Deconstructing Paranoia: An Analysis of the Discourses Associated with the Concept of Paranoid Delusion

David J. Harper PhD Thesis June 1999

Back to title page

Chapter 4

Analytic theme I:

The construction of the cluster of emotions, actions and beliefs associated with paranoia

... to treat depression, anxiety or fear as emotions from which people involuntarily suffer is to have far different implications than to treat them as chosen, selected, or played out as on a stage.

Gergen (1985, p.268)


In chapter 3, one of the binary oppositions I identified was that of reason/unreason. The Cartesian dualism that underpins this is related to a number of other oppositions including mind/body and thought/feeling. In this chapter, the thought/feeling opposition will be explored further. Chapter 6 will see some attention given to the mind/body opposition implicit in much of the discourse about psychiatric medication.

The privileging of rationalism that is seen in the construction of paranoid delusion is also a privileging of mind over body and of thought over feeling. Most professional texts transform emotions into symptoms, defining delusions with reference mainly to characteristics of form like 'plausibility', 'conviction' and so on. However, I was struck in the interviews by how many of the service users and professionals talked of fear and suspicion as primary concerns. This emotional aspect seems silenced and subjugated in most professional accounts. For the users I interviewed, their fear was not merely a matter of mental experience but was felt bodily and resulted in a number of effects including them taking physical action consistent with their fear (eg withdrawing from social contact with others). Of course, the meanings fear had varied within and between professionals and service-users.

Issues such as these formed the focus for this chapter. If the main concern to users was not so much their beliefs which were considered delusional but rather their fear, then what effects might this have had on their experience and their description of it? What effects might they feel this fear had had on them? What actions did they take in response to the fear? How did they see themselves in relation to it and how did professionals view such fear? Did they construct it in particular ways? How did professionals relate this experience to the interventions they offered and was the experience of fear more or less of a concern than the 'delusion' in their eyes?

In responding to these questions I will present a four-part argument. In the first part, I selectively review recent and current social constructionist work on emotion in order to provide a theoretical framework for the later analysis. In the second section I move on to examining some of the interview extracts. Here I am concerned with how 'paranoia' and 'distress' are made manifest in talk. We will see that in such extracts, notions of 'thoughts', 'feelings' and 'actions' are invoked. In the final two sections I will examine the relationship between these notions: in the third part I analyse extracts where the link between distress and paranoia are important and use them in examining the dominant psychiatric view that paranoid delusions cause distress in a linear fashion; in the fourth, and final, part I examine extracts where the link between paranoia and action are important and contrast these with dominant conceptions that such delusions cause (violent) action.

Questions about emotions often imply particular notions of those emotions -- take some of the questions about fear I asked rhetorically above. There is a need to question some of these taken-for-granted meanings of emotion prior to the analysis to see how they were taken up by professionals and service-users in the interviews. Commentators like Hallam (1994) and Harré (1991) have taken a conceptual approach to the issue of emotion, teasing out the various distinctions between feelings. One could debate, for example, the various meanings of the term 'fear'. One could also attempt to differentiate 'fear' from 'anxiety' and attempt to make 'paranoia' distinct from 'suspicion', 'mistrust' and 'self-consciousness'. Attempts to police firm boundaries around concepts of emotion often fail, however. Russell & Fehr (1994) utilised the notion of 'fuzzy concepts' in their empirical study, arguing that the concept of anger was not well characterised from the classical perspective. They hypothesised that emotion concepts 'vary in their degree of overlap' and 'are mentally represented as scripts that allow different instantiations in different contexts' (p. 186).

Another alternative to the categorical approach is to view concepts like paranoia as surrounded and constructed by a cluster of emotions, thoughts, actions and so on and, as I have argued in Part I, we can then see paranoia as a discursive complex. Others have viewed categories like paranoia in similar ways -- see Stenner & Stainton Rogers (1998) in relation to jealousy. Such clusters form certain representations of paranoia which circulate in popular and professional culture. Examples of the kinds of elements which form such clusters can be gained through surveying the professional literature. Rawlings & Freeman (1996) analysed the contents of a number of pencil and paper paranoia scales and suggested that five factors clustered together: interpersonal suspiciousness/hostility; negative mood/withdrawal; anger/impulsiveness; mistrust/wariness; and perceived hardship/resentment. Kreitler & Kreitler (1997) have noted the importance of themes of jealousy, suspicion, feelings of self-importance and a sense of danger and have reported an empirical study of people diagnosed as paranoid. They argued that participants' scoring on a list of questionnaire items revealed a two factor 'cognitive orientation'. The first factor was characterised by getting respect from others, maintaining control and an emphasis on strength, guilt and achievement. Their second factor was characterised by seeking total information, being controlled, having a negative view of the world, an emphasis on weakness and an absolutistic ambivalent connectionism (eg the existence of hidden meanings). In a sense to analyse one of these aspects is also to analyse the others since they intersect and work together. At the same time, certain social categories (eg gender, class, race and so on) cluster around paranoia producing particular subject positions and I will analyse these aspects in more detail in chapter 5.

The clusters of emotions surrounding and constructing paranoia -- and, indeed, the concept itself -- are culturally specific. Harré (1986b), notes how there may be a wide cultural diversity in emotion: that one culture's concepts of emotion may be similar in some respects but also different in other respects to concepts used in another culture; that emotions suppressed in one culture may be encouraged in another; that emotions experienced strongly in some cultures may be experienced weakly in others; that there may be historical changes in the emotional repertoire of one culture; and that terms used may involve not only 'emotions' but also other aspects like physical feelings such as coming down with flu.

Whilst concepts like paranoia and the emotions seen to be clustered around it are likely to be culturally specific it is important to recognise, as Sabini & Silver (1986) do, that the very concept of emotion serves a number of purposes: to refer to particular emotions; to provide an explanation or reason for certain actions; or to denote an action as transgressing the moral order. They have argued for the importance of treating an emotion -- in their case envy -- as 'a word in language' (1986, p. 179) since this avoids essentialising emotions and treating them as causal agents. They note, for example, that the persuasiveness of much empirical investigation of emotions is actually due to research simply telling us what we would expect from our implicit definition of the emotion being investigated. Such research, they argue, is simply conceptual clarification and, one could argue, quite tautological(1). Sarbin (1986) similarly comments that the questioning of what 'emotion' is results from the reification of the term in the first place since it is really a metaphor. It may well be that the meanings and distinctions considered by theorists to be important are not reflected in the actual use of emotional descriptions in talk. Moreover, as Stenner & Stainton Rogers (1998) note, an emotion may be 'read' in different ways according to context, position and so on. In order to avoid these dangers here I will be less concerned with definitional distinctions of paranoia-relevant emotions and concepts per se but more concerned with the way these terms are used. In order to provide a theoretically-grounded framework for such an analysis it is helpful to review recent social constructionist psychological research on emotion.


Schachter & Singer (1962) marked a turning point in the theorising of emotion by demonstrating that physiological change alone (ie an injection of epinephrine) was not sufficient to determine emotional state in the absence of some social and cognitive meaning. Studies such as this began to question essentialist and crudely biological notions of emotion, focusing attention rather on its varied contexts, influences and meaning. A further turning point came with the development of the dramaturgical model described by Sarbin (eg Sarbin, 1986), Harré and others. Harré summarises this view by defining an emotion as 'a bodily enactment of a moral judgement or attitude in accordance with the conventions of local dramatistic codes' (1991, p.142). As a result then,

there is no such thing as 'an emotion'. There are only various ways of acting and feeling emotionally, of displaying one's judgements, attitudes and opinions in an appropriate bodily way

(1991, p.142)

Work drawing on similar theoretical foundations formed part of Gergen's (1985) social constructionist agenda emphasising the importance of context, meaning and language in the experience of psychological life. For instance, constructionism questioned the essentialist view of the history of emotion by detailing changes in the way emotions have been described and experienced over time -- for example emotions dominant in Medieval times, like accidie (a kind of spiritual laziness), seem to have disappeared in present times (Harré & Finlay-Jones, 1986)(2). In chapter 1 I discussed some of the difficulties in writing histories of a concept like paranoia since the cultural meanings of apparently similar psychological objects vary over time according to developments in wider thought and the introduction of certain words and concepts like the self and so on (see also Hallam, 1994). Constructionism also questioned the idea that emotions were simply expressions of private feelings. The notion that this idea was in itself tied to a particular historical and cultural moment led to a number of further questions: how do we currently describe and account for emotions? Indeed, are emotions used as an explanatory resource in accounting for other things, like actions? What are the social and cultural effects of these views?

Other constructionist work of interest has suggested that emotions are rule-bound and have regulatory and moral functions (Averill, 1986). Armon-Jones (1986) argues that emotions may regulate certain attitudes and that the judgement about whether an emotion is warranted is a social and cultural one. This is not merely idle philosophising. Sand for example, herself someone who has suffered from severe panic attacks for much of her life, has described how certain notions of emotion affected her:

I have always felt so defeated by the belief that there is something wrong with me. Biochemically or psychologically, it doesn't matter, the problem is me, in my cellular structure. There's no escape, I am to be held hostage by Anxiety and Depression for the rest of my life(3).

(1993, p.21)

Dominant views of emotions serve a number of social and cultural purposes and in chapter 2 I discussed how the attribution of paranoia to the self or another can have a number of effects. The same is true of anxiety and fear. Smail (1984) argues that the current experience of anxiety in Western culture is related to the development of a panoptical society where we have increasingly come to experience ourselves as the focus of an objectifying gaze. This raises the possibility that anxiety and fear serve regulatory functions. However, he notes how the dominant view of anxiety as a disorder of the individual is maintained by popular culture and much of the professional literature. Hallam (1994) comments that defining anxiety as a consequence of deviation from cultural norms serves certain ideological functions in that the consequences of deviating are 'commonly attributed not to any defects of a system of regulation (or scale of values) but to natural disorders of individuals for which socially sanctioned remedial help (counselling, pharmacotherapy) is available' (p.148). Thus an individualised notion of anxiety serves to locate pathology in the individual rather than society and serves to warrant certain kinds of professional intervention. Indeed, he describes how some psychological theories locate the cause and therefore solution to anxiety as within the person.

