This paper [Note 1] is about the field of sociological works that claim to be taking a "narrativist" approach to biographical accounts. It is a complex field offering much variety and concealing some schisms.
A recent textbook on biographical sociology (Miller 2000) characterises narrativism by its eye for the fluid interactive process in which a life story gets told on page 17 and by its eye for the biographical structure of meaning that guides the biographer on page 134. While there is nothing wrong per se with either vision, they may contradict one another in one important point: only the second takes the biographer's subjectivity to be a relatively stable structure of perceptions and outlook, and a researchable topic.
My examination of the field naturally starts with what defines its unity. I identify two attributes of narrativism, which seem to me to form the common denominator of what otherwise turns out to be a wide range of data-collection and analytic practices. Both harbour the ambiguity intimated by Miller (2000), in that they assume and indicate either an author-subject and hermeneutics, or a situated interaction and discourse analysis.
Secondly, I situate narrativist approaches to biographical accounts within the broader field of qualitative methods in sociology, by drawing on Gubrium and Holstein's (1997) discussion of its two realist and two constructivist "idioms". I argue that narrativist work with life stories can be, rather counter-intuitively, of the realist as well as the constructivist kinds.
Thirdly, I offer for discussion a critique of those narrativist approaches that claim or assume one of the following: (a) a notion of "deep structure"; (b) that certain procedures of data collection and analysis should be decoupled from their original purpose and codified to constitute a scientistic methodology; (c) a notion of the psychological subject; and (d) a notion of the romantic subject.
And in conclusion, I suggest four sociological research themes in which narrativist analyses of life stories could be especially productive.
Throughout, my concern is with sociology, the scholarly discipline dedicated to the collective dimensions of human life. What sociological knowledge can we produce by reading a life narrative carefully?
The first attribute of narrativism, on which there seems to be wide consensus, is its focus on the story – its composition and its telling – in preference to a focus on the sociological practices and conditions to which the analysed biographical account bears witness.
It is not that narrativists ignore the referential content of the autobiographer's recollections. The related facts are of interest, but not of a qualitatively different kind from the interest aroused in a non-academic hearer. We are engaged by what we learn about the realities of the recounted life, responding to them perhaps with feelings of surprise and recognition in turn, as they evoke the exotic in our own midst or the familiar in times and social locations we took to be outside our own. We may also run sceptical credibility checks in our minds on the facticity of what we are told.
The narrativist professional focus, however, is on the arts of emplotment, rhetoric and interactive performance that weave the recollected facts into a more or less compelling message. Like non-academic hearers or readers of an autobiographical account, narrativists are aware that the reading or hearing has left them with an overall impression. The story was gripping or bland, it had the ring of authentic truth or contrived artifice, it inspired, intimidated or bewildered by its morals. Unlike the non- academic hearers or readers, however, the narrativists analyse the recording, transcript or text to understand how such impressions come about. They suspend the commonsensical assumption that a story sounded authentic because the narrator was sincere, contrived because s/he had things to hide, or bland because s/he had led an uneventful life. They analyse the story and its telling for itself, as a structure and process that endows experiences with significance and the narrator with identity.
Narrativists study the cultural accomplishment of moments in which narrator and audience share in an outlook on life, single out some of its realities for acknowledgement, and orientate to each other as the co-producers of a more personal relationship. They are interested in the criteria by which narrator and audience appear to judge what to select for recollection and what coherence rules to apply to make the recollections relevant to one another; the criteria by which they judge a life story to be good and true to the occasion in which it is told.
The second attribute of narrativism, on which there is wide consensus, is a logical corollary of the first. Since it is the construction of the story that is under analytical focus, the time to which the narrativist findings refer must be the time in which the narratives were constructed, although the narratives that are under analysis are mainly in the past tense themselves. Narrativism generates knowledge of phenomena that exist in the present.
These two consensual attributes of narrativism in analytical work with life stories – that focus is on story construction and the present time – on closer inspection harbour an important methodological and theoretical division. Story construction in fact turns out to have two different meanings.
In its first meaning, story construction means the composition of the autobiographer's story – its emplotment and its use of contrastive structures and metaphorical imagery – which expresses an overall message about the life and its narrator's view of it. We analyse the story to see how each of its elements fits into the functional whole that defines the author's message and viewpoint. Our analytical work of identifying (or deconstructing) the construction then resembles textual analysis. It has a long tradition to draw on, in hermeneutics, which is as old an approach to the rigours of achieving valid textual interpretation as scholarship itself, and in the modern literary disciplines as well.
The study of written autobiographies as literary texts courts no controversy. That is different in the case of life stories elicited and orally performed in interviews, because the focus on story emplotment, imagery and authorial message tends to leave the oral and interactive quality of the original communication out of focus. It limits attention only to the interviewee (the autobiographical author), and indeed to only those of his or her utterances which appear to belong to the life story itself, with conversational interchanges with the interviewer, digressions and ungrammatical noises left out. This is unacceptable to the narrativism that, in taking story construction to be its prime object of analysis, takes it on board in its second meaning.
In this second meaning, story construction is the process of interaction within which the story got told. If the story was told in the course of a research interview, it was constructed interactively, even if the interviewer adopted a passive style and allowed or encouraged the interviewee to speak in long monologues. The interviewer's presence was of supreme importance to the story production. It defined the social setting in which the life story got told. The forms and content of what got told had their function within the interactional process within which interviewer and interviewee monitored each other and acted towards one another to create situationally appropriate identities vis- à-vis one another. The story production was a collaborative process, a production of situated meaning. An editorial manipulation of the recorded data that is geared to bringing out an internally coherent and transportable story, i.e., a story that could make sense outside of the social setting that produced it, loses record of that paramount social activity – the production of situated meaning.
What is the nature of the social knowledge that can be produced by paying analytical attention to story construction in either sense? It will concern the present time or, more precisely, a continuous present time. It will show social phenomena that exist not only within a particular story (or a particular set of stories) and the momentary act(s) of collaboration that created it.
The narrativism that takes the life story primarily as being performed in the course of an interaction has a different view of the present time than the (much more common and developed) narrativism of the hermeneutic kind. The present time it focuses on has in fact become a past event by the time a data reduction and analysis takes place. It is bounded by the beginning and end of the interactional process that produced the life story. The sense of a continuous present in the knowledge production (that is, generalisable phenomena existing not only in the particular moment) is constituted by an interest in the repertoire of social skills on which participants in the scrutinised interactional episode evidently drew to accomplish what they needed to accomplish in that setting.
The knowledge of the (continuous) present time that the hermeneutic work with a life story produces can be of two kinds.
