Researching family practices in everyday life: methodological reflections from two studies

Abstract

The study of the everyday is recognised as central to the understanding of identities, agency and social life. Yet, attempts to research everyday life often fail to capture the complexity of the mundane. This paper draws on findings from two studies: fatherhood across three generations and adult narratives of childhood language brokering to illuminate that complexity. The purpose of bringing these studies together is methodological; in particular, it is to examine how the storied material generated by narrative approaches can contribute to understanding of the everyday practices of family life, practices that are often unacknowledged, hidden or assumed. One study adopted a narrative form of interviewing while the second combined narrative photo-elicitation techniques with narrative accounts. Methodologically, the two studies illustrate that no one method produces ‘objective’, comprehensive knowledge of family practices. Together, however, they produce new insights into family practices around fatherhood, migration and language brokering and how participants ‘do’, ‘display’ and commemorate family. The paper argues that narrative approaches, sometimes alongside visual methods, can assist holistic analyses of family practices from sociocultural, intergenerational and life course perspectives.

Keywords:

• family practices
• fatherhood
• language brokering
• life stories
• narratives
• photo elicitation
• migration

Introduction

This paper focuses on two family studies conducted separately by each of the authors. One looked at how adults from different family backgrounds negotiate their identities as they re-evaluate earlier family experiences, while the second examined fatherhood over the generations. The purpose of bringing these studies together in this paper is methodological; in particular, it is to examine how the analysis of storied material generated by narrative approaches can contribute to understanding the everyday practices of family life; practices that may often be unacknowledged, hidden or assumed. Methodologically, the study of many family practices poses particular challenges. As Bourdieu (1979) observed, tastes and dispositions are often not conscious once they have become habitual. Moreover, the minutiae of everyday lives are not readily recalled and observed by others (Boddy & Smith, 2008). Memories are constructed with hindsight, and narrative accounts are created and performed as co-constructions between participants and researchers – in accordance with current and past individual and collective identities (Freeman, 2010; Phoenix, 2004). Narrative approaches that are increasingly employed across the social sciences have been particularly useful in enabling people to reflect upon practices that have become habitual or are taken for granted (see Martens, 2012).

The data from the two studies are part of the Parenting Identities and Practices project of the ESRC’s NCRM Novella Node (Narratives of Varied Everyday Lives and Linked Analyses)¹ in which we are re-using data in order to advance secondary narrative analysis. The boundaries between primary and secondary analysis are not always straightforward and, although we are revisiting the studies, the material presented here has not previously been analysed. In the paper, we set out some theoretical aspects of the concept of family practice. We then outline some key aspects of narrative approaches (Brannen, 2013). Next, each of us describes and discusses the approach applied in our studies, the theorisation of family practices, and how successful the methods employed were in illuminating how family life is routinely lived. The first study, with adults looking back on childhood practices of interpreting and translating for their parents, employed a narrative interview approach that started from the common narrative principle that interviews should be sufficiently open ended to allow participants to express themselves as fully as possible and in the way they choose. It drew on Hollway and Jefferson’s (2012) Free Association Narrative Interview method and on Ricoeur’s (1991) notion of emplotment. The Free Association Narrative Interview is guided by Freud’s psychoanalytic principle of free association and employs open questions to encourage research participants to make any associations that come to their minds between events in their lives, in order to gain insights into how these are emotionally and meaningfully linked (see Hollway & Jefferson, 2012 for the detail of this method). The Free Association Narrative Interview also allows the application of Ricoeur’s notion that analysis of how participants plot their stories gives a good sense of the ways in which the stories are significant to the participants (Kirkman, Harrison, Hillier, & Pyett, 2001). The approach of the second project, a three-generational study of fathers, was also narrative. It combined a biographic-narrative approach (Wengraf, 2001) and a semi-structured interview with a photo-elicitation technique where the latter was used as a further prompt to narrative about family practices. In both cases, the methods were chosen to provide for a holistic and complex focus on individual agency and identities forged in sociocultural context and an intersectional, intergenerational and life course perspective. The new focus on family practices resulted from bringing the studies together in the NOVELLA project.

