The continuous decrease in the average size of households is often considered an indication that families are increasingly of the nuclear type and that the extended family will inevitably disappear. In the past several decades, however, historians and sociologists have consistently challenged this view. Historians have pointed out that households limited to parents and children were the norm in the past, and sociologists have shown that there exist considerable flows of exchange and contacts between relatives that do not live together, and that these are not showing any sign of decreasing. The article by Catherine Bonvalet offers an operational definition of the “local family circle”, a term which designates a certain type of family system characterized by spatial and emotional proximity, but also by frequent contacts and genuine mutual help between members of a family not living under the same roof. The author combines a quantitative study with a qualitative analysis, which reveals the diversity of the social and biographical processes leading to this type of family organization. This organization may represent a reproduction of a situation experienced by the previous generation, but this is not always the case. The process may be intentional or accepted by the persons concerned; conversely, and more often than one might think, it may be felt as an imposition, for lack of any alternative.
1Current research on the sociology of the family tends to focus on couples with children and on one-parent or blended families, i.e. on family groups deﬁned by co-residence. The notions used to describe the contemporary family are those of individualization, autonomy and distancing from one’s relatives, and are thus based on a logic of contraction of the family (de Singly, 1990 and 1993). However, in the past ten years, there has been a great deal of research on the extended family, and large surveys have been conducted on relations and exchanges within the kinship, including the Proches et parents survey  (Bonvalet et al., 1993).
2According to François de Singly, “the misunderstandings that exist in the sociology of the family concerning the role of family and kinship in contemporary societies is due to the fact that specialists reason in terms of the alternative ‘either…or’” . Even though intergenerational relations are better acknowledged today, as witnessed by many studies (Segalen, 1993; Attias-Donfut, 1995), the role of the family is often presented as instrumental, and the family and kinship are reduced to the function of provider of services and social capital. Even if we do know, thanks to L. Roussel’s (1976) and C. Gokalp’s (1978) research, that family life lasts long after the children have left home, we know little about the different types of relationships that exist between couples and their parents. The functioning of the extended family is never studied as such. In order to grasp the diversity of extended families, individuals and households must be considered in the context of their local, family and occupational environment.
3The crisis of the welfare state, the decline of ideologies, and the development of new family forms have certainly encouraged a “new way of looking at families”, as called for by anthropologists, ethnologists and historians who have shown that families have always come in many shapes and forms. Indeed, there is no reason to believe that the diversity of family situations is limited to cases where parents and children live under the same roof. The same diversity should necessarily be found in the study of the extended family.
4One of the most fruitful approaches to the study of extended families is to look at the geographical area in which exchanges take place within the kinship group. In the early 1960s , attempting to describe the breakup of families that according to Parsons was happening in urban environments, Chombart de Lauwe and his team were surprised to discover that family ties were on the contrary quite resilient in the city . At the same time in Great Britain, Michael Young and Peter Willmott (1957) were observing the same phenomenon, and described it in a book that has become a classic.
5Today — and this may be the foundation of a new approach — the individuals can no longer be de?ned through conventional family relations, since they shape their own environment of family and friends by manipulating space, distance, and proximity. To study how space is involved in the dynamics of links of af?nity is to understand the relationship that people establish with their family, the family they have chosen for themselves, because geographical proximity contributes to the construction of socialties by facilitating contacts and exchange (Bonvalet et al., 1999). The way the family organizes its “territories” by concentrating in one area or, conversely, by spreading out, tells a great deal about the links between relatives as well as about the strategies developed in order to reinforce or loosen them.
6Just as the “domestic group”, in other words the household, is de?ned on the basis of dwelling units, we may assume that the location of related persons in the same geographical area makes it possible to de?ne and analyse the composition of the kinship group consisting of several households. Under what conditions does this aggregate of persons represent a coherent and meaningful group ? If this question is not relevant in the ?rst case , it is in the second. Geographical distance between households related by kinship is not a suf?cient indicator to measure the intensity of the relationship. Conversely, the strength of some ties highlighted by the large-scale surveys (Degenne and Lebeaux, 1997; Attias-Donfut, 1995; Crenner, 1998; Ortalda, 2001) does not enable us to “conclude that a new type of extended family exists” (Hammer et al., 2001). This is why Peter Willmott, on the basis of survey results, suggested three large types of family forms in contemporary urban England (Willmott, 1967 and 1991):
- the “local extended” family, consisting of two or three separate households that live close by, meet almost daily and offer each other mutual support;
- the “dispersed extended” family is also made up of two or three households, but they do not live close by. They meet less often, but contacts remain frequent, and they help each other out;
- the “attenuated extended” family, whose members are much less in touch, independently of the households’ location.
7For this purpose, we will extend Peter Willmott’s work by examining how various types of families are operating, on the basis of the Proches et parents survey. We will ?rst use the quantitative part of the survey to determine whether any strong family entities are standing out. In the second part, we will attempt to understand, through interviews, the processes that constituted these units, and how they operate.
I – A strong family entity: the local family circle
8The “Proches et parents” survey, conducted by INED in 1990 on a sample of 1,946 persons representative of the French population, enabled us to study the strength of the ties that link individuals to their kinship network . The aim of the survey was to improve our knowledge of the extended family, to explore the networks of af?nities and study the social practices of the network of relatives and friends. The questionnaire is divided into three parts. The ?rst concerns help received and offered by ego during his or her lifetime, the second describes the universe of persons ego considers as close relatives or friends, and the third lists all the members of the family of ego and of ego’s spouse. Thus three types of networks are identi?ed: the extended kinship network, the network of close relatives and friends, and the mutual help network, and the three do not necessarily coincide.
9In this article, we focus on the persons considered close by the respondent, that is the relatives and friends with whom the respondent entertains ties of af?nity. The information collected on every one enables us to determine the types of relationships entertained by these persons with ego. Exchanges, contacts, as well as af?nity and spatial proximity reveal the existence of speci?c family groups. Taken separately, such variables as frequency of contact, providing help or not, geographical proximity or distance do not mean anything. However, when these variables are cross-tabulated, the speci?c con?gurations that appear become signi?cant (Gribaudi, 1999). The next step is to evaluate, through this array of indicators, the strength of the relations linking different households within a family, and to isolate systemic elements. However, this analysis is limited by the very nature of the survey. The respondent describes his or her family universe from which we infer the existence of a family group by crosstabulating the indicators, instead of the members of the same family reporting the way they live more or less together.
1 – The different types of family organization in the Proches et parents survey
10In the Proches et parents survey, the relationships between ego and his or her relatives were characterized on the basis of four indicators: being mentioned as “close”, living in the same commune or a bordering one, being in contact at least once a week , being part of a mutual help network . We may note that these indicators correspond in fact to four types of links out of the six in the micro-social model of intergenerational solidarity developed by Vern Bengtson and his colleagues of the University of Southern California in the 1970s (Bengtson et al., 1976; Bengtson and Roberts, 1991): emotional solidarity (feelings of affection) which, to a certain extent, can be assimilated to the feeling of being close to a person; structural solidarity (living together or nearby); associative solidarity (frequency of contacts); and functional solidarity (the extent of help provided or received). Consensual solidarity (sharing opinions) and normative solidarity (values pertaining to intergenerational obligations) could not be measured on the basis of this survey. In the present study, the family is viewed as a “microcosm with a capacity to maintain an internal cohesion through its commitment to these various elements of solidarity” (HillcoatNallétamby et al., 2002). Our aim is to understand the notion of solidarity at the micro-social level.
