Volume 10, No. 3, Art. 24 – September 2009
"This is for Life": A Discursive Analysis of the Dilemmas of Constructing Diagnostic Identities
Abstract: This paper takes a discourse analytic approach to the construction of identities formed through reception of a psychiatric diagnosis (I will refer to these as "diagnostic identities" throughout) as dilemmatic, and the subsequent negotiations of identities in light of that dilemma. More specifically, it is the diagnosis schizophrenia that is of interest, and how people who receive that diagnosis construct their identities. A key feature of receiving a schizophrenic diagnosis is the potential to see one's identity as under threat from the many negative, and predominantly stereotypical ideas, that persist regarding schizophrenia. Drawing on literature emerging from the field of service user research in mental health, the paper attempts to go beyond the boundaries of a psychiatric biomedical perspective of diagnosis, in order to illuminate how such classifications impact upon those who receive them. In this paper the discursive re-workings of individual diagnostic identities included strategies of resisting diagnosis, attempts to distance oneself from diagnosis, existentialising diagnosis, and recognising but resisting suggestions that people with diagnoses of schizophrenia are a social "risk".
Key words: schizophrenia diagnosis; dilemma; risk; discourse analysis; service users
Table of Contents
3. Interactive Kinds
4.3 Interview technique
4.4 Analytic coding
5.1 The problem of acceptance: Long term identities
5.2 Attending to notions of risk
This paper is concerned with analysing how people react to receiving a psychiatric diagnosis of schizophrenia, and construct identities in light of such a diagnosis. A key part of the identity of mental health service users1) can come through societal understandings and knowledge of their particular diagnosis (e.g. schizophrenia). Diagnosis is a central part of psychiatric practice as it enables a set of presenting experiences to be classified in a way that allows a program of care and treatment to be implemented. For the people who receive psychiatric diagnoses, the process of categorisation can become an important means through which they become socially visible. This process of being identified as someone with a psychiatric diagnosis means service users can become recipients of the effects of the labels they receive (BHUGRA, 2006; DAVIDSON, 2003; DINOS, LYONS & FINLAY, 2005). As such, diagnosis can have important consequences for identities (SADLER, 2005). 
Of all the different forms of diagnosis it is the categories of schizophrenia that have been shown to be amongst the most challenging to service users (KNIGHT, WYKES & HAYWARD, 2006; PINFOLD et al., 2003; SCHULZE & ANGERMEYER, 2002). Media reports involving schizophrenia very often link people diagnosed to notions of risk and violence (HARPER, 2002)—not only risk in relation to potentially being a danger to themselves, but also a threat to others, through a suggested propensity to commit violent crime. Links between schizophrenic diagnosis (in the form of psychosis) and violence have been repeatedly reported (e.g. ANGERMEYER & SCHULZE, 2001; DAY & PAGE, 1986; ROSE, 1998; SIEFF, 2003). This can result in those diagnosed with schizophrenia being perceived as posing a disproportionate risk to society through committing violent crime (LAURANCE, 2003). 
Analysing identity has become an important and prominent task for social psychology (DEAUX, 1993; TAJFEL, 1982; TERRY, HOGG & WHITE, 1999). Cognitive approaches have claimed that identity-formation is a reasonably stable process in which ascriptions are made regarding the categories within which we are placed. These become cognitively placed, and drivers of our post-identified actions. Discursive and conversational approaches have sought to elucidate the fluid contextual nature of identity (BROWN & LUNT, 2002; COUPLAND & GWYN, 2003; HORTON-SALWAY, 2001). Far from perceiving identity as a stable cognitive entity, discursive models have considered identity as produced through interactional "work". The everyday business of discursively producing knowledge is the practice of "doing" identity work. As with other psychological phenomena, identity is taken as an achievement, produced through discursive (inter)action. Multiple identities can be produced according to the social contexts in which we operate at any given time. An identity worked up in one context (e.g. as a teacher) could be markedly different to that formed in alternate contexts (e.g. as a disgruntled shopper). Accordingly, identities are not fixed but are transient, malleable products of our lexical economies. 
An issue with psychiatric diagnoses is the potential pervasive nature of their power to form identities that exist across multiple contexts. Whilst identities can be transient and multiple, differing according to context, diagnostic identities can have the power to be the dominant identifying feature in multiple contexts, and become the prime identifying feature (KNIGHT et al., 2006; PINFOLD et al., 2003; SCHULZE & ANGERMEYER, 2002). This makes it important to address the effects of diagnosis on service user identity. 
Adopting an action-oriented approach to the diagnostic identities of service users involves identifying the discursive features of narratives utilised to negotiate entering diagnostic categories. Harvey SACKS' (1995) work on the negotiation of category membership in discursive interactions proves valuable here. Particularly the concept of "category entitlements" in analysing how service users discursively relate to the diagnostic categories placed upon them. Given that issues facing service users are already bound up in being a member of a category, it is not how they create categories per se that is of analytic interest, but how they re-work those categories. With category entitlement SACKS attended to the different ways that narrative "rights" work for each member of the interaction. His now well-known example was the interaction between a witness to an accident and a friend, in which the friend did not have equal "rights" to express the same feelings of distress regarding the accident, as the friend had not actually witnessed it. This shares the concerns of service user research (this will be discussed in greater detail later in the paper), in terms of emphasising the need to address people's own perspectives of their experiences, as opposed to, for instance, addressing the issue and possible effects of diagnosis solely from the perspective of mental health professionals. 
This raises questions such as "what does it mean to enter a diagnostic category of schizophrenia, and how are such categories taken on and re-worked through the localised concerns of individual service users?" In mental health research the use of category entitlements has been demonstrated to be useful in analysing the meaning associated with aggression in psychiatric wards (BENSON et al., 2003). Discursive approaches have additionally addressed a wider set of issues in mental health research. For instance, professional accounts of the use of electroconvulsive therapy (STEVENS & HARPER, 2007); narratives of recovery from severe mental distress (MANCINI & ROGERS, 2007); men's experiences of depression (GALASINSKI, 2008); and analysing the role of interpersonal psychotherapy in recovery from depression (CROWE & LUTY, 2005). Discursive research has also focused on analysing important mediums for the construction and representation of mental health knowledge and understanding, for example, discourse analysis of the Diagnostic and Statistical Manual (CROWE, 2001), and reporting of mental health issues in the print media (HAZLETON, 1997). An aim of this paper is to further develop discursive approaches in mental health through focusing on diagnosis and identity from the perspective of service users. 
