LONGMAN

Stl Gv

of SPOKEN and WRITTEN

ENGLISH Douglas Biber Susan Conrad Geoffrey Leech

Jangal

Publicati

on

Pearson Education Limited Edinburgh Gate Harlow Essex CM20 2JE England and Associated Companies throughout the World.

Visit our website: http://www.Iongman.com/dictionaries

© Pearson Education Limited 2002

The rights of Douglas Biber, Susan Conrad and Geoffrey Leech to be identified as authors of this work have been asserted by them in accordance with the Copyright, Designs and Patents Act 1988.

All rights reserved; no part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical. photocopying, recording or otherwise, without the prior written permission of the Publishers.

First published 2002 Third impression 2004 Reprinted 2013 Words that the editors have reason to believe constitute trademarks have been described as such. However, neither the presence nor the absence of such a description should be regarded as affecting the legal status of any trademark.

ISBN 0 582 237270 (cased edition) 0 582 237262 (paper edition)

Library of Congress Cataloguing-in-Publication Data A catalog record for this book is available from the Library of Congress.

British Library Cataloguing-in-Publication Data A catalogue record of this book is available from the British Library.

Designed by Sally Lace Graphs by Robert Escott Set in Minion and Frutiger by Mendip Communications Limited, Frome, Somerset Printed in China EPC/03 Ja

ngal Public

ation

Preface This book is a coursebook and a reference grammar for students and their teachers. Although it is based on the Longman Grammar of Spoken and Written English (LGSWE), the Student Grammar is much more than just an abbreviated LGSWE. Rather, we have simplified and reorganized the content of the LGSWE, avoiding much technical detail while retaining the strengths of the larger book. For example, like the LGSWE, the Student Grammar contains over 3,000 authentic English examples, and contrasts the major patterns of use in spoken and written registers. However, we have simplified the presentation when possible; for example, we have tried hard to find new and simpler extracts from our corpus data, often replacing more complex examples in LGSWE.

Of the five authors of LGSWE, we three owe a debt of gratitude to our two other co-authors: Stig Johansson and Ed Finegan. We benefited from the work that Stig and Ed did on the LGSWE during our writing of the Student Grammar. We would also like to thank the students in English 528 at Northern Arizona University, who used a draft of the Student Grammar as their grammar textbook and provided numerous helpful comments and suggestions for revision.

To support the educational use of the Student Grammar, we have also written a Workbook, which can be used as a classroom textbook or for self-study by students. The Workbook provides exercises based on authentic English texts and sentences, linked directly to the Grammar Bites in the present book.

DB SC GL

Jangal

Publicati

on

Abbreviations, symbols, and conventions Corpus examples and data All examples and text extracts are taken from the LSWE Corpus, and are marked as coming from one of the four main subdivisions of the corpus:

(ACAD) (coNY) (Ficr)

academic text

conversation transcription

fiction text

(NEws) news text

Some examples are truncated to save space. These are marked with the t symbol:

Truncated examples, showing an incomplete sentence or conversational turn, are used only when the omitted parts are judged to have no bearing on the grammatical point being illustrated, and where overly long examples might distract rather than help the reader. For example:

Ia Every atom has a dense nucleus. (ACADt) lb Every atom has a dense nucleus that contains

practically all of the mass of the atom. (ACAD) Example la is a truncated version of the complete sentence in example lb. Note that even with truncation, dispensable material is almost always omitted from the beginning or end of an example, not from the middle. In this sense, virtually every example quoted is a continuous 'slice of linguistic reality'. Occasional cases of medial omission are shown by the insertion of< . . . > at the point where the omission occurs:

< . . . > part of an example where words have been omitted

Other abbreviations used to label examples are:

ArnE American English

BrE British English

In a few cases, examples have been invented to show a contrast between accceptable and unacceptable usage. Such unacceptable examples are marked *:

* [preceding] an example which is unacceptable in English

Some examples have a following editorial comment, marked < >:

I < > an editorial comment

Conventions with i n examples Typographic features are used to highlight particular features of an example:

bold type highlights the main item for attention

underlining or brackets [ ] highlight further elements

A dash at the end of a word signals that it is incomplete.

Grammatical abbreviations

A adverbial OP object predicative

DO direct object p predicative

IO indirect object s subject

L VP long verb SP subject phrase predicative

NP noun phrase v verb (phrase)

0 object VP verb phrase

Glossary terms There is a Glossary at the end of the book, containing definitions of many grammatical terms used in the book. These terms are flagged in bold at their first instance, or at a subsequent early instance.