Firstly, it refers to cultural genre. The life story may be patterned to be similar to an established, popular story form, thus inviting a certain kind of (popularly practiced) reading or hearing. Cultural genre is known to be a historical creature, some of its forms rising and falling with the ebbs and flows of history, unlike the social skills identified in empirical studies of the "mechanics" of interactional events, which tend to be implied to be timeless. Relative to the passing moment in which the life story got told, however, the evoked cultural genre has a continuous presence.
The second type of knowledge produced by hermeneutic work with life stories refers to the subjectivity of the author. Again, the latter is known and freely acknowledged to be subject to change, but it does not change overnight every night. Relative to the passing episode that produced the life story, the story's author, his or her subjectivity, is a continuous presence.
Narrativists of the hermeneutic kind frequently write of dialectic between author and genre, the individual subject as a reflexive user of cultural resources, in the communicative act of self-definition.
It could be tempting at this juncture to link hermeneutic narrativism with cultural studies and social psychology on the one hand and, on the other hand, the narrativism oriented to interactive performance of life story with interactionist sociology. However, the matter is complicated by the fact that some elements of the symbolic-interactionist tradition emphasize inner dialogue and the formation of self-concept. That links them with the hermeneutic type of analytical work with life narratives.
Everett Hughes (cited in Becker 1963, 102), for example, coined the concept of subjective career,
…the moving perspective in which the person sees his life as a whole and interprets the meaning of his various attributes, actions, and the things which happen to him.
Although the perspective on life is "moving", Hughes' suggestion here is that it is stable and continuous enough to transcend particular situations, that it is indeed the one social object common to all situations in which we take part – the anchoring point from which we make judgements and subsequent plans of action in each specific situation (Athens 1994; Ezzy 1998; McCall and Simmons 1966).
It is a small step from these reflections to the viewpoint that a life story, its core thematic structure and emplotment if not its detail, is the direct product of the author's self, of the present stage in his or her subjective career, rather than of the social setting in which it got told. In this viewpoint, analytical attention to the interactional process tends to function merely as ground-clearance for the hermeneutic work on the life story (and its author). The transcript is studied for the point from which the interviewee "took control" and commenced developing his or her narrative line regardless of interviewer's interjections; memory is scanned for the periods in which the interviewee talked almost as if the interviewer was not there, as if the talk was addressed to a "generalised other" rather than to the interviewer (Rosenthal 1993; Burgos 1989). The authorial story that those moments left on record is then taken to reveal the subjectivity, the self, the self-concept, the self-identity, or whichever of the plethora of self-terms, bequeathed by the symbolic-interactionist tradition in particular as well as by modern western individualism in general, a life-narrative researcher happens to favour.
By contrast, the sociological interactionism that starts with Blumer's (1969) development of Mead's ideas, regards the self as very much a process of situated interaction. The self and its theoretical variants indeed have no more than a marginal existence in this approach, in so far as they tend to mean a structure of the mind that defines an individual subjectivity as continuous across specific interactional situations. It is, of course, our everyday social practice to talk and act as if we naturally were such continuous, situation-transcending and authentic subjects. In the Blumerian and later ethnomethodological perspective, however, the analytical focus is on how we do this, how we practice socially the collusive implicature of an individual with a coherent and authentic self that exists somewhere within but also beyond our evident actions, although they always take place and have their meaning constituted within a specific setting. The analytical focus is on social practice evident in specific situations, not on the structure or attributes of an authentic self that the social practice may imply to exist transcendentally across the specific situations.
In this perspective, narratives of personal experience are of special interest because they imply a coherent subject and draw the hearer into colluding with this implicature.(Alasuutari 1997; Linde 1993). They are demonstrably important in the social, collaborative practice of creating coherent and accountable selves and making them look naturally pre-given at the same time. In this analytical perspective, the claims that the emplotment of a life narrative is constitutive of, or analogous with, an author's authentic, situation- transcending self, is rejected as a romantic delusion. Thus Paul Atkinson:
I repeat, narrative does not provide a hyper-authentic version of actors' experiences or selves. A backdoor smuggling in of romantic constructions of the self will not do." (Atkinson 1997, p.343.)
The consensual viewpoint, that narrativism in biographical research consists of analytical focus on the told story and on the present time, thus hides a schism between two views of what a life story is and what knowledge it can yield. The first view is textual and hermeneutic, the second is radically interactionist or ethnomethodological.
In addition to this schism-covering consensual viewpoint, narrativism has been also characterised as a single-case method. Bertaux (1996) criticises narrativism for creating an academic climate in which it becomes possible to submit a sociological dissertation based on a single life-history interview, while Wengraf (2000; 2001, 302- 10) tries to construct a justificatory argument for such practice.
It is, of course, true that attention to thematic construction, emplotment and the gamut of linguistic phenomena that might be termed "narrative strategies" (Chase 1995) require a close reading of each transcript. This does not, however, prevent systematic comparisons with other cases, on the basis of which generalisations about shared, collective patterns or phenomena could be made. Many narrativists do just that, although eventually they may choose to present their analytical findings in a detailed discussion of a selected typical or deviant case.
Chase (1995) identified four narrative strategies of dealing with racial and gender inequality issues, in career-history accounts of American minority female educationalists in senior executive positions, on the basis of 27 interviews. Ginsburg (1989) identified life-narrative patterns constitutive of pro- abortionist and anti-abortionist activist commitment, partly by using the Russian formalist distinction between story fabula and syuzet, on the basis of 35 life-story interviews (Riessman 1993, 26-30). Rosenthal (1991) compared German veterans' War I and War II narratives by analysing 36 interviews. Cortazzi (1991; 1993), who admittedly researched other kinds of narratives of experience than life-story interviews, used the Labov & Waletsky (1967) analytical schema in cross comparisons of almost 1000 stories told by teachers in natural settings as well as in over 100 interviews.
This arbitrary list of examples could go on. The point is that narrativist approaches can apply or generate analytic schemas that facilitate cross- comparison of cases and the identification of typical or collectively shared patterns. Narrativism does not spell a commitment to making generalisations from a single case; and neither does it need to confine its result to recognition of an individual subject.
Narrativist or realist, sociological work with life stories belongs to that growing branch of sociology that could be labelled as qualitative methods. That has been a prolific but also very diverse branch, a veritable Babylon of languages for the definition of empirical research ends and means. Gubrium and Holstein (1997) make an interesting attempt to introduce some order into the diversity by sorting it into four "idioms", or "isms": naturalism, ethnomethodology, emotionalism, and postmodernism.
