Volume 5, No. 1, Art. 30 – January 2004

Paradigm Wars Around Interview Methodologies: Constructionism and Postmodernism "on tap" or "on top"?

Tom Wengraf

Review Essay:

James Holstein & Jaber Gubrium (Eds.) (2003). Inside Interviewing: New Lenses, New Concerns. Thousand Oaks: Sage, 557 pages, ISBN 0-7619-2851-0, £29

Jaber Gubrium & James Holstein (Eds.) (2003). Postmodern Interviewing. Thousand Oaks: Sage, 287 pages, ISBN 0-7619-2850-2, £29

Abstract: Two overlapping volumes extracted from the Handbook of interview research are seen as providing a considerable variety and depth of useful technical advice for qualitative researchers and a plethora of ideological warfare and confusion which helps nobody. Chapters on a range of types of more or less reluctant respondents and on phone and computer-aided interviewing (on and off the Web) are followed by chapters on transcription and on computer-aided (or not) qualitative analysis and interpretation from a variety of positions. Questions of representation are addressed. The review argues that post-modernist and constructivist practice can be a useful tactic in a struggle for a more reflexive and subtle researched understanding, but (as a philosophy competing for dominance) they make up a counter-productive "American ideology" which is self-contradictory, and personally and politically unhelpful. "On tap", yes; "on top", no!

Key words: constructionism, postmodernism, realism, ethnomethodology, paradigm, ideology, reflexivity, scientificity, Bachelard, interview, ethnography, craft, interviewism

Table of Contents

1. On Addressing the Volumes

2. Content-Focused Review

3. General Points






1. On Addressing the Volumes

The first thing that the reader should note about these two volumes is that they are both extracted from a larger earlier volume: GUBRIUM and HOLSTEIN's Handbook of interview research (Sage 2001). So if you have the bulky but even more comprehensive original volume, don't buy either of these, especially as it costs the same amount for roughly half the pages! [1]

The second thing to note about the two volumes under review is that, although the publisher describes them as "companion volumes", this does not quite do justice to their originality. Over half the chapters in Post-modern Interviewing are already printed in New lenses. Hence, if you have the New Lenses (550 pages), don't buy the other one (380 pages). [2]

The third thing to note about the two books under review is that this bizarre "overlap" between the contents of the two "companions" is not signalled by the publisher on the back covers or by the editors in either of their two prefaces. The hasty student or librarian might even find themselves ordering all three on the grounds that they seem to be distinct books. Certainly the notion of two "companion volumes" does not suggest that one of the companions has over half of the contents of the other companion already inside them! Frankly, this seems like sharp practice, whether post-modern or otherwise. Caveat emptor. For this reader, a slightly nasty taste was left in the mouth. [3]

On the other hand, the editors' and the publishers' care not to describe the relation between the two companions may be seen as a lesson in post-modernity. The modernist purchaser/reader like myself discovers that their notion of "companion" is a category that has to be completely "unsettled" and "deconstructed" in order to fit a situation. The situation is one in which one companion has devoured over half of the other companion. In addition, by searching in vain for an "authoritative editorial voice" to explain how the selection of the two later volumes was made from the 2001 volume and how the two companions relate to each other, the modernist reader is forced to confront a total absence. He or she is thus forced to decide that the search for a single meaning or an authoritative agent is itself a residue from an earlier epoch, and/or has to have their own un-checkable fantasies about this, "actively creating meaning", as the editorial suggests we do in interviews. What is or are the "subject position(s)" behind the names (GUBRIUM, HOLSTEIN, Sage, from now on referred to as GSH) that have produced this product? [4]

They have set up a virtual collective interviewee response to my perhaps too open-ended interview question "What is going on in interview methodology discussions in the USA these days?", and I am faced with making sense (as reviewer-reader) of their gigantic "multi-logue" of a response.1) Unfortunately, this is not an interactive interview in which I can actively interrogate GSH about their initial multi-logue; faced with this collective transcript of voices, I can get no further clarifications. So, as a solitary analyst, I have to make my sense of this text, and have then to produce "constructions" for the audiences I imagine may read this review in FQS. Following the postmodernist injunction to try different forms of representation, and not to worry too much about reality,—since we all have (only) our local "realities"—, I shall try different formats. [5]

