South Asian relative-correlatives and unexpected particles

Abstract

Crosslinguistically many languages use morphemes from two series of particles — split roughly into additive/conjunctive/universal morphemes (Sinhala -t, Dravidian um, Japanese mo) and interrogative/disjunctive/existential morphemes (Sinhala *da, Dravidian ō, Japanese ka) — in a wide range of ‘quantificational’ contexts. [Cf. Kratzer & Shimoyama 2002; Szabolcsi 2010, 2015; Slade 2011; Mitrović 2014.] Relative-correlative constructions [RCC] in Dravidian and literary/classical varieties of Sinhala (as well as in Mangalore Saraswat Konkani and Dakkhini Urdu, cf. Hock 2016) use a quantifier particle as a ‘closing particle’ in the relative clause. Unexpectedly, the quantifier particle involved in these languages is part of the disjunctive/existential group, as opposed to the case in Nepali, where a member of the additive/universal group appears (as also in Burushaski, Basque, cf. Hock 2005). Thus the use of the particle da in classical/literary Sinhala relative clauses (which often have a free-choice flavour) like (1), Nepali pani in free choice constructions like (2) (cp. also Hindi bhī, cf. Dayal 1996). Since RCCs often have a free-choice interpretation the presence of particles associated with universal quantification like Nepali pani &c. is not unexpected, but the existential-associated particles of Sinhala and Dravidian RCCs are. I provide an analysis which treats existentially-associated particles (Sinhala da, Tamil/Malayalam ō) as variables over choice-functions carrying an anti-singleton presupposition (cf. Alonso-Ovalle & Menéndez-Benito 2010), which accounts for their use in the formation of epistemic indefinites, and can be extended to explain their use in the formation of relative clauses.

Benjamin Slade

Associate Professor of Linguistics

My research interests include formal semantics and syntax, historical linguistics, South Asian and Caribbean languages, and the use of computational concepts in formal linguistics.