Abstract
This article examines the notion of ‘family’ to consider how it may be understood in people's everyday lives. Certain recurrent and powerful motifs are apparent, notably themes of togetherness and belonging, in the context of a unit that the person can be ‘part of’. At the same time, there may be important variations in the meanings given to individuality and family, evoking differing understandings of the self and personhood. I consider these ideas further through globally relevant but variable cultural themes of autonomy and relationality, suggesting the term ‘social person’ as a heuristic device to distinguish the sense of ‘close-knit selves’ that may be involved in some understandings of personhood. I argue that this version of personhood may be powerfully expressed through ‘family’ meanings, with a significance which can be at least provisionally mapped along lines of inequality and disadvantage within and between societies around the world. These forms of connectedness may be hard to grasp through those theoretical and methodological frameworks which emphasize the (relational) individual. I argue that, in affluent English speaking societies, 1 there may be little alternative to the language of ‘family’ for expressing such forms of relationality and connection.
Keywords
Introduction
I mean, you just take it for granted really don't you that you're in a family. (Pat Burrows)
How can you put it? Just being there. Just being part of a family's what's important. I don't know. It's hard to say really. No, I just can't think of anything to describe that one. It's just being there basically. Just being part of a family. (Sean Carlton) (Langford et al., 2001: 13)
While Pat Burrows and Sean Carlton might take it for granted that ‘family’ is something that is ‘there’ that you can be ‘part of’, scholars have questioned the term and its usefulness as a sociological concept for the last two decades at least (Ribbens McCarthy and Edwards, 2011; Ribbens McCarthy et al., 2008/2012). Wilson and Pahl (1988) for example, wrote more than two decades ago about premature sociological moves away from ‘family’ as an analytical category, pointing out how ‘family’ needs to be distinguished analytically from ‘household’, and the possibilities for ethnographic research in this regard. In more recent years, a variety of other analytic approaches and terms have been proposed and discussed (Morgan, 2003), including ‘family’ as an adjective (as in ‘family practices’, Morgan, 1996, 2011), or an alternative language altogether such as ‘intimacy’ (discussed in Jamieson, 1998, 2004), ‘relatedness’ (Carsten, 2004), or ‘relationships’ (discussed by Brynin and Ermisch, 2009). Through such alternatives, some scholars argue for the importance of decentring ‘family’ and refocusing academic scrutiny in ways that make apparent other forms of relationships and their meanings (Roseneil and Budgeon, 2004). Smart (2007), for example, argues that we may want to prioritize the concept of ‘personal life’ as broader and more inclusive than the concept of ‘family’, which she suggests is particularly important in countering the potential risk that the notion of family ‘subsumes’ the individual within the collectivity. Mason (2008) argues for the term ‘kinship’ over ‘family’, ‘because of its greater capacity to capture the dynamism and fluidity of people's lived relationships’ (Mason and Tipper, 2008: 137).
These are important considerations in debates on contemporary domestic and relational lives, but there are also some drawbacks to this shift away from the usage of the language of ‘family’ in sociological analysis. Gilding (2010) contends there is a risk of obscuring the inequalities and power dynamics around the continuing ‘conventions’ of family, through a sociological over-emphasis on the ascendancy of reflexivity in personal lives. Gilding thus argues the continuing importance of ‘family’ as an institutionally embedded social form, such that ‘the family is best understood as an institutional regime’ (2010: 774). This perhaps echoes the point made from a more qualitative perspective by Widmer et al., when they suggest that individual narratives of family experiences ‘are embedded in relational structures that exist beyond individual knowledge’ (2008: 7).
Gilding also makes another important point in passing – which he does not pursue – when he notes that, ‘the family is a stubbornly pervasive point of reference in everyday language’ (2010: 774), which is indeed the focus of this present article. While I recognize that the contemporary academic debates, referred to above, offer important and very useful contributions, my project here is to draw on those sociological and anthropological approaches which have long argued the importance of paying attention to everyday language and the topical concepts through which social actors construct their social worlds and give meanings to their experiences – sometimes described as a grounded, or an emic, analysis. 2 It is important, in the various debates about how to understand ‘family’ in contemporary societies, to include in this discussion a close consideration of how people themselves use this term. This clearly implicates a much larger question: why should, and how can, sociologists take people's accounts of their lives seriously, in the present instance, through the everyday usage of the language of ‘family’?
One possible response is to view people's usage of the language of ‘family’ as evidence that they are the cultural dupes of a wider dominant discourse, unwittingly invoking a dominant ‘trope in the cultural imaginary’ (Budgeon and Roseneil, 2004: 127). In the process, in their everyday lives people are seen to be reproducing moral discourses that are in fact socially structured, and that lead to the marginalizing and ‘Othering’ of some groups in systematic ways. However, drawing on the work of Gubrium and Holstein (1990, 2009), the approach for which I am arguing here is to suggest that, as sociologists, we should consider these everyday understandings and discourses of family as ‘real’, if only in terms of their consequences for social life. This requires us to approach such meanings seriously and respectfully while also maintaining a critical stance that considers both the circumstances in which these meanings are embedded as well as their consequences for social life in its broadest sense; thus, an emic as well as an etic analysis is needed, seeking to understand cultural meanings from the ‘insider’ perspective as well as to consider and critique them from an ‘outsider’ perspective, a both/and approach (Headland et al., 1990). Such a sociological project, I suggest, may reveal how the language of ‘family’ is used in everyday lives to express a sense of relationality that may be at odds with the dominant – almost hegemonic – individualistic cultures of Western capitalist economies and liberal democracies. If sociologists do, then, take such family meanings seriously, it may be appropriate to see ‘family’ as a central repository for culturally and personally meaningful understandings of human connectedness, for which – in contemporary affluent English-speaking societies – there is little alternative language.
