Abstract: The paper offers an analysis of how narratives may be reflexively used at different stages of the research process as a tool to access the interpretative frameworks that actors use to construct their accounts of events and to make sense of their action. Inter-subjectivity is a determinant of the action research process as it is in the interactions between participants (professional researchers included) that certain versions of knowledge are produced. Action research can be seen as an ongoing process in which different narratives are co-produced allowing different interpretations to be actively constructed by participants. A case study is briefly presented where reflexivity is used both to clarify how accounts (narratives) are constructed (identifying concepts and categories used by participants to make sense of their action) and to allow different forms of knowledge to be developed by participants. Reflexivity is intended here as being inherently connected to action and as a part of the sense-making process in which both participants and the researcher are engaged. The problem of power relations between professional researchers and participants is discussed.

Key words: reflexivity, narratives, inter-subjectivity, interpretative frameworks, action research

Reflexivity has become a legitimate topic for discussion for qualitative research as the "linguistic" and "interpretative turn" has emerged in philosophy, psychology and social theory. Norman DENZIN and Yvonne LINCOLN (1994) argue for three broad divisions of qualitative approaches in social science: 1) positivism and post-positivism; 2) constructivism and critical theory; and 3) interpretative perspectives. There has been a focus both on language and discourse as well as on practice as inquiry at the heart of constructivism and interpretative perspectives. As Eric STEWARTS (2000, p.726) points out, language "as both constituting and being constituted by social practices, and as spanning the conceptual divide between individual and culture, private and public, becomes both the object of and a vehicle for social science research". [1]

According to this point of view, knowledge is not sustained by its correspondence to an objective reality, rather it is inherently constructed and sustained by social processes, or, in other words, by communal practices. As Vivien BURR (1995, p.160) argues, objectivity is

"an impossibility, since each of us, of necessity, must encounter the world from some perspective or other (from where we stand) and the questions we come to ask about that world, our theories and hypothesis, must also of necessity arise from the assumptions that are embedded in our perspective (...) The task of researchers therefore becomes to acknowledge and even to work with their own intrinsic involvement in the research process and the part this plays in the results that are produced. Researchers must view the research process as necessarily a co-production between themselves and the people they are researching". [2]

From this point of view, experience is interpreted by means of open processes dependent on the context, which are therefore variable according to the situations and interactions, which are then mediated by means of the language, discourses and narrations organising this experience. Here, reflexivity is strictly connected to the notion of inter-subjectivity since it is in the interactions between participants in the research process (professional researchers included) that certain versions of knowledge are produced. It is no coincidence, for example, that hermeneutics places at the centre of its practice the metaphor of the conversation between the researcher and his materials (GADAMER, 1983). At the same time, the notion of reflexivity means for researchers the application of theory back onto itself and its practices. [4]

This "interpretative turn" which touched the social and psychological sciences led them to redefine goals and research practices within the area of the narrative approach as well (see BRUNER, 1986, 1990; GEERTZ, 1973; SARBIN, 1986). Various theoretic outlooks in various disciplines emphasise in fact the essentially interpretative character which distinguishes the analysis of narrative material (cf. DENZIN, 1994)¹⁾. Still, it is important to distinguish the narrative approach from the narratological approach (FORMENTI, 1998). The former aims at identifying the generative processes of the narration and ways in which experience is discursively constructed by means of consensual linguistic practices; the latter refers to the analysis of narrative as text with its constituent elements and structures. It is without a doubt the narrative approach which, in my opinion, allows one to face the problem of the relationship between research and reflexivity. [5]

