LIFE AS FICTION
PhD by Kevin Murray
Department of Psychology, University of Melbourne, 1990
1. Introduction; 2. Narrative making sense; 3. Life manuals; 4. Life construction groups; 5. Travel space; 6. Travel and personal change; 7. Travel talk in context; 8. Narrative partitioning; References; A. Appendix
2 Narrative making sense
This chapter attempts to explicate a mode of making sense that can accommodate forms of agency. First, research in the institutional practices for regulating agency is presented. This is followed by a formal argument for narrative as a mode of making sense that allows for agency. This argument is then extended in a review of various ways narrative has been employed in the social sciences.
2.1 Sense making practices of agency
For an institution to function it must contain practices by which different levels of responsibility are distributed among its members. Goffmans (1968) observational study of the asylum provides a useful example of how this might happen. Goffman attends to the institutional practices which serve to deprive agency in the members under its care. Interestingly, these practices include not only formal bureaucratic rules that restrict the mobility of patients, but also entail the less formal regulation of conversations between staff and patients. For example, Goffman notes that information about a patient which would normally be private, restricted to case files and the patients family, is often openly circulated among staff. The effect of this is to discredit both the accounts which patients offer of their behaviour and any contribution they might make to conversations. Goffman finds that the standing of patients within institutions is largely expressed by the degree to which their words are taken seriously, and any attempts by them to assert this could be undermined by alternative interpretations supported with facts from their case history, often unknown to the patient.
In a later synthesis of his studies, Goffman (1983) proposes that the institutional regulation of agency often involves people‑processing encounters, such as job interviews, whose function is to render candidates readable according to a set of institutional character profiles. Such a concept cannot be accounted for in the social cognition perspective because there is no allowance for such a construction of information prior to its analysis. Goffman (1983. p.8) goes further to claim that the same process occurs at a less formalised level in ordinary interactions:
Everyone is a gatekeeper in regard to something. Thus friendship relationships and marital bonds...can be traced back to an occasion in which something more was made of an incidental contact than need have been.
Goffman asserts that this gatekeeper function is largely implicit, often contained in a micro dot of information. Explicating the content and format of this micro dot is a way of looking at the method of the current study. Goffmans observations bring to light the institutional processes by which claims to agency are screened: it provides the focus in which more systematic accounts of screening practices may be viewed.
Goffmans account suggests that agency is an attribute which is granted and taken away by othersthat one has little opportunity to set out on a project to increase ones agency. The possibility of creative individual acts is largely restricted to the task of saving face, such as the construction of a sad tale by asylum patients which attributes their misfortune to an unpredictable set of circumstances. Goffman excludes those acts which may be seen as active attempts to assert agency. Marsh, Rosser & Harré (1978) conducted an observational study of soccer violence which takes such acts as its focus. These take the form of tests of hazard in which the respect of others is wagered against their contempt. Physical battle is an obvious example of this; other instances include sitting for educational examinations, engaging in verbal baiting, and running a marathon (Murray, 1985a). Marsh, Rosser & Harrés study demonstrates that though agency can be given or taken away according to certain routine practices, it also can be won or lost through events that entail risk, in which one demonstrates a particular capacity. Harré (1979) pursues the idea that such tests of hazard are part of a moral order which governs the allocation of powers in the community. In this understanding of agency, one needs to obtain the respect of others by expressing ones claim through conventional means.
There is a problem with this framework: How can a persons powers in society increase when the only means available entails conformity to a set of conventions? Such a problem involves a mistaken assumption: i.e., behaviour which is seen to be governed by conventions is predictable. Harré argues that these behaviours act as proofs of autonomysituated in displays of unpredictable, seemingly random behaviour. Harré (1979, pp.255‑6) outlines the position as follows:
We shall see that the legitimate claim to freedom as randomness has a place in human life. But its force is to be located among expressive activities. Demonstrations of freedom as randomness are illustrations of the kind of person one takes oneself to be rather than amongst practical ways of acting to bring about ones plans and projects.
Random behaviour defies the attempt to predict ones actions. Harrés point is reflected in the example given earlier (see Chapter One.doc - car) of choosing the colour of a car. By their nature, expressive actions cannot be predicted and therefore become the medium for the demonstration of individual agency. The role of conventions here is marking certain spaces where random behaviours are seen to occur and providing interpretations for what happens.
Though Goffman and Harré provide valuable clues concerning the practices that mediate personal change, there is a gap in the understanding of how this change is made sense of. This is particularly critical when one considers that because agency is a negotiated power, it requires forms of representation to make its claim outside of the immediate setting in which it was gained. How is agency recorded such that a history of ones demonstrations of autonomy is sustained in the minds of others? While social cognition would propose the operation of an implicit hypothetico-deductive system in which people were represented, this renders behaviour in terms of predictable action and thus is contrary to the logic of agency itself. If not this, though, what forms serve to represent making sense of agency?
2.2 Narrative and paradigmatic modes of sense
As anticipated at the end of the last chapter, the form of making sense which accommodates agency will be separate from that posited under social cognition which grants control. In demonstrations of randomness, Harré has set aside the expressive sphere for certain representations of agency. In this section, narrative will be proposed as the means of representing the agency in the expressive domain.
2.2.1 Bruners distinction between paradigmatic and narrative modes
Bruners (1986) later work grooves a dichotomy similar to the split between practical and expressive spheres. His dualism has the advantage of outlining modes of making sense. Bruner refers to the inductivist mode characteristic of scientific explanation as paradigmatic thinking. The paradigmatic concerns how the truth is known. Its primary function is to refer to the worldboth the objects within it and the principles that govern their appearance. As a mode of thought, the paradigmatic urges consistency and predictive capacity.
Bruner sets up as a contrast to the paradigmatic, the narrative mode of thought. Rather than reflecting directly on the world, narrative thought is concerned with the experience of that world through someones eyes. There are two main distinguishing features of the narrative mode of thought. While the paradigmatic employs the indicative mode to represent the world, the narrative mode of thought is expressed in the subjunctive. One refers to a single presence in the world and the other evokes the possibilities that are contained in the world. The subjunctive mode serves the division cast in story between the landscape of actionhow the world offers itself as assisting or preventing certain projectsand the consciousness of characters within the storytheir way of going about things, their goals, their point of view. Narrative enables the conceptualisation of what could be done, as much as what is done. Unlike the scientific theory understood by social cognition, narrative incorporates a sense of temporality: the present is seen as conditioned by the possibilities of the future and the events of the past.
The second distinguishing feature of narrative is what Bruner calls implicature. This term is borrowed from conversation theorists who use it to denote the information that must be assumed in order to follow what is being said. While the paradigmatic mode endorses explicitness, narrative thinking favours a saving of information that requires a tacit understanding between speaker and listener. How implicature serves to strengthen the bond between participants in a conversation is referred to below.