Corpus data Sections of text that draw most heavily on the data from the LSWE Corpus are flagged:

a section of text presenting corpus patterns

Jangal

Publicati

on

Contents Preface Ill 4.15 Indefin ite pronouns 99

Abbreviations, symbols, and 4.16 Other pronouns 101

conventions iv

Grammar Bites in each chapter viii Verbs 1 02

5.1 Introduction 103

Introduction: a discourse 5.2 Verb functions and c lasses 103

perspective on grammar 1 5·3 Semantic categories of lexica l

1.1 Introduction 2 verbs 106

1.2 A corpus-based grammar 3 5·4 The most common lexica l verbs llO

1.3 More about language variation 4 5·5 Regu lar and i rregu lar verb

1.4 More deta i l on the LSWE Corpus 7 endings ll5

1.5 Overview of the book 9 5.6 Verb formation ll8

1.6 Conventions used i n the book 10 5·7 Valency patterns ll9

5.8 Multi-word verbs: structure and

Words and word classes 1 2 mean ing 123

2.1 Introduction 13 5·9 Phrasa l verbs 127

2.2 What are words? 14 5.10 Prepositional verbs 129

2.3 Survey of lexical words 20 5·11 Phrasa l-preposit ional verbs 132

2.4 Survey of function words 26 5.12 Other m u lti-word verb

2.5 Specia l classes of words 32 constructions 134

2.6 Word-class ambiguities 35 5.13 Be 135

5.14 Have 136

5.15 Do 137

37 5.16 The copula be, and other copular

3.1 Introduction 38 verbs 140

3.2 Phrases and the i r characteristics 38 5.17 Functions of copu lar verbs 141

3·3 Types of phrases 41 U Variation in the verb 3·4 A preview of clause patterns 46

3·5 Clause elements 47 phrase: tense. aspect. voice. and modal use 1 48

Nouns. pronouns. and the 6.1 Introduction 149

simple noun phrase 55 6.2 Tense and t ime d isti nctions:

4-1 Introduction 56 s imple present and past tense 150

4·2 Mai n types of nouns 56 6.3 Perfect and progressive aspect 156

4·3 Proper nouns v. common nouns 59 6.4 Perfect aspect i n use 159

4·4 Package nouns 60 6.5 Prog ressive aspect 162

4·5 Determi ners 65 6.6 Active and passive voice 166

4.6 The articles 67 6.7 Associations between verbs and

4·7 Other determiners 72 passive voice 170

4.8 Number: s ingular and pl u ra l 78 6.8 Voice and aspect combinations 173

4-9 Case: the gen itive 79 6.9 Modals and sem i-modals 174

4.10 Gender 85 6.10 Personal and log ica l mea n i ngs of

lf.11 The formation of derived nouns 88 modals 178

4.12 Personal pronouns 93 6.11 Be going to and used to 182

4-13 Reflexive pronouns 97 6.12 Modals combined with aspect or

4-14 Demonstrative pronouns 98 voice 183

Jangal

Publicati

on

vi Contents

6.13 Seq uences of modals and semi- Complex noun phrases 263 modals 184

9.1 Introd uction 264

9.2 Survey of noun modifier types 265 Adjectives and adverbs 1 86 9·3 Noun phrases with premodifiers

7.1 Introduction 187 and postmodifiers across reg isters 267 7.2 Characteristics of adjectives 188 9·4 Types of premod ifiers 272 7·3 The formation of adjectives 190 9·5 Mean ing relationsh i ps expressed 7.4 The function of adverbs 193 by noun + noun seq uences 273 1·5 The formation of adverbs 193 9·6 Noun phrases with m u lt ip le 7.6 Semantic categories of adjectives 197 premod ifiers 276 1·1 More about attri butive adjectives 1 99 9·1 Restrictive v. non-restrictive 7.8 More about pred icative function 279

adjectives 200 9.8 Postmod ification by relative 1·9 Other syntactic roles of adjectives 202 clauses 281

7.10 Syntactic roles of adverbs 204 9·9 Postmod ification by non-fi n ite 7.11 Semantic categories of adverbs 208 clauses 29 1 7.12 Comparative and superlative 9.10 Postmodification by

forms of adjectives 215 preposit ional ph rase 294 7.13 Comparative and superlative 9·11 Postmodification by a ppositive

forms of adverbs 218 noun ph rases 297 7.14 Comparative clauses and other 9.12 Noun ph rases with m u lt ip le

degree complements 2 1 8 postmodifiers 298

9·13 Noun complement c lauses 300 G Exploring the grammar 9·14 Head nouns with noun of the clause 221 complement c lauses 302

