Abstract: The paper argues that while qualitative research has been flourishing in many fields of the social sciences, it has become unhelpfully fragmented and incoherent. Equally, there have developed a number of specialist domains of qualitative research that are too often treated in isolation. It is argued that we need to return to some fundamental principles of ethnographic inquiry that recognise the multiple modalities of social action and cultural representation, while locating them within a wider ethnographic framework. We need to recognise the intrinsic, indigenous principles of order and organisation that permeate social forms—discursive, visual, and material. Such formal ethnography provides a way of renewing classic ideas such as "grounded theory", "triangulation" and "thick description".

Key words: ethnography, thick description, triangulation, grounded theory, discourse, visual ethnography, material culture

There is no question that qualitative research of many varieties has flourished on a global scale over the past twenty years or so. From a personal perspective, I know that when Martyn HAMMERSLEY and I first made the Open University course on ethnography that subsequently led to our co-authored book (HAMMERSLEY & ATKINSON 1983) there was a very sparse literature from which to draw. There was a series of local oral traditions, but little by way of explicit methodological reflection. Since then, the position has changed out of all recognition. By the time of our second edition (HAMMERSLEY & ATKINSON 1995) the methodological literature had expanded exponentially. Since then the growth in qualitative methods has continued: methodological reflection and writing have flourished in recent years. Qualitative research, in a variety of forms, has been advocated and discussed in an ever-increasing number of publications. From its bases in such disciplines as anthropology and sociology, qualitative research has become prominent in many disciplinary contexts. Emergent disciplines such as cultural studies are thoroughly grounded in qualitative research, while it has penetrated very many substantive fields of research—such as educational research, organisational research and nursing studies. Cultural geography, discursive psychology, feminist scholarship and many other disciplinary fields have developed and contributed to distinctive strands in qualitative research. There are major networks and groups of qualitative research practitioners and methodologists in many national contexts. While the global character of academic publishing has meant that English-language research communities have dominated much of the discourse, and American work has been especially prominent, in fact there are strong and distinctive national as well as disciplinary traditions. The character and diversity of such work can be mapped most readily by inspecting the overall scope and details of contents of a series of major edited collections and handbooks (e.g. ATKINSON, COFFEY, DELAMONT, LOFLAND & LOFLAND 2001; DENZIN & LINCOLN 1994, 2000; SEALE, GOBO, GUBRIUM & SILVERMAN 2004; GUBRIUM & HOLSTEIN 2002). [1]

Ironically, despite classic ethnographic appeals to holism, context and similar ideas, qualitative, ethnographic research seems to have become increasingly fragmented. As the methodological literature has expanded, it has also diversified. Different authors adopt and promote specific approaches to the collection and analysis of data. Equally, particular kinds of data become celebrated in the process: personal narratives, life-histories and other documents of life; film, video and photographic images; texts and documentary sources; material culture and technological artefacts; spoken discourse. In the process types of data and corresponding types of analysis are elevated to occupy a special status. The implication often seems to be that, say, documents of life provide especially privileged insights, or that visual materials are especially significant, or that talk is the form of social action par excellence. Consequently, types of data and their associated analytic strategies are promoted as the single preferred method for social inquiry, rather than strategies within a broader ethnographic approach. Indeed, the enthusiasm shown for particular methods of data collection and analysis sometimes seems odd. There seems in principle little or no reason for social scientists to develop their research programmes on the basis of one technique or one research strategy exclusively. It seems equally perverse implicitly to regard methods as being in competition. [2]

I have no quarrel with attempts to define and practise appropriate strategies for the analysis of particular kinds of data. Indeed, I want to insist on the proper, disciplined approach to any and every type of data. Equally, I want to insist that data should be analysed, and not just reproduced and celebrated (as sometimes happens with life-histories, and some visual materials). My main message, however, is that the forms of data and analysis should reflect the forms of culture and of social action. In other words, we collect and analyse personal narratives and life-histories because they are a collection of types or forms—spoken and written—through which various kinds of social activity are accomplished. They are themselves forms of social action in which identities, biographies, and various other kinds of work get done. One accords importance to narratives and narrative analysis because they are important kinds of social action. In the same spirit we pay serious attention to visual data insofar as culture and action have significant visual aspects that cannot be expressed and analysed except by reference to visual materials. This is by no means equivalent to the assumption that ethnographic film or video constitutes an especially privileged approach to sociological or anthropological understanding. The same can be said of other analytic approaches: documentary analysis is significant insofar as a given social setting is self-documenting, and in which important social actions are performed. Texts deserve attention because of their socially organised and conventional properties, and because of the uses that they are put to, in their production, circulation and consumption. The same is true of other material goods, artefacts, technologies etc. The analysis of dramaturgy, likewise, is important insofar as social actors and collectivities engage in significant performative activities—but it should not be treated as a privileged way to approach all of social life. [3]

