A history of the English Language

250.
The American Dialects.
Certain features of pronunciation characteristic of a part of New England and others 
associated with many parts of the South are so easily recognized and so well known that 
A history of the english language 354

---

for a long time it was customary to distinguish three main dialects in American English—
the New England dialect, the Southern dialect, and General American, meaning the 
dialect of all the rest of the country. Such a division, in a broad way, is not unjustified 
because each of the dialect types is marked by features that distinguish it clearly from the 
others. But it is not sufficiently exact. Not all of New England shares in the features—
such as the so-called “broad a” and the loss of [r] finally and before consonants—that are 
thought of as most characteristic. Parts of the South were settled from Pennsylvania and 
are not typically southern in speech. And finally, General American itself shows regional 
differences which, although not so obvious to the lay person, can be recognized by the 
linguist and charted. 
Our ability to distinguish more accurately the various speech areas that exist in this 
country is due to the fact that we now have a large mass of accurate data gathered by field 
workers for the 
Linguistic Atlas of the United States and Canada
(see page 399) and a 
growing number of detailed studies of regional pronunciation and other features. These 
have contributed greatly to a clearer understanding of some of the speech areas of the 
country.
27
In 1949 Professor Hans Kurath published a study of the first importance, 
A Word 
Geography of the Eastern United States
. On the basis of lexical evidence, mainly in the 
Atlantic Coast states as far south as South Carolina, he distinguished eighteen speech 
areas, which he grouped into three main groups: Northern, Midland, and Southern. 
Positing a Midland dialect had the effect of taking parts of what had been considered 
General American and
26 
See J.S.Kenyon, “Flat 
a
and Broad 
a,
” 
American Speech, 
5 (1930), 323–26. 
27 
The following studies may be mentioned by way of illustration: Hans Kurath and Raven I. 
McDavid, Jr., 
The Pronunciation of English in the Atlantic States
(Ann Arbor, 1961); E.Bagby 
Atwood, 
Survey of Verb Forms in the Eastern United States
(Ann Arbor, 1953), and the same 
author’s 
Regional Vocabulary of Texas
(Austin, 1962); and Craig M.Carver, 
American Regional 
Dialects: A Word Geography
(Ann Arbor, 1987). An account of the more important dialect areas 
will be found in the chapter contributed by McDavid to W.Nelson Francis, 
The Structure of 
American English
(New York, 1958). An excellent overview that includes social dialects is by Walt 
Wolfram and Natalie Schilling-Estes, 
American English: Dialects and Variation
(Oxford, 1998). 
The newest description of American dialect areas is by William Labov in the Atlas of North 
American English: www.ling.upenn.edu/phono_atlas. For the publication of regional atlases and 
dictionaries, see § 255. 
The english language in america 355

---

THE DIALECTS OF AMERICAN 
ENGLISH 
Southern and carving out a third major dialect, straddling the traditional boundary and 
extending from the Middle Atlantic area to the Mississippi and beyond. This area was 
divided into North Midland, which comprised most of Pennsylvania and the central areas 
of the Great Lakes states; and South Midland, which continued to be referred to as the 
Upper South or the Southern Uplands, and which included the southern Appalachians, the 
Ozarks of Missouri and Arkansas, and parts of Oklahoma and Texas. Subsequent studies 
have supported or modified particular isoglosses and dialect boundaries, the results 
varying from study to study with the phonological and lexical criteria used. The main 
point of controversy has been whether the selection and weighting of isoglosses supports 
a distinct Midland dialect. Although Kurath’s tripartite division was widely accepted for 
forty years, recent investigations have led some dialectologists to reemphasize a primary 
North-South linguistic boundary. Craig M.Carver, for example, proposes