A null result is not necessarily a failure. The present eclipse expedition may for the first time demonstrate the weight of light; or they may confirm Einstein’s weird theory of non-Euclidian space; or they may lead to a result of yet more far-reaching consequences—no deflection.
—Arthur Eddington, Observatory, March 1919
THE SECOND-CLOSEST STAR cluster to earth is 153 light-years away and is believed to be about 625 million years old. This rich group of stars is in the constellation Taurus, the bull, whose name dates back to ancient times, when bulls were worshipped in the Middle East. The bull’s head is marked by a V-shaped cluster known as the Hyades and is cataloged as Melotte 25. For the ancient Greeks, the Hyades were the daughters of Atlas, who was condemned to hold the sky above his head for eternity. Some legends have them as five sisters, but in reality, 15 stars in the cluster shine brighter than fifth magnitude and can be seen with the naked eye. Hold binoculars on the Hyades, and you will see more than 130 of its stars that are brighter than ninth magnitude. According to one legend, Hyas, a brother to the Hyades, was killed while hunting. His death made the sisters start crying from grief, and their tears turned to rain. Thus, the Hyades are known as “the rainy ones” and are associated with wet and stormy weather. It was this bright cluster of stars that would be closest to the sun during the 1919 eclipse.
The last thing that Eddington and Cottingham hoped for as they waited for the eclipse on Príncipe was rainy weather. On the other hand, in Sobral, rain would have been a gift from the gods for that drought-stricken area, despite certain defeat of Crommelin and Davidson’s scientific plans. Although May would ordinarily mark the end of the rainy season in northeastern Brazil, the area had seen scant rain since the astronomers arrived, other than a few brief afternoon showers. And brief showers during any drought meant only temporary relief. The impending natural disaster is most likely what had prompted writer Almeida Brito to criticize the very idea that two Englishmen were visiting in the midst of a devastating drought and yet were wishing for no precipitation on eclipse day. But just as Eddington hadn’t selected a path of totality that threw the moon’s shadow over six hundred “imported labourers” on Príncipe, neither had Crommelin and Davidson chosen a place of parched desolation that was killing thousands. As always, nature had made this decision all on its own.
Several newspapers in northern Brazil had published articles about the expeditions to Sobral. One was Correio da Semana, a Catholic publication pleased to inform its readers that both Andrew Crommelin and Charles Rundle Davidson were devout followers of the faith. Other papers encouraged locals to be on their best behavior while these esteemed guests were in town. Morize had also given interviews, reminding citizens that what was about to happen was a natural phenomenon. He acknowledged that “ignorant men of the wilderness” would panic and resort to beating on cans and drums—the can beating was to encourage plants to continue growing—all to dissuade “the infernal deities” not to destroy the sun god. “The truly civilized man, however, does not commit any of these absurdities,” he said. Morize also cautioned against the use of firecrackers since their bright sparks would interfere with the cameras when the scientists were taking photographs. During his visit in 1912, Eddington had noticed and commented that the setting off of firecrackers was “the regular Brazilian way of demonstrating.” But the local authorities had agreed to restrict the general public from entering the racetrack area during the eclipse.1
In the predawn hours of May 29, at Saboya’s house, Crommelin and Davidson woke to clouds filling the dark sky overhead. The partial eclipse would begin there at 7:46 a.m., so they needed to be up early to prepare. There had been rainfall in Sobral four days earlier, which locals had welcomed for the needed precipitation. Crommelin welcomed it, too, since the rain helped control dust at the racetrack. When they were taking experimental photographs, he and Davidson had noticed that the coelostats and clocks were sensitive to the winds that periodically swept over the grounds of the jockey club. From their experience at previous expeditions, they were well aware that gusts of wind during totality were not unusual, given the sudden temperature drop when the moon blocks out the sun. As the open racetrack was already vulnerable to wind, they had arranged for windscreens to be positioned around the hut in places where the screens wouldn’t interfere with the observations. The screens would also protect the parts of the telescope tubes that would be exposed to direct sunlight.
After an early breakfast, the Englishmen made their way out to the racetrack. Carnegie’s Daniel Wise, from Williamsport, Pennsylvania, stayed behind. He would spend those exciting minutes of totality in the basement, surrounded by his magnetic instruments. But Andrew Thomson, born in tiny Dobbington, Ontario, near Owen Sound, would make a direct observation on the racetrack, not far from Crommelin and Davidson. Since dawn, the churches in Sobral had begun filling up, jammed with superstitious souls who knelt and prayed for divine protection. The more courageous crowded into the town square to wait. One journalist wondered where the citizens had found so many little telescopes, the smoky shards of glass now well known at eclipse gatherings around the world.
Most of South America would witness this eclipse as a partial. The narrow path of totality, however, would begin before sunrise in the dark waters of the South Pacific. The moon’s umbra would move onshore at the southern tip of Peru and the northern tip of Chile. It would then cut across the center of Bolivia, clipping the southern reaches of the Amazon rain forest and traveling at an average speed of twenty-five hundred miles per hour, before it reached Sobral. As Davidson and Crommelin inspected their instruments, they had no way of knowing that across the ocean, Eddington and Cottingham were watching a heavy rainstorm beat down on Roca Sundy. This was a rare thing for that time of year on the island of Príncipe and the worst storm they had seen since arriving there a month earlier. But there was still time for a change in the weather, as the total eclipse would not reach Africa until the early afternoon.
