The first talk in our McGill Linguistics Colloquium Series will be given by Lefteris Paparounas (UQAM), next Friday, October 2nd, at 3:30pm in Arts Building, room 150. The details of the talk are given below.
Building verbal reflexives in Turkish and beyond
Many languages derive reflexive interpretations from ostensibly intransitive verbal forms (e.g. English ‘John bathed’; compare ‘John bathed himself’). Constructions of this general sort – verbal reflexives – raise important questions on the mapping between the syntax and the interpretation of arguments: what is the arity of these verbs; where does their surface subject originate; how does the reflexive reading arise; and what sort of reflexive reading is it?
This talk will begin by arguing, based on joint work with Faruk Akkuş, that Turkish verbal reflexives provide an unusually clear window onto these questions. We will first see that Turkish verbal reflexives are syntactically intransitive and interpretively monadic: their surface subject is their only argument. Diagnosing the derivational history of this argument, we will find a striking instance of mixed behavior. The surface subject of verbal reflexives originates as a deep object; this much brings Turkish in line with the many languages that assign to verbal reflexives the same basic syntax (and morphology) as passives and unaccusatives. Crucially, however, the surface subject of verbal reflexives transits through an intermediate VoiceP-internal position on its way to becoming the grammatical subject; in this respect it is like the subject of transitives and unergatives, and unlike that of unaccusatives and passives. Turkish thus clearly evidences a movement-based derivation of verbal reflexivity, whereby the identification of thematic roles reduces to movement between the positions associated with these roles.
Time permitting, we will ask whether this derivation should be understood as *the* analysis of verbal reflexivity across languages. A set of focused cross-linguistic comparisons informed by the Turkish findings will suggest a negative answer, reinforcing the conclusion that notional reflexivity does not receive a unique encoding in the grammar.
~~