Abstract

Introduction

Narrative creates multiple ways of knowing the world as a place in motion, as defined by change and yet as renderable, agreeable as a story but not necessarily as a story of precisely specified causes and effects. The brief oral stories told during health research interviews are contingent articulations which make sense out of that situation, they are accounting stories or stories of accountability (Radley & Billig 1996): in this case Harry has to account for his smoking. Much has been written on lay accounts of illness and narrative (Hydén 1997, Mattingly, 1994, Mishler 1984, Mogensen 1996, Radley 1993, Williams 1984) perhaps because it is a period in life when people struggle to account for how they came to be in their present position of suffering and what future might now lie ahead of them. A great deal less has been written on lay accounts of health and narrative since by a similar turn, it is understood in part as the absence of illness, the lack of a tellable event (Pierret 1993). However being called to account for one’s health still produces stories and is in turn a by-product of those stories, though as the following article will suggest, this account is likely to be more fragmentary since it is infrequently plotted and still tends to rely on relating health as normal life’s ‘background’ to illness as an event:

Picking up the lines of argument advanced by conversational analysts (Sacks 1992, Polanyi 1989), social psychologists (Shotter 1993, Murray 1997) and ethnomethodologist (Garfinkel 1967, Lynch 1993) this article will question the broad sweep of narrative analysis by reflecting on narratives gathered during interviews which are situationally co-constructed. In answering this question it will argue that a degree of caution should be exercised in treating 'stories about health and illness [that] are everywhere in books, in magazines, on television and in everyday conversation' (Murray 1997, page 17) as amenable to the same forms of analysis and interpretation.

If asking about health does not always lend itself to a good story since it often lacks an event to recount, asking about something else might end up turning into a revealing story about health. When Billig (1992) interviewed a series of respondents about the Royal Family, he ended up with a very rich collection of stories of nationalism, gender and the meaning of family life, a collection that would have probably been considerably poorer had he asked about nationalism, gender and the family directly. In a similar way, the complex reflections and partial narrations on life, death and health which this article will discuss arose out of interviews that were understood to be about smoking. Like the Royal Family smoking is a publicly charged topic, and is frequently the focus of argumentative stories which invoke health as one of several formal structurings to everyday behaviours, not least of course in conversational tales about what happened when for instance one person asked the teller to put their cigarette out or to smoke elsewhere. That tellable moment of argumentation about cigarettes and conduct may seem trivial but the importance of argumentation as it is told, and as it is used as part of another argumentation according to Billig (1992) is that we can learn about what it is that constitutes common-sense about smoking and health and it is generally such common sense that sets the rules around what doing being healthy consists of. Smoking is interesting in a further way since it sets up a problematic relation to health for the smokers (and perhaps a temporary resolution for the ex-smoker). As long as someone is smoking they are indeed breaching the ordinary conduct of being healthy and are likely to reflect upon what doing being healthy means and how it might be extended (or not) to their unhealthy habit.
 
 

Situating narratives

In this section I will sketch a brief history of narrative to consider whether more attention needs to be paid during analysis to the situated nature of stories gathered during research interviews. My sketch pencils in the early outlines of narrative analysis in the practices of folk-lorists and literary critics, where the object of their analysis were printed texts. Although any attempt to make a once and for all division between narrative achieved through writing and narrative achieved through talk is both misguided and unfeasible since each constitutes the other, infects the other and to divide them is to risk creating the kind of oppositional logic heavily critiqued by de-constructive approaches to speech and text (i.e. Derrida 1979). Each are engaged in a scriptural economy, they are 'social actions embedded in social worlds' (Plummer 1995, page17).

Early work on the analysis of narratives is associated with the Russian Formalist school who studied folk tales and children’s stories to establish a typology of possible patterns in the construction of a story (Propp 1917). They divided narrative into two elements fabula (story) and szjuhet(plot). The story is the sequence of events and actions as they actually happened and the plot is sequencing of events and actions as they are told. Formalism has some major drawbacks for analysing the content of any narrative since it was predominantly concerned with the uncovering of universal or meta structures underlying all story-telling. Ricouer (1991), one of the most important contemporary theorists of narrative, simplifies the confusing Formalist labels of story and plot into life and plot and narration as the activity which struggles to bring the two together. Although a phenomenologist and arguably still in search of underlying unchanging essences to story-telling, Ricoeur’s analyses of narratives are less reductive than formalist approaches and offer greater possibilities for researchers interested in the form and content of narratives. Apart from his more philosophical speculations on the relations between life and narrative, Ricoeur’s main objects of his analyses are narratives drawn from textual sources, in particular from literary novels. Indeed as noted by Potter & Wetherell (1987) there are plenty of textual sources such as novels, magazine articles, newspaper stories and autobiographies where health and illness are narrated and these have provided a rich source of material for tracing the ways in which health and illness are storied.