From this position, it is necessary to ask if it is possible to assert a mode of making sense that is independent of scientific activity. In the terms of a competence theory (see Section ?), one could present the use of the subjunctive mode and implicature that characterise narrative sense making as inessential differences. Relative to the paradigmatic mode they may be cast as signs of a lack of both certainty and explicitness respectively. According to this view, the narrative mode is a lesser version of the paradigmatic. An alternative to this normative comparison depends on the presence of a radical difference between the two modes of sense making.
Bruner goes some way to establishing this difference by reference to the developmental conditions of mental life. He poses the question of whether the narrative way of making sense is a wired‑in cognitive property or a culturally acquired convention for communicating events. Though the manifestations of the two modes are culturally determined, Bruner presents an account of the initial inscription of narrative sense that suggests it is not. Bruner refers to the narrative mode as ...a primitive category system in terms of which experience is organised. (1986, p.18) He sees the perception of causation by six‑month old infants as integral to the later understanding of intention that finds its full development in the narrative mode of ordering reality. In a way similar to Gibsons act of partitioning, this structuring delineates a background of circumstances from out of which the story character stands. The particular disjunction that demarcates character is termed plight by Bruner. Plight refers to the disruption of the normal state of affairs that initiates the narrative action; it is when reality becomes problematised.
The narrative mode can be conceived as a process by which this problem is worked out: according to the temporal pattern of steady state, breach , crisis, and redress, the character sets out to restore some semblance of order between expectation and reality. Bruner marries the paradigmatic with the natural world and the narrative with the social (he cites the U.S. Navy adage to illustrate this: Salute if it moves, otherwise paint it.). This entails an incommensurability in the way physical and social objects are subject to thought. Bruners distinction does much to indicate the uses of language in developing modes of thought for different purposes. However, a way of distinguishing the logic of narrative explanation may be found more directly within a philosophical approach.
2.2.2 The narrative break
One project, which assumes that the narrative can be reduced to the paradigmatic, is the structuralist method of Lévi‑Strauss (e.g., 1968). Identifying the limits of this project is one means of ascertaining the space particular to narrative. Lévi-Strauss method entails the reduction of myths into a synchronic structurea set of binary oppositionsfrom which it should be possible to predict further myths. Such a project questions the association of narrative with agency; once it is mapped out into sets of oppositions the unpredictability of the agent is reduced. Ricoeur (1985) confronts this issue and presents an argument for the irreducibility of temporality in narrative forms. His method is to pursue the reductionist project as far as it will go by attempting to synthesise the abstract narrative schemas proposed by Propp, Greimas, and Bremond. Ricoeur claims that beyond the sets of oppositions, agents, quests, etc., there exists an element that resists absorption into a synchronic reading. This element is the undecidable event contained in each narrative scheme; e.g., a quest, which cannot be reduced to a predictable structure. This event opens the narrative to events outside the world of the story, and provides the possibility at stake towards which a hero can project goals and ambitions. Here the crucial event is the test, which the hero must negotiate before attaining success. Fundamental to narrative, therefore, is the suspenseful apprehension of the world: a framework where the possibilities of action that are presented extend beyond the events that occur.
This understanding of narrative corresponds to Aristotles theory of plot as a mimesis of human action. In Poetics, Aristotle distinguishes two characteristic features of plot: first, a moment of unexpected development ( peripeteia) such as a reversal of fortune; and second, some recognition of the changed circumstances, ( anagnorisis). Aristotle conceives of plot, especially the tragic variety, as a move from ignorance to awareness. This narrative schematism entails a break between event and knowledge of an event: it holds apart the act of presenting information and showing its response. Such is the break that Bruner points to in the separation of landscapes of action and consciousness.
Labov & Walestsky (1967) distinguish the same operations in everyday storytelling. They determined that the basic requirement of a narrative is that it makes some reference to an event that is combined with an evaluation of its occurrence. The sense of that event rests in the point given to the story by the audience and narrator. This evaluation reintegrates the action into the understanding from which the event is first seen to deviate. An unexpected occurrence is thus brought back into the regular order; it is re‑framed within the point of the story, the familiar moral order; (compare the story-telling coda: ...and it just goes to show).
By figuring the act of referencing within the mode of representation, such that the unexpectedness of the event is rendered with the event, the narrative mode of making sense can be seen as a different species to the paradigmatic. Yet, as a medium for partitioning event and response, narrative is more than simply a mode for the expressive sphere of social life. Narrative contains both the possibilities given to a character and the responses thus evoked. It therefore contains a partitioning between the expressive and practical spheres: what is given and what can be taken from it. It is the mode which incorporates a consciousness within the world of action.
2.3 Narrative and explanation
Having established the grounds for the autonomy of the narrative mode, it is now possible to present the reverse side of the competence theory: not how narrative may be a lesser form of paradigmatic, but how scientific modes of thought may be reduced to story forms. This section presents examples of attempts to identify narrative forms in the existing practices for determining claims about the world.
2.3.1 Scientific theories
The search for narrative constructions in the practice of science exposes breaks in the official picture. Science on its own terms provides a paradigmatic account of the corpus of laws that govern nature; i.e., a set of abstract principles that enable the scientist to predict the outcome of certain events. However, in order to fulfil its public role of making sense of the world, these laws must be communicated in a way that lends coherence, relevance, and therefore conviction to the variety of scientific work. Lyotard claims that scientific practice must resort to narrative forms of representation if it is to command the necessary respect in the community. Lyotard (1984, p.29) describes the situation ironically:
Scientific knowledge cannot know and make known that it is the true knowledge without resorting to the other, narrative kind of knowledge, which from its own point of view is no knowledge at all.
Any new development of science is seen to relate to a superordinate macronarrative which locates science in human affairs. Lyotard describes two macronarratives and their place in different cultural settings. He describes the French situation as one which grants science a role in liberating people from toil and inequalitya product of the Enlightenment project of democratisation through the spread of truth. Lyotard contrasts the emancipatory macronarrative with the more insular German tradition. Academic practices in Germany are seen to encourage syntheses, particularly in the project of speculative philosophy: knowledge justifies itself. The forms of legitimation in the French and German traditions are largely divided according to different implicit end points: the French quest for freedom and the German search for systematic knowledge. Macronarratives, therefore, are partly the generalised ends towards which specific immediate events are seen to be directed. 
Historians, even more so than scientists, seem to require the use of narrative tools in the process of constructing a picture of the past. White (1973) explored this question in the work of nineteenth century historians. Besides finding differences in the style of argument and ideologies they brought to their material, White distinguishes between historians according to the narrative schematisms they employ; he refers to these as modes of emplotment. Using Fryes (1957) categories of comic, romantic, tragic, and ironic myths, White analyses histories in order to find similar constructions of real events. White makes two points of relevance. First, White claims that these plots are culturally specific ways of constructing reality, or as he writes (1978, p.85):
...the encodation of events in terms of such plot structure is one of the ways that a culture has of making sense of both personal and public pasts.
