The contents of this volume originated in a conference held in Munich in November 2012, on the occasion of the tenth anniversary of the international doctoral program on “Sprachtheorie und Angewandte Sprach-wissenschaft.” The ten papers include all but one of the papers presented there, with the addition of two invited contributions on selected syntactic features of German dialects of northern Italy. The theme of these papers—as of the conference—is the particular kind of microvariation which can be observed in regional dialects. The overarching goal is to evaluatethe potential contribution of close comparative investigation of such microvariation to our knowledge of language and, in particular, to our understanding of Universal Grammar.

In fact, the value of dialects as a testing ground for linguistic theory has long been recognized, despite some suggestions to the contrary at various places in this volume—as, for example, in the polemical title “Dialektologie in neuem Gewand” [Dialectology in a new guise]. However, I recall hearing, as a graduate student in Manchester in the mid-sixties, a fascinating guest lecture by William G. Moulton on “Dialects as a language laboratory”—in practice, a revised version of Moulton Reference Moulton1962. This term, and the topic, is picked up in another recent collection, De Vogelaer & Seiler 2012. In its introduction, the authors provide a concise and informative survey of the contribution dialectology has made to the development of linguistic theory since the days of the Neogrammarians.

Nevertheless, theoretical linguistics has sometimes been uncomfortable with the fact of linguistic variation and change, as this hason occasion proved difficult to reconcile with the postulate attributed to Saussurethatlanguage is “un systèmeoù tout se tient” or with Chomsky's (Reference Chomsky1965:3) “ideal speaker-listener, in a completely homogeneous speech-community.”Footnote ¹ De Vogelaer& Seiler (2012:1), though, take the view that “the study of dialect variation has indeed the potential […] to play a central role in the process of finding answers to fundamental questions of theoreticalhistorical linguistics.”In their introduction to the present volume (p. 7), Werner Abraham and Elisabeth Leiss similarly emphasize that the kind of microvariation found in related dialects could throw light on the processes involved in language change. However, the tension between the fact of variation and some crucial theoretical assumptions about language becomes apparent when, later on the same page, they ask the question:

This is almost as if competence were to be exclusively associated with written rather than spoken language or with codified standard languages. After all, thelatter are “cultural artefacts” that consist ofprescriptive norms thatmay deviate significantly from speakers'actual grammars, as Davies (Reference Davies, Hüning, Vogl and Moliner2012) makes clear.

However, what does incontrovertibly constitute a “new look” in this collection is the focus on syntax, as all ten of the papers deal with syntactic issues. This area had been notoriouslyneglected in traditional—especially German—dialectology, which rarely ventured beyond phonology and morphology. In the first one “Philologische Dialektologie und moderne Mikrovarietätsforschung. Zum Begriff des Erklärstatus in Syn- und Diachronie” [Philological dialectology and modern micro-variation research]. Considering the status of explanation in synchrony and diachrony](pp. 9–27), Werner Abraham presents the contrast between traditional dialectology and modern approaches in a rather different light. He admits that certain modern linguistic theories have difficulty accounting for linguistic change, but he shows that certain features of Cimbrian dialects can be most satisfactorily explained in terms of their internal parameters, as proposed by modern approaches, and not through contact, as proposed in traditional research. Thus, the clitic pronouns of Cimbrianare not the result of contact, because they are proclitics and not enclitics (as in Italian and other Romance languages). As encliticsthey are essentially incompatiblewith the V2 (or SVOV) structure, which Cimbrian, like all varieties of German, has retained. Thus, Abraham provides an internal explanation for the structure of Cimbrian on the basis of Universal Grammar—albeit in this case, it is essentially a retention of inherited structures rather than a change in the strict sense.

However, two aspects of his account may be open to question. First, Abraham suggests (pp.20–21) that linguistic change takes place more readily in orally transmitted dialects than in written varieties. However, if, as he accepts, linguistic change takes place during childhood acquisition, that is, presumably before the achievement of literacy, then it is surely irrelevant whether a variety has a written form or not. Second, he asserts (p. 16) that “In einer lebenden Dialektsprache gibt es Wandel unter Sprachkontakt bloßdort, wo solcher Wandel auch autonom stattfinden hätte können, ”in other words, where it is compatible with the structure of the language.Footnote ² He then claims (p. 22), in support of this view, that if Old English, with its SVOV structure, changed to SVO by Middle English, this was due to the catastrophic effect of massive contact with Norman French, comparable to creolization. However, this contention is highly controversial among linguists specializing in the history of English, and it is, for instance, rejected outright by Fischer & van der Wurff (Reference Fischer, van der Wurff, Hogg and Denison2006:185–188).

