Publication |Sophia Linguistica

A Corpus-based Study on Phonetic Differences between “Can” and “Can’t” in American English

AUTHOR

Tomohiko Ooigawa

ABSTRACTS

The present study examined phonetic differences between can and can’t in American
English using Buckeye Corpus (Pitt et al. 2007). The purpose of this study is to develop
a set of training programs to help learners of English to correctly perceive and produce
the sounds of the phonological contrasts which are difficult to learn. Depending on the
context, it is difficult to correctly discriminate/identify American English can and can’t
for native speakers of languages which have no syllable-final /n/-/nt/ contrast (Ernestus et
al. 2017, Ooigawa 2018). Ooigawa (2018) claimed that can is realized with a clear long
nasal consonant, and that can’t is realized without any nasal consonants or with a very
short one. According to the corpus analyses in the present study, 94.4% of the can tokens
were realized with a nasal consonant, and 43.5% of the can’t tokens were realized without
any nasal consonants. The duration of nasal consonant in can was significantly longer than
that in can’t. The claim in Ooigawa (2018) was consistent with the results. The findings of
the present research can contribute to instruction studies on perception and production of
the sounds of the phonological contrasts which are difficult to learn.