Implications of Lexically Specific Variation in Dialect Coaching for Performance: Realization of again(st) and (n)either in Australian English

Understanding lexically specific phonological variation in dialect coaching is important to ensure an empirical basis for acceptable accent performances. However, these changes may not follow the expected substitution rules between the varieties. This paper uses a spontaneous speech corpus of Australian English to investigate variation and change over time in the realizations of two sets of word forms: again/against (termed again(st)) and either/neither (termed (n)either). Over 1,000 tokens of these words were extracted from the speech of more than 200 Australian English speakers from Sydney (born 1889–2001) recorded in the 1970s and in the 2010s, with tokens auditorily coded as dress (/e/) or face (/æɪ/), for again(st), and fleece (/iː/) or price (/ɑe/), for (n)either. Analysis reveals that for again(st), dress is the dominant form across all age groups, with only the oldest speakers displaying some variation with face. For (n)either, there is robust variation between fleece and price across all age groups, but a gradual shift toward a preference for fleece over price among speakers in the 1970s; subsequent changes in price among the 2010s are conditioned by ethnicity and social class. The paper uses these findings to provide dialect coaching advice to support sociolinguistically grounded portrayals of Australian characters.