At its simplest, shifting narrative approaches from texts to talk involves holding onto a definition of narrative as ‘an entity that is distinguishable from the surrounding discourse and has a beginning, a middle and an end’ (Hyden, p50, 1997). The Formalists suggested that the sequential ordering of events is of critical importance since it is by such ordering that other orders are created by the narrator. Shifting away from a simple sense of language as a transparent medium which ‘tells it as it is’ there is a double emphasis on how the story is constructed and at the same time what it is representing, which becomes ‘it is as it’s told’. In the representational notion of narrative there is one problem worth highlighting which is that the narrative is still treated as a reflection of the real world of ‘lay respondents’. Narrative is treated as a distorting mirror held up to a world in motion, held up by storytellers to reflect reality as it approaches, passes them by and recedes into the distance. If we are willing to accept ‘it is as it’s told’, then this is to say that narration is the realising, interpretation and organisation of health and illness. Narrative in talk although not detachable from its present situation is part of a shifting out of a narrow and strongly contingent present tense of a conversation to a wider time span that may embed other stories directly or may borrow their formal structurings. In conversational situations a constant activity is cross referencing, validating, looking for mistakes and inconsistencies in a performance, so creating an agreed upon real is a normal part of the work of being involved in a conversation, reality is brought off by the conversationalists rather than being an entirely given and determining structure. Even telling a lie about health or illness, if it has to remain hidden has to be constructed from appeals and references to a shared set of assumptions, so the narrative analyst will still be left with relevant material on appealing to everyday discourses of health and illness (Goffman 1974). The central point that I wish to assert about conversational narrative, is that it is a negotiated and dialogical process for making up what is healthy and what is not.

The shift, as I noted earlier, that I wish to assert here is between kinds of narrative analyses still based in literary theory and the potential of a narrative analysis arising out of a conversational situation (Mishler 1986, Sacks 1992). Where in unstructured conversation ‘every story is so full of comments and interruptions, negotiations and discussion’ (Polanyi 1989: 107) and as a result does not have the discrete bindings of a book. As Ochs et al. (1996) succinctly put it, ‘stories often do not come in neat packages.’ Telling is an action carried out interactively through turn-taking in the conversation, though not necessarily more ‘democratically’ as Polanyi (1989) suggests when it is in an interview situation. Oral stories are situated and situating activities. They are produced by and productive of who tells, how and what they are telling, who is listening and where and when it all happens (Shotter 1993). Beyond their situated or contextual nature, narratives produced in conversation have a multitude of oral qualities such as variations in accent, pitch, pausing, intonation and the use of overlaps and silences which for the trained linguist provide endless fascinating data and for the ordinary participants provide important clues in the interpretation of often very brief stories (Gee 1991).
 
 

A brief story about the project

Research projects have their own guiding stories and the one related to this project that the grant-holder and the researcher told each other was about inheritance, disciplinary differences and competencies. The project was designed by a psychologist with an expertise in statistics venturing into 'qualitative methods' in health for the first time and it drew on a statistically powerful but fairly reductive model of addiction and behavioural change. It was designed drawing on her experience of statistics and psychology, but also of the interests and assessments of a funding body. A funding body that urged ever greater number of interviews to be carried out to ensure a representative population was created so that statistical analysis could be carried out on the interview data. When during the first quarter of the project the designer, and the main researcher, left for other jobs it was inherited by another member of the disciplinary team, a sociologist, and the author of this article, a cultural geographer, was appointed as the new researcher on the project.

