N e w prospects f o r t h e study o f English dialect syntax: impetus f o r m syntactic theory a n d l a n g u a g e typology
Bernd Kortmann Freiburg
0 § INTRODUCTION
At the turn of the 21st century there are new prospects for the study of dialect syntax. These are primarily due to developments outside dialectology, more exactly in linguistic theorizing. What is responsible for the currently observable rise of dialect syntactic research in several European countries is, in the first place, a broadening of the perspective in recent generative theory and language typology. No longer is it cross-linguistic variation only that matters. Variation within individual languages, too, is increasingly attributed important theoretical significance. One of the consequences of this is that a strong need is felt to improve the empirical basis for reliable descriptive generalizations and for drawing conclusions for linguistic theory. In other words, due to the rising interest in variation across dialects within generative linguistics and language typology, we are witnessing a period in which a much improved data situation will allow us to make substantial advances in exploring dialect grammar and integrating the findings into a larger theoretical frame. The two major aims of this paper are the following. First of all, the theory-internal developments will be sketched which are responsible for this relatively recent interest in dialect grammar within the generative framework, on the one hand, and the framework of language (more exactly: functional or Greenbergian) typology, on the other hand (Part 2.2). In a second step, the significance of various phenomena of English dialect syntax will be demonstrated when looked at from a generative and functional-typological perspective (Part 4). Before looking at the new prospects, though, the role of dialect syntax in past and, especially, present Anglo-American dialectology will briefly be evaluated.
1 § DIALECT SYNTAX IN PAST AND PRESENT DIALECTOLOGY
The study of dialect grammar, especially of the syntax of regional varieties, neither plays nor ever has played a major role in English dialectology. Numerous statements to this effect can be found in the recent literature on English dialects. Just compare Toon on dialect research for the Old English period, Fischer and Lass for the Middle English period, G??rlach for the Modern English period and, for example, Wakelin or Trudgill and Chambers on dialect research in the 20th century: ?? OE: ?? ME:
"Syntactic variation between dialects has scarcely been studied and in any event the material is relatively meagre." (Toon 1992: 451) "The Linguistic Atlas of Late Medieval English (1986) provides an extensive survey of dialectal differences in the fields of phonology, morphology and lexis, but it has nothing on syntactic variants. In the introduction it is stated that 'it may well be that syntax will perforce remain the Cinderella of Middle English dialectology' (McIntosh et al. 1986: 32)." (Fischer 1992: 208)
?? EModE: ?? PrDE:
"… evidence of Early Modern English dialect syntax is almost nil." (G??rlach 1999: 492) "… modern dialectology has until recently largely neglected the field of syntax…" (G??rlach 1999: 493) "Syntax is an unwieldy subject which dialectologists have fought shy of." (Wakelin 1977: 125)
A major part of the problem in the second half of this century has been the absence of reliable data and, to start with, a sufficient amount of data. Take, for instance, the Survey of English Dialects (SED), which was compiled in the course of the 1950s and has served as the most important data source for English dialectologists and dialect geographers in the last four decades. This major survey was simply not geared to the systematic collection of syntactic data, despite the good and serious intentions Harold Orton and Eugen Dieth, whose brainchild the SED was, may originally have had (cf. Kirk 1985: 130). Only a fraction of the 1322 questions in the SED questionnaire was explicitly
SYNTACTIC MICROVARIATION §
186
designed to collect morphological and syntactic information (cf. Klemola 1996: 39). Thus, for example, Upton et al. (1994) compiled their SED-grammar on the basis of breadcrumbs, as it were, making use in particular of the (relatively scarce) grammatical information to be obtained from the answers collected for the bulk of questions geared towards phonology and lexis. In addition, they drew upon parts of the so-called incidental material of the SED, i.e. utterances that the fieldworkers picked up during their interviews or from more informal conversations with the informants and took down in their notebooks. The latter type of material “proved to be specially rich” (Upton et al. 1994: 479) for finding out more about syntactic variation across English dialects as displayed by traditional dialect speakers (the well-known NORMS).1 Even today, when we look at current Anglo-American dialect research, there is no denying that the study of dialect syntax still constitutes no more than a sideline, despite the statement by Trudgill/Chambers (1991: 1) that there has been a noticeable "upsurge" in attention to dialect syntax since the 1980s, especially in the British Isles.2 For instance, none of the papers in the proceedings volume edited by Upton and Wales (1999) addresses a problem of dialect syntax, although this volume is meant to provide a survey of "current research projects in urban as well as rural dialect studies" (1999: vii). In the last 20 years morphology and syntax have regularly played a role only in connection with the substrate question of the so-called Celtic Englishes (like "Which features of Irish English go back to Irish Gaelic?") and in studies concerned with the genesis of African American Vernacular English (just witness Poplack's volume on The English History of African American English (2000)). What still dominates dialectological research are phonological studies, these days especially in connection with urban varieties (for a recent publication cf. Foulkes/Docherty (1999) Urban Voices or Labov's studies (e.g. in Labov 1994
1
But, to be fair, we should not forget one of the major aims of the SED. Along with documenting regional differences in the middle of the 20th century, "Orton and Dieth were explicitly interested in modern dialects as evidence for [the] study of historical varieties of English,..." (Kretzschmar 1999: 274), thus continuing the 19th century tradition of English (and German) dialectology (just cf. A.J. Ellis' survey of the phonology of English dialects (1889)). And since syntax played no or no more than a minor role in historical dialectology, the problem that the SED-material poses for dialect syntacticians was to be expected. 2 The full statement runs as follows: "In recent years there has been a move, as far as interest in dialects of English is concerned, from a concern with purely phonological, lexical and historical issues towards a deeper interest in the grammar of different English varieties. This is especially true of syntax, which has traditionally received much less attention from dialectologists than morphology. This upsurge has been particularly noticeable, in the 1980s, in the British Isles." (as opposed to North America; Trudgill/Chambers 1991: 1)
SYNTACTIC MICROVARIATION §
187
and to appear) on the Northern City Shift in the United States). New research trends are linked to this: studies on language (or in this case: dialect) and identity, and studies within the framework of perceptual dialectology, i.e. on how people perceive speech features and dialects (Preston 1999; cf. also Kretzschmar 1999: 279). This is not to deny, of course, that there exist quite a number of publications on English dialect syntax. For the most part, though, they concentrate on just one particular phenomenon in one particular dialect or dialect area, are based on a very small database and are purely descriptive. Furthermore, the small size of the database often makes it very difficult to arrive at valid descriptive generalizations. Virtually non-existent in English dialectology are systematic comparative studies, whether theory-driven or not, of individual grammatical subsystems across a selection of dialects (like comparative studies of the tense and aspect systems, pronominal systems, relativization or complementation patterns). This is the rather sobering situation of dialect syntax in past and current dialectology. There are, however, new prospects for the study of the morphology and especially syntax of regional varieties of English (and, I should add, of other languages). These prospects are primarily due to the two developments mentioned in the introduction: a growing interest in syntactic variation both from a generativist and typological point of view and, largely as a consequence of this, a much improved data situation.