Naturalism is exemplified by participant observation studies of specific settings, of the nooks and crannies of society, which the researcher enters in a manner reminiscent of social anthropologists who research exotic communities. The researcher learns the norms, beliefs and practices that order everyday life in that world, describes that world as it functions in the participants' perspective, and explains it as an internally meaningful social order to the rest of us, the folks back home.
Ethnomethodology is exemplified by studies of the ways in which "members" (participants in interactional processes) make themselves accountable, manage the range of interpretations that others could make of their actions, account for their actions and deal with others doing it. The researcher seeks not the social norms that supposedly regulate conduct, but the ways in which implicit or explicit references to social norms are used in specific instances of members' accounting. It is in this activity of accounting, in our normatively oriented representations of reality to one another, that we make our world social, shared and ordered, despite the myriad contingencies upon which we improvise what we do.
Emotionalism is the approach to research that seeks to get in touch with social realities as experienced at an emotional level. It takes emotions to constitute the bedrock of the authentic subject, and the authentic subject as the target of understanding.
Postmodernism puts under critical scrutiny researcher's accounts, the ways of writing that constitute an authoritative knowledge of social reality. It seeks reflexivity and innovation in the practices of sociological representation.
Gubrium and Holstein characterise naturalism and emotionalism by "what" questions, and ethnomethodology and postmodernism by "how" questions. The "what" questions ask what the data under analytical focus report about social reality. The "how" questions ask how the communicative acts that are recorded in the data get to be practically effective; how they constitute social practice. Naturalism and emotionalism take the data in a referential, realist vein. Ethnomethodology and postmodernism take the data in a constructivist vein. So far, so good: a division reminiscent of the one created by the "language turn" and the general rise of enquiries that take "the surface level" of representation itself as their topic. The authors exemplify each "idiom" by a discussion of selected works, and their four-fold typology does seem to introduce some order into the Babel of languages in which qualitative researchers have couched their preferred versions of the craft.
The authors, however, go on to advance an argument that the best work is to be done in the middle ground between realism and constructivism, where analysis duly takes on board both the "whats" and the "hows" of the data in hand. This may seem a reasonable position to advocate, for indeed it is the case that neither form nor content can be adequately studied in total disregard of the other. However, a close scrutiny of the authors' selection of the works they cite, to exemplify the four approaches on the one hand and the virtues of the middle ground on the other hand, reveals that the selection was guided by rules of considerable vagueness.
Thus Chase (1995) is discussed as an example of the virtuous middle ground rather than of ethnomethodology, although the discussion picks out, entirely appropriately in my view, that the work primarily addresses the problem of how ethnic-minority women in elite positions account for their success in a situation where two public discourses – an established discourse of professional competence and a contested discourse of social inequality – feature prominently in the situational relevance frame. Gubrium and Holstein correctly point out that Chase's work also conveys information about the "whats" of the conditions under which minority women achieve elite careers in US society. It is not clear, however, why they think that this informativeness concerning the "whats" of a social setting is a quality that Chase (1995) offers in a markedly greater degree than Wieder (1974), the work they cite as a classic example of the ethnomethodological approach. Gubrium and Holstein advance their argument for the middle ground at the expense of depriving their four-fold typology of its analytical utility.
They advance it at a commonsensical level where, as I have already argued in the case of the distinction between realists and narrativists in life story research, narrativists respond to the content of a life story in the same way as any other hearer. At this level, however, the whole range of questions about our data surely occurs to us in the course of a research project, and most of us can claim the middle-ground virtue, leaving research in one or the other of the four "idioms" to extremist oddballs.
Holstein and Gubrium's fourfold typology, however, regains its analytical sharpness when we take it as relating not to the presence or absence of certain questions during our engagement with our data, but to the endpoint of a specific, empirically based argument such as constitutes the main theme of a single article.
In examining a specific empirically based argument, we can ask what is the nature of the main finding here: does it contribute to our understanding of a practice, condition or outlook existing independently of their representations ("whats"), or to our understanding of "artful" representational routines themselves ("hows")? This is an either-or choice. The main theme and endpoint of a particular empirical argument or line of enquiry cannot be "a bit of both". Observations appertaining to the questions that are dispreferred by the choice of main analytical focus have to be organised into a theme that has a clearly secondary status within the text, if not altogether bracketed and set aside for special attention in another piece of writing.
Gubrium and Holstein's advocacy of the "middle ground" makes sense as a plea to qualitative sociologists not to get locked into a dogmatic cast of mind in which a choice of analytical focus is forever, because it is deemed to be the only valid one. It is, indeed, desirable that a researcher working on a substantial data set relating to a broad theme pursues a number of successive lines of enquiry about it, which alternate between realism and constructivism in their choice of main analytical focus. Ideally there should be realist as well as constructivist articles emanating from a good qualitative data set. The line between the two kinds of analytical focus, however, can be straddled within a single article only at the cost that the article loses clarity and will be misread by its readers. Derrida's (1981) injunction that one ought not to mix genres in a single text applies.[Note 2]
My argument is that, when applied to the choice of analytical focus that is evident at the endpoint of a line of empirically grounded argument, Gubrium and Holstein's fourfold typology holds true and their advocacy of the middle ground is unsustainable. The works they discuss as examples of middle-ground virtue fall from the middle ground into one or another of the four categories.
Of the two life story researches they cite in their discussion of the middle ground, I have already argued that Chase (1995) belongs to the ethnomethodological camp. The other life story research that is given as an example of middle ground virtue is Gubrium (1993). This work contributes an understanding of the various subjective viewpoints, or "horizons of meaning", in which residents of a nursing home consider their residency in the home and evaluate the care they get there. The "horizons of meaning" are evident in the "narrative linkages" of life stories told by the residents. Gubrium is aware that the life stories were told in a setting where the role of nursing home resident was of inescapable significance. The narrative linkages and the horizons of meaning they outline were not a fixed entity; they were the product of a co-operation between researcher and interviewee, which took place in a certain time and place. Gubrium is critical of a narrative interview methodology where the researcher approaches the interviewee as if there was a life narrative already there, stored in the interviewee's mind but available to be brought to light by the researcher using the right kind of data-collection method (1993, 181; Holstein and Gubrium 1997). In that sense, he appears to be placing his work in the ethnomethodological camp. The significance, however, that Gubrium convincingly claims for the narrative linkages and horizons of meaning created in the course of his (and Carol Ronai's) nursing home interviews lies, in the last analysis, beyond the particular hours of narrators' meetings with their audience. It lies in making available for empathetic understanding "biographically active subjects" who in effect defy well- meant institutional assumptions concerning the role of nursing home resident and the links between quality of life and quality of care (1993, 177-88). The book demonstrates a variety of individual life perspectives in the context of which the residents experience and evaluate what the nursing home and its staff do for them.
