Pattern substitution in Wuxi tone sandhi and its implication for phonological learning

Abstract

Tone sandhi in Wuxi Chinese involves “pattern substitution,” whereby the base tone on the first syllable is first substituted by another tone, then spread to the sandhi domain. We conducted a wug test to investigate native Wuxi speakers’ tacit knowledge of tone sandhi and found that the substituion aspect of the sandhi is not fully productive, but the extension aspect is, and sandhi productivity is influenced by the phonetic similarity between base and sandhi tones. These results are discussed in the context of how phonological opacity, phonetic naturalness, and lexical frequency influence phonological learning, and a grammatical learning model that can predict Wuxi speakers’ experimental behavior is proposed.

A Preliminary Study on the Productivity of Mandarin T3 Sandhi in Mandarin-speaking Children

“…It is noteworthy that although the Mandarin-speaking children have not reached the adult-like level, their tone sandhi application rate in pseudowords is still higher than adults' application rate in pseudowords where the sandhi items are operated lexically. For example, the sandhi application rate of adults in AG items was below 30% for Southern Min and Wuxi Wu [6] [29], but it was above 70% for Mandarin children aged 4 to 6. This may suggest that even though children adopt lexical mechanism at the early ages, they did not entirely rely on the lexical mechanism, for the productivity is apparently higher than the items that were purely operated via lexical mechanisms.…”

Using Nonce-probe Tests and Auditory Priming to Investigate Speakers' Phonological Knowledge of Tone Sandhi

“…Take the Wuxi pattern as an example: the transparent spreading aspect of the sandhi can potentially be accounted for by the interaction between markedness constraints such as *TONE, *CONTOUR and faithfulness constraints such as MAX(Tone) and FAITH-ALIGN (which penalizes spreading, see [4]), but the opaque aspect of the sandhi needs to be encoded with listed sandhi tones on various levels of representation (e.g., syllable, morpheme, word) and constraints that force the listed forms to be used [49]. An implementation of this, which allows both the lexical patterns and the experimental results to be modeled, can be found in [42]. Other works that appeal to listedness to account for opaque tone sandhi patterns include [37] and [38].…”

“…Wuxi, therefore, has elements of both the Taiwanese and Shanghai types of sandhi. The productivity of the Wuxi tone sandhi pattern is tested in [42], and I summarize the results for T1-X, a cognate of the Shanghai 51-X pattern, here. Interested readers should consult [42] for more details.…”

“…The f0 pattern here looks markedly different from that in Shanghai. There are large differences in f0 on both syllables between real and novel words (see [42] for statistical comparisons). It appears that the substitution sandhi only occurred in the real words; in the novel words, the initial fall was spread instead, and there are also traces of the base tones in the surface tone patterns, e.g., the low dipping on the second syllable of 53-323.…”

“…These seem to correspond to the Shanghai pattern we observed in 2.2. I again refer the readers to [42] for statistics on these results. The f0 and the classification results together indicate that the two aspects of Wuxi tone sandhi have different productivity properties: similar to Taiwanese, opaque substitution lacks full productivity and often categorically fails to apply to novel words; and similar to Shanghai, transparent spreading, either from the base tone or from the substituted tone, is generally productive, but may apply incompletely in a phonetically gradient fashion.…”