In this paper, Ohala provides a nice case study in the style that he has become so well known for. He presents experimental data indicating that a regular phonological process, the direction of assimilation, is grounded in facts about speech production and perception. Such results are significant not only because of the light they shed on particular phenomena, but also as examples of research methodology. They point out the importance of seeking the right sphere of explanation for observed patterns in sound structure.

The type of explanation which is featured in this paper is phonetic explanation, or explanation based on the physics and physiology of speech. Phonetic explanations are especially attractive because of their reductionist character; it is very satisfying to reduce psychology to biology, and biology to physics. Ohala's comparison of phonetic and nonlinear phonological accounts of assimilation links reductionism (“None of the terms of the explanation are unfamiliar, other-worldly entities”) with generality (“ A few primitives go a long way”). It is not clear to me that this link is well-founded, especially with respect to ongoing research. Nonlinear phonology has identified a number of principles which have great generality, although their physical basis is unclear. In particular, the principle of hierarchical organization has been shown to be a factor in the lexical inventory, phrasal intonation, and allophony rules of many languages. On the other hand, some parts of phonetics are extremely particular, from a scientific point of view. For example, there is no reason to suppose that the specific nonlinear oscillator responsible for vocal fold vibration has any generality from the point of view of physics.