Thus, White argues, there can be no history independent of the setting in which it is written.
More importantly for White, though, the historical treatises themselves can become objects of inquiry; e.g., how was history constructed in nineteenth century Europe? White proposes that the organising element in the histories stems from a prior poetic act that is independent of the facts themselves. White (1973, p.31) articulates the extent of this choice: In the poetic act which precedes the formal analysis of the field, the historian both creates the object of analysis and predetermines the modality of the conceptual strategies he will use to explain it. White argues that one needs to reveal the narrative schematism which grants a history order before that history can be understood.  Unlike Lyotards approach to science, White does not introduce narrative as a component of the dissemination of history in the wider community, rather, narrative intercedes in the initial act of laying out the inquiry.
2.3.3 Legal procedures
The possible uses of history are diverse, including entertainment, instruction, legitimation, and rationalisation. The practice of law, while sharing with history an official responsibility for obtaining a truthful account of the past, has a more focused end. Officially, it has two functions: judgement of guilt and allocation of punishment. Law, like science, may be viewed as an agonistically organised institution whose contests are decided largely by rhetorical power. Like science, it appears divided into two processes: the practice of showing, involving laws of evidence, respect for the bench, etc.,; and the practice of telling, governing how evidence is used to best advantage to convince the jury of its trustworthiness. Research into the practices of law is thus presented with an alternative path to the study of verdicts. What are the rhetorical resources drawn on in constructing and interpreting cases? The lines of inquiry that have branched from this question include the manner of story construction that is specific to lawyers tales (Kurzon, 1985), a conceptualisation of how precedence is taken into account (Robinson & Hawpe, 1986), and an analysis of the types of agonistic contexts that favour particular arguments (Holstein, 1985). The focus on the construction of lawyers accounts alerts one to the inevitably narrative representation of past events in the law process. Such a representation relies not only on factual verification but also the status of the narratorswearing on the Bible provides some guarantee of the narrators claims. In attempting to have the members of the jury imagine themselves in the defendants shoes, the lawyer must work on the storied quality of his account. This is how narrative may be seen to intervene in the legal process. The narrative approach within legal studies therefore presents the problem more within Lyotards terms, than Whites: it is an attempt to win immediate credibility for ones claims rather than an originary poetic act of constellating meaning.
2.3.4 Biographical studies
The focus on credibility is particularly evident in Kelleys slogan: person as scientist. This appears to promise an ordinary actor the authority normally lent to the professional scientist. The selection of an object for scrutiny ultimately indicates a belief that it is worthy of examinationthat it holds some mystery above other objects. Thus a concentration on the narrative practice by which institutions make sense of human affairs, although qualifying the official story of objective truth, still implicitly devalues the ways of making sense outside of those institutions, especially by lay actors. In this case, the examination of narrative competence among ordinary actors opens up institutional boundaries to popularist concerns. Such a push has been taken up by various European sociologists, such as Bertaux & Kohli (1984).  Denzin (1986, p.17) outlines the basis of this research: The goal of life story investigation is to reveal how ordinary people give meaning to their lives, within the limits of the freedom given them.
Like Whites study of historians, the goal of much life story research is identification of the narratives employed by ordinary actors. For example, Hankiss (1981) examines the means adopted by people to maintain a coherent life story while accounting for success and failure that is and is not expected. By relating perceived expectations to outcome she is able to identify four narrative strategies that deal with each instance. Hers may be viewed as a historiographical study in the spirit of Whites, though aimed at the lay sense of history. Its concern is how people construct their experience as participants of historical events and how they thread their life story into an historical fabric. 
The fundamental assumption of the European approach is best summarised by Gülich & Quasthoff (1986, p.228):
An experience becomes an experience by virtue of its being processed according to specific interpretive schemata. An event is a particular constellation of reality seen by a particular person as an instantiation of a particular schema.
This is an assumption that I aim to contest here. I would argue for the possibility of experience that is not housed within a narrative. Gülich & Quasthoff preclude the possibility of an experience that is not ...processed according to specific interpretive schemata. However, the processes of making sense that have been developed in this chapter contain the dual function of revealing events and evaluating them. It is this former function which an approach like Whites does not account for: it is within the revelatory moment that different information can be granted the capacity to be significant. The sorts of narrative at play in the interpretive schemata can be seen to be part of the process of evaluation. The act of perceiving events as disruptions to ordinary reality exists beyond the narratives given. Therefore what sets the narrative process in train exists outside of its mechanisms. This of course does not imply that such an experience is not eventually housed in narrative form. It merely asserts that some theoretical space is left open for unconscious experiences that exist prior to being processed into publicly conventional form. 
Though this issue does not constitute the focus of study here, it is important to make explicit the stand being taken. In this study, the assumption is that practices governing the revealing of the world are specific to ritually defined partitioning of space‑time. This partitioning enables the practices of referring and evaluating to be separate. Show and tell time in early school life, the casting of clues in a detective story, the graphics accompanying pictorial news, scene‑setting descriptions at the beginning of storiesall these exercise a duality of demonstrating and commenting on events. An understanding of making sense should be able to point to both activities. The European school of biographical research almost exclusively focuses on one: the strategies of telling a life. This ignores the settings in which these lives are revealed. It assumes that the research interview is a neutral theatre for the revealing of self. Yet the particular conditions of naturally occurring theatres, such as obituaries, television interviews, testimonials, award speeches, and gossip, must require different rhetorical conventions for the telling of self. The assumption that one can abstract the story of self from these regions is supportable only given the proposition that the biographical narrative operates simply to provide a consistent picture of the world for the subject. The metaphor of person as ideal scientist is the ground for such an inquiry. The current path, for reasons given above (see page Error! Bookmark not defined. ), is guided by the realist picture of scientist. Here the anthropological understanding of sense making is relevant. Rather than seeing the individual as the locus of consciousness, if one presents it as a product of sense making then it is necessary to take into account social context, particularly the material practices for revealing character and the evaluation of the outcome in terms of moral arguments.
2.4 Narrative and social context
The emphasis shifts in this section from the institutional forum for representational practices, to the everyday process of making sense. Given that this is closer to the focus of the current study, narrative based research in this area, particularly in the psychological sciences, will be examined more closely.
The concept of narrative in anthropology has perhaps its most notable expression in the processional model of culture, as displayed in the work of Victor Turner. Turners research is governed by a model of society as temporal construct, rather than as a static set of structures. His approach to narrative is particularly richit brings together many levels of explanation. According to Turner, narrative has its dynamic in the problems that beset the structured relations in a community. His separation of reflective and material spheres of cultural lifetermed anti-structure and structure respectivelyparallels the differentiation of expressive and practical domains. What is unique about Turners work is that it ties the activity of the expressive domain to the limits of practical life.
