This volume is a collection of 18 papers, mainly arising from the workshop “Dialectology and Typology” organized by Bernd Kortmann at the conference METHODS XI in Joensuu, Finland (5–9 August 2002). Ten of the contributors were invited speakers at this workshop and, according to the editor, the remaining eight “share a pronounced interest in looking across the fence which still separates dialectology and typology.” Typology in this volume is understood in functional terms in the Greenbergian tradition – that is, as the study of structural similarities between languages resulting from common design features and general properties of communicative systems (Muysken Reference Muysken, Gilbers, Nerbonne and Schaeken2000: 263). The chapters are arranged in alphabetical order of the authors (Lieselotte Anderwald, Peter Auer, Raphael Berthele, Jack Chambers, Östen Dahl, Gunther de Vogelaer, Jürg Fleischer, Bernd Kortmann, Yaron Matras, Jim Miller, Günter Rohdenburg, Guido Seiler, Peter Siemund, Peter Trudgill, Johan van der Auwera & Annemie Neuckermans, Susanne Wagner, Björn Wiemer), although the volume starts out with the contribution by Walter Bisang in a nod to his integration of sociolinguistic models of diffusion with contact linguistics in typology and dialectology. In addition, there is a 10-page preface by the editor.

The volume has a clear geographic and genetic bias with respect to the dialects and (non)standard varieties discussed: The overwhelming majority of the papers deal with contingent areas and induced language-contact situations in Europe. It discusses (morpho)syntactic phenomena of varieties of Baltic, Dutch, French German, Greek (sound changes), Romani, Scandinavian, and Slavic, in addition to English. As such, it is a convincing counterbalance to Chambers's claim (142) that studies of vernaculars are virtually nonexistent for varieties other than English.

The volume is very diverse in regard to the grammatical phenomena discussed, which include negation (Anderwald, van der Auwera & Neuckermans), complementation (Matras), complementizer agreement (De Vogelaer), clitic pronouns (Trudgill), resumptive pronouns (Rohdenburg), relative clauses and relativizers (Auer, Fleischer), and subject doubling (De Vogelaer), verb clusters (Seiler), pronominal reference (Wiemer), gender (Rohdenburg, Wagner, Wiemer), articles (Auer), agreement (Chambers, Wiemer), motion and posture verbs (Berthele), tense and aspect (Kortmann, Miller, Rohdenburg, Siemund), and do-periphrasis (Auer, Kortmann).

Both the editor and the authors support the view that dialectology in an integrated approach may benefit from the use of typology as a reference frame to study cross-dialectal and intradialectal variation in individual syntactic phenomena, whereas the use of dialect and nonstandard data in typology make it possible to formulate more fine-grained parameters, generalizations, and hierarchies.

Let us begin by emphasizing that discussing this volume within the scope of one short review is doomed to fail. It is impossible to do justice to each article in an appropriate way since they differ considerably with respect to the frameworks used, the concepts, and the analyses of the data. Rather than summarize and comment on each of the individual chapters, I will highlight themes that cut across the volume.

To start with, this volume is extremely valuable for its reflection on the problematic aspects of the use of standard data in language-internal grammatical theories, be they functionalist or formalist. The volume as a whole expresses a strong need to establish the empirical sources that can serve as crucial correctives for typological research (Kortmann, 2) and support claimed generalizations and hierarchies at best (cf. Bresnan Reference Bresnan2005). It is convincingly argued that in language-internal grammatical theories relevant variation will be overlooked when focusing exclusively on standard varieties (see Berthele, Bisang, Kortmann, Matras, Miller, Seiler, van der Auwera, and Neuckermans & Wagner) but, worse, this may also lead to an incorrect typological picture. Wagner (480) points out that while the Accessibility Hierarchy concerning relativization predicts that gapping or zero relativization should be possible in subject position, it is not possible in standard (written) English but it is in English dialects. Similarly, Fleischer shows that standard German is fairly unrepresentative and even atypical compared to dialect varieties with respect to relative clause formation (236). It is even argued by Fleischer (237) that the typology of Standard Average European (SAE) looks so strange compared with the rest of the worlds' languages because it is based on standard languages.

Further, a written standard may differ considerably from spoken standard varieties. Miller (310), for example, notes that in reference grammars standard English is usually presented as having a prototypical Perfect, one structure with four major interpretations, including the experiential/indefinite anterior. But, as Miller argues, speakers and writers producing spontaneous UK standard speech frequently use the simple past combined with ever and never to express experiential/indefinite anterior. This has consequences, since any researcher comparing the classical prototypical perfect of standard English with nonstandard English varieties should also take the simple past into account.

Milroy (Reference Milroy2001:545) explains why written standard varieties do not always behave according to typological predictions. In a European context, functional and other formal theories that are based on standard languages always imply an empirical base that is externally characterized by numerous social and ideological criteria. Standard languages are used in writing; they have “educated” status and literary functions. The most general assumptions conditioned by the standard-language ideology are that standard varieties are uniform in structure, that they are stable, and that they are finite-state entities. However, as Milroy (Reference Milroy2001:545) suggests, these are not properties of “real” languages but properties of idealized states of languages. Probably it is this idealized or “constructed” state of a standard that explains why the linguistic conditioning of (written) standards is different than in the case of nonstandard varieties.

