Processamento do preposition stranding em orações relativas por brasileiros usuários de inglês como L2

Abstract

Studies in Psycholinguistics and Bilingualism investigate relationships between language knowledge in the mind of L2 users (COOK, 2016). Such relationships are usually considered capable of occurring at various linguistic levels, from phonological to pragmatic, with the directionality of these effects reported in the literature largely occurring from L1 to L2. L2-to-L1 effects are also found, mostly in L2-immersed bilingual populations, with little to no use of the L1. However, there are investigations (SOUZA et al., 2014) that show that even in situations of non-inversion of dominance, it is possible to find L2 acquisition effects on the L1, especially in syntactic structures that do not have equivalence between them, such as the resultative construction (OLIVEIRA; SOUZA; OLIVEIRA, 2017) and the induced movement alternation structure (FERNÁNDEZ; SOUZA, 2016). In the linguistic pair Brazilian Portuguese (BP) and English, besides de previous structures, there is the Preposition Stranding (PSt) (ROSS, 1967; RIEMSDIJK, 1978; HORNSTEIN; WEINBERG, 1981), which occurs, among other structures, in prepositioned relative clauses. In PSt there is the movement of the object of a preposition to head the relative clause while this preposition remains in its place of origin. Descriptions of relativization in BP (TARALLO, 1983) do not admit this type of construction, considering them ungrammatical, while in English the use of this structure is highly productive in speech situations. In this way, we seek to investigate the processing of Preposition Stranding in prepositioned relative clauses by Brazilian speakers of English as L2. Through three experiments, a self-paced reading task in English, a self-paced reading task in Portuguese, and a speeded acceptability judgment task also in Portuguese, we found suggestive results that advanced Brazilian bilingual speakers of English: (i) process Preposition Stranding similarly to native speakers of English, given certain conditions; (ii) process Preposition Stranding in BP more easily than BP monolinguals; and, (iii) rate Preposition Stranding higher in BP than BP monolinguals. These results point to the existence of effects caused by the presence of bilingualism influencing the processing of L1, as are compatible and discussed through proposals that admit the existence of relationships between the linguistic knowledge of the bilinguals’ languages in their minds (COOK, 2016; AMARAL; ROEPER, 2014), as well as consider that individual differences (DUSSIAS et al., 2019) may act in language processing tasks.