What other effects do notions of fear and anxiety have? Croft & Beresford (1995) note how individualised conceptions like agoraphobia serve to undermine legitimate concerns about the threatening, dangerous, discriminatory, inhospitable and unsympathetic nature of public space and thus act to minimise attempts for social change. Lopez (1991) has described how notions of fear can lead to the dominance of a 'text of fear' which then organises the experience of life with people increasingly retreating to the private space of home, guarded by the technology of the security industry. This has a number of effects which are: economic (witness the growth in personal and home security alarm systems); cultural (with society becoming dominated by suspicion and observation -- the development of Neighbourhood Watch schemes in the UK is symptomatic of this); raced, gendered and age-related(4). Noam Chomsky has made a similar point in a comment on the US international War on Drugs policy:

The more you can increase fear of drugs and crime and welfare mothers and immigrants and aliens and all sorts of things, the more you control people. Make them hate each other, be frightened of each other and think that the other is stealing from them. If you can do that you can control the people.

Noam Chomsky in López et al. (1996, p.14)

Of course, as well as focusing on the negative effects of certain conceptions of fear and anxiety there is a need to recognise that such concepts are productive, that is they construct certain kinds of objects and certain shapings of subjectivity and kinds of positions for subjects to adopt. Thus Lopez and Chomsky illustrate how the construction of a notion like fear leads to the development of a reality in which other aspects are constructed including certain objects (people or things to be feared) and certain subjects (fearful individuals or sectors of society). The narrative of fear in which these objects and subjects are located has certain cultural effects -- we have seen in an ealier chapter how the adoption of a paranoid narrative can mobilise unity against a perceived threat. Allwood (1996) has described how discourse about another emotional concept, depression, constructs a number of oppositions, for example of mind/body and reason/emotion(5). Thus some lay accounts of depression talk of emotion overpowering reason and some psychological accounts argue that the mind can produce changes in the body. Conversely, medical accounts stress how the body can affect the mind. These views construct particular kinds of intervention such that, for example, cognitive therapists might wish to help their clients to use reason to control emotions seen as problematic and to help the mind affect the body in a positive manner. Alternatively, doctors might wish to prescribe medication to correct what they might see as chemical imbalances in the brain. As Allwood points out, these competing ideologies fail to challenge the underlying oppositions including that of individual/society. Even a self-help ideology (found, for example in popular psychology books) whilst appearing to 'empower' those seen as suffering from depression, may also serve to obscure the need for social change since agency and responsibility in depression are constructed at the site of the individual subject rather than at the site of society.

The view that the fear and lack of trust characterising paranoia are pathological disorders of the individual appears to be taken for granted in popular and professional culture. This perhaps links with a more general assumption that trust is a 'generally desirable, static, measurable attitude held by the individual' (Willig, 1997, p. 211). However, as we have seen in Part I, a body of work points to such a view being simplistic. In many situations, suspicion may be the best way of proceeding (eg Case, 1987; Wedow, 1979). Thus optimism and trust are not universally valued in all situations. Nor is trust solely a matter of individual decision-making. Evans et al. (1996) and Simpson (1996) have demonstrated how perceptions of fear, for example of crime, safety and danger are developed not only individually but also socially by communities through active constructions and reconstructions of events. Moreover such fears may be more or less socially sanctioned. Thus Simpson (1996) notes that one distinction between the person considered paranoid or phobic is the lack of social sanction for their fears as opposed to, say, a fear of crime.

Willig (1997) argues that rather than reifying 'trust' we should see it as a discursive resource and examine when it is called on and for what purposes. In her research on the rhetoric and practice of sexual risk-taking in heterosexual relationships she found that a range of discursive constructions of trust were employed to warrant the practice of unprotected sex, with participants using variants of the 'if you really love me, trust me' move. She concluded that trust could be reconceptualized as 'situationally specific, negotiated and purposeful social action' (p.220). With the focus here on suspicion, anxiety and fear we might explore how such constructions were employed in interviews.

I have, therefore, argued for a constructionist view of emotion. How might such an approach orient us to the analysis of paranoia and the feelings of mistrust, fear, anxiety and threat associated with it? It raises certain questions about how these feelings are described, discussed and so on. For example, what notions of paranoia are invoked in the interviews? How is distress made manifest in these discussions? Do particular ideas of paranoia and distress invite and warrant particular modes of professional intervention (eg medication, counselling and so on)(6)? How is a causal link made between paranoid delusions and distress? In what ways is a causal link made between delusions and both actual and potential actions? In the next three analytic sections, I will attempt to respond to some of these questions by exploring how paranoid beliefs/delusions, distress and action were constructed in the interviews.


A constructionist view of emotion and a focus on accounts of feelings does not mean that emotion is all talk. As I stated in chapter 3 distress can be seen as 'real' but as always mediated through concepts which are historically and culturally contingent and the aim of this work is to explore the processes and tensions involved in this mediation. In the analysis that follows I will be using extracts from the interviews to examine how paranoia and the feelings and thoughts associated with it are used in accounts. In much of the social constructionist work reviewed earlier, accounts of emotion were seen to have an important moral aspect. In this first extract we will see how paranoia provides a powerful, yet contradictory moral warrant.

4.3.1 Paranoia and moral accounting

Sharon: Well, <pause> in the past I have been paranoid quite a lot really, quite a lot of my life but erm <pause> I didn't take it as seriously as I do now. It seems to have got worse, I don't know why. I think it's because I'm, it might be because I'm doing nothing all day, because I'm unemployed, well on the sick and erm I've got nothing else to think about all day so my imagination runs a bit wild really, that might be why it is <pause>. And I do feel like frightened, sometimes I'm frightened of going outside, I get a bit agoraphobic as well. <Dave: Right> 'cos I think these people are going to be saying something <Dave: Right, right> and I just can't relax and I feel like I'm setting myself up because I can't relax and everything and that's, it gives people something to say about me.

Sharon Harrison (Service user) Moral positioning

A moral connotation is invoked in line 132 where Sharon says that she didn't take paranoia seriously in the past. This implies that she does take it seriously now and perhaps might be read by a mental health professional as an indication of motivation and insight. This moral aspect also seems significant in lines 132-133 where Sharon says that she does not know why the paranoia has got worse. This provides a useful innoculation against anyone who might challenge that she was now serious because she was not forthcoming. This account thus positions Sharon as insightful and co-operative with professionals. Paranoia as filling a lack and as needing to be controlled

Sharon explains why the paranoia has 'got worse' by noting, in line 134 that she is unemployed and that this means that she has 'nothing else to think about all day' (line 135) thus meaning that her 'imagination runs a bit wild' (line 135). This is interesting for two reasons. Firstly, this demonstrates that thoughts and feelings are not disembodied but are constructed in an embodied reality where class, employment status, gender and so on are powerful shaping forces(7). Secondly, Sharon draws on notions of control here, particularly of the mind having power over her and her feelings (eg being frightened) and her actions (eg going outside). This draws on the oppositions both of mind/body and thought/feeling. Stainton Rogers et al. (1995) have commented on how accounts of emotion in the psychology literature have been dominated by a view of them as negative in some way and in need of control. Sharon links this lack of control over her imagination with a lack of occupation. This evokes a notion of paranoia as filling a lack, a notion which occured elsewhere in the interviews. As Paul Dench put it 'if your mind's busy, you never get paranoid' (lines 515-516). Such notions construct the mind as a vulnerable and defenceless empty vessel.

The issue of control arises again at line 139 where Sharon says she 'can't relax'. This implies she has tried to relax but can't. Again, this has an effect of demonstrating she is motivated and insightful and again suggests that she does not have power over her own body. In placing agency outside of herself as a moral subject, Sharon is thus able to justify her difficulties and thus deflect perceived criticism of her. The position constructed here is of a person serious about her difficulties but who is faced by an imagination out of control. The functions of moral accounts

The mention of unemployment at line 134 implies that if Sharon were employed her imagination would not be like this and she would have something to think about. This effect is also achieved in line 139 where it is implied that control lies elsewhere too. I am not arguing here that Sharon is intentionally doing these things, merely that it is possible to read the text in this way, seeing certain positions and seeing what effects those positions might have have. That such a moral reading can be made of the extract raises a number of questions. For example, for what implied problem could this account be seen as a solution?

