The Production and Perception of Optional Persistent Tone 3 Sandhi as a Social Variable in Mandarin Chinese

Joshua L Perry, Purdue University

Abstract

Tone 3 Sandhi (T3S) in Mandarin Chinese is the productive tonal alternation whereby a falling-rising tone (T3) becomes a rising tone (T2) when followed by another falling-rising tone. This alternation can also undergo persistent application across syntactic and prosodic boundaries, and several phonological analyses show whether the persistent application of T3S is obligatory or optional. While different phonological analyses claim that optional persistent T3S is applied due to rapid speech or casual speech, there are no studies that use sociolinguistic methods to corroborate these claims. This thesis seeks to understand which social factors are associated with a higher production of optional persistent T3S in native speakers of Mandarin Chinese and whether optional persistent T3S is a product of a more casual style of speech. Additionally, it seeks to understand the attitudes that are formed by native listeners of Mandarin Chinese. The first study in this thesis used a modified sociolinguistic interview to elicit data from native speakers of Mandarin Chinese. Participants were shown stimuli of four different styles which each contained strings of words in Mandarin Chinese with more than two T3 syllables in a row: phrases, a list of numbers, short paragraphs, and comic strips. While the stimuli within each task type was presented in random order, block order was maintained in order to simulate a sociolinguistic interview. While technical error prevented responses from some task types to be evaluated, the data from the study suggests that more casual task types resulted in higher production of optional persistent T3S. Additionally, male participants were more likely to produce optional persistent T3S, and there was no effect of age of the participant nor the length of residence on the production of optional persistent T3S. The second study for this thesis used a matched-guise experiment to evaluate the attitude formation by native listeners of Mandarin Chinese for speakers who produced only obligatory T3S, optional persistent T3S, and no T3S alternations. The stimuli were controlled so that the only difference in each stimulus was the type of T3S that the speaker produced. Native listeners of Mandarin Chinese rated the speakers in the stimuli for their level of education, intelligence, successfulness, wealth, poshness, friendliness, kindness, trustworthiness, inherent goodness, and reliability. Two analyses were conducted on the data: one comparing the ratings for obligatory and optional persistent T3S, and a second which looked at all three T3S types. The findings from these analyses showed that participants were more likely to rate speakers who produced only obligatory T3S as being more trustworthy and reliable than those who produced optional persistent T3S. Additionally, participants were more likely to rate female speakers who produced optional persistent T3S as being more kind than male speakers who produced optional persistent T3S. Furthermore, female participants were more likely to rate speakers who produced optional persistent T3S as less wealthy than male participants rating for the same type of T3S production. Finally, participants over the age of 25 were more likely to rate speakers who produced optional persistent T3S as less reliable than those who produced only obligatory T3S, and participants who had resided in the US for less than one year rated speakers who produced only obligatory T3S as being more reliable than those who produce optional persistent T3S. The findings from first study suggest that optional persistent T3S is influenced in some part by the gender of the participant, and that future studies in the production of optional persistent T3S are necessary to confirm these hypotheses. Additionally, the findings from the second study suggest that speakers are able to perceive different types of T3S production and that they form attitudes of reliability, trustworthiness based on the production of optional persistent T3S. Finally, both studies indicate that further studies exploring the sociolinguistic factors of Mandarin Chinese are necessary in order to better understand how native speakers differ in their production and how native listeners perceive these differences.

Degree

M.A.

Advisors

Niepokuj, Purdue University.

Subject Area

Linguistics|Modern language|Sociolinguistics|Language

COinS