“New” Evidence on the Zhou Conquest - Dartmouth …earlychina/docs/earlychinajournal/ec6...dered the white banner from which the head of Shang king Zhou was suspended and the red

57

“NEW” EVIDENCE ON THE ZHOU CONQUEST

Edward L. Shaughnessy Department of Asian Languages

Stanford University Stanford, CA 94305

The inscription on the Li gui 利𣪘, a ves-sel unearthed in 1976 in Qishan 岐山 county of Shaanxi province, by substantiating certain key as-sertions of traditional Chinese historiography, has once again focused attention on King Wu's conquest of the Shang and his consequent establishment of the Zhou dynasty. It is the expectation, and ever increasing reality, of discovering ancient records, usually in the form of bone or bronze inscriptions, that makes the study of early China so vibrant and open to new interpretations.

Unfortunately, however, in the excitement over these new inscriptional discoveries, some scholars have rejected, or at least neglected, other writ-ten records of the period that have come to us by way of the more circuitous route of historical transmission. This is not exclusively due to the prejudices of the modern "scientific" historian. In the case of at least one document, the "Shifu"

世俘 chapter of the Yizhoushu 逸周書, the

fullest account of the Zhou conquest was rejected outright as early as 2,300 years ago by one of the first known historians of ancient China——Mencius

孟子 (-371–289). (The text was known to Mencius as the "Wucheng" 武成; see below, Pt. II, Sec. 1.) And with the prestige accruing to his thought from other considerations, Mencius's rejection of the account of the conquest given by the "Shifu" and the idealized history he proposed in its place have continued as more or less orthodox until the present.

But the modern historian has a responsibil-ity to consider objectively all of the evidence at hand. The "Shifu" is, I believe, worthy of such careful consideration. Therefore, I offer here a translation of the text and a discussion of its authen ticity, in the belief that ancient records are discoverable in places other than the soil of north China.

Part I: Translation and Critical Text

The Great Capture1

I It was the fourth month, day yi-wei (day 32); King Wu achieved rule over the four directions and went through the countries that Yin had commanded.2

II It was the first month, ren-chen (day 29), the (day of) expanded dying brightness. On the next day, gui-si (day 30), the king then in the morning set out from Zhou and went on campaign to attack the Shang king Zhou.3 On jia-zi (day 1), five days after (the day) "after the dying brightness" of the following second month,4 (they) arrived in the morning and defeated the Shang,5 thence entirely decapitating the Shang king Zhou and shackling (his) one hundred evil ministers.

III Grand Duke Wang was ordered to secure the area.6 On the coming day, ding-mao (day 4), Wang ar-rived and reported about ears taken and captives.

《周書》《世俘》寫定本

I 維四月乙未日,武王成辟四方, 通殷命有國.

II 維一月壬辰旁死霸, 若翌日癸己, 王乃朝步自周, 于征伐 (商王) 紂, 越若來二月既死霸越五日甲子, 朝至于商, 則咸劉商王紂, 執夫惡臣百人.

III 太公望命禦方, 來丁卯, 望至告以馘俘.

* * * * *

58

On wu-chen (day 5), the king then performed a yu- exorcism and an inspection tour, and (then) made a commemorative sacrifice to King Wen.8 (On) this day the king established (his) government.

IV Lü Ta was ordered to attack Yue and Xifang. On ren-shen (day 9), Huang Xin arrived and reported about ears-taken and captives.

Hou Lai was ordered to attack Miji and/at Chen. On xin-si (day 18), (he) arrived and reported about ears-taken and captives.

On jia-shen (day 21), Bai Ta was ordered to address the Tiger Vanguard, being ordered to attack Wei. He reported about ears-taken and captives.9

V On xin-hai (day 48), presentation of the captured caldrons of the Yin king. King Wu then reverently displayed the jade tablet and the codice, making an announcement to the heavenly ancestor, Shangdi. Without changing his robes,10 the king entered into the temple. Holding the yel-low axe,11 he spoke and regulated the many states. The flutists (played) nine refrains. The king's honored ancestors from Taiwang, Taibo, Wang Ji, Yugong, and King Wen to Yi Kao12 were arrayed and elevated, in order to report the crimes of Yin. The flutists entered. The king, holding the yellow axe, confirmed the elders of the countries.

On ren-zi (day 49), the king, wearing the royal attire and displaying the yan-tablet, entered the temple. The flutists entered. The king, holding the yellow axe, confirmed the rulers of the states.

On gui-chou (day 50),13 presentation of the 100 captured nobles of the Yin king. The flutists entered. The king was displaying the yan- tablet, holding the yellow axe, and grasping a halberd. The king entered; the bell was struck;14 the "Great Sacrifice," one refrain. The king folded his hands and touched his head to the floor. The king (lit. settled =) sat; the bell was struck;15 the "Great Sacrifice," three refrains.

On jia-yin (day 51), inspection of the mili-tary Yin at Muye.16 The king suspended red and white pendants (from his sash). The flutists played "Wu."17 The king entered and presented the myriadgifts:18 "Brightly, Brightly," three refrains.

On yi-mao (day 52), the flutists played "Venerable Yu (i.e., Yu the Great) begat (Kai =) Qi," three refrains.19 The king sat.

VI On geng-zi (day 37), Chen Ben was ordered to attack Mo, Bai Wei20 was ordered to attack Xuan-fang; and Huang Xin21 was ordered to attack Shu.22

On yi-si (day 42), Chen Ben and Huang Xin ar-rived (from) Shu and Mo23 and reported the netting of the Archer-Lord of Huo,24 the capture of the Lord of Ai,25 the Lord of Yi and minor ministers, and the netting of 803 chariots, reporting about ears taken and captives. Bai Wei arrived and reported about the netting of Xuanfang and the netting of 30 chariots, reporting about ears-taken and cap-tives. Bai Wei was ordered to attack Li: he re-ported about ears-taken and captives.

戊辰, 王遂禦循, 追祀文王, 時日, 王立政

IV 呂他命伐越戲方, 壬申, 荒新至告以馘俘, 侯來命伐靡集于陳, 辛巳, 至告以馘俘, 甲申百弇命以虎賁誓, 命伐衛, 告以馘俘.

V 辛亥薦俘殷王鼎, 武王乃翼矢琰矢憲, 告天宗上帝, 王不革服, 格于廟; 秉黃鉞, 語治庶國; 籥人九終, 王烈祖自太王, 太伯、王季、虞公、文王、邑考以列升, 維告殷罪, 籥人造; 王秉黃鉞, 正國伯.

壬子, 王服袲衣, 矢琰, 格廟, 籥人造; 王秉黃鉞, 正邦君.

癸丑, 藨俘殷王士百人, 籥人造; 王矢琰, 秉黃鉞, 執戈: 王入; 奏庸: 「大享」, 一終, 王拜手稽首, 王定; 奏庸: 「大享」, 三終.

甲寅, 謁戎殷于牧野, 王偑赤白旂, 籥人奏「武」, 王入進萬獻: 「明明」, 三終

乙卯, 籥人奏「崇禹生啟」, 三終, 王定.

VI 庚子, 陳本命伐磨, 百韋命伐宣方, 荒新命伐蜀.

乙巳, 陳本, 荒新, 蜀、磨至; 告禽霍侯, 俘艾侯、佚侯, 小臣四十有六, 禽禦几百有三十兩; 告以馘俘, 百韋至, 告以禽宣方, 禽禦三十兩, 告以馘俘, 百韋命伐厲; 告以馘俘.

59

VII King Wu hunted and netted 22 tigers, 2 panthers, 5,235 stags, 12 rhinoceri, 721 yaks, 151 bears, 118 yellow-bears, 353 boars, 18 badgers, 16 king-stags, 50 musk-deer, 30 tailed-deer, and 3,508 deer.27

VIII King Wu had pursued and campaigned in the four directions. In all, there were 99 recalcitrant countries, 177,779 ears-taken registered,28 and 310,230 captured men. In all, there were 652 coun-tries that willingly submitted.

IX It was the fourth month, six days after (the day) after the expanded growing brightness, geng-xu (day 47),29 King Wu arrived in the morning and performed a burnt-offering sacrifice in the Zhou temple, intending that "I, the small one, bring peace to the glorious ancestors."30

King Wu descended from (his) chariot and caused Scribe Yi31 to intone the document in the declaration to heaven. King Wu then shot the hun-dred evil ministers of (Shang king) Zhou.32 He beheaded and offered their jia little prince, the great master of the cauldron, and beheaded the leaders of their forty families, masters of the cauldron.33 The commander of foot-soldiers and the commander of horses first (attended) to their decla-ration of the suburban sacrifice,34 then the southern gate was flanked with the captives to be sacrificed,35 all of whom were given sashes and clothes to wear. The ears-taken were first brought in. King Wu at-tended to the sacrifice and the Great Master shoul-dered the white banner from which the head of Shang king Zhou was suspended and the red pennant with the heads of his two consorts. Then, with the first scalps, he entered and performed the burnt-offering sacrifice in the Zhou temple.36

On the next day, xin-hai (day 48), he performed a sacrifice in his position (as king), therewith mak-ing a yue-offering37 to heaven.

Five days later on yi-mao (day 52), King Wu then sacrificed in the Zhou temple the ears-taken of the many countries,38 (declaring), "Rever-ently, I, the young son, slaughter six oxen and slaughter two sheep. The many states are now at an end." (He)reported in the Zhou temple, saying, "Of old I have heard that (my) glorious ancestors emulated the standards of the men of the Shang; with the dis membered body of (Shang king) Zhou, I report to heaven and to Ji (i.e., Hou Ji)." He (i.e., sacrificed) the minor offerings, sheep, dogs and boar, to the hundred spirits, the water and the earth,39 declaring to the altar of the earth, saying, "It is I, the small one, who brings peace to the glorious ancestors; may it reach to the small one." He sacri ficed 504 oxen to Heaven and to Ji and used 2,701 sheep and boar, the minor offering, to the hundred spirits, the water and the earth.

X The Shang king Zhou was in the suburb of Shang on that jia-zi (day 1) evening. Shang king Zhou took the "Heaven's Wisdom" jade and jewels and, wrapping them thickly around his body, immolated himself. In all, four thousand (pieces of) jade were reported to have been fired. On the fifth day, King Wu then caused one thousand men to seek them. The four thousand pieces of jade40 were burnt (but) the "Heaven's Wisdom" jade and jewels41 were unburnt in the fire. King Wu then treasured

VII 武王狩: 禽虎二十有二, 貓二, 麋五千二百三十五, 犀有二, 氂七百二十有一, 熊百五十有一, 罷百一十有八, 豕三百五十有二, 貉十有八, 塵十有六, 麝五十, 麋三十, 鹿三千五百有八.

VIII 武王遂征四方: 凡憝國九十有九國. 馘磨億有七萬千七百七十有九, 俘人三億萬有二百三十, 凡服國六百五十有二.

IX 維四月既旁生霸越六日庚戌武王朝至燎于周廟, 維予沖子綏文考.

武王降自車, 乃俾史伕繇書于天號, 武王乃廢于紂夫惡臣百人, 伐右厥甲小子鼎大師, 伐厥四十夫家君鼎師, 司徒, 司馬初厥于郊號乃夾于南門; 用俘皆施佩衣衣, 先馘入, 武王在祀, 大師負商王紂縣首白旂, 妻二首赤旗, 乃以先馘入燎于周廟.

若翌日辛亥, 祀于位, 用籥于天位.

越五日乙卯, 武王乃以庶國祀馘于周廟; “翼予沖子，斷牛六，斷羊二，庶國乃竟.” 告于周廟曰: “古朕聞文考脩商人典; 以斬紂身告于天于稷.” 用小牲羊, 犬, 豕于百神, 水, 土; 誓于社曰; “維予沖子綏文考至于沖子.” 用牛于天于稷五百有四, 用小牲羊, 豕于百神, 水, 土二千七百有一。

X 商王紂于商郊, 時甲子夕, 商王紂取「天智」玉琰, 環身厚以自焚, 凡厥有庶玉, 四千告焚, 五日武王乃俾千人求之, 四千庶玉則銷, 「天智」玉琰在火中不銷, 凡「天智」玉, 武王則寶與同, 凡武王俘商舊五億有八萬.

60

and shared all of the "Heaven's Wisdom" jade (pieces). In all, King Wu captured 180,000 pieces of old, Shang jade.

Part II: The Authenticity of the Text

The significance of a text such as this would, of course, be enormous if it could be demonstrated to be a genuine product of the early Western Zhou period. In attempting to determine the authenticity of a transmitted document, three factors must be considered: the history of the text's transmission, its linguistic usage, and whether the content dis-plays internal consistency. In the following, each of these factors will be considered in turn, and all three will, I believe, demonstrate beyond any doubt that the "Shifu" is indeed a document con-temporary with the events it describes.

