Adverbial Clauses in Scots: A Semantic-Syntactic Study 9783110803822, 9783110156430

Citation preview

Adverbial Clauses in Scots

1749

1999

Topics in English Linguistics 27 Editors

Bernd Kortmann Elizabeth Closs Traugott

Mouton de Gruyter Berlin · New York

Adverbial Clauses in Scots A Semantic-Syntactic Study

by Martina Hacker

W DE

G

Mouton de Gruyter Berlin · New York 1999

Mouton de Gruyter (formerly Mouton, The Hague) is a Division of Walter de Gruyter & Co., Berlin.

© Printed on acid-free paper which falls within the guidelines of the ANSI to ensure permanence and durability. Library of Congress Cataloging-in-Publication-Data Hacker, Martina, 1956Adverbial clauses in Scots : a semantic-syntactic study / by Martina Hacker. p. cm. - (Topics in English linguistics ; 27) "Revised and much extended version of a doctoral dissertation submitted to the University of Freiburg in 1995" Acknowledgements. Includes bibliographical references and indexes. ISBN 3-11-015643-1 (alk. paper) 1. Scots language-Adverbiale. 2. Scots language—Semantics. 3. Scots language-Clauses. 4. Scots languageSyntax. I. Title. II. Series. PE2104.H33 1998 427'.9411-dc21 98-42857 CIP

Die Deutsche Bibliothek - Cataloging-in-Publication-Data Hacker, Martina: Adverbial clauses in scots: a semantic syntactic study / by Martina Hacker. - Berlin ; New York : Mouton de Gruyter, 1999 (Topics in English linguistics ; 27) Zugl.: Freiburg (Breisgau), Univ., veränd. Diss., 1995 ISBN 3-11-015643-1

© Copyright 1998 by Walter de Gruyter GmbH & Co., D-10785 Berlin All rights reserved, including those of translation into foreign languages. No part of this book may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher. Printing: Druckerei Hildebrand, Berlin. - Binding: Lüderitz & Bauer, Berlin. Printed in Germany.

Acknowledgements This book is a revised and much extended version of a doctoral dissertation submitted to the University of Freiburg in 1995. Much of the collection of the data was undertaken during a stay in Scotland in the spring of 1992, supported by a research grant from the German Academic Exchange Service, which I gratefully acknowledge. Thanks go also to the School of Scottish Studies for making material available for analysis and to James Robertson, who generously provided me with a collection of Scots short stories he was editing at a time when they were not yet available in print. Above all, I would like to thank my native speaker informants, particularly those three families who not only allowed me to record them, but also extended their hospitality to me. The manuscript has profited at various stages from constructive criticism and helpful suggestions from colleagues and friends. I should like to mention my supervisor Lilo Moessner, Bernd Kortmann and Jim Miller, who read an earlier version of the manuscript. Pius ten Hacken and Flemming Andersen answered questions about Dutch and Danish respectively, and Flemming also proofread part of the manuscript. I should also like to take this opportunity to thank Cornelia Tschichold and David Allerton, my former colleagues at the University of Basel, for the co-operation which made the difficult task of combining teaching with the writing of a dissertation much easier. My greatest debt of gratitude is to Alan Murray, who not only patiently answered all my questions about his mother tongue, but also helped me with the glosses and indices, and read the whole manuscript and discussed aspects of it at various stages of completion. Without his continual encouragement and support this book might not yet be finished. Odense, Denmark

MARTINA HACKER

Table of contents Acknowledgements

v

List of figures

xii

List of tables

xii

PART I: SCOPE AND METHODOLOGY

1

1. Introduction

3

1.1. Aims and scope of the present study 1.2. The history of Scots 1.3. Perceptions of Scots and attitudes towards it 1.4. The state of Scottish language studies 1.5. The grammar of Scots: A neglected field of linguistic research 1.6. Methodology 1.6.1. Data collection and description: preliminary considerations 1.6.1.1. Spoken and written language 1.6.1.2. Introspection, elicitation and corpus studies 1.6.1.3. Dialect in speech and writing 1.6.2. The database: Collection and description 1.6.3. Methods of analysis

3 5 6 9 11 12 12 12 17 18 18 19

2. Adverbial clauses: Definitions and classifications

21

2.1. The syntactic status of adverbial clauses in grammatical theory 2.2. A function-based definition and classification of adverbial clauses 2.3. General structural characteristics of adverbial clauses 2.4. Clause patterns of adverbial clauses 2.4.1. Finite clauses 2.4.2. -ing clauses 2.4.3. -ed clauses 2.4.4. Verbless clauses 2.4.5. Tae-infmitive clauses 2.4.6. Γ/7-infinitive clauses 2.4.7. Bare infinitive clauses

21 23 26 27 27 30 31 32 32 33 33

viii

Table of contents

2.5. A function-based definition of adverbial subordinators 2.6. The morphological structure of adverbial subordinators 2.7. Adverbial clauses and polyfunctionality 2.7.1. Syntactic polyfunctionality 2.7.2. Semantic polyfunctionality 2.7.3. Polyfunctionality and the development of adverbial subordinators

34 37 38 38 46 47

PART II: DESCRIPTION

49

3. Semantic categories

51

3.1. Clauses of place 3.1.1. Locative relationships 3.1.2. Definite place 3.1.3. Indefinite place and direction 3.1.4. A Iternative constructions 3.1.5. Differences from Standard English 3.2. Clauses of time 3.2.1 Temporal relationships 3.2.2. Posteriority 3.2.3. Overlap of main clause situation and subordinate clause situation 3.2.4. Anteriority 3.2.5. Alternative constructions 3.2.6. Differences from Standard English 3.3. Clauses of condition 3.3.1. Conditional relationships 3.3.2. Condition proper 3.3.2.1. Open condition 3.3.2.2. Presupposed condition 3.3.2.3. Hypothetical condition 3.3.2.4. The verb phrase in conditional sentences 3.3.2.5. Alternative constructions 3.3.3. Rhetorical condition 3.3.4. Alternative condition 3.3.5. Indirect condition 3.3.6. Differences from Standard English

52 52 52 53 55 55 56 56 56 65 79 92 93 98 98 98 99 107 110 112 117 118 119 121 122

Table of contents

3.4. Clauses of restriction 3.4.1. Relationships of restriction 3.4.2. Restriction of reference 3.4.3. Direct restriction of knowledge 3.4.4. Indirect restriction of knowledge 3.4.5. Differences from Standard English 3.5. Clauses of exception 3.5.1. Relationship of exception 3.5.2. Differences from Standard English 3.6. Clauses of comparison 3.6.1. Relationships of comparison 3.6.2. Manner 3.6.3. Similarity 3.6.4. Dissimilarity 3.6.5. Indirect comparison 3.6.6. Differences from Standard English 3.7. Clauses of preference 3.7.1 Relationship of preference 3.7.2. Alternative constructions 3.7.3. Differences from Standard English 3.8. Clauses of concession 3.8.1. Concessive relationships 3.8.2. Alternative constructions 3.8.3. Differences from Standard English 3.9. Clauses of reason 3.9.1. Reason relationships 3.9.2. Direct reason 3.9.3. Indirect reason 3.9.4. Differences from Standard English 3.10. C lauses of purpose and prevention 3.10.1. Relationships of purpose and prevention 3.10.2. Direct purpose 3.10.3. Prevention 3.10.4. Indirect purpose 3.10.5. Differences from Standard English 3.11. Clauses of result 3.11.1. Relationship of result 3.11.2. Alternative constructions 3.11.3. Differences from Standard English 3.12. Clauses of means 3.12.1. Relationship of means 3.12.2. Differences from Standard English

ix

124 124 124 125 125 126 127 127 128 129 129 129 130 132 134 134 135 135 137 137 138 138 146 148 150 150 150 165 167 169 169 169 178 181 182 184 184 189 189 190 190 193

χ

Table of contents

3.13. Clauses of accompanying circumstance 3.13.1. Relationship of accompanying circumstance 3.13.2. Differences from Standard English

193 193 196

PART III: ANALYSIS

199

4. Variability of patterns

201

4.1. Variation within Scots 4.1.1. Regional variation 4.1.2. Written versus spoken Scots 4.1.3. Change in progress 4.2. The diachronic perspective 4.2.1. Introduction 4.2.2. Adverbial subordinator inventories in English and Scots 4.2.3. A comparison of the formation patterns of adverbial subordinator systems from Old English to present-day English and Scots 4.3. The cross-linguistic perspective: Structural comparison 4.3.1. Formation patterns of adverbial subordinators and adverbial clauses in some European languages 4.3.2. The morphological structure of adverbial subordinators in comparison 4.4. The cross-linguistic perspective: Semantic comparison 4.4.1. The semantic composition of adverbial subordinator systems 4.4.2. Cognitive semantics and the predictability of subordinator meaning

201 201 202 204 206 206 209

214 217 217 220 221 221 227

5. Summary and conclusion

236

5.1. 5.2. 5.3. 5.4. 5.5. 5.6.

236 237 240 241 242 244

Variation in Scots: Qualitative and quantitative differences Scots and Standard English: Qualitative and quantitative differences Adverbial clauses in Scots: Possible and probable developments Adverbial clauses in Scots as a case study in dialect syntax Adverbial clauses in Scots: Implications for linguistic theory Areas for future research: Some suggestions

Table of contents

xi

Appendix: The database

247

1. The spoken corpus 2. The written corpus

247 250

Notes

255

References

267

Maps

287

A note on maps Map 1: Scots dialect regions according to the SND Map 2: Regional divisions based on the present study Map 3: Regional distribution of spoken texts Map 4: Regional distribution of written texts Map 5: Regional distribution of written and spoken texts combined

287 288 289 290 291 292

Author Index General Index

293 296

xii

Table of contents

List of figures Figure 1. The centrality scale of adverbial clauses Figure 2. Subordinates—co-ordinator—conjunct triangle

24 45

List of tables Table 1. The gradience of clause integration in linguistic theory Table 2. The centrality scales of adverbial clauses in different linguistic theories Table 3. Temporal subordinators introducing finite clauses in Scots and English Table 4. The division of the temporal semantic field in Scots and English Table 5. Frequency of major temporal relationships in written and spoken Scots and English Table 6. The temporal-conditional semantic field in Scots and English Table 7. Frequency and function of for all (that) and as far as in LOB, Brown and LLC Table 8. Concessive clauses introducing finite clauses in Scots and English Table 9. Frequency of because, for, as, and since introducing clauses of reason in Scots and English Table 10. Frequency of clauses of purpose in Scots and English Table 11. Frequency of clauses of result in Scots and English Table 12. Frequency of zero-introduced clauses of means in Scots and English Table 13. Frequency of clauses of accompanying circumstance in Scots and English Table 14. Proportional Relationships of subordinators of time, reason, condition, and concession in a selection of European languages Table 15. Proportional relationship between polyfunctional and monofunctional subordinators of time, reason, condition, and concession Table 16. The most frequent adverbial relationships in -ing clauses, -ed clauses, and verbless clauses in Scots and English

22 24 95 96 97 123 126 149 168 183 190 193 197 223

224 240

In days when mankind were but callans, At Grammar, Logic, an' sic talents, They took nae pains their speech to balance, Or rules to gie, But spak their thoughts in plain, braid lallans, Like you or me.

ROBERT BURNS

PARTI SCOPE AND METHODOLOGY

1. Introduction

1.1. Aims and scope of the present study This study investigates the semantics and syntax of adverbial clauses in Scots. It thus contributes to the study of English dialect grammar, a long neglected field of linguistic research, which has only recently begun to gain recognition among linguists. While a description of any aspect of dialect grammar has its own merit, given the scarcity of such descriptions and their relevance for education (see, for example, Cheshire 1982, Cheshire and Edwards 1993, Cheshire and Milroy 1993, Edwards 1989, and Williams 1989), the study of dialect grammar can also make an important contribution to other areas of linguistic research, a fact which has been largely overlooked. The relevance of non-standard grammar, in particular that of regional dialects, for other linguistic disciplines is related to the following factors: (i) Synchronic variation to a great extent mirrors diachronic variation, (ii) Dialects show a lower degree of standardisation compared to the standard language, (iii) Dialects are predominantly used in the spoken medium. The close relationship between synchronic variation and diachronic variation implies that an increase in knowledge in the former area necessarily leads to an increase in the latter; thus new insights into processes of language change may be gained from the study of dialects. The lower degree of standardisation in dialects makes it possible to observe processes of language change more clearly than in their respective standard languages, where they tend to be obscured by the prescriptiveness of standard grammars. The study of dialect grammar may therefore be of a similar interest for research into grammaticalisation processes as Creole languages are. In addition, dialects can be used as test cases for linguistic theories. Do their structures confirm universal claims about the relationship between language structure and the degree to which a particular language exists in written form, i.e. does their structure reflect the functional restriction to informal, predominantly spoken use? Do they support typological claims made on the basis of standard languages? The study of dialect grammar here may thus function as a corrective, preventing overgeneralisations based on incidental developments in the standard language or refuting untenable claims derived from them. A case in point is Chomsky's (1989) claim that language-specific rules can apply only when universal principles are blocked, which is based on the fact that unstressed tfo-support in English occurs in interrogative sentences, but not in declarative sentences. This claim is refuted by Ihalainen's (1991) analysis of the use of periphrastic do in the dialect of Wessex, where do in declarative

4

Part I. Scope and methodology

sentences has no other function than that of tense-carrier (see Chambers and Trudgill 1991 for a detailed discussion). Scots is a particularly promising candidate for the study of dialect grammar because of its history as an independent variety until the Early Modern English period and its extended contact with a number of other languages, namely Gaelic, Scandinavian, French and Dutch. Adverbial clauses were chosen as the subject for analysis in this study for two main reasons: firstly, because of an impressionistic notion that there are structural and semantic differences between Scots and Standard English adverbial clauses; secondly, because the historical development of subordinating conjunctions, which shows them to be extremely flexible elements continuously extending and changing their meanings and structural scope, suggested adverbial clauses as a likely construction to have retained or developed features different from Standard English. With few exceptions, such as the subordinator gin, adverbial clauses do not contain "overt" Scotticisms. This makes the researcher's task more challenging. Being a non-native speaker who has consciously acquired the grammatical rules of Standard English may be an asset in such a task. It is selfevident that covert Scotticisms will be more frequent in Scottish Standard English than overt ones, and therefore many of the features noted for Scots adverbial clauses will also be found in Scottish Standard English. The study has the following major aims: (i) to provide a comprehensive syntactic and semantic description of adverbial clauses in present-day Scots in both written and spoken forms, (ii) to compare the Scots system of adverbial clauses with that of its closest linguistic relative, Standard English (in both spoken and written forms), (iii) to place Scots in this respect within a wider geographical context by providing comparisons with other European languages, in particular those whose ancestors may have influenced Scots through language contact, and (iv) to relate the results of these comparisons to claims and predictions made in language typology. The comparisons will focus on the following aspects: clause structure, formation patterns of adverbial subordinators, syntactic polyfunctionality of items serving as adverbial subordinators, and polysemy patterns of adverbial subordinators. The restriction of the study to the syntax and semantics of adverbial clauses was made to keep the scope of the study within a manageable size. This means that prepositional phrases have largely been excluded despite their great functional similarity to adverbial clauses, as have other constructions expressing similar semantic relations, such as conjunctive adverbs. It should be noted, however, that the study is not restricted to finite adverbial clauses, but also includes non-finite and verbless adverbial clauses and that the relationship between conjunctions and functionally similar word classes, such as prepositions and conjunctive adverbs, is extensively discussed in the section on the structural formation of subordinators and the polyfunctionality of subor-

7. Introduction

dinators. With respect to the detailed semantic description, the study is restricted to present-day written and spoken Scots, but references to earlier stages of Scots and English are made where they help to explain structures and meanings of the present system.

1.2. The history of Scots The ancestor of Modern Scots is Old Northumbrian, which was spoken in the kingdom of Bernicia.1 The Angles had settled as far as the Firth of Forth in the course of the seventh century and according to place-name and chronicle evidence spread over south-eastern and southern Scotland as well as along the Solway. In 973 part of this area, Lothian, was lost to Kenneth II, king of Scotland, and until the twelfth century Northumbrian and Gaelic probably coexisted in southern Scotland, although Cumbric, a P-Celtic language, survived in parts of Strathclyde up to this time (Macafee—O Baoill 1997); the rest of Scotland was Gaelic-speaking with the exception of Shetland, Orkney and Caithness, where Old Norse was spoken. During the reign of King David I (1124-53) the language of the Scottish court changed to Norman French while that of the people remained predominantly Northumbrian and Gaelic. In the eleventh and twelfth centuries there was an influx of immigrants from northern England, whose speech contained a strong Scandinavian element, as in 867 the Danes had established a Norse kingdom at York. Their language had merged with the local vernacular and by the fourteenth century this had become the dominant language of all Lowland Scotland, that is, the regions south and east of the Highland Line. In the fourteenth century what is termed Early Scots, though still hardly distinguishable from Northern English, began also to enter the Norse-speaking areas of Caithness, Shetland and Orkney. This linguistic division of Scotland between the Scots-speaking Lowlands and northern islands, and Gaelic-speaking Highlands and Western Isles remained roughly the same until the late eighteenth century, when the linguistic border receded further into the Highlands. Since Gaelic was supplanted by Gaelicised English rather than Scots after the eighteenth century, the areas affected are not counted as Scots-speaking. The common division into four broad dialect areas, primarily based on phonetic and lexical differences, is illustrated by Map 1 (cf. The Scottish National Dictionary [henceforth SND], vol. I: ix-xli; also CSD: xvi). The gradual development of English and Scots into separate languages was encouraged by the existence of two political centres at London and Edinburgh, which served as linguistic points of orientation, and the recognition of a fixed political frontier. The fifteenth and sixteenth centuries were the heyday of the Scots language. Scots was not only the language of the Scottish court, but also

6

Part I. Scope and methodology

a language of high literary standards as is amply attested by the work of the poets William Dunbar, Gavin Douglas and Robert Henryson. Its decline is blamed on various factors having an anglifying effect: (i) the loss of a separate Scottish court as a result of the Union of the Crowns, which took place when James VI of Scotland succeeded to the throne of England in 1603, (ii) the advent of printing which brought English books, in particular the English Bible, to Scotland, at a time when there was no Scots Bible available, and (iii) the Reformation, which strengthened links with Protestant England at the cost of traditional links with Catholic France. In 1844 Lord Cockburn made the following observation concerning the changes affecting the speech of the upper classes: English has made no encroachment on me; yet, though I speak more Scotch than English throughout the day, and read Burns aloud, and recommend him, I cannot get even my own children to do more than pick up a queer word of him here and there. Scotch has ceased to be the vernacular language of the upper classes, (cited from Kay 1993: 112) Nevertheless, despite the decline lamented by Cockburn, Scots has continued to be spoken to the present day and has even extended its functions from a language restricted to the home and (in its written form) poetry by being extended to other literary genres, such as drama and prose, and by entering the media and the classroom.2

1.3. Perceptions of Scots and attitudes towards it Some distinctions need to be drawn between the different terms used to refer to the Scottish language. Beside "Scots" we find "Scotch", "Scottish English" and "Lallans". The original term was "Inglis", while from the fifteenth century "Scottis" gained more ground. In the eighteenth century the form "Scots", which derived from "Scottis", was gradually replaced as a result of English influence by "Scotch", which was the English reduced form of "Scottish". As is apparent from early studies on Scottish dialects (e.g. Wilson 1915), "Scotch" was used up to the beginning of this century. When "Scotch" used to refer to the Scottish people began to acquire derogatory connotations it was replaced by "Scots" or "Scottish". The designation for the language followed suit, being replaced by "Scots", or "Scottish" by Scottish non-linguists. The use of "Scottish" is illustrated by the following extract from a caller's opinion, expressed in the phone-in programme "Speaking Out" (BBC Radio Scotland, 27 February 1992), which was dedicated to the use of Scots in education, arts and the media.

1. Introduction

1

They couldn't talk to the Scottish soldiers, the Americans, because they didn't even know what they was sayin although they liked them very much ... and 1 think we should be progressin and teachin the kids uh all types ο languages because you're still gonna take time up in school teachin the Scottish, but I see it as backwards and not forwards. It seems that the CSD's statement (entry Scots) that "Scotch is still the regular vernacular form" may need to be modified at least with respect to the designation for the language. This is also suggested by the fact that Kay (1993: 156-57) only mentions it as being used by older speakers. "Scotch" is, however, still used in food terminology, such as "Scotch whisky" and "Scotch broth", where it has no negative connotations, but denotes quality. "Scotch" on its own as a synonym for whisky is, however, only used outside Scotland. The distinction between "Scots", and "Scottish English", or "Scottish Standard English", is more recent and restricted to linguistic use. There was no need for this term as long as linguistic research concentrated on local rural dialects. The term "Scottish English" is mainly used with respect to pronunciation, that is to refer to Standard English as pronounced by Scots. Scottish Standard English has a different vowel and diphthong system from (English) Standard English. In addition, is pronounced as a flap or trill irrespective of its position, in Scottish words, such as loch, as a velar fricative, and is generally distinguished from (cf. Abercrombie 1979; Aitken 1979; Giegerich 1993: 45-46, 53-57). Scots differs from Standard English not only with respect to pronunciation, but also vocabulary and grammar. Speaking Scots therefore means speaking a dialect of English, or, some would argue, a different language, whereas speaking Scottish Standard English means speaking Standard English with a Scottish accent. For many individuals in Scotland both Scots and Scottish Standard English are available options. It also has to be borne in mind that all speakers of Scottish Standard English to a greater or lesser degree unconsciously use vocabulary, phrases and constructions that are geographically restricted, i.e. not used in English Standard English, such as outwith as the opposite of within, and that there are differences in the density of Scots, too (cf. Aitken 1979: 104-109; McClure 1979). These exist not only between different speakers, but also vary with the same speaker depending on the linguistic and extralinguistic context. Therefore there is no clear-cut boundary between Scottish Standard English and Scots, but a gradient. Another recent distinction is that between "Broad Scots" (also "Braid Scots") and "Lallans" ("Lowland Scots"), also referred to as "Synthetic Scots", "Ideal Scots" or "Plastic Scots", depending on one's attitude towards the literary language that draws from various dialects and includes extinct words from the language of Burns and earlier poets. The term "Synthetic Scots" was coined by

8

Part I. Scope and methodology

the poet Hugh MacDiarmid for the language he used in his own work. "Synthetic" was meant in the original sense of the word, i.e. a combination of different elements to form a whole. Only later did it acquire negative connotations and was used in the same sense as "Plastic Scots" by those who felt that it was artificial and unnatural. To distinguish their own use of Scots from "Synthetic Scots" they used the existing term "Broad Scots". Aitken's (1981: 80) term "Ideal Scots" contrasts the literary language with the "real" Scots of the spoken vernacular. None of these terms is used by Scots speakers not involved in the linguistic discussion. They would speak of themselves as speaking, for example, "Hawick", "Fife", or the "Doric", thus using terms associated with localities or regions rather than a de-regionalised Scots. The term "Doric", originally referring to the oldest and simplest style of Greek architecture, was extended to language, where it likewise stands for the "old and simple". From a general term for broad rustic dialects it has become increasingly restricted to the designation for one of the broadest dialects, that of north-eastern Scotland. The use of terms referring to localities and regions reflects the frequently uttered opinion that the different Scottish accents or dialects are so different from each other as to be mutually incomprehensible. Kay (1993: 156) states that "for many Scots speakers and writers the local dialect is more important, and more clearly defined as an entity than something called Scots". This is confirmed by the views expressed in the abovementioned phone-in programme by a caller from Aberdeen. Well, my point is that there is no possibility for having the Scots language taught, because the south of Scotland and the north of Scotland speak it terribly differently ... Well, I'm just saying there's no possibility of teaching a national Scots language, I mean. ("Speaking Out", BBC Radio Scotland, 27 February 1992) Another widespread view in Scotland is that the people of the urban centres do not speak Scots. The urban varieties are stigmatised and often described in derogatory terms, such as "Gutter Scots" etc. (cf. Robertson 1994: vii). The need to distance oneself from these varieties leads to the frequently uttered belief on the part of educated people that they themselves do not speak Scots at all. Thus in the same phone-in programme Alan Jack, Assistant Head of Radio Scotland, gives the following answer to the moderator's question, "Are you a Scots speaker yourself, do you think?" No, definitely not — uhm — My father has some Scots words in his — uhm — vocabulary from — I think childhood holidays in Lanarkshire ... But no, I'm born and bred in Glasgow, and I do not have very many Scots words at all. ("Speaking Out", BBC Radio Scotland, 27 February 1992)

Λ Introduction

9

This statement is interesting in two respects. Firstly, it shows that the common opinion that dialect is primarily a question of vocabulary and only found in rural areas is also current among educated people. Secondly, Alan Jack is obviously unaware of the fact that he is using a Scots construction when he says "I'm born and bred in Glasgow", rather than "I was born and bred in Glasgow". The Scots perception of their own language raises two questions, namely, to what extent urban speech and rural speech have common features and whether the different varieties of Scots have more in common with each other than with Standard English.

