The ancient Greeks had valid reasons for believing the Earth to be at rest at the centre of the Universe. Geocentrism fit with their understanding of gravity, and while parallax shifts were observed for the Moon and planets, ancient astronomers did not detect any noticeable shift among the fixed stars — though with their instruments and conceptual framework, such shifts would likely have been imperceptible. This is much the same as Einstein’s reasoning for one of the basic principles of special relativity: no change could be detected in the speed of light due to the difference in Earth’s velocity when it is on opposite sides of the Sun, so Einstein postulated that the speed of light is a universal constant that does not depend on one’s motion. Similarly, given their observational limits and prevailing beliefs, many ancient astronomers reasonably concluded that geocentrism was the more coherent hypothesis in their time.
Eudoxus of Cnidus (408–355 BCE), who, along with Aristotle, was a student of Plato’s, developed a mathematical description of a geocentric universe involving a system of 27 nested spheres rotating at different rates about different axes, in order to account for the apparent motions of the Sun, Moon, stars, and the five planets. Aristotle updated this to a system of 55 crystalline spheres made of aether. The large number of spheres was needed to account for the observed complexity of planetary motion.
Aristotle considered everything that the Greeks knew about the world in the fourth century BCE and carefully addressed the alternatives that had been proposed. It is therefore unsurprising that his theory eventually came to be adopted as the standard, self-consistent physics of the era. Revised forms of Aristotelian physics survived for nearly 2000 years, even as alternative cosmological models were occasionally proposed. Aristotle’s framework remained dominant because it provided a comprehensive picture that fit well with everyday experience and with the observational data available.
One of the most famous challenges to Aristotle’s geocentric theory came from Aristarchus of Samos (c. 310–c. 230 BCE). Aristarchus proposed a heliocentric theory of the Universe, in which the Sun and stars remained fixed while the Earth and planets followed circular orbits around the Sun. Regarding the question of how distant the stars might need to be relative to Earth’s orbital radius, Aristarchus claimed that the Earth’s orbital radius “bears such a proportion to the distance of the fixed stars as the centre of the sphere bears to its surface” (Archimedes, 1897, p. 222). This implies either that the Earth’s orbital radius is precisely zero — the exact centre of a sphere — or that the fixed stars must be at an essentially infinite distance.
Aristarchus’ original works have not survived, but his theory was discussed by Archimedes of Syracuse (c. 287–c. 212 BCE) in The Sand-Reckoner, where Archimedes aimed to estimate the number of grains of sand required to fill the entire universe. Archimedes noted Aristarchus’ hypothesis as well as his overstatement of the ratio of Earth’s orbital radius to the size of the Universe. He then attempted a more reasonable estimate for the size of the Universe, arriving at a value for the distance to the fixed stars that was on the order of 1 parsec — comparable to the actual distance to the nearest stars — and concluded that 10⁶³ grains of sand would be needed to fill it.
Although Archimedes indicated that Aristarchus’ heliocentric theory was not the prevailing one, the fact that he used it as the basis for his estimate rather than the common geocentric theory is often taken as evidence that the ancient Greeks were not ideologically opposed to heliocentrism. A number of such instances suggest that Aristarchus’ proposal remained an open possibility for centuries.
The question that is important for us to ask, in our search for a well-balanced scientific method, is this: why did Aristarchus’ heliocentric theory not gain traction in ancient Greece?
Some have speculated that the answer lies in Aristotle’s personal authority, but this is not sufficient: Archimedes himself was willing to consider Aristarchus’ hypothesis alongside the Aristotelian view. A better explanation is that Aristarchus’ model faced three serious obstacles. First, the dynamics of a rapidly spinning and orbiting Earth were implausible within ancient physics: if the Earth rotated, objects thrown into the air ought to fall far to the west, and violent winds should sweep constantly eastward. Second, the Ptolemaic refinements to geocentrism that came after Aristarchus produced extremely accurate predictive tables; no comparably precise heliocentric tables existed to compete with them. And third, Aristarchus’ theory required a radical epistemic shift: the apparent daily rotation of the heavens and the looping motions of the planets would have to be treated not as literal motions but as optical effects of Earth’s own motion — a leap that violated the prevailing expectation that appearances correspond to physical reality.
