Present and Absent in Troubling Ways: Families and Social Capital Debates *

Why social capital?

A range of reasons have been put forward as to why the concept of social capital has come so rapidly to the fore in recent years. Tom Schuller and colleagues (2000) review many of these. They note: a ‘politico-psycho-anthropological’ explanation that sees social capital arguments (in the ‘social capital lost’ vein) as resonating personally with the experiences of those in the heartland of intellectual and political debate; explanations focusing around a concern with the excesses of individualism; and the notion that ideas live in cycles, with social capital as a revival or rebranding of longstanding themes. They themselves, however, place primary emphasis on ‘a growing concern to revalorize social relationships in political discourse; to reintroduce a normative dimension into sociological analysis; to develop concepts which reflect the complexity and inter-relatedness of the real world’ (Schuller et al., 2000: 2).

Another explanation, though, especially in the context of considering families, relates to intellectual and political conceptions of contemporary society as undergoing radical social change, producing a strong impulse to instill certainty. ¹ We are said to be living in a ‘risk society’, marked by a pervasive and inescapable sense of concern about what is arising from past and current actions, and uncertainty about what will be encountered in the future (for example, Beck, 1992; Giddens, 1999). Intellectually, politically and popularly, social capital offers a particular sort of explanation of, and remedy for dealing with, perceived changes in the way we live, work and relate to each other.

As several commentators have pointed out (for example, Fine, 2000; Gamarnikow and Green, 1999), the social capital theorists favoured in policy development tend to give primacy to the economic or political effects or outcomes of family and social relationships. In so doing, they move towards explanations that view society as prior to and causative of the production of the economy, taking attention away from the economy to families and culture as the focus of policy intervention. Consequently, families are not viewed as necessarily interesting in themselves, but as significant for economic and political stability and order.

The focus on the need to impose order in the face of change, and the positioning of family values as a key causal feature that requires intervention, is writ large in the current Home Secretary, David Blunkett's remarks on releasing the recent Respect and Responsibility – Taking A Stand Against Anti-Social Behaviour White Paper (March, 2003). The White Paper includes measures to deal with wayward parents who do not take responsibility for socialising their children into respect for community and society – firstly through offering them support (intensive advice and training), and then if needed through compulsion (parenting contracts and orders, fixed penalty notices, residential parenting classes, and fostering). Blunkett is quoted as arguing that world insecurity (speaking at a time of concern about rogue states and terrorism, and the brink of war with Iraq) was creating a thirst for security at home:

Implicitly then, dysfunctional (working class) families are the internal rogue states with their yobbish children and young people as the internal terrorist threat.

This take on the condition of society keys into many of the concerns of the social capital thinking of theorists such as Coleman and Putnam that is influencing policy. Social capital points to a comfortable way forward. It provides a link between ordered families, ordered communities and an ordered society through shared values. In this, however, it is cross-cut with troubling presences and absences around gender and generation. This is not to say, however, that more critical work in the field, such as Bourdieu, entirely escapes these problematic issues either.

Families and social capital: strong but fragile

Both ‘the family’ and social capital are posed as fundamental and strong bases for social cohesion, but also as easily eroded and in need of protection and encouragement. These positions are located within the breakdown and erosion versions of both fields of work pointed to in the introduction.

A functionalist perspective on ‘the family’ – a complimentary nuclear conjugal unit of breadwinning husband, home-making wife and socialised children – has long treated it as positively meeting the needs of its members and carrying out vital functions for a successful society: an economic unit that provides shelter, food and care, regulating stable adult (hetero) sexual relationships, producing the next generation and socialising it into a set of common social values (eg Parsons and Bales, 1955; Fletcher, 1966; Goode, 1963; Murdock, 1949). Linked to this, more recent New Right and revisionist thinking also sees ‘the family’ in this way, as both a norm and an ideal – but as under threat. In particular, the ‘breakdown’ of the ‘traditional’ family and the rise of a range of diverse family forms (such as lone mother, same-sex, cohabiting, step, dual worker) are regarded as bringing irresponsible behaviour, social fragmentation and social disorder in their wake (eg Anderson and Dawson, 1986; Davies, 1993; Dennis, 1993; Dennis and Erdos, 1993; Halsey, 1993; Morgan, 1995; Murray, 1990, 1994). Mothers are seen as increasingly placing their own needs above those of their children and husbands, facilitated either by their entry into the labour market or the availability of welfare support, both enabling them to live independently. In this view, ‘the family’ thus requires protection from state intervention, especially in the form of welfare benefits that undermine family responsibility, and/or it requires normative and policy encouragement and support (see, for example, the policy initiatives laid out in Home Office, 1998).

