Published online by Cambridge University Press: 06 February 2009
Most intonation analysts would consider it a truism to insist that any model of the formal properties of a language's intonation system has to be relativistic in character. By this one would mean that the linguistic constants in the system are the contrasts between the features involved (pitch, loudness, or whatever) and not the values of the features themselves, as defined in any absolute, physical way. The point hardly needs quotation to support it, but it will be useful to refer to one person's formulation of the relativity claim as a reminder of how the position is typically presented. Abercrombie, for example, says (1967: 107):