Copyright Sociological Research Online, 1997

Opie, A. (1997) 'Teams as Author: Narrative and Knowledge Creation in case Discussions in Multi-Disciplinary Health Teams'
Sociological Research Online, vol. 2, no. 3, <http://www.socresonline.org.uk/2/3/filename>

Abstract

Narrative has been described as a universally used means for ordering experience. Although the narratives of medical teams have received recent attention, those produced by health professionals in multi-disciplinary health care teams in the course of their everyday work in team reviews and case discussions about service users have not. This paper, then, presents a discussion of an under-investigated area of narrative in the social sciences. The analysis is developed from the narratives produced during team reviews conducted over several weeks about 2 users - one a quadriplegic, the other, a psychiatric patient in a medium secure unit. The major issues with which the paper is concerned are: (i) the identification and explanation of significant differences between the narratives produced by medical and multi-disciplinary teams; (ii) the identification of a suppressed dimension (both in the literature on health care teams, and in the practice of these teams) on the management of difference in the development of complex multi-disciplinary team narratives; and (iii) how members of MD teams work with the different professional knowledges represented by their members. The final section of the paper defines team work as primarily a process of knowledge work and knowledge creation, and it discusses some of the organizational conditions which facilitate such work.

Keywords:

Effectiveness; Multi-Disciplinary; Narrative; Post-Modern; Qualitative Methodology; Team Work

Introduction

1.1

Narrative is the most common method for ordering and re-focusing experience (White, 1981). Brooks, describing narrative as inseparable from human experience, wrote, 'Our lives are ceaselessly intertwined with narrative, with the stories that we tell, all of which are re-worked into that story of our own lives. ... We are immersed in narrative' (quoted in McCloskey, 1990: p. 7). That my fieldwork with multi-disciplinary teams in health care has left me immersed in narratives should not, therefore, be surprising. Human service workers (see for example, Pithouse, 1987) within their professional worlds attend to others' narratives and create/author their own as part of their everyday work. In this domain representing a service user narratively allows service workers to present and re- define issues, to speculate or entertain possibilities, and to re-position their work and the user as their narratives develop. However, narrative is not necessarily or only a facilitative means of forming and communicating knowledge; its universal currency and assumed 'naturalness' may well suppress its problematic dimensions (Miller, 1995). Furthermore, expectations about narrative based on traditional, "well-formed" narratives may prevent recognition of the existence and role of narrative in situations, like those of multi-disciplinary teams, which demonstrate many features now recognized as characteristic of postmodern narratives.

1.2

In order to ground and develop my narrative of health professionals' narrations I have drawn on Hunter's (1991) and Atkinson's (1994; 1995) recent sociological writing in this domain. Part of the significance of their work is that it addresses an underdeveloped area of inquiry within medical sociology: medical narratives developed and exchanged between doctors during ward rounds, case conferences and in informal work settings. My interest in professional talk and its role in knowledge production touches upon but is different from Hunter's and Atkinson's. The focus of my research is not on medical narratives, but on another equally neglected area of health work, the textual productions of multi-disciplinary health care teams (Sands, 1993). My attention to the teams' narratives and their relationship to knowledge creation moves beyond issues of disciplinary transmission of medical knowledge to focus on multi-disciplinary teamwork's positioning between and working across disciplinary knowledges, and on the organizational conditions which facilitate such work. As with those medical teams involved in Hunter's and Atkinson's research, so the multi-disciplinary teams in my research were also engaged in extensive processes of narration; in contrast to Atkinson and Hunter, my work is located in current narrative theory, a position which has much to offer given the discursive nature of the verbal and written texts which these teams authored/produced.

1.3

The first section of this article presents my reading of key issues in Hunter's and Atkinson's work. The second section opens with a brief account of the multi-disciplinary teams involved in my research. It presents my analysis of team narratives about two users developed by two teams over several weeks. The final section moves onto a discussion of knowledge creation and the organizational conditions necessary to facilitate such work.

Narratives of Medical Narrations

2.1

One reading of Atkinson's and Hunter's work is that it underscores the functionality of medical narrative. Both writers present narration as a critical factor in the process of evaluating and weighing an often extensive range of (competing) complex medical evidence when decisions about patient management are being made, and in the organization and transmission of medical knowledge.

2.2

Although these two writers' analyses are concerned with the function and the organizational shaping of narrative, Hunter's emphasis is on interpretation while Atkinson's (here working from ethnomethodological approach) is on the complex, skilled and situated work associated with the production of medical narratives, an area of work which is ethnographically insufficiently mapped and delineated. In her account, Hunter defines parsimony as a distinctive feature of medical narrating, arguing its importance in sense-making (sifting out the unnecessary or extraneous detail) and suggesting that achieving the ability to produce (appropriately) parsimonious narratives is a critical part of professional development. In commenting on the processes of transformation of the patient's narrative into a medical narrative, she argues that this process of re-working reflects the different narrator positioning of patient and doctor - the patient's story being told from a position of being subject to a medical condition, the doctor's story being told from a position external to that condition but including an obligation to resolve or attend to that state because of the possession of relevant expert knowledge. The resultant highly purposive medical account, Hunter maintains, develops in relation to strict criteria of inclusion and exclusion. Its structure represents psychological as well as professional needs. 'The achievement of the regular, patterned, self-effacing plot of medical narrative is to control insofar as possible the subjectivity of its observer-narrator and the variables of its telling' (p. 62); further 'the narrative organization of the case is shaped by the physician's quest for an understanding of the patient's illness .... the plot not only reveals to its audience the meaning of the puzzling events it recounts but is also the narrative of discovery of that meaning' (pp. 65 - 66). The defect of such narratives is that the 'tradition works well in difficult or puzzling cases ..., but it is insufficient for long-term patient care .... chronic illness and dying tend to be "uninteresting" to many physicians ...'(p. xxi). While highly medically functional, these narratives are nonetheless problematic because of the marginalization or absence of patients' life- worlds. Hunter suggests that the development of 'enriched' (p. 166) narratives incorporating patients' subjectivity may become more necessary, particularly in the areas of chronicity and aged care.

2.3

Hunter's work, then, is grounded in a long-standing preoccupation in medical sociology and anthropology (see, for example, Good, 1994^[1]) about the lack of attention given in medicine to the life-world of the patient. Her foregrounding of the interpretive dimension of narrative, however, opens the possibility of a postmodern reading. Her concern with interpretation, and her identification of parsimony, inclusion and suppression as shaping factors in the composition of medical narratives questions the concept of narrative as 'reporting' or revealing the world as it really is and challenges the assumption of a high degree of correspondence or mirroring between the event and any narration of that event (Miller, 1995). What is produced is always a representation of that event. Current theories of representation emphasize how a representation of a person, event or situation can never be complete because it is a function of the perceiver's position or location in time, space, culture and discourse. Moreover, narratives are linguistic events. Narrators do not speak from a neutral or objective position but from within language, which is embedded in power relations, shapes thought and perception, and provides models for the composition of appropriate narratives^[2] (Kritzman, 1988; Nicholson, 1990; Clifford and Marcus, 1986; Smith, 1990; Trinh Minh-Ha, 1992).