Family practices

Theoretically, family and personal lives offer a useful frame for understanding everyday habitual practices within wider social structures (Smart, 2007) and at the intersection of time and space as they produce and reproduce identities (Shove, Pantzar, & Watson, 2012). While neither study originally foregrounded family practices, the concept is critical to the study of family life in several ways. The routinisation of practices in families makes life easier: it takes for granted the scheduling of household tasks, caring responsibilities and other routines. This saves effort that would otherwise be spent in negotiation and assists with time compression and producing a ‘comfortable groove’ of order, repetition and coherence that avoids the necessity for ‘inventive thinking’ (Felski, 2002, p. 612). The importance of habitual practices to the construction of family identities is demonstrated by the readiness with which families claim practices and their centrality to ‘identity projects’ (Giddens, 1991). For example, ‘reconstituted families’ quickly develop narratives suggesting that they have entrenched, shared, family practices (Burgoyne & Clark, 1988). That this is consequential is illustrated by the (sometimes difficult) negotiations new couples engage in about the elements that constitute practices such as Christmas celebrations (Mason & Muir, 2013). Practices are composed of multiple, interdependent elements that last beyond specific moments of enactment and that can be reconfigured over time, both recruiting practitioners and ‘losing’ recruits (Shove et al., 2012). Family practices are embedded in culture and history (Gross, 2005), in ways that mean the personal and social are inextricably linked (Smart, 2007). They are also about time, linking biography to history and configurations of intergenerational relations (Morgan, 2011) and are a site for both family change and reproduction in that myths, scripts and stories that get repeatedly retold serve to maintain a family’s ethos and idealised notion of itself and facilitate change towards the ideal (Byng-Hall, 1995; Samuel & Thompson, 1990). The ‘families we live by’ are often not the same as the ‘families we live with’ (Gillis, 2002).

Drawing on both practice theory and theories of practices in families, ‘family practices’ can be said to consist of multiple, interdependent elements (Shove et al., 2012), with an emphasis on doing, action and social action. They serve to define and establish practitioners as family members (Morgan, 2011). The new analyses presented below give insights; methodologically, into how it is possible to ask new questions of data generated for other purposes, by deploying different theoretical concepts and methods and substantively, into the ways in which participants construct their family practices.

Some key aspects of a narrative approach

The term narrative denotes the activity of narrating or telling stories that relate to sequence, and generally, temporality (Andrews, Squire, & Tamboukou, 2013; Elliott, 2005). One of the main functions of narrating is to ‘make present’ life experience (Schiff, 2012). Narrative speech is, moreover, part and parcel of everyday speech in which one may choose to tell or not to tell particular stories or experiences. The social context is, therefore, a key aspect for the researcher to bear in mind in analysing narrative, particularly since narration and storytelling are co-constructed within the interview between interviewer and interviewee (Phoenix, 2013). Narrative speech depends on having an audience. It is therefore a form of social action performed in the research encounter (Atkinson, 2005) and so involves performativity (Riessman, 2008). For while narrators have privileged access to the floor, they must ensure the audience stays engaged, clarifying what they believe their audiences need to know in order to understand and placing limits on what they say. They deploy strategies to re-enact the past as if they and their audiences were present at that time, for example by deploying emphasis, repetition and direct appeals to the audience’s attention and judgment. Tellers may employ dramatic techniques of metaphor and metonymy to convey meaning and pay attention to aesthetics and emotions (e.g. Denzin, 1997).

Narratives are both personal and generic. They reflect shared cultural conventions (Atkinson, 2005) and are enacted in accordance with structured conventions of narrative and performative genres (Denzin, 1989). Narrators do not, therefore, discover the rules of narrative for themselves but employ genres suited to their aims, whether or not they are not cognizant of the narrative frames they are using (Maybin, 2006). Narrative speech is not pure; interviewees typically intersperse narrative in the strict sense with argumentation and evaluation which are recounted from present time perspectives and with both hindsight and anticipation of the future (Freeman, 2010). Some narratives, for example, ‘big stories’ such as life stories, provide holistic and processual accounts both through the concept of the life trajectory and the hermeneutic aspects of the life: ‘interpretative understandings of biographically unfolding subjectivity’ (Atkinson, 2005, 2009). Like stories in books and plays, they have plots composed of temporal and spatial features (Abbott, 1997). They are also created horizontally through the juxtaposition of the self with others (Edgar, 2009). They involve beginnings, middles and ends through which the audience and the teller make sense of the events, persons and places that constitute the plot. Other stories are ‘small stories’, told in passing as ‘narratives-in-interaction’ (Bamberg, 2006, p. 146) in which it is possible to analyse how narrative is performed and accomplishes particular tasks, including identity construction (da Fina & Georgakopolou, 2012; Phoenix, 2013).

Narratives are mediated through memory and Antze (1996) observed that memories are monuments that we visit, but they are also ‘ruins’ that are subject to restoration. They are retrospective accounts of decisions, actions and events, often relating to distant periods of the life course and located in particular contexts, situations, relationships, emotions and moral judgments pertaining to these. While people’s recall of the past falters, their evaluations are made with reference to present time frames, making it impossible for the raconteur to stand outside the present when considering the past (Brannen, 2013; Brannen et al., 2004, p. 84).

