Discourse analysis is concerned with the ways in which language constructs and mediates social and psychological realities. Discourse analysts foreground the constructive and performative properties of language, paying particular attention to the effects of our choice of words to express or describe something. Discourse analysis involves the careful examination of talk and texts in order to trace the ways in which discourses bring into being the objects and subjects of which they speak. Discourse analysis is based on the premise that the words we choose to speak about something, and the way in which they are spoken or written, shape the sense that can be made of the world and our experience of it. Discourse analysts are acutely conscious of the power of discourse, and they consider our social and experiential worlds to be the product of our discursive construction of them.
A discourse analytic approach challenges the idea that the accounts people provide of their thoughts, feelings and experiences are comparable to a mirror image of what is going on inside of them, in their hearts and minds. Such an approach is the intellectual product of what is often referred to as ‘the turn to language’, itself a consequence of a philosophical reappraisal of the role of language in human interaction and experience. This reappraisal involved considering the social effects of language, its action orientation and its constitutive power. Ludwig Wittgenstein (1953) and John Austin (1962) are perhaps the most well-known early proponents of this perspective on language. Wittgenstein's (1953) argument that the meaning of words is constituted by their function in particular ‘language games’ and Austin's (1962) assertion that speech is a form of action and that we ‘do things with words’ have been enormously influential within this context.
A discourse analytic approach to qualitative research adopts what Forrester (1996: 32) describes as the ‘language-dominant view of language’. According to this view, language and thought are inextricably bound up with one another because language produces ‘versions and visions of reality as codes and conventions [Page 342]embedded within particular cultural contexts’ (Forrester, 1996: 33). From this point of view, language is organized into discourses which are culturally specific and whose availability depends upon social, historical and cultural contexts. Discourses may be defined as ‘sets of statements that construct objects and an array of subject positions’ (Parker, 1994: 245). Discourses make available particular interpretative repertoires, which provide us with ‘a lexicon or register of terms and metaphors’, which can be ‘drawn upon to characterise and evaluate actions and events’ in particular ways (Potter and Wetherell, 1987: 138). Discourse analysis is concerned with understanding these processes of discursive construction and their social consequences.
A concern with the role of discourse can be identified across a wide range of disciplines within the social and human sciences (e.g. sociology, philosophy, cultural studies, linguistics, social psychology), and as a result a variety of approaches to the analysis of discourse have emerged. Although all of them emphasize the importance of processes of discursive construction in the organization and management of social life, they differ in the emphasis they place on different aspects and dimensions of these processes. For example, some discourse analysts are particularly concerned with the ways in which institutional discourses are implicated in the maintenance of power relations within a society, for example by obscuring or mystifying power inequalities (e.g. Wodak, 1996; Fairclough, 1995), thus taking a ‘critical’ approach to the analysis of discourse. Others are more interested in the micro-level processes associated with the use of discourse in relatively mundane, everyday conversations (e.g. Schegloff, 1968; Drew and Heritage, 1992) and their role in creating and maintaining the social worlds that speakers inhabit; this approach is referred to as ‘conversation analysis’ (see Toerien, Chapter 22, this volume). Another approach, perhaps best described as ‘socio-linguistic’, involves the close examination of the ‘language behaviour’ of different social groups in order to understand the differences between them and their implications for social (and power) relations (e.g. Labov, 1966; Tannen, 1990). Yet another way of studying the role of discourse involves the examination of cultural representations (see Winter, Chapter 17, this volume), for example in the media (see Hodgetts and Chamberlain, Chapter 26, this volume), and the ways in which they make available and thus perpetuate shared meanings. Here, the focus is upon ‘the production and circulation of meaning through language’ (Hall, 1997:1) and its role in the production of ‘culture’. There is also narrative analysis (see Esin et al., Chapter 14, this volume) which is concerned with the structure, content and function of the stories people tell about their experiences, in terms of both their social impact and their psychological effects (e.g. Riessman, 1993; Mishler, 1995). There are many ways in which researchers have addressed questions about the role of discourse in social life, and in this chapter we are going to examine in more detail only one of these. However, Wodak (1996: ch. 1) provides a helpful historical overview of a range of perspectives on the study of discourse.