Section 1. Textual Transmission

The occurrence in the text of three full-date notations (i.e., month, lunar phase, and cyclical day; see sections II and IX in the text), while constituting a source of confusion for later commentators (see below, Sec. 3), has proven to be a happy coincidence; the paucity of such dates in the transmitted literature has made each of them extremely valuable to scholars of ancient China's chronology. The first of these scholars, Liu Xin

劉歆 (c. -46–+23), quoted these three dates verbatim and at rather considerable length in his Shijing 世經 (as quoted in Han shu, "Lüli zhi"42), all the while identifying the source of the dates as the "Wucheng" 武成 chapter of the Shangshu

尚書. Comparing the two texts (see fig. 1) their virtual identity leaves no doubt that they repre-sent one and the same text. Except for two or three minor inversions of word order, the only differ-ence of note lies in the first full-date notation.43 But the fact that the cyclical day specified in the "Shifu" date (惟一月丙辰旁生魄, 若翼日丁巳 ... "It was the first month, bing-chen (day 53), the (day of) "expanded growing brightness." On the next day, ding-si (day 54), ...) is incompatible with the date given for the battle at Muye, while the date given in the "Wucheng" is both mathematically compatible and independently verifiable (see n. 3), leads to the conclusion that the "Shifu" text is here corrupt, and not a separate text, and should be emended in light of the "Wucheng" text.

The coincidence of this quotation is by no means the only evidence that the present "Shifu" text does in fact represent the original text of the "Wu cheng" chapter. The "Preface to the Docu-ments" (Shuxu 書序) states with regard to the "Wucheng": "King Wu attacked Shang; he went out and attacked, returned and hunted; he made known the affairs of government and composed the 'Wu-cheng'."44 While this description is inconsis-tent with the contents of the guwen 古文 "Wu-cheng" text now extant, it corresponds perfectly with the "Shifu" text. Especially noteworthy is the mention of hunting, which figures so promi-nently in the text of the "Shifu" (Sec. VII), but which in the guwen "Wucheng" has been transposed into a pasturing of the war animals, symbolizing King Wu's pacifistic nature.45

Another early reference to the "Wucheng" is also illustrative of the content of the "Shifu," and not the "Wucheng" now existing. Mencius, for whom ancient history was a vital philosophical topic, apparently was unable to reconcile his view of a moral imperative manifesting itself through the Zhou conquest with the narrative given in the "Wucheng" text available to him. He voiced his opinion in a well-known passage.

Mencius said, "If one were to believe everything in the Book of History (Shu), it would have been better for the Book of History not to have existed at all. In the "Wucheng" chapter, I accept only two or three strips. A benevolent man has no match in the Empire. How could it be that the blood flowed with pestles (xue zhi liu chu), when the most benevolent waged war against the most cruel?"46

By the time of the Han dynasty, Mencius's view of history (and the attendant selection of historical sources) seems to have prevailed along with the Confucian school. Even the generally conscientious Sima Qian 司馬遷 -145-86) made no use of the "Wucheng" qua "Shifu" in his account of the Zhou victory, relying instead on the "Taishi"

泰誓 chapter of the Shangshu and especially the "Ke Yin" 克殷 chapter of the Yizhoushu, both of which contain linguistic features demonstra-bly anachronistic to the Shang—early Western Zhou period.47 Apparently only the skeptical pragmatist Wang Chong 王充 (+27—97) was not seduced by the Mencian view of history.

"Wu cheng": 惟一月壬辰旁死霸, 若翼日癸巳, 武王乃朝步自周, 于征伐紂 ...

粵若来三月既死霸粵五日甲子 , 咸劉商王紂 ... 惟四月既旁死霸粵六日庚戌, 武王燎于周廟, 翼日辛亥,

祀于天位 . 粵五日乙卯乃以庶國祀馘于周廟。

"Shifu": 惟一月丙辰旁生魄, 若翼日丁巳, 王乃步自于周, 征伐商王紂, 越若来二月既死魄越五日甲子朝至接于商則咸劉商王紂 ... 時四月既旁生魄越六日庚戌, 武王朝至燎于周 ...

若翼日辛亥祀于位, 用籥于天位, 越五日乙卯武王乃以庶國祀馘于周廟.

Figure 1

61

Now there are those who say that, when King Wu defeated Zhou, the blades of his weapons were not stained with blood. When a man with such strength (as Zhou) that he could twist iron and straighten out hooks, with such sup-porters as Fei Lian and E Lai tried issues with the army of Zhou (the people), King Wu, however virtuous he may have been, could not have de-prived him of his natural abilities, and Zhou, wicked though he was, would not have lost the sympathy of his associates....

King Wu succeeded King Zhou, and Gao Zu took over the inheritance of Ershi Huangdi of the house of Qin, which was much worse than that of King Zhou. The whole empire rebelled against Qin, with much more violence than under the Yin dynasty. When Gao Zu had defeated the Qin, he still had to destroy Xiang Yu. The battle field was soaked with blood, and many thousands of dead bodies lay strewn about. The losses of the defeated army were enormous. People had, as it were, to die again and again, before the Empire was won. The insurgents were exterminated by force of arms with the utmost severity. Therefore it cannot be true that the troops of Zhou (the people) did not even stain their swords with blood. One may say that the conquest was easy, but to say that the blades were not stained with blood is an exaggeration.

... According to the "Wu cheng," the battle in the plain of Mu was so sanguinary that the flow of blood floated pestles (xue zhi liu chu) and over a thousand li the earth was red. Ac-cording to this account the overthrow of the Yin by the Zhou must have been very much like the war between the Han and Qin dynasties. The statement that the conquest of the Yin territory was so easy that the swords were not stained with blood is meant as a compliment to the virtue of King Wu, but it exaggerates the truth.48

In addition to its general approbation of the "Wu-cheng"—"Shifu" text, Wang's statement also throws light on another feature of the identification prob-lem; that is, Mencius's description of the "Wucheng" as "the blood flowed with pestles" (xue zhi liu chu

血之流杵), which commentators of the Mencius have assumed to be taken verbatim from the "Wucheng," and the absence of which in the "Shifu" has led some scholars, Cheng Tingzuo 程廷袏 (1691-1767), for example, to dismiss the identifi cation between the "Wucheng" and the "Shifu." But it is at least as likely that with this four character phrase Mencius was simply using a euphemism for wide-scale bloodshed to describe the chapter—just as it would seem that Wang Chong was doing when he used the phrase "over a thousand li the earth was red" (chidi qianli 赤地千里) to describe the text. That the "Shifu" does not contain these exact words should not imply that it was not the text consulted by Mencius--its record of 177,779 deaths, 310,230 captives taken, 99 countries defeated, and routine human sacrifice, fits better than anything else the description "blood flowing with pestles."50

All of this suggests the following process for

the identification of the "Wucheng" with the "Shifu." The text was originally an integral part of the early Western Zhou documents linked together as the Shu. But having been denounced by Mencius, whose idealized view of history came to prevail, it was expunged from the Shu. By chance it was included in the col-lection of "left-over" texts, the Yizhoushu, which appears to have taken shape in the mid- -3rd cen-tury.51 With the subsequent appearance of the forged guwen chapters of the Shangshu, including a spuri-ous "Wucheng" chapter consistent with the romantic notions of Mencius, even the title of the text was compromised.

Fittingly, this matter of title is the final piece of evidence linking the "Shifu" with the "Wu-cheng." Two methods of entitling ancient documents were current during the late Warring States– early Han period: a title could either characterize the general import or contents of the text, or it could be based on the first words or the most sig-nificant words of the first line. In the case of the "Shifu" qua "Wucheng" text, both of these methods seem to have been employed. "Shifu," the "great capture," evokes the main theme of the text—the capture of the Shang domain, the Shang and their allies' subjects, of the Shang treasures, and also the capture of the animals in the victory hunt; in all, an extremely appropriate title. But the germ of the title "Wucheng" is also to be seen in the first line of the text, "Wuwang chengbi sifang, tong Yin ming youoguo" 武王成辟四方, 通殷命有國, where "Wuwang chengbi" ("King Wu achieved rule") must be considered the most important phrase of this preface. Taking the first words of each compound, a common practice of the Warring States title- givers,52 the result is, of course, "Wu cheng."53

In summary then, at least the following four points about the "Wucheng"

--- the Hanshu, "Lüli zhi" quotation, which accords, with the exception of one cor-ruption, exactly with the "Shifu" text;

--- the Shuxu description, which is consis-tent with the contents of the "Shifu";

--- the characterizations of the text given by Mencius and Wang Chong;

--- the possibility of deriving the title "Wucheng" from the first line of the "Shifu,"

leave little doubt that the "Shifu" text in front of us now is in substance the "Wucheng" text known to Mencius at the end of the -4th century.

Section 2: Linguistic Usage

While the identification of the "Shifu" with the "Wucheng," and especially its attested existence as such as early as the time of Mencius, suggests the authenticity of the text, the most reliable criterion of its antiquity lies in its linguistic usage. Not only is the text free of the sort of tell-tale linguistic anachronisms that frequently mar forgeries, but language similar to that in the early western Zhou chapters of the Shangshu has long attracted the attention of commentators.54 More important, features of the text apparently anomalous within the context of the transmitted

62

literature have appeared also in the inscriptions on oracle bones and bronze vessels. In the following, I will discuss some of these idiomatic constructions that seem to me to verify the text's contemporaneity with the events it describes.

The significance of the full date notations is discussed elsewhere in this chapter with respect to other considerations (see Part II, Sec. 1 and Sec. 3), but let us again examine the date for the battle at Muye, this time from the standpoint of linguistic usage. The line reads: 越若來二月既死霸越五日甲子, 朝至接于商, 則咸劉商王紂 "On jia-zi (day 1), five days after (the day) "after the dying brightness" of the following second month, (they) arrived in the morning and defeated the Shang, thence entirely decapitating the Shang king Zhou." Compare this with-a line in the "Shao gao" 召誥 chapter of the Shangshu: 越若來三月惟丙午朏, 越三日戊申, 太保朝至洛卜宅; 厥既卜, 則經營. "In the following third month, on bing-wu (day 43), the day of the moon's appearance; three days later on wu-shen (day 45), the Taibao arrived in the morn-ing at Luo and divined about establishing residence; after he had obtained the divination, he thence laid out the encampment." Not only are the general struc-tures of the two sentences similar, but such idiom-atic features as yueruolai 越若來, yue X ri 越 X

日, zhao zhi yu 朝至于, and the usage of ze 則, "thence," are identical. Moreover, in the "Junshi"

君奭 chapter of the Shangshu, the compound xian liu 咸劉, "entirely decapitate," occurs in a simi-lar context: 咸劉厥敵, "Entirely decapitated his enemies" (said of King Wu and his subordinates). When it is considered that the date notation ji siba 既死霸 is standard in Western Zhou bronze inscriptions (but, significantly, never appears after that period), virtually every segment of this "Shifu" sentence rep-resents attested early Western Zhou usage.55

Despite these parallels with the Shangshu, it is now recognized that a more valid authenticating methodology is to compare usage with oracle bone and bronze inscriptions, materials presumably unavailable to the aspiring forger. With this in mind, let us then consider the passage which immediately follows in the text. It reads : 太公命禦方來丁卯望至告以馘俘, and has been interpreted by several modern scholars as "The Grand Duke Wang ordered the Yufang to come. On ding-mao ... "56 Three features demonstrate the fallacy of this interpretation.

The word ming 命 occurs throughout the "Shifu" in a passive mode; e.g., 陳本命伐磨, "Chen Ben was ordered to attack Mo" (Sec. VI). Although this usage has no parallels in the Shangshu or Shijing, it does occur in the contemporary inscriptional literature. Consider, for example, the following pair of oracle bone inscriptions:

乙 8898 甲申卜: 命啄宅正Crack on jia-shen (day 21): "Order Shi to make a residence at Zheng."

乙 8893 甲申卜: 啄命宅正 Crack on jia-shen (day 21): "Shi is ordered to make a residence at Zheng."

This inversion clearly shows the possibility of ming being used as a passive. A perhaps more sub-stantive example occurs in a later Western Zhou bronze inscription, that of the Jing you 競卣.57 The first line of this inscription reads: 隹伯屖父以成即東命伐南尸 "It was when Boxin fu took the Cheng troops and proceeded to the east, being ordered to attack the Southern Yi." In this line, ming must be construed as a passive.58 It is also interesting to note that it occurs in the same construction as throughout the "Shifu"; i.e., ming fa 命伐 ("ordered to attack").

