Sujeitos nulos, tópicos e a satisfação do EPP na aquisição de segunda língua: a influência do português brasileiro no inglês

Abstract

This study focuses on language transfer, more specifically, on the interference of Brazilian Portuguese (BP) as the native language (L1) on the acquisition of English as a second language/foreign language (L2/FL). The phenomenon under investigation is the use of null vs. overt pronominal subjects. The theoretical framework adopted is the generative linguistic theory based on the Minimalist Program (Chomsky, 1995) and psycholinguistics, responsible for providing procedural models in order to explain how production and comprehension of sentences can be understood. There are different proposals in the literature on L2 acquisition. Considering the phenomenon under investigation, the satisfaction of an EPP feature, responsible for the projection of syntactic subjects in natural languages, would be differently parametrized in BP and in English. BP has become similar to English for referential subjects, but not in respect to expletive subjects. Moreover, it presents topic-subject constructions that may interfere in the acquisition of the negative parametric value for null subjects in English. In order to investigate the changes that have affected BP and observe how learners of English as FL that have BP as their native language are sensitive to the grammaticality of sentences with null subjects in English in different contexts, two experiments were conducted, both with a grammaticality judgment task . The pilot experiment was conducted with learners of basic and advanced levels and presented two distinct types of sentences (Type 1: null subject and type 2: topic + null subject), and a final experiment with learners of basic, intermediate and advanced levels, with three types of sentences (Type 1: null subjects, Type 2: topic + null subject and type 3: conjunction + null subject) .Given the complexity of BP grammar, our hypothesis is that full transfer will not be observed, but rather the emergence of an interlanguage that sometimes is closer to the target language, sometimes differs from it, reflecting processing difficulties imposed by the heavy demands of dealing with the two grammars. The results support the hypothesis indicating that (i) the value of the null subject parameter seems to be transferred from L1 to L2 since we found a high number of incorrect responses, (ii) the interference occurs more strongly in the context of expletive subjects, (iii) there is interference of general restrictions of the L1 grammar (V1 restrictions) in L2, and (iv) the interference decreases with the increasing proficiency in L2. Furthermore, in type 2 sentences, there seems to be a possible interference of BP which leads to the omission of the expletive, indicating a difficulty in integrating information from the constraints imposed by the need to process two languages at specific times in order to prevent the interference from the unwanted language, in this case L1, which requires more resources to be inhibited once it is still the dominant language (SORACE , 1996, 2011).