2.4

Atkinson, in his valuable but somewhat celebratory analysis of medical narratives seeks to foreground that which has been sociologically 'backstage' (1994: p. 118) - the complex professionally acquired skills involved in selection, differentiation and evaluation of evidence from different sources in medical settings, and the rhetorical structures underpinning the resultant narratives. In particular he posits that 20th century medicine produces a multiply sited body (Atkinson, 1995: p. 89), a body produced/appropriated by a range of disciplines, each with different methods of measurement requiring particular knowledge for their interpretation. This body, then, is a:

...series of representations. These representations are themselves dispersed in time and space within the complex organization of the clinic. They are inspected, interpreted and reported by different cadres of specialized personnel. ... , one should think of the modern clinic as producing a disembodied body. ..., a body ... divorced from the body of the patient. The body may therefore be read at different sites ... It [the patient's body] is, however, one among may possible versions of the body that may be assembled in the modern clinic. The patient thus may have a multiple existence within the clinic. ..., the various fragmentary aspects of the patient and his or her body are brought together under the auspices of the case. (Atkinson, 1995: p. 89).

2.5

Describing and theorizing the production of this complex, disembodied body, however, requires more attention than has hitherto been given to the analysis of a number of key, core activities 'in the accomplishment of medical work' (p.37) that occurs away from the patient, and it is to this task that Atkinson turns his attention. His analysis focuses on the processes by which intricate and multi-faceted medical knowledge is articulated in everyday practices through the development of 'plausible accounts' (p. 90), whose rhetorical structure is responsive to the organizational context and confers credibility to the narrator. Plausibility, he suggests, is accomplished through attention to relevance, through the employment of a number of rhetorical features (such as 'hedging' and 'shielding' (Atkinson, 1994: p. 126)) and through the production of a 'good account'. The distinguishing features of a 'plausible' account are that it: (i) has sufficient detail to allow the audience to follow the chronological development of the illness; (ii) is 'sufficiently eventful' to enable the audience to reconstruct the most significant findings and follow the management of the case; (iii) provides evidence of clinical reasoning and decision-making; and (iv) demonstrates professionally appropriate grasp of detail, thus providing evidence that the clinician had grasped the case, and that the selection of detail is related to others' 'need to know'. (Atkinson, 1995: p. 97), a model not without its parsimonious dimensions. He also argues that many clinical medical presentations follow Labov's evaluative model of narrative events, having the following dimensions: abstract, orientation (locating in time and place); complication (what happened); evaluation (how speaker views the events; conveying the point of narrative); result (resolution); and coda (optional- closing summary) (p. 103).

2.6

However, there is some tension, on which Atkinson remarks, between the formality of Labov's model and his (Atkinson's) comment on the incompleteness of many medical narratives - these are often provisional and open to revision and negotiation, and they disrupt the Aristotelian unities of actor, duration and setting underpinning Labov's model. Since the theoretical and textual implications of such disruptions are significant it is worth quoting White (1981: p. 1) at some length on the problems with traditional narrative forms. He wrote:

I have sought to suggest that this value attached to narrativity in the representation of real events arises out of a desire to have real events display the coherence, integrity, fullness, and closure of an image of life that is and can only be imaginary. The notion that sequences of real events present the formal attributes of the stories we tell about imaginary events could only have its origin in wishes, daydreams, reveries. Does the world really present itself to perception in the form of well-made stories, with central subjects, proper beginnings, middles and ends, and a coherence that permits us to see "the end" in every beginning? Or does it present itself more in the forms that the annals and chronicles suggest, either as mere sequence without beginning or end or as sequences of beginnings that only terminate and never conclude? And does the world, even the social world, ever really come to us as already narrativized, already "speaking itself" from beyond the horizon of our capacity to make scientific sense of it? Or is the fiction of such a world, a world capable of speaking itself and of displaying itself as a form of a story, necessary for the establishment of that moral authority without which the notion for a specifically social reality would be unthinkable .... Can we ever narrativize without moralizing?' White (1981: p. 1)

2.7

The significance of this quotation is twofold: first, expectations of what constitutes 'narrative' need to be flexible (Nash, 1990) and may not accord with formalist aesthetics of narrative; second, narrative cannot be separated out from issues of representation and power, albeit that that power may be productive and not necessarily repressive or prohibitory (Kritzman, 1988). As Miller (1995) has noted narrative contains within it a policing function; it is embedded in dominant discourses. Equally, one of the cultural functions of narrative is to critique the dominant representational practices sustaining the existing social order. This paradoxical positioning places narrative in an interrogatory role. Narratives themselves, through their capacity to represent/ re-present difference, become vehicles for change (albeit that 'change' cannot necessarily be read as progressive or linear). Highlighting the possibility of challenge to the existing social order, they demonstrate its tenuousness and its potential for re-negotiation (Mumby, 1993).

The Narrative Productions of Multi- Disciplinary Health Teams

Scene Setting

3.1

I draw here on the work of two of the three ward-based teams participating in my current research project on the practices of multi-disciplinary teams working in different domains of chronic and long-term care in different Crown Health Enterprises (CHEs) in New Zealand. One team provided assessment, rehabilitation and long-term care for people with severe physical disabilities; the second worked with committed psychiatric patients on a medium secure unit. The third team whose work is not directly represented here for reasons of space worked on an assessment, treatment and rehabilitation ward for older people.

3.2

I undertook the research with teams working with these particular service users for several reasons. The users have tended to be seen as unrewarding to work with because much of the work is on habilitation, rather than rehabilitation. Their conditions may often only improve gradually, not improve, or deteriorate. Their needs are often complex and require other than a medical response - their illnesses are as much a social and emotional event as they are a medical one, and their care needs to be coordinated over a number of different disciplines. There are, then, often no 'fast' results, contact with some of the service users is likely to be on-going, and the complexity and intractability of some of their situations can leave health professionals feeling powerless, or that as a group they are not adequately progressing the user. Since the provision of a holistic response to care issues is a major dimension in the rationale for multi-disciplinary team work, and because the work raises issues of how the team manage 'stuckness', teams working with these populations seemed very appropriate to involve in the research.

