The Processing of Affixation and Compounding in Chinese

• Main Authors: Wang, Wen-Ling, 王汶鈴
• Other Authors: James Myers
• Format: Others
• Language: en_US
• Published: 2004
• Online Access: http://ndltd.ncl.edu.tw/handle/75769361670610121961

碩士 === 國立中正大學 === 語言學研究所 === 92 === The present study investigates the processing of affixation and compounding in Chinese, as well as the effects of stimulus context during morphological processing. The distinction between affixes and roots has long been proposed to be problematic in Chinese, due to the lack of orthographic and phonological diagnostics. For example, 性 —xing, 度 —du, and 化 —hua are such controversial morphemes. This consequently leads to the controversy of identifying complex words ending with these morphemes. While various studies have been conducted to explore the processing of affixed words in Indo-European languages, studies investigating affixation processing is virtually nonexistent in Chinese, which motivates the current research. A relevant study by Andrews (1986), which investigated suffixed and compound word processing in English, showed that morpheme frequency effects were not revealed during suffixed word recognition unless they were presented together with compound words, whereas compound words were consistently found to produce such effects. Andrews ascribed this to the effects of stimulus context on suffixed word processing. Following the design of Andrews, we conducted a series of experiments with different paradigms. Experiment 1 was an investigation of visual priming on suffixed words, which revealed full priming effects. The same task was used in Experiment 2 to investigate compound priming, and partial priming effects were found. Experiment 3, with both suffixed and compound stimuli included, showed full priming effects for both types of words. Even though the overall findings from Experiments 1-3 turned out to be semantic rather than morphological effects, it was proposed that suffixed and compound words were distinguishable, at least in the semantic aspect. Experiments 4-6 were replications of Andrews, which used morpheme frequency as diagnostics for suffixed and compound processing. The overall results from Experiment 4-6 also replicated Andrews. To avoid any effects observed as due to visual overlap, Experiment 7 was conducted, using cross-modal priming, with auditory primes and visual targets. In this experiment, priming between two morphologically related suffixed words pointed toward the inhibitory direction, while priming between two morphologically related compound words was more of facilitation. Despite the fact that these effects did not reach significance, we still could observe the different patterns of processing regarding the two morphological types. Moreover, suffixed and compound words in this experiment were not found to be significantly primed by their constituents either, which was attributed to the activated homophones as competing candidates when auditory primes were heard. Overall, the present study suggested that suffixed and compound words in Chinese can be distinguished and are processed differently. In addition, nonlexical factor such as stimulus context also plays a role in morphological processing.