2 § WHY NEW PROSPECTS NOW?
2.1 § The data situation The data situation has improved and continues to improve in three ways. First, of the SED data, there is more material than the so-called “Basic Material” (elicited by the questionnaire) and the so-called “Incidental Material” (in the field workers’ notebooks) which will soon be ready for linguistic analysis. These are the 8-10 minute sound recordings from all SED localities, which will shortly be published on CD-ROM (Klemola 2001). Here we have stretches of fairly natural informal discourse (however brief) which, together with the incidental
SYNTACTIC MICROVARIATION §
188
material, may at least augment grammatical analyses of the individual dialects. In both types of material we run a much greater chance of encountering certain constructions that are far less frequently found in the rather formal questionnaire-oriented interview situation (e.g. unstressed periphrastic d o in affirmative sentences, which occurs very frequently in the fieldworkers' notebooks for the English Southwest). Second, more and more efforts are made at collecting interviews with dialect speakers which provide us with long stretches of conversational, more or less informal discourse. As is well-known, a much larger amount of data is necessary for the study of syntax than for the study of phonology and lexis. One important source that is currently being exploited are interviews conducted in oral history projects across the British Isles in the course of the 1970s and 1980s. These discourse data are being made available in the form of computerized corpora, which shortly will, and to some extent already do, lend themselves to all the corpus-analytic methods and studies familiar from research based on, for example, the British National Corpus (which itself provides some dialect material). A computerized corpus of English dialects has been built up over the last three years in the English department at the University of Freiburg (FRED). The result already constitutes the largest corpus of English dialects that has ever been computerized (currently some 1.5 million words). So the Freiburg research team will soon be able to formulate much more reliable descriptive generalizations than in the past and to perform, at least for high-frequency phenomena, quantitative analyses, which according to Kretzschmar (1999: 282) "will be the hallmark of future analysis in dialectology". Third, the need is felt for collecting new data, data which represent the status of regional and urban varieties of English around the year 2000 and, on top of that, do not neglect issues of dialect grammar as they were neglected in the SED. There are currently two major models for this kind of data collection: One large project is being planned in England (the Survey of Regional English (SuRE); cf. Kerswil/Llamas/Upton 1999) whose aim it simply is to record as much natural discourse material as possible, without following a strict questionnaire method. The second type of project intends to collect data specifically for the purpose of syntactic analyses, using elicitation questionnaires. This method is currently being explored in Switzerland for
SYNTACTIC MICROVARIATION §
189
Swiss dialect syntax (cf. the contributions by Glaser, Fleischer, and Seiler, this volume) and in The Netherlands and Belgium for the SAND project on Dutch and Flemish dialect syntax. This method is also anticipated in the second phase of the Freiburg research project on the syntax of English dialects.3
2.2 § DIALECT SYNTAX IN CURRENT LINGUISTIC THEORISING
Even more important with regard to the present state and especially the future of dialect syntactic research is the growing interest in dialect syntax outside dialectology, more exactly in generative syntactic theory and, as yet much less pronounced, in functional approaches to grammar. This is an extremely astonishing development in view of the often stated lack of interest or even averseness of traditional dialectological research to linguistic theorising (cf. Kretzschmar 1999: 274, Chambers/Trudgill 1991) and, vice versa, the lack of interest of syntactic theory in issues of variation, at least up to the 1980s. But the changed role of variation in more recent generative theory is indeed the key to understanding the new development: "... variation [has indeed become, B.K.] highly relevant for the theoretical grammarian" (Haspelmath 1999: 191). 2.2.1 § Syntactic theory: Principles & Parameters and Optimality Theory In generative linguistics, variation seriously started to matter with the advent of the Principles & Parameters approach in the 1980s, i.e. the idea that Universal Grammar (UG) is an invariant system of highly abstract principles some of which, within a given language, permit at most a specified degree of variation. The (core) grammar of any particular language is considered to consist of these universal principles and the language-specific settings for a small number of parameters or, as Chomsky puts it (1995: 129), "… the rules of L are the principles of UG as parameterized for L." The concept of parametric variation thus accounts for variation observable across languages. Important for this paper is the crucial step that was taken in generative studies from the study of parametric (more exactly, macroparametric) variation
3
Note, incidentally, that the data situation for Early (and Late) Modern English may also soon be brightening up given, for example, the Corpus of Early English Correspondence that is currently being compiled at the English Department of the University of Helsinki (Terttu Nevalainen, Irma Taavitsainen, Matti Rissanen).
SYNTACTIC MICROVARIATION §
190
to the study of microparametric, i.e. language-internal, variation. Relevant research is strongest for Italian, Dutch and Flemish dialects; less frequent are studies on microparametric syntax in German dialects (with the exception of Bavarian and individual dialects of Swiss German), and least frequent for English dialects.4 With regard to the further development of generative theory, the study of microparametric syntax is expected to yield more insights into, for example, the form and range of syntactic parameters as well as into the effects which variation along a single parameter may have. Compare the following quotations: on the form of syntactic parameters: "It seems reasonable to expect work in microparametric syntax to play a privileged role in the future in answering the more general question concerning the form that syntactic parameters may take. Chomsky's recent work, for example, suggests that all syntactic variation might be expressible in terms of strong/weak features on various functional heads; microparametric work will enable to test this kind of hypothesis in a particularly interesting way." (Kayne 1996: xiv) on the range of syntactic parameters: "Note that there has in effect been a shift in linguistic theory recently, though one that has been largely unacknowledged,... [...] ..., we have moved from a conception of parameters as very wide-ranging to one where,..., they cover a much smaller range of phenomena. What we have found out about dialect variation suggests that this is the right approach. [...] However we formalise these, we have to admit that something like construction-specific differences between dialects, and optionality of movement, must be able to be part of the variation between grammars permitted by UG, something which we might have missed if study was confined to the larger-scale differences typically found between those varieties generally characterised as different languages." (Henry 1996: 91f.) on variation along a single parameter: "In a linguistic group of interrelated dialects with little differentiation, we can expect to find realized only those
4
Cf. for example, Abraham/Bayer (1993), Benincá (1989), Black/Motapanyane (1996), Chambers/Trudgill (1991), Penner (1995).
