The Reversal of a Sound Change in Cincinnati

Abstract

One of the most important variables for characterizing and distinguishing American English dialects is the phoneme /'iJ:./, or short-a, the sound of mat, mass and man. Many dialects have two main allophones of l'i1:-/: a tense ['iJ:.h], characterized by phonetic raising along the outer perimeter of the vowel space and the development of a central in-glide; and a lax, monophthongal ['iJ:.] more or less in the position of cardinal ['iJ:.]. In some dialects these allophones have split into two phonem~. Labov (1991:12), in putting forth a tertiary typology of English dialects, calls this split one of two pivot points in American English. Some of the oldest American dialects, those on the Atlantic Coast and in the South, have tense /'iJ:./ in a restricted set of environments, most commonly before front nasals, voiceless fricatives and varying subsets of the voiced consonants. The tensing rule in some of these dialects also interacts with syllabicity. The dialect of the large industrial cities around the Great Lakes, known as the Northern Cities dialect, tenses in every environment. The speech of the West Coast exhibits tensing only and always before nasals, while Canadian English shows no tensing at all. While /'iJ:./-tensing has been thoroughly studied in Atlantic Coast dialects (Trager 1930, 1934, 1940, Cohen 1970, Ferguson 1975), and has been well described as part of the Northern Cities Shift, much less is known about the phonology and dialect geography of /'iJ:./-tensing in the lower Midwest. This broad and populous region, intervening between the Northern Cities and the South, is part of the Midland of traditional American dialectology (Kurath 1949). The status of ('iJ:.) in most of the major Midland cities is virtually unknown. One of the oldest and largest of these cities is Cincinnati, located on the northern bank of the Ohio River, across from Kentucky, in southwestern Ohio. Preliminary research on Cincinnati phonology was carried out as part of a Telephone Survey of Change in Progress in North American English, or TELSUR, now underway at the Linguistics Lab of the University of Pennsylvania. The TELSUR data revealed a unique situation that demanded further study. In the first major study of Cincinnati speech, part of which was presented at the 1995 Penn Linguistics Colloquium (Boberg and Strassel1995), we examined a wide range of phonological, syntactic and lexical variables. Of the phonological variables, the most interesting proved to be short-a. Our data on this variable revealed not only a unique set of tensing environments, distinguishable from every other regional system hitherto described, but an apparent reversal of an earlier change toward tense short-a. As this finding, if confirmed, would represent one of the few documented cases of a historical change reversing its direction, we decided to study it in greater detail, using more data from a wider range of speakers. This expanded database would also allow us to resolve apparent discrepancies from the first study, particularly a lack of consistency among the youngest speakers.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 19 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.