Title
Linguistic Geography in the United States
Author(s)
Chinen, Hirozane
Citation
沖大論叢 = OKIDAI RONSO, 2(1): 1-13
Issue Date
1962-03-01
URL
http://hdl.handle.net/20.500.12001/10764
Linguistic Geography in the United States
Hirozane Chinen
CONTENTS
Ch. I
Introduction
Ch. II
Early Dialect
Investigation~Ch.m Linguistic Atlas Project
Ch.
IVValidity of the Atlas
Ch.
v
Defects or Limitations of the Atlas
Ch.
VIProgress of the Atlas
Ch.
w
·
Principal Dialect Areas
Ch.
VliiApplication of the
DataBIBLIOGRAPHY
-The purpose of this paper is to introduce and examine American Lin· guistic Geography with a special attention to "Linguistic Atlas of the United States and Canada".
The specific aims of the study are
1 ) To make a study of the Atlas project.
2) To pursue how far the Linguistic Atlas project has been carried out.
3) To investigate the merits and demerits of the project and method. Ch. II Early Dialect Investigations
Early dialect investigations in the United States were rather of varying quality. Some were merely a kind·:of .w:opdlist, while the others were serious investigations. One of such serious s.tudies is 0~ F. Emerson's The Ithaca dialect (1891)1
The American Dialect Society has aimed at the compliation of an American dialect dictionary comparable to the major study of British dialects, Joseph Wright's English Dialect Dictionary (1895-1905) , since its establishment in 1889.
According .to Marckwardt2 , the American Dialect So~iety remained rela-tively small, and though its journal Dialect NoteS,: presented some valuable information, a systematic investigation of regional American English has not yet come out of the society.
1 Francis, Structure of American English, P. 788. 2 Marckwardt, Americag English, P. 132.
3 Dialect Notes~ published by the Americ<'.n Dialect Society, ed. by the suc-cessive secret<'ries of the Society (by M. L. Hanely since 1930) , Norwood, Mass., 1890-99; New Haven, 1900-.
Consequently, the Linguistic Atlas project was significant because "it became apparent that a linguistic atlas of the United States and Canada was not only more feasible as an immediate objective, but would actually simplify the work of the proposed dialect dictionary by delineating the areas to which lexical investigators should devote the greatest amount of time" .I
Ch. Irr Linguistic Atlas project
The Linguistic Atlas Project was set up in 1930 under the sponsorship of the American Council of Learned Societies and the general direction of Hans Kurath.2
1. Selection of Communities
The principles on whitch the selection of communities are based are as follows. In each area, a network of communities is selected for · investiga_ tion. In selecting these communities the study of economic and cultural histocy of the region is fundamental. The communities must be picked at relatively even intervals yet reflecting the population density. Some other criteria for the selection of a community for investigation are (1) its early permanent settlement history, (2) its historical role as a key station of migration, (3) its economic or cultural status, (4) its relative isolation, (5) its bilingual status, and (6) the degree of hom.eogeneity of its settlement populated from England, the Continent or other foreign countries.
1 Francis, ibid, P. 788.
2 Hans Kruth has been one of the leading scholars on the study of American linguistic geography. His principal works on linguistic geography are: Handbook of the Linguistic Geography of New England, 1939: A Word Geography of the Eastern United States, 1949. He has also been compiling the Middle English Dictionary at the Univercity of Michigan •.
-2. The Investigator
· In selecting the data, an investigator with some training in gei).eral linguistics works as .a .fielp-worker. The field ·worker. shold also . have knowlecJ.ge of phonetics and phonemics and some. understanding of the history and culture of the ~ea in which he is doipg resear~ch.
The following are the field-workers who worked on the maps of New England: Bloch, Bernard. Chapallaz, Marguerite. Hultzen, Lee S. Harris,. :Rachel .S . . Hanley,. Milles L. Joos, Martin. Kurath, Hans. Lo:\Vman, Jr., Guy
S.
