ITL - International Journal of Applied Linguistics - Volume 165, Issue 1, 2014

The Common European Framework of Reference for Languages (CEFR, Council of Europe, 2001) currently functions as an instrument for educational policy and practice. The view of language proficiency on which it is based and the six proficiency levels it defines lack empirical support from language-use data. Several issues need to be investigated collaboratively by researchers working in the fields of first and second language acquisition, corpus linguistics and language assessment. These issues are concerned with (i) the CEFR’s failure to consistently distinguish between levels of language proficiency (static aspect) and language development (dynamic aspect), (ii) with the CEFR’s confounding of levels of language proficiency and intellectual abilities, and (iii) the potential problem of mismatches between second-language learners’ communicative and linguistic competences. Furthermore, from a more theoretical perspective, this paper proposes (iv) to investigate which CEFR proficiency levels are attainable by native speakers and (v) to empirically delineate the lexical, morpho-syntactic and pragmatic knowledge shared by all native speakers (called Basic Language Cognition).

High proficiency in L2 partly depends on acquiring many formulaic sequences (FSs), yet post-childhood learners find this difficult. Ways of accelerating the acquisition of FSs would be welcome. Small-scale studies have indicated that assonance (e.g., strong bond) makes studied FSs especially retrievable if, during exposure, assonance is made the object of teacher-instigated awareness-raising and attention direction. However, questions remain about effect size and duration. In two new experiments a mnemonic effect of assonance was detected after 5–10 minutes. This was despite a sorting task thought likely to direct participants’ attention particularly to the control collocations. The effect appeared to fade over an hour and disappear after a day. A small-scale meta-analysis indicates the effect is initially of medium size. We discuss how short-term operation of such an effect could facilitate the fuller acquisition of partly learned assonant FSs. We propose avenues for research into means whereby the mnemonic effect of assonance might be exploited in learning materials. We touch on effects of item frequency, mutual information, and concreteness-imageability of meaning.

Recent literature in second language acquisition shows that syntax-driven structures give way to successful modal interpretation and morphological production, while discourse-dependent environments do not (Sorace, 2005; Tsimpli & Sorace, 2006, Borganovo, Bruhn de Garavito, & Prévost, 2008; Iverson, Kempchinsky, & Rothman, 2008). It has also been suggested that discourse-dependent environments involve both structural and pragmatic knowledge of L2, which intersect at the syntax-discourse interface (Sorace, 2005; Iverson, Kempchinsky, & Rothman, 2008), thereby requiring a multi-layered understanding of the target language. The present study contributes to this line of research by further examining morphological variability (Prévost & White, 2000; Sorace, 2000; Sorace, 2005; Iverson, Kempchinsky, & Rothman, 2008; Slabakova, 2009) in L2 acquisition at the morphosyntactic-semantic interface, following the work of Borganovo, Bruhn de Garavito, & Prévost (2008). The results of our study reveal that learners, even at advanced stages of acquisition, perform poorly in epistemic environments where syntax and discourse intersect. In such environments, there appears to be an interaction with pragmatics (cf. Iverson, Kempchinsky, & Rothman, 2008) that causes learners to opt for the indicative mood, even when the subjunctive is prescriptively required. Unlike deontic modality, which is essentially syntax-driven, epistemic modality requires structural knowledge, as well as knowledge from other domains (Sorace, 2005). Our study reveals that learners at all levels of instruction performed better in “purely syntactic” environments of deontic modality than in pragmatically challenging epistemic environments.