8.1 Introduction 222

8.2 Devices of elaboration and Verb and adjective condensation 223 complement clauses 307

8.3 Subordination and dependent 10.1 Introduction 308 clauses 224 10.2 Types of complement clauses 308

8.4 Coord ination 227 10.3 Grammatica l posit ions of 8.5 E l l ipsis and structura l complement c lauses 3 1 0

condensation 230 10.4 Discourse fu nctions of that-8.6 The su bject-verb concord rule 232 clauses 3 1 2 8.7 Verbs a s operators 238 10.5 Post-pred icate that-c lauses 314 8.8 Negation 239 10.6 Verbs contro l l i ng extraposed 8.9 Major types of i ndependent that-clauses 317

clause 248 10.7 That-cla uses control led by 8.10 Declarative clauses 249 adjectives 317 8.11 Interrogative clauses 249 10.8 Discourse choices with that-8.12 Exclamative clauses 254 clauses 319 8.13 Imperative clauses 254 10.9 Structure and fu nction of wh-8.14 Non-clausa l material 255 clauses 322 8.15 Fin ite dependent clauses 256 10.10 Post-pred icate wh-clauses 8.16 Non-fin ite dependent clauses 259 control led by verbs 324 8.17 Subjunctive verbs i n dependent 10.11 Interrogative c lauses with

clauses 261 whether and if 326 8.18 Dependent clauses with no main 10.12 Overview of i nf in itive c lauses 328

clause 262

Jangal

Publicati

on

� 10.13 Post-pred icate to-clauses 11.17 Semantic categories of l i n king

control led by verbs 328 adverbia ls 389

10.14 Subject pred icative to-clauses 334 11.18 The most common l i nk ing

10.15 Post-pred icate to-clauses adverbia ls in conversation and

control led by adjectives 335 academic prose 393

10.16 Rais ing 337 11.19 Positions of l i n king adverb ia ls 394

10.17 Extraposed to-clauses 338

10.18 Rais ing v. extraposition 341 397 10.19 To-clause types across reg isters 342 12.1 398 10.20 Overview of ing-clauses 344 12.2 Word order 400 10.21 Post-pred icate ing-clauses 344 12.3 Fronti ng 400 10.22 E l l ipsis and substitution i n 12.4 Invers ion 405

complement clauses 348 12.5 Existentia l there 412 10.23 Choice of complement clause 12.6 The verb in existentia l there

type 349 clauses 414

12.7 The notional subject 414

353 12.8 Adverbia ls in existentia l c lauses 415

354 12.9 S imple v. complex existentia l

11.2 The main classes of adverbia ls 355 clauses 415

11.3 Syntactic forms of adverbia ls 358 12.10 Discourse fu nctions of existentia l

11.4 Positions o f adverbia ls 359 clauses 416

11.5 The relationship between 12.11 Dis location 418

adverbia ls and other c lause 12.12 Clefting 419

elements 361 12.13 Word order choices after the verb 423

11.6 Semantic categories of 12.14 Summary: syntactic choices i n

ci rcumstance adverbials 362 conversation v. academic prose 425

11.7 Extremely common ci rcumstance

adverbials in conversation 368

11.8 Positions of c ircumstance 427

adverbials 369 Introduction 428

11.9 Add itional semantic categories of 13.2 The d iscou rse c i rcumsta nces of

c ircumstance adverbia l c lauses 373 conversation 429

11.10 Syntactic forms of adverbia l 13·3 Performance phenomena i n

clauses 376 conversation 436

11.11 Positions of adverbia l c lauses 378 13·4 Non-clausal u n its 440

11.12 Subord i nators with c ircu mstance 13.5 E l l i psis 441

adverbia l cla uses 380 13.6 Lexical bundles in speech 443

11.13 Semantic categories of stance 13.7 Inserts 449

adverbials 382

11.14 The most common stance Glossary of terms 455

adverbials 386 A-Z list of irregular verbs 462

11.15 Positions of stance adverbia ls 387 Index 464

11.16 Other discourse functions of

stance adverbia ls 388

Jangal

Publicati

on

Grammar Bites in each chapter 2

A Introduction to words 14

B Lexical word classes 20

c Function word classes 26

3 A Introduction to phrases 38

B Clause elements and clause patterns 46

4 A Types of nouns 56

B Types of determiners 65

c Number and case in nouns 78

D Gender and noun formation 85

E Types of pronouns 92

5 A Verb functions and classes 103

B Single-word lexical verbs 106

c Lexical verbs: structures and patterns llS

D Multi-word lexical verbs 123

E Primary verbs 135

F Copular verbs 140

6 A Tense 1 50

B Aspect 156

c Voice 166

D Medals and semi-medals 174

7 A Characteristics of adjectives and

adverbs 188

B Adjectives: roles and meanings 197

c Adverbs: roles and meanings 204

D Comparative and superlative forms 215

8 A Subordination, coordination, and

ellipsis 223

B Subject-verb concord 232

c Negation 238

D Independent clauses 248

E Dependent clauses 256

9 A Types of noun modification 265

B Premodification 272

c Relative clauses 279

D Other postmodifier types 291

E Noun complement clauses 300

10 A Types and positions of complement

clauses 308

B That-clauses 312

c Wh-clauses 322

D Post-predicate infinitive clauses 328

E More on infinitive clauses 337

F /ng-clauses, ellipsis/substitution, and

review 344

11 A Overview of adverbials

B Circumstance adverbials

c Circumstance adverbials that are

clauses

D Stance adverbials

E Linking adverbials

12 A Fronting and inversion

B Existential there clauses

355

362

373

382

389

400

412

c Other topics in word order 4 1 8

13 A A functional overview of

conversational grammar 429

B Grammar tailored to real-time

construction 436

c Grammar, lexis, and discourse 443

Jangal

Publicati

on

1 Introduction: a discourse perspective on grammar Subjects in this chapter

� Introduction to the concept of corpus-based grammar � Language variation: registers, dialects, standard and non-standard

