Introduction

Q_Lex is an exploratory program that assesses how easy it is for an L2 learner to access a small set of high-frequency words. The program is basically a word recognition task, but it uses an innovative methodology that gets round some of the problems that arise in standard L2 word recognition tasks.

Around the time that the first versions of Q_Lex emerged, most researchers who thought about vocabulary development thought in terms of a single ‘continuum’ that ranged from passive/receptive vocabulary at one end to active/productive vocabulary at the other. It was never exactly clear how the various stages on this continuum were actually defined, or what caused words to move along the continuum. In spite of this, the metaphor was widely used in the vocabulary research of the 1980s (e.g. Crow, 1986; Melka Teichroew, 1982, 1989). The emphasis in this work was very much concerned with the individual words a learner knew, and where these words fell on the ‘continuum’, and there was very little interest in describing the larger, global properties of learners’ vocabularies, and what these properties enabled learners to do.

However, a few of us were beginning to think about whether it might be possible to go beyond the properties of individual words, and describe learners’ lexical competence in terms of a small number of dimensions that were more or less independent of each other. One dimension was obviously going to be vocabulary size, and we thought at the time that we more or less had this dimension sewn up: Paul Nation’s Levels Test and Paul Meara’s Yes/No tests seemed capable of generating workable measures of how big learners’ vocabularies were, and it had been shown that these measures correlated moderately well with other aspects of proficiency. The question that we were interested in at this time was whether vocabularies could also be characterised by other dimensional measures, and if they could, what these additional measures would be.