Einstein finally set foot on British soil on June 8, 1921, descending from the White Star liner Celtic into Liverpool. He stopped there on his way back to Germany after a long trip to the United States to raise money for a Zionist organization. Freundlich joined him in Great Britain to help with translation (his mother was English). There was great excitement and anticipation about the visit, with lectures, receptions, and dinner parties planned everywhere. Of course, not everyone was happy to see him. Henry Brose sighed at the anti-Germanic screeds still put forward by “a few ‘irreconcilables’ who, needless to say, have never seen a field of battle.”
Einstein’s trip would be a whirlwind three days. Young women fainted as he entered the room. He gave lectures in German, with proceeds going to the Imperial War Relief Fund. At Westminster Abbey he placed flowers on Newton’s grave. In rooms full of dignitaries he was almost always the youngest person there. At a dinner party the Archbishop of Canterbury asked him what difference relativity should make to the way we thought about morality. He replied, “It makes no difference. . . . It is purely abstract science.” Einstein either did not realize or did not care that his host, Lord Haldane, had just written an enormous book claiming the exact opposite.
The newspapers followed every awkward step of the shabby scientist’s sockless feet. Depending on which paper you read, Einstein might be described as absentminded, patient, or playful, and wearing an “ill-cut” or possibly “well-cut” morning coat. Everyone commented on his black hair. Punch said his mane “takes its own course through space and is not subject to gravitation.”
Upon his arrival by train in London he was driven by car to Burlington House, the home of the Royal Astronomical Society, where an overflow crowd waited. As the door was opened for him, he saw a slim figure waiting at the end of the hall. Sharply dressed and preparing to preside over the event as president of the RAS was Arthur Stanley Eddington. They shook hands; we don’t know who reached out first. Three years after the cannons went silent, sixteen years after the first paper on relativity, Einstein had won his war. Relativity had triumphed. Einstein, reminiscing later, told Elsa that Eddington was “a splendid chap. He had to endure so much that I would have admired him even without his theories.”
* * *
WHEN EINSTEIN STEPPED on the stage at the RAS, he had already become something more than himself. He was a walking myth. He had become an icon of science for the entire world, a symbol of the best of what humanity had to offer. William Carlos Williams named him “St. Francis Einstein of the Daffodils.” His name was now synonymous with genius. Einstein’s mind had become disembodied (literally, after his death, when his brain was removed and eventually taken on a cross-country road trip).
But really he was an intensely embodied person, with stomach pains, enjoying cigars, scrunching his toes in the sand, falling in and out of love. Relativity, too, was a tangible thing (for a theory anyway). It came from clocks and rulers and elevators. Its equations were written in well-worn notebooks, crossed out, scrawled over, and remade. It became real on a hot, rainy rock off the coast of Africa.
Today we think of relativity as a fantastically abstract theory, just a few elegant equations on a page. But its emergence was a messy, tangled story. There was no single moment of discovery, no guarantee of fame, only years of struggle and failure and challenge. Einstein had to persist through failure and skepticism; he had to trust his friends. Eddington had to hold to his pacifism against overwhelming pressures; he needed to have faith in both God and physics. Any of a dozen turning points could have waylaid relativity, leaving Einstein no more recognizable a name than Lorentz or Noether, leaving the equations more curious than earthshaking. He might be remembered as one of a dozen or so people who contributed to early quantum theory; relativity might be mentioned as an odd side project of his. We like to forget how hard it is to do science, and how it could have been different. A simple story of science seems more true, more convincing.
We like to simplify the scientists too. The Einstein scholar John Stachel reminds us that the most persistent myth about Einstein was that he was born old. We imagine that he always had gray hair and a lined face, a grandfatherly sage beloved by all humanity. We forget the wartime Einstein, starving, scrappy, a socialist radical battling to make sense of his own ideas, much less persuading anyone else. Einstein was just forty years old in 1919. We project our elderly Einstein back in time because we want him to have always been the great sage.
But he wasn’t. Our mythical genius came out of bloody, devastating years of war. It was only in contrast to those horrors that Einstein’s triumph was so striking—a victory for pure thought, scientific beauty, and world peace at a time when civilization itself seemed to be in peril. Relativity’s sudden explosion, and Eddington’s zealous evangelism for it, would never have happened in quieter times. The theory had few applications for decades and, even if it had been confirmed, would likely have languished in dusty journals until cosmologists or GPS engineers realized they needed its delicate adjustments. Without the war, relativity would have been just one more theory, true but obscure. Without the war, Einstein would be just one more name for bored schoolchildren to memorize. Instead, his name is now an idea, an icon, a personification of everything we want science to be.
EPILOGUE
The Legacy of Einstein and Eddington
What kind of politics does science have?
THE 1919 ECLIPSE has lasted a long time. Not the eclipse itself; that was only a handful of minutes. Its legacy has lived on for a century. Every generation has used the story of Einstein and the eclipse to explain what science is, how it works, and what it means. Exactly what those lessons are has changed, and will change. But the 1919 eclipse has become an exemplar for the essence of science—good or bad.
