In a post called “Bridging the Gap” (see under the SLA section of pages) I summarised an article which addressed the question of how to bridge the gap between two different styles of research into second language learning and teaching. On one side are those researching linguistic-cognitive issues, using quantitative research methods and statistical analysis, and on the other side are researchers working on the basis of sociocultural or sociocognitive views, using qualitative research methods including case studies and ethnography. Despite the dramatic claim that the whole community is in danger of disintegrating, no post on this blog has ever aroused as little interest as this one: just about nobody read it, nobody “liked” it, and only Mark left a comment. So rather than comment on the differences between two different styles of research, I’d like to comment on differences between the achievements of the two sides.
First, cognitive research in SLA. Learning English as a second language is usually seen as involving the acquisition of two kinds of knowledge: declarative knowledge, which involves knowing that something is the case – that Paris is the capital of France, for example – and procedural knowledge, which involves knowing how to do something – how to swim, for example. A similar distinction is made between explicit and implicit learning. A “traditional” information-processing model of SLA suggests that first you learn declarative knowledge (the past of “go” is “went”; “see you later” means “goodbye”) through attention-demanding controlled processes, and then, through practice, you transform it into implicit (procedural) knowledge. The teaching implication of this model is presentation and practice. A more recent view of SLA sees it as a process involving the development and restructuring of learners’ mental representation of the target language: their interlanguage. The construct “interlanguage” is used in a theory which sees explicit and implicit learning differently, and answers the question “Why don’t learners learn what teachers teach?” The answer is that interlanguage development involves the acquisition of the L2 in a more or less fixed order, which is impervious to instruction. Many would say that the shift from presentation and practice to a more task-based approach to ELT represents progress, and that it’s the result of a better understanding of SLA.
Regardless of what methodology they adopt, or what epistemological views they hold, how have researchers working on the basis of sociocultural or sociocognitive views contributed to improving our understanding of second language learning and teaching?
The first candidate is “the ethnography of communication”, which studies “the social roles of languages, in structuring the identities of individuals and the culture of entire communities and societies.” (Mitchell and Myles (1998: 164). Examples of studies of speech events are phone conversations, shopping, and job interviews, and themes dealt with include “gatekeeping and power relations in L2 communication”, and “speakers’ social identity, face and self-esteem”. I know of nothing in this area which has helped explain SLA, or helped teachers. Nothing.
How about variable competence models? These take a sociolinguistic approach to SLA, abandoning Chomsky’s distinction between competence and performance and viewing competence as variable, not homogenous. I won’t bother to outline these theories here; it’s enough to note that they both make use of the construct of interlanguage without providing any explanation of the acquisition of linguistic knowledge. By erasing the distinction between competence and performance, “the variabilist is committed to the unprincipled collection of an uncontrolled mass of data” (Gregg 1990: 378). What’s interesting here (in light of the “Bridging the Gap” article) is that Tarone (1990), in reply to Gregg, labels Gregg’s approach “rationalist”, and complains that “such scholars, perhaps motivated by “physics envy”, are trying to turn the study of language into an exact science” (Tarone, 1990: 395).
Just as bad, IMHO is Schumann’s (1978) Acculturalisation/Pidginisation approach, which claims that SLA is “just one aspect of acculturation” and that the more a learner acculturates to the target language group, the better he will acquire the second language”. The two essential problems of this approach are the use of ill-defined key constructs (social and psychological distance), and the unwarranted, and unsupported assumption that L2 users make use of a “simplified grammar”. It has led nowhere.
Finally, attempts to explain “incompleteness” in SLA investigate, among other things, aptitude and motivation. Sawyer and Ranta (2001) suggest that “the clearest fact about SLA that we currently have” is that L2 learners “differ dramatically in their rates of acquisition and in their ultimate attainment” (Sawyer and Ranta, 2001: 319). Unfortunately, as Sawyer and Ranta admit, despite its importance, L2 research into the sources of individual differences has lagged far behind research in other areas. The problem is partly due, as Sawyer and Ranta say, to the reliance on correlational research designs, and partly to the inherent difficulty of finding reliable and valid measures of the traits examined. Saywer and Ranta (2001) have attempted to revitalise Carroll’s work (1974) on aptitude, and Dörnyei and Ushioda (2009) have done something to rectify the problems with Gardner’s work (1985) on motivation, but, yet again, I suggest that progress is very slow.
What some contributors to the Bridging the Gap argued was that the two sides should get together, but it’s a tall order. DeKeyser says that the combination of longitudinal research and mixed-methods research “is largely unheard of” and most acknowledge a fair degree of incomensurability. So why bother? I’m being deliberately provocative; I can see the value of case studies and of small-scale studies (like those done by MA students) using qualitative methods to help validity through so-called triangulation, but I honestly can’t see the value of most of the work done by those in the sociocultural or sociocognitive camp. I suggest that those researchers working in the area of psycholinguistics who take what Hulstijn and I, following Popper, refer to as a critical rationalist approach really don’t need to work with those trying to articulate how members of “a thought collective” are affected by different “thought styles”.
Just in case I’ve given the impression that those taking a cognitive-linguistic approach are all working nicely together, I should quickly say that there is, of course, an awful lot of disagreement within the camp. While most accept that teaching can affect rate but not route, some say that explicit learning plays a major role while others say it plays an extremely limited role; some say it’s a process where controlled processes become automatic through practice, while others say you just get better at accessing declarative knowledge; some say you just need lots of input, others say you need output too; some say L2 adult learners have access to UG, some say they don’t; some say language is learned in a special way, others say it can be explained by general learning theories. But they at least agree on the best way to do their research, namely by testing hypotheses through the use of empirical data.
Dörnyei, Z., & Ushioda, E. (Eds.) (2009) Motivation, language identity and the L2 self. Bristol: Multilingual Matters.
Mitchell, R. and Myles, F. (1998) Second Language Learning Theories. London: Arnold.
Sawyer, M., and Ranta, L. (2001) Aptitude, individual differences, and instructional design. In Robinson, P. Cognition and Second Language Instruction. Cambridge: CUP.
Schumann, J. (1978) The pidginization process: A model for SLA. Rowley: MA: Newbury House.
Tarone, E. (1983) On the variability of interlanguage systems. Applied Linguistics 4,2,
Tarone, E. (1990) On variation in interlanguage: a response to Gregg. Applied Linguistics 11,1, 392-400.