The Myth of Disappearing American Dialects
"As everyone knows, because of the influence of the mass media, people's regional accents have almost completely disappeared. Nowadays, everyone sounds the same wherever you go..." How many times have you heard people echo this assertion? A few weeks ago I attended a local screening of a documentary about language. In the post-screening discussion, a middle-aged man stood up and delivered the above remark with great authority. Several people in the audience nodded their assent. Although intuitively this might seem to be the case, in fact people's characteristic regional accents are alive and well, and are growing ever more distinct from each other and from Standard American English, which is what we call the neutral-seeming, vaguely Midwestern-sounding speech patterns that the general public associates with network news anchors and educated speakers. And yet this specious postulate persists in the face of vast and easily observable dialect diversity all around us, and in direct contradiction to what has been studied and published by respected linguists. So what is really going on with American English?
The Speech Community
A speech community is a group of speakers who share a code or set of rules about how speech is articulated and language is used. These concepts, originally identified by ethnographers of communication John Gumperz and Dell Hymes, are key to understanding why the proliferation of mass media may not pose as great a threat to regional dialects as we might assume. In contemporary U.S. culture, most speakers simultaneously belong to many different speech communities, defined in terms of class, ethnicity, gender, level of education, or, in this case geographic region and are thus fluently conversant in several different codes. We all know people who deliberately and conspicuously change the way they pronounce certain words in order to advance at work, or be accepted within a desirable social set. Most of us continually and unconsciously make minute adjustments to our speech depending upon our interlocutors (i.e., partners in conversation), seamlessly shifting between registers, or modes of speaking (loosely corresponding to particular social identities), without even thinking about it. Mastery of a repertoire of codes is a basic social survival skill in a complex society. However, ethnographers of communication assert that it is the speech community in which we interact the most, especially over the course of a lifetime, that most shapes our primary speech patterns. Although they are exposed to many different manners and registers of speech all the time, people can pretty much be expected to talk like the people they spend the most time talking with--their social peers--be they college students, Wall Street executives, Maine lobster fishermen, Alabama middle-schoolers, neighbors from South Philadelphia, or Boston cops.
Take the coast of Maine, which at one time represented a relatively insular cultural area. Maine is a poor state, and movies, television, art and literature have classically depicted its local residents as provincial, even backward. However, since the 1980's or so, all but the very poorest of homes boast cable TV or a satellite dish and lately high-speed Internet access as well. "Mainers," as Maine people call themselves, can get every major newspaper. They (and presumably dwellers in other hinterlands around the U.S.) now shop at the same big box stores as the rest of America, watch the same movies, listen to both Howard Stern and NPR via satellite radio, and generally enjoy exactly the same access to news media and pop culture as anyone living in Boston, New York or LA. In spite of this, the characteristic local speech pattern in rural Maine is instantly identifiable by its palatalized sibilants, particular diphthongs and mergers, drawling cadence, and dry, ironic humor. In fact, all one needs to do in order to satisfactorily disconfirm the notion of dialect homogeneity is to go hang around a general store during hunting season (when there are likely to be hunters coming in for hot coffee, doughnuts, and bragging rights) and pay attention as they discuss the morning's luck, local politics, the economy, their families, and the price of gas. You will not need your GPS to know you're in Maine. The same goes for the local TJ Maxx, where you are more likely to find women talking with each other about the circumstances and ideas that concern them. In either case an attentive eavesdropper is likely to hear versions of typical Maine speech, flavored by the nuances of gender and class perhaps, but endemic to Maine nonetheless.
Variation and Change
Language is dynamic. Words change, phonemes (the units of sound known as vowels and consonants) change, and grammar changes, even within a speech community, and sometimes abruptly. This kind of punctuated equilibrium runs counter to our expectation that language changes gradually and imperceptibly, over great stretches of time. Dialect geographers study the boundaries between regional dialects as defined mainly by vocabulary, and try to identify isoglosses, coincidences of the outermost reaches of stable and emblematic vocabulary items. The distributions of usage for variants like hoagie/sub/grinder and shake/cabinet/frappe, to name a couple, help to define major dialect areas and act as cues for sociolinguists, who want to study the accompanying variations in phonology. They then employ ethnographic methods, linguistic theory (in particular historical linguistics and phonology) and sophisticated recording and analysis technologies to identify, analyze and map systematic changes in the way speakers produce certain sounds, mainly vowels. Why, they want to know, does language change at all, especially when changes in the way that words are pronounced may impede, rather than facilitate, communication? What are the primary mechanisms of change?