In this respect, Gubrium's (1993) endpoint is not dissimilar from, for example, Jones and Rupp (2000), who present their findings as the product of the "biographical- interpretive method" (of which more below), which otherwise takes an ostensibly different view of appropriate data collection, in that it requires the interviewer to be what Gubrium would call, in a critical vein, a "passive inquirer" taking care to bring the interviewee's own story out, as if it had been already lying dormant in the subject's mind. Jones and Rupp show a "thematic field analysis" of a life story interview with the mother of a severely handicapped son, who had cared for him in the condition of keeping a guilty family secret, until the recent birth of a healthy son gave her a positive maternal identity and opened her world up. Like Gubrium's attentiveness to his interviewees' "narrative linkages", Jones and Rupp's "thematic field analysis" offers an insight into the subjectivity of an individual caregiver and potential social service recipient, a biographically constructed viewpoint that defines the individual's agency vis-à-vis her institutional roles, which it is now possible to understand with empathy. It is a thematic, hermeneutic analysis of an interview talk, which summarizes a viewpoint that has a relevance to the interviewee's actions, at least those related to dealing with welfare agencies, beyond the interview itself.
Neither Jones and Rupp (2000) nor Gubrium (1993) could be accused of appearing to assume that there is emotional bedrock of subjectivity for researchers to apprehend. Both recognise subjectivity as biographically constructed, not as determined by emotional experience taken to exist prior to reflexive thinking.[Note 3] The former as well as the latter appear to be in broad agreement with the theories that view personal identity as narrative construction, an achievement of narrative practice. In that sense, neither would appear to be a contribution in what Gubrium and Holstein (1997) call the emotionalist idiom.
And yet, it could be argued that none of the other three idioms permit what emotionalism clearly does: the presentation of individual subjectivity for empathetic understanding as the main endpoint, the prime contribution of a line of analytic inquiry. Naturalism requires the prime contribution to be about social practices, norms and ideologies that make a social location or milieu. Ethnomethodology requires it to be about "artful" accounting practices by which members constitute their membership and deal with threats to it that may occur in the course of interaction. Postmodernism requires it to be about textual practices that make sociological knowledge.
Here it is interesting to note that Silverman (2001, 38- 9), in adopting Gubrium and Holstein's typology for his own expositional purpose, characterises emotionalism by a focus on "subjectivity" as well as "emotion", and by personal biography constituting its favoured kind of data. Gubrium and Holstein's own discussion of emotionalism is indeed less satisfactory than the chapters dealing with the other idioms, in that it takes somewhat dated and outlandish texts (Douglas 1977; 1985) as its prime illustrative example. There is a case for tweaking their typology in the way almost suggested by Silverman. I suggest that their emotionalism be renamed subjectivism; and that it be defined, rather than by romantic declarations of a primacy of authentic emotions over socialised reasoning, by individual subjectivity per se constituting the prime endpoint of the piece of analytical work under consideration.[Note 4]
Let me reiterate here that the typology is useful for thinking about particular pieces of qualitative analysis rather than for labelling authors and complex projects. It is entirely in order that Gubrium (1993) is arguably a contribution in the subjectivist idiom while Gubrium and Holstein (1995) and Holstein and Gubrium (1997; 2000) are ethnomethodological.
There is another important point to add. Once we broaden the category of emotionalism into subjectivism, where the latter includes such excellent works as Gubrium (1993) and Jones and Rupp (2000), there is no need to regard any of the four idioms in itself as superior to the rest. Worthwhile work can be done in any of them. It may be that postmodernism and ethnomethodology are more contemporary in their style of theoretical thought, but that does not mean that radiant academic future belongs to them alone. There will always be good "intellectual market" also for naturalism and subjectivism, and rightly so. I certainly do not propose either to be a pejorative term.
However valid in itself, Silverman's observation that subjectivism tends to take personal biography as its data should not lead to the conclusion that all life story research belongs to this category. Realist analyses of life stories fit into the naturalist category, as they contribute knowledge of social practices, conditions, norms and ideologies that define a social locale, milieu, or class. Narrativist analyses are much less consistent in their endpoints. They can be found in all four categories.
We have seen above that narrativism of the hermeneutic kind can contribute findings concerning shared life-story emplotment patterns and genre. These can have direct relevance to our understanding of norms and ideologies specific to certain social locales, groups and milieus (Miller and Glassner 1997). In that case, narrativist work with life stories belongs to the naturalist idiom of qualitative sociology. Ginsburg (1989), Rosenthal (1991), Plummer (1995) and Maruna (1997) can serve as examples.
Other works of hermeneutic narrativism show primarily individual case(s) of personal identity construction suggestive of a self that has a continuous agency transcending the boundaries of particular interactional situations. They also show that public identities or roles serve as objects in the personal construction, which are indispensable, but to which the narrative achievement of personal identity is not reducible – at least not in the modern era. In some cases they also demonstrate the difficulties of narrating a coherent story in times of crisis, when personal identity is necessarily in a state of confusion and flux. They constitute a contribution to qualitative sociology in the subjectivist idiom. Riessman (1991; 1993), Rosenwald and Ochberg (1992), Frank (1995) and Rosenthal (1997) can serve as examples, in addition to the above-mentioned Gubrium (1993) and Jones and Rupp (2000).
Some narrativist work with life stories, however, highlights primarily the researcher's mindfulness of the problems involved in interpreting a personal narrative reflexively and for a public arena; and of the ways in which this knowledge production is shaped by a style of its representation, which constitutes a practical (political) act. This kind of work makes its contribution in the postmodern idiom. Richardson (1992), Stanley (1993) and Frank (2000) can serve as diverse examples.
Finally, there are the narrativist works that have their analytical focus on the oral telling of life stories or personal experience anecdotes in face-to-face interactions, the arts of self-accountancy that are evident therein, and on the ways in which a social setting occasions self-accounts in a particular style, making narrative identity local in its message and practical function. These works make their contribution in the ethnomethodological idiom. Linde (1993), Irvine (2000) and Middleton and Hewitt (2000) can serve as good and varied examples.
