I’ve been trying to watch Cosmos by Carl Sagan. I’ve never seen it and it’s proving to be a bit of a struggle. He definitely can write. Some of the sequences are fantastic, but some of it is badly dated. The thing that really grates to me is his dismissal of Ptolemy and his geocentric universe. For Sagan at best Ptolemy’s system held back astronomy by 1,500 years. At worst he’s only worth mentioning to say he’s dead wrong, like in the first episode.
It’s not really fair to lay into Sagan for his attitude to Ptolemy. His work is a product of its time and it was written over thirty years ago. But the idea that Ptolemy was clearly wrong seems to the popular understanding of Renaissance astronomy. The question here is Why did some people oppose the heliocentric theory of the universe? not Who in their right mind would accept it? It overlooks the power of the Ptolemaic system. If you followed Ptolemy’s work you could predict where the planets would be with enough accuracy for naked-eye astronomy. If Copernicus had only used simple circles, then his model might have seemed better, but he too needed to add epicycles and fudges to make his system match the observable sky. It needed fewer epicycles, but it was hardly perfect.
Popular belief is that the problem was solved when Galileo picked up his telescope and proved the heliocentric theory. In fact a recently published paper by Christopher Graney, The Telescope Against Copernicus: Star Observations by Riccioli Supporting a Geocentric Universe in the Journal for the History of Astronomy shows that the telescope could have dealt a serious blow to the Copernican model of the universe.
One obvious problem was the lack of stellar parallax. If the Earth was going round the sun, then the closest stars would seem to shift position relative to the more distant stars on an annual basis. No shifting was seen. In reality that’s not a surprise. Even the closest star has a parallax of around an arc second. This is a tiny angle. Imagine one degree divided into sixty parts. Each of these parts is an arcminute. Now divide one of these arc minutes into sixty parts and one of these parts will be an arc second. There simply wasn’t the equipment to measure a star’s position with that kind of accuracy in the seventeenth century.
The answer, said the Copernicans, is that the stars are a very, very far away. That’s why no shift due to parallax was visible. If that was the only data they had then no one would be able to say if this were true or not. But not only was there an argument about the size of stars.
Tycho Brahe had measured the sizes of stars. He did this by naked eye, measuring the size of the apparent disc of the star. If you know the angular diameter of the star, and how far away it is, you can work out the size of a star. It’s this fact that the Jesuit astronomer Riccioli used to debunk the Copernican model. Graney points out a table compiled by Riccioli that says that Tycho thought the stars were 14,000 Earth radii away, but they could be up to 40,000 Earth radii away according to some Ptolemaic models. Sirius was measured as 18 arc seconds across. That meant its diameter was between 0.6 to 3.5 the size of the Earth’s roughly. He also had his own estimate of 210,00 Earth radii for the distance to the stars, but even this only made Sirius 17.5 times the diameter of the Earth.
He then made a table of the required distances to the stars proposed by some Copernicans, to account for the lack of stellar parallax. In the case of the astronomer Wendelin, this distance to the stars was over 600,000,000 Earth radii. That made Sirius’s diameter over 50,000 Earth radii. Comfortably bigger than the entire universe according to Tycho or Ptolemy. Even the Sun was only thought to be 180 times bigger than the Earth. The stars would have to be completely unlike anything else in the universe, or else Copernicus was wrong.
The standard narrative in the history of science is that Galileo solved this by stating the stars appeared at points, and so could not be measured. In his first mention of the subject in the Starry Messenger of 1610 he says:“The fixed stars are never seen to be bounded by a circular periphery, but have rather the aspect of blazes whose rays vibrate about them and scintillate a great deal.”
Graney points out this is simply not the case. Through telescopes stars appear as discs. This is true with modern equipment. Above is the constellation Delphinus compiled from exposures taken by the ESO as part of the Digital Sky Survey. The major stars are visibly discs. This is something that happens with telescopes, it’s an effect known as the Airy disc, and it’s due to the diffraction of light. Graney can also show that his later work, when he was more adept at using a telescope, refers to stars being round and not points.
Graney takes this idea forward. It’s not simply enough to show that Riccioli was poking holes in the weak parts of Galileo’s idea. Graney goes on to ask if this was a fundamentally flawed attack on the Copernican system. Eventually the true nature of the discs would be discovered, and Graney also notes that while Riccioli was measuring these discs, Horrocks was watching stars wink out instantly when the Moon passed in front of them. That meant that the stars must have been points, even if they appeared to be discs. There are a few publications that say the stars cannot have large discs in the seventeenth century, but Graney also notes they don’t reference each other. It shows that the discovery was inevitable if it was being made independently in several places, but he also points out that it must have also been needed to be made again and again for that to happen. Clearly the idea that stars were points wasn’t getting traction.
Graney’s paper makes seventeenth century science a bit more comprehensible. The caricature of the period is that Galileo turned up on the Pope’s doorstep and said “Guess what your holiness! The Earth goes around the Sun!” and Pope Urban VIII stuck his fingers in his ears and yelled “LA LA LA I CAN’T HEAR YOU!” Obviously there was a political dimension to the heliocentricism debate, but Graney shows that it wasn’t purely about politics and there were sound scientific reasons, given the state of instruments and knowledge, for holding to a geocentric universe.
If you frame the whole debate in terms of who was right and who was wrong then you could argue this elevates Ptolemy at the expense of Copernicus. But science isn’t a zero-sum game. Graney’s work shows that the Ptolemaic system was supported by a lot of observable evidence. If that’s the case, then the achievements of Copernicus, Galileo and Kepler are even more impressive. There’s no difficulty in disproving something that’s obviously wrong. Spotting there’s something not quite write about an answer when everyone else is happy with it is much more difficult. Downplaying the arguments for a geocentric universe denies some of the high-powered brainwork that was needed to develop a heliocentric system. Us and Them histories of science can bring modern disputes into a historical setting and risk losing sight of what was unique about a time.
Graney’s paper doesn’t have a DOI, but it is currently (Jan 2011) on Ingenta Connect. If you have a subscription to JHA you can download it from there.
Graney, C.M. (2010). The Telescope Against Copernicus: Star Observations by Riccioli Supporting a Geocentric Universe Journal for the History of Astronomy, 41 (4), 453–467
And just after finishing up this post and wondering when to schedule it for I see Christopher Graney is in the news with another Riccioli paper, this time on the lack of evidence for a Coriolis force.