3.3

The teams were responsible for in- patients and community-based users with a very high overlap of personnel across the community and ward-based teams. Both ward teams had been set up as multi- disciplinary teams. The physical disabilities team had been established 10 years ago; the psychiatric team, five years ago. The members of the first team were: two occupational therapists (OT1 & 2), a speech language therapist (SLT), a ward-based doctor (Dr), a rehabilitation specialist (RS), a senior medical clinician (Senior Dr) , the unit manager and clinical nurse specialist (CNS), a registered nurse (N), a social worker (SW), a psychiatric registrar (who occasionally and briefly attended the meetings), a senior community nurse (SNT) and a locum physiotherapist (11-12 people). A dietitian joined the team towards the end of my fieldwork. The majority of this team had had considerable experience in the field, and five (all women) had been in the team for over two years. The only males in the team were the rehabilitation specialist and the senior clinician who alternated attendance at meetings because of other commitments.

3.4

The second team was larger, consisting of a psychiatric consultant (CP), two psychiatric registrars (later reduced to one) (PR1 & 2), a consultant clinical psychologist (CCP), a senior clinical psychologist (SCP), a psychology intern, two senior social workers (SSW1 & 2), a social worker (SW), the Unit director (UM), her deputy (DM) (both nurses) , two clinical nurse specialists (CNS1 & 2), a psychotherapist and a part-time occupational therapist (14-15 people). However, some meetings had up to two additional members, and different nurses came and went in the process of presenting patients for whom they were the key worker for that day. While the psychiatric team had more original team members than did the other, it had also had had a number of recent staff changes among senior clinicians, including three recent changes of the consultant psychiatrist. However, the team collectively had a considerable amount of experience in their various fields. In gender terms, it was more balanced than Team 1. Women held the positions of senior clinical psychologist, one of the senior social work positions (SSW2), social worker, the unit director and deputy, one of the clinical nurse specialist positions (CNS1) and occupational therapist (OT).

3.5

Neither team had a clearly identified leader. In Team 1, the role of chair was rotated weekly through non-medical staff; the chair's responsibilities were defined almost exclusively in relation to timekeeping. In Team 2 the Unit Director tended to take responsibility for time watching and, along with her deputy, to try to ensure the team addressed patients' requests.

3.6

Both teams enjoyed good reputations in their fields, and members brought more rather than less experience of working in teams (although some, especially in the psychiatric team had had very negative previous team-based experiences). Only the psychiatrists, psychologists and the social workers had had any training in team work. While the extent of that training varied very considerably, the focus in all cases was on group dynamics and management of interpersonal relationships, a focus which Mohrman and Mohrman (1993) suggest is not necessarily productive. It is also pertinent to note that since 1988, the New Zealand health system has been through two major re- organizations, the second of which sought to make substantial changes to the delivery of health care (Salmond et al, 1993), and that individual CHEs have had different degrees of financial problems and continuing internal re- structuring, leaving remaining staff feeling highly insecure, demoralized and disaffected. Such organizational pressures are extremely likely to affect the development of team work.

3.7

Despite the recognized complexity of teamwork (a substantial literature addresses particularly the interpersonal pitfalls), the absence of any team protocols, including reviews to enable the teams to evaluate their work, the taken-for-granted (ie. unproblematized) statements in job descriptions about participation or contribution as a team member, and the lack of resourcing of , or reward systems for, team work suggest that organizationally the CHEs did not understand their role in the development of team work, although giving lip service to its value. A few team members recalled brief discussions at job interviews about the importance of taking an active role in the team. My impression was, however, that their interviews had concentrated primarily on issues of disciplinary competence and experience and, at two of the three sites where I undertook the fieldwork, it appeared that members primary allegiance was to their discipline rather than to the team.

3.8

Since the focus of my research is to produce an analysis of team work, my primary data gathering has concentrated on the audio taping of team reviews and case conferences held to discuss clients' progress and issues and to formulate on-going plans, and meetings between the teams and the users and their families. I attended over 40 such meetings, in the course of spending 8 - 10 weeks at each site. The ward meetings were in all instances significantly longer than the community meetings^[3]. Team 1's meetings lasted about 90 minutes, during which time they reviewed up to 16 patients, eight of whom were long-term residents on the ward. Team 2's were significantly longer, lasting generally between two and two and a half, hours, and once, nearly three. They reviewed 15-16 patients, some of whom had been on the ward for years rather than months, and a few of whom had spent most of their adult (and sometimes adolescent) lives in psychiatric institutions.

3.9

Atkinson (1995) remarks on the constitution of the medical field by competing specialisms with different frames of reference. Nonetheless, that field is constituted by a range of commonly held assumptions relating to the nature of scientific evidence as informing interpretation and decision-making and by a shared knowledge base. These teams could be described as operating within a master narrative (Lyotard, 1984). In contrast, multi-disciplinary teams bring together professionals whose training calls on highly diverse assumptions and very different knowledge bases, which, as with the training within each discipline, are accorded different status by different disciplines although the discourse of team work foregrounds disciplinary equality and defines competence in moral terms (Burgoon et al, 1988).

3.10

Such teams, then, are grounded in heterogeneity, where each discipline provides the means to compose a fully plausible account or, in other words to tell the 'whole' story of the user; in practice only fragments of that fuller knowledge can be and are presented in team meetings yet it is from working with these heterogeneous fragments that the team is required to develop a plan. The intersection of disciplinary knowledge or knowledge systems, a requirement in a multi-disciplinary context, also complicates the application of 'parsimony' in any evaluation of the efficacy of team discussions since the definition of an 'extraneous' detail is a disciplinary, not a self-evident, issue. In an MD team, members need to talk through their different perceptions of the user and their situation as part of a necessary process of displaying how each discipline shapes knowledge and determines action. Rather than regarding redundancy as an irrelevancy, it can be regarded as necessary to this mode of team work. These teams, then, were not concerned with the internal transmission and learning of disciplinary knowledge but (ideally) with the intersection of different knowledges in order to develop an appropriate care and management plan for and with (many) long-term patients. Instead of typically foregrounding a detailed chronology of the illness/condition and its management, they focused on their work in relation to the demands of long-term care, quality of life issues, and the provision of holistic care.

3.11

Relevance can become a contested issue in an MD context. Interviews with members pointed to an absence of shared agreement about what was significant information and the purpose of the team's discussions in light of the teams' responsibility to achieve positive outcomes for users. One of the consequences of chronic illness or long-term care is the users' body may become very vulnerable to an unnecessary exposure to the team gaze. Extensive discussion of elimination processes and toileting problems is arguably unnecessary in a multi-disciplinary meeting. The discussion, too, can move too easily to an evaluation of users', or their families', moral worth. Situations where staff members used the meetings to develop detailed and frequently unchallenged negative narratives about users highlighted the problematic nature of representations of users developed in team work and teams' lack of reflexivity (Opie, 1977). In this context, accessing the concept 'relevance' acquires not just a functional, but also ethical dimension: protection against violation of privacy through unnecessary or ethically questionable exposure.

