Article 7 - Stine (TRYKKEKLAR) 181 The Disavowal of Difference: Accessing Class and Symbolic Boundary-Drawing in Interviews Stine Thidemann Faber Abstract: In this article I recount and reflect on methodological issues raised in my research about class representations and symbolic boundary drawing. The article discusses different aspects of accessing class by way of interviews (thereby privileging ‘talk’ as a source of data), the dilemmas involved in conducting the interviews (including the need for paying attention to classed power inherent in the research setting), and the challenges and associated problems of representing the experience of others, especially when researching across difference. Keywords: class distinctions, positioning, moral judgement, interview encounters. Please cite this article as: Faber, S. T. (2012). The Disavowal of Difference: Accessing Class and Symbolic Boundary Drawing in Interviews. Qualitative Studies, 3(2): 181-197. Introduction Often the emphasis in research is on the finished product (Bourdieu & Wacquant, 1992). However, the purpose of this article1 is to open the black box of the research process to present a reflexive account of methodological lessons learned from a study on social class in Denmark, highlighting the drawing of symbolic boundaries and examining the ‘commonsense’ categories used to describe and explain class-based differences. The kind of empirical research that I have done is full of dilemmas and challenges that are rarely confronted, and it raises questions of how to deal with difference and power in social research. In the article I thus discuss different aspects of accessing class by way of interviews and I deal with several strands of dilemmas: about the practicalities of accomplishing the interviews, about being aware of class in analysing the interview data, about being sensitive to issues of power and about the challenges of representing the experience of others when writing up the research. I agree with Richardson when she states that as researchers “we have an opportunity – perhaps even an ethical duty – to extend our reflexivity to the study of our writing practices” (Richardson, 1995: 191). Rather than hiding the dynamic, and at times ‘messy’, processes related to doing research, thereby concealing the hard effort that goes into creating the texts, we should, argues Richardson, reflect on and share with other researchers how we struggled, the effect that we had on the outcome and which factors influenced our interpretation of the data and the writing of the research – that is “how we came to construct the particular texts we did” (Richardson, 1995: 191). Studying Class and Class Representations in Denmark The article draws on qualitative data that originates from a research project based on twenty face- to-face in-depth interviews with women living in Aalborg [a city in Denmark], selected on account 1 The article is based on my PhD research. I am grateful for the valuable comments made by the anonymous reviewers and the editorial team on an earlier version of this article. 182 of their responses to a telephone survey conducted by the author of this article together with three colleagues (for elaboration see Faber, 2008, 2009, 2010; also see Faber, Rosenlund, Prieur & Skjøtt- Larsen, forthcoming).2 The women in the study were all strategically selected on the basis of different background information (including age, marital position and residence), and according to their answers to a number of questions about political stance and cultural preferences, inter alia, and they were roughly split in groups primarily on the basis of their level of education, work situation and overall household income. The group of women selected to represent the working class all had little or no formal education and worked in jobs with low status and low income; all had husbands with similar or less privileged working situations and all were living in deprived areas of the city. The group of women selected to represent the middle class all had high levels of formal education and worked in jobs with high status and high income; all had husbands with similar privileged working situations and all were living in privileged areas of the city. The women selected were between 29 and 52 years old, all were mothers, all were ethnically Danes, and with a single exception, all were working at the time of the interview. Within my study Bourdieu’s work provides a context of examining the impact of social class position combined with a feminist influenced perspective. My ambition was to look at what Reay (2005) calls ‘the psychic landscape of class’ - that is, to understand issues of social belonging, identities and subjectivities in their lived complexity, and the ways in which the women understand, negotiate and transform class-based meanings in their everyday lives. The study is closely connected to the growing stream of research rethinking class through a concern with other intersecting identity issues (including Skeggs, 1997; Reay, 1998; Bettie 2003). It is also connected to the growing stream of research focusing on lay normativity, especially morality, and ‘symbolic boundary-drawing’; a notion derived from Lamont (1992) to capture the idea that distinction is a process central to the formation of class differences, and one which relies on people drawing boundaries between themselves and other social groups whose interest are supposedly different from (and usually opposed to) their own. Following Bourdieu (1984), class is understood as being about people trying to show distinction. Class is viewed as a cultural space; “a space of ideas, values, goods, practices and embodied behaviors” (Liecthy, 2002: 16) that encompasses processes of inclusion and exclusion, and that is constantly tested, affirmed and renegotiated. Underlining that it is the process, not the product that constitutes class, Bourdieu’s work is said to account for the ways in which class is lived and accomplished. This point of view underlines that class is not only understood as ‘a structure’ or as ‘a category’ but as something that is constructed in human relationships (Thomson, 1963) - a view which of course also has implications when conducting research about class. Some of the challenges inherent in such research are to do with the difference between the researcher and the researched, and with dimensions that may reflect differential positions of power in the wider society or that may represent an ideological divide of some significance. 2 The interviews that form the core of this article were only one aspect of a multidimensional study that also included a large survey, interviews in addition to the ones I discuss in this article and register data. For more, see the COMPAS project (Studying Social Differentiation in Contemporary Societies: The Case of Aalborg) (www.socsci.aau.dk/compas). 