Corpora: corpora in the ESL classroom

Date: Thu Nov 08 2001 - 19:25:20 MET

  • Next message: LITTLECHILD Peter: "Re: Corpora: History of plagiarism"

    In a message dated 11/1/01 3:36:58 PM Central Standard Time, writes:

    > But the question is, whose interpretation? Whose intuitions?
    > TEFL Teachers, please tell: do students learn better if presented
    > with A, or with B? -
    > A) Pre-sorted sets of concordance lines (maybe with carefully
    > crafted explanations already attached), or
    > B) Unorganized concordances, which they have to wrestle through,
    > forming their own hypotheses and imposing their own order.
    > Obviously A is quicker -- but, if time is not a problem, is B more

          I find this an interesting question, especially since I am writing a
    dissertation now using Cobuild's US Books corpus to analyze sense extensions
    of the verb HOLD via basic level images, image schematic structures,
    metaphor, and metonymy. As a researcher and native English speaker, I have
    'wrestled through' the corpus, attempting to form my own hypotheses and to
    impose my own order on the mish mash. It is quite a challenging undertaking.
     I teach university level ESL and would not impose this task upon my
    students. If it's hard for me, I can imagine that it would be well nigh
    impossible for them.
          After fumbling towards a solution to the question of how best to carve
    up the corpus and to uncover some kind of category structure, I am now
    thinking in terms of domains for my own research, and wonder if this wouldn't
    be a workable approach for ESL students as well. What I think would be most
    interesting to do as an instructor would be to first identify domains that
    are participating in the word sense distribution of the target word. For
    example, with HOLD, I see a clear grouping of senses within what I would term
    a cognitive domain. These are senses that refer to 'holding a belief',
    'holding a judgment', a tradition that one 'holds dear' and so forth. I see
    another clear grouping of senses associated with people coming together to
    accomplish some task. These are uses that refer to 'holding a meeting',
    'holding a seminar', holding a demonstration', and so on. I am calling this
    a communication domain, for want of a better term right now.
          At any rate, suppose students were given a simple but thorough
    explanation of the meaning of a domain - background knowledge that you can
    use to understand a word; areas of human experience, etc. with several clear
    examples. Following this, they could be presented with a small corpus,
    perhaps 20-25 examples, and asked to identify domain groupings. As an
    instructor, my thinking would be to arrange the corpus such that no more than
    three very obvious domains are represented across, say, 21 examples (7 per
    domain). Depending upon how structured the instructor wishes the lesson to
    be, he or she could propose a possible list of domains for the target word,
    or, with more time, ask students to identify their own domains. The second
    task has the potential for uncovering fascinating construals based on
    cultural differences. Some of the construals, I would be willing to bet,
    would also be based on misreading stemming from low language proficiency.
    That, too, would be useful to uncover and discuss. The instructor would
    already have his or her analyses on hand, and could use them as a reference
    point for comparison/evaluation of the students' analyses.
          The key, I think, would be to keep the corpus small, and to give
    students a clear set of criteria for their groupings. Domains are one idea.
    I'm sure other teachers could think of more.
    Paula Bramante
    University of New Mexico

    This archive was generated by hypermail 2b29 : Thu Nov 08 2001 - 19:35:51 MET