1.4. The state of Scottish language studies It seems appropriate in a work that analyses an aspect of Scots grammar to give a brief overview of the state of studies in the field of Scottish language. Early work concentrated on lexicography (cf. Murison 1987 for a detailed survey). It arose from the need for glossaries of Scottish words to read Scottish literature and from the philological and antiquarian interests of the eighteenth and nineteenth centuries. Scots was seen as particularly interesting as it had preserved many lexical items from Old and Middle English which no longer existed in contemporary English. In the second half of the nineteenth century scholars began to describe local dialects. Pioneers in this field were Gregor (1866) and Murray (1873). Descriptions of local dialects focusing mainly on phonetics and lexis, including some morphological but little syntactic information, remained the main topic for individual research until the 1970s (e.g. Dieth 1932; Wettstein 1942; Speitel 1969). Exceptions are Wilson's Lowland Scotch as spoken in the Lower Strathearn district of Perthshire (1915) which besides phonological and lexicographical information contains a section on grammar of some eighty pages, and Grant and Dixon's (1921) Manual of Modern Scots, which consists of some 500 pages of a grammar and reader. The reader contains extracts from Scots prose, poetry, ballads and songs dating mainly from the eighteenth and nineteenth centuries; the grammar is written in the tradition of nineteenth-century philology, containing sections on phonetics, parts of speech (articles, nouns, pronouns, adverbs, prepositions, conjunctions, interjections), and word formation (prefixes, suffixes and compounds), but not on syntax. While Grant and Dixon's work on grammar found no followers until much later, the interest in lexicography has been undiminished, producing The Scottish National Dictionary (1931-76) [henceforth SND], A Dictionary of the Older Scottish Tongue (1937-) [henceforth DOST], which covers the language up to 1700, The Concise Scots Dictionary (1985) [henceforth CSD], The Pocket Scots Dictionary (1988), and the Scots Thesaurus (1990). In the

10

Part I. Scope and methodology

field of word-geography the most important publication was The Linguistic Atlas of Scotland (1975-86), while work on a linguistic atlas of Older Scots is being undertaken at the Institute of Historical Dialectology attached to the School of Scottish Studies at the University of Edinburgh (cf. Williamson 1992-93). In addition, linguistic and literary studies on Older Scots appeared, focusing on the language of the makars and Middle Scots lexicology and orthography. In the 1970s linguistic interest in Scotland, as elsewhere, turned to sociolinguistics. This is reflected in studies of the urban speech of Scotland's two largest cities, Glasgow and Edinburgh (e.g. Macaulay 1977; Macafee 1983; Reid 1976; Romaine 1975). In connection with new insights in pedagogy and with the political situation, particularly the movements for devolution and independence, the interest in the social aspects of language developed into several strands: (i) Scots and education (e.g. Low 1974; Borrowman 1977), (ii) the status of Scots as a language or a dialect of English (e.g. McClure 1980; Aitken 1981; Moessner 1993) and (iii) attitudes towards Scots (Aitken 1982; Sandred 1983). The 1980s saw an increase in studies on the use and function of Scots in literature (e.g. Tulloch 1980; Letley 1988), and also the first detailed descriptions of aspects of present-day Scottish grammar (e.g. Brown and Millar 1980; Miller and Brown 1982; Kirk 1987; Moessner 1988). It is no coincidence that these focus on negation, auxiliaries and subject-verb concord. They thus deal with frequent syntactic structures which show noticeable differences from Standard English (i.e. overt Scotticisms):3 the clitic negative adverb -na, double modals and the ending -s used where Standard English would require a zero-ending. Aspects of the discourse structure of spoken Scots have been analysed by Miller (1984) and Macaulay (1991). Other recent studies investigate the possibilities of applying computer technique to the teaching and analysis of Scottish language (e.g. Zettersten 1987; Kirk 1992-1993). Interest in the grammar of Scots has recently received a new stimulus from the changing attitude towards non-standard English, as apparent from the title "Real English" of a recently published book on dialect grammar (Milroy and Milroy 1993), which contains a brief, though well-informed, "Grammar of Scottish English" (Miller 1993). A similarly short grammar of Older Scots was published by Macafee (1992-93). The most recent addition to the study of Scots grammar is The Edinburgh history of the Scots language (Jones 1997), including chapters on "The syntax of Older Scots" (Moessner 1997c), "The inflectional morphology of Older Scots" (King 1997), and "Syntax and morphology" (Beal 1997). All of these publications are very welcome, but at the same time they emphasise the need for more detailed research on the subject, both for Older and Modern Scots.

/. Introduction

\\

1.5. The grammar of Scots: A neglected field of linguistic research There are several reasons why there is no comprehensive grammar of Scots or Scottish English. Most important of these is that Scots grammar for a long time was viewed as a corrupt version of English grammar, which needed to be corrected, but not described. That this is not true can be seen from the few detailed studies on aspects of Scots grammar available to us. A second factor, not unrelated to the first, is the lack of resources to carry out research on Scots grammar. The cuts in university funding during the 1980s in Scotland particularly affected the studies of Scots. The philosophy of Thatcherism did not encourage the study of subjects whose economic value was not immediately recognisable. As a consequence all but one of the departments of linguistics at Scottish universities were closed, while departments of Scottish language did not exist until recently. The emergency of the situation can be inferred from Kirk's (1992-1993: 100) appeal: Every opportunity for staffing and funding must be encouraged, with more — not fewer! — university posts for scholars and researchers willing to specialise in Scots — and quite definitely more posts in Scottish Language in relation to the seemingly ever larger number of posts in Scottish Literature. Kirk touches on an additional problem here, that of the relationship between Scottish literature and Scottish language. Scottish literature was never stigmatised in the way Scottish language was, and as a result it is easier to persuade both the authorities and the people of Scotland of the need for a Department of Scottish Literature than it is to persuade them of the need for a Department of Scottish Language. The dichotomy between the attitudes to Scots literature and language is not new. The historian John Pinkerton states in his preface to a collection of Scots poems in 1786 that ... none can more sincerely wish a total extinction of the Scottish colloquial dialect than I do, for there are few modern scoticisms which are not barbarisms ... Yet I believe, no man of either kingdom would wish an extinction of the Scotish dialect in poetry, (cited in Kay 1993: 106). The question arises, therefore, whether there is any need at all for a grammar of Scots. Views on the grammar of Scots expressed by Macafee (1983) and Kirk (1987) contrast sharply with those of Miller and Brown (1982: 16) and Moessner (1992), who hold the view that the differences between Scots and Standard English are systematic. Thus Macafee (1983: 47) states that "the grammar of Scots differs from that of Standard English mainly in superficial

12

Part l. Scope and methodology

features such as the form of inflections and function words, and rules of concord'1, and Kirk (1987: 179-180) is of the opinion that there is no autonomy [of Scots], for Traditional Scots, Glasgow Scots and Standard English all share the same "grammar", and so the same identity. All are subject-verb-object dialects, in which the sentence has the same set of constituents and the same principles and patterns of ordering and embedding these constituents. While confirming the differences in the system of auxiliaries described by Miller and Brown (1982), Kirk (1987: 181, note 6) denies their significance, "None of these differences detract from the contention that the auxiliary verb systems in Scots and StE [Standard English] are not significantly different, or from the existence of a common core between the two". This means that there is perhaps not only a tendency to underestimate the extent of the differences between Scots and Standard English, as Miller and Brown (1982: 16) claim, but also a tendency to undervalue them. This undervaluation constitutes an additional obstacle for the study of Scots, as scholars naturally do not want to devote valuable research time to "insignificant" matters. All these factors not only explain why there is no grammar of Scots, but also suggest that the lack of it will probably continue for a long time. However, irrespective of the ideological views on Scots as a language or a dialect of English, the few existing studies on Scots grammar suggest that it shows equally systematic differences from present-day Standard English as those noted for Middle English. Like Middle English, Scots therefore deserves to be described as a system of its own.

1.6. Methodology 1.6.1. Data collection and description: preliminary considerations 1.6.1.1. Spoken and written language Comparisons between spoken and written language are a fairly recent phenomenon in linguistic research. For centuries linguistics, or philology as it was called then, concerned itself with the study of past stages of languages and therefore necessarily with language in its written form. The shift of linguistic interest from diachrony to synchrony did not immediately result in the study of spoken language, as written language was simply seen as a "way of recording language by means of visible marks" (Bloomfield 1935: 21). It therefore sufficed to study language in its more easily accessible written manifestation.

1. Introduction

13

Such views were no longer tenable when the improvement in recording techniques facilitated and encouraged the study of spoken language in the latter half of this century. Chafe's (1982) seminal article "Integration and involvement in speaking, writing, and oral literature" relates structural differences between spoken and written texts to two basic differences in the discourse situation: (i) that speaking is faster than writing (and slower than reading) (ii) that speakers interact with their audiences directly, whereas writers do not (Chafe 1982:36). The question linguists are concerned with now is no longer whether a comparison of speech and writing is unnecessary because they are "merely two manifestations of something fundamentally the same" (Hockett 1958: 4), but to what extent a comparison is possible, given the different circumstances under which written and spoken discourse is produced. One difficulty arises from the fact that sentences, which are an easily recognisable unit in written texts, are a problematic category in speech. Some linguists have therefore abandoned the term "sentence", or restricted it to the orthographical sentence of written language. Thus Halliday (1985) uses the term "clause complex" for the grammatical unit associated with the orthographical sentence. A string of arguments in favour of abandoning the notion of sentence as a grammatical unit are presented by Miller (1995). He argues that the conventions that orthographically mark sentences have changed over the centuries, that children have to be taught the rules of orthography to recognise sentences, and that people's intuitions differ as to what constitutes a sentence. These arguments certainly confirm the view that the basic unit of language is the clause rather than the sentence, but they do not rule out a larger unit, such as Halliday's clause complex, whose definition as a "head clause together with other clauses that modify it" (Halliday 1985: 192) corresponds to the sentence of traditional grammar. If such larger units exist, what are the means available to speakers to indicate them? Firstly, they could use grammatical subordination, secondly, they could use prosody (intonation contours, changes in tempo and volume, and pauses), and thirdly, they could use extralinguistic means, such as facial expression and gestures. The linguistic controversy about the recognition of sentences in speech focuses on intonation and pauses. Chafe and Danielewicz (1987: 103104) argue that the sentence should be retained as a grammatical unit of spoken language despite a certain amount of ambiguity in speakers' signalling of sentence boundaries:

14

Part l. Scope and methodology

The function of sentences in spoken language is problematic, but speakers appear to produce a sentence-final intonation when they judge that they have come to the end of some coherent content sequence ... It is difficult to predict when a speaker will decide that a chain of intonation units worthy of recognition as a complete sentence has been produced. Speakers are sloppy in this respect, often producing a sentence-final intonation before they mean to, or neglecting to produce one when they should. Premature sentence closures are often followed by afterthoughts; delayed ones give the effect of run-on sentences. What Chafe and Danielewicz term "sloppiness" is, of course, the effect of spontaneous production. Halliday (1987) points out that the "imperfections" that seem to be characteristics of spoken language are by no means restricted to this medium, but are equally present in writing, where they are, however, eliminated in the final version that reaches the reader, with the exception of personal letters, which undergo less editing. He adds that spontaneous speech shows a great deal less fragmentation than self-monitored speech, such as that typically found in academic discourse (cf. Chafe 1987). Indeed the examples of spoken discourse cited by him can easily be presented as written sentences.4 Why then is there so little agreement about sentence boundaries in spoken language? In her study of spoken German, Wackernagel-Jolles (1971) asked a group of test persons to mark sentence boundaries in texts that were simultaneously played to them on tape and given to them as unpunctuated transcriptions. Some sentence boundaries were marked almost in complete agreement by all informants, while in other cases roughly 50% indicated a sentence boundary, while the rest did not. People disagreed predominantly in the following cases, which are given in the order of frequency: (i) where a clause begins with the prototypical co-ordinating conjunction und, (ii) where parenthetic clauses are interpolated, and (iii) in enumerations of zero-introduced finite clauses. German grammar is highly prescriptive with respect to the use of co-ordinating conjunctions. It seems safe to assume that all of WackernagelJolles's informants were taught at school never to begin a sentence with und, which was an essential part of the elementary grammar teaching in primary schools up to the mid-1960s. At the same time und functions as an all-purpose connective device in spoken German. This means that in speech it is also employed at transitions that in written German would be sentence boundaries. It seems therefore plausible that the divided opinion of Wackernagel-Jolles's informants reflects the division between those who have internalised the prescriptive grammar teaching of their early schooldays to the extent that they strictly obey the rules they were taught and accordingly indicate no sentence boundary before und, and those whose judgements are less influenced by prescriptive grammar teaching. Interpolated parenthetic clauses are likewise instances where disagreement is to be expected, as they are typically embedded

1. Introduction

15

in another clause with respect to linear sequence, but not with respect to syntactic function. Enumerations are borderline cases, as the mere fact that they are part of a list suggests a close relationship with the other members of the same list, but in the case of full finite clauses there is no overt marking to signal such a close relationship on the structural level. Should evidence such as that presented by Wackernagel-Jolles be taken as an indication for the non-existence in speech of a grammatical unit corresponding to the orthographic sentence? Miller (1995: 133) argues strongly in favour of such a view. He sees "very little evidence to support either text-sentences or system-sentences in spontaneous spoken language."5 Interestingly, interpolation is also a frequent feature in the examples he cites to point to the discrepancy between intonation contours and dependency relations. This suggests that interpolation may be a special category, which does not fit into the system of linear sentence structure, irrespective of the medium of production. The punctuation convention of marking off interpolations by dashes supports such a view. Another frequent problem in the analysis of texts is illustrated by the case of German und. Here it seems that the difficulty in marking sentence boundaries originated not so much in the transference of a written structural unit, the sentence, onto spoken language, but the transference of the categories assigned to grammatical words in prescriptive grammar, which restricts people's perception of functional polysemy. If we allow for fuzzy edges of the category "sentence", a feature which is shared by most grammatical categories, it can be said that the semantic and syntactic relationships between clauses are mostly the same in spoken and written language. Thus, the following example, cited by Greenbaum and Quirk (1990: 460 [1]), is characterised as spoken, not by a combination of clauses that does not allow us to analyse it as a sentence, but by the way the clauses are combined to serve the speaker's purpose of apologising for and justifying his or her behaviour: Although I know it's a bit late to call, seeing your light still on and needing to get your advice if you'd be willing to help me, I parked the car as soon as I could find a place and ventured to come straight up without ringing the bell because, believe me, I didn't want to add waking your baby to the other inconveniences I'm causing you. The major differences between spoken and written language are the frequency with which different types of relationships are employed: the number of clauses that form a structural unit, the way structural units are marked off and the proportion of different clause types and the proportion of metalinguistic utterances. In spoken language intonational marking may be in conflict with structural marking, as speech is on-line processing and therefore even minor ad

16

Part I. Scope and methodology

hoc changes, such as the addition of an afterthought, will be noticeable in a careful analysis of spoken discourse. Major changes, such as anakolutha, which may obscure the structural organisation of clauses into larger units, are mainly restricted to self-monitored speech and are rare in spontaneous narratives (Halliday 1987: 68). Halliday's statement seems to contradict Miller (1995: 132). However, the apparent contradiction can be resolved if a distinction is drawn between spontaneous discourse and informal narrative. Miller's term "spontaneous discourse" includes conversations where the participants give directions or try to follow directions. They necessarily pause to think, which is in one instance even verbalised as "hold on let me see" (Miller 1995: 125), and correct themselves. This relates to another problem concerning a comparison of spoken and written language. There is no simple dichotomy between spoken and written texts, but a variety of different text types exist in both media. Biber (1988) distinguishes as many as seven dimensions which determine the structure of discourse: (i) (ii) (iii) (iv) (v) (vi) (vii)

Informational versus involved production Narrative versus non-narrative concerns Explicit versus situation-dependent reference Overt expression of persuasion Abstract versus non-abstract information On-line informational elaboration Academic hedging

He lists "narrative versus non-narrative concerns" as the second of seven dimensions determining the structure of discourse, preceded only by "informal versus involved production". Spoken and written narratives also range closely on the scales of Biber's dimensions (i), (iii), and (v), where spoken and written texts typically have different patterns. Thus fiction has an intermediate value with respect to dimension (i) and a low, i.e. oral, value with respect to dimensions (iii) and (v). Beaman (1984), who compares written and spoken spontaneously produced narratives, retellings of the same film, likewise found a comparatively high correspondence between the structure of the written and spoken versions. Therefore the choice of text type will inevitably influence the result of any comparison of spoken and written language. More and greater differences will be found if the respective texts are maximally different with respect to Biber's seven dimensions. In the reverse case, the results will be less striking, but such a choice will ensure that the differences are related to the medium of production rather than to other factors.

1. Introduction

17

1.6.1.2. Introspection, elicitation and corpus studies Grammatical description is either based on the linguist's intuition, also referred to as introspection, or on the analysis of a text corpus. Introspection is a highly subjective method and should therefore not be used without a database as a corrective.6 Nevertheless, it is indispensable for the interpretation and evaluation of linguistic data. Several methods are available to set up a database. The methods traditionally used in linguistic fieldwork are elicitation, questionnaires and recordings of spontaneous speech. Two kinds of elicitation need to be distinguished: (i) the researcher supplies part of a syntactic structure which is then completed by the informant; (ii) he or she directs the conversation in a way that obliges the informant to produce the required item. Both methods are frequently used in lexicography. There is, however, a danger of the researcher influencing the results in both methods, because the informants might guess the researcher's purpose and produce the "right" answer, i.e. the answer they feel the researcher would like to hear. Elicitation of the first type is also common in the form of a questionnaire. A different type of questionnaire requires the informant to compare his or her own active and passive use of certain lexical items or constructions to a given set. Compared to elicitation this method has the advantage of not restricting the informant to one answer. It restricts the informant's answers, however, to those provided in the questionnaire, i.e. those already known to the researcher. The method is therefore best suited to confirm the existence or absence of certain items in the informant's idiolect. While the use of questionnaires may be indispensable for large-scale projects involving the compilation of data by a number of researchers, this was rejected for the present study as the presentation of a Standard English questionnaire might have influenced the informants' answers. The questionnaire method of offering several possible answers was likewise unsuitable, as the purpose of the study was to gain new linguistic information rather than to confirm existing knowledge. Elicitation is difficult to apply to grammatical constructions and was therefore likewise rejected. The use of a text corpus as a database has several advantages. The data are not influenced by the researcher in the case of written texts or less influenced than by other methods in the case of spoken ones. This does not mean that the presence of a tape recorder does not affect the speech of an informant. This is, however, of greater consequence in phonetic analysis than in the analysis of unmarked grammatical constructions. A text corpus allows quantitative analyses, which is important for studies whose objective is not merely the synchronic description of a language or a variety, but which are concerned with variation and change, as here differences tend to be quantitative rather than qualitative. A text corpus was therefore chosen as a database for the present study.

18

Part I. Scope and methodology

1.6.1.3. Dialect in speech and writing Dialects are predominantly the medium of everyday informal speech. Whether a dialect exists at all in a written form depends largely on the prestige it carries not only among its native speakers, but also within the society of the country where it is spoken as a whole. The position of Scots is fairly complex in this respect. It is a variety that is perceived as difficult to understand, but it is not one that is antipathetic to most Britons. Writing in Scots has a long tradition, but one that has mainly been restricted to poetry. Those who write in Scots today were taught to write in a different variety, namely (Scottish) Standard English, and most of their non-literary writing will still be done in Standard English. This means that writing in Scots is a conscious process, which tends to be influenced by the three models available to the dialect writer: (i) the standard language, i.e. Scottish Standard English, (ii) the literary tradition of the dialect, i.e. literary Scots, and (iii) the dialect in its spoken form (spoken urban or rural varieties of Scots). In most cases the influence of all three models will be at work to varying degrees. The model of Standard English will mostly be present at a subconscious level, as an author who chooses to write in his or her dialect consciously opts for this rather than the standard variety. A typical effect of this subconscious influence is inconsistent spelling, where non-standard spelling predominates but standard spelling crops in unintentionally. With respect to the other two models authors may consciously decide for the more literary or the more colloquial style. However, even those authors who consciously choose the literary model will draw on their everyday experience of the spoken dialect. These considerations suggest that a comparison of spoken and written dialect texts may confirm the following predictions: (i)

Written dialect texts show a greater similarity to the standard language than spoken ones, (ii) The difference between written and spoken dialect texts is smaller than that between written and spoken texts in the standard variety, (iii) Written dialect texts may show a number of features that are no longer part of the spoken dialect.

1.6.2. The database: Collection and description The analysis presented in this study is based on two corpora, of written and spoken Scots narrative prose respectively. Narratives were chosen as a text type naturally occurring both in the spoken and the written medium (cf. Givon 1990: 951). The inclusion of spoken material is particularly important, as Scots

1. Introduction

19

is predominantly used as a spoken medium. Although Scots people do not normally write in Scots, Scots was preferred to Scottish Standard English for the written corpus as well, as it is thus possible to identify differences between Scots and English on the one hand and differences between the style of writing and speech on the other. The database consists of two corpora (written and spoken texts) of approximately 100,000 words each. Their geographical distribution is given in Maps 3-5. Most of the data were collected during a stay of three and a half months in Scotland early in 1992, funded by the German Academic Exchange Service. A number of principles guided the selection of texts. They should be (i) Scots, (ii) contemporary, (iii) prose, (iv) narrative (v) representative of the different dialect regions within Scotland, (vi) representative of different age groups. Ideally the two corpora (written corpus and spoken corpus) should be balanced with respect to the last two principles. This was not entirely possible, for pragmatic reasons. Not all regions of Scots-speaking Scotland are equally productive as far as narrative in Scots is concerned, nor are all age groups. Nevertheless, the database covers all the major dialect regions of Scotland, except Orkney and Galloway, for which insufficient material was available. It was possible to obtain spoken material from all regions with the exception of Caithness. All age groups are represented in both corpora, with the exception of children, whose speech is only recorded in one spoken text. The selection of informants thus differs from the traditional one, described by Wilson (1915: 10), who states that he recorded "the unthinking speech of the oldest residents" in order to obtain "the true dialect of the locality". The written corpus consists predominantly of short stories published in books and journals, most of them being 3000-4000 words in length. The spoken corpus was taken from the following sources: (i) published transcriptions of spoken texts from books on oral history, (ii) recordings made by myself during a stay in Edinburgh in 1992, (iii) recordings and transcriptions of recordings made available to me by the School of Scottish Studies at Edinburgh University. In addition, a few examples were added which were collected ad hoc in pencil and notebook fashion. Such examples were, however, not included in any of the quantitative analyses. The Appendix contains a detailed description of the database and, in the case of the spoken texts, where available, additional information on the speakers.

1.6.3. Methods of analysis The database was analysed manually. This was preferred to the timeconsuming process of entering onto computer and subsequent checking of the spelling which could not have been done automatically because of the great

20

Part I. Scope and methodology

variety in the spellings of the same words. While a computer would have provided more accurate statistics and would have been superior in providing a concordance of all instances of a given lexical item, it could, of course, only do this if it were supplied with all the relevant information, i.e. all variants of spelling occurring in the data. Moreover, unless the texts were tagged first, the concordance would list homonyms, such as the adverb and preposition since beside the subordinating conjunction (henceforth subordinator), and it would not be able to deal with zero-categories. While statistical information is relevant for the present study, it must be borne in mind that the number of occurrences of certain items depends on so many factors, such as individual style, topic, repetitions etc. that even in a much larger corpus differences of a fraction of a percent, which computer statistics could supply, would be irrelevant. The approach to the description of the data was guided by the following requirements: (i)

It had to be able to deal with a semantically and structurally heterogeneous grammatical category, (ii) It had to be applicable to Scots as well as English, (iii) It had to be suitable for written as well as spoken material. The categories established in the analysis are illustrated by examples from the database, followed by Standard English glosses in inverted commas, as not all readers may be familiar with Scots. Where it seemed helpful, additional information is provided in square brackets. All source texts are denoted by abbreviations, the key to which is given in the Appendix. To enable the reader to distinguish between written and spoken texts, the abbreviations relating to written texts are marked by italics, while abbreviations relating to spoken texts (both published and unpublished) are given in Roman type. Unpublished spoken texts are transcribed as orthographic sentences, which is the form in which the published spoken texts appear. In cases where intonation and structure are in conflict, the intention of the speaker will be respected. This means that afterthoughts which are grammatically marked as subordinate will be treated as forming one complex with the preceding clause(s) and not be treated as separate sentences even if the preceding clause finishes with a sentence-final intonation. On the other hand, finite clauses which lack an overt subordinator will be treated as subordinate rather than as a juxtaposed main clause, if the prosody clearly suggests such an interpretation. The spelling of the texts has not been standardised, which means that, for example, to will be found beside the traditional Scots spelling tae.

2. Adverbial clauses: Definitions and classifications

2.1. The syntactic status of adverbial clauses in grammatical theory There is general agreement in both traditional grammar and contemporary grammatical theories that adverbial clauses are dependent clauses with a modifying function. However, grammarians up to the turn of the century commonly held the view that subordinate clauses were not part of the superordinate clause, but formed a complex together with it (cf. Sweet 1891 [I960]: 161-171). This view has been totally abandoned for nominal clauses in contemporary grammatical theory and to a varying degree for adverbial clauses. Clauses realising obligatory clause constituents, such as subjects, are now generally assigned the same syntactic status as their phrasal counterparts and clauses modifying noun phrases are treated as part of the respective noun phrases. The syntactic status assigned to adverbial clauses varies considerably in contemporary grammatical theories. The two opposing positions are held by mainstream generative theory, which treats adverbial clauses on a par with adverbial prepositional phrases and thus as part of the superordinate clause and by systemic functional grammar, which draws a major distinction between nominal, relative, and comparative clauses on the one hand and adverbial clauses on the other. In functional systemic grammar two dimensions are relevant for the description of the relation between clauses in clause complexes:7 the dependency relation and the logico-semantic relation. With respect to the first dimension Halliday distinguishes between parataxis and hypotaxis. Parataxis is defined as the linking of elements of equal status, and hypotaxis as the linking of elements of unequal status, i.e. a dominant and a dependent element. The term "embedding" is used for clauses which are not only hypotactic, but also part of a superordinate construction. On the logico-semantic dimension, "projection" and "expansion" are distinguished as major categories, with the subcategories "locution" and "idea" (projection) and the subcategories "elaboration", "extension" and "enhancement" (expansion) respectively. The position taken by adverbial clauses in this system is described as follows (Halliday 1994: 235-236): "The combination of enhancement with hypotaxis gives what are known in traditional formal grammar as 'adverbial clauses' ... They may be finite or non-finite."8 In generative grammar the categorisation of adverbial clauses rests on the category assigned to adverbial subordinators as either complementisers (i.e. they are grouped together with subordinators introducing obligatory clauses, such as that and whether] or as prepositions, which is the position now held by

22

Part [. Scope and methodology

most generativists. In the first case the adverbial clause would appear in S' position; in the second case it would be an S'-complement with obligatory deletion of COMP [complementiser]. A third option is presented by Emonds (1985), who suggests a common category for prepositions and complementisers. A different approach to the categorisation of adverbial clauses is taken by Haegeman (1985), who distinguishes three different levels of centrality in adverbial clauses: adverbial clauses on V" level, which are part of the verb phrase, adverbial clauses on S level, which are clause constituents, and adverbial clauses on the E level (expression or utterance level), which are on a level above the sentence. According to Haegeman (1985: 37), the E node establishes "the link between Sentence and Text". The adverbial clauses on the verb phrase level are manner clauses, the adverbial clauses on the S level are central adverbial clauses, such as clauses of time, and the adverbial clauses on the E level are peripheral adverbials, such as speech-act adverbials and adverbial clauses expressing inferences. Functional typology takes an intermediate position. Functional typologists generally recognise a hierarchy of structural dependence, but while there is general agreement that adverbial clauses are dependent clauses, there is no agreement as to their degree of dependence or integration, nor is there in the terminology used to denote these different degrees. The gradience of clause integration and the position assigned to adverbial clauses, as well as the respective terminology used is illustrated in Table 1: Table 1. The gradience of clause integration in functional typology independent

+ dependent - embedded

+ dependent + embedded

Foley—Van Valin (1984)

co-ordination9

co-subordination subordination non-finite adv. clause finite adv. clause

Hopper—Traugott (1993)

parataxis

hypotaxis subordination finite adv. clause non-finite adv. clause

Lehmann (1988)

parataxis

hypotaxis finite adv. clause

embedding non-finite adv. clause

What comes nearest to a consensus view is presented in Thompson—Longacre (1985). The authors describe both finite and non-finite adverbial clauses as

2. Adverbial clauses: Definitions and classifications

23

sentence margins which modify the sentence nucleus. The nucleus may be one clause or a combination of clauses. They distinguish different degrees of centrality without, however, stating explicitly which adverbial clauses belong to which category or whether modifying implies embedding.

