The reanalysis of the plates, though, left out a critical part of the events of 1919. Eddington and Dyson had made the case that the Sobral astrographic results should not be taken seriously because of the coelostat problems. They had been worryingly close to the Newtonian half-deflection; if they had been included it would have been difficult to claim a confirmation for Einstein. In 1980, two philosophers, John Earman and Clark Glymour, argued that the Sobral astrographic results should not have been excluded. If those results were dropped, then the Principe results (which were far from perfect) should have been dropped as well. Since Eddington did not do that, he must have been biased—Eddington admitted he thought relativity was true even before the expedition, and he had political reasons for wanting a positive result. Earman and Glymour conclude that Eddington only won the debates because he ended up writing the textbooks afterward. They reassured their readers that, while this might cause “despair on the part of those who see in science a model of objectivity and rationality,” it was not a deep problem because we had other reasons to think relativity is true. Indeed, today we have a dizzying number of different experiments confirming relativity—it is one of the best-supported theories of all time. The gravitational redshift can now be seen in any laboratory. The deflection of light is so well established today that it is used as a basic tool for exploring the universe (in the form of “gravitational lensing”). From the movements of galaxies to spinning spheres in orbit to the GPS system in your pocket, physicists keep looking for failures of relativity’s predictions and find none.
Earman and Glymour’s argument found its way to a much larger audience than typical for an academic paper thanks to a bestselling 1993 book, The Golem: What Everyone Should Know About Science. The authors were Harry Collins and Trevor Pinch, two sociologists interested in showing that science is a social construct—that is, its conclusions come from social and cultural processes rather than providing an objective perspective on the physical world. Their second chapter used Earman and Glymour to make an even stronger claim about relativity. It was not just that Eddington was biased; the confirmation of relativity was itself simply a social construct. The 1919 expedition, for Popper the model of how to do science in the most rational and reliable of ways, was now a demonstration of the impossibility of objectivity. Even the greatest of scientific feats—the experiment that made Einstein famous—showed that science was just another series of myths. According to this book, the importance that scientists placed on the expeditions did not come from revelations about the physical world, it was because “science needs decisive moments of proof to maintain its heroic image.” It was about the story scientists told one another.
* * *
IT IS NOT unusual today to meet physicists who have accepted Collins and Pinch’s (really Earman and Glymour’s) critique of the 1919 results. They will talk about error bars and bias. This version of the story has become a kind of scientific folklore, passed around at water coolers. The Golem was a fantastically successful scholarly publication (more than a dozen printings and many sequels) and its arguments have filtered out into many other books in many different disciplines. Very few physicists repeating this version of the story know where it came from, though. They would probably be quite shocked to learn they were passing along arguments aimed at undermining the very foundations of their field.
Generally scientists have strong negative feelings toward social constructivists like Collins and Pinch. The battles among them even earned a special term—the “Science Wars.” In the 1990s there were ferocious (by scholarly standards anyway) debates about whether theories like relativity should be seen as concrete parts of physical reality or as mere sociopolitical negotiations. One part of the question was whether fields like sociology and the humanities could have anything important to say about the natural sciences. Also under debate was the nature of science. What were the forces that made science work? Physical ones like gravity and electricity? Or social ones, like personal bias and political affiliation? And who was allowed to speak about science?
N. David Mermin, a Cornell physicist, wrote a reply to The Golem for Physics Today that walked an interesting line. He accepted that there were social elements to science (anyone who has set foot in a lab knew that). But he rejected Collins and Pinch’s exclusive focus on those social elements. Surely, he said, science could be the product of both physical and social forces.
Mermin did not follow up on this suggestion, but other scholars have. Daniel Kennefick has stressed that even though Eddington and Dyson made important decisions about what data to accept and what to reject, that does not mean those decisions were wrong. Context, he writes, is crucial for understanding any experimental result. The astronomers in 1919 had good reasons for attributing systematic error to the Sobral astrographic but not to the Principe one—and they presented those reasons publicly. The vast majority of scientists qualified to judge those reasons were persuaded that this was the correct choice by the standards of the day. They understood that recognizing the difference between good and bad data was hardly a controversial or suspicious thing. It was an ordinary part of doing science.
It is not always easy to tell the difference between good and bad data. Generally you need lots of experience producing and looking at specific kinds of data. In 1919, it was good to trust people who knew telescopes very well. Conversely, don’t trust what a particle physicist has to say about fossils. Direct experience produces a special kind of understanding. The Germans, naturally, have a special, lengthy word for this—Fingerspitzengefühl. Literally this is the feeling in your fingertips; more loosely, it refers to that special awareness that comes with long experience. It is how your car mechanic figures out what is wrong with the engine just by listening, or how the chef knows what spice to add a pinch of. It would be very complicated for either one to explain the specific details on which they made their decision, though your car will run better and your meal will be delicious. Science is similar. On May 30, 1919, Davidson and Crommelin glanced at the Sobral astrographic photo and immediately knew something was wrong. Their colleagues in astronomy were able to do the same. Recognizing good from bad was not easy; neither was it mysterious.
