Karpathos (1996). The kinship system is characterised there by the existence of cleanly separated sexual lines, both masculine and feminine, based on first names and indivisible estates. The first-born male continues the masculine line of his father, which he in turn inherited. As such, he revived his paternal grandfather whose first name he took, legitimising his right to his inheritance. Similarly, the first-born female belonged to the female line of her mother, which she in turn inherited. She took the name of her maternal grandmother which she revived. The other children were excluded from inheriting, and less neatly classed by their first names and their resemblances. If the youngest boys emigrated, the youngest girls stayed. Remaining unmarried for life, these females served as unpaid helpers to their elders. The result was that solidarity between brothers was weak and indeed weaker than that which bonds the youngest girls to the elders for whom they worked. This emotional family economy, generated by bonds of intrafamilial symbolic forces, politics and economics – from which emanate feelings of ‘obligation’, and which strongly aligns individual behaviour, gives rise to a specific horizon of expectation which indicates what the agents rightfully think they have at their disposal.
69The horizon of expectation defines the stock of links and the means of use, which each agent has at his disposal. He knows that he can count on such and such member of his family more than such and such other. Hence, he can make use of a maternal uncle in order to migrate, for example, more than he can a paternal uncle, because he knows that the first has a moral obligation towards him. Here we can reconsider the case of Turkey where, if one follows Altan Gokalp, people expect various kinds of behaviour from different family members. As such, the relationship between the kayın birader [the wife’s brother] and the enişte [the sister’s husband] is characterised by rivalry and distrust. This is expressed by the subconsciously uttered phrase that one uses when confronted by somebody one judges too friendly to be sincere: ‘bayram değil, seyran değil; eniştem beni niye öptü’ [‘We are not on display nor are we having a party. Why did my step-brother embrace me?’] (Gokalp 1980: 116). Similarly, the maternal uncle, representing the lineage of donors, appears as a rival figure of authority to Ego’s father, while the paternal uncle is a closer person with whom relations are more relaxed and more emotional (1980: 96). Let us also consider bacanak. These are the men who have married sisters. As such, they share a common position in the exchanges that link them to the same wife-giver. This equality grants them equivalent status, abolishing any age hierarchy (Gokalp 1980: 110). However, this equality does not necessarily lead us to postulate any privileged cooperation. As Nükhet Sirman showed (2004), in a highly competitive social structure as is rural Turkish society, individuals look to preserve this equality by avoiding entering into relationships of dependence, hence avoiding the need to have recourse too the other bacanak. Here we will rediscover the ideological frameworks and the hierarchies that they support together.
70The importance of these ideologies can find a confirmation in the use agents make of their parents, as we can see in Bernard Vernier’s article (Vernier 2006). It is because one expects some given rights and obligations that one makes milk and blood parents. ‘Fictitious’ kinship is a kinship where the expectations to one another is well defined, as these people are often more useful than parents by alliance or consanguineous ones. In Nigeria, Vernier shows that the Apari is a mediator of exceptional efficiency in case of conflict within a group, for example between a man and his wife. One expects him to play a protective role, to offer hospitality in this kind of situation within your own tribe or during attacks of sorcery (of which the most dangerous comes from close relatives, such as father or mother).
71Finally, an analysis of social practices of kinship requires a discussion of the expertise and competence of the agents, and also the forms of action they have at their disposal. The aim is to expose the methods of manipulation of kinship.
72One of the first techniques used is to strengthen those kinship links that are the most useful (Vernier 1996: 140). There are many ways of doing this: matrimonial unions, constantly maintaining links with visits (Ekal 2006), exchanges of goods and services etc., spiritual kinship (Vernier 2006).
73Kinship also serves the purpose of reinforcing links that did not necessarily exist before. Claire Autant-Dorier, in her work on Turkish families in France, demonstrates how ancestries are social constructions having a practical significance, even if they are, from a historical point of view, false. These ancestries are used and give rise to effective and fundamental relations in the migratory experience. A genealogical knowledge is therefore at work here, an essential asset in the use the individual can make of the various resources provided by the members of the kinship (Autant-Dorier 2006). Following the same logic, Jenny White demonstrated how, in Istanbul, the working classes used a fictitious kinship to include those persons not related in an imaginary network of reciprocity, and to euphemise the relations of economic power (1994, 2000). In the same perspective, we can see here the article of Bernard Vernier, who shows how and why kinship is constructed through humours (blood and milk) (Vernier 2006). One of the main ideas of Bernard Vernier is to show that kinship is so important that it is necessary to construct it if it does not exist. Spiritual kinship or humour kinship are forms of kinship constructed with a means to an end. So that they function like alliance kinships and consanguineous kinships, it is necessary to install strong prohibitions, like incest for example: ‘It is one of the possible ways of affirming that these kinships (like alliance kinships in the case of those closely connected) are as important as the close kinship referred to as consanguineous to which one firstly associates the ‘forbidden’ person. (Vernier 2006). It is obvious that these types of kinship are social constructs.