2.2. A function-based definition and classification of adverbial clauses The basic syntactic function of the heterogeneous category "adverbials" is that of a modifier. Its modifying function extends to all levels of constituents from individual words to sentences, with the exception of nouns, pronouns, and noun phrases. Adverbial clauses are thus a structural subcategory of adverbials, whose functional diversity they share to some extent. The term "adverbial clause" is sometimes used in the literature in a wider sense, including also obligatory clauses functioning as predicative complements which typically denote semantic concepts such as time, place or direction (cf. Quirk et al. 1985: 2.15). As their syntactic function is that of predicative complement rather than modifier, the term "adverbial clause" is restricted in the present study to optional subordinate clauses modifying a clause or the predication of a clause, whereby "predication" is understood as comprising all obligatory clause constituents other than the subject and auxiliaries expressing modality, aspect, or tense.10 A difference in centrality between different types of adverbial clauses has been recognised by linguists from various linguistic schools. They have accordingly established subcategories of adverbial clauses along a centrality gradient. These subcategories are based on different criteria ranging from logical semantic, cognitive, and pragmatic ones to syntactic ones. Table 2 illustrates the great diversity in the systems of subcategories, which exists on two levels: (i) the number of categories established, (ii) where the divisions between the categories are drawn.

24

Part I. Scope and methodology

Table 2. The centrality scales of adverbial clauses in different linguistic theories

Quirk et al. (1985)

Predication adjuncts manner, place

Sentence Sentence adjuncts content disjuncts place, time, concession reason

Sentence style disjuncts

Haegeman (1985)

VP-adverbials manner

S-adverbials time, place

E-adverbials

Sweetser (1990) Hengeveld (1993)"

content adverbials

Epistemic Speech-act adverbials adverbials

Zero order Second order Third order manner simultaneity cause concession

Fourth order reason explanation

There appears to be almost general agreement with respect to the two endpoints of the centrality scale (with the exception of Hengeveld's zero order): manner adjuncts as the most central adverbial clauses and speech-act adverbials and adverbials expressing inferences as the most peripheral. With respect to the intervening categories, there is practically no correspondence. Thus Hengeveld's second order and third order categories do not correspond to the sentence adjuncts and sentence disjuncts of Quirk et al. respectively. It seems probable that there is so little agreement because there are no clear divisions, as is indeed suggested by Quirk et al. (1985: 2.15). In other words, we have to recognise a gradient between more central and more peripheral adverbial clauses, as sketched in Figure 1, using the terminology of Quirk et al. (1985):

—r

Predication content

Clause content

arliimrt

nrliimnt12 /

ΓιΡΛΠΤΡΝΤ

Figure 1. The centrality scale of adverbial clauses

·>

l ^ujjili^iai

Clause content

Clause style

Hi locative subordinators => ators modal subordinators => modal subordinators => ators temporal subordinators => ators

temporal subordinators causal, conditional or concessive subordintemporal subordinators causal, conditional or concessive subordincausal, conditional or concessive subordin-

Such a development can be perceived in the history of English while, since, and as, German weil and während, Danish siden and medens, or French comme. However, it should be noted that unidirectionality does not imply that a partial reversal of the process is impossible. As is illustrated by the semantic development of German seit, a derived meaning (in this case a causal one) may be lost again and only the original, in this case temporal, one survive.85 The semantic composition of the subordinator inventory of a dialect can be expected to share general linguistic tendencies. However, with respect to tendencies that relate to the use as a written language, a dialect, which is certainly more used as a spoken medium, should lag behind. Therefore a

4. Variability of patterns

223

comparison of the semantic composition of Scots with Standard English can serve as a test-case for the third of Kortmann's tendencies. Out of the 68 adverbial subordinators introducing finite clauses in presentday Standard British English (see section 4.2.2. above),86 21 introduce temporal clauses, 15 conditional clauses, 6 concessive clauses, and 10 clauses of reason. Out of the 60 adverbial subordinators introducing finite clauses in present-day Scots, 21 introduce temporal clauses, 9 introduce conditional clauses, 8 concessive clauses, and 10 clauses of reason. While the proportion of temporal clauses is slightly higher in Scots, the same is true of the percentage of clauses of reason, and even more so of clauses of concession, whereas the percentage of conditional clauses is noticeably lower. The results of the comparison are not conclusive. They neither confirm Kortmann's third tendency nor contradict it. At best, it can be said that the causal-conditionalconcessive network appears to be fairly heterogeneous. It seems therefore appropriate to compare these figures with those of other languages. Table 14 shows the total number of adverbial subordinators, as well as the number of the respective subordinators of time, reason, condition and concession for presentday English, Scots, Danish, Icelandic, Dutch, German, French, Latin, and Gaelic. The languages selected for comparison are those described in 5.2.1., with the addition of Latin and Icelandic, chosen as ancestor of French and more conservative relative of Danish respectively. The figures for languages other than English and Scots are based on Kortmann (1994).87 Table 14. Proportional Relationships of subordinators of time, reason, condition, and concession in a selection of European languages Language

Total number of subordinators

English Scots Danish Icelandic Dutch German French Latin Gaelic

68 60 45 47 55 54 82 48 24

Time

Reason

Condition

Concession

21 21 16 17 15 17 30 12 5

10 10 8 7 7 7 17 8 3

15 9 10 5 10 11 13 11 3

6 8 6 2 7 10 5 4 2

In this cross-linguistic comparison the Scots figures for adverbial subordinators denoting reason, condition and concession are certainly not lower than those of the other Germanic languages on average. What is more surprising is the difference in the proportional relationship of concessive subordinators. For the

224

Part III. Analysis

majority of the languages under investigation the figures for concessives are the same or slightly lower than those for subordinators denoting reason. They are, however, much lower in French and Icelandic (roughly a third) and considerably lower (half) in Latin. According to Kortmann (1997: 132), French has a particularly high rate of monofunctional subordinators, which certainly accounts for the fact that the total number of subordinators in French is exceptionally high. Nevertheless the figure for concessive subordinators in French is not different from that of the majority of the other languages analysed. This raises the question whether the differences in proportion are related to the question of polyfunctionality. Are the absolute figures for concessives more alike because concessive subordinators form a more uniform category than other subordinators? Is there a difference between concessive and other subordinators in the relationship between monofunctional and polyfunctional subordinators? Table 15 compares the proportional relationship between semantically monofunctional (MF) and polyfunctional (PF) subordinators for the categories of time, reason, condition and concession. The term "polyfunctional" is here taken in a fairly narrow sense, including polyfunctionality within the microstructure of semantic networks. Table 15. Proportional relationship between polyfunctional and subordinators of time, reason, condition, and concession

monofunctional

Language

Time

English Scots Danish Icelandic Dutch German French Latin Gaelic

PF 10 10 10 5 7 11 14 9 3

MF 11 11 6 12 8 6 16 3 2

Reason

Condition

Concession

PF MF 7 3 3 7 5 3 2 5 5 2 4 3 6 11 4 4 1 2

PF 2 4 2 4 3 6 3 7 0

PF MF 3 3 1 7 1 5 2 0 1 6 2 8 0 5 1 3 0 2

MF 13 5 8 1 7 5 10 4 3

Table 15 illustrates the fact that the Germanic languages have predominantly monofunctional concessive subordinators. Apart from Icelandic, English is the only exception. In English whereas, while, and whilst may in formal registers be used as subordinators of concession, which seems to be an extension of their contrastive function.88 That the number of concessive subordinators is highest in German is due to the coexistence of obwohl, obschon, obgleich, and obzwar. This points to the general importance of formation patterns for the semantic composition of adverbial subordinator inventories. The productivity of a certain formation pattern may considerably increase the percentage one

4. Variability of patterns

225

semantic category takes of the total inventory without, however, affecting its frequency of use.89 The coexistence of four similar concessive subordinators does not imply that concessive adverbial clauses occur more frequently in German. Other figures that require an explanation are those for temporal subordinators in English and Scots, which are higher than in the other Germanic languages. It appears that this is largely due to comparatively new monofunctional subordinators. In Standard English these are reanalysed adverbs ending in -ly (immediately, directly, and instantly), which express immediate anteriority, and subordinators originating in adverbial phrases followed by a restrictive relative clause: the minute (that), the moment (that), by the time (that). Only the latter type is also found in Scots, where it seems to be particularly frequent. This formation pattern is, of course, not new in the sense that it had not existed before. English while and its Dutch and German cognates wijl and weil go back to this pattern. It is, however, new in the sense that it had not been productive as a formation pattern for temporal subordinators in any of the languages under investigation since the Middle Ages, and it seems that it had been dormant in English and Scots as well, as Jespersen's first example of this type dates from the beginning of the nineteenth century, while the immediately type dates from the end of the eighteenth century (Jespersen [1965] V: 21.2.7.- 21.2.8.). With respect to conditional clauses the subordinator systems appear to be as diverse as for concessive clauses. In some of the investigated languages the proportion of monofunctional subordinators is as high as it is for concessive clauses (Danish), in others the difference is less significant (German), and in Scots the figures are almost equal. The total number of conditional subordinators is exceptionally high in English, compared to the other Germanic languages, while that of Scots is roughly the same. Again, it seems that the unusually high number in English is due to the productivity of one formation pattern: that of reanalysed participles, in particular present participles. In the case of provided and providing present and past participles coexist with exactly the same meaning. In Scots this pattern is much more restricted, with only two conditional subordinators of this type, compared to seven in Standard English. Another fact that accounts for the different patterns is the high number of polyfunctional conditional subordinators in Scots, which are all subordinators of time as well as of condition. As has been pointed out in chapter 3, the division of the temporal-conditional space is different in Scots and English. There is a tendency in Scots for habitual, generic and potential adverbial clauses to be introduced by subordinators whose primary semantic role is that of a temporal subordinator, while in Standard English this is only the case with habitual adverbial clauses, generic and potential adverbial clauses being more frequently introduced by if, whose primary semantic role is that of a conditional subordinator.

226

Part III. Analysis

As a possible explanation for this difference, the coexistence of the two etymologically different homophonic subordinators gin in Scots has been suggested (cf. section 3.3.6. above). As Thompson—Longacre (1985: 193) point out, not all languages have a distinction between if- and when-c\auses. It seems that the distinction between the domains of //and -when was less clearcut in earlier stages of the English language, as Shakespeare could still use when in a hypothetical conditional clause:90 (510) and what a thing should I have been when I had been swelled (Shakespeare, The Merry Wives of Windsor, III, 5, 14; Franz 1939: 435; my italics, MH) 'and what a thing would I have been if\ had swollen up [as a corpse after drowning]' The distinction is particularly frequently neutralised for predictive clauses and future reference. Such a partial neutralisation is also found in German, which uses als for specific past situations and wenn for recurrent past situations, the latter also serving as a subordinator denoting overlap of situations located in the present or in the future. Similarly, Dutch uses toen for specific past situations and nu for present situations and recurrent past situations, Danish uses da for specific past situations and när for present situations and recurrent past situations, and the same distinction is also found in Old English, where pa is restricted to specific past situations. In German the distinction was maintained even when the subordinator denoting overlap for specific past situations in Middle High German, da, lost its temporal meaning. It was gradually replaced by als, which eventually ousted it in the nineteenth century (Paul 1920, IV: 228, 239-240). This is different in English, where this function was lost from the language with the loss of the subordinator pa. I suggest that this is no coincidence, but is determined by functional needs. The Old English subordinator denoting this specific function was not replaced because the language had other means of distinguishing between the two functions. Iteracy can either be expressed by means of an adverbial or by marking the verb phrase as iterative. The first strategy is the origin of the Modern English subordinator whenever, the second one of the modal use of used to and would. The use of would as a marker of iteracy arose in early Old English (Mustanoja 1960: 599); that of used to is a later development. For Chaucer the primary marker of a habitual past situation seems to have been the modal rather than the subordinator, and if could still be used to denote habitual action: (511) She wolde wepe, if that she sawe a mous Caught in a trappe, (my italics, MH) 'She used to cry whenever she saw a mouse ...' (Chaucer, Canterbury Tales, A 144; Eitle [1970]: 77; Jespersen [1965] V: 19.1.6.).

4. Variability of patterns

227

The hypothesis that the loss of a specific subordinator indicating simultaneity of specific past events in English is closely linked to the development of a modal denoting iteracy is further supported by the fact that there are no modals denoting this function in German, Danish or Dutch.91 Scots prefers would over used to as a marker of habitual past. This suggests that Scots in its division of the temporal-conditional domain is not only more conservative than Standard English, preferring the modal that was current in Chaucer's time to denote habituality, but also takes a middle position between Standard English and other Germanic languages in its subordinator use, the major division being that between non-factual situations and specific past ones. The semantic composition of the Scots adverbial subordinator system suggests that the proportional relationship between temporal subordinators and those denoting a reason, condition and concession may partly follow Kortmann's third tendency, but that other factors also play a major role. These are (i) the availability and productivity of certain subordinator patterns, which may lend themselves more easily to one semantic category than another, (ii) grammatical features of the linguistic system outside the system of subordinators itself, such as mood, modality, tense and aspect, and (iii) the time that has elapsed since the latest restructuring of the system, which may account for the French system being different from that of its neighbours.92 4.4.2. Cognitive semantics and the predictability of subordinator meaning It has been pointed out above that polysemy patterns are to some extent predictable. Thus subordinators whose basic meaning is to express anteriority or a terminus a quo will frequently develop the additional function of denoting reason (Kortmann 1997: 189). Cognitive semantics accounts for such polysemy patterns by claiming the existence of cognitive relatedness between the two meanings. Cognitive relatedness is either intuitively accessible or may be established by cross-linguistic studies on the assumption that meanings must be cognitively related if the same pattern of semantic polyfunctionality reoccurs in many genetically and areally unrelated languages (cf. the work of Haiman on conditionals, in particular Haiman 1974, 1978, 1986, and Traugott and Sweetser's application of this concept to grammaticalisation, in particular Traugott 1982 and 1988 and Sweetser 1988 and 1990). It is important to relate semantic polyfunctionality of subordinators to the lexical material subordinators consist of. Thus the polysemy pattern anteriority/term/>jMs a quo exists in the Germanic and Romance languages only for subordinators that are derived from temporal adverbs and prepositions (e.g. English since and now that). Other source categories for subordinators of reason are nouns denoting 'reason', 'means' or 'way', such as English because, or

228

Part III. Analysis

participles, predominantly of verbs with the basic meaning 'see', such as English seeing that (cf. Handke 1984: 125). The basic meanings in connection with the formation structure determine how easily subordinators develop additional meanings and which meanings are likely to be derived from them. Thus a subordinator denoting a total overlap of two events will easily develop into a subordinator of condition if it is of the so long as type. Examples are German solange, Danish sälcenge, English as long as, while a subordinator whose basic meaning is all the time that is most likely to develop into a subordinator denoting contrast, such as English while, and less frequently into a subordinator denoting reason, such as German weil, which, although now exclusively denoting reason, retained its temporal meaning well into the Modern German period, as can be seen in the well-known lines from a German song: (512) Freut Euch des Lebens, weil noch das Lämpchen glüht, (my italics, 'Enjoy life, as long as the light [of life] is glowing.'

)

The difference between the so long as type and the all the while that type is that the first one lends itself easily to the extension only so long as, which can easily be interpreted as a condition, while this is not the case with the second type. In Scots as long as is much more frequent as a subordinator of condition, with only one example of the temporal use in the corpus. As Scots uses phrasal subordinators like the minute, a likely substitute for temporal as long as is all the time (thai), which makes use of the same formation pattern as the minute (that). Scots frae, fae, faem and from derive from the formally identical prepositions. From was used as a temporal subordinator denoting a terminus a quo in Middle English (Mustanoja 1960: 284; Jespersen [1965] V: 21.2.4.). Its survival in Scots may be due to reinforcement through contact with languages which have the same formation pattern (Scandinavian fra and Gaelic (bh)o). If we consider external influence as a decisive factor, the etymology of the subordinator suggests that Scandinavian had a major impact, as the predominant form in Scots is that derived from Scandinavian fra and not from Old English from, which developed into modern English from (Mustanoja 1960: 384). Both fra and til are still used as temporal subordinators in Danish (Hansen 1967: 415-417), especially in combination, as in the following standard imprint on receipts from the Odense University selfservice restaurant: (513) Abningstider: Alle dage^ra vi äbner /// vi lukker. (my italics, MH) Opening times: every day from [when] we open until [when] we close.'

4. Variability of patterns

229

There is also one instance offae 'from' as a subordinator of reason in the spoken Scots corpus. This use is not surprising, as it corresponds to one of the most frequent polysemy patterns of subordinators (cf. Kortmann 1997: 189). Moreover, the preposition from had a causal reading until Early Modern English (Franz 1939: 375). Till is not only a temporal subordinator in Scots, but also implies purpose. Although as a subordinator of purpose //'// seems to be losing ground, its existence calls for an explanation. This polysemy pattern is comparatively rare in the European languages (Kortmann 1994, II: 142-144; 176-179). Other subordinators of purpose formally identical or closely related to temporal subordinators denoting a terminus ad quern are Danish til (cf. Hansen 1967: 416), Latin dum, donee and quoad (Rubenbauer—Hofmann 1975: 309-311), Gaelic gus gun (Calder [1990]: 317-318; Gillies 1993: 218), as well as Italian affinche (cf. Kortmann 1997: 363). The main difference between Scots till and Standard English //'// is that in Scots its meaning is spatial as well as temporal, meaning 'to a place' as well as a 'to a point in time'. A comparison of the lexical material incorporated in the other subordinators with the same polysemy pattern shows that the same applies to Danish /// (Allan—Holmes— Lundskaer-Nielsen 1995: 420), Gaelic gu (Calder [1990]: 297; Gillies 1993: 183), and Latin quoad which is literally translated 'whereto-at', while the etymology of dum and donee (Old Latin donicum) is obscure. In these the second or last element respectively are an accusative singular masculine form of a pronoun, while the first element is explained as a noun denoting duration or an adverb (Rubenbauer—Hoffmann 1975: 309). However, the meaning of Gothic du appears to be 'to' (Kortmann 1994: 177), which suggests that the du- and do- in the Latin subordinator may go back to the same root. If this were the case, all the subordinators in question contain a spatial preposition of the meaning 'to a place'. According to Traugott (1982), the English and Danish prepositions go back to the general Germanic noun tilam 'goal', which is also the meaning of Latin finis, the main lexical source of Italian affinche. This suggests a similar extension of meaning as that noted for from: the meaning 'to a spatial goal' develops into 'to an abstract goal'. It seems thus essential for the development of the meaning of purpose that the spatial meaning is not lost, as purpose is not a likely extension of a purely temporal meaning. Another subordinator that is practically absent from present-day British Standard English,93 but which occurs in Scots is^or all (that), which expresses concession and restriction. It was common in the concessive sense in Early Modern English (Franz 1939: 457; Partridge 1969: 151), but appears to have dropped out of use in the second half of the nineteenth century. The first entry in the OED (entry for A. prep. 22. b.) dates from 1523 and the last one from 1866. The restrictive use (OED entry for A. prep. 26. b.) seems later (the first entry dates from 1731) and rarer and seems to have replaced an earlier for

230

Part III. Analysis

aught and for what in this function. According to Mustanoja (1960: 382), for aught was used in the concessive sense in Middle English. Although Mustanoja does not mention the restrictive sense, the example from Chaucer given as an example of a non-concessive sense (Mustanoja 1960: 382, footnote 1), shows a use of for aught which corresponds exactly to that of the later restrictive subordinatoryör all (that): (514) For aught I woot, he was of Dertemouthe. (Canterbury Tales A Prol. 388; my italics, MH) Tor all [= as far as] I know, he was from Dartmouth.' According to Kortmann (1997: 198), the combination of concession and restriction is not a frequent polysemy pattern. Concessive subordinators are most frequently derived from conditional ones, through the intermediate stage of concessive condition, i.e. the typical development of meaning extension is: if => even if => even though => although (cf. König 1985a, 1985b). This raises the question how the two uses of for all (that) are related. The two meanings are mutually exclusive: 'As far as I know' cannot imply 'although I know'.94 This suggests that in this case there is no primary meaning although that is extended to include as far as or vice versa, but rather that the lexical material incorporated in the subordinator allows extension in two directions, that of concession and that of restriction. For is most frequently incorporated in subordinators denoting cause or purpose. However, there is also a less frequent meaning offor, that of'in proportion to, considering' (OED entry for, A. prep. 27) which may shade into concession. Consider the following example from Richardson (Grandison I.ii.6), cited in the OED: (515) A man of an excellent character for a Lawyer [sic] (my italics, MH) Here the meaning of for 'considering that he is' can easily be extended to 'although he is'. This suggests that it is the proportional reading of for that developed into the notion of concessiveness. An alternative explanation is given by Burnham (1911: 114): "whenyor is accompanied by eall the contrast between the 'cause' referred to and its 'ineffectiveness' becomes more explicit, and the meaning of the preposition shifts to 'in spite of." Franz (1939: 456) attributes a reinforcing force to all, but does not claim that all determines the concessive character of the sentence, which he derives from the local meaning of the preposition for. something that is 'in front' may be 'in the way' and hence present an obstacle (Franz 1939: 374). Handke (1984: 114-116), who discusses the use of English all and German all or words containing all, such as albeit and allerdings 'though' in concessive expressions, sees the origin of all in concessive contexts as an emphatic assertion, which serves as a device

4. Variability of patterns

231

"to minimise a possible conflict with the interlocutor." This is in line with the use of al as a concessive subordinator in Middle English, which derives from a zero-introduced concessive clause, in which al had the function of an intensifying adverb (Mustanoja 1960: 468). Subordinators containing words meaning 'all' are also found in Greek, as well as Slavic and Romance languages (Kortmann 1994, II: 167-172), which suggests that a concessive reading is more frequent with all than wither among the European languages, but that in the subordinator for all (thai) the concessive meanings of both for and all are combined. The restrictive meaning of for all thai is more difficult to explain. It is related to the notion of degree, but the lexical meaning of all seems at odds with "restrictive", as the latter seems to imply a limited degree. However, a limited or restricted knowledge may at the same time be a person's whole knowledge. The importance of the notion of degree for restrictive clauses can be seen in the alternative formation pattern for subordinators introducing restrictive clauses in Germanic and Romance languages, which derives from clauses of comparison, such as English as far as, German soweit and French autant que. While polysemy in the meaning of the preposition for explains the coexistence of concessive and restrictive for all (thai), it does not explain the regional restriction. Two factors seem to play a role here: (i) The concessive meaning of for is of greater importance in Scots than in Standard English, as is evident from the concessive subordinators listed in Grant—Dixon (1921: \10):for all, for a', for a' as, for as ... as. (ii) For all that is derived from a prepositional phrase containing a restrictive relative clause. The main relative pronoun for restrictive relative clauses in Scots and American English, as well as a great number of English dialects, is that.95 That as a relative pronoun was, however, replaced in British Standard English by wh- pronouns, probably under the influence of Latin. In Addison's famous "Humble petition of who and which", written in 1711, the two pronouns make the following historically ill-informed complaint: "We are descended of ancient families, and kept up our dignity and honour many years till the jacksprat that supplanted us." (cited in Jespersen [1965] III: 4.3.2.). The humble petition seems to have been extremely successful, as that, which was the most frequent relative pronoun in Early Modern English, was outnumbered by far in Jespersen's ([1965] III: 4.3.5.) frequency analysis of the relative pronouns in eight nineteenth-century texts. The proportions of who, which and that were 182: 212: 125. It is worth noting that the two authors avoiding which were a Scot (Stevenson) and a Lincolnshireman (Tennyson).96 It seems probable that the avoidance of that had a retarded effect on subordinators incorporating relative that. This may also explain why two other subordinators originally incorporating relative that, everywhere (that) and anywhere (that) are restricted to Scots, and were pre-

232

Part III. Analysis

dominantly used by non-English authors in the nineteenth and the early part of the twentieth century, judging from Jespersen's examples ([1965] III: 8.6.5.). A number of Scots subordinators are restricted to Shetland. These are aless introducing an infinitive clause of exception, and da wye dot introducing a finite clause of reason. Robertson—Graham (1952: 18) list in addition whaarsay 'as if and whaarby 'whereas', of which the latter occurs in Oreström (1986: 99-100), but neither in my data. The history of Shetland, where Norn was spoken until fairly recently, raises the question whether these are borrowings or loan translations from Norn or whether they are Scots subordinators which simply survived longer in Shetland due to its geographical isolation. The lexical input is certainly Scots, as aless is an aphetic form of unless, which is also used as a preposition denoting exception in Shetlandic (Robertson— Graham 1952: 21). According to the OED (entry unless 4.b.), unless was used as a subordinator introducing finite clauses of exception in seventeenth century England. The infinitive construction may reflect Scandinavian influence, as exception is expressed by infinitive clauses in Scandinavian languages (e.g. Danish uden at). This is, however, by no means certain, as /o-infmitives are also found with but and except introducing clauses of exception in the nineteenth and twentieth centuries (Jespersen [1965] V: 14.3.5.). Da wye dat is not only absent from Scots dictionaries, but also from Robertson—Graham 1952. Icelandic has a subordinator of reason that incorporates the lexical element 'way': vegna pess ad (Einarsson 1945: 176; Kress 1982: 247). The internal structure of the subordinator is, however, different from the Shetlandic one: it derives from the preposition vegna. Causal prepositions of the same formation pattern are also found in German (wegen) and Dutch (wegens), and in nonstandard Southern German a subordinator is derived from the preposition, wegen dem daß, which corresponds in structure to the Icelandic one. This suggests that words originally denoting 'way' may be extended to denote 'reason' in the Germanic languages and that two different subordinator formation patterns coexist: (i) a preposition is derived from the noun, which then functions as source category for a subordinator (Icelandic and dialectal German), and (ii) the subordinator is a reanalysis of a noun phrase followed by a relative pronoun (Shetlandic). The structure of the Shetlandic subordinator is more likely to be of Scots origin, corresponding to the well-known pattern of the time that 'when' and the place that 'where'. The etymology of whaarsay 'as if seems to be a combination of'where' and 'say'. However, it is difficult to see how the meaning 'as if could be derived from this combination. I propose an alternative, which needs, however, to be tested against historical data. In Middle English and Early Modern English there existed a subordinator as who say with the meaning 'as if (Visser 1966: 922-923). According to Visser this subordinator may be related to Old French come qui die, which developed into Modern French comme qui dirait. The northern equivalent of

4. Variability of patterns

233

the English subordinator is as hwa say, which is in one instance spelled as one word askwasay. If this subordinator had undergone the same kind of reduction as as soon as in North-Eastern Scots (that is seen's) it would have been reduced to whasay. This means that the origin of the subordinator would most likely have become obscure to the majority of speakers, particularly as in Shetlandic the relative pronoun is at, whereas who is only used as an interrogative pronoun (Robertson—Graham 1952: 5-6). It seems possible that folk etymology related whasay to the subordinator whaarby and accordingly inserted an etymologically incorrect /r/ and changed the spelling to whaarsay. The weakness of this theory is, of course, that in Shetlandic, unlike Southern British English, postvocalic /r/ is pronounced. However, if we consider other cases of folk etymology, such as the well-known case of dialectal sparrow grass for asparagus, this theory may not be too far-fetched. This would mean that Norn had no or very little influence on the subordinator system of Shetlandic. This is not surprising given the rapid decline of Norn and the fact that its grammatical structure appears to have become Scots at a time when its vocabulary was still largely Norn (Barnes: 1984: 355-357), the one exception being the retention of the second person singular personal pronoun. A subordinator that is not exclusively Scots, but much more frequent in Scots than in Standard English, is and, which introduces verbless clauses. Its meaning may be either that of an emphatic subordinator of concession or an emphatic subordinator of reason, but may also stand in a concessive relationship with respect to the immediately superordinate clause and a causal one with respect to the main clause. In clauses of reason and implies that the content is well-known to the addressee, while in concessive ones it implies in addition a sharp contrast with expectations. As an emphatic subordinator introducing clauses of reason it is mentioned by Franz (1939: 471-472) for Early Modern English and as a concessive subordinator by both Franz (1939: 458) and Partridge (1969: 151), but it is already found in Middle English (cf. Hacker 1994b). The use of emphatic and increased dramatically in the second half of the nineteenth century for reasons as yet unknown. As this use goes against our notions of and as the prototypical co-ordinator, it is largely neglected by grammars. However, this phenomenon is not restricted to Scots or English. An emphatic use of the prototypical co-ordinator, which likewise seems to imply contrast to expectation, is also known in Danish: (516) Og jeg som stolede pä ham! (Allan—Holmes—Lundskaer-Nielsen 1995: 455) 'And I who trusted him!' Similarly in post-classical Latin et could imply emphatic contrast. A famous example first appeared in the seventeenth century under a skull in a picture of a

234

Part 111. Analysis

pastoral scene by the painter Giovanni Francesco Barbieri. The same phrase appears in Poussin's painting "Les bergers d'Arcadie", where it seems to represent the presence of death even in an idyllic pastoral environment. (517) Et in Arcadia ego! 'Even I [Death] am in Arcadia/But I am in Arcadia, too' A Latin use of et that even more closely resembles the Scots subordinating and occurs in a high medieval chronicle from Andes in Artois (Heller 1879: 705), where the author complains about the bodily defects of people who had been accepted as monks: (518) Quidam enim claudi, quidam contract!, quidam monoculi, quidam strabones, quidam ceci, quidam vero manci inter eos apparebant, et hii fere omnes genere nobiles existebant. 'For some were obviously lame, some crooked, some one-eyed, some had squints, some were blind, some were crippled, and [= although] almost all of them of noble descent.'' [My translation, MH] According to Grevisse (1975: 1077), this use of emphatic et is also attested in French. German uses wo ... doch as a subordinator introducing emphatic clauses of reason or concession. Engel (1988: 269) states: Vorwiegend in der gesprochenen Sprache werden Kausalsätze auch mit dem subjunktiven Element wo eingeleitet. Dieses wo verlangt gewöhnlich zusätzlich die Abtönungspartikel doch, die Zustimmung heischende Bedeutung hat. [Predominantly in the spoken language causal clauses are introduced by the subordinating element wo. This use of wo typically requires the modifying particle doch, whose function (literally "meaning") is a request for confirmation.] [My translation, MH]. He adds that this clause type normally follows its matrix clause. Behagel (1928: 351) mentions the concessive use of wo (doch) in Southern German dialects. That wo ... doch can indeed be concessive and that the causal or concessive reading is derived from the context can be seen in Engel's example, where the meaning of wo ... doch changes from denoting reason to denoting concession if the matrix clause is turned into a question and the adjective brav 'good' is changed into its opposite unfolgsam 'naughty': (518) Die Kinder dürfen heute noch aufbleiben, wo sie doch den ganzen Tag so brav waren. (Engel 1988: 269) 'The children may stay up late today, as they have been so good all day.'