Others (Rapp, 1982; Bernardes, 1987; Gillis, 1997) have discussed the ways in which people's variable family meanings may be seen at one level as contradictory and shifting, but at another level as a way of expressing some of the tensions apparent in personal relationships. If we build on this work to understand the multiple meanings of ‘family’, perhaps we (as sociologists and social scientists) can see how powerful is the language of ‘family’ precisely because it does act as a repository and expression for deep but ambivalent desires for – and, sometimes, fears of – belonging and connection. Indeed, such emotions led Walkover (1992) to describe ‘family’ as constituting ‘an overwrought object of desire’. In this sense, then, attending to the language of ‘family’ may provide us with insights about the human condition – about social relationships, separation and connection – which we might indeed want to take seriously.
In order to do this from a critical and culturally reflexive stance, however, we need to consider further how meanings of ‘family’ may express variable understandings of individuality 3 and of relationality, embedded as these are in particular historical, material and political contexts. This then implicates broader interdisciplinary discussions of how to understand human belonging and connectedness, which may help us to see how different family meanings implicate variable understandings of relationality and personhood. While everyday meanings may convey idealized or (sometimes, but less commonly) vilified understandings of ‘family’, in the discussion that follows I focus, not on the moral, emotional and evaluative connotations, but on what is being evoked in the relational language of family togetherness. This may be a particularly difficult task in the context of Western cultural assumptions which wed public discourses to assumptions and ideals of autonomy and individuality which are hard to evade, so I look also to anthropology for insights into underlying notions of personhood and self.
I will start, however, from a discussion of ‘family’ meanings as evidenced through empirical sociological work. There is now a significant body of robust qualitative research 4 – based in various localities, regions and countries, and ranging across age, generation, gender, social class and ethnicity – that offers insights into the topical meanings of ‘family’, and some researchers have explicitly considered what ‘family’ means to people themselves in their everyday lives. Much of this empirical material (although not all) is based in the UK, where the conceptual debate about ‘family’ has been particularly vigorous. Here I want to pull together some threads from these various sources to consider what they may tell us about the language of ‘family’ in everyday lives.
In empirical work focused on close personal relationships and the life course, the language of ‘family’ may occur quite spontaneously through unprompted talk in research interviews and it is such talk that I prioritize here. At the same time, various studies (some of my own included) have explicitly asked interviewees about their understandings of ‘family’ and what it means to them (as with the quotes from Sean and Pat from which I started), which is revealing in other ways. There are significant methodological issues concerning how and where ‘family’ talk is researched (Gubrium and Holstein, 1990), such as: how far and in what ways the language of ‘family’ is prompted or spontaneous; whether related individuals are interviewed jointly or separately; how talk between related people is theorized and analysed (Ribbens McCarthy et al., 2003b); and the interplay of family representations and practices (Cowan, 1999; Sameroff and Fiese, 1999). In what follows, I try to indicate some of this briefly, but a full methodological discussion of these issues is beyond the scope of the present paper (although I do return to the implications of methodologies towards the end of my discussion). My guiding question is to explore through published research what meanings are invoked when people talk about ‘family’.
It is, of course, an established approach for qualitative research to consider how far the findings of any particular study may also resonate (or not) with other qualitative work conducted with different samples, in different localities, with different researchers' approaches to analysis. In what follows, I draw most heavily on my own qualitative empirical work over three different studies (some with colleagues), since this is the work where I have most knowledge of the circumstances of the research and the interviews from which I draw. Additionally, I will include discussion by other researchers drawn from studies relevant to my concerns here. My focus will be upon research concerned primarily (although not exclusively) with heterosexual ‘mainstream’ or ‘ordinary’ relationships and family lives, since it is here that the language of ‘family’ may be used in the most taken-for-granted ways, as something that is unremarkable while also highly significant. Studies that have focused specifically on the meanings of ‘family’ in bisexual, lesbian, gay and transsexual relationships reveal that its usage in such contexts is often self-conscious and politicized (Weeks et al., 2001; Ribbens McCarthy and Edwards, 2011), while also engaging in many ways with ‘mainstream’ understandings of ‘family’.
The language of ‘family’ in contemporary Western lives
Togetherness and belonging
I first became aware of the significance of the language of ‘family’ in my doctoral research (Ribbens, 1990, 1994), when I found that I could not make sense of mothers' accounts of their lives with their children without also considering how the women framed their lives and aspirations in terms of what it means to be ‘a family’ – in very taken-for-granted ways – such that motherhood for my interviewees was inextricably linked to ‘being a family’.
I: Could you say in what ways you think perhaps you make a good job of being a mother yourself?
Susan: … Giving them loving is the main thing.
I: And loving involves all the things you mentioned?
Susan: Yeah, giving them a good family life I think. (Ribbens, 1994: 59)
Amongst these interviewees, there were many shared assumptions about ‘family’, and this language occurred spontaneously throughout all the interviews. The arrival of the first baby was thus not just a simple addition of another person to a ‘household’, but an essential ingredient in creating ‘a family’. For many people, then, a core issue in the meanings of ‘family’ is that it signifies something more than a collection of individuals, or a set of relationships, constituting a ‘unit’, to which the individual can ‘belong’ – much as Pat Burrows and Sean Carlton, quoted earlier, refer to being ‘in a family’, and being ‘part of a family’.
