Read an Excerpt

Chapter Three
Astronomy: Sky Watchers and More China

Chinese astronomy resembles most other premodern sky technologies in that it was driven by divination. Yet Chinese astronomy differed from all others. It was run solely by a government bureaucracy and based on a worldview that said the ruler was "emperor under all heaven" -- a divine appointment. Yet the connection between celestial events and human fate was perhaps even more profound. The link was not just between heavenly deities and the emperor; the earth, the emperor, and the entire cosmos were bound together in one gigantic entity, a superorganism in which the five elements, or "phases" -- fire, air, wood, earth, and water -- were in constant interaction as they sought their affinities with one another.

Yet in China as elsewhere, portent astrology called for careful and regular observations of celestial events. The cosmic importance of every omen in the sky demanded that its results be noted down in detail. As a consequence, the Chinese possess the longest unbroken run of astronomical records in the world, observations of considerable importance to modern astronomers, whose research requires data about long-term celestial events.

China developed astronomy very early in its history. Evidence goes back five thousand years. The ancients wrote stars-laden texts in many forms -- on wine jugs, tortoise shells, and silk. The earliest records from archaeological sites in Qinghai Province consist of ceramic fragments on which are painted images of rayed sun disks and moon crescents. A piece of bone found to be thirty-five hundred years old contains writing showing that the Chinese already knew the length of the year to be 3651Ž4 days. There is evidence of star observation from before the twenty-first century B.C.186

The first recorded astronomical inscriptions date from the sixteenth to nineteenth centuries B.C. in the Shang kingdom of Henan Province. These artifacts are examples of an astronomical-divination system, technically called scapulimancy, a technique going back to Neolithic times. Selecting an ox or deer shoulder blade (scapula) or a tortoise shell, the diviners then dried, polished, and drilled the material with holes. They inserted a hot metal brand into one hole and examined the pattern of resulting cracks in the bone or shell. The diviner noted both the prognostication and later results on the cracked material.

The oracle bones' existence was lost to the modern world until 1899, when a scholar from Peking became ill and sent his valet to a drugshop for medicines. One ingredient in the potion the pharmacist sent him was labeled "dragon's bones." The scholar realized it was bone chips with words inscribed on them in ancient Chinese -- oracle bones.

During the following decades the bones were traced to a field near An-yang, around three hundred miles southwest of Beijing. During the 1920s and '30s, some twenty-five thousand oracle bones were excavated there, from what may have been a palace archive. At least 135,000 more pieces have been excavated since, forming a treasury of information going back to Shang times. This vast library recorded on the bone texts has enabled modern historians of astronomy to backtrack regularly occurring celestial events with computers to match sky phenomena inscribed millennia ago.

Recently, NASA astronomers used fourteenth-century B.C. oracle bones to help determine how much the earth's rotation is slowing down. Based on analysis of the tortoiseshell inscriptions, Kevin Pang and his colleagues at the Jet Propulsion Laboratory at Pasadena reported they had fixed the exact date and path of a solar eclipse seen in China in 1302 B.C. That, in turn, led them to calculate that the length of each day was 47Ž1,000 of a second shorter in 1302 B.C. than it is today.

A cache of five thousand pieces of oracle bones excavated in An-yang in 1972 yielded a series of divinations of sky events. The Chinese astronomical historian Zhang Peiyu found that six dates recorded in the inscriptions matched perfectly with a series of solar eclipses visible from the Henan area in the twelfth century B.C., half a millennium earlier than records of such events obtained from Babylonia or Egypt. Other Shang bones yielded inscriptions of lunar eclipses.

A reconstruction of another bone recording from around the same time revealed the observation of a supernova. The supernova inscription, perhaps the most ancient extant record of a nova sighting, says, in part, "On the seventh day of the month...a great new star appeared in company with Antares." The Chinese called these supernovas "guest stars." Thus the Chinese knew well what they were observing when, in June 1054 (A.D.), a star in the constellation Taurus blew. Chinese sky watchers reported it to be as bright as Venus, apparent during daylight, and visible for twenty-three days. The remnant of this explosion can be seen today and is called the Crab Nebula. (The Greeks have no record of the supernova.) Experts today have compiled detailed descriptions of supernova explosions that coincide with contemporary X-ray and radio sources.

In the Greek-influenced West, the sun and heavens were supposed to be immaculate. But Chinese astronomers saw spots on the sun. The earliest surviving record of a sunspot observation is by the astronomer Kan Te in the fourth century B.C. Kan Te assumed that these spots were eclipses that began at the center of the sun and spread outward. Although he was wrong, he recognized the spots for what they were -- solar phenomena.

