LINGUIST List 19.2201

Wed Jul 09 2008

Review: Discourse Analysis: Englebretson (2007)

Editor for this issue: Randall Eggert <>

        1.    Marian Sloboda, Stancetaking in Discourse

Message 1: Stancetaking in Discourse
Date: 09-Jul-2008
From: Marian Sloboda <>
Subject: Stancetaking in Discourse
E-mail this message to a friend

Announced at

EDITOR: Englebretson, Robert TITLE: Stancetaking in Discourse SUBTITLE: Subjectivity, Evaluation, Interaction SERIES: Pragmatics & Beyond New Series 164 PUBLISHER: John Benjamins YEAR: 2007

Marián Sloboda, Charles University in Prague

SUMMARY The volume deals with the activity of taking stance, both evaluative and epistemic (i.e. a person's expression of knowing or not knowing something). The book is not a psycholinguistic enterprise: the 10 contributions approach stancetaking as a discursive and social interactional activity. It is not surprising, therefore, that most of them used corpora of spoken conversational language as the data. To answer the question of what ''stance'' actually is the editor and author of the Introduction, Robert Englebretson, in a methodologically congenial way, examines corpus data in order to find out what ordinary language users themselves mean by saying ''stance''. He concludes that stance: (1) refers to physical embodied action, personal belief/attitude/evaluation, and social morality, (2) is a public act, (3) is a relational, interactional phenomenon, (4) is consequential for stancetakers, and (5) is socioculturally embedded.

The Introduction outlines the arrangement of the volume in the following way: The first four chapters (following the Introduction) each present a different approach to stance research. The chapter ''Using a corpus to investigate stance quantitatively and qualitatively'' by Susan Hunston is a meta-methodological discussion of stance research using linguistic corpora. In her chapter ''Linking identity and dialect through stancetaking'', Barbara Johnstone analyzes the negotiation of epistemic stance in a qualitative research interview about a link between social and linguistic identities, an issue in sociolinguistic research. In his chapter ''Stancetaking in colloquial Indonesian conversation'', Robert Englebretson deals with the grammatical encoding of stance and the use of stance-expressing grammatical means. The chapter ''Subjective and intersubjective uses of generalizations in English conversations'' by Joanne Scheibman is a discourse linguistic approach to stancetaking. This chapter is an attempt at an analysis of categorization and extreme case formulations independent of the Membership Categorisation Analysis research tradition which also deals with these topics (e.g. Jayyusi, 1984; Lepper, 2000; on extreme case formulations, see Pomerantz, 1986).

In the sixth chapter ''The stance triangle'', John W. Du Bois extensively theorizes about stance and provides a triangular model of stancetaking. This chapter also marks the transition to the conversation-analytic approach (e.g. Hutchby & Wooffitt, 1998) adopted in the remaining four chapters. These four chapters analyze stancetaking by means of the expression ''I guess'' (in ''The role of _I guess_ in conversational stancetaking'' by Elise Kärkkäinen), of the Finnish expressions ''minun mielestä / minusta'' ('I think' in ''Stance markers in spoken Finnish: _Minun mielestä_ and _minusta_ in assessments'' by Mirka Rauniomaa), of yes/no interrogatives and tag questions (''Stancetaking as an interactional activity: Challenging the prior speaker'' by Tiina Keisanen) and, finally, the last chapter deals with answers to interviewers' difficult questions in news interviews (in ''Positioning and alignment as activities of stancetaking in news interviews'' by Pentti Haddington).

EVALUATION What should be evaluated positively about the book under review is, among other things, the editorial work, e.g. the ordering of the chapters and cross-referencing which contribute to a high level of integration. A minor disappointment on the part of the reader, however, could occur in two cases: the sociolinguistic approach of Johnstone's chapter announced in the Introduction does not differ much from discourse analytic approach present in some other chapters, and the discourse linguistic approach promised to be present in Scheibman's chapter may be difficult to find due to the weakness of discourse analysis in that chapter (see below).

As well as other multi-authored volumes, this book has its stronger and weaker chapters.

Hunston's chapter ''Using a corpus to investigate stance quantitatively and qualitatively'' is a very good meta-methodological reflection on the qualitative and quantitative aspects of research into stance using linguistic corpora. However, the author fails to draw all the consequences from her own statement that ''context is crucial in identifying stance [in the corpus data], and that this must be borne in mind when relatively crude quantitative measures are used'' (p. 36). Namely, she does not include her, i.e. the analyst's, own knowledge of language or of dictionary definitions of linguistic expressions as a type of qualitative data which also should be taken into account in quantitative analysis of those expressions. A paragraph on page 42, for example, can illustrate this point. The paragraph is preceded by this extract from the Bank of English corpus:

''This assessment, given by one of the nine members of his committee, reflects a genuine warmth and respect for their chairman. _Unfailingly courteous, to the point of gentleness_, Lord Nolan also displayed a fierce independence and skill in both bringing together his disparate committee members...''