3.12

This suggests that defining team work (as did most members and their organizations) primarily in terms of contributions of information is a somewhat simplistic account of a more complex process. In recognition of how contributing some information about service users gained in the course of her work with them may result in the formulation of representations of the users with which she was unsympathetic, one social worker remained substantially silent about her work with users. Because the team had no structure which enabled it to discuss how it carried out its work, it was therefore unable to discuss how different information contributed to different representations of users which affected the orientation of the team's work. The social worker's silences were misinterpreted by some other members who believed she should be contributing more (but did not say so).

Presenting the Teams' Reviews

3.13

Because the full texts were too long to reproduce in the body of the article, the original version of this article which I submitted to Sociological Research Online included quite detailed summaries of the case reviews of the two users. Producing what were, nonetheless, substantially abbreviated and somewhat crude representations of the team discussions was clearly problematic. One of the reviewers' recommendations was that the full discussions about these users be made available using a hypertext link, either to text or to audio tapes. Incorporating the audio tapes into article presented practical difficulties in light of the need to edit out real names and change aspects of the users' circumstances to preserve confidentiality. There were also ethical problems, because the ethical approval given the project over two years ago did not take the development of electronic journals into account, and therefore the ability to include participants' voices, not just transcribed material, into an article. For these reasons, the text of the discussions is available in written, not oral, form.

3.14

My editing of the hyperlinked texts has been informed first by the need to reasonably protect the anonymity of the service users and their families, I have changed some of the users' material circumstances. Doing this without having to substantially modify sections of the discussion was a relatively simple matter in respect to one user. The circumstances of the other user, though, required more alterations. Since these texts are highly contextual, my alterations may have caused some incoherence which was absent from the original discussion.

3.15

Secondly, my editing practice has been informed by the demands of working with oral texts, texts which create practical as well theoretical issues. Audibility, a factor in audio taping group meetings or discussions, became much more of an issue because of the environments in which the teams worked. There were the penetrating and frequent noises of vacuum cleaners moving up and down the halls, zip heaters being filled, trolleys passing the door of the meeting room, steel buckets being dropped, shouting and, on the medium secure ward, doors being locked and unlocked. I have not marked the external noise factors on the transcripts because simply noting the presence of noise is very different from having that noise as a constant in the background. In addition to these noises, the teams contributed their own problems of audibility -rustling papers, talking simultaneously, laughing and chatting to each other. I have marked especially the overlapping talk because it exemplifies team practices.

3.16

The audibility of some members was also a problem when their voices trailed off or they spoke very softly. Transcribing the speech of two members in particular presented considerable difficulties, and one of these was largely inaudible during the meeting, not just on tape. I have included what I could transcribe while indicating the sections (and their approximate length) which were inaudible. These features at times contribute to a certain incoherency in the transcript, which is less marked on some sections of the tape, where the flow of language assists in sense-making. This factor and the noise contamination can affect issues of interpretation. For instance, the minutes of one meeting refer to a decision that the psychiatric registrar see the user. Although there is no reference to such a decision in the transcript of the previous team meeting, such a decision could have been made but been inaudible on the tape.

3.17

The edition I have produced of these transcripts is certainly not the only one that could have been produced. I have not edited the texts to produce what would be an acceptable written text. As is clear from these discussions, team work is not conducted in well-formed sentences. These speakers hesitated, changed direction, did not complete a train of thought, or were interrupted by others; some frequently emphasized words although it is not always clear why a particular word is given the weight it is. Producing a more polished text which cut off access to the 'shape' of the discussions, and turning such speech into good writing would have involved substantial editorial licence. Moreover, team work is conducted as an oral exercise and it is an oral 'text' of which teams (and researchers in team work) must make sense.

Team Reviews Transcripts

3.18

My decision to produce a more complicated and less elegant text requiring active reading is further grounded in the recognition that the peculiarities of spoken texts can be regarded as "good speaking rather than bad writing" (Tedlock, 1983: p. 6, his emphasis), and their possibilities explored. Fabian too has noted how the reproduction of oral texts into 'straightforward writing' has the tendency to result in an impoverished text because of the suppression of evidence which would enhance the very phenomena which the researcher is intending to analyze or "de-scribe" (Fabian, 1993: p. 85). For example, there are at least two ways of editing one utterance. In contrast to Version 1, Version 2 suppresses the community nurse's sense of urgency and frustration and her desire to galvanize her colleagues (and Henry's wife) into action across a number of different areas.

Team Reviews as Narrative

3.19

Multi-disciplinary team reviews, as I have argued, are distinguished from accounts of service users spoken within the context of meetings held by a single discipline by their heterogeneous sources in different disciplinary perspectives and the fragmented representations of knowledge which result from the flow of conversation among the various members of the team. My purpose is here is to show that, in this situation, the most common mode of representing knowledge adopted by team members is narrative. Because the texts which result from this shared discursive activity are not "authored" by any one member of the team, and yet are the sum of all the team members' attempts to provide an adequate and purposeful account of the situation of the service user, I will describe them as "team narratives". It will be the business of this section to elaborate on the significance of this term.

3.20

The examples of team narratives which I am presenting here were typical of the way each of these teams went about its work. Their reviews of service users were not characterized, to refer back to White's (1981: p. 1) problematizing of narrative form, by well-formed accounts, a single speaker, a 'central subject', or 'proper beginnings, middles and ends'. They were not coherent. As I have noted, members interrupted each other, their voices trailed off, some spoke inaudibly. Different agendas were inserted into the conversation, issues were not developed. Many of the discussions were provisional and inconclusive. Their outcome was dependent not just on the teams' work but on a series of factors beyond the teams' control: the ability and speed of the individual body to heal; the speed of the return of tests and conclusiveness of results; the agreement of other organizations to proposals; the availability of alternative placements; the extent to which these teams' work with their clients may be compromised by other organizations' problematic input into different areas of work with the family . The teams could not 'conclude' their narratives about users too quickly - these were dictated by the events which would unfold in the fullness of time. There were no strong beginnings and endings, no easily claimed coherences, there were possibilities for the team to develop its work in various directions other than that of the service user (what of the work potentially to be done with parents or spouses?; to which 'client' should the team turn its gaze?). These stories, as White (1981: p. 23) would have it, were 'sequences of beginnings that only terminate and never conclude'.

