書傳會選

Shū zhuàn huì xuǎn 書傳會選

Selected Compilation of Documents Commentary by 劉三吾 (fèng chì zhuàn 奉敕撰)

About the work

The early Míng imperially-commissioned Shū jízhuàn 書集傳 (KR1b0017) corrective sub-commentary in 6 juǎn, compiled by a team of Hànlín scholars under the directorship of Liú Sānwú 劉三吾 (1312–1399) and submitted in Hóngwǔ 27 / 1394. The work was prompted by the Hóngwǔ emperor’s own astronomical study, which identified specific errors in Cài Shěn’s commentary; sixty-six corrections were ultimately documented. The compilers were ordered to follow the precedent of the Hàn-period Shíqú 石渠 and Báihǔ 白虎 conferences — i.e., to operate as a court-convened classical-scholarship colloquium. The whole enterprise — from imperial decree (Hóngwǔ 27.IV) to completed manuscript (Hóngwǔ 27.IX) — took just five months, although Liú Sānwú’s preface notes that he had been working privately on the questions for the previous seventeen years.

The work’s broader political-curricular significance is large. The Yuán Yánȳòu curriculum (1313+) had elevated the Cài zhuàn into an unrivaled monopoly, suppressing the Sòng-end and early-Yuán correction-of-Cài tradition — Zhāng Bǎoshū 張葆舒’s Shàngshū Càizhuàn dìng wù 尚書蔡傳訂誤, Huáng Jǐngchāng 黃景昌’s Shàngshū Càishìzhuàn zhèng wù 尚書蔡氏傳正誤, Chéng Zhífāng 程直方’s Càizhuàn biàn yí 蔡傳辨疑, Yú Bāoshū 余苞舒’s Dú Càizhuàn yí 讀蔡傳疑 — all of which had perished with the institutional consolidation of the Cài zhuàn. Even Chén Lì’s 陳櫟 own earlier Shū shuō zhézhōng 書說折衷 had to be self-suppressed (see KR1b0027). The Hóngwǔ-imperial Huì xuǎn re-opened the discussion: not by replacing the Cài zhuàn but by institutionally documenting where it must be corrected, with imperial authority. The Yǒnglè-era Wǔ jīng dàquán 五經大全 (1415) subsequently re-closed the discussion by re-canonizing the Cài zhuàn alone — but the Huì xuǎn preserved the official record of where Cài was wrong.

Tiyao

Imperially Authorized Sìkù Quánshū. [Classics, division 2.] Shū zhuàn huì xuǎn. [Books-class.]

Précis. Your servants etc. respectfully submit: the Shū zhuàn huì xuǎn in six juǎn is the work of the Míng Hànlínxuéshì Liú Sānwú and others, by imperial commission. On investigation: although the Cài Shěn Shū zhuàn had its source in Master Zhū, on many points he draws from his own opinion. Already at the time of its first circulation there were many divergent voices; in late Sòng and early Yuán, Zhāng Bǎoshū wrote a Shàngshū Càizhuàn dìng wù; Huáng Jǐngchāng wrote a Shàngshū Càishìzhuàn zhèng wù; Chéng Zhífāng wrote a Càizhuàn biàn yí; Yú Bāoshū wrote a Dú Càizhuàn yí — successive challenges. By Yuán Rénzōng’s Yánȳòu 2 (1315), when the discussion to revive the tribute-examination set the Shàngshū yì on the basis of the Cài, the writings of Bǎoshū and the others were lost altogether and not transmitted.

Chén Lì at first wrote a Shū zhuàn zhézhōng, in which he rather extensively discussed Mr Cài’s errors; once the legal regulations were settled, he reworked the project as the Zuǎnshū, developing the meaning of Cài, and the Zhézhōng likewise was lost. His autograph preface saying “in the imperial age, with the examination system established in motion, the Shū’s veneration of the Cài zhuàn is therefore as it should be” — clearly there was a [political] purpose behind the formulation. Until Míng Tàizǔ began to investigate the heavenly phenomena and recognized that they were not consistent with the Cài zhuàn, he made wide call to those of accomplished scholarship and settled this present compilation. Where the Cài zhuàn is convergent he keeps it, not setting up his own opinion in advance to wantonly heap on slander; where it is divergent he corrects it, but he does not rigidly hold to factional positions in clever apologetic protection. The corrected items in total are sixty-six.