11Even if the goal is not to model solidarity links between elderly parents and adult children, our approach is in the line of an English-language body of research that is opposed to the theses on the primacy of the nuclear family as the form that is most adapted to contemporary society, as well as to the conclusion that the family is declining (inferred from the increase in one-parent families, in divorce rates and female labour force participation), and tries to establish a typology of families.
12In a recent study, we de?ned the “family circle” on the basis of the combination of three criteria (Bonvalet and Maison, 1999):
- af?nities (to be among relatives considered “close”);
- frequency of contacts (at least once a week);
- mutual help (the close relative has been helped by ego, or has helped ego).
13Among the 1,946 respondents, 904 or 46% of the total sample belong to a family circle de?ned in these terms. However, it must be noted that since the questionnaires are individual, the family circle is de?ned in reference to the respondent, and not to the household as a whole .
14Within this group, two family types can be distinguished on the basis of geographical distance: the “dispersed family circle” and the “local family circle”. The ?rst category corresponds to respondents who maintain strong ties with a relative but do not live in the same commune or in an adjacent commune. They represent 17% of the total sample (see Table 1). In the course of migration, families of origin were split, even though a subsequent tendency to get back together mitigated these situations. Geographical distance does not necessarily mean a loosening of bonds: some families, in spite of distance, continue to maintain strong relations by keeping in contact with one or several close relatives at least once a week and through mutual help.
Proportion of individuals belonging to a family circle according to geographic proximity and type of relative involved (in%)
Proportion of individuals belonging to a family circle according to geographic proximity and type of relative involved (in%)
15The second category includes respondents living near a close relative with whom they have close ties. In this case, the criteria are intentionally restrictive, since the objective is to identify a strong family cohesion. In order to belong to a local family circle, the respondent must not only maintain a relatively active relation (that is to say, consider that person to be close) with at least one relative (father, mother, father-in-law, mother-in-law, adult child not living at home, brother, sister, brother-in-law or sister-in-law), but also live close by, be in regular contact, and have helped or been helped by that person. Of the 1,946 respondents, 579 (30%) belong to a local family circle de?ned according to these criteria. If one takes into account the family of the spouse, the results are modi?ed and the proportion of individuals belonging to a local family circle increases from 26% to 30% (Table 1). The concerned households are mainly those of parents and of adult children. Only one quarter of the local family circles include the household of a brother or sister. In this type of family, vertical lineage is predominant (taking into account the fact that the only collateral relatives taken into consideration were siblings, while cousins, nephews and nieces were not ). The local family circle seems to be a kind of extension in time and space of ego’s own original nuclear family of origin, since in-laws remain in the background (5.4%).
16In approximately nine out of ten cases, this type of family is made up of two households, including that of ego. The others are most often made up of three households, with several possible con?gurations in addition to the household of the respondent: the household of his or her parents and that of a brother or a sister; of the parents and of a child; of a brother or a sister and of a child. Family circles including ego, his or her parents, and adult children and siblings are practically non-existent. In-laws may sometimes be added to these con?gurations: 3% of the respondents include both sets of parents in their family circle.
17Among these families, there is a speci?c sub-group which we have called the “semi-cohabiting family circle”. These families enjoy daily contacts. Whereas cohabitation in the strict sense of the term among generations has become quite rare in France, the fact of living nearby and remaining in close contact remains true for 15% of the respondents of the Proches et parents survey. This, given the chosen criteria, is considerable for an almost totally urbanized society. This family lifestyle may be called “living together at a distance”. In fact, this type of semi-cohabitation may well have replaced the domestic cohabitation of past generations.
18In the end, it turns out that nearly half of the respondents enjoy relationships with their relatives that involve relations on a weekly basis, emotional closeness and exchange of mutual services. The isolation of the nuclear family or the decline of the family described by some sociologists seems rather relative. This proportion is all the more striking as it is probably underestimated. The qualitative interviews con?rm that many respondents had not mentioned, in the part of the questionnaire devoted to close relative and friends, that they had helped or had been helped by a close relative or friend, whereas such help is mentioned in the section devoted to dif?cult times or to mutual aid provided on a regular basis. For some, indeed, to exchange services between the members of a family is so much part of the relationship that they do not mention it. As for others, they have not, as yet, felt the need for such help, but this does not mean that they will not receive any if they experience dif?cult times. Thus the very notion of mutual help is of limited application (Ortalda, 2001). If we set aside the help criterion, 45% are local families and 25% are dispersed families (Table 2).
Distribution of families by type, according to whether or not the aid criterion is taken into account (in %)
Distribution of families by type, according to whether or not the aid criterion is taken into account (in %)
2 – Local family circles
19Are local family circles different from other families ? To answer that question satisfactorily would have required interviewing several members of the same family. The individual responses can nonetheless be made to yield some elements of an answer. Persons who belong to local family circles have speci?c demographic, social and family characteristics that can easily be identi?ed by means of a logit model (Table 3). As would be expected, this type of family system involves women more often than men (the marginal effect amounts to 4%). Apart from the effect of sex, however, the purely demographic variables, age and household type in particular, have no impact. The local family circle is not more frequent among the pivotal generation aged 50-64 than among the younger generations. All age groups are involved in the system of mutual help, the young and the elderly being usually on the receiving end, and the members of the pivotal generation being more often givers (Attias-Donfut, 1995). Similarly, one-parent families or persons living alone are no different from couples. Since women are at the centre of family relations, the presence of a spouse does not in?uence matters either way; and the mother-daughter bond in particular endures whatever the vicissitudes of marital life (Bonvalet and Maison, 2002). In fact, the truly discriminating variable is not the type of household but the sex of the respondent. It would seem that one-parent families are more intensely involved in the mutual help system simply because in a vast majority of cases, the reference person is a woman.
Probability of belonging to a local family circle, by selected characteristics of the respondent and his or her family (logit model)
Probability of belonging to a local family circle, by selected characteristics of the respondent and his or her family (logit model)
20Rather than demographical determinism linking the position in the life cycle to the functioning of the kinship group, it is the past that provides one of the keys to the existence of local family circles. Past history and the succession of places of residence over time play a decisive role in the way of life of these families. Regardless of the number of moves, the fact of remaining in the same département as at the time of leaving home is one of the crucial factors. Geographical roots and local family circle go hand in hand (the marginal effect with reference to those who left the département is 28%). Even though, as Anne Gotman has described, a local family circle can be the result of a “family migration”, sedentary life strongly favours the development of this type of family organization. The same is true for couples who lived for a time with their parents, a sign of closeness between parents and children that extends during adult life despite residential separation (the marginal effect is 8.02%). It seems that this sequence in a residential trajectory is not a random event but on the contrary reveals a certain type of domestic organization where the local family circle is an extrapolation of the extended family (or multiple household) de?ned by P. Laslett (1972a, 1972b). Homeowners tend to live in local family circles more often than others (the marginal effect amounts to 5.36%). These results con?rm the studies that have shown a very strong correlation between home ownership and the wish to “create a family” (Bourdieu, 2000). As we shall see in part II, this project is not restricted to forming one’s own family but sometimes involves settling grown-up children near their parents’ home.
21Certain features of family history can also strengthen the ties between the members of a family. Thus, the death of the father reinforces family ties (the marginal effect comes to nearly 10%) . But contrary to a widely held view, large families do not necessarily generate strong ties: local family circles more often involve small than large families. In addition, the size of the family network can promote a strong sense of cohesion. The more close family members a respondent reports, the higher the probability of maintaining strong ties with them.
22These features of residential and family life are so powerful that social position seems to play a secondary role. However, a contrast does seem to exist between blue-collar workers on the one hand and “managers” , intermediate occupations and other white-collar workers on the other  (the likelihood for managers to have a local family circle is 8% lower than for blue-collar workers).