Some scholars who are primarily interested in answering this last question concern themselves mainly with three kinds of processes common to the behavior of vowels: shifts, mergers and splits. A shift is a change in the place of articulation, within the oral cavity, of a particular vowel. Vowels can be articulated higher or lower, more front or more back in the mouth. If one pictures a roughly football-shaped space, the high front vowels like /i/ and /e/ occupy the prime real estate up there near the lips. /u/ is a high back vowel (Try saying the vowel in the word 'shoot.' The vowel is shaped towards back of your palate, but up high, which is why the lips are pursed. Now say the vowel in the word 'oar,' noting how you must lower the back of your tongue and round the inside of your mouth to achieve it. Thus, vowel shifts are talked about as being directional. They are also sometimes systematic, involving entire classes of words, or sets of vowels in what Labov (1994) identifies as a chain shift. A good example of this would be the Great Vowel Shift, in which a series of changes in the long stressed vowels of Early Modern English produced the diphthongs we modern English speakers use in words like 'mouse' and 'wife,' and in which an initial shift prompted articulation of the surrounding vowels to shift in reaction along the same directional track.
A merger is when words that sound the same except for one vowel start being pronounced as homonyms, as has been the case in some dialect areas of the U.S. with the vowels in the word 'dawn' and the name 'Don' both being pronounced as /don/. This is referred to as the Don/dawn merger, or the merger of /o/ and /oh/ (in linguistic notation). Ordinarily such mergers do not reverse themselves, and they can cause misunderstandings between interlocutors in natural speech. This is one of the paradoxes of language change, that such changes can actually inhibit communication!
A split is the opposite of a merger--i.e., a phoneme develops two distinct variants, or allophones, and this alternation in pronunciation gains vigor within a speech community. Ultimately one of the two begins to have more prestige associated with it than the other, and the other may even drop out of common usage altogether. This kind of variation is a necessary condition for linguistic change to occur.
As to the question of identifying a catalyst for these kinds of phonological changes, most scholars agree that the answer should lie in studying the intersection of language and social behavior. They are interested in understanding the different social norms that influence speech, as well as the various social identities that speakers embody. They want to know what speakers lead most changes in progress (women, the young, and upper-working to lower-middle class according to Labov (1994 and elsewhere), whether changes within a community tend to be gradual or abrupt (abrupt), and how much individuals’ speech systems ordinarily change over the course of a lifetime (not much).
Language and Social Life
Humans are symbolically oriented beings, and although the most basic evolutionary function of language was probably to communicate descriptive information about the material world, language became an essential element of culture, our species’ main adaptive trait. A speaker’s tone, gestures, inflections, choice of words, register, dialect, and style send both blatant and subtle (and sometimes mixed!) signals about him or herself, the listener, the relationship between them, and the social context of the utterance (that is, the social world). Thus, in addition to referential practicality, language also functions as a sophisticated indexical system, delivering many layers of information at one time, reflecting the social landscape, but also shaping it.
We might logically ask at this point, wouldn’t this support the assumption of homogenization of American speech? After all, we know that people tend to affect the habits and symbols of others whom they perceive to be savvy, connected, or powerful. Why wouldn’t speakers mimic the media standard to the detriment of their native vernaculars? First, prestige and power are not straightforward, but instead represent fluid and nuanced concepts and structures. What represents a prestige form in one social context may not be in another. Language is an emotional issue for most people, tied in as it is with ethnicity, class, region and education. Second, whereas Standard speech often represents a tool for maintaining social distinctions, people actively cultivate their local accents (and native languages in multilingual situations) as a way of creating solidarity. If people ‘speak the same language,’ we assume they probably also share some other common normative ground, and this may even be based on indexing the contrast between their local speech and the Standard. Thus, although language and culture are not one and the same, language is definitely emblematic of culture and identity for most people. Just as the cultural landscape presents a picture of rich diversity in defiance of the ideologically supported 'melting pot' metaphor so commonly evoked in social discourses of the twentieth century, the current vigor of America's regional dialects and observed trajectories of change defy the primacy of a generic Standard.
Gumperz, John and Dell Hymes. 1972. Directions in Sociolinguistics: The Ethnographyof Communication. New York: Holt, Rinehart and Winston.
Labov, William. 1994. Principles of Linguistic Change Volume 1: Internal Factors.Malden, MA: Blackwell Publishers.
Labov, William. 2001. Principles of Linguistic Change Volume 2: Social Factors. Malden, MA: Blackwell Publishers.
Carver, Craig M. 1987. American Regional Dialects: A Word Geography. Ann Arbor:University of Michigan Press.