Applying Phonetic Methods to Language Variation

Since Labov, Yaeger, and Steiner (1972) introduced acoustic analysis to dialectal variation a generation ago, sociolinguists have increasingly utilized instrumental techniques. While the number of studies involving acoustic analysis has grown steadily, nearly all of them have been concentrated on one narrow area: variation and change in the production of F₁and F₂ (the first and second formants) of vowels. Studies in that area have revealed a great deal about sound change. However, more instrumental research is needed on variation in other aspects of vowels, such as duration, phonation, and F₃ (third-formant) variation; on variation of consonants and prosody; and, most of all, on speech perception. Sound change should not be the sole focus of phonetic research on variation, either. It could be applied more often to studies of identity, and it could allow variationists to say more about the mental organization of sounds than they have recently.

The contributions of instrumental analysis to the understanding of sound change are substantial. Labov, Yaeger, and Steiner (1972) and subsequent studies described two shifting patterns found in English. One is the Northern Cities Shift, found in the Great Lakes region of the United States, which they related to developments in Scotland and Northern Ireland. The other is the Southern Shift, found in the southern United States and southern England and in Southern Hemisphere English. By comparing these patterns to historically attested vowel shifts in other languages, they identified four general shifting patterns. They proposed that vowels located on the periphery of the vowel envelope behave differently from those located on the inside, and from this conjecture they posited several principles of vowel shifting--for example, tense (or peripheral) vowels rise and low nonperipheral vowels become peripheral. Further testing of these principles is needed, and the reasons for them remain unclear. Nevertheless, numerous other researchers have utilized these findings as a basis for sociolinguistic studies of dialects influenced by either the Northern Cities Shift or the Southern Shift. In fact, there has been a veritable blossoming of instrumental studies of the F₁/F₂ patterns of vowels in those and other dialects.

In contrast, phonetic variation studies have largely passed over other features. There are only a few studies of duration or phonation of vowels. Other vocalic traits, such as intrinsic pitch and steady-state structure, have received even less attention. Consonantal variation remains in the domain of impressionistic studies. Intonation and other aspects of prosody have largely been ignored. Part of the reason is that measurement of some of these factors, such as phonation and intonation, is complicated. Nevertheless, language variationists need to begin investigating them.

Likewise, perception has not received the attention it deserves. Although the number of studies of variation in perception has grown, it is far outstripped by the number of production studies. This imbalance is serious because perception is as important cognitively as production. Numerous types of perception experiments involving various tasks, such as identification, discrimination, or judgment of stimuli, are possible. For some, natural stimuli are appropriate, while for others, filtering or other synthetic modification is more appropriate. They can be used to investigate such issues as stereotypes about dialects or particular variants, cues used for ethnic identifications, dialectal differences in perceptual boundaries between phonemes, different cues used by different dialects for the same contrast, and whether speakers recognize contrasts. Sociolinguists certainly need to investigate perception more intensively.

Most of the previous instrumental work on linguistic variation has been directed at the question of how sound change occurs. Many of these studies also touch on issues of ethnic, gender, generational, social group, or other identity. For the most part, these instrumental studies have variation and change of the sounds themselves as their main focus. Sociolinguistic studies focusing primarily on identity still rely largely on impressionistic transcription. However, studies of identity could employ instrumental techniques profitably. Such methods might uncover important but hitherto unnoticed variables or refine the understanding...