23These results con?rm numerous studies on differences in sociability between social classes (Héran, 1987 and 1988). Moreover, if executives and managers are less likely to belong to a local family circle than blue-collar workers, this is not only because they have a greater ability to sustain relationships from a distance, but also that they are more mobile than workers or lower level staff (Blum et al., 1985).
24 Many sociological studies, and large surveys such as those conducted by INSEE, have shown the predominant role of women in maintaining family relations (Crenner, 1998; Blanpain and Pan Ke Shon, 1999a and 1999b). Gender roles have been invoked. The family is assumed to be the special sphere of women, while friends and social relations are the world of men. The results of the model point in that direction, since women are more often involved in a local family circle system. However, not all the variables that are used in the general logit model have the same impact when men and women are studied separately (Table 3). Age and household type play no role among either women or men. On the other hand, although residential mobility has a negative impact on the likelihood for men to belong to a local family circle, it has no impact on women, as if, regardless of how often they moved, women managed to maintain the geographical proximity necessary to a local family circle. This is also true as regards the environment: the fact of living in a rural or urban environment has no impact whatsoever on the relationship between women and their circle, unlike what is observed for men.
25Women also react very differently to demographic events or to the characteristics of their family of origin. For women, the death of the father or the fact that their mother is still alive are favourable to family life as a local family circle (in both cases the marginal effect exceeds 10%), whereas these factors have no impact for men. In addition, being the eldest child or having many brothers and sisters leads men to have more distant relationships with the family, at least in geographical terms, whereas the woman’s rank among siblings will have a lesser in?uence on her special relations with her relatives.
26Social class has a different impact depending on the sex of the respondent. For men, working-class households are differentiated only from the households of managers, whereas working class women differ from both middle and upper class women. This may be due to the speci?c nature of relations between women in the working classes . Olivier Schwartz’s ?ndings (1990) are grounds for not dismissing this hypothesis out of hand. The strength of the mother-daughter relationship is even clearer in the case of semi-cohabiting families (Bonvalet and Maison, 1999).
27At this stage of our analysis, we are able to con?rm the existence of several types of extended families, as suggested by Peter Willmott, and can subject them to quantitative analysis. However, since the local family circle is a statistical construct obtained by cross-tabulating several indicators that link one individual with another belonging to a related household, the question to be addressed is how to get closer to the reality that lies behind this category. Is the local family circle a real entity made up of several households belonging to the same family, that together constitute a tight-knit group (Bengtson and Roberts, 1991), or is it merely a network of exchanges within the kinship group ? In other words, can the notion of local family circle be considered comparable or equivalent to the more anthropological notion of “houseful” (maisonnée) ?
II – The local family circle, as seen through life histories
28The statistical results presented in part I are no more than a description of how these families function. For this reason, the quantitative survey was supplemented with about one hundred semi-directive interviews, in which the respondents were asked about their family and residential history . Of the 99 persons who were re-interviewed, 37 had a way of life that corresponded to a local family circle. The material thus collected allows a better understanding of how the “topography” of the kinship group and the functioning of the local family circle are constructed over time. Thus, the next task is to get closer to the real life situation or rather situations that underlie family relationships, as described by the respondents .
29The life histories reveal considerable variation between the networks of close relatives and friends, both in their composition (proportion of family members, inclusion of friends) and in their geographical location (whether the entire network lives in one neighbourhood, or the same commune, or conversely if only one person lives nearby). The local family circle, which stood out as a clearly de?ned entity in the statistical analysis, in fact encompasses many different realities, independently of social class. The results of the logit model, which made it possible to identify several signi?cant variables, point to family history as an initial line of inquiry for studying the various modes of formation of local family circles. In a subsequent stage, we will see that the way in which they are constituted in?uences the actual nature of the resulting family forms
1 – Different modes of formation of local family circles
30The history of respondents and their parents, or even their grandparents, enable us to distinguish several types of local family circles. It is this genealogical depth, obtained by simultaneous observation of several groups of relatives, that reveals the diversity of the different family lifestyles that take shape, at each generation, on the basis of multiple heritages (Bertaux 1987).
31From the outset, a contrast is observed between individuals who belong to a family circle without deliberately choosing to do so, and those who prefer this way of life and de?ne themselves in relation to their families of origin. A ?rst difference concerns the attitude of respondents towards their own family history, viewed in terms either of reproduction or, conversely, of distancing. A second difference arises over the attitude towards the spouse’s family: it is either adopted or on the contrary kept at a distance or even rejected. Many combinations are therefore possible: reproduction of the model transmitted by both families in the case of homogamy; adoption of one family and rejection of the other, depending on personal history and on the social and cultural differences between the two families; or the wish to keep the preceding generations at a distance while compensating by a strong emotional investment in the children.
32The local family circles observed at the time of the survey result from four different processes: reproduction, creation, adoption of the spouse’s family or constraint/survival. In the ?rst case, the individuals reproduce the local family circle way of life of their parents and grandparents, whether they have stayed in the same place or migrated as a family. In the second and third cases, by contrast, they are trying to make up for an unhappy childhood or the consequences of migration or a war, though the solutions vary depending on the spouse . Some people ?nd that their in-laws provide the family circle they would have liked to have in their childhood and they adopt the same way of life. Others build from scratch a family circle in which they can feel secure. These three types of family circles bring together individuals who have chosen this way of life in which residential autonomy is combined with the preservation of strong family relationships. The last type, on the contrary, concerns individuals who in fact had no real choice: at the time of the survey, they were compelled by economic or family reasons to live close to their parents’ home.
Reproducing the family circle
33Of the 37 interviews of persons belonging to a local family circle, nearly half illustrate the phenomenon of reproduction. In this type of con?guration, the respondents have spent their youth internalizing the family’s way of functioning and the relationship maintained over this period with the grandparents both by the respondents themselves and by their own parents. This genealogical depth affords an insight into how the notion of what it means to be “a family” is transmitted. There is a genealogy of people, but also of places. Loved ones are very often identi?ed with the places where close relationships have developed. And these places possess a meaning that must be included in the set of complex interactions that have occurred in the past and which can be reactivated throughout a person’s history. As Anne Gotman (1999) wrote: “Can places be inherited, passed on, taken over ? Or are they not merely a re?ected image of family relationships that are de?ned elsewhere ? These places represent a whole that is inseparable from family relations, but there is more to them than that. The social and environmental attributes of places also play a role”. Thus, the family cannot be studied without reference to these places, and in turn, setting and sociability must be studied together (Maillochon, 1999).
Reproducing the family circle as the result of a choice…
34By accident, the selection from the sample for the Proches et parents survey included a woman and one of her daughters. This means that we have questionnaires and in-depth interviews for both of them. These interviews have been the subject of a more detailed analysis (Bonvalet, Maison, 2001). The mother (Christine) and her daughter (Valérie) belong to a family system of the local family circle type. Christine is 59 and comes from a well-to-do family with ?ve children; she reproduced this large family model by having ?ve children herself. Her husband’s career in international management led him to work abroad, and she lived in Hong Kong, Tunis and ?nally Amiens where, after his death, she remained with the two younger children. Some years later, she decided to move back to Paris where her parents and four of her children were living. Valérie is married to a journalist and has a little boy. She works as a public relations specialist in a large commune of the Paris suburbs.
35Their stories allow us to observe the mechanisms at work in local family circles that correspond to the reproduction-appropriation model. The model is transmitted essentially between women, as Christine relates:
“I am still closer to my sisters than to my brothers, even though I have a very good relationship with my brothers…I have a completely different relationship with each of my ?ve children … My eldest son is very much the big brother and is a bit straight about some things…. He is both very conservative and traditional but also rather unpredictable, he does things off the top of his head, without warning. That’s Laurent, we’re very fond of him but the girls laugh about it, they say ‘well there goes Laurent again’… Valérie is perhaps the pivot for everybody.”