4. Variability of patterns

235

(518) a. Warum dürfen die Kinder noch aufbleiben, wo sie doch den ganzen Tag unfolgsam waren? 'Why are the children allowed to stay up, although they have been naughty all day?' Both sentences have in common the fact that the content of the subordinate clause is presupposed, i.e. known or assumed to be known by the addressee, and that the clause is emphatic. This suggests that there is a pragmatic need for a subordinator that primarily expresses emphasis and whose basic lexical meaning is rather unspecific to allow a context-dependent causal or concessive interpretation or a combination of both. It seems that languages differ with respect to the source category of these subordinators. They may be general relativisers, as is the case in German, which uses the non-standard relative pronoun wo, or co-ordinators, as is the case in Scots, English, Danish, Latin and French. More research is needed in this area, which exemplifies the fact that besides the interplay of form and lexical meaning, pragmatics plays an important role in the development of adverbial subordinators and may overrule other principles, such as iconicity.97 As has been illustrated in this section, regional differences in subordinator meaning are not simply due to differences in the structure or meaning of individual subordinators, but may reflect more wide-ranging differences in the vocabulary and grammar of the respective varieties. Thus it could be shown that the different semantic scope of the subordinator //'// in Scots is based on a difference that already exists in the meaning of the preposition till, and that the higher frequency of for all that may well be related to a difference in the use of relative pronouns. Moreover, the importance of pragmatic needs, which should be seen on a par with other principles, such as iconicity and economy, is illustrated by the case of and and its functional equivalents in Danish, Latin, French, and German.

5. Summary and conclusion

This chapter will summarise the evaluation of the regional differences within Scots and the differences between Scots and Standard English. It will then turn to the question of the prospective development of adverbial clauses in Scots. In conclusion, it will discuss some problems in linguistic theory, for which theoretical claims need to be modified on the basis of the analysis. Finally, potential fields for further research will be pointed out, which have suggested themselves in the course of this study.

5.1. Variation in Scots: Qualitative and quantitative differences Admittedly, a much larger corpus would have been preferable for a study of the regional differences within Scots. However, general tendencies can also be observed in a smaller corpus. While there are considerable differences in vocabulary and pronunciation between rural and urban areas, no such differences can be observed in the use of adverbial clauses. There is not one instance of a subordinator or structural type which occurs in the major cities (Edinburgh, Glasgow, Dundee), but not elsewhere, nor do we find the opposite, i.e. structural types or subordinators exclusively found in rural areas, excepting those which are used by only one writer or speaker. Regional differences within Scotland are comparatively rare and restricted to peripheral areas. Region-specific use of subordinators introducing adverbial clauses is found in Shetland, Caithness, north-eastern Scotland (i.e. from Buchan to the Firth of Tay), and western Scotland, and, in addition, tilinfinitives replace toe-infinitives in Caithness. As pointed out above (4.1.1.), these differences are not only marginal in the sense that they occur predominantly in peripheral areas, but also in that they are of an extremely low frequency. All the differences can be described as lexical differences. There is no evidence of structural differences within Scots. The most important result of the present study with respect to the regional variability within Scots is that the regional divisions based on the evidence of the grammatical category "adverbial clauses" do not correspond to those established primarily on the basis of pronunciation in the SND, where Northern Scots is characterised by features such as the vowel [i] in words like "moon", while Southern Scots is characterised by its diphthongisation of the vowels in words like you and me. The latter pronunciation is certainly much more restricted now than is indicated by the dialect division in the SND. In the Scottish Borders today this feature is primarily associated with the speech

5. Conclusion

237

of Hawick rather than other towns in the area, such as Jedburgh or Selkirk. A division based on differences in the grammatical categories "adverbial clauses" and "adverbial subordinators" suggests different dialect regions: North Insular Scots, Caithness Scots, Northern Scots, Central Scots and Western Scots. It is worth noting that these divisions seem to correspond to those based on historical phonology, as Johnston (1997: 56, 60) defines the "Anglian Core" as "Southern and East Mid Older Scots" and the "Burgh-borne Dialects" as "North of the Tay - Moray Firth", excluding Caithness as not Scots-speaking. The major dialect regions according to the SND and established here on the basis of adverbial clauses are illustrated in Maps 1 and 2. The geographical distribution of adverbial subordinators confirms that the linguistic notion of centre and periphery not only works for large areas (cf. Kortmann 1997 for Europe), but also for comparatively small ones such as Scotland. Thus the peripheral areas of Caithness, the Northern Isles and Western Scotland have a number of subordinators absent from the rest of Scotland. A further point that may be important for the question of the status of Scots relates to the differences between spoken and written Scots. The differences are comparable to the differences found between spoken and written Standard English, which reflect the different discourse situation, and they are in some cases even bigger. This shows clearly that written Scots is not simply a written version of spoken Scots, and thus refutes claims that Scots is only suitable as a spoken medium.

5.2. Scots and Standard English: Qualitative and quantitative differences The differences between Scots and Standard English adverbial clauses are greater than those between different regions within Scotland, with respect to both quality and quantity, and they are of a more diverse nature. The differences can be grouped into three major sets: (i) differences in the inventory of adverbial subordinators, (ii) differences in the clause structure and polyfunctionality patterns of adverbial clauses, and (iii) differences in the distribution of structural types.

(i) Differences in the inventory of subordinators The differences in the inventory of subordinators are not coincidental. Certain types of subordinators are rare or totally absent from Scots:

238

1. 2. 3. 4. 5.

Part III. Analysis

subordinators of the participial type, in particular ending in -ing phrasal subordinators beginning with the prepositions in and on phrasal subordinators derived from correlative structures beginning with so subordinators derived from adverbs ending in -ly subordinators derived from locative -where by adding -ever

The first two of these types are subordinators based on French patterns, which only entered the English language in Late Middle English and Early Modern English (cf. Kortmann 1997: 300, 308-309). The absence of the fourth type is probably based on the fact that -ly never acquired the status of a marker of adverbs in Scots. That this may have been a prerequisite for the development of this type is also strongly suggested by its absence from American English, which likewise marks adverbs less frequently by the addition of -ly than Southern British English does. The adverbs in -ly are modelled on the French adverbs ending in -ment, so that this type of subordinator ultimately may also go back to French influence on the English language. The absence of the other two types cannot be as easily explained. Here it seems that Scots and Standard English have developed different types without any external influence being at work. The preference for initial every or any to final ever may, however, be related to the Scots preference for interrogative adverbs of the type what time, what place, what way, which likewise consist of a sequence of pronoun and in this case lexical head (see 3.1. and 3.2. above). The differences between the inventories of Standard English and Scots subordinators introducing adverbial clauses are greatest in the field of conditional subordinators, where Standard English uses a great variety of participlederived subordinators and phrasal subordinators beginning with in. An additional difference between Standard English and Scots exists in the inventory of subordinators introducing non-finite and verbless adverbial clauses. Here subordinators which also introduce finite clauses are rarer in Scots. This may be described as a conservative feature of Scots, as the introduction of non-finite and verbless clauses by subordinators is a comparatively late development in Standard English (cf. Curme 1931: 177). (ii) Differences in clause structure With respect to finite clauses the following differences can be observed: 1. Apart from its use in conditional clauses and as a marker of habitual past, the modal would is rare in Scots adverbial clauses and should is generally absent. This may be a conservative feature of Scots, as the modals in present-

5. Conclusion

239

day Standard English usage occur where the subjunctive was used in earlier stages of the English language. 2. The present perfect and past perfect are rarer in Scots than in Standard English. It seems that Scots relies more on the use of subordinators to express completion, using in particular once for this purpose. 3. Conditional clauses have a number of patterns in the verb phrase that are less frequent in Standard English or totally absent from it. These are in particular the marking of both protasis and apodosis by would in hypothetical clauses referring to simultaneous situations and the use of the "plu-pluperfect" in the protasis of hypothetical clauses for hypothetical past conditions. 3. Scots uses zero-introduction with finite clauses more frequently than Standard English, which may be interpreted as a conservative feature, as zerointroduced adverbial clauses occurred also in Early Modern English. 4. Amalgamations are found not only in spoken Scots, where they could be explained as slips of the tongue, but also in written Scots. These amalgamations occur where a correlative or co-ordinating construction is a frequently used alternative to the construction containing a subordinate clause, as for example in the combinations of nae sooner with when, or gin with and. This suggests that the boundaries between subordination and co-ordination on the one hand and subordination and correlative structures on the other hand are not as clear-cut in Scots as they are in Standard English. This is likewise a feature that can be found in earlier stages of English. 5. Polyfunctionality in Scots is not restricted to the grammatical categories subordinator and preposition, but also occurs in the categories subordinator and conjunctive adverb. It is found with subordinators denoting a causal relationship in the widest sense (for, because, sae, and though). In present-day Standard English this type of polyfunctionality is only described for so, and to a certain extent for though, where the function as conjunctive adverb is, however, only recognised if though occurs in non-initial position.

(iii) Differences in the distribution of structural types A comparison with Kortmann's Table 9.5. (1991: 141) suggests that there is also a difference in the percentage that the various semantic types take of the sum of all non-finite and verbless adverbial clauses in Scots and Standard English. Table 16 gives the four most frequent adverbial relationships

240

Part III. Analysis

expressed by -ing clauses, -ed clauses and verbless clauses in the spoken and written Scots corpora and Kortmann's corpus of fiction and his spoken corpus consisting of texts from the "Survey of English Usage". Relationships not treated in the present study, such as exemplification/specification are excluded. Table 16. The most frequent adverbial relationships in -ing clauses, -ed clauses, and verbless clauses in Scots and English Spoken Scots

Written Scots

Spoken English

Written English

ace. circumstance reason means temporal overlap

ace. circumstance means reason temporal overlap

reason ace. circumstance temporal overlap condition

ace. circumstance temporal overlap reason anteriority

"Ace. circumstance" is used as an abbreviation for "accompanying circumstance".

These differences seem to reflect different structural preferences: unlike Standard English, Scots rarely uses non-finite or verbless clauses to express conditional relationships, reserving non-finite clauses predominantly for the relationships of accompanying circumstance, reason, and means, which also outnumber temporal relationships both in the spoken and the written corpus. It is worth noting that in other Germanic languages conditional relationships are rarely expressed by zero-introduced non-finite and verbless clauses. Non-finite conditional clauses are typically restricted to formal registers and structurally to subordinator-introduced -ed clauses and verbless clauses (cf. German falls möglich and Danish om muligt 'if possible'). Thus it seems that once again Scots is closer to its Germanic neighbours than Standard English.

5.3. Adverbial clauses in Scots: Possible and probable developments One question to emerge from the present study is that of the prospective direction of changes in Scots adverbial clauses. Do the results of the present study suggest that Scots is moving closer to Standard English? The answer is "yes" and "no". "Yes", if we think of subordinators like ////, which is gradually losing distinctively Scots meanings, i.e. those which fall into the scope of other subordinators in Standard English. The answer is "no", if we think of the development in the temporal-conditional field. When has certainly a wider semantic scope than in Standard English, and there is no indication of a change. The differences between Scots and Standard English in the scopes of

5. Conclusion

241

semantically closely related subordinators, such as once, since, and after likewise do not seem to be affected by any changes. These seemingly conflicting tendencies can, however, be explained: the change towards Standard English generally consists of a loss of overt Scotticisms, while the developments in new directions not shared by Standard English predominantly occur where an overlap or polysemy of subordinators exists in Standard English and Scots, or where words belonging to other word OS classes take over the function of subordinator, as is the case with say. It may be expected that this dichotomy in the development of subordinators will continue. A loss of archaic Scots subordinators is thus extremely likely. The extensions and shifts in the semantic scope of individual subordinators, or new additions to the inventory of subordinators are less predictable, although, of course, Scots can be expected to conform to universal cognitive and semantic principles, which predict that temporal subordinators are likely to develop additonal causal, conditional or concessive readings.

5.4. Adverbial clauses in Scots as a case study in dialect syntax The analysis of adverbial clauses in Scots makes an important contribution to the new linguistic field of dialect syntax. The study confirms that differences between dialect grammar and standard English grammar are indeed "systematic and far-reaching" (Miller 1993: 137). The study has established that differences from Standard English, which viewed through the lens of Standard English (prescriptive) grammar might have been termed "incorrect use of subordinators and tenses", are not unsystematic if they are seen within their own system. Thus the use of whenever in clauses that do not refer to repeated action is correct in Scots, as is its cognate wanneer in Dutch, while it seems odd in Standard English. Similarly the use of would in both protasis and apodosis of hypothetical clauses does not correspond to Standard English grammar, but corresponds to the rule of using the same tense and mood in protasis and apodosis, which is obligatory in Latin as well as many other languages. Moreover, would + infinitive rather than the simple past tense in the protasis may distinguish a hypothetical clause from a clause denoting past habituality in Scots. As in Standard English, would in Scots may denote either a past habitual or a counterfactual situation, but in Scots the same subordinator (cf. Scots gin) may introduce temporal clauses and conditional clauses. This means that within its own system the double marking of counterfactuality is not superfluous in Scots, but serves to avoid ambiguity. Another fact to emerge from the present study is that a dialect as a predominantly spoken variety may use prosodic features to mark subordination of clauses, where the Standard variety relies exclusively on structural markers

242

Part III. Analysis

such as subordinators, non-fmiteness and word order. According to Polikarpov (1996: 154), Russian linguists claim that asyndetic clause linking may be coordinative and subordinative and that this phenomenon is not restricted to Russian, but is also found in other languages, such as German. His defining criteria for subordination include besides the canonical ones, namely the use of subordinators, non-fmiteness and inversion, the following: intonation, fixed order of clauses, the occurrence of a corresponding adverb in the superordinate clause, and the logical relationship and dependence between the respective clauses. (Polikarpov 1996: 157). The debatable Scots zero-introduced adverbial clauses that are not marked by inversion each fulfil all four of these criteria and would thus qualify as subordinate. This suggests that it may well be advisable to reconsider the defining criteria with respect to subordination for the spoken language, as it is predominantly the spoken language that makes use of zero-introduced clauses that are intuitively felt to be subordinate, but do not fulfil the canonical criteria for subordination.

5.5. Adverbial clauses in Scots: Implications for linguistic theory The Scots use of adverbial subordinators illustrates the fact that the adverbial system of an English dialect need not be simpler than that of Standard English. Scots has a greater number of subordinators denoting concession and it has not only a greater number of subordinators denoting immediate anteriority than Standard English (as soon as, once, whenever, the minute versus as soon as, once and immediately) in comparable texts (i.e. fiction and informal face-toface conversations), but there is also a higher frequency in the actual use of subordinators denoting this temporal relationship in the Scots corpora than the respective Standard English ones. This is the more surprising as the level of education of the speakers is higher in the Standard English spoken corpus (LLC), consisting exclusively of academics, while the Scots spoken corpus consists of working-class speakers and academics with a working-class background. Thus the analysis of spoken and written Scots texts establishes a high level of semantic differentiation in the Scots system of adverbial clauses, which refutes claims that relate subordinator systems directly to the degree to which a variety is used in writing or the level of education, unless we assume that the level of education is generally higher in Scotland." On the other hand, Scots makes comparatively frequent use of non-finite -ing clauses and uses an and-c\ause that allows a causal and a concessive reading. It is important to note that this does not imply that the meaning in context is ambiguous or extremely vague. In the cases of -ing clauses there are no examples that allow a temporal as well as a causal reading. It is only with respect to the closely related categories of reason and means that alternative interpretations are possible,

5. Conclusion

243

although even in these cases the two readings are not equally acceptable, but one is more plausible. These clauses are therefore in no way more ambiguous than for example so f/zfrt-clauses, which may likewise allow two interpretations, namely as clauses of purpose or of result. Here, too, the two interpretations are not normally equally acceptable, but one is preferred to the other in the given context. The verbless awJ-clauses are never ambiguous, as the content of the awrf-clause is presupposed as known, and its main meaning is emotive, denoting emphasis, rather than semantic. This means that the claim frequently put forward by functional typologists that the system of adverbial clauses is less sophisticated in languages that are predominantly used as a spoken medium does not hold for Scots. Semantic polyfunctionality of adverbial subordinators or adverbial clauses is not a feature that indicates a lower degree of sophistication of the subordinator system, but seems to be a deliberate choice that is built into the system to fulfil functional needs, such as economy and the expression of emotive and pragmatic meanings. This view is supported by the evidence of both diachronic and cross-linguistic comparisons of adverbial subordinator systems. Both the development of the English system of adverbial subordinators for finite and non-finite clauses as well as that from Latin to present-day French show that there is not a linear development from fewer to more and more complex subordinators, but possibly a cyclic development, in which periods of extension are followed by periods of reduction and simplification of structural complexity, and periods of restructuring alternate with comparatively stable periods. The difference between my hypothesis which proposes that adverbial subordinator systems undergo cyclic changes and the hypothesis that proposes a linear development is not as great as it appears. The latter hypothesis implies a unidirectional development, and, as Croft (1990: 229) points out, "processes that are unidirectional are also cyclic when viewed as changes from one language state to another. The 'endpoint' of the sequence of changes is the beginning stage, arrived at through a different process [Croft's emphasis]". Unidirectionality of language change has been claimed to be exceptionless, a view that has recently been challenged by Hopper and Traugott (1993: 184), who adduce the case of although used as a synonym of however as a counterexample to the "normal" development "from more to less paratactic clause combining". While the development from more lexical to more grammatical functions of words or phrases is certainly unidirectional, it is more difficult to see unidirectionality as an absolute rule if different grammatical categories are involved, such as co-ordinators, subordinators, and conjunctive adverbs. The history of the English prototypical co-ordinator and shows that it had acquired the additional function of subordinator in Middle English, and the use of subordinating and in Scots suggests the same direction of change, namely from co-ordinator to subordinator, which would be in line with the development

244

Part HI. Analysis

"from more to less paratactic". The history of but is, however, quite different. According to Mitchell (1985, II, 815-825), but had the functions of preposition and subordinating conjunction before its co-ordinator function, which means that here we have a change "from less to more paratactic". The acquisition of the function of conjunctive adverb of the subordinators ybr and because would likewise suggest a development "from less to more paratactic", whereas the use of but in Scots as a sentence-final conjunctive adverb illustrates yet another type of change, which does not quite seem to fit into this kind of categorisation. It is not clear whether the linkage by conjunctive adverbs should be considered more paratactic than that by co-ordinators, especially as the defining criteria for these two categories vary considerably from language to language. Thus it seems that with respect to the relationship between adverbial subordinators, co-ordinators and conjunctive adverbs, it is impossible to claim unidirectionality of change, whereas this seems possible with respect to the relationship between subordinators, adverbs and prepositions, where the most convincing model is the cyclic one proposed by Hansen—Nielsen (1986: 264) for except, where they speak of a "cyclic swing: adverb > preposition > conjunction > adverb". This model could be used as a general model, under the assumption that not all lexemes need to pass through all stages. It would then account for subordinators derived from adverbs via the intermediate stage of preposition, for those derived directly from adverbs and those derived directly from prepositions, and it would also account for cases like the subordinators for and because, which have developed the function of conjunctive adverbs.

5.6. Areas for future research: Some suggestions A number of possible areas for future research have suggested themselves in the course of this study. One would be an analysis of adverbial clauses in Irish English, which could perhaps explain the patterns found exclusively in western Scotland, as it seems possible that these may be due to the influence of Irish immigrants or seasonal farm labourers. In addition, an analysis of Scottish English texts, such as journalistic and scientific texts written by Scots, might be of interest to establish how far Scots constructions are also found in Scottish Standard English. In such a study, it would be essential to establish where the authors of the respective texts spent their formative years. With respect to Scots, a diachronic analysis of adverbial clauses would complement the present study. It might explain some of the differences between Scots and Standard English for which no adequate explanation could be provided in the present study. A study of prepositions seems likewise a promising task. Not only can certain differences between Scots and Standard English in the inventory and semantic scope of subordinators be related to

5. Conclusion

245

similar differences found with related prepositions, for example the concessive meaning of for all (that) and the instrumental meaning of with, but also a number of differences have already been attested in the smaller existing studies (Grant—Dixon 1921 for Modern Scots up to the beginning of the twentieth century and Miller 1993 for present-day Scots), which could be used as a starting point for an in-depth study. A similar study of adverbs in Scots would also be desirable, as there are systematic differences between Scots and Standard English in this category: Scots uses, for example, to a greater extent adverbs identical in form to prepositions and subordinators, i.e. after rather than afterwards. Corpus-based studies of grammatical aspects similar to the present one may also provide an answer to the question whether the dialect divisions based on adverbial clauses hold also for other grammatical structures. This might then lead to a reconsideration of the traditional dialect boundaries, as given in Map 1. This study has demonstrated that adverbial clauses in Scots show systematic differences from Standard English adverbial clauses and that in this grammatical category regional differences within Scots are only of minor significance. The detailed syntactic and semantic description and analysis also indicates that in determining the relationship of Scots to Standard English, the study of syntax and semantics is of equal importance to that of phonology and lexis. It has also demonstrated that in many cases the study of one aspect, such as adverbial subordinators, is not enough to account for the differences between Scots and Standard English adverbial clauses. A wider perspective is needed encompassing other aspects, such as tense and modality and also alternative constructions, such as correlative and co-ordinated clauses (cf. also Harris 1984). Moreover, this study has raised important questions for linguistic theory, and on the basis of the analysis suggestions for modifications of frequently voiced claims in functional typology have been made: the historical development of subordinator systems should be seen as cyclic rather than linear, and with respect to grammaticalisation processes subordinators are not necessarily the endpoint of the grammaticalisation chain. The study has thus also demonstrated the importance of more work on dialect syntax not as an end in itself, but as a complementation of the work that is being done in other fields of linguistics.

Appendix: The database l. The spoken corpus Three types of sources provide the basis for the spoken corpus: (i) Transcriptions of texts from published works on oral history, dating from 1980 onwards. These originate from Edinburgh, Glasgow, Dundee, Shetland, and the Mearns, and reflect mainly the speech of older speakers. The topics are reminiscences of work and social life. Though varying considerably, the orthography can generally be said to be a modified Scots, sometimes closer to Standard English than Scots, which has evidently been employed to facilitate reading. (ii) Recordings and transcriptions of recordings made available by the School of Scottish Studies. The transcriptions and recordings were made by students as part of oral history projects in Scottish Ethnology. The topics were work, traditions, early life, living conditions, and children's games. For all texts the name of the interviewer is given, and, where available, the title of the project and additional information on the informants (such as age group, geographical origin) is added in square brackets. All speakers are of a working class background. The texts reflect the speech of Edinburgh and Musselburgh in Midlothian, the Gorbals and Govan (Glasgow) in Renfrewshire, Hawick in Roxburghshire, and that of the Tweeddale in Peebleshire. Transcriptions were only used where no recordings were available. The orthography used in the student transcriptions is based on Standard English rather than Scots, as the students had no phonological training. The Scots grammar is, however, rendered faithfully. The quality of the recordings varies and unclear passages had to be excluded from the analysis. (iii) Recordings made by myself early in 1992: The recordings are from Charlestown of Aberlour in Morayshire, Port Glasgow in Renfrewshire, Niddrie (Edinburgh) in Midlothian, and Galashiels and Selkirk in Selkirkshire. All speakers are from a working-class background and all had grown up in the place they were recorded, with the exception of one speaker recorded in Galashiels (GA/A), who spent her formative years in Moffat (Dumfriesshire) and one speaker recorded in Selkirk (SE), who was brought up near Ettrick (Selkirkshire). The speakers recorded together are either members of the same

248

Appendix

family or friends. The topics covered in the recordings are reminiscences from the speakers' own lives and work, holidays, and some anecdotes. The quality of the recordings is influenced by the fact that places for the recordings were chosen where informal speech was most likely to occur, such as home or pub and that, with the exception of one of the recordings from the Borders (SE), two or three speakers were recorded together. Therefore unclear passages had to be excluded from the analysis. Different speakers are marked by initials. Thus GA/A indicates that the speaker of the extract is speaker A from tape GA. All additional information on the speakers is given in square brackets. The geographical distribution of all spoken texts is illustrated in Map 3. The additional examples cited as EX were collected ad hoc, and are therefore not on a par with the other examples, which originate from texts analysed as a whole. They are not included in any quantitative analysis, which accounts for minimal differences between the number of occurrences given in tables comparing Scots and English usage and that given for each subordinator in the main text. * marks texts which appeared without pagination and for which I have provided page numbers. 1.1. Transcriptions in Works of Oral History [published] DMB

Kay, Billy (1980), "Da merry boys: Shetlanders at the whaling", in Billy Kay (ed.), Odyssey: Voices from Scotland's recent past, Edinburgh: Polygon, 47-57.

