<?xml version="1.0" standalone="yes"?>
<Paper uid="J88-4002">
  <Title>NATURAL LANGUAGE QUERYING OF HISTORICAL DATABASES</Title>
  <Section position="20" start_page="0" end_page="0" type="relat">
    <SectionTitle>
3.6 RELATED WORK
3.6.1 INTRODUCTION
</SectionTitle>
    <Paragraph position="0"> Two common threads run through much of the recent work on formalizing a theory of questions. The first is the idea that all questions should be defined so as to denote objects of the same type. Generally, this has meant propositions or sets of propositions, but it seems that even before the choice of just what questions denote was made, this &amp;quot;single semantics&amp;quot; viewpoint had been adopted. The other, as we have already pointed out, is that some account of the answer(s) to a question should be included at least as a component of its semantics. When combined with other factors these two biases have led to somewhat different results. Thus Hamblin (1973) suggests that a question denotes the set of all &amp;quot;propositions that count as answers to it&amp;quot;; Karttunen (1977), &amp;quot;the set of propositions expressed by\[its\] true answers&amp;quot;; Bennett (1977, 1979) and Belnap (1982), who worked with Bennett on the theory, &amp;quot;sets of open propositions:functions from sequences of individuals to propositions.&amp;quot;</Paragraph>
  </Section>
class="xml-element"></Paper>