Home Page Science Frontiers

No. 126: Nov-Dec 1999

Issue Contents

Other pages











It's All In The Believing

Returning to the subject of solar eclipses, it seems that in the past eclipse phenomena have been employed to promote an appealing theory even when the observations were of poor quality. Scientists have been known to "spin" data like politicians!

A classic case of scientific "spin" occurred in connection with the total solar eclipse of 1919. British astronomer A. Eddington had mounted expeditions to Sobral, Brazil, and the island of Principe off the west coast of Africa. He had telescopes set up at these two locations to measure the bending of starlight by the sun, as predicted by Einstein's Theory of Relativity. In 1919, Relativity was not yet the cornerstone in the Temple of Science that it is today. Eddington "believed" in Relativity and wished to make it more acceptable. Eclipse photos showing the shifting of star images by the gravitational influence of the eclipsed sun might do the job.

On the day of the eclipse, Principe was bedevilled by clouds, and only 2 photographic plates were deemed marginally acceptable. At Sobral, 18 poor plates and 8 better plates were obtained. The problem was that the 18 poor plates yielded a deflection of starlight much smaller than predicted by Relativity, while the 8 better plates produced a much higher value. By adding the 2 plates from Principe to the mix, Eddington managed to come up with a number close to that required by the Theory of Relativity. It was not the clear-cut victory for Einstein that the textbooks proclaim.

Yet the spin was on! The New York Times trumpeted:

"Lights Askew in the Heavens. Men of Science More or Less Agog; Einstein's Theory Triumphs."

Everywhere scientists began to take Relativity more seriously; Einstein's star rose rapidly. Of course, it all turned out well in the end, because we now have many other, more convincing data supporting Relativity. But this happy ending does not subtract from the fact that Eddington let ideology affect his conclusion. Even today, the results from the 1919 eclipse are still proclaimed to be proof of Relativity.

(Morton, Oliver; "Science in the Dark," Wall Street Journal, August 11, 1999. Cr. E. Fegert.)

From Science Frontiers #126, NOV-DEC 1999. � 1999-2000 William R. Corliss