The dominant strand of social capital theorising mirrors this functionalist assessment of the state of contemporary family life. Like ‘the family’, social capital is often said to be the ‘glue’ that holds society together. But it is also said to be the ‘oil’ that lubricates social life. This metaphorical tension between static sticking and friction-free fluidity is addressed in the next section. Whatever its consistency, however, social capital is portrayed as being eroded, and family change is placed at the root of this ‘social capital lost’ story, implicating particular forms of intergenerational and gender relations.

The socialisation of children is a key but underacknowledged issue in social capital thinking. ² For Coleman, for example, social capital ‘inheres in the structure of relations between persons and among persons’ (1990: 302), including obligations and expectations, and norms and effective sanctions. In particular, he identifies social capital as a resource within ‘the family’ that inheres in the structure of intergenerational relationships between parents and children. Parental interest in and support for their children enables those children to increase their human capital (educational achievement). Thus the more children that parents have, the more this attention is diluted. The process of social capital building within families integrally links to social capital as a resource outside ‘the family’, where parents and children are embedded in close community relationships. Parent-child relations and social ties outside families come together to create a dense social structure of cohesive norms, extensive trust and obligations (which Coleman calls ‘intergenerational closure’). Similarly, Francis Fukuyama (1995), and Eric Uslander (2001), each regard trust as a key component of social capital, and one that children are socialised into through their parents' attitudes to others and co-operation with them – although for Uslander this is unchangeably set within the early years of a child's life. Putnam (2000: 314) draws in the socialisation of children when he argues that ‘good families have a ripple effect by increasing the pool of good peers’, proving effective against youth crime and drug and alcohol abuse in neighbourhoods.

Bourdieu also focuses on generational socialisation within families in his thinking about social capital. He sees social capital as ‘made up of social obligations (‘connections’) … [it] is the aggregate of the actual or potential resources which are linked to the possession of a durable network of more or less institutionalised relationships of mutual acquaintance and recognition’ (1997–1986: 47–51). People derive their social capital from their membership of a group, such as a family or kinship group, but it has to be continuously worked at rather than merely being constituted in ‘the geneological definition of kinship relations’ (1997–1986: 52). Material and symbolic exchanges that produce obligations and mutual recognition of group (family) membership are built up over time and can be transmitted over generations. Bourdieu, however, is quite different from the theorists above because, although social capital derives from family relationships, he sees its type and content as inevitably shaped by the material, cultural and symbolic status of the individual and family concerned (1990). Thus, for Bourdieu, social capital is ubiquitous and continually transmitted and accumulated in ways that produce and reinforce social inequality, rather than undergoing a deterioration instigated by features of contemporary family life that fracture the ‘proper’ socialisation of children.