Diagnosis is primarily a practice done to people, in a top-down sense. Knowledge of how practices of diagnosis operate in a top-down way is important, and BOYLE (2002), and BENTALL (2003) amongst others, have demonstrated the fragility of certain diagnostic categories in terms of their scientific validity and reliability. These critiques have questioned the very heart of psychiatric practice, namely diagnostic categories as the primary mechanism through which psychiatric knowledge is practiced. This critical work is valuable, but insight can also be gained regarding how diagnostic categorisation works through analysing the ways that service users "own" categories. For instance, how they relate and discursively account for their experiences in light of diagnostic categories, and attempt to overcome the challenges of being obliged to discuss one's identity in relation to diagnostic knowledge. Through focusing directly on the accounts of service users themselves, analysis can highlight the ways that becoming and recognising oneself as identifiable through a diagnosis of schizophrenia can impact upon everyday life. 
The impact of awareness of social positioning can be significant. Feelings and actions are constituted, in part at least, by recognition of how we are socially classified. We are constantly being made (and remade) through "feedback loops" (HACKING, 1999), which are processes of being classified, and then recognising our classifications. People act differently in recognition of the way/s they are classified, which involves not only a base awareness, but also one built through interacting with the world in a particular way because of classification. Indeed, HACKING's concept could be seen as a move on from traditional "labelling theory" that pointed to the effects of the labels placed upon us for identity. Classic studies in that field have suggested labelling can be a positive factor in terms of providing an "explanation" for severe mental distress, but can additionally be a significant negative influence on identity due to post-labelling influence of discrimination (THOITS, 1999; LINK & PHELAN, 1999). A notion of interactive kinds not only incorporates the effects of classification on those classified, but additionally provides a theoretical route on post-classification interaction between social knowledge and individual behaviour. 
In exemplifying this HACKING utilises BOYLE's (2002) argument that the diagnosis schizophrenia fundamentally lacks scientific reliability and validity, and as such the diagnosis should be understood as a "social construct", rather than a label for a clearly defined mental illness. Whereas mainstream psychiatry's diagnostic model posits the diagnosis of schizophrenia as a scientific entity, which defines an underlying mental disorder, HACKING suggests it is actually an "interactive kind", namely a classification that is organised through the ways that people interact with it post-awareness of their classification. HACKING uses the historical changes in the diagnosis schizophrenia to exemplify this, namely how hallucinations became a "first rank symptom" in Kurt SCHNEIDER's classification of psychiatric disorders due to their absence in the earlier (and foundational) model of Eugen BLEULER, in which hallucinations were not a prime symptom. For HACKING one effect of the absence in BLEULER's model was that people were not wary of displaying hallucinatory behaviour, as it would not result in being labelled as schizophrenic. In turn, this overtness of hallucinatory behaviour subsequently "fed back" into future classifications of schizophrenia (e.g. Kurt SCHNEIDER's). HACKING argues the term "schizophrenia" is not solely defining an underlying illness, but is organised according to the relation between social knowledge and feedback of people's actions (as influenced through their awareness of being classified) post-classification. The diagnostic category has been moulded through the interaction of altered actions of people classified through awareness of their diagnosis. 
A strength of HACKING's concept is that it enables theory to consider the relation between social knowledge and its impact upon people and the localised re-working of knowledge in people's everyday lives. We become aware of the classifications placed upon us (e.g. as parent, sibling, academic) and this impacts upon our subsequent behaviour. Altered post-classification knowledge subsequently feeds back into future knowledge. In mental health this nicely frames the endeavours of "service user research" (COFFEY, 2006; GODFREY & WISTOW, 1997; NEWNES, HOLMES & DUNN, 1999, 2001; PINFOLD, 2000; POWELL, SINGLE & LLOYD, 1996; SALVI, JONES & RUGGERI, 2005; TRIVEDI & WYKES, 2002), which, although varied, share a desire to identify the effects on people of becoming and being a service user, with emphasis on their own perspective. Whilst knowledge about service users (e.g. diagnostic knowledge) is available, what cannot be known are the multifarious localised re-workings of such knowledge by service users, unless their perspective is gained. 
Service user research has emerged from a desire for, and recognition of, the value of incorporating the experiences of people who use psychiatric services in research around mental health. As we have seen, mental health knowledge is predominantly organised through mainstream psychiatry, and its diagnostic-biomedical approach of viewing mental health difficulties as "illnesses". In this sense, service users' experiences are primarily understood through psychiatric theory and practice (i.e. the practice of diagnosis). Diagnostic classifications work to categorise people's experiences, reifying them as diseases and illnesses, as things that can be identified, singled out, and treated. Critiques of this position have been long known (BOYLE, 2002; COPPOCK & HOPTON, 2000; GOFFMAN, 1968; LAING, 1969; PARKER, GEORGACA, HARPER, McLAUGHLIN & STOWELL-SMITH, 1995; SZASZ, 1974) and although varied, have placed great emphasis on de-medicalising service users' experiences. They have argued such experiences are not purely symptoms of illness, but real-life experience, grounded in everyday life concerns that need to be listened to, addressed, and fed into mental health research. 
Consequently it is important to analyse how service users themselves relate to potentially challenging diagnostic identities. Undertaking this approach will facilitate a greater insight into how service users view, perceive, manage and feel about their diagnoses. This will help illuminate the ways that service users construct identities in light of the multiple forms of knowledge that are culturally available to draw upon. Given the previously mentioned negative connotations that can be associated with these forms of knowledge, service users face particular challenges in relation to identity. Additionally, given the current debates regarding the role of diagnosis in mental health (see BOYLE, 2007; CAMPBELL, 2007; CROMBY, HARPER & REAVEY, 2007), it is important a service user perspective continues to be investigated. 