SYNTACTIC MICROVARIATION §
191
possibilities which are admitted by the theory. It is evident, then, that the more the dialects are similar to one another, the more possible it becomes to find, for a specific grammatical area, the ideal case of some dialects differing only in respect to phenomena that can be traced back unambiguously to a single parameter." (Benincà 1989: 3) Note that in generative linguistics of the 1990s, the study of cross- and intralinguistic variation continues to matter not only within the Principles & Parameters approach, but also in one of the latest developments of generative linguistics, namely Optimality Theory (for an overview cf. Archangeli/Langendoen 1997). Optimality Theory views UG as a set of constraints that can be violated. It further holds that each language makes use of the same set of constraints, but has its own rankings for these constraints, so that the differences between constraint rankings give rise to systematic variation between and within (!) languages (in synchrony and diachrony; cf. also Dahl 1999: 209).5 Optimality within this theory means that a particular constraint ranking chosen in any given language will satisfy the relevant constraints best. This can be achieved, for example, by tolerating violations of lower ranked constraints in order to help avoid violating some higher ranked constraint (cf. Archangeli 1997: 15f.). 2.2.2 § Functional typology Like the generative approach, functional typology pursues the aims of identifying language universals and, more modestly, determining the patterns and limits of cross-linguistic variation.6 Like the generative approach, its major focus is on morphology and syntax. Like the generative approach, it also has an
5
Of course, historical variation is also accounted for in this way. As Traugott (1999: 238) puts it: "It is a given of OT that differences among dialects and stages of a language can be accounted for in terms of different rankings [of constraints, B.K.]." Compare also Kayne's statement on the significance of the study of microparametric syntax for the study of diachronic syntax: "It is also clear that the study of minimal syntactic variation is bound to provide crucial evidence bearing on questions of diachronic syntax (which involves the study of the minimally different stages in the evolution of the syntax of a language)." (1996: xiv) 6 Just compare the striking similarities between this characterization and the following definition of linguistics from a recent introduction to Optimality Theory. Anyone who does not know the source would rather expect this to be a definition entertained in language typology. "There are two central research objectives in linguistics. The first is to determine and characterize universal properties of language,... ... the second research objective in linguistics [is] to determine and characterize the range of possible language variation." (Archangeli 1997: 2)
SYNTACTIC MICROVARIATION §
192
important diachronic dimension (cf. Kortmann 1997, Croft 1990: 203-245 or, most recently, Croft 2000). Among the major differences between these two approaches are the following: functional typology has a much stronger empirical orientation and, above all, advocates the hypothesis that form, i.e. linguistic structure, should primarily be explained in terms of (linguistic, communicative) function (cf. Croft 1990: 2). A rather minor difference, which is important though in the context of this paper, is the following: much earlier than functional typology, namely in the 1980s, did generative theory acknowledge the importance of studying intralinguistic variation (i.e. microparametric syntax) besides cross-linguistic variation. In functional typology we can only observe first steps in that direction. In fact, as yet these steps have primarily been taken by the present author’s own research group, which aims to study English dialect syntax from a typological perspective (cf. Kortmann 1999).7 Among other things, this means that the cross-dialectal variation observable in individual domains of grammar (e.g. negation, relative clauses, pronominal systems) is judged against the crosslinguistic variation described in typological studies: How much of the range of variation across languages is found in varieties of a single language? Adopting a typological approach also means that the Freiburg research group will soon be using questionnaires for the collection of dialect syntactic data which will follow the format of such questionnaires in recent typological research, also taking into consideration, however, the experiences Glaser’s research team on Swiss dialect syntax has made with their questionnaire method (cf. Glaser 2000, Glaser/Bucheli this volume). Both dialect syntacticians and typologists are bound to profit from this kind of approach. On the one hand, dialectologists can draw upon a large body of typological insights in and hypotheses on language variation, which simply helps them to look at dialect data from a different perspective; no longer is it just the historical perspective or the comparison with the standard variety that matters in evaluating the observable cross-dialectal variation. On the other hand, typologists will get a broader and, perhaps, more adequate picture of what a given language is like when no longer ignoring dialectal variation. To start
SYNTACTIC MICROVARIATION §
193
with, they will encounter variation at their doorstep, as it were, variation that otherwise can only be observed, if at all, in languages in other parts of the world. Moreover, it is a much ignored problem that, for languages that do have standards, typology is typically concerned with the standard varieties only and, which makes it even worse, with the written standards.8 Especially the grammars of languages with a long literary tradition (like many European languages) may have been subject to non-natural language change in the course of their history, be it due to prescriptivist grammarians or kind of “levelling processes” that affected particularly the written language. This is an important factor to be reckoned with in the shaping of the Standard Average European language complex. This concludes the scene-setting first part of this paper. In the second part (sections 3 and 4) the focus will be on the presentation and discussion of a selection of phenomena in English dialect syntax, especially when looked at from a generative and functional-typological point of view.
7
Consult the following homepage for information on the current research activities within the project and the current state of the computerized Freiburg Corpus of English Dialects (FRED): http://www.unifreiburg.de/philfak3/eng/. 8 The reason for this is of course not one of principle, but one of practicality, or simply convenience. The major source of information for the relevant languages typically is a (set of) reference grammar(s), and these are based on the standard varieties.