(vernacular) English �The Longman Spoken and Written English Corpus � Overview of the book � Conventions used in the book

Jangal

Publicati

on

2 Chapter 1 : Introduction: a discourse perspective on grammar

1.1 Introduction Every time we write or speak, we are faced with a large array o f choices: not only choices of what to say but of how to say it. The vocabulary and grammar that we use to communicate are influenced by a number of factors, such as the reason for the communication, the setting, the people we are addressing, and whether we are speaking or writing. Taken together, these choices give rise to systematic patterns of choice in the use of English grammar.

Traditionally, such patterns have not been included as part of grammar. Most grammars have focused on structure, describing the form and (sometimes) meaning of grammatical constructions out of context. They have not described how forms and meanings are actually used in spoken and written discourse. But for someone learning about the English language for the purposes of communication, it is the real use of the language that is important. It is not enough to study just the grammatical forms, structures, and classes. These tell us what choices are available in the grammar, but we also need to understand how these choices are used to create discourse in different situations.

The year 1999 saw the publication of a large-scale grammar of English with the aim of meeting the above needs: the Longman Grammar of Spoken and Written English (LGSWE) . This was the outcome of an international research project which lasted seven years. Now, we have written a revised, simplified, and shortened version of LGSWE for use by advanced students and their teachers.

LGSWE made important innovations in the method of grammatical study. It was based on a large, balanced corpus of spoken and written texts. These texts were electronically stored and analyzed with the aid of computers. The analysis produced information about the frequency of grammatical features in different kinds of language. (We use the term 'feature' broadly in this book, to refer to any grammatical form, structure, class, or rule. ) The results of the analysis were then studied by the team of grammarians. The goal was to explain not just what is possible in English grammar, but what is more or less probable in different situations.

This book, the Student Grammar of Spoken and Written English (SGSWE) , presents the insights and discoveries of LGSWE to advanced students of English and their teachers. SGSWE is designed to be used with the accompanying workbook as a textbook (for use in class or for self-study) , or alternatively as a reference grammar: • SGSWE is a pedagogical textbook on grammar. For students systematically

studying English grammar, SGSWE begins with the 'basics' in the opening two chapters, and moves progressively into more advanced territory. It ends with a chapter devoted to the special characteristics of conversational grammar-an important topic which has generally been neglected in grammars up to now.

• SGSWE can also be used as a reference grammar. It covers all major features, structures, and classes of English grammar, together with their meaning and use. With the help of the glossary of grammatical terms and the index, students and teachers can use this book as a reference guide. For more detailed information about a feature or its use, you can consult corresponding sections in the larger LGSWE.

Jangal

Publicati

on

A corpus-based grammar� The greatest innovation of SGSWE is that it is a corpus-based grammar. The

new methodology of large-scale corpus study developed in LGSWE produces results which are very useful for students. In general, advanced students of English want to understand not only the structural rules of English, but how the language is used for communication. For this purpose, new insights, explanations, and information from corpus-based studies of English are an important advantage.

1.2 A corpus-based grammar A corpus is a large, systematic collection of texts stored on computer. The corpus used for LGSWE and SGSWE-the Longman Spoken and Written English Corpus (the LSWE Corpus)-contains approximately 40 million words of text, providing a sound basis for the analysis of grammatical patterns. Because the SGSWE is based on analysis of this corpus, it offers a number of advantages over traditional grammars: • Real examples: The book contains over three thousand examples of English

in use. These are authentic examples from the corpus, showing how real people use real language. Invented examples that sound artificial-a familiar feature of many other grammars-are entirely absent. At the same time, we have taken care to avoid corpus examples that are overly difficult because they require understanding a complicated context.

