Constructions and language change (review)

This book is a collection of articles that are concerned with the role that grammatical constructions play in determining or shaping ways in which language changes. It arose from a workshop on the topic at the seventeenth International Conference on Historical Linguistics in 2005, held in Madison, Wisconsin. Rather than providing individuated accounts of the component parts of the book, in this review I outline some of the themes and major issues that cut across the individual contributions. Unequal space devoted to the contributions in the volume should not be viewed as a reflection on the quality of individual articles; I believe each contribution to have significantly furthered our knowledge about various aspects of the mechanisms of, and motivations for, constructional change. As a whole, the book presents a catholic approach to the nature of the construction and its role in change: certainly no one variant of construction grammar is privileged in the book. Since many of the papers are particularly concerned with the relationship between constructional change and grammaticalization, this will be the point of departure for the review.

Adopting a radical construction grammar (Croft 2001) approach, Elizabeth Closs Traugott ('The grammaticalization of NP of NP patterns') analyzes the development of English quantifiers and degree modifiers (such as Modern English a bit or not a jot) that originate in binominal expressions (NP of NP) where the first noun is, for example, Old English bita 'bite' or Middle English schrede 'a portion cut or broken from something'. She stresses that constructions have always been important in grammaticalization research, but that the notion of construction in earlier work is not the same as that used by practitioners of the different variants of construction grammar. In much of the earlier work on grammaticalization, the term 'construction' was used for the syntagmatic string in which grammaticalizing morphemes were contextualized, an issue that the editors of the volume raise in their introduction.

Traugott's article discusses important issues regarding the nature of constructional knowledge and of constructional taxonomies, matters that are further addressed by Mirjam Fried in her article on the development of Old Czech participial adjectives, 'Constructions and constructs: Mapping a shift between predication and attribution'. Like Traugott, Fried observes that language change does not originate in constructions (abstractions) but in constructs (understood as particular instances of use). Particularly, she claims that 'constructions are understood as "blueprints" (generalizations over constructs) and as such presuppose variation and change as an inherent part of grammar' (74). Similarly, in 'Constructional idioms as products of linguistic change: The aan het + infinitive construction in Dutch', Geert Booij focuses on the notion of this particular microconstruction as forming part of a hierarchy. This further refines the theme of constructional taxonomies, their nature, and their role in a synchronic grammar as a product of particular diachronic processes, a theme that runs through the volume. Booij's discussion includes an analysis of the important notion of constructional layering, which may be linked to the three-way division proposed by Traugott of constructional types as micro-, meso-, and macroconstructions. His presentation of aan het + infinitive as part of a constructional inheritance tree (98) can be fruitfully compared to the position taken by Malcolm Ross's article, 'Negative verbal clause constructions in Puyuma: Exploring constructional disharmony'. Ross suggests a network approach to constructional organization. Although Booij's article is not concerned specifically with diachrony, but rather with the role played by constructional idioms in the synchronic grammar, the contribution is nonetheless of great value to those working on language change, since the development of constructional idioms is a characteristic of diachronic lexicalization and grammaticalization (see further Brinton & Traugott 2005). Idiomaticization and its relationship with grammaticalization and lexicalization is a process also addressed by Wallace Chafe's short but sweet article, 'Syntax as a repository of historical relics', on the (quasi)functionality of elements in syntactic structure. Chafe's example (that's gonna...