Reynard, Cassil.l 3. The QuestionnaireQuestionnaire items are composed of questions on pronunciation, gram·
mar, .or vocabulary. In selecting the items, the staff of the Atlas followed three principles. (2) Questionnaire items ought to be well known to the majority of people in the given area under examination: (b) they ouglit to be comparatively simple to bring into a conversation: (c) they ought
to haveregional or social variants.2
Even with these precautions, it has not been possible· to employ a single uniform questionnaire throughout the United States. Many items pertaining to climate, produce, or culture are not nation-wide terms but are restricf ed to specific areas. Also a pronunciation item in one region may become a lexical item in other regions. In length, the questionnaire used along the Atlantic Seaboard consists of about 800-850 items; however, the ques_ tionnaire used in other regions include from 520
to
700 · items. With a skilled field~ worker and a cooperative infotrriimt; ·these questiortnaire~ can1 The field workers of other areas are stated in Chapter Vf. "Progress of
the Atlas".
be filled out in from four to seven hours;
Though it has not been .possible to u~ a single uniform questionnaire thoughout the United States, the basic list is essentially the same. There-fore; the collections
of
the whole country can be compared when the investigations are complete.4: Iriforinants
The field-worker consults representative infOrmants who are natives of the region 'under examination. It · was obvious that the irivestigation of American di~lects would not be satisfactory if only the oldest and least educated people were examined. For, in the United StateS, there is no' hard line between folk speech and cuitured speech~ :Moreover, the cultured speech has regional differences. Furthernior~, there is common sJ)eech between folk speech and cultured speech. In consequence, field"workers who worked ori the maps of
New
England corisiilted at least two inform-ants in every community. One of these informants was of the oldest generation, whose education beyond the elenieri.tacy school was 'not usual and had a minimum of reading and travel, while the other was a middle-aged person who had a secondary school education and more coni 1:ct with the outside world~ .Besides, inyestigatqrs interyieweq cultured infm;mantsin aP,prqxjmately a fifth of t~e communitie~ e:x;:;unined. This is cons~dered
to be in contrast tp the usual method of the European field-workers, :who might be satisfied with one.l
in some regions, one of the informants stands for a native American, while the other a native speaker of English ·represents a foriegn language group that has influenced much on the early history of the region. There-fore the tingi.Iistic Atlas data may indude · ·.such information as differ-ences between cultured speech
and
uneducated speech; agreement between cultured speech ·and cortiinon speech; older proriuric'iation, granimatieal irregularities, and historical develoimtent in a terms ·of pronunciation, grammar, and syntax.1 American English, P. 134.
-5---5. Method of recording the data
Linguistic Atlas data has been recorded on the spot, in a phonetic alphabet.
While conversation between an informant and a field-worker was going on, phonemic contrasts were closely examined but the transcription were phonetically written.! The phonetic alphabets used for Eastern part of the United States are explained and discussed in Chapter IV, handbook of the Linguistic Geography of New England.
It is apparent that there will be differences between investigators in the way they handle the phonetic alphabet for the Atlas. However, the differences will become less, if they have worked together for a consider-able length. of time. "There is, in fact, a remarkconsider-able degree of correlation between the transcriptions of field-workers who have had the same kind of training" .2
Therefore the transcriptions of a trained field-worker are worthy of considerable confidence.
Since each response is recorded in a phonetic transcription, all responses have value as pronounciation evidence.
6. Recording Situation
The interview between the field-worker and the informant is carried on in a conversational situation--or as close to a conversational situation as possible--in order that the informant will use his normal speech.
Ch. IV Validity of the Atlas
1. The Atlas matt-rials are the most dependable data on American speech.
First of all, as we have already observed in the previous chapter "Linguistic Atlas Project", no books, dictionaries, textbooks, etc. have been so well planned, examined, and recorded as the materials for the Linguistic Atlas.
2. In consequence the Linguistic Atlas is very useful for preparing
1 Francis: ibid P. 492. 2 Francis: ibid P. 493.
-materials for courses on all levels.' Without the information obtained through the Atlas, no publishers of textbooks dealing• with American speech can feel· easy about their jobs·.
3. The .:Atlas is also very useful for historical problems,. such. as the study of the cultural and settlement history. Specific problems are· 1) the development' of· · English • phonology, 2) the tracing of settlement mdvet'nent, 3) culturally. isolated areas, and 4) changes in the• distribuJ tion of certain aspects.
Ch.
v
Defects or Limitations of ~he .Atlas1. First of all the Atlas materials do not provide enough evidence about . supra·se~ment~l phonology, such
as
stress, pitch, and juncture .. It is because most of the interview is concerned with single words or short phrases.2. · Though the number of informants is ·great, it .is .not large. enough in proportion to .the . total population for. determining the. speech character-istics of American people.!
3. · The preponderance of rather old .informants , should . be paid sp~cial
attention; That is, the informants, generlly !iP~:;Il\:ingj were .. older than the population average. This means. that the Atlas indicates the most stable elements in each local area. :On the other hand, however, the Atlas materials are not extensive enough for .us to1 know the transient elements in some communities." ... ; moreover, the Atlas ·recor,ds would
:not reflect the most recent trends, fads, and innovat~ons--.. some of. which are extremely rapid, others .extremely slow".2
Let us examine the nature of the informants. Three. types of inform-ants: were interviewed.
