No description available.

This dissertation investigates (direct) object clitics and object omission in the acquisition of French as a first language. It reports on two original empirical studies which were designed to address aspects of object omission in child French that have remained unexplored in previous research. Study 1 investigates the incidence of object omission in the spontaneous speech of French-speaking children aged three and above, an age group for which no analysis, and only little data, have been available so far. Findings show that object omission continues to occur at non-negligible rates in this group. A comparison with age- and language-matched groups of English- and Chinese-speaking children (from Wang, Lillo-Martin; Best & Levitt 1992) suggests that French-speaking children omit objects at higher rates than their English-speaking peers, yet at lower rates than children acquiring a true null object language, such as Chinese. Study 2 was designed to investigate whether French-speaking children would accept null objects on a receptive task, an issue that has not been previously investigated. A series of truth value judgment experiments is developed, adapting an experimental paradigm that has not been used previously in the context of null objects. Results from English- and French-speaking children show that both groups consistently reject null objects on these tasks, a finding that constitutes counterevidence to proposals which attribute object omission in production to a genuine null object representation sanctioned by the child grammar. Overall, the pattern of results turns out not to be consistent with any developmental proposals made in the literature, suggesting that a novel approach is required. Proposing a minimalist adaptation of Sportiche's (1996) analysis of clitic constructions, and taking into consideration the recent emphasis on 'interface' requirements imposed by language-external systems, I put forward a hypothesis for future research, the Decayed Features Hypothesis (DFH), which locates the source of object (clitic) omission in child French in a specific language-external domain, namely the capacity of working memory.

This dissertation is a contrastive study of the semantics, pragmatics, and discourse functions of demonstratives in English and Chinese.It is shown that there is a metaphorical relationship between the basic semantic properties of demonstratives and their various uses in the two languages. The proximal demonstrative tends to be used for spatial, temporal, or emotional closeness, or for a foregrounded referent, whereas the distal demonstrative is usually used for spatial, temporal, or emotional remoteness, or for a backgrounded referent. However, details of the metaphorical extensions in the two languages may vary. Functional differences between demonstrative pronouns and neuter pronouns in English and Chinese are also discussed. It is shown that demonstrative pronouns tend to code a higher degree of topic discontinuity or topic change, and neuter pronouns a greater degree of topic continuity in the two languages.In addition to contributing to an understanding of the basic factors governing the uses of demonstratives in English and Chinese, which could be used as a basis for further cross linguistic study, this research should also have some pedagogical value for teaching both English and Chinese as foreign languages. / Department of English

In this thesis we shall investigate the kinds of information which the Human Sentence Processing Mechanism employs during the resolution of local syntactic ambiguity in written texts. The thesis is in three parts. In Part I we consider some current models of syntactic ambiguity resolution. On the one hand, we consider the structural approaches, in which the processor considers only syntactic information when choosing between alternatives. On the other, we consider the interactive approaches, in which different kinds of information are brought to bear during the resolution process. In Part II, we describe a number of experiments which contrast the predictions of these two approaches. In particular, we investigate the processing of sentences which are locally ambiguous between a simple noun phrase analysis and a complex noun phrase analysis. Frazier (1979) predicts that the simple noun phrase analysis is chosen because it utilizes fewest phrasal nodes in its associated phrase marker. Crain and Steedman (1985), however, predict that the interpretation of the noun phrase is determined by referential factors, such as whether or not a unique referent can be identified for the noun phrase. The results support Crain and Steedman's interactive hypothesis. In Part III, we consider some theoretical issues concerning the timing of the processor's decisions. Crain and Steedman's original model is modified in the light of such considerations. We explore the implications of the modification for the status of syntax and semantics within our model of sentence comprehension. In the final chapter, we attempt to explain the existence of parsing preferences in sentences which are presented in isolation, and for which no explicit contextual information has been provided. We conclude that contextual considerations, such as the distinction between what is and what is not already known to the hearer, are of fundamental importance during the resolution of local syntactic ambiguity by the Human Sentence Processing Mechanism.