[The tíyào continues with detailed substantive observations: Zhù Yǔnmíng’s 祝允明 Qián wén 前聞 records that the imperial-circulated declaration cited only two: the Yáo diǎn “sun-and-moon revolve leftward” and the Hóng fàn “xiāng xié jué jū” 相協厥居 — these are general examples. Gù Yánwǔ’s 顧炎武 Rì zhī lù 日知錄 says: “this book holds that the heavens revolve to the left, and the sun-moon-and-five-planets contrary to the heavens revolve to the right (following Chén Xiángdào 陳祥道); the Gāozōng róng rì is held to refer to Zǔ Gēng 祖庚’s offering at the Gāozōng temple (following Jīn Lǚxiáng); the Xī Bó kān Lí is held to refer to Wǔ Wáng (also following Jīn); the ‘Zhōu Gōng vastly preserved WénWǔ’s mandate for seven years’ is held to mean the seventh year of Zhōu Gōng’s regency for Chéng Wáng (following Mr Zhāng [Zǎi] and Mr Chén [Lì]) — all of these are reliable readings. Likewise: Yǔ gòng ‘the tribute fields are the sound ones’ (jué fù zhēn 厥賦貞) follows Sū Shì’s reading that ‘tribute and field are exactly mutually corresponding’; ‘Jīng belongs to Wèiruì’ follows the Kǒng zhuàn that ‘water-north is ruì’; Tài jiǎ ‘from zhōu there is the end’ follows Jīn’s reading that ‘zhōu’ should be read ‘jūn’; Duō fāng ‘[he] could not bring forth the people’s bondage’ follows Mr Yè 葉; only on Jīn téng’s ‘Zhōu Gōng dwells in the East’ it refutes Kǒng [in arguing] that this is not the eastern campaign — but in Luò gào it again adopts the eastern-campaign reading: thus self-contradictory.“]

[The tíyào further notes:] At the end of each commentary entry [Liú Sānwú attaches] a jīng zhuàn yīn shì 經傳音釋 (canonical-text and commentary phonological-and-graphical gloss), distinguishing pronunciation, character-form, and character-meaning very thoroughly. Where the commentary draws on ancient persons’ surnames or ancient book titles, it always provides the full reference; alongside, also, it verifies historical events. The SòngYuán Confucian framework is still in place, and the writers of this book — having all from childhood pursued wù běn 務本 (“attending to the root”) learning, and not having risen by means of the bā gǔ — therefore the books they composed, although not equal to the earlier Confucians, have still served the later students well.

[Conclusion of the tíyào:] As Yán [Ruòqú] in the Shàngshū gǔ wén shū zhèng, on account of the entry on Shì shuǐ 澨水 reaching the boundary of Fùzhōu Jìnglíng [where the character zhě 者 had been miscopied as lái 來], unleashed venomous abuse — this is not a sound judgment.

On investigation: the Míng Tàizǔ shílù records the imperial discussion of the Cài zhuàn’s errors as taking place in Hóngwǔ 10 / 1377, third month; the imperial decree to compile this book is in Hóngwǔ 27 / 1394, fourth month, bǐngxū day; and the work was completed in the ninth month, jǐyǒu day — only five months in all. Looking at Liú Sānwú’s xù, which says: “Your servant Sānwú serves on the staff of the Hànlín; I have repeatedly brought these arguments to His Majesty’s attention; His Majesty has assented; thereupon he ordered the Confucians of the empire, on the precedent of the Shíqú and Báihǔ, to collate and settle the work alongside us” — so over the seventeen years [from 1377 to 1394], Sānwú had already done the verification and inquiry, and arrived at his settled views; the further survey of multiple voices was in order to perfect the result.

Only: the named compilers listed in the Shílù differ from those listed at the head of this present recension. Zhū Yízūn’s Jīngyì kǎo says that Xǔ Guān, Jǐng Qīng, Lú Yuánzhì, Dài Déyí and others all died in the Jiànwén disasters and were therefore deleted [from the Shílù’s list] — that is correct as far as it goes. But Hú Jì’ān, Mén Kèxīn, Wáng Jùnhuá, and so on — eleven persons in all — why are they likewise deleted? And further, Jìn Guān, Wú Zǐgōng, Sòng Lín — three persons whom this book does not list — why are they additionally inserted? Because in the Yǒnglè era the Tàizǔ Shílù was redacted afresh, and its purpose was to slander the Hóngwǔ-loyal Huìzōng régime [i.e., the Jiànwén emperor], in order to legitimate the Yǒnglè usurpation as not-by-choice. The remainder was carelessly handled, with errors flowing in profusion, not deserving of authority. This [present] book is the original recension of its time; the names listed in it should be taken as fixed. Respectfully submitted, Qiánlóng 46 / 1781, fifth month.

— Director-General, Jì Yún 紀昀, Lù Xīxióng 陸錫熊, Sūn Shìyì 孫士毅. — Director of Final Collation, Lù Fèichí 陸費墀.

Abstract

The Shū zhuàn huì xuǎn is the most consequential institutional-imperial intervention in Shàngshū exegesis between the Yuán Yánȳòu canonization of Cài Shěn (1313) and the Qīng Kāngxī Shū jīng zhuàn shuō huì zuǎn (early 18th c.). Compiled in Hóngwǔ 27 / 1394 in five intensive months by a team led by Liú Sānwú 劉三吾 — and prompted by the Hóngwǔ emperor’s own astronomical investigations — the work documents 66 specific errors in Cài Shěn’s Shū jízhuàn 書集傳 (KR1b0017), preserves the alternative readings drawn from the major SòngYuán Cài-correction tradition (Sū Shì 蘇軾, Jīn Lǚxiáng 金履祥, Chén Lì 陳櫟, Zhāng Zǎi 張載, Yèshì 葉氏, Chén Xiángdào 陳祥道), and re-opens the institutional door to Càizhuàn skepticism that the Yánȳòu reform had nailed shut.