Bringing this information to bear on the sentence under discussion, another oracle-bone inscriptional idiom, yu fang 禦方, appears. Although yu is regularly defined as "to defend," because of its combination here with fang 方, ("country") it has been interpreted as a place-name by several renowned oracle bone scholars.59 Analyzing the following examples, however

後下 41.16 卜師乎御方于商Crack-making, Shi (divining); "Call to yu fang at Shang."

後下 42.19 壬午卜師貞玒命多視御方于...Crack-making on ren-wu, Shi divining: "The king orders the many Shi (?) to yu fang at ..."

佚 348 .. 巳卜王貞于中商乎御方Crack on ..-si , the king divining: "At Zhong Shang call to yu fang."

外 30 己卯卜王命御方Crack on ji-mao: "The king orders to yu fang."

南坊 3.62 丙辰卜御方Crack on bing-chen : "Proceed to yu fang."

庫 595 余勿乎御方"I should not call to yu fang."

前 5.2.7 貞, 冓于入御方Divining: "Have Gou go to yu fang."

63

甲 3539 辛亥卜貞, 令冓以御方于陟Crack on xin-hai, Zhong divining: "Order Gou with to yu fang at Zhi ."

one notes that the compound never acts as the sub-ject of a sentence nor precedes an active verb. What is more, the several examples here of yu following hu 乎 are comparable to such usages as hu fa 乎伐, hu qu 乎取, hu cong 乎從, hu lai 乎來, and hu chu 乎出, where hu is always followed by a verb. And while 前 5.2.7 has been noted as evidence of yu fang being a place-name, following yu 于 as it does,60 comparing this inscription with 甲 3539, where gou 冓 is certainly a personal name, it would seem that this yu 于 must have its verbal sense of "to go" rather than its prepositional sense "at." Thus, yu fang should not be interpreted, in any of these examples, as a place-name, but rather as a verb-object compound.

An example of yu 御 in a Western Zhou bronze inscription confirms this verbal usage and also pro-vides sufficient context to determine its meaning. The inscription on the Buqi gui 不𡢁𣪕, which dates from the reign of King Xuan 宣, reads:

It was the ninth month, first auspiciousness, wu-shen (day 45); Bo shi said: "Buqi, secure the area (yu fang 駿方).61 The Xianyun widely attacked Western Yu. The king commanded us to follow up in the west. I came in return to pres-ent the catch.

"I ordered you to mop up at Luo (yu zhui yu Luo 御追于 ).You took our war-carts and thoroughly attacked the Xianyun at Gaotao. You cut off many heads and shackled many captives. The Rong greatly converged to counter-attack you, and you were endangered. The Rong greatly pressed the attack but you were victorious, not taking our war-carts and sinking them in dif-ficulty, but cutting off heads and shackling cap-tives."

Bo shi said,"Buqi, you little child; you have begun adeptly in military merit. I present you one bow and a quiver of arrows, five families of servants, and ten fields of land, to be used in fulfilling your affairs."

Buqi folded his hands and touched his head to the ground, herewith making "my illustrious grandparents Gongbo and Mengji" precious gui-vessel, and herewith entreating many blessings and long life without limit, eternal purity and a spiritual end. Would that my children's chil-dren and grandchildren's grandchildren eternally treasure and use it in sacrifice.62

This inscription is related to that on the Guo-jizi Bo pan 虢季子白盤, which provides the setting for the action described here. The Guo-jizi Bo pan has a full date placing it in the 12th year of King Xuan, and narrates the court ritual marking a victory by Guojizi Bo. It ap-pears that this victory, though conclusive, was not quite final, for while the Buqi gui inscrip-tion lacks a year notation, the otherwise com-plete date corresponds exactly with the 11th year of King Xuan. The content of the inscription shows this date to be correct. It begins with a reference to Bo's (i.e., Guojizi Bo) recent vic-tory, which is followed by an order to his lieu-tenant Buqi to complete the pacification of the area. This order, first mentioned in the opening, prefatory paragraph, is further elaborated by the phrase "yu zhui yu Luo" 御追于 , where yu combines with zhui to give the sense "to defend against by chasing after," which is to say, to conduct a sort of "mopping-up" operation.

This is also the meaning that the context of the "Shifu" calls for. Significantly, it is also the meaning that Kong Zhao 孔晁 (c. +265), the earliest commentator of the text, apparently intended by his gloss: 太公受命追禦紂黨方來 "The Grand Duke received an order to chase after and yu (the Shang king) Zhou's allies, the Fanglai." But because Kong, writing in the late +3rd century, was unfamiliar with the language of Shang Chinese,he was led to make one minor mis-take in his comment.

Two words, yi 翌 and lai 來, are used in oracle bone inscriptions to describe future time.63 The following examples show the distinc-tion between them:

乙 6385 甲寅卜貞,翌乙卯易日.Crack on jia-yin (day 51), Que divining: "On the next yi-mao (day 52), it will be a clear day."

乙 7258 癸卯卜 , 翌甲辰酒大甲.Crack on gui-mao (day 40), Que divining: "On the next jia-chen (day 41), perform a wine libation to Da Jia."

粹 605 乙酉卜賓貞, 翌丁亥不其易日.Crack on yi-you (day 22), Bin divining: "On the next ding-hai (day 24), it will not be a clear day."

前 7.27.2 戊辰卜爭貞, 來乙亥不雨Crack on wu-chen (day 5), Zheng divining: "On the next yi-hai (day 12), it will not rain."

續 3.15.1 丁酉𣪗貞, 來乙巳王入于 . Ding-you (day 34), Que divining: "On the following yi-si (day 42), the king will enter at ."

64

粹 692 自今辛至于來辛有大雨From this xin (day) all the way to the coming xin (day), there will be a great rain.

Yi is used for either the next day (as it is also used elsewhere in the "Shifu" [Sec. II] and in the early portions of the Shangshu), or occasion-ally two days hence, but never more. Lai, on the other hand, always indicates a day separated by a longer interval. This is also the case with the construction in the "Shifu," lai ding-mao, where ding-mao (day 4), the day of the Grand Duke's return, is four days distant from jia-zi (day 1), the day on which he received his orders. This correspondence with Shang usage demonstrates that lai here is neither a verb meaning "to come" (as so many oracle bone scholars have assumed) nor part of a compound place-name (Fang lai, as interpreted by Kong Zhao). Rather, it is a marker of future action, which, when combined with the other idiomatic usages discussed above (ming, yu, fang), shows that the sentence should be trans-lated, "The Grand Duke Wang was ordered to secure the area. On the following ding-mao, Wang arrived and reported about ears-taken and captives."

The above two examples both demonstrate clus-ters of Shang—Western Zhou idioms. The text also abounds with individual vocabulary items with char-acteristics dating from that linguistic period. I will briefly illustrate just three or four of these.

The word fu 俘, of course, occurs often in the text. Although in later usage fu is generally nominal, in this text it is used in three dis-tinct ways: as a noun meaning" captive" (gao yi guo fu 告以馘俘; Sec. III, IV, VI), as a verb for the capture of humans (fu Aihou, Yihou 俘艾侯佚侯; sec. VI), and as a verb for the capture of inanimate objects (fu Yinwang ding 俘殷王鼎; Sec. V; Wuwang fu Shang jiu yu 武王俘商舊玉; Sec. X). The latter two uses were common in the language of the Shang-early Western Zhou, as the following examples (first a Shang notational inscription, and then two bronze inscriptions) demonstrate.

菁 5 四日庚申亦㞢棧艱自北, 子告曰: 昔甲辰方征于 , 孚人十又五人, 五日戊申方亦征, 孚人十又六人, 六月在. . .Four days (later), on geng-shen (day 57), there again came trouble from the north. Prince Tuan reported saying: "Last jia-chen (day 41), the Fang were on campaign in You and captured 15 men. Five days (later), on wu-shen (day 45), the Fang were again on campaign and captured 16 men. Sixth month at...

Xiao Yu ding 小盂鼎: 執嘼二人, 隻八百□十二 , 孚人萬三千八十一人, 孚馬□□匹, 孚車兩, 孚牛三百五十五牛, 羊八羊I took prisoner two of their chiefs, obtained 4,8X2 scalps, captured 13,081 men, captured... horses, 30 chariots of war, 355 oxen, and 38 sheep.64

Shi Yuan gui 師㝨𣪕: 毆俘士女羊牛, 孚吉金We routed and captured men, women, sheep and oxen; we captured auspicious metal.65

As significant as the record of the captures is in the text, it is clear that for the Zhou composer of the narrative, the sacrifices, particularly the human offerings, which followed the captures held much greater symbolic significance. It goes with-out saying that the parallel between the liao 燎 burnt-offering sacrifice here with that in the Xiao Yu ding inscription is striking. (See above, n. 34.) More subtle, perhaps, but equally characteris-tic of contemporary usage are the other terms used in the "Shifu" for sacrifice. Fa 伐, which mid-way through the Western Zhou period came to be used exclusively to indicate military attacks, was the most commonly used specific human-sacrifice term in Shang oracle bone inscriptions. As the character graphically depicts, it indicated the decapitation of a prisoner.66 The use of fa in the "Shifu," as for example 伐厥四十夫家君鼎師 "(He) beheaded their forty family, masters of the caldron" (Sec. IX), certainly indicates the same type ritual.

In addition to specific sacrifice terms such as liao and fa, oracle bone usage also included a pair of general sacrifice terms, you 㞢, 又 (i.e., 侑)67 and yong 用.68 Let us consider here just the case of yong.

續 1.44.1 丁卯卜: 用𠬝于兄已Crack-making on ding-mao: " Yong cap-tives to Xiong Ji."

南明 525 其用人牛十又五"Let us yong men and oxen, 15."

This sacrificial sense of yong is also evident in such early Western Zhou bronze inscriptions as that on the Ling yi 令彜: 乙酉, 用牲于康宮 "On yi-you, yong offering in the Kang Pavillion."69 To these examples compare the sentence of the "Shifu,

用牛于天于稷五百有四 "He used 504 oxen to heaven and to Ji," where yong is equally unmistak-able as a sacrifice term.70

Added to the features discussed above in the notes to the translation, such as the general and

65

specific similarities with the Xiao Yu ding inscrip-tion (see n. 36), the parallels with Shang hunting records (see n. 27), and the place-names mentioned in the text (and especially their attested loca-tions; see notes 9 and 22), such linguistic features as these strongly suggest that the text was indeed composed at or about the time of the events it de-scribes.

Section III: Content and Textual Integrity

The final criteria to be considered in judging the historical worth of any document are appropri-ate content and textual integrity. In the preced-ing discussion, specific points of linguistic usage have been shown to be appropriate to the Shang-Zhou transition era, but in general flavor as well, it seems self-evident, as Wang Chong noted during the Han period (see above, p. 61), that the "Shifu" is a realistic account of the events surrounding the conquest of one dynastic power and the investiture of a new people with that power. Granted, the obser-vation of Wang Chong was based in large part on a common-sensical comparison with contemporary events, but such common-sense generally makes for better historiography than the idealism characteristic of Mencius, for example.

It is within the sphere of textual integrity, however, that most scholars have presumed difficul-ties with the text. These difficulties revolve around the three full date notations. Disregarding for the moment the problems with the first of these dates which has been discussed at some length above (p. 3), the correlation between the second and third dates and the events that could have occurred within their interval is the crux of a perceived textual disorder. The second date, that for the battle at Muye, corresponds to the 28th day of the second month, while the third date, that for the burnt- offering sacrifice in Zhou, corresponds to the 16th day of the fourth month.71 The interval between these two dates is generally consistent, allowing one minor modification, with the interval of 47 days indicated by the cyclical days mentioned, jia-zi (day 1) and geng-xu (day 47).72 If these two dates were the only evidence available, as they are in the Hanshu quotation of the "Wucheng," it would be a simple matter to make the minor calendrical modifica-tion required.73 But the blessing, and to some the bane, of the "Shifu" is that, to a great extent, it fills in the events that occurred during the inter-val (unfortunately not with full-date notations, but always with cyclical days), and this chronicle demonstrates that such a simple 47 day interval is impossible.