The approach to discourse analysis adopted in this chapter is informed by a concern with the availability of discourses and interpretative repertoires to individuals when they speak about their experiences. As such, the approach adopted here can be located within the social psychological tradition of discourse analysis as it was developed within the UK over the last 25 years or so (e.g. Potter and Wetherell, 1987; Parker, 1992). The type of discourse analysis described in the remainder of this chapter seeks to generate insights about how [Page 343]speakers draw on available discursive resources in order to construct particular versions of their experiences. It is also interested in the action orientation of the deployment of discursive resources within a particular context; for example, we may want to understand what may be achieved, socially and/or interpersonally, by describing something in a particular way. Thus, the analytic method described here is influenced by a Foucauldian approach to discourse analysis as well as by ideas drawn from conversation analysis and ethnomethodology as they are incorporated into their analytic approach by discursive psychologists (see Willig, 2008a: chs 6 and 7, for accounts of Foucauldian discourse analysis and discursive psychology, respectively; see also Wetherell, 1998, for an account of how to integrate the two).
Approaching data from a discursive perspective means focusing on language. The purpose of a discursive analysis is to gain a better understanding of how the use of language (that is to say, the choice of words, grammatical constructions and various rhetorical strategies) is implicated in the construction of particular versions of events. Discourse analytic research is very much concerned with the effects of discourse, with what discourse can do and, as a result, discursive research is primarily interested in discourse itself rather than in the individuals who use it and whose speech or writing constitutes the data to be analysed. In other words, the research questions which drive discourse analytic research are about the (social, institutional, psychological) effects of discourse, and not about the thoughts and feelings within the individual speakers which may give rise to the words they utter. A discursive analysis always starts with discourse. Discourse analysts can go on to ask questions about how discourses may construct subjectivities, and this is a particular concern of Foucauldian versions of discourse analysis; however, here subjectivity is conceptualized as the product of internalized discursive constructions and positionings, never as an entity that pre-exits the use of discourse.
Any text constitutes suitable data for discourse analysis. Indeed, Parker advises that ‘we consider all tissues of meaning as texts’, including ‘speech, writing, non-verbal behaviour, Braille, Morse code, semaphore, runes, advertisements, fashion systems, stained glass, architecture, tarot cards and bus tickets’ (1992: 7). In Critical Textwork Parker presents discursive analyses of a wide range of ‘symbolic systems that are not usually thought of as textual’ (1999: 1) including material from visual media and physical settings such as cities and gardens. Hall (1997) reminds us that the signs and symbols, which constitute language as a representational system (see Winter, Chapter 17, this volume), include more than words; they can take the form of sounds (see Maeder, Chapter 29, this volume), images (see Banks, Chapter 27, this volume), musical notes or objects. This means that discourse analysis can be conducted on ‘texts’ in the widest sense. Some examples of such research can be found in Hall (1997) and Reavey (2011). Non-linguistic texts are particularly suitable for analyses, which are concerned with the production of cultures and social identities (e.g. Edwards, 2011; Gill, 2011). Versions of discourse analysis which are particularly concerned with how speakers deploy discursive resources within specific conversational contexts, however, require data which capture the to and fro of discursive engagement between speakers. Potter and Hepburn (2005) argue that ideally data for discursive analysis should consist of naturally occurring conversations, rather than written narrative accounts or semi-structured interviews (see Roulston, Chapter 20, this volume). Transcription conventions (see Kowal and O'Connell, Chapter 5, this volume) for preparing data for discourse analysis also vary depending on the approach taken. Foucauldian versions of discourse analysis, for example, require less detailed transcription of the various non-linguistic features of speech than does discursive research [Page 344]inspired by conversation analysis. The analysis presented in the second part of this chapter draws on both Foucauldian and discursive psychology strategies for analysis using an extract from a semi-structured interview. The nature and quality of the data are suitable for the analytic approach adopted as they provide evidence of a range of discursive resources deployed as well as their strategic deployment within the context of the research interview.
