<?xml version="1.0" standalone="yes"?>
<Paper uid="W98-0313">
  <Title>Discourse Relations versus Discourse Marker Relations</Title>
  <Section position="3" start_page="0" end_page="0" type="intro">
    <SectionTitle>
1 Introduction
</SectionTitle>
    <Paragraph position="0"> The idea that discourse markers (DMs) like then or anyway signal underlying discourse relations (DRs) like cause, opposition, contrast, etc., has been adopted in a certain number of works on text and conversation structure (see Roulet 1985, Martin 1992, Knott 1996 for various examples). In itself, the idea is reasonably intuitive and appealing and seems empirically true to a large extent (Knott 1996).</Paragraph>
    <Paragraph position="1"> However, the linking between DRs and DMs is more intricate than is currently assumed. We show here that some French consequence DMs akin to therefore ( donc, par consequent, alors) are difficult to describe in terms of DRs. We argue that such clashes are due to the semantic profiles of DMs, that is to the way DMs 'see' the left and right argument of the semantic relation they denote. We offer an analysis of the profile of the donc class DMs along the lines of Veltman's update semantics (Veltman, 1996). We conclude that the compatibility of DMs with DRs must be studied by identifying first the relational core of DMs, that is, the semantic relation they denote and the types of arguments selected by this relation) ll.n this paper, we consider only the deductive use of donc, in monologual written speech, a use illustrated for example by Paul opened the tuindow, DONC we 90t some fresh air.</Paragraph>
    <Paragraph position="2"> We ignore here other uses of doric. We will also ignore the</Paragraph>
  </Section>
class="xml-element"></Paper>