To summarise from grant proposal document, what was inherited was : to provide a degree of stratification interviewees were sampled from three different socio-spatial zones : one inner-city, one affluent urban district and one mixed sub-urban rural. With the co-operation of local G.P.s (UK doctors providing primary healthcare) a list of around four hundred smokers and ex-smokers between ages 18 and 40 was generated and they were invited by letter to participate in the study - all of this had been completed and was thus inherited when the research project underwent a shift in its personnel. The legacy of earlier stage of the project was that each interview - of which only around fifty were agreed to - was a one-off encounter between an interviewer who had been identified in advance in the letter of invitation as coming from a 'health centre' which while helping in accessing interviewees simultaneously caused an entrance-strategy obstacle since the interviewer was then assumed to have a medical role (and potential expertise), the interview was then associated with the power-laden role definitions of medical encounters (see Mishler 1984, 1986). In the brief time left available the project underwent a process of (limited) adaptation by the inheritees which attempted to shift it toward explanations and understandings of smoking more firmly rooted in the interview-derived accounts of its ex-smokers and smokers rather than the numerical development of a model of behavioural change and addiction.

As the interviews were actually carried out certain perhaps unsurprising social trends in the 25% of the original ‘sample’ who had agreed to be interviewed on the subject of smoking, health and behavioural change, became apparent: the majority of the respondents to the mailshot were middle class, women were more likely than men to respond, and home-owners than renters. However the project which had initially been concerned with statistical representativeness was now seeking out the shared and divergent discourses between a variety of individuals. Moreover in common with many ethnomethodological responses to discourses around the statistical sampling populations it retained a suspicion of a particular statistical use of terms such as 'representativeness' (Mishler 1986, Lynch 1993)

Pre-specified by the original project investigators and contractually agreed with the funding body, the interviews were to last an hour, they were to be semi-structured around a list of topics identified in advance by the senior researcher, the principle investigator (again see Mishler (1986) for a critique of the semi-structured interview). They were mostly carried out by the author of this article, as was the case with Harry's interview which is drawn upon here. What should be born in mind is that the interviews were not designed to elicit extended stories about smoking and health, though fortunately the interview provides a highly suitable occasion for co-telling of stories and it was the interviewer’s aim to attempt to re-frame the interview as more conversational, informal and less medical. Each interview was thus pre-faced by the interviewer introducing himself as a social researcher and not a medical doctor (see quote 1 below). It was during the post-fieldwork analysis of the interview transcripts that a narrative approach was brought to bear on suitable empirical material. Part of the logic behind exploring the transcripts through narrative was derived from the interviewer's gradual substitution of question prompts such as 'when did you first start smoking' with stories about 'when my brother first started smoking' which led to an exchange of information via a more collaborative interview rather than a structured, controlled and well-defined list of questions from the interviewer and correspondingly limited responses from the interviewee. As a consequence of the interview inescapably being framed as an interview (that was the invite to which the respondents answered) the number of well-formed narratives produced was fairly limited. In choosing what transcript material to represent the interviews any kind of selection process inevitably silences numerous other possible stories about health and cigarettes and leads into questions as to 'why choose this respondent and not others?' Choosing just one narrative has even further ramifications, some of which are picked up later on in this section, at this point though it is the selection of Harry that has to be justified and not just on the grounds of typicality, since, though not a house-working mother, he was a middle class home-owner. For the interviewer Harry was one of the least intimidated of the interviewees, and importantly, was one of the quickest to re-position the interviewer as a social scientist rather than a medical doctor. Part of the explanation for this lies in the similarities in the social backgrounds of the interviewee and the interviewer (two white, male, middle class, young professionals). Nevertheless Harry's skill in bringing off stories during the interview should not be under-estimated since he provided a particularly rich source of plots, which in the empirical space of an article can only be skimmed over. What is perhaps of greatest interest in Harry’s interview material is his frequent use of humour to achieve a number different aspects of his performance as someone competent in managing his health whilst also allowing some of the moralistic and mortal charge surrounding cigarettes to be played with.

Excerpts from several interviews rather than solely Harry’s might have been extracted from the project’s data archive via coding, which is a method for tracing commonalties across large numbers of narratives and by that means revealing various shared discursive structures. In the interviews on smoking for instance, there was the ‘walking under a bus’ (Harry’s ‘I could walk out into my car...’ [and have a fatal accident]) discourse of mortality and chance, where several interviewees either said directly they might die of cancer from smoking in later life but on the other hand they might be knocked down by a bus the next day. The possibility of a bus tomorrow weighed against cancer in a distant future, being one of those ways of sharing out the cause of their death to maintain an ‘equilibrium’ (as Blaxter might put it).