One view might be that this reflects her pathology, as Dr Smith noted in his interview:

Dr Smith: /Yes, never fully co-operated with treatment. <Dave: Right> Erm <coughs> <pause> erm shes erm rather insecure erm has got a very poor sort of erm self-image

Another view might be that for the person positioned as paranoid they are faced with a three-fold tension. First they need to warrant behaviour that others regard as problematic. Since one alternative explanation for such behaviour is willfulness which might lead to social disapproval, this can be achieved by being seen as suffering a psychiatric 'illness'. Second, given the moral ambivalence about whether those diagnosed with psychiatric problems are seen as wilful and responsible (eg psychiatric patients may both be seen as not responsible for their actions and, on the other hand, be seen as 'resistant', 'manipulative' and so on) there is a need to provide further evidence against a wilful interpretation, the person needs to be seen as motivated -- as a 'good' patient. Third, since dominant Western cultural views of people with psychiatric problems regard them as completely irrational, unpredictable and potentially violent the person needs to present themselves as an ordinary person who can be rational and so on. It is obvious that these three imperatives pull in different directions and have contradictory effects. Some of the comments in this account could be seen as attempted discursive solutions to these problems but of course, a solution to one of these dilemmas may construct another problem. Such an analysis illustrates why a moral theme may run through this extract in that Sharon is attempting to negotiate her identity through these three poles. Agency, ambiguity and contradictory imperatives

In this extract imagination and feeling are linked together by using 'and' (line 136) as a link word which is ambiguous with respect to cause. The term 'frightened' is used both as a general term (line 136) but also to refer to a particular object, 'going outside' (line 137). Agency in this extract is constructed in an ambiguous and contradictory manner. The line 'and I do feel like frightened' (line 136) could imply that agency is present in the feelings as separate from Sharon. 'I just can't relax' (line 139) gives a stronger impression of this in that it suggests Sharon is trying to relax but cannot and once again notions of control are invoked here. On the other hand, Sharon goes on to say that 'I feel like I'm setting myself up' (line 139) and throughout has used the pronoun 'I' when referring to thoughts and feelings. Paranoia as personal trait/external medical condition

There is a similar variability in the construction of paranoia. Thus in line 131, Sharon talks of having 'been paranoid', drawing on a notion of paranoia as a description of her character. However, at other points, paranoia is constructed as an external object. In line 132 Sharon says she does not take 'it' as seriously and that 'it' seems to have got worse -- the latter reminiscent of how one might talk about a medical condition rather than a character description. That the discourse about paranoia can be used flexibly produces a variable construction of agency too. Constructing paranoia as a personal trait simultaneously constructs agency as within the person. Constructing it as an external object simultaneously constructs agency as outside the person. Weinberg (1997) has noted how psychiatric categories may be treated by professionals and service-users as agentive, especially at times when a person's actions are not easily explained as intentional. Thus professionals might talk of 'paranoia acting up' (p.220).

Such variable construction of agency could be seen to have a number of effects. By constructing agency as elsewhere, Sharon is positioned as not responsible for her own distress -- thus deflecting implied criticism as I noted earlier. However, through the simultaneous construction of agency and responsibility as within Sharon (eg in line 139) she is also positioned as a motivated and insightful patient. This ambiguity then, does the job of negotiating two potentially contradictory imperatives. It also mirrors the ambiguity in much of the cognitive-psychiatric literature which, on the one hand, asserts that service-users are not responsible for problems like 'schizophrenia' whilst simultaneously suggesting ways in which they can cope better with their symptoms through the use of cognitive 'coping strategies'. This ambivalence is also seen in Family Management models of schizophrenia which, on the one hand construct the cause of 'schizophrenia' as biological whilst, on the other hand, suggesting that families' ways of relating lead to relapses (Johnstone, 1993a, 1993b). Paranoia as a warrant for professional intervention

Of course, one of the effects of agency (and therefore cause) being constructed as elsewhere is that an external solution is also implied. This implication is heightened when more technical terms are used. Thus Sharon talks of being 'a bit agoraphobic' (line 137). The use of a technical term here helps to construct Sharon's difficulties in such a way as to warrant professional intervention.

We have already seen how discourse about paranoia can construct it as a personal trait and as an external agent. In the previous extract there was some movement between these two oppositions. In the next extract, however, paranoia is constructed as a personal trait and we will be able to trace some of the effects this construction has.

4.3.2 Paranoia and moral accounting: paranoia as a self-description

Paul: <Pause> It's [inaudible], it's very hard to describe being paranoid and that, its symptoms, it's very hard describe <Dave: Right>. I know what they are <Dave: Yeah> [inaudible] but it's very hard to describe <Dave: Right> in words <Dave: Right, right>. <Pause> Basically being paranoid is, <pause> you're looking over your shoulder <pause> you know, and <pause> you're scared of walking out in the street 'cos you're paranoid <pause> you're scared of walking through the front door <pause> er you're scared, you're listening for sounds <pause> other people messing, other people may <pause> say different things <Dave: Uh-huh>. [inaudible]

Paul Dench (Service user) Paranoia as a way of being

In this extract, paranoia is not seen as something external. Rather, Paul talks of 'being paranoid' (line 491). Agency is seen as lying within Paul and he gives an, as it were, insider's view of what being paranoid is like, for example, what feelings are experienced. Paranoia as (obstructed) action

It is striking that Paul's account focuses less on emotion and mental events and more on actions: looking over your shoulder, walking in the street, going through the front door, listening and so on. Emotions are constructed as happening in relation to actions. Thus Paul is 'scared of walking out in the street' (lines 495-496, emphasis added). Paranoia is constructed here not as about disembodied thoughts and feelings but as something that is intimately bound up with embodied everyday actions. The construction of distress through implicit comparisons

Although Paul talks of paranoia as a way of being this account does not convey that this is something that he is content with or that he can easily change. How does this extract achieve these effects? First, by focusing on difficulties in relation to even the simplest of daily activities this account conveys how being paranoid affects the totality of life -- it is omnipresent, there is no break from it. Second, by focusing on these activities a construction of normality is implicitly assumed. We might ask, following Sampson (1993) what is the 'absent standard' that we are invited to compare Paul's difficulties with? What normal person is the implicit comparison here? Such a person would not look over their shoulder, would be confident of walking through the front door and out into the street and would not be vigilant for signs of danger (eg criticism by others). Thus the construction of distress involves comparison with implicitly assumed versions of normality. Paranoid tautologies: diagnosis as explanation

In this extract, paranoia is presented as an explanation for itself as in line 496 where Paul talks of being scared of walking out in the street 'cos you're paranoid'. This reflects the tautologous nature of much psychological and psychiatric research which has already been noted. In professional discussions, diagnoses serve both as shorthand descriptions of 'problems' and as explanations for those problems (Boyle, 1990). In this respect, paranoia acts as a warranting device for interruptions of or re-shapings of everyday activities -- ie being scared of engaging in these everyday activities. Such warrants may be necessary since there may be criticism by others which indeed seems to be revealed in accounts from people in this position. Thus Sharon talks of people saying things about her in the previous extract (lines 138 and 140) as does Paul in line 498. However this may owe as much to their moral status as patients than to their positioning as paranoid. Of course, professionals might regard statements about and sensitivity to such criticism as reflective of paranoid pathology. Furthermore, criticism may be seen to be much more likely if one draws on a notion of paranoia as a personal trait since, by implying agency lies within the person, certain personal obligations are constructed, for example that one should simply 'think rationally' or 'have courage'.

Although this extract illustrates how paranoia could be seen as a description of the self, paranoia could also be seen as an external entity. In the next extract we will see how paranoia can be constructed in this way, what kind of powers it is then seen to have and what effects this kind of description has.

4.3.3 Delusions as external and powerful

Dave: Uh-huh. I,I suppose I was wondering w-, would you see medication having a part to play in, in exploring somebody's delusions or would you prefer to work with them without any medication at all or?

Julie: Depends how distressing, as I say, the delusion is to them <Dave: Uh-huh> and their, their views about medication. If somebody is so paralysed and frightened that they can't even get out of the house then they're not going to be in a position to start to look and analyse the nature of their delusion and past experiences so I think you've got to look at it in a very practical way to start with. You know, if they're totally terrified <laughs> and I think there's some sort of almost onus on me as a health worker that, a nurse, constantly think "Well this person is clearly frightened and terrorised" <Dave: Uh-huh> "Is there something almost as a first-aid measure that you can do to get them to a reasonable level of, of, you know calmness". Not for me I, I mean <Dave: Uh-huh> traditionally [inaudible] but, I mean, for that person so that they're not in such a state of fear. <Dave: Uh-huh> But with people where it's not a, you know, terrorising their delusion then I don't know erm <pause> I would again say it's about their preference for taking medication <Dave: Uh-huh> rather than just medicating them to get rid of it <Dave: Uh-huh> because they're things people shouldn't have. I, I wouldn't necessarily go along with that, you know, if it's not causing them extreme distress <Dave: Uh-huh> and they don't want medication I don't feel they should be forced to have them in the hope this will, you know, erase the belief?

Julie Mason (CPN) Delusions as powerful

In this extract delusions (not specifically noted as paranoid) are talked of in strong agentive terms. They are seen as potentially distressing (line 229), paralysing (line 229), making service users frightened (line 230), terrified (line 233), terrorised (line 235) and leaving them in a state of fear (line 238). Julie talks of them as terrorising and as, literally, paralysing people so that they can't get out of the house. This latter aspect was noted elsewhere in the interviews -- Mike Sullivan talked of experiencing a 'paralysing suspicion' (line 257). The term paralysing is an interesting one since it is ambiguous and can imply physical paralysis -- it thus serves to embody what might otherwise be seen simply as a mental problem. Here then, the body is used as a metaphor for the mind. In using this term Julie thus implies that delusions can have serious and embodied effects. Delusions as external

I have already noted how, in some of the interviews, paranoia could be constructed as an agent external to the person(8). Such a view is similar to how Sand (1993) discussed anxiety and depression. In the professional literature delusions are already constructed as separate by virtue of their abnormality. Although delusions are not constructed as external explicitly in this interview it is implied at times, for example by using the term 'the delusion' (line 228) and by using strongly agentive descriptive terms. Such a view has a dual effect. By constructing delusions as both external and powerful, the service-user can be positioned as a victim. In this extract, we have noted how strong terms like terrified, terrorised and paralysed are used. Thus Julie constructs a potential position of victim for those with delusions. However, another view of paranoia as an external force was present in the interviews. Thus in Don's Community Mental Health Team Review I talked of Don's 'thoughts': 'I mean, you'd, you'd fought against the thoughts quite a bit' (line 329). In talking of fighting the thoughts, Don was given agency and this was made explicit at lines 333-335: 'there is the potential there for you to kind of fight back against them and stop them kind of dominating your life so much'. One effect of such a view might be to construct service-users not as simply passive victims but as active combatants who are not seen as to blame for their problems. Although subsequent failure in fighting them might position the user even further as a victim. Thus the view of delusions or paranoia as external need not necessarily invoke a victim position. Narrative Therapy (White & Epston, 1990) is an approach which externalises problems in this way and is discussed more fully in chapter 8. Constructing distress

A further feature of interest here is how the notion of distress is deployed in the account. First, as we have already noted, it is seen as having a major effect on everyday life -- it is paralysing (line 229). Second, it is seen as an obstacle to talking treatments (line 231). Third it is constructed as a priority -- something which needs sorting out first. Thus Julie talks of the need for practical, first aid measures (line 236) to get someone to a reasonable level of calmness (lines 236-237). Such a construction has a number of effects. Distress becomes objectified, as if it were something that could be affected separate from the other aspects of the person's life. This separation of distress also constructs a particular kind of professional intervention -- that is, one that could address a decontextualised distress. Medication is an obvious candidate here and is the only option mentioned in this extract (line 241). It is seen as having primacy (as a first aid measure) over talking treatments. The relationship between these two modes of intervention will be discussed in more detail in chapter 6. Another candidate for dealing with decontextualised distress which emerged in the interviews was 'anxiety management' a term used to describe a range of cognitive-behavioural techniques for coping with panic attacks. Thus in an extract we will examine in more detail below, Edward Jackson (CPN) sums up two common options in work with such service-users as 'probably up the medication, maybe do some work around anxiety management' (lines 295-296). This separation of distress from the meaning and context of user's lives and 'delusions' was common in the interviews.