FMY

Kay, Billy (1980), "They fairly mak ye work: Dundee and the jute industry", in Billy Kay (ed.), Odyssey: Voices from Scotland's recent past, Edinburgh: Polygon, 37-45.

LE*

[no editor] (1987), "Leith lives — It wisnae a' work" Edinburgh: Leith Local History Project [unpag.].

UOC

Faley, Jean (1990), Up oor close: Memories of domestic life in Glasgow tenements. 1910-1945, Wendlebury: White Cockade Publishing.

WPR

Sprott, Gavin (1980), "A weel plou'd rig: The era of the working horse on the farms of the east coast of Scotland", in Billy Kay (ed.), Odyssey: Voices from Scotland's recent past, Edinburgh: Polygon, 99-109.

The database: Written and spoken corpora

249

1.2. Transcriptions in Works of Oral History [unpublished] ED

Ewans, Christopher (1989), "Life as a tailoress", project in Scottish Ethnology MP 1989.10 (School of Scottish Studies), 2-7. [informant female, from Edinburgh, age group 60-80].

GOV

Hambley, Paul M. (1991), "World's apart", project in Scottish Ethnology MP 1991.23 (School of Scottish Studies), Appendix A, 112. [informant male, spent his formative years in Govan (Glasgow) and Ayr, age group 80+].

MU*

McNicholl, Mike (1990), "1 met a man who wasn't there", project in Scottish Ethnology MP 1990.21 (School of Scottish Studies) [unpag.] [informants male and female, from Musselburgh (Midlothian), age group 6-12].

1.3. Recordings AB

Hacker, Martina (1992) [informants father and son from Charlestown of Aberlour (Morayshire): AB/W male, age group 60-80; AB/C male, age group 20-40].

GA

Hacker, Martina (1992) [informants father, mother, and son from Galashiels (Selkirkshire): GA/V, male, age group 60-80; GA/A, female, spent her formative years in Moffat (Dumfriesshire), age group 60-80; GA/AV, male, age group 20-40].

GOR

Kerr, Kirstie (1987), project in Scottish Ethnology MP 1987.19 (School of Scottish Studies SC 1987.14) [informants married couple from the Gorbals (Glasgow), age group 80+]. No initials are added as only the male speaker is cited.

HA

Bonsor, Alison (1990), "Oor industry: The rise of the Hawick hosiery industry", project in Scottish Ethnology MP 1990.02 (School of Scottish Studies SC 1990.06) [informant male, from Hawick (Roxburghshire), age group 60-80].

NI

Hacker, Martina (1992) [informants friends from Niddrie (Edinburgh): NI/X male, age group 20-40; ΝΙΑ" male, age group 20-40].

250

Appendix

PG

Hacker, Martina (1992) [informants grandfather, father, and son from Port Glasgow (Renfrewshire): PG/G male, age group 80+; PG/F male, age group 60-80; PG/S male, age group 20-40].

SE

Hacker, Martina (1992) [informant male, from Selkirk (Selkirkshire), grew up near Ettrick (Selkirkshire), age group 40-60].

TW

Haddon, Mark (1992), project in Scottish Ethnology MP 1987.19 (School of Scottish Studies SC 1987.09-10) [informants married couple from Tweeddale (Peeblesshire), age group 60-80]. No initials are added as only the female speaker is cited.

2. The written corpus The authors use the dialects of Shetland, Caithness, Aberdeenshire, Angus, Dundee, Fife, Leith (Edinburgh), Stirlingshire, Glasgow, Dumfriesshire, and Ayrshire. The geographical distribution of the texts is illustrated in Map 4. In the case of four authors from northern Aberdeenshire (Buchan), Ayrshire, Selkirkshire, and Lanarkshire more than one text or a longer text was used. The topics of the stories correspond closely to those of the spoken corpus. * marks texts which appeared without pagination and for which I have provided page numbers. AD

Rae, William (1994), "Antic disposition", in James Robertson (ed.), A tongue in yer held: A selection of the best contemporary short stories in Scots, Edinburgh: B & W Publishing, 114-119.

AWT*

Gregor, Bill (n.d.), "A winter's tale", in Oor ain tongue, Banchory: The Rotary Club of Banchory-Ternan [unpag.].

BAT

Fortune, Pete (1994), "Big Alex's turn", in James Robertson (ed.), A tongue in yer heid: A selection of the best contemporary short stories in Scots, Edinburgh: B & W Publishing, 153-157.

CH*

Blackball, Sheena (n.d.), "The clockin hen", in Oor ain tongue, Banchory: The Rotary Club of Banchory-Ternan [unpag.].

DB

Johnson, Laureen (1994), "Dead branches", in James Robertson (ed.), A tongue in yer heid: A selection of the best contemporary short stories in Scots, Edinburgh: B & W Publishing, 164-173.

The database: Written and spoken corpora

251

DF

Toulmin, David (1994), "The dookit fairm", in James Robertson (ed.), A tongue in yer heid: A selection of the best contemporary short stories in Scots, Edinburgh: B & W Publishing, 71-83.

DOD*

Duff, John (n. d.), "Dod", in Oor ain tongue, Banchory: The Rotary Club of Banchory-Ternan [unpag.].

DSD

Ogston, David (1988), Dry stone days, Edinburgh: The Ramsay Head Press.

DWD

Begg, J. A. (n.d.), "Tales", in J. A. Begg and J. Reid, The dipper an the three \vee deils: Tales and poems in Ayrshire Scots, Barr: Luath Press ["Angling club minutes (unofficial) item 1", "Angling club minutes (unofficial) item 2", "Anyntit", "Butch and sundance", "Communication gap", "Cops an robbers", "Double anyntit", "Fair cop", "F rst bluid", "Puggie", "The technicality"], 11-13, 29-33, 4446, 59-63, 75-77, 89-93, 107-111, 122-123, 127-129, 137-139, 148149.

EN

Kermack, Mary (n.d.), "Lang syne in the east neuk ο Fife", ed. George Philp (Text accompanying Cassette SCOTSOUN SSC 069 "Mary Kermack's early childhood in the east neuk ο Fife", read by the author).

GH

Fenton, Sandy (1994), "Glory hole", in James Robertson (ed.), A tongue in yer heid: A selection of the best contemporary short stories in Scots, Edinburgh: B & W Publishing, 39-46.

GTA

Hershaw, William (1994), "Guid ti his ain", in James Robertson (ed.), A tongue in yer heid: A selection of the best contemporary short stories in Scots, Edinburgh: B & W Publishing, 47-49.

HAM

Miller, James (1994), "The hamecoming", in James Robertson (ed.), A tongue in yer heid: A selection of the best contemporary short stories in Scots, Edinburgh: B & W Publishing, 110-113.

INF

Kay, Billy (1994), "Inrush at nummer fower", in James Robertson (ed.), A tongue in yer heid: A selection of the best contemporary short stories in Scots, Edinburgh: B & W Publishing, 141-146.

LBD

[anonymous] (1991), "Laeves fae Beenie's diary", Shetland Life 123: 33-35.

252

Appendix

LL

Robb, Lydia (1987) "The lest lauch", Lallans 28: 29-31.

LRH

Fräser, Willie (1981), "The lang road harne", Keith Festival 12-14 June 1981 (Programme), The Traditional Music and Song Association of Scotland [privately published], 12-14.

LS

McLellan, Robert (1990), Linmill stories, Edinburgh: Canongate ["The aipple", "The Auchenheath races", "A drive to Lanark", "The kittlins", "The Saubbath", "The trap"], 9-16, 45-53, 63-70, 80-87, 98-107, 159-168.

MAN

Robertson, James (n.d.), "The Man up the stair", [unpublished].

MRS

McKenzie, Douglas (1994), "Mister and missus", in James Robertson (ed.), A tongue in yer heid: A selection of the best contemporary short stories in Scots, Edinburgh: B & W Publishing, 1-12.

NBN

Kelman, James (1983), "Nice to be nice", in James Kelman, Not not while the giro and other stories, Edinburgh: Polygon, 30-37.

PA

Herbert, W. N. (1994), "Patience", in James Robertson (ed.), A tongue in yer heid: A selection of the best contemporary short stories in Scots, Edinburgh: B & W Publishing, 158-163.

STE

Fitt, Matthew (1994), "Stervin", in James Robertson (ed.), A tongue in yer heid: A selection of the best contemporary short stories in Scots, Edinburgh: B & W Publishing, 174-182.

TCT

[Elliot, Walter] (n. d.), The thochts of chairwoman Tibbie: A collection of "opeenions" by Tibbie Tamson, columnist for "The Selkirk Weekender" with illustrations by Cath Rutherford, Selkirk: Weekender Graphics.

TT

Murray, John (1994), "Thrie tales", in James Robertson (ed.), A tongue in yer heid: A selection of the best contemporary short stories in Scots, Edinburgh: B & W Publishing, 99-103.

VIC

Paisley, Janet (1994), "Vices", in James Robertson (ed.), A tongue in yer heid: A selection of the best contemporary short stories in Scots, Edinburgh: B & W Publishing, 29-36.

The database: Written and spoken corpora

WT

253

Kermack, Alison (1994), "A wee tatty", in James Robertson (ed.), A tongue in yer heid: A selection of the best contemporary short stories in Scots, Edinburgh: B & W Publishing, 37-38.

Notes

1. The section on the history of Scots is primarily based on Murison (1979), Templeton (1975) and the introduction in The Concise Scots Dictionary (henceforth CSD): ix-xvi. 2. Journals that publish articles and literature in Scots and on Scots are Chapman and Lallans, which go back some twenty years. The increasing interest that Scots literature and language enjoy in schools is reflected in Laverock, a recent addition to publications on Scots language and literature, which states as its explicit aim "to increase awareness and appreciation of Scottish literature and language with particular reference to the school curriculum" (Laverock 1: 1), and which likewise includes articles written in Scots. Most recently there is Cairn: The historie jurnal in the Scots leid, first published in 1997 by the Aiberdeen Univairsitie Scots Leid Quorum, with the aim of using Scots as a medium for historical scholarship. 3. The term "overt Scotticisms" is taken from Aitken (1979: 107), who uses it to designate the "special diction of Scottish-tagged locutions used selfconsciously by many Scottish speakers as a kind of stylistic grace and as a way of claiming membership of the in-group of Scotsmen". Its opposite is "covert Scotticisms", defined as "more or less exclusively Scottish usages employed by many Scottish speakers without their being very much or at all aware that in so doing they are revealing their Scottish origins" (Aitken 1979: 106). 4. Both the complexity and the well-structured nature of sentences in spoken narratives is illustrated by the following example: "The other man who kicks is the full-back, who usually receives the ball way behind the rest of his team, either near his line or when somebody's done what the stand-off in the first example was doing, kicked over the defenders; the full-back should be able then to pick it up, and his job is usually to kick for touch — nearly always for touch because he's miles behind the rest of his side, and before he can do anything else with the ball he's got to run up into them, before he can pass it, because he can't pass the ball forward, and if he kicks it forward to another of his side the other man's automatically offside." (Halliday 1987: 58 [ii]) 5. This distinction, which is based on Lyons 1977, uses the term "text sentences" for sentences occurring in actual texts and "system sentences" for sentences that "belong to the hypothesised language system" (Miller 1995: 117).

256

Notes

6. That introspection and corpus analysis should not be seen as alternatives, but rather as methods that complement each other, is illustrated by the Nijmegen grammar projects (cf. Aarts 1991). 7. The term "complex" was already used by Sweet (1891 [1960] II: 161): "Two or more sentences may be joined together to form a single complex sentence, or complex, as we may call it for the sake of brevity." German grammars express a similar concept by the term Satzgefüge. 8. Matthiessen and Thompson (1988) adduce as an additional argument against the treatment of adverbial clauses as embedded the fact that they can only be replaced by prepositional phrases consisting of prepositions and nominalisations, but not prepositions and "ordinary nouns". While this is true for their example Before leaving Krishnapur, which corresponds to Before the departure from Krishnapur, there are cases where the prepositional phrase may contain an "ordinary noun": He cried for joy is a near synonym of He cried because he was (so) happy. 9. Foley and Van Valin's (1984: 247) notion of co-ordination is strikingly different from that of other functionalists in that it includes the "accusative cum infinitive", a clause type that other typologists would analyse as embedded. 10. In Quirk et al. (1985: 15.22) the category "adjunct" includes obligatory clauses. As a result happy and as if he was happy in "he looks happy" and "he looks as if he was happy" are allocated to different functional categories, i.e. that of complements and adjuncts respectively, which seems problematic to me. My uneasiness with the terminology seems to have been shared by Greenbaum, as Greenbaum 1996 (66-68) uses the term "subject predicative" for all obligatory complements of copular verbs, irrespective of their morphological and syntactic structure. 11. Hengeveld's classification is based on Lyons (1977). 12. This term is preferred to "sentence adjuncts" (Quirk et al. 1985: 15.22), as the matrix clause may be subordinate to another clause. 13. The punctuation in the Standard English gloss, which differs from that of the Scots original, is based on the rules set out in Quirk et al. 1985 (Appendix III, in particular III. 10, 18). 14. Traugott (1988) proposes three semantic—pragmatic tendencies: (i) Meanings situated in the external described situation > meanings situated in the internal (evaluative/perceptual/cognitive) situation, (ii) Meanings situated in the described external or internal situation > meanings situated in the textual situation, (iii) Meanings tend to become increasingly situated in the speaker's subjective belief-state/attitude toward the situation. 15. Free adjuncts and absolutes have been the subject of detailed studies. See in particular the work of Kortmann (1991, 1992) and Stump (1985).

Notes

257

16. Zandvoort's example is taken from the Concise Oxford Dictionary (henceforth COD, where it illustrates a use of so, described as: "often in sentence appended as explanation". 17. For Standard English this is no absolute rule, as the following counterexample of an -ing clause that is concessive or contrastive rather than temporal occurs in LOB (K13 124-126): " It was far wiser to treat Vicky as an ignorant, hot-headed girl, and while appearing to tolerate Fritz, to estrange the King from him by various subtle means." 18. They are identical in written language, but distinguished by stress in spoken language, the adverb being 'trotzdem, while the subordinator is trotz'dem. 19. Abraham (1993) does not take verb position as the most important criterion, and classifies the controversial weil with main clause word order as a conjunctive adverb on the basis of the cliticisation of the personal pronoun to the preceding element and the pause preceding it. 20. Earlier scholars who made this the primary criterion for their classification, such as Curme (1935: 96), Jespersen ([1965] V: 21.8.8. ) Poutsma (1929, part I/II: 605), classified for, but not so that as a co-ordinator. Sweet ([I960] I: 150) is a notable exception. 21. An exception is Harris (1989), who presents an analysis which is in line with the treatment of clause-linking elements in languages, such as German, where the strict application of the verb-second criterion has led to a double categorisation of an increasing number of lexical items traditionally categorised as subordinating conjunctions. Opinions are, however, divided as to the classification of the respective items as co-ordinators (Primus 1997: 146; Wegener 1993) or conjunctive adverbs (Abraham 1993). 22. The phenomenon of verb-second position in connectors that take verb-final word order in the standard language was first noticed for southern German and Austrian dialects (see for example Eisenmann 1973; Labouvie 1938; Wessely 1981). The phenomenon is, however, becoming a widespread feature of informal speech and is now entering the written language (see Clyne 1975; Gaumann 1983; Giinthner 1993; Keller 1993a, 1993b; Hacker 1994a; Willems 1994; Pasch 1997). 23. The example in Quirk et al. (1985: 1.26) corresponds to the Scots, Irish English and Tyneside English usage: The job 's still not done; I'll finish her this arvo, but. ['... it this afternoon, however'] [Quirk et al.'s gloss]. 24. A similar relationship seems to be suggested for adverbs, conjunctions and prepositions by Hartig (1976: 230). 25. This use of but is not mentioned in any of the Scots dictionaries or in Todd's (1990) Northern Irish Dictionary, but is frequent in (particularly Western) Scots, which would suggest a (Northern) Irish origin, as does its occurrence in Tyneside, (cf. Beal 1993: 211).

258

Notes

26. The phenomenon of every and any neutralisation and -ever deletion is discussed in a more general context in Haiman 1974. 27. The non-standard present tense ending in -s is a frequent feature in dialects, in particular in narratives using the historic present (cf. Edwards—Weltens 1985: 108). 28. This use of never is also frequent in English dialects (cf. Cheshire, Edwards and Whittle [1989: 197]). 29. This use is different from the cum inverswn type of temporal clause, as the latter implies some sort of unexpectedness of the subordinate clause situation, which is not characteristic of the non-finite a/ore-clauses. 30. For a detailed discussion see Couper-Kuhlen (1989: 359-372), who characterises this type of when-clause as typical of "formal narrative style" (p. 362). 31.cf. Givon (1990: 828), who states: "Quite often, a more general ('unmarked') time subordinator, such as 'when' in English, is used when more specific temporal relations are in fact signalled. Such less-marked coding strategy is successful when the semantic specificity of the temporal relation can be inferred from other features of the two clauses — most commonly from the verb and its tense-aspect." It is, however, not only syntactic features that allow such inferences, but also the extra-linguistic context. 32. The SND has separate entries for the two subordinators. It gives the meaning of the temporal subordinator as "by that time that, when; before; until" and states that conditional gin "is generally supposed to be a reduced form ofgien, given ... used absol. [absolutely] in its concessive meaning". This is also given as the most likely etymology in the OED, which likewise has separate entries for temporal and conditional gin, but lists temporal gin only as a preposition. 33. According to the OED entry again (section "Forms"), in Scotland and northern England against was not adopted. The subordinator again, whose meaning is given as "in preparation for the time that" by Grant—Dixon (1921: 172) and illustrated by a quotation from Scott's Old Mortality, is not found in the Scots corpus. 34. The two examples illustrate at the same time how the density of Scots varies with the same speaker. It is probably no coincidence that the broader Scots form necht occurs when the speaker cites direct speech. 35. The Scots use of likely differs from that of Standard English, where "he is likely to ..." is only used if the meaning is "it is likely that he ...", whereas in Scots likely is a synonym of probable and probably in all its uses. 36. The tap realisation of the phoneme /o / in intervocalic position is a typical feature of Glasgow speech (cf. Macafee 1983: 33). As illustrated by this

Notes

259

example, it is not restricted to the city itself, as Macafee (1983: 33, note 22) seems to suggest. 37. The same option exists for Danish, where the choice between efier and da is likewise determined by the amount of time that elapses between matrix clause and subordinate clause situation, while the use of German als ('when' with reference to past situations) to express anteriority is more restricted compared to nachdem 'after': "Vor allem aber lassen sich nur bestimmte Verbalausdrücke in Obersatz und Untersatz kombinieren. Diese Restriktionen hängen mit der zeitlichen Einordnung des Obersatzgeschehens zusammen." (Engel 1988: 261) [Most important, however, is that only certain verbal expressions in the matrix clause and subordinate clause can be combined (if als is used). These restrictions are related to the temporal orientation of the matrix clause situation.] [my translation, MH] 38. The use of us with reference to a single person is the norm rather than the exception in Scots, but is also found in non-standard English (Quirk et al. 1985: 6.18, note [b]; Trudgill 1990: 82). 39. The formfaem shows the same loss of M asfae. The /r/-less forms seem to be typical of northern Scots, while southern Scots tends to retain the /r/. 40. This sentence also illustrates the omission of a relative pronoun functioning as subject, which is a widespread feature in English dialects (cf. Petyt 1985: 238 for Yorkshire; Beal 1993: 208 for Tyneside; Wakelin 1986: 48 for South-Western British English; Edwards 1993: 229 for SouthEastern British English; Harris 1993: 149 for Irish English; Miller 1993: 112 for Scots). 41. Stanley Robertson in The Story of English, IV The Quid Scots Tongue (BBC Videos for Education and Training). 42. The use of the feminine personal pronoun she to refer to objects is a typical feature of North-Eastern Scots. 43. This corresponds to Edgren's (1971: 132-133) analysis of English temporal clauses: "In clauses opening with as soon as the use of the perfect form of conclusive verbs (join, stop, cut off...) indicated a less immediate sequence of the activities; in clauses opening with the moment, the minute, immediately there were no instances of perfect forms to counteract the immediacy." 44. Harris (1993: 165) gives the following example: "My husband died whenever I was living in New Lodge Road", while Todd (1990: entry whenever) uses the following: "Whenever he was in Belfast last week, he called to see us." Thus the Irish use of whenever appears to correspond to that of Dutch wanneer, which likewise does not refer to repeated actions. 45. Present tense and past tense of the verb run are identical in Scots (cf. CSD entry rin "run")

260

Notes

46. A change from generic we to generic you is more common in Scots than in Standard English. 47. This does not indicate that the police car has been already mentioned, but is a feature typical of vivid spoken narratives (cf. Quirk et al. 1985: 6.43, note [a]). 48. The subordinates occurring in the sections of LOB and LLC selected for comparison with the Scots corpus are more restricted in comparison to the inventory listed in Quirk et al. (1985: 15.25). 49. The etymology of conditional gin is uncertain, but it is generally assumed that it is not related to temporal gin (see gin in section 3.2.2. above). 50. For the use of an and-clause in Scots where Standard English would use a relative clause see Miller (1993: 112). 51. The paraphrase "only if... not" suggested by Quirk et al. (1985: 15.35 note [fj) and Fillenbaum (1986: 183-185) only works for preposed unlessclauses, while in postposed unless-clauses only "only not... if" is possible. 52. Mines is the Scots form of the first person singular possessive pronoun in non-determinative function. 53. Forms like went, had gone, etc. are actually interpreted as subjunctives for Standard (American) English by Fillmore (1990: 138). 54. In some German dialects würde could be replaced by täte, thus neutralising the differences in the meanings of the two auxiliaries for hypothetical use. 55. For a discussion of cliticisation of auxiliaries in Scots see Brown—Millar (1980: 103), who explicitly state that "subject-operator and operatornegative cliticisation cannot both occur." 56. The usual order of this idiomatic phrase is "rain, snow, or hail", according to Jim Miller (personal communication, September 1995). 57. It is a frequent phenomenon that the same word has the double function of subordinator and contrastive adverb. Jespersen ([1965] V: 21.4.6.) gives an example of however used as a subordinator, and in southern German dialects trotzdem, a concessive adverb, is used as a subordinator, while the German subordinator obwohl is increasingly used as a contrastive adverb. 58. Shetlandic uses the masculine personal pronoun he for objects in the same way as North-Eastem Scots uses the feminine personal pronoun she. 59. According to the CSD the crap the wa is 'the space between the top of a wall and the roof of a building'. 60. Us refers to the speaker and is thus coreferential with the matrix clause subject/i/z T. This use of us is not restricted to Scots, but is also a general feature of colloquial English. According to Hughes—Trudgill (1996: 28), its use is, however, limited to a few fixed phrases "outside the north-east". 61. Standard English requires the interrogative pronoun which if reference is made to a specified or known set of items, whereas what is used if there is no restriction on the items referred to. In Scots this functional division does

Notes

261

not apply; what is also used in questions referring to a limited specified set, as is well illustrated by the continuation "kennin fine" in (344) (cf. also Miller 1993: 125). 62. For a discussion of the possible origins of this construction and its history see Hacker (1994b). 63. The term "common case" is chosen to avoid awkward combinations of morphological and syntactic terms, as illustrated by Macafee's description: She uses "objective personal pronoun" to refer to the subject of the clause, but "subject complement" to refer to its complement. 64. This restriction is also noted for English. Curme (1931: 314) states that "as is now as in older English peculiarly appropriate where it is desired to give an easy, natural, almost self-evident explanation of the statement in the principal proposition". 65. This difference in meaning is confirmed by A. V. Murray (Personal communication, 5 December 1994). 66. The two historical personages mentioned are James Graham, Marquis of Montrose (1612-1650) and Alexander Leslie (1580-1661), or possibly General David Leslie, who defeated the royal troops at Marston Moor (Donaldson 1965: 332-333). 67. The corresponding entry in the OED (what: D.I1.2. suggests that the construction derives from co-ordinated phrases introduced by relative what (= and that) followed by a w/7/7-phrase expressing a reason or means, and subsequent ellipsis of what with in the second co-ordinate. 68. For a more detailed discussion of this development in Standard English see Hacker (1994a). 69. The possessive personal pronoun their is not only employed to be politically correct, but is generally used in Scots to indicate that the person referred to may be of either sex. In the above example this leads to a certain amount of apparent contradiction because "he" is the term used for the person fulfilling a specific role in games. 70. Abd. = Aberdeenshire refers to the pre-1975 county. 71. Some is used here as an indefinite pronoun corresponding to Standard English one (cf. the similar use of ony in Scots). 72. Mosse (1968: 101) suggests that the for + noun + /o-infinitive construction should be seen as a possible origin of the for /o-clause. 73. The difficulty in establishing rules which explain the non-occurrence of certain clause constituents in Standard English infinitival purpose clauses is discussed by Mair (1990: 210-216), who emphasises the importance of the semantic class of the matrix clause verb. 74. When stating that "clauses of result do not require a modal auxiliary" Quirk et al. (1985: 15.49) seem to refer only to modal would expressing

262

Notes

volition. To my knowledge, could as a modal expressing possibility is also found with clauses of result in Standard English. 75. Quirk et al. (1985: 15.49) seem to refer only to written sentences when they state that "when that is omitted in the result clause, the conjunction [i.e. subordinator] so is indistinguishable from the conjunct [i.e. conjunctive adverb] so in asyndetic co-ordination". 76. Franz (1939: 433) states: "Die Freiheit des Sh.schen [Shakespeareschen] Gebrauchs geht noch über die der modernen Verkehrssprache hinaus; that fehlt im elisabethanischen Englisch sogar in Konsekutivsätzen nach so und such ... sehr häufig ..." [The freedom of Shakespearian usage goes even beyond that of the modern usage; thus that is in Elizabethan English (very) frequently omitted even in clauses of result] [my translation, MH]. 77. This classification is not in line with the distinctive features given for nominal structures and clauses in Quirk et al. 1985 (14.5 and 15.14). 78. Kortmann's (1997: 294) figures are: for Old English 37 items, for Middle English 74 items, for Early Modern English 96 items, and for present-day English 63 items. 79. The importance of economy as a principle governing syntactic change is pointed out by Moessner (1997a: 362-363), who relates both the extension of the inventory of adverbial subordinators by creating new subordinators using that as a general subordinator marker, and the later deletion of that to this principle. 80. The restriction to verbless clauses and -ed clauses respectively should be seen as a tendency rather than an absolute rule, as the corpus is not large enough to make absolute statements in that matter. 81. The case of say is interesting, as unlike the other subordinators that are found in present-day Scots and earlier stages of the English language, there is no evidence that say was simply retained in Scots. It seems possible that it dropped out of use, as it did in Standard English, and was "reinvented" only recently. This would explain why it is not mentioned anywhere in the literature. 82. Unlike south-eastern and central Scotland, where the Scandinavian influence was mainly Danish, northern mainland and insular Scotland was in contact with Norwegian rather than Danish. As the differences between Danish and Norwegian are predominantly phonological rather than grammatical, it seemed reasonable to restrict the comparison to one Scandinavian language. For reasons of comparability Danish was chosen rather than Norwegian, as the Danish, but not the Norwegian subordinator inventory is listed in Kortmann 1994: II. 83. The development is characterised by Paul—Moser—Schröbler (1975: 450451) as follows: "Die Verbindung von daz mit einem Adverb muß nicht am Anfang des Weges des betreffenden Adverbs zur Konjunktion stehen.