A dilemma inherent in reading across narratives to gather together shared discursive structures is that it is the interpreter that is creating one order out of all the chaotic instances, as the Formalists sought to do with folk tales and children’s stories. A further and more fundamental dilemma is whether speakers who often only uttered parts of that discourse in their narrative would follow that order or whether their very disorderliness offers alternative orderings of the world. In a critique of the work of Bourdieu and Foucault, it is de Certeau (1984) that most forcefully argues against this totalising tendency in making everyday practices and everyday speech fit into theoretical structures. This critique is played out closer to the concerns of this article in Riessman (1993) and Polkinghorne’s (1995) suggestions for a different practice of dealing with transcribed narratives than coding them (see also Lucas 1997, Psathas 1992). Beginning with Polkinghorne (1995) first, he distinguishes narrative analysis from analysis of narratives, which is to contrast looking for paradigmatic elements (i.e. codes) across narratives, with looking at the unfolding of one plot in all its complexities (what the semiotician would call the syntagmatic). What is lost in the abstracting of coding is ‘the flow and flux of experience... the uniqueness and diversity of each experience’ in favour of categorising, ordering and abstracting. Polkinghorne and Riessman do not discount such approaches, instead they argue for their supplementation, in part to restore a sense of human subjectivity which is often diminished in grounded-theorising.

Riessman’s work on the practicalities of dealing with transcribed narratives are to consider ‘how’ interviews are ‘transformed into a written text’ (hence the earlier comments about the coding of the interviews) and whether during the process of determining the narrative meanings the narrative analyst leaves open the possibility of alternate meanings being derived from the narrative. Polkinghorne is more explicit in his demands, the number of narratives being dealt with have to be dramatically reduced. Riessman, as she closes her guide to narrative analysis, reminding the reader that sometimes only one narrative may be all that is required, citing Breuer’s ‘Anna O.’, Garfinkel’s ‘Agnes’ and Piaget’s children.

In what follows, it is thus Harry’s narrative that will bear weight of an argument around the analysis of conversational narratives, one of their potential methods of interpretation and indeed about what kind of ordinary narratives of health emerge from discussing smoking. There are several complex systems for transcript notation (Atkinson & Heritage 1984) used to mark up various important phenomena that occur such as speaking simultaneously, overlapping speech emphasis, doubt over transcribed words and non-verbal elements. For two reasons the quotes from the transcript of Harry’s interview that is used throughout this article are simplistically transcribed. The first is a banal legacy of the research process, wherein the large quantity of interviews were transcribed for thematic coding and isolated quotation rather than for conversation or/and narrative analysis. Interview tapes were also reused thus erasing the original audio recordings, and so Reissman’s (1993) warning to always keep the original interview tape for later more detailed transcriptions was learnt the hard way, through experience rather than fore-shadowing. The second reason however is of a more conceptual nature since it is one which is in sympathy with the over-examination of formal properties of conversation. Just as the Formalists were heavily criticised for their inattention to context, as well as content in favour of form (Holquist 1994, Hodge & Kress 1988), so it is that this article seeks to still address the transformations achieved by tellers of their situations and their lives. What this has meant during the analysis was still paying attention to how the stories are told (jointly, singly, heavily interrupted, ironically) yet balancing that with attention to what is being told, though clearly the two can never be truly separated anyway.

In practical terms this means that all non-verbal comments are put inside square brackets, lengthy pauses are indicated by three dots and the speech is uncorrected but partially punctuated for the sake of easier comprehension. The role of the researcher in this case has been in part interpretation and in part translation (Bauman 1989), shifting understandings from one interpretative site (the interview) to another (the academic article). So my apologies to the conversational analysts for over-simplifying their method and to the interviewees for both over-summarising their arguments and lifting them out of their common-sense situation (Radley & Billig 1996).
 
 

Doing ‘being healthy’

What I want to use Harry’s tale to examine is just how his conversational narrative diverges from a novelistic narrative since it is not directed to the same ends. As Sacks (1992: 215-221) suggests narratives worked upon for a novelistic text or autobiographical text will spend a lot of space and time describing objects and characters, motivations and feelings and often thereby achieving a great deal of precision, while narratives worked upon during a conversation with a stranger or even an acquaintance will often be remarkably unremarkable, frustratingly vague (though not for participants) and indeed stray toward the banal. Sacks argues that there is in effect a struggle between the tellability of daily events as worthy of interest and the tendency towards ‘non-production of stories’ (p216, ibid.) since speakers in demonstrating the ‘utter usualness’ (p219, ibid.) of their lives (i.e. ‘what’s been happening?’/ ‘just the usual’) follow one rule which then demands special manoeuvres during the telling of unusual events such as Harry’s panic attacks (quote 2) and brother’s death (quote 3). When a conversational story is told it may well be to demonstrate with extra-ordinary economy a knowledge of how far an event deviated from the usual run of things and how it might still be recuperated to show an awareness of the correct conduct of being healthy.

