Phonetic details of coronal consonants in the Italian spoken by Italian-Australians from two areas of Veneto

BACKGROUND AND AIMS. In an age of globalization that is seeing movements of large groups of people across vast distances, immigrants must adapt to their new home across multiple dimensions. Immigrant groups moving to a new country with a different dominant language tend to maintain their heritage language(s) as part of their cultural identity, while at the same time they must acquire and use the language of the host country for everyday communication. The focus of this study is how people negotiate between their home language(s) and their new language even at the level of the fine-grained features of their speech in each. Specifically we will address: 1) the extent to which fine-grained details of their mother tongue(s) may continue to provide unique markers of their linguistic origins; 2) if, in the constant interaction between the native linguistic repertoires (Auer, 2005) and the new language, any of their linguistic repertoires shows evidence of phonetic drift and, if it does, in which direction. As for the latter question, we ask whether, with respect to L3-English, the local dialect as L1 will show more resistance to attrition than Italian as L2; whether both L1 and L2 drift towards English and, if they do, whether they show the same type of drift. With respect to L1-dialect and L2-Italian, we ask whether any convergence has emerged, resulting from dialectization of Italian and/or italianization of dialect (Cerruti & Regis, 2011). To this aim, we tested whether Italian emigrants who settled in Australia maintained, lost or changed specific phonological and phonetic features of their Italian after five decades of living in the new country. Italian-Australians constitute a trilingual community of speakers: the local dialect as L1 coexists with Italian as L2 (usually learned at school in Italy) and English as L3, learned at their arrival in the host country. Among a corpus of Italian-Australians we have been collecting during the past years, we selected for the present analysis four first generation speakers: speakers who emigrated to the area of “Greater Sydney” as young adults during the second mass immigration wave (1947-1976). The phonetic features of their dialect have been the topic of a previous study of ours; in the present study we focus on their second heritage language: the regional variety of Italian spoken in Veneto. We will discuss the maintenance/change/loss of a given set of phonetic and phonological properties of their regional Italian comparing the results with what we already know about the conservation of the the detailed speech characteristics in their native Dialect. METHOD. Our subjects are 2 males and 2 females originating from Veneto, who have a similar number of years of experience of English (50-57). Two of them (one male and one female) are native speakers of the Northeastern Veneto dialect (centered on the towns of Belluno, Treviso, Feltre) and the other two are speakers of the Central Veneto dialect (centered on the towns of Padova, Vicenza, Rovigo). The two sub-systems of the Veneto dialect have been selected because they show interesting differences in the phonetic properties of the class of coronal obstruent consonants. These consonants are especially useful for examining interlanguage effects in immigrant groups, as they show a wide range of variation in fine-grained details across languages and regional accents and, relatedly, tend to undergo phonetic shifts over time and distance (language change). The Northeastern Veneto system presents the following class of coronal obstruents: /t, d, θ, ð, s, z, ʧ, ʤ/; the Central Veneto system has /t, d, s̪, s, z, ʧ, ʤ/ (Trumper, 1972; Mioni & Trumper, 1977; Zamboni, 1988); in both systems the coronal stops are dental, as in Standard Italian. Notably, in neither phonological inventory are dental affricates present, while they are part of the inventory of SI; conversely, in SI (inter)dental fricatives do not occur. English lacks dental affricates too (even if they can be produced as result of phrasal-level phonological processes), and presents voiced and unvoiced interdental fricatives. Our corpus amounts to 619 occurrences of stop, fricative and affricate coronals spontaneously produced in Veneto Italian by the speakers. The set of coronals that are the object of our analisysis is: /t, d, ʦ, ʣ s, z, ʧ, ʤ/. The speakers, recorded in Australia, were shown on a computer screen a set of images. Their task was to describe the images and freely comment on their use and/or narrate any memories they had related to them. As the referent of each picture is designated by a name with the desired coronals, the speakers were solicited to spontaneously produce several repetitions of a target word. This task is part of an interview in which the speaker is guided through different steps. After a phase of attunement to the local dialect via the presentation of speech excerpts of Veneto monolingual speakers and an interview in Veneto dialect, the speakers were presented on a computer screen the set of images depicting the target words. Their first task was to describe them in dialect. In the second step, the same images (in a different random order) were presented again, but the task was to describe and comment them in Italian. This step was preceded by an attunement phase to Italian. Each occurrence of the target words was segmented at phone level using Praat, and each target consonant was IPA transcribed and acoustically analysed. For the class of fricatives and for the fricative release of the affricates we measured duration, RMS, and the four spectral moments. RESULTS. Results indicate that the process by which intervocalic dental stops get lenited to dental fricatives or approximants in dialect is not transferred to Italian by any of our speakers: neither [ð] nor [ð̞̞] occur in our corpus. Italian appears to be resistant to a potential drift toward dental fricatives that both the local dialect and English could exert, given that in both phonological inventories such phones do exist. As for [θ] and [s̪], the two coronal fricatives that act as identity marks of Northeastern and Central Veneto dialect respectively, our data show only 1 occurrence of [θ] attested for each of the 2 Northeastern Veneto speakers and 3 occurrences of [s̪], attested for one of the 2 Central Veneto speakers. However, the Italian speech of all speakers shows signs of marked variability in the set of dental affricates: out of 52 cases of expected affricates in Standard Italian, only 24 are realizes as such, while 28 tokens are realized as fricatives. Among those, Northeastern speakers use [θ, s, z], Central speakers use [s̪, s, z]. Such a strong variability restricted to this specific set of consonants opens up to different explanations. It can be seen as conservation of Veneto Italian, where deaffrication of [ʦ, ʣ] is attested as one of the possible realizations of such consononants, on a par with their production as a stop-fricative sequence [t-s; d-z] (Canepari, 1984). Alternatively, it could be interpreted as a selective attrition exerted by L3-English that enhances and reinforces the allophonic tendency of Veneto Italian. Moreover, we will present data on spectral properties of the regional Italian fricatives; referring to spectral moment values, we will compare the Veneto Italian coronals with those produced by our speakers in dialect. The comparison will allow to discuss whether fine-grained details of the informant’s speech in their mother tongue(s) continue to provide unique markers of linguistic origins.