聞分館君 Ask BunKanKun

山海經

14 min read

Text ID
KR3l0090
Date
-400–100
Dynasty
戰國至漢

Shānhǎijīng 山海經

Classic of Mountains and Seas with commentary by 郭璞 (注)

About the work

The Shānhǎijīng 山海經 (Classic of Mountains and Seas) is the great early-imperial Chinese compendium of mythical geography, paradoxography, and proto-ethnography — a mytho-geographic anthology that, in Anne Birrell’s formulation, “constitutes a fabulous bestiary, a botanical thesaurus, a dictionary of natural science, a catalog of geological substances and rare jewels, a guide to portents and omens, a register of medical ailments, an apothecary’s handbook, and a medley of folkloric and ethnological material.” The transmitted text is in 18 juàn, traditionally divided into two principal portions: the Wǔzàng shānjīng 五藏山經 (Classic of the mountains of the five treasuries, juàn 1–5), describing in a near-formulaic itinerary the mountains of the four quarters and the centre with their flora, fauna, minerals, divinities, and sacrifices; and the Hǎijīng 海經 group (juàn 6–18) — the Hǎiwài 海外 (Beyond-the-seas) and Hǎinèi 海內 (Within-the-seas) classics, the Dàhuāng jīng 大荒經 (Classic of the great wilderness), and the closing Hǎinèi jīng — describing the peripheral lands, fabulous peoples, deities, and mythic events.

The standard received text is that established with the Eastern-Jìn commentary of 郭璞 Guō Pú 郭璞 (276–324, zì Jǐngchún 景純; see 郭璞), composed in the early 4th century and the editio princeps of the work’s exegetical tradition. Guō Pú was the leading philologist of his age — commentator on the Ěryǎ 爾雅, the Mù tiānzǐ zhuàn 穆天子傳, the Fāngyán 方言, the Chǔcí 楚辭, and the Cāng Jié piān 蒼頡篇 — and his Shānhǎijīng zhù 山海經注 is at once a glossarial gazetteer, a recovery effort for an already half-dead text (“its mountain and river names are in many places corrupted and disagree with what is current; teacher’s instruction has not been transmitted, and the work is on the verge of being lost”), and a vigorous philosophical defence of paradoxography against the rationalist scepticism associated with Sīmǎ Qiān 司馬遷 and Qiáo Zhōu 譙周. Guō also composed a now-lost Shānhǎijīng túzàn 山海經圖讚 (Encomia on the illustrations of the Shānhǎijīng) in 2 juàn, of which the zàn (encomia) survive in his collected works while the illustrations were already lost by Sòng times.

Tiyao

Your servants report: Shānhǎijīng in 18 juàn, with the Jìn Guō Pú commentary, has at its head the memorial of Liú Xiù 劉秀 (= Liú Xīn 劉歆) on submitting the collated text, claiming the work was composed by Bóyì 伯益. We observe: the name Shānhǎijīng first appears in the Shǐjì’s Dàwǎn zhuàn 大宛傳, where Sīmǎ Qiān merely says, “as for the monsters spoken of in the Yǔ běnjì 禹本紀 and the Shānhǎijīng, I do not dare to speak of them” — and does not state who composed it. The Lièzǐ 列子 says, “the Great Yǔ travelled and saw them; Bóyì recognised them and named them; Yíjiān 夷堅 heard and recorded them” — apparently pointing to this book, but without naming it. Wáng Chōng’s 王充 Lùnhéng 論衡, in the chapter Biétōng 別通, says, “Yǔ was in charge of moving the waters, Yì was in charge of recording strange things; the mountains beyond the seas and the lands beyond the borders, there was none they did not reach; from what they saw and heard, they made the Shānhǎijīng.” Zhào Yù’s 趙煜 WúYuè chūnqiū 吳越春秋 says the same. Only the Suíshū Jīngjízhì 隋書經籍志 says, “when Xiāo Hé 蕭何 acquired the maps and books of Qín, he afterwards also acquired the Shānhǎijīng; tradition is that the Xià Yǔ recorded it” — its account differs slightly, but all seem to derive from the Lièzǐ’s words, expanded out. Observing that the book records placenames such as the Xià sovereign Qǐ 啟, Zhōu King Wén 周文王, and the Qín–Hàn Chángshā 長沙, Xiàngjùn 象郡, Yújì 餘暨, Xiàsuí 下嶲, it cannot have been written before the Three Dynasties — most likely composed by men of the Zhōu–Qín period, with later lovers of strange things adding to it. Observing that the Chǔcí Tiānwèn 天問 agrees with it on many points, had there been no such matter in antiquity, how could Qū Yuán have fabricated it? Zhū Xī’s Chǔcí biànzhèng 楚辭辨證 says, on the contrary, that the Shānhǎijīng was made because of the Tiānwèn — this seems not to be so. As to Wáng Yìnglín 王應麟, in his Wánghuì bǔzhuàn 王會補傳, who cites Zhū Xī’s words that “the Shānhǎijīng records the various strange things and birds-and-beasts, often saying ‘facing east’ or ‘east-headed’, as if it had originally been described on the basis of paintings — antiquity had this kind of study, and the Jiǔgē and Tiānwèn are of this same class” — this gets at the truth.