• Coverage of language variation: The core corpus we have used for this book represents four major registers of the language: conversation, fiction writing, news writing, and academic prose. In the past, grammars have usually presented a single view of the language, as if the grammar of English were one fixed and unchanging system. This clearly is not true. Although there is an underlying system of grammar, speakers and writers exploit that system very differently to meet their communication needs in different circumstances. The corpus-based information included in SGSWE describes differences between a spoken register (conversation) and three written registers (fiction, newspapers, and academic prose) . In addition, our corpus contains data from both American and British conversations and newspapers. Where they occur, important differences between these geographical dialects are pointed out. (See 1 .3 for more about registers and dialects. )

• Coverage of preference and frequency: This grammar gives information about the preferences speakers or writers have for one grammatical choice over another. Specifically, we discuss the frequency of alternative structures and the conditions that are associated with them. This is clearly a major matter of interest to advanced learners of the language and future teachers of the language. For example, it is not enough simply to describe the structural differences between active and passive constructions, because students and teachers need to know how users choose between these two options. For example, passive verbs are ten times more frequent in academic prose than they are in conversation. And there are good reasons for this, which we discuss in 6.6. 1 .

• Interpretations of frequency: context and discourse: Information about frequency needs to be explained by human interpreters. That is, a corpus

Jangal

Publicati

on

4 Chapter 1 : Introduction: a discourse perspective on grammar

grammar needs to present the evidence of the corpus as a means of exploring why users of English make one choice in one situation, and a different choice in another. Usually several factors are relevant, including register, expressing personal attitudes, giving informational emphasis, or other more specific contextual conditions. The SGSWE gives attention to the conditions under which grammatical choices are made, so the grammar has a discourse orientation.

• Lexico-grammatical patterns: Another distinctive feature of the SGSWE is that it brings together the study of grammar and vocabulary. Traditionally, both in theory and pedagogical practice, grammar has been separate from vocabulary, as if they were two totally independent aspects of language and language learning. This separation is artificial, as becomes evident to anyone who uses a large corpus for studying grammar. What becomes clear is that, when they use a language, people bring together their knowledge of word behavior (lexis) with their knowledge of grammatical patterns. These two aspects of language interact in lexico-grammatical patterns. For example, there is one set of verbs that commonly occur with a that-clause (e.g. think, say, know) and a different set of verbs that commonly occur with a to-clause (e.g. want, like, seem) . In addition, each register prefers different verbs with these clause types. These patterns help to explain the typical meanings and uses of each clause type in each register.

1.3 More about language variation We have mentioned the coverage o f language variation as one o f the strengths of this grammar. It is important to recognize that there are two major types of language varieties: registers and dialects.

1.3.1 Reg isters

Registers are varieties of language that are associated with different circumstances and purposes. For example, Table 1 . 1 compares the circumstances and purposes of the four registers compared in SGSWE.

Comparisons between registers can be made on many different levels. For example, the most general distinction can be made on the basis of mode: conversation is spoken, while the other three registers are written. If we consider a more specific characteristic-the main communicative purpose-the registers fall into three categories. Conversation focuses on personal communication,

Table 1 . 1 Circumstances of the four main registers in SGSWE CONV FICT NEWS ACAD

mode spoken written written written

interactiveness and yes restricted to no no rea l-time production fictiona l d ia log

. . . . . . . . . . . . . . .

shared situation yes no no no

main communicative persona l pleasure reading i nformation/ a rg u mentation/ purpose/content communication eva l uation explanation

audience i ndividual wide-publ ic wide-publ ic specia l i st

Jangal

Publicati

on

More about language variation� fiction on pleasure reading, and newspapers and academic prose share a more informational purpose. (We sometimes call these last two expository registers. ) However, even between newspapers and academic prose there are differences. Academic prose is more concerned with building an argument than newspaper writing is. What is more, academic prose has a more specialized audience than a newspaper, which is written with a wide audience in mind.

Register variation is the main aspect of language variation we consider in this book. We choose this focus because registers differ greatly in their grammar usage, reflecting their different communication circumstances. In other words, the circumstances of a register have a direct impact on which grammar habits are common in the register. To take an easy example, consider how pronouns and nouns are used in conversation and news. Conversation is interactive as a form of personal communication. It is not surprising, then, that conversation shows a frequent use of the first-person pronouns I and we, and the second-person pronoun you. In contrast, newspaper writing is not directly interactive: it is not addressed to an individual reader, and it often does not have a stated author. But it has a function of conveying general information of current interest. Not surprisingly, in newspaper texts, first- and second-person pronouns are relatively rare. Instead, proper nouns, referring to people, places, and institutions, are particularly common.

The four registers that we concentrate on are, of course, far from a complete picture of register variation in English. However, they have the advantage of (a) being major registers, likely to be frequently encountered by any advanced student of the language, and (b) being sufficiently different from one another to show important differences as well as important similarities.

Within each register there are also sub-registers. For example, fiction can be broken down into different sub-categories-detective fiction, fantasy fiction, romance fiction, etc. The compilers of the LSWE Corpus took pains to represent all the categories and to obtain as good a balance as possible between different sub-varieties within each category. (More details are given in 1 .4 below. ) In order to limit the size of this book, though, we make only a few references to sub­registers, such as commenting on a feature common in, say, sports reporting or detective fiction.