The impossibility of this simple calendrical sequence is shown most clearly by the dual occur-rence of the cyclical date xin-hai (day 48), the first time marking the presentation to King Wu of the captured Yin caldrons in the Shang capital (Sec. IV);74 and the second time, when King Wu ritually assumed his throne in the Zhou capital (Sec. IX). It is naturally inconceivable that a person 3,000 years ago could have performed actions in two places separated by some 550 km. on the same day. The in-escapable conclusion is that these two xin-hai day notations must have been separated by a full sixty-day cycle, which is both consistent with the se-quence given in the text,75 and compatible with the

logistical requirements of moving an army from the environs of Anyang to Xi'an.76

Allowing for this extra 60 days between the battle at Muye and the victory ceremonies at Zhou, the interval is not 47 days but 107 days. When the cyclical days indicated for these events (i.e., jia-zi and geng-xu are inserted into a lunation chart covering this 107-day interval (see the ap-pended "Calendar of Events"), it can be seen that the lunar-phase notations for the two dates coin-cide exactly with their standard definitions. Such a calendar also brings to light another interest-ing feature: both the victory celebration at Zhou and the parallel ceremonies in the Shang capital begin on the 16th day of their respective lunar months. This is not coincidental, for the 16th day of a luna tion marks the day after the full-moon, and evidence from Western Zhou bronze inscriptions confirms that it was during the pre-dawn hours of this day (i.e., still during the 15th night, when the moon is at its fullest) that ceremonies were commonly held.77

But as appropriate as this correspondence between the lunar-phase notations and the cycli-cal days looks, there is one major problem: the sacrifice in the Zhou capital is stipulated as taking place in the fourth month, while this in-terval of 107 days beginning from the 28th day of a second month satisfies the requirements for a sixth month. Stymied by this, Gu Jiegang 顧頡剛 (1893–1980),the most thorough modern commentator of the "Shifu," despaired of being able to make sense of the text's calendrical system.78 But I believe that, as is so often the case, a very simple pro-posal resolves the problem. On wu-chen (day 5; and corresponding to the 3rd day of the third month), five days after the climactic victory at Muye and the day after the Grand Duke Wang reported his suc-cess in eliminating the last enemy resistance in the capital area, the text states laconically that King Wu "established his government" (Sec. III). I propose that at this time, he also promulgated a new calendar, declaring the beginning of a new year. Thus, what according to the calendar in ef-fect before the conquest would have been designated as the third month, was thence designated the first month. In like fashion, what would have been the sixth month became the fourth month in the new Zhou calendar. In this way, both the arrangement of the text and the two full-date notations are mutually corroborating, and combine to confirm the integrity of the text.

In addition to being logically required by the text, there is also support for this calendar revision in later tradition. Since at least the devel opment of the san-tong 三統 theory by Liu Xin, it has been believed that the custom in an-tiquity was to promulgate a new calendar upon the establishment of a new dynasty. The most explicit statement of this tradition is found in the "Zhou yue" 周月 chapter of the Yizhoushu, a chapter which dates from very late in the Warring States period.79

The ten-thousand things are born in spring and grow in summer, are collected in autumn and stored in winter: this is the correctness

66

of heaven and earth, the epitome of the four seasons, and the unchanging Way. The numera-tion of the Xia dynasty accorded with heaven, and the hundred kings have shared in it. With Tang of the Shang dynasty, troops were used against the Xia: he dispelled the misfortunes of the people, and in accord with Heaven over-turned the mandate; he revised the first month, changed the clothing and distinguished the titles. One refined (the Shang) and one sub-stantial (the Xia), the sacrifices were incom-patible, and the chou month (i.e., that after the month containing the winter solstice) was taken as the first month.

Changing the views of the people, like the great changes in the heavenly time, is also the affair of a single dynasty. Coming then to our Zhou kings, they were brought to attack the Shang. They revised the beginning of the year (gai zheng 改正) and differentiated the standards, leaving to posterity the santong (the three systems).

As for respectfully granting timeliness to the people and maintaining the sacrifices, it is still the Xia (calendar) that is the best. That which is called the Zhou months is used in government annals (yi ji yu zheng 以紀于政).

From this passage it is clear that there was indeed a tradition concerning a change of calendar at the time of the Zhou conquest, but it would seem that the original simple declaration of a new year by King Wu had been later elaborated into a systematic change, consistent with the philosophical system-atizing of the period.80

Finally, the calendrical reconstructions of Dong Zuobin 董作賓 (1895–1963) also offer a striking confirmation of this calendar of the "Shifu." The calendar required by the "Shifu," with lunar months beginning on cyclical days ding-mao (day 4), ding-you (day 34), bing-yin (day 3), bing-shen (day 33), yi-chou (day 2), and yi-wei (day 32), respectively (see the appended "Calendar of Events"), coincides precisely with that of the year -1045,81 the year which Professor David S. Nivison has, on the basis of several independent factors, determined to be the year of the Zhou conquest.82 This firmly establishes that the "Shifu" can be nothing other than an authentic account of the events immediately following that conquest.

Part III: The Conquest Campaign

Being thus satisfied of the authenticity of the "Shifu" text, it is possible to use it as the basis for the following chronicle of the Zhou conquest campaign.

Although the text of the "Shifu" begins with King Wu's departure from Zhou on gui-si (day 30), the 27th day of the first month (pre-conquest calendar; Dec. 15, -1046), there is reason to believe that this did not mark the beginning of the campaign. The "Taishi" 泰誓 chapter of the Shangshu includes the curious statement, "It was on bing-wu (day 43; correlated with the "Shifu"

calendar, the 10th day of the second month [pre-conquest calendar]; Dec. 28 -1046) that the king reached the troops." Combined with the specificity of the "Shifu" that King Wu, and presumably only King Wu, departed on gui-si, this shows that the Zhou army must have departed at an earlier date (as the logistics of the campaign would dictate; see above, n. 3). When did the army depart? A well-known tradition in the "Zhouyu" 周語 chapter of the Guo yu 國語 gives a precise astronomical date. The passage reads: 昔武王伐殷, 歲在鶉火, 月在天駟, 日在析本之津, 辰在斗柄, 星在天黿 and has been translated by Prof. Nivison as, "Long ago, when King Wu attacked Yin, the Year Star was in Quail Fire, the moon was in the Heavenly Sky Team of Four, the sun was in the Ford at Split Wood, the (next) conjunction (of sun and moon) was in the Handle of the Dipper, and the Star was in the Sky Turtle."83 According to Prof. Nivison, the combination of locations for Jupiter, the sun, moon, and the lunisolar conjunction given here, is satisfied by only one day during the years -1046–45; that day is Nov. 16 1046. Correlating this date with the dates required by the "Shifu," it corre-sponds to the 27th day of the twelfth month (pre-conquest calendar). Computing then the cyclical day for this date, one finds that it is, certainly not by coincidence, a jia-zi day. Deeply concerned with temporal portents it is to be expected that the Zhou would select a particularly auspicious day on which to begin their conquest campaign. As the decisive battle at Muye, which the Zhou apparently timed to occur on a jia-zi day, shows, there could be no day more appropriate for this than jia-zi, the first of the cycle.

This concern for timely actions conceivably also dictated the departure date of King Wu. It is well known that the governmental custom during the late Shang (particularly period V) was for the king to divine on gui days, the last day of the ten-day week, as a means of ensuring good fortune in the coming ten days. It is plausible that the Zhou shared this divination tradition, in which case one might assume that on the morning of gui-si (Dec. 15, -1046), one month after the departure of the army, King Wu performed the final formal divination in the Zhou temple. Having thus ensured that future actions would be fortuitous, he set out to join his troops. Whether in chariot or on horseback, there is no doubt that the king and his entourage would have moved far more rapidly than the large Zhou army. That he overtook the troops on bing-wu (day 43; Dec. 28, -1046) adds a practical verification of the date Nov. 16 for the departure of the army, for this shows that King Wu in 14 days covered the same distance that it took the army 43 days to traverse, a ratio which practical experience might indeed suggest.

The next mention of the conquest army's move-ments has them completing the crossing of the Yellow River at Mengjin 孟津 on wu-wu (day 53; Jan. 9, -1045).84 From there, a march of six days brought them on gui-hai (day 60; Jan. 14 -1045) to Muye, where they deployed in anticipation of the battle which would take place the next morning. Everyone now knows that the battle resulted in an over-

67

whelming triumph for the Zhou army; but as the "Shifu" shows, and as common sense should lead us to ex-pect, it was by no means the end of all hostilities. Still to come would be a series of "mopping-up" operations before the Zhou could be secure in their victory. The first of these operations was initi-ated as soon as the victory at Muye had been won; the Grand Duke Wang, the commanding general of the Zhou forces, was ordered to secure the area of the Shang capital. Wang accomplished his mission in four days, and on the following day, wu-chen (day 5; Jan. 19), King Wu formally declared the "estab-lishment of government."

It is one thing to establish a government; it is quite another thing to ensure that it is a lasting one. The next section of the "Shifu" men-tions campaigns, beginning on ren-shen (day 9; Jan. 23), against a number of Shang vassal states lo-cated to the north and east of the former capital. But immediately before these campaigns began, King Wu presided over a sort of victory ceremony in an important cult center near the Shang capital. The Li gui, the vessel discovered in 1976 which has excited so much interest because of its verifica-tion of the jia-zi date for the battle of Muye, was cast after the presentation of metal to Youshi Li

有使利, its maker, by the king.85 This presenta-tion took place on xin-wei (day 8; Jan. 22) at Jian

, which inscriptions on late Shang bronze vessels show to have been a powerful Shang ally situated near the Shang capital, and where a "great temple" (taishi 太室) was located.86 When correlated with the narrative of the "Shifu," this ceremony appears to have been intended to reward the soldiers after one great victory and to encourage them in the battles to come.

The "mopping-up" campaign against the states Yue, Xifang, Chen, and Wei then lasted until jia-shen (day 21; Feb. 4; for a full discussion of the location of these place-names and the possible ramifications, see above, n. 9) After the report of the successful conclusion of this campaign, the "Shifu" record is blank for nearly a month. We can only conjecture that the time must have been filled attending to the routine duties of an occupation army. But on the first full moon af-ter having thus put the former capital firmly under its control, the Zhou celebrated their victory in the Shang capital. The events of days xin-hai (day 48; March 2) to yi-mao (day 52; March 6) are suf-ficiently well described in the text that they need no amplification here. Suffice it to make one general remark: these ceremonies display a certain diplo-matic and psychological intuition on the part of King Wu. They seem to have been designed to in-voke in the conquered Shang people awe and respect for the somewhat barbarian Zhou conquerors, while at the same time attempting to minimize any feel-ings of enmity. It is undoubtedly for this reason that although King Wu was there presented with the captured Shang officials, he did not execute and sacrifice them immediately, but preferred instead to transport them back to the Zhou homeland.87 Al-though Sec. IX shows that he did indeed execute and sacrifice them there, this was exclusively a Zhou affair, well out of sight of the Shang people. It

was perhaps in part because of this seeming policy of clemency that the Shang army was willing to sub-mit to Zhou command, as the events of jia-yin (day 51; March 5) seem to suggest they did.

With the conquest now complete, not only in military terms but just as importantly politically (note the "confirmation" of the heads of states on days xin-hai [day 48] and ren-zi [day 49]) and psycho-logically, King Wu and the Zhou army (or at least that portion of the army that was not to be garri-soned in the former Shang domain) were free to return to Zhou. Considering the logistics of the march, the departure must have taken place very soon after the end of these ceremonies. But, even though the Zhou were now masters of the Shang and their subjects, that was no longer sufficient to content the people who considered themselves the recipients of Heaven's mandate to rule all of China. It was now incumbent upon King Wu to bring even those states in the west, hitherto enemies of both the Shang and the Zhou, under his control. This was effected by means of a final "mopping-up" operation, conducted against the states of Mo, Xuanfang, Shu, and Li, while the Zhou army was making its return to the Zhou homeland.

With this western campaign successfully con-cluded (on yi-si [day 42; April 25]) and all recal-citrant states now submissive, King Wu, once again in the Zhou capital, was then free to celebrate his victory at home. The celebrations again began on the first full-moon after this occasion (geng-xu [day 47; April 30]), and this time were marked by a rather liberal shedding of human blood. Despite the protestations of later Confucians that the founding fathers of the Zhou were too virtuous to engage in such violence, the "Shifu" leaves no doubt that they engaged in the practice with all the vigor for which the Shang, their more "civilized" pre-decessors, were noted. Whether or not the victims' blood propitiated the Zhou ancestors, it must cer-tainly have demonstrated to the Zhou people that they were indeed now the rulers of China.

* * * * *

In conclusion then, modern scholars are jus-tifiably excited whenever a bronze vessel bearing an inscription is unearthed. But earth is not the only thing that can bury authentic records. In the case of this "Shifu" text, Confucian idealism has been nearly as obscurant. Let us not succumb to our own prejudices against "unattested" evidence, but instead examine this text just as we would a newly discovered bronze inscription. Whether in terms of chronology, military affairs, or court ritual, it has much to tell us about the very early Zhou.