Type I Consisted of older, less educated, and more isolated persons who are regarded as maintaining the archaic speech characteristics and who show the minimum effect of school education. Persons of this type made
1 Summer lves, Applied English Linguistics, P. 151. 2 E. Bagby Atwood, Applied English Linguistics, P. 166.
-up about one half of the whole number of informants. However, people of this group are not that· common in the total population. I
Type li Consisted of persons with average schooling. These were generally a little older than the population average but younger than Type I informants. When the usage of Type I and Type li is in accord with each other, it may perhaps be considered representative of the speech -of::the area under investigation although some-what old-fashion-ed.Z
Type ][ Consisted of those who had received more education than Type li and whose social contacts were more or less limited to others of high educational and social standing. The speech characteristics of this group may be considered representative usage among the older educated people.
Type ][ informants. constituted about 10 percent of the total number of informants.3
4. A certain miniber of people took part in the field work. Therefore, differences in the transcription and in the selection of informants shoud be taken into consideration in evaluating the data.
5, The Atlas does not try to record the usage of non-native speakers, or even of those who have lived for long time outside their home towns or villages. Such speakers are rather scarce in some communities, while considerably numerous in others. In a few of the latter the Atlas may reflect the usage of a small group.
6. The preponderance of rather old informants should be paid special attention. Since the interviews of the New England states were made about a decade ago, some of the usages, pronunciations, and grammati-cal structures may now be rare or archaic.
Also the Atlas materials do not seem to reflect the most recent trends and innovations.
i Swruner Ives, ibid, P. 151. 2 Ibid
3 Ibid
-Ch. VI Progress of the Atlas
About twenty-five years have already passed since the Linguistic Atlas of the United States and Canada was started. During the previous two and a half decades field workers have had approximately three thousand interviews.
1. New England
The Linguistic Atlas of New England was the first program and is the only part edited and published so far. In 1931 the field. work of the program was started and completed two years later. The first book and the Handbook of the Linguistic Geography of New England were pub-lished in 1939, and the publication continued until 1943. "The maps are arranged according to semantic associations, roughly following the plan of the questionnaire, except that maps presenting grammatical data .are grouped together in the second part of the third volume" .I
The Handbook includes such materials as the phonetic alphabet, questi-onnaire used, a survey of settlement history, bibliographies of linguistic geography and New England history. Also included is information about methodology and communities and informants used in the study. 2. Middle and South Atlantic States
The second regions investigated were the middle and south Atlantic States. Investigation was started by GuyS. Lowman, Jr. in 1933. After his death, Lowman was succeeded by R. J. Me David,Jr., who complet_ ed the field work in 1949. Data have been brought together at the University of Michigan for the editing and publication of the Atlas of Middle and South Atlantic States. Pending publication of this Atlas, two volumes appeared: Kurath's Word Geography of the Eastern United
~(Ann Arbor, Michigan, 1949) and E. Bagby Atwood's Survey of Verb Forms in the Eastern United States (Ann Arbor, 1953) Kurath and McDavid's The Pronunciation of English in the Eastern United
~ is close to publication. 3. North-Central States
The Linguistic Atlas of the North-Central States, which include Wisco-1 Francis: ibid, P. 494.
-nsin, Michigan, Southwestern Ontario, Illinois, Indiana, Kentucky, and Ohio, under the direction of Albert H. Marckwardt, University of Michigan, was begun in 1938. However, the fieldwork of only Wiscons-in was fWiscons-inished before World War: JI by Frederic G. Cassidy. In 1948, investigation was begun again and completed in 1956' chiefly • by McDavid, Cassidy, and A. L. Davis. The collections have been asseirtbled at the University of Michigan for editing and publication.
4. Upper Midwest
The Linguistic Atlas of the Upper Midwest, which inCludes Mines<:>ta, Iowa, the Dakotas, and Nebraska, was started by Harold B. ·Allen, University of Minesota, in 1947. Investigation was completed in 1957, more than half 'by himself. In the meantime, Mrs. Virginia ·McDavid
has finished a dissE:rtation on A Survey of Verb Forms in the North Central States and the Upper Midwest (Minesota, 1956) .
5. Rocky Mountain Region
In 1950, fieldwork was begun in the Rocky Mountain area, whieh includes Colorado, Utah, Arizona, New Mexico, and Montana, under the direction of Professor Marjorie Kimmerle of the University of Colorado as director and T. M. Pearce of the University of New Mexico as associate director. Francis observes that the field work has been even spottier in this region than in most other . places because of the kmg distances and the less dependable financial aid:I
Field work was finished for Colorado (mostly by Miss Kimmerle) and Utah, more than two-thirds for Arizona, and aiJ. indefinite amount for New Mexico.
