As part of the preparation for the meeting on the revision of the International Phonetic Alphabet, I have been asked to write a short summary of my correspondence with a number of members of the Association on the principles on which the alphabet should be based. To this end, I have sought the views of the 34 respondents who gave first or second rank to this topic on the preregistration form for the 1989 Convention.

In experimental phonetic research, we often come across cases of multiple perceptual cues to a phonemic distinction, even though this distinction may be seen traditionally as dependent on some single phonetic feature. The question arises as to the relative power of these cues. How equally do they share the burden of communicative relevance? For example, among the several cues that emanate from the timing of the valvular action of the larynx (Lisker and Abramson (1965); Abramson (1977)), fundamental-frequency perturbations (House and Fairbanks (1953)) have been shown by some studies, apparently starting with Haggard, Ambler and Callow (1970) and Fujimura (1971), to help in the perceptual differentiation of voiced and voiceless stop consonants; however, recent work (Abramson and Lisker (1985)) suggests that this cue has very limited efficacy compared with other acoustic consequences of voice timing.