Publicação
Chinese causative resultative V-Vs and their acquisition by L1 european portuguese learners
| Resumo: | The Chinese Causative Resultative V-Vs (CR V-Vs) express caused-result events, with the component denoting the Manner (causing eventuality) and the one encoding the Result in adjacency. They constitute an interesting construction since they exhibit both lexical and syntactic properties, show thematic flexibility, and sometimes are semantically ambiguous. In previous studies, authors generally fell into one of the following two groups: some claim that CR V-Vs are formed on the lexical level, while others claim that CR V-Vs are formed on the syntactic level. In this study, within the framework of the Minimalist Program and under the assumptions of Distributed Morphology, we attempt to provide an account that can explain CR V-Vs’ properties holistically, including lexical properties such as V-V integrity and the “small size” constraint, and syntactic properties such as productivity and semantic compositionality, as well as other observed constraints such as the semantic constraint on V2 and the constraints related to causative alternation. More importantly, the syntactic structure we propose can account for the thematic flexibility and the semantic ambiguity of some CR V-Vs, which have attracted a lot of attention and imposed difficulty for explanation in the literature. In particular, we propose that Chinese CR V-Vs involve the head vCAUSE. The Mannerdenoting root conflates to it as an adjunct, and the Result-denoting root is incorporated to it as its Complement. Contrastingly, Portuguese does not allow either Manner Conflation or rootselecting vCAUSE. Therefore, for L1 Portuguese learners to acquire L2 Chinese CR V-Vs, parameter resetting will be required. To find out the accessibility of Universal Grammar (UG), the role of L1, and the attainability of parameter resetting in L2 acquisition, we conducted an experiment with 27 L1 Portuguese speakers learning Chinese as L2 (intermediate to advanced level) and 27 Chinese native speakers. The experiment includes three tasks: a semi-elicited production task (SPT), a grammaticality judgment task (GJT), and a comprehension task (CT). Results show a general positive developmental trend in CR V-Vs’ production and comprehension and successful acquisition of some CR V-V constraints, which is a strong argument for UG access. Nativelike performance is observed in some L2 learners’ responses in the SPT and CT, showing the attainability of parameter resetting. L1 transfer of both lexical properties and functional categories has been detected. However, some apparent evidence of L1 transfer of functional category properties is ambiguous because they can also be interpreted as evidence of UG access (similar to L1 acquisition). In general, the results of our study support an argument in favor of the Full Transfer Full Access hypothesis (Schwartz & Sprouse, 1994, 1996) of L2 acquisition. Nevertheless, the results of our study are not exclusively successful. We find that different aspects of the L2 grammars form a hierarchy of acquisition difficulty. While the L2 learners were successful in acquiring properties such as the V2 constraint, the V-V integrity, and the causative alternation constraints, they did not show sensitivity to constraints such as the V-V adjacency, and proficiency effect was not detected. It seems that parameter resetting does not guarantee successful acquisition. We attribute the L2 grammar variations to factors such as the Feature Reassembly (Lardiere, 2005, 2008, 2009a, b), processing difficulties, and the input quality. The L2 acquisition process is more complicated than selecting the correct values for parameters. The learners should also learn how the bundles of features are realized in L2, namely, the Feature Reassembly. Moreover, since frequency plays an important role in efficient acquisition (see Yang, 2010), the processing difficulty and the ambiguous input may complicate the acquisition process and decrease the acquisition efficiency. When the natural input quality is poor, explicit instruction may be needed to help. Furthermore, we hypothesize that the Bottleneck Hypothesis (Slabakova, 2009, 2014, 2016, 2019) and the distinction between macroparameters and microparameters (Baker, 2008; Slabakova, 2019; Tsimpli, 2014) may also provide an explanation: the acquisition of microparameters is more difficult than that of the macroparameters; the core syntax and semantics come easily, but the functional morphology imposes the most difficulties. |
|---|---|
| Autores principais: | Yao, Jiaojiao |
| Assunto: | Língua chinesa - Estudo e ensino - Falantes do português Língua chinesa - Aquisição linguística Língua chinesa - Morfologia (Linguística) Língua chinesa - Sintaxe Teses de doutoramento - 2022 |
