University of Virginia Library

Appendix II

THE HAN DYNASTY'S EARLIER CALENDAR

In ancient times, several calendars were used in China. Months were
always counted from new moon to new moon, a month occupying the
time of a lunation, but the year did not always begin at the same period.
For astronomical purposes, months were numbered by the twelve horary
characters, beginning with the month which normally contains the
winter solstice. The calendar anciently used in the feudal state of
Chin, said to be that of the legendary Hsia dynasty, put the first month
in the third [OMITTED] astronomical month. The Yin calendar, used in the
state of Sung, began the year one month earlier, putting New Year's
day in the second [OMITTED] astronomical month. The royal calendar of the
Chou dynasty began the year one month earlier, New Year's day coming
in the first astronomical [OMITTED] month. (Cf. H. Maspero, La Chine antique,
pp. 222-223.) The Ch'in dynasty put New Year's day one month
farther forward, into the twelfth [OMITTED] astronomical month. The early
Han rulers continued this practise, until, in 104 B.C., the Han Emperor
Wu put New Year's day back to the period it occupied in the Hsia
calendar, in the third [OMITTED] astronomical month, where it stayed (with
unimportant exceptions) until the time of the Chinese Republic. (For a
fuller account, cf. Havret, in the T'oung Pao, vol. 8, p. 399.)

The SC and HS were both written after the Han calendar reform
in 104 B.C., but they record events occurring at a time when the year
began three months earlier than at the time they were writing. Which
calendar did they use? What names did they give to the months, those
in use in later Han or in Ch'in times?

The true answer to the above questions is a very peculiar one: the
Ch'in dynasty kept the same names (numbers) for the months as those
used in Hsia (and later Han) times, but merely shifted the date of New
Year's day and the attendant official ceremonies. That is, the Ch'in
dynasty made New Year's day occur in the month which they called the
"tenth" month, so that the month which they called the "first month"
was the fourth in order from the beginning of the official year! The Han
Emperor Wu merely restored New Year's day to the beginning of what
he and his predecessors had been calling the "first" [OMITTED] month, but which
month had not previously begun the year.

The existence of such an anomalous calendar as one beginning with
the "tenth" month is so unnatural that only quite conclusive evidence

should make us accept it. This evidence is even more important because
Yen Shih-ku (581-645 A.D.), the outstanding commentator on the
HS, contradicts our view—he said that the Ch'in dynasty had really
called the month with which their year began the "first" month, and
numbered the months consecutively beginning with that one; and that
the authors of the SC and HS had changed the names of the months to
correspond to the different names given them in Han times after the
calendrical reform. The evidence for the contrary view is presented
below:

I. In the first place, there is ample evidence from the SC, HS, and
other documents to show that in Ch'in and early Han times the year did
actually begin with what the historians called the "tenth" month and
that the "first" month was the fourth in the order of the months. Since
exactly three months are reckoned to a season, the Ch'in calendar then
made the year begin with the first month of autumn and made the
"first" month begin the spring season. The pertinent points in the
following passages are accordingly: the order of the months in the year,
the seasons at which certain months come, and the month and season in
which official ceremonies occurred. Italics are mine.

a. In chap. 5 of the SC we read (cf. Mh. II, 91): "In the 48th year of
King Chao-hsiang, in the tenth month, [the state of] Hanh offered [to
Ch'in the city of] Yüan-yung. The army of Ch'in was divided into three
armies. . . . In the first month, the troops were disbanded." Note that in
the same year, the "first" month comes after the "tenth" month.

b. In the monthly tables of the SC (chap. 16; cf. Mh III, 59) during
the second year of the Second Emperor, the months are enumerated
beginning with the tenth month, next the eleventh and twelfth months,
and then only the first month. It is here called the [OMITTED], for [OMITTED] was
tabooed, since it was the personal name of the First Emperor.

