!

of five lines, in I. iv. VI. 1, 11. 3-7; of six lines, in I. v. III. 4, 11. 2 -7; of seven lines, in I. v. IV. 6, 11. 2—8; of eight lines, in I. v. IV. 1, II. 1—8; of nine lines, in III. ii. VI. 1, ll. 2—10; of ten lines, in II. vi. V. 2, II. 1—10; of eleven lines, in IV. iii. II. II. 12-22; and even of twelve lines, in IV. ii. IV. 4, II. 1—12.

[ii] Where the rhyming lines are interrupted by one or more lines intervening which do not rhyme with them.8 Thus in I. i. I. 1, ll. 1, 2, and 4 rhyme, separated by 1. 3, which does not; and in I. xv. I. 5, ll. 1—5 rhyme; 1. 6, not rhyming, intervenes; and the rhyme is resumed in 11. 7-9. Then come two lines, not rhyming, and 1. 13, which closes the stanza, resumes the rhyme again.

The rhymes are sometimes wide apart, the intervening lines not rhyming at all, or rhyming differently together. E. g., in III. iii. II. 3, a stanza of eight lines, only 11. 2 and 8 can be said to rhyme, though Twan-she makes out an irregular rhyme between 11. 4 and 6. In III. ii. I. 3, ll. 2 and 6 rhyme, two of the intervening lines, 3 and 4 being assonances, and 5 not rhyming at all; and in st. 8, 11. 4 and 8 rhyme, with intervening lines all rhyming differently together.

[iii] Where the stanza contains only one rhyme, as I. i. I. 1.10 Sometimes two stanzas succeed each other, with the same rhyme in both, as stt. 7, 8 of II. iii. V., and 3, 4 of III. i. VIII.

[iv.] Where the stanza contains two or more rhymes,11 as I. i. I. 2; II. vii. VI. 1.

Where the different rhymes alternate12;-with more or less regularity or irregularity. In I. i. VII. the stanzas are quatrains proper, 11. 1 and 3 rhyming together in each, and also 11. 2 and 4. In I. ii. VI. 3, containing six lines, II. 1 and 3 rhyme, and also 11: 2 and 4, whose rhyme is then continued in l. 5, 6. So in I. ii. X., the stanzas of which are of five lines, 11. 1 and 3, rhyme, and then 11. 2, 4, 5. In I. i. II. 1, II. 2 and 5 rhyme, and then 11. 3, 4, 6. In III. iii. VII. 1, 11. 2, 4, 6 rhyme; ll. 3 and 5; and then 11. 8, 9, 10, 12. [vi.] Where one or more lines at the commencement of the dif ferent stanzas in a piece, or their concluding lines, rhyme with one. another. 13 The former case occurs in I. xv. III.: II. vi. VIII.: III. iii. I. 2-8; the latter, in I. i. XI.; ii. XIV.; iv. IV.; vi. III.; vii.

XIII.; xi. X.: III. i. X.: IV. ii. II. But in all these instances we

8閒句韻 隔數句遙韻, and 隔韻遙韻

11

章

13 隔章

10

韻

章易韻: 12 隔韻 and 三句隔韻 章首遙韻 and 隔章尾句遙韻

have the repetition of the whole lines, and not of the rhymes in them only.

[vii.] What we call medial rhymes are found occasionally.1 E. I. iii. I. 5, 1. 1; IX. 2, 1. 2 (doubtful); XVI. 1, 2, 3, 1. 5; iv. III. 1, l. 3; xiv. II. 4, ll. 1, 2: II. v. VI. 1, 2, 1. 1: IV. iii. I., l. 1. Keang gives two instances under this case, where the members of different lines in the same stanza rhyme:-I. ii. X., 2, 11. 2, 4, and III. ii. VIII. 9, II. 5, 6. Without specifying any additional characteristics of the rhymes, which the minute research of native scholars has pointed out, it is to be observed that in all the Parts of the She, there are multitudes of lines, sometimes one, and sometimes more, which do not rhyme with any others, in the same stanza, while in Part IV., Book I., there are at least 8 pieces in which there is no attempt at rhyme at all. Even in the 4th and 5th stanzas of III. i. VI., and the 4th stanza of iii. XI., it is only by a violent exercise of poetic license that we can make out any rhymes. We may consider such disregard of rhyme as an approach in Chinese to the structure of blank verse; but while every other irregularity in the ancient odes has met with imitators, I am not aware that this has received any favour. So far from the Chinese having any sympathy with Milton's contempt for rhyming as 'a jingling sound of like endings,' ‘a troublesome bondage,' they consider rhyme as essential to poetry.

poems and the tones.