In the second paper “Klitisierung, Reanalyse und die Lizensierung von Nullformen: Zwei Beispiele aus dem Bairischen” [Cliticization, reanalysis, and licensing zero forms. Two examples from Bavarian] (pp. 29–45), Josef Bayer deals with two features of Bavarian dialects. One of these, namely, complementizer agreement, is now very well known and has attracted much theoretical attention: It is difficult to explain how (or why) the agreement of a finite verb with its subject should be extended to the C position (and that, in many dialects, only for the 2nd person singular and plural). The suggestion (p. 30) that Universal Grammar might make provision for complementizer agreement in all languages, but that it is then overtly expressed in very few, would seem a not awfully insightful application of the notorious “Duke of York gambit” (Pullum Reference Pullum1976). Besides, as Bayer subsequently shows, this suggestion cannot be upheld in light of data from other Bavarian dialects, where the complementizer agreement takes the form of the 1st person plural pronoun -ma, and it could thus be generally taken as independent of the agreement of subject and verb (and may license prodrop). On considering further data (including some from West Flemish), Bayer concludes, quite plausibly, that -ma may ultimately be traced back to the cliticization of a following pronoun, and this may be a vocative rather thena subject.

The second feature of Bavarian dialectsthat Bayer deals with has been commented on much less: Itis the almost obligatory cliticization ofthe downtoner particle denn in W- questions (in the form -n, before the subject), for example, Wowohnst-n-du? ‘Where do you live, then?’ or Wos is-n do los?‘What's up, then?’ A variant of this feature in Bavarian is the optional deletion of the interrogative wos in such sentences,[_] is-n do los? This deletion can be explained through its similarity with complementizer agreement: If the latter involves marking C for the features [person] and [number], then -n can be taken as indicating the unmarked interrogative wos and thus licensingits deletion.

The next two papers, both in English, are by Italian colleagues whose contribution to the volume was invited. Like the paper by Werner Abraham, they deal with island dialects of German in northern Italy. The first, by Ermengildo Bidese, Andrea Padovan, and Alessandra Tomaselli, “Bilingual competence, complementizer selection and mood in Cimbrian” (pp. 47–58), explicitly challenges Abraham's view that “language contact should be considered as a last resort, both in terms of work methodology and as justification strategy” (p. 48) and asserts that certain syntactic features of the dialect of Lusern are indeed based on calques from Italian. In particular, they show that the two complementizers in the dialect, the inherited ke(borrowed from the surrounding Italian dialect) and az, select different moods after nonfactive verbs such as gloam‘believe’ (High Germanglauben) in a rather complex pattern. On the basis of this, they concludethat the unexpected co-occurrence ofke with a following subjunctive “must be parasitic on the Italian pressure” (p. 55). However, one must point out that this conclusion tacitly relieson the potentially quite problematic assumption that linguistically naïve bilingual speakers are in a position to draw a parallel between the instances ofan abstract morphosyntactic categorysuch as mood in their two languages. What makes this assumption even more problematic is that the functions of the mood in Cimbrian dialect and in Italianare rather different. Instances of such transfer are in practice quite difficult to find.

The second invited paper, “Limits of syntactic variation and Universal Grammar. V2, OV/VO and subject pronouns in Mòcheno” (pp. 59–83), by Federica Cognola, deals with syntactic variation in Mòcheno, the dialect of the Fersina valley (Fersental) in the province of Trento. In a thoughtful and well-detailed account with full and excellently presented data, the author shows that what appears to be rather unsystematic deviation from the canonic V2 of most varieties of German is actually rule-governed. This variation has been ascribed to linguistic contact, but Cognola rejects this, since the word order of the neighboring Italian dialects is rather different, and the apparent variation (and deviation from canonic V2) in Mòcheno is actually quite similar to that found in some other Germanic languages, notably Norwegian and Old English. Thus, the variation in the position of pronoun subjects is systematically related to the distinction, which can be independently verified, between three types of pronoun: strong, clitic, and weak. In addition, different positions of DP subjects are not to be analyzed as optional, as the DP subject position is wholly dependent on the rules of interaction between syntax and information structure. Thus, the problem posed by these data is not how to account for supposed optionality in syntax—which the author explicitly dismisses—but how to systematically integrate information structure into general syntactic theory.

The paper by Cecilia Poletto, “On the internal structure of q-words” (pp. 85–101), is the only one which is not primarily on dialects of German, but covers French and Italian (and some regional forms of these) together with some examples from Cimbrian and Icelandic. The author presents an explanation for why bare quantifiers may behave in a different way to quantified expressions. In French, for example, some bare quantifiers may intervene between the auxiliary and the past participle in compound tenses, as inJe n'ai rienvu ‘I didn't see anything’. This is not possible with a quantified expression: *Je n'ai aucungarçonvu‘I didn't see any boy’ is ungrammatical. The author maintains that this contrastcannot be ascribed to the fact that bare quantifiers might be weak. She argues instead thatthey are not paired with a full DP or NP but with a covert classifier of the kind which surfaces in English in forms with -thing or -body (and similarly in several varieties of Italian). This explanation finds support in its mirroring the well-known phenomenon of the split of a quantificational portion from its lexical component (see Kayne Reference Kayne, Broekhuis, Corver, Huybrechts, Kleinhenz and Koster2006), and the fact that in her account, bare quantifiers frequently display adverbial properties.

