2 Inflected and Uninflected Adjectives in Japanese 653.3. UNINFLECTED ADJECTIVES AND NOUNS
Overlap is also found between uninflected adjectives and nouns. Uehara (1998:
215) finds that 151 (57 per cent) out of 264 items which combine with na or wit
both na and no also occur followed by case particles; in addition, as indicated earl-
ier, Uehara does not consider adjectives which combine only with no, and these
are noted by Martin as commonly belonging to at least one other class, includ-
ing noun (1975: 822). Examples can be found in Martin (1975:180-1, 822-5), and in
Uehara (1998:100-2).
While sizeable overlap clearly exists, many common uninflected adjectives have
no noun counterparts. Furthermore, where nouns are found, they are often highly
restricted in distribution. To give one illustration, inspection of texts indicates that
the noun taboo (cf. taboo (nalno) 'very busy'), given as an example in Uehara (1998:
101-2), occurs overwhelmingly in the phrase taboo o kiwameru 'become extremely
busy' and a few other expressions including gotaboo no naka and the derivative
gotabootyuu (both 'while (you) are very busy'), taboo ni tuki, and taboo ni yori
(both 'due to pressure of work'). Nouns like these thus function to a greater or
lesser degree as elements in larger lexicalized units. The deadjectival noun in -sa
(taboosa) is also readily attested, and this too is by no means untypical for these
cases. Elsewhere, some adjectives and nouns show unpredictable semantic differ-
ences: cf. nounyooki 'weather conditions' vs. adjective yooki (nalno) 'jovial'; geNki
'energy' vs. geNki (na) 'energetic' but also 'well, in good health'; heiwa 'peace' vs.
heiwa (na) 'peaceful' but also 'carefree, careless, inattentive' (Uehara 1998:122). How
widespread such factors are must be determined by further detailed research, but
the evidence suggests that a variety of relationships are found between adjectives
and nouns involved in this type of multiple membership.
4. Wider linguistic features of adjective types
4.1. LEXICAL MAKE-UP
A variety of linguistic factors appear to be relevant to the membership of adjec-
tive types, and they are summarized here. Lexical semantic relations across types,
and the distribution of adjectives across semantic domains, are illustrated specif-
ically in §4.2.4.1.1. Inflected adjectives
Inflected adjectives show strong phonological restrictions (Backhouse 1984: 179-
180). All stems must end in a, u, o, or i (overwhelmingly in the inflected adjective-
forming element -si); i.e. stem-final e and AT (the latter common in Sino items) are
excluded, and stem-final i is dispreferred outside of -si. Lexically simple inflected
adjectives are native, and (C)V syllables are preferred in these stems.
In relation to semantics, Nisio (1972:160) notes that basic native inflected adjec-