Abstractness and Ordering
We have said that the phonological component is an ordered block of all of the phonological rules of the language. With this conception of the way the rules are organised, we need to look in a little more detail at the way in which rules apply, and at the effects of rules upon one another, that is, at rule interaction. Let us look again at the Russian data presented in Chapter 5: We said that this data exemplifies a rule of Voicing Assimilation (VA). That rule expresses the generalisation that all of the obstruents in a sequence of obstruents assimilate in voicing to the final obstruent. Another rule of Word-Final Devoicing (W-FD) operates on obstruents, as in /sad/ (‘garden’), which is [sat] in the nominative singular, but [sada] in the genitive singular. Other examples are nominative singular/dative singular pairs like [xlep] / [xlebu] (‘bread’), [storoJ]/[storo3u] (‘guard’), [rok]/[rogu] (‘horn’), and nominative /genitive pairs like [ras]/[raza] (‘time’).
Unable to display preview. Download preview PDF.
- Among the detractors of standard generative phonology, Derwing (1973) and Ohala (1974) insist on ‘external’ evidence, from psychological and instrumental phonetic experiments, respectively. Foley (1977), on the other hand, argues that standard generative phonology is not abstract enough, and that the primitives of phonological representation must be entirely non-phonetic. Both camps insist that generative phonology cannot otherwise be considered’ scientific’. For an interesting reply to the Ohala position, see Anderson (1981), and for discussion of the issue of whether generative phonology may be said to be’ scientific’, see Carr (1990: 6.3).Google Scholar
- There is a rather large literature on rule ordering and application, and on the abstractness issue. For detailed discussion on problems of rule ordering and application, see Kenstowicz and Kisseberth (1977), Chapters 4–6, and Kenstowicz and Kisseberth (1979), Chapter 8. See also Anderson (1974), Chapters 5–9. Lass (1984a), Chapter 9, gives a critical overview of the problems and issues in the abstractness debate; his conclusions concerning generative phonology are much more critical than those adopted in this book.Google Scholar