c. In chap. 8 of the SC (cf. Mh II, 393) we read, "In the tenth
month of the tenth year [of Kao-tsu], Ch'ing Pu, King of Huai-nan,
P'eng Yüeh, King of Liang, Lu Wan, King of Yen, Liu Chia, King of
Ching, Liu Chiao, King of Ch'u, Liu Fei, King of Ch'i, Wu Jei, King of
Ch'ang-sha, all came to court at the Ch'ang-lo Palace. In the spring and
summer, nothing [of note] happened. In the seventh month, the Grand
Emperor died at the Yo-yang Palace." The great court reception at the
beginning of the official year then occurred in the tenth month, which
was in winter, for spring came afterwards.

d. In the HS, 4: 15b f, we read, "In the spring of the fifteenth year, a
yellow dragon appeared at Ch'eng-chi. The Emperor issued an edict
ordering a discussion of a sacrifice in the suburbs. . . . In the fourth month,
in the summer, the emperor visited Yung and then first sacrificed to the

Five Emperors." Since the SC says that in ancient times the sacrifice
to the Five Emperors was always made in the summer, the fourth month
came in summer; hence the first month came in the spring.

e. In HS 6: 6b we read, "In the fourth year [of the period Yüan-kuang
(131 B.C.)] Tou-yin, the Marquis of Wei-chi, who had committed
a crime, was executed publicly. In the spring, the third month, on the
day yi-mao, the Lieutenant Chancellor [T'ien] Fen died." But the SC
107: 12b, says, "On the last days of the twelfth month, [the Marquis of
Wei-chi] was sentenced to be publicly executed in the market-place in
Wei-ch'eng. In that spring the Marquis of Wu-an [the chancellor] died
of illness." Then the execution occurred before the spring, according
to the HS, and in the twelfth month, according to the SC; the death
occurred in the spring and in the third month.

f. In SC 16: 9b, 10a, b, 12a we read, "[In the first year of the state of
Hans], in the first month, Hsiang Yu divided Kuan-chung. . . . The
second month was the beginning of the [reign of] the King of Hans, the
former Lord of P'ei. . . . In the third month he made Nan-cheng his
capital." Now in 22: 1b we read, "In the first year of the Emperor
Kao-tsu, in the Spring, the Lord of P'ei became the King of Hans and
went to Nan-ch'eng." Thus the first, second, and third months came
in spring.

g. In chap. 25 of the SC, the explanation of the twelve musical tubes
begins with the tenth month (cf. Mh III, 303), "Among the sonorous
tubes, [it, i.e. the tenth month, corresponds to] the Yin-chung." The
explanation proceeds month by month, and ends with the ninth month
(cf. ibid. 313), "Among the musical tubes, [it, i.e. the ninth month,
corresponds to] the Wu-yi." Thus the order of the tubes follows that
of the official year, from the tenth to the ninth month.

h. In HS 90: 7b, 8a, b we read, "When Wang Wen-shu . . . became
the administrator of Ho-nei, . . . by the end of the twelfth month, in all
the commandery there was not a thief to make any dogs bark. Those
few [thieves] whom he did not catch and who fled to neighboring commanderies,
he pursued. Meanwhile spring had come. [Wang] Wen-shu
stamped his feet and sighed, `Alas! If the winter months should be prolonged
one more month, it would be sufficient [for my task].' . . . The
Emperor considered him an able [person] and promoted him to be Palace
Military Commander." Now according to HS 19B: 18b, he was appointed
Palace Military Commander in 119 B.C., 15 years before the
reform of the calendar. In his time, the twelfth month came in winter.

i. In the Book of Rites, chap. 17, p. 4a we find the following, "In the
third month of autumn [OMITTED] . . , it was [ordered] that the nobles should
unify their practises and customs and that [the officials of] all the

districts [should come] to receive [instructions for their government] at
[the grand reception] on the first day of the month in the coming year."
The commentary tells us that this passage refers to Ch'in practises.
Since the great court reception was held on New Year's day, that festival
came in winter.

j. In the HS 1B: 17a we read, "In the eleventh year . . . the second
month, an imperial edict said, . . . `The vassal kings and marquises
shall regularily pay court and make offerings in the tenth month." The
reference is again to the grand court reception on New Year's day, here
stated to occur in the tenth month.