Once a suitable text for analysis has been obtained, discourse analysis proceeds by working through the text line by line. Although there are a number of step-by-step guides to discourse analysis (e.g. Parker, 1992; Kendall and Wickham, 1999; Langdridge, 2004; Willig, 2008a), it is important to bear in mind that discourse analysis is not so much a recipe as a perspective from which to approach a text. It is a perspective on language which allows the researcher to produce a particular kind of reading of a text, a reading which foregrounds the constructive and performative properties of language. As Potter and Wetherell put it:
There is no analytic method … there is a broad theoretical framework which focuses attention on the constructive and functional dimensions of discourse, coupled with the reader's skills in identifying significant patterns of consistency and variation. (1987: 169)
One way of generating a discursive reading is to approach the data with a set of questions in mind, and to interrogate each line of text as well as the text as a whole with the help of these questions. Helpful questions with which to approach a text include the following (see also Holt, 2011):
- What sorts of assumptions (about the world, about people) appear to underpin what is being said and how it is being said?
- Could what is being said have been said differently without fundamentally changing the meaning of what is being said? If so, how?
- What kind of discursive resources are being used to construct meaning here?
- What may be the potential consequences of the discourses that are used for those who are positioned by them, in terms of both their subjective experience and their ability to act in the world?
- How do speakers use the discursive resources that are available to them?
- What may be gained and what may be lost as a result of such deployments?
It is important to bear in mind that when analysing interview transcripts, a discursive analysis requires that as much attention is paid to the interviewer's contribution to the conversation as to the interviewee's. This is because the interviewer's questions and comments constitute the discursive context within which the interviewee's contributions are made and to which they will inevitably orient themselves.
The epistemological position associated with discourse analytic research is social constructionism (see Esin et al., Chapter 14, this volume). Here, the researcher adopts a relativist position whereby the data are of interest not because they inform the researcher about ‘how things really are’ (e.g. what people are really thinking or feeling, or what happened in a particular social context), but rather because they tell the researcher something about how people construct meaning around events using the discursive resources that are available to them. The researcher is not concerned with the truth value of what participants are telling him or her; rather, the aim of the research is to generate an understanding of what people are doing when they talk about something in a particular way. In other words, the aim of the research is to gain a better understanding of the social dimension of participants' meaning-making activities.
Here, the type of knowledge sought is not knowledge about the world itself or knowledge about how things are experienced by research participants, but rather knowledge about the process by which such ‘knowledge’ is constructed in the first place. Such an approach to research is based upon the [Page 345]assumption that all human experience is mediated by language and that all social and psychological phenomena are discursively constructed in one way or another. The discourse analytic researcher is interested in how socially available ways of talking about the phenomenon of interest are deployed and what the consequences of this may be (see Willig, 2012a, for a discussion of the epistemological bases of different qualitative approaches; also see Willig, Chapter 10, this volume).
Discourse analytic research focuses on the role of language in the construction of social and psychological phenomena. It is concerned with the effects of discourse rather than with human experience as such, and it constitutes a profoundly non-cognitive form of social psychology. Discourse analytic research has been criticized for privileging discourse over ‘the person’ and for failing to theorize subjectivity (e.g. Langdridge, 2004; Burr, 2002; Butt and Langdridge, 2003; Nightingale and Cromby, 1999) including our sense of self, intentionality, self-awareness and autobiographical memory. While discourse analysts have argued that speculation about mental entities is not relevant to discourse analytic research (e.g. Potter and Wetherell, 1987), it could be argued that the concern of discourse analysis with action orientation does beg the question of why it may be that particular individuals or groups of people pursue particular discursive objectives. In other words, while discourse analysis is very good at generating insights into how speakers deploy discursive resources and with what effects, it is not very good at telling us what motivates them to do so. It has been argued that ‘discursive psychology brackets, and yet relies upon, a notion of motivation or desire, which it is incapable of theorising’ (Willig, 2008a: 107). Another criticism of discourse analytic research concerns its claim that the analysis draws on nothing outside of the text itself. However, it could be argued that it is impossible to make sense of what is going on in a text without importing ideas and concepts from outside of it. For example, the very assumptions about the role and function of language, which underpin discourse analytic research, are themselves brought to the text from outside of it (see Willig, 2012b, for a more detailed discussion of these issues; also see Willig, Chapter 10, this volume).
