<?xml version="1.0" standalone="yes"?>
<Paper uid="W97-1413">
  <Title>Constraints on the Use of Language, Gesture and Speech for Multimodal Dialogues</Title>
  <Section position="2" start_page="0" end_page="0" type="concl">
    <SectionTitle>
6. Conclusions
</SectionTitle>
    <Paragraph position="0"> In this paper, we argue on the one hand for a unified account of gesture and perception, and on the other hand for a matching between contrastive conditions required by referential 4 There can be of course many different contexts projected upon a ~iven object, as this depends upon the intention that the speaker wants to convey about it.</Paragraph>
    <Paragraph position="1"> expressions and a pre-existing perceptual contrast. We show that such a contrast has to be localized and to exhibit traces of these contexts through dialogue structures as well as specific orientation properties related to such perceptual contexts.</Paragraph>
  </Section>
class="xml-element"></Paper>