Notes

263

Aus der vorstehenden Darstellung der Konjunktionen ... scheint vielmehr hervorzugehen, daß der unmittelbare Übergang vom Adverb zur Konjunktion, ohne daß daz als Stütze erscheint, das Gewöhnlichere ist, und es gibt ein offensichtlich erst spätes Hinzutreten von daz. Andererseits zeigt es sich auch , daß es die jüngere Sprache ist, welche die Neigung hat, die Konjunktion daz aus Verbindungen dieser Art wieder auszustoßen, nach dem gewissermaßen die Adverbien mit Konjunktionscharakter gesättigt sind." [The combination of daz with an adverb need not be the initial stage of the development of an adverb into a conjunction. The preceding description of conjunctions suggests rather that at the transition stage of an adverb developing into a conjunction there is no cfaz-support and that daz apparently is a later addition. On the other hand, recent developments of the language show that daz is dropped again in combinations of this kind, after the adverbs have been, as it were, saturated with conjunctional character.] [My translation, MH] 84. Yiddish seems to be an exception (cf. König—van der Auwera 1990: 349). However, as Yiddish has a unique history which can in no way be compared to that of the other Germanic languages, no attempt will be made here to venture any hypothesis as to the reason for this. 85. Braunmüller (1978: 113) lists this subordinator as a counter-example to the typical shift in subordinator meaning, namely from temporal to causal, claiming that seit developed from a causal into a temporal subordinator. This is, however, not supported by historical facts. Behagel's (1928: 248249) earliest examples for the temporal as well as the causal reading are from the same text. 86. Slight discrepancies between my figures and Kortmann's are due to different defining criteria for the category adverbial subordinator. 87. Apart from the omission of correlatives a number of changes and additions have been made to Kortmann's (1994: II) lists for the sake of consistency or completeness. For Danish: til and indtil are treated as two separate subordinators like the corresponding English //// and until, sälcenge is treated as a temporal subordinator in line with Hansen (1967: 426-427) and Allan—Holmes—Lundskaer-Nielsen (1995: 463), endskent (concession) is added. For German: Southern German dialectal derweil(st) and wo (time, cause and concession) are added. For Scottish Gaelic: ged a (concession introducing a positive clause) is added. 88. Contrastiveness is an inherent quality of concessiveness, i.e. there are no concessive subordinators that do not imply a contrast between matrix clause and subordinate clause (cf. Handke 1984: 146). 89. Of the four German concessive subordinators obwohl, obschon, obzwar, and obgleich, only obwohl occurs in Eaton's (1940 [1967]) frequency list, in group 2a, i.e. among the first 500 of the second thousand words, while

264

Notes

the others are not listed, which means that the other three subordinators were not among the first 6,474 words. Obwohl, obgleich, obschon, and trotzdem have been analysed statistically for written Swiss German (Gelhaus 1972: 114-118), the relationship between them being 66.4% obwohl, 20.0% obgleich, 12.7% obschon, and 0.9% trotzdem. 90. Franz (1939: 435) notes: "Die Mischung von Temporal- und Konditionalsatz kennzeichnet sich bisweilen auch formell in der Doppelkonjunktion when //" [The amalgamation of temporal and conditional clauses is sometimes also indicated structurally by the double conjunction when if.] [My translation, MH] 91. It is worth noting that, like Scots and unlike Standard English, these three languages also allow double modals. 92. Schlieben-Lange (1992: 350) is convinced that the present state of the French subordinator system is only an intermediate stage and predicts a further change in the French subordinator system. 93. For the one exception of a Northern English speaker in LLC see 3.4. above. 94. Franz (1939: 374) does, however, derive the restrictive meaning from the concessive one: "In for aught I know (I see, can tell), woraus die jüngeren Ausläufer for anything I know, for what I know, for all I know hervorgegangen sind, ist die ursprüngliche Bedeutung ganz verblaßt. Der Ausdruck hatte von Haus aus den Sinn: spite of anything that I know of,' woraus sich weiter ergab 'there being nothing to the contrary that I know of; so far as I know'." [The original meaning of for aught I know (I see, can tell), which is the source for the more recent for anything I know, for what I know, for all I know, is totally bleached. The original meaning of the expression was: in spite of anything that I know of, from which the following meaning derived: there being nothing to the contrary that I know of; so far as I know.] [My translation, MH] 95. Despite its shortcomings this term is used for that, for reasons of simplicity and comparability with the wh- pronouns. 96. The Lincolnshire dialect shares also other features with Scots, such as the use of and as a subordinator introducing emphatic verbless clauses of reason and concession (cf. Hacker 1994b: 44-45). 97. Apart from the special case of emphatic subordinators, polysemy between subordinators of reason and concession exists also in other languages, such as Latin (cum) and substandard varieties of Balkan languages (cf. Buchholz 1993: 13-14). It may also be argued that in Standard English while denotes reason in speech act adverbial clauses, such as "Could you get me a glass of milk while you are in the kitchen?" 98. As there is no positive evidence to support the existence of the subordinator say after the Early Modern English period until it is found in present-day Scottish texts, it is assumed that its occurrence in present-day

Notes

265

spoken Scots is best interpreted as a case of a subordinator being reintroduced rather than retained. 99. As Scotland has an independent education system, which is different from the English one, this cannot be ruled out a priori, but would need to be investigated in a separate study.

References

Aarts, F. G. A. M. 1979 "Time and tense in English and Dutch: English temporal since and its Dutch equivalents", English Studies 60: 603-624. Aarts, Jan 1991 "Intuition-based and observation-based grammars", in Karin Ajmer—Bengt Altenberg (eds.), English corpus linguistics: Studies in honour of Jan Svartvik, New York: Longman, 44-62. Abbott, E. A. 1870 A Shakespearian grammar: An attempt to illustrate some of the differences between Elizabethan and Modern English, London: Macmillan. [1966] New York: Dover. Abercrombie, David 1979 "The accents of Standard English in Scotland", in A. J. Aitken— Tom McArthur (eds.), Languages of Scotland (Association for Scottish Literary Studies, Occasional Papers 4), Edinburgh: Chambers, 68-84. Abraham, Werner 1993 "Einleitung zum Thema dieses Bandes. Grammatikalisierung und Reanalyse: Einander ausschließende oder ergänzende Begriffe?", Folia Linguistica Historica 13: 7-26. Aitken, A. J. 1979 "Scottish speech: A historical view with special reference to the Standard English of Scotland", in A. J. Aitken—Tom McArthur (eds.), Languages of Scotland (Association for Scottish Literary Studies, Occasional Papers 4), Edinburgh: Chambers, 85-118. 1981 "The good old Scots tongue: Does Scots have an identity?", in Einar Haugen—J. Derrick McClure—Derrick S. Thomson (eds.), Minority languages today: A selection from papers read at the First International Conference of Minority Languages held at Glasgow University from 8 to 13 September 1980, Edinburgh: Edinburgh University Press, 72-90. 1982 "Bad Scots: some superstitions about Scots speech", Scottish Language 1:30-44. Allan, Robin—Philip Holmes—Tom Lundskaer-Nielsen 1995 Danish: A comprehensive grammar, London—New York: Routledge.

268

References

Altenberg, Bengt 1984 "Causal linking in spoken and written English", Studia Linguistica 38: 20-69. 1986 "Contrastive linking in spoken and written English", in Gunnel Tottie—Ingegerd Bäcklund (eds.), English in speech and writing: A symposium (Acta Universitatis Upsaliensis, Studia Anglistica Upsaliensia 60), Uppsala: Uppsala University, 13-40. Bache, Carl—Niels Davidsen-Nielsen, 1997 Mastering English: An advanced grammar for non-native and native speakers (Topics in English Linguistics 22), Berlin— New York: Mouton de Gruyter. Bäcklund, Ingegerd 1986 "Beat until stiff. Conjunction-headed abbreviated clauses in spoken and written English", in Gunnel Tottie—Ingegerd Bäcklund (eds.) English in speech and writing: A symposium (Acta Universitatis Upsaliensis, Studia Anglistica Upsaliensia 60), Uppsala: Uppsala University, 41-55. Barber, Charles 1997 Early Modern English (second edition), Edinburgh: Edinburgh University Press. Barnes, Michael 1984 "Orkney and Shetland Norn", in Peter Trudgill (ed.), Language in the British Isles, Cambridge: Cambridge University Press, 352-366. Barry, Michael V. 1982 "English in Ireland", in R. W. Bailey—Manfred Görlach (eds.), English as a world language, Ann Arbor: University of Michigan Press, 84-133. Bayer, Josef 1993 "Zum in Bavarian and Scrambling", in Werner Abraham—Josef Bayer (eds.), Dialektsyntax, Linguistische Berichte, Sonderheft 5, 50-70. BBC 1992 "Speaking out", Radio Broadcast, Radio Scotland, 27 February 1992. Beal, Joan 1993 "The grammar of Tyneside and Northumbrian English", in James Milroy—Lesley Milroy (eds.), Real English: The grammar of English dialects in the British Isles, London—New York: Longman, 187-213.

References

1997

269

"Syntax and morphology", in Charles Jones (ed.), The Edinburgh history of the Scots language, Edinburgh: Edinburgh University Press, 335-377. Beaman, Karen 1984 "Coordination and subordination revisited: Syntactic complexity in spoken and written narrative discourse", in Deborah Tannen (ed.), Coherence in spoken and written discourse, (Advances in Discourse Processes 12), Norwood, New Jersey: Ablex, 45-80. Behagel, Otto 1928 Deutsche Syntax: eine geschichtliche Darstellung, vol. Ill: Die Satzgebilde, Heidelberg: Winter. Biber, Douglas 1988 Variation across speech and writing, Cambridge: Cambridge University Press. Bliss, Alan 1984 "English in the south of Ireland", in P. Trudgill (ed.), Language in the British Isles, Cambridge: Cambridge University Press, 135-151. Bloomfield, Leonard 1935 Language (revised edition,), London: Allen & Unwin. Bolinger, Dwight 1984 "Intonational signals of subordination", in Claudia Brugman— Monica Macaulay (eds.), Proceedings of the tenth annual meeting of the Berkeley Linguistics Society, February 17-20, 1984, Berkeley: Berkeley Linguistics Society, 401-413. Borrowman, Lorna 1977 "The Scots tongue in education", Akros 11/33: 107-118. Boyle, Daniel 1973 "Ach and agus as coordinate and subordinate conjunctions in Gaelic", in Claudia Corum—T. Cedric Smith-Stark—Anne Weiser (eds.), You take the high node and I'll take the low node: Papers from the comparative syntax festival. The differences between main and subordinate clauses, Chicago: Chicago Linguistic Society, 220-228. Braunmüller, Kurt 1978 "Remarks on the formation of conjunctions in Germanic languages", Nordic Journal of Linguistics 1: 99-120. Brown, Keith—Martin Millar 1980 "Auxiliary verbs in Edinburgh speech", Transactions of the Philological Society, Oxford: Philological Society, 81-133.

270

References

Buchholz, Oda 1993 "Zu konzessiven Subordinatoren in den Balkansprachen", in Uwe Hinrichs—Helmut Jachnow—Reinhard Lauer-—Gabriella Schubert (eds.), Sprache in der Slavia und auf dem Balkan. Slavische und balkanologische Aufsätze (Norbert Reiter zum 65. Geburtstag), Wiesbaden: Harrassowitz, 11-20. Burnham, J. M. 1911 Concessive construction in Old English prose (Yale Studies in English 39), New York. Calder, George 1923 A Gaelic Grammar: Including a chapter on personal and place names, Glasgow. [1990] Glasgow: Gairm. Chafe, Wallace 1982 "Integration and involvement in speaking, writing, and oral literature", in Deborah Tannen (ed.), Spoken and written language: Exploring orality and literacy, Norwood, New Jersey: Ablex, 35-53. 1984 "How people use adverbial clauses", in Claudia Brugman-Monica Macaulay (eds.), Proceedings of the tenth annual meeting of the Berkeley Linguistics Society, February 17-20, 1984, Berkeley: Berkeley Linguistics Society, 437-449. 1987 "Cognitive constraints on information flow", in Rüssel S. Tomlin (ed.), Coherence and grounding in discourse: Outcome of a symposium, Eugene, Oregon, June 1984 (Typological Studies in Language 11), Amsterdam—Philadelphia: Benjamins, 21-51. 1988 "Linking intonation units in spoken English", in John Haiman— Sandra Thompson (eds.), Clause combining in grammar and discourse (Typological Studies in Language 18), Amsterdam— Philadelphia: Benjamins, 1-27. Chafe, Wallace— Jane Danielewicz 1987 "Properties of spoken and written language", in Rosalind Horowitz—S. Jay Samuels (eds.), Comprehending oral and •written language, San Diego: Academic Press, 83-113. Chambers, J. K.—Peter Trudgill 1991 "Dialect grammar: Data and theory", in Peter Trudgill—J. K. Chambers (eds.) Dialects of English: Studies in grammatical variation (Longman Linguistics Library), London—New York: Longman, 291-299.

References

271

Cheshire, Jenny 1982 "Dialect features and linguistic conflict in school", Educational Review 34: 53-67. Cheshire, Jenny—Viv Edwards 1993 "Sociolinguistics in the classroom: Exploring linguistic diversity", in James Milroy—Lesley Milroy (eds.), Real English: The grammar of English dialects in the British Isles, London— New York: Longman, 34-52. Cheshire, Jenny—Viv Edwards—Pamela Whittle 1989 "Urban British dialect grammar: The question of dialect levelling", English World-Wide 10: 185-225. Cheshire, Jenny—Jim Milroy 1993 "Syntactic variation in non-standard dialects: Background issues", in James Milroy—Lesley Milroy (eds.), Real English: The grammar of English dialects in the British Isles, London— New York: Longman, 3-33. Clyne, Michael G. 1975 Forschungsbericht Sprachkontakt: Untersuchungsergebnisse und praktische Probleme, Kronberg, Taunus: Scriptor. COD = The Concise Oxford Dictionary of Current English 1976 6th edition ed. by J. B. Sykes, Oxford: Oxford University Press, Comrie, Bernard 1986 "Conditionals: A typology", in Elizabeth Closs Traugott—Alice ter Meulen—Judy Snitzer Reilly—Charles A. Ferguson (eds.), On conditionals, Cambridge: Cambridge University Press, 77-99. CSD = The Concise Scots Dictionary 1985 Mairi Robinson (ed.), Aberdeen: Aberdeen University Press. Couper-Kuhlen, Elizabeth 1989 "On the markedness of 'narrative temporal clauses'", in Olga MiSeska Tomic (ed.), Markedness in synchrony and diachrony (Trends in Linguistics: Studies and Monographs 39), Berlin— New York: Mouton de Gruyter, 359-372. Croft, William 1990 Typology and universals (Cambridge Textbooks in Linguistics), Cambridge: Cambridge University Press. Curme, George O. 1931 A grammar of the English language, vol. Ill: Syntax, Boston: D.C. Heath. Declerck, Renaat 1991 Tense in English: Its structure and use in discourse, London: Routledge.

272

References

DOST = A Dictionary of the Older Scottish tongue. 1937- Vol. I-VII, William A. Craigie, A. J. Aitken et al. (eds.), Aberdeen: Aberdeen University Press. Dieth, Eugen 1932 A grammar of the Buchan dialect (Aberdeenshire), vol. I, Phonology—accidence, Cambridge: Cambridge University Press. Donaldson, Gordon 1965 The Edinburgh history of Scotland, vol III: Scotland: James V to James VII, Edinburgh—London: Oliver & Boyd. Edgren, Eva 1971 Temporal clauses in English (Acta Universitatis Upsaliensis, Studia Anglistica Upsaliensia 9), Uppsala: Uppsala University. Edwards, Viv 1989 "A postscript", in Jenny Cheshire—Viv Edwards—Henk Münstermann—Bert Weltens (eds.), Dialect and education, Clevedon: Multilingual Matters, 317-323. Edwards, Viv 1993 "The grammar of Southern British English", in James Milroy— Lesley Milroy (eds.), Real English: The grammar of English dialects in the British Isles, London—New York: Longman, 214242. Einarsson, Stefan 1945 Icelandic: Grammar, texts, glossary, Baltimore: Johns Hopkins. Eisenmann, Fritz 1973 Die Satzkonjunktionen in gesprochener Sprache: Vorkommen und Funktion. Untersucht an Tonbandaufnahmen aus BadenWürttemberg, Bayrisch-Schwaben und Vorarlberg, Tübingen: Niemeyer. Eitle, Hermann 1914 Die Satzverknüpfung bei Chaucer (Anglistische Forschungen 44), Heidelberg: Winter. [ 1970] Amsterdam: Swets & Zeitlinger N.V. Emonds, Joseph E. 1985 A unified theory of syntactic categories, Dordrecht: Foris. Engel, Ulrich 1988 Deutsche Grammatik, Heidelberg: Groos. Faley, Jean 1991 Up oor close: Memories of domestic life in Glasgow tenements, 1910-1945, Wendlebury: White Cockade Publishing.

References

273

Fillenbaum, Samuel 1986 "The use of conditionals in inducements and deterrents", in Elizabeth Closs Traugott—Alice ter Meulen—Judy Snitzer Reilly—Charles A. Ferguson (eds.), On conditionals, Cambridge: Cambridge University Press, 179-195. Fillmore, Charles 1990 "Epistemic stance and grammatical form in English conditional sentences", in Michael Ziolkowski—Manuela Noske—Karen Deaton (eds.), Papers from the 26th Regional Meeting of the Chicago Linguistic Society, I (The Main Session), Chicago, 137-162. Filppula, Markku 1991 "Subordinating and in Hiberno-English syntax: Irish or English origin?", in P. Sture Ureland—George Broderick (eds.), Language contact in the British Isles, Tübingen: Niemeyer, 617-631. Fischer, Olga 1992 "Syntax", in Norman Blake (ed.), The Cambridge history of the English language, vol. II: 1066-1476, Cambridge: Cambridge University Press, 207-408. Foley, William A.—Robert D. Van Valin 1984 Functional syntax and universal grammar, Cambridge: Cambridge University Press. Fortune, Pete 1994 Author's self-description, in James Robertson (ed.), A tongue in yer held: A selection of the best contemporary short stories in Scots, Edinburgh: B & W Publishing, 200. Franz, Wilhelm 1939 Die Sprache Shakespeares in Vers und Prosa unter Berücksichtigung des Amerikanischen entwicklungsgeschichtlich dargestellt (4th revised and extended edition of ShakespeareGrammatik), Heidelberg: Carl Winter. Gaumann, Ulrike 1983 Weil die machen jetzt bald zu: Angabe- und Junktivsatz in der deutschen Gegenwartssprache (Göppinger Arbeiten zur Germanistik 381) Göppingen: Kümmerle. Geerts, Guido—Walter Haeseryn—Jaap de Rooij—M. C. van den Toorn (eds.) 1984 Algemene Nederlandse Spraakkunst. Groningen: WoltersNoordhoff.

274

References

Geis, Michael L. 1973 "If and unless", in B. B. Kachru (ed.), Issues in linguistics: Papers in honor of Henry and Renee Kahane, Chicago: Chicago University Press, 231 -253. Gelhaus, Hermann 1972 Vorstudien zu einer kontrastiven Beschreibung der schweizerdeutschen Schriftsprache der Gegenwart: Die Rektion der Präpositionen trotz, während und wegen, Bern—Frankfurt am Main: Peter Lang. Giegerich, Heinz J. 1993 English phonology: An introduction (Cambridge Textbooks in Linguistics), Cambridge: Cambridge University Press. Gillies, William 1993 "Scottish Gaelic", in Martin J. Ball (ed.), The Celtic languages, London—New York: Routledge, 145-227. Givon, Talmy 1984-1990 Syntax: A functional-typological introduction, 2 vols., Amsterdam—Philadelphia: Benjamins. 1990 Syntax: A functional-typological introduction, vol. II, Amsterdam—Philadelphia: Benjamins. Grant, William—James Main Dixon 1921 Manual of Modern Scots, Cambridge: Cambridge University Press. Greenbaum, Sidney 1980 "The treatment of clause and sentence in A Grammar of contemporary English", in Sidney Greenbaum—Geoffrey Leech—Jan Svartvik (eds.), Studies in English linguistics for Randolph Quirk, London: Longman, 17-29. 1996 The Oxford English Grammar, Oxford: Oxford University Press. Greenbaum, Sidney—Meyer, Charles F. 1982 "Ellipsis and coordination: Norms and preferences", Language & Communication!: 137-149. Greenbaum, Sidney—Quirk, Randolph 1990 A student's grammar of the English language, London: Longman. Gregor, W. 1866 The Dialect ofBanffshire, London. Grevisse, Maurice 1975 Le bon usage: Grammaire francaise avec des remarques sur la langue fran^aise d'aujourd'hui (10th edition), Gembloux: Duculot.

References

275

Günthner, Susanne 1993 "'... weil — man kann es ja wissenschaftlich untersuchen' — Diskurspragmatische Aspekte der Wortstellung in we/7-Sätzen", Linguistische Berichte 143: 37-59. Hacker, Martina 1994a "The death of English for and German denn: Linguistic change in progress", Grazer Linguistische Studien 42: 29-35. 1994b "Subordinate ami-clauses in Scots and Hiberno-English: Origins and development", Scottish Language 13: 34-50. Haegeman, Liliane V. M. 1985 Subordinating conjunctions and X'-Syntax (Studia Germanica Gandensia 2), Gent: Studia Germanica Gandensia. Haiman, John 1974 "Concessives, conditionals, and verbs of volition", Foundations of Language 11: 341 -359. 1978 "Conditionals are topics", Language 54: 565-589. 1986 "Constraints on the form and meaning of the protasis", in Elizabeth Traugott—Alice ter Meulen—Judy Snitzer Reilly— Charles A. Ferguson (eds.), On conditionals, Cambridge: Cambridge University Press, 215-227. Halliday, Michael K. M. 1985 An introduction to functional grammar, London: Arnold. 1986 "Constraints on the form and meaning of the protasis", in Elizabeth Closs Traugott—Alice ter Meulen—Judy Snitzer Reilly—Charles A. Ferguson (eds.), On conditionals, Cambridge: Cambridge University Press, 215-227. 1987 "Spoken and written modes of meaning", in Rosalind Horowitz— S. Jay Samuels (eds.), Comprehending oral and written language, San Diego: Academic Press, 55-82. 1994 An introduction to functional grammar (second edition), London—New York—Sydney—Auckland: Arnold. Hancock, Cecily Raysor 1993 "If he would have and if he didn't", American Speech 68: 241-252. Handke, Jürgen 1984 Descriptive and psycholinguistic aspects of adverbial subordinate clauses, Heidelberg: Goos. Hansen, Aage 1967 Moderne Dansk, vol. Ill: Sprogbeskrivelse, no place: Grafisk Forlag.

276

References

1994

1997

history and cognition. Papers from the international symposium at the University of Duisburg, 26-28 March 1990, Frankfurt am Main: Peter Lang, 429-453. Adverbial subordinators in the Languages of Europe: Towards a typology and history, vol. II (Eurotyp Working Papers V, 8), Strasbourg: European Science Foundation. Adverbial subordination: A typology and history of adverbial subordinators based on European languages (Empirical Approaches to Language Typology 18), Berlin—New York: Mouton de Gruyter.

Kress, Bruno 1982 Isländische Grammatik, München: Hueber. Labouvie, Erich 1938 Studien zur Mundart von Dillingen an der Saar, Marburg: Elwert. Lehmann, Christian 1988 "Towards a typology of clause linkage", in John Haiman— Sandra A. Thompson (eds.), Clause combining in grammar and discourse (Typological Studies in Language 18), Amsterdam— Philadelphia: Benjamins, 181-225. Letley, Emma 1988 From Gait to Douglas Brown: Nineteenth-century fiction and Scots language, Edinburgh: Scottish Academic Press. The Linguistic Atlas of Scotland 1975-86 J. Y. Mather—H. H. Speitel—G.W. Leslie (eds.), London: Croom Helm. Low, John Thomas 1974 "Scots in education: the contemporary situation", in J. Derrick McClure (ed.), The Scots language in education (Association for Scottish Literary Studies, Occasional Papers 3), Aberdeen: Association for Scottish Literary Studies, 17-27. Lyons, John 1977 Semantics, vol. II, Cambridge: Cambridge University Press. Macafee, Caroline I. 1983 Glasgow (Varieties of English Around the World, Text Series 3), Amsterdam: Benjamins. 1992-93 "A short grammar of Older Scots", Scottish Language 11-12: 10-36. Macafee, Caroline I.—Colm 0 Baoill 1997 "Why Scots is not a Celtic English", in Hildegard L. C. Tristram (ed.), The Celtic Englishes (Anglistische Forschungen 247), Heidelberg: Winter, 245-286.

References

277

McArthur, Tom 1992 The Oxford companion to the English language, Oxford: Oxford University Press. Macaulay, Ronald K. S. 1977 Language, social class, and education. A Glasgow study, Edinburgh: Edinburgh University Press. 1991 Locating dialect in discourse: The language of honest men and bonnie lasses in Ayr (Oxford Studies in Sociolinguistics), New York—Oxford: Oxford University Press. McClure, J. Derrick 1980 "Developing Scots as a national language", in J. Derrick McClure—A. J. Aitken—John Thomas Low, The Scots Language: Planning for modern usage. Edinburgh: The Ramsay Head Press, 11-41. Mair, Christian 1990 Infinitival complement clauses in English, Cambridge: Cambridge University Press. Matthiessen, Christian—Sandra A. Thompson, "The structure of discourse and 'subordination'", in John Haiman—Sandra A. Thompson (eds.), Clause combining in grammar and discourse (Typological Studies in Language 18), Amsterdam—Philadelphia: Benjamins, 275-329. Miller, Jim 1984 "Discourse patterns in spoken English", Sheffield Working Papers in Language and Linguistics 1: 10-39. 1993 "The grammar of Scottish English", in James Milroy—Lesley Milroy (eds.), Real English: The grammar of English dialects in the British Isles, London—New York: Longman, 99-138. 1995 "Does spoken language have sentences?", in F. R. Palmer (ed.), Grammar and meaning: Essays in honour of Sir John Lyons, Cambridge: Cambridge University Press, 116-135. Miller, Jim—Keith Brown 1982 "Aspects of Scottish English syntax", English World-Wide 3: 3-17. Mitchell, Bruce 1985 Old English syntax, 2 vols., Oxford: Oxford University Press. Mitchell, Bruce—Fred C. Robinson 1994 A guide to Old English, Oxford: Blackwell.