From an ethical standpoint (see Mertens, Chapter 35, this volume), one could question the acceptability of analysing research participants' accounts through a discursive lens when their accounts were provided in good faith with the participants, assuming that the interviewer was genuinely interested in the nature of their experiences rather than in how they deployed discursive resources. It could be argued that for ethical reasons certain types of accounts, such as those which are concerned with suffering and distress and which were provided by participants who believed that the interviewer was genuinely interested in the experiential aspects of their account (rather than the discursive ones), should not be subjected to discursive analysis (see also Willig, 2004).
The Foucauldian approach to discourse analysis does attempt to address some of these limitations. Here, research participants are not simply seen as strategic users of discourse but rather as historical subjects who are themselves constructed through and positioned within discourse. From this point of view, discourse is directly implicated in the process by which ‘human beings are made subjects’ (Foucault, 1982: 208). The argument here is that the availability and uptake of subject positions in discourse give rise to different kinds of selves and to possibilities of subjective experience. However, there are still unanswered questions about the extent to which subjectivity can be theorized on the basis of discourse practices alone, and what may be the role of other factors and structures in this process.
The data from which this example is drawn were originally collected as part of a phenomenological study of the experience of taking part in extreme sport (Willig, 2005; 2008b). Extracts from one of the participants' transcribed interviews will be used here in order to illustrate what is involved in a discursive analysis (see Willig, 2012b, for an extended version of this analysis). In the interview extract the participant, whom we shall call Anna, is asked to describe an occasion on which she took part in a form of extreme sport. In her response to this question Anna produces a lengthy account of not one but two such occasions. In her account she describes an initial white-water rafting expedition which did not satisfy her desire for an ‘adrenaline rush’ and which was followed by a second trip which did deliver the desired ‘rush’. Although the data had not been collected specifically for the purpose of discursive analysis, it was felt that the nature and quality of the data were suitable for the discourse analytic approach described in this chapter. The data are rich in evidence of a range of discursive constructions of the participant's experience, and the research interview constituted enough of a conversation between the interviewer and the participant to elicit clear evidence of a variety of action orientations.
In order to protect the participant's identity, all identifying details have been either changed or removed from all quotations used.
The process of analysis of the data involved careful reading and rereading of the transcript. This reading was done with an awareness that the focus of the analysis was the text itself; this meant that this was not a reading ‘for gist’ but rather a reading which was concerned with the properties of the language used in the account. The questions listed on p. 344 were used to guide the initial encounter with the text.
The second phase of the analysis involved a line-by-line analysis of the transcript. Here, analytic notes were written in the margins of the text in order to capture systematically what was talked about (construction), how it was talked about (discursive strategies) and with what consequences (action orientation). These notes were then reviewed and after a process of cross-referencing and integration of analytic observations, a discursive reading of the extract was produced. In what follows two analytic observations are presented and discussed in order to provide some illustration of what is involved in a discursive analysis and what kinds of insights can be generated on the basis of it. We will be focusing on Anna's use of a discourse of addiction to frame her engagement with extreme sport and her deployment of dualistic constructions of self in her account.