But Sobral was now minutes away from totality, and the sky still heavily clouded. One newspaper, Camocin’s Folha do Littoral, wrote of watching the tall, slender figure of Henrique Morize, striding back and forth among the huts and tents, nervously waiting. After all the funding, all the months of preparation, and thousands of miles traveled, would he get the photographs he hoped for of the solar corona? At 7:46 a.m., the partial eclipse began. All the astronomers were in their places at the instruments, staring up at the cloudy heavens. Earlier, in darkness, the English team had loaded the plates for the sixteen-inch telescope, which Davidson would oversee. And the smaller plates had been loaded for the four-inch aperture, in Crommelin’s care. The smaller glass had been decided on almost as an afterthought when Cortie advised that it be taken along as a backup telescope, with the square brown box that served as its tube. Loaned to the Greenwich team by the Royal Irish Academy, it had a nineteen-foot focal length and had been used successfully by Cortie in Sweden, when the Jesuit was banned from Russia in 1914.
But it seemed at the outset in Sobral that neither telescope would be useful if the sun stayed hidden. At first contact, when the shadow appeared to touch the sun’s outer edge, the sky was nine-tenths covered in clouds. This was not an auspicious sign. The total eclipse would begin just an hour and thirteen minutes later. Of all the teams representing multiple nations that were spread out from continent to continent along the path of totality, only two would test for the light deflection predicted by one Professor Einstein. The first team was now clouded over on a dusty racetrack in Brazil. The second team was waiting thirty-three hundred miles away on an island cocoa plantation, watching it rain.
A minute before second contact, with the total eclipse seconds away, luck shifted for the astronomers. According to Morize, a sigh of relief was heard coming from all the spectators. “As totality approached,” Crommelin wrote, “the proportion of cloud diminished, and a large clear space reached the sun about one minute before second contact. Warnings were given 58s., 22s. and 12s. before second contact by observing the length of the disappearing crescent on the ground glass.” When the crescent vanished, Araujo, their faithful interpreter, shouted, “Go!” and set the metronome in motion. Araujo then called out every tenth beat as totality occurred. The exposure times were recorded in terms of those beats. The gusts of wind they expected did not arrive; nor did the temperature drop noticeably, as often happens when the moon’s shadow covers the earth below. The sky at totality was not as dark as Crommelin had witnessed during previous eclipses. It appeared as it might a half hour before sunrise. Everything around them was deadly calm except for the sounds of cameras clicking inside the canvas huts on the racetrack. For a minute during midtotality, a ragged cloud floated over the sun. Other than that, the astronomers were able to use about four of the expected five minutes and thirteen seconds promised them by nature. Davidson managed nineteen photographs with the Greenwich telescope, and Crommelin eight with the smaller four-inch lens.
The total eclipse was over at 9:03 a.m. local time. The shadow had long moved on to the coast of Brazil and was now crossing the Atlantic Ocean. The collective mood of the astronomers on the racetrack was upbeat, but no one would really know until the photographic plates were developed. Morize, so helpful to his fellow astronomers around the world despite speaking what Eddington called “the worse English I’ve ever heard” and a man said to be “timid and discreet” by his colleagues, nonetheless described this wonder of nature in very poetic terms:
At that moment, everyone, even the simple onlookers who surrounded the camp, felt moved by the magnificence of the spectacle that was manifesting itself. The sky darkened, becoming as before dawn, but without the gilded and reddish clarity of the East, and it was a dark blue that quickly brightened in the vicinity of the now completely blackened solar disk. Around it, one could see the crown of changing color, with shades of mother-of-pearl… on which stood out in intense red a beautiful protuberance, which is one of the largest that has been observed.2
This was the same magic that had frightened Paleolithic hunters and gatherers as they peered heavenward. It had amazed the early Chinese, Egyptians, and Babylonians as they made calculations and drew up their charts. It had summoned modern men and women of science to travel the globe, no matter the difficulties or the dangers. At 10:28 a.m. local time in Sobral, at fourth contact, the moon’s shadow disappeared from the sun and it returned in its full glory. It was finished, as if the magic had never happened. In just over two hours, the full eclipse would first reach the southern tips of Liberia, French West Africa, and the British Gold Coast before touching the tiny island of Príncipe.3
From Sobral, Crommelin sent a telegram to Dyson: “Eclipse Splendid.”