Turner uses narrative to formulate the processional form of what he terms social dramas. These dramas are expressive episodes in which conflicts endemic to the structured order of a community are acted out and resolved. The separation and reintegration of this episode into normal life is the function which narrative performs. Like Bruner (see page 5), Turner assumes a basic narrative progression of breach, crisis, redressive action and reintegration. Turner sees these practices as designed to fix diffuse and acrimonious conflicts into the form of a collective event, where their business can be completed through an act of public judgement. Turner not only describes these practices but also examines the ways that their scripts are learnt and reproduced. Tales such as Icelandic sagas are interpreted by Turner (1974) as providing the repertoire of motifs by which real life dramas are carried through and concluded. In Turners scheme narratives are a necessary accompaniment of the practices of conflict resolution; they maintain its scripts, grant seriousness, and enable a sense of continuity of society despite changes to its structure. Thus in Turners scheme, an ostensive function of narrative is to separate social drama off from everyday life: in a space distinct from normal life, a drama is enacted that settles conflict with an authority that is beyond everyday dealings between people.
Turners concerns also include the passage of the individual through the public life of a social drama. He argues that the lives which are represented in the narratives of figures in social dramas provide the community with the repertoire of paths for gaining a position within society. He refers to relevant aspects of these narratives as action paradigms (Bruner, 1962, anticipates Turners idea in his concept of a library of scripts). Turner writes (1980, p.155): Just to be in the cast of a narrative drama which comes to be taken as exemplary or paradigmatic is some assurance of social immortality. The emergence of an individual into the status of a recognised and legitimate member of society is seen by Turner as a species of social drama.
Thus for Turner, the expressive domain contains both forms which resolve conflicts related to the practical sphere of life and rituals of separation and merger which mark the boundary between the two domains. What is critical for this study is the suggestion by Turner that this differentiation can be acted out in a persons life: that individuals might orient themselves towards certain expressive forms by which their status is measured.
This individual function of social dramas is represented in the concept of rite of passage, a phrase coined by van Gennep (1960). According to van Gennep, changes in status, especially entry into adult society, are managed ritually through a drama that severs the dependent ties of the candidate from the home, subjects him/her to an ordeal or mystical rite that leads to the revelation of sacred power, and reintegrates him/her back into the new role. The temporal structure of breach, crisis, redress, reintegration, is clearly evident in the rite of passage. The narrative practices that orchestrate the rite of passage may be seen as asserting sameness with difference: both integrating developmental shifts into a familiar plot structure and revealing change in power to act. If one thus implicates the practice of making sense in the allocation of agency, then the relation of biographical narratives to rites of passage in anthropological research is of central relevance.
Other examples of this approach include Jules‑Rossettes (1986) study of prophecies in an African community and Herzfelds (1985) investigation of the role of story‑telling in adult socialisation in Cretan society.  Both these studies grant a significant place to the performance elements of narrative; in Herzfelds study the character of the story told about a goat-theft determines the amount of credit which a young boy will obtain from his exploit.
The immanent nature of the narrative performance is also the object of Stemples (1986) paper on the functions of narrative. He allies the captivating outcome of storytelling with the positioning of the narrator within the community. Stemple (1986, pp.213-4) expresses a point that extends Herzfelds position:
The process of formulation, the particular choice of expression, individual elaborations or other features of the narrative discourse become interesting for the listener as characteristic expressions of a subject which also presents himself/herself in many different ways by assuming the dominant monologic role of speaker for a comparatively long period of time: as an individual with particular qualities concerning behaviour, knowledge, abilities, etc., as the author of a qualitatively preferred contribution to light entertainment, and in the case of a story experienced by the narrator, both as historical subject and his own judge (mostly in an indirect way, i.e., orientating himself by seemingly objective facts.)
These ethnographic approaches bring into consideration what is revealed through story-telling practices: it is not simply the content of the narrative but also the qualities of its performance. The act of storytelling not only conveys information from time past, but also enables particular relations to exist within the present: more directly, it allocates a special place for the narrator standing out from the others who form the audience. This has many relevant social implications. While the audience is separated from the storyteller, it is brought together and can share the experiences that are evoked by the narrative. As such the act of storytelling can be seen to accompany the regions of social life that highlight commonality between people, characteristic of the expressive sphere, what Turner terms communitas. 
Once the storytelling setting in brought into play, it is possible to take into account the specific conditions which made it possible: the desire of the narrator to tell and the willingness of the audience to listen. Given the emphasis in this study on the regions in which personal change is revealedespecially the region of the conversationit is important to be open to the types of exchange necessary for the region of storytelling to be possible.
2.4.2 Social exchange
Outside the ethnographic field, there is a variety of approaches to the relation of storyteller and audience. Each of these offers an account of the exchange that occurs within the context of narrativeit examines interests of the participants. Benjamin (1973) suggests that a role for the storyteller is to evoke an experience that exists in the realm of possibilities for the audience. For Benjamin (1973, pp.108‑7), the ultimate role of the storyteller is to tell of a life:
His gift is the ability to relate his life; his distinction to be able to tell his entire life. The storyteller: he is the man who could let the wick of his life be consumed completely by the gentle flame of his story.
The place Benjamin assigns to the storyteller is what Turner and other anthropologists call the liminal space, where possibilities that arise from normal life are explored as if they could be true. The narrator thus exposes for the audience experiences related to anticipated possible events: adventure, death, etc.
A more conservative version of this exchange is proposed by Kotre (1984). Kotre employs Eriksons developmental scheme to situate the role of storyteller within the age-related goal of generativity. Here narrative is a means by which one contributes to the stock of knowledge in the community in order to create something which is seen to outlast ones life. Kotre claims that this contribution primarily reproduces the already present cultural values. He writes (1984, p.225):
In telling a good story one retells one of the cultures enduring stories. One provides a particular for a universal and ensures thereby that the culture will carry on.
While Kotres approach provides a way of looking at storytelling as a reproduction of significant cultural values, his analysis is abstracted from the life setting in two ways. First, his surveys of life stories provide a readymade format for autobiographies: participants respond to the question, If your life consisted of chapters, what would they be? And second, he provides little context for the type of exchange proposed: Why is there this focus on generativity? and Why does it take the conservative form of reproducing existent values?