37According to her mother, it is Valérie who has most closely reproduced the family model although this has not prevented her from adapting it to make it more open to the outside world. Her version is rather an “equivalent” reproduction than an “identical” one (Bertaux and Bertaux-Wiame, 1988). In this case, the strong relations between the members of the family coexist with an equally strong level of sociability outside the family. Far from becoming a source of competition, the entire network of friends starts to function along the lines of the family model, as Valérie’s mother explains:
“All in all, the children demonstrate that this generation has introduced something that didn’t exist for my generation, or at any rate not in my milieu, it’s the idea of belonging to a generation. Our way of life was, I don’t know, vertical… And my son Laurent has held on to that image, very much so, whereas his sisters very quickly became much more dependent on their group of friends, and even now that they are married, two types of lifestyle show up, and they spend a huge amount of time with their friends.”
39Christine is pleased to see that in the next generation, her daughters are doing things as before: “They are very close to one another, they see each other often, their children go to the same nanny, the same thing is starting over again, very much so…”
40Valérie’s role as driving force clearly appears in the family’s pattern of residence in Paris. Although her maternal grandparents live in the west of Paris, Valérie chose to live in the east, near her mother, and this is where her sister Céline and friends from Amiens later came to live.
“I found a ?at for my sister because we wanted to live near each other… Her ?at is not far at all. It’s nice, mother lives not far away… I’m the ?rst, after mother, to have come to live in this neighbourhood (…) Patrick [her husband] joined me, and other friends, so I didn’t take me long to like Paris, not long at all. But I did bring half of my life at Amiens with me! It’s the people that count, more than the city itself.”
42Lifestyles both inside and outside the family take shape gradually within households according to the experiences of the spouses. Thus, each generation rede?nes its own form of family life in relation to the parental models in the husband’s and the wife’s families. In the case of Valérie, the model reproduced is that of her parents, not of her in-laws. Valérie’s husband seems to have adopted his wife’s large family model:
“… he regrets being an only son. Oh yes. He doesn’t want to repeat that, and besides he gets on very well with my brothers and my sister, so he’s taken our family, he’s adopted a new family, there’s no doubt about it.”
44In fact, the two families differ in every respect. One is a large upper middle-class family with an extensive network of relations and numerous geographical attachments; the other is a small family unit, of modest social background, and whose residential space Valérie ?nds somewhat limited. Indeed, it is exactly this point that she uses to illustrate the contrast between the two families:
“… with a tiny little garden in a row of houses. The house isn’t at all… I like it because it’s cosy and they’re so sweet and there’s always lots of things to eat, it’s the complete opposite of my family, there’s a social difference. It’s a different way of life.”
46The history of Valérie and Christine must be placed in a wider context. They belong to the Parisian well-to-do. It does seem that in their case, in M. Gribaudi’s (1999) words, “the centrality of the family and of the kinship system appears to be directly linked to very powerful mechanisms of spatial implantation and social reproduction, and also of social discrimination”.
… or the consequence of an asset
47If reproduction is often the result of the appropriation or adaptation of the family model through a process of identi?cation with one of the parents or grandparents, it acquires an additional dimension if this parent is the head of a business or farm. The very existence of this economic capital leads the related households to adopt a speci?c lifestyle in which private life and work are intertwined. We observe in this case the pattern of semi-cohabiting family circles, since for professional reasons the members of the same family meet daily and live near the business.
48In the type of family being studied here, this constraint is accepted by the individual members who have made a choice to remain with the family business. Georges Menahem (1988) has described this type of property-owning family lifestyle through the domestic organization and the type of relations these households maintain with their relatives. In his words, “the patrimonial type of family organization, which is tied to the importance of safeguarding and extending the family wealth, creates a hierarchical domestic organization which mobilizes all the energies of a household, or even a lineage, under the authority of a paterfamilias who both owns the property and controls the process of production and reproduction”.
49Paul spent all his childhood in a family dominated by the personality of the grandfather and memories of Italy. The persons he considers close are family members and in-laws. His grandfather, of Italian origin, was a businessman who, with the help of his brothers, built up the company they had founded into a thriving enterprise. He seems to ?t the image of the paterfamilias to perfection, with all his children, nephews, and grandchildren “at his beck and call” both at home and at work. Everything about this man seems exceptional, even his longevity:
“He came to France from Italy at the age of 10 or 12, speaking no French… He was really someone quite exceptional and he died at 98.”
51This man’s life story, as related by Paul, resembles an epic saga. Despite early hardship as an Italian immigrant in France, he climbed the social ladder and founded a genuine dynasty. In a way it was the “American dream” of professional success come true:
“… many members of the family work in the company. At one time there were ten of them, including sons, grandsons, daughters-in-law and sons-in-law of the three brothers…”
53Even if the elders do not speak Italian, their lifestyle, behaviour, culture and eating habits are all reminiscent of Italy, as Paul recalls:
“… I remember our family reunions at my maternal grandmother’s… you got there and there was nowhere to sit down because of the ravioli that was hung out to dry. There was ravioli drying everywhere!”
55In Paul’s view of his in-laws, we observe the same disparity already observed with Valérie. On the one hand, we have a large family of businessmen of Italian origin, a family full of charm, memories, and with a heroic dimension; on the other hand, a family which, as Paul describes it, is “a traditional French family, no fun. I’m just joking — they’re from the Beauce region”.
56In family circles based on the reproduction rationale, vertical lineage is very important since the members of the family accept the family model and transmit its values, practices, and in some cases, technical know-how. Transmission rarely means repeating the model in an identical fashion: each generation adapts its way of life. Valérie does so by opening her network of relatives to friends, others on the contrary by focusing exclusively on the family. But even when friends play a large role, the family comes ?rst: “friends are ?ne, but family is always more important”. (Valérie)
57But for models to be reproduced, even to the point of equivalence, the in-law family has to be taken into account given that family ways of life are the result of a combination of lineages. For some, this is not a problem because of social homogamy and similarity or compatibility between family cultures. The in-law family is in effect integrated into the family. There is even a desire for equality: one must do for one’s mother-in-law what one does for one’s own mother. At times, however, reproduction of the family model is to the detriment of the in-laws, who then appear to be lagging behind. Reading the interviews, we can even perceive a desire to keep a distance from this “different” family and a lack of understanding of its way of doing things. The family of ego is perceived to be superior and its lifestyle and references are those that ego wishes to transmit, after adapting them. This leads to a kind of exclusion of the family-in-law that must not be allowed to disrupt the reproduction mechanisms. This representation must of course be quali?ed, since we are dealing with a narrative, whereas the reality of everyday life always involves varying degrees of compromise, as numerous studies have shown (Lemarchand, 1999; Attias-Donfut et al., 2002).
The local family circle as a compensation mechanism
58“The renewed interest in the family encourages a tendency to overestimate its positive aspects. Its role as a source of support, mutual help, integration, and provider of resources is often highlighted, but we tend to forget that the family can also be a source of problems and disabilities which hinder the social success of individuals (Bonvalet et al., 1993). In the Proches et parents survey, nearly half the respondents reported having experienced serious family crises during their youth. The main causes of disruption came not from the internal functioning of the families but were exogenous. Death or disease affected nearly one third of the families.
