Sociophonetic variation in Hong Kong and Heritage Cantonese: Understanding phonological representation and sound change in multilingual communities

GRF 2023/2024
(PI Jonathan Havenhill)
General Research Fund (GRF), University Grants Council (UGC), Hong Kong
Amount: 1,059,558 HKD

Abstract
All languages exhibit variation, with multiple ways of saying the same thing (eg, walking vs. walkin’). No two people speak exactly the same way, nor does one person use the same speech patterns in all contexts. This type of variation is far from random, and does not reflect poor linguistic command or laziness. Rather, speakers make productive use of variation to convey social information, construct personal identities, and navigate and shape their societies. Research on sociolinguistic variation has predominantly been carried out in monolingual, English-speaking communities. Yet the majority of the world’s population is multilingual, changing not only their speech style within one language, but also switching between languages appropriate to various contexts. Determining how multilingual speakers perceive and make use of variable linguistic forms is essential for understanding how speech sounds take on social meaning and how they are organized in the mind.

This project will investigate sociolinguistic variation in the multilingual environment of Hong Kong. The first study will collect spontaneous conversational speech from Cantonese speakers of different ages and backgrounds. The sibilant sounds “s” (as in Sai Kung) and “ts/ch” (as in Tsuen Wan or Wan Chai) have recently experienced change in their pronunciation, leading to a partial restructuring of the Cantonese sound system. This study will compare how sibilant pronunciation differs according to age, gender, socioeconomic status, and English proficiency, as well as how these sounds are produced under various topics and speech styles.

The second study will use ultrasound tongue imaging to observe the tongue shapes that are used in sibilant production. Participants will recite words in Cantonese and English, to test how speakers produce sounds that are similar in their first and second language. In addition to Cantonese-English bilingual Hongkongers, this study will include Heritage speakers of Cantonese in Canada. These individuals have spoken Cantonese with their families from an early age, but have high English exposure and proficiency through living in an English-dominant community.

The third study will experimentally test which social meanings are associated with Cantonese sibilant variants. Listeners will provide qualitative descriptions and quantitative ratings on a range of social scales, to examine how they evaluate speakers who use specific pronunciations.

Together, these experiments will provide deeper understanding of how speakers use and understand linguistic variants across multiple languages, how first and second languages exert both social and linguistic influence on one another , and how language contact contributes to variation and change.