My first is a content-focused review, for those who have not yet realised that to worry about the mere "truth content" of what is alleged in a so-called report is to define oneself as hopelessly lost in modernism and a belief in empirical reality and truthfulness. Post-modernly speaking, I know that this is not the thing to do since I have the warrant of ATKINSON and COFFEY's chapter in the book in which they argue against interviewers posing a certain question to themselves or to anybody else:

"As Silverman (1993) points out, we cannot approach interview data simply from the point of view of 'truth' or 'distortion', and ... by the same token we cannot rely on our observations to correct presumed inaccuracies in interview accounts ... Part of the reported comparison between participant observation and interviewing has revolved around the ironic contrast between what people do and what people say (they do). This has also fed into the equally hoary question posed by and to field researchers: How do you know whether your informant is telling the truth? These related problems equally reflect the position we have characterised as naïve: the contrasts between actions and accounts, and between truth and dissimulation." (New Lenses, pp.442-443; italics added, TW) [6]

I notice ruefully the terrible habit I have still not shed of giving in my account a page-reference (no less) so that you might, when reading my account, check to see whether I am citing Atkinson and Coffey correctly. If I had fully acquired the New Lenses that they promote, I would not worry (and nor would you) about being asked the question: "How do you know if the reviewer as informant is telling you the truth about ATKINSON and COFFEY?" Still, at the moment of my review, I am committed to rhetorical devices at least suggesting that I have fulfilled a "duty to truth", even if I have invented the quotation and the page reference. Please note, I also still refer to myself as a "unitary self", but eventually I hope to "go beyond" such superstitions.—So, now to the alleged content of this alleged publication. [7]

2. Content-Focused Review

New Lenses has a substantial Introduction by the editors (some 30 pages, most subsequent chapters average about 20 pages) that is followed by four parts. Each part is composed of some 4-7 chapters. (I note a further terrible positivist hangover in which I have succumbed to thinking that the quantification of qualitative data might be of some value: I have failed to recognise the "positivist quantifying enemy within", but no doubt you will be able to recognise this enemy and see it as wholly Other. If you are Christians, you can "cross yourself" at this point; otherwise take subculturally local and appropriate action to avert the evil eye or functional equivalent for qualitative researchers). [8]

Part I is entitled "Subjects and respondents" and has some excellent chapters on how to interview different sorts of people. [9]

There are separate chapters on interviewing people of different gender (one on men, one on women), people of different ages (one on children and adolescents, one on older people), people of different sexualities ("Queering the interview") and people of non-white races ("Race, subjectivity and the interview process"). In addition, I would take from GSH's "Part 2" ADLER and ADLER's excellent chapter "The reluctant respondent" and put it in my version of "Part 1" (as a post-modernist, I know that I have a perfect right to produce for you another "version" no less privileged than GSH's own; as a modernist, by telling you this, I'm leaving an audit trail), since, together with some parts of the editors' Introduction, the ADLER and ADLER chapter provides a useful more general concept of the "subject positions"; and "subjectivities" behind the formal role-position of "respondent" and "interviewer" that can lead to types of reluctance, co-operation and collusion in the co-produced interview process. [10]

Part 2 is entitled "Technical concerns". Ignoring the ADLER and ADLER chapter, this Part might be called "Hi-technical concerns". Apart from one chapter on "Standardisation and interaction in the survey interview", it deals primarily with technologies that can be used by researchers once they go beyond a tape recorder and pen-and-pencil. [11]

The earlier chapters of this Part move gradually up from "In person versus Telephone interviewing", through chapters on computer-assisted interviewing and internet interviewing. Not having used any of these technologies myself, I found them well-written and informative, making me feel like exploring them further. The last two chapters deal with questions of transcription quality (Blake POLAND) and a good review of CAQDAS (Computer-assisted qualitative data analysis software) and its diverse possibilities and limitations by Clive SEALE. [12]

Part 3 is entitled "Analytic options", dealing with how you analyse the well-transcribed interview material you gained using the technologies of Part 2 in the co-produced co-constructed active or passive interview events with the different sorts of respondents discussed in Part 1. [13]

