Life of the Mind

No Shadow of a Doubt: The 1919 Eclipse that Confirmed Einstein’s Theory of Relativity

Daniel Kennefick. Princeton University Press, 2019. 403 pages. $29.95.

By Jay M. Pasachoff

An expert on the work of Arthur Eddington and others in proving Einstein’s General Theory of Relativity, Daniel Kennefick has written a masterwork that graces this centennial year. Eddington and Frank Dyson, from England, took telescopic equipment to Principe, an island off the coast of what is now Gabon, and Sobral, respectively, for the total solar eclipse of May 29, 1919. When the results were released at a joint meeting of the Royal Society and the Royal Astronomical Society on May 9, 1919, news of the discovery was covered in headlines around the world, and Albert Einstein became a celebrity.

As Kennefick points out, “Given the expense and the difficulty and the likelihood of failure, one might wonder why astronomers bother with eclipses at all. The reason is that although eclipse expeditions are apt to end in disappointment, every once in a while they can change the world.” The expeditions of 1919 did.

Kennefick provides a course in the evolution of science and of ideas relevant to relativity in the early 20th century. In the first chapter, he provides the background of the test of relativity, and provides a wonderful photo of Eddington and Einstein together.  In the second chapter, he gives a background on solar eclipses. At this writing, we are looking forward to the 2 July 2019 total solar eclipse whose path of totality will cross the Pacific, with only one small island included, and then in the late afternoon reach Chile and Argentina. But eclipses are no-longer needed to test Einstein’s theory; the positions of quasars when the solar disk passes near them provide more precise tests of the same “deflection” effect of the Eddington/Dyson expeditions, and the discovery of merging black holes and merging neutron stars, with data reaching Earth starting in 2015, provide a triumph of which Einstein would be proud—especially since he once doubted that gravitational waves existed.

In the next two chapters, we learn about the twin conscientious objectors involved: Eddington (a Quaker) and Einstein, though “CO” status was not as acceptable as it was to many young people in the Vietnam-war era and since. Indeed, we see that Eddington had apparently waived his CO status by having a deferment, and he wound up justifying his further deferment by the need to have him prepare for the 1919 eclipse expedition. We also learn briefly about Einstein’s marriage to Mileva Maric, herself the subject of a new book, and his moving in part to be close to his cousin Elsa.

In Chapter 5, we learn how the War affected the preparations for the expedition. As it turned out, and as Naomi Pasachoff and I discussed in "This Isn't the 1st Eclipse to Occur on Aug. 21, and It Won't Be the Last," (space.com, 2017), the German scientists who had agreed to test Einstein’s theory (with the American scientists from the Lick Observatory deciding to wait for the next American total solar eclipse) reached Crimea just before war broke out and were interned. Though they were released weeks later, their equipment remained in Russia for years, which affected future possible tests. The weather was not good in Principe; unfortunately, they had dismissed the idea of going to Liberia, from which I just saw coronal images from an expedition of the Carnegie Institution of Washington under clearer skies. The Carnegie expedition, however, made magnetic measurements, not testing relativity. (The August 21 date in 1917 had also been relevant, since attempts were made to observe the bright star Regulus near the Sun, but the sky remained too bright.)

I am amused to learn about the shipping and transportation difficulties that Eddington and his instrumentalist, Edwin Cottingham, had in getting to Principe and that even only two months in advance it wasn’t clear he (and/or his equipment) would get there, given revolution and shipping problems. And many of the problems they discuss continue to this day. The change in focus resulting from temperature changes the late Prof. Menzel and I dealt with at the 1970 total solar eclipse in Miahuatlán, Mexico (for which 50th anniversary celebrations are being planned in Oaxaca) was dealt with brilliantly by the late Prof. James Baker, perhaps the best astronomical-optical physicist then practicing by having temperature-insensitive “Invar” rods controlling the spacing in the spectrograph he designed for us. The reflection provided by coelostats (“see-lo-stats”; etymology: sky steady) in both Principe and Sobral to allow horizontal, fixed mounting of the telescopes was imperfect, and Kennefick points out that the quickly-tarnishing silvering of the time on glass surfaces wasn’t replaced by longer-reflecting aluminum until the 1930s. (I myself am coauthor of two papers on the use of coelostats at solar eclipses.)