The following passages occurred in response to a direct question of what ‘family’ means to interviewees in a study of the family lives of young people aged 16–18:
A unit, to be together. You know, which I know a lot of people haven't got. (Susan, white, working-class mother) 5
A sense of identity, belonging. I think that's very important. Em, shared values, shared things like humour that you have just in your family … shared memories, the real sense of belonging is the strongest I know. (Hugh, white, working-class father)
Family means to me speaking with one voice. You know, if you see one, the other will represent the same thing. And to me that's family. You have the similar sort of frame of mind. Like, if you see my brothers, and how they behave, and how they, their manners, will be similar to myself, you have a trait. Like your family has certain traits, the conduct of your family (Otis, African Caribbean, middle-class father). (Gillies et al., 2001: 26–7)
The references here to ‘identity’, ‘shared memories’, and ‘speaking with one voice’ hint at something that is deeply rooted, perhaps shaping or linking individualities so that they merge, to the extent that ‘… if you see one, the other will represent the same thing’. It is also noteworthy that these speakers are all from less advantaged groups, in terms of social class or ethnicity, pointing to the possibility that a collective sense of family-based identity may be more common in situations of disadvantage or reduced resources over generations – a point to which I return later.
The general theme of ‘togetherness’ is extremely pervasive across numerous research studies, in various affluent English speaking societies (and maybe beyond, eg Jallinoja, 2008), as part of general talk about everyday lives. In the following quote, Shaw is discussing her study of the leisure activities of middle-class parents in the USA:
Many parents … place emphasis on … children learning about ‘the family’ as a value and the importance of family togetherness. In this sense, the purpose of family leisure is not simply something that is done for the sake of the children and/or to enhance child development, but also for the sake of the family as a whole and for shared family ideas and family cohesion … Family leisure is seen as a way to ‘cement’ relationships and ensure the stability of the family unit. (Shaw, 2008: 6–9, emphasis added)
Richards (1990) similarly found a widespread emphasis on ‘togetherness’ in her study of Australian suburban family lives, while Jallinoja (2008) found that an emphasis on family ‘togetherness’ has strikingly increased in Finland since the turn of the twenty-first century. Gillies et al. summarize the pervasiveness of the language of ‘togetherness’ found in their UK interviews with parents and young people aged 16–18:
The words ‘closeness’ and ‘togetherness’ recurred throughout interviewees' discussions of family … such as ‘living together’, ‘doing things together’, ‘going out together’, ‘sticking together’ or ‘pulling together’. (Gillies et al., 2001: 26)
Care and support
Another recurrent theme refers to care and support that can be relied upon over time, indicating something of the quality of the ‘togetherness’ that is expected in a ‘family’, powerfully expressed here by Paula, interviewed as part of a study of parenting and step-parenting after divorce and separation:
Security, stability. Um, loyalty. Not necessarily getting on all the time but knowing that you've got family who are there for you. Yeah, security and, er, somewhere to go when all else fails. (Paula – white, middle-class mother and non-resident stepmother). (Ribbens McCarthy et al., 2003a: 40)
In the next quote, from women living on a deprived housing estate in South Wales, we hear an exchange between a (grand)mother and her son's partner, Margaret, whose ties remained strong even during Margaret's (temporary) separation from the son.
Mother: I mean at the end of the day, couples break up, marriages break up
Margaret: The family is always there.
Mother: The family is always there. (Becker and Charles, 2006: 114)
As Becker and Charles comment:
The women also make clear that their children come first, and that children and ‘family’, meaning the female network around them, are more permanent, more reliable and more important than their relationships with men. (2006: 114)
Here, then, the gendered bonds around children are seen as the defining characteristic of ‘family’, as these ties are felt to be stronger than the bonds between partners and more capable of providing care and support. So here the language of family is used to indicate these (stable) connections, rather than the (more unstable) household-based nuclear family structure.
Emotions and ideals
The meaning of ‘family’ may thus entail expectations of a caring and relational unit sustained by togetherness and a sense of belonging. At the same time, it may also evoke a particular quality of experience with powerful consequences for people's self-evaluations. It is thus clear that many people, in very varied circumstances, imbue the term ‘family’ with their highest ideals and powerful emotions.
Family is the most important thing – the most important thing in the world you know (Somera, Pakistani, middle-class mother). (Gillies et al., 2001: 27)
Langford et al., in their UK study of parents and younger teenage children, consider some of the ideals and associated feelings of fragility involved in the language of ‘togetherness’, which is seen to be at risk as children grow older, as they elaborate in relation to Peter, a father of younger teenage children:
Paradoxically, nostalgia for a lost family togetherness, visible in many parents' accounts, allows the construction – or, perhaps the reconstruction – of the family as an ideal family. High days and holidays … are significant because they evoke a time when the family really were all together … Peter's desire to create opportunities for ‘togetherness’ appears to arise in response to his actual experience of a household where nobody in fact does anything together very much … Only by deliberately creating opportunities for ‘quality time’ can Peter experience the togetherness that reassures him that his family is, and will continue to be, a family, and not fragment entirely in the face of ‘all the other stresses that we've got’. (Langford et al., 2001: 48)
So what happens if everyday life not only threatens to expose the fragility of such idealized experiences, but actually directly contradicts these hopes and expectations? Barnard's Scottish study provides some sense of the power of the language of ‘family’ even in the face of very difficult circumstances, in this case, when the younger adult generation are heavily dependent on illegal drug use, leaving (grand)parents struggling with the effects they see on their children and grandchildren. As one mother put it: ‘Drugs … it destroys a family, so it does’ (Barnard, 2007: 28). Other parents also recounted very difficult experiences, such as having money stolen from their house by their children, and Barnard comments:
As this parent, like others, went on to explain, the only real route to recovering the money was to involve police and have their child legally charged with fraud. However, to do so was so humiliating and shameful and so far at odds with their notion of being a family, that most parents would resist this course of action … (2007: 30, emphasis added)
Nevertheless, Alyx, a young homeless woman (interviewed as part of the study on the family lives of young people aged 16–18), felt so let down by her family that she no longer wanted to refer to her parents and relations as ‘family’ at all. For Alyx, then, a family that utterly fails to care for its members ceases to exist, which becomes a powerful expression of rejection:
… (family) doesn't mean nothing to me any more, ‘cos as far as I'm concerned I ain't got no family (Alyx, white, working-class young person). (Gillies et al., 2001: 28)
Such quotes point to the complexity and depth of the underlying emotions and evaluations of ‘family’ at stake, which can also be ambiguous, ambivalent and context dependent. The emotions implicated in family meanings can comprise something that is much desired and longed-for, and/or something that is lost, and thus mourned. In these senses, then, the language of ‘family’ may act as a key repository of meanings and desires for togetherness and belonging.