Everyone wants a simple explanation for why things turn out as they do. Popper thought the expeditions were extraordinary and made them exemplars of good science. Everitt thought the expeditions were biased and made them exemplars of bad science. Collins and Pinch thought the expeditions were shaped by politics and authority, and made them exemplars of socially constructed science.
Einstein’s War has been a story about how none of those are enough. Einstein and relativity’s victory involved good science, bad science, politics, and personal authority. Any episode in science does. None of those mean relativity is wrong (it has been confirmed many, many times since then) or that Eddington fudged the numbers (there were good reasons to trust the 1919 results). Science is done by people. That means it will be inherently complicated and often confusing. People will make mistakes, equipment will break, poor decisions will be made because of political or personal bias. But people will also have flashes of insight, they will have friends who make crucial suggestions, they will take up a cause because of political or personal beliefs.
We do not have to be forced into extremes. The presence of human scientists does not make science unreliable. We need to understand, though, what science-done-by-people actually looks like and how it works. That means leaving behind some comforting myths about the dispassionate, purely rational, always-objective nature of science. The deeply human, sometimes chaotic story of relativity is not an exception. It is an exemplar. Science is messy; it is also a powerful way to learn about the real world around us.
* * *
BEYOND DATA POINTS and error analyses, most everyone agrees about the broader historical importance of the 1919 expeditions. They were a great victory for the higher values of science. They showed that scientists could rise above petty
nationalism, that science could help one escape the shackles of nationalism and war. We saw how Eddington and Einstein intentionally spread this interpretation; they wanted to use the moment to change the way scientists were behaving. They never let up, either. After Dyson died in 1940 in the darkest days of World War II, Eddington used his obituary to again remind everyone that the eclipse expeditions had “opportunely put an end to wild talk of boycotting German science. By standing foremost in testing, and ultimately verifying, the ‘enemy’ theory, our national Observatory kept alive the finest traditions of science; and the lesson is perhaps still needed in the world today.” The expeditions were a model not just in the epistemological sense of Popper, but also in a political and moral sense.
Einstein and his international allies, September 1923. Back row: Einstein, Paul Ehrenfest, Willem de Sitter. Front row: Eddington, Hendrik Lorentz.
EMILIO SEGRÈ VISUAL ARCHIVES
This modeling sometimes becomes self-congratulatory. On the Einstein centenary in 1979 the distinguished British astronomer William McCrea gave a rousing speech celebrating how fortunate it was for Einstein that British science had held to scientific internationalism during and after the Great War. His generation had completely forgotten the brutal battles that had been fought over whether German science would ever be welcome on British shores. Eddington had been successful in portraying the expeditions as triumphs of internationalism, so looking back, it seemed to someone like McCrea that internationalism must have been the obvious and natural way to handle German science. In hindsight everything seems inevitable. Of course relativity would be confirmed; Einstein was a genius. Of course politics would not have gotten in the way; scientists always transcend nationalism.
Even those who attacked the expeditions’ scientific value acknowledged its importance for the postwar world. Hawking called them a triumph of international reconciliation in the same sentence that he rejected their data. Physicist Clifford Will in 1986 described the expeditions as something to strive for in his own era:
In our present time, when cold-war politics sometimes obstructs the free flow of scientific information and interaction, we would do to remember this example: a British government permitting a pacifist scientist to avoid wartime military duty so he could go off and try to verify a theory produced by an enemy scientist.
Again, the internationalism of 1919 was presented as something uncontroversial that all scientists would naturally accept. If Soviet and American scientists couldn’t get along, they should look to Einstein and Eddington as an example.
As always, though, “internationalism” is a complicated category that means different things to different people. To celebrate the ninetieth anniversary of the observations a plaque was placed on Principe at the very spot (still accessed by dirt road) where Eddington and Cottingham observed the eclipse. There was some controversy over who should install it—the British or the Portuguese? Was the nationality of the spot itself important or the nationality of the people who worked there? In echoes of the eclipse expedition equipment in 1919, the fifty-kilogram plaque was only allowed through customs because local officials made special arrangements. The plaque is a celebration of internationalism, though it is hardly postcolonial: some of the local children were herded away from the installation, being told it was only for whites.
The right connections between science and politics are not obvious. Some say they should be totally separate—scientists should not be involved with politics and vice versa. During the First World War this idea of separation was almost completely abandoned. Who was allowed to subscribe to a journal was a political question. Who received funding from the government was a political question. What words were used to describe an experimental apparatus was a political question. One might look to this and say that it was war, and nothing was working as it should. It was an anomalous kind of science.
