2.5 Front vowels

The front vowel shift of the SVS includes the high vowels /i/ and /ɪ/ and the mid vowels /e/ and /ɛ/. The nuclei of both of these pairs are alleged to approximate switching places in acoustic space⁵ under Labov’s (1994: 209-212) “Pattern 4”, with the tense vowels becoming nonperipheral and lowering and the lax vowels becoming tense, peripheralizing and raising.

However, while the suburban Huntsville area does participate in the front mid vowel shift, the situation regarding the high front shift of the SVS is much less straightforward, with /i/ and /ɪ/ often remaining stable. This is corroborated by research that Fridland (2000) conducted in Memphis, Tennesee, with speakers interviewed in 1996 and by Bailey (1997) looking at Southerners (from various places, but especially Texas) over a period of over 100 years (mid-19^th century to the 20^{th century), whereby the shift in front high vowels was not very advanced. In contrast, Feagin} (1986) and Labov (1997) did find this shift in Alabama, in Anniston in the early 1970’s and in Birmingham in the late 1980’s, respectively.

As part of the Cross Dialectal Comprehension project, Labov and Ash (1997) interviewed 15 participants from the Birmingham, Alabama, area, in 1988 and 1990. While this study focused primarily on the comprehensibility of words that were shifted according to the SVS, it closely examined three females from whom data were recorded: Wendy P., Melanie O. and Alison K., who were age 18 in 1988, 24 in 1988, and 16 in 1990 respectively. Relevant here is that, for all three females, the high and mid front vowels have exchanged relative position in acoustic space according to the pattern of the SVS; the lowering of /e/ was particularly advanced.

Feagin (1986) and (2003) reports SVS findings in speakers from Anniston, Alabama, recorded in the late 1970s. In the former, she looked at seven speakers from her data pool and the latter, twenty. Each study examined the progress of the SVS based on apparent-time analyses. Her acoustic measurements showed many of the same patterns Labov found: the gradual reversal of both high and mid front vowels, with the oldest man showing nascent signs of front vowel movement and the oldest woman showing completion of the reversal of /e/ and /ɛ/ and some movement of /i/ and /ɪ/. Most importantly, the youngest speakers of both sexes showed a completion of the shift of both high and mid front vowels.

Fridland (2000) examined the behavior of the Southern Shift in Memphis, Tennessee. Her study was carried out using speech from 25 participants who were recorded in 1996. Despite the above findings, her results show that Memphis speakers displayed an extremely active reversal of /e/ and /ɛ/, but rarely showed any signs of movement in front high vowels, with no one having anything close to an exchange of these vowels.

Further, Bailey (1997) uses old records of Southerners dating from the mid to late 19^th century, additionally bringing in analyses of 20^th century speakers from Texas. There is evidence that the time period of the early 20^th century through the present has been the window for the change involving backing and lowering of /e/. The high front vowels from speakers of birth dates ranging from 1896 to 1976 seem to have remained in the traditional places, i.e. unshifted, much like Fridland’s findings. The variability of the behavior of the high front vowels is indirectly noted in Tillery and Bailey (2008: 124). While elaborating on the parts of the Southern shift, the sub-shift involving the high front vowels is qualified as occurring only “in some parts of the South….”

It is important to additionally note that the front vowel exchange may not occur if the vowel precedes syllable-final /l/. In fact, this environment often results in a merger or near- merger, where both pairs – /i/ and /ɪ/ as well as /e/ and /ɛ/ - are merged, with the resulting sounds being those of the lax vowels. This merger among Southerners has been noted by several researchers. For example, Feagin (1986) notes a merger before tautosyllabic /l/ in Anniston, Alabama (141). Thomas (2008) and Tillery and Bailey (2008) all describe this merger as well. Thomas (2008) describes it as a feature of young, rural, white Southerners and claims that the high front vowels are merged before /l/ more often that the mid front vowels are (316). Tillery and Bailey (2008) discuss current mergers affecting the urban South, including those before /l/. They point out that, as a result of the mergers, there are only three front vowels in Southern English that occur before /l/: /ɛ/, /ɪ/, and /æ/. While the area from which the data for the current project was collected straddles the line of urban and rural, this phenomenon may actually surface in all types of Southern areas, whether urban or rural, which is borne out in the data below.

Interestingly, the merger has also been found outside of the Southeast. It is described in some depth by Di Paolo and Faber (1990) using data from Utah in the West. Di Paolo and Faber (1990) conclude that it is only an apparent merger since listeners were usually successful in distinguishing the pairs in tests. Upon investigation, the distinction was due to creaky voice in one phoneme and not in the other. Labov et al (1972) also discuss a merger of the tense and lax pairs of vowels in Salt Lake City, Utah, and in Albuquerque, New Mexico.

While Di Paolo and Faber (1990, 20) suggest that it may be the case that “Utah English is beginning to participate in the Southern Shift,” the various findings seem to suggest that the mergers are either not necessarily indicative or no longer indicative of the SVS. The mergers may be more widespread than they were previously or than they were previously believed to be. Evidence for the mergers is available via popular media as well. For example, in the song, “Damn it Feels Good to be a Gangster,” the artist group Geto Boys rhymes bills with deals, which suggests the merger exists in AAVE as well. Additionally, in an informal survey young women and girls from across America show variable merging of these front vowels. The young women and girls were participating in a model search for the clothing boutique Wet Seal. The contest entries were available on the website You Tube⁶and in each entry, the young females used the word seal at least once and up to five times. The 157 contestants whose videos were viewed self-reported being from all over the United States, ranged from 13-25 years of age, and included several different ethnic or racial backgrounds. While Southerners and African Americans appeared more likely to pronounce seal as sill, not all of them did, or it varied within the video. More interesting is that females from various backgrounds and regions (not only Southerners and African Americans) pronounced (completely or variably) seal as sill. This informal information combined with the work of linguists such as Di Paolo and Faber (1990) suggests that the merging or near-merging of front vowels may variably occur nationwide.