文章緣起

Wénzhāng yuánqǐ 文章緣起

The Origins of the Literary Genres by 任昉 (撰), 陳懋仁 (註), 方熊 (補注)

About the work

The Wénzhāng yuánqǐ 文章緣起 (also transmitted as Wénzhāng shǐ 文章始) is a brief proto-treatise traditionally ascribed to Rèn Fǎng 任昉 (460–508), the great Liáng stylist. It is the earliest surviving Chinese attempt to catalogue the origins of named literary genres: 85 short entries, each in the form “X (genre name) — originated with so-and-so” (e.g., “biǎo 表 originated with Lǐ Sī 李斯’s Shàng shū Qín wáng 上書秦王”; “wǎn gē 挽歌 originated with Miào Xí 繆襲”; “duì 對 originated with Sòng Yù 宋玉’s Duì Chǔ wáng wèn 對楚王問”). The book is a single short juǎn with no preface; the Suíshū Jīngjízhì records it under the title Wénzhāng shǐ with the note “yǒu lù wú shū” (catalogue entry but no surviving text). The transmitted version is a Táng reconstruction by Zhāng Jī 張績 (noted under the entry in the Xīn Tángshū Yìwénzhì); the Sìkù editors therefore treat the work as not quite Rèn’s. The Sìkù recension carries two layers of commentary: Chén Màorén 陳懋仁 (late Míng, zì Wúgōng) supplied the original zhù 註; Fāng Xióng 方熊 (early Qīng) added the bǔzhù 補注. Lines of commentary opening with “zhù says” are Chén’s; those opening “bǔzhù says” are Fāng’s.

Tiyao

Wénzhāng yuánqǐ. The transmitted text is attributed to Rèn Fǎng of the Liáng. Examination of the Suíshū Jīngjízhì shows a Wénzhāng shǐ in one juǎn by Rèn Fǎng, with the note “catalogue entry, no surviving text” — so the work was already lost in the Suí. The Xīn Tángshū Yìwénzhì records the same Wénzhāng shǐ in one juǎn by Rèn Fǎng, with the note “supplemented by Zhāng Jī.” Who Zhāng Jī was is unknown, but a Táng supplement is on record — so in the Táng there was already no original. The Sòng compilers of the Tàipíng yùlǎn drew on 1,690 titles, including Zhì Yú’s 摯虞 Wénzhāng liúbié and Lǐ Chōng’s 李充 Hànlín lùn and the like — but they did not record this title.

Now examining the surviving list, the citation is rather careless. Biǎo 表 and ràng biǎo 讓表 are separated into two genres; sāo 騷 and fǎn sāo 反騷 are made distinct forms. The note says wǎn gē 挽歌 began with Miào Xí 繆襲 — but it is unaware that Xiè lù 薤露 (the funeral ballad) precedes it. The Yù piān 玉篇 (the rhyme-tone-glossary tradition) is said to have begun with Fán jiāng 凡將 — but the Cāngjié is yet older. Cuī Yīn’s 崔駰 Dá zhǐ 達旨 is just of the kind of Yáng Xióng’s 揚雄 Jiě cháo 解嘲 — yet a separate genre zhǐ 旨 is set up. Cuī Yuán’s 崔瑗 Cǎo shū shì 草書勢 is in fact a treatise on the dynamic of cursive script — yet a genre shì 勢 is forced into existence. None of this can be relied on. As for the entry xiè ēn yuē zhāng 謝恩曰章 — the Wénxīn diāolóng explains this clearly — the two characters xiè ēn are wrongly taken as a genre name. It looks like an apocryphal text.

Including Hóng Shì 洪适’s postface at the end — that too looks copied from his Pánzhōu jí anthology. Yet Wáng Déchén 王得臣, a Jiàyòu-era man (mid-eleventh century), in his Zhèn shǐ 麈史 writes: “Rèn Fǎng of Liáng compiled the names from the QínHàn down, calling them Wénzhāng yuánqǐ: from shī and fù and Lí sāo down to shì yuē 勢約, eighty-five titles in all, that may be called comprehensive. While Sīmǎ Xiàngrú’s Yù Shǔ is included, Yáng Xióng’s Jù Qín měi Xīn is not; Jiě cháo is included, Hán Fēi’s Shuō nán is not; Liú Xiàng’s Liènǚ zhuàn is taken but Chén Shòu’s Sānguó zhì píng is missed.” He also says: “Rèn places the three-character poem starting with Xiàhóu Zhàn 夏侯湛 of Jìn; the Táng Liú Cún 劉存 places it instead with Lù yú fēi zuì yán guī 鷺于飛醉言歸 [ode]. Rèn places the sòng with Wáng Bāo 王襃 of Hàn; Liú with the Duke of Zhōu’s Shí mài. Rèn places the xí (proclamation of war) with Chén Lín 陳琳’s Xí Cáo Cāo; Liú with Zhāng Yí 張儀’s Xí Chǔ. Rèn places the bēi with Emperor Huì of Hàn’s Sìhào bēi; Liú quotes Guǎnzǐ — that Wúhuái-shì封 sealed Tàishān and inscribed the stone — as the first bēi. Rèn places the míng with the First Qin Emperor’s Huìjī ascent; Liú with Cài Yōng’s 蔡邕 commentary on Yellow Emperor’s jīnjǐ inscription.” All of these match the present text. So the text was already in circulation in the Northern Sòng — presumably the Zhāng Jī supplement, which later readers wrongly thought was Rèn’s original.

