Published online by Cambridge University Press: 01 February 2019
Dutch partitive constructions of the type one of the (few) X who Y show a striking pattern of singular subject-verb agreement in their relative clause. This paper presents a corpus study showing that the prescriptively “incorrect” singular agreement pattern is in fact the dominant pattern in Dutch. In order to explain this, we argue that this type of partitive construction often has a specific function in context, namely, to point out that the subject is special or extraordinary, usually for the reason presented by the relative clause. We apply a usage-based approach to this construction within the framework of Construction Grammar, arguing that the prevalent implicature of the subject’s specialness has become a conventionalized part of the meaning of the construction. This analysis then can be used to explain the syntactic pattern of singular agreement within the relative clause. A similar albeit less pronounced pattern can be found in German.*
We are most grateful to Roeland van Hout for his help with the statistical analyses of the data, and to Lukas Reinarz, who collected and annotated the German data. We would like to thank Sebastian Collin, Geertje van Bergen, Ad Foolen, Johan Kobben, Sander Lestrade, Vera van Mulken, Kees de Schepper, Wessel Stoop, Peter de Swart, Thijs Trompenaars, Ruti Vardi, Puck Wildschut, and Martine Zwets for helpful discussions and comments. A special word of thanks goes to Johan Kobben and Vera van Mulken for translating the Dutch fragments. Versions of this paper were presented at the conferences Structural alternations: Speaker and hearer perspectives (Groningen, August 2011), TIN-dag 2012 (Utrecht, February 2012), and New ways of analyzing syntactic variation (Nijmegen, November 2012). We thank the audiences for their questions and comments. Finally, we thank two anonymous reviewers for their helpful comments on an earlier version of this paper.