SYNTACTIC MICROVARIATION §
194
3 § DIALECT SYNTAX IN THE SOUTHWEST OF ENGLAND
Rather than culling data from a wide range of English dialects, it will be syntactic phenomena from one particular dialect area that will be discussed, this area being the Southwest of England. According to Wakelin (1986: 3) this area is mainly constituted by the counties of Cornwall, Devon, Somerset, South Avon, Wiltshire and Dorset, with East Cornwall, Devon and (West) Somerset forming its core. Anyone who knows a little bit about English dialects will not be surprised by this choice. Since medieval times the rural dialects of the geographically rather secluded Southwest of England have constituted one of the most distinctive and best known dialect areas (best known not only by specialists but, to the present day, by their fellow countrymen and women; cf. e.g. Ihalainen 1998 and Inoue 1999). In this respect the Southwest is only rivalled by the dialects of East Anglia and the North (especially Northeast) of England. Noteworthy from a phonological point of view, for example, is the rhoticity of the southwestern dialects and the voicing of the voiceless fricatives [f,s,∫], which are pronounced [v,z,Z], as in [v??ks] fox, [zi:] see, or [Z∪r] sure. There are three major reasons for choosing the English Southwest for the purposes of illustration in this paper: (a) As in their phonology, the relevant dialects exhibit a relatively large number of grammatical properties which make them distinct as a dialect group among the regional varieties of the British Isles. Several of these properties are almost impossible to observe outside this area and are extremely interesting to look at from a theoretical perspective. In addition, of that, we can of course observe features of grammar here that are found in other dialect areas, too, so that for these features the SW is as good a point of departure as any other dialect area. (b) It is the Southwest for which we can find a considerable number of recent publications offering fairly detailed syntactic analyses based upon carefully collected and relatively recent material (notably those by Ossi Ihalainen and Juhani Klemola in Helsinki). (c) It is the Southwest (mostly Somerset) for which the Freiburg research group has compiled one of the largest subcorpora of the Freiburg English Dialect Corpus (currently about 160,000 words including many unpublished interviews
SYNTACTIC MICROVARIATION §
195
of dialect speakers born around 1900).9 So the data situation is very good, relatively speaking, and allows us to put to test many previously made claims which were based on far fewer data. In fact, at various points of this paper it will be seen that claims made in the relevant literature (even in relatively recent publications) cannot be upheld in light of the FRED data. To give just one example: In a study of relative clauses in the Southwest, Ihalainen (1980: 189) claims that the relativizer that is only used for relativizing subjects and prepositional objects, but not for relativizing NPs serving as the direct object in the relative clause (for instance I know the man that you just greeted is claimed to be out). This would be extremely astonishing since it contradicts the predictions of the well-known Noun Phrase Accessibility Hierarchy by Keenan and Comrie (cf. also section 4.2.2). But this problem is really simply a problem of too few data that Ihalainen had to work with. In the FRED data from the Southwest there is ample evidence for that relativizing NPs in the direct object function, too. The grammatical features to be observed in the dialects of the Southwest fall into three main groups: exclusively southwestern features (Group 1), features also found in other dialect areas (Group 2), and features that are no longer regionally bound and rather constitute part of a general spoken nonstandard (Group 3). The lists for the second and third group are non-exhaustive; also the boundaries between these two groups are anything but hard and fast.10 The five features in bold print (1a, 2a, 3e, 3f, and 4a) will be returned to in section 4, where they will be discussed with regard to their significance in syntactic theory and functional typology.
§ G r o u p 1 exclusively southwestern features: these are essentially to be found
in two domains of grammar, namely verbs, as in (1), and pronouns, as in (2) (cf. also Wakelin 1986: 33 and Ihalainen 1991: 105):
9
The author would like to thank Juhani Klemola (Helsinki) for the permission to make use of his data from the English Southwest, which formed the central basis for Klemola (1996). These data approximately account for 20% of the relevant subcorpus of FRED.
10
Compare, for example, Cheshire/Edwards/Whittle (1993) for a set of grammar features characteristic of a large number of British urban dialects. Thus the Group-2 features in (3a,b) and (3e,f) are candidates for Group-
SYNTACTIC MICROVARIATION §
196
(1)
(a)
unstressed d o as simple tense-carrier i n affirmative sentences We do breed our own cows. This man what do own this,... We've been up milking at 6 o'clock in the morning, and then we did go on haymaking,...
(b) (c)
be-paradigm: e.g. invariant be in the present tense, clitics: I'm/you'm/ we'm/ they'm, negation: bain't special morphological marking (-y) of intransitive verbs or intransitively used transitive verbs (extremely rare): Then you had to stand in and milky,... (Somerset) Can you zew up thease zeam? vs. Can you zewy? (Dorset)
(2)
(a)
g e n d e r d i f f u s i o n : gender system primarily sensitive to mass/count distinction (it for mass nouns, h e / s h e for count nouns): Pass the bread - it's over there. vs. Pass the loaf - he's over there. Sewing machine? You'd better take he back where you had him from.
(b)
pronoun exchange: the use of subject pronouns in object position and, occasionally, the converse:11 O-Pro in S-position: Her don't like it! Us be a-goin. S-Pro in O-position: I used to go up to my uncle's and help he. Work didn't frighten w e , we knew we had to do it,... He were along with we.
3 membership: regularized reflexives (3a), unmarked plurals after numerals (3b), were/was-generalization (3e), and what as a general relativizer (3f). 11 Note that pronoun exchange is an extremely rare phenomenon. It accounts for less than 1% of all cases in the FRED subcorpus for the Southwest analyzed so far (some 160,000 words, with a strong Somerset bias). Moreover, there are hardly any instances of object pronouns in subject position. For most speakers in our corpus no instances were found, at all; and those speakers who do use object pronouns in subject position do this primarily in questions and, especially, question tags (cf. similarly Ihalainen 1991: 106), i.e. employ the object form as a kind of weak, unstressed form of the subject pronoun. It seems more adequate to speak no longer of pronoun exchange, but rather of a pronounced tendency for 'subject' pronouns to do service for all grammatical functions: subject, direct object, and prepositional object. Thus in the Somerset dialect(s) we observe a further reduction of case-marking in English.
SYNTACTIC MICROVARIATION §
197
(c) (d) (e)
they instead of t h e m and those: to/with/under they, I used to go down there couple hours and mind they cows, in they days special pronominal forms: e.g. thou, thee, unstressed u n / e n for him, unstressed ee for h e (in South Devon also for y e / t h e e ) special demonstrative pronouns (extremely rare): e.g. thick/thicky (here), thuck/thucker (there)
§ G r o u p 2 features also found in other dialect areas:
(3)
(a) (b) (c)
regularized reflexives-paradigm (possessive pronoun + self/selves) unmarked plurals after numerals (e.g. three year) irregular use (omission or insertion) of article: omission: insertion: I had nice garden..., they had awful job...; Father rented the farm under Squire, take them to market about a three fields or about a seven inches square on a board; I had the toothache
(d) (e)
regularized Past and Past Participle forms of irregular verbs regularized be-paradigms: e.g. generalization of were or was ( S o u t h w e s t : w e r e- g e n e r a l i z a t i o n i n p o s i t i v e contexts; i n negative contexts generalization o f w e r e n ' t o r ( ! ) w a s n ' t)
(f)
s p e c i a l r e l a t i v i z a t i o n s t r a t e g i e s : e.g. - gapping in subject position: There was one or two people___ made their living by this. - shadow (or: resumptive) pronouns: I jumped in and bought this, which I were lucky in a way to get it . - Southwest: what ... they'd put a couple in the old anchor boat what we weren't using it
SYNTACTIC MICROVARIATION §
198
§ G r o u p 3 features no longer regionally bound, constituting part of a general
spoken non-standard English: (4) (a) (b) (c) d o u b l e / m u l t i p l e n e g a t i o n (I've never been to market to buy no heifers) t h e m instead of those (e.g. one of them things what run around) m e instead of m y / I
4 § CASE STUDIES
In the present section five of the grammatical features in Groups 1 to 3 will briefly be discussed in order to demonstrate what can be interesting about dialect data when looked at from a generative or typological point of view. 4.1 § Unstressed periphrastic do from a generative perspective The use of unstressed periphrastic d o as a simple tense-carrier in affirmative sentences, as in the examples in (1a), seems to be a truly southwestern innovation documented already in the 13th century (Fischer 1992: 272, Tristram 1997: 413, Garrett 1998: 284). Among the dialects of English today, those of the Southwest have gone furthest in grammaticalising d o as a tense marker.12 From a generative perspective the existence of this construction causes a significant problem for an argument advanced by Chomsky in a 1995 article (first published in 1989) on the “Economy of Derivation and Representation”.13 Chomsky constructs an argument why English needs the rule of d o -support in interrogatives and why at the same time the use of unstressed d o as a tense carrier in declaratives is ungrammatical or "illegitimate", as he calls it. English needs the clearly language-specific rule of d o -insertion in interrogatives lacking auxiliaries because otherwise the "legitimate D-Structure representation" in (5) would not permit a legitimate, i.e. grammatical, output.
