This is a corpus-based study of Nigerian Pidgin English (NPE) among the educated in the urban center of Lagos. Deuber's report on the use of NPE in Lagos makes for interesting comparison with the spread and use of other contact varieties in urban areas in Africa (e.g., Sheng in Nairobi; see Fink 2005 and references therein) and its impact on indigenous languages. It is also comparable with the discussions of urban varieties of creoles reported in Patrick's (1999) work and more recently in Hackert 2004. One of the obvious issues is the functioning of the variety in new public formal domains. Issues related to both corpus and status planning are discussed (cf. Devonish 1986). To help the reader navigate the vast amount of data, there are several maps, sample questionnaires, and complete transcriptions from elicitations and interviews. The included CD contains sociodemographic data for all the informants and background information on the texts, translations, and sound samples. The book should prove beneficial to sociolinguists of varying persuasions, especially creolists, variationists, and discourse analysts.

In chap. 1, Deuber reviews the origin and development of NPE, reaffirming its birth through English contact with indigenous languages and other contact varieties like Krio. Deuber notes that NPE, an expanded pidgin, developed as a community language from the outset, a situation atypical for prototypical pidgins (Thomason & Kaufman 1988, Winford 2003). The book focuses on three areas: determining the scope of English influences on contemporary NPE as spoken by educated speakers, investigating whether a continuum situation analogous to that reported for Caribbean Creoles exists between NPE and English, and examining the linguistic consequences of expanded functions of NPE in public formal domains. Unlike that of previous studies, the data discussed in the book come mainly from the speech of educated speakers of NPE in an urban setting. Deubner uses Labovian sociolinguistic methodology, and most of the data are from questionnaires, interviews and radio broadcasts.

Chap. 2 discusses theoretical and methodological issues in language contact, variation, and change. Here NPE is referred to as a “creolizing expanded pidgin.” There is no discussion of this concept, so it is not clear whether the author is claiming a theoretical difference between “creolization” and pidgin expansion (see DeGraff 2003, Winford 2003, among others, for alternate views). In this chapter, Dueber reviews the existing body of research on NPE, which yields different claims as to the existence of an NPE continuum. To test the continuum hypothesis, Deuber mines the data for “typical” features of a mesolect. However, the author's own position on the NPE situation is not explicitly stated and the rubrics for creole continua are applied wholesale.

As with any discussion of a continuum, Dueber faces the choice between analyzing varieties of NPE as discrete systems or as variations within a single system. As such, code-switching (CS), borrowing, and interference are discussed as alternative explanations for variation. Dueber also discusses social function within the framework of di-/triglossia and the linguistic and social functions of code-switching. Dueber appears to favor an amalgam of several approaches to code-switching, as discussed by Myers-Scotton, Poplack, and Gumperz.

Chap. 3 orients the reader to the sociolinguistic background of NPE as spoken in Lagos and focuses on the status and function of the variety as reported by interviewees and published sources. The overarching sentiment is that while NPE plays an important role in Nigerian life, English is the (overt) prestige variety. This is echoed in family planning posters published by the Lagos State Ministry of Health and Social Welfare, which reflect the attitude that poor families use NPE and prosperous families use English.

Chap. 4 is meant to discuss elements of language variation and change in NPE. However, at only 11.5 pages, there is not much discussion of language change per se; the discussion focuses only on CS. CS is a function of the speaker's level of education, so four different types of NPE were identified: educated NPE (rudimentary and non-rudimentary), and less educated NPE (rudimentary and non-rudimentary). Deuber's research focuses on the non-rudimentary types.

Chap. 5 is the heart of the book and its longest, and presents details on the data collection and analysis. The analysis is based on the speech of 40 speakers aged 16 to 49 who self-report very good to moderate knowledge of NPE. There are also results of a grammaticality judgment questionnaire from 22 educated NPE speakers who did not participate in the recordings (Appendix A). The results showed that nouns, lexical verbs, and prepositions were the most frequent English elements in the texts (Figure 5.1). Deuber compares spontaneous speech among educated speakers (no English-to-NPE translations) to radio broadcasts in which translations or explanations are provided for the target audience. Interestingly, judges interpreted some of these texts as not fully satisfactory and as “artificial.” Some of English items in the NPE texts are explained as CS, but there is no detailed conversation analysis (section 5.6).

The frequency and distribution of several grammatical items were examined as well. For some of these, both frequency data and the Varbrul results are given. Hard-core variationists should be aware that the author does not claim to provide a strict variable rule analysis but instead examines the extent to which the variables are used in the data. This yielded some new, interesting results. For example, plural marking by dem was more likely to occur with [+human] nouns than [–human] nouns. Summarizing over all the grammatical variables examined, Deuber found that for each type, the forms in the texts were either fully grammatical English forms or part of the core NPE system. Further, there were no intermediate forms analogous to mesolectal-type features reported for Caribbean English creoles, and in some cases (e.g., 3sg pronoun) speakers produced forms analogous to basilectal/more conservative creole forms. Dueber's explanation is that the variation must instead be attributed to alternation between two discrete systems.

Chap. 6 presents texts and analysis of NPE spoken by less educated speakers. This includes a speaker with no formal education, one with only primary school education, and one in the process of completing high school.

Chap. 7 discusses language-planning problems in Nigeria. Three main issues are reviewed: use of the language in education, its use in the media, and official recognition. Deuber's main conclusion is that without official interest in the standardization or modernization of the variety, there is little prospect for a change in the linguistic status quo. Repeated reference is made here and throughout the text to the “broadening range of functions” of NPE, but only its use in the media (radio) is discussed at length.

Chap. 8 summarizes the main findings of the book, and Dueber reiterates her point that the observed variation between English and NPE in Lagos is not attributable to a decreolization model of a creole continuum. At the same time, there is a stable triglossic situation involving NPE, English, and indigenous languages.

One thing missing from this book is a discussion of what is expected at the different levels of the putative continuum. I think this is an essential part of the inquiry, by which the reader can judge the results presented. Further, Dueber claims that features of a putative basilect are stable for the educated speakers. I am not sure I follow this argument, since I question whether we would expect a truly basilectal variety from educated speakers. As for the acrolectal level, though this is unequivocally English, there is still need for discussion, given previous research on variation in Nigerian English itself, which is separate from the NPE variety (see discussion on p. 65). In addition, I would have liked to see a bit more discussion of the main conclusion: that the variation results from the interaction of different systems. This would have been a good place to compare the results for NPE with known cases of intersystemic variation. Finally, a lingering question for me: In a developing system, how difficult is it to tell what is borrowing, what is interference, and what is code-switching?