The next documentation of sunspots was in 165 B.C., when it was reported the Chinese character wang appeared in the sun -- shaped like a cross with a bar across the top and bottom. It is accepted as the world's earliest precisely dated sunspot. The West's earliest reference to sunspots is Einhard's Life of Charlemagne, around A.D. 807. Joseph Needham found 112 instances of sunspots recorded in Chinese histories between 28 B.C. and A.D. 1638. In other Chinese books he found hundreds more notices, "but no one has had time or stamina to collect them into a body," comments sinologist Robert Temple. Nonetheless, the sunspot records constitute the oldest continuous series of such observations. And again, these are of great use to modern astronomers. Sunspot cycles, for instance, affect the earth's ionosphere and weather (magnetic storms are related to sunspots). Analyzing available records, Japanese astronomer Shigeru Kanda reports he has detected a 975-year sunspot cycle. If so, it may have significant implications for weather cycles.

The Chinese were also careful observers of comets. They computed the approximate orbits of about forty comet trajectories with such precision that many of their trajectories could be drawn on star charts simply from reading ancient texts. They were interested in the precise position and direction of the tail of each comet.

In the year 240 B.C., astronomers officially documented the appearance of a comet today known as Halley's. Another comet recorded in 467 B.C. is also thought to be Halley's. In the 600s A.D., they observed that comets shine by reflected light like the moon. They noted that comet tails always pointed away from the sun, suggesting this phenomenon was the result of a solar "energy." Today it's known that this cometic tail direction is caused by the force of "solar wind," the sun's radiation. It wasn't much of a stretch, says Temple, for the Chinese to formulate the idea of solar wind. It is congenial with their cosmological assumptions, Chinese literature being filled with references to the ch'i of the sun's radiation. Ch'i, translated as something like the "emanative or radiative force," comes from the sun. To Chinese astronomers it would have been obvious that the sun's ch'i was strong enough to blow the tails of comets as if in a strong wind. The Chinese conceived of space as being full of strong forces.

As a consequence of the emperor's divine connection with the cosmos, it became traditional after important changes of rulership, and always after a switch to a new dynasty, for a fresh calendar to be drawn up. This custom was well established by Han times (206 B.C. to A.D. 220) and led to some forty new calendars made up between early Han and the beginning of the Ming dynasty in 1368.

According to the theory of monarchy, the ruling dynasty remained fit to rule because of the accord the emperor maintained with the heavenly order. His special status in the order of nature allowed him to maintain a parallel order in the political realm, for the state was a microcosm. If the emperor lacked virtue or was careless in his duties, disorderly phenomena would appear in the sky as a warning of potential political disaster. Thus astronomers had to incorporate as many phenomena as possible in a "correct" calendar. The calendar, issued in the emperor's name, became part of the trappings of power that demonstrated his dynasty's right to rule, a function, writes sinologist Nathan Sivin, "not entirely different from that of economic indicators in a modern nation."

The importance of astronomical observing in this world of extreme politics, then, made secrecy absolutely necessary. Because the data could be easily manipulated, it could be dangerous in the hands of someone trying to undermine the current dynasty. It was therefore state policy that the proper place to do astronomy was the imperial court. In certain periods it was illegal to do it elsewhere. With this information virtually classified as top secret, the astronomer became a high-level administrative functionary in a country that developed the most elaborate bureaucracy in the ancient world. The databases resided in a state observatory deep within the bowels of the palace.

If not the greatest astronomical mathematicians, the Chinese were the greatest star charters before the Renaissance. Their earliest star chart goes back to at least 2000 B.C., to a carving on a cliff at Jiangjunya in Jiangsu Province. The carving contains many stars, as well as human and animal heads. There are disks indicating the sun in seasonal positions and where a number of bright stars and the moon appear over the seasons. This bright region is recognizable as the Milky Way by its position and appearance; the Milky Way displays gaps and divisions that are depicted on the carving.

China, being in the Northern Hemisphere, fixed itself on the northern circumpolar stars, both for orientation and to express its concept of divine rulership. The circumpolar stars in the higher latitudes are raised quite high in the sky as they rotate about the pole, so the fixity of the polar axis became an apt metaphor for the divine right of emperors. The pivot point about which this rotation occurs is known as the north celestial pole. The emperors were clever to adopt the stars of the north, such as Cassiopeia and Cepheus. These stars are located near the celestial pole, so that in the temperate latitudes of most of China they are visible eternally in the sky, never hidden by the horizon.