12
13
Cf. Klemola (1996) for a detailed account of do-periphrasis as in (1a) for the English Southwest. Note that even in the fast-changing world of of syntactic theory Chomsky's account is still relevant today (cf., for example, Radford 1997 or Collins 2001). An alternative argument for why English needs do-support has been suggested and has "gained some support" within an economy-based framework in the meantime (cf. Collins 2001: 46 on Bobalijk 1995 and Lasnik 1995), but from a dialectological point of view it suffers from the same fundamental problems as Chomsky's account does. This alternative argument is based on adjacency and assumes that the inflectional head I(NFL) contains an affix. Under this approach the use of unstressed do in John did write books is blocked because do-support is legitimate only "when the derivation would otherwise lead to a stranded affix" (Collins 2001: 46) and thus to an ungrammatical output.
SYNTACTIC MICROVARIATION §
199
(5)
Q John I Agr write books
In Chomsky’s words, "To permit an output from the legitimate D-Structure representation ... [in (5), B.K.], English makes use of the dummy element d o to bear the affix,..." Only this rule of d o -support yields the grammatical English question Did John write books?. The interesting point Chomsky (1995: 139 f.) now makes is this: "The same device, however, permits the illegitimate form John did write books (d o unstressed) alongside John wrote books, both deriving from the declarative form corresponding to (5) (lacking Q)" [i.e. from the DStructure representation in (6), B.K.]. (6) John I Agr write books
Chomsky continues (1995: 140): "In fact, this option is not only available but i n f a c t a r g u a b l y o b l i g a t o r y [my emphasis, B.K.] if shorter derivations are always preferred. The reason is that the illegitimate form requires only the rule of d o -insertion and raising, whereas the correct form requires overt lowering and subsequent LF raising." So Chomsky has three problems: Problem 1: The do-insertion rule that is necessary for English interrogatives is not allowed in the case of declaratives although, within the economy-based approach that Chomsky himself is advocating, it represents the more elegant, because more economical rule ("... shorter derivations are always chosen over longer ones", 1995: 139). Problem 2 is concerned with the competition between principles of Universal Grammar (UG) and language-particular rules. In the case of interrogatives, English needs a language-specific rule "... to 'save' a D-Structure representation yielding no [grammatical] output,..." (1995: 140). In other words, here a language-specific rule, namely d o -support, is necessary, a “last resort” as it were (Radford 1997: 219), to save a representation that strictly abides by UG principles. But why does not the same happen for d o -support in declaratives? Because, according to Chomsky (ibid.) “... UG principles are applied wherever possible, with language-particular rules used only to 'save' a D-Structure representation yielding no output:... UG principles are thus 'less costly' than
SYNTACTIC MICROVARIATION §
200
language-specific principles.” In simple words: You could not ask a grammatical question in English if it was not for the d o -support rule. But since there exists a grammatical declarative sentence John wrote books, which is formed according to UG principles, d o -support does not apply in the case of declaratives. It would be unnecessarily costly. Problem 3: Chomsky evidently does not know about the dialects of the English Southwest and their use of unstressed d o as a simple tense carrier in declarative sentences. What then does the existence of the "illegitimate" construction John did write books along (!) with the legitimate construction John wrote books in these dialects mean for Chomsky's line of reasoning? On the one hand he can be happy, since this solves Problem 1. Within his economy-driven approach, it makes perfect sense for a variety of English to have the insertion of unstressed d o in declaratives. On the other hand, there is really bad news for the solution that Chomsky suggests for Problem 2. In the Southwestern dialects a language-/dialect-particular rule seems to permit overruling principles of UG even if they yield legitimate output; these dialects allow both the 'less costly' and the 'more costly' solution, as it were. Languages or dialects, it seems, may happily let universal principles and languageparticular rules compete, allowing also for rivalling outputs. Haegeman/Guéron’s (1999: 633) distinction between ‘core grammars’ of a language (governing the unmarked variety) and ‘peripheral grammars’ of a language (governing special registers and styles) also offers no solution to this problem. Surely, Chomsky’s model of the division of tasks between UG principles and language-particular rules should apply to peripheral grammars, too. In the face of this situation, Chambers/Trudgill (1991: 295) have a point when concluding that, with regard to this division of tasks, “..., the generality of Chomsky's claim simply dissolves, rendering it vacuous.” But maybe this is the kind of problem that Optimality Theory can help to solve: the dialects of the English Southwest, it could be argued, exhibit a different constraint ranking from Standard British English.14
14
It could also be argued that, given the existence of unstressed periphrastic do in the Southwest, in the relevant dialects INFL is a stronger head than in other varieties of English, notably Standard English, since this position is often (though not obligatorily) filled even in unemphatic declarative sentences (cf. Radford 1997: 218f. and 226ff. on Chomsky’s strength metaphor).
SYNTACTIC MICROVARIATION §
201
4.2 § Three domains of grammar from a typological perspective In this section three domains of dialect syntax will be commented on from the point of view of functional typology. The three domains are gender diffusion in the Southwest, relativization and negation (cf. Kortmann 1999 for a fuller account).15 4.2.1 § Gender diffusion Gender diffusion as exemplified in (2a) essentially involves a semantic system of gender marking. Unlike the system in Standard English, though, this gender system is not primarily sensitive to the animate/inanimate-distinction, but rather to the distinction between mass and count nouns: it is used only for mass nouns, count nouns take h e unless they refer to female humans, in which case s h e is used (cf. (7 and 8). (7) it: he: she: (8) a. b. only for mass nouns only for count nouns (male humans or nonhumans) only for count nouns (female humans) Pass the bread - it's over there. Pass the loaf - he's over there.