183 Approaching the Field When preparing to do the interviews for the research project, I started thinking further about the power-dynamics in the research process. Although having conducted several interviews throughout my professional life, it was the first time I found myself wondering about my own physical appearance to this extent. I began to think of myself as classed in ways that I had not previously done, and it felt interesting and awkward. Suddenly, I felt acutely aware of how I looked, what I was wearing, how I styled my hair, which signals about my social belonging I gave through my appearance, inter alia. To mention just one example, I began to reflect on whether it was expedient to wear my Dior glasses during the interviews or to display the fact that I drove a Volvo. Subconsciously, at least, it seemed that I aspired to appear classless to my interviewees, as I was concerned about putting on display possessions that could signal wealth or status. In the end, I settled with what I felt was a class-neutral look (as if that even exists): wearing contact lenses, discreet make up, minimum jewelry, subdued clothes, and I abstained from parking my car right outside the house of the interviewees. Yet, I imagine that, just as I ‘checked them out’, my interviewees also ‘checked me out’ in terms of sameness as well as difference, and I did experience that the interviewees sometimes pointed out differences between us by referring to: my age, my level of education (discussed below) and to my car! In the interview where I had chosen to park right outside the house due to being late for the appointment, the working class woman I interviewed that day said: I can’t help looking at my fellow human beings. People are just different. It is not insignificant if you are… I don’t know why, but when I see you arriving in a huge Volvo I immediately think: ”Oh, it pays well to be a sociologist”. That is my first thought. We are all loaded with prejudices in some way or the other, right? (…) It is almost the only thing I know about you, and yet I reduce this to something that might be totally crazy. You see? [Laila, social and health care assistant, early forties] Despite my acute self-awareness and concerns about my ‘presentation of self’ (Goffman, 1959) I was left with the impression that I established good relationships with all interviewees, and in most interviews the atmosphere was open, comfortable and non-threatening, although for some interviews it took a while to achieve this (discussed below). When asking for interviews, I announced the research topic as being about the women’s own life histories, their views and their choice of lifestyle. Class was not explicitly mentioned, neither when recruiting participants nor when conducting the interviews, as I was aware that “class is not an innocent descriptive term but a loaded moral signifier” (Savage, Bagnall & Longhurst, 2001: 889). Instead, I asked the women questions about their social background, upbringing, schooling, family, kinship and friendship, parenting and mothering practices, working life, leisure activities and opinions. All of the interviewees were given the same themes during the interviews, but at the same time, like Rittenhofer (2010), I left it “almost entirely to the interviewees how they narrated these themes, and the way these themes and related situations were signified” (Rittenhofer, 2010: 4). While the repertoire of topics relating to class still seems broad, the interviewees have evidently put more emphasis on certain topics and less on others. For instance, the fact that all of the women were mothers proved to be important, as much of the work involved in being mothers and bringing up 184 children involves negotiating, repeating and reciting not only gendered and heterosexual, but also classed norms (Lareau,2003; Skilbrei, 2003; Gillies 2007). Studying Class by Way of Interviews Studying class by way of interviews and thereby privileging ‘talk’ as a source of data, actualizes the need for paying attention to the classed power inherent in both language and in the “telling of the self” (Skeggs, 2004). Addressing these issues remains an ongoing challenge, which is also why sociologists, and especially feminists, continually invite researchers to explore the interactions between the researcher and the researched. Yet, as Taylor (2005) notes, social class dynamics are not prominent in these debates, and more work is needed to address the ways in which class may be working to silence and/or to promote particular discourses and accounts (some researchers have interrogated this: see for instance Taylor, 2004 and McDermott, 2010). According to Krumer-Nevo: ...the power relations that determine every research are intensified in research settings where gaps in the social ladder between researcher and researched are evident. This is especially so when the researcher belongs to the middle-class while the researched belongs to the lower class or to a very low class (Krumer-Nevo, 2002: 305). When composing the research design I had thought about the point raised by Krumer-Nevo above, and initially I was concerned that the interview setting might limit the working class women in the telling of their stories as they can be said to lack “linguistic capital” (Bourdieu 1984). To some extent my concerns were well founded. From the very beginning I was confronted with classed differences in the behavior of the interviewees, ranging from astonishment, mixed with a little skepticism, that I would want to know details of their life biographies and everyday lives (mostly among the working class women), to not questioning that I would be interested in their personal histories and almost taking it for granted that they were worth consulting and listening to (mostly among the middle class women) (see also Taylor, 2005 and McDermott, 2010). I did also find that the working class women often answered my questions with very short sentences (some to an extent that made me wonder why they even consented to doing the interview in the first place), and often they did not follow up with ideas or elaborate on my questions in the same way as the middle class women did. Some of the working class women even appeared vigilant, almost on guard, and I remember that one woman especially sought to explain and justify herself to me continuously. After she had just told me about her brother, she said: “Well, none of us have an academic education like you, but I really do not think we would feel comfortable with it either. We feel good about the lives we have chosen” [Margit, kitchen assistant, late thirties]. However, it was rare and only sporadically that differences in terms of socio-economic and cultural