Twentieth-century changes

The twentieth century brought different demographic movements and associated linguistic change. Initial settlement of the western part of the country by home-steading was essentially complete, and demographic change then occurred by internal migration. In the first half of the century Southerners both black and white left the untenable agricultural conditions of their region and looked for new opportunities in the North and West. In the second half of the century Northerners sometimes moved away from the Rust Belt in search of opportunities in emerging industries in the South. These population movements often created speech islands in the regions to which the migrants traveled, such as African American or Southern White neighborhoods in Northern cities.

The greater change, however, stemmed from an essential change in the urban demographic pattern from residential neighborhoods within cities to the model of an urban core surrounded by suburbs. Suburban housing changed the essential interactions of the community, because people no longer lived with the people they worked with: in sociolinguistic terms, suburban social networks often became characterized by weak ties (i.e., the density and multiplexity of linguistic interactions decreased. In addition, because American suburban housing has most often been economically stratified, the social networks that did develop were more likely to be class-bound, unlike the situation in older cities where there was more mingling on a daily basis between people of different economic registers.

At the same time that suburban residential patterns were developing, improvements in transportation (highways, airlines) created a super-regional marketplace for the highly educated. While the American population has always been mobile, the most highly educated segment of the population has become nationally mobile to a much greater extent than the working and lower-middle class population, which tends to move around locally, often within the same metropolitan area or the same state. This change has led to the growth of the notion that highly educated speech should not show evidence of regional affiliation. Highly educated speakers in formal settings tend to suppress their regional features (to the extent that they have them in the first place, owing to suburban housing patterns. The typical speech of national news broadcasters is symptomatic – not a cause of the change, as many suppose.

The contemporary situation for StAmE pronunciation, then, is that the most highly educated speakers in formal settings tend to suppress any linguistic features that they recognize as marked, i.e., regionally or socially identifiable. Many educated speakers therefore think that language variation in America is decreasing. On the other hand, the economically-stratified suburban residential pattern promotes the continued existence, even expansion of local varieties (cf. Labov and Ash 1997: 508), though perhaps varieties with fewer strongly marked characteristics than were maintained before in the previous era of stronger, denser ties in social networks. American English, paradoxically, in some ways has more local variation than ever before, at the same time that in other ways it has less variation than before. The linkage between demographic trends and education remains the central fact for any discussion of standards in American English: those with the resources to proceed the furthest in the educational system have the greatest commitment to and investment in the idea of linguistic standards, now expressed particularly through their suppression of marked regional and social characteristics, while those with fewer resources and less investment in the educational system generally accept the idea of formal educational standards but do not routinely enact them in their own linguistic behavior. That said, it is of course true that many educated speakers value their regional affiliations and refuse to suppress, or even take pride in the display of, their regional speech characteristics, while some speakers without a high level of educational achievement may choose to suppress their regional features.