A number of changes in family life are posed as key in the ‘social capital lost’ story. In Coleman's view, changing family structures, specifically in terms of increases in lone mothers, ‘absent’ fathers and mothers working outside the home, have led to a deficit in social capital, as has geographical mobility. The rise of youth-oriented media and leisure activities and a focus on children's rights leads to a generational divide and further undermines parent-led norms in children's socialisation. In addition, the existence of welfare state benefits and services promotes ‘free-rider’ norms and negates parental and community trust-based obligations (1988a and b; 1990). Putnam also identifies some similar culprits in his portrayal of decreasing levels of social capital in society, even though his focus is on civic engagement through membership of self-sustaining voluntary associations in social capital building: ‘The most fundamental form of social capital is the family, and the massive evidence of the loosening of bonds within the family (both extended and nuclear) is well known. This trend, of course, is quite consistent with – and may help explain – our theme of social decapitalisation’ (1995: 73). Related possible features of contemporary family life that are causal in social capital decline that Putnam considers include women's rising labour market participation and dual earner households, a decline in marriage, increased television viewing, and an exodus of middle class and respectable working-class families from particular neighbourhoods that erodes the social capital of those left behind in the ‘ghetto’. Generally, an older, civically active generation is being replaced with a younger one whose ‘stocks’ of social capital are much reduced. In contrast, while Bourdieu (1997–1986: 54) also identifies ‘the mothers’ free time’ as a feature of intergenerational social capital transmission, as noted earlier he treats this as an aspect of social inequality rather than an indicator of social capital erosion (i.e. the family has to have enough income for the mother to be ‘free’ at home rather than in employment, and/or for buying in domestic help). Thus Bourdieu's conceptualisation of social capital is an aspect of his concern with the reproduction of privilege, rather than the theory of inadequacy championed by most social capital theorists.

As with the New Right and revisionist thinking on ‘the family’, the dominant ‘breakdown’ versions of social capital regard it as in need of protection and encouragement. Coleman is ambivalent about state intervention in this respect, posing substitute forms of organisational social capital building as a policy dilemma. Putnam, though, regards intervention to regenerate social capital as the central challenge for social policy to meet, including through encouraging family-friendly and community-congenial workplaces and youth volunteering schemes. ³ Indeed, Putnam (2000) poses social capital as a correlate of egalitarian policies. Such equality, however, encompasses material distribution across family households, not within them.

Within the dominant portrayal of the foundations of social capital then, ‘traditional’ nuclear families are present, accompanied by gender relations in which married fathers are breadwinners, and their wives are home-makers and dedicated to the care and socialisation of their (two joint biological) children as well as community activity. ‘Non-traditional’ families, as the other side of this coin, are also present, but in a troubling and destructive role. Such families do not conform to the longstanding gender and generational order that underpins a stable, social capital rich society, but are ‘changing’ them in negative ways. Their influence needs mitigation normatively and/or through policy interventions.

What is absent or silenced, however, are issues of power and conflict (other than in Bourdieu's policy-marginalised work – and see below for further discussion of these issues). The divisive and oppressive side of the ‘traditional’ family and gender relations, such as an unequal division of labour and domestic abuse, is ignored (eg Blaxter and Hughes, 2000; Levi, 1996; Molyneux, 2002). The complexity and ambivalence of family life, and its social class and ethnic specificities (see eg discussions in Allan and Crow, 2002; Reynolds, 2001; Ribbens McCarthy et al., 2003; Segal, 1995) are also obscured. Indeed particular ethnic and class groups are invoked in the family structures posited as social capital deficient. Lone motherhood, for example, is a common family form amongst African-Caribbeans, while larger families (more children than average) are more common amongst Bangladeshi and Pakistani minority ethnic groups (Platt, 2002). Both Black mothers and White middle class mothers are implicated in the idea that ‘dual earner’ families are undermining social capital generation, since it is these groups who are most likely to be undertaking substantial paid work ⁴ , while working class families are conjured up in Coleman's reference to low social capital ‘family deficiencies’: ‘poor, uneducated and disorganised, structurally weakened or broken by the personal disorganisation of the parent or parents’ (1988b: 385–6, his emphasis).

Also absent are conceptions of children as active participants in shaping the nature of family life and caring relationships, and in generating social capital for themselves, parents, siblings and other family members (eg Brannen et al., 2000; Edwards, 2002; Morrow, 1999, 2001). Moreover, as Ginny Morrow (2003) has pointed out, ‘children’ are not an homogenous category. For example, children's own social engagement appears to mirror or prefigure gender relations and practices in adulthood in many aspects. In social capital theorising, however, socialisation is seen as a one-way street, with children as passive ‘tabula rasa’ recipients. It draws upon traditional psychologically-based models that construct childhood as a state of ‘becoming’, in terms of their future capacities, not their ‘being’ in the here and now. In turn, this links to the increasing ‘familialisation’ of childhood, which focuses on children as the responsibility of their parents and their dependency on them (Edwards, 2002). Additionally, the ‘upside’ of increasing family diversity and changing relationships in terms of new forms of association is dismissed, an issue to which I now turn.

