Part II
8.4.1 Confirmation of 1040, #1: Oracle inscriptions for Di Xin's campaign against the Ren Fang (or Yi Fang), in his 10th and 11th years, fit the years 1077-76, and these years only. (As I explained in Chapter Five, these inscriptions taken together require (1) that the "10th year" end with day jiawu (31), yiwei (32), or bingshen (33); and that this year had a "9th month" followed by a lunar month lacking a "qi-center," therefore being intercalary. 1077 is the only possibility.) Therefore Di Xin 1 = 1086. But the Bamboo Annals date for Di Xin 1 is 1102, 16 years earlier. The best explanation of this is that at some time in early Warring States the de facto first year of Zhou 1040, as Wu Wang 1, was confused with the de jure first year 1056, which also came to be seen as Wu Wang 1. This would make Di Xin's 46th year appear to be 1057 rather than 1041. This 16-year shift is seen in earlier Shang reigns in the Annals; e.g., Wu Yi's reign, actually 1145/43-1109,is made to be 1159-1125; and Zu Geng's reign, actually 1188-1178, is made to be 11 + 16 years earlier. (It was supposed that his 11 years must precede Zu Jia's 33 years, instead of being the first 11 years of Zu Jia's "33 years." See Chapter Four.)
8.4.2. Confirmation of 1040, #2: Oracle inscriptions show that Di Yi 1 = 1105. (See Chapter Six. The "8th year" inscription set consisting of Ku-fang 1661 and other fragments requires an intercalary 3rd month, and it fits only the year 1098, a date confirmed by several other inscriptions.) A frequently quoted or paraphrased line from an unknown source, sometimes said to be the Da Dai li ji, says that "Wen Wang in (the/his) 15th year produced Wu Wang." The meaning probably is the 15th year of Di Yi, i.e., 1091. The Bamboo Annals says that Wu Wang died in his 54th year, which would have to be 1038. If this was two years after the Conquest, the Conquest date must be 1040.
8.4.3. Confirmation of 1040, #3: In the Bamboo Annals the enfeoffment of Cheng Wang's younger brother Yu as lord of Tang, later Jin, is dated to 1035. The compilers must have believed that Jupiter was in station Da Huo, Great Fire, in that year, which was 3 x 12 years after the conjunction, said to be in Fang (the middle lunar lodge of Da Huo) and in 1071. The Guoyu, "Jin Yu" 4, says explicitly that when Jin began Jupiter was in Great Fire. But Jupiter actually was in Great Fire in 1043 and 1031. If in 1043, Wu Wang was still alive (even if the Conquest was in 1945; he is said to have died at the very end of the year of his death); and the story of the event in the Shiji shows that it was Cheng Wang's action, when he was still a minor but de jure king, hence during the Regency. So it must have been in 1031. This is confirmed by Zuo zhuan, Xi Gong 15.14, which implies that the shy Shang prince Ji Zi, enfeoffed in Korea, was present at the ceremony. Ji Zi almost never came to court, but would have attended the great convocation recorded in the Annals in the summer of the last Regency year, the probable occasion of Tang-shu's enfeoffment. This can only be so if the Regency was 1037-1031, and the Conquest was in 1040.
8.4.4. Confirmation of 1040, #4: The "Lu Shijia" in the Shiji says that Wu Wang granted Lu to Zhou Gong right after the Conquest, and that Zhou Gong gave it to his son Bo Qin soon after Cheng Wang's succession. Only then does the account take up the revolt of Lu Fu and the royal uncles. Liu Xin as quoted in the Han shu says that Bo Qin's reign in Lu was 46 years; and that it began in the first year of Cheng Wang's 30-year calendar (i.e., for me, his accession calendar). The Shiji, "Lu Shijia," has a chronology of Lu dukes that implies that Bo Qin's death was in 999; so his first year would be 1044 (the first year of Cheng Wang's 37 years, in the Annals), but this is impossible. The difficulty is resolved by careful analysis of the Bamboo Annals, which reveals that Bo Qin's actual death date must have been 990. (The date given, 989, is due to distortions in the text, generated by (1) mourning intervals; and (2) the displaced three-year slip. By (1), Bo Qin's first year was backed from 1035 to 1037; and by (2) his whole tenure was at first moved down three years., making his death date 990 + 2 – 3 = 989. Later, most other 3-year displacements of dates were reversed, but Bo Qin's death date was not. See Chapter Twelve.) The trouble can be traced to the reign of Lu duke Xian Gong, who died in 856. The Shiji gives him a reign of 32 years; the Annals account implies a reign of 23 years. Thus Cheng Wang's dates are 1037/35-1006, 2 + 30 years; therefore the Conquest was in 1040.