In Anna's account of her engagement with extreme sport activities she positions herself as an active seeker of ‘thrill and excitement’, somebody who goes to great lengths in order to generate a sufficiently powerful ‘rush’. With reference to her decision to repeat the white-water rafting expedition in order to access a more intense experience, she says, ‘even though it had cost me a fortune, but I was gonna do it again’. In her narrative we can identify repeated constructions of herself as someone who is strongly drawn to experiences which provide her with an ‘adrenaline rush’; in fact, she explains that her attraction is so strong that her ability to resist this ‘urge’ is compromised. Having described the second white-water rafting trip as involving a complete loss of control over the raft and as being ‘probably the most terrifying thing I've ever done in my life’, Anna concludes that she ‘had to love’ the experience (‘I just loved it. I had to’). In her construction of herself as someone who is so powerfully attracted to courting danger that this disables any cautionary impulse within her, Anna uses a series of extreme case formulations (Pomerantz, 1986):
[Page 347]So I went down with pedals, steering ourselves down and it was probably the most terrifying thing I've ever done in my whole life (laughs) I actually thought I was going to die but we went down, we were completely out of control the whole way and one of the guys very nearly drowned (laughs) …
Extreme case formulations (italicized in the quote above) are formulations which take claims or evaluations to their extremes; in this way, they provide an effective warrant for the speaker's actions and thus legitimize them. In our case, Anna's claim (that she is irresistibly drawn to extreme sport) is supported by the extreme nature of the risks and dangers she exposes herself to and which she evokes very powerfully with the help of the extreme case formulations deployed in the quote. The compulsive dimension of Anna's relationship with extreme sport is affirmed at the end of the extract when she concludes, ‘I need to do these things. It's something I just have to do.’
In her construction of herself as irresistibly drawn to extreme sport and her frequent references to the impact of the extreme sport experience on her physiology (e.g. when she talks about the ‘adrenaline rush’, her ‘frayed nerves’, the ‘buzz’, the ‘huge amounts of energy’ generated, and the observation that ‘your heart's going a million miles per hour’), Anna invokes a discourse of addiction. ‘Addiction’ as an explanatory construct attributes the compulsion to engage in a behaviour to its powerfully rewarding physiological effects which in turn have a positive impact on mood thus leading to the desire to repeat the behaviour. Anna's use of a discourse of addiction is particularly evident in her description of the pleasurable aftereffects of an episode of extreme sport:
The feeling, the rush that you get is just so, it takes a good day to come down, at least for me it does. Even longer, actually, ‘cos I'm just thinking of when I first went on a big hike that lasted over a week and for about a month and a half after the walk, I was just bouncing off the walls. I just had such a great time. This made everything in my body was just going, I just loved it. My sister thought I was completely nuts when I got back because I just couldn't sit still. So when I finish things like this it's just (intake of breath) you know.
The terminology used by Anna in this extract (her references to ‘the rush’, ‘coming off it’ and ‘coming down’) is identical to that used to describe drug users' experiences. These commonalities confirm that Anna does indeed position herself within a discourse of drug addiction when she talks about her experiences of extreme sport.
In addition, Anna's incomplete sentence (‘It's just …’) followed by a sharp intake of breath emphasizes the embodied quality of her experience, constructing it as something which cannot be adequately captured in words. Again, here we can see parallels with the way in which drug addiction is constructed: that is, as something which is seen as irrational and overwhelming and which neither the addict nor those close to the addict can make sense of; Anna ‘just loves it’ while her sister thought ‘she was completely nuts’.
Positioning herself within a discourse of addiction allows Anna to offer a vivid and captivating account of her high-risk activities without having to justify her risk-taking in a way that appeals to rational considerations. Through drawing on a discourse of addiction Anna is able to talk openly about her extreme sport activities without having to take responsibility for choosing to engage in such dangerous pursuits, as her engagement with extreme sport is constructed as a compulsion rather than a choice. Within a context which invites accountability (after all, she is being interviewed about her extreme sport activities) the deployment of a discourse of addiction serves to disclaim responsibility for taking part in a potentially socially undesirable activity and to ward off potential criticism.
