Was Starlight Deflection Important for the Acceptance of General Relativity?
A Response to Critics
August 2, 2004
A criticism has been posted on several places on the Internet concerning our claim in The Privileged Planet that total solar eclipses were important to the confirmation and rapid acceptance of Einstein’s General Theory of Relativity.
The basic criticism is as follows. Observations of the deflection of starlight near the Sun during total eclipses were not very important for the acceptance of General Relativity. The advancement of the perihelion of Mercury had already proven the theory. Therefore, this example of the link between habitability and measurability has been oversold in chapter one of The Privileged Planet.
Both the anomalous advancement of the perihelion of Mercury and the deflection of starlight observed near the Sun were important confirmations of Einstein’s General Theory of Relativity. In the mid-nineteenth century, Leverrier pointed out that the perihelion of Mercury’s orbit was advancing at a rate inconsistent with Newtonian mechanics. Thus, it was well known to scientists long before Einstein proposed his theory in 1915 that this was a problem needing to be solved.
The known anomalous behavior of Mercury is what is called in science the problem of “old data.” When a new theory is proposed, scientists try to confirm it both with existing observations and with proposed novel observations. Since the anomalous advance of Mercury was already known prior to 1915, General Relativity is said to “retrodict” it. But Einstein also proposed a novel test of his theory--an observation that had never been made. This was the famous test Eddington conducted in 1919 during a total solar eclipse. For this reason historians of science consider the eclipse observations so important historically. The solar eclipse observations, unlike the advance of Mercury’s perihelion, confirmed General Relativity.
But, the critic might respond, perhaps the prior knowledge of Mercury’s perihelion advance was merely an historical accident. Scientists might just as well have discovered the bending of starlight before the advance of Mercury’s perihelion. Therefore, the fact that total solar eclipses were more important to the acceptance of General Relativity is just an accident of history. So this plank of our argument is weakened.
At first reading this criticism sounds plausible, but if you think about the progress made in understanding the Solar System, it seems less likely. Astronomers discovered the specific value of the advance of the perihelion of Mercury simply because they acquired more knowledge about the motions of the planets. One of Newton’s successors was bound to notice that its motion didn’t fit the predictions. Observing the deflection of starlight near the Sun, however, was very different. No one is likely to attempt it unless one had a specific theory in mind about the motion of light near massive objects, which one wanted to test. So the historical order of these observations is not surprising.
Solving the problem of the advance of Mercury’s perihelion was not as clear-cut as solving the deflection of starlight. Even as recently as the 1980s Dicke was still arguing that part of the advance of Mercury’s perihelion might be explained by a small non-sphericity of the Sun. (See, for example, R.H. Dicke (1974), “The oblateness of the Sun and Relativity,” Science 184, 419-429.) If true, this would dissolve the agreement between the observed advance of Mercury’s perihelion and General Relativity.
But even granting all the critics’ claims--so what? We state in Chapter 5 that the observation of the advance of Mercury’s perihelion was one of the two important tests of General Relativity. We argue that both solar eclipses and our observation of the advance of Mercury’s perihelion are better from Earth than from most other places in the Solar System:
The outer planets are quite spread out, making it difficult to observe the other planets. And the inner planets would be largely lost against the bright glare of the Sun, making the first two tests of Einstein’s General Theory of Relativity—the precession of the perihelion of Mercury and the bending of starlight—more difficult. – p. 88.Interestingly, both tests of General Relativity would have been more difficult earlier (less habitable) in the history of the Solar System. The Sun’s greater activity early on would have made the corona brighter, which, in turn, would have made it more difficult to image stars near the Sun. As it is, early observations were limited to angular distances greater than two solar radii. Also, early on the Sun would have been rotating faster, making its oblateness greater. Thus, astronomers would have had to disentangle the effects of solar oblateness to test General Relativity with the advance of the perihelion of Mercury.
You might think that observations from Mercury would be better for discovering its anomalous perihelion advance. But consider the prerequisites for discovering the true nature of the planets’ orbits. It is a bit more involved than merely transporting modern humans with all our accumulated knowledge to Mercury and then asking whether it would be easier for us to determine the properties of its orbit. As we note in footnote 7 of Chapter 6, Mercury’s high eccentricity and spin-orbit 3:2 resonance would have made Celestial dynamics a more difficult affair for Mercurians (if there were any). Contrast that with Earth’s simple circular orbit and very different day and year lengths. And, let’s not forget the presence of our Moon, which has aided astronomers several times throughout history to understand our place in the Solar System and celestial dynamics. Neither Mercury nor Venus has a moon.
Of course, as we say several times in the book, we don’t argue that the Earth is optimal for observing every particular type of phenomenon we cite. Rather, it’s optimal in the constrained sense of providing the best overall place for discovery. So, even if it were easier to measure the perihelion advance from Mercury or Venus (neglecting its thick cloudy atmosphere), this would have to be balanced with other important scientific discoveries that depend on distance from the Sun. For instance, Mercury and Venus would offer poorer platforms for measuring stellar parallax because they orbit closer to the Sun.
Finally, we don’t argue that Earth is unique. Discovering another planet with complex life around another star in the Galaxy would be quite compatible with our hypothesis, so long as that planet is genuinely Earth-like. Finding a fundamentally different planet with (native) complex life on it, in contrast, would contradict our argument that the conditions for life and scientific discovery correlate in the universe.