In the narrative framework, the metaphor of the story refers to the idea that the most important means with which individuals give meaning to an experience consists in inserting it in an narrative structure (BRUNER, 1986; MISHLER, 1986). This idea has been developed in various ways in different approaches, but comes from two basic ways of thinking about the functions of stories: on the one hand, it has been found that narrative constitutes one form of interpreting actions, and on the other, there is the fact that by means of narrative the individual constructs the continuity of self on a temporal level, bringing together the past, present and the future. The role of narration as the organising principle of action is specifically emphasised on a psychological level, starting with the distinction introduced by Jerome BRUNER (1986) between paradigmatic thought and narrative thought. These two forms of thought have their own characteristics and functions; the thought procedures vary in different fields of knowledge: with regard to the action, they are interpretative rather than causative, associative rather than deductive, thematic rather than categorical, and descriptive or prescriptive rather than predictive (FELDMAN, 1991). [6]

In this perspective, narratives may be considered as a tool to access the interpretative frameworks that actors are using to construct their accounts of events and to make sense of their action. According to Theodore SARBIN (1986), narration includes not only the description of actions and events but also the actor's reasoning and motives. Within stories, in fact, actors inscribe a multiplicity of complex conceptual structures and transform transitory events and actions into stories that can be re-examined and revisited: "In this way, events and actions which are punctiform with regard to space and time, and also highly changeable, reveal under observation their complex specificity, their dense structure" (LANZARA, 1993, p.239). [7]

These versions of events can be considered in relation to the context in which they are produced as constructions oriented toward action. In this framework, narratives not only provide a thick description (GEERTZ 1973; 1983) of events, they are an inherent part of the events. One of the central theses developed by Derek EDWARDS and Jonathan POTTER (1992), within the area of the discursive approach²⁾ is, in fact, that descriptions and accounts are supplied in contexts characterised by the presence of a controversial question (see also DREW, 1984; POMERANTZ, 1984) or of a dilemma; one of the characteristics of accounts is that they offer possible alternative, implicit or explicit versions. This has important implications on an analytic level: instead of trying to explain the nature of an account in relation to an external reality predefined by the researcher, it is necessary to consider the account in relation to alternative versions supplied by the participants. The term "interpretative repertoire" introduced by Jonathan POTTER and Margaret WETHERELL (1987) refers to a stock of available linguistic devices from which people may construct accounts. According to Jonathan POTTER and Margaret WETHERELL (1987, pp.33-34) the term "construction" is "apposite for three reasons. First, it reminds us that accounts of events are built out of a variety of pre-existing resources, almost as a house is constructed of bricks, beams and so on. Second, construction implies active selection: some resources are included, some omitted. Finally, the notion of construction emphasises the potent, consequential nature of accounts. Much of social interaction is based around dealings with events and people which are experienced only in terms of specific linguistic versions. In a profound sense, accounts "construct reality". [8]

The perspective of research and intervention presented here refers as a whole to these orientations and it avails itself—to different degrees and at different levels—of conceptual and methodological instruments developed in these areas. [9]

Although there are differences of opinion about what constitutes action research and practice as inquiry (see TORBERT, 1991; REASON, 1994; NEWMAN, 2000), the various forms of research have in common the fact that they are fundamentally based on the idea that performing skills can be translated into more general knowledge by means of a circular process which connects, at various levels, declarative-propositional knowledge with practical knowledge and experiential knowledge. In this sense, it underscores the fact that by means of the research process it is possible to undertake a circular comparison between knowledge and actions which allows for the bringing forth of habits, cultural models and styles of thought. [10]

In the perspective which we have adopted, it is a question of instituting a research context as a place in which it is possible to construct a program of analysis and reflection around the problems which are met in (work) practice with the aim of sustaining sense-making processes, i.e., that of the continuous construction which occurs when actors make sense retrospectively of situations and events (WEIK, 1995). This type of formulation is characterised by the attempt to avoid separations of theory and practice, by the interdisciplinary approach and by the central role of problem-setting, i.e., of the possibility of constructing and defining problems starting from the interpretative frameworks stated and developing more increasingly complex interpretative categories. [11]

