The Acquisition of Sign Language: The Impact of Phonetic Complexity on Phonology

Abstract

Research into the effect of phonetic complexity on phonological acquisition has a long history in spoken languages. This paper considers the effect of phonetics on phonological development in a signed language. We report on an experiment in which nonword-repetition methodology was adapted so as to examine in a systematic way how phonetic complexity in two phonological parameters of signed languages — handshape and movement — affects the perception and articulation of signs. Ninety-one Deaf children aged 3–11 acquiring British Sign Language (BSL) and 46 hearing nonsigners aged 6–11 repeated a set of 40 nonsense signs. For Deaf children, repetition accuracy improved with age, correlated with wider BSL abilities, and was lowest for signs that were phonetically complex. Repetition accuracy was correlated with fine motor skills for the youngest children. Despite their lower repetition accuracy, the hearing group were similarly affected by phonetic complexity, suggesting that common visual and motoric factors are at play when processing linguistic information in the visuo-gestural modality.

ACKNOWLEDGMENTS

This work was supported by the Economic and Social Research Council of Great Britain (Grant RES-620-28-6001), Deafness, Cognition and Language Research Centre (DCAL), a City University London Research Fellowship awarded to the first author, and a Leverhulme Early Career Fellowship awarded to the second author. All the authors would like to thank Cathy Green and Tanya Denmark for their help in the pilot version, Gary Cutmore for modelling the nonsense signs, and Katherine Rowley for assistance with data coding. We are very much indebted to the three anonymous reviewers and the editor of this journal, Susan Goldin-Meadow, for their insightful comments on earlier versions of this paper. Finally, we would like to express our gratitude to all the children and teachers who took part in and supported this research study.

Notes

¹Following the conventions of the sign language literature, we use Deaf with an uppercase (D) to refer to members of the community that use BSL. We use deaf with a lowercase (d) only when discussing the effects of hearing loss.

²Marked structures are those that are less common cross-linguistically, and such structures tend to be acquired later by children (Jakobson, 1941). As a first approximation, marked phonological structures tend to be more complex than unmarked, notwithstanding disagreements over exactly how complexity is best measured (Gierut, 2007).

³The convention is for handshapes to be named after the letters they represent in the American Sign Language alphabet or counting system.

⁴These handshapes occur in American Sign Language, where they are also considered unmarked, for example, by Brentari (1998) (who uses the term ‘1’ handshape instead of the ‘G’ handshape).

⁵The degree to which this statement is true depends on how word-like the stimuli are. Although nonwords are by definition not stored in the lexicon phonotactic probability is an important predictor of how accurately children will repeat them (see Coady & Evans, 2008, for a review).

⁶This test assesses the comprehension of selected aspects of BSL morphology and syntax (negation, plurals, verb morphology and the distinction between nouns and verbs) in a picture-pointing paradigm. There is an initial vocabulary check for the signs that are used in the test to avoid the possibility of the child making errors because of unfamiliarity with individual lexical items.

⁷The correlation here is negative because the score for the motor skills task is measured in seconds – lower scores represent faster bead threading, and hence better fine motor skills.

⁸This already shows a difference between the Deaf and hearing children, as the hearing children could not do the task at the younger ages whereas the Deaf children could.

⁹A notable exception is de Bree, Wijnen and Zonneveld (2006), who tested Dutch children on the repetition of nonwords with different stress patterns, including stress patterns prohibited in Dutch.