3.21

The team discussions are the product of a distinctive situation - the reviewing of service users' progress by members of multi-disciplinary teams - and they result in the production of distinctive texts. These texts are in the first instance oral productions, generated through a mini narrative process and achieving more permanent, (ie, written) status because of the presence of a researcher. As written texts, they become available for classification, description and analysis. They were not, however, solely oral texts. For example, the Team 1 reviews were minuted and the minutes were then used to start the team discussion the following week

3.22

Although Team 1's introductions to each user were very short, they demonstrated a multi-disciplinary authorial process. Each introduction, which focused largely on different tasks that were to be discharged with particular reference to medical, nursing and paramedical concerns, was used to focus the team on its direction and immediate treatment goals (hence attending to issues of accountability), and constituted a partial summary of members' input the previous week. The processes of compilation of the minutes were also multi-disciplinary, with the note-taker role circulating between all non medical members. (Medical staff did not take minutes as, with one exception, they did not attend all meetings because of overlapping responsibilities). Within Team 1, however, the activity, minute-taking, was taken-for- granted and not understood as invoking an authorial role involving processes of selection, like foregrounding, suppression, or exclusion of aspects of the team's discussion, nor, as a result of these features, as weighing subsequent input and so orienting the team's gaze and structuring its work. As is apparent from the transcripts, significant dimensions of the team's discussion were recorded in a way which did not well reflect the discussion or were omitted.There was no procedure to enable the team to attend to the adequacy or fullness of the notes' representation of the team's work or discussions, or a place where the role of the notes and their appropriate structure could be discussed.

3.23

In Team 2, the review of each user was initiated by a nursing report read by the user's nurse for the day. This report was a written summary of what the key worker had defined, in reviewing the week's nursing notes, as the main issues for the user over the past week. Some non-nursing team members were dissatisfied with this mode of introduction because it was so explicitly discipline focused and appeared to contradict the multi-disciplinary mode of conduct of the team. Team 2's introductions were more complex and extensive than those of Team 1, although the comments about the user, Pat, whose reviews inform this article, were brief in relation to some introductions of other users. Each summary was presented under headings: current issues, physical, mental, management, social and family, and patient's requests. Three of the summaries about Pat explicitly requested team discussion on particular issues or identified the need for the development of management strategies. Overall, however, the summaries offered little in terms of team or organizational accountability.

3.24

These summaries, the starting point of the weekly reviews, can be described as 'narratives' because they represent in a conventional, shared mode some team members' knowledge about and experience of users in order for that knowledge to become available to other members. They possess the primary narrative components of events, actors, and time (future, past and present); they include inference and speculation; they are expository. Together with the discussions which they generate, they contribute to the developing 'story line' about each user which can be thought of as composed by the team as a corporate author. Making their shared and unique knowledge explicit in this evolving story about the service user is a fundamentally important means by which team members affirm, question and re-focus that knowledge, and by which they can make manifest the team dimension of their individual work. In defining such accounts as narratives, the conventional association of narrative with literary or fictional accounts of events and people, and with complex stories of loss, violence, courage, determination, intransigence, marital breakdown and family problems, played out against an organizational backdrop, has to be put to one side. Rather, the crucial factor is that the health professionals who have produced these texts in a multidisciplinary team setting have habitually used the distinctive features of narrative representation to organize and communicate their knowledge, including selection (and exclusion) of factual material and interpretation of that material. A brief account, 'We had a lovely meeting', contains the basic narrative elements of place, time, actors and point of view, and is as much an act of interpretation and representation, however abbreviated, as is the psychotherapist's much more elaborate narrative of an earlier meeting.

3.25

The language or team discourse which is used by the speakers is also significant insofar as it closely parallels that of ordinary conversation, where speakers typically engage in narrative and quasi-narrative forms. Each contributing account to the developing narrative about a service user is produced in substantially unself-conscious language, that is, the language used is descriptive, not 'analytic' in that it does not particularly overtly engage with concepts or seek to integrate, examine minutely, or critique either the individual or team narrative about a specific user. Each team narrative is contributed to by various team members at different times, who have different experiences and perspectives of the service user. These narrators at different points complement and contradict each other's narratives, creating competing representations of the user (or, in Team 2's narrative, of informal caregivers).

3.26

To note the largely unanalytical mode of the teams' discourse is in no sense to suggest that the narratives lack purpose. They are not spoken by a random collection of individuals or a group of professionals chatting about users over a drink, but by a group with a specific organizational task to develop and modify care plans and monitor users' progress in the light of two organizational objectives: achieving an improvement in the users' quality of life through attention to those areas of their condition that can be modified or adjusted, and working towards their eventual discharge.

3.27

I do not intend to suggest that the discussion should from the outset be conducted in a more 'analytic' or conceptual mode, nor to privilege narrative above concept. Each mode of discussion generates and is dependent on the other. The processes of selection, organization, and interpretation which are consistently at work in the composition of these texts, convert individual team members' experiences into shareable knowledge. The significance of a team narrative is that it allows a mapping of multiple interpretive possibilities, enables unpredictable connections to be made, and generates alternative frameworks. More explicit conceptualization occurs when team members introduce disciplinary knowledge to interpret further or critique the adequacy of the differing narrative representations of the service user's situation. The effect of such discursive shifts is not to permanently anchor a particular interpretation but, because of the ways in which teams' narratives are shaped by the work to be undertaken, to facilitate through that (temporary) stasis in the production of narratives the generation of other narrative possibilities and modes of conceptualization. The metaphor is rhizomic rather than aborescent (Deleuze and Guatarri, 1983).

3.28

The model team narrative draws on individual members' narratives but does not necessarily mirror them; it is built up during a series of discussions and is shaped by the team's need for purposeful action; and it can always be adapted in response to changing circumstances. The team narrative may contain, with varying degrees of explicitness, reference to several interrelated but distinct professional modes of narrative formation - administrative, medical, nursing, psychological and social, and organizational - which are embellished and played out in the ensuing discussions. The rehearsal of a detailed chronology of the user's condition, while assisting in the identification of team issues, questions, and plans about engaging with a user in some situations, may not be necessary in all cases where the team has longer-term on-going contact with that user. However, the development of a team narrative may well make explicit the different representations of its clientele and of the team's role vis a vis a user which may be circulating within a team. Open identification of such differences (which is not without therapeutic implications for the user/family) in the process of generating the team narrative shows that a significant aspect of team work is the recognition of interpretative differences between team members and developing processes which allow the team to explore the implications of following particular narrative and conceptual paths in order to arrive at a workable plan of action.

The Team Narratives

3.29

I want to exemplify my argument by reference to team reviews of two service users. Henry, the user reviewed over four weeks by Team 1 had became a quadriplegic following a sporting accident about two years ago. He had come on to the ward initially for two weeks as an respite care patient. The respite programme is designed to give informal carers a break; equally, the return to hospital allows the team to re-assess users' care levels and changing needs. Part of the issues for the team was finding a different 'type' of bed (respite beds were only available for two weeks) to prolong his stay in hospital in order to further stabilize his condition. Shortly after his admission, another young woman, who had been a continuing care patient on the ward for some years, died, an event with particular effect on other continuing care users.