Individuals and families
Themes of family as a supportive unit stretching backwards and forwards in time, providing a sense of belonging and care, and evoking deep emotions, are thus highly pervasive in the studies I have been considering here, even against lived experiences that are strongly at odds with such a view. Nevertheless, some understandings of family provide more scope than others for the expression of individuality. While the mothers interviewed for my doctoral research were living in similar circumstances (in terms of incomes and household structures), there were also important variations in how women described ‘family’. Thus, while most of the mothers emphasized family as a ‘unit’, others – less commonly – understood ‘family’ more as a collection of individuals who live together and are closely related. In some accounts of everyday lives, then, we can see how issues of individuality and togetherness, separation and connection, weave backwards and forwards through variable family meanings. These differences arguably implicate varying understandings of relationality, personhood and the self.
Thus, for some people there is an evident sense of tension between ‘being myself’ as an individual free to pursue my own preferences (sometimes couched in the language of ‘selfishness’), and ‘being a family’ with obligations towards the care of dependants – resonating with Askham's (1984) discussion of the tensions between identity and stability for individuals in couple relationships. While family as togetherness may thus evoke strong ideals and desires, family may also be described as constraining. Such differing understandings (and valuing) of individuality and togetherness might be played out – in these and other studies – in various concrete ways, including the naming of children (Gillis, 1997), the organization of space within the home (Edwards, 1993), and the display of family photographs (Rose, 2004). This variable emphasis on ‘the individual’ and ‘the unit’ is thus a key route into seeing how everyday taken-for-granted meanings of ‘family’ vary, with significant practical and material manifestations.
Feelings of family as constraining can be seen in the following quotes:
Having children means you have to give up being selfish … I mean I can't sort of do all the things that I would like to do, erm, because I can't just say to Sue, ‘Oh, do you fancy going to Paris this weekend?’ or something (John – white, middle-class half-weekly resident stepfather).
And my feeling of entrapment, if you like, in the [work] field and the [geographical] location and so on, which weren't actually what I'd been aiming for. I'd much rather have been elsewhere in the country (Jonathan – white, middle-class resident father). (Ribbens McCarthy et al., 2003a: 40)
By contrast, in the next quote, from an interview with an older couple living in a prosperous area of South Wales, the speakers seem united in their understanding of the significance and desirability of creating a ‘family’ unit:
Well I think, we have talked about this a number of times, Sue and I, that when we got married, we knew we wanted to have a family. A family. It wasn't just having kids, it was having a family. And making a family, and that's what we wanted.
As Becker and Charles go on to elaborate:
Having a family here means for Sue and Richard not only having children but bringing them up in a certain way … Changes in family life are seen as negative because ‘these days’ ‘the modern generation’ are not prepared to ‘give anything up’ and, by implication, are inclined to park their children off on other people if they have them at all. There seems to be an implied moral judgment of the ‘modern family’, and particularly the ‘modern woman’ who pursues a career at the expense of ‘having a family’. (Becker and Charles, 2006: 109)
For Sue and Richard, then, family and individuality should not be (and in their own experiences apparently were not felt to be) in tension. For others also, identity and selfhood may be closely bound up with experience of ‘being a family’.
I know me sisters and me brother and me mum so well that I can just act how I want to in front of them … I can just be how I want with them (Jean Mason age 12).
I like everything about my family … I feel pride in being queen of the house (Fazana Bokhari, mother).
It's proof of how you've lived your life … It's proof that I've worked hard (Ed Finch, father). (Langford et al., 2001: 14)
Overall, then, this brief discussion of contemporary research in affluent English-speaking societies demonstrates the powerful nature of the language of ‘family’, at both discursive and emotional levels, in which strong (if variable) themes of togetherness and belonging are apparent, along with expectations of reliable support. Indeed, it is precisely because ‘family’ is so laden with ideals that scholars may argue against its use as a sociological concept, while politicians may see advantages in the invocation of such powerful language. So how can a critical sociological analysis respond and attend to this language, and take account of the deep emotions which it conveys, while also seeking to avoid the potential pitfalls of a sociological reflexivity which might unthinkingly reproduce such ideals and expectations in the very process of researching them?
The individual, the self and the (social) person
Anthropologists have long debated how cultural understandings of personhood and the self may constitute variable responses to pervasive human dilemmas about connection and separation, interdependence and independence. Drawing on such work, Lukes suggests that there are two versions of the person discernible, one of which is universal but the other of which is culturally variable. The universal version is captured through the use of the personal pronoun, referring to an entity that has both states of consciousness and bodily characteristics, but the more variable version comprises inter-personal attitudes or a ‘structure of sentiments’ (1985: 298).