| Ano: | 2022 |
| País: | Portugal |
| Tipo de documento: | tese de doutoramento |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | Universidade de Lisboa |
| Idioma: | inglês |
| Origem: | Repositório da Universidade de Lisboa |
| Resumo: | The Chinese Causative Resultative V-Vs (CR V-Vs) express caused-result events, with the component denoting the Manner (causing eventuality) and the one encoding the Result in adjacency. They constitute an interesting construction since they exhibit both lexical and syntactic properties, show thematic flexibility, and sometimes are semantically ambiguous. In previous studies, authors generally fell into one of the following two groups: some claim that CR V-Vs are formed on the lexical level, while others claim that CR V-Vs are formed on the syntactic level. In this study, within the framework of the Minimalist Program and under the assumptions of Distributed Morphology, we attempt to provide an account that can explain CR V-Vs’ properties holistically, including lexical properties such as V-V integrity and the “small size” constraint, and syntactic properties such as productivity and semantic compositionality, as well as other observed constraints such as the semantic constraint on V2 and the constraints related to causative alternation. More importantly, the syntactic structure we propose can account for the thematic flexibility and the semantic ambiguity of some CR V-Vs, which have attracted a lot of attention and imposed difficulty for explanation in the literature. In particular, we propose that Chinese CR V-Vs involve the head vCAUSE. The Mannerdenoting root conflates to it as an adjunct, and the Result-denoting root is incorporated to it as its Complement. Contrastingly, Portuguese does not allow either Manner Conflation or rootselecting vCAUSE. Therefore, for L1 Portuguese learners to acquire L2 Chinese CR V-Vs, parameter resetting will be required. To find out the accessibility of Universal Grammar (UG), the role of L1, and the attainability of parameter resetting in L2 acquisition, we conducted an experiment with 27 L1 Portuguese speakers learning Chinese as L2 (intermediate to advanced level) and 27 Chinese native speakers. The experiment includes three tasks: a semi-elicited production task (SPT), a grammaticality judgment task (GJT), and a comprehension task (CT). Results show a general positive developmental trend in CR V-Vs’ production and comprehension and successful acquisition of some CR V-V constraints, which is a strong argument for UG access. Nativelike performance is observed in some L2 learners’ responses in the SPT and CT, showing the attainability of parameter resetting. L1 transfer of both lexical properties and functional categories has been detected. However, some apparent evidence of L1 transfer of functional category properties is ambiguous because they can also be interpreted as evidence of UG access (similar to L1 acquisition). In general, the results of our study support an argument in favor of the Full Transfer Full Access hypothesis (Schwartz & Sprouse, 1994, 1996) of L2 acquisition. Nevertheless, the results of our study are not exclusively successful. We find that different aspects of the L2 grammars form a hierarchy of acquisition difficulty. While the L2 learners were successful in acquiring properties such as the V2 constraint, the V-V integrity, and the causative alternation constraints, they did not show sensitivity to constraints such as the V-V adjacency, and proficiency effect was not detected. It seems that parameter resetting does not guarantee successful acquisition. We attribute the L2 grammar variations to factors such as the Feature Reassembly (Lardiere, 2005, 2008, 2009a, b), processing difficulties, and the input quality. The L2 acquisition process is more complicated than selecting the correct values for parameters. The learners should also learn how the bundles of features are realized in L2, namely, the Feature Reassembly. Moreover, since frequency plays an important role in efficient acquisition (see Yang, 2010), the processing difficulty and the ambiguous input may complicate the acquisition process and decrease the acquisition efficiency. When the natural input quality is poor, explicit instruction may be needed to help. Furthermore, we hypothesize that the Bottleneck Hypothesis (Slabakova, 2009, 2014, 2016, 2019) and the distinction between macroparameters and microparameters (Baker, 2008; Slabakova, 2019; Tsimpli, 2014) may also provide an explanation: the acquisition of microparameters is more difficult than that of the macroparameters; the core syntax and semantics come easily, but the functional morphology imposes the most difficulties. |
|---|