1.

This book, introduced by Ian PARKER, brings together 16 chapters grouped into four parts identified by the nature of texts analyzed: Spoken and written texts (interviews, letters, fiction, and lessons); visual texts (comics, advertising, television, and film); physical texts (cities, organizations, gardens, and sign language); and subjectivity in research (bodies, ethnography, silence, and action). Each of the rather brief chapters (10-15 pages) can be said to have followed a very strict format of introduction/exposé, discussion of the text ("Text"), a plate with the text to be analyzed, an analysis ("reading"), as well as "disadvantages" and "advantages" of the approach. [1]

2.

The authors begin their chapters with an untitled section, where they present a brief and far from comprehensive review of the methodological literature on their chosen topic (meaning in groups, romance, absence of children's view on literature, etc.) and methodology of research and analysis (discourse analysis, phenomenology, activity theory, etc.). Subsequently, the authors use a framed text box to present their data, and a description of this data in a section titled "text." They analyze their data in the sections "reading," and then describe, in the final two sections, the shortcomings and strengths of the chosen theory or method. Soon after beginning the book, the line "resté sur ma faim" (left hanging) surged and occupied me, as if asserting itself, the French much better expressing the difference between the great anticipation ("hunger") with which I had started this book and what I experienced while reading it. The shortness of the chapters and the rigidity of the representation left underdeveloped many of the ideas that I thought needed to be addressed, particularly at this point in the history of discourse analysis, that is, more than a decade after Opening Pandora's Box (GILBERT & MULKAY, 1984) or the ground-breaking work Discourse and Social Psychology (POTTER & WETHERELL, 1987), or Discursive Psychology (EDWARDS & POTTER, 1992) all of which I had tremendously enjoyed and used in the past. [2]

The rigid format generates interesting tensions, for it forces authors to present in textual form what might otherwise be difficult to represent in textual form. In other words, being forced to present their data in textual form, the authors are forced to generate data that turn everything into text—bodies, lived experience, film, or silence. It might have served the authors and editor well to make a crucial distinction between data and data sources. Data sources may take all sorts of form, for example, the audio recording of an interview or a telephone call, the lived (unreduced) experience of walking through a 19th-century Victorian garden, or the visits to a historical site and Granada Studios in Manchester. From these data sources, the author constructs the data that are subsequently analyzed. The contributors to Critical Textwork (CT) would have definitely benefited from an explicit discussion of the transformations involved in going from the phenomenological dimension of an event to the ultimately published chapter. (Pertaining to conversation analysis, ASHMORE and REED [2000] have conducted such an analysis.) [3]

That is, the authors do not explicitly discuss that what the reader gets to see as data has already gone through at least one level of filters and analysis. This is an important step to make salient, for the reader can no longer reproduce the same analysis that was done by the author. As my own research among high school physics students showed, just what the observer sees when looking at a demonstration may differ among spectators (ROTH, MCROBBIE, LUCAS, & BOUTONNÉ, 1997). Thus, what the students described as having occurred after watching a teacher, sitting on a rotating stool and spinning a bicycle wheel differed—some said that the teacher moved; others suggested that he didn't move. Based on these observations, they subsequently used a variety of theoretical models to explain why the teacher should have (not) moved. Because their data were different, it is not surprising that they used different theoretical frameworks in order to produce consistent readings. [4]

Thus, when the chapters move to different forms of "text," such as bodies, cities, films, television, or gardens (see my version of it in the insert), the number of problems increases. There is a curious absence in the text of the things being talked about. This absence begins with PEARCE's chapter concerned with advertising, but the data of which it is a description of a poster rather than a copy or an iconic depiction of the poster. Throughout reading his analysis, I wanted to see the poster in order to check the primary reading, which had produced descriptions such as "unusually thin," "she looks somewhat expressionless at the audience," or "seemingly untidy but paradoxically regular." (In several cases, the "data" actually contain analytic commentaries [e.g., CT, p. 83, 120].) While reading, I annotated (in the margins) RUSSELL's analysis also with the comment "same problem as before ... lack of opportunity for the reader to disagree with the analysis." I thought readers might enjoy a discussion of the difference, differance, differal, of (phenomenologically) analyzing primary, lived experience and analyzing secondary, tertiary, or, for that matter nth-order texts. [5]

3.

Even if I was to overlook these aspects of the book, there were many other ways in which I was disappointed. One of the important disappointments resided in the frequently problematic relationship between the "data" and the "reading" that the authors provided. Repeatedly, I could find no evidence for a particular reading, for the required data had not been presented. There was therefore the tension that the authors presented four pieces of transcripts (BEVAN & BEVAN) or two letters (PHILLIPS) but in their analysis really required much more from their data to support the claims actually made. For example, PHILLIPS conducts a reading of Letter 1 in which the writer makes reference to a person named Lizzie. He then suggests that "Lizzie's inclusion is always significant, as she is perceived as a rival" (CT, p.33). Thus PHILLIPS reads not just this letter but interprets it in terms of his reading of all of the other letters. Here, the notion of intertextuality of a particular text in the body of texts analyzed, the intertextuality of the texts analyzed to other parts of the author's text (e.g., LEMKE, 1992) is insufficiently theorized. I would have thought that a reading of JOYCE's Ulysses or Dubliners might have given rise to tremendous insights about reading and discourse analyzing letters. [7]

4.