278

References

Kirk, John M. 1987 "The heteronomy of Scots with Standard English", in Caroline Macafee—Iseabail Macleod (eds.), The nuttis schell: Essays on the Scots language presented to A. J. Aitken, Aberdeen: Aberdeen University Press, 166-181. 1992-93 "Computing and research on Scots", Scottish Language 11-12: 75-131. Klemola, Juhani—Markku Filppula 1992 "Subordinating uses of and in the history of English", in Matti Rissanen—Ossi Ihalainen—Terttu Nevalainen—Irma Taavitsainen (eds.), History of Englishes: New methods and interpretations in historical linguistics. Papers given at the 6th International Conference on English Historical Linguistics May 1990, Berlin—New York: Mouton de Gruyter, 310-318. König, Ekkehard 1985a "On the history of concessive connectives in English: Diachronie and synchronic evidence", Lingua 66: 1-19. 1985b "Where do concessives come from? On the development of concessive connectives", in Jacek Fisiak (ed.), Historical semantics — historical word-formation (Trends in Linguistics, Studies and Monographs 29), Berlin—New York—Amsterdam: Mouton, 263-282. König, Ekkehard—Johan van der Auwera 1990 "Adverbial participles, gerunds and absolute constructions in the languages of Europe", in Johannes Bechert—Giuliano Bernini—Claude Buridant (eds.), Toward a typology of European languages, Berlin—New York: Mouton de Gruyter, 337-355. König, Ekkehard—Elizabeth Traugott 1988 "Pragmatic strengthening and semantic change. The conventionalizing of conversational implicature", in Werner Hüllen—Rainer Schulze (eds.), Understanding the lexicon: Meaning, sense and world knowledge in lexical semantics, Tübingen: Niemeyer, 110123. Kortmann, Bernd 1988 "Freie Adjunkte und absolute Konstruktionen im Englischen und Deutschen", Papiere zur Linguistik 38: 61-89. 1991 Free adjuncts and absolutes in English: Problems of control and interpretation, London: Routledge. 1992 "Reanalysis completed and in progress: Participles as source of prepositions and conjunctions", in Günter Kellermann—Michael D. Morrissey (eds.), Diachrony within synchrony: Language

References

1994

1997

279

history and cognition. Papers from the international symposium at the University of Duisburg, 26-28 March 1990, Frankfurt am Main: Peter Lang, 429-453. Adverbial subordinators in the Languages of Europe: Towards a typology and history, vol. II (Eurotyp Working Papers V, 8), Strasbourg: European Science Foundation. Adverbial subordination: A typology and history of adverbial subordinators based on European languages (Empirical Approaches to Language Typology 18), Berlin—New York: Mouton de Gruyter.

Kress, Bruno 1982 Isländische Grammatik, München: Hueber. Labouvie, Erich 1938 Studien zur Mundart von Dillingen an der Saar, Marburg: Elwert. Lehmann, Christian 1988 "Towards a typology of clause linkage", in John Haiman— Sandra A. Thompson (eds.), Clause combining in grammar and discourse (Typological Studies in Language 18), Amsterdam— Philadelphia: Benjamins, 181-225. Letley, Emma 1988 From Gait to Douglas Brown: Nineteenth-century fiction and Scots language, Edinburgh: Scottish Academic Press. The Linguistic Atlas of Scotland 1975-86 J. Y. Mather—H. H. Speitel—G.W. Leslie (eds.), London: Croom Helm. Low, John Thomas 1974 "Scots in education: the contemporary situation", in J. Derrick McClure (ed.), The Scots language in education (Association for Scottish Literary Studies, Occasional Papers 3), Aberdeen: Association for Scottish Literary Studies, 17-27. Lyons, John 1977 Semantics, vol. II, Cambridge: Cambridge University Press. McArthur, Tom 1992 The Oxford companion to the English language, Oxford: Oxford University Press. McClure, J. Derrick 1980 "Developing Scots as a national language", in J. Derrick McClure—A. J. Aitken—John Thomas Low, The Scots Language: Planning for modern usage. Edinburgh: The Ramsay Head Press, 11-41.

280

References

Macafee, Caroline I. 1983 Glasgow (Varieties of English Around the World, Text Series 3), Amsterdam: Benjamins. 1992-93 "A short grammar of Older Scots", Scottish Language 11-12: 10-36. Macafee, Caroline I.—Colm 0 Baoill 1997 "Why Scots is not a Celtic English", in Hildegard L. C. Tristram (ed.), The Celtic Englishes (Anglistische Forschungen 247), Heidelberg: Winter, 245-286. Macaulay, Ronald K. S. 1977 Language, social class, and education. A Glasgow study, Edinburgh: Edinburgh University Press. 1991 Locating dialect in discourse: The language of honest men and bonnie lasses in Ayr (Oxford Studies in Sociolinguistics), New York—Oxford: Oxford University Press. Mair, Christian 1990 Infinitival complement clauses in English, Cambridge: Cambridge University Press. Matthiessen, Christian—Sandra A. Thompson, "The structure of discourse and 'subordination'", in John Haiman—Sandra A. Thompson (eds.), Clause combining in grammar and discourse (Typological Studies in Language 18), Amsterdam—Philadelphia: Benjamins, 275-329. Miller, Jim 1984 "Discourse patterns in spoken English", Sheffield Working Papers in Language and Linguistics 1: 10-39. 1993 "The grammar of Scottish English", in James Milroy—Lesley Milroy (eds.), Real English: The grammar of English dialects in the British Isles, London—New York: Longman, 99-138. 1995 "Does spoken language have sentences?", in F. R. Palmer (ed.), Grammar and meaning: Essays in honour of Sir John Lyons, Cambridge: Cambridge University Press, 116-135. Miller, Jim—Keith Brown 1982 "Aspects of Scottish English syntax", English World-Wide 3: 3-17. Mitchell, Bruce 1985 Old English syntax, 2 vols., Oxford: Oxford University Press. Mitchell, Bruce—Fred C. Robinson 1994 A guide to Old English, Oxford: Blackwell.

References

281

Moessner, Lilo 1988 "Is the subject-verb concord in Scots a number concord?", in Richard Matthews—Joachim Schmole-Rostosky (eds.), Papers on language and mediaeval studies presented to Alfred Schöpf (Neue Studien zur Anglistik und Amerikanistik 37), Frankfurt am Main: Peter Lang, 177-191. 1993 "Scots — a regional variety of Standard English?", in Hans Ulrich Seeber—Walter Göbel (eds.), Anglistentag 1992 Stuttgart (Proceedings of the Conference of the German Association of University Teachers of English 14), Tübingen: Niemeyer, 154-163. 1997a "Economy as a principle of syntactic change", in Raymond Hickey—Stanislaw Puppel (eds.), Language history and linguistic modelling: A Festschrift for Jacek Fisiak on his 60th birthday, vol I: Language history, 357-372. 1997b "Middle English ^^-constructions", [Unpublished manuscript]. 1997c "The syntax of Older Scots", in Charles Jones (ed.), The Edinburgh history of the Scots language, Edinburgh: Edinburgh University Press, 112-155. Mosse, Fernand 1952 A Handbook of Middle English, translated by James A. Walker, Baltimore: Johns Hopkins. Murison, David 1979 "The historical background", in A. J. Aitken—Tom McArthur (eds.), Languages of Scotland (Association for Scottish Literary Studies, Occasional Papers 4), Edinburgh: Chambers, 2-13. 1987 "Scottish lexicography", in Caroline Macafee—Iseabail Macleod (eds.), The nuttis schell: Essays on the Scots language presented to A. J, Aitken, Aberdeen: Aberdeen University Press, 17-24. Murray, James A. H. 1873 The dialect of the southern counties of Scotland, London: The Philological Society. Murray, John 1994 Author's self-description, in James Robertson (ed.), A tongue in yer heid: A selection of the best contemporary short stories in Scots, Edinburgh: B & W Publishing, 204. Mustanoja, Tauno F. 1960 A Middle English syntax (Memoires de la Sociote Neophilologique de Helsinki 23), Helsinki.

282

References

Odlin, Terrence 1991 "Syntactic variation in Hiberno-English", in P. Sture Ureland— George Broderick (eds.), Language contact in the British Isles, Tübingen: Niemeyer, 597-615. OED = The Oxford English Dictionary 1989 James A. H. Murray et al. (eds.), 2nd edition prepared by J. A. Simpson and E. S. C. Weiner, Oxford: Oxford University Press. Oreström, Bengt 1985 A corpus of Shetland English (A eta Universitatis Stockholmiensis 64), Stockholm: Almqvist & Wiksell. 0 Siadhail, Micheäl 1984 "Agus (island: A shared syntactic feature", Celtica 16, 125-137. Partridge, A. C. 1969 Tudor to Augustan English: A study in syntax and style from Caxton to Johnson (The Language Library), London: Deutsch. Pasch, Renate 1997 "Weil mit Hauptsatz — Kuckucksei im Z)ew2-Nest?", Deutsche Sprache 25: 252-27 . Paul, Hermann —Hugo Moser—Ingeborg Schröbler 1975 Mittelhochdeutsche Grammatik (Sammlung kurzer Grammatiken germanischer Dialekte A. Hauptreihe 2), Tübingen: Niemeyer. Pawley, Andrew—Frances Hodgetts Syder 1983 "Two puzzles for linguistic theory: Nativelike selection and nativelike fluency", in Jack C. Richards—Richard W. Schmidt (eds.), Language and communication, London: Longman, 191-226. Petyt, K. M. 1985 Dialect and accent in industrial West Yorkshire (Varieties of English Around the World, General Series 6), Amsterdam— Philadelphia: Benjamins. The Pocket Scots Dictionary 1988 Isabeail Macleod—Ruth Martin—Pauline Cairns (eds.), Aberdeen: Aberdeen University Press. Polikarpov, Aleksandr [given as Polikarpow, Alexander] 1996 "Zum Problem der asyndetischen Subordination in der Syntax der gesprochenen deutschen Sprache", Deutsche Sprache 24: 154168. Poutsma, H. 1905 A grammar of Late Modern English for the use of continental, especially Dutch, students, Part I: The sentence, Section II: The composite sentence, Groningen: P. Noordhoff.

References

283

Primus, Beatrice 1997 "Der Wortgruppenbau in der Geschichte des Deutschen: Zur Präzisierung von synthetisch vs. analytisch", Sprachwissenschaft 22: 133-159. Quirk, Randolph 1995 Grammatical and lexical variance in English, London: Longman. Quirk, Randolph—Sidney Greenbaum—Geoffrey Leech—Jan Svartvik 1972 A grammar of contemporary English, London: Longman. 1985 A comprehensive grammar of the English language, London— New York: Longman. Reid, Euan C. 1976 "Social and stylistic variation in the speech of some Edinburgh schoolchildren" (unpublished M. Litt, thesis, University of Edinburgh). Robertson, James 1994 "Introduction", in James Robertson (ed.), A tongue in yer held: A selection of the best contemporary short stories in Scots, Edinburgh: B & W Publishing, vii-xvi. Robertson, T. A.—John J. Graham 1952 Grammar and usage of the Shetland dialect, Lerwick: The Shetland Times. Romaine, Suzanne 1975 "Linguistic variability in the speech of some Edinburgh schoolchildren" (unpublished M. Litt, thesis, University of Edinburgh). Rubenbauer, Hans, J.—J. B. Hofmann 1975 Lateinische Grammatik, revised by R. Heine, Bamberg: Buchner—München: Lindauer—München: Oldenbourg. Sandred, Karl Inge 1983 Good Scots or bad Scots? Attitudes to optional lexical and grammatical usages in Edinburgh (Acta Universitatis Upsaliensis, Studia Anglistica Upsaliensia 48), Uppsala: Uppsala University. Schiffrin, Deborah 1987 Discourse markers, (Studies in Interactional Sociolinguistics 5) Cambridge: Cambridge University Press. Schlieben-Lange, Brigitte 1992 "The history of subordinating conjunctions in some Romance languages", in Marinel Gerritsen—Dieter Stein (eds.), Internal and external factors in syntactic change, Berlin—New York: Mouton de Gruyter, 341-354.

284

References

Schmidt, Alexander 1971 Shakespeare-Lexicon: A complete dictionary of all the English words, phrases and constructions in the works of the poet, revised and enlarged by G. Sarrazin, Berlin: De Gruyter. Schrodt, Richard 1993 "Von der Diskurssyntax zur Satzsyntax: Reanalyse und/oder Grammatikalisierung in der Geschichte der deutschen Nebensätze", Folia Linguistica Historica 13: 259-278. The Scots Thesaurus 1990 Iseabail Macleod—Pauline Cairns—Caroline Macafee—Ruth Martin (eds.), Aberdeen: Aberdeen University Press. SND = The Scottish National Dictionary 1931-76 William Grant—David Murison (eds.), Edinburgh: Scottish National Dictionary Association. Speitel, Hans Henning 1969 "Some studies in the dialect of Midlothian", 2 vols. (unpublished Ph.D. thesis, University of Edinburgh). Stump, Gregory T. 1985 The semantic variability of absolute constructions, Dordrecht: Reidel. Svartvik, Jan—Randolph Quirk 1980 A corpus of English conversation (Lund Studies in English 56), Lund: W. K. Gleerup. Sweet, Henry 1891 New English grammar: Logical and historical, Part I: Introduction, phonology, and accidence, Oxford: Clarendon Press. [ 1960] Oxford: Oxford University Press. Sweetser, Eve 1988 "Grammaticalization and semantic bleaching", in Shelley Axmaker—Annie Jaisser—Helen Singmaster (eds.), Proceedings of the fourteenth annual meeting of the Berkeley Linguistics Society, Berkeley: Berkeley Linguistics Society, 389-405. 1990 From etymology to pragmatics: Metaphorical and cultural aspects of semantic structure (Cambridge Studies in Linguistics 54), Cambridge: Cambridge University Press. Templeton, Janet M. 1973 "Scots: an outline history", in A. J. Aitken (ed.), Lowland Scots (Association for Scottish Literary Studies, Occasional Papers 2), Edinburgh: Edinburgh University, 4-19. ter Meulen, Alice 1986 "Generic information, conditional contexts and constraints", in Elizabeth Closs Traugott—Alice ter Meulen—Judy Snitzler

References

285

Reilly—Charles Ferguson (eds.), On conditionals, Cambridge: Cambridge University Press, 123-145. Thomason, Sarah Grey—Terrence Kaufman 1988 Language contact, creolization, and genetic linguistics, Berkeley—Los Angeles—London: University of California Press. Thompson, Sandra A.—Robert E. Longacre 1985 "Adverbial Clauses", in Timothy Shopen (ed.), Language typology and syntactic description, vol. II: Complex constructions, Cambridge: Cambridge University Press, 171-234. Thomson, A. J.—A. V. Martinet 1980 A practical English grammar, Oxford: Oxford University Press. Todd, Loreto 1990 Words apart: A dictionary of Northern Ireland English, Gerrards Cross: Smythe. Tottie, Gunnel 1986 "The importance of being adverbial: adverbials of focusing and contingency in spoken and written English", in Gunnel Tottie— Ingegerd Bäcklund (eds.), English in speech and writing: A symposium (Acta Universitatis Upsaliensis, Studia Anglistica Upsaliensia 60), Uppsala University, 93-118. Traugott, Elizabeth Closs 1982 "From propositional to textual and expressive meanings: Some semantic-pragmatic aspects of grammaticalization", in Winfred P. Lehmann—Yakov Malkiel (eds.), Perspectives on historical linguistics (Current Issues in Linguistic Theory 24), Amsterdam: Benjamins, 245-271. 1988 "Pragmatic strengthening and grammaticalization", in Shelley Axmaker—Annie Jaisser—Helen Singmaster (eds.), Proceedings of the fourteenth annual meeting of the Berkeley Linguistics Society, Berkeley: Berkeley Linguistics Society, 406-416. Trudgill, Peter 1990 The dialects of England, Oxford: Blackwell. Trudgill, Peter—Jean Hannah 1994 International English: A guide to the varieties of Standard English (3rd edition), London—New York—Melbourne—Auckland: Arnold Tulloch, Graham 1980 The language of Sir Walter Scott: A study of his Scottish and period language (The Language Library), London: Deutsch, van Kerckvoorde, Colette M. 1993 An introduction to Middle Dutch, Berlin—New York: Mouton de Gruyter.

286

References

Visser F. Th. 1966 An historical syntax of the English language, vol. II/I, Leiden: Brill. Wackernagel-Jolles, Barbara 1971 Untersuchungen zur gesprochenen Sprache: Beobachtungen zur Verknüpfung spontanen Sprechens, Göppingen: Kümmerle. Wakelin, Martyn F. 1986 The Southwest of England (Varieties of English Around the World, Text Series 5), Amsterdam—Philadelphia: Benjamins. Wegener, Heide 1993 "Weil — das hat schon seinen Grund. Zur Verbstellung in Kausalsätzen mit 'weil' im gegenwärtigen Deutschen", Deutsche Sprache 21,289-305. Wessely, Gerda 1981 Nebensätze im spontanen Gespräch: Dargestellt an der Mundart von Ottenthal im nördlichen Niederösterreich, Wien: Braumüller. Wettstein, Paul 1942 The phonology of a Berwickshire dialect, Bienne: Schüler. Willems, Klaas 1994 "we/7 es hat mit Bedeutung nicht viel zu tun ... Zum Sprachwandel einer Konjunktion", Deutsche Sprache 22: 261-279. Williams, Ann 1989 "Dialect in school written work", in Jenny Cheshire—Viv Edwards—Henk Münstermann—Bert Weltens (eds.), Dialect and education, Clevedon: Multilingual Matters, 182-189. Williamson, Keith 1992-93 "A computer-aided method for making a linguistic atlas of Older Scots", Scottish Language 11-12: 138-172. Wilson, James 1915 Lowland Scotch as spoken in the Lower Strathearn district of Perthshire, Oxford: Oxford University Press. Wright, Joseph 1905 The English dialect dictionary, Oxford: H. Frowde. Zaandvoort, R. W. 1966 A handbook of English grammar, London: Longmans [sic]. Zettersten, Arne 1987 "Teaching Middle Scots by computer", in Caroline Macafee— Iseabail Macleod (eds.), The nuttis schell: Essays on the Scots language presented to A. J. Aitken, Aberdeen: Aberdeen University Press, 124-128.

Maps A note on maps Map 1 illustrates the major dialect divisions as they are given in the Scottish national dictionary and the Concise Scots dictionary. Central Scots, which is the designation used in the CSD, is preferred to Mid Scots, which is the corresponding term in the SND. The "Highland Line" is the division between the Scots-speaking Lowlands and the Gaelic-speaking or formerly Gaelicspeaking Highlands. Map 2 represents regional divisions as established in the study of adverbial clauses. The boundaries shown are not meant as cartographic representations, but rather as indications of regional differences. Although no data was available from South Ayrshire, it seems probable that the features specific to Western Scots occur also south of Ayr, as this area is geographically separated from south-eastern Scotland by a sparsely populated area of hills and moors and socially and culturally oriented towards Ayr and Glasgow. The area for which no material was available is marked by dotted lines. Maps 3 and 4 locate the regional affiliation of authors and speakers respectively. In some of the spoken texts, therefore, the location marked on the map differs from the place of recording. The regional affiliation of one text could not be established with certainty, and this is indicated by a question mark. For one text two locations are given because the speaker in his formative years continually moved between them. Map 5 gives an overview of the regional affiliation of all texts.

288

Maps

Map 1: Scots dialect regions according to the SND

Maps

Map 2:

Regional divisions based on the present study

289

290

Map 3:

Maps

Regional distribution of spoken texts

Maps

Map 4:

Regional distribution of written texts

291

292

Map 5:

Maps

Regional distribution of written and spoken texts combined

Author Index Aarts, F. G. A. M. 84 Aarts, Jan 256 Abbott, E. A. 208 Abercrombie, David 7 Abraham, Werner 257 Aitken, A. J. 7, 8,10, 255 Allan, Robin 229,233,263 Altenberg, Bengt 135, 148-149, 168 Bache, Carl 117 Bäcklund, Ingegerd 106 Barber, Charles 208 Barnes, Michael 233 Barry, Michael V. 34 Bayer, Josef 219 Seal, Joan 10,44, 116, 175, 216, 257, 259 Beaman. Karen 16 Behagel, Otto 234, 263 Biber, Douglas 16 Bliss, Alan 34 Bloomfield, Leonard 12 Bolinger, Dwight 26 Borrowman, Lorna 10 Boyle, Daniel 34 Braunmüller, Kurt 37, 46, 263 Brown, Keith 10-12, 121,260 Buchholz, Oda 264 Burnham, J. M. 230 Calder, George 229 Chafe, Wallace 12-14, 26 Chambers, J. K. 3,48 Cheshire, Jenny 3, 258 Clyne, Michael G. 257 Comrie, Bernard 115-116 Couper-Kuhlen, Elizabeth 258 Croft, William 243 Curme, George O. 93-94, 97, 126, 142, 175, 187, 193,196,238,257,261 Danielewicz, Jane 13 Davidsen-Nielsen, Niels 117 Declerck, Renaat 76, 79 Dieth, Eugen 9 Dixon, James Main 9, 48, 192, 230, 244, 258 Donaldson, Gordon 261 Eaton, Helen S. 263

Edgren, Eva 58, 71-72, 75, 85, 89, 92, 259 Edwards, Viv 3,258-259 Einarsson, Stefan 232 Eisenmann, Fritz 257 Eitle, Hermann 208, 226 Emonds, Joseph E. 22 Engel, Ulrich 234, 259 Faley, Jean 62 Fillenbaum, Samuel 117, 260 Fillmore, Charles 114-116, 260 Filppula, Markku 34 Fischer, Olga 208, 214 Foley, William A. 22, 256 Fortune, Pete 148 Franz, Wilhelm 53-55, 76, 107, 193, 208, 226, 229-230, 233, 262, 264 Gaumann, Ulrike 257 Geerts, Guido 218 Geis, Michael L. 104 Gelhaus, Hermann 264 Giegerich, Heinz J. 7 Gillies, William 229 Givon,Talmy 18,48,258 Graham, John J. 201, 232-233 Grant, William 9, 48, 192, 230, 244, 258 Greenbaum, Sidney 15, 35, 41, 256 Gregor, W. 9 Grevisse, Maurice 219, 234 Günthner, Susanne 257 Hacker, Martina 34-35, 233, 257, 264 Haegeman, Liliane V. M. 22, 24 Haeseryn, Walter 218 Haiman, John 100, 117, 227, 258 Halliday, Michael K. M. 13-14, 16, 21, 255 Hancock, Cecily Raysor 114, 116, 205 Handke, Jürgen 228, 230, 263 Hannah,Jean 94 Hansen, Aage 228-229, 263 Hansen, Erik 244 Harris, John 34, 84, 88, 116, 175, 216, 245, 259 Harris, Martin 257 Hartig, Matthias 38, 257

294

Author Index

Heller, J. 234 Hengeveld, Kees 24, 256 Henry, P. L. 34 Hockett, Charles F. 13 Hermann, J.B. 229 Holmes, Philip 229, 233, 263 Hopper, Paul J. 22, 37, 48, 243 Huddleston, Rodney 41 Hughes, Arthur 127, 260 Ihalainen, Ossi 3 Jacobsson, Bengt 114, 135 Jespersen, Otto 29, 38, 44, 55, 80, 85, 88, 103, 114-115, 126, 134, 137, 182, 189, 208, 215, 225-226, 228, 231232, 257, 260 Johnston, Paul 237 Jones, Charles 10 Kaufman, Terrence 208 Kay, Billy 6-8, 11,61 Keller, Rudi 257 King, Anne 10 Kirk, John M. 10-12, 121 Klemola, Juhani 34 König, Ekkehard 26, 230, 263 Kortmann, Bemd 26, 34, 37-39, 46-47, 97, 193, 196, 206-208, 214, 221-222, 224, 227, 229-231, 238-240, 256, 262-263 Kress, Bruno 232 Labouvie, Erich 257 Lehmann, Christian 22 Letley, Emma 10 Longacre, Robert E. 22, 37-38 Low, John T. 10 Lundskasr-Nielsen, Tom 229, 233, 263 Lyons, John 255-256 Macafee, Caroline I. 5, 10-11, 34, 145, 258-259, 261 Macaulay. Ronald K. S. 10, 84, 174 Mair, Christian 175, 182,261 Martinet. A.V. 157-158 Matthiessen, Christian 256 McArthur, Tom 145 McClure, J. Derrick 7, 10 Meyer, Charles F. 35 Millar, Martin 10, 121,260 Miller, Jim 10-13, 15-16, 80, 86, 115-116, 121, 174, 205, 242, 245, 255, 259-261 Milroy, Jim 3, 10

Milroy Lesley 10 Mitchell, Bruce 44, 114, 208, 244 Moessner, Lilo 10-11, 103, 262 Moser, Hugo 220, 262 Mossd, Fernand 44, 175, 208, 261 Murison, David 9, 255 Murray, James A. H. 9 Murray, John 131 Mustanoja, Tauno F. 208, 216, 226, 228, 230-231 Nielsen Hans Frede 244 0 Baoill, Colm 5 Odlin, Terrence 34 Oreström, Bengt 201, 232 0 Siadhail, Micheäl 34 Partridge, A. C. 30, 148, 208, 229, 233 Pasch, Renate 257 Paul, Hermann 220, 226, 262 Pawley, Andrew 76, 104, 120 Petyt,K. M. 116,259 Polikarpov, Aleksandr [given as Polikarpow, Alexander] 27, 242 Poutsma, H. 196, 257 Primus, Beatrice 257 Quirk, Randolph 15, 23-24, 29, 34-36, 41, 44-45, 56, 59, 64, 66, 71, 76, 79, 9394, 98, 103-104, 112, 114, 116-118, 121, 125-127, 134-135, 143, 148, 157-158, 162, 168, 175, 182, 184, 187-190, 193, 196, 208, 216, 256257, 259-262 Reid, Euan C. 10 Robertson, James 8 Robertson, T. A. 201, 232-233 Robinson, Fred C. 208 Romaine, Suzanne 10 Rooij, Jaapde218 Rubenbauer, Hans J. 229 Sandred, Karl Inge 10 Schiffrin, Deborah 25 Schlieben-Lange, Brigitte 206-207, 264 Schmidt, Alexander 54, 85, 208 Schröbler, Ingeborg 220, 262 Speitel, Hans Henning 9 Stump, Gregory T. 256 Svartvik, Jan 127 Sweet, Henry 21,256-257 Sweetser, Eve 24, 227 Syder, Frances Hodgetts 76, 104, 120