Given that many of the software programmes for analysing interview material are oriented to retrieving segments in and across interview by code or by "string character", the CAQDAS article in Part 2 can be usefully placed here, next to CHARMAZ's chapter on "Qualitative interviewing and grounded theory analysis". The first part of CHARMAZ's chapter deals with the practice of "doing GT interviews" and can be usefully read in conjunction with Part 1. The second part of her chapter distinguishes interestingly between an "objectivist GT" tradition and a "constructionist GT" tradition; she places herself firmly in the latter. Three of the following articles are also pretty "constructionist". Two study "narratives" (one on personal narratives, and one on oral history narratives) and the last, on the ethnomethodological analysis of all interview material, demonstrates rather well the value of a modified "conversational analysis" (CA) approach to understanding movements of cooperation and negotiation within the interviewer/ee dialogue inside the interview. [14]

However, those who want to make inferences from interview material to objective realities about the external world (admittedly, the notion of a "referent" is a relic of a modernist attempt to distinguish between a word and the thing it refers to, the verbal production in an interview and the different realities that the interviewee may talk about; but let's keep it as a minority taste and amusing relic of a naïve past) are not completely ignored in Part 3. DEVAULT and McCOY's chapter "Institutional ethnography: using interviews to investigate ruling relations" provides an excellent update into interview-using research into local manifestations of trans-local ruling relations, based on the approach initiated by Dorothy SMITH. In its concern to recognise (nay even "investigate", such an empiricist word) relations of power, oppression, resistance and struggle, it relates well to the chapter in Part 1 by DUNBAR Jr., RODRIGUEZ and PARKER on the racialisation of the interview and in society as a whole. The concern for oppression and empowerment of the "Institutional ethnography" chapter also looks forward to two essays in Part 4, one by NARAYAN and GEORGE on folk and personal narratives and the other, the final essay of the volume, by Charles BRIGGS, on "Interviewing, power/knowledge and social equality", of which more anon. [15]

Part 4 is entitled "Representational Issues" but has only one or two chapters that deal with what might expect from such a title. [16]

The first contribution is another postmodernist struggle against the delusions of modernism. ATKINSON and COFFEY (already cited) attempt to persuade us to avoid in their eyes naïve questions such as "How do you know whether what your informant says is true?" or "How accurate is the interviewee's report of what he did on a given occasion in the light of what direct observation at the time noted that he did?" (This last is my reformulation of their formulation of the "hoary ironic contrast" between: "what people say they do and what they do".) [17]

As a modernist who thinks it important to distinguish (a) whether there were or were not weapons of mass destruction in Iraq ready for use in 45 minutes time, (b) whether Bush and Blair acted as if they believed that this was true (they didn't), and (c) whether they are bringing "democracy" to Iraq—as they say—or just extending colonial conquest and puppet regimes on a world scale to control oil and other key natural resources, I am not convinced by their arguments. But struggling to be post-modern, I must assure you that I would never say that they were "wrong", just that they live in a different "reality" from my own. As a modernist, though, I prefer journalistic books about Bush and Blair called "Lies: and the lying liars who tell them" to academic tomes that attempt to efface the difference between truth and lies. [18]

Predictably, I really liked the last chapter by Charles BRIGGS on "Interviewing, power/knowledge and social inequality". Working within a perspective informed by BOURDIEU and FOUCAULT, he argues that the "interview society", heralded as an advance towards the "democratic recognition of all persons" by some (including certain subject positions behind GSH), is in fact a cover for increased surveillance and control by those with power who are accelerating the increase of social inequality and powerlessness at the planetary level. His rather complete pessimism about the "empowering" nature of interviews (except for the already powerful) is a useful corrective to the apolitical technicism of some authors and to the giddy optimism of yet others: it too, however, seems in its turn a bit extreme. [19]

BRIGGS notes that the relatively poor and dependent find it difficult to protect themselves from being interviewed by middle-class researchers, whose "results" can be then re-circulated (after being contextualised differently) by those with power, while the powerful can easily protect themselves from being interviewed except in a way that supports their public relations. As a sociologist (name forgotten) of an earlier generation said, "The eyes of the sociologists are on the down people; their hands are turned towards the up people". BRIGGS's argument about the interviewees individual and collective lack of control over their "re-contextualised interview material in circulation" links very well with the "Institutional ethnographers" identifying how local text-mediated practices serve "the relations of ruling". [20]

An article on "Cross-cultural interviewing" by Ann RYEN is thorough but unexciting, and should really have gone into Part 1 dealing with different sorts of folk that get interviewed. She makes useful points about there being no "simple insiders" in complex divided societies—to be in one fraction of that society is to be out of another fraction. Her discussion is useful on the possibilities of mis-recognition and mis-understanding across "boundaries" and on questions of apparently common symbols or linguistic or non-verbal behaviours which mean subtly or grossly different things. As she says, such misrecognition and misunderstanding across boundaries should not be thought of as unique to "societal" or "cultural" boundaries. [21]