Compounded by these factors — dynamic implausibility, predictive sufficiency of geocentric models, and the conceptual leap required to treat appearances as illusory — heliocentrism never became the standard model in antiquity.
In fact, a better estimate of the Earth’s circumference was found by Eratosthenes of Cyrene (c. 276—c. 194 BCE), a contemporary of Archimedes. Eratosthenes knew that at local noon on the summer solstice, the Sun would appear at the zenith (directly overhead) in the city of Syene, as its light would reach the bottom of a deep well. He measured the angle of a shadow cast at local noon on the summer solstice in Alexandria, and found it to be 1/50 of a circle. Assuming Alexandria and Syene lie on the same meridian (i.e. at the same longitude), he could take these measurements to occur at the same time. Assuming the Sun is distant enough for light rays at the two locations to be parallel, he needed only the distance between Alexandria and Syene to determine the radius of the Earth (see Figure 2-2), which he determined to within a few percent of its actual value of 40,000 km. This would increase the rotational velocity of a person on the equator to more than 450 m/s.
In addition to the Earth’s daily spin, the heliocentric system also postulated that the Earth should orbit the Sun once a year, which, at such a great distance, amounts to another very large velocity. Aristarchus’ estimate for the distance to the Sun, which was quoted by Archimedes in The Sand Reckoner, was 10,000 Earth radii (the actual value is 23,000); therefore, the circumference of Earth’s orbital radius, according to Aristarchus, must have been 10,000 times the Earth’s circumference. In order to cover that distance in a year, the Earth’s orbital speed would have to be 13 km/s (it’s actually 30 km/s)!
Nothing seems more stable than the earth beneath our feet. As Aristotle noted, apart from the celestial objects, everything seems to move vertically, no matter where on Earth one is located, and all earth and water seems to be attracted towards the centre of the Earth. If it were spinning, why should its atmosphere spin too; and if it were moving tens of kilometres each second around the Sun, why should the air and the clouds go along with it? As Claudius Ptolemy (c. 90—c.168 CE) eventually put it, “although there is perhaps nothing in the celestial phenomena which would count against [the hypothesis that the stars are fixed and the Earth is spinning], at least from simpler considerations, nevertheless from what would occur here on earth and in the air, one can see that such a notion is quite ridiculous” (Almagest, H24).
Yet despite the arguments that stood against the heliocentric system, there were also arguments in favour. The Sun was known already to Aristarchus to be much larger than the Earth, and it would have made sense for the smaller body to orbit the larger. The spinning Earth, with its tilted axis, offered a simple explanation of the seasons and the variation in the Sun’s altitude throughout the year, as well as its path through the zodiacal constellations. Perhaps most importantly, the heliocentric model offered a simple explanation for the planets’ tendency to occasionally pause, and for a while wander back along their path through the apparently fixed stars (their so-called retrograde motion), an important phenomenon in the geocentrism vs heliocentrism debate which will be discussed in more detail below.
The ancient Greeks were confronted with fundamentally different hypotheses, each with its own set of merits and drawbacks. While geocentrism was the commonly adopted theory, it is clear even from the arguments Ptolemy offered centuries after Aristarchus, that geocentrism was not merely accepted on an ideological basis, by people too closed-minded to consider other possibilities.
The best explanation for why heliocentrism never gained wide acceptance in ancient Greece seems to be: that at this point the early philosophers had done all they could to provide working hypotheses that would possibly explain the phenomena; that what Greek science really needed the most was a good observational record and a mathematical model that would describe observations; and that it would be a very long time before either of these would come to a point where creative thinking about the nature of reality, that challenged concepts people had become accustomed to, would even be relevant. Simply put, there was too much science to be done—and the people who did the work, such as Ptolemy, favoured the geocentric model for reasons they had given.
In this way, by about the second century BCE, Alexandrian astronomers finally began to practice “hard science”—and they arrived only centuries later at a sophisticated description of all celestial motions that agreed well with detailed observations and coincided with the known physical laws. The significance of their achievement can only be appreciated with a better knowledge of the changing positions of the Sun, Moon, stars and planets; so watch the following video, which demonstrates all the different motions that had to be carefully mapped and accounted for.
cosmiCave.org 