A Multifactorial Analysis Of Relativizer Omission By Chinese EFL Learners

Relativizer omission, though an important syntactic phenomenon, has received insufficient attention in academia. Previous studies with regard to relativizer omission were carried out with intuitive or experimental approach. More importantly, few studies have ever dealt with relativizer omission by Chinese EFL learners.This study, therefore, adopting a usage-based perspective, highlighting a variety of statistical analyses, addresses the cognitive mechanism underlying relativizer omission by Chinese EFL learners including their shared salient constraining variables with NS of English and their idiosyncratic ones. Three questions have been proposed:a) Do Chinese EFL learners share with English native speakers similar constraining factors in relativizer omission? b) Are there significantly different constraining factors governing relativizer omission between Chinese EFL learners and English native speakers? c) What about the greatest discrepancy between the two groups under comparable context?1726 non-subject restrictive relative clauses have been retrieved from Chinese EFL learners corpora (WECCL 1.0, WECCL 2.0) and English native speakers corpora (LOCNESS, NESSIE, ICNALE), with a variety of linguistic factors annotated, and statistically investigated in four respects:monofactorial analysis, binary logistic regression, double regression, and distinctive collexeme analysis.The results of regression analyses (binary logistic regression and double regression) show that a) relativizer omission are constrained by a variety of linguistic factors; Chinese EFL learners share nine constraining factors with native speakers: Head contentfulness, Head definiteness, Head animacy, the number of Head modifiers, the Syntactic role of Head in main clause, POS of relative clause subject, definiteness of relative clause subject, relative clause negation, relative clause length; and b) possess four unique salient factors:relative clause length, Head concreteness, RC subject concreteness, RC predicate; c) a more salient inter-group discrepancy arises from explicit relativizer construction, i.e., an inverse correlation between the inter-group discrepany of explicit relativizer construction and the cognitive load of constraining variable levels has been verified.Distinctive collexeme analysis demonstrates that a) zero relativizer witnesses a greater overlap of collexemes between Chinese EFL learners and English native speakers; b) Chinese EFL learners rely more heavily on those shared collexemes.Theoretically, it has been proved that, through statistical analyses, there exists a continuum of relativizer constructions ranging from item-based concrete constructions to schematic constructions. Schematic constructions arise out of item-based constructions through a competition-emergent approach. Compared with native speakers of English, Chinese EFL learners’ use of relativizer omission is more geared towards item-based constructions.The multifactorial analysis of relativizer omission by Chinese EFL learners has empirically falsified the two features of interlanguage (systematicity, permeability), verified the default prototype relativizer construction, and thus supported the feasibility of usage-based approach in SLA studies.