A New Angle on L2 Writers' Texts: A Statistical Approach to Translation Universals

摘要

Studies on the second language (L2) writing from the angle of translation universals (TU) have substantial empirical research prospects, but as yet only limited literature explains what linguistic factors shape non-nativeness in L2 writing. This article claims to prove that key TU indices may predict non-nativeness, more particularly translationese in L2 writers' texts. The ultimate aim is thus to classify texts using the indices of translationese, which will, in turn, signify shared, universal features detectable in L2 texts. To that end, we used a collection of multi-factorial analysis methods to compare native scholars' L1 corpora, respectively with two varieties of non-Anglophone scholars' non-translated L2 corpora (L1 English vs. Quasi-L2 English vs. L2 English). The results provided evidence that the TU indices were valid to spot translationese as a signal of non-nativeness in expert non-native writers' journal abstracts. Additionally, the behavioral profiles of TU indices demonstrated that the two variant L2 texts were clustered in higher mutual proximity due to intergroup homogeneity when compared to their native counterparts.