ABSTRACT
It has often been remarked that the essential auditory identity of a language, the immediate impression conveyed upon a first or a passing encounter, derives from the character of its prosody (in a broad, Firthian sense). Evidence on the point comes from the way young children (before the end of the first year) are able to differentiate prosodic features from the auditory soup which surrounds them and to introduce them into their own discourse, long before segmental distinctions come to be discriminated and used. The evidence is somewhat mixed, as regards the role of intonational features, but is substantial with respect to rhythm. For example, the latest study I have seen (Levitt and Aydelott Utman 1992) compares the syllable durations of French and American infants, and shows that the Frenchlearning infant produced more regularly timed non-final syllables and showed significantly more final syllable lengthening than the Englishlearning infant. In other words, there was evidence of the emergence of the ‘machine-gun’ rhythm typical of French syllable-timing, while the English child maintained the ‘morse code’ rhythm typical of English stress-timing (see Lloyd James (1940: 25) for the former pair of terms, Pike (1945: 35) for the latter pair).