* * *
IF YOU ASK a scientist today what makes an idea “scientific,” you are likely to get a blank look (that’s not a question they need to think about much). If you do get an answer, it will probably be something like this: an idea is scientific if it is falsifiable. That is, it is scientific if it can be proved wrong (by an experiment, usually). Science, then, is not about proving good ideas true—it is about proving bad ideas wrong. By eliminating all the inadequate ideas, scientists will gradually get better and better understandings of the world.
This position has the awkward name of falsificationism. It is extremely widespread among the people who actually do science. It is the brainchild of the Austrian-born philosopher Karl Popper, one of the most influential philosophers of science of the twentieth century. Like Einstein, Popper is one of those figures who seemed to have been born old—heavy jowls, well-retreated hair, elephantine ears. But to understand his connection to the 1919 eclipse we need to see the young, handsome, dashing Popper. Born in 1902, he was just too young to fight in the war but old enough to see its injustices. That drove him to become a Marxist by age fifteen. When he was seventeen, the 1919 eclipse spread the Einstein phenomenon across the world and he found a new intellectual hero. Young Karl heard the genius speak in Vienna the winter of 1919–1920 and found himself “dazed.”
It was not time dilation and curved space-time that startled him. Rather than the science, he was more interested in how Einstein talked about his science. What struck him was Einstein’s “intellectual modesty”—that the physicist had specified the conditions under which relativity could be refuted. No gravitational redshift, no relativity. No deflection of light, no relativity. This combination of boldness (these are my predictions, go check them) and tentativeness (my theory is only provisional and can be proven wrong at any time) enormously impressed Popper.
He had been dissatisfied with claims that Marxism was scientific for exactly these reasons. Marx’s predictions of revolution, it seemed, could only be proved right and never proven wrong. Any world event was claimed by the Marxists as confirmation of Marx’s ideas; there was no evidence that could convince them otherwise. Popp
er noticed something similar about Freudian psychology, one of the other great intellectual frameworks of the age. No matter what sort of dream you had, a Freudian would explain how it supported their theories. It was irrefutable. The Marxists and the Freudians seemed to have immensely powerful theories that could explain anything. Surely that made them scientific?
Popper had been frustrated by this, and Einstein helped him understand why. The mark of a good theory was not that it made predictions; the mark of a good theory was that it made predictions that could be refuted. A scientific theory should declare a rigorous, severe test that would show it is wrong. If light is not deflected by gravity, relativity is not true. Concentrating on evidence for a theory, as the Marxists and Freudians did, inevitably would lead one to only look for (and only see) what you want to see, rather than what is really there. That marked them as pseudosciences instead of true science. The ability to falsify a theory was how one found the boundary of the scientific (what philosophers call the demarcation problem).
So the 1919 eclipse was, to Popper, not about demonstrating that Einstein was right. It was a test to see if relativity was wrong. It passed the test and thus one could have provisional confidence in the theory. This became the exemplar of science for Popper’s philosophy. Otto Neurath said Popper turned Eddington’s experiment into “a scientific model.” Popper himself said that all he had done was “to make explicit certain points which are implicit in the work of Einstein.” All science, he said, should follow Einstein and Eddington’s model.
Popper’s falsificationism has become tremendously popular among scientists and science educators, particularly those looking for a clear standard with which to convince courts to keep creationists out of the classroom. It gave a model for how to do their own science (be like Einstein!). It also gave good reasons to keep doing science—since a theory cannot be ultimately proven, there is always more work for scientists to do. That gradual approximation to truth means there is, as Popper wrote, “no point of rest in science.”
The 1919 eclipse observations fit nicely into this scheme. Falsificationism demands that the theory be checked again and again, and that is indeed what astronomers did. The Lick Observatory repeated the observations at the 1922 eclipse. W. W. Campbell, perhaps embarrassed by his team’s unreliable results from 1918, carried out extensive preparatory work to find exactly the equipment and arrangement that would work best for measuring the deflection. The results strongly confirmed Einstein. Erwin Freundlich finally had the chance to carry out the test himself in 1929 in Sumatra. Each expedition could only show that relativity had been non-disproved once more. It was triumphant . . . until the next test.
Popper’s ideas are still regularly invoked by scientists to police radical ideas or shape their fields. Calling an idea unscientific is perhaps the most devastating critique one can level (remember such attacks on relativity), and falsifiability is a convenient way to do that. Nowadays cosmologists talk about multiple universes—but is that a falsifiable idea? Theoretical physicists have been pursuing string theory for decades—but they have yet to propose a Popperian decisive test to see if they are wrong. With Popper’s criterion—does this follow the model of 1919?—hypotheses can be discarded without further consideration, the whole trajectories of scientific research can be directed. Creationists have even tried to attack Darwinian evolution by claiming it fails Popper’s test of falsifiability (it does not; it is perfectly falsifiable).
