No CrossRef data available.
We use cookies to distinguish you from other users and to provide you with a better experience on our websites. Close this message to accept cookies or find out how to manage your cookie settings.
We use cookies to distinguish you from other users and to provide you with a better experience on our websites. Close this message to accept cookies or find out how to manage your cookie settings.
Published online by Cambridge University Press: 24 December 2009
Since Chinese is the crucial example of a purely isolating language, linguists have long been interested in any phenomena which suggest that some classes of Chinese words may once have been infected. A point which has attracted especial attention is the existence of a pair of 1st person pronouns, wu/*ngo- 
and wo/*ngấ 
, of which the latter has free distribution but the former is generally subject or possessive down to the second century B.C. Half a century ago Karlgren pointed out the same terminations in a pair of 2nd person pronouns, ju/*ń
o' 
and erh/* 
, and suggested that proto-Chinese had pronoun declension, with -o marking the nominative and genitive and -a the accusative (but his evidence of restricted distribution was inconclusive except in the case of wu). Later Kennedy preferred to explain the distribution of wu by classing it with various other pronouns and particles in which the level tone is associated with uncompleted utterance (so that either, like wu, they are never phrase-final, or they are phrase-final but interrogative, as with hu/*g'o- 
and the other interrogative particles). A further important contribution is Chou Fa-kao's discovery of a possessive termination -ǝg, which he ascribes to fusion with the possessive particle chih/*
ǝg- 
. In 1969 I myself proposed a classification of the pronouns into two systems, the pre-Classical and the Classical.
1 Karlgren, 1920. For list of references, see p. 298. Being ignorant of phonology, I continue to use the archaic forms as reconstructed by Karlgren. It may be noticed that, although the morphological regularities of the pronouns are generally constant in different reconstructions, whether one locates them in the terminations (and therefore describes the pronoun as inflected) may well be affected by the choice of reconstruction.
2 Kennedy, art. cit.
3 Chou, 13.
4 Graham, 1969, 54.
5 Graham, 1972, 95–7.
6 Graham, 1969, 31–5.
7 For the generalizing effect of fu on common nouns, cf. Graham, 1972, 98 f.
8 Tao-chuan 
, Duke Hsiang 
4, 14.
9 I am not sufficiently at home with the bronze inscriptions to offer these interpretations as more than suggestions which appear to me plausible when the passages are read in context. But the phrase 
, only interpretable as ‘our ruler’, appears in an oath formula in tablets from the state of Chin 
excavated in Hou-ma 
and published in 
1966, [No.] 2, 1–6. They are discussed by Ch'en Meng-chia (art. cit.), who dates them in the late fifth century B.C. The formula recurs in more recent Hou-ma discoveries (Wen-wu, 1972, [No.] 3, 5–7). I am indebted to Mr. Lee Yim 
for this additional example, and to Professor C. C. Shih for further advice on the inscriptions.
10 In favour of the fusion hypothesis I pointed out that judging by the prosody of the Ch'u tz'ŭ 
the graphs 
represent not two syllables (*d o-*ǝg-) but one, presumably the *d ǝg- written 
in the ‘Documents’; cf. Graham, 1969, 55. However this is evidence of the current pronunciation (suggesting that the particle could be pronounced indistinguishably from an -ǝg ending) rather than of the etymology.
11 For the question of the tones of *d o- and *ńo', cf. Graham, 1969, 28 f.
12 Since in Classical usage this pronoun is added to the neutral demonstratives to make them possessive (
X ‘the X of this man/thing’), there are advantages in treating it as a possessive common to all demonstratives rather than as a 3rd person pronoun.
13 Graham, 1969, 52 f.
14 For wang cf. Yang, 26; Graham, 1961, 174–6.
15 Graham, 1969, 32.
16 Graham, 1972, 89.