Interestingly, in this volume two alternatives are presented to avoid as much as possible bringing typological inaccuracies along with the selection of a standard variety. First, it is possible to choose a variety that has no standard and/or globally accepted prestige dialect and is an oral variety for the great majority of speakers, as Matras discusses for Romani (279). Second, Auer's exploration (see also Berthele) of which type of data is the most accurate for syntactic typologies and generalizations is extremely important. It is suggested that analysts select those syntactic features that have a wide areal distribution and do not belong to the written and oral standard variety – the so-called nonstandard, nondialectal features. Examples include double negation and periphrastic do, which occur throughout the German area (and cross national borders too, since these constructions also appear in non-spoken standard in the Netherlands and Belgium (Barbiers et al. Reference Barbiers, Cornips and Van der Kleij2002)). In contrast, dialect features with a restricted areal reach may in fact constitute idiosyncrasies that do not contribute to a typological perspective on, for instance, German as a whole (whereas they are of course relevant for any linguistic theory). Moreover, Auer emphasizes the importance of keeping separate the structural consequences of orality, such as prolepsis, elliptic constructions, and congruence violations, from geographically restricted syntactic features. Also, Cheshire Reference Cheshire, Cornips and Corrigan2005 argues that the way speakers use the forms of spoken language may result as much from interactional and social factors as from syntax, and typologists need to come to a principled decision about whether such forms should be seen as generated by the grammar or attributed to performance mechanisms.

This volume is also extremely valuable in that it convincingly reveals that “the present-day political borders between the countries do not correspond well to the internal structure of the dialect continuum and the diversity within the countries may be greater than the differences between the national standards” (Dahl, 159).

There are three issues I would like to address, however. The first issue is that concepts like simplification/complexification, (un)markedness, and (un)naturalistic tendencies play a very important role in the theoretical analyses in several contributions in this volume, but their definitions are left implicit and the authors' attitudes remain unclear (with the exception of Anderwald, 48; cf. Kusters Reference Kusters2003:1–17). These concepts are extremely complicated and well founded from the perspective of functional grammar, but without explicit characterizations a precise insight into their explanatory adequacy is impossible.

Second, why did it take so long to include nonstandard/dialect data in typology (or other formal theories)? The reason for exclusion cannot be a grammatical one, since structural differences between a dialect and a related language are of the same nature as differences between two related languages. This is so because the question of what makes a “dialect” and what makes a “language” is not linguistically motivated but rather social in nature (Bisang, 13). Bisang (25) doesn't exclude a priori the existence of major structural differences between closely related varieties. Indeed, many authors in this volume show that dialects may differ in many (sometimes surprising) respects from the standard variety. One answer might simply be the shortage of morphosyntactic dialect data available for typological research, since it was a neglected topic in dialectology in the past (Barbiers et al. Reference Barbiers, Cornips and Van der Kleij2002, Cornips & Poletto Reference Cornips and Poletto2005). Nevertheless, computerized storage of spoken-dialect morphosyntactic data with annotation and retrieval has become available recently, and two such databases are drawn on in this volume: the Freiburg English Dialect corpus (FRED) (Anderwald) and the Romani Morphosyntactic database (RMS) (Matras). (See also the Syntactic Atlas of the Dutch Dialects, SAND; Barbiers et al. Reference Barbiers, Cornips, Kunst, Beal, Corrigan and Moisl2007.) It is a pity that some authors in this volume still concentrate their research on traditional dialect grammars and other (written) descriptions of very local dialect varieties (in the past).

The final issue concerns the notion of inherent variability. Although the empirical base in this volume is constituted of dialect and nonstandard varieties rather than standard varieties, there is no mention or focus on variability within the grammars. It is, of course, an important empirical and theoretical question whether we consider nonstandard varieties in contemporary Western communities as homogeneous and stable, since they rarely exist nowadays in absolute isolation. Indeed, most competent speakers of dialects and nonstandard varieties can resort to a range of varieties along a continuum from standard to nonstandard, and stylistically from more formal to less formal, depending on social and discourse contexts (cf. Cornips & Corrigan Reference Cornips and Corrigan2005). In addition, dialect speakers may be raised “bilingually” from birth in the local dialect and a supralocal variety, thus being, in effect, simultaneous L1 acquirers. In my view, issues as these should be a focus for further investigation and research.

In sum, this book is admirable for its ambition to answer the ultimate question that the inclusion of dialects in European areal typology might answer: “what the ‘real’, i.e. spoken non-standard, landscape of Europe looks like” (Kortmann, 6). It is certain that many readers will be inspired by the wealth of carefully presented data and thoughtfully argued work to embark on such (international) projects.