A more immediate approach to this question is presented in the exchange theory of storytelling offered by Chambers (1984). Chambers aim is to identify the narrative strategies that are deployed in literary texts in order to secure the attention of the reader. He argues that previous literary analyses have ignored the manner in which works of fiction mediate human relationships (1984, p.4). In Chambers scheme, the power of the narrator to re‑create experiences within the audience depends upon his or her ability to capture its attention. Chambers outlines several strategies that serve this purpose: maintaining suspense by withholding dramatic information; calmly understating the significance of the events being related; and laying claim to an inherited position of storytelling authority within the community. Clearly these tactics have different effects in different contexts.
In the storytelling situation Chambers assumes that the audience demands a story with an acceptable point. He draws this assertion partly from Labov (1972), who claims that the narrative performance has as its interrogatory background the question, So what? Labov states that, Every good narrator is continuously warding off this question (p.366). This question, So what?, contains an uncertain depth: it seems to be both a request that one come to the point, and a demand that something out of the ordinary be presented. It is difficult to account for the latter command. Why should an audience be willing to give its time to a narrator who will undermine its sense of order? As Agar (1980) points out in an ethnographic study of life history narratives, sometimes the violation of the audiences thematic expectations is a useful rhetorical device for asserting ones presence. One alternative way into this problem is to see these breaks as Turner would, that is as guides for dealing with situations where normal social rules of acting do not apply. The narrator in this case is someone who has witnessed an unexpected event and is willing to cast it into familiar terms for an audience. The story therefore promises to enable the audience to face similar situations. Yet we must also ask the converse question. What does the story offer the narrator?
The reciprocal nature of the relationship between narrator and audience is completed for Chambers by proposing the desire of the storyteller to control the experiences of others. As he writes (1984, p.214):
Narrative moves, then, produce authority, and authority is an acknowledgement that the story has a point. But another name for narrative moves addressed as they are to another whose adherence to the story must be gained, is seduction, as a way of inhabiting the space of the other without possessing it.
The desire to seduce functions in Chambers scheme to reiterate certain beliefs about the world that provide the legitimate moral ground of the story reality. The strategy of seduction is to provoke a sense of anxiety in the audience which predisposes it to the narrative point. In taking liberties with the audience, the narrator is seen to offer safe familiar meaning. The gain for the narrator is a sense of power over the feelings of the audience.
Though Chambers approach appears to incorporate a more immediate context, the form of exchange set up seems to command for the audiences attention a similar reproduction of cultural values as suggested by Kotre. Examined in comparison with Benjamins version of the storytellers value, it is possible to propose a mixture of unknown and familiar meanings which might be of value to the audience.  In general, the perspective of exchange theory provides a motivational account of the conversational region.
2.4.3 Narrative and psychology
The attempt to place narrative in an academic inquiry has thus far established four propositions. First, that this mode of meaning is relatively autonomous from the paradigmatic or scientific form of reasoning. Second, that narrative entails a partitioning of the practical and expressive order of meaning, in which the act of referencing is presented and claims to agency are possible. Third, that the expressive and practical modes can be related in a variety of ways; e.g., the expressive is a means of releasing anxiety caused by the practical order. And fourth, the context of storytelling has its own conditions, in which the narrator exchanges for the attention of the audience a story which reflects the possible and real conditions of its lives.
While these propositions are critical to the role which narrative has in explanation, they are not sufficient for an understanding of making sense of personal change. In order to describe the relevant referential practices and the contexts in which they occur it is necessary to undertake empirical investigation. Psychology is a candidate discipline for such work. Given the general interest in narrative as a means of constructing reality in the social sciences, one would expect its presence to be noticed within the discipline of psychology, particularly with its cognitivist orientation towards processes that deal with information about the world and the effect of social contexts on behaviour. There have been many expressions of the possible sympathy between the literary and psychological points of view.  The fact that Freud received the Goethe Prize for Literature is a sign that psychoanalysis in particular warrants investigation as a storytelling activity as well as a therapeutic practice.  The review of narrative and psychology will begin here as the first point of entry for questions of representation in the study of individual behaviour.
Ricoeurs (1970) interpretation of psychoanalysis as a mixture of hermeneutic and energetic discourses overlaps with the expressive-practical dualism used in this study. As an expressive technique, psychoanalysis allows for the disclosure of blocked information; and as a practical action, it effects release from control of defence mechanisms. The narrative approach to psychoanalysis has concentrated on the former emphasis: therapy is seen as a process of story construction. The link between this and agency is instructive in this current study. The two main exponents of this approach, Schafer and Spence, grant different degrees of structuredness to the narratives at play in therapy.
In a manner similar to Whites historiography (see Section 2.3.2), Schafer (1976) seeks to uncover basic schematisms at work in psychoanalytic narratives by reference to Fryes scheme of four myths. Like White, these schematisms are prior dispositions that order the way events are presented; Schafer terms these visions of reality. They include the comic, romantic, tragic and ironic visions. Each vision contains different assumptions about the nature of time and the possibilities for change. Schafer ties the comic and romantic visions to the active curative disposition of psychoanalysis for which hope is a necessary element. Elsewhere (Murray, 1987), I have extended this analysis outside therapy and linked the romantic vision to the popular appropriations of psychoanalysis. These schematisms, however, are not characteristic of the official practice of psychoanalysis, which is seen to manifest a love of truth that is more appropriate to the tragic and ironic visions (note the similarities here with the competence theory of social cognition).
Schafers analysis has the advantage over Whites that the schematisms are related to specific audiences. Yet what is of more interest to the current study is Schafers understanding of the dynamic relation between narrative and personal change. Schafer places these schematisms in an analytic setting ...whose concern is to construct life histories of human beings (1978, p.6). Within this frame, the phenomena of transference and resistance manifest themselves as expressions of narrative conflict (Schafer, 1980). Therapy is the setting for a confrontation of opposing psychic forces, and the interpretive process wrestles with confusion until insight is gained. The outcome of analysis is seen to be a story of the person that accounts for hitherto excluded details in a manner which grants the analysand greater agency in determining his/her destiny. From a story of radical discontinuity, in which the analysands life is subject to more powerful forces, it becomes possible to identify a persistent self. Schafers use of narrative in psychoanalysis is thus tied to the practical domain: it fixes a broken structure so that it can become more powerful. Like Malinowskis (1925) explanation of magic, it ritualises optimism and provides the working basis for active participation in the environment.
Given the similarity with Whites project, there is a clear problem with Schafers approach. That visions of reality can act as schematisms for ordering experience according to human interests is plausible, but why these schematisms? Frye would argue that they are part of the narrative tradition one has inherited from the age of agrarian community, tied as it was to the cycle of four seasons. The dubious nature of this grounding has been the source of much criticism of Fryes approach (e.g., Grob, 1983). At this stage it is best not to be tied to this system, but to hold it as a source of reference. The approach of Spence attempts a narrative version of psychoanalysis that is independent of literary genre.
