Shuōwén xìzhuàn 說文繫傳
A Commentary in Sequence on the Shuōwén by 徐鍇 (Xú Kǎi, 撰) with fǎnqiè readings by 朱翺 (Zhū Áo)
About the work
The earliest sustained philological commentary on the Shuōwén jiězì — predating his elder brother Xú Xuàn’s 徐鉉 DàXúběn recension by a generation — known as the XiǎoXúběn 小徐本 in the standard typology of Shuōwén studies. Xú Kǎi 徐鍇 (920–974) of the Southern Táng presents the Shuōwén in eight thematic divisions across forty juàn: tōngshì 通釋 (30 juàn — the Shuōwén itself, with each of Xǔ Shèn’s 許愼 15 piān split into upper and lower), bùxù 部敘 (2), tōnglùn 通論 (3), qūwàng 祛妄, lèijù 類聚, cuòzōng 錯綜, yíyì 疑義, xìshù 系述 (1 each). Xú Kǎi’s editorial additions are marked Chén Kǎi yuē 臣鍇曰 and Chén Kǎi àn 臣鍇案. The fǎnqiè readings are by the Southern-Táng cháosǎn dàfū Zhū Áo 朱翺.
Tiyao
Shuōwén xìzhuàn in forty juàn. — Composed by Xú Kǎi of the Southern Táng. Kǎi’s zì was Chǔjīn 楚金; he was a man of Guǎnglíng 廣陵 and rose to yòu nèishǐ shèrén. When Sòng troops crossed south into Jiāngnán, he died in the besieged city; his life is recorded in the NánTángshū biography. The book has eight divisions: opening tōngshì (30 juàn) — Xǔ Shèn’s fifteen piān split into thirty; whatever Xú Kǎi has clarified or annotated from the Classics is marked “Chén Kǎi yuē” or “Chén Kǎi àn”; followed by bùxù 2 juàn, tōnglùn 3 juàn, qūwàng, lèijù, cuòzōng, yíyì, xìshù, each one juàn. Qūwàng refutes Lǐ Yángbīng’s 李陽冰 baseless reconstructions; yíyì gathers cases where the Shuōwén radicals are present in the bù but the head graph is missing, and cases where the seal-script form has minor stroke variants in transmission. Bùxù models the Yì xùguàzhuàn in declaring the rationale for the order of the 540 bù. Lèijù groups graphs that take meaning from each other (e.g., yī èr sān sì). Cuòzōng expands by liùshū analysis to ordinary human affairs to draw out the meaning. The closing xìshù is the equivalent of the Shǐjì’s authorial postface. — Xú Kǎi himself separately produced the Shuōwén zhuànyùnpǔ KR1j0021 in five juàn; in the Sòng Xiàozōng era Lǐ Tāo 李燾 used it to produce the Shuōwén jiězì wǔyīnpǔ. Lǐ Tāo’s preface says: “the Yùnpǔ should circulate alongside the Xìzhuàn. Today the Yùnpǔ is sometimes printed by the schools, but the Xìzhuàn has remained in obscurity. Long-collecting only seven or eight parts in ten, missing juàn and graph-corruptions I cannot correct — every time, I sigh.” So already in the Sòng the Xìzhuàn was incomplete. The present transmitted copy is a manuscript only — Qián Zēng’s Dúshū mǐnqiú jì marvels at it as a “secret-treasury” book — but it has many lacunae and errors. The end of the juàn records a Xī-níng-period note by Sū Sòng: “the old book lacks juàn 25 and 30; awaiting a separate copy for supplementation.” The present text has juàn 30 intact (perhaps later supplemented from another copy); but juàn 25 simply transcribes Xú Xuàn’s DàXúběn, having stripped the xīnfùzì. — i.e., later copyists, unable to find the original, made up the gap from Xuàn’s text — like the Wèishū tiānwénzhì, which was lost and supplemented from Zhāng Tàisù’s book. Other lacunae are likewise filled from Xuàn’s edition. — Now Xuàn’s edition uses Sūn Miǎn’s Tángyùn, but Kǎi’s edition uses readings supplied separately by Zhū Áo 朱翺 (cháosǎn dàfū, mìshūshěng jiàoshū láng). Xuàn’s calls the gloss “X-X qiē”; Kǎi’s calls it “X-X fǎn.” Where the present text agrees with Xuàn’s, the gloss-content too is identical — an obvious sign of textual transposition. The clearest case: at shì 示 bù, three xīnfù graphs from Xuàn — tiāo 祧, xiān 祆, zuò 祚 — are inserted (mis-collated to the end of the piān). The text shows clear traces of having been cut down and