Anna's account of her engagement with extreme sport is shot through with dualistic constructions of the self. Here, the self is constructed as being composed of distinct and separate parts, which do not necessarily communicate with one another. For example, a dualistic construction of the self, which constructs the material body (with its powerful [Page 348]urges and desires) as being in conflict with the moral self (which attempts to overrule the material body), is part and parcel of a discourse of addiction. Another way of constructing the self dualistically involves invoking conscious and unconscious parts of the self which do not have direct access to one another. Such a construction emerges when Anna is asked to reflect on her motives for taking part in extreme sport. In her reply she takes up the position of an outside observer of herself. She constructs a dualism between herself as the observer and herself as the object of her observations. In this way, she disavows an insider perspective and finds herself speculating about her motives in much the same way that somebody else may speculate. For example, in the following quote Anna disclaims any knowledge of why she does what she does:
Actually, it's funny when you think about it ‘cos I don't know what makes me do that. When I went bungee-jumping and I was standing there, I was the first person to go and I didn't see anyone jump before me and I didn't know, I hadn't even seen it on TV, I'd never seen anyone bungee-jump before. And standing on the edge of this platform and the guy just said, I'm gonna count down from 10 and then you just jump, alright? And then he counted down and I just jumped and I can't tell you what made me jump. I don't know what went through my head. It was sort of not an option. I was there, I made myself get to the edge of the platform and I just was gone …
Here, a dualistic construction of the self allows Anna to disclaim knowledge of her motives. Something other than her conscious self (‘I don't know what made me do that’; ‘I can't tell you what made me jump’; ‘I don't know what went through my head’) is responsible for her decision to jump and this means that Anna is unable to account for her actions. A little later, she goes on to develop a hypothesis about her motives for giving up sky-diving:
Yeah, but I'm just thinking, it's interesting ‘cos I wonder if the events, the sports that I've done I've always got that faith in somebody else? I'm just wondering, maybe when I stopped doing things like my sky-diving, when it's just me. I have to rely on myself. I don't have that much confidence in myself to keep on going. You know, when you get to a stage, ‘cos when I started sky-diving I was doing (partnered jumps) … so I was completely relying on somebody else to hook me up to the plane and everything. But when it got to the stage where I could jump on my own, I just stopped going. I mean there were a lot of reasons for that but I just wonder …
Again, although she identifies a possible motive (‘I don't have that much confidence in myself to keep on going’), Anna presents this as a hypothesis which she has arrived at through observation and reflection rather than as an insight based on direct self-knowledge (‘It's interesting ‘cos I wonder …’; ‘I'm just wondering, maybe ‘I mean there were a lot of reasons for that but I just wonder …’). Anna constructs her insights as the product of a dualistic scrutiny of herself as an object of interest, thus creating a gap between herself as a conscious agent and her motivations for taking part in extreme sport activities. Again, as was the case in relation to the deployment of a discourse of addiction, Anna's use of dualistic discourse allows her to distance herself from her engagement with extreme sport and to disclaim responsibility for her actions and their consequences.
