Learning to talk: A non-imitative account of the replication of phonetics by child learners

How is it that an English-speaking 5-year-old comes to: pronounce the vowel of seat to be longer than that of sit, but shorter than that of seed; say a multi-word phrase with stress-timed rhythm; aspirate the /p/s of pin, polite, and spin to different degrees? These are systematic features of English, and most people believe that a child replicates them by imitation. If so, he is paying attention to phonetic detail in adult speech that is not very significant linguistically, and then making the effort to reproduce it. With all the other communicative challenges he faces, how plausible is this? An alternative, non-imitative account of the replication of these features relies on two mechanisms: (1) emulation, and (2) the conditioning of articulatory activity by the developing characteristics of speech breathing. The phenomena above then become no more than expressions of how a child finds ways to warp his phonetic output in order to reconcile conflicting production demands. The criteria he uses to do this make the challenges both of learning to talk and then of managing the interaction of complex phonetic patterns considerably more straightforward than has been imagined.