This book presents a new theory of grammatical categories - the Universal Spine Hypothesis - and reinforces generative notions of Universal Grammar while accommodating insights from linguistic typology.
Date: Wed, 22 Sep 2004 01:17:25 -0600 From: Fang-yi Chao <Fang-yi.Chao@colorado.edu> Subject: Sentence Processing in East Asian Languages
EDITOR Nakayama, Mineharu TITLE: Sentence Processing in East Asian Languages PUBLISHER: CSLI Publications YEAR: 2002
Fang-yi Chao, Department of East Asian Languages and Civilizations, University of Colorado at Boulder, Assistant Professor
OVERVIEW The book contains a selection of the papers given by participants of the International East Asian Psycholinguistics Workshop held at the Ohio State University on August 4, 1999. The editor of the book indicates that this is the first book of its kind that devoted solely to sentence processing issues in East Asian languages, particularly Cantonese, Japanese, Korean and Mandarin Chinese. This is a valuable resource for investigating East Asian languages in psycholinguistics, computational linguistics, and theoretical linguistics.
SYNOPSIS The book consists of ten chapters, each of them being an independent article. Each chapter manifests the special characteristics of the language(s) under discussion as well as the problems pertaining to the processing of those unique features in the language(s).
The first chapter (pp.1-29) by Kathleen Ahrens entitled "Time issues in lexical ambiguity resolution" examines the effects of homophones on lexical ambiguity in Chinese. Ahrens based on modularity hypothesis discusses the significance of the immediacy as well as the automaticity of lexical access in processing lexical ambiguity, and argues that time as well as content play significant roles in ambiguity resolution in Mandarin Chinese.
Chapter Two (pp. 31-52): "Resolution of reanalysis ambiguity in Japanese relative clauses: early use of thematic compatibility information and incremental Processing" by Yuki Hirose. Hirose's chapter discusses the effect of the thematic role ambiguity on processing Japanese relative clauses. Hirose argues that the parser makes use of thematic compatibility information as early as the application of structure-based preference principles, and parsing is strictly incremental in the reanalysis stage as well as in the first-pass stage.
Chapter Three (pp.53-83): "The nature of categorical ambiguity and its implications for language processing: a corpus-based study of Mandarin Chinese" by Chu-Ren Huang, Chao-Jan Chen and Claude C. C. Shen. In their chapter, Huang et al. discuss the nature of categorical ambiguity in human language and establish the baseline performance for categorical ambiguity resolution for Chinese. They propose that categorical ambiguity might be primarily motivated by conceptual necessity.
Chapter Four (pp.85-110): "Syntactic and positional similarity effects in the processing of Japanese embeddings" by Richard L. Lewis and Mineharu Nakayama. Lewis and Nakayama argue for the similarity-based interference theory; a proposal that claims syntactic similarity of NP arguments is a significant determinant of processing difficulty in unambiguous embedded constructions. They extend their theory to include positional interference, and present experimental data in support of this theory.
Chapter Five (pp.111-29): "Lexical ambiguity in sentence processing: evidence from Chinese" by Ping Li, Hua Shu, Michael Yip, Yaxu Zhang and Yinghong Tang. Lexical ambiguity has been one of the most important topics in the studies of context effects in word recognition. As the Chinese language has a massive number of homophones at the lexical and morphemic level, Li et al. investigate the processing of homophones in both Cantonese and Mandarin. Their findings support the context-dependency hypothesis, which argues that language processing is a highly interactive form of information processing, and that prior sentence context can influence lexical access at an early stage.
Chapter Six (pp.131-66): "Costs of scrambling in Japanese sentence processing" by Reiko Mazuka, Kenji Itoh and Tadahisa Kondo. In their chapter, Mazuka et al. explore from psycholinguistic perspectives the cost associated with processing scrambled sentences. They challenge previous generalization about processing scrambled sentences and argue that scrambled sentences are indeed more difficult to process than canonical sentences.
Chapter Seven (pp. 167-88): "Source of difficulty in processing scrambling in Japanese" by Edson T. Miyamoto and Shoichi Takahashi. Similar to the previous chapter, Miyamoto and Takahashi also deal with the problems related to processing scrambling in Japanese. Miyamoto and Takahashi propose that the processing of non-canonical word orders involves searching for the gap required by a displaced element, thus causes extra time during the processing procedure.
Chapter Eight (pp.189-221): "Processing filler-gap constructions in Japanese: The Case of Empty Subject Sentences" by Tsutomu Sakamoto. Based on Fodor (1978) Sakamoto proposes a "Multi-level parsing model" to account for different levels of information processing. Supported with evidence from experiments involving the processing of filler-gap constructions, he claims that there exist at least two distinct levels of processing. However, the findings in his experiments do not confirm the sequential ordering of these levels.
Chapter Nine (pp.223-55): "Effects of accentual phrasing on adjective interpretation in Korean" by Amy J. Schafer and Sun-Ah Jun. The purposes of this study are to investigate the effects of accentual phrasing on sentence comprehension in Korean as well as to compare the effects of prosodic phrasing in Korean and English. Shafer and Jun argue that sentence comprehension in Korea, like English, shows effects of acoustically subtle prosodic phrase boundaries. They also claim that the discrepancy between Korean and English in prosodic effects stem from the grammatical constraints governing prosody and the mapping between prosodic structure and syntactic and semantic structures in the two different languages.
Chapter Ten (pp.257-87): "Center-embedding problem and the contribution of nominative case repetition" by Keiko Uehara and Dianne C. Bradley. Uehara and Bradley's chapter discusses the effect of nominative case (-ga) repetition, particularly the contrast between three times and twice, on processing doubly center-embedded sentences in Japanese. Based on the results of testing 1) ga- repetition cost in double embedding, 2) phonological repetition effect, and 3)ga- structural ambiguity effect, they argue that neither the phonological repetition nor the syntactic ambiguity of the nominative marker (ga-) contributes to the processing difficulties associated with doubly center-embedded sentences. As the results of their experiments indicated that there was a strong preference to take NP-ga as the subject of a clause, they suggest that the processing difficulties are mainly caused by the repeated subjectness.
CONCLUSION As language processing is relatively new in the studies of East Asian languages, most chapters in the volume conclude their discussions with further directions for the investigation of topics related to their studies. This collection of papers is not suitable for beginners, because some papers require background knowledge of computational linguistics or psycholinguistics. This book is indeed a valuable contribution to the study of language processing, as it provides empirical results from processing East Asian languages for the construction of a universal theory of human language processing. It is to be hoped that it will stimulate a whole range of follow-up studies in the area of processing non Indo-European languages.
ABOUT THE REVIEWER:
Fang-yi Chao is an Assistant Professor in the Department of East Asian Languages and Civilizations at University of Colorado at Boulder. Her research interests include Chinese historical linguistics, Chinese Dialectology, Sociolinguistics, and language pedagogy.