status constituted this kind of context-setting, explicit positioning. Although class seemed to be present in the interviews, for the most part, it remained unspoken or perhaps even unspeakable. And yet, just as in ‘real’ life, class was being challenged, confirmed and enacted as the conversations between me and the interviewees, of course, did not occur “in a vacuum” (Archer, 2002), but were contextually produced, and classed positions and identification were discursively negotiated and struggled over throughout the interviews (Fine & Weis, 1996). In the end, I cannot say with any certainty in what way my class background and/or my (classed) appearance affected the interviews, just like I cannot say in what way my greeting, manner of speaking and body language were perceived. Though I am convinced that differences between the researcher and the 185 researched, such as class, gender, age, ‘race’ and sexuality enter the interview, it happens in ways that are not necessarily transparent or predictable, making it difficult, even impossible, to unscramble (see kennedy-acfoy and Pristed Nielsen, this issue, for a discussion of the unpredictability in researcher-researched interactions). Although more reluctant than the middle class women, the women from the working class were not passive or silent during the interviews; several spoke persuasively of their lives and their experiences of growing up, schooling, job opportunities, family and community. Some also became very emotional during the interview, underlining that seen from below, class appears “not just visible but almost tangible” (Ortner, 2003: 41) bringing into focus my responsibility, not only as listener, but also as conveyor of these powerful accounts. One woman, working as a kitchen assistant, said during the interview:”If I am in a company where I do not feel at home, I get a stomach ache because I feel inadequate. I just feel wrong. I don’t even reach their ankles”, she said, illustrating how objective social structures are inscribed in body and mind as subjective mental experiences (Bourdieu, 1984). Dealing with people placed higher than themselves in the social hierarchy can prompt defensiveness and shame, and make people feel inferior. It is, therefore, no wonder that some of the women from the working class seemed to be on guard, and were reluctant to speak during the interviews. Entering into a conversation with an academic surely risks activating (classed) discomfort, especially when it takes place in the private space of the home, which could potentially be viewed (by both interviewer and interviewee) as “another devalued signifier of class position” (Taylor, 2005: 496). Anyone who has tried it, knows that it is demanding to enter into a conversation with someone who does not really want to talk, and as a researcher it is not unusual to experience that some interviews ‘feel better’ than others (Skilbrei, 2003). However, form, not only content, is important. As Byrne argues “producing a ‘storried narrative” in the context of an interview is an uncertain process and there may be as much to be learned from those instances where a storied narrative is not produced as where the story of the self is easily told” (Byrne, 2003: 30). Or as Letherby phrases it: “Silences are as important as noise in research and the interpretation of silence is as important as the interpretation of what is being said” (Letherby, 2003: 109). Statements that might seem empty of content can represent important evasions or silences. Looking back on the transcripts of the interviews, I can see that I was, to some extent, too cautious about not pressing the interviewees to expand on answers when they reacted defensively, adopted short answers or where there were inconsistencies. This was so with both the interviews with women from the middle class and women from the working class. I found, to my cost, that talking about class was difficult for me too, and I remember feeling inhibited to ask what I perceived might be more difficult questions, and to prompt for clarity over inconsistencies or illogical statements. During the interviews I worried at times that there would be nothing to report but evasions and silences. However, close attention in the process of analysis revealed the degree to which the accounts were in fact explicitly classed (see also Byrne, 2006). Class Ambivalence and the Silencing of Class While interviewing the women, I found that they were reluctant to talk about class and their own position in the social structure, yet, at the same time, they seemed to have a quite an infallible sense of class. In other recent works on class cultures and self-identities the same tendency has also been observed. Thus Savage, Bagnall and Longhurst (2001) also find that their British informants were 186 reluctant to use class in personal terms even while using it to explain wider social conditions. They conceptualized this behavior as “class ambivalence” or “defensiveness”. As described, I also found that speaking about class differences (although mostly framed as social inequalities) tended to lead to, if not evasions or silences, then at least complex answers. Doing this kind of research raises questions about how to deal with topics like class, which “seems to exist and not exist, to be everywhere and nowhere, to have a kind of now-you-see-it, now-you-don't existence” (Ortner, 2003: 10). However, you could also argue, as do Payne and Crew, that the reason why the interviewees in my study - as well as in other similar studies - expressed their views about social class in a somewhat confused way, is simply because “they are being asked to handle a genuinely multifaceted concept at short notice” (Payne & Crew, 2005: 903). Maybe the women simply had as much difficulty dealing with the concept of class as researchers do within sociology where class, on one hand, is seen as “a thing of the past” (Reay, 2005), and on the other hand is considered to be increasingly relevant, although class consciousness may not be in evidence in contemporary societies.3 According to Payne and Crew, inarticulateness about complex concepts like class does not necessarily mean lack of salience. Within interviews, class, they argue, is often used extensively albeit not in a precise way and this does not reflect ambivalence, rather it reflects the fact that the interviewees have a different frame of reference than sociologists, and in fact they are often using “what they mean by class, in a consistent rather than ambivalent way” (Payne & Crew, 2005: 893, emphasis in original). Or maybe seeming ambivalent, defensive or reluctant is not as much about genuine confusion as it is about protection and resistance, as class and class identification are emotive issues that may make people feel uncomfortable (Taylor, 