Tight Constraints on WH-movement in Lankan English? - A Preliminary Inquiry into Subjacency and Grammatical WH-movement

Abstract

fPTi-movements in English involve both Subjacency and grammatical w/z-movements. Subjacency is one of the non-parameterised principles or language universals, which has constraints on w/j-movements. Although this language universal is strictly observed in languages like English, it is not so in some languages. The knowledge of Subjacency and grammatical w/j-movements is believed to be part of a person's unconscious linguistic knowledge (I-language). The knowledge of Subjacency tells a person as to from what types of structures wA-elements can be extracted or not. This study tried to find out the nature of the Ilanguage related to Subjacency and grammatical wA-movements in first language (LI) Lankan English (LE) users. For this purpose, the study employed a grammaticality judgement task (GJT), which comprised sentences with Subjacency violations and those with grammatical whmovements applying the norms of Older Varieties of English (OVE) - the token sentences used in the GJT were similar to those sentences, I mean, followed the same syntactic patterns, that were used in the studies carried out on Subjacency among modem native speakers of English (OVE) as control groups. The results of the OVE groups suggest that they agree on (un)grammaticality of w/i-movements. The subjects of my study use L E as their L I or one of their L i s— the subjects are mostly Colombo-based and use Sinhala as one of their L i s or second language with one or two exceptions, where the subjects use Tamil as one of their L i s. Based on their linguistic intuition, the subjects gave values for the grammaticality of the token sentences using a gradient scale ranging from (-2) to (+2) assigning the highest negative value for complete ungrammaticality and the highest positive value for complete grammaticality. Results show that there is strict adherence to Subjacency observation among the LI L E users; however, there is a marked reluctance among many of the L I L E users to extract w/i-elements across complex structures even when these extractions can be considered completely grammatical in OVE. A pattern observed in the finding is that L E users feel more comfortable with moving adjuncts as wA-elements across complex structures rather than arguments. Whether this observed syntactic phenomenon should be considered as an indicator for a variety specific linguistic feature of LE, or, considering the fact that several of the subjects have considered the movements as grammatical or somewhat grammatical, as an indicator for language change in LE, or the different levels of complex syntactic structures the subjects are exposed to, is a matter for future in-depth study and analysis.