Given these premises, reflexivity is intended here as being inherently connected to action and as a part of the sense-making process in which both participants and the researcher are engaged. Consequently, the aim of inquiry moves towards a consideration of how certain phenomena or forms of knowledge are achieved by people in interaction: "Knowledge is therefore seen not as something that a person has (or does not have), but as something that people do together" (BURR, 1995, p.8). Problems and theories are not given right from the start of the research, but are constructed and reconstructed during the research by means of continuous transactions among the actors. [12]

With regard to research, this becomes a question of exploring the temporal evolution of these interpretations and to understand how the actors involved in the process—including the researchers themselves—construct and reconstruct the events and how these reconstructions direct their actions: "The actors interpret events and actions not so much by imposing on them a fixed structure as by "translating" the events and actions over time and by developing new frameworks and "stories" for their (re)interpretation" (LANZARA, 1995, p.227)³⁾. Action research can be seen as an ongoing process in which different narratives are co-produced allowing different interpretations to be actively constructed. This process unfold on two levels: one regarding the use of individual and collective stories, tales and narrations as heuristic sources and instruments for generating clarifications and for constructing meanings with regard to the actions and events; the other being inherent to the research process itself, and being understood overall as a narration which is created by means of this interplay of connected constructions and the negotiation of meanings by those involved in the research. [13]

In order to analyse how narratives may be reflexively used at different stages of the research process as a tool to access the interpretative frameworks that actors are using to construct their accounts of events and to make sense of their action, we briefly present here a case study where reflexivity is used both to clarify how accounts are constructed (identifying concepts and categories used by participants to make sense of their action) and to allow different forms of knowledge being developed by participants. I was asked to design an action research which would involve the social workers at a residential community for adolescents. The team is composed of a coordinator and five social workers. The community is mixed and accepts Italian and foreign minors between 14 and 18 years of age. The need presented by the director of the community was to construct an accompanying framework for the team which would allow the social workers to systemise the knowledge acquired during their work experience and to enhance their particular skills, and also to rework the education project. [14]

In line with these premises, the intervention had as its aim the development of a reflexive analysis about problems which, on a group and individual level, the social workers experience during their daily work in the context in which they operate. This implies taking into consideration the totality of knowledge, skills, theoretical frameworks and methodological instruments which is adopted to carry out the task, by using the existing or past experiences, both positive and negative, as material to analyse the existing problems and to enhance the knowledge generated through these experiences. This is possible by encouraging the social workers to process the information present in this context to decide which work can be effectively carried out and actually realised, and by giving value to daily experience as a source of knowledge. [16]

This research was carried out during ten monthly meetings, each lasting three hours. The research group included two researchers along with the team of social workers. The contents of the group discussions were transcribed. From a methodological point of view, the research technique was distinguished by its reflexive procedure with questions repeated to the actors so that their respective points of view were made explicit by means of a constant questioning and self-questioning during the entire process. This self-questioning also made possible the re-programming during the research by the researchers and a self-evaluation of the research methods adopted. In this sense, in fact, the researchers were also involved in the task of revision of the data emerging during the group discussions which allowed them, on the one hand, to widen their knowledge of the organisational context and the group process, and, on the other hand, to construct an increasingly complex interpretative framework with regard to the problems dealt with in this research. As will be seen, this work permitted a constant reorientation of the research on the basis of the data and the interpretative hypotheses which gradually emerged. [17]

Moreover, with regard to methodology, we decided to produce two documents after each meeting, one prepared by the team of social workers, and the other by the researchers. These documents, which were the subject discussed during the following meeting, synthesised the most meaningful data which emerged during the discussion, the problems presented and the interpretative hypothesis gradually formulated by the social workers and the researchers, thereby providing the chance for all the actors to meet. In this sense, it was also possible to assume as a subject of discussion and interpretation of the research process itself. [18]

The reflexive analysis took place on two levels: that of the declared theories and that of the theories in use (ARGYRIS & SCHÖN, 1995). The declared theories refer to the account which the actors supply with regard to the reasons for their actions, while the theories in use constitute inferences about the theories underlying the actions observed, and can be made explicit by thinking about the action. This allowed the social workers' team to explore the explicit and implicit representations which orient and make sense of their action. The definitions were accompanied by accounts of episodes and events connected with both the life in the community and the experiences of the social workers themselves. [19]