6. Pacific Coast
The Linguistic Atlas of the Pacific Coast, which in<:;ludes Washington, Oregon, California, Idaho, Nevada, and Western Mont~a, has advanced very speedily, for several experienced field-workers are available arid two institutions have given sufficient financial support. The
invesdga-1 I . . . .· .. .. · : ' . .
tion is being carried on by David Reed of California and Caroll Reed of Washington as co-directors, with John Moncur of U. C. L.
A.
and David Decamp of Washington State as associate directors.Approximate-1 Francis: ibid, PP. 493-497.
-10-ly two-thirds of the field work was completed. Decamp has finished his disseration on the speech of the San Francisco area.
7. Texas and the Gulf States
"In Texas and the Gulf States work has been sporadic. A file of vocab· ulary check lists has been growing at Texas under Atwood's direction. Students of C. M. Wise, ,;K Louisiana State University, has completed more than a hundred fiefd records.
At University of Texas check lists have been increasing under Atwoods instruction. At Louisiana State University C. M. Wise has been collect· ing field records with the help of his students, work in Arkansas and Missouri has recently been begun.!
8. Canada
"In Canada, except for border communities investigated in connection with the American regional atlases, there are only incomplete prelimi. nary survey of the Maritime Provinces by Henry Alexander, the nearly complete intensive study of Queens Country, Nova Scotia, by Rex WUson, and the projected check-list investigation of eastern Ontario by the Rev. Brother Pius".2
Ch. VII Principal Dialect Areas
Generally speaking, we can distinguish the three principal dialect areas: Northern, Midland, and Southern, Midland being further divided into North Midland and South Midland.
"Other recognizable but not clearly set-off areas have been found in the Northern Plains, the San Francisco bay area, and parts of the Rockies clearer subdivisions will be recognized as a result of more extensive field work and more intensive analysis of the available data".3
The dialect regions are most clearly identified along the Atlantic Coast, where people settled earliest.
As the settlement advanced westward, local variances became increasing-ly less, and even the boundary lines between principal speech areas are
1 Francis: Ibid, P. 497. 2 Francis: Ibid, P. 497. 3 Francis: ibid, PP. 511-512.
-11-difficult to identify,
Ch. WI Application of the Data
1. The. immediate application .. of :~he data is. in tile revision, of state-.·ments about moder1;1 EI}glish usage. 'The teaching ~f standad forms,must
be done in full awareness of frequen~;y .and ,distribution of the contrast-ing non-stanqard form".l
English teachers are supposed. to use the data :in lectures, dr-ills, discu~
sions, etc., . :fhis is also true of the people :who m~ke txtboo~, dictiona-ries, ete.
2. Another application is in the recognition of the standard form~ .which are regional, and not nationaL Although local American speech, cap ):>e considered as standard it is also important to . recogni;z;e tl;J.e existence of . the standard speech of other areas.
3. · ~'Then, finally,. a.Jnrther utilization of the Atlas data, possible
in
bothcollege and secondary school, would be for ·the aim of developing aware-ness that language is a complex, changing, and always. relative . struc-ture, not a set of absolutes" .2
1 Summer lves, Ibid, P. 149. 2 Ibid
. BIBLIOGRAPHY
Allen, H. B., ed., Readings in Applied English .. Ling1,1istics, Appletol1:Cen-tury-Crofts, Inc., New York, 1958.,
Bloomfield, Leonard, 1~!5uag!:, Henry Holt & <;:q., Inc., New Yor}{, 1~33.
Francis, W. Nelson, . The . structure of American English,. The .. ~onald
P.ress ComPan>:•. New York, .1!:!58~
Gleason, H. A., An Introduction to Descriptive Linguistics, Nenry Holt &
Co., Inc., New York, 1955.
Hall, Robert A., Jr., Leave Your Language Alone!. Linguistica, Ithaca,
Kurath, Hans, A Word Geography of the Eastern United States, Univers-ity of Michigan Press, Ann Arbor, 1949.
Kurath, Hans, et al, Handbook of the Linguistic Geography of New England, American Council of Learned Societies, Washington, D.
c.,
1939.Marckwardt, Albert, American English, Oxford University Press, New York, 1958.
Mencken, H. L., The American Language: An Inquiry into the Develop-ment of English in the United States, 4th ed. ,rev., Alfred A. Knopf, Inc., New York, 1936.
