Work

Contrast and Complexity in Chinese Tonal Systems

Year: 2017
Type: dissertation
Abstract: Lexical tone has been investigated from a diverse variety of perspectives, with significant disagreement as to whether a feature-based analysis can account for the synchronic and diachronic patterning of tonal systems. This thesis investigates the tonal systems of 13 Chinese languages from the Yue and Mandarin groups, and presents a feature-based model which accounts for their respective inventories and patterns of phonological alternations, and traces their respective diachronic trajectories back to the feature specification of a reconstructed Middle Chinese tonal inventory. The analysis here employs hierarchically arranged contrastive features. Whereas analyses postulating a universal set of phonetic features tend to incur a number of descriptive and theoretical problems when tasked with the description of tonal systems, utilizing language-specific features grounded in tonal alternations and phonetics sheds significant light on synchronic and diachronic patterning. The reconstruction of a common Middle Chinese proto-language enables the description of a precise phylogenetic tree as well as the postulation of a number of contrastive reanalyses. It is found that both reanalysis and the specific phonetic content of tonal inventories are accounted for by postulating a principle of complexity minimization: the cognitive apparatus chooses representations which minimize the algorithmic entropy or Kolmogorov complexity of phonologies and maximize the reuse of structure. At each point in the diachronic trajectory of a language, speakers choose the phonological representation with the lowest total complexity. Phonetic naturalness can also be formalized as the minimization of complexity, and a general tendency towards an inverse relationship between the naturalness of tonal inventories and the amount of activity in a given language is described. The cognitive preference for the minimization of complexity represents a parsimonious solution to the problem of induction which avoids the stipulation of hard-coded universals pre-specified in the cognitive apparatus.
Language: English
Field: Psychology
Open Access status: green
OpenAlex ID: W2621009770
Cites: 96
Cited by: 6
Related to: 10