Hunston comments on this data extract in the following way:

''It can be deduced in this example that gentleness, in the context of committee chairmanship, is a bad thing, because _unfailingly courteous to the point of gentleness_ occurs as a concession countermanded by positive evaluation (_fierce independence and skill_). However, the phrase _to the point of_ provides the reader with a short cut to arriving at this interpretation.''

Hunston demonstrably used her knowledge of English in the analysis of that example, but failed to explicitly recognize this: it is not clear (at least to me) whom the expression ''the reader'' refers to in the last sentence, and the author did not show any evidence for her claim that the phrase _to the point of_ indeed provided reader X with a shortcut to such an interpretation of the extract. In other words, the author's interpretation presupposes certain knowledge of English expressions which, however, remains unspecified in the chapter. That knowledge, however, should also be treated as a set of (qualitative) data which enter the analysis and interpretation of corpus extracts in order to make their interpretation stronger.

Scheibman's chapter ''Subjective and intersubjective uses of generalizations in English conversations'' belongs to the weaker chapters of the book, at least with respect to data analysis. Its author makes attempts at data analysis but the result is more musing or ruminating over data extracts than analysis: namely, the author fails to show which specific features support her interpretations of the data extracts. In addition, a large part of her interpretations are based on data which are not presented in as authentic and raw of a form as possible but are merely retold, i.e. already substantially processed, interpreted, by the author. The evidence is thus unavailable for readers' inspection. Statements on pages 128-9 can serve as an example. They concern the following data extract (p. 128):

1 CAROLYN: You know, 2 -> *they come in with that .. attitude,* 3 and they go, 4 ((THUMP)) ' I've always wanted to teach math. 5 SHARON: '' stands for laughter.) Scheibman claims, for example, that ''_they_ here [in line 2] indexes a class of people with social and institutional relevance who are viewed derisively by the conversational participants,'' and that ''the simple form of _come_ in the generalization in line 2 expresses habitual meaning that generalizes over supposed specific instances, thereby intensifying the speaker's expression by portraying the behaviour she's satirizing as repetitive'' (p. 128). Although this interpretation of derisiveness and satirizing may become more convincing after inspection of the data that follow this extracted conversational sequence but were not shown; this extract, as it stands, is insufficient for making such an interpretation. Scheibman also claims that the laughter (cf. line 5) was ''a sign of recognition as well as amusement (Coates 1996)''. She uses reference to a different book (Coates 1996) rather than reference to features of her data that would support this claim. It must be stressed here that the authors of the last four chapters, in contrast, do manage to provide strong evidence for their interpretations.

Du Bois' chapter ''The stance triangle'' makes many basic things about stance explicit but, in my opinion, it unnecessarily reformulates many rather banal ideas in a complicated and formalized manner. For example, it is not difficult to maintain that stance is dialogical and intersubjective (as the chapter emphasizes, passim) using conversational (i.e. interactional and dialogical) data. The author further claims that an isolated sentence removed from its original context and transplanted into the context of the book (or in the author's inadequate wording: a ''context-free'' sentence) cannot be interpreted from the perspective of its original context from which it was removed (p. 146). This, however, is not surprising. Concerning the expression ''I (don't) know + Object'', Du Bois says that ''the point is that people do not normally present themselves as knowing (or not knowing) in the abstract. Rather, they know (or don't know) particular things [hence the Object]. Generally the precise specification of what they know, if not present in the sentence itself, is already there in the immediate prior discourse'' (p. 157). This, again, is not surprising. Other examples can follow. In sum, the author seems to reinvent the wheel, spending 43 pages on claiming that linguistic meaning is contextually bound, an idea that could have been ground-braking several decades ago but not in 2007.

Du Bois emphasizes that stance should be analyzed as a social action in its dialogical and contextual embededdness. At the same time, however, he himself does not base his claims inductively on data analysis, but uses data extracts rather as illustrations for his theorizing. Part 3.1 (p. 146-7) is an example of what the author can do without data and simply postulates ideas without showing how he arrived at them. Intersubjectivity of stance, skillfully showed in a thorough data analysis and interpretation in, e.g., the last chapter, is one of the issues that are rather postulated here.

Moreover, the attempts at data analysis in this chapter ''The stance triangle'', in contrast to other chapters, are unconvincing. Let us take, for example, extract 27 on page 150:

1 JAN; Take it downstairs. ((28 LINES OMITTED)) 30 MELISSA; I totally agree. 31 I should go downstairs.

The author comments on the extract as follows:

''In saying _I totally agree_, Melissa is not agreeing in the abstract, but specifically agreeing with her mother Jan's directive that she should take her homework downstairs. [...] Melissa makes it explicit in her next utterance (line 31), yielding a composite stance that is roughly paraphrasable as _I totally agree (that) I should go downstairs_'' (p. 150).

