Copyright Sociological Research Online, 2003

Julia Brannen (2003) 'Towards a Typology of Intergenerational Relations: Continuities and Change in Families'
Sociological Research Online, vol. 8, no. 2, <http://www.socresonline.org.uk/8/2/brannen.html>

Abstract

This paper focuses on 'beanpole' families, that is those with several living generations; it analyses patterns of care and paid work across the generations and the resource transfers which take place between generations. Drawing on a small-scale study of four generation families, it provides a typology of intergenerational relations with respect to the transmission of material assets, childcare and elder care, sociability, emotional support, and values. It examines two a fortiori conditions which are considered to shape intergenerational relations: (a) occupational status continuity/ mobility and (b) geographical proximity/ mobility. Four types of intergenerational relations are generated by this examination: traditional solidaristic; differentiated; incorporation of difference; and reparation in estrangement. The paper looks at families holistically and draws on the concept of ambivalence to describe the forces which push family members to carry on family patterns and those which pull them apart and lead them to strike out on their own. It shows how, whatever the type of intergenerational pattern, each generational unit seeks to make its own mark.

Keywords:

Family Continuity And Change; Family Resource; Intergenerational Relations

Introduction

1.1

The paper and the study ^[1] upon which it draws set out to explore change and continuity between generations within families, in relation to paid work and care and the ways in which employment, breadwinning, motherhood, fatherhood, grand parenthood and great grandparenthood are both enacted and constructed intergenerationally. It sheds light on broader sociological and demographic questions concerning family change and increased longevity, in particular whether the prevalence of multi-generational bonds^[2] represents a valuable new resource for families in the 21st century (Bengston 2001). The study focused upon four generation families in which young children constituted the fourth generation. Such families, while they may be less common and less long lived in middle than working class families where age of mothers at first birth is higher and still rising, constitute a significant proportion of the population^[3]. Moreover, such families while providing considerable resources are also likely to be make heavy demands upon resources notably with respect to childcare and elder care; these may fall disproportionately on the grandparent or 'pivot generation. In this paper I examine these families in relation to a wide range of social processes concerning the transmission of resources of different kinds and the ambivalences within families which constrain transmission. I look at both reproduction and innovation across family generations.

1.2

The broad theoretical approaches adopted in the study were as follows. First, the study took a generational perspective - with a focus on family generations and historical generations. According to Mannheim (1952) generation units are created, especially in the so-called 'formative part' of our lives. The process of becoming a generation unit is two fold: (a) through sharing a similar social location notably relating to social class and (b) through the process of collective exposure to the same historical set of cultural and political events and experiences. Adults in the three generations interviewed were born in three historical periods: the great grandparents born 1911-1921; grandparents born 1940-1948; parents born 1965-1975. These generations grew up in very particular times: the great grandparents experienced the 1930s Depression.

1.3

The grandparents were children of postwar reconstruction and the welfare state while the parent generation experienced the neoliberal economic policies of the 'Thatcher' period.

1.4

Second, the study adopted a realist perspective; in an ontological sense that we assumed that reality exists and had an interest not only in the accounts family members gave but also of events in their lives and the patterning of the life course. As well as suggesting what the past has to offer the future, we also had an interest in the past for its own sake; so often it seems that today's sociologists 'prefer an image of adults creating their futures' (Rossi and Rossi, 1990:20). Thus we wished to limit time 'discussing the telescope' through which the past is often viewed. (Abbott 2001: 223)^[4]

1.5

Third, the study took a gender perspective on employment, family relations and care since we expected to find clear differences between the ways in which men and women conducted their lives in these areas over the generations while also expecting these patterns to be shaped differently in each historical generation. Finally, we adopted an interpretative perspective. Thus, despite a realist ontology, we were also interested from an epistemological point of view in issues of interpretation from the actor's perspective. Indeed, we were interested in the ways in which people's accounts of the past were mediated by time, meaning and audience (research context). But we were also interested in the 'facts' of their lives as we deduced them from their accounts and those of other members of their families.

Study Design

2.1

The study was small scale and intensive and was carried out with my three colleagues (see note 1). It adopted biographical methods^[5] in order to generate full life histories of work and care, on the one hand, and accounts of people's lives in the format in which they chose to narrate them, on the other. This then was a set of cases, the 'case' being the kinship group consisting of four generations (of which the adult members of three generations were interviewed). We considered this an appropriate research design to examine how employment and care play out over the life course and across the generations. We make no claim to the representativeness of our sample and cannot generalize from it. Our aim was rather to provide a thorough description of the families, to identify patterns, and to generate theoretical interpretations. Twelve case study kin groups were theoretically sampled where the youngest generation had at least one young child (originally intended to be under 5 but revised to under 10 years) and the other generations were, from the vantage point of the youngest: the parent, grandparent and great grand-parent generations. The kin groups were also sampled to 'represent' a diversity of different employment and occupational statuses at different ends of the socio-economic spectrum, at least in the grandparent generation. In recruiting the families, we approached the grandparent generation first. We sought to divide the sample according to those grandparents employed, either currently or in the past, in a professional or managerial occupation, and to include an equal proportion who worked, or had worked, in lower status or manual (skilled and unskilled) occupations. We also sought to ensure that the grandparent generation would be split between those in employment and those outside the labour market, so that we might vary their respective roles in relation to employment and care. A final criterion related to marital status. We decided to rule out further complexity by including only those grand parents who were still in the same relationship as when they were bringing up their own children and, similarly, to include only those in the current parent generation who were living with the parent of their children. The study has therefore little to say about the impact of divorce on intergenerational relations.

