<?xml version="1.0" standalone="yes"?>
<Paper uid="W98-0504">
  <Title>r g ! How to define a context-free backbone for DGs: Implementing a DG in the LFG formalism</Title>
  <Section position="7" start_page="117" end_page="117" type="concl">
    <SectionTitle>
6 Conclusion
</SectionTitle>
    <Paragraph position="0"> We have presented a new approach to word order which preserves traditional notions (semantically motivated dependencies, topological fields) while being fully lexicalized and formally precise (BrSker, 1997). Word order domains are sets of partially ordered words associated with words. A word is contained in an order domain of its head, or may float into an order domain of a transitive head, resulting in a discontinuous dependency tree while retaining a projective order domain structure. Restrictions on the floating are expressed in a lexicalized fashion in</Paragraph>
    <Paragraph position="2"> terms of dependency relations. V~re have also shown how the order domains can be used to define a context-free backbone for DG, and used a grammar development environment for annotated phrase-structure grammars to encode the DG.</Paragraph>
    <Paragraph position="3"> A number of questions immediately arise, some of which will hopefully be answered until the time of the workshop. On the theoretical side, this work has argued for a strict separation of precedence and categorial information in LFG (or PSG in general, see (BrSker, 1998a)).</Paragraph>
    <Paragraph position="4"> Can these analyses and insights be transferred? On the practical side, can the conversion we sketched be used to create efficient large-scale DGs? Or will the amount of f-structural indeterminacy introduced by our use of functional uncertainty lead to overly long processing? And, last and most challenging, when will the first large treebank with dependency annotation be available, and will it be derived from XLE's f-structure output?</Paragraph>
  </Section>
class="xml-element"></Paper>