NOTES

1. Due to a loan possibility of "shi" 世 for "da" 大 in archaic Chinese, commentators of the text generally interpret the title to mean "da fu," i.e., "The Great Capture." (See e.g., Gu Jiegang, 1963:2.) In terms of modification, this reading seems preferable to "world's capture," and is here

(Notes continue on p. 70)

68

建戍, 12 月 (10/21-11/18) 建亥, 1 月 (11/19-12/18) 建子, 2 月 (12/19-1/16, -1045)1 丁卯

4丁酉 34

1

2 戊辰 5

戊戌 35

2

3 己巳 6

己亥 36

3

4 庚午 7

庚子 37

4

5 辛未 8

辛丑 38

5

6 壬申 9

壬寅 39

6

7 癸酉 10

癸卯 40

7

8 甲戌 11

甲辰 41

8

9 乙亥 12

乙巳 42

9

10 丙子 13

丙午: King Wu reaches Zhou army. 43

10

11 丁丑 14

丁未 44

11

12 戊寅 15

戊申 45

12

13 己卯 16

己酉 46

13

14 庚辰 17

庚戌 47

14

15 辛巳 18

辛亥 48

15

16 壬午 19

壬子 49

16

17 癸未 20

癸丑 50

17

18 甲申 21

甲寅 51

18

19 乙酉 22

乙卯 52

19

20 丙戌 23

丙辰 53

20

21 丁亥 24

丁巳 54

21

22 戊子 25

戊午: Army fords Yellow River at 55 Mengjin.

22

23 己丑 26

己未 56

23

24 庚寅 27

庚申 57

24

25 辛卯 28

辛酉 58

25

26 壬辰 29

壬戌 59

26

27 甲子: Zhou army departs. 1 (Nov. 16, -1046)

癸巳: King Wu departs from Zhou. 30

癸亥 60

27

28 乙丑 2

甲午 31

甲子: Zhou victory at Muye. Grand Duke 1 Wang ordered to secure capital.

28

29 丙寅 3

乙未 32

乙丑: 2

29

30 丙申 33

30

Calendar of Events

69

Calendar of Events (2)

建丑, 3-1 月 (1/17-2/16) 建寅, 2 月 (2/17-3/16) 建卯, 3 月 (3/17-4/16) 建子, 4 月 (4/17-5/16)1 丙寅

3丙申 33

乙丑 2

乙未 32

1

2 丁卯: Wang reports success. 4

丁酉 34

丙寅 3

丙申 33

2

3 戊辰: King Wu performs rituals, 5 establishes government.

戊戌 35

丁卯 4

丁酉 34

3

4 己巳: 6

己亥 36

戊辰 5

戊戌 35

4

5 庚午 7

庚子 37

己巳 6

己亥 36

5

6 辛未: Ceremony at Jian; Li gui 8 cast.

辛丑 38

庚午 7

庚子: Attacks on Mo, Xuanfang, 37 Shu.

6

7 壬申: Victory over Yue, Xifang, 9 attack on Chen.

壬寅 39

辛未 8

辛丑 38

7

8 癸酉: 10

癸卯 40

壬申 9

壬寅 39

8

9 甲戌 11

甲辰 41

癸酉 10

癸卯 40

9

10 乙亥 12

乙巳 42

甲戌 11

甲辰 41

10

11 丙子 13

丙午 43

乙亥 12

乙巳: Victory reported. 42

11

12 丁丑 14

丁未 44

丙子 13

丙午 43

12

13 戊寅 15

戊申 45

丁丑 14

丁未 44

13

14 己卯 16

己酉 46

戊寅 15

戊申 45

14

15 庚辰 17

庚戌 47

己卯 16

己酉 46

15

16 辛巳: Victory over Chen. 18

辛亥: Presentation of Yin 48 caldrons.

庚辰 17

庚戌: Liao-sacrifice in Zhou 47 temple.

16

17 壬午 19

壬子: Ritual ceremonies. 49

辛巳 18

辛亥: Sacrifice as king. 48

17

18 癸未 20

癸丑: Presentation of Yin 50 captives.

壬午 19

壬子 49

18

19 甲申: Attack on Wei led by 21 Bai Ta.

甲寅: Inspection of Yin army 51 at Muye.

癸未 20

癸丑 50

19

20 乙酉 22

乙卯: Musical ceremony. 52

甲申 21

甲寅 51

20

21 丙戌 23

丙辰 53

乙酉 22

乙卯: Feudal lords come to 52 court.

21

22 丁亥 24

丁巳 54

丙戌 23

22

23 戊子 25

戊午 55

丁亥 24

23

24 己丑 26

己未 56

戊子 25

24

25 庚寅 27

庚申 57

己丑 26

25

26 辛卯 28

辛酉 58

庚寅 27

26

27 壬辰 29

壬戌 59

辛卯 28

27

28 癸巳 30

癸亥 60

壬辰 29

28

29 甲午 31

甲子 1

癸巳 30

29

30 乙未 32

甲午 31

30

70

adopted.

The recension and these textual notes are based on the Sibu congkan 四部叢刊 text, Yizhoushu (the "Shifu" comprises chapter 40, 4.9a–12a), which in turn is a photo-reprint of a Ming Jiajing era edition (printed in 1543), the oldest extant text. (In these notes it is this text that is intended when I refer to the "original text.") I have also extensively consulted the critical recension by Gu Jiegang (1963), which not only presents Gu's origi-nal research but also synthesizes the various Qing editions and scholarship.

2. It would seem that this summation is a later addi tion to the basic text, akin to the "Preface to the Documents " (Shuxu 書序). The record, however, does seem reliable when correlated with the remain-der of the chapter. According to my chronologi-cal reconstruction of the events, yi-wei marks the first day of the fourth month, a time at which King Wu and the Zhou army were returning through former Shang territory to the Zhou homeland. (See below, p. 67, and especially the appended "Calendar of Events.")

3. The original text reads 維一月丙辰旁生魄若翼日丁巳王乃步自于周征伐商王紂; the emendation is based on the quotation in Han-shu, "Lüli zhi" (1015-16) of Liu Xin's 劉歆 Shijing 世經, which in turn quotes the "Wucheng" chapter of the Shangshu. The occurrence here of a full-date notation (i.e., month, lunar-phase, and cyclical day) makes it possible, when correlated with the two other full-date notations (Sec. II, Sec. IX), which are identical in the "Shifu" and Hanshu texts, to determine the correctness of the "Wucheng" date.

In his Shengba siba kao, Wang Guowei demon-strated on the basis of bronze inscriptional evi-dence that the traditional definitions of Liu Xin for the lunar phase notations (defining "shengba" as the full-moon and "siba" as the new moon) were in-correct. Wang, on the other hand, divides the month into four quarters, with chuji 初吉 representing days 1 to 7 or 8, ji shengba 既生霸 (note that there is a functional equivalence between the terms ba 霸 and po 魄 which occurs in the "Shifu") days 8 or 9 to 14 or 15, ji wang 既望 15 or 16 to 22 or 23, and ji si ba 既死霸 days 23 to the end of the lunation. In his essay Wang discusses the dates of the "Shifu" (as given in the Hanshu quotation), concluding that this first full-date designates the 26th day of the first month, and that the date given for the battle at Muye designates the 27th day of the second month.

While generally reliant on Wang's thesis, the calendrical reconstructions presented in this paper represent a slight modification in detail. Instead of interpreting ji siba as designating the first day of the third quarter, as does Wang, I interpret it to be the following day, "ji" having a common mean-ing of "after." This entails a one day divergence

in the day placements within the lunation, my calen-dar giving this first full-date as the 27th day of the first month, and the date for the battle at Muye as the 28th day of the second month. (When correlat-ing these two dates, however, the interval between them, 32 days, is identical to that given by Wang.)

After having determined by means of the lunar phase notation the placement of the days within the lunation, it is a simple matter to count the cycli-cal days to see that the two dates, as given in the Hanshu, are in perfect correspondence. (It is per-haps worthy of note that they are also in perfect correspondence according to Liu Xin's definitions, but with his definitions the date for the battle and the fourth month date for the burnt-offering sacrifice in Zhou can be made to correspond only through the insertion of an intercalary month.) Such is not the case with the date given in the original text of the "Shifu." The lunar-phase notation there (旁生魄若翼日) designates either the 10th or the 11th day of the lunation, but the cyclical day ding si (day 54) is completely out of synchro-nization with the date for the battle (see, "Calen-dar of Events.")

Despite the calendrical correctness of the Hanshu "Wucheng" date here, Qu Wanli (1965: 317-19) has argued that it is also suspect because the inter val of 31 days thus afforded between the depar ture from Zhou and the battle at Muye would be insufficient time to move an army the roughly 550 km. between the Xi'an area and Chaoge. (Qu mentions 700 km., relying on Liu Xin's estimation of the distance instead of a map.) Qu's suspicions with regard to the logistical problems are well-taken, yet a correct reading of this line resolves the difficulty. The date does not mark the departure of the Zhou army; rather, it is explicit in mentioning only King Wu's departure. Other sources indicate that the Zhou army began its march 30 days earlier, and that it was only 14 days after the date given here that King Wu joined his troops en route to the battle. (For a full discussion, see below, p. 66) The marching time of 60 days thus afforded is quite consistent with the speed of troop movements in an-cient times.

Having said this, it is necessary to digress slightly and consider more fully this problem of logistics. Included with the passages from the "Wu-cheng" in the Hanshu quotation is a remark by Liu Xin that troops moved 30 li per day, or roughly 15 km. per day. This figure seems to have been widely accepted in ancient sources, being cited by Du Yu

杜預 (+222-284) in his commentary to the Zuozhuan (Chunqiu jingzhuan jijie, 7 .5a), and in the Baihu tong (p. 628). Moreover, Mao Chang 毛萇 (-2 c.) gives the same gloss, interpreting the poem "Liu yue" 六月 (Mao, 177) of the Shijing as testifying to this same tradition at a much earlier time: 我服既成, 于三十里 "Our clothes being ready, we marched 30 li" (Maoshi, 10.6b-7a).

According to this tradition, the marching time between Xi'an and Chaoge would have been about 35 days. But this type of tradition must be authenti-

71

cated by actual campaign records. Surprisingly, there exists a nearly day-by-day account of a 10-month campaign virtually contemporary with that of King Wu's against the Shang (and it is even more temporally comparable in that it was also conducted during the inclement winter months.) This campaign was led by Di Xin 帝辛, the final Shang king, against the Renfang 人方, a campaign which though successful is credited by the Zuozhuan (Zhao 11; Legge, CC V: 633-4) as having so weakened the Shang state that it prepared the way for the Zhou con-quest. During this campaign, Di Xin routinely performed divinations, and a corpus of 150 of the divination inscriptions has been compiled by Dong Zuobin (1945: Xia 9.48a-63b). Because nearly all of the inscriptions record the date (giving month and cyclical day) and location, it is possible to plot the course of the campaign rather precisely. (See e.g., Chen Mengjia, 1956: 301-9 and esp. the ap-pended map, pl. 10.) Analyzing the most reliable segments of the march, the following table of troop movements can be compiled. (See fig. 2)

Depart Point

Arrival Point

Depart Date

Arrival Date Distance Time km/day

1. 雇雷 4-9 27-28 150 18-24 6.25-8.33

2. 亳攸 50 6-10 110 17-21 5.24-6.47

3. 攸𣴛 11 21 130 10 7

13.00 18.57

4. 商云 44 10 200 25 8.00

Figure 2

The table divides naturally into two sections, 1, 2, and 4 showing an average speed of 7-8 km/day, and 3 showing a revised (discounting three days spent in one encampment) speed of 18 km/day. There is a simple explanation for this divergence. The march described in 3 was in enemy territory, pre-sumably in pursuit of the enemy. With the exception of the three day encampment, it can be assumed that the army was on the march every day. 1, 2, and 4, on the other hand, represent segments of the march to and from the battle area, and the inscriptional record shows that they were not sustained marches. In general, the practice seems to have been to march a day and camp a day. To account for this 50% of the campaign time spent encamped, the average rate of troop movement/day for 1, 2, and 4 should thus be approximately doubled, giving an average marching speed of about 15-16 km/day, corresponding exactly with the later tradition.

This is just as we should expect. A large army on the move cannot sustain over a long period the pace it can keep for one day. Applying the logisti-cal information from this record to the campaign described in the "Shifu," we should likewise not expect the Zhou army to have covered the 550 km distance in the 36 days a daily march of 15 km would permit, but should rather expect something closer to the 60 days re-constructed here for the march on Shang, or the 55 days for the return to Zhou.