2005). Sayer writes: “Insofar as actors recognize the arbitrariness and injustice of natal class, and the ways in which it influences individuals’ lives – and it is hard for them not to – it can prompt guilt, shame, resentment and defensiveness” (Sayer, 2005: 201-202). Sayers’ description of the affect and emotions attached to class is not new. In earlier works, Senneth and Cobb (1972) pointed to the personal “injuries of class”, Willis (1977) to “class refusal’” and Bourdieu et al. (1999) to “positional suffering”. Euphemizing Class Indeed, talking about class differences can be awkward. Often the women, both from the middle class and from the working class, resorted to euphemisms: “It’s like they are on another frequency than we are” (about rich people), or “It’s a completely different world they live in” (about socially marginalized relatives). Bourdieu says about euphemisms, that they make it possible to say something all while one does not say it, and they “permit the naming of the unnameable” (Bourdieu, 1998: 98). 3 Within sociology the theoretical debates, whose chief proponents are Beck (1992) and Giddens (1991), the emphasis has increasingly been on individualization, reflexivity, detraditionalisation, plasticity and self-fashioning. In line with this, class has been marginalized as a mode of analysis, as class is seen as a matter of old, ascriptive ties and in the postindustrial societies classed identities are correspondingly held to have disappeared. However, others argue that class relations still have a causal effect on people’s lives, even when they are not articulating this in class terms (see for instance Skeggs, 1997, 2004; Savage, 2000; Sayer, 2005). 187 As described earlier, I selected the informants in my study on the basis of a larger survey, which was helpful, as I knew quite a lot about the interviewees in advance (such as family income, residence, and political stance.), and it gave me the possibility to ask about the answers previously given in the survey. When I asked one of the interviewees, a wealthy woman living in an expensive neighbourhood, why she declared that she belonged to the upper-class in the survey, I was taken by surprise by her answer, as with a few sentences, she completely withdrew from her original answer: Of course we have more money than ‘Jones, the painter’, or his like, right? But many people have a lot of money in their pockets nowadays, don’t they? Yes, they do. They just use a lot of money these days. In the past, this neighbourhood used to be upper- class (...) but I don’t feel it is like that nowadays. There are people … quite ordinary people who buy houses here now…. I don’t think one can generalise about it. Of course there are still people who earn more than others, but I wouldn’t walk around telling people that my husband is a lawyer, and that he earns well and such things. I wouldn’t dream of doing that. It doesn’t concern anybody. We have what we need, and then it doesn’t concern anybody how I spend my money, does it? They don’t pay for me. No, I don’t think … I don’t think there’s anything upper-class about us. (…) I really don’t. Our home it quite ordinary [Solveig, stay-at-home mother, early fifties] When talking about this during the interview I clearly felt her unease. This was a topic she considered to be private, which according to Ortner (2003) is not out of the ordinary. Not wanting to talk about class is often caused by a personal embarrassment about talking about money, about personal income, family resources or both. Yet, of course one also has to consider the fact that when filling out a survey you “talk to” a generalized other as opposed to the face-to-face interaction during an interview, in which we continuously adjust our answers to the person we are talking to. This might explain why the woman adjusted herself presentation; it could simply be the case that she chose to withdraw her previous answer so as to not alienate me, the interviewer. Yet, it also exposes an important methodological lesson: Had I not known from the survey the magnitude of the family’s resources (besides their house, the woman and her husband also owned two cars, a summer residence, and several investment and tenanted properties), data from the interview would not have revealed it. However, what I find most interesting is the way in which the woman tries to dissociate herself from the elite and the way she treats my questions as normative; that is, as if she was asked whether she considered herself superior or inferior to others (see also Sayer, 2005). My question about why she categorized herself as upper class seemed to be interpreted as an accusation of being conceited and it was obvious that she feels obliged to convey that she is not pretentious: she does not think she is better than most other people. This message is probably of particular importance in a society like the Danish one, where many are influenced by a particular Nordic ideology of equality, which tends to cause differences of class and other differences to be toned down as much as possible (Lien et al, 2001). The working class women also portrayed themselves as ordinary in the interviews, although for different reasons. Most of these women seemed to use the word ordinary to mean that, like everybody else, they worked for a living. At the same time, both implicitly and explicitly, they made accusations about middle class pretentiousness. Explicitly separating money and societal 188 status from personal values and integrity (see also Lamont, 1992, 2000), they accentuated the importance of being decent, wholesome and hardworking people who chose family over career. This was illustrated in moral statements such as: “I would also like a new bath room with a spa, but I believe that the money for such things is hard-earned money”, as one working class woman said at one point in the interview; at another point, she made a similar statement while talking about her children and the social intercourse within her family: “We don’t need to travel to Mallorca to be able to have a good time and relax together” [Heidi, gardener, mid-thirties]. For both the middle and the working class women, the presentation of oneself as ordinary seemed to indicate an aversion to be judged and identified through the markers of class (see Reay, 1996; Skeggs, 1997), but as Savage writes, the idea of ordinariness betrays class: “It is, after all, members of ‘other’ classes who might not be ordinary” (Savage, 2000: 117). Boundary-Drawing and Affiliation with Moral Communities In the interviews, explicit class identities were almost never voiced, and similarities with most other people were highlighted more often than differences. Still, more or less subtle dis/identifications were indeed voiced, often materialized in the ways in