Guō Pú’s commentary on this book is recorded in his Jìnshū biography. The Suí and Táng bibliographic treatises both give 23 juàn; the present text is 5 juàn short — we suspect that later editors combined some juàn to match the count of 18 piān given in Liú Xiù’s memorial, and that the missing material is not actually lost. The Suí and Táng treatises also list Guō Pú’s Shānhǎijīng túzàn 山海經圖讚 in 2 juàn; the zàn still survive in Pú’s collected works, but the illustrations were already not recorded in the Sòngshǐ treatise — so the illustrations have long been lost. The old text contains Liú Xiù’s memorial saying the book is 18 piān; this does not agree with the Hànshū Yìwèn zhì 漢書藝文志 figure of 13 piān. The Qīlüè 七略 was made by Liú Xiù himself, so he should not contradict his own count — we suspect a forgery. Yet because Guō Pú’s preface already cites the memorial and the tradition is long-standing, we have nonetheless included it.

The book’s accounts of mountains and waterways are much mixed with the spiritual and the monstrous, so the Dàozàng incorporated it into the Tàixuán section under the Jìng call-number. But examining its true purport, it is in truth not the doctrine of Huáng–Lǎo. Yet its principles of geography and mountains-and-rivers are mostly jīkǎo (inferred speculation) — checked against what eyes and ears can reach, scarcely one in a hundred is true. The various commentators all consider it the chief of geographical works — this too is not exact. Determining its name in truth, it is the most ancient of works in the shuōbù (novelistic) genre. Respectfully checked, Qiánlóng 46 (1781), 1st month. Chief Compilers: Jì Yún 紀昀, Lù Xīxióng 陸錫熊, Sūn Shìyì 孫士毅. Chief Collator: Lù Fèichí 陸費墀.

Abstract

The Shānhǎijīng’s composition history is layered and disputed. Modern scholarly consensus, building on Bì Yuán’s 畢沅 Qīng analysis and consolidated by Yuán Kē 袁珂 (1916–2001) in his 1980 critical edition, distinguishes at least three textual strata: (1) the Wǔzàng shānjīng (the “Shānjīng” proper, juàn 1–5), generally dated to the late Warring States, with a probable composition window of c. 4th–3rd century BCE; (2) the Hǎiwài and Hǎinèi classics (juàn 6–13), dated to the late Warring States and early Hàn; and (3) the Dàhuāng jīng and closing Hǎinèi jīng (juàn 14–18), traditionally regarded as the latest stratum, dated to the early to mid Western Hàn, with some material possibly as late as Eastern Hàn. Yuán Kē assigned the Dàhuāng section to the Hàn explicitly. The placenames Chángshā 長沙, Xiàngjùn 象郡 (a Qín commandery), and the Yújì / Xiàsuí counties — flagged by the Sìkù compilers as terminus-post-quem markers — secure the received recension’s lower bound in the Qín–Hàn period.

Riccardo Fracasso (1996) and Marc-William Henri Strange more recently have argued for an even later compositional centre of gravity, with the Hǎijīng portions reflecting Western-Hàn cosmography and the entire received text effectively a Western-Hàn compilation projected back into pre-Qín mythic deep time. The terminus ad quem is the memorial of Liú Xīn 劉歆 (劉秀; fl. c. 6 BCE – 23 CE), which announces the imperial collation of the text in 18 piān and ascribes its authorship to Bóyì 伯益, Yǔ’s minister. Modern scholarship does not accept the Bóyì ascription, nor the alternative YǔYì tradition reported in the Lùnhéng and the WúYuè chūnqiū — these are projections back onto the mythic past from the early-imperial reception, not historical authorship claims.

