Márcio Martins Leitão

A influência da referencialidade no processamento de orações relativas em português brasileiro / The influence of referentiality on relative clause processing in Brazilian Portuguese

Resumo: Neste artigo apresentamos dois experimentos que foram conduzidos para investigar, no portugues brasileiro, o papel da referencialidade no processamento de oracoes relativas associadas a Sintagmas Nominais complexos (SNs complexos) do tipo substância (como “a bolsa de (do) couro”). O experimento de leitura automonitorada revelou uma preferencia imediata pela aposicao nao local, evidenciando uma forte influencia da referencialidade do N1 no processamento da oracao relativa. O experimento de rastreamento ocular, por sua vez, revelou um efeito imediato da referencialidade do N2: uma aposicao local e favorecida quando o N2 e referencial e uma aposicao nao local e preferida quando o N2 nao e referencial. Nossos dados vao na direcao do que e proposto pela Hipotese de Construal (FRAZIER; CLIFTON, 1996). Palavras-chave: processamento; oracoes relativas; referencialidade. Abstract: In this paper we report two experiments that were conducted to investigate the influence of referentiality on the processing of relative clauses associated with complex noun phrases of substance reading (like “the bag of (the) leather”). The self-paced reading task showed an immediate preference for high-attachment, revealing a strong influence of N1 referentiality on relative clause processing. The eyetracking study, in turn, revealed an immediate effect of N2 referentiality: low attachment is favored when the N2 is referential and high attachment is preferred when the N2 is non-referential. Our data are in tune with the general proposal of the Construal Hypothesis (FRAZIER; CLIFTON, 1996). Keywords: processing; relative clauses; referentiality.

A influência da animacidade no processamento de cláusulas relativas no Português Brasileiro

In this paper, two self-paced reading experiments address the question whether the parser accesses and compute semantic features during sentence processing. Both experiments aimed to compare the processing of subject relative clauses and object relative clauses. In the first experiment, these clauses were not manipulated in terms of its animacity features - both subject and object were animate DPs. The results of this experiment demonstrate that subject relative clauses were read significantly faster when compared to object relative clauses. The second experiment did manipulate the animacity features: subject and object of relative clauses were divided in 50% of animate and 50% of non-animate DPs. Under these circumstances, the results reveal that the reading times of subject and object relative clauses are statistically equivalent. The syntactic asymmetries between subject and object relatives alone cannot explain these results, therefore the experiments lead to the conclusion that the parser is likely to access and compute semantic features during on-line sentence processing.

O PROCESSAMENTO DE ORAÇÕES RELATIVAS COM E SEM O PRONOME “THAT” POR BRASILEIROS APRENDIZES TARDIOS DE INGLÊS COMO L21

The purpose of this study is to investigate the linguistic processingof L2 English by a group of Brazilian native speakers. In order to achievethis goal, we analyze the processing of unambiguous object relative clauseswhich contain contrasting structures in the two languages studied here, that is, that have two structure options in English (with or without the pronoun“that”) but only one in Portuguese (containing “that”). Thus, in an onlineself-paced reading experiment, we examined the extent to which Braziliannative speakers would show advantages in processing sentences containingthe relative pronoun “that”, once this structure resembles the one found inPortuguese. Results show that basic, intermediate and advanced learnersseem to process object relative clauses more quickly when they present “that”explicitly, which is taken as an indication of the presence of L1 transfer ef-fects in language processing. Such results are discussed within the predic-tions of the Shallow Structure Hypothesis (Felser et al. 2002).