This concluding chapter returns to focus on theoretical questions about phonetic grammar, a controversial object of description. It is a fact that physically motivated phonetic processes are usually not physically necessary, and can vary stylistically, so that the most natural, physically easiest and most simplified phonetic forms are restricted to certain styles. There are similar differences in these details across dialects. ``Hard'' coarticulation does exist, since the tongue cannot move infinitely fast. However, natural phonetic processes, while physically simpler or easier, are not physically necessary.
The tongue is nearly all muscle, and the only mass it must move is itself. Further, a considerable part of the human brain is devoted to the physical control of the tongue. So it should not be surprising that the coordinated movements of the human tongue are among the most complex and rapid large-scale physical actions naturally performed in the animal kingdom.10.1 The physical limitations of the human apparatus of speech -- the finite mass and amount of force that may be exerted on that mass -- are real, but there is considerable leeway for linguistic control (and conscious control) of details of articulation at normal speaking rates. The phonetic effects which lie within this realm of ``soft'' coarticulation are not due to absolute physical constraints. People learn coarticulations that are often physically easy but not always physically necessary. The control of these often intricate details is part of the human language faculty, because it is part of what differentiates one dialect or language from another.
The system by which surface phonological structure is interpreted phonetically is a partly linguistic system (Liberman & Pierrehumbert 1984). This system includes the learned rules of soft coarticulation and of prosodically governed (stress) reduction, as well as the interpretationof abstract tonal specifications as F0 contours. One of the primary elements of this system is a specification of the target, or average, phonetic vowel qualities; these qualities do not follow from phonological structure and universal principles of phonetic implementation alone, unless the many effects described in the preceding dialect studies can be accounted for by some aspect of phonological structure heretofore unknown.