This paper aims to consider the ways that knowledge of the diagnosis schizophrenia is taken on and re-worked in the localised discursive workings of service users. In doing so it aims to contribute to service user research through analysis of service users' own versions of diagnosis, in which they are obliged to "enter" such classifications when discussing the diagnostic categories within which they have been placed. Understanding of future knowledge of schizophrenia requires addressing how people who receive that diagnosis react to it, which form ongoing "feedback loops" through which future incarnations of the category are produced. This is particularly important for schizophrenia given its contentious nature (BOYLE, 2002). 
The methodological approach in this paper appreciates the arguments made by WETHERELL (1998) that discourse analysis can benefit from drawing on concepts from both discursive psychology and Foucauldian discourse analytic traditions. 
In terms of discursive psychology (EDWARDS, 1997; EDWARDS & POTTER, 1992; POTTER, 1996), it is the focus on the localised micro interactional use of language that is of interest. An example would be to analyse the ways that the notion of "implausibility" is constructed by mental health professionals when discussing how certain experiences are categorised as "delusional" (GEORGACA, 1996). In this area, "implausibility" could be used in multiple ways in managing the "stake and accountability" (POTTER, 1996) of the individual. One may want to justify the administration of diagnosis of delusion, and thus provide professional justification for one's decisions. Here, "implausibility" is not conceptualised as a distinct cognitive entity, a feature of somebody's mental attributions, but rather as a rhetorical "tool" used as part of the production of a particular version of events. 
Foucauldian discourse analysis (FDA) on the other hand approaches the role of discursive practices in the operation and production of knowledge from a different level than discursive psychology (see WILLIG, 2001, for a broader description). Rather than look no further than the use of categories and concepts in interactions, FDA draws attention to the role of societal-level discursive formations (or discourses). A feature of this is the assumption that people draw upon particular discourses in context-dependent ways. In doing so they position themselves, and are positioned, by discourses, with the view that such positioning exposes them to societal knowledge associated to the position they occupy (see HARRÉ  for broader explication of positioning theory). For instance, people who receive a diagnosis of schizophrenia are exposed to societal knowledge regarding that classification, and subject to institutional practices associated with it, namely, psychiatric services, and the illness/medical model they mostly work within. 
WETHERELL's (1998) approach has become popular in the area of discourse analysis (e.g. LaFRANCE, 2007; STEVENS & HARPER, 2007), and this kind of approach is valuable in terms of the analytic aims of this paper. For instance, there is a need to address the positioning practices at work in interactions with service users. Additionally, these need to be analysed in terms of how they operate, impact upon, and are taken on and re-worked at a localised everyday level. This approach enables analysis of how discourses position people, but are additionally used and re-framed through localised discursive interactions. 
Thirty-eight participants were interviewed for this project, with fifteen female and twenty-three males, all over eighteen years of age. Recruitment occurred through a variety of voluntary and charity-run day centres (non-NHS2)) across the East Midlands in the UK. I, as the interviewer, was introduced by centre staff prior to making myself available to speak to service users, and subsequently arranging suitable interview times for those interested in taking part. A purposive sampling strategy was utilised in that service users approached had all received a diagnosis of schizophrenia, were in contact with local community psychiatric services (not in-patient) and consequently were receiving treatment. Thus, participants were recognised by services and themselves as "service users". It should be noted that since the move from in-patient hospital care as the primary means of treating service users to "care in the community" in the UK, most service users (unless in states of acute illness) will have the opportunity to access day centres such as those approached for this study. A large proportion of day centres are provided by the voluntary and charity sectors. As such, the participants interviewed in this study are broadly typical of service users across the UK. Ethical approval was gained from the university I was based at the time, and all service users gave written informed consent to take part in the research. 
Undertaking an interview approach means accounting for the role of the context of the interview itself. In being interviewed, the participants are placed in a space that requires them to provide an account of their experiences, but one that addresses the interviewer, as well as their everyday lives. Arguments have been made that interviews are overused in discursive work (POTTER & HEPBURN, 2005), in that interviews produce a certain kind of interaction that is not appropriate for all kinds of discursive analysis (e.g. where accounts of "naturally occurring" talk are taken as more appropriate). Despite these concerns, interviews, and the context they provide, are considered beneficial for the aims of this study, as they provide service users an opportunity to discuss their experiences, in a flexible space, with a concern to allow any variability and complexity to emerge. Alternative methods could also have facilitated this (e.g. focus groups), but interviews were used so as to provide a safe confidential space for users. 
In-depth semi-structured interviews were used, which allowed the participants to elaborate and introduce new aspects of experience, whilst still covering a set number of key areas. These were around forty-five minutes in length, and focused on receiving a diagnosis, understanding of diagnostic terms, and overall feelings in regard to diagnosis. The use of semi-structured interviews offered a flexible interview space, within which service users could discuss other aspects of their lives, and potentially open up new avenues, other than those directly addressed by the interview schedule itself. Interviews were then transcribed using a technique based upon that designed by JEFFERSON (1985), but to a less detailed degree (e.g. POTTER & WETHERELL, 1987). This can be seen in the Appendix. 
Analysis began with repeated reading and re-reading of transcripts. This is the common procedure for developing in-depth familiarity with transcripts in discourse analysis (GILL, 1996). After this the coding process began, which involved identifying all parts of the transcripts that referred to the service users' own diagnostic history. This approach was utilised so as to include all sections of transcripts that referred to the diagnostic categorisation or classification of mental health difficulties by psychiatric services. Coding was conducted in a very inclusive manner, so as to include all references to diagnosis, and the process of categorising mental distress, even those that do not explicitly mention health difficulties. Following WOOD and KROGER's (2000) guidelines analytic categories were developed through noting of repeated and interesting features of the transcripts. Once categories for coding had been completed they were cut and pasted into separate documents, and these formed the basis for analysis. The interpretative phase of analysis that followed focused on identifying wider discursive formations of positioning drawn upon (DAVIES & HARRÉ, 2001), as well as notions of stake, accountability and blame (POTTER, 1996), in participants' accounts of receiving and living with a diagnosis of schizophrenia. This followed WETHERELL's (1998) approach of utilising aspects of both FDA and discursive psychology. 