Among varieties of English, such or similar gender systems have only been observed in the English Southwest, from where it was exported, as it were, to Newfoundland and, possibly, to Tasmania (cf. Pawley 1994). Interesting from a typological perspective is, first of all, that a semantic gender system which is primarily conditioned by the mass/count distinction is extremely unusual and has not been described for a European language so far. So here we have a case of variation at our doorstep which typologists otherwise have to go a long way for. Secondly, Siemund (2001) has looked at a range of languages with semantic gender systems and comes to the conclusion "... that the inanimate nouns treated as animate in this way come from relatively homogeneous semantic domains and are surprisingly similar in languages that are completely
15
These domains are only three of many that could be discussed from a typological perspective. Periphrastic do, for example, has recently also been investigated from exactly this angle (cf. van der Auwera 1999).
SYNTACTIC MICROVARIATION §
202
unrelated in genetic terms. Among the domains recurring in language after language are higher and lower animals, heavenly bodies, plants, body parts, instruments... and household utilities". Moreover, he formulates the hypothesis that a high degree of individuation may be the crucial clue for understanding assignment of animate gender to inanimate nouns, "i.e. the use of he and s h e with inanimates occurs predominantly with those inanimate nouns that denote highly individualised entities. The animation of mass nouns, by contrast, is an extremely rare phenomenon". It remains to be seen to what extent this hypothesis will stand the test of time and further data (e.g. the 'gender' behaviour of abstract nouns), but Siemund's study itself is a very nice (and, to the present author’s knowledge, the first) example of a typological study that was triggered by studies on dialect syntax, in this case on gender diffusion in the English Southwest. 4.2.2 § Relevatization There is one particular relativization strategy that is very common in English dialects (including the Southwest) but not found, at all, in Standard English: zero-relativization (or: gapping) of the subject position, as in (3f). Noteworthy from a typological perspective is that in having this relativization strategy, English dialects conform to basic constraints on and predictions of Keenan and Comrie's Accessibility Hierarchy (1977, 1985) where Standard English does not. (9) Accessibility Hierarchy (AH) in the original version: subject > direct object > indirect object > oblique > genitive > object of comparison According to the AH, if a language can relativize any NP position further down on the hierarchy, it can also relativize all positions higher up, i.e. to the left of it. This is what Hawkins (1999: 256), from a processing point of view, recently formulated as the Relative Clause Gap Hierarchy. For gapping there is thus a clear prediction that the relativized NP is most likely to be gapped if it is the subject of the relative clause, next most likely if it is the direct object of the relative clause, etc. (cf. Keenan 1985: 154). English dialects conform to this
SYNTACTIC MICROVARIATION §
203
prediction whereas Standard English does not: here only the (direct, prepositional) object position can be relativized by gapping. Another basic constraint on the AH is that any relative-clause forming strategy must apply to a continuous segment of the hierarchy. It is this constraint that made Ihalainen's earlier mentioned observation on relative clauses introduced by that look rather suspicious right from the start (cf. section 3). Ihalainen (1980) claimed that in the Southwest such relative clauses are used only for the relativization of the subject and indirect object/oblique position. This would have meant a non-continuous segment on the AH and would thus have run counter to it. Another relativization strategy often found in dialects is pronoun retention, i.e. the use of resumptive (or: shadow, copy pronouns). Here Keenan/Comrie made the prediction that this strategy is more likely to be used for positions further down on the AH. Compare similarly the prediction made by Hawkins' (1999: 258) "Relative Clause Copy Pronoun Hierarchy: If a relative clause is grammatical in position P on a complexity hierarchy H, then copy pronouns will be grammatical in all lower positions that can be relativized at all." This remains to be tested for varieties of English, but is most likely to be confirmed.16 4.2.3 § Negation Negation is yet another area of grammar where English dialects conform to prevailing cross-linguistic tendencies whereas Standard English does not. Double (or: multiple) negation is a well-known case in point. What is less known is that the pronounced tendency to use invariant negation markers (don't, ain't) in many non-standard varieties of English seems to follow a pattern that has been observed by Bernini/Ramat (1996) for Finnic languages, for example. Within Europe, these are the only languages besides Standard English (!) with inflected verbs or auxiliaries that must be used in negation. For the Finnic languages, Bernini/Ramat identify "a growing tendency to transform these verbs or negative auxiliaries into invariable NEG markers" (1996: 111).
16
The use of resumptive pronouns violates the Silent Trace Constraint in Optimality Theory ("Don't pronounce the trace of a moved constituent."). Far more on this constraint and under which conditions it can be violated (e.g. in conversational English), cf. Pesetsky (1997: 153ff.).
SYNTACTIC MICROVARIATION §
204
This is exactly what we can observe in many non-standard varieties of English. In fact, this is part of the supra-regional spoken non-standard English.17 In her monograph Negation in Non-Standard British English, Anderwald (to appear) makes another interesting observation that has hardly been noticed yet. This observation links up with what was said above about the regularization of the be-paradigm in many traditional and modern dialects: for the past tense, for example, as illustrated in (3e) many dialects use either was or were for all persons in the singular and plural. The dialects of the Southwest generalize were. Anderwald investigated among other things these two generalization strategies for negative sentences and found the following three patterns: (10) a. b. c. was-generalization in positive and negative contexts (was - wasn't) were-generalization in positive and negative contexts (were – weren't) mixed system: was-generalization in positive contexts and weren'tgeneralization in negative contexts The patterns in (b) and (c) are characteristic of non-standard varieties of current British English, with the mixed system in (c) being the more frequent pattern of the two. 18 According to Anderwald, varieties making use of this mixed system have "remorphologized" 19 the was-were distinction: the number distinction for the was/were choice familiar from Standard English has been replaced by a polarity distinction. This Anderwald interprets from a functional point of view. In the relevant varieties we can observe a maximization of the phonological and morphological difference between the positive and the negative form of past tense BE (was vs. weren't). This maximized formal distinction, in turn,
17
Note that in African-American Vernacular English ain't is often used for negative forms of do, too. Compare Kautzsch (to appear) on this characteristic in Earlier African-American Vernacular English. 18 The pattern in (a) is found only in very few British dialect areas, and there only very marginally. Only in positive contexts is was-generalization clearly preferred over were-generalization. In negative contexts, on the other hand, the use of generalized weren't over wasn‘t is much more common and pronounced than is the use of generalized was over generalized were in positive contexts. In general, Anderwald found that in nonstandard varieties of English generalization in negative clauses is three times as likely as in positive clauses. 19 This term was used first by Wolfram and Schilling-Estes (1996) for the same phenomenon in the dialect of Ocracoke in the Southern US.