For the date bracket of the work itself we follow the modern consensus: notBefore c. -400 (the upper bound of the Shānjīng core) and notAfter c. 100 (an outer Hàn-period bound for the received recension’s assembly, accommodating both Liú Xīn’s collation and the possibility, argued by Fracasso, of post-Liú-Xīn material). The Bóyì attribution is pseudepigraphic; the work is best understood as a layered Warring-States-to-Hàn anthology of mythological-geographical lore.

The Hànshū Yìwèn zhì 漢書藝文志 lists the Shānhǎijīng under the Xíngfǎ 形法 (physiognomic) category in 13 piān. The discrepancy with Liú Xīn’s 18-piān memorial — flagged by the Sìkù compilers as suspicious and inviting the charge of forgery — is now generally explained as a divergence between the count of the Qīlüè compilation tradition and the later transmitted division. The Suíshū Jīngjízhì and the Jiù Tángshū and Xīn Tángshū bibliographic treatises record Guō Pú’s commentary in 23 juàn; the received text in 18 juàn is a later editorial reconfiguration to match the Liú Xīn memorial’s piān count.

Guō Pú’s commentary, written in the Tàixīng 太興 — Tàiníng 太寧 reigns (c. 318–324) shortly before his execution in 324 at the close of the Wáng Dūn 王敦 rebellion, is the foundation-stone of the exegetical tradition: glossing rare graphs, identifying flora and fauna where possible, cross-referencing the Ěryǎ, the Mù tiānzǐ zhuàn, the Chǔcí, and emerging local geographical lore. Guō’s preface is a programmatic defence of paradoxography: “what the world calls strange — we do not yet know why it is strange; what the world calls not-strange — we do not yet know why it is not strange. Things are not strange of themselves; they wait upon us, and then they are strange — so the strangeness lies in us, not in the things.” This strangeness-is-relative argument grounds the work’s epistemology and was canonical for later paradoxographic writing (the Bówùzhì 博物志, the Sōushén jì 搜神記).

The commentary tradition after Guō Pú develops in waves: Sòng-period notes are preserved in the Yǒnglè dàdiǎn fragments and the various Sòng quotations; Míng sees the woodblock-printed illustrated editions of 1597 (the basis of Strassberg’s 2002 reproduction); Qīng sees the major critical edition of Bì Yuán 畢沅 (1730–1797), Shānhǎijīng xīnjiàoběn 山海經新校本, which systematically collates Guō Pú’s commentary against the Sòng and Míng witnesses and supplies the Qīng-philological apparatus that became the basis of all subsequent editions. Hǎo Yìxíng 郝懿行 (1755–1825) produced the Shānhǎijīng jiānshū 山海經箋疏, a fuller annotated edition that remains highly cited. The modern critical standard is Yuán Kē 袁珂, Shānhǎijīng jiàozhù 山海經校註 (Shanghai Gǔjí, 1980; rev. edn. 1981), which incorporates the Qīng tradition and adds modern philological and mythographic apparatus; it is the base text of the ICS electronic concordance (no. 22) and of the Birrell, Strassberg, and Fracasso translations.

