Vol. 4. No. 3 R-14 May 2000

Return to Table of Contents Return to Main Page

In Longman Grammar of Spoken and Written English, Biber, Johansson, Leech, Conrad, & Finegan present an important, corpus-based description of form and function of grammar in use for four representative registers for both British and American English. This massive reference book has many new findings, some of which will challenge conventional thinking about what grammar is and how it is used for many years to come.

This large-scale grammar owes much of its terminology and grammatical framework to A Comprehensive Grammar of the English Language (Quirk, Greenbaum, Leech, & Svartvik, 1985). Those readers familiar with that grammar will have few problems with the terminology or concepts in this grammatical analysis. However, it is in the exemplification and quantitative analysis of grammar across the varieties of spoken and written English that this grammar is more comprehensive than the earlier one. In spite of this, the authors look at this grammar as a companion to the earlier volume.

This book uses a computer-aided, corpus-based approach to look at the use of grammatical features in four registers (conversation, fiction, news, and academic prose) from American English (AmE) and British English (BrE). Each of the core registers consists of approximately 5 million words. The core conversation corpus is from BrE, fiction from AmE and BrE, news from BrE, and academic prose from BrE and AmE. In addition to the four registers, the full corpus includes AmE texts for conversation and news for dialect comparisons and two supplementary registers: non-conversational speech (BrE) and general prose (AmE and BrE). The two supplementary registers are used for two kinds of analyses: for the overall findings from the complete corpus, and for a few analyses that specifically target one or the other of these registers. The total corpus has over 40 million words. The majority of the texts were spoken or written after 1980. All of the findings are normed to frequency of occurrences per 1 million words.

Although this analysis was done using computers and large electronic corpora, the process was not fully automated. It has human analysis at every step of the process, from deciding on the 5-10% of word forms not recognized by a tagging program that used dictionaries, probabilistic information, and grammar rules, to analysis of the meaning distinctions, and to the functional interpretation of quantitative findings. Checking the texts to ensure that the automatic analysis was not skipping occurrences of the target feature was done manually, as well as other tasks. [-1-]

The book is divided at its beginning into a foreword, a very useful abbreviations and symbols section, and a preface. There is also a contents summary which is very useful for finding specific grammatical forms, such as types of phrases, ellipsis, tense, post-modification by finite relative clause, stance adverbials, lexical bundles, and so on. After the general contents summary section there is a contents in detail section, which gives the details of each grammatical topic very well. Next there is a symbols and notational conventions section, which gives the symbols and notational conventions used in the book in a concise and well exemplified fashion.

The actual grammar analysis consists of five major sections, A through E, and is composed of fourteen chapters.

Section A consists of one chapter and is the introduction to the book and the corpus-based approach to English grammar. In this section, the authors discuss the parameters of the many choices they had to make in completing this project. The introduction discusses structure and use in English, varieties of English, representativeness of varieties in the corpus, grammatical analysis in the corpus, quantitative findings, and functional interpretation of the quantitative findings. It also gives an overview of the grammar and lists potential users and uses of the book. Those benefited include English language teachers, especially ESL/EFL teachers, English language students and language researchers, and also researchers and practitioners in 12 other sub-disciplines.

In each part of this section, and in the whole of the book, the authors are concerned that readers understand the impact of their approach and findings. Concepts such as register, grammatical feature, dialects, varieties, and many others are well explained. Also, a table on page 39 explains occurrences per million (which is what the grammatical features for each register are normed to); that is, how often 10, 20, 40, 100, 200, and 1000 occurrences per million happen in minutes for spoken English or pages for written English. This is done succinctly and clearly, so this conceptually rather difficult idea is easily understood.

In the following sections, B through E (chapters 2-14), the analysis of English is done. The format for this is similar throughout the book. First the targeted feature is introduced and its form and structure are discussed. Then corpus findings, frequencies, and a discussion of findings are given in varying lengths, depending on the characteristics of the targeted feature. The distribution of targeted features is regularly exemplified throughout the text using graphs comparing frequencies in the four registers. The corpus findings section includes important findings about the targeted feature from a comparison of registers. The discussion of findings consists of functional interpretations of the targeted feature's use in a register--the why and the how. These last sections are what make this innovative grammar book different from all others.

To make this format clearer, an example is taken from Section B (coordination of phrases, topic 2.9, pp. 113-117). First its basic forms are given (noun, verb, adjective, and adverb phrases or parts of phrases). Unique problems (in this case, ambiguity) and other uses of the targeted feature (phrasal versus clausal coordination and coordination tags) are discussed. Then the corpus findings are given (in this case, simply, "Most coordination tags are particularly frequent in conversation" (p. 116)), along with a chart of the most frequent coordination tags in conversation and a table which represents the frequency of other coordination tags across registers and occurrences per million words. Finally, a discussion of the targeted feature's findings is given. [-2-] One example from this topic was the coordination tag and so on, which is commonly used in writing to signal that a list is incomplete. The functional explanation includes the reason for its use: users of and so on desire to indicate that they are being brief or are not sure of the complete listing. This functional description is highly useful for teachers' lesson planning or materials development, and to answer possible student questions.