In his extremely wide-ranging article, “Infinitivprominenz in deutschen Dialekten” [Infinitive prominence in German dialects] (pp. 103–140), Oliver Schallert touches on many of the principal themes raised in this volume. He begins by looking at evidence for syntactic variation in the material of the Deutscher Sprachatlas, which, characteristically for the history of German dialectology, has been available since Wenker's survey, but with the traditional focus on phonology and morphology it had been completely ignored. Thus, he shows that areas can be clearly identified where subordinate clause constructions, such as Du bistnochnichtgroßgenug, dass du eine Flasche Wein austrinkenkannst‘You aren't big enough yet to finish off a bottle of wine’ are recorded rather than Du bistnochnichtgroßgenug, um eine Flasche Weinauszutrinken. He also shows that dialects exhibit considerable variation with respect to the much-discussed coherent infinitive construction. For example, constructions such as Woerihnzutreffennichtversucht hat‘Where he hasn't tried to meet him’ are often allowed, which would be problematic in standard German (as well as theoretically challenging). One also finds other constructions where it is not wholly clear whether they contain nominal or verbal elements, for example,Eshörtglei auf van schneibn‘It will stop snowing in a minute’ (Traunstein).

However, his perhaps most challenging and important point concerns the nature of linguistic data, in thatseveral constructions which have occasioned much theoretical debate are simply not attested in any dialect or spoken variety, but are halbnatürlich ‘semi-natural’ features of written standard. These include obvious cases like extended epithets and piedpiping (die Ratten, die zufangensichniemandtraute ‘the rats which nobody dared to catch’), but possibly even relative pronouns, of which Fleischer (Reference Fleischer2005:184–185) has said “Das System der neuhochdeutschen Standardsprache hat keine vergleichbare dialektale Basis und ist in diesem Sinn unnatürlich.”Footnote ³ This, naturally, raises the absolutely fundamental issue of whether data can appropriately be obtained through acceptability judgments deriving from introspection by highly educated and literate linguists of their own speech, or whether it may be more reliably obtained from observing actual spontaneous production.

Gunther De Vogelaer and Jan Klom's paper, “Mikrovariation beim Erwerb des niederländischen Genussystems” [Microvariation in the acquisition of the Dutch gender system] (pp. 141–169), begins by stressing that the most important distinction between traditional dialectology and microvariation is the latter's concentration on I-language rather than E-language, and notits adherence to some version of the generative paradigm. Thus, their aim is to explain rather than describe, with particular reference to recent developments in the Dutch gender system. They discuss modern sociolinguistic studies that have called into question the notions of language and dialect (or variety), since individuals typically have competence over a wide linguistic repertoire, and such constructs are essentially artificial generalizations over a range of speaker behavior. Children acquiring language in the Netherlands and Flanders, as well as most European countries, are typically exposed to an extremelyvaried range of input from standard (or near-standard) to dialect. Thus, changes in the gender system among children in the southern Netherlands—the system that had maintained a three-waygender contrastas opposed to the two-way gender contrast in the North—do not simply involve the abandonment of the regional distinction between masculine and feminine, in line with standard Dutch;such changes involve resemanticization, that is, children's reinterpretation of the variable inputs to which they are exposed. For example, children can reinterpret mass nounsas het-words (that is, neuter) and count nouns as de-words (that is, non-neuter), and shift to semantically-based rather than grammatically-based anaphora. Such developments are taking place in various ways in different regions of the Dutch-speaking area, and, interestingly, they could possibly involve post-critical threshold learning.

Helmut Weiß's main aim in his paper, “UG und syntaktische Mikro-variation” [UG and syntactic microvariation] (pp. 171–205), is to demonstrate that syntactic variation cannot be used to challenge the existence of Universal Grammar. Nevertheless, the fact of variation is a challenge for syntactic theory. Chomsky's (Reference Chomsky1988:187) view, quoted here, that “every human being speaks a variety of languages. We sometimes call them different styles or dialects, but they are really different languages” effectively implies uniformity for each of these languagesand is ultimately difficult to reconcile with observations of language variation in use. As Weiß points out, individuals appear to have access to a range of options, even within what one might consider the “same” variety. He presents a number of examples of this, mainly from a project on the syntax of the dialects of Hesse. The first issue the projectlooked at (and the one Weiß deals with in this paper) wasthe order of clitic pronouns and null subjects—again, features present in the material of the Deutscher Sprachatlas but largely neglected by traditional dialectologists. In a further section, Weiß presents a number of examples to test the view that syntactic variation is either reducible to the lexicon or may be ascribed to the level of phonological form. He finds that although this is the case, some variation, for example, variation that involvescomplementizer agreement and possessive constructions, may originate in the morphology.