When Eddington and Cottingham had finished their breakfast on the morning of May 29, their counterparts were just rising in Sobral, in the dark hours of predawn, to prepare. If someone had given the six hundred “imported” workers a scientific explanation of what would occur that afternoon on Príncipe, if they were told that infernal demons were not devouring the sun god, as Morize had clarified in Brazil, would the explanation have brought them comfort? Would they have a better knowledge than did the citizens filling up the Catholic churches in Sobral? In the past forty years, the people of Africa had seen many partial eclipses. The path of totality for at least three total eclipses had swept over Angola, the homeland for most of the Roca Sundy workers. In addition, an annular eclipse in 1900 and a more recent one in 1918 had cut across the heart of that country, the moon blocking the sun’s center and leaving what would look like a burning ring of fire in the sky. The laborers on the plantation, whether free or not, had experienced the disruption of the natural world when the universe seemed to misbehave. They would have learned and repeated rituals passed down for centuries from their ancestors. The beating of drums was likely heard that day at Roca Sundy.
Eddington had no way of knowing that a layer of clouds was covering the sky over Crommelin and Davidson when he had welcomed excited guests to Roca Sundy that morning. Coming to his plantation to stay throughout the eclipse was Jeronimo Carneiro, the owner. With him were the curador and the judge, the two men with whom Eddington and Cottingham had enjoyed afternoon games of tennis. The British Mr. Wright from the cable station, who translated for the Englishmen, and three local physicians were also in tow. They arrived in the midst of a thunderstorm, with a tremendous downpour of rain at their heels. In England, the Hyades were once known as the April Rainers. But the deluge was most unusual for the gravana season. The rain didn’t stop until one o’clock in the afternoon, when rays of sun shone down. Eddington noted that the rain had probably cleared the sky, but the clouds soon collected again overhead. The total eclipse was due in two hours.4
It would have been an anxious morning for Eddington. With an annular due late that same year, and two partials coming in 1920, the next total eclipse would not occur until October of 1921. Useless to astronomers, it would reach its maximum in the waters of the Southern Ocean before touching land at Antarctica. The next total eclipse would not occur until September 1922, with five minutes, fifty-one seconds of possible observation at Ethiopia and Somalia, before it crossed the Indian Ocean to traverse all of Australia. If this eclipse of 1919 didn’t give the British an answer to Einstein’s prediction of light deflection—and if William Campbell was unable to secure positive results from Goldendale, Washington—then it was likely that a multitude of national expeditions would be spread out along the path in 1922 to tackle the relativity question.
Hoping for the best, despite the blanket of clouds overhead, Eddington and Cottingham had the Oxford telescope carried outside to the chosen spot beyond the bedroom windows. As had been done in Sobral, a canvas screen was arranged around the instrument to protect the lens and tube from wind and direct sunlight. The coelostat was readied onto the brick pier that had been built for that purpose by the workers a month earlier. Five hundred feet below, sloping down from the three-million-year-old volcanic plateau where the main house stood, the waves of the South Atlantic washed against the sand. This was the path the sun would take. The astronomers and their guests waited.
The total eclipse would occur at 3:13 p.m. At 1:46 p.m., the partial began, amid clouds. At 2:45, the sun’s crescent peeked in and out from behind them. At 2:55, the observers could see the shrinking crescent almost continuously, but still through drifting clouds. Around the sliver of sun, clear patches of blue sky now appeared. As totality began, the astronomers could do nothing but take photographs and pray for the best. Cottingham called out the exposures and managed the driving mechanism of the coelostat. Eddington changed the plate slides. “We had to carry out our programme of photographs in faith,” he wrote home. “I did not see the eclipse, being too busy changing plates, except for one glance to make sure it had begun, and another half-way through to see how much cloud there was.” They managed to take sixteen photographs in all. In five minutes and nineteen seconds, it was over. A few minutes later, the sun floated free in a perfectly clear blue sky. Soon, the mainland of Africa, where the partial had already begun, would receive the maximum eclipse. The shadow would then move on across several countries on that continent to die near sunset in the waters of the Indian Ocean, before reaching land at Madagascar.
With the clouds having cleared away before the full eclipse began, Crommelin and Davidson had taken quite a few photographs. Transporting glass photographic plates to the other side of the world was worrisome enough. But the handling of them onsite had to be done carefully to prevent any breaking or cracking of the glass. Because of the warm temperature in Sobral and lacking any supply of ice, development was carried out during the late hours of the night leading up to dawn. The plates had to be loaded in total darkness, and they were now unloaded the same way. A safelight could be used to watch the times during development. With locally made earthenware pots as water coolers to lower the temperature—it was seventy-five degrees just before daybreak—Davidson began the job the first night after the eclipse. For added cooling, the developing dish was placed inside a larger dish filled with melting hyposulphite of soda. Because the films would soften in the warm solutions, formalin was used to harden each one and minimize any possible distortion.5
Before the teams left England, Eddington had ordered two brands of plates for the expeditions, Ilford and Imperial. When Davidson had developed the check plates days earlier in Sobral, he had already discovered that only the Ilford plates would respond well enough in the warm temperature. He and Crommelin began with the Greenwich telescope, which had taken nineteen photographs with its wide-ranging field of view.6 By 3:00 a.m. on May 30, they finished the first four of the larger plates. When they were examined, they showed twelve stars. But the images were out of focus, the stars fuzzy and diffused. Since this was not the case for the check plates that had been done on nights before and just after the eclipse, the astronomers concluded that the sun’s heat on eclipse day had affected the curvature of the coelostat mirror. The several star images on the remaining large plates also turned out to be hazy.