2.2

We screened and recruited families through a variety of means: by surveying current and recent employees of public sector organizations in London; advertising in several cities outside London including in local newspapers; and using our own social networks. The sample is based in England since we had limited resources for travel but includes persons living in the South West, the North, East Anglia, as well as the South East and London. Despite great efforts to make contact with minority ethnic families, there was none which matched our criteria in terms of numbers of generations and sufficient proximity. We had to exclude both majority and minority ethnic grandparents on account of divorce. Employment criteria did not always match the neat research design. Our completed sample included: five families where both grandparents were employed full-time at the time of interview; four families in which the grandfather still worked full-time and the grandmother was not employed; two families in which couples were retired or both working part-time; one in which the grandfather worked part-time and the grandmother was retired. In terms of occupation, we ended up with five couples where the grandfathers and grandmothers had held, or were still in, professional or managerial jobs; two grandfathers in lower level nonmanual occupations; two in skilled manual and three in low skilled jobs.

2.3

We interviewed between eight and five members in the 12 families, each member separately: one set of grandparents; one child of the grandparents (with at least one child under 10 years) and his or her partner; both sets of the grandparents' parents, where available (i.e. both the grandfather's and grandmother's parents). This meant that we did not try to interview the whole family. Nor did we interview the fourth (i.e. the youngest) generation, the young children. In the event, we interviewed 71 family members: all the parents and grandparents, but only 23 great-grandparents, of whom one was a step-great grandfather. Of the 25 'missing' great-grandparents, 21 were dead (7 great grandmothers and 14 great grandfathers) and 4 refused. The refusals included one couple where the husband was very ill, and two great grandfathers whose wives agreed to be interviewed.

Reproduction and Innovation in Intergenerational Relations

3.1

There is a creative tension between change and continuity, between processes of reproduction and innovation. In intergenerational families, values and practices are transmitted, while each generation may also develop or subscribe to its own. Parenting is passed on while new practices are adopted in different generations. For example, parents' educational aspirations for their children and the availability of educational opportunities may work against occupational transmission. As family members' life chances change across the generations, some generations may lack the resources to repay their parents or to provide similar resources to the next generation down, for example employed grandmothers may not be able to provide care for their grandchildren in the way their mothers had provided care for grandmothers' own children. Bertaux and Bertaux- Wiame describe such tension between continuity and change in the following way and suggest how one resource may be transformed into another:

'Because transmission of sameness reifies the heir (treats him as an object), it seems to carry the kiss of death. To become the subject of the heritage, the heir must act upon it by leaving his or her mark upon it. ....The new element involving both the rejection of the past and innovation, enables the heir to take possession of something that actually was passed on to him. The point is not simply that he must 'make something of what was made of him' as Sartre put it so aptly, but that he make something of what has been passed on to him' (Bertaux and Bertaux-Wiame 1997, 93)

3.2

The tension between change and continuity generates ambivalence. Luscher (2000) identifies ambivalences in which different generations are caught up in a tension: between on the one hand, the reproduction of some aspects of their 'family systems' , and innovation of other aspects, on the other hand. Ambivalence has to be managed; it is not resolvable. As Luscher (2000) and Luscher and Pillemer (1998) suggest, ambivalences may be expressed structurally e.g. via a change in occupational status across family generations. They may be reflected in strategies as for example when a family seeks to put geographical distance between different family generations or chooses to remain geographically close at hand. Ambivalence is expressed through feelings and in the social interaction and interpersonal relations; it may be expressed in values.

3.3

These different aspects may not however work in tandem. Structural aspects of people's lives may pull in one direction, for example towards the reproduction of aspects of family systems while, at a strategic level or, in Bourdieu's terms, their habitus (Bourdieu 1986) creates a divergent lifestyle. Thus, over the generations, some families may reproduce the life chances of the older generation, as when wealth and educational capital are transmitted. However, younger generations may, at the same time, also seek divergence from older generations despite the transmission of assets and wealth which cushion their life chances. They may differentiate themselves, for example with respect to values and life styles.