I believe, therefore, that it is important to avoid the essentially reductionist view that treats one type of data or one approach to analysis as being the prime source of social and cultural interpretation. We should not, in other words, seek to render social life in terms of just one analytic strategy or just one cultural form. The forms of analysis should reflect the forms of social life: their diversity should mirror the diversity of cultural forms; their significance should be in accordance with their social and cultural functions. This may seem obvious. But while few social scientists would explicitly claim otherwise, implicitly in much current writing and discussion, the reverse seems to be true. [4]

In reviewing an array of different analytic approaches I do not merely celebrate diversity; nor do I endorse a vulgar version of triangulation through methodological pluralism and synthesis (cf. COFFEY & ATKINSON 1996). The reverse is true: I stress the importance of rendering the different formal properties of culture and social action and preserving their distinctive qualities. I want, therefore, to affirm that aspects of culture and the mundane organisation of social life have their intrinsic formal properties, and that the analysis of social life should respect and explore those forms. In doing so, I am reacting against some analytic tendencies that have under-valued anything that smacks of formal analysis. Major commentators like DENZIN and LINCOLN (2000), or ELLIS and BOCHNER (1996) have promoted an image of contemporary qualitative research that is relentlessly innovative, allied to postmodernist views of social inquiry, and radically distant from its intellectual origins. As my colleagues and I have suggested elsewhere (e.g. DELAMONT & ATKINSON 2004; ATKINSON, COFFEY & DELAMONT 2003), appeals to postmodernism have, in many influential quarters, de-valued the systematic analysis of action and representations, while privileging rather vague ideas of experience, evocation, and personal engagement. Yet discourse, narratives, performances, encounters, rhetoric and poetics all have their intrinsic, indigenous modes of organisation. So too do visual, textual, material and other cultural embodiments. It is not necessary to endorse a narrowly structuralist analytic perspective or endorse unduly restrictive formalisms in order to recognise the formal properties of talk, the codes of cultural representation, the semiotic structures of visual materials, or the common properties of narratives and documents of life. [5]

The current state of qualitative research and research methods is confused. There is a gratifying proliferation of research methods, and they have been spreading to a wide range of substantive research areas. Equally, there has been a variety of rationales, justifications and theoretical underpinnings for qualitative research. Researchers have become increasingly wedded to particular methods of data collection and strategies of data analysis. So we have people who are restrictedly expert in, say, visual methods, or the methods of visual anthropology. Likewise, there are researchers who are wedded to particular ways of reporting social research—through forms of poetic writing, or through multi-layered texts, or realist styles. There are now several contexts in which "alternative" forms of representation, such as "auto-ethnographic" reflections, poems and other genres of creative writing (for examples see: ELLIS & BOCHNER 1996; GOODALL 2000; for discussions see ATKINSON & COFFEY 1995; SPENCER 2001). The journal Qualitative Inquiry, edited by Norman DENZIN, is one key site for the publication of such innovative materials. [6]

Taken overall, the field of qualitative research presents a confusing picture. The manifest variety is not always related systematically or in a principled fashion to any particular disciplinary, theoretical or substantive concern. Some of the current methodological positions seems to advocate strategies of research without reference to the indigenous modes of social organisation they are designed to address. Consequently, it is necessary for social researchers to have an understanding of a variety of research methods, in order to do justice to the equivalent variety of cultural forms. The paper will therefore outline and explore some particularly important modes of social analysis, in order to examine how they construct and reflect specific cultural forms. I am not unmindful of our international, comparative theme. The tendencies I refer to are not uniformly distributed across different national and disciplinary contexts. It is clear that the so-called postmodern turn has been especially marked in American social science, where different forms of experimental and innovative social inquiry have developed, not least in the context of communication studies and cultural studies. They have certainly had an effect in the United Kingdom and continental Europe, but to a much lesser extent. While American cultural anthropology has been pervasively influenced by the so-called "crisis of representation" precipitated by the literary and discursive turn in anthropology (cf. CLIFFORD & MARCUS 1986), British social anthropology—while by no means immutable—has been far less deflected from its prior practices by such fundamental critiques. European scholars are to some extent protected from undue influence from American "post" enthusiasms. They have their own indigenous intellectual traditions. There are, for instance, strong traditions of French discourse analysis and German hermeneutics that can provide a durable matrix for the reception of Anglo-American ideas. Indeed, I want to suggest that the disciplinary and national traditions of European research can provide a strong basis for a renewed synthesis of qualitative research. In the following sections of this paper I do not attempt to undertake a comprehensive overview of qualitative research strategies. In a selective review I highlight some of the ways in which methodological particularism can lead to weak research, and how a generic methodological attention to the indigenous organisation constitutes forms of culture. [7]