4. As outlined above (n. 3), this date corre-sponds to the 28th day of the month.

5. The Jingdian shiwen 經典釋文 (2.14b) glosses an occurrence of jie 接 in the ZhouYi

周易 as a loan for 捷 ("to defeat"). For other examples of this loan in classical texts, see Gao Heng, 1947: 120; and for a phonological discussion of a "ji-jie" word family meaning "to attack, to conquer," see Ping Xin, 1979: 55-60.

6. For a detailed discussion of this sentence, see below, pp. 62-64.

7. According to tradition, the left ear of a slain enemy was severed to prove the kill; the Shuowen definition reads: 聝軍戰斷耳也 ...

聝或从首 "Guo, the severing of ears by troops in battle.... It may also be written with 首 ('head') signific." (Shuowen jiezi gulin: 5361a)

8. Emending 自祀 to 追祀 with Kong Zhao 孔晁 (c. +265)'s commentary.

Both yu 禦 and xun 循 occur commonly in oracle bone inscriptions, meaning an exorcistic ritual (see Xu Jinxiong, 1963: 3b) and a kingly inspection tour (see Li Xiaoding's comments in Li Xiaoding, 1965: 567-9) respectively. Previ-ous commentators have apparently not been aware of these usages, stating, for instance, only that yu is the name of a sacrifice (Kong Zhao), or worse, have misinterpreted the nature of the ritual, with such suggestions as a sacrifice to the chariot god (Chen Pengheng 陳逄衡, Yi-zhoushu buzhu) or a variant for zhai 柴, a form of burnt-offering sacrifice (Gu Jiegang; for all of these interpretations, see Gu Jiegang, 1963: 6, n. 5.)

Interpreted correctly, these ritual actions on the part of King Wu can be seen to constitute an integral part of this section. It is not co-incidental that yu is used twice in the section, once in its temporal usage of "driving out" enemies, describing the Grand Duke Wang's mili-tary actions, and once in its spiritual usage of "driving out" inauspicious influences, the ritual responsibility of the king. Having thus spiritu-ally purified the area, the king would naturally set off to inspect the new territory.

9. The place-names indicated by the "Shifu" are Yue 越, Xifang 戲方 (perhaps Miji 靡集), Chen 陳, and Wei 衛. Qu Wanli (1965: 329-330) has identified these places with the following Chunqiu period locales: Yue, the southwestern part of the state of Wei 衛; Xi, in northeast-ern Zheng 鄭; Chen, the state by the same name to the west of Song 宋, and Wei, the state of the same name to the northeast of the former Shang capital. While I have been unable to cor-relate these identifications with evidence from oracle bone inscriptions, as would be prefer-able, they seem generally reliable.

72

The case of Wei, however, calls for some elaboration.

Historical sources record that Feng 封, the youngest brother of King Wu, was, at the time of the conquest, enfeoffed at Kang 康, and hence was known in early Zhou texts as Kang hou, the Archer-Lord of Kang. After quelling the rebellion of the Shang scion Wugeng 武庚 and his two Zhou overlords, Guan shu 管叔 and Cai shu 蔡叔, both of whom were also broth-ers of King Wu, King Cheng 成 ordered the Archer-Lord of Kang to oversee the Yin area, appointing him the Lord of Wei 衛. This appointment is clearly recorded in the inscription of the Kang hou gui 康侯𣪕: 王朿伐商邑, 征令康侯, 啚于衛 "It was when the king (i.e., King Cheng) attacked the city Shang; he ordered the Archer-Lord of Kang to be en-feoffed at Wei." (See Chen Mengjia, 1955: 25; Shi-rakawa Shizuka, 1964: 4.144-152.)

There are however two traditions regarding the location of Wei. The Hanshu, "Dili zhi" identifies it with Chaoge, the capital of the last Shang king, Di Xin: "Chaoge was the capital of Zhou (Xin), and the feudatory of King Wu of Zhou's younger brother Kang shu. Its name was changed to Wei." (Hanshu: 1554) On the other hand, both the Zuozhuan (Ding 4;Legge, CC V:754) and the "Wei Kangshu shijia" 衛康叔世家 chapter of the Shiji (1589) identify it with the "Waste of Yin" (Yinxu), now known as the capital dur-ing most of the late Shang. That the Kanghou gui and its various related vessels were discovered, about 1931, in northern Henan in the vicinity of Anyang, lends credence to this latter identification.

This passage of the "Shifu" further substanti-ates that identification. Chaoge having been captured immediately after the battle at Muye, it would be redundant to now, twenty days later, make it the target of a campaign. The passage makes perfect sense however, when Wei is seen as the former capital, and presumably still an important cult center, at Anyang, approximately 100 km to the north. The importance of such an objective would explain why the "Tiger Van-guard," the crack troops of the army, were explicitly ordered to make the attack. This consideration was probably also applicable in the choice of commanders for the mission. The general here ordered to lead the attack is Bai Ta 百弇. I believe that this is the same man who is said in Sec. 6 to have led the attack on the Xuanfang, although there the name is Bai Wei

百韋 (note the graphic similarity between 弇 and

韋, which combined with the identical first charac-ter leaves little doubt about this identification.) Moreover, both phonetic and graphic considerations lead me to identify this figure with Bai Da 百達, a deputy of King Wu who figures prominently in the post- conquest events narrated in the "Ke Yin" 克殷 chapter of the Yizhoushu (4.3b).

This is the only general mentioned in the

"Shifu," with the exception of the Grand Duke Wang (for whom there is a rich tradition; see e.g., Shiji, 1477-81; and Allan, 1972-73), who can be so identified in other texts, suggesting that he was of a higher stature than the other generals, and hence suitable for leading so important a campaign.

10. Emending ge 格 to ge 革 on the basis of Kong Zhao's commentary: 不改祭天之服 "He did not change his clothes for sacrificing to heaven."

11. Following Zhu Youceng 朱右曾 (Yizhoushu jixun jiaoshi 逸周書集訓校釋 and Gu Jiegang in add-ing huang yue 黃鉞 on the basis of parallel ex-amples in the following text; see Gu Jiegang, 1963: 9, n. 6.

12. Tai Wang, great-grandfather of King Wu, was the illustrious Gugong Danfu 古公亶父, the ruler who moved the Zhou people to the Qishan 𡵨山 area. Taibo was his eldest son and Yugong (i.e., Yuzhong

虞仲 ) his second son, both of whom acceded to Kangji (i.e., Jili 季歷), their younger half-brother and the father of the future King Wen, when Taiwang made known his wishes that Wen one day be king. Yikao was the eldest son of King Wen who pre-deceased his father, thus enabling King Wu, the next eldest son, to be the next king. (For this ac-count, see Shiji: 112-5.)

Gu Jiegang has pointed out that this form of collateral ancestor sacrifice, including not only the kingly ancestors but also other members of the royal family in a given generation, is akin to the Shang ancestor sacrifices displayed in the oracle bone inscriptions. He regards this as evidence of the authenticity of the "Shifu." (Gu Jiegang, 1963:9, n. 9; 29.)

13. The original text here reads gui-you 癸酉 (day 10), which,from the sequence of cyclical days, is obviously corrupt, and should read gui-chou 癸丑.

14. Following Gu Jiegang in adding ru 入 after wang 王, as in similar constructions below; 1963: 11, n. 20.

15. Following Lu Wenchao 盧文弨 (Yizhoushu) and Gu Jiegang in reading yong 庸 (i.e., yong 鏞) for qi 其; ibid. 11, n. 23.

16. Following Lu Wenchao in reading rong 戎 for wo 我. The usage of ye 謁 here is somewhat of a problem, with most commentators glossing it as:

告也 "to report." While a "report" is certainly a part of what ye means, it is a formal reporting given in person, usually to one's superior. I believe that the line should be interpreted to mean that the defeated Shang troops were drawn up

73

on the former battlefield of Muye, and that King Wu, their new sovereign and commander, there re-ceived a report of their condition, i.e., "in-spected" them. (For further remarks on this line, see below, n. 74.)

17. Zuozhuan (Xuan 12; Legge, CC V: 320) states

武王克商... 作武 "When King Wu had subdued Shang,... he composed 'Wu' (Shijing, Mao 285)." The identification of this poem seems relatively firm, but it is my feeling that any attempt to identify the other poems mentioned in the text would be mere conjecture.

18. Gu Jiegang interprets this line quite differ-ently, marshalling considerable evidence to show that wan 萬 was the name of a dance. (Gu Jiegang, 1963: 12, n. 29) He may be correct but I provi-sionally interpret the word in its typical sense.

19. Deleting zhong 鍾. With the reign of Han Jingdi 漢景帝 (-156–141), the word qi 啟 be-came taboo and this line must have been changed to read kai 開 at that time. Qi, of course, was the son of Yu the Great, the progenitor of the Xia dynasty.

20. To be identified with Bai Ta in Sec. IV, as noted above; see n. 8.

21. The original text here reads xin huang 新荒; I reverse the order on the basis of the reading in Sec. IV. It is of course possible that Xin Huang is the correct reading, in which case the first oc-currence should be emended.

22. I here retain the original sequence of the text. Gu Jiegang (1963: 14, n. 13), however, rear-ranges the text, thinking with previous commenta-tors that this campaign occurs in the same locale as that chronicled in Sec. IV. To the contrary, the states here enumerated lie far to the west, in the vicinity of the Zhou capital, and on the basis of the cyclical dates given, this campaign would had to have taken place in the fourth month, just days before the victory celebration in the Zhou capital.

Of the place-names in this section, Mo 磨, Xuanfang 宣方, Shu 蜀 and Li 厲, I have been unable to make any identification of Mo and Li. And despite its occurrence in Shang oracle bone inscriptions, I have also been unable to locate Xuanfang. (Qu Wanli (1965: 329-30) notes that the absence of this place-name in later literature makes its occurrence here strong evidence for the authenticity of the "Shifu.") But it is possible to determine that Xuanfang was an enemy of the Shang state; for example 綜圖 22.4: 甲申卜貞及亘方 "Crack on jia-shen, divining: 'Cap-

ture the Xuanfang.'"; suggesting at least that it was located outside of the sphere of direct Shang influence.

This suggestion is confirmed by the final place-name, Shu. A state of this name also occurs

in the Shang oracle bone inscriptions, but with evidence of having had a long-term vassalage rela-tionship with the Shang; (e.g., in Per. I the Shang divined about Shu's harvest [乙 5280; 6422], and in Per. IV performed divinations in Shu [庫方 993; 981]).

This vassalage relationship notwithstanding, Shima Kunio (1958: 378) has located Shu far to the west of Shang, just to the south of the great bend in the Yellow River. Shima's placement, in addi-tion to being firmly based on contemporary evidence, has recently received further verification from the Western Zhou oracle bones discovered in 1977 at Qishan, Shaanxi province. Inscription Hll: 68 of that find reads 伐蜀 (𢆶) "Attack Shu. (This)" (Wenwu 1979.10:40; a photograph of the shell is on. pl. 5, no. 4, and a drawing can be found on p. 43, diagram 11. See also, Xu Xitai, 1979: 189.) If these oracle bones were in fact inscribed by the Zhou during the time of kings Wen and Wu, this inscription demonstrates two things: first, Shu was indeed in the general vicinity of Zhou, and second, a short time before the Zhou conquest of Shang it was an enemy of the Zhou. Judging from this passage in the "Shifu," the adversary relationship presum-ably continued and was resolved only after the Zhou conquest of Shang.

(N.B. It is true that Shu is listed in the "Mushi" 牧誓 chapter of the Shangshu, as one of eight major allies in the Zhou conquest campaign. But the "Mushi" is a late composition, and when such evidence is contradicted by the "hard" in-scriptional sources, it must be discounted. It is interesting however, that this tradition confirms the location of Shu in the west near the Zhou home-land.)

23. The original text is here corrupt, reading 陳本命新荒蜀磨至. I follow the emendation of Gu Jiegang, 1963: 13, n. 7.

24. In the oracle bone inscriptions, there is a Per. V Shang vassal state called 靃, which is pre-sumably the same as this Huo 霍; see Shima, 1971: 234.1.

25. Ai 艾 is possibly the 乂 of the oracle bone inscriptions (since both graphs are read ai and are defined as a type of mugwort, the later elaboration of the "grass" signific would be natural), which 前 14 1.44.7: 貞王狩于乂. "Divining: 'The king will hunt in Ai'," shows to have been a territory under the control of the Shang king.