Methodologically, it is necessary to take account of the fact that narratives, however, seductive, are, like all accounts, partial. They are shaped by the researcher’s and the interviewee’s interests and by the research encounter itself. Thus, rather than treating them unproblematically as ‘true’ records or as narrators’ authentic voices, narrative analysis demands the same methodological rigour and ‘methodological scepticism as we would apply to any other acts and social forms’ (Atkinson, 1997, p. 341). For example, just as what people choose to narrate is often telling, so too are the researchers’ choices in selecting a particular case or interview extract for analysis.

Study one: family language brokering practices

The first study that we propose to discuss, led by Ann Phoenix, was part of a larger study of adult retrospective narratives of their ‘non-normative’ childhood experiences.² It consists of three projects: (i) serial migrants who came from the Caribbean to rejoin their parents in the UK (ii) members of visibly ethnically different households (iii) childhood language brokers, who interpreted and/or translated for their parents in childhood. This paper focuses on the third project. Forty adult language brokers (27 women and 13 men) were interviewed individually in five countries and three group discussions were held in the UK and USA. The participants were asked to look back on their childhoods and tell the stories of their language brokering in any way they chose. The aim was to understand the ways in which they re-conceptualised their experiences over time and the impact of those experiences on their identities. The interviews were first analysed thematically. Since narrative analysis is time-intensive, only some were analysed narratively, and on the basis of theoretical sampling of the themes they evoked. The initial invitation to tell their stories frequently elicited mention of the themes elaborated in the rest of the interview and constituted the rationale for conducting narrative analysis on the first part of the interview.

In order to analyse the place of childhood language brokering in family practices from the narratives of adults looking back, it is important to see how the adults’ accounts relate to practice theory and theories of family practices as well as to narrative analysis. Unlike dishwashing or showering practices (Martens, 2012; Shove et al., 2012), there are no normative guidelines for language brokering as a practice. How it is done, who does it, for whom, what is interpreted (orally) and translated (from writing), for how long all have to be established in each family. The two examples below have been chosen because the narrative accounts present different ways of dealing with language brokering as a family practice and, hence, their analyses draw on different aspects of practice theory and theories of family practices as well as narrative analysis (as in the discussion above). The following extract is taken from the opening of an interview, conducted via Skype, with Fong, a woman in her mid twenties, who is of Chinese origin and lives in California.

… And I remember at that time it was … when my mother was trying to petition her family, her two sisters and her brother over from China. So you know, … And you know, the (.) the entire immigration process is (.) is (.) takes a long, long time, I think I remember maybe it took ten years, the entire thing, … so I remember im- you know, translating immigration paperwork, calling immigration several times, um and (.) and being put on hold (laughs) several times for long periods of time. But I think having them arrive you know, you’re seeing the fruits of your labour, I think it’s (.) it’s kind of rewarding in that sense um but you know, didn’t see it for a long time … that was a big project um I think some of the smaller projects were just you know, writing cheques and things like that. Um and as a matter of fact, I think when my brother and I both went to [university] um she (.) my parents were you know, at home (.) without you know, anyone to language broker for them and that (.) that year was actually very difficult because um they didn’t (.) they don’t know how to use email quite … I think that’s the reason why I (.) … I chose to move back down. Um just so that someone was there uh but do writing cheques, I also translate for (.) for the relatives now that they’re here … I do more for the relatives now than I do for my parents …

One of the points on which narrative scholars agree is that narrative is ‘defined … by sequences with a specific order’ (Andrews et al., 2013, p. 13). As Riessman (2008, p. 3) explains, this connecting of events into a sequence is not accidental, but is accomplished because it is ‘consequential for later action and for the meanings that the speaker wants listeners to take away from the story’. In narrative analysis, then, one of the key issues to which researchers need to pay attention is how participants sequence their accounts and the meanings that particular sequences allow and produce. At the start of her interview, Fong situates her account chronologically, explaining that she cannot remember doing very much language brokering until middle school. She then explains the process of how she came to do a lot of language brokering until she went to university and that, after her brother also moved away to university, she transferred to a local university in order to help her parents by language brokering.

The relevance of this narrative temporal sequence for family practices is illuminated by the bringing together of narrative analysis and practice theory. As Shove et al. (2012) suggest, practices do not emerge simply as a matter of individual choice, but have to compete with other practices in order to recruit practitioners who have finite amounts of time and space. In Fong’s case, studying competes with her language brokering practices but, unlike her brother, she eventually negotiates a way both to study and to do language brokering because she considered language brokering her responsibility and something that she wanted to do, even though she also makes it clear that it is work and can be onerous.

While Fong’s narrative suggests that it is routinely expected that she (and her brother) will do language brokering for her parents and other relatives, the nature of what is interpreted or translated shifts over time and according to context, family needs and Fong’s and her brother’s maturity and positioning. In keeping with practice theory, Fong’s narrative suggests the dynamism and finitude of a language brokering career and how habitual practices entail emotional engagement with ‘doing’ family life. Her account is evaluative, documenting how she currently feels, what she found difficult, what she enjoyed and improved at doing. Morgan’s (2011) theory of family practices allows a further analytic layer here in that Fong’s chronology of language brokering, her delimiting of the people for whom she does it and her reported feelings of responsibility bring family relations into the present and are performative, setting boundaries around her family.

