To be clear: No remotely good reason was found. And the results presented are obviously nothing more than numerology
But to damn Eddington and others for looking is to abuse hindsight.
Completely disagree.
But to claim it was absurd from the start seems not only incorrect but unfair.
No, it is both correct and fair. Eddington ca. 1930s and was a crank. Bohr, Fowler, Gamow, Kramers, von Neumann, Rosenfeld, and Wigner all knew it at the time, although they generally used more polite terms out of respect for Eddington's prior work [arXiV:
hep-th/9411233].
This was the first time he had ever talked about these theories to a scientific audience. Many in the audience were waiting to ambush him. Everyone jumped on him, including Kramers, von Neumann, Rosenfeld, Wigner, Gamow, Fowler and Bohr. Everyone said, very politely, that the way he approaches all parts of physics, including quantum mechanics and relativity, is in contradiction with the ordinary theory of quantum mechanics and relativity.
Kramers was elected by the younger members, especially Gamow, to deal with Eddington and he gave the longest discussion in which he criticized Eddington’s views. When I was at the anniversary meeting in Kazimierz the organizers showed me an illustration, that came from their private files, in the form of a medal that Gamow presented to Kramers after he had performed this service to the community. The medal reads: “For the masterpiece of polite scolding.” For most of the the participants this talk and the following discussion was the highlight of the meeting
Here's what Born thought Eddington's programme:
Eddington connects the dimensionless physical constant with the number n of the dimension of his E spaces and his theory leads to the function f
= n²(n²+1)/2 which, for consecutive even numbers n = 2,4,6,... assumed the values 10,136,666... Apocalyptic numbers, indeed. It has been proposed that certain well-known lines of St. John's Revelation ought to be written in this way: "And I saw a beast coming up out of the sea having f(2) horns and his number is f(6)..." but whether the figure x in "...there was given to him authority to continue x months..." is to be interpreted as 1✕f(3)-3✕f(1) or as [f(4)-f(2)]/3 can be disputed.
and Kramers' more private reaction:
Goudsmit and Kramers were both in the audience. Goudsmit understood little but recognized it as farfetched nonsense. Kramers understood a great deal and recognized it as total nonsense. After the discussion, Goudsmit went to his friend and mentor Kramers and asked him, "Do all physicists go off on crazy tangents when they grow old? I am afraid." Kramers answered, "No Sam, you don't have to be scared. A genius like Eddington may perhaps go nuts but a fellow like you just gets dumber and dumber."
On one point I was wrong: I said that Eddington switched to 137 because of more accurate experiments. This was not the case; from the start in late 1928, Eddington was aware that the experimental value was 1/137.1. Eddington acknowledged this in the paper but predicted 136 anyway because it is the number of independent components in a symmetric 16x16 matrix. This kind of thing became his
modus operandi; e.g., in 1933, he published a paper "deriving" the proton-to-electron ratio based on the roots of the equation
10m² - 136m + 1 = 0
with handwaving to "explain" why the equation predicts 1847.1:1 while the experiments said 1836:1.
And gives the wrong impression of science always being "right from the start," ...
I'm not saying that science is always right from the start. I'm saying that
this particular nonsense was wrong from the start.
My apologies if my failing to distinguish 136 and 137 was the heart of your complaint. I'd hold the distinction does not matter in terms of confirmation bias.
No. My main complaint is that there never was any good scientific reason to believe Eddington or any other similar numerologists. It was not even the phlogiston of the time. It was rubbish from its very inception.
The distinction certainly plays a role in terms of arguing when one "should" let go of a pet theory. I should have been more clear.
Did you disagree beyond this?
Yes, I do. For the issue of "letting go" to come up, what you start with would have to have been sensible at one point in time. I do not believe that to be the case for Eddington's numerological programme.
