Anne CutlerJul 1, 2013
Language Experience and the Recognition of Spoken Words
MIT Press 2012
One of the risks of a telephone interview is that the sound quality can be less than ideal, and sometimes there's no way around this and we just have to try to press on with it. Under those conditions, although I get used to it, I can't help wondering whether the result will make sense to an outside listener. I mention this now because Anne Cutler's book, Native Listening: Language Experience and the Recognition of Spoken Words (MIT Press, 2012), is an eloquent and compelling justification of my worrying about precisely this issue. In particular, she builds the case that our experience with our native language fundamentally shapes the way in which we approach the task of listening to a stream of speech - unconsciously, we attend to the cues that are useful in our native language, and use the rules that apply in that language, even when this is counterproductive in the language that we're actually dealing with. This explains how native speakers can typically process an imperfect speech signal, and why this sometimes fails when we're listening to a non-native language. (But I hope this isn't going to be one of those times for anyone.) In this interview, we explore some of the manifestations of the tendency to use native-language experience in parsing, and the implications of this for the rest of the language system. We see why attending to phonologically 'possible words' is useful in most, but not quite all, languages, and how this helps us solve the problem of embedded words (indeed, so effectively that we don't even notice that the problem exists). We consider how the acquisition of language-specific preferences might cohere with the idea of a 'critical period' for second-language learning. And we get some insights into the process of very early language acquisition - even before birth - which turns out to have access to richer input data than we might imagine.