Another common, and related, theme was the separation of the intellectual and affective. In the literature on delusions this separation is maintained and elaborated with the intellectual (belief) being seen to cause the affective (emotional distress). Another separation is maintained between thought/feeling and action. Here again a causal link is seen. The first separation will be explored in the next section whilst the second will be examined in the third and final analytic section.


In the previous section we have seen some of the various ways in which thoughts, feelings, actions and beliefs are used in accounts of paranoia. However, in Western psychiatry, these concepts are seen as conceptually and causally distinct. Gelder et al. (1989) note:

As a rule, when a patient first experiences a delusion he [sic] also has an emotional response and interprets his environment in a new way. For example, a person who believes that a group of people intend to kill him is likely to feel afraid. At the same time he may interpret the sight of a car in his driving mirror as evidence that he is being followed. In most cases, the delusion comes first and the other components follow

Gelder et al. (1989, p. 17)

Descriptions such as this imply certain notions of psychological and causal hierarchy. Buchanan et al. (1993) reported that the 'testing out' of beliefs regarded delusional was linked with users' feelings of sadness, fright or anxiety. They interpreted these feelings as a consequence of the delusional beliefs and they were thus seen as aspects of the 'phenomenology' of the delusions. For Gelder et al and Buchanan et al there is an assumption that, at least most of the time, thoughts or beliefs precede feelings and actions. Indeed Gelder et al specifically suggest that interviewing psychiatrists 'describe what the patient has experienced [and] record the order in which changes have occured in beliefs, affect, and the interpretation of sense data' (1989, p.17).

Such a view sits uneasily alongside social constructionist views of emotion. Thus Stainton Rogers et al. (1995) argue that dominant psychological views of emotions are predicated on the setting up of a variety of binary oppositions such as mind/body and feeling/ thought. Different traditions might assume the dominance of one of these poles over the other. Thus cognitive approaches would assume a dominance of thought over feeling whilst psychodynamic approaches would have the opposite assumption. For Stainton Rogers et al. this reproduces the 'artefactual dismemberment' of these concepts in the first place. For constructionists, they argue 'emotions are not just (perhaps not even) a "feeling" mixed with a "thought" but ways of being -- ways of course, that vary over historical time and from culture to culture' (1995, p.184). They quote Harré's (1986a) warning of the danger of talking of emotion as a thing, rather what there is

are angry people, upsetting scenes, sentimental episodes, grieving families and funerals, anxious parents pacing at midnight and so on. There is a concrete world of contexts and activities. We reify and abstract from that concreteness at our peril

(1986a, p.4)

Stainton Rogers et al. argue that it is impossible to distinguish arousal and cognition, indeed it is difficult in many encounters to separate out discrete emotions at all. Rather emotions are inseparable from the location, positioning or footing of the participants and these positions are not given but 'negotiated, worked at, resisted at some points, foisted upon others at other points' (1995, p. 187). For them, an emotion is, then

always a complex assemblage of bodily organization and transformation, bodily experience, thought, judgement, evaluation, perception, all of which takes place in specific cultural locations at particular historical junctures within the context of given social arrangements and practices and in the light of particular personal circumstances.

Stainton Rogers et al. (1995, p.188)

Moreover, they argue that the 'way' of any given emotion is not fixed or pre-established, rather its form and direction are largely influenced by conventional script-like understandings of how we 'do' various emotions and we learn this from our culture through novels, films, television and so on(9). Some descriptions of the way of being paranoid are available in popular culture and these have been described in chapter 2. Colby (1981) has even been able to simulate a narrow range of 'paranoid' verbal responses with a computer program called PARRY although, of course, these are acontextual and disembodied. However, the view that thoughts and emotions can be distinguished and that, in the field of delusions, intellectual beliefs cause emotions dominates in the mental health literature(10).

In the dominant psychiatric view what is not considered is the possibility that beliefs, feelings and so on are all certain kinds of accounts. Rather than reifying 'action', 'belief' and 'emotion' and seeing them as causally linked they could be seen differently. What if 'beliefs' and 'emotions' were seen as accounts which might or might not warrant certain kinds of action? What if the notion that beliefs cause emotions which cause actions was seen as a particular kind of account which warranted certain kinds of explanations and professional interventions? In this section I will tease apart the accepted psychiatric conception of thoughts and feelings in delusions and present extracts which illustrate the limitations of such a conception.

4.4.1 Beliefs, distress and rationality: the traditional view

In a long extract, which I have edited for reasons of space, Edward Jackson (CPN) draws on this accepted and dominant psychiatric view of the separation between cognition and affect and the view of the primacy of the cognitive element, ie, the 'delusional belief'.

Edward: Erm <pause> well the beliefs are the cause really, it's they that cause the symptoms of sort of the anxiety and the distress and the erm the fear the belief's cause.

Edward Jackson (CPN)

In this portion, Edward notes how 'the beliefs' cause distress and impact on people's lives and activities. A number of objects are constructed: 'beliefs', 'symptoms', 'anxiety', 'distress' and 'fear'. However, no subject is constructed. Instead a technical and mechanical world of objects (ie symptoms and so on) is implied. Of particular interest here is the notion of a belief or beliefs. As we saw earlier a belief can be, and is here, constructed as an agentive and separate object. One might think that feelings might be intimately involved with beliefs -- for example, if we feel strongly about something we might talk of 'having strong feelings' or 'caring passionately' about it. However, in this kind of account, beliefs are seen as solely intellectual. Moreover there is only one causal direction: beliefs cause certain feelings, not the other way around. In this kind of account then, beliefs are seen as separate from feelings and as having causal primacy. We will consider some of the effects of that view in this next portion.

Edward: ... Yeah. I mean if you can, if you can modify the beliefs then you'll modify the actual symptoms of that. <Dave: Uh-huh> I think it's very difficult to do anxiety management and just work on the symptoms. But when that belief is so strong it's probably a better approach to see if you can modify the belief in some way. If you can't then I think the choices are limited <Dave: Right, right> probably up the medication, maybe do some work around anxiety management and sort of, you know, they're feeling better, they're coping better with the [inaudible word].

Edward Jackson (CPN)

The idea that beliefs lead to distress leads to a further assumption: that modifying the beliefs will modify symptoms (lines 290-291). This constructs two options for professional intervention: either modifying the belief, seen as the cause, or 'work on the symptoms' (lines 292-293). Thus feelings (constructed here as 'symptoms') are separated from thoughts (constructed here as 'the belief') not only in terms of formulation but also in terms of actual professional intervention. There is an ambiguity in the text about whether working on the beliefs (the cause) is better than working on the symptoms (the effects). Focusing on the symptoms is seen as 'very difficult' and it is implied that this is a second-best option ('the choices are limited' -- lines 294-295). On the other hand, it is implied that one would only focus on work on the beliefs if they were strong (line 293).

Elsewhere in the interview Edward talks of how one might modify beliefs through the use of cognitive-behaviour therapy techniques. A consequence of this view of professional intervention is that other treatment options are seen pessimistically as 'limited' (line 295). Interestingly this is not when these treatments have been tried and failed but when a judgement has been made that the belief is too strong. Intervention options are then limited to increasing the medication and/or anxiety management (lines 295-296). Such a view sees these interventions as only focusing on symptoms and not their cause. The idea that some treatments can only manage symptoms and not lead to a cure is part of the discourse around medication as we will see in chapter 6. One effect of such a view is to give a defence against the argument that treatments do not lead to a cure. We might see the separation of thought and feeling, then, as part of a wider discourse explaining the difficulties in 'curing' people of their delusions.

One of the major reasons given for such failure is that the belief is strong. This is a little tautological but is of further significance in that the notion of a strong belief is ambiguous. What does strong mean here? It could be read as implying the existence of strong feelings, however, and could be seen as undermining the straightforward separation of beliefs and feelings present in this account. Moreover earlier in the interview, Edward talks of the possibility that supposedly delusional beliefs might be giving service-users 'some sense of self-esteem' (lines 280-281). Thus even in a traditional account it is difficult to police a firm distinction between thoughts and feelings. This discursive variability will be seen in further extracts where I will explore its effects in more detail.

4.4.2 Critiquing the traditional view Asking impossible questions: The presence/absence of thoughts

Dave: <Pause> Did it make any difference to the thoughts themselves? Were the thoughts still there but just bothering you less or...?

John: Well the thoughts were there but they just didn't bother me.

Dave: So it's not that they actually make the thoughts go away as such?

John: To some extent they do, yeah. But if it comes across my mind I just ignore it. <Dave: Right. And you/> /I have no feelings towards it whatsoever.