"Isms" are associated with intellectual forms of drum beating. Partisan claims are often made on the basis of hasty arguments or even less. The repertoire of constructivist drum beating, for example, includes casual references to realist works as "naïve realism". I have never heard a user of this descriptor to report that s/he had read something and found it to be a sophisticated (or thoughtful and carefully argued) example of realism. The usage pattern implies that a realist line of enquiry is always naïve while the user of the phrase by contrast thinks thoughts that are sophisticated and constructivist. In the field of life story research, the word "narrative" and its derivatives can be heard as a fashionable drumbeat, although we have seen that analytical orientations to narrative construction are by no means unison, and not even always in cadence with what would be recognized as "constructivist" in the wider field of qualitative sociology. I wish to single out for critical reflection two interrelated notions that seem to me to lie behind partisan claims of sophistication for narrativist works with life story data, and two assumptions on which defensive arguments for narrativist analyses of the subjectivist kind often seem to be made.
In normal English usage narrative simply means, to quote my trusty Concise Oxford Dictionary, "a spoken or written account of connected events in order of happening". Literary analysts, sociolinguists and, indeed, "narratologists" have added to this definition a number of scholarly refinements.
For many, narrative is primarily a story. Burke (1945) defines it as a structure consisting of an agent, an action, a goal, a setting, an instrument, and trouble. It is trouble that drives the drama; it is generated by a mismatch between at least two of the other five constituents.
Labov and Waletzky (1967), whose interest is in everyday speech rather than literature, see a narrative as consisting of an abstract, an orientation, a complication, an evaluation, a result and a coda.
Propp (1968) sees a story as having the aspects of fabula (a mythological element evoking a timeless human theme), syuzet (a particular sequence of actions giving an individual twist to the timeless tale), and forma (genre). He identifies 31 functions (i.e., types of significant action) and 7 spheres of action (roles played by story characters in the emplotment) as comprising the repertoire of story elements and relations on which all folktales draw.
Greimas (1966; 1971) modifies Propp's schema to make his own contribution to the structuralist movement, i.e, the search for the "deep structure" of a relatively small number of elements and binary relations capable of accounting for the apparently infinite number of complex phenomena that can be registered at the "surface level" of "appearance". In Greimas's scheme, the "deep structure" of every narrative is made of a subject and an object, a sender and a receiver of the object, and a helper and an opponent of the subject. Like in Propp's scheme, several characters may be making up just one of these elements in any particular story, and conversely, one character may make a contribution to more than one of these elements.
These theories do not contradict one another. The overlap between Burke and Labov and Waletzky is large. Labov and Waletzky's concept of orientation can be taken to subsume the Burke's agent and setting; Labov and Waletzky's complication comes close to Burke's action and trouble.
The difference between them is mainly in the emphasis that Labov and Waletzky place on story-telling involving evaluative comments, in which the narrator makes explicit the point of the telling, as if to ward off hearers' asking that damning question: "so what?" This difference can be put down to the fact that Labov and Waletzky deal with oral accounts of personal experience. Given as they normally are in the course of conversation, such accounts require certain conversational rules to be temporarily suspended, to allow a speaker the privileges of a narrator (Jefferson 1978; Sacks 1972). Evaluation completes the moves that had secured the temporary and provisional privilege, in that it shows the privilege to have been justified.
Greimas's scheme, which is a further abstraction of Propp's, is so abstract as to be easily accommodating of the distinguishing features of a story pointed out by Burke. Every story indeed has a subject (agent) who acts on an object that has come from somewhere and has some significance (setting, goal, action, trouble), and who finds things and people that help (instrument) as well as those that have to be overcome.
All these contributions, including those by Jefferson and Sacks, are useful in alerting us to the ways in which we recognise a speech or text as a narrative and attune our reading or hearing of it accordingly. This basic recognition is a members' accomplishment, an everyday act of common sense. The theories "merely" spell out explicit and decontextualised versions of what we all know.
In addition, the contributions that address narratives proper rather than the placing of narratives within conversation alert us to the ways in which it is possible to do the scholarly work of comparing narrative emplotments systematically and recognising emplotment patterns and types. A focus on Burke's agent in the European novel, for example, enables us to chart a progression from the folktale figure that is all action and no subjective experience, through Austen's persons defined by roles, and Trollope's selves, who compete for roles to earn rights, to Beckett's individuals, who resist society to "rip off" their rights (Rorty 1976). Or, it can serve to compare orally produced life stories for the degree of self-agency on display, and perhaps to cross-compare these with the social settings in which they were told (Ezzy 2000). A focus on Labov and Waletzky's evaluation enables us to reduce large narrative data sets into manageable summaries for further sorting, without losing information on the evaluative standpoints claimed by the narrators (Cortazzi 1991; 1993, 47-51, 120-2). The range of particular ways in which it is possible to carry out the analytical work of emplotment pattern recognition and further sorting is, I believe, virtually endless.
The narrative-defining theories or schemata help us to be aware of possible starting points in our search for patterns that will have something to say to the research question that got us going in the first place. They are not themselves a starting point of some cumulative science that will produce discoveries of deep-seated formulae that make the world turn around, but that can be known only to the science. Greimas's schema is not deeper than other narrative inquiry starting points, although it is couched in structuralist noises about deep structure. It is simply more abstract. Famous observations on sociological "grand theory" come to mind, which point out that high-level generalisations can be curiously uninformative compared to "middle range" theories.
The informativeness of analytic inquiry depends on the creative act of matching research goals with knowledge fields and available data sets, and finding analytic procedures that work with the mix. Thus Maruna (1997) adopts none of the above-mentioned narrative schemata when faced with autobiographies of reformed criminals and a question concerning popular conceptions of reform. Instead, he generates his information about a telling emplotment pattern with the help of McAdams' (1993) observation that life stories have nuclear episodes, imagoes, and themes. It does its job. It yields an emplotment pattern that is telling, as far as Maruna's research goal is concerned. There is no more and no less depth to McAdams' schema than to Greimas'. Both are close to what we already almost know, being as we are practised everyday users of the narrative form. The claims that there is a deep structure to narratives are based on a borrowing from concepts of deep grammar in generative linguistics. The borrowing asserts analogy between syntax and narrative; it makes no compelling argument for it (Cortazzi 1993, 87).
In addition to structuralist allusions to deep-level syntax, narrativist partisanship tends to draw on arguments associated with the "narrative turn" of the 1980s.
The "turn" was an aspect of the demise of social theory, which intellectual endeavour it gave less heroic character, by proclaiming it to be just attempts at spinning a "grand narrative" of progress. It was nicely presaged by White's (1973) analysis of selected works of history, which were generally deemed to be landmark achievements in the scholarly professionalisation of the discipline. It showed the works to be works of literature, in that the powers of historical understanding they conveyed were generated by strategic use of tropes and dramatic emplotment genres. The professional discipline prided itself in methodical work with documents and categorical rigour, but its most influential works created historical understanding by their narrative artfulness.