NARAYAN and GEORGE's chapter on personal and folk narratives talks interestingly about the way that any given narrative may be a personalisation of a given folk-narrative or a cultural-genre inflected expression of a personal narrative and argues that the two categories of "folk" and "personal" are in fact extreme points on a spectrum. Folk narratives are selected and tailored by individuals for an audience and a moment; personal narratives must draw on folk-categories and genres to be effectively communicable to others. [22]

This chapter goes well with Richard Cándia SMITH's chapter on oral history in Part 3. NARAYAN and GEORGE discuss the value of eliciting people's stories about stories and story-telling, citing Alan DUNDES (1966) concern for the meta-level beyond that of just eliciting the primary stories themselves. Their and his suggestions can be expanded to consider the art of improving our understanding by getting interviewers and interviewees to tell stories about interviewing and being interviewed. Much of Part 1 in this book and the concern for methodological reflexivity is precisely an attempt to do that. If in Part 3, RIESSMAN's mode of analysis of personal narratives suggests that trying to get at their "truth" is not a particularly productive exercise and Richard Cándia SMITH (largely following Luisa PASSERINI) sees oral history testimonies from oppressed and struggling working-class communities as likely to be factually wrong for reasons of "keeping hope alive", NARAYAN and GEORGE's chapter seems to me to be in a more productive middle position, combining an optimism about factual accuracy with a concern for cultural genres. [23]

There is an author index and a subject index, and that's all very useful. [24]

I feel a great sigh of relief at having done my duty and made points about each article—some factual material, some prejudiced comment—. Hell! I then realise I have forgotten the one article in Part 4 which is really about representation, Carolyn ELLIS and Leigh BURGER's "Their story/my story/our story: including the researcher's experience in interview research". It's the one—I now realise—that made me feel free to experiment slightly with the "form" of this book review for FQS. Thanks, Carolyn and Leigh—though the readers of FQS may not thank you! Indeed, its influence can be seen in the first four sentences of this very paragraph. [25]

Irritatingly subjective as I found its lengthy descriptions of the domestic life of the autoethnographic authors (dogs leaping up in warm living rooms; I myself am writing in a cold farmhouse in Central France with snow around me and a paraffin heater to keep me warm), it moves between a number of components: autoethnographic self-description of the two authors trying to put the chapter in question together, negotiating uneasily the fact that one believes in Jesus as the Jewish messiah and the other can't bring herself to tell the believer that she can't share this belief; well-constructed literature reviews (this has a wonderful I hope self-conscious parody of mass referencing with some 28 authors cited by name; date format with no other comment); flash-backs to earlier autoethnographic descriptions about abortion seen through male and female eyes and about food and bulimia between one of the authors and three other women; as well as the interviewing of a Rabbi and his wife in the "Jesus as Messiah" group. ELLIS and BURGER also provide a typology of the stories that individual researchers write—reflexive dyadic interviews, interactive interviews, mediated co-constructed narratives and unmediated co-constructed narratives—and provide exemplars of each. The chapter's autoethnographic presentation of its contents—which can be as difficult to assimilate as this paragraph—concludes with an ethnography of the meeting at which the two authors decided the actual title of the chapter; the typing of the title is the last line of the chapter itself. [26]

So much for the attempt to give you, dear readers, factual information on New Lenses.—I restrain my sense of boredom at the thought of doing the same for the "companion volume", Postmodern interviewing. I think I'll take a postmodernist's license. As said before, GUBRIUM and HOLSTEIN and Sage have produced two volumes of which half the second volume is the same as the first. I've decided, therefore, that I'm producing a review of both volumes, which will really only be a review of one of them. Post-modernism is nothing if not a license to be cavalier. [27]

I will just say that two new chapters in the latter volume which I liked were HOLSTEIN and GUBRIUM's "active interviewing", and ROSENBLATT's meditation on "Interviewing at the border of fact and fiction" in which the division between modernist conscience and post-modern identity is first formulated in terms of himself as postmodernist struggling with interviewees and audiences who are all modernists wanting truth, and then in terms of his choice of taking the "truths" he's found by research and expressing them in novels, poems, plays, etc. This discovery of the jouissance of being "more truthful to the truth by inventing truths" in literary forms is one he shares with DUNBAR Jr. et al. (Part 1, Chapter 7) who also have experimented with powerful non-academic forms for empowering and activating activists and others. A third chapter in this volume which I also liked—and liked more than the title made me think I would—was Norman DENZIN's "The cinematic society and the reflexive interview". He writes