The eight-inch mirror feeding the four-inch lens, however, had performed well. This smaller telescope, with its eight-by-ten-inch plates—the exposure times were twenty-eight seconds—succeeded in getting eight photographs. Of those, all but one showed seven well-defined stars. The single flawed photograph had been rendered unusable because of cloud covering. Although the solar corona was not the subject of attention for this expedition, the plates still revealed some detailed, even spectacular images. Crommelin noted that to the naked eye during the eclipse, the corona appeared three times brighter than the moon. The two astronomers worked diligently over the next few days, as did the Carnegie team. When both Davidson and Wise came down with a fever during this time, this symptom was most worrisome, given the many cases of yellow fever in the vicinity. Protecting the visitors from this disease had been Saboya’s foremost concern in hosting the foreigners. A doctor was quickly called to the house and pronounced the fevers as symptomatic of the common cold. The men went back to work.
By June 5, Crommelin and Davidson were finished with developing the plates. They were now exhausted from the late-night work and the taxing climate. Before they would take a second set of check plates in mid-July, they were ready for the cool ocean breezes at the very urban Fortaleza, the state capital of Ceará. On the racetrack, workers dismounted the driving clocks after Davidson had marked the mirrors so they could later be returned to their original positions. They were then carried into Saboya’s house to keep them safe from the blowing dust. The telescopes and coelostats were covered over and left in place in the hut. On June 7, Crommelin and Davidson bid their colleagues farewell since Wise, Thomson, and Morize and his entourage would be gone by the time the two Englishmen returned to Sobral. The chauffeur had already been careening around town to use up the oil in the Studebaker’s engine so that it could be later stored. The two men boarded the train and rode north again to Camocim. There they caught the steamer Turyassu on to the coastal city of Fortaleza. They needed to recharge.
Since their housemates Wise and Thomson had spent a month in Fortaleza before the eclipse, the Englishmen knew what to expect. The city had a population of seventy-five thousand, numerous cafés and cinemas, and green parks and gardens that overlooked the sea. The brightly colored mud-brick houses with their red-tiled roofs hugged the streets, the windows lit by electricity. Quality shops offered handmade Brazilian laces made on wooden bobbins and spindles. There were plazas, town squares, and cathedrals. Pagoda-shaped refreshment stands sold mangoes, dried bananas, green coconuts, and palm wines. Even automobiles existed in Fortaleza, and they didn’t frighten pedestrians when they rolled past, a wild driver at the wheel. Electric tram tracks were laid out for transportation, and bands played music at night in the open-air plazas. Compared with the five weeks they had just spent in Sobral, a mere 125 miles west as the crow flies, this was heaven. With the hotels packed because of scheduled government meetings, the Englishmen were invited to stay as guests at the Catholic Seminário da Prainha, the beach seminary, with its two spires rising above the city. They settled in to rest and recuperate.
Because Eddington had brought along the micrometer, which could make precise measurements, his intention was to evaluate the plates on Príncipe, rather than risk damage or loss during the long steamer journey home. In their makeshift darkroom, on the evening of the eclipse, he started developing, working on two glass plates per night, with Cottingham assisting. Things began well enough, despite the water temperature remaining at a warm seventy-eight degrees. For the comparison plates they had done after arriving at Roca Sundy, they were fortunate to have access to ice, supplied by Gragera. But the supply had trickled out soon after they began work on the actual eclipse photographs. Thus, as had occurred in Brazil, Eddington realized that of the two brands of plates he had brought in sealed cans, only the Ilford Special Rapid could withstand the warm temperature. Again like Crommelin and Davidson, he was left using formalin to harden the film. And alcohol was necessary to speed up the drying of the plates.
Over the next six nights, Eddington developed twelve of the sixteen photographs. Since four were taken with Imperial plates, they would have to go back to England to be developed there. It was a grueling job in the island heat. Accustomed to the mild maritime climate back in England, he wiped sweat from his eyes as he stared down at the images appearing in the dishes before him. But night after night, the results were disappointing. There were some very good exposures of the sun, but not of the all-important other stars in its vicinity. Because of cloud interference, there were almost no star images on the first ten plates. Judging from the last two, Eddington assumed that the clouds had cleared away considerably during those final seconds of totality. The star images on those plates appeared to be much better. One was quite good, and the other looked hopeful.
On the seventh day after the eclipse, as Cottingham peered over his shoulder, Eddington continued to measure the plates. “The cloudy weather upset my plans and I had to treat the measures in a different way from what I had intended,” he wrote in a letter home. He was already coming down with a fever from the strenuous work. But he needed to know if he had been done in by clouds. Or was there an answer waiting for him on at least one plate? He brought out the pre-eclipse plates they had taken in mid-May, for comparison. As he studied both sets, rough as the measurements were at that point—there would be ample ones done in England under Dyson’s supervision—Eddington thought he saw an answer. Like Einstein, he had already seen it in the mathematics. Now he hoped to see it in the stars. “Consequently,” he wrote, “I have not been able to make any preliminary announcements of the result. But the one good plate that I measured gave a result agreeing with Einstein and I think I have got a little confirmation from a second plate.” It was his first inkling that this eclipse expedition would change the rest of his life.
