“Come here” in Teiwa (Papuan)

Marian Klamer

In my presentation I would like to discuss the verb ma 'come (here)' in Teiwa, a Papuan language spoken on the island of Pantar, just north of Timor island in Indonesia. Teiwa is a head-final language (e.g. OV order, and final negations) that makes extensive use of serial verb constructions. The serial verb constructions often use direction verbs, one of which is the directional verb ma 'to come (here; towards the speaker)'. Ma is descriptively characterised as having developed a number of additional functions, including the following:
(i)     It marks a set of events as being in temporal sequence, as in:
    People catch goat, people go do finish, come stable come put =  The people catch a     goat, they go and then put it in a stable.
(ii)     It marks hortative and irrealis modality, as in:
    Come me-talk! = Talk to me!; I come talk? = Shall I tell (it)?
(iii)     It functions as an adposition-like item when it introduces an adjunct of e.g. location, as     in: Child wood close come stand = A child stands close to the tree
(iv)     It introduces a change-of-state nominal predicate, as in:
    Bamboo not, it come sword = It's no longer a bamboo, it became a sword.
The question I will address is how we can relate the grammaticalised functions of ma to its (original) verbal use. I propose that ma should be analysed as being subcategorised for a single argument that is either [+animate] or [-animate]. Depending on the animacy of its argument, the verb then sets off into two distinct grammaticalisation directions. I argue that the grammaticalised functions (i) and (ii) are derived from ma with an animate argument, while the functions (iii) and (iv) are derived from ma with an inanimate argument. A general conclusion will be that the feature animacy appears to be crucial to account for the morpho-syntactic patterning of arguments and adjuncts in Teiwa, while the thematic (semantic) roles of arguments are largely irrelevant.