Skip to page Content.

Libby Coggshall’s successful dissertation defense

Libby Coggshall has successfully defended her dissertation, Short-a in the Sixth Borough: A Sociophonetic Analysis of a Complex Phonological System in Jersey City. Her committee was John Singler (chair), Renée Blake, Carmen Fought (Pitzer College), Gregory Guy, and Susannah Levi (Communicative Sciences and Disorders, Steinhardt, NYU).


The study of short-a (e.g., the vowel in words such as bat, bad, bang, ban) in New York City English (NYCE) has a long history, and with many different descriptions of this complex system (e.g., Babbitt 1896; Trager 1930; Labov 1966/2006; Cohen 1970; Labov 2007). It is complex due to the fact that it is not just a matter of following environment but instead includes a number of phonological, grammatical, and lexical conditions that determine which tokens of short-a are tense (fronter and higher) and which are lax (backer and lower). There are several other short-a systems in varieties of North American English. Most common perhaps is the nasal short-a system where short-a is tense before nasals and lax elsewhere.

While older work found the complex NYCE short-a system to be used only by white speakers, recent work has shown that some people of color also use this system (e.g., Becker 2010; Newman 2014). Work in Manhattan suggests the complex NYCE short-a system is being lost (e.g., Becker and Wong 2009), but work done in other boroughs suggests that language change is taking a different or at least a slower path (Newman 2014). More work in parts of the NYCE dialect area outside of Manhattan is needed to understand the breadth of variation. To this end, I interviewed twenty English speakers from Jersey City, NJ, born between 1918 and 1984, including three Filipino Americans, one Latino, four black speakers, and twelve white speakers. They were all either born in Jersey City or moved there by the age of five.

While recent research assumed a single, invariable system for the NYCE short-a, the body of work that first described the system showed immense amounts of variation. Assuming that there is a single, unerring version of the NYCE short-a system gives improper results in regards to how much the NYCE short-a system may have changed. For instance, Labov (2007) says that short-a before velar nasals is always lax, but Labov, Yeager, and Steiner (1972) said that this environment varied so much that they could not put it in the tense or lax category with full certainty. Recent work has used variability before /ŋ/ as evidence of loss of the system, but past work shows this variation to have been there all along.

In order to study the complex short-a system of NYCE fully, I composed a paradigm of 33 word classes of short-a words based on environments that past research had found to be important. These included following environment, lexical status (lexical versus function word), syllable status (open versus closed syllable), position in the word (word-initial versus not), morphemic status, and lexical exceptions. In creating this word class paradigm, I noticed a pattern: Environments that had been described as invariable, like those before voiced stops or voiceless fricatives in closed syllables, are the environments where short-a occurs frequently; in contrast, environments that had been described as variable, such as those before voiced fricatives or voiced affricates, are ones where short-a is much less common. This finding suggests frequency may play a role in determining which word classes are invariable and which are variable.

The type of short-a system used by the speakers in Jersey City fell along ethnic lines. White speakers used a complex NYCE short-a system while the speakers of color used a nasal system. Variation was found within both systems. My research shows the need to take a detailed, historically informed approach when studying the NYCE short-a system. If we fail to take into account past variability, we risk misinterpreting variation within our data. If we fail to account for all present variability, we risk overlooking important differences between and among speakers.