The first catalogs of star positions appear to have been drawn up by Shi Shen, Gan De, and Wu Xian, the earliest notable astronomer in China, who worked between 370 and 270 B.C., two centuries before Hipparchus. Together their lists enumerated 1,464 stars grouped into 284 constellations. (The West made bigger groups, with only 88 constellations.) In A.D. 310, during the Western Chin dynasty (A.D. 265-317), this early work was collated by the astronomer royal Qian Luozhi, who cast a bronze celestial globe with the stars on it colored in red, black, and white to distinguish the listings of the three astronomers. As early as the Han dynasty, astronomers prepared star charts. Carvings and reliefs show individual constellations or asterisms depicted as dots or small circles connected by lines to delineate the constellation itself. This ball-and-link convention did not appear in the West until the late nineteenth century.

Star maps need a means of specifying positions of heavenly bodies with reference to one another. The science of mapmaking took a leap forward in the second century B.C. when Chang Heng invented what's now called quantitative cartography. Chang, the inventor of the seismograph and a leading scientist, applied a grid system to maps so that positions, distances, and itineraries could be calculated and analyzed. Chang Heng's own works are lost, although an official history of the Han dynasty stated, "He cast a network of coordinates about heaven and earth, and reckoned on the basis of it." Copies of these maps were never made, since the information on them was too dangerous to risk its falling into the wrong hands. Meanwhile, in Europe, mapmaking had degenerated under the influence of religion, says Robert Temple, "to a point scarcely credible."

Drawing actual charts of the sky means finding a way to depict positions as if one is drawing a map. Preparing maps also involves the problem of mapping the curved surface of the celestial sphere on a flat surface, just as mapping the near-spherical surface of the earth requires the use of map projection. This is made more difficult if the sky is seen as a dome curving above one's head. In both China and the West, projection goes back a long way for mapping the earth. But for mapping the stars, Chang Heng was first, drawing up in Han times a chart that was a "Mercator" projection.

Mercator projection was "invented" in Europe by the Flemish mathematician and geographer Gerhard Kremer, a.k.a. Gerardus Mercator, and first published in 1568. But this projection system had been used by the Chinese centuries before Mercator. The projection works by means of a cylinder. If one inserts a transparent globe of the earth (or other celestial sphere) in the center of a hollow cylinder and turns on a light inside the globe, the features of the sphere's surface will be thrown, or projected, onto this cylinder, and will reflect a certain distortion. The higher up and lower down from the sphere's center, or equator, the more the features are distorted. Virtually useless for land travel, this projection has the odd property that a navigational course drawn on it will come out as a straight line, whereas with other maps such courses are arcs.

The oldest surviving projection chart depicting the whole of the visible sky is painted on paper and now resides in the British Library. Dating from about A.D. 940, it comes from Dunhuang in Gansu Province and gives a flat representation of Qian Luozhi's (the astronomer royal's) tricolored chart, working from his celestial globe. It presents the celestial globe as projected onto a surface by the cylindrical projection technique, displaying over 1,350 stars in thirteen sections. One section is a planisphere -- that is, in a kind of Mercator projection it depicts the circle of the sphere on a flat map centered on the north pole. The remaining twelve are flat maps centered on the celestial equator.

A century later, in 1094, Su Sung published further Mercator-style map projections in his book New Design for a Mechanized Armillary Sphere and Celestial Globe. One map had a straight line running across the middle as the equator and an arc above it, the ecliptic. The rectangular boxes of the lunar mansions are clearly seen, with the stars near the equator being more tightly packed together and those near the poles spread farther apart.

The evolution of Chinese instrumentation parallels that of the West. Notched jade disks and cylindrical sighting tubes date back to the fifth century B.C., and probably functioned as means of computing rudimentary celestial cycles. The Chinese used the gnomon as far back as 1500 B.C. Just as the Chinese had begun to standardize weights, measures, and other practical details in the sixth century B.C., and more extensively in the next three centuries, they standardized the gnomon. In addition to timekeeping, they used the gnomon to determine the terrestrial distance corresponding to an arc of the meridian. Such a determination of this north-south line was vital for precise calendar making, because precision calendars required measuring the latitude of those stations where the relevant observations were made. The gnomon was also significant in mapmaking and in the Chinese fascination with determining the size of the earth -- nearly a millennium before Eratosthenes!

Between A.D. 721 and 725, under the auspices of the Buddhist astronomer and mathematician Yi Xing and the Chinese astronomer royal Nangong Yue, Chinese scholars set out to do this. To measure the size of the earth, they selected nine locations covering the prodigious distance of more than 3,500 kilometers (2,175 miles) on a nearly north-south axis. They made simultaneous measurements of shadows at the summer and winter solstices at all nine stations. The main outcome of this feat: they determined that the distance on earth corresponding to 1 degree latitude was 155 kilometers (97 miles). This is larger than today's value of 111 kilometers (69 miles), but far more accurate than previous attempts. Indeed, they found that the variation on shadow length with changed latitude was four times the value previously thought. There was no piece of research like it carried out anywhere else during the Middle Ages. In making his tabulations, Yi Xing used "tangent tables." This was thought to have been a Muslim invention of the ninth century, but it turns out that the Chinese discovered the use of tangents and tabulated them at least one hundred years earlier.

