Independent evidence: historical restructuring

Paul Kiparsky’s seminal 1968 paper “How abstract is phonology?” raises the question whether limits on abstractness are possible and desirable. Kiparsky’s concern is the postulation of segments which are never realized, where a language is assumed to have an underlying distinction between two segments which are always phonetically merged. A classic example is Hungarian, which has a vowel harmony rule where suffix vowels agree with the preceding vowel in backness, e.g. ha:z-am ‘my house,’ fylem ‘my ear,’ vi:z-em ‘my water.’ A small number of roots with the front vowels [i: i e:] always have back vowels in suffixes, e.g. he:ɟ-am ‘my rind,’ ɲi:lam ‘my arrow.’ The abstract analysis is that these roots have underlying back vowels [ɨ ɨ: ə:], which later become front vowels. This move makes these roots phonologically regular. The reasoning is that since these front vowels seem to act as though they are back vowels, in terms of the vowel harmony system, maybe they really are back vowels at a deeper level.

Kiparsky terms this kind of analysis absolute neutralization, to be distinguished from contextual neutralization. In contextual neutralization, the distinction between two underlying segments is neutralized in some contexts, but is preserved in others. Final devoicing in Russian is contextual neutralization because in the words /porok/ and /porog/, the distinction between k and g is neutralized in the nominative singular [porok], but is maintained in genitive [poroka] vs. [poroga]. With absolute neutralization, the distinction is eliminated in all contexts, and thus in Hungarian, /ɨ/ is always neutralized with /i/. Kiparsky argues that while contextual neutralization is common and has demonstrable psychological reality, absolute neutralization is a theoretically constructed fiction.

In arguing against absolute neutralization, Kiparsky faces the challenge that a number of cases of such abstractness had been postulated, so good reasons for rejecting those analyses must be found. Kiparsky focuses on the extent to which the psychological reality of theoretical constructs can be measured – this is an important consideration since linguistic theories are usually intended to be models of the psychological processes underlying linguistic behavior. The problem is that it is impossible to directly test whether linguistic constructs are psychologically valid by any simple or obvious tests. Linguistic properties are highly abstract, and not easily tested in the same way that one can experimentally test the ability to perceive touch or distinguish colors or sounds. Kiparsky argues that one can, in certain circumstances, use the pattern of language change as a theory-external test of grammatical theories. It is argued that historical sound change can provide just such a test.

An abstract phonological distinction cannot be justified on the basis of the fact that two historically distinct sounds merge in the history of a language, so even if it were shown that Hungarian he:ɟ ‘rind’ and ɲi:l ‘my arrow’ derived from earlier *hə:ɟ and *ɲɨ:l, this would not be evidence for an abstract underlying form in modern Hungarian. A child learning the language has no access to this kind of historical information. What Kiparsky points out is that you can inspect a later stage of a language to learn about the analysis of a language that was actually given at an earlier stage of the language, and then adduce general principles about grammars based on such independent evidence.