Ancient Chinese Calendar: Dynasty Reforms and Astronomers

Contents

The Xia Xiao Zheng, a Chinese astronomical text from roughly 2000 BCE, contains the earliest surviving record of monthly observations of the Big Dipper used as a seasonal indicator. Oracle bone inscriptions from the Shang dynasty (around 1300 to 1046 BCE) already use the sexagenary day-count cycle that modern Chinese calendars still rely on. Across the next 3000 years, Chinese astronomers built and rebuilt the lunisolar calendar through more than a hundred dynastic reforms, named after the eras that issued them and signed by named astronomers whose work survives in the Twenty-Four Histories.

This article traces that chronology dynasty by dynasty, naming the astronomers and the calendars they wrote, the observatories they built, and the moments when foreign methods entered the imperial system. For the mechanics of how the modern lunisolar system functions, see our Chinese lunar calendar guide. For the zodiac symbolism layered on top of the calendar, see our Chinese zodiac signs overview, and for the broader political context behind these reforms, our ancient China timeline sets each calendar against its dynasty.

The Earliest Records: Xia Xiao Zheng and the Oracle Bones

The Xia Xiao Zheng (Lesser Annuary of Xia) is the oldest surviving Chinese astronomical text. Preserved in the Da Dai Liji compilation, it lists monthly indicators tied to the rotation of the Big Dipper handle: when the handle points east at dusk, spring has begun; when it points south, summer; west, autumn; north, winter. The text describes plant phenology, bird migrations, and insect behaviour month by month, anchoring agricultural work to celestial observation.

Shang dynasty oracle bone inscriptions, dated to around 1300 to 1046 BCE, push the record further. Diviners burned tortoise shells and ox scapulae to read cracks, and the questions and answers they carved onto the bones include lunar eclipses, planetary positions, and the sexagenary day-count. The Shang sexagenary cycle already runs Jia Zi, Yi Chou, Bing Yin, and so on through 60 distinct day labels, identical to the modern sequence. The same oracle bones constitute the earliest known stage of Chinese script, a topic our history of Chinese writing traces in detail.

Two Shang oracle bone records describe solar eclipses, one of which has been astronomically dated to 1226 BCE. These are among the oldest written eclipse observations from any civilisation. Inscriptions also report sightings of supernovae and comets, with the 1054 CE Crab Nebula supernova later recorded in Song dynasty observations using the same continuous tradition.

Pre-Imperial Calendars: Xia, Shang, Zhou, and Qin

Pre-imperial Chinese dynasties each set their own start-of-year month, an act with strong political symbolism: declaring a new calendar meant declaring a new mandate from heaven. The succession of pre-imperial start-month conventions was:

Xia (around 2070 to 1600 BCE): year began on the first day of the first lunar month, the convention the modern lunisolar calendar still uses
Shang (around 1600 to 1046 BCE): year shifted to the twelfth month of the Xia system
Zhou (around 1046 to 256 BCE): year moved to the eleventh month, anchoring the calendar on the winter solstice
Qin (221 to 206 BCE): year moved again to the tenth month, a deliberate political break from earlier traditions
Early Han (206 BCE onward): retained the Qin start-month for nearly a century before the Taichu reform restored the Xia convention

The Qin used the Zhuanxu calendar, named after the legendary emperor Zhuanxu. The system survived the brief Qin dynasty and ran into the early Han, where it was used for almost a century before Emperor Wu of Han ordered a comprehensive reform that restored the first-month start and consolidated regional variants into a single imperial standard.

The Taichu Reform of 104 BCE

Emperor Wu of Han issued the Taichu calendar in 104 BCE, the first systematic Chinese calendar built on documented mathematical rules. The reform team was led by the astronomer Tang Du and the calendrical mathematician Luoxia Hong, who designed the underlying calculations. The historian Sima Qian documented the reform in the Shi Ji but did not author the mathematics, a point Chinese historical sources are explicit about and that English-language summaries often miss.

The Taichu calendar fixed three things that the modern system still uses: the year begins on the first day of the first lunar month, the 19-year Metonic cycle of seven leap months governs leap month placement, and the winter solstice anchors the eleventh lunar month. The reform team measured the synodic month at 29.53086 days, within seconds of the modern value of 29.53059 days. The reform also formalised the pairing of Wu Xing elements with Heavenly Stems and the twelve-branch hour system covered in our Chinese zodiac elements and Chinese zodiac hours guides.