During the planning phase, the research staff established the first framework with regard to the context in which the research would be carried out so as to allow them to orient their actions. It consisted of some general thought on the problems connected with work in a residential community for adolescents, which were clarified and discussed with the participants during the initial phase of the research. In fact, we considered it important to share with the participants the premises and the assumptions which would give meaning to our actions during this process⁴⁾. In this way, right from the start, the meaning that this research might have in this context was made visible: the action research of shared meanings. Moreover, the goal was to allow the participants to evaluate the results of the observations and the descriptions made by the researchers, i.e., the stories which the researchers would gradually tell during the project, starting from the materials brought by the participants. [20]

This feature seemed to us to be a central element with regard to the problems which social workers face in their work. Work in residential communities is characterised by the complex problems to manage and by the high degree of uncertainty which the social workers must face; also, in these organisations, it is often a complex matter to define clear work tasks. The declared goals are often extremely general ("re-educate", "rehabilitate", "treat") and it is difficult to find a confirmation that the goals presented are actually reached by the community. Each community is based, explicitly or implicitly, on a theory of person-environment relationships from which emerge a series of operative criteria which orient the actions of the social workers and define its function on an organisational level. It is important to determine on which theoretical principles, as well as on which values and general goals, the intervention of the community is based in order to help the staff overcome the difficulties which they often encounter when actually defining the task which the community, understood as an organisation, can carry out. [21]

In the case of communities the task often derives from two essential contents which are seemingly contradictory. On the one hand, in fact, the task of the community is to "control" deviant behaviour, problems, existential difficulties and destructive urges; and, on the other hand, to "grow up", i.e., encourage development, autonomy and the ability to relate with others and the environment of subjects which hosts them (MANOUKIAN, 1992). Numerous research projects have shown that the presence of both these features often creates problems and that communities tend to structure themselves, in terms of model of internal function, so as to take on, or concentrate on, just one of the components of the task (see CARDONA, 1992). With this general interpretative framework in mind, one can deduce that the social workers act within a contradictory situation deriving from the presence of various interpretative frameworks and orientations, which, in turn, give rise to alternative versions of events. [22]

During the initial phase of the research, definitions of education were gathered. It was therefore possible to find a dominant narration (RAPPAPORT, 1998)⁵⁾ based on the following position: to educate = to transmit values. This narration corresponded to one of the explicit theories and constitutes one of the frameworks prevalently used by the participants to interpret and give meaning to their actions. Now we will present some of the definitions produced during the discussion which refer to this idea of education: "Here we offer an alternative to negative values." "We can't pretend to have no ideas with regard to this situation. If for them (minors) money is everything, our model is that of a poor community. We give a relative value to money. We express a choice and we want to transmit it." This discussion on values seemed to derive on the whole from the problem of controlling potentially deviant and antisocial behaviour of the minors by the adoption of strategies and practices for their re-education and re-socialisation on an essentially ethical and moral level ("If it were not for this education regarding values ... these kids must know how to interact with others, and know what the rules are, even if they don't agree with them. We intend to and try to communicate another concept: know the rules of the society in which you live"). [23]

The researchers focussed their attention on the signs that pointed to the presence in what the social workers said of other positions and narrations in order to support the construction of a more complex vision of the educational action. Other accounts, in fact, contained reference to a different vision of educating: to educate = to help acquire capacities/skills, or in other words, to accompany these individuals in their growth and the acquisition of greater autonomy ("We try to develop their ability to think critically by offering something new." "We try to assist them in learning to think critically, to know how to distinguish rights and duties." "To help them to represent their surroundings ... Go outside and see what's there. It's important to know what's outside"). [24]