Practice theory suggests that practices are not unitary but consist of multiple elements and bundles of activities that shift over time, many of which can belong to other family practices (Shove et al., 2012). In order to understand the practice of childhood language brokering, we therefore have to consider not just its sequencing and trajectory, but the activities that constitute it. Fong establishes that childhood language brokering is closely linked with other practices such as negotiating with immigration services. The bundle of activities Fong identifies as constituting language brokering consists of writing cheques for financial transactions, translating paperwork, answering mail, making and taking phone calls and interpreting for family members other than her parents. As Shove et al. (2012) suggest practices form clusters that compete and/or support each other. It is clear from the first few minutes of her interview that Fong’s narrative constructs her language brokering as supporting various family members’ negotiation of their immigration status and so was of crucial economic and familial importance. Fong’s negotiation with immigration services also served to delimit who belonged in the family in terms of who received and was entitled to such extended (‘maybe … ten years’) language brokering services (c.f. Morgan, 2011).

The second example is taken from a written and emailed account produced in response to a request to explain experiences of childhood language brokering. Husniyah, a young woman in her mid twenties, was born in the Middle East and lives in Sweden. Her language brokering was between Arabic and Swedish.

When we were newly arrived [aged eight years] and none of us spoke Swedish, translators usually addressed my mother when translating, even if the topic was about school and us kids. I didn’t need to take responsibility and my mom handled the information. After a while I had to tell my mom what had happened in school and what the teachers told us to prepare for the next day. This was my first encounter with translating for my parents and I found it quite amusing. I was also rather proud of myself for knowing things that my parents didn’t. At the same time, I didn’t want people to know that my parents didn’t understand Swedish or that their language skill was poor. To me, it was kind of a family-secret that I usually enjoyed but did not want to reveal to others … So whenever I got a letter that I didn’t understand I was reminded about my lack of skill in the Swedish language … after a while I understood that at my age (about 11–12 years) I wasn’t supposed to know anything about taxes, company rules, and so on … I remember feeling that my family and I were marginalized because we lived in a society where we didn’t know all the rules or all the words … I felt bad for being able to read better than my parents but still not being good enough … My father’s frustration with the Swedish language affected me in a negative way. I felt bad for him, and for myself, when I wasn’t able to help him, and when I did help him he didn’t show whether he was impressed or not … After refusing to translate a few times he stopped asking me and I was actually relieved.

Perhaps, not surprisingly, since it is a written account, chronology and cause and effect are very much part of Husniyah’s narrative. An analysis of the place of sequence in her narrative shows that she narrates how (in terms of practice theory), she came to be recruited into the practice of childhood language brokering and how she came to stop the practice.³ She thus documents her career as a family language broker and her emotional evaluation of it. The analysis of temporality is central to Husniyah’s written accounting that it is possible to analyse how language brokering as an everyday practice emerged, burgeoned and faded for her over a period of six years. In addition, the emotional changes she identifies over time inform the analysis of the difference between Husniyah’s and Fong’s language brokering practices in that Husniyah explains that her ability to interpret and her tolerance of being chided when doing it both changed over time. Over the same period, her father and mother changed in their orientation to language brokering, becoming more impatient and demanding, which she disliked. While Husniyah calls attention to amusement, frustration and relief, her narrative allows the analyst to infer further emotions, such as embarrassment (and hence wanting to keep language brokering secret), regret and dejection as well as anger, vituperation and obduracy (in response to her father’s criticism of her language brokering). In contrast, Fong indicates that, long projects of language brokering (e.g. in relation to immigration) sometimes brought her satisfaction, rather than frustration.

The evoking of emotion cannot demonstrate when, or whether, these episodes happened as Husniyah describes them. However, the allusion to various emotions in an account produced through free association, as in the Free Association Narrative Interview discussed earlier (Hollway & Jefferson, 2012), signals the emotional tone and, in so doing, gives an indication of the nature of language brokering in Husniyah’s childhood. In this regard, the analysis of emotions in narrative accounts arguably allows better purchase on, and more insights into the meanings evoked by everyday practices such as language brokering, than do observational methods.

While Husniyah’s narrative suggests that there were several years when the practice of language brokering was habitual for her, an application of Morgan’s (2011) theory of family practices, helps to underline the importance of analysing how family practices performatively serve an inclusionary purpose. For Husniyah, as for many in the study, language brokering was reported to be private family business; relationships between family members that produced emotions of both pleasure and inadequacy. In Brannen’s study, research participants were invited to bring photographs to the interview and delimited and displayed family. In contradistinction, however, language brokering set boundaries around the family without being a site of family display (although it was often a site where family practices were made temporarily visible).