1.3.2 Dialects

Dialect variation interacts with register variation. Dialects are varieties according to the identity of speaker(s) or writer(s)-their geographic area, gender, socio­economic class, and so on. Dialect is less important for grammatical purposes than register. From the grammatical point of view, dialect differences are arbitrary, while register differences are functional, reflecting the way that grammar varies according to communicative purpose.

Like register, dialect distinctions can be made at different levels of specificity. 'American English' or 'British English' is a very general level; 'the speech of female teenagers in the South Bronx area of New York City' would be a more specific dialect. In the SGSWE we focus only on the high-level distinction between American English and British English (although the spoken corpus was developed following sampling methods for geographical regions, socio-economic class, gender, and age, so more detailed dialect studies are possible) .

Jangal

Publicati

on

6 Chapter 1 : Introduction: a discourse perspective on grammar

The corpus used for this book also includes some representation of Australian, Canadian, Caribbean, Irish, and West African English in the fiction subcorpus (see details in 1 .4) . For full coverage of dialect variation, we would need to have coverage of many more world varieties of English. The contrasts between British and American English in this book serve as just one example of regional dialect variation. We have chosen these two dialects because they serve as a target for many learners and teachers of English.

1.3.3 Standard and non-standard {vernacu lar)

Engl ish

There is no official academy that regulates usage for the English language, but there is still a prevailing world-wide view that there is a 'standard English': the language variety that has been codified in dictionaries, grammars, and usage handbooks. However, in the corpus, especially in conversation, usage regarded as non-standard (also called the vernacular) is also found. For example, consider the vernacular use of what as a relative pronoun in 1, and of ain 't and multiple negative constructions in 2 and 3:

1 They were by the pub what we stayed in. (CONY) 2 I ain't done nothing. (coNY) 3 'There ain't nothing we can do. ' (FICT)

As 3 shows, vernacular forms also occur in fiction texts, particularly in representing the speech of fictional characters. However, in general, vernacular forms are rare in the written corpus.

When vernacular forms are discussed or illustrated in the rest of the book, they are generally noted as 'non-standard' . In general, our description of grammar is limited to standard English. However, it is worth bearing in mind that conversation contains many vernacular features of language, and so such features can be expected in the conversation part of our corpus, whereas they are strictly avoided in written language (except where written language deliberately mimics speech, as in fictional dialog) . In a similar way, conversational transcriptions often contain informal or non-standard spellings like gonna and cos (see 1 .6.2 below) .

A Variation in standard Engl ish

The term 'standard English' can be misleading. 'Standard' in some contexts means 'uniform, unchanging', and so it is assumed that only one form of a grammatical feature is accepted in standard English. But this is clearly false. For example, both of the following conform to standard English, even though one begins a relative clause with that and the other with which:

1 I could give you figures that would shock you. (FICT) 2 This chapter is devoted to a discussion of various flow processes which

occur in open systems. (ACAD) The relative pronouns that and which could exchange places in these sentences, although there are a number of factors that favor one over the other (see 9.7, 9.8. 1 ) .

Little of the variability within standard English i s due to dialect differences between American English and British English, which actually show very few

Jangal

Publicati

on

More deta i l on the LSWE Corpus�

grammatical differences in their standard dialects. We point out such differences where important in SGSWE. For example, one well-known difference is that American English has two past participles for the verb get (got and gotten) , whereas British English has only one:

Angie, I think we've got a leak. (BrE CONY)

They've got money. (ArnE coNY)

He must have got to the door just as the bomb landed. (BrE coNYt)

And we still haven't gotten a damn pumpkin. (ArnE coNY)

Note: Another aspect of vernacular English is illustrated by the last example: the word damn (or damned) , which is a mild 'swearword' or taboo term. Swearwords can cause offence or be considered impolite, especially where used in the wrong context, and so we point them out when they occur in examples.

B Prescriptive v. descriptive g rammars

Most cases of variation within standard English (e.g. that and which in I and 2 above) do not attract attention from ordinary language users. However, speakers do tend to be aware of some aspects of disputed usage and sometimes have strong opinions about what forms are 'correct'. Thus while the use or the omission of that is rarely noticed or commented on, the choice between who and whom can rouse strong feelings (see 9.8 .2 ) . Prescriptive grammars dictate how people 'should' use the language. For example, a prescriptive grammarian would insist that only whom should be used when the pronoun refers to a human and functions as an object or prepositional complement. In contrast, speakers in conversation regularly prefer who in actual usage:

There's a girl who I work with who's pregnant. (coNY)

In fact, many speakers would find the use of whom unusual in any informal, conversational situation.

In this grammar, we do not argue that any one alternative is correct in cases like these. Rather than a prescriptive grammar, the SGSWE is a descriptive grammar. We focus on describing the actual patterns of use and the possible reasons for those patterns. However, we do refer from time to time to some cases of disputed usage. Although these may not be so important from the viewpoint of communication, they often play a significant role in people's judgments of what is 'good grammar'. These judgments, in turn, may have an influence on actual patterns of use.