8.4.5. Confirmation of 1040, #5: Yi Zhou shu 45, "Wu Jing" (Wu Wang warned) begins, "12th year, 4th month. The king reported a dream. on day bingchen (53) it was divined...." (We are supposed to understand that this dream was an omen portending the king's imminent death.) "An order then was given for Dan, Duke of Zhou, to appoint the successor, and to give Prince Song the text (of the order), and a copy of the "Bao Dian" (Treasured Document)." When a date is thus incomplete, normally the first of the month is meant, which should be yimao (52). If the 12th year is 1038, the 4th month begins with yimao if one supposes that the day counted as winter solstice was two days late. Shang oracle inscriptions suggest that this was the practice, i.e., the autumn equinox day was determined by observation, and the interval to the winter solstice (89 days) was assumed to be 91 days. The "Bao Dian," which is Yi Zhou shu 29, opens with a complete date: "It was the King's 3rd cult-year, 2nd month, day bingchen (53), first of the month...." This should be the same year, for if the 2nd month began with bingchen (53), the 4th month (30 + 29 days later) should begin with yimao (52). The only year that could be "the King's 3rd year" and also, in another calendar (counting from the year following his father's death) his 12th year, is 1038, and 1038 only if the Conquest was in 1040.
8.4.6. Confirmation of 1040, #6: The "Luo Gao" chapter of the Shang shu ends with an account of a sacrifice by Cheng Wang, called cheng ji sui, on day wuchen (5), said in the pseudo- Kong Anguo commentary to be the last day of the month; and the "Luo Gao" dates the event in the 12th month of the 7th year of the Regency. Two dates must be considered: 1036 and 1031. 1036 requires that the 12th month be the second month before the solstice month (possibly followed by an intercalary 12th month), the date being 27 November. 1031 requires that the date be 31 December, and that the next month begin one day earlier than indicated in Zhang Peiyu's tables (the syzygy was at 03:20, so this is likely). The solstice was on dingmao (4), 30 December, and one may assume as before that a 91-day count from the equinox specified jisi (6), 1 January. Thus if 1031 was the date, Cheng Wang thought he was sacrificing on the eve of the solstice, giving a plausible meaning for ji sui, "sacrifice for [the end of] the year," as well as a more satisfactory meaning for cheng, supposed to be a "winter sacrifice for royal ancestors." In 1036, the date could not be at the end of the year, unless the next year were taken as beginning with the pre-solstice month, making the date the last day of autumn, rather than any date in "winter." So 1031 seems to be the date, requiring that the Regency be 1037-1031, and that the Conquest be in 1040.
8.4.7. Confirmation of 1040, #7: The fact that the astrological
text in the Guoyu can be best explained (as above, 8.1(b)) as calculated
in the early fifth century BC, during the last years of Confucius' life,
by someone who believed the Conquest to have been in 1040 BC, is itself
a confirmation that 1040 is probably the correct date.
8.4.8. Confirmation of 1040, #8: Pankenier (1983 p. 241) has noted another text giving the location of Jupiter at the time of the launching of the Conquest campaign. A commentary to the "Ju Xiao" chapter of Xunzi says that at that time Jupiter was in the "north." Chun Huo would be due south, on an astrologer's chart. Having explained "Chun Huo" as a calculator's error, one must consider the alternative "north." If the campaign started in January of 1040, as would be required if one interpret the "Zhou Benji" data as in the Xia calendar, one finds that at that time Jupiter was in lunar lodge Xu, the middle lodge of station Xuan Xiao, due "north" on an astrologer's chart.
