On the basis of data collected during a year-long study of a Congressional campaign in California in the mid-1990s, this article uses semantic, pragmatic, and narrative analysis to show how candidates for political office construct and defend the coherence of their actions, including their choice to run for office. First, semantic and pragmatic analysis is used to discuss two charges of lack of coherence against one candidate. Second, three discursive strategies used by candidates for building existential coherence are identified: (i) constructing a narrative of belonging; (ii) casting the present as a natural extension of the past; and (iii) exposing potential contradictions in order to show how to solve them. After examining the extent to which each strategy is common across candidates and situations, it is shown that candidates who frame themselves as “independent” tend to use these strategies more than those who choose to identify more closely with a party's platform and ideology.The research on which this article is based was in part supported by two small grants from the University of California, Los Angeles (UCLA) in 1996–1997 and 1997–1998, and by a John Simon Guggenheim Foundation Fellowship supplemented by funds from UCLA during the 1999–2000 academic year. An earlier version of this article was presented at the Discourse Lab in the Department of Anthropology at UCLA on 2 June 2004. I thank my colleagues and students for their generous feedback and comments. Among my research assistants over the years, special thanks go to Jeff Storey, Sarah Meacham, and Jennifer Reynolds for their help in transcribing the talk in dozens of videotapes I recorded. I am also indebted to Anjali Browning for her careful reading of the first draft of this article. Some of the data and ideas presented in this article were first introduced in a number of seminars, workshops, and conferences at the University of Rome “La Sapienza,” the University of Florence, and the University of California at Santa Barbara. I would like to thank the participants in those events for their engagement with this material and their comments. I am also grateful to Jane Hill, former editor of Language in Society, and three anonymous reviewers for specific suggestions on how to improve the organization and content of the article. A number of people made the project on which this article is based possible and a rewarding experience. First and foremost, I am deeply indebted to the late Walter Capps and to his wife Lois Capps – now Rep. Lois Capps (D-California) – and to their extended family for letting me enter their home and giving me access to their lives as they experienced an extraordinary series of events. I am also very grateful to Walter's brother, Doug Capps, who was Walter's campaign manager in 1996 and has continued over the years to be my liaison with the rest of the Capps family. Others members of the Capps-for-Congress campaign staff I could rely on for information include Bryant Wieneke, always most generous with his time, Steve Boyd, Thu Fong, and Lindsey Capps. After Walter Capps's death, I benefited from conversations with Capps's colleague and friend Richard Hecht, professor and former chair of the Department of Religious Studies at UCSB. I am also grateful to the 1995–1996 Independent candidate Steven Wheeler, who, in June 1998, consented to meet with me and to being interviewed. This project was born out of conversations with Walter Capps's daughter Lisa while she was a graduate student at UCLA. She remained a strong supporter of my efforts to capture her father's adventure in politics after she accepted a position in psychology at the University of California, Berkeley, and even during the last year of her life, as she struggled with cancer. This article is dedicated to her memory.

This article links the use of you know in a group of mostly working-class London adolescents (WCG) and I know in a mostly middle-class group (MCG) to Foucault's notion of power/knowledge and, by extension, to group-specific (re)production of dominant discourses and power relations. Speakers in WCG use you know clause-finally to mark the information in that clause, which tends to be about the deviance of others, as dominant. Speakers in MCG use I know either clause-initially or as a stand-alone item in reaction to the previous speaker's utterance and to mark that utterance as dominant. Thus, the use of you know shows relatively active identity construction while the use of I know shows relatively reactive identity construction, but both groups construct their identities indirectly: WCG because they talk about others, and MCG because they react to others. Relations between these processes and class and gender identities are discussed.

This article sketches the effects of 100 years of missionary presence on how people in the San Carlos Apache community regard language and the idea of a “language expert.” Evangelical Christian practice demands an Apache language emptied of all indexical associations with non-Christian Apache cultural practices. The reservation is home to perhaps two dozen missions and churches, each of which takes a slightly different view of the role of Apache language and culture in religious practice. In an exploration of the translation practices of Phillip Goode, a San Carlos Apache interpreter, and of early Lutheran missionaries on the reservation, it is argued that Bible translation is a key factor in shifting ideas about language as a purely referential system on the reservation. This shifting language ideology has repercussions on how people in the community consider the prospects for language revitalization.This article was delivered in earlier versions to anthropology colloquia at Yale University, at Hamilton College, and at the University of Massachusetts. I thank Joseph Errington, Bernard Bate, Harold Conklin, Eleanor Nevins, Bonnie Urciuoli, Enoch Page, Jacqueline Urla, and Roy Wright, both for helping to arrange the presentations and for their generous critiques of the content. Willem de Reuse offered cogent comments and morphosyntactic advice. James Wilce gave careful comments on two versions of the manuscript. Thanks to Jane Hill and Barbara Johnstone, as well as the anonymous reviewers from this journal, for their splendid and productive advice. I also heartily thank the teachers and staff of the St. Charles Elementary School, Peridot Lutheran Day School, and Rice Elementary School for their insights and assistance in the preparation of this essay. As always I remain grateful to the Goode family for their continued friendship and encouragement. All errors and omissions remain my own.