The accounts that emerged during the discussions constantly showed this oscillation between these two positions. Even the goals of the educational project stated by the participants during the initial phase of the research derived from these two macro-categories: that of values and that of the capacities/skills which the adolescents can acquire from their experience in the community. Still, these two ideas tended to develop by means of a seemingly rigid alternation between the two positions without leaving room for any possible integration. Since the terms used during the discussion on the theme of the goals of the educational action derived from a group of representations implicitly connected to the theme of learning (i.e., to one of the theories in use), the researchers chose to clarify that connection. So, it was possible to undertake further analysis with the aim of bringing forth and exploring these representations by paying attention at the same time to the problem of the congruency between the object (what is learned) and the processes (how one learns). This discussion then developed from two questions: How do we learn values? How do we learn capacities and skills? And, to be even more specific: What and how do we learn in this community? [25]

Overall, these different representations of learning expressed by the participants can be seen to derive from the following ideas:

The clarification of the reasoning underlying their ideas on learning allowed the social workers to recognise the connection between educational practices and the theories which, often implicitly, are behind them. During the following phases of the research, the group gradually realised that these ways of reasoning put into play representations and prefigurations which are different not only with regard to the educational relationship and the role of the educator, but also with regard to the cognitive aspects and emotional dimensions during the learning processes, with different concepts regarding development, change and the processes of constructing identity in adolescents. Moving discourse from education to learning allowed them to widen the interpretative framework within which to place their action and to discover the connections between the two positions initially defined in terms, as we said above, basically rigid. In this way a new position gradually emerged: "You cannot teach values but you can only educate within an affective relationship. What should be considered is my personal relationship with values." With this affirmation we can observe a passage from a linear/transmitted conception with regard to the learning of values (the initial position) to a more complex position that has been developed. Moving discourse from education to one on learning also moved the social workers off of centre stage, which made the minors themselves the principal actors within their discourse. [27]

This work of re-formulation also meant that these passages were not just explicit but also systemised within a wider discussion on the overall strategies adopted in education ("The passivity of the learner intrinsic in learning by models becomes active when they reason on the meaning of what is offered. In general, when they intersect with other models of learning. A single learning model/strategy is not used, but several models. You are constantly moving from one model to another"). [28]

During the following phase it was also possible to reconnect this reasoning on education to the interpretative repertories regarding the community itself. In fact, the discourse of the social workers referred to a vision of the community based on a cultural model which was defined as a "mission" (ORSENIGO, 1992): the social workers felt that the meaning of their work, besides the particular activities carried out while in service, consisted in "bearing witness" to the possibility of a different kind of society and in sharing the suffering of the minors hosted in the community; in this model, daily problems are placed in the background, and the organisation of the work is often approximate; the element on which is founded the possibility itself of working together is the affective closeness and the sharing of a common mission (With regard to these aspects, the participants stated, for example: "The kids live in a community. It is a state of passage, transition ... precariousness ... We too have a precarious job. Nobody can promise you that this will be a lifelong job. This brings us closer to them. We are all on the same side of the barricades." "It is important to strongly state what we feel is right, otherwise others will do it by affirming negative values." "The kids feel that the staff is really close-knit"). During the discussions this interpretative repertory (once again, we could say "dominant narration") turned out to be substantially irrefutable, and, therefore, "resistant" to reflection and re-thinking. [29]

The generation of stories turned out to be a central feature to carry out the research process: not just the stories told by the participants to describe and interpret their actions and the events, but also the stories produced by the researchers and the participants themselves to recount the research work. In this sense, by means of this (autobiographical) story intentionally told by the researchers (a visit to a residential community for adolescents based on a different cultural model, characterised by the presence of professional expertise, by a basically executive attitude and by an extremely reduced emotional involvement of the social workers with regard to their work, where the tasks are clearly defined and bounded) it was possible to point out the fact that during the discussion the identity of this community was defined by participants in terms of opposition with the other model, to explore the chance of discovering other possible models and, again, to construct narratives which refer to a more complex vision of the community itself ("In this community our work is based on the idea of living together because it is important to integrate diversity and to create a small open world. This community is a little world"). [30]