Spence, while working on a similar project to Schafer, puts less faith in a pre‑determined framework such as Fryes four visions. He claims that the subject of negotiation between analyst and analysand is the tacit understanding about the world that governs the psychoanalytic practice. Therefore he does not look for the emergence of timeless archetypes. Spence describes the way in which the progress of analysis concerns the development of shared background assumptions that will enable experience to be registered in language. As in the classical method, the analysts task involves evoking unexpressed experiences, though here the ideal system of reference as a literal description is replaced by the notion of the interpretive act as a tactic for evoking pre-linguistic experiences. Transference in this case is essentially the relationship of audience to narrator, the interpretation being the narratological move of showing something to the world. Spence (1982, p.289) writes:
An interpretation may produce the desired result: the patient, supported by a belief in the analyst and reinforced by the power of the transference, may allow himself to suspend disbelief in the literal meaning of a given interpretation and thereby make himself accessible to its artistic and rhetorical surround.
By disrupting the normal routines of conversation in risking confusion and silence, the analytic space provides an appropriate setting for experience that has hitherto been silent. Accordingly, the goal of therapy is to give the experience of the analysand a narrative home by granting it expression in language. The interpretive manipulation of language is essential to the revelation of experience that is denied outside the analytic setting. This contrasts with what Spence calls Freuds archaeological metaphor: Freuds guide to truth as an essence waiting to be uncovered in the ruins of the past. Spence offers an alternative notion of truth as an element tied to the interpretive process. Truth here is consonant with its place in Harrés depiction of scientific investigation.
By contrast to Schafer, therefore, Spence tends towards the expressive mode of meaning, seeing psychoanalysis as the creative space in which previously occluded memories and feelings can be reflected upon. Spence sees language as the end point of therapy, whereas for Schafer it is a means by which one can extend the ground for action. A positive outcome, for Spence, is a sense of self‑coherencebut not because it is a necessary preliminary for action, rather it serves a general need for understanding. Given the approach taken so far, the positing of this need seems an unnecessary theoretical commitment, lacking substantiation. As the current focus is on the forms of agency granted by others, this need would be more appropriately expressed in terms of an understanding of ones experience by others. Schafer is also exposed to this shortcoming: agency in his terms has similar properties to locus of control in social cognition (see Section Error! Reference source not found. ). 
The psychoanalytic perspective thus presents the question of narrative fit for an individuals life, suggesting problems that may emerge with an unworkable story and also techniques directed towards the re‑construction of that story. In terms of the problem of how people make sense of others, it re‑emphasises the centrality of narrative understanding to agency, while opening up the problem of ownership of self‑narrative. It also raises the question of the agency for granting agency. (One could say that it is the authority of the analyst that enables the change in self-narrative.) But there are problems in translating these developments in psychoanalytic theory into this study. This study concerns the realisation of agency as a reciprocal event that requires public dramatisation: the present gaze is on the region of the conversation, not the clinic. From this perspective the question is: How does the newly constructed narrative begin to penetrate the social context of the analysand? A possibility that would be consistent with the realist picture is that the decision to undergo analysis itself becomes part of a negotiated narrative, representing a publicly perceived crisis point in personal development.  Fictional representations of the analytic journey would suggest this (see Murray, 1987). The pertinence of psychoanalytic theory rests partly on the way the analytic setting can be presented as a region for revealing elements of character, where certain interpretive practices govern the emergence of normally marginal experience into the open, into language. The notion of insight as a source of change indicates the place of the psychoanalytic setting in a larger narrative of personal journey, a narrative that is shared outside the clinical situation. This discussion will be continued when the identity of a psychological narrative is canvassed in Chapter Three.
So far reference to research on narrative within the discipline of psychology has been limited to the work of Bruner. However, the discipline has recently witnessed the emergence of a field of investigation termed narrative psychology. Sarbin (1986, Manusco & Sarbin, 1983) has done much to introduce this field into the discipline by providing the rationale for a contextualist metaphor in psychology. Sarbin judges how well the different root metaphors of formism, organicism, mechanicism and contextualism are able to represent humans as active agents; these are agents who are capable of forming intentions rather than passively suffering the impact of forces upon them. It is only the contextualist metaphor that enables this. In relating event to context Sarbin argues that one approaches human action as a historical event, allowing for certain possibilities of action to emerge given an understanding of the situation. Sarbin assigns to narrative the task of guiding agents through the world (Narrative is an organising principle of human action. 1986, p.19) Its primordial nature is demonstrated in the projection of storied events onto the random movements of non‑human stimuli. To know persons, Sarbin maintains, it is necessary to know the story in which they are participating. Making sense of others thus requires knowledge of the world from their point of view. The means by which this awareness is gained remains an open question.
Like Bruner, Sarbin presents a brave challenge to the discipline of psychology, much in sympathy with the project outlined here. However, there are a number of points that require clarification, given the framework of this study. Sarbin refers to an experimental set‑up that demonstrates the inevitability of narrative in sense making. It provides a screen upon which subjects can project their base level interpretations. The disengagement of the subject from the stimuli is thus intended to create a neutral space. Yet it is necessary to question the possibility of any wholly neutral space with the same ecological concern that Gibson (1979) expressed towards the use of the headrest in the perception laboratory. The two‑dimensional representation of stimuli creates a stageframed off from normal flow of actionthat renders a changing nebulous world observable. Just as Harré & Secord (1972) express scepticism about the relation of the neutral setting of the laboratory, so one needs to judge this study in the light of events that occur outside the laboratory context. Certainly there are stages for the demonstration of character in theatre, television, books, etc. But the objects of the current search include the ways of making sense that occur in exchange between active agents. As such, a distanced neutral position cannot be taken by default. One cannot successfully employ the experimental logic of the natural sciences in order to control for the setting of interpretation. Every revealing of another has a setting that limits the ways of making sense, whether is it a current affairs interview, a religious confession, a job interview, or a dinner party discussion.
The second query concerns the means suggested by which people make sense of others: i.e., discover the narrative that the other employs to order his/her experience. While there is much to say for this proposition, it raises the awkward question of how a narrative is understood: by another narrative, by a precis? The notion of narrative as an inner cognitive structure that can be prised out is vulnerable to the criticism of social cognition that it ignores context. If such context is take into account, Sarbin can be seen to have laid out a direction of research with much potential future work.
Another major contribution to narrative psychology comes from Kenneth and Mary Gergen (1983, 1984, 1986). Like, Sarbin, they present narrative as a process of making sense. They depict this process as attributing coherence to events by figuring them as accelerating or impeding progress towards a particular goal. They illustrate this process within the discipline of psychology, describing how different developmental theories imply different schemes of progress. Gergen provides an economical account of the practical orientation of narrative, though the act of determining the goal is left unspoken for: there is no allowance of a persons recognition within the story of what his/her quest consists of. For current purposes, it is worthwhile considering the valuable insights offered concerning the social context of narrative. Gergen & Gergen (1984) extend the exclusively individual focus of much psychological research to consider the role of narrative in defining a relationship. Further, Gergens (1989) attention to the conditions in which it is possible to be granted a voice, in the concept of conventions of warrant, provide insight into the broader philosophical assumptions at stake in this negotiation.