When Husniyah’s account is analysed alongside those of other language brokers, it becomes possible to see that there are potentially important absences. Some language brokering narratives, for example, demonstrate how the career of language brokering loses recruits when parents become sufficiently skilled in the local language to be able to manage many situations by themselves. Husniyah does not say, but it seems possible, that her father had become sufficiently good at doing his own translation not to insist on her contribution or for her to feel that she had to continue to help (as Fong did, even though she suggests that she had to do less for her parents over time). This raises the methodological point that researchers are given only as much contextual background in narrative accounts as a participant chooses to report or remembers. Husniyah’s account gives less indication than does Fong’s of how practices external to the household impact on her language brokering. The analysis of the particularity of Husniyah’s narrative is, therefore, richer when brought into intertextual engagement with the narratives of other participants so that it is set in sociohistorical context (Shove et al., 2012).

Taking Fong’s and Husniyah’s narratives together, and analysing them alongside other narratives from the language brokering study, narrative analysis gives insights into the particular bundle of activities that form language brokering and constitutes a family practice. While narrative accounts necessarily raise issues about the accuracy of memories, we know from other studies that examine what participants say they do in relation to what was observed, that there can be crucial differences in accounts and understandings of practices and that observation cannot fully capture meaning (see Martens, 2012). Nonetheless, in this study, we argue that narrative accounts of language brokering, brought together with family practice theory, give a flavour of the boundaries set around particular families, as well as the ‘doing’ of family relations and the emotional tone of particular practices that, while freely discussed in interviews, are displayed to society often reluctantly and incidentally. Applying practice theory to narrative analysis of language brokering also gives insight into the temporality of the bundles of activities that constitute the practice of language brokering and so how they emerge, burgeon and lose recruits within each particular family, rather than (as is more usual in practice theory) for society in general.

Family photographs and narrative: prompts to reflecting on intergenerational relations

The second study, led by Julia Brannen,⁴ examined changes in fatherhood and men’s understandings of father–son relationships (Brannen, 2012, Brannen, Mooney, Wigfall, & Parutis, 2013). Using a biographical narrative approach, biologically related grandfathers, fathers and grandsons of white British origin, Irish origin and recent Polish origin were interviewed (30 chains). As noted earlier, a biographic-interpretive interview was carried out, followed by a semi-structured interview. In addition, as another lens on father–son relations, men were invited to bring to the interview a couple of photographs of themselves with their fathers and with their sons, an exercise that needs to be considered in the context of the overall narrative method.

Photographs act as prompts to memory in which people ‘make meaning with and from pictures’ (Chalfen, 1998, p. 229). The idea was to make visible those memories spoken about, stimulating further reflection and enabling the past to be retrieved in the present (Harper, 2002). The photographs were also used to help interviewees ‘break the frame’ of the normative and habitual aspects of family relations.

Forty of the 60 fathers and grandfathers brought two or more photographs to the interviews. Some forgot, others said they could not find any photographs, and some probably chose not to bring any. In a few cases, the interviewer failed to mention the photographs or judged that the time available for the interview would preclude this activity. The absence of photographs seem to reflect in some cases fraught or not close relations while in others they reflected the context; new migrants said their photographs were either packed away or had been left behind. In some cases, the only photographs available were for identity or passport purposes.

Visual images, like verbal accounts, are selectively organised for replay (Chalfen, 1998). In contrast to the ways in which many narrative interviews in the study unfolded, as lengthy monologues, the photographs generated a dialogue between interviewer and interviewee as the interviewer sought clarification from the interviewee about the identities of the persons in the photographs, their ages at the time, where and when the photographs were taken and by whom. In order to interpret this talk about the photographs the interviewer’s field notes became crucial, not only concerning the researcher’s interpretations of the photographs but also substantively as to what the photographs displayed.

Talk around photographs had the potential to generate different accounts from those produced elsewhere in the interviews. As a methodological practice, photo-elicitation was significant in highlighting intergenerational family practices in several ways.

