Page 48 - Fall2020

TRENDS IN SOUTHERN US ENGLISH
be interpreted as evidence of a decline in the robust cues to identification of Southern US English (i.e., dialect endangerment) or as evidence of a change-in-progress (i.e., dialect evolution). Research so far suggests that the distribution of unique speech elements, dominating the narrative of standardized Southern US English in the 20th century, shrinks, yielding way to mainstream pronun- ciations. However, other, less known elements that may contribute to the perception of Southernness (Gunter et al., 2020) and new elements that are not uniquely South- ern but may develop social meaning unique to the region are also attested. An example of the latter is a change in the pronunciation of the /au/ diphthong, first described in relation to Canadian English (think of stereotypical Cana- dian pronunciation of out, about, house) but then observed as far south as New Orleans (Carmichael, 2020).
Distribution of Nonunique Southern
Speech Markers
Although variation in the distribution of standardized, unique features of Southern speech such as /ai/-monoph- thongization is well-established in current research, the variation in nonstandardized but similarly widespread characteristics are less overtly discussed. An example of
such a feature that often flies under the radar of public consciousness is the degree of prevoicing of word-initial consonants /b, d, g/. Voice onset time (VOT) is one of the acoustic-phonetic characteristics of these consonants that varies across languages (/b, d, g/ in different languages are not acoustically the same) and within a language (the variation in /b, d, g/ may be socially meaningful). Figure 2 illustrates that the VOT duration (long or short) and quality (presence or lack of voicing) may result in differ- ent acoustic variants of these consonants.
This acoustic feature may serve as a sociolinguistic marker, for example, distinguishing between those who earn their livelihood off land (and tend to have par- ticular sociopolitical beliefs) and those who do not in inland California (Podesva et al., 2015), between those from North Carolina and those from Wisconsin (Jace- wicz et al., 2009), or between Black and white speakers in Mississippi (Herd, 2020). Although this marker is not talked about in the public sphere and is not reflected in the folk Southern orthography, American English speak- ers are aware of this distributional pattern at some level because they produce more prevoiced /b, d, g/ under the Southern guise than under the Standard English guise
Figure 3. Mean vowel plots of two Southern speakers. Isaac exhibits the Southern vowel shift in his speech (left) and Brittany does not (right). The acoustic quality of each vowel is represented by plotting a trajectory based on three sets of F1 and F2 values (presented in normalized units) recorded at three time points during vowel productions where the latest F1/F2 value is represented by an arrow. Color was added for vowels discussed in the text. Reproduced from Gunter et al., 2020, with permission of the
Acoustical Society of America.
48 Acoustics Today • Fall 2020