SOME DISTRIBUTIONAL FACTS ABOUT FRICATIVES AND A PERCEPTUAL EXPLANATION

Across and within languages voiced sibilants tend to be disfavored rel ative to voiceless ones. This paper explores the claim that voicing mo re adversely affects the distinctive acoustic properties of sibilants than those of nonsibilants. One prediction associated with this claim is that voicing differentially lowers the amplitude of frication noise for sibilants and nonsibiliants so that amplitude differences between the two classes are reduced. Acoustic measurements confirm this predi ction. A second prediction is that voicing has a greater negative effe ct on the identification of sibilants than nonsibilants. Perceptual re sults from this and previous studies are somewhat variable, but averag ed data support this prediction. The findings suggest that voiced sibi lants are disfavored in part for perceptual reasons.