In part, the problems are created because the authors voluntarily write themselves out of the narrative. Their agency becomes a mediated agency, or the agency is attributed to an abstract analysis ("[the reading] has tried to show" [CT, p. 76) or chapter ("This chapter is concerned with" [CT, p. 129] or "this chapter had argued" [CT, p. 150]) and section ("The above section raises questions" [CT, p. 209]). This then leads to disembodied readings that anyone could or perhaps should be experiencing. The readings become normative without revealing the community that adheres to this norm, leading to transcendent meanings available to readers in general. Throughout my reading of this book, I became and remained uncomfortable with the apparent assumption that the texts rallied as data would lead to a particular, the author's favorite reading. [8]

Of course, the problems I perceived are not problems in some absolute sense but are related to my own developmental trajectory, having enacted discourse (and conversation) analytic studies of my own whenever the topic of my interest lend itself or required such a form of analysis (e.g., ROTH, 1993; ROTH & ALEXANDER, 1997). The problems exist in reference to my interpretative horizon, a contingent moment of a continuously changing and evolving scholarly Self to which I indexically and self-referentially refer to as "I." I did not come as a novice who felt the need to read an introductory text for students of discourse across the social sciences. Whether I would have been disappointed had Critical Textwork been the first book on discourse analysis is a mote question. Overall, I finished the book feeling that it did not hold what I felt it had promised me; or rather, there was a gap between the actual reading experience and the one I had expected. The one lesson I thought it might teach newcomers to discourse analysis is that "discourse" pertains to situations more broadly rather than text. On the other hand, because of the format of presenting "data" in largely textual form, a naïve reading might come to the conclusion that "everything is text." [9]

5.

I was wondering how I could best encapsulate my feeling about this book. Given that it was announced as an introduction for newcomers to discourse analysis, I thought that an appropriate evaluation would be whether I would consider using the book as a central or peripheral resources. Would I recommend these books to the students in my Interpretive Inquiry courses, that is, the ones I just finished teaching and, more importantly, the other one that I will be starting within weeks from writing these lines? Whereas I could ponder this question in the abstract, this pondering would not bring me closer to resolving a practical question. Not being a person of lengthy (disabling) considerations, I decided not to use these chapters as resource materials because the tight theory-data-interpretation integration that I wish to get my students to attend to is absent from most of the chapters in this book. In my reading, the chapters were in many respects somewhat naive in the way I might expect it from relative newcomers in the research community. [10]

References

Ashmore, Malcolm & Reed, Darren (2000, December). Innocence and Nostalgia in Conversation Analysis: The Dynamic Relations of Tape and Transcript [45 paragraphs]. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research [Online Journal], 1(3). Available at: http://www.qualitative-research.net/fqs-texte/3-00/3-00ashmorereed-e.htm [Date of access: April 17, 2001].

Edwards, Derek & Potter, Jonathan (1992). Discursive psychology. London: Sage.

Gilbert, G. Nigel, & Mulkay, Michael (1984). Opening Pandora's box: A sociological analysis of scientists' discourse. Cambridge: Cambridge University Press.

Lemke, Jay (1992). Intertextuality and educational research. Linguistics and Education, 4, 257-268.

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

Roth, Wolff-Michael (1993). Metaphors and conversational analysis as tools in reflection on teaching practice: Two perspectives on teacher-student interactions in open-inquiry science. Science Education, 77, 351-373.

Roth, Wolff-Michael & Alexander, Todd (1997). The interaction of students' scientific and religious discourses: Two case studies. International Journal of Science Education, 19, 125-146.

Roth, Wolff-Michael, McRobbie, Cam, Lucas, Keith B. & Boutonné, Sylvie (1997). Why do students fail to learn from demonstrations? A social practice perspective on learning in physics. Journal of Research in Science Teaching, 34, 509-533.

Citation

Please cite this article as follows (and include paragraph numbers if necessary):

Roth, Wolff-Michael (2001, May). The Politics and Rhetoric of Conversation and Discourse Analysis. Review Essay, Supplement I: Ian Parker and the Bolton Discourse Network (1999). Critical textwork: An introduction to varieties of discourse and analysis [10 paragraphs]. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research [On-line Journal], 2(2). Available at: http://www.qualitative-research.net/fqs-texte/2-01/roth/2-01review-roth-intro1.htm [Date of Access: Month Day, Year].