Task effects in the articulation of coronal stop deletion
The various tasks for eliciting speech that form the sociolinguistic toolkit differently influence speakers’ stylistic choices. However, very little is known about stylistic variation in the articulatory domain. This study uses EMA to explore task effects in the articulation of tokens eligible for Coronal Stop Deletion: the variable surface absence of word-final coronal stops after consonants in English, e.g. ol’ ~ old (Labov et al., 1968). A classic approach to sociolinguistic style would predict participants to converge towards a standard form and eschew casual speech processes like Coronal Stop Deletion as their level of self-monitoring and metalinguistic awareness is increased (Labov, 1972). Despite this, speakers produced the least tongue tip raising in a Wordlist task, which would normally be considered to induce the greatest amount of attention to speech. I consider some potential explanations for this effect, including engagement with an interlocutor, prosodic factors, and speaker fatigue.
