perfective and imperfective aspect; iterative and semelfactive verbs; semantic underspecification

What is it about?

This paper is a contribution to a long-standing discussion related to the domain of aspectual interpretation. More precisely, it focuses on the impact of the degree of specificity and morphological complexity on the time course of processing of perfective (prefixed perfective and semelfactive perfective) and imperfective (simple imperfective and iterative imperfective) verbs in Polish. In two experiments, eye-tracking during reading and self-paced reading, we tested a hypothesis based on Frisson & Pickering (1999), Pickering & Frisson (2001) and Frisson (2009) that the interpretation of semantically underspecified verbs should be delayed to the end of a sentence. As predicted, in both of the reported experiments significantly longer reading measures were observed for aspectually underspecified simple imperfective verbs as compared to aspectually more specific perfective verbs in the sentence-final region. Our second major prediction was that morphological complexity of aspectual forms should cause computational cost directly on the verbal region. As predicted, significantly longer reading times were observed on morphologically complex (prefixed) perfective verbs and (suffixed) semelfactive perfective verbs as compared to their morphologically simple imperfective counterparts in the eye-tracking experiment. This effect was not confirmed in the self-paced reading experiment. This difference between the results obtained in the two reported experiments is attributed to the differences between the methods used.

Featured Image

Why is it important?

Both psycholinguists and theoretical linguists have recently focused on cross-linguistic differences in the domain of aspectual composition (see Bott & Hamm 2014; Husband & Stockall 2014; Filip & Rothstein 2006; Rothstein 2015). What is at the center of this discussion is the question of whether aspectual meaning is computed on a verb in languages in which aspect is grammaticalized and on a VP in languages in which it is not or whether a complete VP is cross-linguistically needed to trigger the derivation of Aspect Phrase where the aspectual meaning is computed. We would like to contribute to this discussion and ask whether the domain of aspectual interpretation in Polish is the same for perfective and imperfective aspect which differ in the degree of their semantic specificity. There are reasons to expect that the interpretation of imperfective aspect is delayed to post-verbal regions (possibly to the end of a sentence) because it is semantically underspecified (see Comrie 1976; Dahl 1985; Batistella 1990; Filip 1993/1999; Klein 1995; Paslawska & von Stechow 2003; Willim 2006). Such a delay in the interpretation of a semantically underspecified imperfective aspect is expected on the basis of the findings of recent eye-tracking studies reported in Frisson & Pickering (1999), Pickering & Frisson (2001) and Frisson (2009), who investigated the role of context in the processing of homonymous and polysemous verbs. They observed that the processor does not select between alternative senses of a semantically underspecified polysemous) verb but rather it initially activates its underspecified meaning and subsequently homes in on the precise sense for the verb. As pointed out in Frisson (2009, 117), the findings of their experiments suggest that “the end of a sentence is a natural choice point to commit oneself to a specific sense of a polysemous verb in the absence of disambiguating information”. Frisson and Pickering (1999) additionally emphasize that the homing-in stage (the time when a specific interpretation is obtained) probably depends on many factors, among them being the requirements of the task (e.g., whether there is time pressure or whether a full understanding of every single word is required), and on the characteristics of the method used (e.g., unlike eye-tracking during reading, self-paced reading does not allow rereading). With these findings in mind, we investigate the impact of the degree of semantic specificity of perfective and imperfective verbs in Polish on the timing of their interpretation.