6. The only other point which it is necessary to consider in this section is, whether the rhymes of the She were affected by what every Chinese scholar knows as the four tones, and an accurate acquaintance with which is now essential, not only to the making of The relation of the ancient poetry, but even to speaking so as to be freely and readily understood. And on this subject there is considerable difference of opinion between those who have most deeply studied it. One of the cases instanced by Keang Yung in regard to the rhymes, and which I have not adduced in the preceding paragraph, is that characters of the same termination rhyme together though they may be in different tones; 15 and this he endeavours to support by reference to more than 200 stanzas where he contends that the rhymes are altogether independent of the tones.16 Keang in

14 句中韻 15 四聲通

(t. 1) and (t. 4) rhyme; in IX. 1, 2, 3,

I. 1, (t. 1) and (t. 3); in iv. V. 2, 駸 (t.1), (t. 2.); in III ii. I. 1,

and 翼 t. 4); in st. 5, 道草茂苞襃秀好(tt. 1, 2, 3).

(t. 4); in st. 3, (t. 3),

this view followed Koo Ning-jin or Koo Yen-woo (A.D.1,603-1682),17 distinguished by his varied scholarship, and especially by his researches into the ancient rhymes. In opposition to them, Twan Mow-tang, or Twan Yuh-tsae (A.D. 1,735-1,815), 18 contends that we ought to acknowledge three tones, the 1st, the 2d, and the 4th, in the She. He says:-The tones of characters anciently were different from what they are now, just as the ancient rhyming endings were different from the present. Examining the compositions of the Chow and Ts'in dynasties, and the earlier portion of the Han, we find that there were then the 1st, 2d, and 4th tones, but not the 3d. During the dynasties of Wei and Tsin (A.D. 227-419), many words in the 2d and 4th tones assumed the 3d, and many in the 1st tone fell into one or other of the others. In this way there were the four tones complete; but in many cases they were different from what they had anciently been. Characters formerly of the 1st tone were now in one of the others, and many formerly in the 2d and 4th tones were now in the 3d. By diligent research the fact and the process of the change can be ascertained.'19 Admitting, as I believe we ought to do, what is here claimed, that the tones of many of the characters were different anciently from that they became in the 3d and 4th centuries, there is not much difficulty in approximating the views of Twan and Koo to each other. The latter says:-'Although the discussion of the four tones arose only when the capital was on the left of the Keang [say in our 5th and 6th centuries], yet the poetical compositions of the ancients had their characters distinguished in pronunciation as slow or rapid, light or heavy, and hence those now in the even tone rhymed together, as did those in the other tones. Yet it was by no means always so. The tones of characters have changed. In fact anciently these tones were simply the variations of pronunciation made by the voice of the singer, now high now low, now repressed now put forth. And thus the four tones could be used to rhyme together. '20 Three tones existed anciently, according to

Twan. 'No,' says Koo, 'there were no tones; but only certain

19 See the

17 顧甯人or顧炎武 18段茂堂 or 段玉裁 書音均表古四聲說; in the 皇清經解卷六百五十六, p. 16. 20 See the 音論古人四聲一貫; in the 皇清經解,卷四, p. 7.

Koo says that the discussion of the four tones arose on the left of the Keang;' i.e., during the time of the various dynas ies, which had their capital in the ancient Kin-ling, thence called Nanking, or the southern capital, during the greater portion of the 5th and 6th centuries. I have translated the rest of the passage according to the sense of it, without attempting to make a literal version.