Why do globes show the constellations the wrong way round?

by Doris Vickers (Vienna)

My hometown Vienna houses a lot of museums filled with famous artworks from several centuries and a myriad of artists. Vienna is also home to a very special museum – the Globe Museum. It holds almost 600 exhibits including both terrestrial and celestial globes, globes of other planets and astronomical instruments relating to globes, of which roughly 200 are on permanent display. One of its most prominent exhibits is the oldest globe in Austria, dating to 1536.

Globes in classical antiquity

We do not know who invented the first globe in history or where it was manufactured, but we know for certain that globes (or globe-like devices) existed in antiquity. The Roman author Cicero (106-43 BCE) reports in De Re Publica that the first globe was constructed by Thales of Miletus (626/623-548/545 BCE):

But when Gallus began to give a very learned explanation of the device, I concluded that the famous Sicilian had been endowed with greater genius than one would imagine it possible for a human being to possess. For Gallus told us that the other kind of celestial globe [sphaera], which was solid and contained no hollow space, was a very early invention, the first one of that kind having been constructed by Thales of Miletus, and later marked by Eudoxus of Cnidus (a disciple of Plato, it was claimed) with the constellations and stars which are fixed in the sky. He also said that many years later Aratus, borrowing this whole arrangement and plan from Eudoxus, had described it in verse, without any knowledge of astronomy, but with considerable poetic talent. But this newer kind of globe [sphaera], he said, on which were delineated the motions of the sun and moon and of those five stars which are called wanderers [the five visible planets], or, as we might say, rovers, contained more than could be shown on the solid globe, and the invention of Archimedes deserved special admiration because he had thought out a way to represent accurately by a single device for turning the globe those various and divergent movements with their different rates of speed. And when Gallus moved the globe [sphaera], it was actually true that the moon was always as many revolutions behind the sun on the bronze contrivance as would agree with the number of days it was behind in the sky.

Cicero, de re publica 1, 21-22. Translation: C. Walker Keyes

Note that Cicero mentions two devices here: a celestial sphere with depictions of constellations (i.e. a globe) and a device which shows the apparent motion of the Moon, Sun and planets (most likely some kind of orrery). Cicero attributes the invention of the globe to Thales of Miletus and considering that his other statement about Aratus having versified Eudoxos is true, we can assume that Cicero relates the common view about the invention of the globe. From the fact that Cicero states that the celestial sphere is “solid”, we can derive that it had no mechanism (as opposed to the other device, which Gallus moved). It is interesting, however, that both the globe and the other device are called sphaera (sphere), which indicates that sphaera was a term for any astronomical device.

While we do not have any of the devices mentioned by Cicero, we have detailed instructions on how to construct a globe by Ptolemy (c. 100-170) in his Almagest:

On the construction of a solid globe
[…] But we also wish to provide a representation [of the fixed stars] by means of a solid globe in accordance with the hypotheses which we have demonstrated concerning the sphere of the fixed stars, according to which, as we saw, this sphere too, like those of the planets, is carried around by the primary [daily] motion from east to west about the poles of the equator, but also has a proper motion in the opposite direction about the poles of the sun’s ecliptic circle. To this end we shall carry out the construction of the solid globe and the delineation of the constellations in the following fashion.
We make the colour of the globe in question somewhat deep, so as to resemble not the daytime, but rather the nighttime sky, in which the stars actually appear. […]

Ptolemy, Almagest VIII.3. Translation: G. J. Toomer

Apart from these theoretical descriptions of globes, we have very few traces of physical globes from antiquity. One of the more famous ones is the so-called Atlas Farnese, a Roman copy of a Greek marble sculpture dating to the second century AD of the Titan Atlas holding a celestial globe:

The celestial sphere contains 41 of the 48 classical Graeco-Roman constellations as seen from the outside (as if the observer were outside the sphere that the stars are attached to). There is, unfortunately, no way to further specify where the constellations on the globe came from, as they are inaccurate and there are no individual stars depicted. But the sculpture proves that celestial globes were very much part of Graeco-Roman culture.

