Adjective Classes - A Cross-Linguistic Typology

212 Greville G. Corbett

alternative source is Ilola and Mustajoki (1989: 7), using 98,728 items, based on
Zaliznjak(i977).

(24) Distribution of the major parts of speech in Russian: types (%)

Noun

Verb

Adjective

Other

Lazova^9

(121,532 words)

46.4

30.7

20.4

2.6

Zaliznjak/Ilola and Mustajoki

(98,728 words)

47.6

28.8

21.1

2.5

In terms of types, the share of the adjective is substantial. All three major parts
of speech naturally have a lower share in terms of tokens. However, for the adjec-
tive there is also considerable variation across genres. This can be illustrated from
Zasorinas frequency dictionary. The corpus was around one million words of Rus-
sian, divided into four approximately equal sub-corpora: I: newspapers and maga-
zines, II: drama (intended to reflect conversational Russian), III: non-fiction (espe-
cially political texts), IV: literary prose. When the corpus is described according to
tokens, we reach the picture in (25), calculated from Zasorina (1977: 927).

(25) Distribution of major parts of speech in a one-million word corpus:
tokens (%)

Noun

Verb

Adjective

Other

Total^10

26.7

17.1

9-4

46.9

Sub-corpora

I

32.8

14.5

12.O

40.7

II

2O.4

2O.9

6.2

52.6

III

31.0

13.5

12.5

43.1

IV

234

19.0

7-4

50.2

There are considerable differences between the genres. However, the most strik-
ing is with adjectives, where the figure for non-fiction is all but double that for
drama.

10 Semantic types

The core semantic types for adjectives, as established cross-linguistically, are all
expressed most naturally by adjectives in Russian. See Raxilina (2000: 104-233)
for discussion of these types. We shall illustrate each type (we cannot give all

(^9) The columns do not total to exactly 100 because of rounding.
(^10) Again, not all columns total to 100 because of rounding.