Researching language learning: theories, evidence, claims
So far, we have interpreted applied linguistics very broadly. In our dual purpose- of developing and describing a social realist theory of language use, and of illustrating its relevance to applied linguistics- we draw throughout on examples of applied linguistic 'problems' of various kinds. These problems include questions such as what is meant by 'real' language, or by 'real' English; the political consequences of classifying Ebonics as a language; how linguistic, cultural and national boundaries are interrelated; how the concept of ethnicity is deployed in linguistic practice and in social research, and many others besides. We focus here on another core applied linguistic 'problem', using it to probe questions about epistemology and methodology in a research field which involves a significant proportion of the people who are identified as applied linguists. This problem is the one identified, in an overview of the subject (Schmitt and Celce-Murcia 2002) in a recent introductory volume (Schmitt 2002), as 'the dominant application' from among the 'numerous areas' with which applied linguistics is concerned: 'Traditionally, the primary concern of applied linguistics has been second language acquisition theory, second language pedagogy and the interface between the two' (p.2). Another survey of the subject, entitled comprehensively The Encyclopedic Dictionary of Applied Linguistics, has as its subtitle, 'a handbook for language teaching' (Johnson and Johnson 1999, emphasis added), and the entry here for 'applied linguistics' observes that for a long time the term 'has been used to refer somewhat exclusively to the field of language teaching and learning, rather than to any field where language is a relevant consideration'. Thus, although applied linguistics is clearly expanding its scope, for many people it is still the process of coming to use one or more languages other than the first that is perceived as its core concern.
Accordingly, the goal is to continue to develop our account of social realism by exploring how the kind of empirical research familiar to those involved in the teaching of additional languages relates to the key ideas about structure, culture, agency and emergence which we have introduced so far. In particular, we examine the issue of causality, since the problem of how learners can best be helped to make progress in the language they are learning entails an interest in how pedagogy and outcomes are related. Thus we begin by drawing out some of the implications of our realist position for social research, in particular for research in applied linguistics, and specifically for research in instructed language learning (ILL). This position implies a distinctive view of empirical research, we compare with the epistemological assumptions underlying various forms of ILL research. Such an enterprise raises the troublesome issue of what counts as research in applied linguistics. What sorts of claims can be made about languages, cultures, human agency and social relations? What sorts of evidence are adduced in support of these claims? What count as warrants for these claims? Our aim here is not to provide a comprehensive overview of this very extensive field, but rather to identify some of the persistent issues in ILL research, and to highlight the influence on them of how different research traditions have understood the process of doing research.
It is necessary at this point to include a note about terminology. For want of an unambiguous label for the area of applied linguistics, we have adopted the phrase 'instructed language learning' research, which Mitchell and Myles (2001) link to Ellis (1990, 1994).1 We are referring to that substantial enterprise which investigates the detail of institution-based instruction in languages which students are learning after they have become speakers of one or more first languages in infancy. This choice of term, in preference to the broader 'Second Language Acquisition', aims to accommodate the fact that many SLA scholars explicitly distance themselves from pedagogy and practice, identifying 'the task of SLA research' as not particularly to evaluate teaching strategies, but rather 'to determine whether or not learners have access to universal principles of language and to determine what the nature of grammatical knowledge is' (Gass 2000: 53). We will develop more fully our proposals for a realist approach to research, taking account of the fact that while the field of instructed language learning is a very significant one in applied linguistics, it is far from the only one.