SYNTACTIC MICROVARIATION §
205
offers crucial processing advantages in that it iconically codes a maximal cognitive difference, namely the positive-negative distinction. In closing, two points are worth noting concerning Anderwald's observations. She explicitly states that in her data of the British Isles only the mixed system in (10c) can be found, not the logical second possibility of varieties with generalizes were in positive contexts and generalizes wasn't in negative contexts. It needs to be added here that Anderwald did not investigate traditional dialects. In the FRED corpus, there are indications that at least individual traditional dialect speakers may use exactly this were-wasn't pattern. For example, although the Southwest is an area with were-generalization in positive and negative contexts, there is one speaker in the southwestern data who employs the were-wasn't pattern consistently at least in the singular, as illustrated in the examples in (11):20 (11) a. b. c. d. I were married in 1920.... During the war, when the trade were bad. In they days there wadn any Frisian about... And when this milk trade were so bad, ..., you had to do as best you could with it, see. Oh, it wadn very much money then, you see. This shows again how important it is to work with a relatively large database for the study of dialect syntax. In a given dialect or dialect area, you may come across interesting phenomena, but, more importantly, you can put these in perspective and do not run the risk of formulating for this dialect or dialect area generalizations on the basis of what appears to be rather an idiolect. On a larger scale, the negative markers and negation strategies found in non-standard varieties, i.e. the invariant forms ain't and don't as well as the was/were-patterns in (10), conform to the following much more general tendencies: (a) the doing away with person and number distinctions for present and past tense forms of be, i.e. a development towards a much simplified beparadigm in line with the simplification of the paradigms of full verbs (e.g. in the present tense: either generalized -s or for no person -s); (b) at the same time
SYNTACTIC MICROVARIATION §
206
this instantiates the elimination of the last remnants of subject-verb agreement in English; (c) from a typological point of view, more exactly from what is known about markedness phenomena across languages, we can observe the reduction of morphological distinctions under negation. From here it is tempting to move on to the fascinating question to what extent the grammars of English dialects can be said to exhibit, in general, a higher degree of regularity and consistency than Standard English (from an English-internal and crosslinguistic perspective). But this will need to be explored in another paper.
20
In the interview of this informant there are unfortunately no negative contexts in the plural. If this speaker followed the pattern consistently, we would expect wasn't (or rather wadn) in the plural, too.
SYNTACTIC MICROVARIATION §
207
5 § CONCLUSION The last two decades have seen an increasing number of attempts at a rapprochement between generative linguistics and functional typology (cf. For example Newmeyer 1998 and Haspelmath 1999). 21 Dialect syntax is the most recent area where we can observe or at least expect a fruitful competition between these two theoretical frameworks. This competition offers advantages for the further development of the relevant theories, but in the first place for the status of syntactic research in dialectology. The much improved data situation in itself offers new prospects for the study of English dialect syntax. Of course, these new research opportunities and research efforts will not remedy all the shortcomings of the past with regard to the collection of data relevant for the study of dialect syntax. Local dialects continue to disappear and "weaken" (dialect levelling) due to an increased mobility, urbanization, and the impact of the standard variety in education and the media. But that there are bright prospects for the study of dialect syntax can hardly be denied. This is amply documented in this volume, which itself is a result of the fact that in several European countries (Italy, The Netherlands, Belgium, Switzerland, and Germany) there currently exist major projects on dialect syntax. In the course of the next years, this opens up excellent perspectives not only for the study of the dialects spoken in the individual countries, but especially for cross-linguistic dialect studies, for instance for comparative studies of the West Germanic dialects.
21
Haspelmath (1999) offers a thought-provoking attempt at reinterpreting the grammatical constraints in Optimality Theory as ultimately functionally motivated and thus as user-based constraints. On a more sobering note, it looks as if such laudable attempts at a rapprochement remain rather isolated cases, though.
SYNTACTIC MICROVARIATION §
208
§ References
Abraham, W./J. Bayer, eds. (1993). Dialektsyntax. Opladen: Westdeutscher Verlag. Anderwald, L. (to appear). Negation in Non-Standard London/New York: Routledge. Archangeli, D. (1997). Optimality Theory: An Introduction to Linguistics in the 1990s. In: Archangeli, D./D.T. Langendoen, eds. 1-32. Archangeli, D./D.T. Langendoen, eds. (1997). Optimality Theory. An Overview. Oxford: Blackwell. Benincá, P., ed. (1989). Dialect Variation and the Theory of Grammar. Dordrecht: Foris. Bernini, G./P. Ramat (1996). Negative Sentences in the Languages of Europe: A Typological Approach. Berlin: Mouton de Gruyter. Black, J.R./V. Motapanyane, eds. (1996). Microparametric Syntax and Dialect Variation. Amsterdam/Philadelphia: Benjamins. Chambers, J.K./P. Trudgill. (1991). Dialect Grammar: Data and Theory. In: Trudgill, P./J.K. Chambers, eds. 291-296. Cheshire, J./V. Edwards/P. Whittle (1993). Non-Standard English and Dialect Levelling. In: Milroy, J./L. Milroy (eds.), 53-96. Chomsky, N. (1995). Some Notes on Economy ed. The of Derivation and Representation. In: Chomsky, N., Cambridge, Mass.: MIT Press. 129-166. Collins, C. (2001). Economy Conditions in Syntax. In: Baltin, M/C. Collins, eds. The Handbook of Contemporary Syntactic Theory. Oxford: Blackwell. 4561. Croft, W. (1990). Typology and Universals. Cambridge: Cambridge University Press. —— (2000). Explaining Language Change. An Evolutionary Approach. London: Longman. Dahl, ??. (1999). Does Adaptation Help Us to Understand Language Change? Zeitschrift für Sprachwissenschaft 18: 209-211. Ellis, A.J. (1889). On Early English Pronunciation, Part V, The Existing Phonology of English Dialects Compared with That of West Saxon (Early Minimalist Program. British English.