The political-curricular framing is indispensable. The Hóngwǔ emperor’s preface and Liú Sānwú’s xù are explicit that the work is not meant to displace the Cài zhuàn (which remained the curricular text); rather it is meant to correct it where empirical-evidentiary investigation (the emperor’s astronomy specifically) had shown it to be wrong. The 66 corrections include several that the Sìkù compilers list explicitly: the rotational-direction question of Yáo diǎn; the Hóng fàn “xiāng xié jué jū” 相協厥居 reading; the Yǔ gòng “jué fù zhēn” 厥賦貞 reading following Sū Shì; the Tài Jiǎ “zì zhōu” / “zì jūn” graphical correction following Jīn Lǚxiáng; etc. Each correction is documented with its source-attribution, restoring credit to the SòngYuán Cài-correction tradition that had been suppressed.

Gù Yánwǔ’s 顧炎武 Rì zhī lù 日知錄 — quoted at length by the Sìkù compilers — gives the most authoritative early-Qīng endorsement of the work, framing it as the product of pre-bā gǔ (pre-eight-legged-essay) classical scholarship and “still useful to the later students.” Yán Ruòqú’s 閻若璩 venomous attack on the work, occasioned by a single mis-copied character (zhě 者 / lái 來) in a footnote on the Shì shuǐ 澨水 entry of Yǔ gòng, is dismissed by the Sìkù compilers as unfair (fēi dǔ lùn yě 非篤論也).

The compilers’ redaction-list controversy is itself an unusual Sìkù contribution. The list of compilers in the present Huì xuǎn recension differs from the list in the Tàizǔ Shílù (as preserved in the Yǒnglè-era redaction). Eleven names in the Huì xuǎn are missing from the Shílù (Hú Jì’ān 胡季安, Mén Kèxīn 門克新, Wáng Jùnhuá 王俊華, etc.); three names appear in the Shílù but not in the Huì xuǎn (Jìn Guān 靳觀, Wú Zǐgōng 吳子恭, Sòng Lín 宋麟). Zhū Yízūn already noted that some of those deleted from the Shílù (Xǔ Guān 許觀, Jǐng Qīng 景清, Lú Yuánzhì 盧原質, Dài Déyí 戴德彝) had died as Jiànwén-loyalists during the Yǒnglè 靖難 (Jīngnán) coup of 1402, and so the Yǒnglè-redactor Yǒnglè court silently removed them. The Sìkù compilers go further: they identify the Yǒnglè-era Shílù redaction as a deliberate political falsification (“its purpose was to slander the Huìzōng régime, in order to legitimate the Yǒnglè usurpation as not-by-choice”) and conclude that the Huì xuǎn’s own list, as the original 1394 recension, is the historically authoritative one.

The composition window is fixed: 1394 (Hóngwǔ 27, IV–IX). The Sìkù submission was Qiánlóng 46 / 1781.

Translations and research

No substantial Western-language translation of the Shū zhuàn huì xuǎn is known. For the institutional-curricular context see Benjamin A. Elman, A Cultural History of Civil Examinations in Late Imperial China (Berkeley: University of California Press, 2000); for the Hóngwǔ emperor’s intellectual interventions in the canonical curriculum see Edward L. Farmer, Zhu Yuanzhang and Early Ming Legislation (Leiden: Brill, 1995). For Liú Sānwú specifically and the Hóngwǔ-period imperial commissions see John D. Langlois, Jr., “The Hung-wu Reign, 1368–1398,” in The Cambridge History of China vol. 7 (Cambridge University Press, 1988).

Other points of interest

The work’s documentation of the Yǒnglè-era Shílù falsification is one of the cleaner cases preserved in the Sìkù tíyào of an editorial intervention against received imperial-historiographical orthodoxy. The compilers’ willingness to call out the Yǒnglè Shílù by name as politically dishonest (qí yú cǎo cǎo, fēi suǒ zhù yì 其餘草草非所注意) shows the Qiánlóng-era Sìkù project’s underlying philological-skeptical posture toward the early-Míng official record.

The 66-correction list itself remains a useful working scholarly resource: each correction names the alternative reading and the SòngYuán precedent, giving a comprehensive Hóngwǔ-era survey of the Cài-correction tradition before its post-1415 institutional suppression.

The five-month compilation timeline contrasts sharply with the multi-year imperial-commission norms of the Yǒnglè Wǔ jīng dàquán (1415, ~6 years) and the Kāngxī Huì zuǎn (early 18th c., multi-year). The Hóngwǔ emperor’s personal urgency drove the schedule, and Liú Sānwú’s prior seventeen years of private investigation provided the scholarly substrate.

Links

• CBDB id 28545 (劉三吾)
• Wikidata: https://www.wikidata.org/wiki/Q11129437 (劉三吾)
• Sìkù quánshū zǒngmù tíyào, Shū lèi, Shū zhuàn huì xuǎn entry (Kyoto Zinbun digital edition)