Do adult learners improve their language competence during peer-to-peer interaction with reciprocal feedback?

The objective of this paper is to reflect on peer-to-peer interactions with reciprocal feedback as a means to promote the improvement of language competence (LC). According to Van Patten (2010), LC consists of mental representation and skills. The former is the abstract, implicit and underlying linguistic system in a speaker’s mind and the latter is the ability to use language fluently (measured by speed and accuracy) and appropriately in both production and comprehension. Neither mental representation nor skills cannot be directly taught since they evolve in learners based on their experiences. It is thus possible to assume that the development of LC is related with learner autonomy, as it has been defined by Little 1991, and output and interaction with reciprocal feedback can successfully promote its development (Swain 1985, Long 1996). Aiming at improving LC, written e-Tandem exchanges with reciprocal feedback have been promoted for several years between Italian and German learners learning each other language. In order to find out if reciprocal feedback really fostered improvements in output, a corpus of learners’ written activities has been collected and examined. Learners’ texts have been investigated according to Pinemanns’ 5 stages scaled implementation of key morphosyntactic features of German (Pienemann 1995, Pienemann et alii 2005), while feedback have been classified according to Sheen’s (2007) taxonomy. Findings show that that both writing texts and giving feedback lead learners on to test their own version of the target language (i.e. their interlanguage) and to develop their mental representation of the target language; direct corrective feedback with metalinguistic comments promotes thus improvements in LC.