Reading morpho-syllabic Chinese is challenging for English-speaking learners of Chinese as a second or foreign language (L2) because of the complexity of Chinese orthography, ambiguity in morphemes and words, and the lack of spaces between words in sentences. Without efficient compound-word recognition and identification skills, not only would proficient reading in L2 Chinese be hard to achieve at the individual level, but also low retention rates in Chinese studies would be exacerbated at the institutional level.
Chinese compound words, which have five different structures, account for two thirds of the Chinese lexicon. Among the five structures, coordinative compound words have unique inter-character orthographic and semantic characteristics. Specifically, the two constituent characters (morphemes) may share identical or similar meaning (e.g., ??, hua1cao3, flower-grass, plant) and may share similar visual shape (the radical ? on the top) with each other.
In Chinese as a second/foreign language (CFL) teaching practice, errors and confusion in verbal identification and written production pertaining to coordinative compound words are observed in (a) reversal (e.g., ??, jia4zhi2, price-cost, value, as *??), (b) missing/dropping (e.g., ??, su4she4, to reside-to stay overnight, dorm, as *? only), and (c) substitution (e.g., non-coordinative ??, dai4jia4, to replace-price, cost, as *?? by replacing the second character ? with ? from the coordinative compound word ??).
Motivated by the real-world learning problems in L2 Chinese reading, this dissertation study took multiple stages to investigate if and how inter-character orthographic and semantic information influenced the recognition of Chinese coordinative compound words by native Chinese readers and by CFL learners. The first two stages explored the decoding mechanism by native Chinese readers. Using a quantitative method and a lexical decision task, I found inhibitory inter-character orthographic similarity effects, intra-word character reversal effects, and inter-character semantic similarity effects during the recognition of transparent coordinative compound words by native readers. Next, I addressed character position encoding behind the character reversal effects in particular. I found that among native readers, character position encoding, as an integral part of visual word recognition, was modulated by the orthographic information of the constituent characters and the inter-character orthographic relationship within coordinative compound words. In the last stage, a mixed methods study explored the processing of inter-character orthographic and semantic similarities, character position encoding within coordinative compound words, and the recognition strategies by intermediate and advanced CFL learners in the United States. The quantitative strand revealed significant character reversal effects and significant inter-character semantic similarity effects. Such effects were validated and complemented by qualitative findings from a semi-structured interview on learners’ thought processes. The quantitative data showed no significant inter-character orthographic similarity effects, and qualitative findings validated that it was the meanings of component characters contributed by the shared radicals that influenced the recognition process. A mixed methods integrative analysis suggested pedagogical implications in (a) radical knowledge at the sublexical level, (b) morphological awareness of constituent characters for self-sufficient vocabulary building, and (c) handwriting in CFL teaching and learning.