In this thesis, I argue that early child second language (L2) grammars allow truncation, on a par with proposals by Rizzi (1993/1994) and Haegeman (1995) for first language (L1) acquisition. This account (the Truncation Hypothesis) holds that Rizzi's (1994) Root Principle, according to which root declaratives are CPs, is initially underspecified in L2 systems (for processing reasons). This means that the root of main declaratives will not systematically be CP. Instead, different types of roots should be projected, such as CP, IP or VP, with VP underlying root infinitives. If one further assumes that functional categories are present in early grammars, the possibility of truncation can thus account for optionality of verb-movement and finiteness in early SLA, and more generally for why such categories seem to be optionally projected initially (Vainikka & Young-Scholten, 1994; 1996; Eubank, 1992; 1993/1994; 1996). / Predictions based on the Truncation Hypothesis were tested against longitudinal spontaneous production data from child and adult L2 learners. There were two child and two adult learners of L2 French (whose L1s were English and Arabic) and two child two adult learners of L2 German (native speakers of Romance pro-drop languages). The findings suggest that the distribution of finite and nonfinite verbs is structurally determined in L2 child grammar, i.e. tenseless verbs only appear when VP is the root, while finite verbs are found when functional categories are projected. This in turn means that children project truncated structures in early L2 acquisition. I argue that no other theory of the nature of early L2 grammars is able to account for the full range of properties of the child L2 data. / The adult data are less conclusive concerning the possibility of truncation in adult L2 grammars. In particular, the learners seem to use infinitival markers as substitutes for finite inflections, which means that nonfinite verbs are found in contexts which are not predicted by the Truncation Hypothesis. The difference between the child and adult learners is attributed to problems that adults may have in mapping the syntactic and morphological systems (Lardiere, 1996), and not to a discrepancy in syntactic knowledge.

No description available.

Research on morphological variability in second language (L2) acquisition has focused on the syntactic consequences of variability: that is, whether or not morphological variability entails underlying syntactic deficits. The interrelationship between morphological features in their own right has been largely ignored. This thesis addresses the representation of L2 features by investigating the use of default morphology---the outcome of systematic substitution errors employed by speakers of L2 Spanish. It is hypothesized that underspecified features act as defaults; by assumption, those features that are unmarked are underspecified. / Evidence to support this hypothesis comes from two sets of experiments conducted on intermediate- and advanced-proficiency L2 Spanish subjects (L1 English). The first set of experiments addresses verbal morphology, and consists of a spontaneous production experiment on person, number, tense, and finiteness, and a comprehension task on person and number. The second set of experiments addresses gender and number in nominal morphology, and consists of a spontaneous production experiment on determiners, an elicited production experiment on clitics and adjectives, and a picture-selection task on the comprehension of clitics. Across tasks and across verbal and nominal domains, errors involve the systematic substitution of underspecified morphology. The observation that morphological variability extends to comprehension, and is qualitatively similar to the variability found in production, counters the suggestion that variability is strictly a product of mere performance limitations on production. Finally, the systematicity of substitution errors suggests that the natural classes of features such as gender, number, tense, and person are acquirable in an L2, regardless of whether or not these features have been instantiated in the native language.

This thesis undertakes to refine our understanding of the syntactic properties of sentential negation. The proposed analyses operate at the juncture of recent innovations to Case, Agreement, and X-bar theories, within a Government and Binding framework. Case is checked in a Specifier/Head configuration whenever possible, and agreement is analyzed as the reflex of a Case-checking operation at S-structure. The proposal that the inventory of functional categories available in Universal Grammar includes a Negation Phrase (NegP) is adopted as a point of departure. / In the context of this investigation, certain syntactic properties are attributed to the head and specifier of NegP. It is proposed that the specifier of NegP provides an A-position in which NPs may be Case-checked. S-structure Case-checking is reflected in agreement marking on the negative head. Evidence for LF Case-checking in this position is derived from the Case-licensing of direct objects in negated clauses in Colloquial Welsh and Russian. Definiteness effects are analyzed as due to constraints on an additional Case-licensing option required if negation blocks Case assignment under government. Variations in agreement patterns in affirmative and negative relative clauses in Literary and Colloquial Welsh are attributed to the barrier status of the head of NegP.

This thesis examines constructions known as resultative constructions. In addition to the well-known adjectival resultative construction in English, I investigate the resultative V-V compound, found in Japanese, and the resultative serial verb construction, found in Edo. / I propose a new classification of these constructions, which focuses on the argument structure of the construction. In Japanese resultative V-V compounds, the argument structure of a compound reflects the argument structure of the second verb only, while in Edo, the argument structure of the construction reflects the argument structure of both verbs involved. With this criterion, English resultative constructions are divided into two classes---a resultative construction containing an intransitive verb is classified with Japanese resultative V-V compounds, and a resultative construction containing an object-selecting verb is classified with Edo resultative serial verb constructions. / Based on the classification provided here, I investigate two types of syntactic operations which license the concatenation of the predicates in resultative constructions. I argue that English intransitive resultative constructions and Japanese resultative V-V compounds are formed by adjoining one of the predicates on the other. The adjunction structure is then interpreted as conjunction called event identification. In contrast, English transitive resultative constructions and Edo resultative serial verb constructions are licensed by treating one of the predicates as a causative predicate. I argue that one of the predicates in these constructions undergoes lexical coercion, and acquires a causative meaning. The newly-formed causative verb takes the other predicate of the construction as its complement. This structure is then interpreted with function-application. I hence argue that the structural difference between the two types of resultative constructions also mirrors the difference in the type of semantic operations used to interpret these constructions.

No description available.