Shuǐjīng zhù 水經注
Commentary on the Classic of Waterways by 酈道元 (Lì Dàoyuán, d. 527) — zhuàn 撰
About the work
Lì Dàoyuán’s commentary on the Shuǐjīng 水經 is the foundational monument of Chinese historical hydrology and a major text of medieval Chinese descriptive geography. The base text Shuǐjīng — anonymous, written ca. mid-3rd century by an author traditionally but spuriously identified as the Hàn-era Sāng Qīn 桑欽 — listed 137 watercourses in cursory entries; Lì Dàoyuán’s commentary expanded each waterway into an elaborate topographical, historical, ethnographic, archaeological, and folkloric treatment, citing some 437 distinct earlier works (many otherwise lost). The result is a 40-juan compendium that is at once a hydrological atlas, a historical-geographical encyclopedia, a textual archive of pre-medieval lore, and a literary masterwork.
Tiyao
We respectfully note: the Shuǐjīng zhù in 40 juan is the work of Lì Dàoyuán of the HòuWèi (Northern Wèi). Dàoyuán, zì Shàncháng 善長, of Fànyáng 范陽; he rose to Yùshǐzhōngwèi 御史中尉. From the Jìn onward, two authors annotated the Shuǐjīng: Guō Pú’s commentary in 3 juan, still seen by Dù Yòu in writing the Tōngdiǎn. Today only Dàoyuán’s commentary survives. The Chóngwén zǒngmù states that 5 juan of it had already been lost; hence the Yuánhé jùnxiàn zhì and the Tàipíng huányǔ jì, citing it for the Hūtuó River, the Jīng River, and the Luò River, all unmatched in the present text. Yet the present text remains in 40 juan — presumably the Sòng reprinters had subdivided in order to reach the original number.
This work, since the Míng, has had no good edition; only Zhū Móuwěi 朱謀㙔’s collation has circulated, and yet errors continue. Now, on the basis of the Yǒnglè dàdiǎn’s citations, comparing each entry under its waterway, not only do textual errors come up tier upon tier, but among the lacunae and dislocated passages there are some from a few dozen to over four hundred characters; and Dàoyuán’s own preface — lost in all extant copies — likewise survives only in the Yǒnglè dàdiǎn. The text upon which the Yǒnglè compilers drew was clearly still a fine Sòng impression. We have therefore aligned the original text against the recent printing: 2,128 characters supplied to fill lacunae; 1,448 characters of spurious addition removed; 3,715 characters of conjectural emendation corrected. The brilliance returned, the old appearance restored: the 300- to 400-year doubts dissolved as if a film had been lifted away. This is wholly attributable to Our August Sovereign’s reverence for antiquity and exaltation of letters: the secrets of Lángxuán and Wǎnwěi in resonant unison brought to manifestation, so that the bequeathed compilations of former ages, fortunately encountering this prosperous reign, send forth their light from amid the bookworms’ destruction — as though awaiting the August Court before they should emerge: this too is an opportunity granted but once in an age.
As to the text of the classic and the language of the commentary — in all extant editions much confused — we now investigate the older prose for clues. Where the watercourse passes through a place, the jīng (classic) says guò 過, the commentary says jìng 逕; the jīng states only the chief towns, while the commentary extends to minor place-names; for any one waterway-name, the jīng announces it in the opening phrase and does not repeat it; the commentary, with its many digressions, must repeat the name to mark a fresh start; for any commandery or county within the work, the jīng gives only the contemporary name, while the commentary supplements with the traces of older walled towns. We have followed these compositional principles in setting the divisions in order, and added our annotations below each entry.
As to the streams beyond the frontier, and the channels south of the Chángjiāng — Dàoyuán’s footsteps had not reached these. Hence on the true source of the Luán 灤 River, the order of the Sānzàng 三藏 waters, and the establishment of Báitán 白檀 and Yāoyáng 要陽, his account cannot avoid forced association and confusion. The most extreme is his arbitrary conflation of the Zhèjiāng with the Yáojiāng — patently a distortion through hearsay. Since Our August Sovereign has dispatched an envoy to inspect on the spot, the full topology of these channels has been ascertained; and Yùzhì Rèhé kǎo (Imperial Examination of the Rèhé) and the Luányuán kǎozhèng essays critically dissect the errors and analyze the cases — sufficient to settle perpetually a millennium of hearsay-corruption. We have respectfully recorded these as crowning the front pages, as a permanent definitive judgment.
Further, the author of the Shuǐjīng — the Tángshū gave it as Sāng Qīn 桑欽; yet Bān Gù once cited Qīn’s account, and it differed from this classic; Dàoyuán’s commentary cites Qīn’s dìlǐzhì, not calling it Shuǐjīng. Looking at the Fúshuǐ entry, where it states that Guǎnghàn has already become GuǎngWèi, this is decidedly not from the Hàn period; in the Zhōngshuǐ entry, where it states that Jìnníng is still called Wèiníng, this had not yet reached the Jìn era. Working through the wording, the author was probably a person of the Three Kingdoms period. Now that we have recovered Dàoyuán’s original preface, and know that it makes no mention of Sāng Qīn, we have on this basis struck out the older attribution — perhaps in keeping with the principle of withholding judgment under uncertainty.