A discursive analysis of Anna's interview extract has generated some insights into her use of discursive resources (such as the discourse of addiction and dualistic constructions of self) and their implication in the construction of a particular version of her engagement with extreme sport. The question of what Anna was doing when she constructed her experience of extreme sport in the way she did was also addressed. It was suggested that by positioning herself within a discourse of addiction and by drawing on dualistic constructions of self, Anna was able to disclaim responsibility for taking part in leisure activities which are considered to be extremely dangerous and which relatively [Page 349]few people engage in. However, in order to better understand the action orientation of Anna's deployment of discursive resources, we need to consider the discursive context within which Anna was positioned when she spoke about her relationship with extreme sport. Taking a closer look at the interviewer's style of questioning, we can see that the questions put to Anna constructed extreme sport as an experience which is composed of distinct elements (thoughts, feelings, sensations) which can be described and understood, and they positioned the interviewee as a self-aware, reflective subject who is willing and able to scrutinize herself. For example, Anna was invited to ‘describe one occasion when you took part in a form of extreme sport’ and a little later she was asked, ‘So if we look at it in terms of just before you did it and then during it and after … can you remember how you felt …?’ On occasion the style of questioning seems to imply that the purpose of the interview was to subject the interviewee's experiences to something akin to almost scientific scrutiny, for example when Anna is invited to dissect the feelings she has during a particular episode of extreme sport (‘So the feelings during it were like a mixture of excitement and fear? Or how would you describe it?’). This style of questioning places Anna at the centre of events; the questions construct Anna as an active agent in the narrative as they are concerned with what she did and what she felt (rather than, say, what happened to her or what other people did). Anna responded to this by providing an account which served to distance her from full responsibility for her actions. If we accept that the style of questioning, and indeed the entire interview situation itself, positioned Anna as having to account for her engagement with extreme sport (an activity which is not universally approved of as it carries serious risks to the life and health of the practitioner), it should come as no surprise that she deployed discursive devices and rhetorical strategies that would help her manage her own stake in the conversation. It follows that from a discursive point of view, our analysis of Anna's account tells us more about the type of situation she found herself in (i.e. an interview in which she was invited to account for her extreme sport practices) than about Anna herself or about the experience of extreme sport.
In addition, the analysis tells us something about the discursive resources which are culturally available to Anna and which can be used to construct ‘extreme sport’. For example, we discovered that the same discourse (a discourse of addiction) can be used to frame extreme sport and drug use. We also observed that a dualist discourse was invoked in the construction of pleasure (e.g. when Anna attributes her love of extreme sport to unconscious forces). Such observations resonate with a construction of the body as the primary site for pleasure and enjoyment, and the assumption that rational thought interferes with feeling good. This is reflected in the strict separation between ‘work’ and ‘leisure’ characteristic of late capitalist cultures where ‘leisure’ is increasingly associated with physical gratification (eating, drinking, the use of recreational drugs, sex) and where the purpose of leisure activities is to help the individual to ‘switch off’ (from work, from worry, from responsibility). It would be interesting to develop this analysis by examining wider social discourses and cultural practices in order to better understand how such a mind-body separation functions, how it arose historically and how it is maintained through various institutional practices.
Another avenue that could be pursued in a Foucauldian-style discourse analysis involves asking questions about how the discursive positionings adopted by Anna in her construction of her engagement with extreme sport may shape her actual experience of herself. By positioning herself within a discourse of addiction, Anna accepts that there are powerful forces at work within her over which she has little or no control. This could be experienced as disempowering; as a result it may be difficult for Anna to stop engaging in extreme sport should she wish to do so (see Eiser, 1984; Gillies and Willig, 1997). Similarly, positioning oneself within a [Page 350]dualist discourse may discourage attempts to integrate experiences which seem to originate in different parts of the self, leading to an increasingly fragmented sense of self.
In the worked example presented in this chapter, discourse analysis was used in order to better understand how a research participant constructs her experiences of extreme sport through language within the context of a research interview, how this positions her and what may be some of the consequences of these constructions and positionings. The analysis has generated insights into the interviewee's use of rhetorical devices and the discourses available to her which she could draw on to construct a particular version of the extreme sport experience. It was acknowledged that a discursive analysis does not allow us to answer questions about why Anna engages in extreme sport, the role of her personality within this or the nature of her motivations. Furthermore, it has little to say about the nature or meaning of the experience of extreme sport as such. Within the context of a discursive analysis questions about the inner world of research participants, their motivations, desires and intentions are suspended; instead the researcher is concerned with how discursive resources are used within particular contexts in order to construct meaning, and what happens as a result of that, interpersonally, socially and in some cases (e.g. Foucauldian analysis) also in terms of the production and availability of particular subjectivities. Some of the limitations and ethical challenges associated with such an exclusive focus on language have been discussed earlier in this chapter (see p. 345). In the final section we will identify some attempts at widening the focus of discourse analytic work, for example by advocating ‘binocularity’ in order to ‘thicken’ a discursive analysis (e.g. Frosh and Young, 2008) or by being more openly interpretative (e.g. Bell, 2011).
