Karsten Schmidtke-Bode, Natalia Levshina, Susanne Maria Michaelis, Ilja A. Seržant (Eds.).Explanation in Typology. Diachronic Sources, Functional Motivations and the Nature of the Evidence

The present volume addresses a foundational issue in linguistic typology, viz. crosslinguistic generalizations and how they should be explained. More precisely, it discusses to what extent statistical universals of linguistic structure can be accounted for by so-called “source-oriented”, mutational explanations, i. e. constraints on language change, or by “result-oriented”, functional-adaptive principles, i. e. forces leading to preferred structural types that facilitate communication and language processing for speakers and hearers. Unlike previous volumes in a similar vein (e. g. Hawkins 1988; Good 2008), the debate in this volume takes place entirely within the functional, usage-based camp. Even so, Explanation in Typology is a major contribution to typology and general linguistics and will be of great interest to any linguist keen to engage with the nature of linguistic universals, regardless of their theoretical position.

After the table of contents, the book starts with an introduction by Karsten Schmidtke-Bode outlining the aim and structure of the volume and briefly summarizing each of the subsequent chapters. This is followed by nine research papers by Martin Haspelmath, Sonia Cristofaro, Jeremy Collins, Matthew S. Dryer, Holger Diessel, Karsten Schmidtke-Bode, Ilja A. Seržant, Susanne Maria Michaelis, and Natalia Levshina, as well as a concluding chapter by Karsten Schmidtke-Bode and Eitan Grossman. Separate indices of names, languages, and subjects round off the book. The volume can be downloaded for free at https://langsci-press.org/catalog/book/220. Alternatively, each chapter can be downloaded separately. Print-on-demand options are available, too.

The first research article is a programmatic paper in which Martin Haspelmath addresses some terminological issues and provides an overview of causal factors for linguistic universals, including functional-adaptive and mutational constraints. Crucially, both explanations involve language change, but in the former, change only serves to fulfill the functional adaptation, whereas in the latter, change itself is the causal factor. Haspelmath acknowledges the importance of mutational constraints in certain cases, but generally favors functional-adaptive explanations.

The subsequent research articles all respond to Haspelmath’s paper in some way. Sonia Cristofaro and Jeremy Collins both argue strongly in favor of a source-oriented approach in which language change is the primary explanatory factor for linguistic universals. For example, Cristofaro claims that synchronic crosslinguistic distributions of accusative vs. ergative case marking and of zero vs. overt marking across singular and plural forms can largely be explained by their historical origins. Similarly, Collins argues that many word-order correlations are best explained in terms of grammaticalization and that historical relatedness between some of these correlations should be taken as an additional confounding variable in typological studies before attempting to motivate them in functional-adaptive terms.

Subsequent chapters argue progressively more in favor of functional-adaptive motivations. Matthew S. Dryer and Holger Diessel take a middle-ground position. Dryer believes that grammaticalization plays an important role in explaining many word-order correlations (e. g. between the order of verb and object and the order of adposition and noun), but cannot explain all of them (e. g. the correlation between the order of verb and object and the order of noun and genitive). Moreover, according to Dryer the source-oriented approach makes wrong predictions in certain cases: even though demonstratives are the most common source of definiteness markers, their positional distributions are complementary. Whereas demonstratives follow the noun more often in VO languages than in OV languages, definite markers tend to precede the noun in VO languages (p. 85). Similarly, Diessel argues that positional patterns of adverbial clauses can be explained by general functional-adaptive and cognitive processes, while the positional tendencies of the subordinators that occur in such clauses are probably a by-product of grammaticalization.

Karsten Schmidtke-Bode’s and Ilja A. Seržant’s articles are concerned with cases in which universal functional-adaptive principles appear to be operative, but are mitigated or even overridden by language-specific, historical factors. Schmidtke-Bode shows how principles of Hawkins’s (2004; 2014) “processing typology” may interact with diachronic developments. Furthermore, he tests Hawkins’s (2014) prediction that OV languages tend not to develop articles. Although a statistical tendency in favor of this prediction, none of the OV language families in his sample show a significant bias towards or against articles (using the Family Bias method, cf. Bickel 2011, Bickel 2013). Schmidtke-Bode therefore concludes that Hawkins’s prediction is “plausible, but currently rather weakly substantiated” (p. 139). Seržant argues that not all universal forces are strong enough to overrule language-specific factors in a wide variety of languages. An example of such a weak universal force is the discriminatory function of case in differential object marking. According to Seržant, this is a universal force, appearing throughout the historical development of Russian case marking as well as in several other unrelated languages, yet it is often mitigated or overridden by other, language-specific factors and therefore not very prominent in a synchronic sample of languages.

The final two research papers argue in favor of the result-oriented, functional-adaptive approach. Susanne Maria Michaelis investigates the universal that independent possessive person-forms (e. g. English mine, yours, hers) are never shorter than the corresponding adnominal possessive person-forms (e. g. my, you, her). She argues that the latter are more frequently used than the former; hence they are more predictable and therefore shorter (cf. Haspelmath 2008 for this line of reasoning). She favors this explanation over a mutational one because this structural pattern is also found in creole languages, i. e. recently developed languages with accelerated grammaticalization processes. Natalia Levshina reports on a language-learning experiment in which she found that participants tend to use shorter causative forms in a fictional alien language to describe more frequent events. This suggests a very strong functional-adaptive constraint, since it emerged within the first generation of speakers.

Finally, the epilogue by Karsten Schmidtke-Bode and Eitan Grossman provides a critical assessment of the discussion in the previous research articles and an outlook on future research. They are slightly more critical of the source-oriented approach, e. g. by pointing out that little to no historical data is available for the vast majority of languages, which is why we cannot know that mutational constraints actually provide the best explanations for synchronic patterns.

Overall, Explanation in Typology provides an intriguing discussion by several renowned researchers on a foundational theoretical issue in linguistic typology. The individual papers present new interesting findings and can be read separately, but since they refer to each other quite often, reading the entire book may be more rewarding. While the volume is obviously aimed at typologists, most articles are also accessible for linguists in other fields. Schmidtke-Bode’s paper, for example, gives a concise but clear introduction into the most important aspects of Hawkins’s (2004; 2014) complex “processing typology”. Technical terms are generally explained on the spot; a rare exception is “flagging”, an umbrella term for case and adpositional marking (Haspelmath 2019), in Seržant’s article.

Despite the lively debate between their respective supporters throughout the volume, the reader may be left wondering how mutually exclusive mutational and functional-adaptive constraints on linguistic universals really are. Such suspicions are confirmed by Schmidtke-Bode and Grossman, who conclude that most typological phenomena are probably best explained by a nuanced mixture of both (p. 234). There is thus a lot of empirical typological work to be done. Most of the chapters are concerned with classic topics such as word-order correlations (Collins, Dryer, Diessel, Schmidtke-Bode) and case marking (Cristofaro, Seržant). Michaelis’s article on possessive person-forms and Levshina’s article on causatives in a fictional language introduce some much-welcomed variety in the empirical domains under discussion. The publication of this volume will hopefully inspire other typologists working on other phenomena to contribute to the discussion.