Section B is comprised of chapter 2 and 3. In chapter 2, such general concepts as the nature of grammatical units, word types and word tokens, three major word classes, and multiple class membership are introduced and explained quite clearly. More detailed analysis is then done into the types and characteristics of lexical words, function words, and inserts. Following a general introduction of phrases, they are then analyzed by types, embedding, coordination, and simple versus complex. All of the analysis is done with regular exemplification. In chapter 3, clause constructions, functions, and use are analyzed. Clause links, elements, patterns and their variations, ellipsis, and dependant clauses are examined. Negation, subject-verb concord, and non-clausal material in writing and conversation are also analyzed.

Section C contains a detailed look at the major word classes and phrases (nouns, verbs, adjectives. adverbs, and their phrases). This section is made up of chapters 4 through 7.

Section D (chapters 8 through 10) contains analyses of more complex structures: complex noun structures, adjective clauses, complements, and adverbials. Section E, comprised of chapters 11 through 14, includes topics looking outward from grammar to larger language issues. Topics such as word order, grammar and stance, lexical expressions in speech and writing, and the grammar of conversation are discussed.

I don't want to lessen the depth and comprehensiveness that sections B through E go into by the brevity of the previous overview of the general topics. The grammar is very comprehensive and goes into great depths of analysis and description of each targeted feature, with new insights in every analysis.

At the end of the book, there is an extensive appendix of contractions in English, endnotes, and a thorough bibliography. In order to find many lexical items and how they are realized in a targeted grammatical structure quickly, the book has a lexical index. If one has the concept for the targeted grammatical item, the book also has a conceptual index. Both of these are very complete, and are useful for finding a particular word or concept quickly, without having to thumb through the book looking for it.

This is an important book about the English language, and it will continue to influence our views of grammar for a long time. This is especially true for English teachers; this grammar should be in every personal and program library. In each part of each section in the whole book, one can uncover empirically mined, not intuitively surmised, gems concerning English. For example, in section A, topic 1.4 (representation of varieties of English), the authors report that one of the important general findings of the project was that ". . . many overall descriptions of general English are incomplete and can be even misleading or inaccurate. Descriptions of general English, based on an averaging of patterns across texts and registers, often obscure important differences and usually do not, in fact, represent the patterns of any register consistently" (p. 24). [-3-]

The potentially most far-reaching chapter of this book is chapter 14. This chapter offers a new look at the grammar of conversation. It shakes, crumbles, and reconstructs the use of the sentence as the foundation for conversational grammar. The authors say that the sentence is not the basis of spoken grammar because "such a unit does not realistically exist in conversational language" because "conversation has no generally recognizable sentence-delimiting marks such as the initial capital and final period of written language" (p. 1039). What the authors propose for a grammar of conversation is C-units and the principles which govern conversation. C-units consist of clausal and non-clausal units. The clausal unit (corresponding to the t-unit of written language) consists of the independent clause and any dependant clauses attached in it. The non-clausal unit consists of segments that are not clausal units or part of clausal units. From the corpus findings of this section, one discovers that non-clausal units account for over one-third of the units of conversation, but their average length is only two words. The constraints on conversation take into consideration the fact that conversation happens in real time and is subject to the participants' working memory. The three principles governing conversation are that participants keep talking, there is limited planning ahead, and what has been said can be qualified. This new view of spoken English should change how ESL/EFL conversation textbooks and classes are written and implemented.

No book is without blemishes, although this one has few. Foremost is its price. Although the book was printed in China, presumably to keep its price low, it still costs $119, a tidy sum for ESL/EFL teachers or trainees, although rather normal for chemistry or physics teachers or students. Unfortunately, this will limit its audience among the people who can use and implement the findings the fastest. Another weakness is that sometimes its explanations make for difficult reading. However, this is true of all grammars. A third may be the gray-scale graphs depicting the frequencies and distributions among features. If one is looking for clear illustrations rather that approximate ratios, then one will dislike some of them, especially the tri-gray scaled ones. These are sometimes difficult to discern. If one only wants to see the approximate ratios, they are adequate.

These are very superficial problems when one considers the wealth of information about the English language that this book provides. For a teacher who wants a definitive answer to a particularly evasive grammar point, this book will give form, function, and use in a very accessible manner. My advice is to buy it quickly before the first edition sells out and another more expensive edition is printed.

Reference

Quirk, R., Greenbaum, S., Leech, S., & Svartvik, J. (1985). A comprehensive grammar of the English language. London: Longman.

Jim Bame
Utah State University
<fabame@cc.usu.edu>