74However, consanguineous kinship or alliance kinship is constructed too, as we can see in Fliche and Massicard’s article on the Ulusoy family (Fliche, Massicard 2006). The Ulusoy are the descendants of an Anatolian saint, Hacı Bektaş, linked with alevism and bektachism. In this lineage, a mürşit is appointed, a spiritual guide of many alevi groups. The Ulusoy are characterised by an important role in politics. Seven family members have been elected MPs since 1920, which makes out of them a ‘family of the Republic’ (Hacımirazoğlu 1998). This lineage, concentrating religious resources and political resources is therefore a particularly salient example of the way in which kinship can link several sectors of social life. Religious resources are transformed into economic and political ones. This shows that kinship is a matrix of change of capital from one kind to another.
75This gives the impression that we are dealing with a family collective action. However, the Ulusoy do not constitute a kinship front (see later). Some individuals do, more than others, ‘produce’ the family, by adopting specific matrimonial behaviours. Élise Massicard and Benoit Fliche attempt to show that the Ulusoy MPs and/or mürşit have specific family practices: they have more children on average than other family members and they preferably marry them into the same lineage (Fliche, Massicard 2006). We are therefore concerned with people who look to strengthen links in the family by forming alliances.
76This impression is reinforced by the fact that these people are, without exceptions, donators: they exchange girls more than boys within the family. In other words, they ‘give’ their daughters to the parents, which makes clear their strategy to bring the family closer together. These endogamous marriages permit them to build alliances with other family branches and therefore to reinforce their own place within the family. They give the impression to create ‘real parents’. The logic behind these constructions is the acquisition of centrality, which seems to take into account the quest of the stake of resources, especially the central one: the mürşitlik, appointed by a ‘family’ vote. (Fliche, Massicard 2006).
77Kinship is also a symbolic resource that certain individuals do not hesitate to put forward in strategies of conquest or perpetuation of capital (Pedroso de Lima 2000). Élise Massicard and Benoit Fliche show how members of the Ulusoy family choose whether or not to use this prestigious patronymic and religious charisma to engage in politics (Fliche, Massicard 2006).
78How are the modes of action that each agent has at his disposal constructed? What are the factors determining it?
79Of course, the atmosphere of expectation contributes to the structuration of forms of action in part, but in part only. Indeed, the latter are also determined by the type of operation the individual wishes to carry out. It is not the same thing to migrate with the intermediary of a relative and to go into business with him: the costs, risks and debts are not the same. For certain actions, it is better to form an alliance with a distant cousin with whom you maintain a good relationship, than with an uncle who looks down on you as unequal, notably if there is a symbolic debt involved. It is sometimes easier to soak up a symbolic debt owed to an equal than to someone of a higher standing with whom you forever stay indebted (see Autant-Dorier 2006; Behar 2006; Valtchinova 2006).
80The modes of action are also determined by a whole host of opportunities that the familial network offers (Autant-Dorier 2006; Behar 2006). Not everyone has a maternal uncle, and even if someone does, it is not always ‘useful’: a maternal uncle from the village is of little avail for one who wants to migrate to Europe. And even if someone has a well-placed uncle or cousin, it might be impossible to ask for help due to a previous disagreement. Indeed, family memory is a central factor within the tactics of kinship networks. Symbolic debts are always owed and as is the case for the matrimonial strategies described by Bourdieu, to understand which individual ‘plays’ what move with whatever relative, it is necessary to take into account previous tactics, sometimes going back to one or two generations. Kinship practices quite necessarily imply a complex genealogy of tactics, confrontations and alliances, legitimising future tactics. Each individual is therefore heir to a heritage with which he has to deal.
