Processing constraints and word order variation in Hindi relative clauses
- Psycholinguistic research has repeatedly found evidence for a deep and non-arbitrary relationship between word order variation and intrinsic constraints on our sentence comprehension and production capabilities imposed by our language processing system. However, the exact nature of these processing constraints and the ways in which they interact with one another are far from understood. One impediment to progress is that most research focusing on word order variation is restricted to only a small set of languages, with limited possibilities for varying word order. Hindi, a language with extremely flexible word order, offers a much richer set of word ordering possibilities than most of these languages but has received comparatively limited attention to date. Thus a systematic analysis of word order variation in Hindi provides an interesting and novel probe into underlying cognitive constraints on sentence production and comprehension. In particular, Hindi relative clause constructions, which have been neglected in psycholinguistic research so far, have a range of rare and intriguing properties that can shed light on the simultaneous influence of multiple cognitive processing constraints on word order. This dissertation experimentally investigates the influence of various processing constraints of current theoretical interest on online word order preferences in Hindi relative clause constructions. More specifically, the dissertation explores the influence of locality, end-weight, and (syntactic and semantic) expectations. Four sub-questions were investigated through a series of six experiments that employed a variety of methods including acceptability judgments, sentence production via fragment-ordering, self-paced reading, and sentence completion: (i) Is the placement of a relative clause influenced by whether or not it overlaps with information from the prior discourse? (ii) Do end-weight and locality both matter in the placement of non-restrictive relative clauses? (iii) How do locality constraints and semantic expectations influence the processing of relativized structures? (iv) Is locality important in the placement of the head-noun being modified? Taken together, the experiments found evidence for the influence of all factors mentioned above, establishing that processing factors can indeed influence online word ordering preferences in Hindi and deserve further empirical attention. In particular, the results demonstrated that multiple opposing constraints can apply simultaneously and the language processor must resolve these competing biases. The results therefore reflect the complexity underlying word order preferences and underscore the importance of studying how competing processing requirements may collectively influence word ordering outcomes. Furthermore, the processing requirements may interact with other aspects of the communicative act that are unrelated to the needs of the processor. The experiments on dependency-locality, for instance, suggested that theories of locality may need to be expanded to incorporate factors like the directionality and type of the dependency or the semantic functions of elements in the dependency. It was only through the availability of multiple relativization options in Hindi, that this insight could surface. As such, this research on Hindi highlights the fact that a complete understanding of the link between processing factors and word order can only be achieved by expanding the range of languages and linguistic constructs being considered.
|Type of resource
|electronic; electronic resource; remote
|1 online resource.
|Stanford University, Department of Linguistics
|Jurafsky, Dan, 1962-
|Jurafsky, Dan, 1962-
|Statement of responsibility
|Submitted to the Department of Linguistics.
|Thesis (Ph.D.)--Stanford University, 2010.
- © 2010 by Anubha Kothari
- This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).
Also listed in
Loading usage metrics...