<?xml version="1.0" standalone="yes"?>
<Paper uid="E95-1040">
  <Title>The Semantics of Motion</Title>
  <Section position="5" start_page="282" end_page="282" type="concl">
    <SectionTitle>
5 Conclusion
</SectionTitle>
    <Paragraph position="0"> The study and the first results we have here presented cover from lexical semantics to discourse structures, with strong interactions between these two ends. Indeed, lexical informations can be used to disambiguate the structure of the discours, as well as discourse relations can be used to disambiguate lexical entries, as shown in (Asher &amp; Sablayrolles, 1994b). Our work is based on systematic and very detailed linguistic studies which lead to rather complex computations for calculating the spatiotemporal semantics of a motion complex. But we have seen that this level of detail and complexity is necessary if one want to understand, to formalize and to compute a right spatiotemporal semantics for motion complexes. We continue our investigations on two directions: 1. we compare our results with similar works in course of realization on the Basquian language (by Michel Aurnague) and on the Japanese language (by Junichi Saito); 2. we use the results presented here for refining the notions of the Aktionsart, where the structuration of the space in 4 zones can be used to distinguish sub-classes inside the traditional well known classes of aspectual studies. null</Paragraph>
  </Section>
class="xml-element"></Paper>