One norm to rule them all? Describing and evaluating learners’ usages in learner corpus research

Gilquin, Gaëtanelle [UCL]

In his seminal paper on interlanguage, Selinker (1972: 212) refers to “those adults who ‘succeed’ in learning a second language so that they achieve native-speaker ‘competence’”. As suggested by this quotation, the native speaker is often used as a model for learners and as a reference against which to evaluate their proficiency (see, e.g., Cook 1997: 38, Cummins and Davison 2007: 8). This was true in the era of Error Analysis (Corder 1981), but this is also the case, to a large extent, in learner corpus research. Thus, Flowerdew (2015: 469) notes that “[a] key facet of learner corpus research is that the learner corpus is usually compared with a native-speaker control corpus”. In fact, one of the most popular methodologies in learner corpus research, Contrastive Interlanguage Analysis (CIA, Granger 1996), involves as one of its components a comparison between learner data and native speaker data. In this talk, I will discuss various issues relating to the question of the norm in learner corpus research. These will include the issue of whether a norm is acceptable at all (see Bley-Vroman’s (1983) comparative fallacy) and what the options are to dispense with a norm. I will also consider whether the norm should necessarily be native (cf. the Quirk-Kachru controversy, as exemplified by Quirk 1990 and Kachru 1991) and what problems a native norm may present. Using authentic examples, I will illustrate that learners who behave differently from native speakers do not necessarily use language incorrectly. As an alternative to a unique, native norm, I will show that a range of norms are available (“reference language varieties” in Granger’s (2015: 17) revised version of the CIA method), but that again some of these norms may be problematic if they are not selected carefully (depending on the learner corpus, the purpose of the comparison, etc.) and handled cautiously. It will be demonstrated that different choices of norms may produce different results and thus lead to different conclusions with respect to learners’ usages (e.g. Chen 2013). Finally, the pedagogical implications of such choices will be examined, with particular emphasis on whether all differences between the learner corpus and the reference corpus should be targeted for teaching intervention.