Jóhanna Barddal: fredagsseminar

The Rise of Dative Substitution in the History of Icelandic

Dative Substitution, i.e. the change from accusative to dative subjects, has been explained in the generative literature as thematic case marking ousting idiosyncratic case marking (cf. Jónsson 2003, Jónsson and Eythórsson 2005). A major anomaly for this account is the late onset of Dative Substitution, not documented in Icelandic texts until the latter part of the 19th century. As the prerequisites for Dative Substitution existed already in Old Norse-Icelandic, the question arises as to why this change did not take place earlier. I show in the present comparative study of Old Norse-Icelandic and Modern Icelandic texts that the semantic structure of the Dative Subject Construction has changed from denoting happenstance and experience-based events equally in Old Norse-Icelandic to a situation where experience-based events are in majority in Modern Icelandic. This change in language use entails that experience-based events are now in the spotlight of the Dative Subject Construction, which in turn makes the construction considerably more coherent semantically in Modern Icelandic, a known precondition for the productivity of argument structure constructions that are low in type frequency (cf. Barðdal 2006a, 2008). This change in the semantic structure of the Dative Subject Construction, i.e. this narrowing and focusing of its semantic scope, is the motivating factor behind the late onset of Dative Substitution in the history of Icelandic. More generally, this investigation illustrates how productivity may increase despite a reduction in the type frequency of a construction, contra claims in the literature that type frequency is the most important factor for productivity (Bybee 1995).