81On the other hand, the consequences, in case of failure or strained relations, have to be evaluated. A too close relationship, if put to the test by a shared initiative, can result in a profound rupture, as we established when studying a Turkish family of Narbonne. A maternal uncle and his nephew went into business together (a ‘döner kebab’ restaurant) but very quickly relations soured, resulting in a split in the family. The nephew finally bought back the restaurant from his uncle, who opened another 300 metres away from the original. This story was an eye-opener for many Turks in Narbonne, as many now prefer to go into business with those socially more neutral than cousins. The prospect of a family argument plays an important role in the decision-making process.
82Finally, the context of the action is vital to our understanding of social networks of kinship. This was evident in the example given by Laurence Fontaine. In the eighteenth century, one did not take the risk of committing to, and investing over several years in an orphan at death’s door. More generally, the characteristics of the society where the action takes place play a role. Also, the extent to which one has recourse (or not) to one’s family varies according to the place in which one lives. Let us take the example of Narbonne again. Many Turks prefer to go into business with strangers rather than relations. Why? Because they are in a country giving them a certain security with the possibility of efficient legal proceedings in case of a problem, which appears to be less so the case in Turkey. Is that to say that Turks do not go into business together as a family in Narbonne (Fliche forthcoming 2007)? Of course not. Families run the majority of shops and businesses because a family provides the greatest flexibility in relation to the various French administrative constraints.
83Consider for example the case of builders. In a highly competitive environment, Turkish families, in particular brothers, adopt the clear strategy of going into business together. One of the brothers forms a company in his own name. If one brother is declared bankrupt, a common predicament, the other brother immediately forms another company with family capital. This results in a veritable turnover of owners in a family, allowing it to bypass various crises and bankruptcies whereas one sole individual would take great pains to recover from them. We also saw a young girl, hardly an adult, using her skills as a craftswoman to form her own business after several members of her family were declared bankrupt. It is not therefore because of an irresistible familialist ‘compulsion’ that Turks in Narbonne favour their relations when choosing who to associate themselves with. It is this way because they live in a very competitive environment.
84If the systems of kinship produce relations, family groups do not exist without an important ideological meaning. Let us bear in mind here that, according to Bourdieu, a family is a ‘social fiction which institutes itself in reality by breeding each of its members with feelings that would ensure the integration of this group and the belief in the value of this group and in its integration’ (Bourdieu 1994: 11). This integration process, which takes place essentially through educational strategies – socialisation – is all the more fundamental that family is first of all a field characterised by power and exploitation relationships (Vernier 1996). It is therefore necessary that an ideological edifice be in place, in which forces of fusion (emotions) counter-balance the forces of fission (Bourdieu 1994, Vernier 1996).
85In other words, and the systems of kinship notwithstanding, the very existence of a family unit supposes that there be an intention to function as a family. In other words, as shown by Fontaine’s or Levi’s works amongst others, the family unit can be considered as the fruit of a mobilisation, along fronts of kinship produced by family businessmen who are, more often than not, the head of the household.
86This remark on the composition of the family group as a collective agent invites us to distinguish between situations where the individuals use their kinship relations to act, and those where it is the agent is family itself. This also implies a significant difference in goal: in one case, the family unit appears as a mere resource, whereas in the other the construction of a front of kinship makes it a collective agent.
87We have seen that to constitute a family as a collective agent, an ideological framing is required. It also implies that the agents (the heads of households) should know how to mobilise family members around a common project. This capacity to organise is not absolutely evident, as indirectly demonstrated in the article of Michel Bauer (1990). Bauer notes the difficulties to transmit a firm from father to son. He distinguishes three different scenarios. The first is the heir’s working next to his father, being forced to play the role of the heir and to silently await his father’s death. The father gives his heir the illusion that his son will one day replace him, but leaves him no real authority and keeps him in the position of a mere second-in-command. If the father is in good health and the difference in age is not significant, we can see fifty year old sons still working under the thumb of their fathers. In this case, the inheritor does not receive his nest egg for some years, making him at best a ‘transitional pope’. The second scenario is the murder of the father. His successor decides to take over. This invariably leads to a conflict caused by the departure of one of them. The third case comprises of a move away from power, by developing a new activity – a diversification which, in the end, might represent a way to take control of a predominant part of the company. This strategy supposes therefore a high level of entrepreneurial spirit. It permits the transmission of responsibility, while at the same time avoiding direct conflict. In addition, through these examples, we can see the economic consequences of the Oedipus complex. It can especially be pointed out that the fronts of kinship are not formed without difficulty, since the familial economic rationale would have it that the training and transition be carried out smoothly.
