What explains the reductions in /s/-clusters: Sonority or [continuant]?

Abstract

This study is a cross-linguistic examination of the patterns of reduction of two-member /s/-clusters. Data from six languages (English, Dutch, Norwegian, Croatian, Hebrew, and Polish) gathered from the productions of typically developing children and from children with phonological disorders are analyzed. While the languages belong to three different families, the results point to a universal tendency. Specifically, in the reductions of “s+stop” and “s+nasal” targets, the preference is the retention of C2, whereas in /sl/ and /sw/ targets, the clear pattern is in favor of the retention of C1. These patterns suggest that the continuancy of C2, not sonority, is the determining factor; C2 prevailed in “/s/+[-continuant]” targets, whereas C1 is retained in “/s/+[+continuant]” targets.

Keywords:

• phonology
• language acquisition
• clusters
• sonority
• distinctive features

Declaration of Interest: The author reports no conflicts of interest. The author alone is responsible for the content and writing of this paper.

Notes

For a detailed account of different explanations, see Parker (2002), where 98 different correlates of sonority are mentioned.

Parker (2002) cites 100 distinct sonority scales in the literature.

This suggestion for #sC clusters has also been made in other languages (Kager & Zonneveld, 1985; Trommelen, 1984, on Dutch; Davis, 1990 on Italian; Steriade, 1990 on Greek).

Gierut acknowledges that “/s/+nasal” clusters, together with “/s/+stop” onsets, may behave like adjuncts for some children, while they may group themselves together with other SSP-following /sl/ and /sw/ for other children.

Gierut's (1999) data come from monolingual English-speaking children, while Yavaş and Someillan (2005) examined the English data in Spanish–English bilingual children.

There are, of course, coalescences (e.g. swim [swim] → [fɪm] whereby the labiality of [w] and the frication of [s] are combined).

There are violations of this prediction in targets other than “/s/+nasal,” mentioned in the literature (Wyllie-Smith, McLeod, & Ball, 2006). However, what we consider here is the consistent pattern shown with a specific #sC target group.