JEngL 27.2 (June 1999)Davis / Changi g Southern Vocabulary From Confederate Overalls to Designer Jeans The Changing Southern Vocabulary

There were three major stimuli for this research. (1) Johnson (1996, i) found that “by 1990, region had become the least influential factor in lexical variation. Rural-ity, education, race, and age all showed equivalent effects on speech. ” In 1990, Johnson interviewed subjects in thirty-nine of the same localities where Guy Low-man had carried out fieldwork for the Linguistic Atlas of the Middle and South At-lantic States (LAMSAS) some fifty years before. Johnson found that southern terms were not as prevalent in her study as they were in the Lowman records. (2) Gregg (1993) found that dialectal differences among younger Kansans may well be level-ing. (3) Davis and Houck (1996) compared lexical and phonological evidence in the Linguistic Atlas of the North Central States (LANCS) to that in The Linguistic Atlas of the Gulf States (LAGS) and found, as they sampled localities from north to south, what looked like an inverse correlation between southern forms and distance south—a conclusion so counterintuitive that they searched for more reasonable ex-planations. One of their hypotheses was as follows: Our results may be a function of when the field work for the two atlases took place. McDavid (1958, 496) reported that “LANCS began in 1938 under the direction of Albert Marckwardt, but only Wisconsin was completed (by Fre-deric Cassidy) before World War II. ” He goes on to point out that all field-work was “pressed to completion by 1956, chiefly by McDavid, Cassidy, and A. L. Davis. ”... For LAGS, Lee Pederson et al. (1986, 1:24) noted that the basic fieldwork began in 1973-75 and that the final fieldwork was undertaken in 1980 and finished in 1983 with the addition of three New Orleans records. AUTHOR’S NOTE: An earlier version of this article was presented at the annual meeting of the Lin