88Besides, the constitution of the collective agent that is the family is not evident. We must understand how related people are driven to form a ‘family’ that acts collectively. In other words we must find the determining factors of the transition from a family group to a ‘household’. This transition can be seen through a change in family actions. Drawing upon Michel de Certeau’s distinction (1990), the process should be considered as a strategy rather than a mere tactic: it is not just about seizing an opportunity to place a son, but to place him in a specific place and not in any other, so that he may become useful in a way determined by the family group. This is not a short-sighted approach, but one calculated in the long run.
89This transition from family group to ‘household’ should not be explained by collective strategies of reproduction which stem from it, but by the role played by certain family-mobilisers, by the structure of bonds of force, domination and emotion (see here Autant-Dorier 2006; Behar 2006; Fliche, Massicard 2006).
90Again, we cannot dispense with the social context. In the case of a group of villagers that I studied (Fliche forthcoming 2007), some individuals exhibited familial behaviour in a highly competitive environment, marked by strong family coalitions, underlining therefore the importance of lineage, and adopting more individualist behaviour in less socially controlled environments.
91The final progression of our work implies to analyse the change of kinship groups and of kinship itself as regards its actions. These social practices of kinship are not without effect on kinship itself. They have feedback effects on different levels. Let us consider two of them: that of the family group, and that of the kinship system.
92Taking risks as a family inevitably leads to the straining of family ties. How does a family group react to being put to the test? Do these tensions lead to conflict, and if so, what are the modes of resolution and management: the forming of a group dynamic in case of violent outbreaks or a procedure for conciliation? To analyse this, we must carefully scrutinise the situations where kinship is put to the test; i.e. situations involving a risk being taken and relations becoming strained, or even jeopardised, by the eruption within the family of conflicts of power, money or prestige linked to the exercise of political or economic activities. Consequently these feedback effects have an influence on the way in which family relationships will organise themselves.
93Benoit Fliche and Élise Massicard analyse how access to the parliamentary sphere from the constitution of the Republic onwards, leads to an upsurge in competition within the Ulusoy bloodline, and even the breaking up of the latter. Individual tactics have had negative effects on family collective capital: the uses of kinship made by the Ulusoys in politics contribute to decrease their religious authority (Fliche, Massicard 2006).
94Gilles De Rapper shows also how the use of the ideological frame by the Albanians has important repercussions for the kinship system and the practices which are linked to it, notably in the domain of matrimonial alliances. Those with a bad background are prohibited to marry others, in turn they are obliged to form a lower class (De Rapper 2006).
95The consequences of social practices of kinship are not limited to just one family group. It can have larger consequences for societies. Let us remind ourselves here of the dynamic character of the kinship systems. Sometimes even the terminology changes quickly. So before our very eyes, Polish and Chinese terminologies of Sudanese origin evolve into an Eskimo terminology, just as was the case 2000 years ago with the Roman terminology. In the geographical sphere that we are concerned with, Gilles De Rapper demonstrated as regards Albania (1998), how and why the terminology had evolved so quickly.
96Leyla Neyzi noted similar processes in Turkey (Neyzi 1991). It is therefore necessary to remember that the evolution of social relationships results in people changing the way in which they manage their alliances and their legacies. As a consequence kinship relations begin to change in their content; after a certain length of time, this may lead to a redrafting of terms describing the relations and positions of kinship, ‘first of all through the abandonment of the ancient terms or the transformation of their meanings, then through a global changing of the architecture of their meanings followed by the architecture of terminology’ (Godelier 2004: 531).
97A kinship system can therefore easily be overturned by social changes which it itself produces. Bernard Vernier gives us the example of the island of Karpathos (1996). This is a system regulated like a clock operating according to a principle of disinheritance of the youngest children. The latter, forced to migrate to Europe or the USA, enjoy a social ascent without precedent in the history of the island during the last half of the twentieth century. They return to Karpathos and completely overthrow the social system. Those traditionally dominant in the system [cancares], find it difficult to understand, or are even incapable of understanding the extent of the change that has occurred. Even if their symbolic capital comes to counterbalance the financial capital of the migrants, their domination slowly crumbles. The social system is therefore reformed with an important phenomena of hysteresis.
98Such is the power of change embedded in kinship. This power is a double-faceted one. It can be a key element in social reproduction, a feature that the majority of studies have tackled, but also an amplifier, an accelerator, even an important factor in mutation. More than a matrix of reproduction, it could also be a matrix of social transformation.