Diversity in families and social capital

In stark contrast to the place of families in the ‘social capital lost’ story, theories of late modernity pose family diversity as pioneering ‘brave new’ forms of association and egalitarian relationships (Stacey, 1990). Post-industrialisation has led to individualisation and democratisation, fluidity and variety in family relationships, and more complexity and choice in who/what constitutes family. This shift is viewed as having positive potential, including opening up space for a flowering of social capital in new and transformed ways. Indeed, the end of the 1960s/beginning of the 1970s as heralding the generational decline that is central to the ‘social capital lost’ story (Leigh and Putnam, 2002), is the same time period identified by some theorists of late modernity as the zeitgeist moment when the optimistic possibilities offered by individualisation and democratisation began to take hold (Giddens, 1999).

Anthony Giddens (1991, 1992) describes a post-traditional society in which men, women and children are progressively freed from definitive gender and generational relationships and obligations. In place of the norms of the traditional nuclear family, people have to make their own family and other lifestyle choices in an uncertain world, and are increasingly seeking intimate connections with others that are sustained on the basis of mutual understanding and respect for autonomy. Relationships are no longer grounded in gender or generation prescribed obligations and responsibilities, providing greater opportunity to negotiate and create more egalitarian, autonomous and fulfilling associations. Ulrich Beck and Elisabeth Beck-Gernsheim (1995, 2002) articulate a similar picture, with the process of individualisation potentially altering the nature of family life, which has an impact on political and ethical frameworks more widely. The ‘gender revolution’, change in family and intimate relationships and, especially, women's entry into the labour market, are driving forces of broader social change. People from a variety of backgrounds are increasingly mixed together in relationships that have to be established and constantly renewed by individuals, in place of traditional and taken-for-granted forms of community. There is the offer of new forms of emancipation, shaping different public repertoires around which identity (rather than place) based communities can mobilise. Gay and lesbian relationships are often placed at the forefront of this cultural shift, innovating ‘families of choice’ and alternative forms of social capital (Stacey, 1996; Plummer, 1995; Weeks et al., 2001). Traditional kinship bonds and family values are seen as characteristically limited and constraining, while new forms of association and communities of interest provide the experimental resources for people to be able to cope with an uncertain and changing world (Pahl and Spencer, 1997).

The ideas put forward in much late modern theorising are largely speculative and, in highlighting adult gender and sexuality relationships, tend to gloss over issues of generation, class and ethnicity. Evidence for them depicting the nature of contemporary family life, and as a contrast to the past, is limited (Crow, 2002; Jamieson, 1998; Ribbens McCarthy et al., 2003). Nevertheless, this strand of family studies provides a challenge to the negative role accorded to non-traditional families in policy-dominant social capital theorising, seeing openness and promise in the rise of diversity and fluidity in family life. It also argues against a picture of collapse in trust and reciprocity as constitutive features of social capital. It provides alternative understandings of how they are now constituted in practice, based on reflexivity and negotiation concerning mutual benefits and shared satisfactions, rather than prescribed and regulated obligations. Social capital theories, though, do involve some aspects of diversity as a positive and necessary feature of social life, in conceptualisations of different forms or mechanisms of social capital, as I now elaborate.

Coleman's work on social capital is largely concerned with advocating a social world characterised by what is termed a ‘bonding’ form, in that social capital (as ‘glue‘), is based on ties between adults and children living in the same sort of (nuclear) family in which children are socialised into an homogenous set of values, and participating in communities made up of networks of these families who share the same values, as discussed above. Family diversity and individualisation thus are not features of Coleman's vision of social capital generation. It could be said that diversity weakly enters the picture for Coleman in the form of his acknowledgement of a need for ‘multiflex’ rather than ‘simplex’ associations; that is, relations occuring in more than one social setting (as neighbours, workers, parents and co-religionists), but this seems to be stretching a point. Bourdieu can also be said to be concerned with bonding social capital within and across generations. People derive their social capital from the obligations and mutual recognition involved in their family, class and other social group connections. As noted earlier, though, Bourdieu is more concerned with exclusionary practices around bonding that create and sustain inequalities, while Coleman sees bonding social capital largely as an unmitigated good. Thus Bourdieu takes strong account of diversity and difference in terms of deep and continually reproduced social divisions and their relationship to power. Indeed, his work distinctively holds together family studies and social capital studies in this respect. This is a strength that has been utilised by some feminist work in the families field, for example Pat Allatt's (1993), Inge Bates' (2002), and Diane Reay's (1998) studies of the work involved, especially from mothers, in middle class parents' transmission of advantage to their children through social capital manipulation.

