<?xml version="1.0" standalone="yes"?>
<Paper uid="J93-2001">
  <Title>Using Register-Diversified Corpora for General Language Studies</Title>
  <Section position="6" start_page="239" end_page="239" type="concl">
    <SectionTitle>
5. Conclusion
</SectionTitle>
    <Paragraph position="0"> In the present paper I have presented evidence from several different structural levels, as well as different languages, showing that there are important and systematic linguistic differences among registers. These data are used to argue for the general point that corpora representing a broad range of register variation are required as the basis for general language studies. In fact, the extent of register differences reported here suggests that overall linguistic characterizations of a language are often inadequate (or even incorrect). That is, since overall generalizations represent a kind of averaging of the linguistic patterns in a language, they often do not accurately represent the actual patterns of any register; in fact, such generalizations can conceal the systematic patterns found across registers. An alternative approach is to recognize the centrality of register differences and work toward a composite linguistic description of a language in those terms.</Paragraph>
    <Paragraph position="1"> Corpus-based analyses of register variation obviously need to be extended in several ways. Future research should be based on larger corpora and include a wider representation of linguistic features and registers. In addition, register distinctions can be made at several levels of abstraction, and the intersection of register and text type analyses needs to be further explored (see notes 6 and 7). The analyses summarized here, though, clearly show the importance of a register perspective, supporting the continuing development and analysis of large register-diversified corpora as the basis for general language studies.</Paragraph>
  </Section>
class="xml-element"></Paper>