The editors' introduction and the first chapter by Omoniyi alone provide an engaging theoretical framework that promises a view of what speakers' social identities involve that goes beyond the standard variables in the Labovian variationist framework. Instead, Omoniyi writes of “identity options” and of speakers who coconstruct identities “in relation to evaluations of the state of affairs in terms of relationships and dispositions” (p. 20). Essentialism, defined as “labelling any number of normative characteristics or practices as constituting the core of an individual or group” is downplayed (p. 16); that is, such features as gender or ethnic group memberships assume less importance than “moments of identification” (p. 21). Block continues these arguments, referring specifically to poststructuralist views of identity.

However, very few of the following chapters realize the promise of a more dynamic perspective on identity. Ironically, many emphasize the essentialism that the editors want to leave behind, in that a number of contributors offer group members' evaluations of linguistic varieties as markers of local identity. For example, Sallabank surveys attitudes on the channel island of Guernsey toward Guernsey French as a marker of Guernsey identity. There are also two studies, based in two northeast English cities, on how residents view their local dialects as differentiating them from speakers in other cities (Llamas on Middlesborough and Burbano-Elizondo on Sunderland).

Few of the chapters are directly relevant to SLA research. I discuss briefly the three chapters of most relevance. In “English Pronunciation and Second Language Speaker Identity,” Jenkins studies nonnative speakers' (NNS) attitudes toward English pronunciation. From studying intelligibility among NNSs, she came up with a core category, lingua franca core (LFC), consisting of features that needed to be pronounced in a way close to the native-speaker (NS) target to make pronunciation intelligible to another NNS with a different first language (L1). Attitudes of respondents toward the LFC were a surprise: They were ambivalent about teaching English lingua franca (ELF). Thus, the intelligibility argument—what teaching and learning ELF would accomplish—does not get many adherents. Attachment to a NS accent remains, and it depends not so much on a question of the identity that the speaker wishes to choose but rather on beliefs about the future social and economic benefits a NS accent versus an ELF accent can promise.

“Negotiating Identities in a Multilingual Science Class” by Vann, Bruna, and Escudero does look upon identities in line with how the editors envisioned the volume. Teacher-student discourse in an Iowa science class illustrates how speakers might “attempt to claim or assign [identities] in the course of social life” (p. 203). The teacher displays an identity of solidarity with Spanish-speaking students by switching at times from her native English to Spanish. The students “construct and resist identities” (p. 207). For example, one student positions himself as more capable and worldly than his peers by telling them about the dangers of playing with a scalpel in the lab.

Spotti writes about ethnolinguistic identity among bilingual students in the Netherlands. However, the subject is attitudes about ethnic identity, not how identities are negotiated. Still, reports on how the two languages in the students' repertoire are used indicate that the students see their identities as fluid. One student says he prefers Dutch with his friends and at school but Turkish when at home with his parents. Also, using “the Dutch language not only emerges as a marker of difference from elderly family members, it also embeds a form of ‘resistance’ that these pupils may employ to challenge parental authority” (p. 193).

In sum, SLA researchers can take from this volume a notion of identity repertoires and social identities as a dynamic that is gaining adherents among some sociolinguists and anthropologists. Categorizing research subjects in the classroom in terms of only their static identities might result in an incomplete analysis.