Power and politeness in action: Disagreements in oral communication (review)

In her analysis of power and politeness in oral disagreements, Locher takes as her premise the perspective that power 'can be exercised in any interaction involving two ormore interactants' (p. 9); power is situated, and emerges, 'in and around relationships' (p. 327). Using a checklist itemizing evidence of the exercise of power in verbal interactions, she focuses particularly on the importance of contextualization and relational work in understanding and analyzing the construction and interpersonal dynamics of disagreements. Locher's study makes theoretical and empirical contributions to the current state of research on politeness and power.

Her first analysis – of a dinner-table argument that takes place among friends and family – discusses disagreement strategies in depth through a chronological examination of a 14-minute segment. Locher uses this as the foundation for her next analysis, in which she investigates power and relational work as they relate to the status, function, and position of each of 11 participants in a physics lab committee meeting. Finally, Locher examines three instances of American public institutional discourse: a critical radio interview that caught then-President Bill Clinton by surprise; a presidential debate between the two leading presidential candidates in 2000, Al Gore and George W. Bush; and segments of the Supreme Court hearing in connection with the 2000 election Florida ballot recount verdict.

One of this book's strengths is that the author provides a thorough theoretical foundation for her empirical findings, drawing on the best-known theories of politeness, which she summarizes, critiques, and then modifies in order to incorporate them into her own requirements for a clear definition and description of power and politeness in disagreements. Locher distinguishes between impolite, polite, and appropriate behaviour. She employs the surplus approach to politeness (Kasper, 1990), in which appropriate speech is viewed as unmarked behaviour described by people as 'normal' or 'okay,' while polite speech is marked, or 'more than the norm' (p. 75). Whereas polite behaviour is always appropriate, appropriate behaviour is not always polite, and overpoliteness is perceived negatively.

Locher analyzes her data quantitatively and qualitatively, through close readings of each transcribed excerpt. To her credit, she makes no attempt to generalize on the basis of her findings; she acknowledges the limitations of working with a homogeneous set of participants ('highly educated, middle class academics,' p. 324); and she recognizes herownuse of insider information and 'hunches' (p. 275) in interpreting her data. Three problematic aspects of her analysis are salient, however.

First, in employing a quantitative approach, Locher should have a way of establishing the reliability of the categories she identifies. Her taxonomy of the functions of phrases such as I don't know is complex and idiosyncratic; without inter-rater reliability, Locher's subjective categorization seems irregular and certainly not replicable. The quantity of her data is small enough and the number of categories large enough that a number of categories are represented by just one instance in the data. The data lend themselves more meaningfully to qualitative analyses.

Second, Locher carefully avoids associating her close readings with any conventionally recognized qualitative data analysis approach, but also does not offer enough information to establish her own approach as a legitimate method of analysis. Much of her close reading resembles conversation analysis, but with inconsistencies in which the author uses her insider knowledge of information not evident in the transcripts; her interpretations are not convincingly obvious enough to assure the reader that other readers would not have their own entirely different interpretations.

Third, in spite of emphasizing the importance of contextualization in analyzing her data, Locher does not provide the reader with enough information about the contexts and ignores the impact that some social factors very likely have on her findings. Other than one footnote referring to gender and language studies (p. 206), for example, Locher disregards gender as a possible (partial) explanation for the imbalance in amount and quality of talk in the dinner argument. She attributes most of...