John Stewart (Service-user)

In this extract, following some earlier discussion of the effects of medication, edited out here for reasons of space, John appears to give a varying account of 'thoughts' (line 195) which are seen as having agency. Thus at line 197 John notes that the thoughts were still present but just bothered him less. However, at line 199 John appears to contradict this view by saying that the thoughts do go away 'to some extent'. How might we understand this variability? One explanation might be that my questions and distinctions do not make sense to John (or at least not in the way I appear to intend). In this extract I ask a number of questions similar to those a psychiatrist might ask (see, for example, Barrett, 1988) enquiring about the effects of medication. In following this model I appear to anticipate in the text that there will be an unproblematic answer: either the thoughts have gone or John is less bothered by them. Thus the thoughts are constructed as objects which are either present/absent and bothering/not bothering. John's responses are more ambiguous and ambivalent -- we can see that the thoughts have gone away to some extent but they do still come back, only now he is less bothered when they do return. John refuses categorisation here: there appears to be acceptance of the separation of thought and feeling I introduce at lines 195-196; however, the account given appears to be a slight contradiction of the received view discussed earlier. Thus feelings appear to be changed even though the thoughts are not always absent. The idea that thoughts are always primary in a linear chain does not appear to be borne out in this extract. There are further problems with the traditional view. Not believing doesn't stop you believing

Julie: But I think Paul would himself acknowledge, he would laugh and say "I know sometimes that they c-, they're not going to attack me" or <Dave: Uh-huh> you know "I could flatten them" <laughs> I mean, the size of him, you know, quite easily he w-, he could see the irrationality at times when he knew there was no basis. <Dave: Uh-huh> But that didn't help him stop believing it and it didn't help him erm lessen the stress those beliefs caused him.

Julie Mason (CPN)

In this extract, Julie gives an account of how Paul continued to fear attack from others even though he could 'see the irrationality' of this at times. The construction of belief here is different from that in the previous extract where beliefs were seen as separate from John and present/absent or bothersome/not bothersome. Here what is constructed is not a disembodied belief but an action of belief. When she says 'that didn't help him stop believing it' (line 140) Paul is constructed as an active agent who is doing the believing rather than passively having a belief. In giving such an account, Paul's view of his own beliefs becomes important. Such a view opens up scope for professional involvement in terms of trying to persuade Paul of the 'irrationality' of the beliefs. This construction appears to fit with my interpretation of the construction in Paul's account earlier where paranoia was seen as a way of being.

The account invokes a number of notions: of reason fighting irrationality; of insight implied (eg in lines 136 and 139); of beliefs causing distress; of control and self-control. In lines 140-141, Julie notes that Paul seeing the 'irrationality' of his belief doesn't 'help him stop believing it'. This phrase is ambiguous with regards to agency. On the one hand Paul is seen as agentive (he can see the irrationality and is seen as engaged in a process of believing). On the other hand agency is seen as elsewhere (Paul can't stop believing what he does despite seeing the irrationality).

In all the extracts discussed in this section so far, speakers have drawn on the thought/feeling opposition, and been concerned primarily with the presence and nature of beliefs which are seen as causative of distress. In other words thought feeling. However, other discourses drawing on notions of the primacy of feelings were at work in the interviews too. The primacy of fear in arriving at a diagnosis as opposed to belief in diagnosis I

Dave: Uh-huh. <Pause> Right. <Pause> What, what do you think erm <pause> the, the doctors er <pause> you know, based on your own experience, what do you think the doctors kind of er look for in deciding whether somebody's paranoid or not?

Geoff: <Pause> Fear.

Geoff Nelson (Service user)

In this brief extract, Geoff's comment implies that, in his case at least, professionals are primarily concerned with fear when arriving at a diagnosis of paranoia rather than the intellectual content or forms of 'beliefs'. Regardless of whether or not professionals would view this as 'true', the fact that Geoff can take this view indicates that such a discourse is culturally available. This suggests that, in some cases at least, a focus on the intellectual content and structure of delusional beliefs may be a post hoc legitimation of decisions. In the following extract the thought/feeling opposition intersects with a further opposition, that of mind/body, in talk about 'agitation'. The primacy of fear in arriving at a diagnosis as opposed to belief in diagnosis II: Agitation talk

Dave: Right, right. Were there other things that made you concerned about him?

Dr Cornwell: <Pause> Basically he wouldn't sit down. <Dave: Right> This was, he was and I can, from what I can remember I think he'd been pacing the floor for most of the night as well.

Dave: Right. If he hadn't been so agitated, if he'd just said that he had these thoughts about his wife would you, w-, would you still have referred him to a psychiatrist/

Dr Cornwell: /Er no. In this, in this part of the world I would have thought he was probably quite correct.

Dr Cornwell (GP)

In this extract Dr Cornwell's concerns focus mainly on John Stewart's behaviour rather than mental events: his not sitting down (line 29) and his pacing the floor at night (lines 30-31). Earlier in the interview, Dr Cornwell has described his behaviour as 'extremely agitated' (line 6). A concern with bodily action was present elsewhere in the interviews. Thus when I asked Paul Dench of ways to tell if someone was paranoid he said: 'the only way you can tell is they's restless <Dave: Right> <pause> standing up and down, walking about' (lines 454-455). In my interview with Edward Jackson, he talked of Geoff Nelson 'running everywhere' (lines 146-147) because of a 'constant fear' (line 146). He noted that Geoff used to 'arrive home er absolutely exhaused and sweating 'cause he'd run all the way from, he couldn't tolerate sitting on a bus' (lines 148-149).

For Dr Cornwell, the belief itself is not the major factor for consideration -- indeed he appears to indicate he might well have believed John Stewart (who had said he thought his wife was having an affair) were it not for the accompanying agitation (lines 35-36). Thus action can be viewed as more important in arriving at a diagnosis than thought. This draws on the mind/body opposition . However, this opposition also intersects with the thought/feeling opposition in another way in that John's behaviour can be read as indicative of agitation. Agitation is an ambiguous term that can be seen as purely descriptive of observable behaviour or can also imply emotion, especially anxiety. So this account moves between these different poles and oppositions to construct different discursive positions.

All the symptoms noted by Dr Cornwell are bodily. The implication is that these bodily actions indicate some internal pathology. We saw in Chapter 3 how the notion of signs and symptoms draws on the anatamo-clinical view of medicine. Such a view constructs John's situation as one where a line of professional intervention is reasonably clearly indicated. Being agitated would not, one might think, be sufficient warrant on its own for professional psychiatric intervention but some internal psychological cause (eg 'anxiety' or 'paranoia') would. Such a linking between internal and external worlds, between thought and action and between mind and body will be explored further in the next section.


We noted above how the dominant psychiatric view is that delusional beliefs cause distress. A concomitant view is that delusional beliefs cause certain actions which are consistent with those beliefs and with distress. In this final analytic section I will examine some of the assumptions implicit in this conception both through analysing examples from the psychiatric literature and through analysing relevant interview extracts.

4.5.1 Reading 'action' as violence

In the psychiatric literature concerned with delusions there is a particular focus on the notion of 'acting on' delusions, that is, that a delusion is seen as a causal trigger for a certain action. Wessely et al. (1993) claim that, based on reports by observers, persecutory delusions are significantly more likely to be 'acted on'. Within the whole sphere of actions that could possibly be focused on, the kind which seems to preoccupy much of the literature is that of violence. Pfohl (1978) notes how 'dangerous delusions' were said to be frequently paranoid and 'it was believed that patients who manifested such psychotic symptoms were likely to act out violently in order to defend their rigid worlds of "unreality" ' (p.108). Often the phrase 'acting on' delusions is used as a euphemistic reference to actions considered violent. In many respects the way violent actions, paranoid delusions, fear of others, feelings of suspicion and so on are constructed and related in psychiatric theories could be seen as an exemplar case and I will briefly explore these relationships here in order to highlight some themes which will be useful in framing an approach to analysing the interview material.

In Wessely et al's (1993) study of action and delusion, actions were divided into four categories: 'aggressive behaviour to others, aggressive behaviour to self, delusional behaviour in response to the principal belief identified on the MADS [Maudsley Assessment of Delusions Schedule], and any delusional behaviour' (1993, p.71).

There are a number of taken for granted assumptions here: that there are entities known as delusions; that these entities cause behaviour; that the behaviour most focused on (and therefore that which is often most expected) is of a violent nature; and that this violence is seen as somehow located within the psychiatric subject. Indeed even where the behaviour focused on is not considered violent it is still regarded as deviant (eg it is defined as 'delusional behaviour') and thus as requiring some kind of warrant(11).

Concepts like 'delusions', 'violence' and 'psychotic symptoms' become reified and come to be seen as explanations in themselves in a wide variety of reports (see also Hiday, 1995 and Swanson et al., 1996). A whole clutch of questions are raised about how 'action' and 'delusion' is defined and how the two can be linked. What is ignored at these points is that the data used throughout such research are accounts of 'symptoms' and of 'violence'. These accounts are given by particular people to certain others in specific contexts for particular purposes. Moreover the accounts concern actions, feelings and thoughts that may have certain meanings and interpretations in context. One important aspect of the way emotion is used in language, as Sabini & Silver (1986) note, is as an explanation or reason for action. Thus rather than a paranoid delusion or distress being seen as a trigger for action we might see them as explanatory resources.

Wessely et al (1993) and Buchanan et al (1993) tend to gloss over such issues. For example, they note a number of discrepancies between the accounts of service-users and those of 'observers' (eg the link between persecutory delusions and action; and delusional phenomenology and action) but do not give an adequate account of this. One possible interpretation of such discrepancies is that the 'reasons' for action have the status of accounts. From this perspective observers may construct service users' actions as either mundane (and therefore requiring no explanation) or bizarre because an acceptable warrant was not provided or culturally available to them. For the service users, the 'causes' of their behaviour (expressed in language through questionnaire, interview and so on) can also be seen as accounts constructed to serve as warrants for their actions(12).

