Shuōwén jiězì 說文解字
Explaining Graphs and Analyzing Characters by 許愼 (Xǔ Shèn, 撰), revised and augmented by 徐鉉 (Xú Xuàn, 增釋)
About the work
The earliest comprehensive Chinese dictionary based on graphic structure: 9,353 head graphs (with 1,163 chóngwén 重文 graphic variants) organized under 540 bùshǒu 部首 (radicals), each entry giving the small-seal form, a brief definition, an analysis of structure (the liùshū 六書), and frequently a phonetic gloss and / or canonical citation. Composed by Xǔ Shèn 許慎 (c. 58–147) of Rǔnán; presented by his son Xǔ Chōng 許冲 in Jiànguāng 1 (121). The standard transmitted form is the DàXúběn 大徐本 — Xú Xuàn’s 徐鉉 revised edition of Yōngxī 3 (986), which adjusted the structure into thirty juàn (each of the original 14 piān split into upper and lower halves, plus the postface), supplied fǎnqiè readings from Sūn Miǎn’s Tángyùn, and added a fresh layer of xīnfùzì 新附字. The WYG version follows this DàXúběn.
Tiyao
Shuōwén jiězì in thirty juàn. — Composed by Xǔ Shèn of the Hàn. Shèn’s zì was Shūzhòng 叔重; he was a man of Rǔnán; he rose to Tàiwèi nángé jìjiǔ. The book was completed in Hédì Yǒngyuán 12 (100); its 14 piān plus the mùlù in one piān total 15 piān; 540 bù; 9,353 wén; 1,163 chóngwén; 133,440 graphs of definition. It pursues the meaning of the liùshū, dividing into bù and grouping by similarity; the technique is most precise. Yet the diction is plain and concise, hard to read at first encounter; and given that pronunciations have shifted, ancient and current readings differing, xiéshēng graphs are likewise often hard to follow — hence transmitted copies have come down with many corruptions. — In Sòng Yōngxī 3 (986), by imperial command, Xú Xuàn 徐鉉, Gě Tuān 葛湍, Wáng Wéigōng 王惟恭, Jù Zhōngzhèng 句中正 and others undertook a thorough revision. Whatever graphs the Shuōwén itself names in its postface or the xùlì but did not include in the bù were now fully retrieved and supplemented; further graphs of canonical use that the Shuōwén lacked were added under the heading xīnfùzì 新附字. Where there was an established orthography but the popular form was a corruption, this is set out in the gloss; where the form contradicted the liùshū, that is gathered separately at the end of the juàn; where the gloss was incomplete, supplementary explanation is appended, marked “Chén Xuàn děng àn 臣鉉等案” to distinguish it. Fǎnqiè readings are uniformly given from Sūn Miǎn’s Tángyùn. Because of the bulk, each juàn was split into upper and lower halves — yielding the present 30-juàn form, which is what circulates today as the Máo Jìn print. In the Wànlì period of the Míng, Gōngshì printed Lǐ Tāo’s Shuōwén wǔyīn yùnpǔ, and Chén Dàkē’s preface mistook it for Xú Xuàn’s edition; Chén Qǐyuán’s Máoshī jīgǔpiān and Gù Yánwǔ’s Rìzhī lù both repeated the error. Was Máo Jìn’s print not yet in wide circulation in the early Qīng? — Lǐ Tāo, following the Táng Lín Hǎn’s view, argued that the gǔwén and zhòuwén (clerical-archaic and Zhòu-script forms) in the Shuōwén were added by the Jìn Wénlìng Lǚ Chén 呂忱. Examining Xǔ Shèn’s own preface — “I now sequence the zhuànwén, joining to it the gǔwén and zhòuwén” — the original wording is unambiguous, and the recorded counts of chóngwén match. The Fǎshū yàolù records the Northern-Wèi Jiāng Shì 江式 Memorial on the Script, saying: “In the Jìn dynasty, the Yìyángwáng diǎncí lìng Rénchéng Lǚ Chén submitted his Zìlín in six juàn. Examining its tendency: it attaches itself to Xǔ Shèn’s Shuōwén and tunes its ànǒu by zhāngjù, separates out gǔ and zhòu, and renders rare and obscure graphs in standard lìshū without losing the seal-script intent” — i.e., Lǚ Chén’s Zìlín did not use gǔzhòu forms; it used standard clerical. Lín Hǎn’s claim that “Lǚ Chén’s Zìlín often supplements lacunae in Xǔ Shèn” is correct only in this restricted sense — Zìlín added graphs the Shuōwén did not contain (e.g., the Guǎngyùn dōng 1 bù’s tóng 烔 and qǐng 谾, the jiāng 4 bù’s nóng 噥 — all Zìlín graphs absent from Shuōwén); it did not add gǔzhòu glosses. Lǐ Tāo took the gǔzhòu forms in the Shuōwén to be Lǚ Chén’s additions — a serious error. — From WèiJìn down, those who treat xiǎoxué all take Xǔ Shèn as ancestor. Lǐ Yángbīng 李陽冰 alone began to stretch and reject Xǔ — his stance is not even-handed. Yet Xǔ’s book takes xiǎozhuàn as its base, and lìshū, xíngshū, and cǎoshū are each their own structure, with zīshēng and transformation, sometimes uniform and sometimes divergent — they cannot all be regulated by xiǎozhuàn. Hence Yán Yuánsūn’s 顏元孫 Gānlù zìshū KR1j0023 says: “Since the change from seal to clerical, the original [forms] have gradually been lost; if we relied solely on the Shuōwén, every stroke would be encumbered. We must shed extremes and find balance.” Xú Xuàn’s Memorial of Presentation says: “For lofty texts and grand records, zhuàn and zhòu should indeed be inscribed in metal and stone; but for ordinary documents and bamboo strips, cǎo and lì are sufficient.” Both are masters of xiǎoxué, and their judgment is the same. Míng Huáng Jiàn 黃諫 in his Cónggǔ zhèngwén would convert all lì back to zhuàn — does he understand the principle of the liùshū? — As for the Shuōwén’s quotations from the Five Classics, they often differ from the present text and sometimes contradict each other internally. Gù Yánwǔ in the Rìzhī lù notes that under fán 汜 it cites “jiāng yǒu fán 江有汜” but under yí 洍 it cites “jiāng yǒu yí 江有洍”; under jǐn 巹 “chìxì jǐjǐ 赤舄巳巳” but under qiān 掔 “chìxì qiānqiān 赤舄掔掔”; so its quotation of “the Shī per Máo’s text” disagrees with the present Máo Shī. Even within one school, sub-lineages diverge and texts no longer match. To take this as authority for emending the Classics would be especially absurd; one must grasp the intent and not be a slave to the trace — that is what it is to read the Shuōwén well. (Translated from the Sìkù tíyào at Zinbun digital tíyào 0084801.html, with the long appended note on the supposed “Kǒngshì” gǔwén Shàngshū citations omitted for length.)