Semin (1999) similarly suggests that all cultures include a sense of ‘self’ in terms of an entity that is linguistically distinguishable from the non-self and that has continuity, including through a sense of willed action and responsibility, but that these elements vary in their cultural framing. In some cultures, then, the self is not distinguishable from the role that results from being part of a community or family: thus, in many parts of the world, ‘people think of the self as a component of an in-group, rather than as an independent entity’ (Triandis, 1987: 81). In such a context, the personal is not something that can be clearly separated out from the collectivity, in which case the question of whether the individual is subsumed within the collectivity does not make sense – a profoundly different cultural view from that of the autonomous individual of Western cultures.
Such cultural understandings of personhood have consequences for expectations of personal goals and group goals, orientations to co-operation and emotional attachments, and varying cognitive activities about how events, persons, selves and others (including the material environment) are understood. Furthermore, the sense of agency involved can be very different; individual behaviour may, for example, be seen as resulting from an ‘external’ power. Carrithers (1999) points to anthropological work based in India and Melanesia, where personhood is constituted by properties, goods and substances, in relationships with others. Writing as a psychotherapist, Kirmayer further points out how different sorts of therapy may be associated with contemporary cultural variations in understandings of personhood; thus, in Japan, Morita psychotherapy is oriented towards cultivating an acceptance of how things are, while Naikan psychotherapy is concerned with ‘recognising the immense and unrepayable debt [the individual client] owes to his [sic] parents and others in his life for their help and care’ (2007: 249).
By contrast, within sociological discussions there has been an increasing tendency to put the individual person centre stage of social theories and methodologies concerned with personal relationships and affiliations in contemporary Western societies, even as those writers who draw on feminist perspectives (amongst others) seek to argue for the importance of seeing the self and the person as inevitably always relational, and not to be seen as equivalent to ‘the individual’ (Lewis and Fink, 2004). In her discussion of ‘personal life’, for example, Smart draws on the work of Mead to lay out the ways in which the personal is always embedded in social relationships.
‘The personal’ designates an area of life which impacts closely on people and means much to them, but which does not presume that there is an autonomous individual who makes free choices and exercises unfettered agency. This means that the term ‘personal life’ can invoke the social, indeed it is conceptualised as always already part of the social … the field of personal life is the ‘me’ compared with the ‘I’ of the individualization thesis. (Smart, 2007: 28)
In considering these variable debates and understandings, it may be useful to distinguish several different potential versions of the person and the self. 6 The first constructs the person as autonomous, a bounded and unique entity that is the outcome of ‘successful’ socialization, with a ‘self’ that is experienced as agentic and self-directing – the quintessential ‘individual’ of Western culture. This view is notably manifested in such academic disciplinary orientations as mainstream psychology (Oyserman et al., 2002), with ‘individuation’ being seen as the goal of ‘healthy’ adolescent development, and well-being occurring through optimizing self-development.
A second version constructs the person as a ‘relational individual’, an entity that is produced through, and continually embedded in, relationships, but experienced as a (largely self-directing) individual; well-being is bound up with the web of relationships to which the person belongs but care of (the unique) ‘self’ is also important. This understanding of the person is more apparent in those academic perspectives that seek to argue for the intrinsic relationality of the person, yet nevertheless retain a focus on the self as distinct from the group.
A third version is associated with an experience of the person as a component of a collective unit, which is experienced as more than, or different from, a set of individuals-in-relationships; well-being is bound up with the unit, since ‘We are one’ – the ‘social person’. Here, personhood is rooted, not in ‘I’ and ‘me’, but in ‘we’ and ‘us’. Some scholars (eg Kağitçibaşi, 1997; Semin, 1999; Oyserman et al., 2002) suggest that this view is often more widespread in Asian and African societies, but may also be apparent in parts of European and New World cultures where it may, however, be pathologized. In systemic therapy in the USA, for example, it might be seen as demonstrating an unhealthy ‘enmeshment’ (Dilworth-Anderson et al., 1993) that prevents individuation occurring.
Setting aside here the notion of the autonomous individual, 7 I next explore further the notions of relationality that are potentially involved in the relational individual and the social person, before reconsidering how these may be expressed through the powerful relational language of ‘family’.
Relationality, autonomy and personhood
A dictionary definition of ‘relationality’ refers to: ‘… our lived relation to other human beings’ (http://www.phenomenologyonline.com/glossary/glossary.html#relationality accessed 31 December 2009). What this definition leaves hanging, however, is how to understand ‘other human beings’. Relationality can invoke a view of persons-in-relationships, drawing our attention to the ways in which particular social actors may trace and act upon affinities and ties, of whatever basis (the relational individual). 8 But other forms of relationality might also be possible, particularly where there is some sense of fusion with the collectivity, such that self-interest is understood through this collectivity – as with ‘the social person’.
The cross-cultural psychologist Kağitçibaşi (2005, 2007) thus argues that developmental psychologists have had too limited a notion of independence and autonomy as foundational to the healthy adult individual. She argues that the academic conflation of autonomy with the separated (individuated) self is a product of a US cultural system (in particular) which values both autonomy and separation, and places such a high emphasis on individuality that no other way of thinking about autonomy is possible. From this perspective, an individual might indeed be at risk – in terms of their agentic autonomy to pursue their personal well-being – if she or he is compromised by their membership of a collectivity. This view may thus underlie the ways in which some research interviewees – as discussed above – express family togetherness as potentially constraining and limiting of individual freedom.
