Fortition and lenition patterns in the acquisition of obstruents by children with cochlear implants

Abstract

This paper investigates patterns of error production in 10 children who use cochlear implants, focusing specifically on the acquisition of obstruents. Two broad patterns of production errors are investigated, fortition (or strengthening) errors and lenition (or weakening) errors. It is proposed that fortition error patterns tend to be related to the process of phonological development, because they are involved with universal implications and notions of markedness. Lenition error patterns, on the other hand, show more context‐sensitive effects and reflect properties related to minimization of articulatory effort. The relationship between fortition and markedness is demonstrated in an optimality theoretic analysis, and it is further demonstrated that the observed characteristics of phonological development in children with cochlear implants are similar to those exhibited by children with normal hearing.

• Cochlear implantation
• child
• phonology
• fortition
• lenition

Notes

1. This simple characterization, however, is somewhat complicated by the fact that nasals appear in the sonority hierarchy but not in most versions of the fortition/lenition scale. The sonority of nasals has been demonstrated to be relevant in those instances where sonority is relevant, such as the sonority dispersion principle (Clements, 1990; for an example from child language, see Chin, 1996). Their role in fortition and lenition processes, however, is less clear. Kirchner (2001), for example, notes that their inclusion in the fortition/lenition scale would incorrectly imply that voiced fricatives can lenite to nasal.

2. In the case of final obstruent devoicing, although devoicing is a strengthening process, it is also possible to view it as a lenition process, in that subglottal pressure quickly weakens in this position. For the purposes of this paper, however, we are considering all strengthening processes (such as devoicing) to represent fortition. Additionally, there is phonological evidence that final obstruent devoicing in English is better characterized as final obstruent fortition, through the addition of the feature [spread glottis], rather than the loss of the feature [voice] (see Vaux & Samuels, 2005; Blevins, 2006; Iverson & Salmons, 2006). The same process would also characterize final aspiration of voiceless stops as seen in (4) through (6) as fortition. Iverson and Salmons (2006) suggest that the synchronic motivation for such processes is the marking of phrase boundaries, which generalizes to word and syllable boundaries.