The Muslim big-observatory concept came to China in the thirteenth century, and during the Mongol dynasty, in 1276, the astronomer Guo Shoujing built a giant gnomon called Tower of the Winds. It was an all-purpose observatory with the tower itself serving as a gnomon. A horizontal rod in an aperture at roof level -- about forty feet above the ground -- cast a shadow on the long low wall extending northward below. A chamber at the top was designed for watching stars, while the inner rooms of the tower housed a water-driven clock and an armillary sphere.

Modern astronomical observatories derive not from the European tradition but from the Chinese. Modern telescopes are oriented and mounted in the equatorial system, which in China goes back at least to 2400 B.C. Equatorial mounting takes the equator as the horizontal circle around the side of the instrument, and the pole as the top point. Europeans originally followed the Greek-Indian-Muslim tradition in which the two circles that were important were the horizontal and the ecliptic, the circle described by the sun's motion in the sky that is in the same plane as the earth's orbit around the sun. This tradition more or less ignored the equator. China, meanwhile, largely ignored the horizon and the ecliptic. In the seventeenth century, European astronomers came to realize that the Chinese equatorial system was more convenient and showed greater promise. It was adopted by Tycho Brahe and his successors and remains the basis of astronomy today.

The Chinese, furthermore, had the skills to build the precision observational instruments to display this system. Having invented cast iron, they built large astronomical instruments of bronze and iron that took the form of armillary spheres -- huge metal rings precisely graduated with the degrees of a circle.

Different rings representing different sky circles were joined together at the two points where they crossed each other. Always with an emphasis on the meridians, one ring would represent the equator, another the sky-circle meridian passing directly overhead and through the celestial pole. These devices had sighting tubes through which astronomers could observe specific stars. The astronomer could move the sighting tube along the equator ring until he found a star. Then he counted the number of degrees marked on that ring back to the meridian ring, which stood up from it at 90 degrees. As soon as he counted the degrees, he could detect the exact position of the star along the equator and tell which sky segment it was in. These instruments aided astronomers in drawing star maps with great precision.

The earliest known instrument of this type was built in 104 B.C., and the instrumentation became increasingly complex until the thirteenth century. Ken Shou-Ch'ang introduced the first permanently mounted equatorial armillary ring in 52 B.C. and in A.D. 84 Fu An and Chia Kmuei added a second ring to show the ecliptic. Chang Heng, the mapmaker, added a ring for the meridian in A.D. 125, as well as one for the horizon. But Chang Heng was not yet satisfied. He made an armillary sphere that rotated by water pressure in about A.D. 132. He used a wheel powered by a constant pressure head of water in a water clock to rotate the instrument slowly. This instrument was a tremendous tool for demonstrating and computing the movements of heavenly bodies.

An advanced version of the armillary sphere is the torquetum, first invented by Arabs sometime between A.D. 1000 and 1200. (Some credit al-Tusi for the invention.) Here all the various rings are no longer nested together in a single sphere but are mounted at various different parts of a set of struts in a way more efficient than that allowed by the constraints of a single sphere. In 1270 Kuo Shou-Ching made a metal torquetum called the "simplified instrument." It was purely equatorial, with all the Arab ecliptic components left out. It survives today at the Purple Mountain Observatory in Nanking. It was moved there from its home site at Linfen in Shanxi during the Ming dynasty when government officials no longer understood that the difference of 33Ž4 degrees in latitude caused by the move would render it useless. Needham called this "simplified instrument" the precursor of all equatorial-mounted telescopes. Needham believed that some knowledge of it eventually reached Tycho Brahe in Denmark three centuries later and led to Brahe's taking up equatorial astronomy for his instruments. Actually, an equatorial mounting of the kind devised by Guo Shoujing wasn't constructed in the West until 1791, when it was used for a telescope made in England, and thus its design became known as the "English mounting."

Astronomy was the first real science practiced by the world's ancient cultures. It was primarily observational (rather than experimental), but it meets most criteria for what a science should be. Next we will continue on to an associated discipline, cosmology. I write "discipline" rather than "science" because, as you shall see, it's not clear what cosmology is. Cosmology is dependent on astronomy, extrapolating its data to a worldview. Which is not to say that astronomers always agree with the tales spun by their cosmology colleagues.

Copyright © 2002 by Dick Teresi