Kağitçibaşi suggests, however, that this is not logically and psychologically inevitable, since the opposite of autonomy is not dependence but heteronomy. Relationality, on the other hand, refers to a different dimension, based on closeness and distance in inter-personal and family relationships. From this perspective, a close identification and fused boundaries within a group such as family may lead, not to enmeshment, but to ‘close-knit selves’ (2005: 411), which do not undermine the agentic self, constituting instead a form of relational autonomy.
If we turn to contemporary Western academic (and particularly feminist) scholarship, there has been a significant focus on re-thinking dominant Western cultural notions of the individual, and the associated concepts of relationality and autonomy. A crucial contribution of feminist scholarship in the late twentieth century was to challenge the notion of ‘the family’ as a biologically based, ‘natural’ unit, lying somehow outside of society and of sociological analysis (Yeatman, 1986; Thorne and Yalom, 1992; Glenn et al., 1994; Allen, 2009). This work exposed the need to analyse ‘families’ in terms of the individuals who comprised them, thus opening up the ‘black box’ of family lives to a whole variety of new questions and research, particularly round the axes of gender, and – somewhat later – generation. However, alongside this analysis, other feminists became concerned about how to understand the nature of the ‘individuals’ who were being brought to light, and particularly argued the need also to re-think individuality from a feminist perspective. This has led to a major body of cross-disciplinary work that has focused on relationality and autonomy, particularly in relation to theorizing ‘care’ (for a review of these debates, see Philip et al., 2012), although Kağitçibaşi (2005) argues that even the feminist work that explores the notion of ‘relational autonomy’ is in danger of reproducing the Western view of relatedness and autonomy as antagonistic. What is most pertinent to my present discussion, from this body of work and others, is to consider those writers who open up the possibility that relationality and autonomy are not unidimensional theoretical objects, so that relationality and relational autonomy are seen to encompass a variety of forms of human connectedness.
The philosopher Donchin (2000), for example, differentiates weak and strong notions in relational approaches to autonomy. In the weak notion, self-hood and capacities for autonomy are developed in the context of social relationships, but such relationships may be seen as entirely voluntary and contingent on agents' self-understandings. Where there is a strong model of relationality, on the other hand, Donchin suggests that autonomy itself is understood to be about more than individuals, requiring collaboration, reciprocity and balances of power. Elsewhere, Donchin (2001) suggests various models available for understanding such relational autonomy, including mothering, friendship and sistering, and considers how different models denote differing relationships of power, with potential for exploitation as well as caring. There is thus variability in terms of whether such relationships can provide equality, and whether or not they work towards fostering autonomy in the other, which Donchin takes to be a desirable goal. This discussion usefully points towards an analysis of relationality that retains a concern with inequalities. At the same time, it is notable that Donchin's examples concern dyadic relationships rather than collectivities or groups. This raises the (primarily empirical) question as to whether the connections between group members and the experience of being part of a ‘unit’ – such as a friendship network, or a church community – can provide the same sense of relatedness and close-knit selves that is sometimes apparent through the language of ‘family’.
Working from empirical research on people's residential histories and choices, the sociologist Mason (2004) outlines a continuum of relationality. At one end of her continuum, relational individualism refers to situations in which a sense of individual agency is clearly expressed, although this avoids appearing as ‘selfishness’ since individual agency is framed as meeting the interests of others. Mason suggests that this form of narrative may put an individualistic ‘gloss’ (particularly by men) on what appear to have been highly relational decision-making processes. Her continuum continues through the categories of relational constraint and conflict and relational participation, and thence to relational inclusion and co-presence. This last refers to narratives which do not address even the possibility that participants in making decisions might have differing interests or perspectives. However, Mason suggests that this position is only possible because there happens to be a convergence or consensus about individual views, in which all are described through a consensual ‘we’ (2004: 168). Mason concludes that:
we need to be able to keep the processes of relating in focus just as much as, if not more than, the individual or the self … both agency and identity need to be understood relationally … the selves that emerge from our narratives are not simply ‘selves in relation’, but relational selves. (2004: 177)
These writers make important contributions in thinking outside the individualistic box of much European and New World thought. However, arguably they do not quite capture Kağitçibaşi's view of close-knit selves, nor the levels of connectedness that may be expressed through the language of ‘family’ where, as Otis expressed it, ‘if you see one, the other will represent the same thing’ (quoted above).
Such theoretical issues also raise methodological and analytic questions. In her article, Mason points to the ways in which personal narratives are ‘sometimes wrongly conflated with the idea of an individualized self or narrator’ (2004: 178), and her solution is to argue for a careful analysis of how social relations, identity and agency are embedded in connections; but is this maybe like trying to overcome Cartesian dualisms by bringing categories into relation with each other, when perhaps the more satisfactory solution is not (always) to split them apart in the first place? In particular, Mason's emphasis on process perhaps draws our attention most clearly to what goes on in the social and inter-subjective spaces between individuals and selves, making it hard to transcend these. As long as sociologists use a methodological focus upon personal narratives and life histories, and an analytic lens rooted in the concept of the self, are we perhaps still bound to reproduce some of the ‘gaze of individualization’ that Mason is seeking to avoid?