The second half of Kennefick’s book is post-eclipse, dealing not only with the reduction of the photographic plates but also with the appropriate philosophy of science, from Karl Popper’s “falsifiability” to Thomas Kuhn’s “paradigms.” We learn about the questions raised whether Eddington had faked some of his results, and why the Sobral expedition had discarded the measurements from one of their instruments on the grounds that it was out of focus. Kennefick’s conclusion, with his reasoning discussed at length, is that “it was Dyson who had the most relevant expertise and it was Dyson who made the key decisions. The decision by many astronomers to accept the eclipse results was, in great part, an endorsement of Dyson’s judgment.”

The New York Times’s headline “Lights All Askew in the Heavens” published shortly after the London November meeting summarized the popular view then and now. It is too bad that Eddington’s plates and notes have not survived the decades. But we have the Sobral plates, and even the re-reduction carried out by the head of the Department of Terrestrial Magnetism of the Carnegie Institute of Washington, which measurements again endorsed the Einstein prediction, as did a re-measurement of the surviving plates in 1978. In fact, the re-measurement led to a closer agreement with the Einstein prediction than the earlier measurements had (though Stephen Hawking had misremembered the circumstances, writing incorrectly in his popular book that the re-measurement had led to huge uncertainties).

A huge anti-Einstein movement had developed early on, and in recent decades a story had spread that Eddington had fudged the results. Kennefick convincingly and elaborately shows the growth of this myth and describes how it has been debunked. The situation has been somewhat similar to the myth that Mileva Maric, Einstein’s first wife, was responsible for an important part of Einstein’s work, debunked by Allen Esterson and David C. Cassidy in a new book Einstein's Wife: The Real Story of Mileva Einstein-Marić, reviewed by my own wife, Naomi Pasachoff, for the journal Metascience (to appear). Jeffrey Crelinsten had earlier discussed the reception of the results in his 2006 Einstein’s Jury; I am glad that both Kennefick and Crelinsten have accepted my invitation to speak for a centennial session of the American Astronomical Society’s Historical Astronomy Division as part of the joint meeting of the AAS’s Division of Planetary Sciences and the European Planetary Science Congress in Geneva, Switzerland, in September.

Popper’s philosophy of science may trace back to the 1919 Royal Society/Royal Astronomical Society report choosing among three choices for how much the positions of stars near the Sun appeared to shift with respect to their positions when their starlight was not bent by the Sun’s warping of space-time: no deflection (no weight for light), the Newtonian amount, or the Einsteinian latest prediction, which was twice the Newtonian amount. Much of the last half of the book is devoted to the pros and cons of various interpretations of the 1919 results about the amount of the apparent deflection, with reports on subsequent eclipse observations—which were all taken with a less favorable background of stars than had been the case in 1919, when the eclipsed sun was in front of a major star cluster—the Hyades—that provided many more stars close to the Sun than is usual at eclipses or even otherwise. As Kennefick writes, “Even those who accepted the results wished to have them confirmed by further observations…. The fact that no one was going to be in a position to repeat the experiment for several years is what set the eclipse results apart.” Further, the same scientists and technicians didn’t observe eclipse after eclipse, so single-point failures were new each time. The sociologist-of-science Harry Collins is cited as holding that “there is a considerable exchange of essentially nonverbal tacit knowledge, which is often required to achieve success in experimentation…. Just as you cannot learn to ride a bike from reading a book, Collins would argue that you cannot learn to perform an experiment from reading a scientific paper.” I have made a similar point in reporting how I watched the optical physicist James Baker tighten screws on a lens mount by hand, in our instrumentation that he designed for Donald H. Menzel’s and my 1970 eclipse expedition to Mexico.  Few of us have used that experience over and over at eclipses, improving the chance that the non-repeatable observations would succeed.

One name that popped out at me was that of Heber Curtis, whose result from the American eclipse of 1918 was so uncertain that his observatory director didn’t report it in London. It was mentioned that Curtis was one of two participants in the famous “Great Debate” a few years later on whether spiral nebulae are independent units of the Universe (that is, galaxies), or mere gaseous nebulae, and Curtis relied too heavily on incorrect observations by someone else, coming out on the wrong side of the issue. One remembers the victor, Harlow Shapley, from that debate (and its subsequent written form). I hope my scientific work is remembered more favorably than Curtis’s is.