More comprehensive works of theory followed, to argue for the centrality of narrative emplotment in our very thought, action and self-construction (MacIntyre 1981; Ricoeur 1984, 1985, 1988; Bruner 1986; 1987; 1995; Polkinghorne 1988; 1991; 1995; Somers 1994). However, compelling as these arguments (and their antecedents, such as Burke 1945) are in themselves, I doubt that they add up to an overthrow of other arguments about the nature of human existence, which put other abilities than story making in the limelight. Neither do they provide much guidance concerning the kinds of empirical research programme that could give the theoretical intimations of narratively constructed existence their deserved follow up.
The "narrative turn" arguments give many reasons why close attention to narrative should pay good dividends in sociological work with life stories. Narratives organize memory; configure experiences and events into meaningful gestalts; provide models for interpreting action; give meaning to the passage of time; facilitate reflexive soliloquy in which we may see ourselves as others would; convey morals and create occasions for their collective affirmation. Narratives create us as continuous and coherent agents of life; as different from others but also co-members with them, accountable and normatively oriented; as individuals overcoming social difference. They enable us to negotiate subtle moral standpoints in morally complex times, because they admit a range of dialogic voices, while affording the narrator freedom of iterative movement between being a narrator, a protagonist in the story, and a participant in an interactional process outside it.
Narratives are useful in social practice. But so are other social forms. And, narratives are everyday folk accomplishments; understanding them and recognising recurrent patterns in them requires intellectual effort that, per se, has no more and no less scientific grandeur than other kinds of sociological work.
We have had an epidemic of qualitative methodology texts. The (British) academic book market now absurdly seems to offer more texts on how to do qualitative research than monographs showing inspirational examples of it. The commercial logic is obvious. Monographs have a substantive topic. They have a good chance of being bought or ordered for a library only by those who already share an interest in that topic. Methodology texts by contrast enjoy wide markets, which are currently widening further, thanks to curriculum harmonisation pressures that require research methods to constitute a large central core of social science teaching and training at every level. They enjoy wide markets because they address no specific research question. The logic of their existence is that the natures of the social world and sociological knowledge are such that the employment of the right kind of procedures for data collection and analysis produces valid knowledge, never mind what the topic is. In some cases, perhaps more by implication than by design, they in effect nominate certain procedures for authoritative status – as methods that ought to be cited to claim professional respectability in research proposals and monographs.
Methods texts can be useful as a source of hints and stimulating thoughts about academic discipline, but they need to be taken with a good pinch of salt – indeed with vigilant scepticism. If, as I have asserted above, good research is a matter of creative matching of disciplinary fields (theoretical discourses and problems), researchable topics, specific research questions, obtainable data sets, and data-reduction and analytic procedures, methodological texts by their nature cannot do justice to the crucial equation between research goals and methods. They highlight the methods at the expense of the goals. They draw their authority either from unexamined theories of "deep structure" that only a certain method can reach, or from an even more obscure source, the one that gives us other social practices distinguished by their ritual adherence to prescribed means in disregard of fuzzy ends, such as those that Robert Merton described as "bureaucratic ritualism".
The methodological publishing boom took a while to reach the field of life story research. In the last decade of the last century, a search for a life story methods textbook might yield Denzin (1989), providing perhaps more a useful theoretical introduction to the biographical field than a methods text, couched in postmodern argument; Riessman (1993), discussing the approach of two other monographs before her own latest work, from which she singles out transcription geared to catching the rhythm of oral speech for special attention; Atkinson (1998), a folklorist celebration of life story interviewing per se as much as a social science text, which my undergraduates love to cite, somewhat to my embarrassment; and Lieblich, Tuval-Mashiach and Zilber (1998), discussing their own recent research in the field of educational psychology, to show some of the ways in which analytical focus can be either on story content or on narrative form, and either on the story as a whole or on some categorical part in it.
The first two years of this century have brought us Miller (2000), Chamberlayne et al. (2000), Hollway and Jefferson (2000b) and Wengraf (2001). These four volumes take up more shelf space than the four cited for the preceding ten years. Miller (2000) is an introductory undergraduate text, organised around a distinction between three approaches – realist, neo-positivist and narrativist. It refrains from far-reaching claims for its methodological arguments. The other three do not. Chamberlayne et al. (2000) claim no less than a Kuhnian paradigm shift in their opening paragraph, on grounds whose confused nature fails to obscure their inability to stand up to examination; Hollway and Jefferson (2000) seek to tool up life story research with Kleinian psychoanalysis, so that researchers can have leverage to open up the ego defences of their interviewees (defended subjects); and Wengraf (2001) is the ultimate authoritative methodology text - in its rhetoric certainly, bristling as that is with tables, diagrams, neologisms and acronyms.
All four texts give pride of place to the procedures developed by Fritz Schuetze and his colleagues in their life history research, which were first outlined in English by Gabriele Rosenthal (1993). Miller draws on them in his discussion of narrative analysis, Hollway and Jefferson give them credit as their major inspiration, Chamberlayne et al. make them the central canon of their Kuhnian revolution, and Wengraf of his textual claim to methodological authority. The canon has been fortified in its status by getting an acronym in the last two texts – BIM in Chamberlayne et al. and BINM in Wengraf, who thought that an initial for the word "narrative" needed to be inserted amongst those standing for "biographical interpretive method".
There is much that BIM shares with acronymless qualitative methods. It favours interviewing based on open-ended questions and encouragement to the interviewee to volunteer narratives of experience. At the analytical stage, it prescribes text segmentation and hermeneutic attention to overall message-defining thematic structure as well as to the sequential ordering of content and the interplay of discursive forms ,[Note 5] with taking the interviewer-interviewee dynamic into account; and it insists on checking the validity of interpretations against what can be found in the text (transcript) itself.
There are, however, other sides to BIM that many a qualitative researcher would have reasons to question: in interviewing, for example, the preference for long silences over offering conversational leads to the interviewee who is at a loss where to start or what to say;[Note 6] or, in analysis, insistence on exploring individual segments strictly in their order, instead of allowing iterative movement between considering the whole text and individual segments. These BIM peculiarities are not necessarily wrong, but whether their rigours are worth following surely depends on research goals.
The immediate and self-evident goal of BIM is to understand the narrator. This quest, as it happens, does not stop at understanding the overall message and viewpoint of the narrative itself. BIM prescribes also a comparison of the sequential order and logic of the narrative with its chronological life history content. This is supposed to enable the researcher to grasp life events as they were experienced in their real time, and also to grasp the principal organising theme of the life.[Note 7] The prize promised by BIM is to understand the narrator's life as it was lived as well as the way it is understood by its subject now, and as having an objective meaning as well as a subjective viewpoint. This all-synthesizing, totalising notion of what there is to understand about an individual life strikes me as bordering on the mystical.