This article examines how the official legal record, presumably an institutional space consisting of Standard American English (SAE), can become a record of a regional variety of English. Utilizing theory from language contact situations, interactional sociolinguistics, and critical discourse analysis, it describes and explains how a prestigious societal institution, often analyzed as imposing its powerful voice on those less powerful, exhibits some permeability as it absorbs at least a few discursive representations of a less dominant bilingual and bicultural group. Traces of the Spanish-English contact situation, biculturalism, and Latino life find their way into the official discursive space via stereotype, topic, lexical items, prepositions, and some verbal constructions. The discussion covers why some legal arenas are more impervious to linguistic and cultural diversity (or “accented English”) than are others. The conclusion discusses what such representations might mean for Latina women.The National Science Foundation's Law and Social Science Program (SBR#-9709938) and the Social Science Research Council's Sexuality Research Fellowship Program provided funding for data collection for this study. I would like to thank two anonymous reviewers and Barbara Johnstone for their comments and critiques on earlier versions of this article. While any remaining errors or oversights are mine alone, the reviewers' and editor's linguistic knowledge and insight have helped to make the work stronger and clearer. Also, I am grateful to Florida State University's Winthrop-King Foundation for paying Ms. Shelley Bayless to help me codify and count data. And finally, I must thank my mother, Angela M. Trinch, for being there when both of my children were born so that I could work on this article.

Discourse analysts come from a variety of intellectual traditions (systemic-functional linguistics, American descriptive linguistics, ethnomethodology, and critical theory among them) and work in a variety of ways. As a result, there is recurring discussion among practitioners about the boundaries that constitute discourse analysis as a field/theory/method in relation to other fields/theories/methods. The following reviews all take up this discussion in treatments of discourse analysis in recent textbooks and other programmatic work. Readers should note that this collection is the result of happy editorial coincidence – Ruth Wodak proposed a review article that came to press about the same time as other book reviews fitting this theme arrived at LiS. It is not the result of a systematic attempt to cover the field, which would require the kind of intellectual boundary-work these reviews report on and represent.

Barbara Johnstone, Editor

Ruth Wodak & Paul Chilton (eds.), A new agenda in (critical) discourse analysis: Theory, methodology and interdisciplinarity. Discourse Approaches to Politics, Society, and Culture, 13. Amsterdam: John Benjamins, 2005. Pp. xi, 322. Hb $138.00.

The title of this book implicitly raises a number of important questions about the relationships among discourse analysis (DA), critical discourse analysis (CDA), and interdisciplinarity: Is CDA one approach to analyzing text and talk, or is it (by implication of the ambiguous parenthetical reference in the book's title) somehow merging with (an increasingly more critical) DA? To what extent does or should (C)DA embrace interdisciplinary approaches to treating language in use? Finally (and perhaps most compelling of all), what is this new agenda, what is wrong with the old agenda, and why is a new agenda needed at this time? Overall, the book does a fair to good job of addressing these questions.

Shi-Xu, A cultural approach to discourse. Basingstoke: Palgrave Macmillan, 2005. Pp. x, 233. Hb $69.95.

This book presents Shi-xu's theoretical and methodological framework for discourse analysis, which he terms the Cultural Approach to Discourse (CAD), and critiques what he sees as the predominantly “Western” canon of social science research so far. Shi-xu argues, often convincingly and engagingly, that culture has a far more important role to play than it has hitherto enjoyed in Western approaches to discourse. He positions himself as a researcher operating from “in-between” cultures. After critiquing Western theories and methodologies of discourse research such as representationalism, universalism, and foundationalism, Shi-xu sets out the theoretical and methodological framework for CAD, and then proceeds to give practical examples of how the approach can be applied to research. Unfortunately, there are a few incongruities between the claims the author makes about his book and more generally about his approach, and what the book actually contains and what CAD is shown to have achieved. I will highlight these in the course of describing the different sections of the book. As a whole, however, A cultural approach to discourse contains much that will interest “Western” social scientists; it could serve as a guide to those who have previously ignored “culture” in their research (at least, in the author's estimation of the term), and will perhaps lead to interesting debates with those who have already incorporated some conception (perhaps an opposing one) of “culture” in their theoretical frameworks.

Norman Fairclough,Analyzing discourse: Textual analysis for social research. London: Routledge, 2003. Pp. vii + 270. Hb $135.00, Pb $31.95.

This latest book by Norman Fairclough is an extension of his earlier work on critical discourse analysis (CDA) (e.g., Fairclough 1989, 1995, 2001). Relying on systemic functional linguistics (SFL) as his linguistic theoretical standpoint on one hand, and on social theoretical themes presented by critical theorists like Bourdieu and Habermas on the other, the author attempts to present a detailed framework of linguistic analysis that links the “micro” analysis of texts to the “macro” analysis of social relations (p. 16). In his introductory chapter, Fairclough specifies two types of audience for the book who may find this framework relevant to their own research: students and researchers in social sciences and humanities with little or no knowledge of language analysis, and those specializing in language studies.

Robin Wooffitt, Conversation analysis and discourse analysis: A comparative and critical introduction. London: Sage, 2005. Pp. ix, 234. Hb $69.95, Pb $31.95.

Robin Wooffitt aims to answer a question in his new book: “Analytically, what is the best way to understand everyday communicative activities?” (2). His answer: “Conversation analysis offers the most sophisticated and robust account of language in action” (2). The remainder of the book proceeds, then, not only as an introduction to Conversation Analysis (CA) and Discourse Analysis (DA) as approaches to the study of language and communication in the social sciences (specifically, sociology and European social psychology), but also as a polemic for CA as a methodology superior to DA. The book is marketed as an introductory textbook; thus, each chapter includes periodic bulleted section summaries, and the early introductory chapters conclude with suggestions for further reading. As a textbook, this volume seems best suited for graduate seminars in linguistics or sociology; it deals with theoretical and methodological disputes that go well beyond most undergraduate students' background knowledge or pedagogical needs. As a scholarly volume, it should attract attention from social scientists already engaged in research utilizing CA or DA, as well as those working with other methodologies who are interested in how CA and DA conceptualize and investigate discourse.

Publications Received (Through 28 February 2006).