As seems to be the case with most qualitative research methods, discourse analysis continues to evolve into increasingly distinctive versions or varieties. Like grounded theory methodology, discourse analysis now encompasses a wide range of approaches with quite different priorities and emphases (see e.g. Glynos, et al., 2009; Wetherell et al., 2001). These reflect disciplinary differences as well as wider theoretical and also political commitments on the part of discourse analysts. For example, in a recent review of developments in discourse analysis in social psychology, Parker (2011) argues that while early discursive work in social psychology was critical of psychology as a discipline, seeking to deconstruct and critically appraise the processes which give rise to the ‘psy-complex’ itself (cf. Rose, 1985), the currently dominant version of ‘discursive psychology’ has become incorporated into the discipline as it offers an alternative (this time, ‘discursive’) account of what motivates and shapes human behaviour. Its preoccupation with the spoken word and its lack of interest in anything that might be happening at the level of emotion or cognition have also meant that this version of discursive psychology is associated with a relatively narrow, almost behaviouristic focus (see Corcoran, 2009, and Billig, 2012, for examples of critical appraisals of some of the assumptions underpinning discursive psychology).
In recent years discourse analysts have become increasingly concerned with the relationship between discourse and subjectivity. While earlier Foucauldian discourse analytic work had already engaged with questions about the ways in which available discourses may shape subjectivities, this work was still based on the assumption that subjectivity is very much a product of discursive structures and processes (e.g. Henriques et al., 1984). In other words, it was concerned with the consequences of discourse, its products and productions, and its power to make us who we are. [Page 351]More recently, questions have been asked about what underlying psychic structures and processes may lead speakers to invest in and commit themselves to the discourses and positionings they deploy when they talk about their experiences. Here, the psychological subject is seen as not (just) a product of discourse practices but as something which both shapes and is shaped by them (see e.g. Frosh, 2010, for a detailed account of such a perspective). Research informed by such a psychosocial approach combines a discursive analysis of participants' accounts with a further reading which attempts to make sense of their discursive actions by developing hypotheses about their deeper motivations and emotional investments (e.g. see Frosh and Young, 2008). Much of this work draws on theoretical resources from psychoanalysis in order to accomplish this.
Another recent development in the field of discourse studies has been to reconnect discourse analytic research with hermeneutics (see Wernet, Chapter 16, this volume), and to embrace and explicate much more openly the process of interpretation (see Willig, Chapter 10, this volume) that is involved in discourse analysis (e.g. Bell, 2011; see also Willig, 2012b). Here, the argument is that discourse analysis – that is to say, the textual analysis of data involving a line-by-line scrutiny of the linguistic, structural and functional characteristics of the discursive material – is really only one part of a wider project which could be referred to as ‘discourse interpretation’ (Bell, 2011: 520). This is because the analysis of discourse forms part of a wider reading of a text within its social and historical context, and this reading is informed by the researcher's own perspective, the assumptions they bring to the analysis, and their theoretical and personal orientation. The interaction between the researcher and the text generates a new understanding which is based upon the researcher's critical reflection upon both their own and the text's claims and assumptions. Bell argues that a purely descriptive approach to discursive work which produces only a ‘structural description of textual features’ (2011: 520) is ultimately irrelevant; however, it could be argued that a purely descriptive analysis is not only irrelevant but also impossible as even the identification of particular textual features requires a commitment to an interpretative lens that attributes a particular significance to such textual features (see Willig, 2012b: ch. 7).
To conclude, it is clear that different versions of discourse analysis are based on different conceptualizations – of human agency, of subjectivity and of the primacy of language (or otherwise), and this means that they address quite different research questions. It is, therefore, important that researchers who consider using a discursive approach as a method of data analysis are clear about the question(s) they are asking of their data and the kinds of insights they seek to gain from the analysis. This will enable them to select the type of discourse analysis which best suits the aims of their study.