The Míng Chén Màorén 陳懋仁 once made annotations to it; in our dynasty Fāng Xióng 方熊 has added supplements. In the present recension entries headed zhù 註 are Chén’s; entries headed bǔ zhù 補註 are Fāng’s. Their notes wander into prose criticism, drawing widely from Zhì Yú, Lǐ Chōng, Liú Xié, and the Míng Wáng Shìzhēn’s 王世貞 Yìyuàn zhī yán — but lack focus. Their textual examination is patchy, and many cases needing correction are left untouched. Their opinions are sometimes weird. They claim that Méi Shèng’s 枚乘 Qī fā derives from the seven piān of Mèngzǐ and Zhuāngzǐ — pure forcing. They claim that village covenants (xiāng yuē) ought to be modeled on Wáng Bāo’s Tóng yuē 僮約 if they are to preserve antique flavor — not realizing that the Tóng yuē is a comic-satirical piece (preserved in full in the Tàipíng yùlǎn); how could it possibly be a model for a village covenant? Especially absurd. Since the original recension carries these notes, however, they are appended here as preserved.

Abstract

The Wénzhāng yuánqǐ is a small but historically pivotal book — the first systematic Chinese effort to attribute genres of writing to historical points of origin. As the Sìkù editors lay out in detail, its transmission is doubly mediated: the Suíshū Jīngjízhì shows that Rèn’s original was already lost in the Suí (yǒu lù wú shū), and the Xīn Tángshū Yìwénzhì records a one-juǎn supplement by the otherwise-unknown Táng figure Zhāng Jī 張績. The Sòng Zhèn shǐ of Wáng Déchén 王得臣 quotes the work at length in terms matching the surviving text — proving the Zhāng Jī recension was circulating already in the Northern Sòng (Jiàyòu era, c. 1056–1063). The transmitted text is therefore not Rèn’s but the Táng supplement, with all the methodological caution that implies; the Sìkù editors classify the work as “yí wéi yī tuō” (suspected of having been faked).

The book lists 85 named genres in genealogical-attribution form: sāo, shī, sānyán shī, sìyán shī, fù, sòng, zàn, biǎo, ràng biǎo, zhuàng, qǐ, fēngshàn shū, lùn, biāo, zǐ, jiě cháo, dá zhǐ, jiě nán, bēi, míng, miáo, lèi, àicí, zhì, fǎ, lìng, mìng, fúshǔ, zhào, cè, jié, jiè, biǎo, zòu, jì, xí, yí, jié, qǐfēng, shuō, shū, zǒubǐ, qièyīn, lùnwén, wǎngē, sìyán shī, etc. (Some of these the Sìkù editors specifically attack as bad classifications — e.g. counting biǎo and ràng biǎo separately, or counting xiè ēn “thanking the imperial grace” as a genre when in fact it is just a function-tag for the zhāng.) Despite these problems the book had real influence: it is the prototype for the more systematic genre catalogues of Liú Xié 劉勰’s Wénxīn diāolóng KR4i0001 and ultimately for the genre-table of Yáo Nài’s 姚鼐 Gǔ wén cí lèi zuǎn of 1779.

The two layers of commentary attached to the Sìkù version are Chén Màorén’s 陳懋仁 Míng zhù — discursive, citing Zhì Yú, Lǐ Chōng, Liú Xié, Wáng Shìzhēn — and Fāng Xióng’s 方熊 Qing bǔ zhù. The Sìkù editors find both layers weak as commentary; the value of the Sìkù recension is mainly that it preserves the work and its two-layered annotation in one fascicle.

Translations and research

• No complete English translation located. The work is mentioned in Owen, Readings in Chinese Literary Thought (Harvard, 1992), 39 n.6, and discussed briefly in James Liu, Chinese Theories of Literature (Chicago, 1975).
• John Marney, Liang Chien-wen Ti (Twayne, 1976), and Marney, “P’i Jih-hsiu and Lu Kuei-meng: Two T’ang Poets” — discussions of late Liáng / Táng literary categorization that take the Wén-zhāng yuán-qǐ as proximate background.
• Lǐ Yán-shòu 李延壽 Liáng-shū and Nán shǐ biographies of Rèn Fǎng — the primary biographical witnesses for the author.
• Lǐ Wěi-fēn 李伟芬, Rèn Fǎng yán jiū 任昉研究 (Shǎnxī shī-dà chū-bǎn-shè, 2010) — modern monograph treating Rèn’s life and works including the Wén-zhāng yuán-qǐ.
• Wáng Yùn-xī 王運熙 and Yáng Míng 楊明, Wèi-Jìn Nán-běi-cháo wén-xué pī-píng shǐ 魏晉南北朝文學批評史 (Shànghǎi gǔjí, 1989) — extended treatment of the work and its place in early Chinese literary theory.

Other points of interest

The work’s transmission profile — original lost in the Suí, Táng supplement attributed to an otherwise-unknown Zhāng Jī, that supplement passed through the Sòng under the original title — is a paradigmatic case of the kind of attribution-instability that the Sìkù editors handled with their characteristic combination of cataloguing accuracy and conservative retention. The book is preserved, but with a clear warning label.

Links

• Wilkinson, Chinese History: A New Manual §30.5.
• Kyoto Zinbun Sìkù tíyào
• Wikipedia 文章緣起