recombined. — This book was completed before Xú Xuàn’s edition; hence Xuàn’s book often cites it. There are minor and substantive divergences. For example: Xuàn’s “fú 福 means hù 祜”; this book “means bèi 備.” Xuàn’s “[plough] gēng with abundant grass”; this “the name of [implements for] ploughing.” Xuàn’s “zǔjié 𥃡, the jié in front”; this, “dùn in front.” Xuàn’s “liù 鷚, large fledgling”; this, taking the Ěryǎ form, “the heavenly yuè 天鸙.” Xuàn under yìng 禜 cites the Lǐjì; under zhōu 禂 cites the Shī; this, Chén Kǎi àn Lǐjì yuē, Chén Kǎi àn Shī yuē — i.e., Xú Kǎi’s citations have been silently absorbed into the body of Xuàn’s edition under Xǔ Shèn’s name. — Likewise under □ 𣍐 graph “(què 闕)” — Xú Kǎi has “the family text has no gloss; Chén Kǎi àn: I suspect this is something Xǔ’s son Xǔ Chōng spoke.” So Xuàn deleted “the family text has no gloss” and substituted què 闕. He has emended on personal whim — but for the survival of this Xìzhuàn, how would we be able to test the matter? — The book originally came down through Sū Sòng 蘇頌; the seal-script characters were written by Jiānchá Wáng Shèngměi 王聖美 and Hànlín zhīhòu Liú Yǔngōng 劉允恭. The end of the juàn gives “Zǐróng” — that is Sū Sòng’s zì. In Qiándào guǐsì (1173) Yóu Mào 尤袤 obtained it from the Yè Mèngdé family and copied it for Lǐ Tāo (Yóu’s postface gives the details). Notes within the text marked “Chén Cìlì àn” 臣次立案 are by Zhāng Cìlì 張次立, Diànzhōngchéng, who once participated in writing out the Jiāyòu Two-Graph Stone Classics; Táo Zōngyí’s Shūshǐ huìyào records his career. (Translated from Sìkù tíyào at Zinbun 0085001.html.) The Sìkù compilers further note: “this book, though earlier than Xú Xuàn’s, is placed after it because Xuàn’s edition takes Xǔ Shèn as principal and Xuàn supplements; this book takes Kǎi as principal — hence cannot precede Xǔ Shèn.”
Abstract
The Shuōwén xìzhuàn is the only deep philological commentary on the Shuōwén jiězì surviving from the medieval period — predecessor to all the great Qing Shuōwén studies (Duàn Yùcái, Wáng Yún, Zhū Jùnshēng, Guì Fù). Xú Kǎi composed it sometime in the Southern Táng era between c. 962 and his death in the besieged Jīnlíng of 974. The textual transmission has been catastrophic: SòngYuánMíng manuscripts already had major lacunae (notably juàn 25 and 30), and the present version is essentially a Qing-era reconstruction in which a fair fraction of the juàn 25 material was silently transposed from the elder brother’s DàXúběn, identifiable by the use of “qiē” rather than Zhū Áo’s “fǎn” notation, and by the persistence of Xú Xuàn’s xīnfùzì in the shìbù radical. The Sìkù tíyào contains a critical demonstration of cases in which Xú Xuàn silently appropriated Xú Kǎi’s analyses (esp. the Lǐjì and Shī citations under yìng 禜 and zhōu 禂), making the Xìzhuàn essential as a control on the DàXúběn. Wilkinson §6.2.1.3 treats it briefly. The dating bracket notBefore 962 to notAfter 974 (Xú Kǎi’s death) covers the years of probable composition; Lǐ Tāo’s Shuōwén wǔyīn yùnpǔ preface is the principal piece of medieval evidence on the work’s transmission.
Translations and research
- Bottéro, Françoise. 1996. Sémantisme et classification dans l’écriture chinoise. Paris: Collège de France, IHEC.
- Liú Yèqiū 劉葉秋. 1983. Zhōngguó zì-diǎn shǐ-lüè 中國字典史略. Beijing: Zhonghua. — Surveys the Xì-zhuàn in the Shuōwén tradition.
- Endymion Wilkinson. 2022. Chinese History: A New Manual, §6.2.1.3.
Other points of interest
The fǎnqiè readings of Zhū Áo 朱翺 in this work are an independent Southern-Táng phonological dataset, somewhat earlier and southern in dialect orientation, and so a useful supplement to the Tángyùn readings used by Xú Xuàn for the DàXúběn.