which the women talked about their families, friends, colleagues, neighbours, and other groups in society, and particularly through the lines of demarcation that the women used to present themselves in opposition to people they wished to be dissociated from (also see Jensen, this issue, for a discussion about the concept of dis-identification). Above all, you could say that in the interviews, class was revealed through boundaries being drawn between ‘them’ and ‘us’, and through a declared affiliation to moral communities. Little by little, stories about experiences and visions of differences surfaced. These stories sometimes dealt with gender, in particular with motherhood and the upbringing of children, sometimes they dealt with territoriality and perceptions of different kinds of neighbourhoods, and sometimes they dealt with ideas about consumption and everyday life. However, talking about social differences and diverse life conditions were often phrased as questions of personalities or presented as expressions of individualistic preferences, lifestyles or choices marking a shift in how class operates now-a-days; rather than being eradicated class identities are re-made, argues Savage (2000), through individualized emotional frames. During the interviews, the women from the middle class often presented stereotypical ideas related to the morality of the working class, whom they perceived as immobile, uncultivated, having a low work ethic, lacking energy and not raising their children in the right way. Likewise, the women from the working class presented stereotypical ideas about the middle class, who were perceived as selfish, focused on their careers, and lacking the ability to cultivate intimate personal relationships. In both cases, neither the working class nor the middle class was defined purely in terms of economic and cultural criteria, but as much on the basis of perceptions of disapproved behaviours and negative characteristics. Such feelings are probably only rarely experienced as having to do with class, but instead about some people being perceived as strange, snobbish, rude, narrow- minded or unsympathetic. However, class identity not only concerns who one is, but also who one is not: it encompasses practices, experiences and feelings, as well as distinctions and the drawing of boundaries. This understanding of class is important, as it implies that class is experienced as a lot of other things than what, in a strict sense, can be seen as class. 189 As noted earlier, looking up from below, class appears to be very visible indeed. For example, one of the women from the working class talked about money and income, although in direct terms, in contrast to the wealthy woman described earlier. She was unemployed at the time of the interview, and she says: “It’s hard to make ends meet. It doesn’t look good. Things are too expensive in general”. Yet, despite economic difficulties, being on welfare, she appears to translate her frustrations into satisfactions outside the work market: Because I’m unemployed, I have twice the energy to give to my son. That is fantastic. Being able to spend time with my child means a lot. I value that higher than going to work and earning money.” (...) I do a lot of stuff with my son compared to other parents. I wish all parents bothered to go to the playground and spend just one hour with the rest of us. [Joan, unemployed desk clerk, on social security, late twenties] Throughout the interview, this woman positions herself as a respectable, caring, attentive mother, at the same time distancing herself from middle class mothers, whom she thinks hand over their children to day care too early and pick them up too late, lacking time and energy. Views on work, children and family entail moral and emotional commitments, defined not only through gendered but also classed schemas. For this woman, motherhood is a loaded moral signifier, which plays a key role in shaping her worldviews and evaluations of self-worth. It is through motherhood and domestic responsibility that she establishes herself as worthy and distances herself from members of the middle class, who, she assumes, do not share her values (also see Skilbrei, 2003). Hidden in the Detail of the Dialogue More than anything else, the main differences between the women I interviewed were about moral matters, although the moral matters were often seen as related to economic or educational differences. Although the norms and values of the middle class women were almost always phrased in a hidden contra-distinction from working class women, and vice versa, the women I interviewed did not name it as being about class, nor did they see it as class dominance or class repression. They also refused to critique social inequalities and the social hierarchy within society, although they did seem to agree that it existed ‘out there’. Retrospectively, this has led me to ask myself one puzzling question: if they would not name the issues I have discussed in this article so far as being about class, should I? Immersed in a similar discussion, Fine and Weis ask if researchers risk neglecting the voices of the research participants that we wish to speak for: “Is this just a theoretical exercise in which we report narrations of denial? Or do we theorize over their voices, giving us little reason for collecting their stories?” (Fine & Weis, 1996: 263). My answer would be no. Though only very few articulated it as class, it was evident that all the women I interviewed, acknowledged, thought in terms of, or at least were familiar with, the existence of a social hierarchy. Hence, they talked about a ‘social ladder’ or about ‘the top’ (for instance, well-to-do people and the upper crust) and ‘the bottom’ of society (for instance, unemployed people or outcasts of society), and the women dissociated themselves from both groups – both the economically and culturally privileged, and the less so. The fact that all of the interviewees made statements about ‘others’, expressing that they were better or more valuable persons or vice versa, underlines distinctions that are about class. This corresponds to Skeggs’ claim that moral understandings of what matters to people includes different mechanisms for exchange that 190 generate “different forms of value that are both attached to people, and that people attach to themselves” (Skeggs, 2010: 31). Skeggs’ immense contribution to thinking about gendered class illuminates the cultural exchanges and practices of class identity and formation, and their moral dimensions. Accordingly, Skeggs argues: “Discerning how positioning, movement and exclusion are generated through these systems of inscription, exchange and value is central to understanding how differences (and inequalities) are produced, lived and read” (Skeggs, 