The earliest representations of the sky were actually globes, on which the constellations were shown as though viewed from a God-like position beyond the stars; this meant that the constellation shapes were represented back to front by comparison with the way we see them from Earth. In the Museo Nazionale, Naples, is a marble statue of Atlas holding on his shoulders not the Earth but a globe of the heavens on which the constellations are depicted in this mirror-image way (see picture at right). The sculpture is called the Farnese Atlas, after Cardinal Alessandro Farnese (later Pope Paul III) who acquired it in the early 16th century and exhibited it in the Farnese Palace in Rome.
It is the oldest known celestial globe, for historians think that the sculpture was probably made in Rome around the second century AD. Even more significantly, it is thought to be a copy of a Greek original from the second or third century BC, around the time that Aratus wrote his Phaenomena. Thus the globe held by the Farnese Atlas provides our only firsthand look at the star pictures that the ancient Greeks imagined in the sky. (Note: In 2005 the American astronomer Bradley Schaefer claimed that the star positions on the Farnese globe were based on the original catalogue of the ancient Greek astronomer Hipparchus, which would be of great interest if true. However, Dennis Duke of Florida State University, among others, has refuted this analysis.)
Flat star charts
An early form of flat star chart was the astrolabe, popular with the medieval Arabs. Usually made of brass, the astrolabe was a disk on which the positions of bright stars were indicated; the same principle lives on in the star-finding devices called planispheres used by present-day amateur astronomers and sailors. The earliest surviving astrolabes date from the tenth century AD, but written evidence shows that they were known much earlier, possibly even in the time of Ptolemy, around AD 150. Astrolabes are a rich source of old star names.
Other than astrolabes, the oldest known flat sky map is a Chinese paper scroll, over 2 metres long, thought to date from the mid to late 7th century AD. It is known as the Dunhuang star chart after the place on the Silk Road trade route in north central China where it was found in the early 20th century; it is now in the British Library, London. Since the Dunhuang chart depicts the Chinese constellation tradition, which was independent of that in Europe and Arabia, most of the constellations are unrecognizable to modern eyes (see, for example, the north polar region of the sky in the box below).
The Chinese sky – a lost tradition
Chinese constellations were smaller than Western ones, and more numerous. Each constellation usually consisted of only a handful of stars which made it easier to specify areas of sky without the need for accurate coordinates. By the end of the third century AD, Chinese astronomers had developed an elaborate system of 283 constellations consisting of a total of 1464 stars. These constellations did not depict myths but facets of Chinese imperial, social and rural life.
For example, we find Dizuo, the throne of the emperor (the star Alpha Herculis); Huanzhe, the court eunuchs (60 Herculis and three neighbouring stars); Lingtai, the astronomical observatory; and even the toilet, Ce, behind a modesty screen, Ping, both in Lepus. In some parts of the sky, groups of constellations on a common theme formed large tableaux depicting scenes such as the autumn harvest, the winter hunt, a cavalry camp, and a celestial market. Some of the same characters popped up repeatedly in different parts of the sky, notably the Emperor and members of his extensive retinue.
Unlike the imaginative artistry of western celestial cartography, Chinese star charts such as that from Dunhuang did not offer pictorial representations of these constellations. Instead, the chart makers simply plotted the constituent stars as dots of similar sizes connected by lines, with no attempt to scale the symbols according to the stars’ brightnesses. This lack of a magnitude scale on Chinese charts adds to the difficulty of identifying the stars involved. The Chinese constellation system was unknown in the West and had no influence on the 88 celestial figures that we know today. It was still in use when Jesuit missionaries introduced Western constellations to China in the 17th century, after which it died out. Information on Chinese constellations is included at the end of each constellation entry in Chapter 3.
Picturing the Ptolemaic constellations
The earliest surviving depictions of the Ptolemaic constellations on paper are found in illuminated manuscripts of the poetic works of Aratus and Hyginus dating from the 9th century and onwards. These illustrations were the creation not of astronomers but of artists who interpreted the figures quite freely, with little concern for the framework of the underlying stars. As a result, the images bore only a loose resemblance to the constellation figures as described by Ptolemy. Woodcuts from later printed versions of these works, such as an edition of Hyginus’s Poeticon Astronomicon published in Venice in 1482, are sometimes reproduced in modern books as examples of early constellation visualization, but they are not true star charts. Their intent was simply to provide a decorative accompaniment to literary texts, with no pretensions to scientific accuracy.
It was actually the Arabs who produced the first scientific depictions of the Ptolemaic constellations over 800 years after the Almagest was written. Around the year 964, the Arabic astronomer Ἁbd al-Raḥmān al-Ṣūfī, usually known simply as al-Ṣūfī (or Azophi in Latinized form), produced a revised and updated version of the star catalogue in the Almagest called The Book of the Fixed Stars. In this, he added illustrations of each constellation with the component stars positioned as Ptolemy had specified. However, the human characters were all drawn in Arabic robes and headdresses, and there are other aspects of the iconography that look unfamiliar to western eyes. So to find the true genesis of modern star charts we must return to Europe.
First printed star chart
A major step towards standardization of imagery of the Greek constellations came in 1515 when Albrecht Dürer, the great German artist, drew the first European printed chart of the heavens with the help of Johannes Stabius, an Austrian mathematician, and Conrad Heinfogel, a German astronomer. Dürer’s chart consisted of a pair of woodcuts, one showing the zodiac and all constellations north of it, the other showing all known constellations south of the zodiac. Both halves are based on the stars and constellations catalogued by Ptolemy in his Almagest, updated and accurately plotted for the year 1500. The constellations are in mirror image, as on a celestial globe or an astrolabe, a tradition that most early maps were to follow. Dürer’s depictions of the constellation figures established an artistic style that was echoed on many later celestial maps. It was superseded by a sequence of four great star atlases that in turn set new standards, both scientific and artistic.
© Ian Ridpath. All rights reserved