3.4

It is important in this discussion of ambivalence not to counterpose structural factors against the agency of actors. The transmission of resources of different kinds is likely to involve processes in which much of what passes on, or is passed on, is taken for granted; cultural transmission of class and family cultures can be implicit as well as explicit (Bernstein 1996). As Bourdieu (1986) suggests in his elaboration of the concept of habitus, the dispositions of individuals and groups are cumulative and not necessarily intentional or strategic.

The Paper's Focus and Questions

4.1

This paper develops and discusses some analysis of these four generation families and suggests a typology of intergenerational relations. It starts out by examining a set of a fortiori conditions which, on the basis of theory and empirical evidence, are likely to shape intergenerational family relations. They are: (a) occupational status continuity/ mobility and (b) geographical proximity/ mobility. In the former case, there is a wealth of evidence that occupational status shapes family resources and life chances; it therefore seems likely to affect the capacity of different generations to provide support, especially material support. Inequalities in resources across family generations are likely also to make reciprocity difficult for both status and resource reasons. Changes in occupational status may also manifest themselves in changes in values and lifestyles. With respect to geographical mobility, which may be associated with occupational mobility, this is likely to make the provision of certain kinds of services between family generations difficult to accomplish at least on a frequent basis and may also affect sociability patterns among kin.

4.2

In the second part of the paper, I plot the twelve case families on a matrix in which occupational status mobility is one parameter and geographical mobility is the other. I then explore the cases in each of the quadrants to see how far each set of cases and how far each quadrant are suggestive of a particular type of intergenerational family relations. In analyzing the life histories and life stories of these families, I examine these accounts with respect to different aspects of intergenerational relations and intergenerational support (Bengston 2001): financial transfers and other material assets, the supply of services including care, the extent of sociability, emotional support, the congruence of values and beliefs, and a commitment to specific norm of family obligations. It is important to recognize that these different forms of 'capital' are often inextricably intertwined, for example the economic with the cultural and symbolic (Bourdieu 1986; see also Allatt 1993), and, as also noted above, that one form of capital may be transformed into another. Thus, while occupational and geographical mobility may shape the conditions for the transmission of capital, they may also shape their transformation into other forms of capital. So, in this paper, I am seeking to draw out wider patterns of intergenerational relations which may be structured by these dimensions and to ask the following questions: How far do multi-generation families pool resources across generations to support one another? How far does each generational unit differentiate itself from other generational units by being self-reliant or reliant on formal services? I also asked further questions framed in terms of ambivalences between generational units in relation to structural factors, social interaction, feelings and values. Thus it is possible to look at these families holistically in relation to the forces which push them to carry on family patterns and those which pull them apart and lead them to strike out on their own.

Occupational Mobility

5.1

Over the twentieth century and up until the late 1970s^[6], there has been an absolute increase in social mobility for both men and women (actual increases in numbers of children in higher status occupations compared with their parents); more space at the top was created for men and more in the middle for women (Goldthorpe and Mills 2000; Aldridge 2001). This is largely due to the expansion in employment opportunities for the middle classes which have substantially grown in size relative to the working class (Heath and Payne 2000). In the analysis of the case study families I concentrate on men's jobs since women in the two older generations had intermittent employment careers following motherhood and typically worked part-time. Moreover, women married to middle class husbands rose later in their life course, that is intra-generationally; they thereby matched the occupational rise which their husbands achieved inter-generationally. By contrast, some wives of manual workers rose into nonmanual clerical employment over the course of their working lives while their husbands remained in manual work.

5.2

In five of the 12 case families no occupational status change took place between fathers and sons (or sons in law); this fits with national data for cohorts of men born in 1910-19 and in 1940-48 (Table 1). In four families the men were upwardly mobile occupationally over the generations while in two families they were downwardly mobile. In one family there was both upward and downward occupational status among the three generations of men. As nationally, intergenerational male mobility is rarely long range: most involved moves in an older generation between skilled manual or clerical work into professional/managerial occupations in a younger generation. No unskilled worker had a son or son in law in a higher status job.

Table 1: Men's Occupational Mobility over three generations (12 families)
Continuity of occupational status: low skilled (3)	2 families of unskilled/ semi-skilled men 1 family of self- employed family builders
Continuity of occupational status: high skilled (2)	1 family of senior managers/ managing directors 1 family of electrical engineers (middle management level)
Downward mobility in occupational status from skilled to unskilled work (2)	1 family with a great grandfather in skilled work and a son and grandson in semi-skilled work 1 family with a great grandfather and grandfather in skilled work and a father (i.e son in law) in unskilled work
Upward occupational mobility from skilled to professional/ managerial work (4)	2 families of great grandfathers and grandfathers in skilled work and a father in a managerial occupation; 2 families of great grandfathers in manual/ clerical work and grandfathers in professional occupations
Upward and downward occupational status (1)	1 family of a great grandfather in skilled manual work/shopkeeper, a grandfather a professional worker, a father (a son in law) in unskilled work