A different set of psychological tools for investigating narrative is provided by McAdams (1985 & 1988). For McAdams, narrative both individualises and socialises. From early infancy, McAdams posits the presence of a narrative tone which develops into more complex forms of biography. For example, a positive tone provides a sense of hope that might flourish into a romantic self-narrative. McAdams provides a set of normative standards for self-narratives, such as continuity and complexity, to which the developmental process is seen to progress. While like many developmental schemas, the events in early childhood are seen to be critical in determining later progress, McAdams allows for the necessary contribution of a social context in providing the narrative structures by which a self can evolve. This context, though, is not conversational. For McAdams, society is a hardware store which equips the individual with narrative structures. Although he identifies power and intimacy as the key dualism in self-narratives, his analysis does not allow for the role of these stories in gaining agency from others.
The project of narrative psychology entails a study of the mediation of narrative structures in the individuals relationship to others. This is a project towards which the current study aims to contribute; though it distinguishes itself from research strategies that situate the effect of narrative purely within an individual locus of control. Certainly, the claim that narrative is a cognitive instrument (Mink, 1978), in which the sense made of the environment is embedded ( Verstricktsein), has a great deal of explanatory power. However, such an approach loses sensitivity both to the negotiated exchange between narrator and audience, and to the partitioning of the expressive and practical domains of meaning. As demonstrated in sociolinguistics, to keep an audience open to ones tale it is necessary to provide it with material that both keeps interest and provides a point relevant to the lives of the assembled group. And as revealed in anthropological studies, the expressive domain, while separated from the practical, can provide a necessary adjunct to it, particularly in releasing and seeming to resolve the tensions it causes.
2.6 The way forward
After this detour through the place of narrative in the study of making sense the questions that I posed at the beginning of this chapter are beginning to take a more definite shape. The question of what one makes sense of in others entails the discovery of referential practices by which capacities of an individual are revealed to others. How sense is made of this is manifest in the narrative forms that record this demonstration. Finally, why one makes sense of personal change is conditioned by the conventions attached to the distribution of responsibility and power in the region of the conversation.
In response to these questions, I propose a plan of research that gradually focuses on a particular setting of making sense. The research plan is constructed to limit as much as possible the psychological assumptions made about the underlying motivation of making sense. I begin with a textual analysis of three life manuals in order to establish a frame for understanding how personal change may be set up in order to reveal particular values. This frame is then used to analyse contemporary life construction, in which groups are required to draw on their conversational resources for representing personal change. From this analysis a specific referential practice is determined as the focus for a collection of accounts of personal change by individuals in groups. This study thus seeks to establish a set of conventions in which forms of agency are allocated within a conversational group.
 . Following Lyotard's approach, one could speculate that the field of social cognition works within the more pragmatic French macronarrative. It assumes that prejudice can be overcome when obstacles to the truth are identified and targeted; when the role of scientist is dominant in a layperson's making sense then misunderstandings between people will disappear. Lyotard's work assumes the inevitability of narrative constructions, even in the most abstract of knowledge practices. The introduction of narrative into an epistemological inquiry extends the purview beyond a formal study of knowledge practice to include the assumed spatio‑temporal context at work in the appropriation of the cultural knowledge.
As well as being communicated to an audience beyond the scientific community, the task of allocating resources demands that scientists convey wider significance of their work to fellow scientists as well. In examining the posterior documentation of scientific work, Harré looks for the presence of a narrative package in which the isolated experimental finding gains resonance to other scientists. He identifies what he terms the `official' story of research that is written retrospectively to maximise the consistency of argument and relevance to other theoretical concerns. Harré (1987, p.22) articulates the nature of this story:
They are describing a rhetoric and an associated set of narrative conventions for presenting a story in which rival teams of scientists appear as heroes and villains.
This rivalry provides the ground for contingency that is essential to narrative. But why not just report the facts? According to Harré the official story exists not just to convey information about discoveries but also to encourage scientific labour which produces facts that conform to the virtues of trustworthy knowledge. Harré here touches on the role of narrative as a `conduct‑guiding force.' The potential of stories to inspire action may lie in the way they constellate the world around the pursuit of an end; this enabling of action may serve as the release that intention requires to seek its realisation. A sensitivity to this question is warranted by the proposition articulated earlier that ways of making sense are tied to the practical field.
One implication of the recent attention given to narrative has been to question the inevitable superiority of paradigmatic forms of knowledge. The positivist hierarchy of knowledge equates story with immaturity. A philosopher of science writing in 1901, Furneaux Jordan claims that the origin of the literary viewpoint lies in primitive fears: one looks to stories for comforting illusions. The gaze of science, however, is seen to spring from a certain immutable desire for truth. But once the perspective is taken that allies narrative with the understanding of all human action, including scientific endeavour, the conduct of social `sciences' is opened to literary analysis.
 . However, given a contextualist perspective, one would need also to take into account the material practices involved in the constitution of historical material, such as documentation and research interviews. In addition one would require some account of the principle governing what narrative schematism is seen as appropriate
to particular historical scenes. For instance, an historian would find it difficult within the current setting to inscribe the bombing of Hiroshima as a comedy. Like the analysis of narrative within science, the projected assumptions of the practitioner of knowledge become the object under focus, however White does not offer an understanding of the place of that story within a practical context.
 . The base for this work is the Centre for Biographical Studies in Paris, and its journal Life Stories comes out once a year.
 . Less historically oriented is the psychological work of Wiederman (1986). Wiederman follows the trail of biographical research towards uncovering ways in which ordinary actors accord sense to their lives. He examines how the metaphors of battle or journey frame certain life stories, and also how a life conceived merely as a possibility (an `unlived' life), figures significantly in the real‑life story. The latter reminds us of Bruner's characterisation of narrative as a subjunctive medium. Leitner (1987) makes a point that ties narrative closely to theories of identity. He proposes that self‑determination requires confrontation with a contingent state of affairs. The apprehension of this through dramatic means touches on Harré's (1983) speculation that there are certain conventions of autonomy.
 . The alternative is a closed interpretive system, lacking the potential for what Lotman & Uspensky (1978) refer to as `deautomatisation.' What is at stake in this argument is the complex issue of the limits of narrative understanding: the Cartesian empiricism denies the imposition of temporal structures while Sartre proposes the inevitability of narrative.
