Universal 1284:
- Original
- If nasality co-occurs in the neighborhood of a vowel or is inherent in the vowel itself of a given height m, then nasality must co-occur also in the neighborhood of or be inherent in a vowel of any lesser degree of height. That is, low vowels favour nasality.
- Standardized
- IF nasality co-occurs in the neighborhood of a vowel or is inherent in the vowel itself of a given height m, THEN nasality must co-occur also in the neighborhood of or be inherent in a vowel of any lesser degree of height.
Instantiations of this general implication:
IF a less low vowel is nasalized, THEN any more low vowels are also nasalized.
IF a more low vowel is denasalized, THEN any less low vowels are also denasalized.
IF a consonant is nasalized preceding a less low vowel, THEN this consonant is also nasalized preceding any more low vowels.
IF a consonant is denasalized preceding a more low vowel, THEN this consonant is also denasalized preceding any less low vowels.
Upon nasalization, vowels tend to lower rather than to rise. - Keywords
- nasalization, denasalization, vowel height
- Domain
- phonology
- Type
- implication
- Status
- unclear whether achronic or diachronic
- Quality
- absolute?
- Basis
- languages mentioned in Chen 1973b, survey of 600 Chinese dialects in Chen 1974; survey of American English, Swedish, French, Amoy Chinese, Hindi, Brazilian Portuguese in Clumeck 1976; experiment with 6 native English speakers in Horii & Monroe 1984
- Source
- Chen 1973b: 185, Chen 1974: 913; Clumeck 1976: 348, Ohala 1971 (favorable nasality of low vowels); Horii & Monroe 1984: 201; Hajek 1993: 116, 145-7, cited in Ferguson 1974: 12
- Counterexamples
- 1. Mentioned (and explained away) by Chen himself:Kunyang dialect of Yunnan, Chinese, where nasality co-occurs with high an mid vowels /i/ and //, but lacks with the low vowel /ae/. 2. For arguments against the implication that vowels tend to lower rather than to rise upon nasalization, see: Rochet, B.L. (1974). About a Pseudo Linguistic Universal: The Nasal Vowels Have a Tendency to Lower. Proceedings of the 11th International Congress of Linguists, Bologna, Il Mulino, vol. 2. 3. Bhat (#1090) claims the opposite, i.e. vowels tend to rise upon nasaliszation.
1. Chen is ambivalent between a diachronic and an achronic interpretation of his implications (steps in the historical spread of (de-)nasalization vs. universal constraints on phoneme systems valid at all times, regardless of the history of a language). However, since he suggests that counterexamples (on the achronic interpretation) can possibly be explained away by assuming overlapping cycles of nasalization and denasalization, his implications are probably to be read diachronically as determining the historical spread of nasalization and denasalization.However, he is clearly inconsistent in explaining away the counterevidence of the Kunyang system. On his own theory of nasalization spreading from low to high vowels and denasalization from high to low vowels, the high vowel ought to be the first to denasalize after all vowels had been nasalized, yielding this system:HIGH ORAL /i/ vs. HIGH NASAL /-/; MID ORAL // vs. MID NASAL /€/; LOW ORAL /ae/ vs. LOW NASAL /ae€/.Instead Cheng suggests in footnote 7 that the scenario is to be one of first-in-first-out, with the vowel first nasalized (i.e., low) also denasalizing first.A phoneme system that would be ruled out on any diachronic scenario is this:HIGH ORAL /i/ vs. HIGH NASAL /-/; MID ORAL // vs. MID NASAL /€/; LOW ORAL /ae/ vs. LOW NASAL /-/. 2. Cf. #1607. 2. Cf. #1866.