Action may, of course, be overdetermined, that is, there may be a multiplicity of reasons for or, meanings, descriptions and interpretations of it. Moreover, explanations and descriptions of actions are accounts of those actions. For example, explanatory statements which locate cause within the subject (and more specifically within the mind of the subject) are simply a particular kind of account which has particular effects and draws on particular metaphors about the mind, emotion and action. Such accounts are not, as it were, direct recordings of reality nor can they be seen simply as 'neutral' expert professional accounts. Rather they form part of a wider societal discourse about violence, paranoia and mental illness.

4.5.2 Linking paranoia and violence in the popular mind

Wessely et al (1993) note that violent behaviour has received most attention in the literature on acting on delusions. The popular image of the violent mentally ill person often includes paranoid aspects and paranoia is one of the categories that is used to account for violent behaviour. The preoccupation with the violent character of paranoia appears to follow the tradition of the State's concern about violence and the project to seek the causes of 'dangerousness' in the mind of the individual (Foucault, 1978; Parker et al, 1995). Once more, we see the operation of a pathologising, individualising and rationalising imperative. This occurs not only at the level of professional cultures but also at the level of popular culture.

The link between paranoia and violence forms part of a wider discourse which links violence with mental illness. This discourse is widely culturally available especially in the news and entertainment media. Philo (1994) notes how the media focus primarily on violence against others when addressing issues relating to 'mental illness' with these items receiving 'headline' treatment. Scott (1994) reports that national newspapers 'frequently promoted the view that dangerousness is synonymous with mental illness' (p.490). This despite the findings of the 1994 Boyd committee that only 22 of approximately 700 British homicides were committed by people who had been in touch with psychiatric services in the previous 12 months (Mihill, 1994) -- Crepaz-Keay (1997) claims that only 6 of these had actually been admitted to a psychiatric hospital in the year prior to the homicide.

Levey & Howells (1995) have noted that, in their study, perceived dangerousness was not as important as the perceived difference and unpredictability of people diagnosed with schizophrenia. Moreover, they reported that reliance on fictional television was associated with higher ratings of unpredictability. Rose (1998a) has argued that television programmes relating to mental health have diverse and multiple meanings. However, although there is variety, especially in genres like soap operas and comedies, the category of danger is very frequent. For example, a third of all camera shots in her collection of TV news relevant to mental health dealt either visually or verbally with danger, violence and crime. Moreover, on the news, nearly two thirds of all stories involving those with psychiatric diagnoses fell into the category of crime news -- crime news itself accounts for only 10% of news coverage. Rose also notes that, when individuals seen as having mental health are filmed in both factual and fictional programmes, close-up and extreme close-up shots are used and individuals are frequently filmed alone -- these shots contrast with the filming of others. She argues that this leads to such individuals being seen as different, as other(13). Allen & Nairn (1997) have commented that the public's reading of the media needs to be seen as interactive and not a one-way street since newspaper stories are often written in such a way that readers have to draw on culturally available discourses in order to make sense of news stories dealing with mental health, on the basis of subtle cues.

This focus on violence constructs a particular discursive location for the paranoid subject: of being someone who is unpredictable, hostile, actually or potentially aggressive and so on. Paranoid delusions, here, are constructed as triggers. We might see them as causal agents to warrant these actions, thus acting as an explanatory repertoire or resource. As I argued in chapter 2, one effect of positioning the paranoid other in a certain way is to construct a particular kind of identity for oneself. In this respect it is interesting to look at the way newspaper reports of Horrett Campbell's machete attack on a group of schoolchildren and their teacher constructed Campbell. Thus the first sentence of the report from The Guardian began with 'a paranoid schizophrenic...' (Chaudhary, 1997, p.10) whilst the comments of the judge, Mr Justice Sedley were illuminating:

In some ways it is a relief to know it was a profoundly sick and deluded individual who committed these offences. To believe such an attack could be carried out by a sane person would shake belief in humanity.

Chaudhary (1997, p.10)

In constituting Campbell as 'sick' and 'deluded' and, therefore, as Other, Sedley constructs 'us' as normal, healthy and rational. The discursive constructing of the actions of the other as deviant and unwarranted and thus delusional is reflected in the manner in which the professional literature constructs and explains 'delusional action'. One interesting aspect of this is the way the literature focuses on highly abstract and disembodied concepts like 'delusions' and 'symptoms' and draws on mechanistic metaphors like 'triggers' and 'drives'. In the next section I will briefly sketch out how attention to one relatively neglected realm of embodiment, that of gender, may shed some light on the links between paranoia and violence.

4.5.3 Action and embodiment: Violence, paranoia and masculinity

So far I have noted how the word 'action' comes to stand for (actual and potential) violence in discussions of paranoid delusions and that explanations and descriptions of those actions should be seen as accounts whose constitution can then be analysed, rather than simply as 'neutral' facts. We have also seen how popular culture is suffused with notions which link mental illness, and paranoia in particular, with violence. In both the popular and professional literature, however, there is a tendency to talk of reified, abstract and disembodied notions. The search is for some abstract predicting variable, either in internal psychological structures (see, for example, Swanson et al's, 1996, concern with 'psychotic symptoms') or societal ones (see, for example, Hiday, 1995). Castel (1991) has noted how new strategies of social administration 'dissolve the notion of a subject or a concrete individual, and put in its place a combinatory of factors, the factors of risk' (p.281, emphasis in original).

There are rarely explicit discussions of major aspects of embodiment which might influence our readings. Gender, for example, was hardly mentioned in Wessely et al's (1993) analysis apart from the fact that 55% of the sample were male, and even though Hiday (1995) notes that diagnosis of major mental illness is far less predictive of violence than being 'young, male, single, lower class, and substance abusing or substance dependent' (p.123). In her paper, she goes on to analyse all of these factors in detail apart from age and gender almost as if the influence of these factors required no analysis.

This absencing of gender should, perhaps, come as no surprise since there is some evidence of a default assumption that violence is committed by men unless otherwise noted. Thus, The Guardian's report, noted above, begins 'a paranoid schizophrenic' with no mention of gender. Another case merited a headline in The Times reading 'Paranoid Killer Jailed' (The Times, 1990). Here, again, gender is not noted although a further headline, this time from The Guardian reads 'Paranoid Woman Killed Girl' (The Guardian, 1990). This might imply that in reports of violence, the fact that it was a paranoid woman is specifically marked whilst gender is unlikely to be noted if the killer is a man. Thus the absent standard is of the paranoid male killer,(14) once again demonstrating the interaction between assumptions and popular culture.

Pilgrim & Rogers (1993) note that 'men are more likely to indulge in behaviour that is antisocial and to be labelled as criminally deviant more than women. This is then reflected within psychiatry, in that men are more likely to have labels which refer to and incorporate the threat of their behaviour' (p.38). Warner (1996), however, comments that women are more likely than men to be dealt with through psychiatry than through the criminal justice system when they offend (eg by referral to a maximum security Special Hospital). What explanations might there be for gender differences? Edley & Wetherell (1995), in a discussion of male sexual violence, argue that gender differences should not be seen as something necessary or essential since 'male sexuality does not have to take violent forms, and the fact it often does reveals something about our particular social organisation around gender divisions' (p.189).

I would argue that notions of gender and, in particular, of masculinity, inform conceptions both of acting on delusions and dangerousness. In what ways could we anticipate that gender might influence conceptions of action/violence and paranoid delusions? Might violence be accepted more in men? Or anticipated more? If women are violent is this more likely to be seen as a sign of mental illness? Some of these questions will be borne in mind as we examine interview extracts that bear on action/violence and paranoia in order to trace the presence and effect of the discourses I have described so far. In the following extracts I will be teasing out the taken-for-granted assumptions about actions and delusions.

4.5.4 Paranoia as an explanation for 'out of character' conduct

Dave: Right, right. And, and <Terry: <coughs>> when things start to go a bit wrong or <Terry: Uh-huh> there's a, there's a relapse say like in Ian's case how, how important are his beliefs and delusions then? I mean like, for example with that situation with you do you think <Terry: Uh-huh> he was just going to be violent anyway or do you think the delusions and beliefs kind of er play a large part in whether he's going to be violent or who he's going to be violent to?

Terry: Well I, I felt at the time that erm that it was <pause> driven by his delusions, you know, that's the impression I got because it was so out of character, particularly with myself, we'd had a, a sort of relationship with him for q-, since he's been here really. <Dave: uh-huh> Erm <pause> and just [inaudible] it may not, it may, could equally have been er <pause> driven by drugs, you know, I don't know what he was on at the time [inaudible] <coughs>. <Pause>

Terry Reid (CPN) Action as driven

In this extract, both I and Terry Reid refer to an incident where he was attacked by Ian Berry. Terry talks of this action as being 'driven by his delusions' (lines 398-399). One effect of such an explanation is to give agency to the delusions and to lessen Ian's responsibility. Delusions are thus seen here as powerful causative agents that can over-ride the agency of the subject as we saw earlier (see also Weinberg, 1997). Another effect is to explain the apparent inconsistency between this attack and the previous relationship between Terry and Ian. Indeed the main reason Terry gives for the attack being driven by delusions was because 'it was so out of character' (lines 399-340). Thus, delusions here provide a moral warrant for the attack. This is reminiscent of Weinberg's analysis that externalising symptoms as agentive provides an explanation for action when the default option of intentionality does not seem to account for it. However, it is also noted that the attack could equally have been 'driven by drugs' (line 402). If these statements were attempted solutions to a problem, what might the problem be? One problem seems to be how to explain Ian's action as unintentional: it seems important both that a cause of the violence be found and also that the cause does not lie with Ian as a moral agent but rather with some other force which, as it were bypasses or over-rides Ian's moral agency, be it delusions or drugs. This might be important if Terry and Ian are to continue with their professional relationship -- a situation which might be more difficult if Terry were to regard Ian's action as intentional. As a result other possible explanations and interpretations of the incident are closed down. The notion that violence has a relatively straightforward cause that can be deduced by experts and is not accidental or random is related to professional discourses about dangerousness (see Castel, 1991; Pfohl, 1978; Rose, 1996).

