Learning and Teaching through Context - A Data-driven Approach
by Ramesh Krishnamurthy

COBUILD (Collins Dictionaries and the University of Birmingham)

This article is based on a talk given at the TESOL Spain 24th National Convention, "A Quest for Teacher Development", Seville, 30th March-April 1st 2001, and was first published in theTESOL Spain Newsletter, Volume 24, Autumn/Winter 2001, ISSN 1575-8044. It is reproduced here with their kind permission.

Abstract

A word may have many potential meanings, but its actual meaning in any authentic written or spoken text is determined by its context: its collocations, structural patterns, and pragmatic functions. Large language corpora offer access to words in a wide range of natural contexts, which can improve and enrich both language learning and teaching.

Introduction

The conference theme is teacher development. So how do teachers develop? Mainly by experimentation and experience. But teachers can also develop by adopting more active strategies, acquiring new knowledge through reading and by conducting their own research, and acquiring new skills by observing colleagues, by attending courses, and of course by thinking and reflecting on their own experiences. Above all, teacher development is a question of attitude. Good teachers (and good students, for that matter) need to be constantly open to new ideas and new opportunities, for example through technological developments.
This paper will focus on one of these technological developments in particular, and show how teachers can use the newly available resources and techniques to supplement and enhance their current teaching practices.

"The single most important task facing language learners is acquiring a sufficiently large vocabulary." (Lewis, 2000:8). For centuries, one major resource for this task has been dictionaries. Dictionaries depend on a simple linguistic fact: words can be easily listed in alphabetical order. Meanings arise when words are used in context, but meanings are too complex to list, so lexicographers listed words and pretended that meanings were associated with them, as a form of shorthand. However, at some point we seem to have forgotten that this is just a conventionalised pretence, and have come to believe that meaning actually is contained in individual words.

Some people talk more accurately about words having potential meanings. Their actual meaning in any authentic written or spoken text is determined by their context: their collocations, structural patterns, and pragmatic functions. If meanings arise from words in contexts, learners will need contexts in order to learn the language. Where can we get these contexts from? We can make them up (as most lexicographers, linguists and teachers did in the past) but, because our memories and intuitions are often inaccurate and incomplete, we usually make up contexts that are inaccurate and incomplete. As John Sinclair has said:
"language users cannot accurately report language usage, even their own" (Sinclair, 1987), and "There are many facts about language that cannot be discovered by just thinking about it, or even reading and listening very intently" (Sinclair, 1995).

Language corpora are collections of authentic written and spoken texts, created in genuine communicative situations, and provide us with many real contexts. Normally, students encounter a particular word, phrase, or grammatical structure on one day, then have to wait for a long period before they come across it again.

Corpora allow students to see many examples of the word, phrase, or grammatical structure at the same time. They also notice patterns of usage and work out rules for themselves, and therefore remember them better.

Many language corpora have become available to the public in the past few years, e.g. for English there are the British National Corpus, and the COBUILD Bank of English corpus. It is also fairly easy to collect smaller corpora oneself, from texts available on the World Wide Web. Students who write their essays on computer can collect a corpus of their own work. Teachers can collect their students' work to form a class corpus. Students can compare their own output with native-speaker texts in the public corpora, improve their own writing, and increase their language awareness.

Public corpora usually provide their own software tools, but a variety of programs are also available separately for use with user-collected data. This paper refers only to English, but corpora are now available for many other languages. Some websites with information about corpora, software, and corpus analysis techniques can be found at the end of this paper.

The data shown below is taken from the Bank of English corpus, but the main facts and figures for most words will be similar whichever large corpus you use. Most corpus software will provide a similar set of standard facilities. Each of these can be used to focus on different aspects of language learning and teaching.
Only two such facilities will be covered here: frequency and concordance, but there are many other facilities such as various collocation tools, and grammatical part-of-speech tags.

To page 2 of 3

Back to the articles index