SYNTACTIC MICROVARIATION §
209
English Text Society, Extra Series 56). Reprinted New York: Greenwood Press. Elworthy, Th. (1877-79). An Outline of the Grammar of the Dialect of West Somerset. Transactions of the Philological Society: 143-257. Fischer, O. (1992). Syntax. In: Hogg, R.M., ed. The Cambridge History of the English Language. Vol. 2. Cambridge: Cambridge University Press. 207408. Foulkes, P./G. Docherty, eds. (1999). Urban voices: Accent Studies in the British Isles. London: Arnold. Garrett, A. (1998). On the Origin of Auxiliary Do. English Language and Linguistics 2: 283-330. Glaser, E. (to appear). Erhebungsmethoden dialektaler Syntax. In: Tagungsband. 1. Internationaler Kongress der deutschen Dialektologie. G??rlach, M. (1999). Regional and Social Variation. In: Hogg, R.M., ed. T h e Cambridge History of the English Language. Vol. 3. Cambridge: Cambridge University Press. 459-538. Haspelmath, M. (1999). Optimality and Diachronic Adaptation. Zeitschrift für Sprachwissenschaft 18: 180-205. Hawkins, J.A. (1999). Processing Complexity and Filler-Gap Dependencies across Grammars. Language 75: 244-285. Haegeman, L./J. Guéron. (1999). English Grammar: A Generative Perspective. Oxford: Blackwell. Henry, A. (1996). Imperative Inversion in Belfast English. In: Black, J.R./V. Motapanyane, eds. 79-93. Ihalainen, O. (1980). "Relative Clauses in the Dialect of Somerset". Neuphilologische Mitteilungen 81: 187-196. —— (1991). On Grammatical Diffusion in Somerset Folk Speech. In: Trudgill, P./J.K. Chambers, eds. 104-119. —— (1998). The Dialects of England since 1776. In: Hogg, R.M., ed. T h e Cambridge History of the English Language. Vol. 4. Cambridge: Cambridge University Press. 197-274. Inoue, F. (1999). Subjective Dialect Division in Great Britain. In: Preston, D.R., ed., Vol. 1: 161-176.
SYNTACTIC MICROVARIATION §
210
Kautzsch, A. (to appear). History of African-American Vernacular English. Berlin/New York: Mouton de Gruyter. Kayne, R.S. (1996). Microparametric Syntax: Some Introductory Remarks. In: Black, J.R./V. Motapanyane, eds. ix-xviii. Keenan, E.L./B. Comrie. (1977). Noun Phrase Accessibility and Universal Grammar. Linguistic Inquiry 8: 63-99. Keenan, E.L. (1985). Relative Clauses. In: Shopen, T., ed., Language Typology and Syntactic Description, Vol. II: Complex Constructions. Cambridge: Cambridge University Press. 141-170. Kerswil, P./C. Llamas/C. Upton. (1999). The First SuRE Moves: Early Steps Towards a Large Dialect Project. In: C. Upton/K. Wales, eds. 257-269. Kirk, J.M. (1985). Linguistic Atlases and Grammar: The Investigation and Description of Regional Variation in English Syntax. In: Kirk, J./S. Sanderson/J.D.A. Widdowson, eds. Studies in Linguistic Geography. The Dialects of English in Britain and Ireland. London: Croom Helm. 130-156. Klemola, K.J. (1996). Non-Standard Periphrastic Do: A Study in Variation and Change. Unpublished Ph.D. thesis, University of Essex. —— (2001). Survey of English Dialects on CD-ROM. The Spoken Corpus, recorded in England 1948-1973. London: Routledge. Kortmann, B. (1997). Typology and Language Change. In: W. B??ker/H. Sauer, eds. Anglistentag 1996 Dresden - Proceedings. Trier: Wissenschaftlicher Verlag. 109-124. —— (1999). Typology and Dialectology. In: Caron, B., ed. Proceedings of the 16th International Congress of Linguists, Paris 1997. CD-ROM. Amsterdam: Elsevier Science. Kretzschmar, W.A. (1999). The Future of Dialectology. In: Upton, C./K. Wales, eds. 271-288. Labov, W. (1994). Principles of Language Change. Vol: 1: Internal factors. Oxford : Blackwell. Labov, W./S. Ash/C. Boberg. (to appear). Atlas of North American English: Phonetics, Phonology and Sound Change. Berlin/New York: Mouton de Gruyter. Milroy, J./L. Milroy, eds. (1993). Real English. The Grammar of English Dialects in the British Isles. London/New York: Longman.
SYNTACTIC MICROVARIATION §
211
Newmeyer, F. (1998). Language Form and Language Function. Cambridge, Mass.: MIT Press. Pawley, A. K. (1994). Austronesian terminologies: continuity and change. Canberra: Dept. of Linguistics, The Australian National University. Penner, Z., ed. (1995). Topics in Swiss German Syntax. Bern/Berlin/Frankfurt a.M.: Lang. Pesetsky, D. (1997). Optimality Theory and Syntax: Movement and Pronunciation. In: Archangeli, D./D.T. Langendoen, eds. 134-170. Poplack, S., ed. (2000). The English History of African American English. Oxford: Blackwell. Preston, D.R., ed. (1999). Handbook of Perceptual Dialectology. Amsterdam/Philadelphia: Benjamins. Radford, A. (1997). Syntactic Theory and the Structure of English. A Minimalist Approach. Cambridge: Cambridge University Press. Ramisch, H./K. Wynne, eds. (1997). Language in Time and Space. Stuttgart: Steiner. Siemund, P. (2001). Pronominal Gender in English. A Study of English Varieties from a Cross-Linguistic Perspective. (Unpublished postdoctorate thesis; Free University of Berlin) Toon, Th.E. (1992). Old English Dialects. In: Hogg, R.M., ed. The Cambridge History of the English University Press. 409-451. Traugott, E. Closs. (1999). From a Diachronic Perspective. Peer Commentary on Haspelmath 1999. Zeitschrift für Sprachwissenschaft 18: 238-241. Tristram, H.L.C. (1997). DO-Periphrasis in Contact? In: Ramisch, H./K. Wynne, eds. 401-417. Trudgill, P. (1990). The Dialects of England. Oxford: Blackwell. Trudgill, P./J.K. Chambers, eds. (1991). Dialects of English. Studies in Grammatical Variation. London/New York: Longman. Upton, C./D. Parry/J.D.A. Widdowson. (1994). Survey of English Dialects: The Dictionary and Grammar. London/New York: Routledge. Upton, C./K. Wales, eds. (1999). Dialectal Variation in English: Proceedings of the Harold Orton Centenary Conference. (Leeds Studies in English, New Series 30.) University of Leeds.
SYNTACTIC MICROVARIATION §
Language. Vol. 1. Cambridge: Cambridge
212
Van der Auwera, J. (1999). Periphrastic 'Do': Typological Prolegomena. In: Tops, G.A.J./B. Devriendt/S. Geukens, eds. Thinking English Grammar: To Honour Xavier Dekeyser, Professor Emeritus. Leuven: Peeters. 457-470. Wakelin, M.F. (1986). The Southwest of England. Amsterdam/Philadelphia: Benjamins.
SYNTACTIC MICROVARIATION §
213