59In addition, it has to be recognized that not every family is necessarily a locus of solidarity. Over 13% of the respondents have experienced the separation or discord of their parents as a grave disruption of the family bond. The corpus of interviews contains the same types of disruptions as those identi?ed in the quantitative survey: death, disease, parental divorce or dissension, bankruptcy or unemployment, as well as abandonment, isolation, poverty, abuse, the consequences of being uprooted, or family dysfunction. In addition, these life histories provide some concrete illustrations of a tragic impact on families of historical situations, an aspect that often does not appear in surveys: in this case, the rise of fascism in Italy, the Spanish Civil War, World War II, the Algerian war… Families were separated, or lost one of their members. One or even two generations later, the traces of these events are still felt, in the sense that life courses have taken a radically different direction.
60For the persons who experienced these trying times during childhood, family life seems to follow two opposite modes, which represent the two extremes of a continuum along which the predominance of one or other attitude is more or less marked: either repetition (reproduction of abuse, divorce or other forms of instability) or on the contrary, compensation. The local family circle that maintains strong ties between related households would then represent a kind of reaction to a dysfunction in the family of origin. Among the 37 persons who belong to local family circles, a dozen had experienced dif?cult times in their youth. They seemed all the more attached to this type of family system as they did not have it during their childhood, and still felt this emotional deprivation as something painful. They intended to give their children the “real family life” they themselves did not have. They gave the impression that by doing this they were trying to mend their own childhood through the one they are creating for their own children. Sometimes, this effort of compensation was directed towards their own parents, since some managed to retain strong ties, or renewed their ties after a break, as if, in spite of everything, they wanted to “secure” grandparents for their children and maintain the link between generations. For others, the break seemed ?nal, at least at the time of the survey.
61Whereas everything seems to develop naturally in the “positive” reproduction pattern studied previously, in this case the situation is created deliberately. The aim is to do things differently and above all to avoid reproduction. In this case, we will speak of “local family circles of the compensation type”.
62The interviews enable us to distinguish between two different ways of creating this type of local family circle, depending on the links that were or were not established with the family of the spouse. In the ?rst case, the spouse had also experienced serious events during youth or had established a certain distance vis-à-vis his or her family , and the family circle had to be built from scratch. This is the “healing-creation” family in which the emotional investment in children and grandchildren is very strong. In the second case, the in-law family was very present, often lived nearby, and provided help and support to the couple. It constituted a second family for ego, who adopted it without reservation. This is the “compensation-adoption” family .
Making up for a difficult childhood by building a family circle
63Through two case studies, we will try to identify the mechanisms that lead individuals to create a family organized as a private micro-community based primarily on kinship.
64Denise is a housewife, married to a senior manager who, like herself, comes from a working-class background. Denise’s childhood was disrupted by her parents’ divorce. The case of Denise has been the object of a more detailed study (Bonvalet and Maison, 2001). Her parents moved from Narbonne to Paris when the father, a civil servant in the post of?ce, was transferred. Cut off from family and friends, the couple had trouble adapting to their new environment and separated when Denise was nine. Her mother stayed in Paris, while her father went back to Narbonne, the city of his birth. During her childhood, Denise lived two years with her father, two years with her mother, and in between with her maternal grandparents: she was, as she puts it “a bit bounced around by family events”.
65Denise currently enjoys close relations with her three children and their families. She sees her present way of life as a reaction against the counter-example of her parents’ behaviour when she was little:
“On the whole, childhood and adolescence were not unhappy but not happy either. Not a ful?lling childhood like the one I tried to give my children later on. I still feel some regret for a messed-up childhood.”
67Denise clearly expresses her refusal to reproduce her mother’s behaviour, which for her represents what not to do:
“I didn’t want to be like my mother, I thought she was… well, that in fact, she hadn’t looked after me and I didn’t want to be like her at all, I wanted to do the opposite of what she’d done, and since I also adored my grandmother, and since my mother… so in a way, my grandmother was the model and my mother the counter-example.”
69The experience of having both lacked affection during their childhood is something that draws Denise and her husband closer. For both, the grandparents represent the stable family reference.
“That is, his mother went to work, so the grandmother brought up the children… It’s the same thing, his family life wasn’t unhappy, but it lacked warmth.”
71For both, therefore, it was important to create ex nihilo a family system that was radically different from what they had known in their childhood.
“Exactly, my approach, if you like, was to provide my children with the family life I hadn’t had, so for me going to work was out of the question, because I wanted to give my children the presence that I hadn’t had.”
73Denise quite deliberately created around her a family world based on “concentricity” (a term used by Olivier Schwartz, 1990), which is the complete opposite of the geography of her childhood, split between Paris and Narbonne. Her network of close persons is for three-quarters made up of members of the family. Denise sees her children and grandchildren, who live in the same commune and some even in the same neighbourhood, every day. Far from being the result of chance, this residential proximity with her children owes much to her commitment to the task of ?nding housing, which she undertook for them. Thus, Denise has recreated a family world that is systematically opposed to her own childhood experience.
74Lucie began the interview by saying: “I was a welfare child”. The story behind those words is that of a person who has suffered, who was robbed of her childhood — which she doesn’t want to talk about — the story of a life spent struggling against the consequences of ill-treatment. Lucie hardly knew her parents. Her mother died two years after her birth, and then her father abandoned her and her two sisters. All her life Lucie has been in poor health because of the bad treatment she received in her foster family. Her four pregnancies were dif?cult and she was forced to stay in bed, either at home or in hospital. Her personality has been marked by this unhappy childhood. She is, as she says, “like a wild animal — we are how we were brought up”; the places she likes are “the most out of the way, the most hidden”.
75Her residential choices, both for main and holiday homes, are determined by this aspect of her personality: “I asked for this place because I’m a bit wild”. The same is true as concerns relationships with friends. Like Denise, Lucie’s main emotional investment is her family life. She wants to create the family world which neither she nor her husband, a worker in the building trade, experienced in their childhood. Her daughters live 20 minutes away from the house; her son also lives nearby. They form, as she insists at several points during the interview “a nice little family”, providing plenty of affection, help and support. Since the family is not particularly well off, exchanges take other forms. Lucie accepted to stand security for her daughter when she rented a ?at and for her son when he bought a car. She does not hesitate to invite her second daughter and her son-in-law to her table every day so that they can save to build their house. Similarly, she did not hesitate to change her mind about her son-in-law, whom she didn’t like, so as to stay on good terms with her second daughter, because “family is more important than all that, family is sacred. Family is very important to us, because I gave them what I didn’t have, at least I hope so”.
76In the qualitative interviews, we ?nd other cases of the “compensation” family type, in particular when there were con?icts with parents during adolescence.
77Making up for an unhappy childhood by adopting one’s in-laws
78Another way of achieving this compensation process for an unhappy childhood is to join the family of one’s spouse. This implies that the latter has chosen a pattern of reproduction of the parental model and that ego accepts this model. In fact, the interviews show that through the spouse’s family, what ego really discovers is the atmosphere of a “real family”.
79Like Denise, 49 year-old Robert, whose mother died when he was ?ve, spent part of his childhood living with different people: with his grandmother in the south of France, at his godmother’s, with his father in Paris, then again with his grandmother in the south. His father remarried with his deceased wife’s sister, and took on a hotel-restaurant in the south, where Robert joined him when he was 14. There, he helped his father with a quarry business while his brother worked in the family hotel. Father-son relations were dif?cult:
“He didn’t realize I was young and that I worked 14 hours a day — I had to help, there was a lot of work. I wasn’t well paid, you know, and sometimes I got fed up working for the family, you know, it’s not always easy.”
81When he got married, he discovered family life and adopted his in-laws, with whom he found the support and reassurance he never had in his own family:
“My in-laws, they did everything for us. With them, I don’t know, it comes naturally to them, they could see right away if something was wrong or what, and they would help us however they could, with what little they had”.