2004: 4). In spite of defensiveness or reluctance, I do believe that I have accumulated substantial data to suggest that the women I interviewed did in fact talk a lot about class, although without directly mentioning class. Time and again, they expressed different distinctions, which can be related to class: they drew on categories that are ultimately categories of class, and they used expressions that implicitly connoted class. Although their understanding of class lay hidden in the detail of the interview dialogue, the issues they talked about were clearly about class. The Writing of the Research When moving from talking to the interviewees to analyzing the data, as a researcher you open another kind of dialogue, a dialogue which in some senses more resembles a monologue. At this point, the interviewee and the context that surrounded the interview are only in the picture as recollected memories, and strictly speaking, it is only the researcher’s loyalty that can prevent him/her in expounding the data freely. The audio file of the interview or the written text (if the interviews are transcribed – mine were) can be listened to/read over and over again and in the process of writing, the data can be cut up, interpreted and understood in different ways; the fact that the researcher now has a monopoly (the final say) over the interpretation of interview data gives her power. As Reed writes: “No matter how much shared experience there has been [within the interview], the sharing stops when the writing begins” (Reed, 2000: 68, referring to Parker, 1995). In order to capture what is at stake in the research process, I find the orchestra metaphor, presented by Nielsen and Rudberg (2006), very illuminating: The sample of research participants constitutes a symphony of many voices and they are allowed to be heard in different ways and with varying strength within the writing of the research. The researcher is not just sitting in the audience listening to the symphony. Rather, it is the researcher who first develops and then writes, produces and orchestrates the symphony. In other words, doing research means recognizing that the research does not necessarily reflect reality but is constitutive of it. When we ask question, listen to answers and interpret what we are being told, we are inevitably marked by our own subjective and material reality, and it will affect what we assume as possible answers, and what kind of knowledge we hear answers to be. As Patai describes it, the interview is a product of interaction and negotiation between the researcher and the researched; it reflects “a point of intersection between two subjectivities – theirs and mine, their cultural assumptions and mine, their memories and my questions, their sense of self and my own, their hesitations and my encouraging words or gestures (or sometimes vice versa) and much, much more” (Patai, 1988:146, cited in Bettie, 2003:22). Consequently, we need to reflect carefully on how we organize our questions and on how we listen, so as to create the best conditions for the co-construction of knowledge. I agree with Presser (2005) that data are not to be viewed as ‘objective truths’, but as mirroring the result of a specific set of social contexts, research processes and interpersonal dynamics (see Jensen, this volume). Accordingly, “the researcher’s goal is not to emancipate the authentic story of the narrator – none 191 exists – but rather to expose as much as she can of the relations that influence the construction of the story that is told” (Presser, 2005:2087). As researchers, we mediate and control the representation of our respondents’ voices. However, recognising the hierarchy within the process of conducting and writing the research does not mean that we cannot work on giving voice to the research participants rather that speaking for them. As a way of addressing this issue, I have tried in particular to reflect on the role of my own social belonging in the production of accounts and in my writings (see Faber, 2008). I have also attempted to use many quotes from the interviews in my writings, thereby displaying the complex ways through which the interviewees produced various discourses about class in relation to me as interviewer. The idea is, thus, that the quotes functions as an empirical test of the credibility of my analysis, so to speak. Like Bech-Jørgensen (1994) I believe that inserting quotes from interviews in the written texts is not only a question of illustrating the theoretical findings that we as researchers attain. It is also a question of letting the research participants ‘have a saying’ within the text. At the same time, the quotes makes it possible for the readers themselves to reflect on what is said, to elaborate on the research conclusions or even to reach different conclusions. Both the conversations that I initiated within the interviews and the social reality in which they were created are open for several interpretations from different perspectives. This is also why, as Gullestad (1996) argues, the interpretation is a process that actually neither can nor should take its ending. The Problem of Representing the Voices of Others When doing research we must continually work towards uncovering “the difference your difference make”, as Reay (1996: 443) phrases it. Our attention should not only be directed towards heightening transparency about positionality (of researcher and researched) and not secreting away the analytical choices, the representational practices and the personal investments, we make as researchers. In order to avoid the risk of reproducing specific discourses of class, thereby covering up or perpetuating structural inequalities, we must also continue to recognize the pitfalls surrounding issues of categorizations, conceptualisations and questioning (also see Jørgensen, this issue). As explained in this article the categories used by the interviewees when positioning themselves were not given, but “invented for the occasion” (what Strauss and Corbin [1998] call in vivo concepts) and handled in ways that gave the most positive presentations of themselves possible during the interview: they were ordinary and middle class, or whatever their social position actually was, had a better work ethic and better mothering practices, inter alia, than other social groups. Within sociological research you can find the same tendency among researchers potentially distorting the analyses with images, or at least euphemistic portrayals, of their own social group (also see Faber & Prieur, 2012). This behavior and the application of hidden class categories within research risk contributing to the veiling of class. Like when researchers write about the educational system and the labour market, where already the use of ‘high’ vs. ‘low’ convey an evaluation, as Vogt (2007) has also pointed out in a critique of a research tradition where scholars may, unreflexively, distinguish between ‘interesting’ and ‘not interesting’ jobs. Likewise, Skeggs (2004) has offered a useful illustration of a similar pitfall with examples from the current sociology of mobility, which tends to disguise who can and who cannot move, and also in Giddens’ sociology of the self with its silence on matters of class, gender and race. 192 The problem of implicit class biasing within sociology merits more thorough investigations. It also raises questions about the need to address the contours of the researcher’s social identity and his/her social belonging or positionality (in terms of gender, race, class, sexuality and other axes of social difference), and it raises questions such as: how can we use research “to produce knowledge across multiple divides (of power, geopolitical and institutional locations, axes of difference, etc.) in ways that do not reinscribe interests of the privileged?” (Nagar & Geiger, 2007). And, indeed, ruptures are possible. Reading the work of Lareau (2003) one cannot help but notice her conceptualisation of the parenting practices of the middle class, which Lareau labels “concerted cultivation”, capturing the way in which middle class parents are providing a structured life for their children constantly trying to guide them, while the working class parents attribute much of their child raising tactics to “the accomplishment of natural growth”. Lareau’s vocabulary is a rare example of a sociological taxonomy that does not give the most positive characteristics to the privileged group. There are no simple answers to the questions regarding representation. These issues are crucially important because our academic work can have ‘real effects’ in different ways and on different levels. On the one hand research can impact upon policy discourses, like Archer, Hutchings and Leathwood emphasize, thereby “effecting what/who are recognised (or not) as important, who are named as ‘problems’, who are constructed as deserving of resources, and so on” (Archer, Hutchings & Leathwood, 2001: 42). On the other hand, research can disrupt the research participants if they feel misunderstood, misrepresented or even subjected to symbolic violence. Whether majority group researchers should seek to represent the ‘others’ (those in the margins of society) has often been discussed within sociology. However, as Griffin argues: “researchers are always ‘speaking’ for Others. This is not something to be denied or avoided: it is a (potential) power and responsibility” (Griffin, 1996:189, cited in Archer, 2002). Back (2007) similarly writes that opening spaces for voices and stories that are otherwise silenced is a sociological responsibility, which is why researchers need to concentrate on both careful listening and critical scrutiny. Summing up In this article I have elaborated on some of the dilemmas and concerns which were raised in my research about class representations. The focus of the article has not been on the research project as a whole or its findings – at least not in a thorough way. Instead, the focus has been on methodological issues when accessing class by way of interviews: the process of gathering the interviews, the complexities involved, the challenges related to analyzing and writing the results, and the challenges of representing the experience of others, especially when researching across difference. Despite the fading of class consciousness and/or reluctance to speak about class, class still informed the experiences and understandings of my research participants – both the economically and culturally privileged and the less so. Yet, as described, I found that utterances about class come in many forms, almost always lurking beneath the surface, and are rarely made explicit. This elusiveness suggests that attention needs to be paid to the detail of the classed discourses produced in interviews, but at the same time, I would argue, it is a central finding about how classes are constructed in contemporary societies. 193 References Archer, L. (2002): It’s easier that you’re a girl and that you’re Asian: Interactions of ‘race’ and gender between researchers and participants, Feminist Review, vol. 72, pp. 108-132. Archer, L., M. Hutchings & C. Leathwood (2001): Engaging with Commonality and Difference: theoretical tensions in the analysis of working-class women’s educational discourses, International Studies in Sociology of Education, Vol. 11, No. 1, pp. 41-62. Back, L. (2007): The Art of Listening. Oxford: Berg Publishers. Bauman, Z. (2001): On Glocalization: Or globalization for some, localizations for some others, in Beilharz, P. (ed.) The Bauman Reader. London: Blackwell. Bech-Jørgensen, B. (1994): Når hver dag bliver hverdag. København: Akademisk Forlag. Beck, U. 1992 [1986]: Risk Society: Towards a New Modernity, Translated by Mark Ritter, London: Sage Publications. Bettie, J. (2003): Women without Class. Girls, Race and Identity. London: University of California Press. Bordo, S. (1990): Feminism, Postmodernism and Gender-Scepticism. In Nicholson, L. (1990) (ed.): Feminism/Postmodernism. New York: Routhledge. pp. 133-156. Bourdieu, P. & L. Wacquant (1992): An Invitation to Reflexive Sociology. Chicago: The University of Chicago Press. Bourdieu, P. (1984): Distinction. A Social Critique of the Judgement of Taste. London: Routledge. Bourdieu, P. (1998): Practical Reasons. Cambridge: Polity Press. Bourdieu, P. et al. (1999): The Weight of the World. Cambridge: Polity Press. Bryson, B. (1996): Anything but Heavy Metal: Symbolic Exclusion and Musical Dislikes. In American Sociological Review, 61(5), pp. 884-889. Byrne, B. (2003): Reciting the self. Narrative Representations of the self in qualitative interviews. In Feminist Theory, 2003, Vol. 4, No. 1. pp. 29-49. Byrne, B. (2006): White Lives. The Interplay of Race’, Class and Gender in Everyday Life, London: Routledge. Castells, M. (2000): The Rise of the Network Society. The Information Age: Economy, Society and Culture. (Vol. 1), Oxford: Blackwell Publishers. Cook, J. (2000): Culture, Class and Taste. In Munt, S. (2000) (ed.): Cultural Studies and the Working Class. Subject to change. London: Cassell. pp. 97-112. 194 Faber, S. T. (2008): På jagt efter klasse. Unpubliched Ph.D.