| Abstrakt: |
Cross-linguistic corpora of speech errors in Dutch, English, German, and Mandarin are examined for evidence regarding the cognitive knowledge of Mandarin speakers in terms of representation and processes of their phonology and mechanisms in language processing. Corpus analysis in Mandarin slips shows that the speech mechanisms in relation to types of errors and classification are universal, whereas the causes of error are language-specific. Evidence from Mandarin slips confirms a number of systematic patterns found in Germanic languages in which the majority of speech errors are phonological errors, substitution errors occur mare frequently than addition and omission errors, the source unit that influences the production of the target unit usually occurs within the context of the utterance, and the distance between them generally does not exceed seven syllables. The discrepancy between Mandarin and Germanic languages shows up most clearly in the treatment of directionality of errors. Evidence from Mandarin slips shows that perseverative errors occur more frequently than anticipatory errors in both segmental and tone errors Issues as to whether such lapses provide evidence regarding the cognitive processes involved in speaking will be suggested. This study aims to show the value of speech-error evidence for use in comparison with cross-linguistic findings, and hors such findings are incorporated in phonological constructs and psycholinguistic models of speech production planning. [ABSTRACT FROM AUTHOR] |