By means of the analysis of how these narratives were constructed it was also possible to consider this specific way of constructing knowledge—by means of divisions and oppositions. The hypothesis which emerged was that it represents an answer to the situation of uncertainty in which the staff had to work: dilemmas are resolved by means of a strategy of thought which represents them in terms of controversy and, which, therefore, generates extremes positions. The recognition of these processes allowed the participants to delineate (re-set) the problem in a different way (it is not a question of choosing one model or another, but of developing diversified strategies). [31]

Within the perspective adopted, it is clear that interaction constitutes the research method; thinking about interaction and about how this conditions the production of knowledge, becomes, therefore, essential to the method. Thinking about interaction means examining, for example, in what way the premises and the results of the research are made valid within these interactions and by what rules these processes are governed, and also in which way the evolution of interaction influences these processes. [32]

During this research, there clearly emerged the fact that, even though a collaborative research method was proposed between the researchers and participants, there prevailed at the beginning an asymmetrical and hierarchical relationship. The researchers, in fact, were perceived by the participants as those who brought knowledge and information different from their own, where "different" means "deeper" and "more systematic", both with regard to theory and methodology. The participants expected that the researchers would transfer this knowledge by once again referring to a model that was essentially linear of the learning processes. During the research, this relationship—and the rules that govern it—became itself the object of research and negotiation among the actors. In fact, the choice of including work areas and discussions lead autonomously by the participants—i.e., the meetings in which they prepared the document for the following one—permitted not only a way of encouraging this work of re-processing, but also the opening of symbolic spaces with regard to the researchers. It was, obviously an ad hoc solution; what seems interesting to us is that in the narrative which the participants constructed of these meetings there clearly emerged the fact that the researchers were, in any case, implicitly represented as interlocutors even in the absence of direct interaction and that often the new interpretations constructed by the social workers were aimed at proposing or supporting—and at times defending—their own version with regard to that of the researchers. The participants' discourse represented a construction oriented towards action since it supplied opposing and alternate versions to other points of reference in order to confute that way of reasoning or to defend their own from possible criticisms and contradictions. By means of the passages which we have described, the development of this research project became an instrument for negotiating their own positions—even internally—and for constructing a situation of dialogue. [33]

How are narrations used when carrying out research? The participants constantly used two different discursive devices (corresponding to paradigmatic and narrative thinking respectively, if we use the distinction of BRUNER, 1986). Their discourse on education and learning was constructed by a continuous shifting from one type to another. The hypothesis which emerged during the research process was that narratives represented the linking element between theory and practice. The analysis of these narratives constructed by the social workers during the discussion had as its aim to bring out interpretations and representations which were only partially systematised. The hypothesis is that the narrative level is mostly used to express knowledge which has not yet been formalised or systematised in a theory that can be stated. The analysis of narrative accounts also started a reflexive process which allowed the actors to construct together new interpretations with regard to the meaning of their actions. [34]

I wish to thank Giusi Incognito for her support and incisive feedback on earlier drafts.

1) In social research it is also possible to distinguish between the various approaches on the basis of the different levels of analysis and observations that they imply. On the one hand, autobiographical narrations are given a documentary value since it is assumed that they allow us to have access to an objective description of the events in questions; consequently, language is seen as an unambiguous instrument for transmitting information. On the other hand, their expression and interpretative value connected to the subjective definition of the situation and the constructive function of the language is emphasised. <back>

2) The origins of discursive psychology derive from the sociology of scientific knowledge (ASHMORE, 1989; GILBERT & MULKAY, 1984) and its applications in psychology (POTTER, 1987, 1988; POTTER & WETHERELL, 1987). The theoretical basis of the approach proposed by EDWARDS and POTTER can be traced back to the philosophy of language (AUSTIN, 1962; WITTGENSTEIN, 1953) and to the development of semiotics and post-structuralism (see ATKINSON, 1990). Within the social sciences the approach to the study of language which the authors adopted is derived from the theory of linguistic acts (GRICE, 1975; SEARLE, 1969), the ethnomethodology (GARFINKEL, 1967; HERITAGE, 1984) and the analysis of conversation (ATKINSON & HERITAGE, 1984; BUTTON & LEE, 1987; LEVINSON, 1983; SACKS, SCHEGLOFF & JEFFERSON, 1974). <back>