It needs to be remembered at this point that one critical framing operating during the interview, was the over-determined public narrative of the relationship between smoking, illness and health which disrupts doing ‘being healthy’ for an interviewee since they were already identified as either a smoker or ex-smoker. Why this is of even greater importance in this case, is the interviewer’s projected role as a representative of health promotion and biomedicine (and perhaps the shared drive on his part to perform his utter usualness rather than an exceptional expertise) which was thus even more likely to elicit narratives that demonstrated competence in saying what being healthy is, while somehow balancing out smoking’s position in this.

Earlier in the interview Harry had firmly tied his smoking to coping with alternately periods of boredom and high intensity at work. His job had been with an electricity supplier, initially in their operation department where he had to work a shift system, sometimes in solid slots of twenty four hours ‘if the system went down’, and at the time of his interview much to his relief he had finally been moved off the shift system. During his shift work he was frequently working alone at night and used the cigarettes to break up the long eventless hours, equally when work got really busy he hardly smoked at all, until there was a break when again he would have a cigarette. Smoking was, like narrative, a device for the marking of an otherwise blank passage of time, or dividing up the ceaseless rush of experiences during a busy stretch of work. Accounting for his ‘bad habit’ went beyond this kind of emotional time management to its importance as a tool in establishing friends at work and in fact acquiring a wider spread of social connections within his workplace than non-smokers. Ostracised from their workplace, the smokers ended up with feelings of solidarity and even some degree of as Harry put it, ‘wildness’ signified in not only their labelling of themselves as the ‘ravers’ but the adoption of this label by other people at their workplace.

The ‘style’ as ‘frame’ (Riessman 1993) of the interview changed, becoming more serious, as the conversation moved on to Harry’s panic attacks (quote 2). Over the past two years Harry had been ‘ill’, suffering from stomach pains when he went out for a meal, symptoms which had been diagnosed by his GP as a stomach ulcer. He had been unhappy with this diagnosis, since Harry suspected his doctor was making the diagnosis on the basis of smoking and ruling out all other factors:

Harry though not actually quitting smoking had cut down and changed to a lower tar brand during the last two years as one ‘responsible’ response to his ill-health even though he remained sceptical about its role. Importantly the story Harry is telling, as quoted here and as told in other parts of the interview, is about his illness interview with his doctor - it is a play within a play. And to return to the central argument about the importance of re-situating this narrative within its conversational context. Harry is, by its close, making an excellent display of his own competence in restoring and restorying his health and managing his illness. Being respectful of the authority of the doctor even though he and I, in the story he is telling me, know that the doctor will probably be wrong, which we know by the way Harry puts emphasis on: I was a smoker so I had a stomach ulcer, and then as he evaluates what is going on later: ‘I tried to tell him that as well but he just didn’t listen to it... he just said, do you smoke? .. and that was the answer’ and also in his shifts from a distant past tense at the opening of his story to an involving (yet also contingent, indicating his optimistic scepticism) present tense by the end (for comparable material on recounting panic attacks, see Capps & Ochs 1995). However, remain inside the summarised and labelled narrative of quote 2 and there is a risk of losing the contextual play within which this story is being told. This is a story about medical authority being wrong, it is further a story about the pre-selection of smoking cigarettes as the cause of any ‘sickness’, it is being told mid-way through an interview in which the interviewer has deliberately distanced himself from medical authority and yet still wants to know about the reasoning behind starting, stopping and restarting smoking. Talking about health and smoking has provided the occasion for a story about ill-health and smoking, that posits smoking as non-causative. That a doctor should suggest cause and be proved wrong adds to the strength of the reasonableness of Harry continuing to smoke even while suffering different kinds of ill-health.