"A responsible, reflexive, dialogic interview text embodies the following characteristics (Trinh 1991: 188):

  • It announces its own politics, and evidences a political consciousness

  • It interrogates the realities it represents

  • It invokes the teller's story in the history that is told

  • It makes the audience responsible for interpretation

  • It resists the temptation to become an object of consumption

  • It resists all dichotomies (male/female and so on)

  • It foregrounds difference, not conflict

  • It uses multiple voices, emphasising language as silence, the grain of the voice, tone, inflection, pauses, silences, repetitions

  • It presents silence as a form of resistance" (Postmodern interviewing, p.152). [28]

I disagree with some of these points as stated, but they certainly seem to be worth considering. They could provide some of the criteria used for evaluating the multi-logue interview text that is the two volumes of GSH. [29]

So, on to my own reflections about all this. I wrote some 12 pages of notes as I went through each of the chapters. What general points do I want to make (retrieve and paste from the 12 pages) to conclude this review? [30]

3. General Points

From my (realist) point of view, the task of developing a more reflexive social science—a project shared by the editors and most of the contributors to the volumes under discussion—is to explore precisely what can be inferred about internal or external reality once the more or less artful local co-production of a well or badly designed interview has taken place. The anti-realist position, which celebrates the difficulties in order to deny that there are any "inner world" or "outer world" realities to make inferences about, or to state or imply that it is impossible to search usefully for data about them in interviews, is the most powerful ideology running through this book, a position we might term "American anti-realism". The text is strewn with constant ideological asides about "going beyond the conventional"; about "naïveté". This strain in the book is a powerful factor for confusing students and researchers. "Qualitative anti-realism" is exactly the enemy that "positivist realists" love to have in front of them, since the position is philosophically and politically self-destructive. [31]

On the other hand, the use of ethnomethodology—and Conversational Analysis (CA) in particular—to disentangle from interviews what is too much "contaminated" by the artful practice of interaction and then use what remains as usable evidence of something about inner and outer world realities and their interlocking both in the interview and particularly beyond is a reflexive methodology of great value. Incidentally, for those interested in such evidence, one strategy of reducing the interviewer's contribution to the emergent talk is to adopt not an "active interviewer strategy" but a "minimal interviewer strategy". Those interviewing strategies that use, for as long as possible, a minimalist strategy (for example the single open-ended initial question asking for biographic narrative, as in biographic-narrative interviews in the Quatext/BNIM approach, see e.g. WENGRAF 2001) delay for as long as possible the moment of "activist" interviewing. They therefore have a baseline of pretty "spontaneous interviewee production" for let us say the first half or more of the interview with which to then explore the more complex later moment of co-production through higher levels of "interviewer input". [32]

The struggle of the interviewists and "verbalists" (interview interaction is the most real thing of all) reaches a peak in COFFEY and ATKINSON's attempt to rubbish the distinction between "what people say and what they do" or even, Heaven forbid, "How do you know if your informant is telling the truth?". They attack what they call the essentially "romantic" notion of the "social actor as a repository of 'inner' feelings and intensely personal recollections". Similarly, they say that "the specific dualism that asserts an authenticity for what people (observably) do and the fallibility of [interview] accounts of [such] action is specifically unhelpful and 'untrue' ... We need to divorce the interview from the myth of inferiority" (pp.424-425).2) [33]

By asserting the truth that participant observation accounts of an event are structured by the sensitivity and the recording practice of the observer and that a later interview of the actors in that "event" is structured by the sensitivity and recollection practices of the actors, they move to say that we should not give any greater authenticity to the "direct observation" than to the "later recall". Both are "forms of action". However, if there is a record on film of a black man being beaten up by police and then in interview or in the courts the police deny this ever happened, would COFFEY and ATKINSON argue so strongly that we should avoid privileging (recorded) direct observation? If the constant planetary practice of US regimes is destroy all attempts at local autonomy by massive use of weapons of mass destruction, should not this observed practice be given privilege as against their spokesmen's denials that they would ever contemplate anything of the sort? The desire to prevent word-action from being considered ironically when contrasted with the track-record of other (more material) actions is characteristic of the public relations experts of those in power. It is in the professional interest of social researchers who fetishise the interview (interviewists) that they would wish to avoid any sense of the limitations of interview methodology. It is a characteristic of verbal intellectuals that they would seek to deny that there is "anything" beyond words, any aspect of "inner or outer realities" for which words may be inadequate and which may exist even if there were no words or wordsmiths. If words or verbalised beliefs are the only reality, as a previous generation argued that "there is nothing outside discourse", and as MARX criticised in The German Ideology, then the supremacy of verbalists in general and interviewists in particular is by definition unchallenged. [34]