1.4 More deta i l on the LSWE Corpus For a corpus-based grammar, the design of the corpus i s an important concern. Detailed discussion of the LSWE Corpus can be found in Chapter 1 of LGSWE, especially pp. 24-35. Here we give only a brief overview of the corpus.

The entire corpus contains approximately 40 million words. Most of the analyses comparing the four registers used a subcorpus of approximately 20 million words. Additional texts for the dialect comparisons and occasional comparisons with supplementary registers account for a further 20 million words. Table 1 .2 provides an overview of the overall composition of the corpus.

Jangal

Publicati

on

8 Chapter 1 : Introduction: a discourse perspective on grammar

Table 1 .2 Overall composition of the LSWE Corpus

core registers

conversation (BrE)

fiction (ArnE & BrE)

news (BrE)

academic prose (ArnE & BrE)

ArnE texts for dialect comparisons

conversation (ArnE)

news (ArnE)

supplementary registers

non-conversational speech (BrE)

general prose (ArnE & BrE)

total Corpus

number of texts

3,436

1 39

20,395

408

329

1 1 ,602

751

1 84

37,244

number of words

3,929, 500

4,980,000

5,432,800

5,331 ,800

2.480,800

5,246, 500

5,7 1 9, 500

6,904,800

40,025,700

The strength of the LSWE Corpus does not just lie in its size-although size can be important, especially for the study of rare grammatical features. More important qualities are the diversity and balance of the corpus. The LSWE Corpus represents a comparatively wide range of register and dialect variation within the language, and each category of texts is represented by a wide range of writers/speakers and 'sub-registers'. In 1 .3 . 1 and 1 .3.2 we mentioned some of the sub-categories within the registers and dialects of the corpus. Tables 1 . 3-6 provide further details.

Like the written registers, conversation is also a diverse register, but no effort was made to identify sub-registers or list all the topics of conversation. Most of the conversations in the LSWE Corpus are private (often domestic) talk. However, occasionally other kinds of talk are included, like service encounters in a store, or one side of a telephone call. Planned speech, such as lectures, speeches, and sermons, are in a separate register of 'non-conversational speech'.

Table 1 .3 Approximate numbers of

gender

female

male

speakers in the BrE and ArnE

conversation subcorpora by

gender

BrE

270 225

AmE

292 1 99

Table 1 .4 Distribution of fiction texts across national varieties

national variety

AmE

BrE

other

number of texts

41 79 1 9

number of words

1 ,095,200 3,347, 1 00

537,700

Jangal

Publicati

on

Overview of the book� Table 1 .5 Breakdown of the British and American news subcorpora by topic

topic BrE ArnE

arts/entertai nment 41 8,400 325,00 business/economics 542,800 1 ,545,000 domestidloca l/city news 1 ,233,900 995,000 foreig n/world news 1 , 1 56,1 00 680,000 sports 1 ,2 1 8,700 260,000 a l l other topics ( inc luding ed itoria l s, law, 862,500 1 ,485,000 social news, science/med ic ine/technology, etc.)

Table 1 .6 Breakdown of the academic prose subcorpus

major categories number of texts number of words

academic book extracts

academic research articles

75 333

2,655,000 2,676,800

Note: Subjects i ncluded are agricu lture, anthropology/archaeology, biology/ecology, chemistry/physics, computing, education, engineering/technology, geology/geography, law/history/politics, l i ngu istics/l iterature, mathematics, medici ne, n u rs ing, psychology, sociology.

1.5 Overview of the book We have organized our discussion of grammar into several major sections: • Chapters 2-3 Key concepts and categories in English grammar

These chapters present a basic introduction to English grammar, providing the foundation for our discussion of particular areas in later chapters. They introduce the basic terms for structures, rules, and classes in English grammar, illustrating them throughout with real corpus examples. It may be useful to refer back to these chapters (as well as the glossary) if you come across a puzzling term later in the grammar. Because they review all of grammar in a simplified way, these chapters have less room for the information about discourse choice found more plentifully in later chapters.

• Chapters 4-7 A close look at the major phrase types Chapters 4-7 cover the major classes of 'content words' or lexical words (noun, verbs, adjectives, and adverbs) together with the related phrases (noun phrases, verb phrases, adjective phrases, and adverb phrases) : 4 Nouns, pronouns, and the simple noun phrase 5 Verbs 6 Variation in the verb phrase: tense, aspect, voice, modality 7 Adjectives and adverbs

• Chapter 8 Clause grammar This chapter introduces the structure and function of independent and dependent clauses. It is a pivotal chapter roughly half way through the book. It looks back to the phrase chapters, Chapters 4-7, showing how those phrase types are used in clauses. At the same time, it looks forward to the later clause

Jangal

Publicati

on

10 Chapter 1 : Introduction: a discourse perspective on grammar

chapters, Chapters 9-12, which describe the use of clauses in more complex constructions.