It had been over two hundred years since Newton asked his question “Do not bodies act upon light at a distance, and by their action bend its rays?” What Eddington saw on the good plate was light from a star bending at 1.61 arc seconds. It was not Einstein’s full prediction, not yet, not until all the plates had been examined carefully back in England. But Eddington had just become the first human being to stare down at the physical proof. Light did indeed have weight. Before him was an answer captured on a glass plate from a distance of ninety-three million miles. It would be an answer formed near those pearly white streamers of the solar corona that were flowing backward into space for millions of miles, an answer born in the gravitational pull of our closest and most brilliant star. Yes, bodies do act on light! It might have been Charles Perrine who saw it first, in 1912, but for clouds. Or William Campbell and Perrine, in 1914, if not for clouds. Or Erwin Freundlich, also in 1914, if not for war and clouds. It was Arthur Eddington, in 1919, who saw the proof.
The intent of these scientists, including Einstein, had never been to overthrow Newton. But thanks to Einstein, gravity had just become far more exciting than humans had ever imagined. From outside the plantation’s open windows came the sounds of a tropical jungle, where monkeys stole the golden cocoa pods and where parrots squawked from the canopies overhead. It was a world most unlike the quiet farmlands and green meadows of England, where a young boy had counted the bright stars over Somerset and first fell in love with astronomy. Light has weight! With this glimpse of success, Arthur Stanley Eddington turned to his team member and said, “Cottingham, you won’t have to go home alone.”
Then he sent a telegram to the anxious astronomer royal, who was waiting for word back in Greenwich: “Through cloud. Hopeful.”7
When the Portuguese government and the shipping company couldn’t agree over passenger rates, boats had been prohibited from leaving Lisbon. A threatened strike would soon cut off steamers from Príncipe. That put an end to Eddington’s plans to measure all the developed plates while he was still at Roca Sundy. If he and Cottingham didn’t leave the island by June 12 on the SS Zaire, they would likely be trapped there until August 1, possibly longer, depending on when the next boat arrived from Lisbon. It was again due to the goodwill of the visiting country that two tickets were “commandeered” by the governor of Príncipe. As Crommelin and Davidson were relaxing in the cool breezes at Fortaleza, the instruments were hurriedly dismantled and baggage packed up at Roca Sundy. The fragile photographic plates were again sealed in tin boxes that would hopefully keep them safe on the journey home.
Meanwhile, back in England, the Observatory journal out of Greenwich had stopped the monthly presses to add some new information submitted by the astronomer royal. He had received cablegrams from the two expeditions. Crommelin and Davidson had had perfect weather; Eddington and Cottingham were hopeful. A later message from Sobral informed the astronomer royal that the Cortie lens had captured seven good images.
Eddington and Cottingham had often wondered how their counterparts were getting on over in Brazil. Had the weather been kinder in South America? Although they wouldn’t know about the Observatory article until after they arrived home, a telegram from Dyson assured them that the Greenwich party had been successful in getting photographs. Eddington was relieved. While he felt strongly that the Einstein prediction was correct, he would only know for certain after he and Cottingham had returned to England. On June 12, five weeks after the Portugal had arrived at Santo António, the SS Zaire sailed out of that harbor with the two Englishmen on board.
On June 2, while Eddington and Cottingham were still developing the photographic plates at Roca Sundy, William Campbell had replied to a letter from Arthur Hinks. Hinks complained that the war had thwarted his ability to stay abreast of his own work, let alone take up with a complex new theory like general relativity, which “is beyond the limits of my comprehension.” Having already lost two years since Einstein’s final publication in 1916, Hinks admitted that he would be “hopelessly outclassed.” Hinks was the man who had resigned from Cambridge rather than become an assistant to Eddington but who had nevertheless done his job in securing information about the island of Príncipe for the expedition. Many scientists the world over would understand this loss of time given to the war effort—time taken away from advancements in any discipline.
