Micro-change and macro-change in diachronic syntax ed. by Éric Mathieu, Robert Truswell (review)

The question of how and why languages change has intrigued linguists for a long time. One of the earliest attempts to address the issue scientifically was made by the Neogrammarians, who, in the 1870s, influentially argued that phonological changes, insofar as they apply mechanically, hold exceptionless (‘Aller lautwandel, soweit er mechanisch vor sich geht, vollzieht sich nach ausnahmslosen gesetzen’; Osthoff & Brugmann 1878:xiii). No such regularity and thoroughgoingness were credited to syntactic change, which was rather seen as an indicator of system-internal weaknesses. As Hermann Paul (1920 [1880]:251, §173), one of the foremost Neogrammarian authorities, put it boldly, ‘there is in language no precaution at all against the imperfections [Übelstände] that penetrate it, but only a reaction against those already present’ (my translation from the German). Paul’s statement sounds surprisingly controversial in the wake of the minimalist program and its foundational hypothesis that the human language faculty is ‘an optimal solution to minimal design specifications’ (Chomsky 2001:1). If the strong minimalist thesis were correct and sentence structures are built incrementally for effective use at the interfaces, why is it that the syntax itself is prone to change, at times with drastic consequences for the grammatical system at large?

Over the past decades, much headway has been made in further understanding the logical problem of language change by integrating the conceptual and analytical tools of generative grammar into historical syntax research. Bridging synchrony and diachrony has been the raison d’être of the Diachronic Generative Syntax (DiGS) conference, which is widely recognized as the main international platform for the formal linguistic study of historical grammar change. The chapters of the volume under review here were first presented at the fifteenth meeting of DiGS at the University of Ottawa, August 2013—four years before their publication in the ‘Oxford studies in diachronic and historical linguistics’ series. The book’s title, Micro-change and macro-change in diachronic syntax (MiCMaCDS), is a tad misleading, as the ongoing debate on micro- vs. macro-parametric variation does not play a prominent role, even though questions about diachronic parametric differences between language stages are dealt with.

In the introductory chapter, the editors Éric Mathieu and Robert Truswell bring up some interesting conceptual and methodological issues. While one may agree with the observation that ‘modern syntactic theory gives us so little to work with’, it does not follow from it that ‘all syntactic change must ultimately reduce to lexical change’ (1). The version of diachronic minimalism presented here raises a concern about its aptness to capture the complexity of syntax change in a descriptively and explanatorily adequate way. Alternatively, one may seriously entertain the hypothesis that the syntax can change endogenously, without interface pressures from the lexicon and the morphology playing any decisive role (Reintges 2009).

Ailís Cournane sets out to revalidate and reposition the acquisitional perspective on historical grammar change, first articulated in Lightfoot’s (1979) groundbreaking work. From kindergarten age onward, the acquisition process is no longer restricted to the family, but also involves peer-to-peer learning. This allows for the possibility that input-divergent analyses survive. The child-innovator approach to grammatical change is not supported by the parallels between developmental patterns in acquisition and diachronic pathways, nor does the absence thereof necessarily falsify it. We are therefore left with a still unproven working hypothesis, albeit a widely accepted one.

The editors did not wish to group the individual chapters of MiCMaCDS into thematic sections, as the reader might expect, insisting that the topics addressed are too intertwined, so that ‘any attempt to draw boundaries just leads to artificiality’ (4). However, it is the present reviewer’s impression that chapters cluster together in both subject matter and approach.

The book includes four methodologically oriented chapters, which use mathematical and statistical tools and...