Reverently collated and submitted, tenth month, Qiánlóng 39 (1774). General Editorial Officer, Reading-in-Waiting (shìdú) Jì Yún 紀昀; Reading-in-Waiting Lù Xīxióng 陸錫熊; Compilation Officer, jǔrén Dài Zhèn 戴震.
Abstract
The Shuǐjīng zhù was composed by Lì Dàoyuán (zì Shàncháng, 范陽 — fl. ca. 515–527) at the height of his Northern Wèi career; he rose to Yùshǐzhōngwèi (Vice Censor-in-Chief), a position from which he was killed in 527 in the suppression of the rebellion of Xiāo Bǎoyīn. The biography in Wèishū 89 (Kùlì 酷吏 zhuàn) characterizes him as harsh and exacting in office. The work was finished by 527 at the latest, and most likely composed in the period of his Inspector-General postings in the early 520s. The Shuǐjīng base text on which Lì comments was anonymous; the Tángshū yìwénzhì attribution to Sāng Qīn 桑欽 (Hàn) is a misidentification — internal evidence (Wèi place-names, Three Kingdoms commandery boundaries) points unambiguously to a 3rd-century date, and the Sìkù compilers correctly removed the Sāng Qīn attribution.
The Sìkù Shuǐjīng zhù edition is one of the great triumphs of 18th-century kǎojù scholarship. The principal recension before the Sìkù was the late-Míng Zhū Móuwěi 朱謀㙔 (d. 1624) edition; the Sìkù compilers — most importantly Dài Zhèn 戴震 (1724–1777), but also Quán Zǔwàng 全祖望, Zhào Yīqīng 趙一清 (KR2k0060), and Shěn Bǐngxùn 沈炳巽 (KR2k0059) — undertook a comprehensive Yǒnglè dàdiǎn-based reconstruction, restoring 2,128 characters, removing 1,448 spurious additions, and correcting 3,715 conjectural emendations. The polemic over the jīngzhù (classic-commentary) separation principle generated one of the most celebrated controversies in Qīng textual scholarship — Dài Zhèn was accused of plagiarizing Zhào Yīqīng’s separation criteria. The recovery of Lì Dàoyuán’s lost preface from the Yǒnglè dàdiǎn is a particularly significant editorial achievement.
The principal modern critical edition is Yáng Shǒujìng 楊守敬 and Xióng Huìzhēn 熊會貞, Shuǐjīng zhù shū 水經注疏 (40 juan, completed 1904, published 1955; reissue Jiāngsū gǔjí, 1989) — a magisterial 9-million-character study integrating archaeological and modern cartographic materials with the Lì Dàoyuán base text. Wilkinson §31 lists the Shuǐjīng zhù among the principal monuments of medieval Chinese descriptive geography.
Translations and research
Partial English translations: Hans van Ess and others have produced selective translations of individual chapters; no complete English translation has yet been published. German: Albrecht Herrmann, Lou-lan: China, Indien und Rom im Lichte der Ausgrabungen am Lobnor (Leipzig, 1931), with extensive use of Shuǐjīng zhù materials. Russian: V. A. Panasiuk, Шуй-цзин чжу (Moscow, 1962, partial translation). Japanese: Mori Kazue 森鹿三 and Yamada Toshiaki 山田俊雄, Suikei chū sho 水経注疏 (Tokyo, 1970–1981, partial). The standard modern Chinese critical edition is Chén Qiáoyì 陳橋驛, Shuǐjīng zhù jiào-zhèng 水經注校證 (Zhōnghuá shūjú, 2007). The principal anglophone study is Robert P. Hymes, “The Shuijing zhu,” in The Cambridge History of China vol. 2 (forthcoming); see also Olivia Milburn, The Glory of Yue: An Annotated Translation of the Yuejue shu (Brill, 2010), with comparative remarks on the Shuǐjīng zhù tradition. For the kǎojù controversy see Hu Shih, Hu Shi quanji (Anhui jiaoyu, 2003) vol. 13, with his major Shuǐjīng zhù monograph; and Benjamin A. Elman, From Philosophy to Philology (Harvard, 1984; rev. 2001), §6 on the Dài Zhèn–Zhào Yīqīng controversy.
Other points of interest
The Shuǐjīng zhù’s textual history — and especially the Qīng-era Dài Zhèn–Zhào Yīqīng controversy over plagiarism in the separation of jīng from zhù — is one of the most celebrated affairs in the history of Qīng kǎojù scholarship. Hú Shì devoted thirty years to the question and produced over 4,000 pages of related research, ultimately defending Dài Zhèn against the Zhào Yīqīng plagiarism allegation. The polemic continues to attract scholarly attention as a window onto early-modern Chinese textual ethics.