8.4.9. Confirmation of 1040, #9: Zheng Xuan
(see Kong Yingda's commentary to the Odes of Bin in the Shi jing) said
that Cheng Wang was born in Wu Wang's succession year. We should
expect that Zhou Gong would return full royal authority to the young king
effective in the king's 20th year. Thus Cheng Wang was born in 1049,
was 13 sui at his succession in 1037, and was 20 sui in 1030. The
Regency therefore was 1037-1031, and this requires that the Conquest was
in 1040. (Zheng Xuan does not derive his date of Cheng Wang's birth
from his date for Cheng Wang's assuming full power, which he puts in Cheng
Wang's 22nd year; and his absolute dates are quite different; so this is
independent evidence.)
8.4.10. Confirmation of 1040, #10: As Liu Xin quotes the
"Wu Cheng" chapter of the Shang shu (in Ban Gu's Han shu, 21B), the
Conquest campaign is said to start in the first month. But if I take
the Shiji to be using a source that dated the events in the Xia calendar,
that source must have made the campaign begin in the 11th month, because
the Yellow River had been crossed by day wuwu in the 12th month.
As Liu quotes it, the "Wu Cheng" continues: "yue ruo lai er yue ... jiazi
..." This is normally misinterpreted by supposing a break (or punctuation)
after "lai." But "lai er yue" is a single phrase, and it seems to
mean "the second month in the next year." Compare, for example, "lai
dingmao," in the "Shi Fu" chapter (#37) of Yi Zhou shu, which has to mean
(though this too is usually misinterpreted) "on day dingmao (4) of the
next month," i.e., the 3rd month, since jiazi, the victory day, was the
next to last day of the 2nd month. (In oracle inscriptions, "lai
dingmao" would normally mean "day dingmao in the next xun," or in a future
xun.) This shows that the "Wu Cheng" text is a modification of a
text that originally did have the campaign beginning in the Xia 11th month.
This is just what my analysis requires, if the date is 1040.
8.4.11. Confirmation of 1040, #11: One might
expect that astrologically weighted days would be preferred for the performance
of important ceremonies or the inauguration of great events: the first
day of the year, for major appointments; full moon day, for a holocaust
sacrifice; the first day of a month and/or season, for the start of a campaign.
Similarly, the first days of the twenty-four weather periods could be expected
to be thus favored. If one starts with the true winter solstice day,
and counts off the weather periods (qi jie) taking that day as the first
day of Dong Zhi, the major events of the Conquest campaign fall on such
days, if the year was 1045. But if one supposes that the true autumn
equinox was taken as Qiu Fen day (making the oberved winter solstice day
two days late), then it is 1040 that satisfies this test:
Qiu Fen
2 Oct 1041 JD 134 1473 +16
Han Lu
18 Oct
1489 15
Shuang Jiang 2 Nov
1504 15
Li Dong
17 Nov
1519 15
Xiao Xue
2 Dec
1534 15
Da Xue
17 Dec
1549 15
Dong Zhi
1 Jan 1040 1564
16
Xiao Han 17
Jan
1580 15 Guisi (30)
Da Han
1 Feb
1595 15
Li Chun
16 Feb
1610 15
Yu Shui
3 Mar
1625 15
Jing Zhi 18
Mar
1640 15
Chun Fen
2 Apr
1655 16
Qing Ming 18 Apr
1671 15 Jiazi (1)
Gu Yu
3 May
1686 15
Li Xia
18 May
1701 16
Xiao Man
3 Jun
1717 15 Gengxu (47)
Mang Zhong 18 Jun
1732 15
Xia Zhi
3 Jul
1747 16
Xiao Shu 19
Jul
1763 15
Da Shu
3 Aug
1778 15
Li Qiu
18 Aug
1793 15
Chu Shu
2 Sep
1808 15
Bai Lu
17 Sep
1823 15
Qiu Fen
2 Oct
1838 16
The key events:
Zhou month 1 / Xia month 11, day guisi (30) = 17 January,
Campaign begins, first day of Xiao Han (Lesser Cold)
Zhou month 4 / Xia month 2, day jiazi (1) = 18 April,
Victory at Muye, first day of Qing Ming (Clear Brightness)
Zhou month 6 / Xia month 4, day gengxu (47) = 3 June,
Celebration in Zhou on full moon, first day of Xiao Man (Grain Ripening)
The sequence of solar periods (qi jie) here follows Huainanzi, "Tian Wen," 12th paragraph. Ideally all periods are 15 days, but there must be five more days in a normal year; so the text says that there are 46 days from Dong Zhi through Da Han, from Chun Fen through Ku Yu, from Li Xia through Mang Zhong, from Xia Zhi through Da Shu, and from Qiu Fen through Shuang Jiang. I interpret this as meaning in effect that the days Dong Zhi, Chun Fen, Li Xia, Xia Zhi, and Qiu Fen are extra days, so that beginning with each of these one counts 16 days.
