Learner Corpora and the Acquisition of Word Order: A Study of the Production of Verb-Subject Structures in L2 English

Though there is a long tradition of research into word order phenomena in Second Language (L2) acquisition, this area of enquiry has recently been given a new impetus both from theoretical developments on the form-function interplay and, crucially, from the emergence of learner corpora This paper focuses on a particular phenomenon which has received considerable attention in the literature: the production of postverbal subjects in L2 English by L1 speakers of languages characterised as allowing ‘free inversion’ of the subject in V(erb) S(ubject) structures, such as Spanish and Italian. In previous research emphasis has been placed on the learners’ production of ungrammatical VS order (see for instance Rutherford, 1989 Zobl, 1989 and, more recently, Oshita, 2004). Our approach, however, seeks to identify the syntactic, semantic and pragmatic conditions, as well as conditions deriving from processing mechanisms, under which learners produce inverted subjects, regardless of errors resulting from the syntactic encoding of those notions. We analyse VS vs. SV structures in the Italian and Spanish subcorpus of ICLE (Grange et al., 2002) and we compare our results with preliminary results obtained from a similar native English corpus (LOCNESS). Thus, we incorporate some of the fundamental tenets of what is known as Contrastive Interlanguage Approach (see, e.g. Granger ,1996 and Gilquin, 2001), which establishes comparisons between: (a) native and non-native data, and (b) different non-native data. Our main purpose is to see if the properties that govern the occurrence of postverbal Ss in native English, as currently analysed in the theoretical and descriptive literature, are the same as those operating in the non-native grammars of Spanish and Italian speakers We first examine the properties of VS order in English vs. Spanish/Italian (Section 2). In Section 3, we review previous L2 studies on postverbal Ss. Our hypotheses are presented in Section 4. Section 5 describes the method used to extract and code data from the corpus and their statistical treatment. Results are presented and