Interviews are increasingly the main mode of qualitative data collection, and biographical narratives are among the most important products of such research. The centrality of the interview is evidenced by the sheer volume of "qualitative research" that is based exclusively or predominantly on extended interviews. The methodological literature on the conduct and analysis of interviews has grown correspondingly (e.g. GUBRIUM & HOLSTEIN 2002; HOLSTEIN & GUBRIUM 1995; RUBIN & RUBIN 1995). The status of the interview as a mode of data collection and the proper analysis of the data generated through qualitative interviewing are, however, problematic. As ATKINSON and SILVERMAN (1997) have argued, social scientists who extol the virtues of personal interviewing, and who base their research exclusively on such data are in danger of recapitulating one of the key features of contemporary society, rather than examining and analysing it. They are, ATKINSON and SILVERMAN argued, complicit in the forms of "the interview society", through which accounts of "experience" and the revelation of a "private" emotional life are expected of potentially any societal member, and actually expected of anyone accorded the status of "celebrity". The fact that interviews and their outcomes are pervasive in certain fields of qualitative research does not of itself guarantee their value. Some researchers and methodologists promote interviews and the accounts garnered from them as enjoying an almost unique status. The popularity of interviewing means that the sorts of individual accounts and narratives that interviews produce are sometimes promoted as conveying a special significance. Moreover, the widespread misuse of interview-derived narratives means that the data are too often treated at "face value", as if personal accounts granted the analyst direct access to a realm of the personal that is not available through other means. This is related to the equally widespread view that it is the goal of qualitative research to represent the personal meanings, experiences and perspectives of individual informants. The consequence can be a version of social inquiry that is devoid of social organisation, in which categories such as "experience" are treated unproblematically. [8]

Such approaches to "narrative" pay insufficient attention to the work of analysts, going back several decades, that treat informants' accounts as accounts, that are performances through which informants enact biographical, self-presentational and explanatory work. This is the analytic perspective promoted by VOYSEY (1975) in her analysis of accounts produced by parents of children with a disability, and by GILBERT and MULKAY (1980) in their analysis of scientists' accounts of scientific discoveries. Each of these analyses, in turn reflecting back the pioneering observations of C. WRIGHT MILLS (1940) on vocabularies of motive, recognising the nature of narrative accounts as forms of speech-act. [9]

We should not collect and document personal narratives because we believe them to have a privileged or special quality. Narrative is not a unique mode of organising or reporting experience, although it is one pervasive and important way of so doing. Narrative is an important genre of spoken action and representation in everyday life, and in many specialised contexts. We should, therefore, be studying narrative insofar as it is a particular feature of a given cultural milieu. Furthermore, narratives are not independent of cultural conventions and shared formats. They are not uniquely biographical or autobiographical materials, and they certainly do not convey unmediated private "experience". This is a perspective powerfully demonstrated by PLUMMER (1995), who demonstrates that even the most "personal" stories display generic properties that reflect collective, shared cultural conventions. Likewise, narratives do not convey "memory" as a psychological phenomenon. Experiences, memories, emotions and other apparently personal or private states are constructed and enacted through culturally shared narrative types, formats, and genres. They are related to story types more generally. There are affinities with other kinds of stories—of history, mythology, the mass media and so on. [10]

We need, therefore, to analyse narratives and life-materials, in order to treat them as instances of social action—as speech-acts or events with common properties, recurrent structures, cultural conventions and recognisable genres. This is, of course, by no means a novel observation in its own right. Many commentators have drawn attention to the formal, structural properties of narrative. Key sources include CORTAZZI (1993), LABOV and FANSHEL (1977), MISHLER (1986) and RIESSMAN (1993). While it would, therefore, be entirely wrong to accuse all proponents of "narrative" analysis as lacking rigour in their own approaches, it remains the case that too many contemporary researchers take their spirit of enthusiasm without their sense of form and structure. Consequently, there is too much social research that collects, reproduces and celebrates individual "stories", without grounding them in a sustained analysis of their forms and functions. [11]

Moreover, we need to regard such accounts as social performances, or forms of social action, embedded within organisational contexts, and socially shared undertakings. Too much of the contemporary deployment of narrative is devoid of social organisation and context. Indeed, it is often not clear in what sense some forms of narrative celebration are social science at all. Lives and narrative voices seem to be recounted in a social vacuum, rather than the products of socially shared conventions, constructed in practical circumstances of everyday life and work, with real consequences for social actors (cf. ATKINSON 1997). [12]

