One of the most basic concepts in phonetics, and one of the least discussed, is that of phonetic similarity. An adequate general phonetic theory should permit any two phonetic events to be compared and rated on a scale of relative similarity, within a coherent account of the physiological, acoustic and perceptual ways in which the events are related to each other. Such a theory has not yet been completely developed. No explicit metric exists to date which would allow such a graded comparison to be carried out across the full range of phonetic entities with any great sensitivity of comparison, though the phonetic schemes for the descriptive featural classification of contoids and vocoids do enable a more limited comparison within each of those broad categories.

Part of the reason for the relative inadequacy of current phonetic theory in this area is possibly that, as noted earlier, the concept of phonetic quality itself is not concrete but abstract, and that the nature of this abstraction has not yet been fully explored. In addition, the detailed nature of the relationship between the phonetic and the organic contributions to speech events is still not well understood. Issues of phonetic similarity, though underlying many of the key concepts in phonetics, are hence often left tacit.

Phonetic similarity and segments

There are two different approaches to the notion of phonetic similarity in speech. The first is the relatively straightforward question of the degree to which any two segments under comparison are similar to each other, at the given level of comparison – articulatory, auditory or acoustic.