Author Index

Templeton, Janet M. 255 ter Meulen, Alice 100 Thomason Sarah Grey 208 Thompson, Sandra A. 22, 37-38, 256 Thomson, A. J. 157-158 Todd, Loreto 88, 257,259 Toom, M. C. van den 218 Tottie, Gunnel 168-169, 183, 190 Traugott, Elizabeth Closs 22, 37, 48, 227, 229, 243, 256 Trudgill, Peter 3, 48, 94, 127, 259-260 Tulloch, Graham 10 van der Auwera, Johan 26, 263 van Kerckvoorde, Colette M. 220 Van Valin, Robert D. 22, 256

Visser F. Th. 76, 127, 208, 221, 232 Wackernagel-Jolles, Barbara 14-15 Wakelin, Martyn 259 Wegener, Heide 257 Weltens, Bert 258 Wessely, Gerda 257 Wettstein, Paul 9 Whittle, Pamela 258 Willems, Klaas 257 Williams, Ann 3 Williamson, Keith 10 Wilson, James 6, 9, 177, 192 Wright, Joseph 64 Zaandvoort, R. W. 29, 257 Zettersten, Arne 10

295

General Index Placenames and language varieties are listed together, e.g. North-Eastern Scots, north-eastern Scotland. Placenames that occur only in examples from the database, e.g. Cyprus, are not included. Adverbs, prepositions, co-ordinators, and subordinators are given in italics; they appear without descriptors when they are present-day Scots, present-day Standard English, Middle English or Early Modern English. In all other cases the relevant language is indicated in parentheses in abbreviated form such as: Dan. = Danish; Du. = Dutch; Fr. = French; Gael. = Gaelic; Ger. = German; Gmc. = Germanic; Goth. = Gothic; Icel. = Icelandic; It. = Italian; Lat. = Latin; OE = Old English.

a(Fr.)219 ab (Lat.) 219 Aberdeen 8 Aberdeenshire 250, 261 absolutes 26, 256 academic discourse see discourse, spoken and written accent 7, 8 accentedness see stress acceptable, acceptability 25, 37, 100, 130, 243 actual situation 56-57, 64, 129, 132 ad (Icel.) 217 ad (Lat.) 219 Addison, Joseph, essayist 231 adjunct 24-26, 52, 129-130, 256 adjunct—disjunct gradience see gradience of adverbial clauses adposition—adverbial subordinator— adverb continuum 47 adverbial subordinator inventories 37, 189, 216, 222, 237, 262; Old English 209; Middle English 209, 211-213; Early Middle English 210, 212-213; present-day English 210-214; presentday Scots 2 -214 adverbial subordinators: complex 36; marginal 36; morphological structure of 4, 36-46, 207-208, 214, 217, 220, 225, 232,244,257,261-262 adverbial subordinator—conjunctional adverb—co-ordinator continuum 47 adverbs 38, 47, 52, 66, 88. 139-140, 214215, 219, 227, 229, 238, 242, 244-245, 260, 263; interrogative adverbs 54 qffinche (It.) 229

afore 56-58, 64, 94, 178-180, 203-204, 215 after 39, 79-82, 95-96, 216, 241, 245 afterthought 14, 16, 20, 104, 203 again/against 70, 258 age of speakers 19, 174-175, 247, 249250 alft) 230 albeit 141, 203 aless 127-128,201,232 all (Ger.) 230 als (Ger.) 226 although 42-44, 138-141, 149,243 a/vore«Jte(Du.)218 amalgamation 93, 102, 109, 118, 239, 264 ambiguity, ambiguous 39-40, 81-82, 116, 171, 184, 186-188, 241-242; structural ambiguity 40, 99, 196; semantic ambiguity 53 American English 93-94, 116, 134, 205, 231,238,260 anaphoric reference 29, 152, 167, 188 and: concessive subordinator 34-36, 45, 138, 145-146; causal subordinator 150, 165; co-ordinator 102, 109, 117, 147 Andes (Artois) 234 Angles 5 Angus 63,250 ansialt daß (Ger.)2\9 anstatt zu/zum (Ger.) 218-219 antecedent 40, 48 any 68, 238, 258 anywhere 52-55, 203, 231 apart from 127-128 aphetic see reduction apodosis 114, 123, 205, 239, 241

General Index Arcadia 234 archaic 1 75, 203-204 areal see geographical Artois 234 as 222; time 65, 67, 95; restriction 124125; comparison 129-130, 134; concession 138, 144, 149; reason 150, 167-168, 202, 204; general subordinator 215 as far as 124, 126,230 ay gin 130, 132,134,202 asi/130-132, 134 as lang as 201, 228; time 65, 75; condition 99, 103; reason 167 as soon as/seen 's 79, 86-89, 93-96, 259 as though 130-131, 134 as who say(s)/as who say 232-233 assuming 123 at (subordinator) 216 Australian English 44 Austrian German 257 autant que (Fr.) 231 Ayr 175, 283 Ayrshire 84, 154, 202, 250, 283 Balkan languages 264 Barbieri, Giovanni Francesco, artist 234 bare(at)(Oan.)2\7 barrin99, 104, 123,203 because/cos 45, 150-152, 165-169, 227, 244; conjunct 151-152, 239, 244 because of 1 50, 163,204 before 56-64, 94, 179-180, 203-204, 215216 oeim(Ger.)219 Bemicia 5 (bh)o (Gael.) 220, 228 Bible 6 Borders (The) 84, 236, 248 borrowing 207, 214-215, 232 Braid Scots see Broad Scots Broad Scots 7-8, 114, 174 Brown corpus (Brown) 55, 126 Buchan 163,236,250 Burns, Robert, poet 6-7 but 42-45, 244; conjunct 140, 244, 257; co-ordinator 147-148 butthat 128

297

bulon/butan (OE) 44,214 by (the) cause (that) 215 by 16, 191 bythat2\4 by the time 65, 75-76, 225 Cairn, periodical 255 Caithness 5, 33, 177, 201-202, 236-237, 250 Campbell, Archibald, 9th earl of Argyll 131 Celtic 35, 220 Celtic influence on English 34-35,228 Central Scots 237 Chapman, periodical 255 Charlestown of Aberlour 247, 249 Chaucer, Geoffrey, poet 216,226-227 Classical Latin see Latin clause adjunct 24-26 clause complex 13, 23, 256 clause linker, clause linking element 4546, 257 clause order 42, 130, 161, 184 Cockburn, Henry Thomas, judge 6 cognate 56, 70, 225 cognitive relatedness 46, 227 cognitive semantics 227 come/came 28, 76-77 comma see punctuation comma intonation see intonation co/nme(Fr.)219, 222 complement, complementation 22, 38, 203, 256; subject complement 142-143, 145, 150, 160, 186, 189; predicative complement 23, 30,46, 52, 195 complement clause 39-40, 256 complementiser 21-22, 38, 40 complementiser—adverbial subordinator—relativiser continuum 47 compound, compounding 9, 47, 53; see also structure of subordinators compound sentence see co-ordinated clause Concise Oxford Dictionary (COD) 162, 257 Concise Scots Dictionary (CSD) 5, 7, 6263, 83, 156-157, 202, 255, 259-260, 260 conditional component of meaning 75, 79, 117

298

General Index

conjunction 38, 263; see also subordinator, co-ordinator conjunctive adverbs see conjuncts conjuncts 4, 38, 40-41, 44-45, 48, 140, 148, 151, 153, 166, 169, 185-188, 190, 239, 243, 257, 262 connective, connector see clause linker conservative 55, 215-217, 239, 243, 262 considering that 2 1 4 content adverbials 24 continuative meaning 64 co-ordinated clauses 29, 35, 41-45, 109, 245 co-ordination—subordination gradient 41 co-ordination—subordination — conjunct triangle 45 co-ordinator, co-ordinate 14, 29, 35-36, 38, 40-42, 44, 135, 233, 239, 243, 257, 261 co-ordinator—subordinator— conjunct gradient 48; see also co-ordination— subordination—conjunct triangle coreferentiality, coreferential see subject correlative 92, 238-239, 245 could \n. 262 counterfactual see factuality court, Scottish 5-6 Creole languages 3 cum (Lat.) 206 cum inversum 66, 258 Cumbric 5 cyclic development of language change see language change da (Dan.) 226, 259 da wye at 150, 156,201,232 da/m?(Ger.)219 danach (Ger.) 39 Danish, Danes 5, 62, 217-218, 221-229, 232-233, 235, 240, 259, 262-263 iAz0(Ger.) 2 18, 262-263 daf(Du.)218 David I, king of Scotland 5 afe(Lat.)219 dependency relation of clauses 22, 35 derweilst (Ger.) 46 despite the fact that 138, 143 dialect grammar 3-4 7, 9-10, 12, 241-242

Dictionary of the Older Scottish Tongue (DOST) 9, 55 dimensions of discourse structure 16; see also discourse, written and spoken directive 117,204 directly 93-94, 225 discourse, written and spoken 3, 10, 1213, 15, 17-18, 97, 115, 135, 187, 193, 201, 202, 203, 237, 242; academic discourse 14, 121, 135 disjunct: content disjunct 24-26, 132; style disjunct 24-26, 51, 167 donee (Lat.) 229 donicum (Lat.) 229 Doric 8 Douglas, Gavin, poet 6, 208 du (Goth.) 229 dum(Lat.)229 Dumfriesshire 247, 249-250 Dunbar, William, poet 6, 208 Dundee 236, 247-248, 250 duration, durative, non-durative 57-58, 71, 73,76, 188,204 Dutch 4, 39, 217-218, 220-221, 223-227, 232,238,241,259 dynamic situation 59, 61 E-adverbials 24 Early Modern English 4, 30, 85, 107, 175, 189, 193, 206-208, 210, 212-213, 216, 221-222, 232-233, 238, 262, 264 economy 207, 235, 243, 262 Edinburgh 5, 91 108, 154, 236, 247, 249250 educated speakers 114, 148, 149, 205, 242; see also education and discourse education 6, 10, 242, 265 efter 79-80, 90, 95 efier(Oan.)2n efter at (Dan.) 217-218 elicitation 17 Elliot, Walter, local historian 202 ellipsis, elliptical 30, 35, 41, 104, 146, 183,261 embedding, embedded 2, 14-15, 22-23, 256 emotive meaning 243 emphasis, emphatic 27-28, 35, 56, 60, 103, 130, 143, 145, 176, 230, 233-235, 264

General Index

299

en (Fr.) 219 England 62 enumeration 14-15 epistemic adverbials 24 ere 56, 62-64, 94,215 et (Fr.) 234 e/(Lat.) 233-234 Ettrick 247 etymology 36-37, 51, 69, 102, 126, 156, 218,220,226,229,232 even for as 138, 141-142 eve« //149 even though 141-142, 149 every 68, 238, 258 every time (when) 65, 68-69, 94 everywhere 52-55, 203, 231 ex(Lat.)219 except (that) 127, 207, 216 excepting that 128, 207 expansion, grammatical 36-37 expected, expectedness 64, 196; see also unexpectedness

formation patterns of subordinators see subordinates fourth order adverbials 24 fra (Dan.) 217, 228 frae see fae/frae/faem/from fragmentation 14 from (OE) 228 free adjunct 26, 256 free construction 35 French, France 4, 38, 206, 208, 214, 216217,219-224,231-232 frequency, frequent 33, 65, 90, 92-96, 146, 149, 154-156, 167-168, 175, 183, 189-190, 193, 196-197, 203, 231, 235, 239, 264 fronting 56, 160; as functional test 42 functional change 47; see also language change and syntactic change functional domain, continua of 47 functional polysemy 3-4; see also polyfunctionality functional typology 3-4, 22, 48, 243

factual, factuality 58, 112, 133, 180, 205, 227,241 fae/frae/faem/from 215, 228; time 79, 8385,201, 203; reason 150, 157,229 /a//s(Ger.)219 Fife 8, 250 finis (Lot.) 229 fixed expression 30, 34, 62, 76, 104, 106, 111-112, 119-120, 180 folk etymology 160,233 far (Dan.) 217 for 41, 44, 45, 150, 152-153, 162, 166169; with infinitive 173-177, 182, 204, 244, 261; conjunct 152-153, 239, 244, 257 for aa/for all (that) 229-230, 235; restriction 124-125, 231-232; concession 138, 141-142, 229-231, 245 for at (Dan.) 217 for aught 229-230 for because 215 for fear 150, 158 for fear οφ 150,183 for what 230 formal vs. informal language 3, 93, 135, 141, 143, 148, 167-168, 193, 196, 240

Gaelic 4, 201, 217, 219-220, 223-224, 228-229, 263 Gaelicised English 5 Galashiels 247, 249 Galloway 19,84 gegen (Ger.) 70 genera] subordinator marker 214-215, 217-220,262-263 generative grammar, generativists 21-22, 38 generic situation 67-68, 70, 80, 99-102, 123, 138, 155,225 genetical relatedness 46, 227 genre 183; see also discourse and dimensions geographical boundaries 201, 245 geographical distribution 35, 124; see also regional restriction and regional variation geographical relatedness 46, 227 geographical restriction 7, 232; see also regional restriction German, Germans 27, 38-40, 45-46, 70, 115, 119, 217-219, 222-224, 232, 234235, 240 Germanic languages 38, 46, 216-217, 220, 223, 225, 227, 229, 240, 263

300

General Index

gerondif219,221 S//107, 109,202,214-215 gin 202, 205, 239, 241, 226, 258; time 65, 69-71, 94; condition 99, 101-102, l ΙΟΙ 11, 121-122; comparison 130-131, 134 given (that) 123 Glasgow, Glaswegian 8, 9, 12, 82, 127, 154,236,247-250,258-259 Gorbals 247,249 Govan 247, 249 gradience of centrality of adverbial clauses 22-26 gradience of clause integration 22, 51, 152 Graham, James, marquis of Montrose 131, 261 grammaticalisation 3, 36-38, 40, 48, 104, 219, 227, 245 grammaticality, grammatical, ungrammatica!40, 176 Grieve, Christopher see MacDiarmid, Hugh gu (Gael.) 220, 229 gun (Gael.) 220 gus (Gael.) 220 gus an (Gael.) 220 gus gun (Gael.) 229 Gutter Scots 8 habitual situation 60, 68, 80, 86, 100, 123, 225, 227 habitually recurrent situation 56, 67-68, 70, 76, 79, 88, 113, 226, 238, 241 had 116 hadda/had have + past participle 115-116 Hawick 8, 237, 247, 249 hedging 121-122, 124, 134, 181 Henryson, Robert, poet 6, 208 Hiberno-English see Irish English Highland Line 5, 283 Highlands 5, 283 historic present 58, 83, 258 however 42-44, 243, 260 hypotaxis, hypotactic 21-22 hypothetical situation 29-30, 57, 70, 82, 91,99, 129,226,239 Icelandic 217, 223-224, 232 iconicity 207, 235 Ideal Scots 8 identity 12

i/225-226; time 71, 85; condition 98-103, 105-108, 110,118-119, 121 ifald (Dan.) 217 immediacy 87, 259 immediately, 93-94, 225, 259 imperative clause 41 ;«(Lat.)219 in as much as 167 in case/in cos 215; reason 157-158; condition 123 in order that 182 in order to 182 in plaats van te (Du.) 218 in so far that 215 in spite of the fact (that) 138, 143 in that 167 in the event that 123 indefinite direction 53 indefinite place 53 indefinite time 69 inden(Oan.)2l7 indtil(Oan.)2ll inference 22, 24, 153, 185 informants 17, innovative 55, 59, 205 insomuch as 215 instantly 225 instead of \35-\36 intensifier 29, 45,186,231 intonation 13, 15, 20, 25-26, 28-30, 152, 166, 187-188, 242; sentence final intonation 14, 20; comma intonation 28; period intonation 166; intonation vs structural marking 15 introspection 17, 256 intuition 17,41,227 inventory of adverbial subordinators see adverbial subordinator inventories inversion 29, 110-111,242 Irish English, Ireland 35, 48 64, 84-85, 88, 116,157,175,244,259 Italian 229 iteracy 226-227, see also habitually recurrent situation Jack, Alan, radio journalist 8-9 James VI, king of Scotland Johnson, Samuel, lexicographer 207 just so 123 juxtaposed clause 41-42, 147-148

General Index

ken\2\ Kenneth II, king of Scotland 5 Lallans 6-7 Lallans, periodical 255 Lanarkshire 8, 88, 105, 136, 250 Lancashire 127 Lancaster-Oslo/Bergen Corpus (LOB) 55, 94, 96, 97, 126, 135, 149, 168, 208, 257, 260 language change 17, 47-48, 201, 204, 221-222, 240, 243; cyclic development of language change 243-244; see also syntactic change language contact 4, 34-35, 46, 207-208, 217,221,228 language variation 17; see also regional restriction and geographical distribution Latin 206-208, 216, 219, 223-224, 229, 231,233-235,241 Laverock, periodical 255 Leith 172, 248, 250 Leslie, Alexander, 1st earl of Leven 261 Leslie, David, general 261 lest Π&, 202 lexicalisation see grammaticalisation lexicography 9 /igesom(Dan.)217 like 130-131, 134 Lincolnshire 35, 231,264 Linguistic Atlas of Scotland 10 literacy 206, 243 logico-semantic relation 21 London 5 London-Lund Corpus (LLC) 55, 94, 97, 126-127, 135, 149, 168-169, 208, 260, 264 Lothian 5 lower class see working-class Lowland Scotland 5, 283 MacDiarmid, Hugh, poet 8 Macdonald, George, author 48 markedness, marked, unmarked 27, 65, 258 meaning, extension of see semantic extension Meams (The) 69, 247 medens (Dan.) 222

301

metaphorical extension see semantic extension Middle English 9, 12, 37, 175, 207-210, 212-213, 215-216, 221, 228, 230, 232233, 238, 243, 262 Middle High German 226 Middle Scots see Older Scots Midlothian 247 mind\2\ misinterpretation see ambiguity misunderstanding see ambiguity mi/(Ger.)219 modal, modality 82, 157, 170, 176, 182, 184, 226-227,238, 245, 261-262 modifier, modifying function 23, 159 Moffat 247, 249 monofunctional subordinates 214, 224225 Montrose see Graham, James mood 114, 227; see also modality Morayshire71,247, 249 multiple functions see polyfimctionality Musselburgh 247, 249 na te (Du.) 218 «acA(Ger.)39,218 nachdem (Ger.) 19, 218 nae sooner... than/when 92-93, 239 nar (Dan.) 226 narrative, as text genre 16, 18-19, 65-66, 105, 148,255,258 narrative when see cum inversum negative, negation 103, 106, 178-180, 202; negative matrix clause 57, 60 network, semantic 222 neutralisation, neutralised 68, 100, 103, 114,156 Niddrie 108, 247,249 no sooner ... than/when see nae sooner ... than/when nominal isation 256 non-factual vs see factuality non-specific situation 67, 86, 99, 124, 138, 175,182 Nom 232-233 North Insular Scots, Northern Isles 237; see also Orkney and Shetland North-Eastem Scots, north-eastern Scotland 57, 62, 75, 83, 88, 124, 136, 142,

302

General Index

164, 185, 189, 201, 204, 216, 236, 259260 North-Northern Scots 182, 216; see also Caithness Northern English, northern England 101, 127, 169, 258, 264 Northern Irish English see Irish English Northern Scots 63, 174, 177, 182, 201202 Northumbrian, Northumberland 5, 48, 64, 116, 175,259 Norwegian 262 now that 90, 227 nu (Du.) 226 ο (Gael.) 220 obgleich (Ger.) 224 object 176 obligatory clause constituent 23, 40, 256 obschon (Ger.) 224 oowo/j/(Ger.)219, 224, 260 Odense. Denmark 228 og (Dan.) 233 og(Icel.)217 oA>w>i/a/8(Ger.)219 ohne zu/zum (Ger.) 218-219 okay 107, 108 Old English 9, 206, 208-209, 214, 221, 226, 228, 262 Old French 234-235; see also French Old Norse 5, 207, 215 Old Northumbrian 5 Old Scandinavian see Old Norse Older Scots 9- 10 twite (Du.) 21 8 omission as functional test 40 on 77-78, 202 on condition 1 23 once/yince/eence 79, 85-87, 91-93, 96, 241 onielime 65, 68-69, 94 operator contraction 116 optional clause constituent 23, 40 or 56, 62-63,94, 201,204, 215 order of events 56, 58, 64-65, 89, 187 Orkney 5

orthography, orthographical 13, 45, 247; orthographical assimilation 174; see also punctuation and spelling ow(Fr.)219 outnome that 214-215 outtaken that 207 outwith 1 overlap of syntactic functions see polyfunctionality Oxford English Dictionary (OED) 55, 6163, 70, 85-86, 88, 102, 128, 138, 188189,215,229-230,258,261 P-Celtic 5 par (Fr.) 219 parataxis, paratactic 21-22, 29, 243-244 parceque(¥r.)2\9 parenthetical clause 14 participle, participial 47-48, 217-220, 225, 228, 238 particle 38 parts of speech see word class passive 31 past perfect 58, 75, 79-81, 86-87, 114; see also tenses past tense 76, 80, 82, 86, 113, 132; see also tenses pause 13, 26, 43,45, 152, 187 Peebleshire 247,250 percentage see proportion perception of dialects 9, 12, 18 period see punctuation personal letters 14 Perthshire 177 phonetics see phonology phonology 9, 29, 32, 51-52, 62, 69, 156, 174,233,236,258,262 Pinkerton, John, historian 11 Plastic Scots 7-8 Pocket Scots Dictionary 9 polyfunctionality; syntactic 4, 27, 34, 36, 38-42. 44, 46, 218, 239; semantic see polysemy polymorpheme, polymorphemous 37, 206; see also morphological structure of subordinators polysemy 46, 208, 224, 227, 229-231, 240, 264 Port Glasgow 247, 250 position of constituents 150-151

General Index positional restriction 66, 79, 170 potential situation 123-124, 132, 157, 163, 225 Poussin, Nicolas, artist, 234 pragmatic 23, 30, 46, 188, 203, 235, 243 predication 25, 27, 130, 143, 150, 165, 186, 189 predication adjunct 26, 129-130 predicative complement see complement prefix 9, 218 prepositional phrases 4, 48, 52, 73, 196, 215,218,221,231,256 prepositions, prepositional 21-22, 30, 3839, 44, 47, 48, 59, 64, 70, 76, 135, 141, 177, 192-193, 205, 214, 216-221, 227229, 231-232, 235, 238, 244-245, 256 prescriptive, prescriptiveness, prescriptivism 3, 14-15, 44-45, 54, 134, 190, 202-204,241 present perfect 82; see also tenses present tense 60, 76, 132; see also tenses and historic present probability 98 productivity, productive 207, 214-215, 227, 232

progressive form 58 promise 179 pronouns: personal 36, 68, 145, 259; relative 47, 214, 231-233, 235, 261; possessive 260-261; demonstrative 214; indefinite 53; generic 100-101, 260; interrogative 38, 40, 233, 260 pronunciation see phonology proportion 222-223, 227, 239; see also frequency prosody 27, 45, 241; see also intonation protasis 114, 123,205,239,241 prototype, prototypical 14, 41-42, 44, 233, 243

providet/provided99, 103,225 providing 225 providing that 123, 225 proximity 68, 79, 89, 92 punctuation 15, 25-26, 28-29, 45, 151, 159, 162, 166-167, 188-190, 256; see also orthography quam (Lat.) 206 quand(Fr.)2\9 questionnaire 17

303

quia (Lat.) 206

Rae, William, minister 202 rafter f/wn 135- 137, 216 Real Scots 8 real situation see factuality reanalysis see grammatical isation reduction: phonological 37, 48, 206; morphological 37, 48, 150 Reformation 6 regional differences see geographical restriction regional restriction 62-63, 77, 100, 189, 201-202,231-232,235-236 regional variation 236-237; see also regional restriction register see style relative clause 39-40, 48, 182, 23 1 relati viser 38 Renfrewshire 154, 201, 247, 250 repetitive, repetition 70, 73; see also iteracy and habitually recurrent situation request 121 resultative meaning component, result 60, 64,80,89,166 Richardson, Samuel, author 230 Robertson, Stanley, traveller 259 Romance languages 28, 220, 227, 23 1 Roxburghshire 247, 249 rural Scots 8-9, 18 S-adverbials 24 ία (Dan.) 217 sae (adverb) 159 safremt (Dan.) 217 salcenge som (Dan.) 217, 228 jarts(Fr.)219 save that 128, 207 say 108, 203-205, 215, 262, 264 Scandinavian 4, 216, 228, 232; see also Danish, Norn, Norwegian, Old Norse School of Scottish Studies 9, 19, 247, 249-250 Scotch 6-7, 9 Scots Thesaurus 9 Scott, Sir Walter, novelist 258 Scotticisms; overt 4, 10, 102, 241, 255; covert: 4

304

General Index

Scottis 6 Scottish Gaelic see Gaelic Scottish English 6-7; see also Scottish Standard English Scottish National Dictionary (SND) 5, 9, 55, 61-63, 70, 83-84, 88, 109, 155, 157, 201-202, 236-237, 258, 283-287 Scottish Standard English 4, 7, 18-19, 244 second order adverbials 24 sedert (Du.) 218 see 121 seeing (at/as/as hoo) 150, 155-156,228 $i/(Ger.)39,218,222 seitdem (Ger.) 39 seither (Ger.) 39 self-monitored speech 14, 16; see also discourse Selkirk 117, 247, 250 Selkirkshire 83, 202, 247, 250 semantic extension 46, 157, 229-230; see also polysemy semantic range see semantic scope semantic scope 97, 189, 192, 204, 216, 240, 244 semantic shift 201, 204, 230 semantic—pragmatic gradience 26, 256 semicolon see punctuation sentence adjunct see clause adjunct sentence boundary 14, 151, 167, 186 sentence margin 23 sentence nucleus 23 sentence: orthographical sentence 13, 15, 20; text sentence 15, 42, 255; system sentence 15, 255; irregular sentence 35, 45; sentence as grammatical unit 13, 15 sequence of events see order of events sequence of tenses see tenses, sequence of Shakespeare, William, dramatist 76, 85, 226, 262 Shetlandic, Shetland, Shetlander 5, 32, 101, 128, 140, 201, 232-233, 236, 247248, 250, 260 ίΛο«/ίΠ12,238 j/ifert(Dan.)217, 222 since/sin 39, 202, 222, 227, 241; time 79, 82-84, 96-97; reason 150, 155-156 sind(Du.)2\S Slavic languages 231 so as; purpose 169, 171, 178. 182; with infinitive 173, 176; result 184-186

so long as 123, 215, 228 so that 41, 243; purpose 169-171, 179, 182, 184; result 184-185, 189, 243, 257 so: purpose 172; result 184, 190, 203-204; conjunct 45,186-188 social class so/er«(Ger.)218 solange (Ger.) 228 Solway Firth 5 som(Oan.)2\7 ioq/j(Ger.)218 source category 37, 47, 218-219, 227, 229, 232, 235 South-Eastem British English 259; see also Southern British English South-Eastern Scots, south-eastern Scotland 35, 283 South-Westem British English 259; see also Southern British English Southern British English 205, 233, 238 Southern German 232, 234-235, 263 Southern Scots, southern Scotland 63-64, 100, 236 sow