Background on Linguistic Atlas of New England
In 1921, members of the Modern Languages Association first conceived of a large-scale project to record linguistic information about regional variations in English language usage in the United States, a project that would gradually evolve into the Linguistic Atlases of the United States and Canada. After several years of discussion and planning, the MLA's Present-Day English group formally proposed undertaking the project in December 1928, and coincidentally less than a month later, E. H. Sturtevant of the Linguistic Society of America proposed a broadly similar project to the American Council of Learned Societies.
With the encouragement of the ACLS, the two initiatives joined forces at a planning conference in March 1929 and a coordinating committee was formed, chaired by Hans Kurath and including other prominent linguists such as Leonard Bloomfield, Eduard Prokosh, and John S. Kenyon. The initial plan was to carry out an ambitious project to canvas speech patterns throughout the United States and Canada, but when faced with a proposed budget of $664,000, the Executive Committee of the ACLS suggested beginning with a smaller-scale study of one region, centered on the dialects of New England. Kurath organized a staff for the proposed Linguistic Atlas of New England (LANE), and by July 1931, a cadre of linguistic fieldworkers was trained and set to work.
The primary objectives of the LANE project were to define the dialect areas of New England and to discover both innovations in speech and relics. In each community the canvassed, fieldworkers sought to include interviewees (informants or consultants) from each of three broad categories:
- An elderly "descendant of an old local family," who might help "establish the regionalism of the pre-industrial era of New England in considerable detail, with the possible exception of the highly urbanized area around Boston."
- A middle-aged person, "native to the community, who had received better schooling" and who had "read more widely or enjoyed contact with the better educated."
- "Cultured informants, with a college education or the equivalent, were to be chosen in most of the larger cities... and a number of smaller communities" (Kurath 1939: 41).
Although it was not always possible in a given community to find consultants for each of the theoretical categories, fieldworkers followed a rigorous protocol based on a highly structured and consistently applied worksheet. Spending from six to twenty hours with each consultant, fieldworkers elicited responses to a questionnaire regarding 814 words and phrases in 711 numbered entries, touching on a range of everyday activities from work and family to domestic goods, numbers and expressions of time, clothing and bedding, animals and calls to animals, agriculture, and social relations. By the time of the last interviews in September 1933, the survey included a total of 416 consultants in 213 communities in New England, whose responses were painstakingly transcribed using the International Phonetic Alphabet to preserve nuances of pronunciation. Although no sound recordings were made in the initial survey, a few aluminum disk recordings were made in follow up work between 1934 and 1939.
After analysis and editing, the interviews resulted in the publications Handbook to the Linguistic Geograpy of New England (1939) and the three volume Linguistic Atlas of New England (1939-1943). The methodology pioneered in New England was replicated in other regions across the country, including the Gulf States, Middle and South Atlantic, North Central, Oklahoma, Pacific Northwest, Pacific West, Rocky Mountain region, and Upper Midwest, as well as the African American and Gullah dialects of the southeast. Work on the Rocky Mountain states project was still in progress as of 2012.