-dissertation. Institut for Sociologi, Socialt Arbejde og Organisation, Aalborg Universitet. Faber, S. T. (2009): En sans for forskelle. Klassetilhørighed og symbolsk grænse-arbejde. In Arbejderhistorie. Tidsskrift for historie, kultur og politik, 2009-10, nr. 3/1, pp 75-91. Faber, S. T. (2010): Den subjektive oplevelse af klasse. In Dansk Sociologi, april 2010, nr. 1, årg. 21, pp 75-89. Faber, S. T. & A. Prieur (2012): Parler des classes dans une société presume égalitaire. Les Représentations des inégalitiés dans une ancienne ville ovriére danoise. In Actes sur la Classes, 191- 192, pp. 114-125. Faber, S. T., A. Prieur, L. Rosenlund & J. Skjøtt-Larsen (forthcoming): Det skjulte klassesamfund. Aarhus: Aarhus Universitetsforlag. Fine, M. & L. Weis (1996): Writing the ‘wrongs of fieldwork: Confronting Our Own Research/Writing Dilemmas in Urban Ethnographies, Qualitative Inquiry, Vo. 2, No. 3, pp 251-274. Florida, R. (2002): The Rise of the Creative Class… and How It’s Transforming Work, Leisure, Community & Everyday Life, Cambridge: Basic Books. Giddens, A. (1991): Modernity and Self-Identity: Self and Society in the Late Modern Age, Polity Press, Cambridge. Gillies, V. (2007): Marginalised Mothers. Exploring working-class experiences of parenting. New York: Routhledge. Goffman, E. (1959): The Presentation of Self in Everyday Life. New York: Anchor Books. Griffin, C. (1996): See whose face it wears: difference, otherness and power. Feminism & Psychology, Vol. 6, No. 2: pp.185-191. Gullestad, M. (1984): Kitchen Table Society: A case study of the family, life and friendships of young working-class mothers in urban Norway. New York: Columbia University Press. Krumer-Nevo, M. (2002): The Arena of Othering. A Life Story Study of Women Living in Poverty and Social Marginality”. In Qualitative Social work, vol. 1, no. 3., pp. 303-318. Lamont, M. (1992): Money, Morals and Manners. The Culture f the French and the American Upper- Middle-class. Chicago: The University of Chicago Press. Lamont, M. (2000): The Dignity of Working Men. Morality and the Boundaries of Race, Class and Immigration, New York: Russell Sage Foundation. 195 Lareau, A. (2003): Unequal Childhoods. Class, Race and Family Life. London: University of California Press. Lash, S. (1994): Reflexivity and its doubles: structure, aesthetics, community, in U. Beck, A. Giddens, and S. Lash (eds.) Reflexive Modernization. Cambridge: Polity. pp. 13-55. Letherby, G. (2002): Claims and Disclaimers: Knowledge, reflexivity and Representation in feminist Research. In Sociological research Online, vol. 6, no. 4. Liecthy, M. (2002): Suitably Modern: Making Middle-Class Culture in a new Consumer Society. New Jersey: Princeton University Press. Lien, M.; H. Lidén, and H. Vike (2001) (eds.): Likhetens Paradokser. Antropologiske undersøkelser i det moderne Norge, Oslo: Universitetsforlaget McDermott, E. (2010): I Wanted to be Totally True to Myself: Class and the Making of the Sexual Self. In Taylor, Yvette (eds.) (2010): Classed Intersections: Spaces, Selves, Knowledges. London: Ashgate Publishing. pp. 199-216. Nagar, R. & S. Geiger (2007): Reflexivity and Positionality in feminist Fieldwork Revisited. In Tickell, A., E. Sheppard, J. Peck & T. Barnes (2007): Politics and Practice in Economic Geography. London: SAGE. Pp. 267-278. Nielsen, H. B. & M. Rudberg (2006): Moderne jenter. Tre generasjoner på vei. Oslo: Universitetsforlaget. Ortner, S. B. (2003): New Jersey Dreaming. Capital, Culture and Class of ’58. London: Duke University Press. Parker, D. (1995): Through Different Eyes: The Cultural Identities of Chinese People in Britain. Aldershot: Avebury. Patai, D. (1988): Brazilian Women Speak: Contemporary Life Stories. New Brunswick: Rutgers University Press. Payne, G. & C. Crew (2005): Unpacking ‘Class Ambivalence’. Some Conceptual and Methodological Issues in Accessing Class Cultures. In Sociology, vol. 39, no. 5, 2005, p. 893-910. Presser, L. (2005): Negotiating Power and Narrative in Research: Implications for Feminist Methodology. Signs: Journal of Women in Culture and Society, vol. 30, no. 4: pp 2067-2090. Reay, D. (1996): Dealing with difficult differences: Reflexivity and social class in feminist research. Feminism & Psychology, vol. 6, no. 3, pp. 443-456 196 Reay, D. (1998): Class Works. Mothers’ Involvement in their children’s Primary Schooling, London: Routledge-Falmer. Reay, D. (2005): Beyond Consciousness? The Psychic Landscape of Social Class. In Sociology, vol. 39, no. 5, pp 911-928. Reed, K. (2000): Dealing with Difference: Researching Health Beliefs and Behaviors of Asian British Mothers. In Sociological Research Online, vol. 4. No. 4. Richardson, L. (1995): Writing-stories: Co-authoring ‘The Sea Monster’, a writing-story, Qualitative Inquiry, Vol. 1, No. 2, pp 189-203. Rittenhofer, I. (2010): Interview without a subject: The Russian doll question and cultural encounters, Journal of Intercultural Communication, issue 23, pp. 35-46. Savage, M. (2000): Class analysis and social transformation. Philadelphia: Open University Press. Savage, M., G. Bagnall & B. Longhurst (2001): Ordinary, Ambivalent and Defensive: Class Identities in the Nordwest of England. In Sociology, 2001, Vol. 53, No. 4., pp. 875-892. Sayer, A. (2005): The Moral Significance of Class. Cambridge: Cambridge University Press. Scharff, C. (2010): Silencing differences: the 'unspoken' dimensions of 'speaking for others. In R. Ryan-Flood and R. Gill (2010): Secrecy and silence in the research process: feminist reflections. London. pp. 1-26. Senneth, R. & J. Cobb (1977): The Hidden Injuries of Class. Cambridge: Cambridge University Press. Skeggs, B. (1997): Formations of class and gender: Becoming respectable. London: Sage Publications. Skeggs, B. (2004): Class, Self, Culture. London: Routledge. Skeggs, B. (2010): The Value of Relationships: Affective Scenes and Emotional Performance. In Fem Leg Stud, vol. 18, 2010, pp. 29-51. Skilbrei, M. (2003): Dette er jo bare et husmorjobb. Ufaglærte kvinner i arbeidslivet. Report 17/03. Oslo: NOVA. Strauss, A. and J. Corbin (1998): Basics of Qualitative Research, Techniques and Procedures for Developing Grounded Theory. London: Sage Publications. Taylor, Y. (2004): Hidden in the small ads. Researching working class lesbians. In Graduate Journal of Social Science, Vol. 1, Issue 2, p. 253-276. 197 Taylor, Y. (2005): Classed in a Classless Climate: Me and My Associates… In Feminism & Psychology, vo. 15, no. 4, pp. 491-500. Thompson, E. P. (1963): The making of the English Working Class. Harmondsworth: Penguin. Trondman, M. (1994): Bilden av en klassresa. Sexton arbetarklassbarn på väg till och i högskolan. Carlsson Bokförlag, Stockholm. Vogt, K. C. (2007): Forskeres språkbruk: dominerende perspektiver på utdanning og arbeid. Replikk - tidsskrift for human- og samfunnsvitenskap, no. 24, pp. 18-21. Wennerström, U. (2003): Den kvinnliga klassresan. Göteborg: Studies in Sociology, no. 19, Department of Sociology, Göteborg University. Willis, P. (1977): Learning to Labor: How Working-class Kids Get Working-class Jobs. Columbia: Columbia University Press. Author Stine Thidemann Faber is Associate Professor in the Dept. of Culture and Global Studies, Aalborg University, Denmark. Email: stf@cgs.aau.dk