This book, the twenty-third in the John Benjamins ‘Studies in Corpus Linguistics’ series, presents extensive analyses of patterns in university language use across ten registers. The linguistic foci include vocabulary use, lexico-grammatical variation, the expression of stance, the distribution of lexical bundles, and the co-occurrence of all of the features taken together in a multi-dimensional analysis. It provides descriptions of university language that incorporate broad frequency trends as well as detailed functional analysis. The descriptions presented would be useful for those seeking starting points for further descriptive or curriculum/materials development research.
In Chapter 1, the author introduces the book by setting the research context, identifying the book's purpose, summarising some previous research and outlining major assumptions. The findings reported in the book come from the TOEFL 2000 Spoken and Written Academic Language Project (T2K–SWAL) funded by the Educational Testing Service. The project entailed the development of a corpus representing many registers of university language, a description of that language, and the development of tools to relate the linguistic characteristics of the corpus and assessment materials to each other. While Chapter 2 is devoted to explaining the development of the corpus, the rest of the book is organised primarily by the unit of linguistic analysis (Chapters 3 to 7), and the author does not truly begin to synthesise the findings until Chapters 7 and 8. The primary purpose of the book is to describe the linguistic characteristics of, and functional relationships among, linguistic features of ten university registers. It does so by comparing the many structural features within and across registers rather than by focussing on how individual features are used within a specific register.
Chapter 2 is co-authored by Susan Conrad, Randi Reppen, Pat Byrd and Marie Helt, and describes the development of the corpus on which all of the subsequent analyses are based. In total, the T2K–SWAL Corpus contains 2,737,200 words from both written and spoken university texts. The registers included are class sessions, classroom management (spoken), labs and other in-class groups, office hours, study groups, service encounters, textbooks, course packs, course management (written) and institutional writing. The samples in the corpus were taken from both undergraduate and graduate contexts from four universities in different regions of the United States.
The linguistic analyses begin in Chapter 3, and focus on vocabulary use in two of the ten registers: classroom teaching and textbooks. These registers are described as the ‘core’ university registers. To describe vocabulary use in classroom teaching and textbooks, the author reports on how many words are used in each of the two registers. He further examines the frequency of these word types, finding that both registers use similar amounts of common vocabulary, but that textbooks use a much higher number of low-frequency, or specialised, words. By examining the parts of speech of the specialised words across registers, the author is able to describe in more detail patterns – that the author ascribes to production circumstances – related to these less commonly used words. Finally, the author finds clear trends related to the variety of words used in different disciplines and offers some related conclusions.
In Chapter 4, a variety of lexico-grammatical structures are explored. The focus of this chapter can be differentiated from the previous chapter, not so much by the objects of analyses (as word class is examined in both), but by the number of registers examined (all ten in Chapter 4). Presentation of relative frequencies provides the basis for some generalisations of how nouns and verbs are used across registers. This investigation is applied to (a) word classes across registers, (b) varieties of noun and verb phrases, and (c) semantic classes of nouns and verbs. The chapter also presents patterns of overall frequency followed by more specific functional trends related to thing, discourse connectors, dependent clauses, adverbial clauses and complement clauses across registers. The amount of data and analyses presented in this chapter is formidable, almost overwhelming.
A variety of features that mark stance in the university language of all ten registers are the focus of Chapter 5. The author begins by providing a rationale for the exploration of stance in university language and describing different ways in which stance is marked. He includes a thorough table of stance-marking features, as well as associated examples of specific stance-marking words for each feature. For each of the modal verbs, stance adverbs and stance complement clause types, the author presents frequencies and functional patterns of use across all ten registers. This chapter is rich in conclusions drawn and provides text samples supporting those conclusions for six registers.
The distribution of lexical bundles primarily in the two ‘core’ registers – classroom teaching and textbooks – is taken up in Chapter 6. The author begins with a brief description of the construct, noting that this analysis is exploratory, and uses frequency per million words and four-word bundles as criteria. The frequencies of lexical bundles by grammatical type are then reported for classroom instruction and textbooks in comparison with previous findings from conversation in a section co-authored with Susan Conrad and Viviana Cortes. Discourse functions of bundles are described. Among these functions are the use of stance bundles, discourse organising bundles and referential bundles. The chapter includes tables that present the most common lexical bundles in all of the university registers organised by function. A more detailed textual analysis by function complements the tables. Finally, the chapter addresses the use of different lexical bundle types across disciplines in textbooks. Biber concludes the chapter by noting that lexical bundles pattern differently across university registers from what would be expected on the basis of the distribution of lexico-grammatical features.
The approach used in Chapter 7 unifies some of the findings discussed in the previous four chapters. The author provides a rationale for using multi-dimensional analysis for the data presented thus far in the book. It is then reported that such an analysis of the university language using his five, well-known dimensions (Biber, 1988) found that the spoken and written registers loaded in a highly polarised fashion on the dimensions. This was unexpected, given the range of communicative purposes within both the spoken and written registers. Based on these surprising results, the author conducted a new multi-dimensional analysis, using ninety features identified as relating to university language in the previous four chapters. The methods for and results of that analysis are then introduced along with interpretations of the four resulting factors: oral v. literate discourse, procedural v. content-focussed discourse, reconstructed account of events and teacher-centred stance. Finally, the author compares the characteristics of the new dimensions specific to university language to those devised from more general English language. A principle finding of this comparison is the prevalence of stance features across all university dimensions.
Finally, Chapter 8 synthesises the vast amount of data presented in the previous five chapters. The author asserts six findings that cut across the data and then discusses each in turn. The findings are: a distinction between spoken and written language that transcends differences in communicative purposes, the prevalence of the advising and management functions across registers, the lack of academic instruction functions across registers, the uniformity of spoken registers with respect to differences in audience and interaction constraints, the prevalence of stance across registers, and interesting patterns related to academic discipline. This chapter, along with Chapter 7, is particularly helpful in creating for the reader a more global perspective based on the detailed, feature-specific analyses in Chapters 3 to 6.
All in all, the book fulfills its purpose to provide a detailed description of university language. It does so through in-depth analyses of the distribution of many features and the functions that more common distributions perform. These analyses would be particularly useful as starting points for future research on the topics of lexical bundles, stance, or grammatical complexity in university language. The final chapters bring these detailed analyses together to form a bigger picture that would be helpful to those interested in curriculum and materials development, or assessment for second language learners bound for American university contexts.
|Biber, D. 1988. Variation across Speech and Writing. Cambridge: Cambridge University Press. Crossref, Google Scholar|