This article investigates the dialect boundary that separates the eastern and western halves of North Carolina using data collected by two very different projects, the Linguistic Atlas of the Middle and South Atlantic States (LAMSAS) and the Corpus of American Civil War Letters (CACWL). Using LAMSAS data collected in the 1930s from speakers born in the late-nineteenth century, Hans Kurath posited that such a boundary extended across North Carolina northward into Virginia and southwestward into South Carolina and that it demarcated the South Midland speech region from the South. An analysis of lexical and grammatical variants in the 2,299 letters (representing one million words) of the CACWL North Carolina sample provides important real-time confirmation of Kurath's claim for the boundary. It also suggests that the boundary was established well before the beginning of the Civil War. No doubt the South–South Midland boundary in North Carolina was in large part a consequence of the migration of Scotch-Irish and German settlers from Pennsylvania into the western half of the state before the end of the eighteenth century. However, while the Scotch-Irish influence appears to have remained strong and perhaps even spread in the late nineteenth and early twentieth centuries, the German influence was already on the decline toward the end of the antebellum period and had almost disappeared by the time the LAMSAS fieldwork was carried out.

The text of this article is only available as a PDF.
You do not currently have access to this content.