The volume can be seen as a "symptom". There is an interesting clash (sometimes in the same chapter) between an intoxicating discourse about the intensifying "interview society" on the one hand and, on the other, a more craftsman-like concern with the ever-reluctant interviewee, whose specificity is increasingly difficult to access (behind the difference) and whose self is defensively presented, and the representation of whom is ethically, legally, and philosophically increasingly restricted. [35]

It is clear that the strain of anti-realist verbalist interviewism is not, most of the time, a fully avowed and explicit ideology. If it were, then its credentials to be a social science would have to be surrendered, since a science is concerned to engage with Reality, and an anti-realist ontology and epistemology that asserts the multiplicity of "realities", the "relativism of all alleged truths" and the absence of an inner or outer world is going to be in serious funding and legitimation trouble. Anti-realism (philosophic idealism) is an ideology that only rarely speaks its name, and never considers its own anti-realist practice in a reflexive way. Indeed, the God-trick means that it says one thing but rhetorically does another. LYOTARD said (or should have said) "Meta-narratives are dead. There are only local narratives", disguising from the reader that such a statement is based on LYOTARD as hidden God, requiring you to accept this new meta-narrative without realising that it is one. This trick is played over and over again in postmodernist and constructivist texts, without the authors realising it. I like to think that, if they realised it, they couldn't do it. [36]

For example. RIESSMAN's "analysis of personal narratives" distinguishes usefully between different uses of the term "narrative"—ranging from a broad account in which the researcher's account uses self-reports through to "accounts of action sequences" embedded in other sorts of talk. She gives an interesting example of talk by an Indian woman, Gita, to demonstrate the value of considering personal narration as "performance". However, she falls into the usual contradictions of "constructionist" rhetoric. Just after she has assured us in the most realist fashion possible that "turning points" fundamentally change the meaning of past experiences, she argues that "the trustworthiness of narrative accounts cannot be evaluated using traditional correspondence criteria". The basis for her realist assertion? The fact that in the interviews women say that they had such turning points. A thorough-going constructionism (anti-realism) would deny that anything could be known about whether such "turning points happened in the life" just from the narratives, and would have to assert that, from the narratives, we could only know that "telling stories with turning points" was a favoured genre.3) The "God-trick" of the post-modernist constructivist operates here as in other chapters. With this exception, the chapter is a very useful introduction to the analysis of personal narratives, even though she gives no special guidance as to how to generate such biographic narratives. [37]

As I have read through the book, my sense of a confused civil war among (primarily qualitative) interview researchers and philosophers grows and grows. The fear of the use of administrative positivism by the business-military complex drives many into a vulgar anti-realist position; the fear of being naïve produces philosophical confusion not just between but within chapters. Post-modern insights used to help a more cautious and more powerful subtle realism are of the greatest value; post-modern dogma used to displace notions of truth and reality and scientificity play wonderfully into the hands of the enemies that their protagonists think they are fighting. [38]

The sceptical desire to strip away the assumptions of simple empiricism by showing the concept-laden nature of all "data" is, of course, a good thing. The critique of commonsense to produce "better science" by overcoming epistemological obstacles and by "rectifying previous scientific and prescientific notions and conceptual frameworks" is one step further. In this respect, though much neglected by English and German researchers, the work of Gaston BACHELARD, especially his Formation de l'esprit scientifique: contribution à une psychanalyse de la connaissance (1938) is central (see also LECOURT 1969). [39]