What I would like to suggest is that Harry during quote 2 and once again in quote 3 is using short conversational stories to build up his entitlement to smoke. As mentioned earlier Harry already reported several different sources of stress: his hours of work, the content of his work, the arrangement of weddings, his stress related illness and, in the following quote, the death of his brother. In the face of all this stress his health is not simply a given it is an achievement. It is an achievement that might well have failed without Harry’s cigarettes, which is one thing that the panic attack story skilfully demonstrates to the interviewer. Part of this entitlement to smoke is constructed by Harry through the stories he is in turn entitled to tell as his own experiences as in quote 3, rather than an embedded story of someone else’s experiences where the interviewer would be on their part be entitled in turn to express some degree of disagreement. Of the many smokers interviewed during the project there were few who did not produce stories about smoking and early death. However it was very different to state succinctly that one could fall under a bus the next day compared to telling a story about the experience of actually seeing your brother crushed by an oncoming lorry.

My gloss is here is that Harry’s story produces his entitlement to assess the experience of death against the duty to be healthy (Lupton 1995). If he were telling someone else’s story about having a sibling die young of a death unrelated to smoking then it would be fairly close to the simple truism of the possibility of being run over by a bus the next day and be open to challenge. As it is through Harry’s entitlement to tell this story to the interviewer he also staked a claim on his entitlement to smoke in several different ways: by allowing his parents to witness him being unhealthy for the first time and by having ‘realise[d] that it comes to us all.’ A realisation based on experience, not on a reported and so more distanced less authoritative awareness of the mortality of the self and significant others.
 
 

Conclusion and discussion

Narrative endings are changeable, mobile and may even be used to reshape the beginning, yet the expected endings are not quite as limited as the Formalists would suggest through their analysis of the possible narrative patterns and as conversational analysts would suggest through their attention to the requirements of ordinary conversation. I introduced this article by quoting Blaxter (1997) on lay respondents moving back and forth between opposing concepts of cause of health and illness, one which emphasises agency - a person’s behaviour and attitude - and the other which emphasises external forces - luck and inevitability. Without overt theorising and without a high degree of competence in medical knowledge, lay respondents use narratives to move back and forth between their own and others’ past experiences and potential futures. Interviewees are, it would seem, hard at work attributing responsibility, stressing their entitlement to break certain codes of healthy behaviour and assessing past, present and future suffering in their dialogues with the interviewer. More than that though they are using their stories to demonstrate their competence in establishing the important aspects of doing being healthy. Cigarette smokers begin from a disadvantaged position with demonstrating the work they are doing to be healthy, as Harry’s story about his encounter with his doctor demonstrated, their competence to act healthy is under question because they have already acted unhealthily. To balance out his account Harry shows how cigarettes assisted in achieving health in the face of a highly stressful job which had caused him to be ill and moreover he displayed his entitlement via the experience of death and loss to disregard the social duty to be healthy.

The curiosity with smoking cigarettes as Klein (1993) frequently remarks is why should people continue with their habit even when they know how bad it is for them socially and in terms of their life expectancy and indeed when they know how much damage it does to their moral career (Goffman 1964). Smoking then is a classic narrative complication (a breach of health norms, an immoral behaviour and an anti-social habit), endlessly productive of narratives offering resolutions (giving up), evaluations (improving health or not, more or less stress, whose moral judgement and whose responsibility) and further complications (allowing parents to witness it, restarting, having health researchers knock at the door). What is specific to the conversational narrative is the situation in which it emerges, so in this case during an interview with a researcher associated with health, in such a situation smoking is not simply a narrative complication it is actually a reason for shifting out of the usualness and non-storyability of life .
 
 

As was noted at the outset of this article when one tries to consider health as a narratable event in its own right there is very little to tell since being healthy is almost by definition a non-event, it is what is taken for granted, as against becoming healthy or becoming ill. Health then tends to be shifted into being active, being as well as others and (in this case narratively) ready for illness . Health is treated as the ordinary state into which illness enters as a complication, yet even health as ordinariness requires work to bring it off. Narration offers the possibility of dealing with major and minor breaches in doing being healthy, so that smoking amongst other things can be the complicating issue which requires repair work for the smoker’s reputation. The reasonableness of smoking can then be resolved narratively in a variety of different ways which will be shaped by who such a course of conversational narrative as social action is co-performed with.