Similar issues of moral warranting and explanation are present in the following extract, this time from the perspective of a service-user.

Dave: Right. <Pause> When you're feeling erm that people are out to get you, I mean, do there, have there ever been times when you've acted on those beliefs, when maybe you've threatened somebody who you thought was following you or something like that? Has that ever happened?

Sharon: There was erm <pause> when I was about eighteen I was on the bus and there was a man there and I thought he was, he was talking to his friends and I thought he was saying something about me, something about going round in a goldfish bowl and all the rest of it and when he was getting up off the bus I just turned round to him and said 'are you talking about me' and he said 'no' and I said 'you'd better stop it <pause> or else' I mean it's not very much I could do <Dave: Uh-huh>. I just, I'd just had enough at that time, everything just snapped and then I just got off the bus and that was that. Erm I didn't see him again after that 'cos of erm I finished work <pause> with depression, that's what it was mainly to do with.

Sharon Harrison (Service user) The dominant view and the closing down of discursive options

In this extract my question is organised by the dominant discourse on action and delusion. Thus I talk of action and belief relatively unproblematically. Moreover I link action with aggression (eg 'when maybe you've threatened somebody' line 332). Such a question closes down discursive options here, thus making it difficult for Sharon to question these notions, to talk of times when actions didn't follow from beliefs or actions other than aggressive ones. Accepting the dominant view

Here, Sharon gives an account which fits with the dominant narrative indicated in my questions. Thus she talks of having certain thoughts (lines 336-337) which then led to her making an aggressive comment to a man on a bus (lines 337-339). Paranoia and warranting action

The account contains a number of warrants for this action. Thus, Sharon talks of what she thought the man was saying -- and the word 'thought' implies that this is no longer seen as the case. The account also notes that Sharon had 'had enough at that time' (line 340), suggesting that things had been building up for a while prior to this incident. Indeed this effect is achieved by the comment that 'everything just snapped' (line 340). Finally Sharon states that some time after the incident she finished work as a result of depression (line 342). Variable agency

However, within these warrants there is some variation with respect to agency. At times, Sharon's account appears to position agency within herself -- referring to her thoughts and her having ' had enough' at that time. At other points agency is seen as lying elsewhere -- with implied hydraulic metaphors of emotion and as something that has a 'breaking point' working to remove agency and therefore responsibility from Sharon. This variation helps once again with the dilemmas noted in the first analytic section of this chapter -- how to warrant 'odd' actions with reference to psychological problems without having to give up being seen as a rational subject. In this respect the phrase 'I just turned round to him' (lines 337-338) is especially interesting since it is ambiguous with respect to cause and agency. On the one hand it could be seen as a purely descriptive account consistent with giving agency to some external agent. On the other hand it could be seen as a strong psychological account, linking with the previous comment about what she was thinking, giving agency to herself.

Both the previous extracts have concerned actions that have actually occured. In the next extract however, the issues of potential action/violence and of risk of violence are raised. These are important issues since, although originally risk factors were social factors like gender and so on (cf Castel, 1991; Rose, 1996), they are increasingly individual and intra-psychic. Thus many studies within the cognitive-psychiatric paradigm seek to predict risk of violence with reference to psychiatric symptomatology like delusions (eg Buchanan, 1997; Swanson et al, 1996).

4.5.5 Risk and potential action/violence

Dr Smith: With erm <pause> Mr Stewart the reason was erm <pause> er sort of risk to others, risk, quite substantial risk that he might act er <pause> because of his belief er there was substantial risk er because he was keeping er er a sharp axe erm and he was keeping it really, sharpening it. There was, there was considerable worry on our part that er he might act er on that basis.

Dr Smith (Consultant Psychiatrist) The construction of risk and the implication of danger

In this extract the notion of 'risk' is used (line 440) in suggesting that John Stewart might 'act ... because of his belief' (line 441). The evidence given for such a view is that John was keeping a 'sharp' axe and that he was sharpening it. The word sharp, used twice here, increases an implication of danger. Moreover the action of sharpening the axe is seen as a sign of other, more potential actions, like using the axe to attack someone. Of course, axes are sharpened all the time but this action is made to signify differently here with recourse to a belief John is said to have (line 441). Rose (1996) notes that whilst dangerousness is seen as a property of the individual, risk 'is a combination of factors which are not necessarily dangerous in themselves' (p. 13). The implication of danger is strengthened by the mention of 'considerable worry on our part' that 'he might act er on that basis' (lines 443-444). Interestingly, neither the belief nor the actual potential actions are made explicit here -- this ambiguity heightens a sense of unfocused danger. The belief constructed here acts as a simple trigger mechanism rather than something which might be complex and varied. Risk and the linking of beliefs and actions

What effects does the notion of risk have in this extract? First it makes a link between certain inferred beliefs and potential, rather than actual actions. Beliefs are seen to cause actions and thus to be predictive of actions. The link between action and belief can be seen as stronger or weaker and risk can be seen as quantitative -- thus Dr Smith talks of 'substantial risk' (line 442)(15). Second, certain actions (eg the sharpening of the axe) are cited as evidence of the likelihood of other actions (attacking someone with the axe) through the warrant of a third underlying factor, the inferred belief. Evidence is cited as an illustration of whether the risk is increased although the threshold of evidence does not appear to be very high -- it is unlikely, for example that the evidence cited here would sustain a criminal prosecution.. The social effects of a discourse of risk

At a wider level the notion of risk has a number of effects. Rose (1986), for example, has noted some of the effects on the values of liberty, democracy and civil rights. He has called for attention to the 'forms of life and the logics of culpability to which those obligations and ambitions are attached' (p.20). The prediction of risk of violence is notoriously problematic since a much wider variety of factors (mainly situational rather than constitutional) are reported to be associated with violence than delusional beliefs. However, given the political and media concern about links between mental health and violence, professionals are probably more likely to act conservatively (Castel, 1991). Current health and local authority policies give the assessment of risk in such situations a high priority which sustains professional legitimacy and is a way of answering media concern despite there being a number of problems with such assessments -- not least the fact that there is a high probability of error (Buchanan, 1997) and that although appearing merely 'technical' they incorporate a wide range of moral and ethical assumptions (Price, 1997). Risk and the construction of subjectivity

Another effect of a discourse of risk is that the subject becomes absenced. In this extract, John becomes constructed as a collection of symptoms, as a cluster of potentialities and signs and as a site of risk. The self disappears here (Castel, 1991) and the subject constructed is Other, an object of mystery, possible of anything, whose internal psychological workings are unavailable or, at the least, need to be explicated. It is perhaps not unimportant that John is a man and the image of the 'mad axe-man' is invoked here. Risk and the recasting of professional obligations

The notion of risk is a major focus in Dr Smith's account -- a primary concern is the safety of others rather than John. The belief is not primarily cast in terms of causing John distress but in raising risks for others. Thus, as Rose (1996) has commented, the concept of risk here reshapes the obligations of professionals: 'risk management and risk reduction, as logics for professional action, [have] come to supplement or replace other forms of professional action and judgement' (p.15). Government of risk takes place then, not only through a transformation of the psychiatric subject but also of the professional (Rose, 1996). Risk and the construction of dramatic narratives

One of the most striking things about the extract is how a dramatic narrative is developed with details like the sharpening of the axe introduced, helping to develop a text suffused with danger and potential (but never explicitly named) violence. This both changes professional roles and provides a moral warrant for the actions of professionals in changing their role and departing from self-consciously consensual and collaborative conduct in order to follow the consequences of a discourse of risk which might involve less consensual action (eg detaining someone under a section of the 1983 Mental Health Act). In the next extract, we will see how there are characteristic patterns in those narratives that highlight delusions, fear and risk.

4.5.6 Paranoid delusions, risk and texts of fear

Dr Chapman: And I've actually been involved in re-admitting him on three occasions. <Dave: Right. Under Section?> Yeah, with police escort usually. <Dave: Right. So, w-, why's that?> Because he's six foot six and weigh about twenty stone.

Dave: Right. I mean, I've met him <Dr Chapman: <laughs>> I mean is he, is, has he threatened violence, does he get quite agitated?

Dr Chapman: He gets very agitated. Erm and he doesn't believe there's anything wrong with him when he's severely paranoid. <Dave: Right> And he's not amenable to <pause> rational argument. Last time I saw him he was just [telephone rings]. Sorry.

[Tape paused then restarted]

Dr Chapman: The last time I went to see him about, that was about three years ago when he needed Sectioning, he'd stopped medication and he was very agitated and there was myself, a female social worker and his wife in the house and the police couldn't come 'cause it was [local town] show. <Dave: Right> They promised to meet us but they didn't turn up and he was extremely abusive erm and very frightening. And he, he didn't do anything violent but his manner was, was, he c-, he called his wife "a fucking cunt" and said "If you get me into hospital again our marriage is over" and he'd had half a bottle of whisky and fortunately he walked out the door and went out walking for half an hour. <Dave: Right> We all escaped <laughter>. It, it can be quite a frightening <Dave: Uh-huh, uh-huh> er situation.

Dr Chapman (GP) Marking Mike out as dangerous

This extract contains a number of features which mark Mike Sullivan out as potentially violent. The adjectives used to describe him and his behaviour are striking: 'very agitated', 'severely paranoid', 'extremely abusive' and 'very frightening'. Other markers used include: commenting that a 'police escort' was used in Sectioning him; his size and weight; and his not being amenable to 'rational argument'.