In his reply to Hinks, Campbell commiserated that “most astronomers” would probably feel the same way. While he was not the strong believer of general relativity that Eddington was, Campbell was nonetheless a committed, curious, and open-minded scientist. “I have not attempted to go through the mathematics,” he wrote, “but the applications have interested me very much in a general way.” He then acknowledged Eddington’s “valuable service in keeping us posted on the applications and implications.” While Campbell still preferred that the Goldendale results would disagree with Einstein, he would continue with an open mind. He hoped he could announce the findings soon.8
Meantime back at Lick, H. D. Curtis had just begun to measure the Goldendale plates as well as the Vulcan plates that Campbell and Charles Perrine had obtained during the 1900 Georgia eclipse. He believed that the 1918 Lick team had captured star images that were measureable. As Eddington and Cottingham were sailing away from Príncipe with their own plates, Curtis believed he had an answer. Time and again, measure after measure, by dividing stars into groups closest and farthest from the sun, he had arrived at the same conclusion. Einstein’s final calculation of 1.7 arc seconds was wrong. As for Einstein’s earlier, Newtonian value, Curtis could “less definitely” pronounce against it. But with more measuring, he finally did just that by proclaiming, “There is no marked deflection of the light ray when passing through a strong gravitational field,” and thus, “the Einstein effect is non-existent.”9
Few scientists would go into mourning over this news, especially in Germany. The well-respected American astronomer George Ellery Hale, to whom Einstein had written in 1913 asking if photographs of stars near the sun might be obtained in daytime, was, like Campbell, ready to welcome negative results. To be sure, Hale was “really delighted.” He had attended a Pasadena meeting of the Astronomical Society of the Pacific, where Curtis’s paper was read. As Curtis had not been in attendance, Hale wrote soon afterward to congratulate him. “I confess that I am much pleased to hear that you find no evidence of the existence of the Einstein effect,” he wrote. The paper was submitted to the society’s journal Publications, to appear in its next issue. The Lick Observatory was ready to share its results with the world.
In June, Campbell headed a small delegation of American astronomers that traveled to Washington, DC, for a meeting. This was a preliminary gathering for a later conference in Brussels assembled to form the International Astronomy Union.10 Before the Brussels event, the delegation sailed to London to attend a special RAS meeting held in their honor. After thanking the Americans for their help in the war effort and praising their contributions to astronomy, Alfred Fowler, the president, asked Campbell to address the members. Campbell began with “the Einstein problem.” He detailed Curtis’s method of measurement over the past months—Curtis had sent him a telegram while he was still in Washington, DC—and acknowledged that the absence of the forty-foot telescope, which was still in Russia, had hindered the Goldendale observation. “For the one we used,” Campbell said, “the stars were too faint and in the long exposure required we suffered from the increased extent of the coronal structure.”
Nonetheless, Curtis had not found a displacement that would support Einstein’s 1.7 arc seconds prediction. In Campbell’s own opinion, he told the listeners, “Dr. Curtis’s results preclude the larger Einstein effect, but not the smaller amount expected, according to the original Einstein hypothesis.” This observation should be astonishing enough, that Campbell could not rule out that light had weight. All the predictions that had been formulated so far, coming from Newton, Soldner, or Einstein, had been based entirely on mathematical calculations. But it’s likely that most of the scientists in the room that day simply rejected Einstein’s description of gravity as a property of space-time. They saw no need to improve on a Newtonian universe that was static and unchanging. Campbell, however, remained cautious.
Dyson spoke next. He confessed that the question Einstein had put before them was not an easy one to answer. But it was certainly a timely one, given the two British expeditions. Eddington and Cottingham were just changing ships in Portugal, three days from arriving on English soil with the Príncipe plates. And Crommelin and Davidson, back from their ocean-side retreat, had just that morning in Sobral begun taking the comparison plates. Dyson then relayed what he had learned from Eddington in a recently received letter. Eddington was disappointed with the plates he and Cottingham had taken “through cloud.” Of sixteen, only the last six had captured any star images, and five of those were flawed. “From his best plate, however, he has some evidence of deflection in the Einstein sense, but the plate errors have yet to be fully determined.” What a quandary for two leading astronomers, from famous observatories in two countries, to be in when seeking the answer to such a monumental question on how the universe worked.
One by one, the remaining eight American astronomers were introduced, and each spoke of his current work.11 The war was still fresh in everyone’s minds. Walter Sydney Adams, of Mount Wilson, apologized for the late US entry into the fight. When the Americans had each presented, Turner was asked by the president to give a vote of thanks to the visiting delegation. Turner mentioned the “wonders” shown to British astronomers by the Americans in 1910. He then replied to the remark made by Adams: “Prof. Adams said that they were later than he had hoped in coming to join in the war. That may be, but, at any rate they are not late in the work of reconstruction.” He then referenced the passing four months earlier of “a great astronomer,” Edward C. Pickering, famous not just for his work and his all-important Harvard bulletins, but also for his female computers.12
When Campbell closed the meeting by standing to recognize the vote of thanks, he also turned back to the scars left by the war. Early on, he had criticized President Wilson’s hesitation to enter the fight, even though Campbell’s pilot son would be shot out of the skies over France. Albert Einstein’s German birth was never mentioned, but the fact seemed to hang in the air. With his wife in attendance, Campbell gave a brief parting speech worthy of the stage:
We have come to assist in reorganizing international astronomical relations, but fundamentally there is something deeper than that in our meeting. We have come somewhat in protest against certain conditions that have existed in international relations, conditions arising from the over-development of militarism. We are not in sympathy with the idea that nations can do no wrong, and that they cannot act towards one another as individuals in a nation are expected to behave to each other. It chances that certain astronomers may be criticized and cut off somewhat from others. We feel, however, that it is clearly our duty to be men first and astronomers later.
