Case innovation and agentive marking: A comparative overview of Central Indo-Aryan

A split-ergative construction had developed during the late MIA period (Bubenik 1998; Peterson 1998) in which subjects of perfective transitive clauses were marked ergative by an oblique form, in contrast with the nominative form for non-ergative subjects. Later in the NIA period, most NIA languages (e.g. Urdu/Hindi) developed a postpositional clitic that was added to the oblique suffix, while others (e.g. Sindhi) continued to mark ergative subjects with a generic oblique suffix. This paper focuses on one exceptional case: the Dehwali language of Gujarat. Dehwali has an ergative marker that is a fusional suffix (i.e. layer I - Masica 1991: 231) and appears to inflect to agree in number and gender with the subject it marks. I will present two possible scenarios as to the origin of the Dehwali ergative marker: that it may be the remnant of an archaic MIA oblique form, or that it may be a more recent innovation as the result of increased contact with neighbouring varieties. Based on theories of grammaticisation, I argue that the former hypothesis is more likely. These theories show that it is not uncommon for oblique case forms (i.e. ablative; genitive) to carry agentive properties.