The interviews demonstrated discursive strategies for managing the risk to identity that can exist through reception of a diagnosis of schizophrenia. A dilemma of seeking to account for distressing events in one's life whilst recognising that a psychiatric explanation brings with it identity-threatening connotations was central to the accounts provided. Additionally, notions of causality were introduced that allowed a partial protection against the negative impact of diagnosis, through discussing the onset of illness as due to genetic factors. In the second half of the analysis the idea of being a risk to others (through propensity to commit violent crime) featured as a concern. Strategies of dealing with this included constructing a sense of agency over personal action, and using a subtle distinction between aggression and risk to position oneself away from being seen as a threat to others. Given the limitations of space the analysis will feature only three service users from the total corpus of data. These three are exemplars of key issues found throughout the data set. 
Analysis took two main thematic directions, firstly, addressing the construction of diagnostic identities as dilemmatic, secondly the negotiating of identity in light of the dilemma. This will involve identifying the discursive resources drawn upon and developed when producing identities that are potentially "risky". Potentially being labelled with a "schizophrenic identity" involves facing the forms of social knowledge that exist about that diagnosis (e.g. positioning people with that diagnosis as "risky" to others). The extracts presented provide examples of some of the discursive strategies utilised in producing diagnostic identities as dilemmatic. This is not to suggest that all service users would experience the same challenges, but that the data shown highlight key challenges service users can face in relation to diagnosis. 
The following section focuses on the problems involved in accounting for mental health difficulties after recognition that a diagnosis has been received, firstly, by exploring strategies of accepting diagnosis, then through focusing on the challenges of acceptance of a diagnostic identity over the long term. In doing so, the focus is on the process of needing to accept (or not) an identity that can have pervasive effects on everyday living, through being long-term and bringing with it potential exposure to negative implications associated with the diagnosis schizophrenia. In the following extract Ben3), who was in his early twenties at time of interview, discusses the situation of receiving a diagnosis around four years previously:
Ian: mm (2) what about your diagnosis then? Have you always agreed with it?
Ben: well I always knew there was something wrong cos you don't hear voices for (.) no apparent reason (I: mm) (3) and I kept, when wh..wh..when they didn't diagnose me at first (.) I kept saying you know (.) I know there's something wrong (I: mm) (1) and then when it was, the diagnosis was given (.) to me it was like (1) finally some closure (.) you know what I mean I..I..I have got something wrong (I: mm) (1) but then I (.) on the other hand it was like schizophrenia (.) I've got to live with that for the rest of my life (I: mm) (1) so it was kind of mixed emotions ... [lines 342-349] 
Ben reports his diagnosis of schizophrenia coming after a period in which he claims he knew something was wrong. His diagnostic identity is presented as something he had a controlling stake in acquiring, through identifying his experiences as problematic, which positions them as something requiring an explanation. He states his concern at the time that he was not provided with a diagnosis immediately, and was subsequently very relieved when a diagnosis was given. In doing this he actively positions himself as a service user through stating his desire and drive to receive a diagnosis ("I kept saying you know [.] I know there's something wrong"). The notion of control over diagnosis is worked up through the use of an extreme case formulation (EDWARDS & POTTER, 1992), namely "I always knew". A sense of agency is formed through a construction of persistence ("I kept"), with the repetition of "I kept" a repair to contrast with "they". Diagnosis is not reported as being involuntarily given to him, but something he actively sought. Diagnosis becomes a welcome explanatory tool in terms of Ben's previous problematic experiences of mental distress. People with health problems can often welcome a diagnosis, particularly if it means they can be effectively treated, as it can provide an explanation for what may well have been a difficult experience (TUCKER, 2004). However, this is constructed as dilemmatic for Ben, as it can result in a long-lasting identity with negative connotations. The formation of diagnosis worked up means his past experiences are re-categorised in a potentially challenging way, namely an identity that Ben potentially has to live with permanently ("it was schizophrenia, I've got to live with that for the rest of my life"). This claim uses an extreme case formulation' ("the rest of my life"), demonstrating the pressures on identity that he faces. He was relieved to receive an explanation for the distressing behaviour ("I knew there was something wrong"), but it came at a cost, namely membership of a challenging category. A strategy aimed at attempting to overcome this dilemma can be seen in the next section of Ben's extract:
Ben: yeah, that's the way I look upon it cos it's not getting any better
Ian: but is that
Ben: it's not getting any easier
Ian: mm (1) have you been told that have you or?
Ben: I haven't been told that no I just, my mum's got it, my my real mum, biological mum (I: mm) she's got it (.) she's got schizophrenia, and my uncle has (I: mm) (.) and my biological dad had a personality disorder (I: mm) so (1) there was an increased chance that (I: mm) (.) I'd get it but (.) only by a little (.) so
Ian: mm (3) so do you think there's sort of a (.) gene.. genetic element to it then?
Ben: yeah (I: mm) (1) yeah definitely (1) sometimes when I'm off on one and I (.) can't come back (1) er (.) come back to (.) society (2) I'll always say, Frances, that's my real mum, shouldn't have had kids (I: mm) (1) cos I'm not going to have kids (.) cos I wouldn't like to pass it on (I: mm) so (1) and e..e...er that's just a decision I've made (.) (I: so you..) I wouldn't like, I wouldn't like to wish this on anyone you know what I mean (I: mm) ... [lines 404-420] 
Ben's "answer" to the aforementioned dilemma is to produce a category of his own based upon both psychiatric knowledge and personal experience. He draws on psychiatric knowledge in terms of framing an increased chance receiving a diagnosis of schizophrenia, through construction of genetic formulation of increased risk due to his mother's mental health status. This turn is a key one. By introducing a family history through a genetic descriptor, a form of biological determinism is alluded to. Ben produces a personal identity that has occurred due to his genes. This works to counter a possible alternative explanation of his mental health difficulties resulting from some form of psychological problem with Ben. This formulation draws on classic mind-body dualistic thought regarding bodies as distinct biological entities, formed, in part at least, by genetics, whose existence and operation are largely outside of personal control. In the extract this version of causality works to limit personal responsibility presently, and, more importantly, over the initial onset of mental distress. 