Any verbal record of a flow of events captured on film (e.g.) will involve some level of conceptualisation; that's quite true. However, there is a distinction between the degree of conceptualisation involved in a minimally-theoretical observation language which attempts to be relatively concrete and uncontroversial, and that involved in an over-conceptualised summary in which—as ATKINSON and COFFEY rightly point out about BECKER and GEER—"we catch very few actual glimpses of say the actual work with patients on hospital wards or in clinics. The book does remarkably little to report what these social actors actually did" (p.420). In passing, note that at this point, BECKER and GEER are criticised for not giving enough "thick description" of what the social actors actually did—a reasonable point for a criticism of bad reporting by observationalists, but one which, against the formal position of some postmodern constructionists, suggests that a better attachment to a less summarising and conceptual language of participant observationalists could produce "actual glimpses of the actual work ... what social actors actually did". Formally, this position contradicts ATKINSON and COFFEY's anti-realism, for which the notion "actual glimpses of what social actors actually did" should have no (privileged) meaning. [40]

The detailed ethnomethodological analysis of interviews shows the considerable scientific productivity of precise analysing of the "interaction activity" within interviews in which the co-production and maintenance of interviewer/interviewee roles is sustained by constant "work" on the part of both parties. Although Carolyn BAKER starts her chapter by the obligatory anti-realist rhetoric and disparagement—she says "such analysis goes well beyond the conventional 'content' or 'thematic' explications, where interviewee talk is seen as information about interior or external realities"—by the end of her excellent exposition, her argument becomes more qualified and useful.

"Studying the social organisation of talk about or around a topic does not mean losing sight of what is said in terms of propositional 'content' in terms of reports on realities external to the interview ... Nor does it meaning losing the 'interior to persons' reports ... However there is much reason researchers should proceed with caution in treating these kinds of reports as 'interview data' without looking closely at the specific local circumstances and conversational interactions that generated these reports ... They represent a far more reflexive social science practice ... In such analyses it becomes impossible not to see the artful practices of interviewer and interviewee in making the interview happen, and consequently it becomes very difficult to unhitch 'answers' from their (em)bedding in an actual, local, situation of production" (p.411). [41]

SEALE's remark that "most practising social researchers recognise the relative autonomy of their craft from the absolutist epistemological and theoretical debates that once appeared to divide them" (p.298) is relevant here. Unfortunately the "passionate absolutism" of strong post-modernism is deeply present in this volume. At one level, the chapters, particularly the experience-based chapters, are full of good technical advice with ample references to further relevant reading. At another level, the ideological struggles and confusions within and between the chapters indicate a collective confusion among researchers interested in (qualitative) interviewing, which suggests to me a great waste of energy in "absolutist battles" powered by ideological currents and contradictions. At this level, I'm tempted to say that, "with a community of scholars like this, who needs enemies?" [42]

More to the point, the volumes desperately need an editorial posture that enables us to appreciate the valuable contribution of moderate constructivism, co-productionism, even post-modernism to the overcoming of vulgar realism in which the words of the interviewer can be taken as simple transparent truths about internal and external past and present reality without oscillating into the corresponding error, of a ludicrous and extremely reactionary trap, of denying the scientific search for truth, more valid accounts, the correction of earlier truths now seen as productive errors, etc. A BACHELARDian analysis would suggest that such a passionate drive for postmodernist purity (the denial of science that dares not speak its name for fear of loss of funding, audience, respectability or point) (a) provides the empiricist positivist enemy with just the enemy it needs to continue its dominance, and (b) makes it very likely that the extreme verbalist rhetoric currently fashionable will suddenly lead to a switch in fashion towards the "vulgar realism of empiricist/empirical work" when the champagne of verbal intoxication goes flat. [43]

What should a Handbook of interview research promise? A "Handbook" (like a car handbook) is something that helps you in doing what you have to do to get better results from a product or practice. It should help us to design our interviewing practices better so as to achieve better raw materials which when subject to better analyses produce findings that can be seen as more scientific and reliable than earlier practices can now be seen to have delivered. [44]

GSH does this, but then undercuts its contribution by a wash of self-contradictory radical constructivism/postmodernism. [45]