8.4.11.1 Lunar months as given by Zhang Peiyu for 1040 begin on the following ganzhi; at three points, I propose months beginning one day later, to avoid having two short (29-day) months in sequence, and two pairs of long (30-day) months in the same year. (In each case, the syzygy given by Zhang is late in the 24-hour day.) I give Zhang's data first, with my regularizations below:
(a) 1040: 05 34 03 33 02 32 01 31 01 30 60 30
04 03 02
Xia month 11 12 1 2 3 4 5 6 7 8 9 10
Zhou month 1 2 3
4 5 6 7
8 9 10 11 12
The major events occur in Xia months 11, 2, and 4. The "Wu Cheng," as quoted by Liu Xin in Han shu 21B, says that the victory day jiazi (in Xia month 2) is five days (inclusive) after jisibo. Therefore I take jisibo (Jsi below, beginning the fourth "quarter") as the 24th, in a month following a long (30-day) month, and as the 25th, in a month following a short (29-day) month (as does Xia month 2). The victory day is the 29th, next-to-last (the month is a long month).
In the following diagram, fei (F) is new moon
day, standardized as the 2nd after a long (30-day) month and the 3rd after
a short (29-day) month. The day zaishengbo (ZSh) is six days
(inclusive) later, as in the introductory narrative at the beginning of
the "Shao Gao" combined with the one at the beginning of the "Kang Gao"
(8.1 (1983) (4) above). Jishengbo (JSh) must be the next day, and
jiwang (JW), full moon day, is standardized as always nine days later (see
below).
(b) Xia months 11 to 4, 23 December 1041 through 17 June 1040
(first days of qi periods are underlined)
Month 11 Month 12
Month 1 Month 2
Month 3 Month 4
1 (05) 23 Dec (34) 21 Jan (04) 20 Feb
(33) 21 Mar (03) 20 Apr (32) 19 May
2 (06) Fei (35)
(05) F (34)
(04) F (33)
3 (07)
(36) F (06)
(35) F (05)
(34) F
4 (08)
(37) (07)
(36) (06)
(35)
5 (09)
(38) (08)
(37) (07)
(36)
6 (10) ZSh (39)
(09) ZSh (38)
(08) ZSh (37)
7 (11) JSh (40) ZSh
(10) JSh (39) ZSh (09)
JSh (38) ZSh
8 (12)
(41) JSh (11)
(40) JSh (10)
(39) JSh
9 (13) PSh (42)
(12) PSh (41)
(11) PSh (40)
10 (14) Dong Zhi(43) PSh (13)
(42) PSh (12)
(41) PSh
11 (15) (44)
JPSh (14)
(43) JPSh (13)
(42) JPSh 1
12 (16) (45)
(15) (44)
(14) (43)
2
13 (17) (46)
(16) (45) Chun Fen(15)
(44) 3
14 (18) (47)
(17) (46)
(16) (45)
4
15 (19) (48)
(18) (47)
(17) (46)
5
16 (20) Jiwang (49)
(19) JW (48)
(18) JW (47)
6
17 (21) (50)
JW (20)
(49) JW (19)
(48) JW
18 (22) (51)
(21) (50)
(20) (49)
19 (23) (52)
(22) (51)
(21) (50)
20 (24) (53)
(23) (52)
(22) (51)
21 (25) (54)
(24) (53)
(23) (52)
22 (26) (55)
(25) (54)
(24) (53)
23 (27) (56)
(26) (55)
(25) (54)
24 (28) JSi (57)
(27) JSi (56)
(26) JSi (55)
25 (29) PSi (58) JSi
(28) PSi (57) JSi 1 (27) PSi
(56) JSi
26 (30) Start (59)
(29) (58)
2 (28) (57)
27 (31) (60)
Li Chun (30) (59)
3 (29) (58)
28 (32) (01)
(31) (60)
4 (30) (59)
29 (33) (02)
(32) (01) Muye 5
(31) Li Xia (60)
30
(03)
(02)
(01)
One must reconcile several accounts, some of them garbled by passing
through hands of persons who did not understand the dates. The "Wu
Cheng" as quoted by Liu Xin says that Wu Wang started from his capital
on guisi (30), the day after pangsibo (PSi), in the "1st month."