In Heike Wiese's paper “What can new urban dialects tell us about internal language dynamics? The power of language diversity” (pp. 207–245), the reader is immediately confronted with the old terminological issue of what is to be labeled a dialect. In the German tradition, the term has usually been restricted to areally defined varieties, whilst linguists in the English-speaking world have often used the term to denote any variety used by a particular social group and speak about certain “social dialects” or “urban dialects” (or even “the standard dialect”). These terms reflect the very different linguistic and political history of the German- and English-speaking countries, but given recent changes in the patterns of sociolinguistic variation in Germany, Wiese feels that it is appropriate to designate the urban variety spoken in multiethnic communities in Berlin, Kiezdeutsch, as a “dialect,” as in her longer monograph, Wiese Reference Wiese2012.

In this paper, she gives an account of three features of this variety, that is, the use of bare local NPs, as in Die muss Bahnhofgehen, ‘She must go station’,violation of V2 order, as in Dann die sindzur U-Bahngerannt‘Then they ran to the underground station’,and the use of invariant gibts as an existential marker. All these features are well documented here for Kiezdeutsch, but the question must be asked whether they are actually exclusive markers of that variety—or, indeed whether this is really a “new dialect” of German. As Wiese says (p. 230, note 54), Kiezdeutsch is “an informal oral variant [sic] of German,” and it would seem plausible that these features are to be found more generally in informal speech in other areas. My own informal observations suggest that this is the case, but the investigation of such varieties has been sadly neglected in the German speech area, where the pressure from normative traditions seemson occasion to have led to many nonstandard variants being completely overlooked (or their existence even denied). Salmons (Reference Salmons2012:322f.), for example, shows thatweil with V2 can be documentedas far back as the early 19th century, and the notion, in the last few years, that it is an instance of Sprachverfall is due to the “recency illusion.” Unfortunately, as Haas (Reference Haas, Glaser, Schmidt and Frey2011:18) says: “Im Grunde ist es ja grotesk, wie wenig wirüber die Formen des Deutschen wissen, die von der überwiegenden Mehrheit der Deutschsprachigen im Alltag tatsächlich gesprochen werden.”Footnote ⁴

The last paper, “Dialect as a spoken-only medium: What it means—and what it doesn't mean” (pp. 247–271), by Werner Abraham, effectively summarizes the main themes of the volume and picks up issues raised in the introductory article. In particular, he stresses again that in looking at microvariation it is essential always to be aware that dialects are oral varieties without support from writing or literacy. However, this is only partially the case for the varieties dealt with in some of the papers in this volume. De Vogelaer and Klom, for example, show how language change is coming about through children's exposure to standard and nonstandard varieties. Similarly, Heike Wiese shows that there is considerable interaction between Kiezdeutsch and near-standard German, in which her informants also have some level of competence. However, Werner Abraham is surely correct in his conclusion that syntactic variation at the micro-level, as illustrated by several of the papers in this volume, cannot per se seriously challenge the assumption of Universal Grammar. Itultimately has to explain the range of possibilities exhibited in spoken varieties, which, after all, precede written language, especially writing in fully codified standard varieties.

The overall presentation of the volume is good, with relatively few misprints or other errors, although one or two matters of detail could have been given more attention. For example, some of the authors use the terms variant/Variant and variety/Varietät inconsistently, without making the usually accepted distinction between them. Furthermore, occasional references are missing. For example, Wiese refers to Trudgill 1992 and Leiss 2010, but these works are not in her bibliography. It is also perhaps strange that the title and cover materialare only in German, given that five of the ten papers are in English. In these, the standard of English is generally high, although the paper by Bidese et al. could have benefitted from judicious editorial attention: It is not always felicitous in its syntax and choice of expression, such that it is in places quite difficult to read. I also feel that the publisher should have insisted on full adherence either to American and British standard orthographic norms, as there is some inconsistency in these papers. Also, not all the papers are preceded by initial abstracts (two are lacking, for no apparent reason).

However, such minor quibbles aside, this volume is to be strongly recommended. The standard of the papers is very high, and although not all linguists would concur with the theoretical approaches espoused by some contributors, they raise a number of fundamental issuesof crucial concern to all linguists, notably as regards the status of varieties and dialects and their place in linguistic research, the nature of linguistic data and the relationship between data and theory, and the essential features of linguistic variation and how it may be accounted for and analyzed satisfactorily.