Other social capital theorists are also concerned with diversity as an aspect of social capital generation, but in terms of a necessary ‘bridging’ across social difference rather than an analysis of how those divisions are accumulated, transmitted and maintained. In the versions of social capital that have been influential on political thinking about issues of social exclusion and economic development (notably, Fukyama, 1995; Putnam, 2000; Woolcock, 1998), bridging social capital (as ‘oil’) refers to horizontal trust and reciprocal connections between heterogeneous people from different walks of life, and is more valuable and effective than bonding social capital. Putnam, for example, characterises the bonding form of social capital as based on homogeneous ties of solidarity between ‘people like us’, inward-looking and reinforcing exclusive identities, and restricted to enabling people to ‘get by’. In his view, it is thus of less benefit to generating rich social capital more widely in a society. It is outward-looking co-operative bridging social capital that enables people to ‘get ahead’ in life and fosters social inclusion (and notably incarnate in voluntary associations). Fukuyama makes a similar case in posing ‘weak’ ties of trust and reciprocity between loosely connected people outside of families as enhancing economic prosperity.

Putnam and Fukuyama, however, diverge when considering the consequences of family diversity for bridging social capital. Fukuyma's stance seems quite ambivalent. While he regards individualisation, manifesting itself in rising divorce rates and births outside marriage, as disruptive for societies (and see also below), he also speculates that such trends may potentially promote beneficial weaker social capital: ‘Although the breakdown of a family itself constitutes a loss of social capital, this breakdown may actually lead some family members to greater levels of association with people and groups outside the family … it is possible that the weakening of bonds within contemporary western families leads to an increase in social ties outside the family’ (1999: 117–8). In this way, Fukuyama has at least some points of agreement with theorists of late modernity. Putnam, in contrast, seems to have shifted from acknowledging tensions or trade-offs between bonding and bridging forms of social capital, with the the former important but potentially exacerbating inequalities (2000), to seeing a link between the two: the more bonding in a society, i.e. stable nuclear families, the more bridging will occur (personal discussion, 20.1.03). For Putnam, then, family diversity firmly remains a feature of the social capital lost story. While diversity is important to him, because heterogeneity is a prerequisite feature of bridging social capital, this only applies to difference ‘outside’ of families, for example class and ethnicity. He does not acknowledge that forms of family diversity may represent and/or encourage bridging social capital, for example for children who are moving between parental families. Moreover, such diversity and any associated inequality and social exclusion is posed more in terms of social plurality, rather than social divisions involving conflict and power relations. This is in contrast to analyses that would point towards the importance of conflict and debate as much as consensus in a diverse society (for example, Amin, 2002, on ethnic diversity).

Michael Woolcock (1998) has also added the notion of ‘linking’ social capital to the typology of forms, to allow for the capacity to access resources from formal institutions. Linking social capital is concerned with ‘vertical’, rather than horizontal, relations. This evades the issue of vertical relations within bonding social capital, as in taken-for-granted asymmetrical gender and intergenerational dependencies within families – a feature that is lacking in most discussions of bonding social capital. (Although it could be said that, while Coleman also takes familial vertical asymmetries for granted, he utilises rather than ignores them) I will return to these issues of hierarchy, conflict and power later in the paper.

