What can diasporic languages teach us about the development of phonological distinctions?: Examples from Somali Chizigula Stops and Toronto Cantonese Vowels

While much research on sound change in progress has focused on phonemic mergers (loss of phonological distinctions) relatively little has focused on splits (either allophonic or phonemic) or on how new distinctions are created (Labov 1994, 2011). The goal of this presentation is to discuss two examples of the development of phonemic or allophonic distinctions in two different diasporic languages (with homelands in two different continents). I will argue that intense contact settings (as defined by Thomason & Kaufman 1988) facilitate the development of new phonological distinctions. Diasporic languages, thus, provide fertile ground for the development of new phonological distinctions because of their development in sociolinguistic settings involving intense language contact. The first example comes from the development of a plosive vs. implosive contrast in Somali Chizigula (Kizigua), a dialect of an East African language spoken in Somali Bantu communities in various US cities including Boise, ID, Columbus, OH, and Pittsburgh, PA. Consultant work shows speakers with minimal pairs such as [basi] ‘bus’ vs. [ɓasi] ‘enough’ and near minimal pairs such as [gasi] ‘gas’ vs. [ɠali] ‘expensive’. Thus, while implosives are found in inherited vocabulary, plosives occur in English loan words. The phonetic production of this contrast has been confirmed by acoustic and aerodynamic studies (Temkin Martinez & Rosenbaum 2017). Tanzanian dialects of Chizigula, however, have only plosives (Kisbey 1897, 1906; Mochiwa 2008) and interestingly these plosives correspond to implosives in Somali Chizigula. The development of this plosive vs. implosive contrast can be accounted for in terms of the migration history of Somali Chizigula speakers. Chizigula speakers from Tanzania arrived in Somalia in the 19th Century where they came in contact with speakers of other Bantu languages, including some that have implosives corresponding to Tanzanian Chizigula plosives. Thus, through contact, the inherited plosives shifted to implosives. As a result of the Somali Civil War, many Somali Chizigula speakers moved to Kenyan refugee camps, where they learned Swahili, and then continued on to the US where they learned English. At this point, loan words with English voiced stops became part of Somali Chizigula speech. Thus, the result of these distinct periods of migration (characterized by contact with speakers of different languages) is a plosive vs. implosive contrast. The second example involves the development of a pre-nasal split in /ɛ/ in Toronto Cantonese. Acoustic data analyzed as part of the Heritage Language Variation and Change in Toronto Project (Nagy 2011) shows that some second-generation speakers (grew up in Toronto) have developed a fronted variant of /ɛ/ that occurs before nasal consonants. This pre-nasal split appears to be influenced by a similar pre-nasal allophonic split found in Toronto English and many other North American English dialects (Boberg 2008). Further supporting an argument for contact-induced change is acoustic data showing its absence among first-generation (immigrant generation) speakers and among Homeland (Hong Kong) speakers. What this second example illustrates is that structural influence from the phonology of the dominant language (Toronto English) can lead to the development of an allophonic split. To conclude, what unites these two cases is intense contact, characterized by multilingual speakers who have access to multiple phonological systems and who are, thus, able to combine elements from their distinct languages in innovative ways. Creating new phonological contrasts and innovating allophonic splits are two examples of changes that can be initiated by multilingual speakers in diasporic settings. Although the extent to which these specific changes persist as features of these two diasporic varieties remains to be seen, what is clear from these two examples is the need to consider diasporic languages in developing models of sound change