83The in-laws ?rst helped by lending them a place to live during the early years of their marriage, then later by giving a holiday home to be shared with the wife’s sister, and ?nally by helping the grandson to buy a ?at. At the time of the interview, the relationship between Robert and his in-laws had been reversed, as he explains: “when we’re ill, they’re the ?rst to come… now we do the same for them because they’re old”.
84The in-laws represent a model of solidarity that Robert and his wife reproduce with their own children: “I did for my children what they did for us. I tried to… well I helped them as much as I could”. This help is ?nancial but also domestic. Both of Robert’s children are divorced. Their daughter, who lives with her son, leaves him with them when she goes to work, and their son, who does not have custody of his daughter, brings her over every other weekend. When a change in occupation forced Robert and his wife to move, their choice of area was determined by the wish to remain near their daughter.
“There were two criteria, ?rst, we had to do it quickly because the company was moving… and second, I didn’t want to be too far from my daughter, since she brings over the little one, we didn’t want to be too far away.”
86It should be noted that this adoption by the in-law family makes it possible to give grandparents to one’s children and provides them with a model of family stability and with roots that were lacking in ego’s own family.
The “passive” local family circle
87In all the cases studied so far, whether reproduction- or compensation-type family circles, the form of family organization and functioning re?ected a conscious choice, a desire for continuity or for creation of a home. However, we also observe another family circle rationale that owes less to an acceptance of the given model. In the present case, the situation is not so much chosen as imposed, either because the family’s burden is too heavy, or because economic conditions prevent individuals from getting free of the group.
88Murielle, a 49-year old civil servant in the national education system, is divorced with a son aged 19. Her relationships with her parents are dif?cult. According to the data of the Proches et parents survey, her family organization is of the local family circle type. She lives a stone’s throw from her parents; she sees them at least once a week, lists them as being close in the questionnaire and mentions several times the help she has received from them: ?nancial help to buy her home, daily help by giving her son lunch every day when he went to school, and moral and ?nancial support during two dif?cult episodes in her life, namely a car accident when she was 30 (her parents then took care of both mother and child) and her divorce.
89Several times during her lifetime, circumstances have brought her back to her parents. She left Nice and her parents in 1973 and moved to Nancy, where she was transferred in her capacity as a civil servant. There, she met her future husband and remained in the Lorraine region until 1981. After her divorce, she sent her son to live with her parents and joined them for one year. Then she left the south of France again to live with her companion in the Paris area where she stayed for four years. In 1987, she went back to live with her parents in the south. Three years later, she requested a transfer to Nancy, so that her son could be close to his father. After two years, however, the relationship between father and son deteriorated, and she decided to move back to Nice.
90Each of her moves back to Nice has followed a failure in her private life: divorce, separation, con?ict between her son and ex-husband. Even though her return to Nice, near her parents, was what she wanted, even though she let them choose a ?at for her and even though her father has provided ?nancial help for doing up her ?at, she ?nds this proximity with them dif?cult to bear: “Never live next to your parents! It’s a disaster!”
91In fact, although as a young woman, alone with a small child, she had valued living near her parents, as a mature adult she is very uncomfortable with the “special protection” she receives from her parents.
“Yes, they live 50 yards away… I thought it would be the same, but things have changed, they’ve grown older, and there’s something else I didn’t think of at the time, but now I know… it’s true they didn’t come to my place but I had lunch with them every day, so they didn’t need to see what I was doing at my place, but perhaps it’s because I’m divorced with a child that they want to overprotect me…”
93Despite these problems, Murielle is not planning another transfer, at least for the time being, since as an only child she feels responsible for her parents:
“At one time I had considered the possibility of going overseas, but I don’t know whether I can do it because there’s the problem of my parents, it’s going to be dif?cult for me with my parents”.
95Her attempt to become independent from her parents has failed, the “urge to drop the family” has not succeeded. Murielle has returned to the bosom of the family but this return is all the more painful for her since she feels sti?ed by intrusive parents. The analysis of the interviews reveals other situations of this type, where ownership of a shop or a family business “chains up” the next generation to the previous one , or when economic insecurity has made a person dependent on other family members. In these cases, instead of being a basis for personal development, local family circles represent an obstacle to autonomy.
2 – The local family circle: group or network ?
96At this stage of the analysis, one might be tempted to follow the suggestion of Jean-Hughes Déchaux and assimilate the categories of local and dispersed family circles to more anthropological notions, such as the kinship group which implies a strong dependence of the domestic groups within the extended family, or the kinship network, which is “a coalition of nuclear families in which intimacy is experienced at a distance, in order to preserve each other’s independence” (Déchaux, 2001) . Do local family circles, especially those of the semi-cohabiting type, form a group that is experienced as such, or do they function as a network ? It is possible to analyse present-day family relations by reference to the notions of houseful (maisonnée) and kin (parentèle), as Florence Weber (2002) has suggested in a recent publication. The kin is de?ned as a kinship network, a “world of reference and interaction governed by rules of behaviour that are based on exchange and reciprocity”, while the houseful is de?ned as a “production and consumption group”. Even if the houseful usually implies co-residence, it may comprise several households. Indeed, recent transformations in family lifestyles (with “dotted line de-cohabitation”, joint custody of children, couples who alternately live together and apart, etc.) have given, as Florence Weber notes, “a new usefulness to the old concept of houseful when analysing the practical aspects of kinship”. In this case, the houseful will be de?ned more by “membership of a group that pools resources on a daily basis”, than by sharing the same home.
3 – Living as a group, while respecting everyone’s independence ?
97Reading the interviews, the notion of group appears most clearly when the local family circle is the product of a family asset, of an industrial or farming property. Paul describes family reunions on Sundays, when the family would discuss the business:
“…every Sunday, we would have lunch with my grandfather and my parents, they would come to our place and the next Sunday we would go to theirs, and inevitably my father and grandfather would talk about the company.”
99One generation later, the weekend is still the time for conversations about the family ?rm.
100This feeling of belonging appears in the family members’ behaviour at the ?rm.
“In the P. company, we must set the example at all levels, that means even if you are a member of the P. family, you have to be there on time, or try to get there before the others, you mustn’t come in as if you’re just strolling through.”
102We observe here the two elements that make up the objective solidarity of the houseful: reputation and collective responsibility (Weber, 2002).
103The sharing of resources and the division of labour reinforce the sense of belonging to a group. This is true for Louise and Raymond, wine growers in the Beaujolais region. Like their parents and grandparents, like two of their three children, like most of their brothers and sisters and uncles and aunts, they are farmers: “we are all close to the earth,” says Louise. Practically all their relatives live within a radius of a few dozen kilometres and they both come from the same village. Their lifestyle perpetuates that of previous generations and corresponds perfectly to the local family circle type.
“My paternal grandparents used to live right near my parents, so we saw each other quite regularly. My maternal grandparents died quite early, but they didn’t live very far either from my parents’ home, so we often saw them too; in those days, it was like a party, to go and see our grandparents.”
105Because they all work in wine growing (except for the daughter who is “anti-vineyard”), the brothers, sisters, brothers-in-law, sisters-in-law, children, sons-in-law, nephews, grandchildren all see each other very often:
106“… we are not always in each other’s homes, but we do have lots of contact with the family.”
107Although the farms are independent, some tasks and tools are shared:
“Let’s say that we have our vines, he has his vines, our other younger son also has his vines, but we do the work all together, that means we have less equipment to buy since we can borrow each other’s tools, and anyway, we are bound together by the family life stuff, so we keep doing it this way.”
109In this context, exchanges, especially ?nancial exchanges to buy farm tools or making an advance on the year’s pro?ts, are part of everyday life, as is helping with childcare:
“… it’s no problem, there’s always someone who’s free, if not one then the other. It’s not a problem, there’s always someone around to care for the children.”