Harry like other interviewees in the study tried out several rich and economical conversational narrative logics in his account, none of which I could discount as either true or false, especially when they were delivered ironically. They are rather several sense-making attempts which can be taken as situated dialogues (in the interview) whilst also attempts to create, sometimes actions, and always endings, good and bad, beyond the interview: for Harry to retain his remarkable unremarkableness and for my understanding of what constitutes a reasonable account of continuing to smoke. However there are strong limits on what can be said during a one-off interview since it is a public account (Radley & Billig 1996) and equally importantly on how things can be said, to fail to perform those limits would be to display an unacceptable level of socio-situational incompetence. What doing being healthy in a conversational narrative sense means with a professional interviewer may be considerably more restricted than the narration of health for kith and kin. Explorations of the varying conversational narratives produced in familial and friendship and other non-interview situations would be a worthy area of future research into understandings of health and unhealth (and it may be worth considering why unhealth seems such a strange category even though unhealthy is not such an odd description).

Returning to the importance of the difference between narrative theory which draws on textual and visual sources (novels, autobiographies, films etc.) and conversational sources, the importance of timing of the conversation is a more fundamental activity in a conversational narrative (Lucas 1997). An inscribed source comes to an end whether it be an arbitrary cut off point implying no real resolutions or alternatively a virtuoso tying up of several plots into a fantastical resolution. The smoking plots told by Harry and other interviewees are picked up and put down many times, new evaluations are linked in and old ones discarded. That such moves are carried out banally hides the complex symbolic and discursive work going on in conversation in a way that tends not to occur in the narratives of magazines, films and advertising (though they have their own styles of banality). As Radley & Billig (1996) emphasise conversational narratives are told jointly and under ‘real-time’ interactive conditions which normally engage participation via various frameworks: people having already had many conversations about and with cigarettes and the interview providing a particular occasion for re-visiting and re-making those conversations. As conversations are continued elsewhere beyond the interview, tellers and audiences are attributing and checking with their co-tellers felt probabilities - ‘there’s more chance of me giving up than you’, ‘you’re at greater risk than I am’ - yet they are also perhaps highlighting how smoking itself can be a vital part of that joint activity of their conversations.
 
 
 
 

Acknowledgements:

For her helpful comments I offer great thanks to Linda Mckie. I would like to acknowledge the support of Drs. Scott Lennox and Ross Taylor (co-grant holders) and the early and invaluable input of Norma Bain, Norma Goodwin, Jenny Henry and Isobel MacPherson. I would also like to express my gratitude to all the interviewees for their participation and patience. This research was undertaken with funding from the Chief Scientist’s Office of the Scottish Office (Grant No: K/OPR/17/2). The ideas and views expressed in the article are those of the author and are not necessarily those of the funding body.

Bibliography

Adam, B. (1990) Time and Social Theory, Oxford: Polity

Ahearne, J.N. (1995) Michel de Certeau, interpretation and its other, Oxford: Polity

Boden, D. (1994) The Business of Talk, Cambridge: Polity

Bauman, Z. (1989) Legislators and interpreters. On modernity, post-modernity & intellectuals. Cambridge: Polity

Bauman, Z. (1994) Mortality, Immortality & Other Life Strategies. Cambridge: Polity

Billig, M. (1992) Talking of the Royal Family, London: Routledge

Blaxter, M. (1990) Health and Lifestyles, London: Routledge

Blaxter, M. (1997) Whose fault is it? People’s own conceptions of the reasons for health inequalities’, Social Science & Medicine, 44, 747-756

Burnard, P. (1995) Intepreting text: an alternative to some current forms of textual analysis in qualitative research, Social Sciences in Health, 1, 236-245

Capps, L. & Ochs E. (1995) Constructing Panic: The discourse of agoraphobia. London: Harvard

de Certeau, M. (1984) The Practice of Everyday Life. London: University of California Press

Derrida, J (1979) Writing and difference. Trans. A. Bass. London: Rouledge

Garfinkel, H. (1967) Studies in Ethnomethodology. Englewood Cliffs, NJ: Prentice Hall

Garfinkel, H. & Sacks H. (1986) (orig. 1969) On Formal Structures of Practical Actions. In Garfinkel, H. Ethnomethodological Studies of Work, London: Routledge & Kegan Paul

Gee, J. P. (1991) A linguistic approach to narrative. Journal of Narrative and Life History, 1, 15-39

Glaser, B. & Strauss, A. (1967) The discovery of grounded theory: Strategies for qualitative research, Aldine Pub., Chicago