26. The number in the original text reads 八百有三百, an obvious corruption.

The anomalous usage here of yu 禦 as "chariot" is dictated by the qualifier liang 兩. In later usage yu commonly indicates "charioteer," which might suggest "pairs of charioteers" here,

74

33. I here interpret fa 伐 in its normal Shang and early Western Zhou sense of "to behead" (being the pictograph of an axe cutting through a man's neck), and you 右 as oracle bone inscriptional

㞢, 又, also a common Shang sacrifice term, which was later elaborated into you 侑 "to offer"; see further, K. Takashima, 1979-80: 52-3.

34. This reading was suggested by David Pankenier (1981: 19, n. 10). Shirakawa observes that the suburban sacrifice was originally a ceremony of initiation, either of a new reign or of a new city (1965: 6.299). Moreover, the Duke of Zhou's per-formance of the sacrifice in the "Shaogao" 召誥 chapter of the Shangshu shows that it was, in its earliest form, not necessarily presided over by the king himself.

35. Following Gu Jiegang in deleting wu wang 武王 (1963: 19, n. 9).

36. Parallels between the "Shifu" and the inscrip-tion attributed to the Xiao Yu ding 小盂鼎 (the vessel is no longer extant) have been noted by many scholars of bronze inscriptions. The Xiao Yu ding inscription also records a victory celebra-tion of sorts, and the language is so consistent, both in general and in technical points, with that of the "Shifu" that it merits extensive quotation.

At dawn, the King entered the Ancestral Temple of the Chou House, .. .. .. .. The Re-ceiver of Guests waited upon the guests of the States (the feudal lords from outside metro-politan Chou). They laid aside their travel-ling clothes (for ceremonial robes) and stood facing east. I, Yu, with all the belted and beflaqged (prisoners) of the Guifang .. .. .. entered the South Gate. I reported "The King commanded me, Yu,together with .. .. to attack the Guifang .. .. .. I took prisoner two of their Chiefs, obtained 4,812 scalps, captured 13,081 men, seized .. .. horses, 30 chariots of war, 355 oxen, and 38 sheep.

I, Yu, .. .. say .. .. .. called out "re-port our campaign." I took prisoner one of their Chiefs, obtained 237 scalps, captured .. .. men, seized 104 horses and 100 (plus ..) chariots of war."

The king .. said "We are pleased."

I, Yu, bowed deeply and saluted and, to-gether with the captured Chiefs, came forward and took my place in the Great Court.

The King commanded Rong (one of the Chiefs of the Guifang) ".. their .. .. .. Chiefs, examine its cause." .. The Earl of Ge .. .. Guifang, Guifang prior to this with the newly .., followed Shang."

The Chiefs were beheaded in the .. . The King called out ".. .. order Yu, with his prisoners and scalps, to come in at the (South) Gate and to present them at the West Walk. With .. to enter and sacrifice them by

especially in that this passage is otherwise an enumeration of human captives. However, there is no parallel for liang being used as a qualifier for humans, and I provisionally follow the Chinese commentators in reading "chariot."

27. The "Preface to the Documents" (Shuxu 書序) for the "Wucheng" chapter states: 武王伐商, 往伐歸獸 "King Wu attacked Shang; he went out and attacked, returned and hunted," one of the reasons for identifying the "Shifu" with the "Wucheng"; see below, p. 60.

Moreover, Qu Wanli (1965: 330) cites this hunt-ing record as one of the reasons he considers the text authentic. He points out that not only does the grammar resemble the hunting notations found on oracle bones, but what is more, the numbers of ani-mals recorded as being caught is consistent, species by species, with the Shang records.

Finally, in an interesting analysis of oracle bone inscriptional evidence, Yu Xingwu (1979: 275-7) has observed that hunting expeditions played an important function after victorious battles. My own study of the inscriptions concerning the Shang Period V campaign against the Renfang confirms that hunts were both frequent and seemingly significant (they being the only specified activity recorded; i.e., non-routine divination) during the return march from the victorious battle. In all these ways, this hunt notation in the "Shifu" accords well with contemporary practice.

28. In the original text this number begins 億有十萬; with yi 億 being defined in antiquity as shi wan

十萬 (100,000), the shi wan following is obviously corrupt. It should probably read qi wan 七萬, the archaic form of qi 十 easily giving rise to this sort of confusion.

29. On the basis of the lunar-phase notation, this date can be designated as the 16th day of the luna-tion. "Shengba," the first day of the second quar-ter, would be the eighth day of the lunation, "pang shengba" (旁, meaning "expanded") would be two days later, or the 10th, "jipang shengba" would be the day after pang, or the 11th, and six days later, counting inclusively, would be the 16th. I should like to mention here that my understanding of these lunar-phase notations owes much to the work of my teacher, David S. Nivison, on Western Zhou chronol-ogy; for a preliminary report of his research, see Nivison, 1980.

For a discussion of the significance of this date, see below, p. 65.

30. Adding kao 考 after wen 文 on the basis of the parallel text below.

31. Scribe Yi also figures in the "Ke Yin" chapter's account of post-conquest activities; Yizhoushu, 4.3b.

32. Emending shi 矢 to fu 夫 on the basis of the parallel usage in Sec. II.

75

fire in the Ancestral Temple of the Chou House."

(Tr. Dobson, 1962), 231–32)

The underlining here represents parallels with the "Shifu." As one can tell at a glance, whole sec-tions correspond in exact detail, offering another piece of evidence for the authenticity of the "Shifu."

37. Yue 龠 offerings are recorded in both Shang oracle bone and Western Zhou bronze inscriptions; e.g., the Chen Chen xiang 臣辰盉: 住王大龠于宗周 "it was when the king performed the great yue offering in Zongzhou." (Luo Zhenyu: 14. 12a) While the context in these inscriptions allows little insight into the nature of the offering, later texts are fairly consistent in defining it as an agricultural (i.e., vegetarian) offering. For ex-ample, the Baihu tong (p. 678) states: 夏曰禴者,

麥熟進之 "In the summer it is called the yue; when the wheat is ripe it is offered."

38. The original text reads 武王乃以庶祀馘于國周朝; the emendation is based on the "Wucheng" quotation in the Hanshu.

39. Reading here yu 于 for yu 於, this occurrence of yu 於 being anomalous in the text.

40. Adding yu 玉 after shu 庶, following Lu Wenchao, whose emendation is based on the Taiping yulan 太平御覽 text.

41. Reading yan 琰 for wu 五, as in the parallel phrase above.

42. Hanshu: 1015–16.

43. It is true that the "Wucheng" date for the battle at Muye reads "third month" (san yue 三月) rather than "second month"(er yue 二月), but this is universally considered a textual cor-ruption.

45. The guwen "Wucheng" states: 歸馬于華山之陽, 放牛于桃林之野, 示天下弗服 "He sent back his horses to the south of Mount Hua, and let loose his oxen in the open country of Taolin, show-ing the empire that he would not use them again." (Legge, CC III: 308)

46. Mencius 7B/3; Tr. Lau, 1970: 194 (modified). While it is clear that the violence of the Zhou conquest detailed in the "Wucheng" was anathema to Mencius' idealistic historiography, elsewhere he seems in fact to have used the "two or three strips" of the text he considered reliable to fill out his history of the conquest.

The Duke of Zhou helped King Wu to punish (the

Shang king) Zhou. He waged war on Yan for three years and punished its ruler; he drove Fei Lian to the edge of the sea and executed him. He extinguished fifty states. He drove tigers, rhinoceroses and elephants to the distant wilds, and the Empire rejoiced.

Mencius 3B/9; Tr. Lau, 1970: 113 (modified)

We have seen that King Wu's victorious hunting excursion was regarded as significant both in his own time (see above, n. 27) and also later (e.g., in the Shuxu), and Mencius here accepts it as his-torically valid. And it goes without saying that the "mopping up" campaigns against recalcitrant states mentioned here by Mencius comprise the major portion of the "Shifu" text. (It would seem however, that by linking the Duke of Zhou with King Wu, Mencius has here confused his history. Traditional accounts [e.g., Shiji: 132] associate the Duke of Zhou with the campaign against Wugeng, Guan shu and Cai shu, during the reign of King Cheng.)

47. The "Taishi" is a guwen chapter of the Shangshu; for a discussion of the anachronisms in the "Ke Yin," see Huang Peirong, 1976: 289–97.

48. Lunheng: 7.14b–16a; Tr. Forke, 1907: 482–84 (modified).

49. Cheng Tingzuo, Wanshu dingyi: 26.4a.

50. It appears however that the Han dynasty com-poser of the text masquerading under the title "Wucheng" also interpreted Mencius's characteriza-tion as a quotation, and in an attempt to camouflage his forgery, incorporated the phrase into his text.

This presentation of Warring States and Han reflections on the "Wucheng" has been adapted sub-stantially intact from the seminal study by Gu Jiegang, 1963: 24–27.

51. Huang Peirong, 1976: 141–236.

52. For a fuller discussion, see Huang Peirong, 1976: 300–302.

53. This also demonstrates that the traditional interpretation of this title, i.e., "The Suc-cessful Completion of the War" (Legge, CC III: 306, following the pseudo-Kong Anguo 孔安國 commentary, 文王受命, 有此功成於克商 "King Wen received the mandate; there was this military success, completed in the conquest of Shang," Shangshu, 6.8a), is incorrect. The title should be translated "(King) Wu's Achievement (of Rule)."

54. See the discussions by such Qing dynasty scholars as Wei Yuan, Shu guwei: 171.17a–b and Cheng Tingzuo (Wanshu dingyi: 26.4a). Modern scholars who have made linguistic studies of the text include Guo Moruo (1929: 269-71, Gu Jie-gang (1963: see esp. 28–29), and Qu Wanli (1965: 327–31).

55. This comparison is adopted substantially in-tact from Gu Jiegang, 1963: 6, no. 5.

76

56. See, for example, Chen Mengjia, 1956: 283; Guo Moruo, 1929: 270; and Xu Jinxiong, 1963: 8a.

57. For this inscription, see Guo Moruo, 1957: Inscription-174, Study-36ff.

58. Shirakawa Shizuka interprets ming here as a noun, being modified by dong 東 and the object of the verb ji 即. He interprets this to mean, "to respond to the Eastern order" (1968: 17.154-6). Despite the awkwardness of his reading, Shirakawa objects that ming is never used as a passive, hence it cannot be so used here. Guo Moruo, on the other hand, states flatly that this is a pas-sive use of ming (1957: Study-36).

59. See above, n. 44. Other scholars identifying yu fang as a place-name include Wang Guowei, Yang Shuda 楊樹達, and Chen Banghuai 陳邦懷; see Shirakawa Shizuka, 1971: 32.817–19, for a summary of their views.

60. Xu Jinxiong, 1963: 8a.

61. The Shuowen notes that yu 馭 (i.e., ) is the ancient form of 御: 馭吉文御 (Shuowen jiezi gulin: 1257a). The occurrence of both forms in this one inscription would suggest a differ-ence in nuance, but taking the ·opening para-graph as a sort of preface to the main body of the inscription, there seems to be no substantive distinction.

62. For a study of this inscription, including a detailed summary of the arguments on yu fang, see Shirakawa Shizuka, 1971: 32.814-840.

63. In his study of early archaic Chinese, W.A.C.H. Dobson notes that "sequence in months and days is indicated by yih (翌) 'next' (before days) or lai (來) 'come' (before months)" (Dob-son, 1962: 88). While this distinction may not be questioned on the basis of the linguistic sample to which Dobson restricted himself (Western Zhou materials), the following oracle bone usages show that this sense is derived from an earlier usage where yi marked short intervals and lai marked relatively longer intervals of time.

64. Dobson, 1962: 231.

65. Guo Moruo, 1957: Inscription-135-36, Study-146.

66. It is well known that a significant percentage of the corpses found at Anyang had been thus de-capitated; see Shih Chang-ju, 1959: 7-8, 297-99.

67. I suspect that this usage of you is also evi-dent in the "Shifu," in the phrase 伐右厥由甲小子鼎大師 "He beheaded and offered their jia little prince, the great master of the cauldron," but the reading of you 侑 for you 右, while adopted in the translation, is too uncertain to serve as evidence of authenticity. See further, above, n. 33.

68. For a discussion of these two general sacrifice verbs, see Yao Xiaosui, 1979: 381-82.

69. See, e.g., Shirakawa, 1965: 6.296. I should note here that vestiges of this technical usage continued well into the classical period, as the following passages from the Chunqiu, Zuozhuan and Du Yu's commentary demonstrate.

Chunqiu (Zhao) 昭, 11) : Winter, eleventh month, ding-you (day 34); the army of Chu vanquished Cai, captured You, the heir-apparent of Cai, and returned with him, "using" (yong) him.