• Chapters 9-12 Building on the clause Building on Chapter 8, these chapters explore additional clause-based or clause-derived structures. For example, complex noun phrases (Chapter 9) may not seem to involve clauses, yet in fact they do, because noun phrases can contain relative clauses and other clause-level constructions. These constructions cannot be explained properly until clauses have been covered. 9 Complex noun phrases

10 Verb and adjective complement clauses 1 1 Adverbials 12 Word order choices

• Chapter 13 The grammar of conversation Chapter 1 3 places the spotlight firmly on spoken language, and shows how the grammar of conversation is adapted to the particular demands of spontaneous spoken interaction. It also highlights some special features of conversational grammar: for example, dysfluencies, discourse markers such as well and okay, ellipsis, interjections, and taboo expressions ( 'swearwords' ) .

• Reference section The book ends with a glossary of grammatical terms and an index. The reference section of the book is designed for all users, but especially to help teachers and students using it as a reference grammar.

1.6 Conventions used i n the book The list of Abbreviations, Symbols and Conventions on p. iv covers many of the conventions used in this book. But some aspects of the transcription of conversation and the use of tables and figures deserve a more detailed explanation here.

1.6.1 Transcri ption of speech

Spoken language must be transcribed before it can be studied. That is, the transcriber must listen to a tape recording and write down exactly what was said. For the LSWE Corpus, the transcribers produced an orthographic transcription. This transcription uses the ordinary symbols of written texts, including the conventional spellings of words (in most cases) . Conventional punctuation symbols-particularly hyphens, periods, commas, and question marks-are used to reflect typical intonation associated with those symbols. Thus, a period reflects falling intonation and a question mark reflects rising intonation.

1.6.2 Spel l ing variations, reduced pronunciation,

and l imitations of the transcription

You may sometimes notice variants which are irrelevant to the study of grammar, such as variant spellings like OK and okay, or American spellings (such as center) in some texts and British spellings (such as centre) in others. These differences are not significant, and can be ignored from the grammatical point of

Jangal

Publicati

on

Conventions used in the book� view. More relevant linguistically are spellings of reduced pronunciations, such as gonna, gotta, and wanna instead of going to, got to, and want to, and cos or cause instead of because. These are semi-conventional spellings, capturing the reduced pronunciations which are very common in casual conversation. They help give an impression of what the speech was like. You will also notice 'words' like urn and er signaling filled pauses (a type of dysfluency-see 1 3 .2 .5) and exclamatory words like hm and ooh which it would be rare to find in ordinary written texts. However, a strict, phonetically accurate transcription was not the goal of the corpus, nor would a phonetically detailed transcription be feasible with a corpus of this size.

Since the corpus was not transcribed phonetically, some features of speech, such as stress and intonation, are not available. In the vast majority of cases, however, the transcription provides plenty of detail for grammatical analyses.

1.6.3 Visible frequency: the use of tables and figures

We have said that frequency is important for understanding how the grammar of English is actually used in different registers, dialects, and situations. At the same time, we recognize that tables of statistics are often not useful to the average grammar student. To overcome this problem, we have used two main ways of representing frequency, both of them avoiding the use of lists of numbers: • The first method of indicating frequency is to use ordinary words such as

often, rarely, common, uncommon. While not precise, these terms are useful in giving a general idea of frequency differences. These generalizations are based on corpus analysis, which is often reported in LGSWE with tables and figures.

• The second method is to use figures (bar graphs) . These figures enable you to compare frequencies in an immediately visible way, by looking at the length of the bars being compared. In these figures, the registers are always presented in the order conversation, fiction, news writing, academic writing. For an example, see Figure 2. 1 on p. 23.

To make comparisons easier in the figures, frequencies are normed to the standard measure of ' occurrences per million words' . Thus, although the sizes of the registers of the corpus are somewhat different, the comparisons are based on a standard measurement of relative frequency. It may be difficult to envision a million words, so (although there is great diversity in print size and rate of speech) handy comparative measures are as follows: • Books average about 350 words/page, so one million words = about 3,000

pages. • Speakers average about 120 words/minute, so one million words= about 140

hours.

Jangal

Publicati

on

2 Words and word classes

GRAMMAR BITES in this chapter

A Introduction to words

� Lexical words, function words, and inserts � The structure of words: morphology � Multi-word units: idioms, collocations, and lexical bundles

B Lexical word classes

� The structure and function of lexical words: nouns, verbs, adjectives, and adverbs

� Comparing lexical word classes in use � Borderline cases in classifying words

c Function word classes

� The structure and function of function words: determiners, pronouns, auxiliary verbs, prepositions, adverbial particles, coordinators, and subordinators

� Special classes of words

Jangal

Publicati

on