The Taichu calendar held for over 180 years before the Eastern Han issued the Sifen (Quarter Remainder) calendar in 85 CE. The Sifen was the first Chinese calendar formally promulgated to the public in printed almanacs, distributed annually through the empire’s prefectural offices. It also introduced the concept of subdividing the 24 solar terms into 72 pentads (hou), each labelled by a seasonal sign such as the arrival of geese or the blooming of peach trees.

Tang Dynasty Innovation: Yi Xing and the Meridian Expedition

The Tang dynasty produced the most important calendrical reform between the Han and the Yuan. The astronomer-monk Yi Xing (683 to 727 CE), born Zhang Sui and ordained as a Buddhist monk at age 21, was commissioned by Emperor Xuanzong to reform the calendar in 721 CE.

Yi Xing’s preparation included the first organised geodetic expedition in world history. In 724 CE, he dispatched imperial astronomers to 13 observation sites stretching from Jiaozhou in modern northern Vietnam (latitude 17 degrees North) to a station immediately south of Lake Baikal (latitude 50 degrees North). At each site, teams measured the noon shadow of a standardised gnomon and the altitude of the celestial pole. The data established the length of one degree of meridian arc and produced the first numerical evidence that the Earth’s surface curves.

The expedition’s central observation site was Yangcheng, near modern Dengfeng in Henan, where a stone Shadow Measuring Platform was erected for direct gnomon work. Yi Xing’s measurement of a meridian degree predated the equivalent European survey by Jean Picard in 1669 to 1670 by 945 years.

The reformed calendar, called Da Yan (Great Expansion), was published in 52 volumes in 728 CE, one year after Yi Xing’s death. Japanese envoys carried the Da Yan calendar back to Heian Japan in 763 CE, where it remained in use for centuries. Korean kingdoms and Vietnamese courts adopted it through the same channels.

Song Dynasty Precision: Yang Zhongfu and the Tongtian Calendar

The Southern Song astronomer Yang Zhongfu issued the Tongtian (Unity with Heaven) calendar in 1199 CE, during the fifth year of the Qingyuan era. His measurement of the tropical year came to 365.2425 days, identical to the value the Gregorian reform would adopt in Europe 383 years later.

Yang Zhongfu also recognised that the tropical year is not a constant. His writings note that the length of the year was slightly longer in antiquity and slightly shorter in his own time, an observation now understood as the secular variation in the Earth’s orbital motion. No European astronomer would document the same effect for another six centuries.

The Tongtian calendar was used by the Southern Song court for 76 years until the Yuan conquest. The Yuan astronomers absorbed Tongtian’s precision into the next major reform, building on Yang Zhongfu’s measurements rather than restarting from earlier values.

Yuan Dynasty Peak: Guo Shoujing and the Gaocheng Observatory

The Shoushi (Season-Granting) calendar of 1281 CE was the work of the Yuan astronomer Guo Shoujing (1231 to 1316), assisted by Wang Xun (1235 to 1281), under direct patronage from Kublai Khan. Guo’s tropical year of 365.2425 days matched Yang Zhongfu’s earlier value, but the underlying observational programme behind Shoushi was on a scale no previous Chinese reform had matched.

Guo coordinated the Si Hai Ce Yan, the Survey of the Four Seas, in 1279 to 1280. Twenty-seven observation stations were established across the Yuan empire, from Pyongyang in Korea in the east to Samarkand in Central Asia in the west, and from the Gobi steppe in the north to Hainan island in the south. Each station ran simultaneous gnomon and pole-altitude observations for a year, producing a calibrated grid of geodetic data.

The central observatory of the network was built in 1276 at Gaocheng, near modern Dengfeng in Henan, on the same site Yi Xing had used five centuries earlier. The brick-and-stone structure stands today: a 12.62 metre tall observation tower with a 31.19 metre horizontal stone scale on the ground for measuring the projected shadow of a horizontal gnomon at the tower’s top. The setup gave Guo’s team gnomon readings accurate to within 2 millimetres on the scale. The Yuan court funded the survey from imperial revenues, an economic context our piece on ancient China economy covers in the broader picture of imperial scientific patronage.

Mount Songshan, rising directly behind the Gaocheng tower, was considered the centre of heaven and earth in classical Chinese cosmology, which is why both Yi Xing and Guo Shoujing chose the location. UNESCO inscribed the Gaocheng Observatory as part of the Historic Monuments of Dengfeng World Heritage Site in 2010. The observatory remains the oldest surviving stone-and-brick astronomical structure in China.

