| Abstract | [Introduction]
By conviction, not apparently contradicted so far by anecdotal evidence, almost any vocal tract, no matter what the ethnic affiliation of its owner, is inherently able to function “natively” in any language community, so long as it, and the ear to which it is attached, are “normal” and have been welcomed into that comminute at a “normal” age, namely in infancy. Linguistic inabilities, including phonetic, which are manifested in later life, are less evenly distributed over individuals, but presumably are in part culturally determined--some Americans, for example, speak more acceptable (to the French) French than others, but there is a recognized American-accented French. The nature of these phonetic inabilities is not all that well understood, for we are still not clear about what is perceptually based and what is a matter of more or less arbitrary category naming. Once acoustic signals are apprehended as speech, their attributes seem to be evaluated by reference to a vocal tract that might have produced them, and beyond that, they are labeled in terms of categories given by the language in which that vocal tract is speaking, which for the naive listener is the language in which he is listening. Comparison of native and non-native labeling of speech samples enables us to map categories of one language on other, and also serves as some check on hypotheses regarding the phonetic basis for category distinctions in one or both of the languages being compared. |
|---|