Ideas around bonding and bridging represent a tension in social capital as a body of conceptual work. Theorists take different stances towards the importance of bonding and bridging forms of social capital, and their relationship to each other, and it is consideration of family diversity that helps highlight this. It is perhaps a paradox, then, that in one respect late modern ideas of individualism, family diversity and new forms of association seem to articulate quite easily with dominant social capital notions of bonding – like-minded people in reciprocal intimate relationships based on trust and equality, albeit that this is regarded as based on contingent negotiation in one and on prescribed obligation in the other. Issues of particularistic bonding and generalised bridging also feature in my next discussion, of analytic approaches to understanding both family life and social capital in terms of economic or moral rationality and how these relate to ideas about ‘the private’ and ‘the social’.

Economic or moral rationality

As noted earlier, several commentators on mainstream social capital theorising have pointed out that it tends towards explanations that view society as prior to and causative of the production of the economy, providing non-economic solutions to social problems (eg Fine, 2000; Garmarnikow and Green, 1999; Portes, 1998). Yet at the same time, dominant perspectives on social capital are imbued with economic rationality in their understandings of how people are motivated to act. Indeed, Coleman explicitly sought to combine economic rationality and social organisation theories in his work: economically rational exchange and ‘cost-benefit’ action, in the form of obligation investment and repayment, produce a structural moral rationality, in the form of cohesive norms and sanctions. ⁵

For Coleman, parents ‘invest’ in their children – as the next generation of the family who will in turn support them in later life – by being physically present, giving them attention and developing an intense relationship with them that involves talk about personal matters and expectations of their educational achievement. Bourdieu's work also runs the danger of falling into this economic rationality mould, in that parents ‘invest’ in their children for the continuance of the family's economic interests across generations – although generally he is critical of understandings of human behaviour that see it as intrinsically rational or utilitarian, since this takes no account of social divisions, cultural environment and historical circumstances. ⁶ In mainstream social capital theorising, in the wider community too, people ‘do things’ for each other. They expect and trust that these actions will be repaid so that, in due course, they will benefit from the ‘cost’ of their helpful effort – an assumption of economic rationality. The existence of state-provided welfare benefits and services thus undermines such economic rationality in producing ‘free-rider’ norms, while women's economic independence through the labour market and notions of children's rights undermine ideas of gender complimentarity and intergenerational obligations, as noted earlier.

Ralph Fevre regards the recent prominance of the concept of social capital as the result of ‘one of those periodic upswings when economic rationality comes to the fore in all aspects of our lives’ (2000: 95). Rather than a decontextualised pendulum effect, however, I have argued earlier that it is a manifestation of a (theoretical, economic, social and political) desire for certainty and non-contingency in a perceived ‘risk society’. ⁷ For similar reasons, economic rationality is also a feature of theorising and policy-making concerning a whole range of aspects of family life, prioritising individual utility maximisation and in particular perceived economic costs and benefits (see discussions in Barlow and Duncan, 1999; Carling et al., 2002; Duncan and Edwards, 1999; Duncan et al., 2002). This understands economic self-interest at the root of, for example, partnership choices, gendered divisions of labour, employment behaviour, decisions to become a parent, and child care choices. For some then, the economic rationality and consumer culture of late capitalism has invaded the domestic sphere, corrupting the way in which family relationships are understood and experienced (Fevre, 2000; Halsey, 1993; Sennett, 1998).

Yet, another explanatory strand of work in the area of family studies shows that prioritising economic rationality misattributes both form of reasoning and values ⁸ . In this work, social ties and moral responsibilities are placed centre stage, with people building identities and reputations as a certain sort of person. People – adults and children – negotiate these moral identities and reputations in understanding how they should behave in relation to others, and notions of the ‘right thing to do’ are enacted in this context. Structural contexts such as gender, social class and ethnicity are important in this, including in relation to prescribed norms in a particular social and local context. (Examples of work in this vein include Duncan and Edwards, 1999; Finch and Mason, 1993; Holland et al., 2000; Holloway, 1998; Jordan et al., 1994; Ribbens McCarthy et al., 2003; Smart and Neale, 1999; Smart et al., 2001) This major strand of theorisation poses family life as deeply moral – an important aspect of understanding that cannot be captured in mainstream social capital's underlying economic rationality.