I apologise to those I have insulted and probably misrepresented in this review essay and to those who find it too full of spleen. I have the sense that I've been unfair to ATKINSON and COFFEY whose less rhetorical research work I find excellent. They don't quite say that there is no reality, but they do appear to me to be saying that there is no difference worth recording between the activity of doing and the activity of saying, that only inferiors are so naïve as to worry about the truth of somebody's report of the past or present. As such, to me their text commits the trahison des clercs, given that oppressed groups and societies need the truth to be told against the lies of their oppressors and the lying liars who tell and re-circulate (as Charles BRIGGS stresses) those lies. Strong post-modernism and constructivism in its ideological manifestation (which I detest, as opposed to its critical work which I respect) seems to function as the academic counterpart to the public relations and propaganda industry of the business-military elites currently destroying the world (see the destruction of Kyoto-consciousness under the pressure of Standard Oil, Enron, etc) and to be desperately repeating great lies to conceal this, great lies of which one is that there is no distinction between lies and truth. Luckily, there are many useful contributions towards "subtle realism" and "a more reflexive science of the reluctant subject and the naïve researcher" in this volume, and so there is much to be gained from reading it, as long as the ideological "framing" of that volume (and indeed of this so-called review, it seems like a New Year rant) is heavily suspected. [46]


1) Not unexpectedly, nobody from Continental Europe (or indeed the non-anglophone world) is considered to have anything special to contribute. Imperial powers are always parochial at heart. <back>

2) This last phrase is interesting. The whole of their argument is about denying that the "interview" (listening to what people say about what they did and do) is at all inferior to participant observation (being there while they did what they did, and experiencing it directly). So I read this point without a qualm. Actually, on re-reading, the context makes it clear that this is a Freudian lapse, for "interiority". The passage runs: "A recognition of the performative action of interview talk removes the temptation to deal with [interview] data as if they give us access to personal or private 'experiences'. We need therefore to divorce the interview from the myth of inferiority [sic]: the essentially romantic view of the actor as a repository of 'inner' feelings and intensely personal recollections". The fight against the suppressed inferiority complex of interviewists is so strong that they write "inferiority" when they meant "interiority" both in the 2001 volume and in this reprint. It is interesting that, in my opinion and those of many of the not-too-post-modern contributors to this book, the strength of the interview is that it allows the socially-silenced to speak, that it allows for intensely personal narratives and the expression of personal experiences and feelings, and so on. Were we to accept ATKINSON and COFFEY's refusal to allow that such things might go on, a verbalist denial of the experiencing reluctant and occasionally enthusiastic respondent, then in my opinion the case for the "inferiority" of interviews compared to participant observation would be strengthened. Methinks the lady and the gentleman do protest too much. Note in the passage quoted in italics above the use of the "myth/reality" distinction and even the reference to "untruth" (admittedly the latter is in single quotation marks) for which their official epistemology has no warrant. <back>

3) Indeed ATKINSON and COFFEY would deny that there are anything like objectively-existing "past experiences" that can have first one meaning and then another. <back>


Bachelard, Gaston (1993). Formation de l'esprit scientifique: contribution à une psychanalyse de la connaissance. Paris: Vrin.

Dundes, Alan (1966). Metafolklore and oral literary criticism. Monist, 60, 505-516.

Gubrium, Jaber & Holstein, James (Eds.) (2001). Handbook of interview research. Thousand Oaks: Sage.

Lecourt, Dominique (1969). L'épistémologie historique de Gaston Bachelard. Paris: Vrin.

Wengraf, Tom (2001). Qualitative research interviewing: biographic narrative and semi-structured method. London: Sage.


Tom WENGRAF is visiting senior lecturer in social research methods at Middlesex University, UK and visiting senior research fellow at the University of East London, UK. He has undertaken fieldwork on agrarian reform in Algeria, on student experience and on social exclusion in Britain, and is currently involved in a mixed method evaluation of the work with older people of a healthy living community centre in East London. With Prue CHAMBERLAYNE, he co-edited Biography and social exclusion in Europe: experiences and life journeys (Policy Press 2002) and runs occasional training courses in biographic-narrative interpretive method (BNIM). He can sometimes be quite reasonable: for example, see his recent Boundaries and Relationships in Homelessness Work: Lola, an Agency Manager.


Tom Wengraf

24a Princes Avenue
Muswell Hill
London N10 3LR

E-mail: tom.wengraf@cheerful.com


Wengraf, Tom (2004). Paradigm Wars Around Interview Methodologies: Constructionism and Postmodernism "on tap" or "on top"? Review Essay: James Holstein & Jaber Gubrium (Eds.) (2003). Inside Interviewing: New Lenses, New Concerns / Jaber Gubrium & James Holstein (Eds.) (2003). Postmodern Interviewing [46 paragraphs]. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 5(1), Art. 30, http://nbn-resolving.de/urn:nbn:de:0114-fqs0401309.

Copyright (c) 2004 Tom Wengraf

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.