There may be no easy solutions here but they are important questions – for both academic sociology and more applied professional and policy debates – that demand our attention, and may require creative conceptual and methodological thinking. Concepts such as ‘family culture’, for example, require attention to the shared historical, material, normative and psychological processes of collective relationships;'family phenomenology’ (Ribbens McCarthy, 2007) may point towards the co-constructions of (more or less shared) realities between family members; ‘family stories’ may evoke identities and narrated realities which persist over generations (Pratt and Fiese, 2004); while ‘family system’ provides another theoretical route into an analysis of close collective relationships from a more therapeutic perspective. Methodologically, the opening of the black box of ‘family’ referred to earlier has meant that researchers have moved away from acceptance of the account of just one family member to tell the story of others, and have developed many creative ways of interviewing, and engaging with, family members jointly as well as separately (eg Doucet, 2006; Gabb, 2008), while others have used observational methods (eg Vetere and Gale, 1987; Thomson et al., 2012). At the same time, it is necessary to recognize that such creative approaches may put the researcher into a more or less powerful position when it comes to interpreting these various materials (Ribbens McCarthy et al., 2003b). We need to find ways to build on such approaches to grasp the complex interweaving of relationality, autonomy and connection in the context of close relationships that persist over time and which may implicate ‘the social person’, expressed through the language of ‘family’.
At times, then, there is ‘something more’ about the notion of family that is hard to grasp within current academic language and methodologies that centre on the individual life, albeit that this individual is also seen as relational. While theoretical perspectives in the social sciences of Europe and the New World may struggle to capture and make visible this ‘something more’, this more collective understanding of personhood and family may be unevenly socially distributed within such countries, and between diverse cultures around the world, in ways that arguably link to levels of resources and affluence. At the same time, however, the language of ‘family’, where it evokes the collective sense of personhood and self, may also be seen to express a widespread human desire for deep connection, in which personhood is enhanced by, rather than (or as well as) constrained through, relationality.
Resources, inequalities and ‘the social person’
Politically, however, some might be concerned that the collective notion of ‘the social person’ neglects power differentials and potential exploitation (paralleling some of the dilemmas discussed in the feminist literature of ‘care’), and risks walking roughshod over the rights or conflicting interests of individuals – and such dilemmas are not to be casually set aside. It is thus crucial to maintain attention to power and inequalities between family members, as well as between differently situated households, even as we may seek to elucidate the connectedness of family ties. Such dilemmas point to the need to take into account the social, material and political contexts in which ‘the social person’ may be most apparent. Drawing on a wide range of empirical evidence, Kağitçibaşi argues that the majority of the world's population, and also less advantaged groups within more affluent societies, tend to prioritize relatedness over separation. Indeed, the collective sense of personhood that underlies ‘the social person’ may perhaps be seen as a realistic response to the uncertainties of survival for the majority of people in the world across history.
The cultural ramifications of ‘the social person’ can be seen, for example, in such key features of family lives as births 9 and deaths. While contemporary European and New World societies may (perhaps increasingly) view birth as creating a unique individual (Bernardes, 1985), in less affluent societies birth may be seen to be significant primarily as contributing to the well-being of the group. Similarly, from one perspective death may be understood as an ending of the self, but from another cultural viewpoint it may be seen as a threat to the survival of the unit (Nordanger, 2007). Bereavement, also, may be viewed as the loss of a unique relationship, while another cultural perspective may frame it in terms of continuity with ancestors who are significant for the group's survival and well-being over generations (Klass, 2001).
Furthermore, the links between resources, disadvantage and ‘the social person’ arguably are not only relevant to broad cultural contexts around the globe, but also to the social distribution of differing family meanings within affluent English-speaking societies. Thus the work of Hill Collins (1990/2008) and Reynolds and Zontini (2006), amongst others, has pointed to the emphasis on the collective among minority ethnic groups, while Ribbens McCarthy et al., 2003a, point to class issues that are involved with different understandings of family as a collective unit. Furthermore, the work of Edwards et al. (2006) points out how working-class children may understand their individuality precisely through their membership of a group – particularly siblings and friendship groups.
Conclusions – extending the analytic lens
Through an exploration of empirical materials, I have suggested that the language of ‘family’ may be used in everyday lives to represent quite differing understandings of relationality and the individual. Thus some everyday topical understandings consider ‘family’ as a set of co-resident individuals, or individuals-in-relationships, but others view ‘family’ in ways that evoke more than this, conjuring a ‘unit’ and a shared sense of identity and personhood that goes beyond the (relational) individual, or dyadic relationships. In this latter perspective, then, ‘family’ may be used to express a sense of ‘close-knit selves’ and belonging, referring to groupings that are not just understood as collections – or even networks – of individuals, or sets of practices or discourses.
Bringing together emic and etic analyses thus highlights that some forms of relationality are obscured through a focus on ‘relationships’, which remains orientated to dyads, or a focus on ‘personal life’, which is inevitably bound up with the individual as subjectively experienced. Such approaches can attend to the relational individual but are much less able to highlight, and make visible, other forms of relationality – notably what I have tentatively termed ‘the social person’. I am concerned, then, that we need to think very carefully about the ways in which particular academic concepts and associated methodologies may obscure or illuminate different aspects of people's everyday lives and sense of personhood, and may reflect particular cultural assumptions about individuality and relationality. The notion of personal life certainly overlaps in many interesting ways with the notion of family lives, and its development by writers such as Smart has been crucial in pointing out how far the personal is – and has to be – also relational rather than individualistic. Similarly the work of Donchin (2000, 2001) and Mason (2004) help to extend the concept of ‘relationality’ to analyse it as a continuum rather than a binary term, such that relationality may be seen to be more or less focused on the individual, while Mason's work on ‘affinities’ (2008) draws our attention to those more elusive – even while ‘tangible’ – aspects of human relatedness that may be particularly hard to grasp. We need to build on this work but take it further with regard to how we understand relationality, to think outside of the framework of the individual altogether, and I have here used empirical materials about family lives to try to explore this.