3.30

Pat was a committed patient on a medium secure ward in a psychiatric hospital. He had been there for over two years following a car crash where he sustained severe head injuries and where both his parents were killed. After his discharge from hospital, he had seriously assaulted a woman and then been committed to care. With his parents' death, his uncle and aunt had become his guardians. The reviews I recorded took place over five weeks.

Team Reviews Transcripts

3.31

Within the course of these reviews the teams developed a series of parallel mini- and more extensive narratives around medical, administrative, psychological and organizational themes. While at first blush, the team discussions seemed to move in what could be regarded as relatively random fashion across these themes, the way the discussions were structured can also be read as pointing to the interrelatedness of events and issues which confront the service user and the team in the course of its range of work with that user. Neither team imposed a strict order of speaking. Although there were no overt rules defining particular areas as the prerogative of particular disciplines, nonetheless, in Team 1 non-medical staff rarely commented on medical issues (such issues were less dominant in Team 2). Each narrative, though, had the potential, if not the actuality, of a heterogeneity of voices/disciplines contributing to its development.

3.32

Nor did these narratives follow a pattern of linear development. Thus, one initial confident narrative of improvement ('his back has improved') was replaced by a more cautious one, constituted by a commentary which moved across different although substantially medical, aspects of Henry's care and his healing processes. This narrative contained moments of uncertainty, or challenge ('What are we going to do'?), hypotheses which were affirmed or rejected, and questions and statements built on observation and past knowledge of the user. Although Team 1's particular medical narrative finally achieved (for the time being) a successful closure, Team 2's considerable confidence at the beginning of the reviews presented here that Pat was going to be transferred to another unit was replaced by their developing awareness of the complexities and uncertainties of inter-organizational issues and funding complexities. As a result of these, the team had to return to their earlier narrative about their own organizational roles and responsibilities.

3.33

Both sets of reviews have examples of abrupt transitions between narratives, where narrative possibilities are left unexplored and where issues that could bear significantly on the development of work remain marginalized. If part of the function of narrative is to assist in a re-working of significances and a (re)- assessment of direction, then these examples point to a seeming difficulty of disrupting and re-orienting the team's gaze through an unreflective narrative process alone. Just as the highly pertinent possibility of Pat's recovery of his memory goes unremarked in relation to Team 2's on-going narrative about how they discuss the problem of his uncle's asking him questions about his memory, so too, the discussion about the team's consistency in its work with Henry is left hanging. There is no clear evidence from the transcript that the team as a whole has subscribed to RS's provisional plan for action. The team scribe fails to identify either the principle or plan as significant 'events' in the team minutes, as is evident from the minutes read at the outset of Henry's fourth review. The question of how the team responds to Henry's psychological needs is converted into a decision that the marginal and absent psychiatric registrar (the registrar changed very six months, this one did not attend all meetings and stayed only briefly, and she was very unclear of her role), is to 'see' Henry. There is, however, no team discussion about the rationale for her involvement or by whom or how she is to be briefed, or of which of the different narratives about Henry circulating in the team she has been apprised.

3.34

In the remainder of this section I want to attend to a related but different issue, that of the competing, indeed polarized, representations generated within the teams about Henry in the one instance and Pat's guardians on the other. That the teams should generate different representations is in itself not surprising, as representations are informed by members' positioning within a team and by their discipline. For example, each discipline's role brings them into contact with users in a manner which is different from that of other disciplines; further, each discipline offers different explanatory models of human behaviour and foregrounds different issues as significant in relation to the work to be done. What is then important is how members elicit, identify and then work with these differences articulated through narrative to achieve a point of provisional closure (a team plan) which then generates further narratives to inform the development and modification of that plan.

3.35

The representations of Henry within Team 1 led to two polarized narratives, each of which drew on competing models of human nature. One, generated by the ward doctor (and one to which she and others returned subsequently in their description of Henry as 'crafty') was of Henry as highly manipulative, aggressive, lacking in insight and emotional control, as deserving to be treated as a child and an organizational nuisance to boot (It is, too, Henry's relationship to the organization which is recorded/repeated in the minutes of the next meeting).

3.36

The competing representation is first articulated by the nurse. It is then picked up and emphasized by the social worker and the community nurse (who moves the narration of Henry's problems beyond the organizational confines). This representation is of a man who is caring and concerned about his fellow ward mates, who is resourceful, and who has made considerable, although unsuccessful, efforts to resolve his personal problems. This narrative about Henry's behaviour finishes extremely abruptly with a return to the (more controllable?) medical problems which continue to dominate the remainder of the team's discussions. At no point does the team formally 'note' the existence of these differing narratives or begin to discuss how such accounts may differently structure aspects of the team's work.

3.37

The second discussion which I wish to attend to more fully is Team 2's complex and lengthy narrative which surfaces throughout the reviews I recorded about Pat's guardians, their relationship to the team and their input into Pat's care. From the way in which the psychotherapist (Miles) constructs his initial story of the meeting, from some comments made to me by team members outside of the meeting, and comments through the reviews it would appear that the dominant team narrative about Pat's guardians is thematized by difficulty, obstruction and a failure to appreciate others' input into Pat's care. In order to open this narrative of 'difficult relatives' (a common enough narrative in human services) to scrutiny and revision, Miles engages on several occasions in some complex work, the process of which reflects his disciplinary orientation, in his delineation of the relevant 'facts', his placing these in different and increasingly intricate relationships to each other as more become available, and his seeking to avoid moral strictures. Further, I would argue that the detailed way in which he goes about his narration can be read as demonstrating a method of narrative construction in which interpretation and closure are suspended for as long as possible in order to encourage the team to identify the problematic issues and to collectively bring these to a provisional point to allow the team to identify a course of action.

3.38

Although Pat has been on the ward for some time, so it could be assumed that most of the team are familiar with his history, Miles both notes his own unfamiliarity with the details of the story (so these could be unfamiliar also to others), particularly the guardians' story of their centrality to Pat's rehabilitation. His full and informative account is significant in a number of ways. He makes a lengthy statement in the context of meetings where members typically comment briefly. He speaks slowly and with deliberation. He takes up team time. His emphases are important - the tiny movement of the knee as that to which the guardians responded and which gave them hope, the assertion of rehabilitation staff about the impossibility of Pat's recovery and the guardians' refusal to accept their verdict, their huge significance as advocates, and Miles' crucial identification of the issue of their positioning - to the team they may be nuisances, but someone in Pat's situation is critically reliant on such determination and intransigence.