111In both cases, a group of related households are linked by common interests, intense exchange of services, and strong emotional ties. The resulting local family circles can be assimilated to housefuls to the extent that, as Florence Weber (2002) writes, “common responsibilities and reputation, division of domestic labour among members, and at least partial redistribution of resources are all elements that de?ne the houseful as an objective or mutually supportive group”.
112Denise’s story shows that in compensation-type local family circles, there is a determination to create a united group:
“Let’s say that my family life, of the family that I created, is very warm, we’re a tribe! A tribe. In the family, we often speak in terms of tribe. We’re very close to one another, and we need to be… that’s what I wanted.”
114Other respondents use the term “clan”. These expressions clearly re?ect the idea of an objectively mutually supportive group, that is the houseful.
115Nonetheless, the statements also reveal the tendency, described by François de Singly (1993), of individuals to seek greater autonomy from the family. A wish to maintain a distance between the generations is visible. For the parents, this means respect for their children’s independence, while for the children, it refers to their own claim to autonomy. For Louise and Raymond, this takes the form of being careful not to interfere in their children’s married life:
“… I don’t meddle with their things, everyone is free to… Sometimes they need me, sometimes they don’t… And the day I need them, it’s the same! We are independent and close at the same time, so we don’t get in each other’s way.”
117This need to be independent from the family is also observed for Paul and his wife, and in their case it is expressed by geographical distance. The family, the company, the references to Italy, all of that seems a little dif?cult for Paul’s wife to bear. During the early years of their marriage, they lived near Paul’s parents in housing provided by the family. The paternal grandmother looked after their daughter every day. However, despite these advantages, Paul and his wife felt the need for some distance. So they decided to move to Nogent-sur-Marne, which is in fact part of the family domain, a place of family reference that enables them to stay within the reach of the family group (some close cousins still live there) while escaping the mother’s control. This represents a good compromise between too great a remoteness from the group, and a degree of proximity that Paul’s wife ?nds suffocating.
“We got out, but we were smart about it, because we went to Nogent which is the town where she [Paul’s mother] spent her childhood. So we knew people she had known… she still had relatives in Nogent.”
119Some decisions are made at the household level, but others involve the family group. Nonetheless, it was not easy to ?nd the right distance — between dependence on the group, tied to the family company, and the desire for autonomy.
The family as a network, to which one “belongs” ?
120In the previous examples, we compared the notion of local family circle to that of the kinship group. The case of Valérie and Christine (daughter and mother) illustrates another type of system that is closer to the kinship network. Here the local family circle goes together with a high degree of independence of the household relative to the families of origin — an independence which is respected as Christine’s interview shows — and a high degree of autonomy within the couple, as explained by Valérie:
“I have a group of girlfriends, I see them without Patrick, we have some groups of friends in common, so we see them together as a family. Then I have girlfriends, as I say, that I see on my own, and Patrick has friends that he sees on his own.”
122Nonetheless, this large degree of independence does not mean that there is no family group. It seems that the family comes together more around the holiday or weekend home. The interviews make very clear the importance of places in family systems and their role in structuring the kinship network:
“… My father’s mother was from Annecy, she had bought for us and for her ?ve children a family home where, as used to be traditional in France, we spent our holidays with our seventeen cousins, there was no question of going anywhere else. Our family’s entire history is bound up with the history of this house.” (Christine)
124For Christine, a family is necessarily a large family gravitating around one or several houses. Having been forced to sell the family home, she reproduced the model established by her grandmother and bought a holiday home in the Annecy region, not far from her sister. The purchase of the house was itself the result of a family decision, as Valérie explains:
“In the summer, we went looking at houses and I remember we would all go together, we’d take two cars, we’d all go, brothers, sisters, children… we wanted to choose all together, as a family project.”
126For all of them, the important thing is to have a place to get together, in order to ensure group cohesion and transmit the family spirit.
“If you like, we repeated what had been done at Annecy, we did it again on the other side of the valley, at V. So it goes on. The idea is that it’s a house, the key is in a box near the entrance, the children can go, our children’s friends can go. My brothers and sisters can go. The aim is to live again like we used to, in a way.” (Christine)
128If, as Bourdieu (2000) writes, “the house cannot be separated from the houseful as an enduring social group, and from the collective project to perpetuate it”, we can infer that in one sense there is a form of houseful that does not rely on the spatial proximity of homes but on the strength of ties and the sharing of a holiday home. “Those who feel at home in a given dwelling are part of the houseful” (Weber, 2002). The case of Christine and Valérie is not unique. Numerous interviews show the role played by holiday homes in preserving and binding together the family group, and this at all social levels. The houseful may also comprise several holiday homes. For example, Lucie bought two small houses in the Corrèze region because she spends the holidays there with her sister. Residential proximity enables both sisters to spend time together quietly in a way that was denied them during childhood:
“So we bought places in Corrèze, the children are happy to come, we like it there, my sister’s not far, so we go mushroom-picking, blackberry-picking, raspberry-picking, or we act the grandmas, I knit, she doesn’t want to knit but I do and I like it.”
130The children have their own “little house all ?xed up”, making it possible for Lucie’s ideal lifestyle to become a reality, with “a home for everyone.”
131The question of whether a local family circle can be assimilated to the anthropological notions of kinship group or kinship network is not a simple one, since the borderline between the two is not static but depends on the individuals, on the way they experience these family relationships, as a form of solidarity or of mutual exchange, as Florence Weber puts it. Here we have probably reached the limits of the exercise, in particular because our analysis is based on the words of just one respondent. Denise’s case is only one example. She has the impression of having created a clan, a tribe, and that is how she experiences it; but what of her children ? According to the research of Vern Bengtson and his team, parents always appear to be more committed to family relations than their children. The former are anxious to preserve certain values and tend to overestimate intergenerational solidarity and minimize con?icts, whereas the latter, who are more concerned with autonomy, tend on the contrary to play down the importance of solidarity and overemphasize disagreements between parents and children (Bengtson and Giarrusso, 1995). To understand the reality underlying the notion of local family circle, it would be necessary to pursue the analysis by interviewing different members of a single family and monitoring the family over time.
132The results of the Proches et parents survey con?rm the persistence of kinship relations shown by many studies on intergenerational solidarity, both in France (Attias-Donfut, 1995; Crenner, 1998; Segalen, 1991; Pitrou, 1978) and abroad (Coenen-Huther et al., 1994), Bengtson and Roberts, 1991; Bengtson, 2001; Hillcoat-Nalletamby et al., 2002). Contrary to the theses of T. Parsons and certain lines of research highlighting the decline of the family, households are not isolated in contemporary urban society. Two thirds of the respondents have at least one weekly contact with a member of their family, half live in the same commune as a parent and three-quarters report being close to their mother (Bonvalet and Maison, 1999). Help and services are commonly exchanged in a vast majority of families (Ortalda, 2001).
133Despite the rise of individualism, the extended family continues to exist in urban society, but the links between parents and adult children take many different forms. Some have distanced themselves, others on the contrary maintain close ties with their family. Just as there is no single type of household, there is no single type of extended family. Statistical analysis has made it possible to distinguish several modes of operations of kinship. The family circle, de?ned on the basis of indicators pertaining to emotional ties, frequency of contact and mutual help, corresponds to a reality which cannot be described as domestic withdrawal: 30% of the respondents belong to a local family circle, that is, they live in the same commune as a relative listed as close, are in contact with this person at least once a week and resort to mutual help and services; 17% belong to dispersed family circles, maintaining strong ties, though not living close by. Overall, nearly half of the respondents live in a family circle mode.