This is not, then, an argument "against" narratives. Indeed, it is not clear that one could in any meaningful sense be for or against any particular form of social activity. But my point here is precisely that: narratives and personal accounts are among a variety of spoken and written social actions. They are, therefore, inescapably part of the subject-matter of ethnographic research. We encounter various forms of story in a multiplicity of social settings, from the domestic settings of family life, through to the everyday settings of work, to the highly specialist settings of science and professional expertise (cf. e.g. CZARNIAWSKA 1997, 1998). We should obviously pay serious attention to the distinctive features of narratives, and how they are used to achieve practical outcomes: we should study form and function, in other words. But we should do so in recognition of the fact that narratives are but one example of structured performance through which everyday life is enacted. [13]

Parallel remarks can be made concerning other forms of spoken activity, although the potential criticism here is rather different from the one I have just sketched in relation to narrative. Here we find a clear danger of over-technical attention to detailed, formal properties with insufficient attention to wider ethnographic interests. The collection and analysis of spoken materials is one domain where over-specialisation is a danger, then. [14]

There is little need in this context to expand upon the very great impact of conversation analysis and discourse analysis. For major sources and discussions, see: ATKINSON and HERITAGE (1984); BODEN and ZIMMERMAN (1991); GOODWIN (1981); JAWORSKI and COUPLAND (1999); POTTER (1996); POTTER and WETHERELL (1987). Conversation analysis has had implications well beyond the confines of its highly specialised research networks, and its early associations with ethnomethodology. The analysis of naturally occurring language and spoken action has become a taken-for-granted feature of social research in multiple contexts. There is, however, a clear danger of treating language analysis as being a self-contained and self-justifying activity. We should not allow speech acts and the organisation of discourse to occupy a self-contained, separate domain of social analysis. We need, by contrast, to ensure that the analysis of spoken language remains firmly embedded in studies of organisational context, processes of socialisation, routines of work, personal transformation, people-processing and so on. Spoken language has its own intrinsic forms of organisation. Indeed, it demonstrates a densely structured organisation at every level, including the most finely grained. It is important, however, that discourse analysis, conversation analysis, discursive psychology and the like are not treated as analytic ends in their own right, and are not intellectually divorced from other aspects of ethnographic inquiry. The expert knowledge required should not be regarded as a specialism in its own right and independent of wider sociological or anthropological competence. The conventions of language use need to be analysed, therefore, in relation to more general issues of identity, the interaction order, moral work and the organisation of social encounters. I do not mean to imply that such applications are entirely missing. On the contrary, there many examples of conversation, discourse or similar analytic attention to spoken activity embedded within more general ethnographic inquiry (e.g. ATKINSON & DREW 1979; MAYNARD 2003; PERÄKYLÄ 1995; SILVERMAN 1997). But I do want to draw attention to the fact that many of even these exemplars pay almost exclusive attention to the organisation of talk, and rest almost exclusively on the analysis of transcribed materials. They therefore demonstrate a single-minded reliance on just one mode of social organisation and one analytic strategy. We also need to remind ourselves that the original inspiration for conversation analysis lay in Harvey SACKS's use of transcribed conversation as an objet trouvé, demonstrating the properties of organisation and order. But they were not intended to occupy a uniquely privileged place in the sociological analysis of pervasive orderliness (SILVERMAN 1998). The example of discourse analysis and conversation analysis demonstrates the recurrent need to pay close attention to the formal properties of social action. [15]

The collection and analysis of visual materials tend, unhelpfully, to be treated as a specialist domain. The production of ethnographic film has, of course, a long history, though it has often been oddly divorced from the mainstream textual practices of the ethnographic monograph. The use of photography for ethnographic purposes has also been relegated to a somewhat specialist sub-field (where it has not been relegated to mere illustration of the written monograph). Specialist commentaries on visual methods have helped to make them especially prominent in recent years (e.g. BALL & SMITH 2001; BANKS 2001; EMMISON & SMITH 2000; LOIZOS 1992; PINK 2000; ROSE 2001; RUBY 2000). In recent years the development of small digital camcorders and the development of digital photography have created an enormous range of possibilities for ethnographers in the field. Two things follow. The first is that "visual" anthropology and sociology should not be treated as separate genres or specialisms. There are many aspects of culture that are intrinsically visual. Many cultural domains and artefacts can only be grasped through their visual representations and the structured properties of their visual codes. There are many social phenomena that can and should be analysed in terms of their appearances and performances that may be captured in visual terms. These are not, however, separable from the social settings in which such phenomena are generated and interpreted. They should not be explored purely as "visual" topics, but as integral to a wide variety of ethnographic projects. Visual phenomena, the mundane and the self-consciously aesthetic, have their intrinsic modes of organisation. One does not need to endorse the most deterministic versions of semiotics or structuralism to recognise that visual culture embodies conventions and codes or representation. There are culturally organised aesthetic and formal principles; there are conventional forms of representation and expression (MANNING 1989, 2001, 2004). Attention to visual culture also implies a serious attention to the ethno-aesthetics of the producers and consumers of visual materials. We need not only to "read" the visual, but also understand ethnographically how it is read by members of the social world or culture in question. [16]

