University of Leipzig
THE SEMANTICS-PRAGMATICS INTERFACE: ASPECT IN RUSSIAN AND TURKISH
The multitude of readings of the Russian aspects (particularly the ipf) and the comparison with aspectual usage in the respective English and Turkish equivalents raises several questions: Are those readings an instance of grammatical polysemy or contextually motivated variations of one invariant meaning? Are the different readings to be attributed to the semantics of the aspectual markers? How can we account for the cross-linguistic differences in asptual usage? This paper argues against a purely compositional, and for a more pragmatic account of aspect, one that uses Levinson?s three heuristics, relevance-theoretically modified by assuming that they act on different levels: the I-heuristic elaborates the underspecified semantic representations to yield truth-evaluable propositions, which are taken as input to purely pragmatic inferences governed by the Q- and M-heuristics. Those different inference processes do not follow strictly sequentially one after the other but rather interact in 'parallel mutual adjustment'.This paper assumes a selectional theory of aspect that regards phases and boundaries as the decisive units accesible for aspectual selection by special markers. Due to language-specific selectors it is necessary, to assume a tripartite structure of events.It will be argued that the various readings of aspect can be systematically accounted for by considering these different factors, and that the principles at work can be seen as cross-linguistically valid. Much of the confusion within the aspectual field can be done away with by taking a more pragmatically oriented approach.Of special interest are the socalled 'atemporal' readings of the Russian ipf aspect. Here a closer look at the Turkish 'Aorist' may help to clear the confuse picture. As the Aorist also expresses certain instances of the pf aspect, the question arises, whether the 'atemporal' readings can be dealt with as an instance of aspect at all, or whether there has to be postulated a separate category.