The 1922 eclipse was observed by Americans, and gave a similar result, as did the 1973 eclipse in Mauritania, also observed by a team pushed by theoreticians. One scientist had earlier said, given the verdict in favor of Einstein’s theory, that “the necessity for this further campaign is now removed, and if any English observers are able to visit America this year it will be no doubt be merely as a return to their old love, the Corona.” Kennefick mentions Don Bruns’s 2017 attempt to measure the effect, citing his 2018 paper claiming an exact agreement with the Einstein prediction but not evaluating the result or any systematic errors.

Kennefick goes on to discuss other observations and experiments relevant to relativity, such as A. A. Michelson’s attempts to detect the ether through studies of the speed of light in different directions.  In that contact, he gives the famous Einstein quotation, “Subtle is the Lord but malicious He is not.” And though eclipse observations are not immediately repeatable, “occultations of quasars are observable anywhere in the world when they occur. Observers remain happily ensconced in their regular facilities and every year a new set of observations provide a chance to improve upon the accuracy of the measurements previous taken.” My 2009 article on “Solar Eclipses as an Astrophysical Laboratory” for Nature magazine (June 11, 459, 789-795) in the International Year of Astronomy refers to several quasar verifications of the Einstein deflection first observed by Eddington and Dyson.

A highlight of Kennefick’s book is a wonderful poem by Berkeley professor W. H. Williams, otherwise unmentioned but cited by S. Chandrasekhar, which, in the style of Lewis Carroll (of whose work Eddington was fond), starts, 

The time has come, said Eddington

To speak of many things

Of cubes and clocks and meter-sticks

And why a pendulum swings

And how far space is out of plumb,

And whether time has wings.

During the writing of this book, I was delighted to be reminded that the advisor to the advisor of one of my two Ph.D. advisors was Eddington, and that the other advisor of mine traces his academic ancestry back to Michelson. Further, I was amazed as I write to be given, by a cousin, a 1939 letter to an uncle of mine signed “A. Einstein,” thanking my uncle for his work in support of refugees. (Diana Buchwald, head of the Einstein Papers Project at Caltech, writes me that this is the 45th such letter she knows of signed and sent that day, and that she is keeping a database. The event was apparently part of Dedication Week of the United Jewish Appeal for Refugees and Overseas Needs.)

The 1919 observations are remembered by most people as Eddington’s expedition, though he had someone accompanying him on Principe and a parallel expedition went to Brazil. Even Dyson’s daughter wrote that her father, the Astronomer Royal, would be remembered mainly as having worked with Eddington, “People will say—‘Dyson? Oh yes—he was Astronomer Royal, when Eddington was Chief Assistant.’” And Chandrasekhar cites that Rutherford “later said to Eddington, “You are responsible for Einstein’s fame.”

Kennefick’s title, No Shadow of a Doubt, puns on the shadow of the Moon falling on the Earth that allowed Eddington and Dyson to see a solar eclipse (the title phrase was due to Dyson). Looking up, they photographed (and barely saw with their own eyes since they were so busy during totality) the silhouette of the Moon against the background of the Sun, much as astronomers using the Event Horizon Telescope revealed on April 10, 2019—in a public press conference akin to the 1919 London conference—another sort of shadow, the silhouette of a 6-billion-solar-mass supermassive black hole in a distant galaxy 55 million light-years away against a background of gas in that galaxy, again a spectacular verification of Einstein’s general theory of relativity. The image was above the fold on the first pages of at least The New York Times, The Wall Street Journal, and The Washington Post the next day, further enlarging Einstein’s public image and success in this 21st century.

Daniel Kennefick’s wonderful No Shadow of a Doubt has something for everyone, and I highly recommend it, especially in this centennial year.


Astronomer and author Jay M. Pasachoff is the director of the Hopkins Observatory and Field Memorial Professor of Astronomy at Williams College. He is a Visitor in the Carnegie Observatories. Williams College is home to the Gamma of Massachusetts Chapter of Phi Beta Kappa.


(Posted on 5/5/2019 )