Zuoihuan: Winter, eleventh month; the viscount of Chu vanquished Cai and "used" (yong) the heir-apparent on Mount Gang.

Du Yu: To "use" yong him is to kill him and sacrifice to the mountain.

(Chunqiu jingzhuan jijie, 22.8b)

For other examples of this usage, see Chunqiu, Xi 僖, Zuozhuan, Zhao 10, and Du Yu's respective commentaries (ibid. 6.3b, 22.6a), and Mozi 墨子 4.14b. I do not believe however that these later archaicisms detract from the contemporaneity of the "Shifu"'s usage.

70. The use of yu 于 in this line, as well as above in 告于天于稷 "I report to Heaven and to Ji" is also worthy of note. Compare, for example,

續 1.42.5: 毋又于且辛于毋辛 "Do not make an offering to Grandfather Xin and to Mother Xin" and 粹 32: 既燎于河于岳 "After performing a burnt offering sacrifice to the River (-spirit) and to the Mountain (-spirit)." Whether the second yu of these lines be interpreted as a re-duplicative preposition or as a conjunction,the parallel be-tween the "Shifu" and the oracle-bone usage remains striking.

71. See above, notes 3 and 28 respectively.

72. For this 47-day interval to be compatible with the lunar-phase notations of the two full-date notations, one would have to assume either two consecutive short (i.e., 29-day) months, or that jipang shengpo refers to the 10th day of the luna-tion instead of the 11th, and that six days later would then be the 15th instead of the 16th.

73. As, for example, Wang Guowei, who in his Sheng-ba siba kao referred only to the Hanshu "Wucheng" quotation. But it is interesting to note that Liu Xin himself, who after all was quoting the origi-nal "Wucheng," did not subscribe to this 47-day interval calendar. Based on his definitions for the lunar-phase notations (see above, n. 3), he dated the battle at Muye to the 5th day of the second month, and the sacrifice at Zhou to the 22nd day of the fourth month (fifth lunation), which was pre-ceded by an inter-calary month. In other words, he interpreted the text to require a 107-day interval.

74. Kong Guangsen has argued that both the cere-

77

monies of sec. V and those of sec. IX took place in the Zhou capital (Jingxue zhiyan: 11.8347).There are two compelling reason why this cannot be the case. However one may interpret the word ye 謁 of the phrase 謁戎殷于牧野 (Sec. V), it must be admitted that it is the only verb in the sentence, and being modified by the preposi-tional phrase yu Muye, the act of ye-ing must have taken place in Muye, just outside of the Shang capital.

Furthermore, it will be noted that in Sec. IX the captured nobles of the Shang are ex-ecuted on geng-xu (day 47). In Sec. V, they are presented to the king on gui-chou (day 50). If these two sections were indeed to be conflated, one would have to explain the rather distasteful presentation of four-day-old corpses to the Zhou king.

75. I mention this, because in addition to Kong Guangsen's proposal to conflate Sec. V with Sec. IX, Gu Jiegang has also proposed a re-arrangement of the text, joining the campaign of Sec. VI with that of Sec. IV (1963: 22-23), thus eliminating any actions between the ceremonies in the Shang capital and those in the Zhou capital. As argued above (n. 22), this is based on a mis-perception of the geographical location of the states named in Sec. VI, Mo, Xuanfang, Shu, and Li, which in fact lie in the vicinity of the Zhou homeland. The attack against these states would naturally have occurred during the return of the Zhou army from the Shang to the Zhou homeland, and the se-quence of the original text is confirmed.

76. See above, n. 3.

77. Note especially that the ceremonies recorded in the Xiao Yu ding inscription take place on ji wang 既望, the 16th day of a lunation. See above, n. 42, and also the discussion in Dobson, 1962: 231–33.

78. Gu Jiegang, 1963: 31. Gu does discuss at some length the very interesting solutions of Kong Guangsen and Chen Yiwang 陳綱. Both of these proposals are ultimately damned, however, because of their reliance on Liu Xin's definitions of the lunar-phase notations.

79. The "Zhou yue" comprises chapter 51 of the Yizhoushu; this passage can be found at 6.2a. For a detailed discussion of the dating of this chapter, see Huang Peirong, 1972: 1-38.

80. According to the san-tong theory, the first month of the Shang year was the chou 丑 month (i.e., the month after the month containing the winter solstice) while the first month of the Zhou year was the zi 子 month (that containing the winter solstice.) But Huang Peirong (1972: 76 overleaf) has organized the calendrical studies of Shinjo Shinzo 新城新藏 on the Chunqiu into the form of a graph, demonstrating that the change in calendar described by the san-tong theory did not take place before about -630.

81. See Dong Zuobin, 1960: I.124.

82. Nivison, 1980.

83. Guo yu 3.22b–23a. Nivison, 1980: 4.

84. See for example, the "Preface to the Docu-ments" (Legge, CC III: 6–7), and Shiji: 121.

85. Wenwu 1977.8:1-12.

86. For a discussion of the relationship between Jian and Shang, see Akatsuka Kiyoshi, 1977: 665, and 138ff.

87. Inscriptional evidence suggests Shang prec-edents for this type of post=-conquest behavior. Consider, for example, the following period V bone inscriptions:

綜圖 16.2: Minor Vassal Qiang partici-pated in the attack, netting (X number of men of) Wei, 24 men of X, 1,570 men of Er, XX (men of) Fan, .. 2 chariots, 183 shields, 50 auxiliary shields (?), and (X number of) arrows. Earl Wen was sacrificed to Da Yi, Earl Pi was sacrificed..., (Earl) Fan (was sacrificed) to Ancestor Yi, Mei was sacrificed to Ancestor Ding.

哲庵 315: Crack on bing-wu (day 43), divin-ing at You: "The king should perhaps call out to .. to bring the manacled leader of the Renfang .., and make a burnt-offering of .. .. captives." In the first month, when (we) are coming (from) the campaign against the Renfang.

Although unfortunately incomplete, both of these pieces allow interesting insights into the be-havior of victors in warfare at the time of the Zhou conquest. 綜圖 16.2 records the capture of 1,694+ prisoners in the course of a battle, but, significantly, only the leaders of the four enemy states were sacrificed after the battle. (For a full discussion of the development of this custom, see Yao Xiaosui, 1979: 385–90.) Likewise, in 哲庵 315, it is apparently only the leader of the Renfang, an enemy of the Shang, who is brought forward to face execution. The place of execution, You 攸, is also noteworthy, for the corpus of inscriptions concerning the campaign against these Renfang (see Dong Zuobin, 1945: Xia 9.48a–63b) shows that You was an ally of the Shang in the campaign and served as a sort of forward base for the Shang army. Thus, the return to You before executing the leader of the Renfang would be comparable to King Wu's return to Zhou with the Shang nobles before executing them.

WORKS CITED

(All citations to classical sources refer to the Sibu congkan 四部叢刋 edition unless other wise noted.)

78

Akatsuka Kiyoshi 赤塚忠1977 Chūgoku kodai no shūkyō to bunka--In

ōchō no saishi 中國古代の宗教と文化————殷王朝の祭祀. Tokyo.

Allan, Sarah1972-73 "The Identities of Taigong Wang 太公

望 in Zhou and Han Literature." Monu-menta Serica 30 (1972-73):57-99.

Baihu tong shuzheng 白虎通疏證 Zhongguo zixue mingzhu jicheng 中國

子學名著集成 (facsimile of Huai nan shuju 淮南書局, 1875 edition.)

Chen Mengjia 陳夢家1955 "Xizhou tongqi duandai" 西周銅器

斷代. Kaogu xuebao 考古學報 1955. 9:137-175.

1956 Yinxu buci zongshu 殷虛卜辭述 Peking.

Chen Pengheng 陳逄衡1825 Yizhoushu buzhu 逸周書補注.

Cheng Tingzuo 程廷袏 Wanshu dingyi 晚書訂疑. Huang Qing

jingjie xubian 皇清解續編, juan 26.

Ding Fubao 丁福保1930 Shuowen jiezi gulin 說文解字詁林.

Shanghai.

Dobson, W.A.C.H.1962 Early Archaic Chinese. Toronto.

Dong Zuobin 董作賓1945 Yinli pu 殷曆譜. 2 vo1s. Nanqi,

Sichuan.

1960 Zhongguo nianli zongpu 中國年曆總譜 (Chronological Tables of Chinese History). Hong Kong.

Forke, Alfred1907 Lun-Heng: Wang Ch'ung's Essays. 2 vols.

Leipzig.

Gao Heng 高亨1947 ZhouYi gujing jinzhu 周易古經今注.

Shanghai.

Gu Jiegang 顧頡剛1963 "Yizhoushu 'Shifu pian' jiaozhu, xieding

yu pinglun" 逸周書世俘篇較注寫定與評論. Wenshi 文史 2:1-42.

Guo Moruo 郭沫若1931 Zhongguo gudai shehui yanjiu 中國古

代社會研究. Peking.1957 LiangZhou jinwenci daxi tulu kao shi

兩周金文辭大系圖录攷釋. Peking.

Hanshu 漢書.1962 Zhonghua shuju 中華書局 edition.

Peking.

Huang Peirong 黃沛榮1972 Zhoushu Zhouyue pian zhucheng de shi-

dai ji youguan sanzheng wenti de yanjiu

周書周月篇著成的時代及有關三正問題的研究. Taipei.

1976 Zhoushu yanjiu 周書研究. Doctoral dis-sertation. National Taiwan University.

Kong Guangsen 孔廣森Jingxue zhiyan 經學卮言. HuangQing jingjie

皇清經解 11.834-83.

Lau, D.C.1970 Mencius. Middlesex, England.

Legge, James1865 The Chinese Classics, Volume III: The

Shoo King or the Book of Historical Documents: Hong Kong and London.

1872 The Chinese Classics, Volume V: The Ch'un Ts'ew with the Tso Chuen. Hong Kong and London.

Li Xiaoding 李孝定1965 Jiagu wenzi jishi 甲骨文字集釋. 8

vols. Nangang, Taiwan.

Lu Wenchao 盧文弨Yizhoushu 逸周書. Baojing tang congshu 抱

經堂叢書.

Luo Zhenyu 羅振玉1936 Sandai jijin wencun 三代吉金文存.

Nivison, David S.1980 "The Dates of Western Zhou." Mimeo-

graphed. Symposium on the Bronze Age of China, Metropolitan Museum of Art, New York. Rev. Dec. 6, 1980.

Pankenier, David1981 "The Chronology of the Conquest Year."

Mimeographed. Jan. 15, 1981.

Ping Xin 平心.1979 "'Bao you ming' xinshi"' 保卣銘新

釋. Zhong hua wenshi luncong "中華文史論, 1: 49-79.

Qu Wanli 屈萬里1965 "Du Zhoushu 'Shifu' pian" 讀周書世俘

79

篇. Symposium in Honor of Dr. Li Chi on his Seventieth Birthday. Taipei.

Shih Chang-ju 石章如 (Shi Zhangru)1959 Xiaotun: yizhi de faxian yu fajue, yi-

bian 小屯遺址的發現與發掘乙編 Academia Sinica, Institute of History and Philology. Nangang, Taiwan.

Shiji 史記.1959 Zhonghua shuju 中華書局 edition.

Peking.

Shima Kunio 島邦男1958 Inkyo bokuji kenkyū 殷墟ト辞硏究.

Hirosaki.

1971 Inkyo bokuji sōrui 殷墟卜辭綜類. 2nd rev. ed. Tokyo.

Shirakawa Shizuka 白川靜1962ff. Kimbun tsushaku 金文通釋. 51 vols.

to date. Tokyo.

Shuowen jiezi gulinSee Ding Fubao.

Takashima, K.1979-80 "Some Philological Notes to Sources

of Shang History." Early China 5:48-55.

Wang Guowei 王國維1923 "Shengba siba kao" 生霸死霸考.

Guantang jilin 觀堂集林 Wucheng.

Wei Yuan 魏源 Shu guwei 書古微. Huangqing jingjie

xubian: juan 171.

Xu Jinxiong 許進雄1965 "Shi yu" 釋御. Zhongguo wenzi 中國

文字 12.

Xu Xitai 徐錫台1979 "Zhouyuan chutu de jiaguwen suojian

renming, guanming, fangguo, diming qianshi" 周原出土的甲骨文所見人名官名方國地名淺釋. Guwenzi yanjiu 古文字研究 1: 184-202.

Yao Xiaosui 姚孝遂1979 "Shangdai de fulu" 商代的俘虜.

Guwenzi yanjiu 1:337-390.

Yu Xingwu 于省悟1979 Jiagu wenzi shilin 甲骨文字釋林

Peking.