Philosophers and historians often point out that despite the appeal and utility of falsificationism, it is deeply flawed. It doesn’t actually help define pseudoscience (astrology can be falsified by the experience of any set of twins). And it is not good at describing how scientists actually do their work. Despite the phrasing that so impressed Popper, Einstein actually was hoping someone would prove him right. Everyone talked about the 1919 results as proving relativity to be true. The later expeditions were indeed important. But the consensus was already that relativity was correct, not that it was a mere conjecture that should be checked.
By the 1950s and ’60s other philosophers had begun to point out that deciding whether a theory had actually been falsified was difficult. It is not always clear, as Popper hoped, whether a given experimental result actually refutes a theory. Thomas S. Kuhn’s famous Structure of Scientific Revolutions argued that one’s paradigm (the framework of ideas through which you interpreted the world) could actually change what you thought an experiment’s result was. An Einsteinian could look at the photographic plates and see the curvature of space-time; a Newtonian could look and see none.
One version of this problem of interpreting an experimental result, sometimes called the Duhem–Quine thesis, states that any conclusion about what a test is actually testing depends on lots of intermediary knowledge and assumptions. Were Eddington’s plates measuring gravitational deflection, refraction in the solar atmosphere, or the uneven heating of a coelostat mirror? We saw those issues explicitly discussed by scientists after the results were first presented. The issues are difficult and absolutely critical to convincing someone that your results mean what you say they do. Data does not speak for itself.
Some skepticism of the idea that the 1919 expeditions had truly confirmed relativity began to appear after World War II. This was when a new generation of scientists had come to prominence who did not remember Einstein’s canonization firsthand and when Eddington’s reputation had been damaged by his widely panned late-career attempts at unifying physics. They had some distance from which it was easier to ask certain questions.
In 1969 the astrophysicist Dennis W. Sciama wanted to celebrate a new era in relativistic physics in which tools such as radio telescopes could provide results undreamt-of by Einstein. Using radio telescopes is a much more precise way to measure gravitational deflection than eclipses (and you can do it anytime you want). By the fiftieth anniversary of Eddington’s observations, the mathematical tools available and the precision expected of an experiment had changed dramatically. The early age of measuring light deflection at a solar eclipse now seemed amateurish. It is an extremely challenging measurement to carry out and even the later expeditions still had a fairly high amount of error. Perhaps Eddington’s results had not been as decisive as once thought. According to Sciama their worldwide influence was “partly because the world was amazed that so soon after the Great War the British should finance and conduct an expedition to test a theory proposed by a German.”
The physicist C.W.F. Everitt went so far as to completely reject the 1919 results. He wrote that “this was a model of how not to do an experiment.” On “cooler reflection” from sixty years’ distance it seemed that the data did not support Einstein at all. Perhaps Eddington, Dyson, and Davidson had always intended to prove Einstein right and manipulated the results to that end. Everitt disputed the idea that there was anything reliable on those photographic plates. “Only Eddington’s disarming way of spinning a yarn could convince anyone that here was a good check of General Relativity.”
Even Stephen Hawking casually dismissed the 1919 results. The errors were, he said, as large as the effect they were trying to measure (like saying you’re sure it is Tuesday, but that it might be Monday or Wednesday). His Brief History of Time concluded that “their measurement had been sheer luck, or a case of knowing the result they wanted to get, not an uncommon occurrence in science.”
Did the photographic plates show what Eddington said they did? A scientist’s reflex in a situation like this is to go check again. So in 1979 (for the centenary of Einstein’s birth), the original photographs were pulled from the Royal Astronomical Society’s archives and checked with modern methods. It is rare for a scientific debate to go back to the original, raw data—but when it does, that is a sign of deep worry. Astronomers are obsessive record keepers for precisely this reason, though. You never know when you will need to go back and look.
The Royal Greenwich Observatory staff reanalyzed the
Sobral plates with computerized measuring equipment and checked whether the 1919 analyses had been done properly. For the four-inch telescope, Dyson had reported 1.98 ±0.18. The modern computer reported 1.90 ±0.11. For the astrographic, Dyson had reported the uncorrected value of 0.93, even though if he applied likely corrections for the distorted mirror he would have had 1.52. The computer, though, could apply those corrections much more reliably, getting 1.55 ±0.34. The new analysis gave a combined result of 1.87 ±0.13, solidly close to Einstein’s 1.75 prediction. It seemed that Dyson’s original analysis had been pretty good, and certainly did not show evidence of tampering in favor of Einstein. Two astronomers wrote a public letter refuting Hawking’s complaint, pointing out that the errors were well below the measured value. Rather than not knowing if it was Tuesday, the errors were more like “I know it is Tuesday between lunch and teatime.” It is true that these errors are large compared to what comes out of CERN today, where results are uncertain to about 1 in 3 million (called “five sigma”). Partly this is because scientific standards change over time—none of Newton’s experiments would pass muster now—and partly because the very first time something is measured, the result is always rough. Precision becomes much easier once you understand what you are looking for.
Einstein's War Page 36