Geographical Mobility

6.1

Historically there has been much movement of the population in Britain. However currently four in 10 people nationally stay in the same local authority where they were born and only one in 100 households moving any great distance (PIU 2002). According to Grundy et al (1999), half of people who have a father or mother or eldest child alive see them at least once a week and half of these live within a half hour's journey time of them. Half of the case study families lived close to one another in the same town or same part of a large city (London), while half were living at a significant distance from the other one or two generational units. Moreover, all but two generational units were living in the same parts of the country as their kin in the study; the exception was one upwardly mobile family in which the grandfather moved from Northern England to the South East and a working class family where the grandparents moved from London to the South West of England. On the other hand, reflecting the historical movement of the population, members of the great grandparent generation in the 'stayer families' were often incomers to an area.

6.2

In this study, as is the case nationally, most moves had rather less to with jobs and rather more to do with significant life events or with improving the quality of life or housing. But graduates were more likely to move than nongraduates, as is the case nationally. Some men in the past travelled to work rather than relocate and some had relocated temporarily (two grandfathers in skilled work), again a common trend in the population. Working class families, particularly among the middle and oldest generations, provided housing for their children when they started married life (see for example Rosser and Harris 1965) thereby inhibiting movement (5 couples in the great grandparent generation, 3 in the grandparent generation, and 1 in the parent generation). Yet it is important to note that, despite the movement of half the families away from close kin, they often maintained 'close ties at a distance' (Mason 1999) and are increasingly able to do so with because of technological advance (Wellman and Berkowitz 1988). On the other hand, it is significant that the two families who moved farthest away were those whose ties were affectively distant (i.e. not close) and where the 'movers' sought to put distance between them and their kin.

Processes of Reproduction and Innovation in Families

7.1

The following data analysis involved plotting the two dimensions of occupational mobility and geographical mobility against one another and locating the case study families accordingly in the four quadrants of a matrix (Figure 1). Intergenerational relations are next explored on a number of dimensions of intergenerational support, as discussed earlier, in respect of each of the cases; and similarities and differences among the cases were examined within and across quadrants. Four types of intergenerational support are identified:

Quadrant 1: Solidaristic Relations

8.1

Families in the first quadrant (1) span the socio-economic spectrum. Each younger male generational unit reproduced the occupations/ occupational status of the older male generations, while geographically the different families remained close to each other^[7]. Intergenerational relations in these families are solidaristic (Luscher 2000) in a traditional sense; there are of course many different kinds of solidarity in families (see Crow 2002). In these families, habits and general dispositions lead them to provide functional support of different kinds to one another - jobs, housing, childcare and elder care. In all four families, the grandmother generation is pivotal in the provision of informal care services which involves specific reciprocities^[8] (Finch 1989), as in the case of the Brand family which most closely approximates to this ideal type. Two generations of mothers in the Brand family helped with their grandchildren when they were growing up and two generations of daughters reciprocated their mothers' care by helping them in old age (see also Brannen et al 2003). Three generations of fathers and sons in this family built up a family building business. Occupations, jobs, business capital and private money were passed from father to son but they also helped one another to build and do up their own houses. Sons help fathers as well as fathers sons. The family business, together with a large amount of informal help and care given by and received from female members of the family, provide for a great deal of family mutuality intergenerationally which in turn serves to reproduce similar lifestyles and life chances across the generations. The family is bound by a strong sense of family obligations which are sharply gendered. The family also engages in similar leisure pursuits across the generations with a strong attachment to 'water'. They are strongly attached to the place where they live. 'Place' figures a great deal in the accounts of the women in the family. While being 'close' means ties of love and care, it also means 'belonging'. The women who marry into the family (i.e. those who took part in the study) admire the entrepreneurial family ethos of the men while retaining their gendered responsibility for home and children. While the grandmother and the mother developed employment careers in contrast to the great grandmothers, these women worked part time while their children were growing up; in fact the grandmother gave up work before retirement age to care for her grandchildren and her parents.

8.2

Yet this type of solidaristic relations is accompanied by ambivalence expressed in differences in values between the generations; there are forces of transformation within the family as well as pressures towards continuity. The father (youngest generation) broke the educational mould by gaining a string of GCSEs. From the current vantage point of history and his status as a father, he feels he was held back by his father from being educationally aspirant. He now sees qualifications as being crucial to 'getting on' in life. Jane, his wife, also wants more for her own children than she had in her own childhood. Like her husband, she wants her children to get on well at school and she wants them to escape from some aspects of the restrictive upbringing provided by her own (working class) parents. This ambivalence did not prevent from choosing her own parents to look after her child while she continued to work part-time following the birth. Jane is torn between identification with the values of her own working class family and those of her husband's family, seeking at one level to emulate the entrepreneurial and risk-taking spirit of the latter, while feeling rooted in the security and For herself, she also harbours educational aspirations 'something to fall back on when the children are older... I'd like to look back and say I've achieved this'. protection of her own family of origin. These tensions are currently contained since Jane will give up work with her second child while still hoping at some later date to pursue further education. For the time being the desire to 'keep close' to both families - geographically, emotionally and in terms of sociability and the provision of support - is over riding. Thus, although such a family pattern, as Harris suggests, may constitute an 'idealization of their mutuality' (Harris 1980:3999 quoted in Crow 2001), a desire in the new generation to put its own stamp upon the resources which are passed on to them is evident (Bertaux and Bertaux-Wiame, 1997).