Ben faces the problem of accounting for his difficult behaviour and experiences. He talks about when he "goes off on one", and "can't come back". He presents such events in quite a vague manner, but nevertheless as requiring some kind of explanation. Not only the difficulty of "going off on one", but also the lack of control he appears to have over such instances ("can't come back"). By stating that "he always knew" something was wrong "in the first place", Ben constructs a level of expertise over the claims that follow. This is a form of "category entitlement" (SACKS, 1995; EDWARDS & POTTER, 1992), in that the expertise he introduces entitles him to make the subsequent claim, and presents Ben as in a position of control regarding his own mental health difficulties. His genetic explanation, which whilst not ideal, at least can be located as part of Ben's identity that he has no initial control over, in the same way as the colour of his hair or sex for example. Towards the end of the extract Ben's position regarding control moves from being externally represented to something he claims back for himself. He regains authority over his life through positioning himself as choosing not to have children, enabling control over genetics to be taken. It may be the case that he has limited control over onset of his mental health difficulties, but he can take responsibility over whether he passes any genetic potential on to anyone else (i.e. children). This again points to the dilemmatic nature of identity for Ben, to retain a sense of agency requires very difficult and potential negativising decisions to be made (i.e. not having children). 
This section will explore how forms of social knowledge can operate as threats to identity, primarily through association with potential risk. Reports in the media of people diagnosed with schizophrenia committing violent crimes form a high proportion of total media coverage of the diagnosis (PHILO, 1996), which can result in cultural (mis) understandings that people diagnosed schizophrenic pose a significant threat to the general public. This issue can be seen in the following extract with a male service user Mark, who was in his mid-thirties, and had received a diagnosis around ten years ago:
Ian: what about what about um (.) your diagnosis has (1) have you always been happy with the diagnosis, has it changed at all or ...?
Mark: cos I mean I read up on it and I (1) I know bits about it and schizophrenic (1) I mean (1) what I've been told about it I'm just saying (1) if you have a person who's schizophrenic (.) you can even sink lower or (.) or be (I: mm) (1) you can be (1) gentle or more aggressive (I: mm) (1) I mean (.) that's all I know (1) I mean I'm not I'm not aggressive to no one (I: mm) (1) I mean I might get mad but I wouldn't lash out (I: mm) I mean cos it's not in my nature ... [lines 127-135] 
Here Mark produces his own tailored definition of what a diagnosis of schizophrenia means. In a similar vein to that of Ben seen earlier, Mark's account firstly works to frame him as someone with authority (and entitlement) to talk about his diagnosis ("cos I mean I read up on it"). The use of "aggression" serves a specific function here. It is noteworthy as to present it as a potential factor in talking about his diagnosis of schizophrenia Mark is possibly creating a problem for himself. To associate a diagnosis of schizophrenia with aggression is to link himself to the notion. The presentation of aggression is lessened somewhat through inclusion of gentleness ("you can be  gentle or more aggressive"). Mark's use of aggression works to relate to, but be distinct from, the notion of risk, which is clearly the main problematic for Mark in positioning himself as having received such a diagnosis. Aggression, whilst having the potential to adversely affect identity, does not necessarily have the same direct impact of admitting one is a risk to others. One can be aggressive, but not towards other people. In Mark's account, he recognises the "risky" nature of his diagnosis, but constructs it in such a way so as to allow recognition, but additionally attempting to minimise adverse impact on identity. 
Mark works to construct a set of categories that firstly works to position him as self-learnt. His statement that he "read up on it" relays a process of drawing on expert opinion and knowledge. This is important to defend against any accusation that his account is produced through self-interest to protect his "stake" (EDWARDS & POTTER  call this the "dilemma of stake"). Mark can resist accusations that his account is based on personal opinion, through including link to credible external sources. Additionally this works to "corroborate" (POTTER, 1996) his version, not because some external person was in agreement, but due to Mark's version being grounded in literature. 
The association between diagnosis and risk featured in other users' accounts. Consider the following extract with Beatrice, who was in her early forties, and had received a diagnosis around ten years ago.
Ian: mm (1) have you then (.) I mean do you agree then with your diagnosis? (1)
Beatrice: I do (.) especially when I get paranoid (.) it's awful (I: mm) (2) you think everything's against you and (1) it's awful being paranoid (I: mm) (.) I'm quite shocked that I am a paranoid schizophrenic because (.) you hear in (.) you hear about these paranoid schiz.., well (.) schizophrenics (.) committing murder (I: mm) don't you (.) in the news (.) so I was quite shocked to learn that I had the (.) the the the (.) disease really
Ian: mm (1) have you always seen it as a disease then?
Beatrice: I think so yeah I think it is (I: uh mm) (1) yeah (1) the..the (.) there's been a couple of times (.) the first time I was ill (.) I wanted to (.) well I I felt like (.) hurting some..well killing someone (1) there was a child (.) a little child at (.) at the newsagent just by the door and I felt like going over and (.) hurting them but I didn't (.) (I: mm) I stopped myself I says no I can't do that I mustn't do that it's wrong (.) and then this time round when I became ill I was in the bus (.) I was (.) waiting for a bus (.) I was in the bus queue (.) and there was this old man in front of me (.) and I thought that in my mind (.) my mind was showing me to put my hands around his neck like that (I: mm) (.) and strangle him (.) and again I had to force myself (.) it was (.) it was a really strong feeling you know (I: mm) I had to force myself not to (.) not to do it (I: mm) (1) but I've (.) that was when (.) I was still under the effects of the side effects (I: mm) (1) of this drug (1) but once I took the Procyclidine4) I..I didn't feel that way anymore (I: mm) (.) I told my psychiatrist about it and he says "you know what will happen to you if you did that don't you?", and I says "yeah I'd go to prison" (I: mm) (1) and he says yes you would (.) and we wouldn't be able to help once you (.) that happens ... [lines 176–201] 
Beatrice states that she agrees with her diagnosis (paranoid schizophrenia) as she has experienced periods of paranoia. She narrates an account in which she positions herself firmly as a paranoid schizophrenic, although this is not immediately comfortable ("I was quite shocked"), another example of dilemmatic nature of diagnostic identity. The introduction of the diagnostic term "paranoid schizophrenic" backs her into a discursive corner, negating any avoidance of the diagnostic identity. Her shock and discomfort is related to an association of paranoid schizophrenic with murder ("you hear about the paranoid schizophrenics committing murder"). The classification is then framed as risk, through positioning in a category that is associated with a possibility to commit violent crime. Interestingly, Beatrice does not seek to position herself away from this understanding as Mark did, but narrates an account in which she could be assessed to be a risk herself. This may be because she did not feel able to move away from the category completely given the firmness of its initial introduction as a defining term for her mental health difficulties. 
