Learning Phonology from Surface Distributions, Considering Dutch and English Vowel Duration

ABSTRACT

In learning language, children must discover how to interpret the linguistic significance of phonetic variation. On some accounts, receptive phonology is grounded in perceptual learning of phonetic categories from phonetic distributions drawn over the infant’s sample of speech. On other accounts, receptive phonology is instead based on phonetic generalizations over the words in the lexicon. Tests of these hypotheses have been rare and indirect, usually making use of idealized estimates of phonetic variation. Here we evaluated these hypotheses, using as our test case English and Dutch toddlers’ different interpretation of the lexical significance of vowel duration. Analysis of thousands of vowels of one Dutch and three English mothers’ speech suggests that children’s language-specific differences in interpretation of vowel duration are likely due to detection of lexically specific patterns, rather than bimodality in raw phonetic distributions.

Acknowledgment

The author thanks Michael Brent and Jeff Siskind for making the English corpus available, and Paula Fikkert and Claartje Levelt for the Dutch corpus.

Disclosure statement

No potential conflict of interest was reported by the author.

Supplemental material

The duration data analyzed in this article can be accessed here. https://www.sas.upenn.edu/~swingley/swingPubs.htm

Notes

¹ The regularity studied by Ko et al. (2009) concerns an allophonic rule, and not just phonetic categories as acoustic objects. We acknowledge, of course, that learning phonology involves much more than learning phonetic categories; e.g., (Peperkamp, Le Calvez, Nadal, & Dupoux, 2006).

Additional information

Funding

This work was supported by NIH grant R01-HD049681 to D. Swingley. Some of the same data, and the argument in quite preliminary form, were presented at the International Conference on Infant Studies in 2006.