Quadrant 2: Processes of Incorporation

8.3

In the second quadrant, with two cases, the generations stay close geographically but there is discontinuity in the men's occupational statuses. In one family the men in two descendant generations are in lower skilled employment while the oldest generation (and his father before him) were skilled workers. However this fact makes little impact on family intergenerational relations since the downwardly mobile male members are readily incorporated. For this reason, this pattern is termed the 'incorporation of difference'. The fact that the father who marries into this family is a labourer is considered unimportant by the two older generations of skilled workers^[9]. Indeed, all the other family members interviewed commented emphatically on the father's propensity for 'hard work', a characteristic which his father in law (the grandfather) suggests may lead to promotion: 'an absolute grafter from that point of view, you couldn't wish for her to have a better partner... the last six months, he's been working seven days a week... if he lost his job ... he wouldn't just sit there and think oh, that's it, I'm not going to bother. He would find a job doing something, I know he would. But it looks like he's going to be made up from, sort of the shop floor to a foreman's job, so' .

8.4

In many respects the two families in this quadrant resemble the families in the solidaristic quadrant; the different generations provide considerable functional support (help with money, housing, employment and so on). The grandparent generation in the last family discussed provides help to the younger generation just as its own parents helped them^[10]. In this family, the inclusiveness of kinship ties is also sustained by a strong commitment to place. However their family practices are less gendered compared with the Brand family, with the women having an equally strong commitment to work as to family commitments. The generational units are also emotionally close and they see a lot of one another socially.

8.5

Compared with the Brand family, they expressed less ambivalence. One reason is that the youngest generation was about to become parents (the baby's birth was imminent). Depending upon whether the mother in this family stays in work after the birth, ambivalences may emerge between the strong commitment of the father to paid work and the mother to both care and paid work. Such tensions may play out intra-generationally rather than inter-generationally. (The grandmother in this family resolved the tension by giving up work when her children were small and after she resumed employment she worked very hard at both work and care.) In the other family in this quadrant, ambivalences emerged between the values of one set of great grandparents and the two younger generations. This reflected the difference in the men's occupational statuses; the great grandfather, a highly skilled worker was disappointed in his son's lack of application at school and his failure to learn a skilled trade. ^[11] In this family, an innovative pattern emerges in household employment patterns among the parent generation who shares breadwinning (both work full- time) and also the parenting of their two young children. The father changed his job and his new hours to enabled him to participate more fully. This pattern was accepted by the other generations and some family help was forthcoming (full-time childcare from the mother's sister).

Quadrant 3: Processes of Differentiation

8.6

The third quadrant contains five families in which there was significant upward mobility among the men: from skilled manual or clerical work to professional/ managerial status. This shift occurred in all but one case in the grandparent generation; the grandfathers took advantage of the increased educational opportunities of the post war period. The main occupations held by the wives also represent upward occupational mobility compared with their own mothers. Moreover, the rise in occupational status of the grandfather generation was followed in three cases by a rise in their wives' occupational status post motherhood. In this group of families, at least one generational unit moved away - three far from their kin; in one case the move away was born out of a clear desire to be distant from them, reflecting differences in values but also a rejection of family 'closeness' and obligations. This pattern I have termed differentiation^[12], since family ties are maintained and are often affectively close. Their relations are however more specialised compared with those in the solidaristic quadrant since there is less functional support provided, especially care, less opportunity for regular socialising and in some families differences in values. There is also less balanced reciprocity (for further examples and discussion of this see Brannen et al 2003). There is also weak commitment in these families to the idea that families should provide the main support for grandchildren's care or for elder care. Instead there is a strong commitment to formal sources of support.