However, the author omitted as many as 28 lines of transcript which precede the utterance ''I totally agree''. He claims that the utterance in lines 30-31 refers to line 1, but we cannot see if this is so just on the basis of the fragmented extract. What if Melissa expressed a stance that is roughly paraphrasable rather as _I totally agree (with what you said in the 28 omitted lines, and that is why) I should go downstairs_? Similarly unconvincing is, e.g., the interpretation of extract 54 (p. 167):

1 KEN; .. I would love to go:. 2 (0.3) 3 LENORE; Yeah. 4 (1.0) 5 JOANNE; Yeah? 6 (0.9) 7 I want to go too.

Du Bois claims that ''Yeah.'' in line 3 is ''displaying her [Lenore's] convergence with Ken's stance in line 1,'' that Lenore adheres to Ken's stance. However, what if the ''Yeah.'' was a mere acknowledgement response token (in the sense of ''I can hear you, go on'')? The author further claims that the rising intonation on the other ''Yeah?'' in line 5 is ''suggesting a questioning of, or at least an ambivalence toward, Lenore's more definite _yeah_.'' However, what if this ''Yeah?'' is a response to Ken and not to Lenore and/or is just a newsmarking response token not expressing any ambivalence? (On response tokens see e.g. Gardner 2001.)

The loose interconnectedness between Du Bois' claims and the data makes some claims rather speculations. This quote from pages 160 and 161 is a striking case:

''1 ALICE; I don't know if she'd do it. 2 (0.6) 3 MARY; I don't know if she would either. [...] As similar as these two stance utterances are, there is a limit to their convergence. The second stance utterance ends with the word _either_, and this is no mere adornment. If Mary had responded to Alice's utterance with lexically identical utterance [...] the effect would likely be perceived as somewhat strange, in part because of the absence of the word _either_. The strangeness cannot be explained away as a problem with an ''echoic'' utterance: saying just _I don't know if she would_ would be pragmatically aberrant as well, in more or less the same way. [...] The word _either_ in this construction serves to index a specific intersubjective relation between two speakers engaged in dialogic interaction. While space precludes full exploration of the detailed workings of this pattern here, the evidence from many similar cases makes it clear that _either_ cannot normally be omitted from the second stance utterance without causing pragmatic anomaly (Du Bois 2004). There is no other explanation for the virtually obligatory presence of _either_ in such sentences in dialogic juxtaposition.''

It is not clear how precisely the author arrived at these conclusions, when there were only the two utterances shown (and analyzed?) without any part of the original context. It is equally unclear what is the basis for his statements about what would happen if something that did not happen happened. His own linguistic knowledge and experience seems to have entered the interpretation of the data extract, but it was not reflected upon and not exposed to the reader. Needless to say, interaction participants use their own knowledge and experience in making sense of the ongoing interaction, not the knowledge and experience of a post-hoc analyst. The analyst should draw consequences from this in his/her explanations.

However, the other chapters of the volume are of high analytical and theoretical quality, and indeed contribute to a deeper understanding of certain phenomena. I would recommend the last four chapters to those who are interested in the conversation-analytic data-driven approach; the chapter by Johnstone to those interested in the consequences of stancetaking for sociolinguistic research on the discursive construction of identity; the chapter by Englebretson to those interested in stance-in-grammar-in-use; and the chapter by Hunston to those interested in the analytical possibilities (not) provided by linguistic corpora. The fact that most of the space was devoted here to just three of the ten chapters allowed the reviewer to evaluate in more detail stances taken there towards the use of discourse data in theory building. Other readers may disagree with the reviewer's subjective stance, if they consider it worth their own stance at all. However, such is the nature of stancetaking which the book under review so clearly and nicely demonstrates.

REFERENCES Coates, J. (1996) _Women Talk: Conversation between Women Friends_. Oxford: Blackwell.

Gardner, R. (2001) _When Listeners Talk: Response tokens and listener stance_. Amsterdam/Philadelphia: John Benjamins.

Hutchby, I. and Wooffitt, R. (1998) _Conversation Analysis: Principles, practices and applications_. Cambridge: Polity Press.

Jayyusi, L. (1984) _Categorization and the Moral Order_. Boston et al.: Routledge & Kegan Paul.

Lepper, G. (2000) _Categories in Text and Talk: A practical introduction to categorization analysis_. London/Thousand Oaks/New Delhi: Sage.

Pomerantz, A. (1986) Extreme case formulations: A way of legitimizing claims. _Human Studies_ 9, 219-229.

ABOUT THE REVIEWER Marián Sloboda works currently as a researcher at the Department of Linguistics and Finno-Ugric Studies, Charles University in Prague, Czech Republic. His research interests lie in the area of interpretive sociolinguistics, discourse analysis, language management, and bilingualism.