What is interesting is how Beatrice negotiates this dilemma. Firstly, she works to produce a factual description of the diagnosis "paranoid schizophrenia". Beatrice introduces this through the rather general claim that "you hear about these", which itself proves inadequate in terms of factual credibility. This is followed by the much stronger "in the news", which draws on understanding of (perceived) impartial reporting of news services about factual instances. The pathological nature of the paranoid category is activated through the association of potential severe criminal activity (i.e. committing murder), and the representation of the category as a "disease". Beatrice's experiences are framed within an objective account of a disease process at work, with an associated severely problematic identity, that of potential murderer. This is further concretised through narrating an instance in which Beatrice felt the urge to cause physical harm to someone else. Indeed, the severity of this is emphasised as it involved two vulnerable groups in society, namely children and the elderly. 
Beatrice's way of managing this problematic identity is to narrate an account of personal agency. Firstly she apportions blame for the potential to commit murder to the disease process, which is something outside of her control. As the narrative develops however, she makes the move of accounting for her agency in terms of overcoming the dangerous urges emanating from the underlying disease. In making this somewhat contradictory move, Beatrice is able to produce a two-pronged defence against the stigmatised identity of being a potential risk to others. 
Recognition of the moral and ethical failings of the dangerous acts works to shore up the claims that Beatrice overcame the forces of potential risk that originate in the illness, rather than herself. This move is continued through apportioning blame for such urges as additionally constituted by the medication she was taking at the time. Illness and medication are documented as the root causes of risk, both whose causality can be positioned as factors that protect Beatrice's identity to an extent that is perhaps the best she could enact once positioned as a "paranoid schizophrenic". 
This paper has developed understanding of three service users' constructions of identity in lives in relation to the operation of diagnosis, and the strategies they produced to discursively manage their experiences in these terms. Through analysis an understanding of the kinds of classificatory knowledge produced was developed, which informed as to the complex issues facing them in making sense of their experiences in relation to the diagnoses they receive. Considering the threats to the identities of those given psychiatric diagnoses, a broad armoury of discursive tools was necessary to maintain identities that work to guard against forms of discrimination, and negative societal perceptions. 
The analysis demonstrated problems that existed for Ben, Mark and Beatrice when classifying their mental health difficulties on several levels. Firstly they are faced with naming something that is a major factor in their lives, but this classification comes at a price. If they classify their difficulties according to psychiatric definitions, this can subject them to negative connotations. This is particularly salient in the cases of paranoid schizophrenia and the associated perception of risk. However, what would it mean to not classify in this way, or to move away from such classifications? This depends on a secondary factor, namely the way that causation is constructed by users. Ben presented his mental health difficulties as being caused by a physical problem, namely genetic predisposition. This serves to construct his own illness as due to a physical problem with his body, rather than some kind of unspecified mental "abnormality". This version of causation maintains that mental health difficulties are physiologically, rather than psychologically, based. Thus, although an initial exposure to potential negative perceptions is risked, Ben's account actually functions to move away from such a position, through a reductive move to neuro-chemical activity. 
What emerges from these analyses are the intricacies and subtleties employed by three service users in the discursive production of classification categories. They are faced with the challenges of becoming members of diagnostic categories that expose them to a range of potential threats to identity. In this paper, discursive strategies have been employed by Ben, Mark and Beatrice to reconstruct categories in ways that aim to limit exposure to threats to identity that can exist through receiving a diagnosis of schizophrenia, namely, genetic causal accounts, and constructing a notion of control over actions post diagnosis, which formulates a perception that personal agency exists in relation to committing actions that could be a risk to others. Although this paper included analysis from only three interviews, these are examples of issues that featured across the wider corpus of data collected for the project for which they were interviewed. 
The analysis in this paper further evidences the value and utility of undertaking service user research, namely, that service users, and their perspectives, provide a vital part of empirical efforts to build knowledge of mental health. Additionally, findings demonstrate that diagnostic categories, in this case schizophrenia, continue to present many challenges and threats to the creation and maintenance of a positive identity. Examples were seen of some of the difficulties in presenting oneself as a service user with a diagnosis of schizophrenia. Crucially, this involved seeing diagnosis as positive in terms of providing an explanation of distressing mental health difficulties. But recognising that with the initial relief of diagnosis came a subject position exposed to negative factors (e.g. "diagnosis for life", and connotations of risk). Whilst these may not exist for all service users with a diagnosis of schizophrenia, the analysis in this paper highlights a set of challenges to identity that service users can become exposed to. Diagnosis is not seen as a straightforward either-or model of bad or good, or right or wrong. Ben, for instance, formulated the reception of his diagnosis as both good and bad. Efforts to rework diagnostic categories in light of individual concerns suggest that the administration and reception of a diagnostic identity is not straightforward. This could be because categories, as they currently exist, provide insufficient explanatory power for service users. Or, it could be that societal knowledge, produced in part by media reports, are somewhat off the mark in constructing sensationalist, and consequently very unhelpful portrayals for the general public to digest. Or it could be that service users are not provided with enough information, or simply, do not understand what categories mean, beyond some "headline understanding" garnered from lay public arena. Looking forward, whatever combination of factors is key, close consideration of the impact of the process of diagnosis on service users is needed. And more broadly, further work is required to increase public awareness and understanding of mental health issues, which is currently overly informed by media reports that disproportionately produce negative, violence related, coverage of schizophrenia. 