k. In HS 4: 6b we read, In the third month of the first year, "an
edict said, `Now it is the time of spring, when [nature is] harmonious,
and the plants and trees are all growing, when things all have means
of enjoying themselves." Then the third month was in spring.
But cf. p. 236, n. 4.

l. In HS 6:26a we read, In the first year of the period Yüan-feng
(110 B.C.) "an imperial edict read, `Let the tenth month [begin] the first
year of the [period] Yüan-feng.' " Hence the year began with the
tenth month. (Altho the word `begin' is not actually in the text, the
passage plainly implies it.)

m. In chap. 15 of the SC we read, "In the thirty-seventh year, in the
tenth month, on the day kuei-ch'ou [Nov. 1, 211 B.C.], the First Emperor
went out on a trip. [cf. Mh II, 184] . . . [In the seventh month] . . .
on the ping-yin day, the First Emperor died in the P'in terrace at
Sha-ch'iu . . . In the ninth month the First Emperor was buried in Mt.
Li" (cf. ibid. 193). In Chap. 15 of the SC (Chavannes did not translate
this passage), it reads, "The first year of the Second Emperor. In
the tenth month, on the day wu-yin [an edict was issued ordering] a
general freeing of criminals. In the eleventh month, he made the Rabbit
park. In the twelfth month, he went to the O-fang Palace. In the
ninth month of that year the commanderies and districts all rebelled."
Thus the historians began a year with the tenth month and ended it
with the ninth month. There are many such passages.

n. In the HS 4: 9a (cf. Mh II, 461), we read, "In the eleventh
month [of the second year], on the day kuei-mao [Jan. 2, 180 B.C.] the
last day of the month, there was an eclipse [Oppolzer's no. 2447]. The
imperial edict read: . . . `Since on the last day of the eleventh month there
was an eclipse—a reproach that was seen in the heavens—how great
must the calamity be!' " In this passage too events in the tenth month
are recorded as preceding this one and events in the first month follow it.

o. The SC 96: 5a (HS 42: 5a), tells that when Chang Ts'ang was

Lieutenant Chancellor, he advised that because Kao-tsu arrived at Pa-shang
(cf. HS 1A: 19b) in the tenth month, and was consequently considered
to have overthrown the Ch'in dynasty in that month, the date
of New Year's day should not be changed from the date set by the Ch'in
dynasty. Thus the conquest of Ch'in was commemorated by continuing
New Year's day on the date of his conquest. In the same chapter we
read, "It was ordered that all the kings and marquises should always
appear at court and make presentations in the tenth month." The same
statement appears in SC chap. 99 and HS chap. 43. Hence the early
Han tenth month was the same as the Ch'in tenth month. An interesting
confirmation is found in the HHS, chap. 14, where it says that
on the first day of each month and at the beginning of the year a
great court was held at which presents and congratulations were received
(from the nobles); the officials (however) congratulated (with presents)
in the first month. In chap. 16 of the HHS it says "The reason that
of all the first days of the months, only on the first day of the tenth month
did they follow the former custom, was because in that month Kao-tsu
subjugated the Ch'in dynasty and began the first year of his reign."

p. In the SC and the HS (before 104 B.C.) the intercalary month
is always called the "later ninth month" [OMITTED]. (Since twelve lunations
do not make quite a solar year, every two or three years an extra, intercalary
month was added). After the calendrical reform in 104 B.C.,
the intercalary month was inserted at various times of the year to keep
the seasons occurring in the proper months. The only adequate reason
for the intercalary month always previously coming after the ninth
month is that thus it was put at the end of the year.