Photographs as displaying family

Family photographs idealise family life and typically portray and ‘display’ (Finch, 2007) how people would like to see themselves and their families and how they would like the world to see them. They also solemnise and commemorate ‘the high points of family life’ (Bourdieu, 1990, p. 19) and constitute ‘both an instrument and index of integration’ (Bourdieu, 1996, p. 46). Taking and showing family photographs reproduce family relations (Haldrup & Larsen, 2003 as cited in Hallman & Benbow, 2007); they reconstitute the distinctiveness of individual members and a particular family as much as sentimental attachments and obligations do (Morgan, 2011). In calling family occasions and persons to mind, family photographs can moreover override the experience of everyday family life as it was lived at the time; thus, a ‘happy moment’ may be lived but only in memory (Morley, 2000; Rose, 2001). They also promote a partial view of life as ‘a preferred version of life that will outlive us all’ (Chalfen, 1998, p. 230). Typically, the photographs chosen were of ‘happy families’ and family groups posed conventionally and smiling bravely at the camera (Morley, 2000). Family photographs rarely ran counter to the dominant happy family narrative or depicted its darker sides (Morley, 2000, p. 60). Some photographs showed men and their families at leisure enjoying celebrations or on holiday. They also underscored particular and often atypical fatherhood practices – the occasions when fathers were ‘present’ in family life. Some photographs showed fathers and sons engaged in active, male-oriented activities, for example football, camping or building a garden shed. Such masculine representations are idealised, suggesting how men wished to see and present themselves though they did not verbalise this. Polish grandparents produced photographs of baptisms, first communions and weddings as evidence of the importance of family rituals and traditions. Showing these photographs to the researcher also seemed particularly important to them in the context of their sons’ migration. They knew that the researcher had already visited their sons and grandchildren and perhaps wanted to make an emotional connection to them via the researcher.

The photographs of fathers and sons that the men were asked to bring displayed families as lineages, and thereby enabled the interviewees to maintain illusions of family continuity in the face of change (Chalfen, 1998; Smart, 2007). A few referred to the importance of continuing the family name and ‘staying together’. Indeed, in one family, all three male generations were photographed together just for the project, suggesting how the research itself reflected, or perhaps even transformed, family practices (Rose, 2010).

While taking photographs serves to constitute ‘the family’ and/or records scenes of family life, the use of photographs did not necessarily prompt narratives of family practices. For example, one grandfather produced a pile of photographs and local newspaper cuttings that confirmed a story about the rise and fall of the family business founded by his father that he had recounted in his life story. He made few comments about the photographs, their significance remaining tacit and embodied. Similarly, Irish fathers who produced photographs of holidays in Ireland did not comment on their possible significance as reflecting and transmitting Irish identities to the next generation but did talk about holidays in Ireland in their life stories. Displaying family relations via the photographs was, as it were, deemed sufficient.

Photographs and relationality

Family practices constitute families as broader constellations (Morgan, 2011). Some research participants interpreted the photographs relationally. Some men used photographs to suggest emotional identification with a parent, while for others the act of looking at the photographs was a powerful reminder of a lack of closeness.

In a few instances, photo-elicitation prompted responses that confirmed, elaborated and raised new questions about what was said elsewhere. In the following, a father’s reflections on a photo of himself as a child and another of himself as a father powerfully convey contrasting generational experiences and family practices that added to the overall picture. Michael, aged 40 at interview, produced a photo of himself aged three or four with his father who, as the interviewer wrote in her field note, was dressed formally in shirt and tie. Noting that his father had taken the photo, Michael said this was one of only two photographs taken with his father when he was young. He recalled what his father said at the time and commented on how looking at the photo gave him ‘physical’, embodied feelings in the interview,

‘Stand there (son) and look at the ball’ – and that’s exact- you know I can almost hear it now (interviewer laughs) slightly goose bumpy – ‘Stand there, don’t move, look at the ball’ (emphasis) Okay I’m looking at the ball, what do I do now? ‘Don’t move, don’t move.’ Okay, and then he’ll have dived in and looked at the ball too. (Interviewer laughs) Cos he’s not looking at me’.

Somewhat surprisingly, given that the photograph stimulated a multisensory memory (Pink, 2011), on closer inspection Michael suspected that the photo might be of his brother, ‘This is going to sound really odd but I am looking at it I’m not sure it’s me’. Michael then volunteered a revealing remark, ‘I just don’t really have any memories of those kinds of times with my dad. It doesn’t conjure up any emotion … it feels a bit alien’. In this instance, there is no sure method of learning definitively what Michael’s family practices were and Michael’s possession of a photo, ostensibly of himself, proves to be an unreliable means of gaining insights into them. In comparison with the narratives given in response to interview questions only, it is not, therefore, that visual methods necessarily give more insights into what people do, rather than what they say.

Yet, Michael’s discussion of his photographs did allow the analysis of what he considered to be his practices as a father, with his own children. Michael, who did not live with his children at the time of interview, turned to the more recent photo taken with his children on holiday. Here, he spoke about the importance of spending focused time with his children. What he said about parenting stands in sharp contrast to how he described his own experience of being parented, ‘So this was very much about you know being with the children – all three of them were there with me. It was just me and the three children’. Looking at the picture of him and his son in a small boat, he observed that his son was taking his turn to steer while he, Michael, was ‘probably quite intensely wanting to let him take hold of the responsibility. So um, you know he’s got hold of the tiller (Yes, he has) so he is controlling it. But my hand’s just above, just in case, I need to sort of step down and correct it, you know’. This reflection then led him to talk about normative parental practices of giving children boundaries and making them feel safe, something which he felt had been lacking in his own childhood.