Translations and research

  • Mathieu, Rémi, tr. 1983. Étude sur la mythologie et l’ethnologie de la Chine ancienne. 2 vols. Paris: De Boccard. The first complete European-language translation, with extensive mythological-ethnographic apparatus; vol. 1 is the translation, vol. 2 a substantial index. Remains the most thoroughly annotated Western rendering.
  • Birrell, Anne, tr. 1999. The Classic of Mountains and Seas. London: Penguin. Complete English translation aimed at a general readership; based on the Yuán Kē 1980 edition. The standard Anglophone reference.
  • Fracasso, Riccardo, tr. 1996. Libro dei monti e dei mari (Shanhai jing): Cosmografia e mitologia nella China Antica. Venice: Marsilio. Complete Italian translation with a substantial scholarly introduction situating the work in early-Chinese cosmography and arguing for a late (Western-Hàn) compilation date.
  • Strassberg, Richard E., ed. and tr. 2002. A Chinese Bestiary: Strange Creatures from the Guideways through Mountains and Seas. Berkeley: University of California Press. Partial English translation organised around the bestiary, reproducing the 76 earliest surviving illustrations (1597 Hǔ Wén-huàn 胡文煥 woodblock edition). The standard reference for the iconographic tradition.
  • Schiffeler, John William. 1977. The Legendary Creatures of the Shan Hai Ching. Taipei: Hwa Kang. An early partial English translation focused on the fabulous fauna; superseded by Birrell and Strassberg but still cited.
  • Fracasso, Riccardo. 1998. “The Illustrations of the Shan hai jing 1: From Yu’s Tripods to Qing Blockprints.” Cina 21: 93–104.
  • Dorofeeva-Lichtmann, Vera. 2003. “Mapping a Spiritual Landscape: Representation of Terrestrial Space in the Shanhai jing.” In Di Cosmo and Wyatt, eds., Political Frontiers, Ethnic Boundaries, and Human Geographies in Chinese History, 35–79.
  • Dorofeeva-Lichtmann, Vera. 2007. “Mapless Mapping: Did the Maps of the Shanhai jing Ever Exist?” In Bray et al., eds., Graphics and Text in the Production of Technical Knowledge in China, 217–94.
  • Chen Lianshan 陈连山. 2012. Shānhǎijīng xué-shù shǐ kǎo-lùn 《山海经》学术史考论 (On the history of the study of the Shānhǎijīng). Beijing: Beijing Daxue. The standard Chinese-language history of the exegetical tradition.
  • Yuán Kē 袁珂. 1980. Shānhǎijīng jiào-zhù 山海經校註. Shanghai: Shanghai Gǔjí. The modern critical edition; rev. edn. 1981. The base text for all subsequent translations and concordances.
  • Ye Shuxian 叶舒宪, Xiao Bing 萧兵, Zheng Zaishu 郑在书. 2004. Shānhǎijīng de wénhuà xún-zōng 山海经的文化寻踪. 2 vols. Wuhan: Hubei Renmin. A cultural-mythographic study.
  • Lewis, Mark Edward. 2006. The Flood Myths of Early China. Albany: SUNY Press. Treats Shānhǎijīng flood material.

Other points of interest

The Sìkù compilers’ classification of the Shānhǎijīng as the “most ancient of works in the shuōbù (novelistic) genre” — that is, refusing the traditional designation of dìlǐ shū zhī guàn 地理書之冠 (chief of geographical works) — is a notable Qīng-philological intervention. Earlier bibliographies (the Hànshū Yìwèn zhì’s xíngfǎ category, the Suíshū Jīngjízhì’s dìlǐ placement, the Sòngshǐ Yìwènzhì’s dìlǐ placement) had treated the work as primarily geographical; the Sìkù reclassification, on the grounds that “checked against what eyes and ears can reach, scarcely one in a hundred is true”, marks the Qīng-evidential-scholarship recognition that the work is a mythographic anthology, not a topographical survey. This is the placement honoured by the Sìkù’s grouping it under 子部小說家類·異聞之屬 (Masters Section, Xiǎoshuōjiā category, Yìwén subdivision), and which the Kanripo cataloging follows by placing it in KR3l (小說家類).

The work’s reception in early-modern Europe began with the Jesuit reports of the 18th century and reached scholarly maturity with Léon de Rosny’s partial French translation in the 1880s and Mathieu’s complete 1983 rendering. The illustrated tradition — Guō Pú had referred to illustrations now lost, and Sòng Wáng Yìnglín 王應麟 had inferred (citing Zhū Xī) that the original was an illustrated work — is preserved only from the Míng 1597 Hǔ Wénhuàn 胡文煥 edition onward, and forms the basis of Strassberg’s 2002 illustrated translation.

The work is one of the principal sources for the figure of Xīwángmǔ 西王母 (Queen Mother of the West), for the Kūnlún 崑崙 mountain cosmography, for the fúsāng 扶桑 tree, for the ten-suns myth, for the jīngwèi 精衛 bird that fills the sea with stones, for the xíngtiān 刑天 decapitated warrior, and for countless other elements of the classical Chinese mythological repertoire. Without the Shānhǎijīng, the Chǔcí Tiānwèn would be in many places unintelligible — a fact noted by Guō Pú in his preface and elaborated by Wáng Yìnglín and Zhū Xī.

Graph View

Backlinks