As the days passed, Campbell considered the possibility that the Lick Observatory might be jumping the gun on arguing against the Einstein prediction. He had a nagging sense that they might have made errors, despite Curtis’s steadfast conviction that Einstein was wrong. From London a few days after Dyson’s talk, Campbell cabled Curtis, telling him to use caution. Curtis wasted no time in writing to the editor of Publications that same day, asking that his paper be held back from the forthcoming issue. The possibility of miscalculation, as Campbell had reminded him, was too large. They would need more accuracy. Curtis would obtain comparison plates in August that would hopefully lessen the scope of error. A few days later, after arriving in Brussels for the IAU meeting that would install both him and Dyson as vice presidents, Campbell again cabled Lick at Mount Hamilton. This time he took no chances: “Delay publishing Einstein Results.”
The cold March night that Eddington and Cottingham had shared with Dyson at Flamsteed House, in front of a roaring fireplace, now seemed a lifetime ago. They were eager to see their colleagues and loved ones again. Most of the English they had heard spoken for the past three months they had spoken to each other. “It seems ages since I started off in a rush in the taxi from the Observatory,” Eddington wrote to his mother. He had received no mail from home since a letter written March 28 had arrived in Príncipe on May 11. Coincidentally, they would pass a ship at St. Vincent, in the Cape Verde islands, carrying the next batch of letters from England. He would have plenty of days before the winter snows to go cycling through the countryside or hiking with his friend Trimble. He missed his world at Cambridge, the observatory with its groomed grounds, his sister’s garden with its chickens and bees. Edwin Cottingham, too, longed for his house and his beloved shop at Thrapston. He was eager to be back amid the clocks and watches.
On board the SS Zaire with them was Jeronimo Carneiro, finished now with hosting the esteemed scientists at his private home and at Roca Sundy. He was headed back to Lisbon for three months, to his exciting world of bullfighting. Four English missionaries serving in Angola, one of them a Quaker, were also aboard the steamer. The ship was not up to par with the Anselm or even the Portugal. It was overcrowded for one thing, which is not surprising, given the dispute and the impending strike. “Of course, a lot of passengers have been ill,” Eddington wrote while at sea. “It is very bad for them being so crowded on the boat. There are lots of children and in some cabins there are as many as seven people. There are three in our cabin—a Portuguese and Cottingham & myself… I shall look forward to the strawberries, which are better than anything they have in the tropics.” Eddington was still ill with a high fever. One night while walking on deck, he fainted. The fresh sea air soon restored him. Yet it must have been a floating paradise compared with the ship that Henry Woodd Nevinson had taken to São Tomé fourteen years earlier, its belly filled with over three hundred captive human beings, fifty of them crying babies.
Eight days out of Príncipe, there was a stop in Praia, the capital of Cape Verde. After changing ships in Lisbon, they reached Liverpool on July 14, just a month after the bloody race riots had taken place there. The men would soon learn that while they were away, the whole world had erupted in turbulence, with strikes, violence, and revolutions flaring up. In England, fighting men had come home from the war—at least those who were lucky enough to have survived—to find unemployment and even poverty waiting for them. West Indians and Liverpudlians clashed. Liverpool was a port city that had grown rich in the mid-eighteenth century on the flourishing slave trade. Its slave-dealing merchants and ships dominated the transatlantic route that had once included São Tomé as a principal stopping port. Liverpool had helped make Maria Correia, whose castle ruins Eddington and Cottingham had visited on Príncipe, wealthy beyond her wildest dreams.
Despite the upheaval in a world recovering from war, Eddington and his traveling companion, the “pottering” clockmaker from Thrapston, had finished photographing stars off the coast of Africa. They were now safely back on English soil.
Crommelin’s and Davidson’s visit to the coast had been courtesy of the Brazilian government. Again, all their needs would be met, gratis, and the same had applied to the Carnegie team. Unlike the isolation of Eddington and Cottingham, marooned as they were on a tiny island and therefore limited to castle ruins and monkey hunts, Davidson and Crommelin socialized with British citizens then living and working in the vibrant city of Fortaleza. After a relaxing month, they returned to Sobral, arriving on July 9. The big house they had shared with the others was quieter now. Wise and Thomson were already doing observations elsewhere in Brazil, and the Rio de Janeiro team was just sailing past the Abrolhos island chain, with its dangerous reefs, and were only two days from home. This eclipse expedition would be Morize’s last.13
The English team was anxious to obtain the check plates so that they could return to Greenwich. Davidson and Crommelin had been gone from their families for over four months now, and the long journey home would take another five weeks. They wasted no time in uncovering the telescopes and coelostats and having the clocks and mirrors remounted. The instruments were now in the same positions they had been on May 29. On July 11, the men began photographing the eclipse field, now distant enough from the sun that they worked in the premorning hours. Crommelin had estimated that twenty-five minutes before sunrise would imitate the same brightness of the sky during the peak of totality. For a full week, they photographed. The local temperature was unforgiving, compared with the cool sea breezes on the coast. It was demanding work.