The study of material goods and artefacts, technology and other physical aspects of material culture deserves systematic attention in many ethnographic contexts, and it is receiving increased methodological attention (TILLEY 1990, 1991, 2001); but is too often relegated to specialised, esoteric studies or to highly specific topics. The latter include studies of technology and inventions, of very particular kinds of physical display—such as museums and art galleries—and highly restricted kinds of artefacts such as religious, ritual and artistic objects. It is vital that the study of physical objects, memorials, technologies and so on be thoroughly incorporated into more general field studies of work organisations, informal settings, cultural production, domestic settings and so on (MILLER 1987; JULIER 2000). Artefacts and technologies are themselves understood, used and interpreted by everyday social actors. They are used to document and record the past—and indeed to construct the past—and there is much to be learned from the local, situated "ethno-archaeology" of the material past (EDWARDS 2001; MACDONALD 2002; MACDONALD & FYFE 1996). Issues of practical utility and aesthetic value intersect. Ideas of authenticity may be brought to bear on artefacts and assemblages. They may be used to display and warrant individual and collective identities: the "collection" (personal or national), for instance, is expressive of taste, identity, commitment and enthusiasm. The material goods of fashion and conspicuous consumption are likewise expressive of status and aspirations. More generally, this leads to a consideration of material culture. The material embodiment of culture and the cultural connotations of things have become prominent in recent cultural-anthropological analyses. Recent examples have included examinations of: home computers, personal stereos, motor cars, photographs, food, memorials. These accounts transcend and transform the mundane material world into domains of signification. It is important for the general ethnographic enterprise to incorporate such perspectives. Equally, it is important to recognise that material goods and artefacts can have their own, indigenous orders of signification and genres of representation. [17]

The material order is also encoded in systems of places and spaces. Most ethnographic reportage seems oddly lacking in physical location. Many sociological and anthropological accounts have sketchy accounts of the built environment within which social events and encounters take place. The treatment of space is too often restricted to aspects of human geography, urban studies, and architecture. It needs to be integrated within more general ethnographic accounts. But ethno-architecture is—as we know from some anthropological accounts—significant in defining the spaces and styles of everyday living. Built spaces provide symbolic as well as physical boundaries. They physically enshrine collective memories as well as more personal biographical and emotional work. Homes are endowed with emotional and cultural value—through the expression of taste and cultural capital, the celebration of historical authenticity, or of modern minimalism. Public spaces also embody tacit cultural assumptions: about the classification and processing of people and things; about commercial and professional transactions; about political processes and citizenship. The ethnographic exploration of places and spaces includes the commercial transformation of them through tourism and heritage work: the transmutation of down-towns and waterfronts; the re-creation of industrial pasts into leisure and entertainment; the construction of replicas and spaces for "experience" (DICKS 2000; LASH & URRY 1994; URRY 1990, 1995). [18]

It is necessary for ethnographers to pay attention to the analytic imperatives of such socially shared codes, conventions and structures. The forms of social and cultural life call for equivalent analyses. I have outlined just some analytic considerations that can and should inform general qualitative, ethnographic inquiry, and should not be restricted to specialised strategies. To them could be added yet more: sound, noise and music provide important temporal and aesthetic components to everyday life, for instance (DeNORA 2000), while senses of smell, taste and touch inform our understanding and recollection of everyday life (cf. STOLLER 1989). I have been at pains to point out that we need to do two things. First, we need to retain a structural, formal sense of the multiple orderings of talk, action, things, places and so on. Secondly, we cannot afford to allow such analysis to become the preserve of small coteries of specialists, while a generalised "qualitative research" proceeds uninformed by such formal analysis. [19]

These methodological principles give us a way of addressing some fundamental methodological precepts in a disciplined way. Herbert BLUMER enunciated the principle that research should be "faithful" to the phenomena under investigation (BLUMER 1954; HAMMERSLEY 1989). In its most general form this methodological precept seems to beg all the important questions, seeming to imply that one can know the phenomena prior to their investigation. A naively naturalist interpretation is clearly inappropriate. My formulation retrieves for BLUMER's principle a more methodologically precise formulation—a more restricted one, but a more fruitful approach. It implies merely that fidelity to "the phenomena" means paying attention to the forms and the media through which phenomena are enacted, encoded or embodied. It means preserving and respecting the different layers of action and representation through which cultures are enacted and social action is performed. [20]