The author would like to thank the participants for giving up their time and energy to take part in the study, and Professor Steve BROWN for insightful comments throughout the analytic process.
short pause, less than one second; numbers used in brackets to indicate number of seconds of pause
square brackets used for brief comments by other person
underlining refers to emphasis
capitals with underlining to indicate severe emphasis
quote marks used when speaker drawing on third party talk
used when that section of talk was not entirely clear; inaudible sections marked by stating inaudible in brackets
1) The people studied in this paper will be referred to as "service users" (a shortened version of mental health service users). This term is used rather than more medicalised labels such as "schizophrenics", "patients", "mentally ill" people. <back>
2) NHS refers to the Government-funded National Health Service in the United Kingdom. The day centres visited in this study were not part of the NHS, but were charity funded and managed centres. <back>
3) All names are pseudonyms. <back>
4) Procyclidine is a commonly used medication for the treatment of unwanted effects ("side effects") of anti-psychotic medication. <back>
Angermeyer, Matthias C. & Schulze, Beate (2001). Reinforcing stereotypes: How the focus on forensic cases in news reporting may influence public attitudes towards the mentally ill. International Journal of Law and Psychiatry, 24(4-5), 469-486.
Benson, Anne; Secker, Jenny; Balfe, Emma; Lipsedge, Maurice; Robinson, Sarah & Walker, Julian (2003). Discourses of blame: Accounting for aggression and violence on an acute mental health inpatient unit. Social Science & Medicine, 57(5), 917-926.
Bentall, Richard P. (2003). Madness explained: Psychosis and human nature. London: Allen Lane.
Bhugra, Dinesh (2006). Severe mental illness across cultures. Acta Psychiatrica Scandinavica, 113, 17-23.
Boyle, Mary (2002). Schizophrenia: A scientific delusion? (2nd ed.). Hove, East Sussex: Routledge.
Boyle, Mary (2007). The problem with diagnosis. The Psychologist, 20(5), 290-292.
Brown, Steve D. & Lunt, Peter (2002). A genealogy of the social identity tradition: Deleuze and Guattari and social psychology. British Journal of Social Psychology, 41, 1-23.
Campbell, Peter (2007). Hearing my voice. The Psychologist, 20(5), 298-299.
Coffey, Margaret (2006). Researching service user views in forensic mental health: A literature review. Journal of Forensic Psychiatry & Psychology, 17(1), 73-107.
Coppock, Vicky & Hopton, John (2000). Critical perspectives on mental health. London: Routledge.
Coupland, Justine & Gwyn, Richard (2003). Discourse, the body, and identity. New York: Palgrave Macmillan.
Cromby, John; Harper, David & Reavey, Paula (2007). Moving beyond diagnosis. The Psychologist, 20(5), 289.
Crowe, Marie (2001). Constructing normality: A discourse analysis of the DSM-IV. Journal of Psychiatric and Mental Health Nursing, 7(1), 69-77.
Crowe, Marie & Luty, Sue (2005). Recovery from depression: A discourse analysis of interpersonal psychotherapy. Nursing Inquiry, 12, 43-50.
Davidson, Larry (2003). Living outside mental illness. New York: New York University Press.
Davies, Bronwyn & Harré, Rom (2001). Positioning: The discursive production of selves. In Margaret Wetherell, Stephanie Taylor & Simeon J. Yates (Eds.), Discourse theory and practice (pp.261-272). London: Sage
Day, David M. & Page, Steward (1986). Portrayal of mental-illness in Canadian newspapers. Canadian Journal of Psychiatry, 31(9), 813-817.
Deaux, Kay (1993). Reconstructing social identity. Personality and Social Psychology Bulletin, 19(1), 4-12.
Dinos, Sokratis; Lyons, Evanthia & Finlay, William M.L. (2005). Does chronic illness place constraints on positive constructions of identity? Temporal comparisons and self-evaluations in people with schizophrenia. Social Science & Medicine, 60(10), 2239-2248.
Edwards, Derek (1997). Discourse and cognition. London: Sage.
Edwards, Derek & Potter, Jonathan (1992). Discursive psychology. London: Sage.
Galasinski, Dariusz (2008). Men's discourses of depression. Basingstoke: Palgrave Macmillan.
Georgaca, Eugene (1996). Exploring "psychotic discourse": The construction and negotiation of reality and subjectivity in language. Unpublished Ph.D. thesis, Manchester Metropolitan University.
Gill, Rosalind (1996). Discourse analysis: Practical implementation. In John T.E. Richardson (Ed.), Handbook of qualitative research methods for psychology and the social sciences (pp.141-159). Leicester: BPS Books
Godfrey, Mary & Wistow, Gerald (1997). The user perspective on managing for health outcomes: The case of mental health. Health & Social Care in the Community, 5(5), 325-332.
Goffman, Erving (1968). Stigma: Notes on the management of spoiled identity. Harmondsworth: Penguin.
Hacking, Ian (1999). The social construction of what? Cambridge, Mass: Harvard University Press.
Harper, David (2002). Storying policy: Constructions of risk in proposals to reform UK mental health legislation. In Brian Hurwitz, Vieda Skultans & Trisha Greenhalgh (Eds.), Narrative research in health and illness (pp.397-414). London: BMA Books.
Harré, Rom (1998). Positioning theory: Moral contexts of interactional action. Blackwell: Oxford.
Hazleton, Mike (1997). Reporting mental health: A discourse analysis of mental health-related news in two Australian newspapers. Australian and New Zealand Journal of Mental Health Nursing, 6(2), 73-89.
Horton-Salway, Mary (2001). Narrative identities and the management of personal accountability in talk about ME: A discursive psychology approach to illness narrative. Journal of Health Psychology, 6(2), 247-259.
Jefferson, Gail (1985). An exercise in the transcription and analysis of laughter. In Teun A. van Dijk (Ed.), Handbook of discourse analysis: Vol. 3 (pp.25-34). London: Academic Press.