Goffman, E. (1964) Stigma. Englewood Cliffs, NJ: Prentice-Hall

Goffman, E. (1974) Frame Analysis. New York: Harper & Row

Goffman, E. (1981) Forms of talk. Oxford: Blackwell

Goodwin, N., Laurier, E., & McKie, L. (forthcoming) Rites, Rituals and Ceremonies: Narratives of Smoking and Smoking Cessation, Social Sciences in Health

Hodge, G & Kress, H. (1988) Social Semiotics. Oxford: Blackwell

Holquist, M. (1990) Dialogism: Bakhtin & his world. London: Routledge

Hutcheon, L. (1994) Irony’s Edge: the theory and practice of irony. London: Routledge.

Hyden, L. C. (1997) Illness and narrative, Sociology of Health & Illness, 19, 48-69

Klein, R. (1993) Cigarettes are Sublime. Raleigh, NC: Duke University Press

Lucas, J. (1997) Making Sense of Interviews: The Narrative Dimension, Social Sciences in Health, 3, 113-126

Lupton, D (1995) The Imperative of Health: public health and the regulated body. London: Sage

Lynch, M. (1993) Scientific practice and ordinary action. Cambridge: Cambridge University Press

Kleinman, S. & Copp, M. (1993) Emotions and Fieldwork, London: Sage

Mattingly, C. (1994) The concept of therapeutic ‘emplotment’, Social Science and Medicine, 38, 811-822

Mogensen, H.O. (1996) The narrative of AIDS amongst the Tonga of Zambia, Social Science and Medicine, 44, 431-440

Mishler, E.G. (1984) The Discourse of Medicine. Dialectics of Medical Interviews. Norwood NJ: Ablex Publishing Company

Mishler, E. G. (1986) Research Interviewing. London: Harvard University Press

Murray, M. (1997) A Narrative Approach to Health Psychology, Journal of Health Psychology, 2, 9-20

Ochs, E., Smith, R.C., & Taylor, C.E. (1996) Detective Stories at Dinnertime: Problem Solving Through Co-Narration. In C.L. Briggs, Disorderly Discourse: Narrative, Conflict, and Inequality, Oxford: Oxford University Press

Phillips, A. (1993) On kissing, tickling and being bored: psychoanalytic essays on the unexamined life, London: Faber& Faber

Pierret, J. (1993) Constructing discourses about health and their social determinants, In Worlds of illness, Biographical and cultural perspectives on health and disease, (ed. Radley, A.), London: Routledge, pp9-26

Plummer, K. (1995) Telling Sexual Stories: Power, Change and Social Worlds. London: Routledge.

Polanyi, L. (1989) Telling the American Story: A Structural and Cultural Analysis of Conversational Storytelling, London: MIT Press

Polkinghorne, D.E. (1995) Narrative Configuration in Qualitative Analysis. In Life History and Narrative (ed. J. Amos Hatch & R. Wisniewski) London: Falmer Press, pp5-24

Potter, J. & Wetherell, M. (1987) Discourse and social psychology. London: Sage.

Psathas, G. (1990) Interaction Competence, Washington, DC: University Press of America

Radley, A. (1993) Introduction. In A. Radley (ed.) Worlds of Illness. Biographical and Cultural Perspectives on Health and Disease. London: Routledge.

Radley, A. (1994) Making Sense of Illness, The Social Psychology of Health and Disease, London: Sage

Radley, A. and Billig, M. (1996) Accounts of Health and Illness: Dilemmas and Represenations, Sociology of Health and Illness, 18:220-240

Ricouer, P (1991) Reflection and Imagination: A Ricoeur Reader. London: Harvester Wheatsheaf

Riessman, C.K. (1993) Narrative Analysis. London: Sage

Sacks, H. (1992) Lectures on conversation. Vol. 1 & 2 (ed. G. Jefferson), Oxford: Blackwell

Schegloff, E. A. (1992) Introduction, in Lectures on Conversation, Vol. 1 (H. Sacks, ed. G. Jefferson), Oxford, Blackwell

Shotter, J. (1993) Conversational Realities, Constructing Life through Language. London: Sage

Stallybrass, P. & White, A. (1986) The Politics and Poetics of Transgression. London: Methuen.

Williams, G. (1984) The genesis of chronic illness: narrative re-construction, Sociology of Health and Illness, 6:175-200