8.7

To take one family as an example, there are several forces for change in the Hurd family. The grandfather, C, (and his wife) experienced a significant rise in occupational status which lifted him out of the skilled working class into the professional class; he grew up in Northern England, took advantage of the increased educational opportunities after the second world war, passed for the grammar school and went away to university in the South of England. Later, as well the incentive to move South because of greater job opportunities there, there was the fact that his wife, whom he met at university, comes from the South. However, upward mobility and the move away were facilitated rather than constrained by his own family. C expressed gratitude to his parents for the 'freedom ... they gave me just to do it...There was no pressure to follow a particular career, to go to a particular school, to go out with a particular girl, to read particular books.' C's affirmation of self-reliance was not however a rejection of his kin but was accompanied by strong support for and from formal institutions - in particular school, church and state which provided him with considerable support over his life. This also made him unwilling to 'interfere' in his children's lives, for example when his youngest daughter (the parent generation) became a mother. In general he said : 'We encouraged (our own children) to make up their own minds and decide for themselves...By all means discuss it with parents but really it had to be their decision.

8.8

This intergenerationally transmitted freedom emerges in the parent generation. His daughter married an unskilled worker and when she became a mother she took on the main breadwinner role, thereby reversing the pattern of work and care. Returning to higher education following motherhood she took a full-time teaching job while her partner, who has few work ambitions, stayed at home to look after the children. C and his wife provided some financial help during their daughter's return to education but little childcare. They visit but don't stay long. It is interesting that their daughter is somewhat ambivalent about her parents' 'standoffishness' , but concludes that 'in some ways it's a good thing because... I couldn't bear to have the kind of parents who... stuck their oar in'.

Quadrant 4: Reparation in Estrangement

9.1

The sole case in the fourth quadrant, the Prentice family, also represents change as well as continuity in inter generational relations. The occupational statuses of both men and women in this family were almost all unskilled^[13]. Remarkably, the grandparent and parent generations together made the break from the inner city to the countryside^[14], with some surprisingly positive consequences. Yet the move away made little difference to the economic situation of these two generations since they exchanged urban poverty for rural poverty, and were no longer able to draw upon family resources. In many respects until the move away, intergenerational relations in this family closely resembled those of a working class family in the solidaristic quadrant. On the other hand, it is clear that family ties across the generations in this family have been long marked by strong ambivalences. These include resentments about the lack of emotional and material support at critical moments in the life course. It is significant that many of these tensions were between mothers and adult daughters, a tension which is repeated over four generations in this five generation family. In close-knit families grandmothers are a key source of intergenerational support for children as well as for daughters (Brannen forthcoming). Moreover, strong grandmothers may typically counter intergenerational interests and support among male family members.

9.2

Through the move away, it is the youngest generation in this family - the young father and his wife - who self consciously is remaking their life. This transformation is not however economic, it is more to do with a change in lifestyles, values and gender identities. At a number of levels and in a number of domains, the parent generation is engaged in a process of reparation, hence the descriptive term for this pattern of relations: reparation in estrangement. They seek not so much to repair kin relations but to break the cycle of transmission of negativity between generations. They appear to be making a conscious attempt to break with the patterns of parents and grandparents, while also remaining close in several senses to the grandparent generation. The father described graphically how his childhood and adolescence were ruptured by his parents having constantly to find new short life accommodation for their large family. Both he and his wife want to create stability in their own children's lives by staying put; they have also developed a strong commitment to the countryside as a 'better place' to bring up children. In contrast to the parenting practices of their parents and grandparents, the young couple is committed to gender equality by sharing the bringing up of their two young children; neither entered employment until the children were at school. At an intra psychic level, the young father has gone through a process of coming to terms with the disruptions and disruptive relationships of his childhood and youth. He spoke of a journey to self understanding: 'My own life, how I perceived that and , from looking at that, trying to see how I should bring up my own children. Watching other people and seeing how they do it.'

Conclusion

10.1

It has not been possible in this paper to present detailed analysis of the families. Instead I have provided brief exemplars to support a typology of intergenerational relations - of solidarity, incorporation, differentiation and reparation - which are several among many possible patterns. A typology is of its nature static. To understand intergenerational family relations, it is crucial to take account of life course and historical time which I have been able to do here only fleetingly. This typology refers to broad processes of resource transmission and support which take place within families over time and which emerged at a single point in time (when the families were interviewed). It emerged through an analysis of patterns of mobility, of occupational status, on the one hand, and geographical (im)mobility on the other. These two factors proved themselves to be robust indicators of variation in patterns of intergenerational relations.

10.2

In this analysis, two types of intergenerational relations stand out - the traditional solidaristic and the differentiated. These represent, in Durkheim's terms, two forms of solidarity - mechanical and organic. In the case of the solidaristic type (representing mechanical solidarity), there is less specialization in the family's division of labour; these families provide and exchange a wide variety of resources intergenerationally. In the differentiated model, there is greater specialisation in the division of labour; the families transmit certain kinds of resources and support intergenerationally but have greater recourse to formal sources of support. These patterns are generated by but also serve to reproduce occupational (dis)continuity and geographical (im)mobility. It is arguable that the other two types of intergenerational relations represent cases in transit or on a continuum between the other two. In the case of 'reparation in estrangement', the youngest (parent) generational unit seems to be moving from a pattern of traditional solidarity towards differentiation. The pattern of incorporation in the other quadrant suggests that these families are resisting forces of differentiation and so are closer to the solidaristic type.