Guo Shoujing’s Shoushi calendar held a year-length error under 26 seconds, the most accurate calendar produced anywhere before the European Enlightenment. The Ming dynasty kept it in service as the Datong (Great Unification) calendar with only minor adjustments for 363 years, from 1368 to 1645.

Late Imperial Reforms: The Shixian Calendar and the Jesuit Encounter

The Shixian (Conforming to the Times) calendar of 1645 CE was the first Chinese calendar built with European methods. The Jesuit astronomer Adam Schall von Bell, who had arrived in Beijing in 1623, led a reform team under the new Qing dynasty’s Shunzhi Emperor. Schall replaced the mean solar terms used in earlier calendars with true solar terms calculated from the sun’s actual position on its elliptical orbit, a concept later codified in modern national standards.

The reform triggered the Kangxi Calendar Case of 1664 to 1669. The Chinese astronomer Yang Guangxian accused Schall of using foreign methods and predicting an auspicious burial date that brought misfortune to the imperial family. The Qing regents sentenced Schall to death, then commuted the sentence after an earthquake during the trial. The young Kangxi Emperor reviewed the case in 1669 and arranged a public astronomical test in which Schall’s Jesuit successor Ferdinand Verbiest correctly predicted a meridian transit while Yang’s team failed by twenty minutes. Kangxi reinstated the Shixian methods and appointed Verbiest as director of the Beijing Observatory.

The Shixian calendar served the Qing dynasty for 267 years and forms the direct ancestor of the modern computational standard. Each iteration since 1645 has refined the precision but kept Schall’s true-position framework intact.

From Imperial Standard to National Standard

The Republic of China abolished the imperial Shixian calendar on 1 January 1912 and adopted the Gregorian calendar for government use. The traditional lunisolar calendar was not banned, only stripped of state authority. Almanacs continued to be printed by private publishers, especially the Tung Shing in southern China and Hong Kong.

The Purple Mountain Observatory in Nanjing assumed responsibility for calculating the official annual lunisolar calendar in 1966 and has held that role ever since. Its publication Zhongguo Tianwen Nianli (Chinese Astronomical Yearbook) is the authoritative reference. The observatory’s New Perpetual Calendar covers conversions from 1840 to 2050.

The Standardisation Administration of China issued national standard GB/T 33661-2017 in May 2017, the first time the lunisolar calendar has been codified as a technical specification. The standard requires astronomical computation to one-second precision, formally completing the move from imperial mandate to scientific specification that Schall began in 1645.

Frequently Asked Questions

How old is the ancient Chinese calendar?

The earliest extant astronomical text, the Xia Xiao Zheng, dates from roughly 2000 BCE. Shang oracle bones from around 1300 BCE record the sexagenary day-cycle still used today, and one Shang inscription describes a solar eclipse dated by modern astronomers to 1226 BCE. The Taichu calendar of 104 BCE marked the first systematic codification with documented mathematical rules.

Who invented the Chinese calendar?

No single inventor exists. The system evolved across many dynasties. The Han-era Taichu reform of 104 BCE was led by the astronomer Tang Du and the mathematician Luoxia Hong; the historian Sima Qian documented their work but did not write the math. The Tang-era Da Yan reform of 728 CE was the work of the monk Yi Xing. The Yuan-era Shoushi reform of 1281 CE was led by Guo Shoujing.

What was the most accurate ancient Chinese calendar?

The Shoushi calendar of 1281 CE, produced by Guo Shoujing under the Yuan dynasty, held a year-length error under 26 seconds and gave a tropical year of 365.2425 days, identical to the Gregorian value Europe would adopt three centuries later. Yang Zhongfu’s Tongtian calendar of 1199 had reached the same value 82 years before Guo, but on a smaller observational base.

Is the Gaocheng Observatory still standing?

Yes. The brick-and-stone observatory built by Guo Shoujing in 1276 at Gaocheng near Dengfeng in Henan stands today. The 12.62 metre observation tower and the 31.19 metre stone ground scale are intact. UNESCO inscribed the site as part of the Historic Monuments of Dengfeng World Heritage in 2010. The site is open to visitors and remains the oldest surviving stone astronomical structure in China.

When did China stop using the ancient calendar officially?

The Republic of China switched to the Gregorian calendar for official use on 1 January 1912. The Shixian lunisolar calendar that the Qing dynasty had used since 1645 ceased to be the state standard, but private almanacs and household use of the lunisolar system continued without interruption. The Purple Mountain Observatory has published the official annual lunisolar calendar since 1966, and national standard GB/T 33661-2017 has governed its computation since 2017.

Ancient Chinese Calendar