- Competing methods for uncovering linguistic diversity:The case of definite and indefinite articles (Commentary on Davis, Gillon, and Matthewson)*
Davis, Gillon, and Matthewson (DG&M; 2014) argue that ‘hypothesis-driven fieldwork has led to a more complete and accurate picture of linguistic diversity than has been produced by methods that rely largely on extracting information from existing descriptive grammars’ (p. e181). They discuss five cases to support their conclusion. In this response, I discuss only one of these cases, since this case involves two chapters of mine (Dryer 2005a,b, 2013a,b) in the World atlas of language structures (WALS; Haspelmath et al. 2005, Dryer & Haspelmath 2013).1 These chapters deal with definite and indefinite articles among the languages of the world (though for reasons discussed in §6, I use these labels in a considerably broader sense than many would expect). I argue that the research for these two chapters uncovered types of diversity that the methodology employed by DG&M is unlikely to uncover. For reasons of space, here I am only able to summarize some of this diversity, which will be described in more detail in Dryer 2015.2
The primary goal of this response is to illustrate the diversity of articles that has been uncovered by the research behind my WALS chapters on definite and indefinite articles. I first discuss some preliminary issues as well as what appears to be a misunderstanding by DG&M about the methodology behind my WALS chapters (§2). I then discuss a classification of noun phrases in terms of what I call the reference hierarchy, which is necessary background both for my illustrating the diversity uncovered by the research behind my WALS chapters and for explaining my criteria for defining definite and indefinite articles for my purposes in WALS (§3). I present a basic typology of articles in §4, based on the reference hierarchy, and this provides a way of summarizing some of the diversity uncovered by my research. Next, I briefly discuss some more unusual articles (§5), providing more evidence of this diversity, and then explain in detail my use of the terms definite and indefinite for the purposes of my WALS chapters, and why, by my criteria, my classification of articles in the Salish languages appears to be correct (§6). In §7, I discuss the general issue of uncovering diversity, arguing that while there are clear limitations on what my methodology can uncover, there are some aspects of diversity that it is well suited for. Finally, I discuss two general methodological issues that DG&M discuss at length (§8), namely the use of falsifiable hypotheses and the need for negative evidence, and argue that in neither case is there much disagreement. First, I argue that my work also employs falsifiable hypotheses. Second, I concede that my methodology is limited in what it can accomplish precisely because of the lack of [End Page e232] negative evidence, but argue that the evidence presented in the earlier sections illustrates how my methodology is well suited for uncovering certain kinds of diversity. My conclusion is that the optimal way to uncover diversity is to use different methodologies, and that there is no unique way to pursue diversity.
I need to be clear from the start that I in no way question the value of the work that Davis, Gillon, and Matthewson have done in this area. There is no question that there are certain kinds of diversity that detailed examination of a single language can uncover that the methodology I used in these studies, based on extracting information from existing descriptive grammars, cannot uncover. In fact, for few of the over 300 languages I examined for these two WALS chapters are the factors conditioning the use of articles as clear as those governing the use of determiners in Skwxwú7mesh as given by Gillon (2006) and St’át’imcets by Matthewson (1998), other than a number of European languages where the use of definite and indefinite articles is fairly well known independently of the research I did. Furthermore, their work clearly has value in documenting languages, quite apart from issues of uncovering diversity. In...