<?xml version="1.0" standalone="yes"?>
<Paper uid="C82-2057">
  <Title>TOWARDS AND E~PLICIT DESCRIPTION OP POSSESSIVITY</Title>
  <Section position="1" start_page="0" end_page="0" type="abstr">
    <SectionTitle>
TOWARDS AND E~PLICIT DESCRIPTION OP POSSESSIVITY
</SectionTitle>
    <Paragraph position="0"> Any grA---ar must aocou~t for the category of possessivity. As far as we ~now this category is not yet exhaustively described for any language. In this paper we want to demonstrate an explicit approach to this complicated language phenomenon. The functional generative grammar proposed by Sgall and worked out in the group of Faculty of Mathematics and Physics, Charles University, is chosen for this description.</Paragraph>
    <Paragraph position="1"> Distinction between the ontologi~,al notion of possessivlty and the language counterpart of it is~ drawn and distinction between meaning of possessivity and forms of its expression is shown. These distinctions are exemplified on the notion of so-called possessive dative, which is not an expression of possessive meaning. Arguments for establishing, this meaning of dative are shown as based on. ontological criteria.</Paragraph>
    <Paragraph position="2"> Various constructions and expressions used for expressing the possessive meaning are enumerated and their syntactic and semantic character is discussed.</Paragraph>
    <Paragraph position="3"> Three main meanings of the verb %0 have are described: I. full lexlcal meaning - a possessive one connected with active - passive transformation, 2. copulative meaning I connected with he__construction transformation, 3. copulative meanin~ connected with any of these transformations. Some differences between copula to b._._.~e and copulative to have are shown. In the b_e-sentences the subject member of the class is - 232 named in the predicate. In the have-sentences the object as a single element is attached to the subject.</Paragraph>
    <Paragraph position="4"> Similarly different meanings of the verb to belonF~are stated. 1. location, 2. appurtenance, 3. possession. Classification is rejected as meaning of this verb and is shown as a content phenomenon. Other expressions for possessivity as genitives, possessive pronouns and possessive adverbs are partially discussed and compared with results obtained for verbal expressions of poesessivity. Finally a list of importsnt meaning distinctions (e.g. possessivity versus appurtenance) is given and a tentative table of homonyms (and synonymy) of forms of their expressing is outlined.</Paragraph>
    <Paragraph position="5"> In addition some simplifications of the complete linguistic description are shown for applications in question-answering systems and some special treatments adjusted to a given pur-I~Sa are suggested. We suppose that from the point of view of various kinds of applications certain simplifications of the theoretical linguistic conclusions are not only acceptable but in some cases even necessary. It seems for example that in certain sublanguage in a limited area of texts some content regularities are met with such a high probability, that we can accept them in these texts as valid arguments for a change in a demcription of meaning. It seems also better to use in q-a systems the genitive construction as a basic expression of the meaning of possession and appurtenance rather than haveor belon~ to-constructions~ which seem to be the best underlying structure in a general theoretical description of grammar.</Paragraph>
  </Section>
class="xml-element"></Paper>