However, even if it were the case that BIM can deliver an objectively valid summary of a real individual life, how does that cognitive potency relate to other possible research goals, such as those that might interest qualitative sociologists in particular? The methods texts do not tell.[Note 8]
A contribution in the subjectivist idiom is the default endpoint of BIM-type hermeneutic. It is a short step from this default to psychological arguments claiming knowledge of the biographical subject such as may have practical applicability in therapeutic dialogue. Sociological theorists who argue, quite compellingly, that frequently revised biographical self-construction is increasingly important in contemporary society for the individual's (re)adjustment and coping, tend to reinforce belief in the psychotherapeutic value of life-narrative interviewing, although they wisely council caution in this respect (Fischer-Rosenthal 1995; Giddens 1991). And last but not least, the seductiveness of the short step into psychological territory is enhanced by the fact that that is where much of the impetus and development of hermeneutic work with life narratives has occurred (Mishler 1986; Rosenwald and Ochberg 1992; McAdams 1993).
However, the short cross-disciplinary step is not necessarily enriching if the grass on the other side in the event turns out to be no greener than the home patch. Psychology is teeming with theories of therapeutic promise, which have been much more successful in gaining adamant believers than in passing therapeutic effectiveness tests. The latter show the application of Freud's scientifically elegant psychoanalytical theory to have no better results than the much less carefully argued mystic new-age avenues to the human soul. Many kinds of talking therapy can prove to do good sometimes, but there is a case for hypothesizing that that is not due to their theories as such, but to the process of negotiated passing into and out of successive stages of the therapist-patient role dyad.[Note 9] It is not the content of the theory that matters, but the process of interaction in which the patient learns to display examples of the validity of the theory, in situationally appropriate ways (Schegloff 1963). Viewed in this way, the history of psychological theorising about the person is not very encouraging to the view that, at last, the right theory has been found in life-narrative hermeneutic.
Apart from these general reservations about letting analytical work with life narratives lead into the psychological terrain, particular instances of that are open to criticism on either of two counts: (a) that psychological theorising adds mere speculative conclusion to otherwise well grounded analysis of life narratives; and (b) that it actually prevents or discourages due consideration to what the data show and how that might be interpreted.
Hollway and Jefferson (2000a) are an instance of the first type. Their hermeneutic work on two life story interview transcripts shows two contrasting perspectives in which the interviewees, both of whom are residents in the same high-crime neighbourhood, assess the everyday risks of living there. Kelly's perspective is defined by categorical divisions of the neighbourhood into risky and safe areas, and her own social and geographical mobility from the former to the latter. Joyce by contrast roots her identity centrally in the neighbourhood, and finds her sense of safety in her personal acquaintance with her neighbours, troublesome adolescents and drug-dependent young men included. These contrasting subjectivities of risk assessment seem well grounded in the respective narratives, and the logic of each is presented well enough in its own terms to facilitate empathetic understanding of it. We may wonder what other perspectives the authors found in the course of their interviewing in the neighbourhood, and whether Kelly and Joyce represent types of narrative emplotment into which most other cases could demonstrably fit. Instead, we get told that the two cases are an instance of a fundamental psychological division in human responses to things, an outcome of our infantile struggles with the objective realities of maternal breast: we can view things in a paranoid-schizoid perspective like Kelly, or in a depressive one, like Joyce, the latter being the one that permits us to view things as mixtures of good and bad.
Lieblich, Tuval-Mashiach and Zilber T (1998) inadvertently show how psychological analysis can actually block sensitive work with narrative data. In their chapter on the kind of analysis that focuses on categorically selected transcript segments and the narrative forms found therein, they claim that particular features of a narrative show the narrator's cognitive skills and emotions. They bring their favourite theorist of cognitive functioning (Carl Frankenstein) to bear on several short excerpts from a life story interview transcript. They adjudge an excerpt featuring a detailed description of recalled experience to be lacking in abstraction, which they take to be a symptom of irrational thinking. The same excerpt, however, could have been read as the kind of concrete and detailed description through which narrators attend to the narrative task of making their recollections heard as authentic.
Lieblich et al. declare another narrator to be cognitively impaired, this time because he asserts logically unwarranted generalisation, on the basis of an excerpt where he concludes his recollection of childhood moods with the statement that everyone experiences such crises. Such an assertion of normality, however, is a frequent occurrence in life story interviews, which can be much more plausibly read as a social act, as an invitation to the interviewer to confirm that s/he is not interpreting what s/he hears as a sign of personal inadequacy or deviance.
The authors likewise ignore the interactional context within which a narrative takes shape when they discuss what they take to be linguistic clues to narrator's emotional states. In an excerpt in which a story of a car accident is told, they note that the narrator does not tell straight away what happened to other passengers, which they interpret as a sign of an emotionally difficult recall. The same excerpt, however, could have been interpreted in terms of a conversational context where the narrator could legitimately expect her interlocutor to help to tie up the threads by asking a question. Again, narrative performance that is geared to drawing the audience into helping out in this way, and displaying interest, is by no means uncommon.
And finally, in the same chapter, Lieblich et al. take an excerpt showing an embellished narrative as an effort to cope with emotionally uncomfortable facts. However, had they been paying attention to narrative construction rather than to psychological theories of cognition, emotion and ego defence, they might have considered the case to be one of a story which, because it plays an important part in the overall narrative, is padded up with markers of authenticity to give it richness and length.
I cannot be sure that my counter-interpretations of the excerpts are more valid than the authors', the chapter does not show enough of the original data for that kind of certainty. It seems to me, however, that the authors' interpretations can be considered compelling only by those readers who share their a priori belief in a particular psychological theory, so strongly that whatever the data show of social action or communicative practice (rather than of cognitive functioning and emotional states) is brushed aside as being of secondary significance.[Note 10]
Another kind of short step from the default subjectivism of life-story hermeneutic is to give up on identifying a disciplinary theoretical discourse to which the finding of a life story's thematic structure or emplotment may constitute a relevant contribution. The implicit or explicit justification of this anti-academic stance is that allowing the narrator to find his or her narrative line, and making the voice thus found heard as an authentic self-assertion in a public arena, is a worthy social act in itself, which has greater humane value than a contribution to the theoretical discourse of an academic discipline. Leaving a life story to speak for itself and presenting it for its human interest alone, however, rather relies on keeping shared beliefs concerning the authenticity of self from reflexive articulation. It romanticises individual subjectivity and offers it for celebration. Where celebration starts, however, intellectual inquiry stops.