Using the "Zhou Benji," we can correct this to "11th month." Yi Zhou
shu, "Shi Fu," dates this action to dingwei (44), the day after pangshengbo
(PSh), in the "1st month." Here "1st month" together with the statement
that Wu Wang was beginning his march repeats the quoted "Wu Cheng" error;
but the rest of this date is meaningful, because Liu (not giving his source)
says that on bingwu (43) Wu Wang "rejoined his army" (or reached his army),
which must have been already in the field ahead of him; and my analysis
shows that pangshengbo in my (i.e., Xia) "12th month" was in fact day bingwu.
This can be known from more of what Liu Xin quotes from the "Wu Cheng":
"In the 4th month, six days (inclusive) after jipangshengbo (JPSh), day
gengxu (47), Wu Wang offered a holocaust in the Zhou ancestral temple,"
i.e., after the victory and back in his capital. Jipangshengbo, "the
expansion of the birth of the (bo =) gibbous moon having occurred," ought
to be the next day after pangshengbo. (The fact that gengxu (47)
in month 4 is dated on jipangshengbo shows that in a long month the 16th
is the last day of the second quarter, which is thus nine days long.
I assume that in the 2nd "quarter," which is nine days, "pang" is two days
after "ji," whereas in the "4th quarter," only six days long, it is the
next day in a short month, but absent in a long month, as is required by
the date of the victory at Muye, "five days (inclusive) after jisibo" near
the end of the "2nd month.") The "Shi Fu" confirms that the 4th month
begins with yiwei (32), because it starts "4th month, yiwei day: Wu Wang
had completed his mastery of the world," etc.; and there can be no reason
for "yiwei" other than to date the beginning of the 4th month, in which
the victory rites were performed.
8.4.11.2 But the victory was on Qing Ming day. This is confirmed by the last line of the "Da Ming" ode in the "Da Ya" part of the Shi jing (Ode #236). That ode narrates Heaven's favor to Zhou through Wang Ji, Wen Wang and Wu Wang, down to Wu Wang's victory. The last line reads, "Si fa da shang, hui chao qing ming." The meaning has escaped all translators and commentators known to me: the line says, "He (Wu Wang) then attacked Great Shang; this occurred in the morning, Qing Ming (Day)." Thus the "Da Ming" ode is a Qing Ming Day hymn.
(It is true that names of other qi periods are not known in texts as early as this. But Qing Ming, the occasion of annual sacrifices to ancestors, is likely to be a very ancient name. And the existence, even earlier than this, of the concept of qi periods (however named) can be demonstrated by an analysis of the earliest known system of 28 lunar xiu. See my article in Aveni 1989.)
8.4.12 Confirmation of 1040, #12: Professor D. N.
Keightley has attempted to estimate the date of the Zhou Conquest by using
an interesting probability argument, in his book Sources of Shang History
(1978), and he reached exactly the dates that I am defending (p. 175):
Estimating an average of 20 years per reign, he counts back from 841 (the
first year of the Gong He regency), getting 1041. Then, counting
forward from 1180, which is the last year he gives Wu Ding (I would have
said 1189), and not counting Feng Xin (nor would I), he gets 1040.