Knight, Matthew T.D.; Wykes, Til & Hayward, Peter (2006). Group treatment of perceived stigma and self-esteem in schizophrenia: A waiting list trial of efficacy. Behavioural and Cognitive Psychotherapy, 34(3), 305-318.
LaFrance, Michelle N. (2007). A bitter pill: A discursive analysis of women's medicalized accounts of depression. Journal of Health Psychology, 12, 127-140.
Laing, Ronald D. (1969). The divided self: An existential study in sanity and madness. London: Tavistock.
Laurance, Jeremy (2003). Pure madness: How fear drives the mental health system. Abingdon: Routledge.
Link, Bruce & Phelan, Jo (1999). The labelling theory of mental disorder (II): The consequences of labelling. In Allan V. Horwitz & Teresa L. Scheid (Eds.), A handbook for the study of mental health (pp.361-377). Cambridge: Cambridge University Press.
Mancini, Michael A. & Rogers, Rebecca (2007). Narratives of recovery from serious psychiatric disabilities: A critical discourse analysis. Critical Approaches to Discourse Analysis across Disciplines, 1(2), 35-50.
Newnes, Craig; Holmes, Guy & Dunn, Cailzie (Eds.) (1999). This is madness: A critical look at psychiatry and the future of mental health services. Ross-on-Wye: PCCS Books.
Newnes, Craig; Holmes, Guy & Dunn, Cailzie (Eds.) (2001). This is madness too: Critical perspectives on mental health services. Ross-on-Wye: PCCS Books.
Parker, Ian; Georgaca, Eugene; Harper, David; McLaughlin, Terence & Stowell-Smith, Mark (1995). Deconstructing psychopathology. London: Sage.
Philo, Greg (1996). Media and mental distress. New York: Addison Weslet Longman.
Pinfold, Vanessa (2000). "Building up safe havens ... all around the world": Users' experiences of living in the community with mental health problems. Health & Place, 6(3), 201-212.
Pinfold, Vanessa; Huxley, Peter; Thornicroft, Graham; Farmer, Paul; Toulmin, Hilary & Graham, Tanya (2003). Reducing psychiatric stigma and discrimination—Evaluating an educational intervention with the police force in England. Social Psychiatry and Psychiatric Epidemiology, 38(6), 337-344.
Potter, Jonathan (1996). Representing reality: Discourse, rhetoric and social construction. London: Sage.
Potter, Jonathan & Hepburn, Alexa (2005). Qualitative interviews in psychology: Problems and possibilities. Qualitative Research in Psychology, 2, 281-307.
Potter, Jonathan & Wetherell, Margaret (1987). Discourse and social psychology: Beyond attitudes and behaviour. London: Sage.
Powell, Richard A.; Single, Helen M. & Lloyd, Keith R. (1996). Focus groups in mental health research: Enhancing the validity of user and provider questionnaires. International Journal of Social Psychiatry, 42(3), 193-206.
Rose, Diana (1998). Television madness and community care. Journal of Applied Community Social Psychology, 8, 213-228.
Sacks, Harvey (1995). Lectures on conversation. Oxford: Blackwell.
Sadler, John Z. (2005). Values and psychiatric diagnosis. Oxford: Oxford University Press.
Salvi, Giovanni; Jones, Julia & Ruggeri, Mirella (2005). Systematic review of the role of service users as researchers in mental health studies. Epidemiologia E Psichiatria Sociale, 14(4), 217-226.
Schulze, Beate & Angermeyer, Matthias C. (2002). Changing perspectives: Stigma from the point of view of people with schizophrenia, their families and of mental health professionals. Neuropsychiatrie, 16(1-2), 78-86.
Sieff, Elaine M. (2003). Media frames of mental illnesses: The potential impact of negative frames. Journal of Mental Health, 12(3), 259-269.
Stevens, Peter & Harper, David (2007). Professional accounts of electro-convulsive therapy: A discourse analysis. Social Science & Medicine, 64(7), 1475-1487.
Szasz, Thomas (1974). The myth of mental illness: foundations of a theory of personal conduct (revised ed.). New York: Harper and Row.
Tajfel, Henri (1982). Social psychology of intergroup relations. Annual Review of Psychology, 33(1), 1-39.
Terry, Deborah J.; Hogg, Michael A. & White, Katherine (1999). The theory of planned behaviour: Self-identity, social identity and group norms. British Journal of Social Psychology, 38, 225-244.
Thoits, Peggy (1999). Sociological approaches to mental illness. In Allan V. Horwitz & Teresa L. Scheid (Eds.), A handbook for the study of mental health (pp.121-139). Cambridge: Cambridge University Press.
Trivedi, Premila & Wykes, Til (2002). From passive subjects to equal partners—Qualitative review of user involvement in research. British Journal of Psychiatry, 181, 468-472.
Tucker, Ian M. (2004). "Stories" of chronic fatigue syndrome: An exploratory discursive psychological analysis. Qualitative Research in Psychology, 1(2), 153-167.
Wetherell, Margaret (1998). Positioning and interpretative repertoires: Conversation analysis and post-structuralism in dialogue. Discourse and Society, 9(3), 387-412.
Willig, Carla (2001). Introducing qualitative research in psychology: Adventures in theory and method. Maidenhead: Open University Press.
Wood, Linda A, & Kroger, Rolf O (2000). Doing discourse analysis: Methods for studying action in talk and text. London: Sage.
Dr Ian TUCKER's research interests span the areas of social psychology, process philosophy, social theory, and organisation studies. He has a long standing interest in community mental health, particularly in regard to theorising and analysing the everyday life experiences of service users. He is currently working on developing an approach to psychology and space, with specific regard to the "spaces" of community mental health.
Dr Ian Tucker
School of Psychology
University of East London
Water Lane, Stratford, London, E15 4LZ, UK
Tel.: +44 (0)20 8223 4349
Fax: +44 (0)20 8223 4937
Tucker, Ian (2009). "This is for Life": A Discursive Analysis of the Dilemmas of Constructing Diagnostic Identities [39 paragraphs]. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 10(3), Art. 24, http://nbn-resolving.de/urn:nbn:de:0114-fqs0903240.