Indeed, because of its assumptions of economic rationality, as this combines with the underlying posed oppositions between the ‘private’ and the ‘social’ or ‘public’ discussed below, social capital theorising often places the ‘bonding’ social capital exemplified in families (described above) as outside morality, as ‘amoral‘ ⁹ . While Bourdieu is an exception here because he does not envisage social capital as a ‘public good’ in the ways that Coleman, Putnam and others do, it is the dominant economic rationality version that is influencing social policy development. In this version, ‘generalised’ trust – that is a ‘disinterested’, impartial trust extending beyond particularistic group attachments to members of other social groups – is posed as moral. Trust within a ‘socially integrated’ group, however, does involve particularistic attachment and is termed ‘amoral familialism’ (see Woolcock, 1998: 171–2). ¹⁰ Amoral familialism, despite or because of the imputation of economic rationality to family members, also undermines the economic efficiency of nation states and regions. Fukuyama (1995), following Tocquevillean assumptions, argues that in societies where family allegiances are elevated above other sorts of social ties and obligations, there is not sufficient broad social capital for economic development and prosperity in global markets. Putnam (with Leonardi and Nanetti, 1993) similarly contrasts the civically vibrant horizontal networks of the north of Italy with the less economically prosperous amoral familialism of the south. Yet, as noted earlier, mainstream social capital theorists also argue that it is in families that children are socialised into trust and other (moral) bases of social capital.

The absence of an understanding of family life as deeply moral is achievable only because of the dichotomous removal of family from conceptions of what constitutes the ‘social’ and ‘citizenship’ into the ‘private’ sphere (see Ribbens McCarthy and Edwards, 2002; Sevenhuijsen, 1998). Morality, from this particular political science perspective, is only applicable within the ‘public’ domain through the collective civic engagement of independent, autonomous individuals. For example, Fukuyama (2001: 11) has argued that ‘excessive individualism’, in the form of a ‘preoccupation with one's private life and family’, as the ‘vice of modern democracy’, can only by countered by involvement in voluntary associations and public, civic affairs.

Again, there are gendered and generational implications to this stance. First, a particular masculinist understanding of the individuals who come together in the public, civic sphere is invoked, as self-contained and self-sufficient – the atomised subject of liberal political philosophy. This understanding of the individual, however, is not equally socially distributed. It is far easier for men to appear to fit its requirements. Least able to are children, who are the prototypes for forms of ‘dependency’, and are institutionalised as dependent in families, education systems, social security systems, and so on (Burman, 1994; Edwards, 2002). This solipsistic version of the individual shuts out, or rather hides, more relational notions of interdependence and connection highlighted by some feminists (eg Ribbens McCarthy and Edwards, 2002; Sevenhuijsen, 1998):

Secondly, women and children have historically been marginalised from the public sphere physically, normatively, politically and conceptually. Concommitantly, it is mothers who are normatively (and usually in practice) responsible for children's emotional and practical care and upbringing. ¹¹ They do not benefit from family as a social capital resource for public sphere engagement in the same way as men, for example in terms of family connections easing the path to financial and/or political power (Lowndes, 2003; Sapiro, 2003). Further, mothers may be involved in informal and small-scale networks of association that are not acknowledged by the dominant social capital focus on more formal, organised civic society, such as child care networks, liaison with health visitors and so on (Lowndes, 2000, 2003; Stolle with Lewis, 2002). Indeed, they may use this social capital to help them balance employment and family life, rather than mothers' employment being part of the destruction of societal social capital. Thus, in social capital terms, unless the activities of mothers and children are conducted within the public sphere, in recognised civic engagement, they remain absent or ‘amoral’, despite their posited key significance to social capital building.

As some feminists have pointed out (Blaxter and Hughes, 2000; Molyneux, 2002; Sapiro, 2003; Stolle with Lewis, 2002) through their gendered social positioning as family and community carers and networkers, women are central producers of social capital. But this very social positioning, and consequent social capital generation, is in itself problematic. These feminist critiques of social capital are informed by work that makes clear issues of gendered inequality, power and conflict in society and families. This would point to questions about any easy assumption of consensual, particularistic trust and reciprocity within families – whether considered amoral or not – and what such assumptions silence.