Globes in medieval times and the Renaissance

Although the history of globes becomes shrouded in medieval times and in the Renaissance, we can find globes depicted in artwork. Two manuscripts in the Abbey Library of Saint Gall contain an illustration of a globe. Codex 902 (below) dates to the 9th century.  It shows a celestial globe with 10 constellations (Perseus, Auriga, Ursa Major, Aries, Taurus, Gemini, Leo, Eridanus, Canis Major and Argo). The globe itself is supported by a centre column, 6 further columns support the horizon ring. Codex 250 dates to the 10th century and appears to be copied from the earlier manuscript.

Other than these depictions in artwork, globes disappear from European history until the scientific investigations of Johannes von Gmunden (1380/84-1442) at Vienna University. He wrote several works about astronomical instruments, including one called Tractatus de ‘sphaera solida sive de astrometro sphaerico’ etc. (“Book about the solid sphere or about the astrometrical sphere”, held as Codex 5412 in the Austrian National Library). By the end of the 15th century, globe manufacturing took off in Europe on a grand scale.

The Aristotelian cosmos

What all globes have in common is that they have been constructed in an Aristotelian-Ptolemaic system, which was valid in Europe until Galileo Galilei (1564-1642) showed that Nicolaus Copernicus (1473-1543) was right in moving the Earth out of the centre of the universe and replacing it with the Sun. Aristotle (384-322 BCE) had proposed a universe that consisted of several celestial spheres, the outermost being the sphere that the stars are attached to. The Earth was considered the immovable centre of the universe. Ptolemy came up with the mathematical calculations that made the movements of Aristotles’ spheres predictable and calculable. Considering this system, a globe is a perfect representation of the celestial sphere – the stars are at an equal distance from the centre and the globe could be used to show the daily and yearly rotation of the sphere. This is also the reason why the ever-changing movements of the planets cannot be portrayed by a globe and that is why no planets are depicted on them.

A problematic viewpoint

While an observer on Earth looks at the apparent celestial sphere of fixed stars from the inside, a globe always presents the sphere from the outside – an observer would have to imagine sitting inside the globe to see the constellations the correct way around. However, astronomical literature from Ptolemy onwards, talked about the constellations and positions of individual stars within them as seen from Earth and most of the constellation illustrations depicting humans were drawn to show them from the front. These constellation illustrations had to be turned around for representation on a globe and were now seen from behind – left and right had to be swapped, so for example the star Regulus, one of the brightest stars in the night sky, sits at the bottom right of the constellation Leo for an observer on Earth (towards West) yet on a globe it sits at the bottom left (towards East). This led to misidentifications when trying to locate a star that was described as “on the right side of the constellation X” on a globe. That is why some globe manufacturers produced images of constellations that were neither seen from the front or back, but from the side – as a sort of compromise.

The problem only persists because we are not really used to celestial globes as much as we are used to terrestrial globes and maps – which show the Earth from an unusual viewpoint, as if you were circling the Earth in a spaceship (something which not many of us have done!).

Further reading

• The Whipple Museum of the History of Science (University of Cambridge) has a brief history of globes from the 16th century onwards.

• The British Library has a nice overview over European globes from the 17th and 18th century, including some pretty 3D-reconstructions.

• If you want to build your own model globe, then the History of Science Museum of the University of Cambridge has got you covered.

About the author

Doris Vickers is a Classical Philologist and currently pursuing a PhD about Wilhelm Schickard’s Astroscopium, a text dating to 1623 that deals with contemporary issues of astronomy, star names and star maps. During her research, she often comes across interesting factoids that she would like to share with you here. She is also an associate member of the International Astronomical Union (IAU) and the Etymology Task Group Chairperson of the Working Group on Star Names (WGSN).