In this instance, photo-elicitation highlighted the constructedness and historically located nature of fatherhood practices. That photographs are no more ‘accurate’ in reconstructing events than narratives is underlined by Michael not being sure who is in the picture with his father. This is partly because the visual is only fully interpretable through the verbal suggesting the need to consider the visual and the textual in tandem. Whether it was Michael or his brother in the photograph was largely irrelevant to the culturally recognisable practice of using photographs to ‘do family’. The practice is relational in that claims are made through the embodied images of Michael with his father and Michael with his children that Michael is the better father. The ‘family gaze’ (Morgan, 2011) here, focused on brief moments in family life, gives Michael the right closely to scrutinise filial relationships and allows the condensation of associated emotions with these for re-presentation to the researcher in narrative form. Once photographs are brought together with narratives, they produce intertextuality, tempting the researcher and reader to provide a linking story (Wiese, 2011). Because the photographs were taken an historical generation and a family generation apart, they also signal the way in which practices change over time (Shove et al., 2012) without Michael necessarily reflecting on this.

Photographs as connecting place and time; mitigating loss

Family photographs are ‘a portable kit of images that bears witness to [a family’s] connectedness’ (Sontag, 1978, p. 8) especially when they live far apart; ‘photographs bring near those far away’ (Rose, 2001, p. 12). Thus for those separated through migration looking at and showing family photographs became additionally significant.

A Polish grandfather whose son and grandchildren were in the UK brought out some passport size photographs of his grandchildren. The grandmother, who was present, noted, ‘They gave us these photographs then,’ that is when the family was leaving Poland. She then went on to describe how her eldest grandson had not wanted to leave. Then, later in the discussion, she mentioned how her son on one of his visits had brought them a present of a framed family photograph taken in a studio; ‘we miss them very much’ she commented. While photographs help interviewees to bring the presence of loved ones into the home (Rose, 2001) and perhaps to cope with their absence, looking at photographs also foregrounded the sense of loss experienced by those left behind – emotions not expressed elsewhere in the interviews. Looking at, and commenting on, photographs in the interview was thus metonymic of ‘doing family’, giving both substance and emotion to transnational family relations and serving to commemorate them (Morgan, 2011).

In all these ways, discussion of photographs constituted researcher/interviewee co-constructions with both parties negotiating ways of displaying, doing, and commemorating family. As long demonstrated, talk is performative (Potter & Wetherell, 1987). Not surprisingly then, while discussions of family photographs evoked past practices, for example migration, they also focussed on current feelings, showing how their performance is central to ‘doing family’ and establishing how it should be done, thereby making visible the distinction and gap between what people say (the normative) and what they do (the practice/ experience). Methodologically, photographs, as snapshots, in some instances prompted more narrative, suggesting affect and performativity not evident in talk alone. In other cases, this did not happen, suggesting either that silence was intentional or that photographs reflect taken for granted representations and everyday practices of family life. However, they cannot arbitrate in disputes about what people really do as opposed to what they say they do.

Conclusion

The purpose of bringing material from the two studies together was to examine the benefits of generating narrative accounts via different methods, a narrative interview and a photo-elicitation method in the context of a narrative interview, and how new theoretical questions can be asked of data collected for another purpose (i.e. the illumination of family practices). We have shown that narrative methods can be productive in providing a picture of how people ‘do’, ‘display’ and commemorate family practices over time (Morgan, 2011) and pointed to some of the challenges and complexities in analysing this. In this undertaking, we have also had to consider the thorny theoretical question of the relation between talk and action.

In the first study concerning retrospective narratives of childhood language brokering, we showed how the bundle of elements that constitutes childhood language brokering was closely linked with other practices (Reckwitz, 2002) and hence had significance for what children in such households did in the form of work and in assisting their parents; these activities may have had profound implications not only for the participants’ schooling but for their ways of ‘doing’ family in emotional terms. The narrative approach employed also enabled insight into how practices like language brokering change over time – in Shove’s terms how practitioners are recruited and repelled from the practice as the individual, familial and societal contexts change (Shove et al., 2012) and how practices have trajectories. The analyses also illuminated how narrative accounts show that the micro processes of everyday family practices of language brokering are also part of the macro social order as, for example in migration. The participants’ accounts delimited those interlinked in language brokering relations and displayed family obligations, entitlements, power relations and affective relations. Analysis of presences and absences from the accounts showed the utility of looking at affective aspects of what is talked about and how.