 . By close observation of the Maranke apostles Jules‑Rossette was able to attend to the generalisable understanding of the world and its expression in practices of detecting crimes and predicting future events. The latter micronarratives are enacted in the course of certain ceremonial performances. Their effect is not only to achieve a desired end, such as triggering confessions through ambiguous statements of blame, but to realise the cosmological macronarrative as it is seen to structure the world of individual members. Such a conceptualisation of narrative mirrors the role of scientific account in Harré's realist theory of science: instantiating ontologies that provide the practical ground for the act of showing.
A similar framework is employed by Herzfeld (1985) in an analysis of the narrative accounts accompanying displays of character in a small Cretan mountain community. Herzfeld identifies the macronarrative in these accounts as the heroic history of the Cretans in their battles against invading Turkish and German armies. This history celebrates the virtue of cunning individual acts of defiance. Such a macronarrative forms the background of the contemporary dramatisations of individual passages to manhood. It is in the individual accounts of goat‑stealing that the Cretan macronarrative finds expression in the lives of its members.
In order to be recognised as possessing the qualities of manhood, the Cretan adolescent normally commits an act of daring theft. These acts not only earn the begrudged respect of the offended goat owner, but also become the subject of the boy's story, told around the village. The dramatic success of the theft depends largely on the effectiveness of the story. The rhetorical gestures that the boys must make to become full members of their society are termed by Herzfeld the `poetics of self.' He further defines this poetics as the dramatic management of a disruption to ordinariness so that it is seen to serve a higher ideal, in this case the manly virtues of cunning and bravery. The very position of an honoured position in the community is thus bound to a narrative process. The relation between recognised identity and narrative unpredictability is the focus of Herzfeld's (1985, pp.10-11) argument:
A successful performance of personal identity concentrates the audience's attention of the performance itself: the implicit claims are accepted because their very outrageousness carries a revelatory kind of conviction. It is in the self‑allusiveness of social performances, and in the concomitant backgrounding of everyday considerations that we can discern a poetics of social interaction. The self is not presented within everyday life so much as in front of it.
Herzfeld ties the achievement of a position in society to the specific context of performance; character is revealed in the exchange of stories around coffee houses. A necessary condition for being granted this place is the capacity to gain the belief of the audience, who are happy to exchange their scepticism for a narrative where formal laws are seen to be broken.
However, the importance of the narrative does not render the initial act of goat theft irrelevant. The breach of order involved in stealing someone else's goats, the process of identifying the culprit and making good the loss is the social drama which is orchestrated by the individual narratives. Herzfeld takes the current inquiry further, though, by suggesting that the information gleaned from the narrative of a goat theft is not solely the details of the exploit, but also the manner in which they are related. His understanding of these social performances as enabling individuals to be revealed `in front of' everyday life significantly alludes to the partitioning process involved in the narrative mode of making sense.
 . So Sacks (1986) describes how an extraordinary event can set going a narrative process of sense making that cuts down the normal hierarchical divisions between people. Thus a `Bullock's' event in a Department store of that name will find the vice‑president talking to the cleaner on familiar terms.
One can speculate that the dramatic mileage of the weather in conversation between strangers can be seen as a way of handling the anxiety of the situation by constellating participants as members of an audience. In switching theoretical attention from the inner cognitive structures that order reality to the social practices enabled by that construction, the referential question has been relegated. As Bakhtin would argue, meaning can be considered as a dialogical phenomenon, and therefore requires an understanding of the interpersonal context for an utterance to be fully explicated. In this Bakhtin reiterates Bruner's argument for linking narrative to the formation of an intimate social group. He writes (Bakhtin, quoted by Clark & Holquist, 1984, p.209): `Every utterance of life is akin to a `password' known only to those who belong to the same social purview.' This general point finds an ally in Wittgenstein's argument for the origin of language in social spacing. As Giddens (1979, p.38) relates it: `To know a language is to be able to participate in the forms of life within which it is expressed, and which it expresses.' The direction, then, turns towards the state of affairs that accompanies the narrative performance.
 . This dualism of information, which is both familiar and strange, is reinforced with a consideration of the demonstrative function of storytelling. If narrative regrooves values by showing their strength or their capacity to allay anxieties, then the form of the story must evoke uncertainty as well as provide the already settled answer. Thus an approach such as Jameson's (1981), which claims that narrative is governed by the operation of ideologeme that contains class anxieties, fails to take into account the dual facetness necessary to account for a convincing truth. Bourdieu's (1977) use of `objectifying principles' which served to inscribe the truth of the group is more appropriate to this dualism.
 . So Henry Murray writes in 1938, `We academic psychologists have yet to discover how much can be learnt from the realists of literature.' (p.609) Indeed, the recent work by Potter, Stringer & Wetherell (1984) argues that much social psychology acts as a form of literature. In one example they compare Bem's gender identity scale with romantic love novels as different ways of telling similar stories about the relations between men and women.
 . Gallop (1984) remarks: `Psychoanalysis is a regional application of literary studies.' The renowned literary quality of Freud's writing has led to many scholars examining his works, especially the case studies, as narratives of investigation, similar to detective novels (e.g., Mahony, 1982) Such an approach reinforces the point that the narrative process can be revealed even in scientific texts. But the inquiry into narrative in psychoanalysis has not only focused on official texts, part of its subject matter is the therapeutic process.
 . There is a variety of further developments of narrative in psychoanalysis. The role of narrative in developing a sense of continuity of self over time has been addressed outside the context of psychoanalysis by Freeman (1984, 1985). Freeman argues for an historical rather than predictive understanding of the life‑span. In a similar vein, Wiederman (1986) figures therapy in general as a way of processing experience in order to establish a consistent point of view. Within Wiederman's terms, point of view is a way of seeing which constellates the world according to certain relevances. In this context therapy can be seen as a training experience for ordering reality; there is no necessity for insight. By implication, if psychoanalytic therapy can be equated with the business of re‑fitting narratives, psychological problems may be conceived as poor stories. This possibility has been taken seriously by several theorists. Cohler (1982) interprets the thought disorders of schizophrenia as desperate attempts to maintain an ordered narrative. And Keen (1986) describes the fantasies of paranoia as the result of an aberrant narrative schematism. The paranoid narrative typically contains an absolute division between good and evil, and between self and other, in conjunction with a cataclysmic time frame. For Keen, the failure of this narrative lies in its unsuitability for grounding interactions with others; working from this schematism does not allow for any possibility of a future toward which to build. The reasons for the emergence of such a point of view are not given by Keen, and we would expect that an understanding of such developments would demand knowledge of the historical conditions of a culture and the trajectory of the person within it.
 . This indeed is the explicit focus of recent developments in family therapy (see White & Epston, 1989; and Epston, 1989) which strategically direct the therapeutic process towards changing the self-narrative of the client in the eyes of his or her social milieu.
Text is copyright Kevin Murray. For reproduction inquiries, email