3.39

His narrative, then, offers the guardians a degree of credibility that it would appear the team has not afforded them to date. And the story is not that simple. His account is immediately problematized by the manager, who then, and later, produces a more rigid narrative about the guardians which highlights their psychopathology. In contrast, Miles' response is to define the dynamics of the situation as a 'fact of life', as a given which (formal) carers will need to understand and manage; and then he himself introduces a further complexity in a speech which highlights his positive contact with the guardians' and a further problematic which bears significantly on the team's work - the guardians' apparently puzzling denial of Pat's role in the attack. But the way in which he does this is important. He persistently foregrounds issues of interpretation of the event rather than the moral character of individuals, in part through his use of the word 'interesting'.

3.40

While it could be argued that the success of this particular narrative strategy in getting the team to attend to its competing perspectives open to question^[4], what the discussions foreground as a significant issue in team work is the concept of undecidability. It is not possible to summarily dismiss as 'wrong' the competing representations of the user and guardians, and the different narratives to which each gives rise. Each representation has evidence to support it. The issue facing the team is how they engage with these different representations and the narratives which they inform, the implications of adopting or working within one or other narrative framework, and the relationship of that engagement with knowledge work and creation.

3.41

Team work assumes that the group thinks of itself not just as a group of individuals coming from different disciplines but as individuals actively bringing those disciplines to the development of work with users. It assumes that members have the language and ability to develop, not only mini- narratives about users, but the conceptual and linguistic skills to respond to the concept 'interesting' and to develop and analyze their own and others' narratives. It therefore bespeaks a familiarity with one's own tacit disciplinary knowledge and a willingness to make such knowledge overt. It also suggests an ability to work with the different implications of this or that representation and narrative focus as part of the process of the team's achieving provisional closure and an orientation for the next focus for its work, a focus which in turn will develop other narratives and different provisional closures.

3.42

The rhizomic model of team work, then, is of the generation of different narratives moving across a surface in an articulation of possibility and their (temporary) meetings at nodal points, in a dynamic process which foregrounds the post-modern concepts of provisionality, incompleteness and undecidability. But, because these teams are operating in the 'real' and not fictional world, they need to be able to develop a plan of action. This foregrounds the twin issues of working across different knowledge domains and the organizational structures which facilitate such work. It is with these aspects that I wish to conclude this paper.

Knowledge Creation in Team Work

4.1

The following account draws heavily on Nonaka's and Takeuchi's (1995) work on organizations and knowledge creation. At the risk of producing an overly simplistic account, the points of their complex analysis which I wish to emphasize are:

1. It is no longer sufficient to foreground knowledge as a significant organizational resource. Instead, the organization needs to engage, through the development of processes and structure, in knowledge creation. This involves regarding knowledge as other than formal and easily transmitted. The re-working of diverse knowledges requires the (difficult) articulation of tacit knowledge (on both its craft and cognitive dimensions); the questioning of one's own knowledge; the presence of productive, because unsettling, ambiguity; the presence of redundancy which enables the identification of shared ground and of difference; and the arrival at a different mode of sense making.
2. 'Knowledge' is not synonymous with 'information'. Rather it is 'the function of a particular stance, perspective, or intention. ... knowledge, unlike information, is about action. It is always knowledge "to some end". And ... knowledge , like information, is about meaning. It is context specific and relational' (p. 58). Information is the material enabling the creation of knowledge; knowledge enables the development of the new position, the achievement of a dynamic truth. Moreover, this positioning of knowledge as active corresponds to Miller's comments on narrative, 'A story is a way of doing things with words. It makes something happen in the real world' (Miller, 1995: p. 69).
3. Knowledge is the interaction between explicit or formal knowledge and tacit knowledge, defining tacit knowledge in Polyani's terms as "in-dwelling" (p. 60). This interaction depends on several processes - those of:
   1. socialization , ie. the sharing of experiences informing tacit knowledge;
   2. externalization, ie. the articulation of tacit knowledge into explicit concepts, a process involving dialogue and reflection;
   3. the development of combinations, ie. the combining of different modes of explicit knowledge; and
   4. internalization, ie. the translation of newly acquired explicit knowledge into tacit knowledge bases to inform work. (pp. 61-72).

4.2

In light of this discussion, part of the significance of the transcripts of the team reviews is the absence of the processes outlined above in the teams' discussions. In relation to Pat's guardians, for instance, if discussing the question of Pat's involvement in the assault with his guardians was necessary, then the team's processes of reflection, dialogue, and identification of tacit and explicit knowledge could have led it to formulate a series of questions about the ways in which different disciplines would approach the issue; their cognitive, behavioural and developmental models and assumptions which underpin each approach; the strengths and limitations of each; and who should undertake this work and when. Following these lines of questioning could result in a shift from the hitherto dominant narrative of 'difficult relatives' to different and more 'interesting' (perhaps more ambiguous) narratives, contributing to different modes of engagement.

4.3

It is axiomatic that such a development cannot occur without a facilitative organizational environment. To borrow Winnicott's phrase, teams require 'good enough' organizations for them to achieve their potential to voice and then work conceptually with the different narratives circulating within the team. A critical first step is a transition from conceptualizing team work within human service organizations as primarily about the development and management of the interpersonal relations of team members, to conceptualizing team work as engaged in knowledge work. Nonaka and Takeuchi (1995) have set out the organizational conditions which promote a 'knowledge spiral' (p.70) as:

1. organizational intention, ie. the organization needs to develop a strategy 'acquire, create, accumulate' (p. 74) and use knowledge, so conceptualizing what kind of knowledge it requires and how this knowledge can best be utilized. This, then, involves problematizing the notion of team work, rather than regarding it as something to be taken for granted, and addressing team processes and organizational structures which could facilitate its development.
2. the presence of autonomy, which enables the development of new knowledge and permits individuals to work creatively.
3. the presence of fluctuation and creative chaos. These terms refer to modes of dialogue and discussion which allow for the questioning of one's own and others' premises, and encourage disciplinary overlap in order for team members to push the boundaries of their own knowledges. In contrast, in information-processing modes, the problem is defined within pre-established boundaries. Doing this 'ignores the importance of defining the problem to be solved. To attain such a definition, problems must be constructed from the knowledge available at a certain time and context' (p. 79);
4. the utilization of redundancy, ie. 'existence of information that goes beyond the immediate operational requirements of organizational members', and assists in 'intentional overlapping' (p. 80). 'Sharing redundant information promotes the sharing of tacit knowledge because individuals can sense what others are trying to articulate. Redundancy encourages concept development and 'enables individuals to invade each other's functional boundaries and offer advice or provide new information from different perspectives' (p. 81).
5. the presence of requisite variety, ie. the organisation's or team's 'internal diversity must match the variety and complexity of the environment in order to deal with the challenges posed by the environment' (p. 82).

