In lieu of an abstract, here is a brief excerpt of the content:

  • Secrets Revealed by Southern Vowel Shifting
  • Erik R. Thomas

Discussions of the vowel variants of Southern English have been extensive and have continued without interruption for over a hundred years. McMillan and Montgomery (1989) list several hundred works on the phonology of Southern English, a large portion of which cover vocalic variation. Dozens of works have appeared since McMillan and Montgomery's bibliography was published. No other region of the United States has attracted this level of interest in its vowels, either from scholars or in the popular press. With such sustained attention, one would expect that research on this variety had reached a point of diminishing returns. In fact, however, it has not. Southern English remains a trove of yet undiscovered dialectal configurations, for vowels as well as for other variables. A clearer understanding of these variants can aid in sociolinguistic descriptions, but in some cases it can also provide insights into the processes of sound change, phonological structure, and even speech perception. The several variables examined here demonstrate how Southern vowels can illuminate each of these latter domains.

Glide Weakening and Monophthongization

The three classes /ai/, as in tide, /oi/, as in boy, and /au/, as in loud, are all subject in the South to modifications of their glides that might be termed lowering, weakening, or, in some cases, monophthongization. For /ai/, this trait has long been regarded as a hallmark of Southern speech: a remark in Johnson (1928, 381) indicates that I and my with weakened glides were considered stereotypical of the South by the 1920s, and more intensive linguistic investigations began in the 1930s (Greet 1931; Edgerton 1935; Evans 1935). Glide weakening of /oi/ is also often mentioned as a Southern trait (e.g., Hartman 1985), but glide weakening of /au/ is only occasionally mentioned, and then often merely as a subregional feature (Pederson et al. 1986–92).

Glide weakening presents numerous problems regarding its description, contextual incidence, diachronic development, phonetic motivation, and sociolinguistic status. Although the sociolinguistic issues are largely [End Page 150] beyond the purview of this paper, the first four issues are closely interrelated. Plots showing the first two formants (F1 and F2) of each vowel can illustrate different variants and thereby shed light on these issues. All of the vowel formant plots in this article except for figures 1b and 3b are taken from Thomas (2001). Methods of acoustic measurement and selection of tokens are described in detail in that source. All the tokens were extracted from connected speech—either reading passages, conversation, or a combination of the two. The squares show mean values of all tokens measured for each vowel nucleus, and the arrows depict the gliding of diphthongs.

Click for larger view
View full resolution
Figure 1.

Diphthongs of Two Northerners: (a) a European American Male, Born 1940, from Lino Lakes, Minnesota; (b) a European American Female, Born 1887, from Westfield Center, Ohio (DARE informant OH 017)

Figure 1a shows a vowel formant plot of a Minnesota native who shows no glide weakening of /ai/, /oi/, or /au/. The mean values of the glides do not quite reach [i] or [u] values, since some truncation is inevitable in connected speech, but the approximation is fairly close. /ai/ is separated into two contexts, that before voiceless obstruents (/aiº/) and that before voiced obstruents (/aiv/). This convention follows Labov (e.g., 1991, 1994) except that contexts before nasals and approximants are excluded from /aiv/. Of import is the fact that the glide of /aiº/ approximates an [i] value more closely than that of /aiv/. Figure 1b plots the values of another Northerner, a native of northern Ohio (informant OH 017 for the Dictionary of American Regional English [DARE 1985–]). She also shows strong glides of /ai/ and /oi/, with the same contextual variation in her /ai/ glides. However, her /au/ is nearly monophthongal, her /au/ glide lowered to the point that it is barely differentiated from the nucleus. Lowering of the glide appears to constitute the main feature of glide weakening and is a necessary precursor to outright monophthongization. Glide weakening or monophthongization [End Page 151] of /au/ is most commonly associated with another non-Southern dialect, that of...


Additional Information

Print ISSN
pp. 150-170
Launched on MUSE
Open Access
Archive Status
Archived 2005
Back To Top

This website uses cookies to ensure you get the best experience on our website. Without cookies your experience may not be seamless.