Corpora: "language engineering": corpus evidence

Date: Fri Feb 18 2000 - 00:14:08 MET

  • Next message: Tim Buckwalter: "RE: Corpora: language engineering"

    Dear Tadeusz and Eric,
    "Language engineering" is *very* rare in general usage outside the linguistics
    community. The Bank of English has only two examples in its 328 million words
    of 1990s texts. However, the first, from the Courier Mail (Brisbane),
    Queensland Newspapers Pty Ltd, of Tue 21 Feb 1995, is used (in what I think is
    probably the "lay" meaning, i.e. as a synonym for "language planning", maybe
    by analogy with "genetic engineering", etc):
    "February 16 Straitjacket for language THE people who are promoting language
    engineering have obviously stopped taking their tablets. Paranoid about
    eliminating sexism and installing gender homogenisation, they are on a futile
    mission to legislate or bureaucratise language into a straitjacket."

    The second example, from the INK newsletter (I don't have the details, but
    a linguistic services group of some kind), uses the term in the more
    technical sense:
    "The second meeting of the International Industry
    Translation Standards Association (IITSA), a recently-formed group initially
    organized by INK International, was hosted by IBM Denmark and brought together
    a growing range of participants from both supply and demand sides of the
    documentation services business. Devoted to the exchange of ideas and
    experience on reducing costs within the multiple-language documentation
    process, IITSA aims to provide a platform for creating industry-wide
    terminology and language technology standards and for developing quality
    training programs. <p> <p> Evidence that IITSA is finding significant echoes
    throughout the public and private sectors was the presence of a Commission of
    the European Communities (CEC) official with responsibilities for the Language
    Engineering Action Program (LEAP)."

    Although the evidence is so scanty, it does seem to point to two meanings,
    one technical and one non-technical. Obviously corpora of a few hundred
    million words are still too small to investigate many features of language.
    Let us hope that corpora of billions of words are not too far away!
    Best wishes

    Ramesh Krishnamurthy
    Honorary Research Fellow
    Corpus Research Group
    University of Birmingham

    This archive was generated by hypermail 2b29 : Fri Feb 18 2000 - 00:12:24 MET