Certainly the language of ‘family’ provides a key discourse in contemporary affluent English-speaking societies through which people may express their understanding of ‘close-knit’ selves, providing (or expressing a desire for) a sense of togetherness and belonging that goes beyond the individual relationships of which it is comprised. This is almost certainly why the language of ‘family’, and ‘family’ relationships, are so commonly used to convey something powerful about a range of other groups that may be experienced in similar ways, such as neighbourhood gangs, strong friendship groups, Trade Unions, or churches. While such groups may also provide an important sense of relatedness, the use of the language of ‘family’ and family relationships to refer to them, is arguably invoked precisely to convey that they comprise ‘something more’ than a network of relational individuals, which other terms fail to capture. If we focus primarily on personal lives and the associated notion of the life course, we have to work at ensuring the relational is also kept in focus. But if we use ‘family’ as our conceptual lens, we are inevitably focusing on relationships and relationality. Furthermore, when we focus on ‘family’ discourse and practices, we may find that for some people, in some contexts, the individual is really not relevant at all, since ‘family’ is understood as a unit that is indeed greater than the sum of its individual parts.
In studying families and relationships in contemporary affluent societies, then, the prevalent methodological and theoretical emphasis on the (relational) individual occludes phenomena that can only be understood through the larger group. An emphasis on the life course may struggle to move beyond a view of ‘family’ as an aggregation of individual persons-in-relationships, such that some ‘family’ issues may require significant re-shaping if they are to be understood through such an approach. For example, themes of heritage, roots and generation within family histories and family stories may be invoked within personal lives but may also have a significance that cannot be captured through a focus on the life course of individuals. ‘Family’ is able to pull many disparate relational experiences together 10 – including the possibility of family culture in its own right, the significance of time past and future, and the sense of being part of something bigger – in a way other terms are unable to do.
These issues are difficult to consider without seeming to stray into moral and/or political minefields, which may risk idealizing connections and the communal understanding of ‘family’, and subsuming individual rights to collective orientations that conceal the material realities of inequalities and power differentials. Indeed, I endorse the vital significance of decades of feminist and other work that has sought to extricate the needs and rights of women and children from being subsumed within a collectivist ‘family’ orientation, and this project must continue. At the same time, however, I hope I have demonstrated that unless we do pay attention to such issues and variabilities, we risk imposing particular cultural orientations that over-ride, and render invisible, ways of understanding our personal lives and relationships that do not ‘fit’ dominant cultural discourses of autonomous – or even relational – individuality. Furthermore, if we listen closely to how topical terms such as ‘family’ are used in everyday contexts, such an approach may be crucial in making visible the implications of how social policy constructs ‘family’ in global contexts, as it is particular (Western) understandings of family that are dominant through globalized activities and organizations, such as international aid agencies, and the UN Declaration of the Right to Family Life (Ribbens McCarthy, 2009). At the same time, I have argued that the understanding of personhood which may be signified as ‘the social person’ is not only found in the majority of developing cultures around the world, but is also sometimes apparent in empirical data from affluent English-speaking societies. In such contexts, some meanings of ‘family’ in everyday lives may thus express forms of connection and relationality for which there is little alternative discursive space available. Perhaps this helps to explain why interviewees such as Sean Carlton, whose words I quoted at the start, can find it so hard to find the language to speak of the meanings of ‘family’.
Footnotes
1
I refer here to the English-speaking affluent societies of the UK, Australia and the USA, which are the countries involved in the empirical work used as the basis for the present discussion.
2
Anthropologists coined the terms ‘emic’ and ‘etic’ to distinguish between those analyses that prioritize ‘insider’ accounts and those which provide an ‘outsider’ perspective. These terms are less widely used in sociology, and are themselves subject to debate. 
, discuss the origins of these concepts in anthropology, and review the related debates. My discussion here sets out from an emic perspective, before moving on to an etic viewpoint, but these are matters of emphasis and are not hard and fast distinctions.
3
4
Such studies include: Becker and Charles's (2006) paper arising from the Rosser and Harris re-study led by Charles, of the family lives of people of varying ethnicity and affluence, living in different localities in South Wales; Langford et al.'s (2001) study of the family lives of young people aged 12–16 living in a city in northern England; Morrow's (1998) study of children's understandings of ‘family’, using multi-ethnic samples obtained through schools in different localities; O'Brien et al.'s research (1996) with ethnically diverse and predominantly working-class children living in London; Shaw's (2008) research on the leisure activities of middle-income US households. The research studies in which I have been involved, and which also present this sort of analysis, include: Ribbens (1990, 1994) on mothers of young children living in middle-income households in the south-east of England; on the family lives of young people aged 16–18 living in various localities in the English Midlands and south-east; and Ribbens McCarthy et al. (2003a) on parenting and step-parenting after divorce or separation, based in various locations throughout the south-east of England. Most of these studies concern heterosexual households, although the study reported in Gillies et al. (2001) included young people of varying sexual orientations. Note that this list of studies which provide relevant data to the present discussion is indicative rather than exhaustive.
5
In the various studies from which these quotes are drawn, the authors made different decisions about how to give interviewees pseudonyms and how to indicate their characteristics. For further insights into these decisions, readers are asked to refer to the original studies.
6
I am conscious of the potential drawbacks of such typologies, which risk being too simplistic and stereotyped. However, on balance they may be heuristically useful in thinking about relationality and personhood in current sociological debates.
7
The autonomous individual is not relevant to my present purposes, and some might argue that it is a political and psychological artefact or myth, eg Hockey and James, 1993; Mason, 2004; Hollway, 2006.
8
This also seems to be the version of the person that is implicated in recent configurational approaches to family and kinships relationships (Widmer and Jallinoja, 2008).
9
I am grateful to David Morgan for pointing out the relevance of this issue.
10
I am indebted to Rosalind Edwards for this insight.