II. The foregoing passages amply prove that in Ch'in and early Han
times the year began with what was later called the "tenth" month.
But did the historians change the names of the months, as Yen Shih-ku
said they did? We have already had evidence that such was not the
case: passages j, k, l, and o quote imperial edicts which fix the months in
the same seasons as those they later occurred in. The cyclical characters
in passages e, m, and n enable us to check the dates; for, with a sixty day
cycle, the same characters would not reappear in a month that came three
months later. There is also the evidence furnished by the following
passages:

q. In HS 6:31b we read, "In the fifth month, in the summer, [in the
first year of the (period) T'ai-ch'u (104 B.C.)], the calendar was corrected,
making the first [OMITTED] month begin the year." If the months
had previously been numbered from the beginning of the official year,
the record should have been different, something like the following:

The calendar was corrected, making the fourth month the first month.
The wording of the HS shows that previously the "first" month did not
begin the year.

r. The great scholar Chia Yi, who lived 200-168 B.C. (before the
calendrical reform) wrote a poem which is reproduced in his biography
in the SC chap. 84 and HS chap. 48. In that poem the word for
summer occurs in the rime, so that the historian could not have changed
it, and it is coupled with the cyclical characters for the day: "In the year
Shan-o [OMITTED] in the fourth month, in the first month of summer, on the
day keng-tzu, when the sun was setting, an owl perched in my house."
The naming of the year as the fourth in the twelve year cycle enables us
to identify it as 174 B.C. If the months were numbered beginning with
the astronomical twelfth month (in which New Year's day then occurred),
the fourth month would have been the first month of spring, not summer.
The cyclical character also enables us to identify the month, for we know
the cyclical character for the day of the calendrical reform, 19 years later,
and a simple calculation (cf. Chinese Social and Political Science Review,
vol. 18, p. 166) enables us to determine the characters for the days of
each month in the year the poem refers to. The day keng-tzu could not
have come in the fourth month after New Year's day of that year;
but it could have come in the seventh month after New Year's day,
which Chia Yi called the "fourth" month. Then the "fourth" month
occurred in summer and contemporary writers numbered the month in
which New Year's day came as the "tenth" month.

s. Liu An, King of Huai-nan, committed suicide in 123-2 B.C., before
the reform of the calendar. In his chapter on astronomy he writes (I
confess that I do not altogether understand the passage), "The cycle of
the universe begins with the first month [OMITTED] which is the third astronomical
month [OMITTED] when the sun and moon have both entered five
degrees into the [constellation] Ying-shih [α, β Pegasus]." Here he says
plainly that the "first" month is the third astronomical month, not the
twelfth astronomical month, with which the Ch'in and early Han dynasty
began the year.

t. In the same book he tells that Mercury appears near the constellations
Andromeda and Aries at the spring equinox in the second month,
near Gemini and Cancer at the summer solstice in the fifth month, near
Virgo at the autumn equinox in the eighth month, and near Sagitarious
and Capricornus at the winter solstice in the eleventh month. Elsewhere
in the book he gives the positions of the sun among the constellations
for the twelve months of the year. Allowing for the precession of
the equinoxes, those positions are the same as for the months called by

the same names in Ch'ing times, altho the book was written before the
calendar reform.

u. A stone inscription known as [OMITTED] found on
a hill near Han-tan, has inscribed on it the date ping-yin in the
eighth month of the twenty-second year of the kingdom of Chao (B.C.
158). This is a contemporary record made before the correction of the
calendar. If the months had been counted beginning with New Year's
day, the eighth month could not have contained a ping-yin day at all.

There is thus ample proof that the Ch'in and early Han dynasties
used a curious calendar in which New Year's day and the official celebrations
connected therewith came in what they called the "tenth"
month, and that the reform in 104 B.C. did not change the names of the
months, but merely shifted the date for New Year's day, altho a court
celebration was continued to be held on the first days of the tenth
month, because that date commemorated the founding of the dynasty.
Hence the SC and the HS use the same names for the months as those
used in Ch'in and Han times, which were the same as those in use in the
time their authors wrote.

The foregoing evidence has been mostly collected by Wang Yin-chih
(1766-1834; Giles no. 2252); his famous reply to Yen Shih-ku is transcribed
in Wang Hsien-ch'ien's Ch'ien-Han-shu Pu-chu, chap. 1, pt. A,
pp. 23-26. Wang Hsien-ch'ien himself added material; further significant
material is found in a paper by Chen Chin-sien, "The Anomalous
Calendars of the Ch'in and Han Dynasties" in the Chinese Social and
Political Science Review for July 1934, vol. 18, p. 157 ff.