10.3

In this analysis I have not attempted to unpack the meanings of solidarity in relation to the understandings of the families themselves. Undoubtedly, whatever the overall pattern of intergenerational transmission, families and family members may view their relations as being characterized by solidarity e.g. mutual trust, sense of belonging, and the transmission of at least some resources (Crow 2002). As Cornwell notes (cited in Crow 2002), we should not assume 'that the formal properties of relationships are a valid indicator of their content and quality' (Cornwell, 1984: 115). In drawing upon the concept of ambivalence, I have suggested that tensions arise and persist in families whatever their pattern of intergenerational relations and may be expressed in terms of interests, feelings, values and interpersonal relations. Even within traditionally solidaristic families, each generation seeks to strike out on its own while their 'strategies' may result in containment of such tensions. In emphasising intergenerational divisions and connections, I have inevitably downplayed gender divisions especially intra-generationally although I have noted how in solidaristic families mother - adult daughter relations are critical in reciprocating resources. Moreover, where such relationships were weak, they may lead not only to a change in intergenerational relations but also in gender roles (as in the case of quadrants 3 and 4).

10.4

In analyzing patterns of intergenerational relations and processes of reproduction and innovation, questions arise concerning the agency-structure dilemma . How far do different patterns of inter generational relations represent strategies which family members adopt and how far do they occur as a consequence of habits and dispositions? This is not an issue which I have sought to address here although I might well have incorporated habits and dispositions into the dimensions of the matrix. In so far as intergenerational relations reflect both structural and strategic aspects of human behaviour, it is perhaps worth quoting Bertaux and Bertaux-Wiame (1997) who suggest

'...socio-structural components may be found in those decisions and acts apparently most clearly powered by will.. the idea that a life trajectory may be determined - or rather, conditioned - much more easily by the supplying of a resource than by the imposition of a constraint lends an entirely new content to the concept of determination: one which includes both the socio-structural dimension and praxis' (p95).

Thus, in so far as innovation takes place in families, I would suggest that it is useful to understand it in relation to the availability and deployment of new resources (for example of place, occupation, ideas), as it is also useful to examine how continuities in families are reproduced and limit change.

Notes

¹ The study Four generations: working and caring was funded under ESRC Future of Work Programme Award No. L21252027 and was carried out with colleagues Peter Moss, Ann Mooney and Emily Gilbert at Thomas Coram Research Unit, Institute of Education.

² Grundy (1999) found in her analysis of the Omnibus Survey that with the exception of 50-59 year olds, close to three quarters of the sample was a member of a three generation family.

³ A third of those over 80 are members of four generation families as are a fifth of those aged 20-29 and 50-59 (Grundy 1999).

⁴ Footnote 9, p. 223 in chapter seven: Temporality and Process in Social Life in Andrew Abbott, Time Matters on Theory and Method, Chicago University Press 2001

⁵ We adopted the Biographic-Interpretive Narrative Interviewing approach (Wengraf 2001) with some adaptations. Following the method in the first part, respondents were invited to give an account of their lives from childhood onwards, with a minimum of guidance and intervention from the interviewer. This provided an opportunity for the respondent to present his or her own gestalt. In the second part , the interviewer invited the respondent to elaborate on salient events or experiences that had figured in the initial narrative. Third, using a more traditional semi-structured style of interview, the interviewer asked questions relating to the specific foci of the study if they had not already been covered in sufficient detail in the first two parts of the interview. This final phase also included the use of a vignette to explore normative views about parental employment and childcare relating to a contemporary situation of parenthood.

⁶ How the picture has changed from the 1970s is a matter of some debate. As Payne and Roberts (2002) suggest, opportunities for men fell while they increased for women especially during the 1990s.

⁷ Three families are clearly within this quadrant while a fourth family is somewhat marginal since the youngest generation had moved away from the region. The three generations in this family are however linked not only by similar occupational interests but by allegiance to a particular religious sect which is based in the city where the two older generations live. In terms of childcare support this grandmother acted in a very similar way to those who lived geographically close to their children. The family stayed 'close' in a number of ways.

⁸ Specific reciprocities involve exchanges between the same set of persons so that a grandmother who receives childcare from her mother when her children are young repays her mother when she required help in old age (Finch 1989).

⁹ The great grandfather was a self employed nurseryman and his son a draughtsman with both of them running their own businesses at some point.

¹⁰ This is a form of imbalanced or generalized reciprocity since repayment of the relevant individual is not (yet) involved (Finch 1989); however there is also evidence of specific reciprocity in this family in caring for the oldest generation who had helped the middle generation during childrearing.

¹¹ He became an unskilled worker as did his own son also. However other members of the family are in skilled work and a daughter in the grandparent generation is married to a professional worker.

