Phrasal Verbs in learner English: a semantic approach. A study based on a POS tagged spoken corpus of learner English

Abstract

Phrasal Verbs (PVs), understood as a verb and a particle, though very common in native speech, are reportedly difficult to learn by non-native speakers (NNSs) of English (see Celce-Murcia and Larsen-Freeman 1999). The hypothesis is therefore put forward that for Polish learners of English too the range of PVs is generally significantly smaller than for English native speakers (NSs) and that their degree of use of the semantic categories of PVs is inversely proportional to the PVs’ level of idiomaticity. In other words, Polish learners have little trouble with transparent verbs, more with semi-transparent and most with opaque ones (see Dagut and Laufer 1985). In order to verify this hypothesis, we have used the evidence from the PLINDSEI corpus, that is, the Polish part of the LINDSEI (Louvain International Database of Spoken English Interlanguage), containing advanced English as spoken by NSs of Polish, and from the LOCNEC (Louvain Corpus of Native English Conversation), which we have used as a reference corpus. The comparison of PV usage by Poles as NNSs of English and by English NSs has been performed employing the scheme of contrastive interlanguage analysis (Granger 1996). We show learner over- and underuse of items and illustrate the searches conducted for identifying patterns of use. The methodology applied consists in a partially automatic extraction and a subsequent manual filtering of PVs from a POS-tagged NNS corpus and its reference NS corpus. A semantic analysis of the extracted PVs based on the notion of compositionality (see Celce-Murcia and Larsen-Freeman 1999; Armstrong 2004) has been performed and the hypotheses verified.