Abstract
The Shuōwén jiězì is the foundational work of the entire Chinese philological tradition. It transmits, organizes, and analyzes ~9,400 small-seal-script graphs under 540 radicals (bù) and supplies for each a structural analysis grounded in Xǔ Shèn’s six-script doctrine (liùshū: xiàngxíng, zhǐshì, huìyì, xíngshēng, zhuǎnzhù, jiǎjiè) — a taxonomy that has dominated Chinese paleography ever since. Xǔ Shèn’s stated motivation was to recover the original meanings of the Classics, which he believed were obscured by the post-Qín shift to lìshū; by analyzing the seal-script structure of each graph he hoped to “stabilize the meaning of the Five Classics.” The work was presented by Xǔ Chōng to Hàn Āndì in Jiànguāng 1 (121) but was substantially complete by Yǒngyuán 12 (100) per the postface. The text passed in manuscript through the medieval period and was substantially revised in Northern Sòng Yōngxī 3 (986) by Xú Xuàn and his team, whose recension — the DàXúběn — is the basis of the WYG and all later printed editions; Xú added the 30-juàn split, the fǎnqiè glosses (from the Tángyùn), the xīnfùzì, and the editorial ànyǔ. Xú Xuàn’s younger brother Xú Kǎi 徐鍇 independently produced the XiǎoXúběn (now KR1j0019 Shuōwén xìzhuàn), the only surviving deep philological commentary from the period. The Sìkù tíyào defends the Shuōwén’s integrity against Lǐ Tāo’s claim that the gǔwén and zhòuwén forms were Jìn additions by Lǚ Chén, by quoting Xǔ Shèn’s preface and Jiāng Shì’s Memorial. Wilkinson §6.2.1.3 surveys the dictionary at length and assesses both the Da-Xú-běn tradition and the principal Qing commentators (Duàn Yùcái, Zhū Jùnshēng, Guì Fù, Wáng Yún). The Shuōwén lemmata count was extensively recalculated by modern scholars; the standard modern figure is 9,449 wén + 1,285 chóngwén (Zhū Mǐnshēn 1998).
Translations and research
- Bottéro, Françoise. 1996. Sémantisme et classification dans l’écriture chinoise: Les systèmes de classement des caractères par clés du Shuowen Jiezi au Kangxi Zidian. Collège de France, IHEC.
- Bottéro, Françoise and Christoph Harbsmeier. 2008. “The Shuowen Jiezi Dictionary and the Human Sciences in China.” Asia Major 21.1: 249–271.
- Bottéro, Françoise. 2015. “Shuowen Jiezi 說文解字.” In Early Chinese Texts: A Bibliographical Guide (ECLL).
- Boltz, William G. 2015. “Shuowen jiezi 說文解字.” In ECLL.
- O’Neill, Timothy Michael. 2013. “Xu Shen’s Scholarly Agenda: A New Interpretation of the Postface of the Shuowen jiezi.” JAOS 133.3: 413–440.
- Serruys, Paul L.-M. 1984. “On the System of the pu shou (部首) in the Shuo-wen chieh-tzu 说文解字.” SYSJK 55: 651–754.
- Duàn Yùcái 段玉裁. 1813–15. Shuōwén jiězì zhù 說文解字注. Reprinted Shanghai: Shanghai guji, 1981. — The standard Qing variorum.
- Tāng Kějìng 湯可敬, ed. 2018 (1997). Shuōwén jiězì jīn-shì 說文解字今釋. Beijing: Zhonghua. — Modern annotated edition.
- Endymion Wilkinson. 2022. Chinese History: A New Manual, §6.2.1.3 (extensive treatment).
Other points of interest
The Shuōwén is the source of (1) the radical-based dictionary tradition (its 540 radicals were reduced to 214 in the Míng Zìhuì and Zhèngzìtōng — the radical set used by the Kāngxī zìdiǎn KR1j0048); (2) the canonical taxonomy of liùshū character-formation; (3) much of the surviving evidence for Eastern-Hàn pronunciation, paleographic forms, and lexical reach. The Sìkù tíyào contains a substantial appended note (BLOCKQUOTE) reconstructing the textual history of the Shàngshū citations in the Shuōwén, defending the position that Xǔ Shèn’s “Kǒngshì” gǔwén Shàngshū is not the present 58-piān recension but rather the Dù Lín 杜林 qīshū line transmitted via Jiǎ Kuí 賈逵 — a critical-textual point taken up later by Yán Ruòjù.