4.4

Team work and the generation of team narratives, then, are about 'making things happen' in a complex world. Team work occurs in an environment productive of different narratives because of the heterogeneity of representations available to the team; representation of situations and issues by the user and their significant others; reporting of those representations by team members; and the team members' various representations of users and issues informed by the different disciplines within the team. The team narrative is the result of a recursive and interactive process which enables the team to explore the possibilities and constraints made available to the team through its different constituent knowledge bases (hence drawing on and working with its tacit and explicit knowledges) in order to develop a plan which allows the team to move on but where the plan is also regarded as provisional, as open to change as the users' situations change. The outcome of team work cannot, therefore, be a single, authoritative narrative or 'case' defined largely by knowledge deriving from one discipline.

Notes

¹ However, this dichotomy between life worlds and medical worlds is contested by Atkinson (1995).

² The permeability of the line between 'fact' and 'fiction' is emphasised in Korobkin's (1996) discussion of the way in which popular TV court dramas, movies and novels have come to shape the conduct of 'real' courtroom battles.

³ I have not introduced the community teams' discussions into this paper because the brevity of the review of each user. Team 2 reviewed approximately 60 users in between 30-35 minutes; in Team 1, the discussion often focused on priority setting for assessment and rehabilitation or respite beds, and the accompanying organisational logistics.

⁴ His narrative about the meeting concludes on administrative and inter-organisation issues, an emphasis picked up by the next speaker. In any group setting, what is spoken last is more likely to be taken up making it correspondingly more difficult for others to reflect back on earlier commentaries. Moreover, the issues about the transfer are of immediate concern to the team because if things work out, then the problem will no longer be their's, ie, there is a certain pragmatism about their approach. Other attempts to raise issues relating to the team's interaction with the guardians are met with a focus shift or a termination of the discussion, and although CP3 in the 5th Review, picks up Miles' phrase about the guardians' significance as advocates, he does so in a somewhat belittling manner.

Acknowledgements

The research reported on here is funded by the Health Research Council of New Zealand. A number of people have contributed in different ways to this article. Commentary from Jane Aronson and the anonymous reviewers encouraged me to reconceptualize aspects of my argument. Particular thanks, however, are due to Brian Opie whose interest, encouragement and engagement with my developing argument has been invaluable. Stuart Peters, at Sociological Research Online has been most helpful in relation to developing the hyper-links to the transcripts.

References

ATKINSON, P. (1994) 'Rhetoric as Skill in a Medical Setting' in M. Bloor and P. Taraborrelli (editors) Qualitative Studies in Health Medicine. Aldershot: Avebury.

ATKINSON, P (1995) Medical Talk and Medical Work: The Liturgy of the Clinic.. London: Sage.

BURGOON, M., HESTON, J. and McCROSKEY, J. (1988) 'Communication Roles in Small Group Interaction' in S. Ferguson and S. Ferguson (editors) (2nd Edition) Organizational Communication. Oxford: Transaction Books.

CLIFFORD, J. and MARCUS, G. (editors) (1986) Writing Culture: The Poetics and Politics of Ethnography. Berkeley: University of California Press.

DELEUZE, G. and GUATTARI, F. (1983) 'Rhizome' in G. Deleuze and F. Guattari (editors) On the Line. New York: Semiotext(e).

FABIAN, J. (1993) 'Keep Listening: Ethnography and Reading' in Jonathan Boyarin (editor) The Ethnography of Reading. Berkeley: University of California Press.

GOOD, B. (1994) Medicine, Rationality and Experience: An Anthropological Perspective. Cambridge: Cambridge University Press.

HUNTER, K. (1991) Doctor's Stories: The Narrative Structure of Medical Knowledge. Princeton: Princeton University Press.

KOROBKIN, L. (1996) 'Narrative Battles in the Courtroom' in M. Garber, R. Walkowitz and P. Franklin (editors) Fieldwork: Sites in Literary and Cultural Studies. London and New York: Routledge.

KRITZMAN, L. (editor) (1988) Michel Foucault: Politics, Philosophy, Culture: Interviews and other Writings 1977 - 1984. London: Routledge.

LYOTARD, J. F.(1984) The Postmodern Condition: Report on Knowledge (translated by G. Bernnington and B. Massouri). Minneapolis. University of Minnesota Press.

MCCLOSKEY, D. (1990) 'Story telling in Economics' in C. Nash (editor) Narrative in Culture: The Uses of Storytelling in the Sciences, Philosophy and Literature. London: Routledge.

MILLER, H. (1995) 'Narrative' in F. Lentricchia and T. McLaughlin (editors) Critical Terms for Literary Study (2nd edition). Chicago: Chicago University Press.

MOHRMAN, S. A. and MOHRMAN, A. (1993) 'Organizational Change and Learning' in J. Galbraith and E. Lawler (editors) Organizing for the Future: The New Logic for Managing Complex Organizations. San Fransisco: Jossey-Bass.

MUMBY, D. (1993) 'Introduction: Narrative and Social Control' in D. Mumby (editor) Narrative and Social Control: Critical Perspectives. London: Sage Publications.

NASH, C. (1990) 'Slaughtering the Subject: Literature's Assault on Narrative' in C. Nash (editor) Narrative in Culture: The Uses of Storytelling in the Sciences, Philosophy and Literature. London: Routledge.

NICHOLSON, L. (1990) 'Introduction' in L. Nicholson (editor) Feminism/Post modernism. London: Routledge.

NONAKA, I. and TAKEUCHI, H. (1995) The Knowledge-Creating Company: How Japanese Companies Create the Dynamics of Innovation. New York: Oxford University Press.

OPIE, A. (1997) 'Thinking Teams Thinking Clients: Issues of Discourse and Representation in the Work of Health Care Teams', Sociology of Health and Illness, vol. 19, no. 1, pp. 259 - 280.

PITHOUSE, A. (1987) Social Work: The Organisation of an Invisible Trade. London: Gower.

SALMOND, G., MOONEY, G. and LAUGESON, M. (editors) (1994) 'Health Care Reform in New Zealand', Special Issue: Health Policy, vol. 29, nos. 1 & 2, pp. 1 - 182.

SANDS, R. (1993) '"Can you Overlap Here?": A Question for an Interdisciplinary Team', Discourse Processes, vol. 16, no. 4, pp. 545 - 564.

SMITH, D. (1990) Texts, Facts and Femininity: Exploring the Relations of Ruling. London: Routledge.

TEDLOCK, D. (1983 ) The Spoken Word and the Work of Interpretation. University of Pennsylvania Press.

TRINH MINH-HA (1992) Framer Framed. London: Routledge.

WHITE, H. (1981) 'The Value of Narrativity in the Representation of Reality' in W. J. T. Mitchell (editor) On Narrative. Chicago: University of Chicago Press.

Copyright Sociological Research Online, 1997