The romantic impulse to celebrate the subject as transcendental agency has been under sustained criticism from qualitative sociologists of more ethnomethodological persuasion (Silverman 2001, 287-90 and 294-7; Atkinson 1997 ;[Note 11] Atkinson and Silverman 1997; Holstein and Gubrium 1997; 2000; Gubrium and Holstein 1997, 57- 74). This sustained criticism carries with it the danger of throwing the baby out with the bathwater, in so far as it may appear to damn the very impulse without which many a project of qualitative fieldwork would have never started. It serves, however, a worthwhile reminder that social scientists' professional responsibility is to keep intellectual inquiry going even when there are seductive (and other) calls to stop.
Contributions in the subjectivist idiom do not need advocacy. They are a natural first stop in the line of enquiry that starts with taking a hermeneutic approach to a life narrative. The narrativist research themes that I wish to advocate belong to the ethnomethodological idiom, which has not been in frequent use among biographical sociologists, and to the naturalist idiom, which has been the home tongue of biographical sociologists who take a realist approach to their data.
My first advocacy is for a theme in an ethnomethodological idiom. The relating of a life story is an exquisitely social process. When it is naturally occurring outside formal institutional realms, it modifies members' identities vis-à-vis one another and constructs a personal relationship between them. As Linde (1993) notes, personal relationships involve rights to know the other's personal biography, and to be told the news without much delay when an event has occurred that might prove significant enough to feature in the other's sense of their life history. Formal institutions have utilised this "doing intimacy" aspect of life-story telling too, to create biographical occasions in which an institution can practice displays of rational regard for "the whole person". In telling a life story, we make ourselves accountable as coherent and continuous persons. Making oneself accountable is of course a universal aspect of interactions. Life story, however, is a type of discourse unit that, in addition to requiring the use of narratives to create a coherent, continuous, related and reflexive self, requires the narrator to be the principal object of evaluation throughout (Linde 1993). The business of making oneself accountable is more upfront and intense than in other types of interaction.
Alasuutari (1997) argues that the telling of a life story involves the Goffmanesque and universal collaborative process of maintaining face, and in addition, the collaborative process encoded in narrative implicature, the presumption of narrator and audience that every narrative has a coherence-giving main theme. We might add to this that the intensely reflexive accounting process also involves the broader joint enterprise of creating a shared sense of cultural co- membership. The arts of face maintenance and the conversational and narrative implicature clearly play an important part in this, but so do, for example, searches for common cultural references and character recognition. The telling of a life story involves the creation of a shared micro-culture in which the narrator and audience are co-members with different but complementary personal identities. This specific achievement occurs by negotiated drawing on more general cultural resources, roles and identities, as well as by references to preceding events in the life-narration process itself.
Interactionally rich though this process may be, it is still very under- researched. There is a potential for generating a wide range of specific empirical topics, and for innovating appropriate data reduction protocols. Research in the ethnomethodological idiom has become dominated by conversation analysis, with its own standardised data reduction protocol, the CA transcript. The latter, however, is feasible to use only for topics relating to the smallest and most universal elements of what Holstein and Gubrium (2000) call "the machinery of everyday conversation" and "the everyday technology of self-construction". Life-story self-accounting is hardly an everyday process, and yet it is an important component of social life; and it is probably subject to greater cultural differentiation than the widely cross-cultural rigours of conversational turn-taking and adjacency pairs. It requires a wider lens, ways of data reduction that will enable systematic examination of the life-narrating process and the testing of interpretations at a level that can take on board its whole-length dynamic. Like CA, it will capture a collaborative process, a joint enterprise in making accountable persons; but it will do it at a level where larger and culturally more specific patterns come into evidence.
This broad theme is a close relative of the one above. Co-membership involves show of orientation to common social norms and to public discourses in which social norms and morals are defined and affirmed or contested. Life narratives, however, also require the narrator to establish his or her individuality. They require that s/he demonstrates moral choice and defines a personal position vis-à-vis the normative frameworks of public discourse. They demand that the narrator establishes a morally adequate identity vis-à-vis his or her audience. This is delicate interactional and narrative work because there is no context-free way in which a person can establish herself to be, for example, diligent without being obsessive, or fun-loving without being frivolous. It requires the negotiated establishment of contexts in which one's actions and moral choices can appear to be naturally placed. The construction of moral adequacy in life-story interviews has already been the main topic of some analytical work (Chase 1995; Andrle 2000), but there is scope for a lot more.
There are many institutionalised settings in which members are required to do biographical work (Gubrium and Holstein 1995). Job interviews, TV chat shows, health and welfare workplaces, voluntary self-help groups, school reunions, corporate webpages, to name just a few. They can be studied in the ethnomethodological idiom. Self-accounting is a general and fundamental social practice, but it is always done locally, in a specific institutional context.[Note 12] These local productions of biographical accounts and identities have been the focus of recent studies (Irvine 2000; Davis 2000; Middleton and Hewitt 2000), but again, there is room for more. Autobiographical occasions (Zussman 2000) can be studied also in the naturalist idiom, with an eye on their wider social control function (Vinitzky-Seroussi 2000).
Gergen (1991) tells us that postmodern times have brought about a change in the way we fashion our biographical selves, in that the self has become fragmented and its narratives less linear.[Note 13] He probably exaggerates, but his argument reminds us that, on the special occasions when we are asked to relate an extended life story outside of a specialised institutional context, we may draw on our knowledge of relevant cultural genres to make our story recognisable as an autobiography and understandable in its emplotment. Some genres appear to be long lasting in history; others are perhaps susceptible to short-term change. And, some genres at certain times appear to be available for use only to certain social classes, genders or ethnicities. Autobiographical genres provide terms in which we describe ourselves as coming from a certain social location, the ways of our social self-anchoring. Empirical research into their usage provides good basis for identifying patterns of social change.
This is easier to do on literary autobiography than on oral life narratives, which are more complex discursively and often rather less tight in their emplotment. Every autobiographical account, however, offers a notion of social coordinates in which the origins of the person are to be understood. The importance of autobiographical genre in life narratives has been often asserted theoretically, sometimes in analytically unhelpful ways, in that too many examples of possibly relevant genre have been mentioned without much discrimination (Bruner 1987; 1995). Systematic and comparative empirical work on thematic and emplotment patterns of extended oral life narratives is still waiting to be done. This will be a contribution to qualitative sociology in the naturalist idiom.
End.