In the second study, we examined the use of family photographs. Photo-elicitation as visual prompts to the verbal initially produced highly occasioned reports of still life images. Several methodological points can be made here. The photo-elicitation exercise was highly interactive between researcher and researched, even though the researcher’s questions did not necessarily generate further narrative. Some participants chose photographs that ‘displayed’ family life in conventional and idealised ways, focussing on celebrations and family members at leisure. Often discussion of the photographs stopped short at the visually embodied images. Much of everyday family lives was, therefore, left implicit and not talked about in the interviews. Silences suggest that either interviewees considered what was shown spoke for itself or what the image elicited, while important to them, was not necessarily something they chose or felt able to talk about. Where the significance of a photo remained tacit, it was important to the analysis that researchers reported on, described and made inferences about the images in their field notes. Talk about photographs in some cases produced dimensions of lived experience that enhanced material produced elsewhere in the interviews; for example, the sense of families as lineages, and feelings of loss engendered by migration, both of which served to ‘display’ and ‘do’ families (Finch, 2007; Morgan, 2011). In a few cases, photo-elicitation increased the possibilities for analysis, by drawing out ambivalence and the emotional tone surrounding father–son relationships, as in the case of Michael.

This analysis further suggests how language and the visual are both performative and can be used in concert. Since family photographs in the fathers study depicted single instances of mundane everyday life or celebratory occasions, they cannot contribute much to debates about changes in fatherhood practices over the generations. Thus, while some family practices can be captured via visual methods, (see Martens, 2012 on washing up), intergenerational relations in particular can only be partially articulated or performed in narratives since they cover large tracts of time and the life course.

Both psychosocial and sociological lenses have been important in developing this paper. Whether the psychosocial is theorised as underpinned by the psychoanalytic (Baraitser, 2009; Hollway & Jefferson, 2012) or situated ‘affective practices’ that constitute subjects (Wetherell, 2012), the analysis of emotional tone, what is said and what appears to be left unsaid (Frosh, 2002) provides insights into practices and what participants and their families have done and habitually do. A sociological lens places importance on putting analyses into context, both in relation to how the talk is generated (or not) in the interview and the wider social context. These lenses, psychosocial and sociological, complement one another.

In a similar way, different methodological approaches to narrative can be integrated and narrative approaches need not be divorced from other methods, such as the visual, although no set of methods can ever be comprehensive. Each study adopted a different interview approach that influenced the data generated. In the first study, direct questions about language brokering produced accounts of routine, habitual practices that in some cases were described as lasting over long periods and interspersed with stories of particular episodes. In the second study, the introduction of photographs into a biographic-narrative interview produced variable results: supporting strongly normative images, views and claims about family practices, commemorating family life and displaying intergenerational relations. In a few cases, it added further flesh to already rich accounts in which ambivalent filial relations were re-enacted in the interview.

Neither study produced ‘objective’, comprehensive knowledge of the past and of family practices, nor would this have been possible using other approaches. Nonetheless, narrative approaches have much to offer in eliciting the hidden and taken for granted aspects of family life and family practices, for example through their emphasis upon temporality, context and fluidity, their attentiveness to memory and to performativity, and their sensitivity to emotional tone and the normative constructiveness of family life.

Notes on contributors

Ann Phoenix is professor at the Thomas Coram Research Unit, Institute of Education, University of London. Her research interests are psychosocial, including motherhood, social identities, young people, consumption, racialisation, gender, transnational families and intersectionality. Recent publications include: ‘Negotiating daughterhood and strangerhood: Retrospective accounts of serial migration’, Feminism & Psychology (2013); ‘Multiculturalism, identity and family placement’, Special issue of Adoption and Fostering (2012); Analysing narrative contexts, Doing narrative research (2013).

Julia Brannen is professor of Sociology of the Family at Thomas Coram Research Unit, Institute of Education, University of London. She has an international reputation for research on family life, work?life issues, intergenerational relations and expertise in mixed methods, biographical-narrative approaches and cross-national research. Her recent publications include: Life story talk: Some reflections on narrative in qualitative interviews, Sociological Research Online (2013); ‘Critical issues in designing mixed methods policy research’, American Behavioral Scientist (2012); Transitions to parenthood in Europe: A comparative life course perspective (2012); ‘Fatherhood in the context of migration: An intergenerational approach’, Zeitschrift fur Biographieforsschung, oral history und lebenverlaufsanalyses BIOS (2012); Work, family and organisations in transition: A European perspective (2009); Handbook of social research (2008).

Notes

1. ESRC number: RES-576-25-0053.

2. ESRC number: RES-051-27-0181-A.

3. Language brokering is central to the conduct of many family lives but continually loses recruits as children grow up. However, the loss of some practitioners does not spell the demise of the practice unlike, for example, the now defunct darning of socks (Shove et al., 2012).

4. ESRC number RES 062 23 1677.