3) G. Francesco LANZARA (1993) talks of first-level and second-class inquiry. For first-level inquiry he means an exploration of the descriptions and interpretations constructed by the actors in relationship to the events and actions which has the aim of supplying an overall description or explanation of the characteristics of an object or a process. For second-level inquiry he means an exploration of the criteria, categories and procedures used by the actors to understand their own actions. This, in other words, is an "interpretation of interpretations". <back>

4) As Eric STEWARTS (2000, p.728) points out, "Our commitment to collaborative, empowering research methods should lead us to converse with, rather than count or survey, those people with whom we work, to aim for intersubjective, emic accounts of their lives and understandings and, to the extent possible, to amplify their voices and foreground their expertise (...) should lead us to write ourselves, our perspectives and positions, into our research; to portray ourselves as part of the process of representing others". <back>

5) The concept of "dominant narration" evidently also refers to the area of power. Julian RAPPAPORT (1998) claims that narrations supply information on identity, characteristics, the history of groups and that through dominant narrations it is possible to control the representation of reality and relationships within a group. Indeed, within the staff, the theme of values was supported mostly by the coordinator with the idea of constantly re-establishing his leadership and of maintaining control. His position, which he stated several times, was that not only the minors must learn the values which the community represents, but the social workers must share them. Below are some comments of the coordinator which reflect this position: "Doing this job does not mean that everyone must transmit his or her own personal values, but those established by the regulations of the community." "It is necessary to make an effort to put down in black and white the values in which we believe." "It is important to be clear about what I bring: I must accept the regulations and make them my own. I created the regulations and I have made them my own. The kids are always trying to note your difficulties (e.g., If you come late or make some mistake). What I bring must be irreproachable." <back>

Argyris, Chris & Schön Donald A. (1992). Theory in practice: Increasing professional effectiveness. San Francisco: Jossey-Bass.

Ashmore, Malcom (1989). The reflexive thesis: Wrighting sociology of scientific knowledge. Chicago: University of Chicago Press.

Atkinson, J. Maxwell & Heritage John C. (Eds.) (1984). Structure of social action: Studies in conversation analysis. Cambridge: Cambridge University Press.

Atkinson, Paul (1990). The ethnographic imagination: The textual construction of reality. London: Routledge.

Bruner, Jerome (1990). Acts of Meaning. Harvard University Press, Cambridge, Mass.

Bruner, Jerome (1991). La costruzione narrativa della realtà. In Massimo Ammaniti & Daniel N. Stern (Eds.), Rappresentazioni e narrazioni (pp.17-42). Bari: Laterza.

Burr, Vivien (1995). An introduction to social constructionism. London, New York: Routledge.

Button, Graham & Lee, John R.E. (1987). Talk and social organization. Clevedon: Multilingual Matters.

Cardona, Francesca (1992). Crescere, educare, curare, contenere: strutture residenziali per adolescenti in grave difficoltà in Inghilterra. In Cesare Kaneklin & Achille Orsenigo (Eds.) Il lavoro di comunità: Modalità di intervento con adolescenti in difficoltà (pp.113-128). Roma: NIS.

Denzin, Norman & Lincoln, Yvonne (1994). Handbook of qualitative research. London: Sage.

Drew, Paul (1984). Speakers' reporting in invitation sequences. In J. Maxwell Atkinson & John C. Heritage (Eds.) Structure of social action: Studies in conversation analysis (pp.129-151). Cambridge: Cambridge University Press.