I am arguing that the campaign began in late 1041 (in the Chinese calendar,
i.e. 17 January 1040), and was concluded on 18 April, 1040.
8.5 The victory was apparently not decisive.
The battle was bloody, with great numbers of "ears taken"; the Shang king
Di Xin was captured and killed. But Wu Wang found it necessary to
confirm Di Xin's son Lu Fu as the new Shang king, known as Wu Geng, in
reality dividing authority with him, while at the same time naming three
of his own brothers as "overseers" of the remaining Shang realm.
Shang dependent lords were accepted as Zhou vassals. Much too soon
after this, late in 1038, Wu Wang suddenly died. His brother Dan,
"Duke of Zhou," assumed authority as regent. Dan persuaded another
brother, Shi (Shao Gong) to support him; but the other brothers abroad,
suspecting usurpation, rebelled, joining forces with Wu Geng. Dan's
suppression of the rebellion in two years, defeating and executing Wu Geng,
was the real conquest. After more military consolidation, a great
assembly of lords was convoked in the summer of 1031, and at the end of
the year Dan surrendered his authority to the young king, Cheng Wang, who
was now of age, at 20 sui.
8.5.1 I add here a resume of the main events, with my dates, of the century ending in Cheng Wang's assuming full royal authority in 1030. Annals dates are on the left, my own on the right:
Birth of Chang (= Wen Wang)
1128
Death of Dan Fu
1139
1127
Ji Li (= Wang Ji)
1138
1126
Zhou victory
1135
1123
Zhou victory
1129
1117
Ji Li at court, given gifts
1125
1113
Zhou defeat
1123
1111
Zhou victory
1121
1109
Wenwu Ding 1
1124
1118, 1108
Zhou victory
1118
1106
Di Yi reign, year 1
1111
1105
Zhou victory
1114
1102
Probable Di Yi accession celebration
1102
Ji Li at court, imprisoned, dies 1114
1102
Wen Wang succeeds
1113
1101
War between Zhou and Shang
1110 ?
1101 ?
Wenwu Ding dies
1112
?
Wen Wang accession year
1099
Earthquake in Zhou capital
1109
1093
Di Xin 1
1102
1086
Ren Fang campaign
1077-1076
Lords assemble in Zhou
1082
1070
Royal hunt in Wei valley
1081
1069
New Shang calendar, probable court celebration
1068
Wen Wang at court, imprisoned 1080
1068
Lunar eclipse; Wen Wang writes sons
1065
Wen Wang released
1074
1062
Conjunction
1071
1059
"Mandate"; Zhou defeats Mi
1070
1058
Zhou victories, Ji, Yu, Cong 1069
1057
Zhou capital moved to Feng
1068
1056
Zhou promulgates royal calendar
1056
Lords assemble in Zhou capital 1067
1055
Shang campaign against Eastern Yi
1055-54 ?
Zhou plans new capital at Hao 1067
1055
Wen Wang dies, age 79 sui
1062
1050
Wu Wang succeeds
1061
1049
Cheng Wang born
1049
Zhou show of force at Mengjin 1052
1045 ?
Ji Zi imprisoned in Shang
1052
1044 ?
Di Xin kills Bi Gan
1052
1043 ?
Wei Zi flees from Shang
1052
1042 ?
Zhou campaign begins
1051
1041
Zhou victory at Muye
1050
1040
Lu Fu made Shang king by Wu Wang 1050
1040
"Yin" overseers appointed
1050
1040
Lords received, enfeoffed
1049
1039
Wu Wang ill
1048
1038
Wu Wang dies, age 54 sui
1045
1038
Cheng Wang succeeds; Regency 1044
1037
Lu Fu (Wu Geng), overseers rebel 1044
1037
Cheng Wang accession
1035
Rebellion put down
1042
1035
Lords assemble in Zhou capital 1038
1031
Tang-shu enfeoffed
1035
1031
Zhou Gong resigns regency
1038
1031
Cheng Wang, 20 sui, has full power 1037
1030