It also gives a particular rendering of the notion of thick description (GEERTZ 1973). Clifford GEERTZ's formulation of that term is susceptible to multiple interpretations and can be translated into various research practices. Some vulgar simplifications of it refer—trivially and erroneously—merely to the richness of detail and concreteness of cultural descriptions on which ethnographic work rests. More sophisticated versions refer to the over-determined character of culture, with multiple frames of reference and perspective. My own gloss is to suggest that whatever else "thick description" could mean, it should include systematic reference to the multiple forms of cultural life, producing cultural descriptions that preserve those distinctive forms. It thus takes GEERTZ's "textual" approach to cultural analysis seriously, by insisting that the "texts" need to be analysed in terms of their material and conventional properties. It also transforms the emphasis on "culture" into an equal stress on social action. [21]

This approach can be extended to a commentary on versions of grounded theory (GLASER & STRAUSS 1967). Again, there are multiple versions of grounded theory, and they have been thoroughly documented. It is noticeable, however, that most of them are more articulate on what being "grounded" means than on the proper analysis of different types of data. In some quarters, therefore, analysis seems to consist of glorified content analysis, translated into different kinds of thematic and theoretical "coding". While grounded theory is clearly not intended to be restricted to any one category of data—and is not even restricted to qualitative research—in practice it used to describe a somewhat amorphous notion of qualitative data, usually field notes and interview transcripts. There is normally little attempt to preserve the narrative structures or other forms of representation. At their worst, vulgar versions of "grounded theory" can result in a kind of analytic blender, generating blandly homogenised categories and instances. We believe that in one sense all productive sociological and anthropological analysis is "grounded": it depends on processes of abductive reasoning in the creative interplay between data and ideas, concrete instances and generic concepts. In a more specific sense, we believe that analyses should be "grounded" in the multiple forms and representations of social life, and should remain sensitive to those forms. In that sense, therefore, grounded theory would be grounded in the multiple layers of codes, conventions, structures and texts of everyday social life. It would preserve their distinctive character and their orderings, not wash out their intrinsic properties. [22]

What is needed is a radical renewal of our sensitivity to forms and modes of organisation that interactionist and interpretative sociology has in principle been addressing for the past eighty years and more (ATKINSON & HOUSLEY 2003). We can retrieve some sense of that analytic tradition by connecting it with contemporary notions of complexity. Contemporary complexity theory provides a powerful set of analytic metaphors for comprehending the emergent properties of social phenomena and their diverse levels of order and meaning. It recalls classic interactionist and interpretative ideas of social emergence and the processes of social life. A contemporary ethnography that is sensitive to the indigenous orders of action, meaning and representation can provide a reflexive and complex vehicle for exploring social organisation and the fluidity of late modernity. Indeed, I want to insist that we continue to need formal methods and formal analyses precisely because they allow us to grasp the complex orders of representation, action, organisation and meaning that constitute contemporary social life. [23]

Whatever the epistemological validity of theories of postmodernism in general, it seems to me that the interpretation of postmodernism in the current methodological literature is in many ways unhelpful, even pernicious. Too many advocates of postmodern qualitative research, and its equivalents, repeatedly rob social life—and hence its investigation—of any sense of order. It is clearly not necessary to espouse anything resembling a positivist intellectual stance in order to recognise that social life has its principles of order, and that those orders can be examined in principled ways. Methodological approaches that wash out those indigenous orders of action and representation empty the social world of many of its most significant (and signifying) phenomena. It would be a pity if the very obvious success of qualitative research in many fields were to result in a weakened version of social science by failing to pay due attention to its disciplinary roots, and its disciplined attention to social forms. [24]

This is intellectual work that can and should be undertaken with especial vigour in Europe. In no spirit of general anti-American feeling, I note that the global production of qualitative methods is dominated by American perspectives. Now the American traditions have undoubtedly provided many of the main foundations for qualitative research. But there are also profound weaknesses in the general climate of American methodology at the present time. Too often, for instance, the social and the political are translated into the personal. Likewise, the restless search for new paradigms and innovation has resulted in a proliferation of methodological pronouncements and prescriptions, that often break free of any disciplinary basis. Qualitative research in such a context is too often treated as a self-justifying activity, rather than amounting to a general approach to doing systematic social science, and addressing intrinsically significant research topics. In Europe, it is possible to develop collective understandings of qualitative research that can re-assert some of the shared strengths of our social-science traditions and disciplines, drawing on the best of Anglo-American work, but also avoiding some of the self-indulgent and ill-disciplined work that too often spoils it. [25]

Atkinson, J. Maxwell & Drew, Paul (1979). Order in Court: The Organization of Verbal Expression in Judicial Settings. London: Macmillan.