¹² I have preferred this term to 'atomisation' which is applied in Luscher's explication of ambivalence (Luscher 2000)

¹³ The great grandmother towards the end of her working life became a clerical worker.

¹⁴ The grandfather uprooted his five children - including four teenage boys and took them to a part of the country where he had no work and no housing in order, he says, to get away from the dangerous influences of the inner city. All the family remained together and after many housing and job difficulties including one return to London they settled.

Acknowledgements

I wish to acknowledge the collaboration and contributions of my colleagues on this project - Peter Moss, Ann Mooney and Emily Gilbert. Thanks are due to Graham Crow for his generous comments on an earlier draft.

References

ABBOTT, A (2001) Time Matters. on theory and method. Chicago: Chicago University Press 2001)Footnote 9, p. 223.

ALDRIDGE, S (2001) Social Mobility: A discussion paper, Cabinet Office: Performance and Innovation Unit.

ALLATT, P. (1993) 'Becoming privileged: the role of family processes', in I. Bates and G. Riseborough, Youth and Inequality, Buckingham: Open University Press.

BENGSTON, V.L 2001 Beyond the nuclear family: The increasing importance of multi-generational bonds, Journal of Marriage and the Family,63, 1-16.

BERNSTEIN, B (1996) Pedagogy, Symbolic Control and Identity: Theory, Research, Critique. London: Taylor and Francis.

BERTAUX and BERTAUX-WIAME 1997 'Heritage and its lineage: A case history of transmission and social mobility over five generations in Bertaux B and Thompson P (eds) Pathways to social class: a qualitative approach to social mobility, Oxford: Clarendon Press.

BOURDIEU, P. (1986) 'The forms of capital', in J.E. Richardson (ed.) Handbook of Theory for Research in the Sociology of Education, Westport, CT: Greenwood Press.

BOURDIEU, P. and WACQUANT, L. (1992) An Invitation to Reflexive Sociology, Cambridge: Polity Press.

BRANNEN, J, MOSS, P and MOONEY, A (2003) Care-giving and independence in four-generation families. In J Brannen and P Moss (eds) Rethinking children's Care, Buckingham: open University Press.

BRANNEN, J (forthcoming) Three generations of childhood: through women's eyes.

CORNWELL, J (1984) Hard-Earned Lives: accounts of heath and illness from East London, London: Tavistock.

CROW, G (2002) Social Solidarities: theories, identities and social change, Buckingham: Open University Press.

FINCH, J (1989) Family Obligations, Cambridge: Polity press.

GRUNDY, E, MURPHY, M and SHELTON, S (1999) Looking beyond the household: intergenerational perspectives on living kin and contacts with kin in Great Britain, Population Trends, 97, pp.19-27.

HARRIS, C C (1980) The changing relation between family and societal form in western society in M Anderson (ed.) Sociology of the Family, 2nd edition, Harmondworth: Penguin.

MANNHEIM, K (1952) Essays on the Sociology of Knowledge, London: Routledge and Kegan Paul.

MASON, J 1999 Living away from relatives: kinship and geographical reasoning in S McRae (ed.) Changing Britain: Families and households in the 1990s Oxford: oxford University Press.

LUSCHER, K 2000 Ambivalence: A key concept for the study of intergenerational relations in S Trnka Family Issues between gender and generations: Seminar Report, Vienna: European Observatory on Family Matters.

LUSCHER, K and PILLEMER, K 1998 Intergenerational ambivalence: A new approach to the study of parent-child relations in later life, Journal of Marriage and the Family, 60, 413-425.

GOLDTHORPE, J and MILLS, C 2000 Trends in intergenerational class mobility in the late twentieth century December (Working paper).

HEATH A and PAYNE C 2000 Social mobility in A H Halsey (ed.) with J Webb Twentieth Century British Social Trends, London: Macmillan.

PAYNE, G and ROBERTS, J (2002) Opening and closing the gates: recent developments in male social mobility in London, Sociological Research Online, 6, 4, <http://www.socresonline.org.uk/6/4/payne.html>.

PERFORMANCE AND INNOVATION UNIT 2002, London: Cabinet Office.

ROSSER, , C and HARRIS, C (1965) The Family and Social Change, A study of family and kinship in a South Wales town, London: Routledge and Kegan Paul (abridged edition).

ROSSI, A and ROSSI, P (1992) Of Human Bonding : Parent-child relations across the life course, NY : Aldine Gruyter.

WELLMAN, B and BERKOWITZ, S (1988) Communities in B Wellman and S Berkowitz (eds) Social Structures : A network analysis Cambridge: Cambridge University Press.

WENGRAF, T 2001 Qualitative Research Interviewing: Biographic narrative and semi-structured methods, London: Sage.

Copyright Sociological Research Online, 2003