Feldman, Carol F. (1991). I generi letterari come modelli mentali. In Massimo Ammaniti & Daniel N. Stern (Eds.), Rappresentazioni e narrazioni (pp.113-131). Bari: Laterza.

Garfinkel, Harold (1967). Studies in ethnomethodology. Englewood Cliffs, NJ: Prentice Hall.

Gilbert G. Nigel & Mulkay Michael (1984). Opening Pandora's Box: a Sociological Analysis of Scientists Discourse. Cambridge, UK: Cambridge University Press.

Grice, Paul H. (1975). Logic and conversation. In Peter Cole & James L. Morgan (Eds.), Syntax and Semantics, 3, Speech Acts (pp.41-58). New York: Academic Press.

Kaneklin, Cesare & Orsenigo, Achille (Eds.) (1992). Il lavoro di comunità: Modalità di intervento con adolescenti in difficoltà. Roma: NIS.

Manoukian, Franca (1992). Le comunità, organizzazioni riabilitative che producono tra dilemmi. In Cesare Kaneklin & Achille Orsenigo (Eds.) Il lavoro di comunità: Modalità di intervento con adolescenti in difficoltà (pp.25-42). Rome: NIS.

Mishler, Elliot G. (1986). The Analysis of Interview Narratives. In Theodore R. Sarbin (Ed.), Narrative psychology: The storied nature of human conduct (pp.233-255). New York: Praeger.

Orsenigo, Achille (1992). I costi emotivi del lavoro in comunità: l'incidenza delle caratteristiche organizzative. In Cesare Kaneklin & Achille Orsenigo (Eds.) Il lavoro di comunità: Modalità di intervento con adolescenti in difficoltà (pp.83-112). Roma: NIS.

Pomerantz, Anita M. (1984). Giving a source or basis: The practice in conversation of telling "how I know". Journal of Pragmatics, 8, 607-25.

Potter, Jonathan & Wetherell, Margaret (1987). Discourse and social psychology: Beyond attitude and behaviour. London: Sage.

Potter, Jonathan (1987). Reading repertoires: A preliminary study of some techniques that scientists use to construct readings. Science and Technology Studies, 5, 112-21.

Potter, Jonathan (1988). Cutting cakes: A study of psychologists' social categorizations. Philosophical Psychology, 1, 17-33.

Rappaport, Julian (1998). The Art of social change: Community narratives as resources for individual and collective identity. In Ximena B. Arriaga & Stuart Okamp (Eds.), Addressing community problems (pp.225-246). London: Sage.

Reason, Peter W. (2001). Handbook of action research: Participative inquiry and practice. Thousand Oaks, CA: Sage.

Sacks, Harvey; Schegloff, Emanuel A. & Jefferson, Gail (1974). A simplest systematics for the organization of turn-talking for conversation. Language, 50, 697-735.

Sarbin, Theodore R. (Ed.) (1986). Narrative psychology: The storied nature of human conduct. New York: Praeger.

Stewart, Eric (2000). Thinking through others: Qualitative research and Community Psychology. In Julian Rappaport & Edward Seidman (Eds.), Handbook of Community Psychology (pp.725-736). New York: Kluwer Academic/Plenum Publisher.

Torbert, William R. (1991) The Power of Balance. Transforming Self, Society and Scientific Inquiry. Newbury Park, California, Sage.

Wittgenstein, Ludwig (1953). Philosophische Untersuchungen. Oxford: Basil Blackwell.

Monica COLOMBO is a researcher in Social and Community Psychology at the Department of Psychology, University of Turin (Italy). Her main research interests are: discourse and narration, identity, qualitative research methods, action research and adult education

Colombo, Monica (2003). Reflexivity and Narratives in Action Research: A Discursive Approach [34 paragraphs]. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 4(2), Art. 9, http://nbn-resolving.de/urn:nbn:de:0114-fqs030291.

Forum Qualitative Sozialforschung / Forum: Qualitative Social Research (FQS)

ISSN 1438-5627

Creative Commons Attribution 4.0 International License