It was not. The war only brought into relief political aspects that have always, and will always, be present in science. Wanting science to be apolitical does not make it so. Instead of ignoring the political aspects of science we would be better off acknowledging them so we can understand them. The question then becomes “What kind of politics does science have?” It is sometimes claimed that science has an inherent connection with certain political frameworks or viewpoints. The sociologist Robert K. Merton, for example, used to argue that science encouraged democracy. You can certainly find connections between scientific values and democratic ones (for example, freedom of speech). But it is not difficult to find connections to, say, anarchism too (for example, there is no formal centralized authority).
During the Great War we saw many different attempts to find these connections. Everyone who signed the Manifesto of 93 thought that the correct politics for science were those of the German Empire. Einstein thought socialism was the most natural politics for science. Eddington argued for internationalism. The practice of science did not carry with it an innate political orientation. Each group, each individual forged their own mixture of scientific practice, national identity, personal beliefs, and past experiences. It was obvious to each scientist that the politics they brought were the best ones for science; what their enemy brought was the worst.
There is not a correct answer here. Being a scientist does not carry with it a natural or default political setting. Politics will be in science regardless, as long as scientists are human. Scientists need to decide for themselves what political values they think are important for the work they do. If you think science needs many different viewpoints and a variety of life experiences, then you should fight for liberal values in science. If you think science suffers from too much government intervention, then you should fight for libertarian values in science. A scientist should not be embarrassed to state political views that they think will make science work better (Einstein certainly wasn’t). And there will probably be other scientists who disagree with those views (as Einstein discovered). A scientist who is wondering whether it is acceptable to lobby their representative, or whether they can march in a protest, should remember Eddington at his conscription hearing or Einstein crossing the revolutionary barricades.
Simply having political views does not make it impossible for someone to do science. Scientists are not emotionless machines, nor would we want them to be. We need scientists who care about things and are willing to take actions to support the things they care about. If Eddington had not cared about pacifism, we would not have had the relativity revolution in 1919. It was because people from all across the political spectrum cared about their science that the world had just the right conditions for Einstein’s sudden catapulting to fame. The horrors of the war, and pacifists’ reactions to them, forged the intricate, fragile network that made relativity what it was. The connections of science to the wider world—politics, religion, culture—are not trivial. How we think about science, how we link it to the other parts of our lives, changes the way science is done. We—scientists and nonscientists—need to choose what values and goals we want to bring to the scientific endeavor. Einstein did.
ACKNOWLEDGMENTS
The publication date on a book creates the illusion that it has a distinct moment of creation (kind of like scientific theories). In reality, of course, it grows slowly and organically. This is more true than usual for Einstein’s War, which relies on and integrates academic research and writing I have done over two decades. This integration might have, in some places, led to some unavoidable similarities with my earlier publications. My thanks to everyone who has supported my work over the years—your efforts made this book possible, even if you can’t see it.
There are a handful of scholars whose work I have relied on particularly heavily in constructing this story, whom I would like to mention even beyond the end notes. Of the nigh-infinite Einstein biographies out there, I have found none that match the balance of detail, clarity, and accessibility found in Albrecht Fölsing’s Albert Einstein:
A Biography, and I have used it as my foundation for this story. Similarly, John Keegan’s book The First World War strongly framed the way I have thought about the conflict. Hubert Goenner and Giuseppe Castagnetti’s research on Einstein’s political activism during the war has been extremely helpful, as has the work of Jürgen Renn, Michel Janssen, and everyone else who was part of the Genesis of General Relativity series. And this book would have been literally impossible without the army of people who have worked on the Einstein Papers Project over the years and made those letters and documents available both in print and online.
In a book of this type I am unable to provide comprehensive citations of all the relevant scholarly literature as would be expected in an academic book. I apologize to the many, many scholars whose work I have not explicitly mentioned here—so much has been written on both Einstein and the Great War that my notes can only scratch the surface. Where possible I have tried to cite widely accessible secondary sources such as The Quotable Einstein or Margaret MacMillan’s Paris 1919 rather than academic publications that may be harder for interested readers to find. Similarly, for the sake of consistency and ease of reference I have generally used published materials for translations from the German.
My deepest thanks to those long-suffering people who read this manuscript front-to-back in various incarnations: Janelle Stanley, Andrew Warwick, Graeme Gooday, Andrew Romig, Matthew Gregory, and Meredith Theeman. Their patience and feedback were essential to the book coming to exist as a coherent whole. David Kaiser and Michael Gordin read huge portions, and I am especially grateful to them for catching many of my stupid mistakes. Thanks to Guy Rader, Bruce Hunt, and everyone in the Department of History at University of Texas at Austin for early feedback on prototype chapters. Professors Damin Spritzer, Jeffrey Johnson, Carsten Reinhardt, Margot Canaday, Luis Campos, and Kitt Price provided invaluable information on specific topics. And I am grateful to Maya and Zoe Stanley for putting up with endless Einstein anecdotes over dinner.
Einstein's War Page 37
