In the preceding chapter we focused on the process of sound change and the initial integration of a completed sound change into the grammar, keeping the discussion of phonological structure to a minimum. In this chapter we investigate the further development of the rules into which sound changes typically evolve – that is, change within the structured phonological system.

These developments can be discussed only in the context of a coherent model of phonology. We adopt a generalized version of generative phonology as developed in the 1970s and 1980s, with ordered rules, autosegments, and metrical structures. We are well aware that this approach has its limitations, but so does every other model of phonology; we have chosen this model because it is exceptionally convenient for the discussion of phonological change. (We have chosen not to work with Optimality Theory because it does not seem well adapted to the description of phonological change; see especially the critique of McMahon 2000: 57–128.)

In the first section below we illustrate some of the advantages of fully articulated modern phonology in describing the effects of sound change; readers who require a fuller introduction should consult, e.g., Goldsmith 1990 or Kenstowicz 1994. We then proceed to consider the evolution of phonological rules that have already become categorical.