Atkinson, J. Maxwell & Heritage, John (Eds.) (1984). Structures of Social Action: Studies in Conversation Analysis. Cambridge: Cambridge University Press.

Atkinson, Paul (1997). Narrative turn or blind alley? Qualitative Health Research, 7, 325-344.

Atkinson, Paul & Coffey Amanda (1995). Realism and its discontents: the crisis of cultural representation in ethnographic texts. In Barbara Adam & Stuart Allen (Eds.), Theorising Culture (pp.103-39). London: UCL Press.

Atkinson, Paul & Silverman, David (1997). Kundera's Immortality: The interview society and the invention of the self. Qualitative Inquiry, 3, 304-325.

Atkinson, Paul, Coffey, Amanda & Delamont, Sara (2003). Key Themes in Qualitative Research. Walnut Creek CA: AltaMira.

Atkinson, Paul, Coffey, Amanda, Delamont, Sara, Lofland, John & Lofland, Lyn (Eds.) (2001). Handbook of Ethnography. London: Sage.

Ball, Mike & Smith, Greg (2001). Technologies of realism? Ethnographic uses of photography and film. In Paul Atkinson, Amanda Coffey, Sara Delamont, John Lofland & Lyn Lofland (Eds.), Handbook of Ethnography (pp.302-219). London: Sage.

Blumer, Herbert (1954). What is wrong with social theory? American Sociological Review, 19, 3-10.

Boden, Deirdre & Zimmerman, Don (Eds.) (1991). Talk and Social Structure. Cambridge: Polity.

Coffey, Amanda & Atkinson, Paul (1996). Making Sense of Qualitative Data: Complementary Strategies. London: Sage.

Czarniawska, Barbara (1997). Narrating the Organization: Dramas of Institutional Identity. Chicago: University of Chicago Press.

Czarniawska, Barbara (1998). A Narrative Approach to Organization Studies. Thousand Oaks CA: Sage.

Delamont, Sara & Atkinson, Paul (2004). Qualitative research and the postmodern turn. In Melissa A. Hardy & Alan Bryman (Eds.), Handbook of Data Analysis (pp.667-681). London: Sage.

DeNora, Tia (2000). Music in Everyday Life. Cambridge: Cambridge University Press.

Denzin, Norman K. & Lincoln, Yvonna S. (Eds.) (1994). Handbook of Qualitative Research. Thousand Oaks CA: Sage.

Denzin, Norman K. & Lincoln, Yvonna S. (Eds.) (2005). Handbook of Qualitative Research, 2nd edition. Thousand Oaks CA: Sage.

Dicks, Bella (2000). Heritage, Place and Community. Cardiff: University of Wales Press.

Edwards, Elizabeth (2001). Raw Histories: Photographs, Anthropology and Museums. Oxford: Berg.

Ellis, Carolyn & Bochner, Arthur (Eds.) (1996). Composing Ethnography. Walnut Creek CA: AltaMira.

Gilbert, Nigel & Mulkay, Michael (1980). Opening Pandora's Box: A Sociological Account of Scientists' Discourse. Cambridge: Cambridge University Press.

Glaser, Barney & Strauss, Anselm L. (1967). The Discovery of Grounded Theory. Chicago: Aldine.

Goodall, H.Lloyd (2000). Writing the New Ethnography. Walnut Creek CA: AltaMira.

Goodwin, Charles (1981). Conversational Organization: Interaction Between Speakers and Hearers. New York: Academic Press.

Gubrium, Jaber F. & Holstein, James A. (Eds.) (2002). Handbook of Interview Research: Context and Method. Thousand Oaks CA: Sage.

Hammersley, Martyn (1989). The Dilemma of Qualitative Method: Herbert Blumer and the Chicago Tradition. London: Routledge.

Hammersley, Martyn & Atkinson, Paul (1983). Ethnography: Principles in Practice. London: Tavistock.

Hammersley, Martyn & Atkinson, Paul (1995). Ethnography: Principles in Practice, 2nd edition. London: Routledge.

Holstein, James A. & Gubrium, Jaber F. (1995). The Active Interview. Thousand Oaks CA: Sage.

Jaworski, Adam & Coupland, Nikolaus (Eds.) (1999). The Discourse Reader. London: Routledge.

Labov, William & Fanshel, David (1977). Therapeutic Discourse: Psychotherapy as Conversation. New York: Academic Press.

Loizos, Peter (1992). Innovation in Ethnographic Film. Manchester: Manchester University Press.