134As could be expected, women organize their family life according to this model more often than men. “They clearly appear as the main builders of family ties and exchange” (Hammer et al., 2001). Even if there are differences between social classes (managers seem to belong to a local family circle somewhat less often than blue collar workers), the residential and family histories are determining factors. To summarize, the local family circle is to a greater extent the result of a family experience covering several generations and lineages than the result of demographic or social determinism.
135The interviews enabled us to re?ne this thesis by trying to understand the process underlying the organization of family life, and showing that there are different ways of reaching a given result. Several modalities are encountered simultaneously. Some correspond to a logic of “home creation”, either through the reproduction of a family model, through adoption of one’s in-laws, or, if there has been a break with both families of origin, through recreation of this type of functioning with the adult children. Others, on the contrary, do not appear to be the result of a genuine choice. In this case, the local family circle seems rather to be the result of economic constraints or of the ownership of capital assets. In fact, the local family circle appears to be both a family lifestyle that respects the independence of each individual and each couple (it might even be said that some degree of individual independence is possible precisely thanks to the existence of the family circle  and an adaptation of the complex family to urban society (some families have chosen a way of life centred on the interests of the kinship group, others did not really choose this way of life and are subjected to its constraints).
136These analyses re?ect the situation at the time of the survey and of the interview. In the course of the life cycle, an individual can at certain times live in a local family circle and at other times, depending on family or job, in a dispersed family circle. For example, this organization can at ?rst be set up with one’s parents, by moving nearby at the time of leaving home; later with one’s children if the latter choose to remain in the same commune, or with one’s parents if there is a decision to move closer in space. The analysis of the interviews of respondents living in dispersed family circles shows a clear separation between those who lay a claim, as it were, to this way of life, inasmuch as distance does not modify the intensity of the relationship, and those for whom this distance is temporary, who wish to come back and live near their parents or children and set up a local family circle.
137Can the notions of local family circle and dispersed family circle be assimilated to the more anthropological notions of kinship group or kinship network, as suggested by Jean-Hughes Déchaux ? According to the interviews, it seems that the opposition is less clear. Certain local family circles, in particular those built around industrial or farm property, function as a houseful. For others, it is sharing the holiday home of one of its members which suggests that dispersed family circles can also be assimilated to housefuls. In this respect, the limits of our analysis have been reached. Extended families come in a great variety of forms, and these change over time. Family circles bear witness to a real phenomenon, at least quantitatively: the family remains present in the lives of individuals. But what does this presence mean ? That individuals adhere to a mutually supportive group, as Vern Bengtson and his colleagues suggest, or simply that they belong to a network of exchange of goods and services, as described by F. de Singly ? How can we reconcile the need for autonomy with this “family need” which comes through so strongly ? This is one of the contradictions that anthropologists, sociologists, psychologists and demographers will have to explore in the coming years.
AcknowledgmentsI thank Dominique Maison for his collaboration in the Proches et parents survey and Arnaud Bringé for his help with computing.
Institut National d’Études Démographiques, Paris.
Translated by Zoé Andreyev.
The French word proche refers to both immediate relatives and very close friends. Thus, the title of the survey might be rendered as “Next of kin, close friends, and relatives”. We preserved the original title when referring to the survey in the text. (Translator’s note)
Either the traditional family has disappeared and been replaced by the “conjugal” family and by other forms such as one-parent or blended families, or it has survived through family relationships expressed through help, donations, and the frequency of visits (de Singly, 1993).
See chapter 2, “En construisant une anthropologie urbaine, P.-H. Chombart de Lauwe bute contre l’État” [“Constructing urban anthropology, P.-H. Chombart de Lauwe comes up against the state”], in Amiot (1986).
Michel Amiot describes this discovery in his book, op.cit.
It must be noted, however, that in Great Britain, a household implies not only that its members share the same dwelling, but also share food expenses and the task of preparing meals. A “household” is de?ned as: “a group of people who live in the same accommodation and share at least some of the catering”.
Drawing our inspiration from Jacques Dupâquier’s historical survey, we selected persons whose last name began with the letters TRA. With the help of the Minitel [the French electronic phone book], we selected a representative sample of the adult population through quota sampling. For this reason, the persons surveyed are either persons of reference or members of the household (see Bonvalet et al., 1993 for a description of the survey).
Contacts include meetings, phone conversations and mail.
The ?rst part of the questionnaire included questions concerning help in the matter of educational or occupational guidance, the search for employment or housing on the one hand, and concerning help provided during dif?cult times or on a regular basis on the other hand. In the part of the questionnaire devoted to close family and friends, one question for each of them asked whether they had helped or been helped by ego and thus identi?ed the persons belonging to the mutual help network.
For instance, a woman may have strong ties with her mother while her spouse is not as close to his mother-in-law.
Uncles, aunts, cousins, nephews and nieces represent only 2.4% of close family members, whereas parents and in-laws represent 22%, children and grandchildren 32% and brothers, sisters, brothers-in-law and sisters-in-law 35%.
In her thesis, Céline Clément (2002) observes that families do not appear to grow stronger around the widowed mother, but around the family they have formed themselves.
The category “cadres” which in the French classi?cation includes not only managers and executives, but also senior civil servants and members of the liberal and higher intellectual professions.
We took into account the profession of the person of reference of the household and not that of the respondent.
In his work Le monde privé des ouvriers, Olivier Schwartz (1990) describes the strong relationship that exists between mother and daughter: “Many women who say that they cannot live without their mother are obviously not referring to material needs. Their need for their mother runs much deeper: they form a couple with their mother: they ‘must’ see her or phone her. Even if no material need is involved, the bond cannot be broken”.
The semi-directive interviews were carried out in the framework of the SRAI (Statuts résidentiels: Approche intergénérationnelle) workshop which includes eight researchers: I. Bertaux-Wiame (CNRS), C. Bonvalet (INED), P. Cuturello (CNRS), A. Gotman (CNRS), Y. Grafmeyer (Université-Lyon II), D. Maison (Université-Paris X/INEDned), L. Ortalda (Ined) and P.-A. Rosental (EHESS).
According to the 99 interviews of the Proches et parents survey, 32 persons lived in families functioning as local family circles — a proportion of 31%. This is practically the same percentage as obtained in the statistical analysis. To this must be added three interviews that were conducted with respondents’ spouses and two with respondents’ sisters. Overall, we have 37 life histories.
M. Coppel and A.C. Dumaret (1995) have shown two opposite tendencies in the choice of a spouse by persons who spent their childhood in institutions: “the ?rst tendency is to choose someone who will help them create the family they were deprived of as children”, the second tendency leads them on the contrary to choose a “mirror” partner, someone who suffered as they did during childhood. (In Que sont-ils devenus ? Analyse d’un placement familial spécialisé, Toulouse, éditions Erès, 1995).
Or else the parents were already dead at the time ego’s family was formed.
In fact, this type represents the “counterpart” of some of the previous cases. Whereas in the examples studied above, ego could ignore the spouse’s family or compensate for its absence through contacts with his or her own relatives, in this case the respondents, having rejected their own family, establish close links with their in-laws, who become their children’s sole grandparents.
See the case analysed by both Isabelle Bertaux-Wiame (1999) and Anne Gotman (1999) — Mr. H, a cafe owner who is unable leave his village — concerning the decision to stay.
See the review by Jean-Hughes Déchaux of the La famille et ses proches: l’aménagement des territoires in the Revue française de sociologie, 2001, vol.42-1.
See the work by Claudine Attias-Donfut, Nicole Lapierre and Martine Segalen, Le nouvel esprit de famille (2002). For them, “kinship and individualism are not incompatible, but complementary, maybe even compensatory”.