The invocation of Mike's size and weight(16) work to imply his potential danger and warrant professionals' concern about him. It also invokes a discourse of masculinity -- would a woman's size and weight be used in this way? Dr Chapman says that 'he didn't do anything violent but his manner was...' (line 88, emphasis added) -- this serves to imply that although Mike didn't do anything, his conduct had the same effect of frightening those near him. The 'but' here serving an innoculatory function which might deflect potential challenge (eg that Mike was not dangerous because he did nothing violent). The development of a dramatic linear narrative

A second feature of interest is that there is clearly a story being told here: Mike gets agitated; he will not listen to reason; when help is sought he gets drunk and abusive and frightens those around him. Additional elements to the story are that although the police were called they could not attend and that the gender of the participants are noted: thus Dr Chapman remarks that the others present beside himself were 'a female social worker and his wife'. The effect of the mention of gender here together with the lack of police presence could imply helplessness. A narrative of fear and danger is thus built up. Indeed when Mike later leaves, Dr Chapman tells of how they have all 'escaped' (line 92). This fear provides a warrant for the Sectioning and an innoculation against possible challenges (eg an alternative reading of the situation might suggest sectioning was unnecessary since Mike had only sworn a little, been drinking and had done nothing dangerous). 'Agitation' and the hydraulic construction of distress

A third feature is the way Mike and his potential danger is constructed. The notion of agitation (which I introduce) is used on two occasions. We saw earlier how this can be read as a sign of emotion. Here there is a stronger implication of a hydraulic model of emotion with Mike being constructed as not being able to contain his feelings and being unpredictable as a result. He notes that Mike then went out walking 'for half an hour' which conjurs up an image of Mike walking to get these feelings 'out of his system'. Mike's danger is also highlighted by his stepping out of the normal boundaries of conduct -- thus he is seen as: not amenable to rational argument; as having stopped his medication; as swearing at his wife; and as drunk. Objectifying the subject

As in the previous extract, Mike is constructed as both an object of risk and of mystery here. There is, for example, no psychological account within the narrative here, as one might expect from a professional, of possible reasons for Mike's anger. The lack of such an account adds to an implication of unpredictability and danger. Moreover, the breaking of normal rules of conduct, as I noted earlier, adds to a view of Mike as Other.

Accounts of paranoia and violence, then, incorporate dramatic narratives which serve to warrant professional fear and a change in professional role. The subject becomes objectified as a site of risk and mystery.

4.6 Discussion

In this chapter we have explored how paranoia (and the beliefs, emotions, distress and actions associated with it) was constructed in the interviews. In doing this I have followed the contours of social constructionist work on emotion highlighting, for example, the moral aspects of many accounts. We have traced the effects of different implicit oppositions, in particular thought/feeling and mind/body and deconstructed some of the dominant psychiatric narratives, for example, that beliefs cause distress and action. It has been necessary to locate the analysis of specific extracts in a wider context to understand what they may signify. For example, the 'absent presence' of masculinity was seen as important in understanding accounts of paranoia and violence and we will see in the next chapter how such absent presences are important in other ways.

Part of this reading has drawn on the fact that there is no clear separation between considerations of thoughts, beliefs, feelings and actions. Rather than focusing simply on the disembodied intellectual contents of beliefs in talking about paranoia, we have seen that professionals and service-users invoke a wide range of other matters and how a range of rhetorical strategies are used in constructing accounts. Moreover, notions of paranoia and of distress often contain implicit assumptions about normality. In the next chapter I will develop these concerns further by focusing on how the notion of plausibility is constructed in talk about paranoia. Specifically, how are accounts of the judgements about whether a belief is delusional constructed? What rhetorical strategies do service-users and professionals use to demonstrate that beliefs are delusional or not? Moreover, given that the dominant psychiatric view of the diagnosis of delusion focuses on the 'form' of beliefs (eg their conviction and veracity) which can lead to relatively disembodied discursive objects, how might a consideration of various aspects of embodiment (eg gender, race, class and so on) influence readings of plausibility?

1. In this respect it is interesting to note that much current cognitive empirical investigation into persecutory delusions simply tells us that those people considered to have such delusions jump to conclusions, are wary of certain kinds of topics or words and are self-conscious. In similar vein, Maher et al. (1987) have noted that when researchers analyse the language features of people with thought disorder (which is diagnosed according to language use) they do nothing more than discover the American Psychiatric Association's DSM. Moreover, as Harré & Van Langenhove (1991) note 'asking a person questions about eg [sic] locus of control or authoritarian behaviour is a form of positioning and has to be understood as such: it tells us something about how people position themselves when answering a questionnaire administered by a scientist' (p.405).

2. Interestingly, Finlay-Jones (Harré & Finlay-Jones, 1986) argues from a different perspective to suggest that accidie still exists but remains undiagnosed. This simply reveals the extent to which the 'existence' of emotions cannot be seen as separate from the language used to express them or the cultural availability of certain concepts, terms and so on.

3. Conversely, she has described how, following some of the tenets of Narrative Therapy (White & Epston, 1990), seeing anxiety as something external to her enabled her to fight back against it as she had done against other oppressive forces in her life. We will examine the effects of viewing paranoia as an external agent later in this chapter.

4. The influence of race, gender and class on constructions of fear will be discussed more fully in the next chapter.

5. Since writing this chapter Stoppard's (1998) materialist-discursive analysis of depression in women has been published examining the mind/body opposition in depression (and alternative conceptualisations) in more detail

6. For example, it could be argued that the biopsychosocial model of anxiety described by Yardley (1996), by representing an attempt to combine cognitive, behavioural and somatic aspects of concepts like anxiety has the effect of warranting eclectic treatment (see Samson, 1995) -- this will be discussed in more detail in chapters 6 and 8.

7. It will, perhaps, come as no surprise to readers that all of the service-user interviewees were unemployed and/or on sickness benefits or retired at the time of interview (see appendix 8). Employment and the lack of it is an important structural factor and also a symbol of the discrimination and inequality of opportunity most psychiatric service-users face.

8. I am indebted to Dick Hallam for pointing out that, of course, for someone who felt their 'beliefs' were true, the external forces would not be 'delusions' but would be malevolent and persecutory others.

9. The script metaphor of emotions is also prevalent in cognitive and other accounts -- see Russell & Fehr (1994), for example.

10. Of course, dominance does not mean that there is no account in the psychiatric literature of emotions causing beliefs. Indeed, Gelder et al note the concept of delusional mood, commenting that 'occasionally the order is reversed: the first experience is change of mood, often a feeling of anxiety with the foreboding that some sinister event is about to take place, and the delusion follows' (p.17). Moreover, 'patients do not find it easy to remember the exact sequence of such unusual and often distressing mental events and for this reason it is difficult to be certain what is primary' (p.15). Such accounts, then, include the possibility both of ambiguity and of both beliefs causing emotions and vice versa -- as such, they innoculate against challenges and are highly flexible. The actual order of feeling and belief is not important for the argument here, rather it is the conceptual separation of the two which is key.

11. There are alternative accounts in the literature that are less individualistic and which have different views on causality. Hiday (1995), for example, has argued that the direction of causality is not mental illness violence but, rather, the opposite way since 'violence may lead to fear, suspicion and distrust' (p. 125). Drawing on the work of Mirowsky and colleagues (Mirowsky & Ross, 1983; Mirowsky, 1985) she has suggested that those in 'low socio-economic positions' are characterized by powerlessness, exploitation and threat of victimization. This produces a belief in external control which, with fear, leads to mistrust and suspicion of others. This can easily be transformed into paranoid symptoms in vulnerable individuals. She concludes that 'mental illness thus produces the delusions and hallucinations, but location in the stratification system produces the attendant distrust, suspicion, and socialization to physical aggression' (1995, p.125). For Hiday then, the cause of actions should not be seen as located solely within the individual. She thus challenges some of the individualizing narrative of mainstream psychiatric accounts. However, her theory still rests on relatively traditional assumptions about mental illness, of pathology and individualism and an underlying mechanistic model albeit epidemiological rather than intra-psychic.

12. Wessely et al (1993) accept that they cannot determine whether or not an action is the result of a delusion since this is 'unknowable' (p.70). However, they still appear to wish to infer causality, although they are careful to talk of actions which are 'congruent with an abnormal belief' (p.70) and to talk of 'links' between actions and delusions. Explanations by users for actions were said not to be considered relevant (p.70). There is a general privileging of the views of 'observers' -- who are seen as neutral -- over those of service-users in this literature (eg Junginger et al., 1998).

13. Interestingly, Rose (1998a) argues that there was a change following the case of Christopher Clunis, diagnosed with schizophrenia, who murdered Jonathan Zito in 1992, leading to a public inquiry. Before 1992/1993, Rose suggests that there were stories linking mental illness with violence and with neglect but few, if any, stories linking violence with neglect. Following this period, however, she notes that blaming the violence of people with mental health problems on neglect as a result of the failures of community care became the dominant category of explanation despite no corresponding rise in the rates of violence.

14. The relationship between gender and paranoia and the absencing of other aspects of embodiment will be developed further in the next chapter.

15. Quantification is noted elsewhere in the risk literature -- Rose (1996) reports on the use of high and low risk categories. The exact manner of quantification is, however, often unclear.

16. Size and weight were used in relation to two of the interviewees by professionals and appeared to be related to discourses about masculinity (see also the discussion of this in chapter 5). They seemed to be used as demonstrations of how frightening and dangerous a big man was and thus how irrational a big man's paranoid delusions were. Thus in the case of Paul Dench, Dr Smith described him as 'a martial art coach <Dave: Uh-huh> erm er he's over er six feet tall and probably weighs now sixteen stone but erm every time you see him er he says "I can't, I can't go on like this. I'm terrified of the gangs" ' (lines 477-479).

Back to title page