Macdonald, Sharon (2002). Behind the Scenes at the Science Museum. Oxford: Berg.

Macdonald, Sharon & Fyfe, Gordon (Eds.) (1996). Theorising Museums. Oxford: Blackwell.

Manning, Peter (2001). Semiotics, semantics and ethnography. In Paul Atkinson, Amanda Coffey, Sara Delamont, John Lofland & Lyn Lofland (Eds.), Handbook of Ethnography (pp.145-159), London: Sage.

Manning, Peter (2004). Semiotics. In Melissa Hardy & Alan Bryman (Eds.), Handbook of Data Analysis (pp.567-587). London: Sage.

Maynard, Douglas W. (2003). Good News, Bad News: Conversational Order in Everyday Talk and Clinical Settings. Chicago: University of Chicago Press.

Miller, Daniel (1987). Material Culture and Mass Communication. Oxford: Basil Blackwell.

Mills, C. Wright (1940). Situated actions and vocabularies of motive. American Sociological Review, 5, 439-52.

Mishler, Elliot (1986). Research Interviewing: Context and Narrative. Cambridge MA: Harvard University Press.

Peräkylä, Anssi (1995). AIDS Counselling: Institutional Interaction and Clinical Practice. Cambridge: Cambridge University Press.

Pink, Sarah (2001). Doing Visual Ethnography: Images, Media and Representation in Research. London: Sage.

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

Potter, Jonathon & Wetherell, Margaret (1987). Discourse and Social Psychology: Beyond Attitudes and Behaviour. London: Sage.

Rubin, Herbert J. & Rubin, Irene S. (1995). Qualitative Interviewing: The Art of Hearing Data. Thousand Oaks CA: Sage.

Ruby, Jay (2000). Picturing Culture: Explorations of Film and Anthropology. Chicago: University of Chicago Press.

Seale, Clive, Gobo, Giampietro, Gubrium, Jaber F. & Silverman, David (Eds.) (2004). Qualitative Research Practice. London: Sage.

Silverman, David (1997). Discourses of Counselling: HIV Counselling as Social Interaction. London: Sage.

Silverman, David (1998). Harvey Sacks: Social Science and Conversation Analysis. Oxford: Polity.

Spencer, Jonathon (2001). Ethnography after post-modernism. In Paul Atkinson, Amanda Coffey, Sara Delamont, John Lofland & Lyn Lofland (Eds.), Handbook of Ethnography (pp.443-452). London: Sage.

Stoller, Paul (1989). The Taste of Ethnographic Things. Philadelphia: University of Pennsylvania Press.

Tilley, Michael (Ed.) (1990). Reading Material Culture: Structuralism, Hermeneutics and Post-Structuralism. Oxford: Blackwell.

Tilley, Michael (1991). Material Culture and Text: The Art of Ambiguity. London: Routledge.

Tilley, Michael (2001). Ethnography and material culture. In Paul Atkinson, Amanda Coffey, Sara Delamont, John Lofland & Lyn Lofland (Eds.), Handbook of Ethnography (pp.258-272). London: Sage.

Urry, John (1990). The Tourist Gaze: Leisure and Travel in Contemporary Societies. London: Sage.

Voysey, Margaret (1975). A Constant Burden: The Reconstitution of Family Life. London: Routledge and Kegan Paul.

Paul ATKINSON is Distinguished Research Professor of Sociology at Cardiff University, and is Associate Director of the ESRC Centre for Economic and Social Aspects of Genomics (CESAGen). He is co-editor of the Sage journal Qualitative Research and was co-editor of the Handbook of Ethnography (Sage). His books include: Everyday Arias (AltaMira); Medical Talk and Medical Work (Sage); The Ethnographic Imagination (Routledge); making Sense of Ethnographic Texts (Sage); Martyn HAMMERSLEY and Paul ATKINSON, Ethnography: Principles in Practice (Routledge); Amanda COFFEY and Paul ATKINSON Making Sense of Qualitative Data (Sage); Sara DELAMONT and Paul ATKINSON, Fighting Familiarity (Hampton); Paul ATKINSON, Amanda COFFEY and Sara DELAMONT, Key Themes in Qualitative Research (AltaMira), Paul ATKINSON and William HOUSLEY, Interactionism (Sage).

Cardiff School of Social Sciences
Glamorgan Building
King Edward VII Avenue
Cardiff CF10 3WT
Wales, United Kingdom

Atkinson, Paul (2005). Qualitative Research—Unity and Diversity [25 paragraphs]. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 6(3), Art. 26, http://nbn-resolving.de/urn:nbn:de:0114-fqs0503261.

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

ISSN 1438-5627

Creative Commons Attribution 4.0 International License