By July 18, three days after Eddington arrived back at Cambridge and Cottingham was at work in his beloved clock shop, Crommelin and Davidson felt assured they had enough plates. Again, the dusty racetrack was filled with workers and porters to begin the job of breaking down the equipment and packing it up. The glass plates went back into the tin cans to travel under Davidson’s legendary care. The heavier packing cases would be left behind. Brazilian agents who had been assigned to help with baggage would see to their shipment on a later boat. On July 22, the Englishmen thanked their gracious host who had been so vigilant in keeping them safe from yellow fever. And who had made sure they had at their disposal “a motor-car, the first that had ever been seen in Sobral.” In the RAS paper they published later, they forgot to thank the chauffeur from Rio.
Crommelin and Davidson said goodbye to the local agents and helpers who had assisted them over the drought-laden months. Then they rode the train north again to Camocim, where they boarded the Fortaleza to Maranhão and on to Pará. They were back to where they had taken their thousand-mile journey up the Amazon. There they booked the Polycarp headed for Liverpool. It was loaded with the cargo one would expect on a steamship leaving northern Brazil: 900 tons of cork, 500 tons of Brazil nuts, 80 tons of rubber, and 50 tons of animal hides. Also put on board the ship were two marmoset monkeys and two parrots. The animals had been bought by Charles Davidson for his children, monkeys for the boys and parrots for the girls. During his and Eddington’s 1912 rained-out expedition to Brazil, Eddington had written home about these monkeys for sale: “Some of the passengers bought little marmoset monkeys at Bahia; they are sweet little things that you could put in your pocket but I was not tempted to go in for one.” He had written this at the beginning of the expedition. Perhaps this is where Davidson had gotten the idea of taking exotic pets home to England at the end of the 1919 expedition. In late July, he purchased the monkeys and parrots before boarding the Polycarp.
On August 25, he and Crommelin were back on English soil. They caught a train from Liverpool to London, and then home to Greenwich. The England they had left behind in white drifts of snow was still green with late summer.14
As Crommelin and Davidson were arriving at Liverpool, William Campbell’s steamer had almost reached New York, his meetings in Europe finished. After a train ride across the country, he was finally back at his small brick house on Mount Hamilton. By mid-September, the Lick Observatory still had no results they could release with confidence concerning the 1918 eclipse. And now the 1919 measurements were under way in England. Eddington had been evaluating his Príncipe results, using the comparison plates that Frank Bellamy had taken that January and February at Oxford’s Radcliffe Observatory. But he believed he still needed the Sobral input before he could more accurately declare for Newton, or for Einstein. Davidson and Herbert Henry Furner, a computer at the Royal Observatory, were now busy measuring the Sobral plates, each man doing it independently to double-check their results. An answer from the astronomer royal would be forthcoming when the time was right. The days dragged on.
In Germany, meanwhile, Albert Einstein had been publishing a flurry of papers on general relativity. At the end of 1918, a third edition of his book had been printed. In it, he undertook to answer a question about his discovery of E = mc2, the equation that relates energy to mass. In the wake of his now completed general theory, would the 1905 equation remain unchanged? He was gratified to learn that it would hold up, and without any modification when considered from the perspective of general relativity. This prolific outpouring continued on throughout the months of 1919. The titles of his papers, aimed solely at the scientific community, show that his major focus was on his finalized theory, now almost three years old. He had learned in mid-June, again thanks to colleagues in Holland, that the British expeditions had obtained usable plates. But what would those plates reveal?
Einstein had focused on his personal life, too, or at least his cousin Elsa had. There had been a monumental event since his divorce became final that spring. On June 2, the day Campbell wrote to Hinks about general relativity and as Eddington was developing the first plates at Roca Sundy, four days after the eclipse, Albert and Elsa were married. Albert had ignored the promise he had made on his divorce papers not to remarry for two years. He moved into the apartment Elsa shared with her two grown daughters. They soon settled into the kind of middle-class life he had never envisioned for himself when he was still in love with his “Dollie.” But Elsa was far removed from the more bohemian Mileva Marić, at least the way Mileva had been in her early years. The new Mrs. Einstein was satisfied to be the wife of a well-respected scientist, seeing that he ate well and was cared for domestically so that he could continue his important work. They converted the attic rooms into a study where he would have solitude. It seemed more like a business partnership than a romantic relationship. Albert was forty years old now, and Elsa three years older. The bride’s bedroom was at one end of the hallway, the groom’s at the other.
Einstein had received at this time a generous invitation to move to Holland and accept a position at the University of Leiden. Colleagues such as Paul Ehrenfest, Hendrik Lorentz, and Willem de Sitter would welcome him there with open arms. They assured him that he would be surrounded by people who loved him for who he was, and not just for his “brain juice.” On September 12, Albert wrote an appreciative reply, turning down the “fabulous” offer and stating that he was surrounded by fellowship in Berlin “and this not only from those who lap up the droplets I sweat from my brainy brow.” But in a serious tone, he could not turn his back on Max Planck and those who had enticed him to Berlin in 1913. He felt a staunch loyalty and would not forgive himself if he abandoned their trust. “I feel like a relic in an ancient church; the old bones are quite useless, and yet…” He ended with an important question. “Have you by any chance heard anything over there about the English solar-eclipse observation?”
That same month, Erwin Freundlich left Potsdam and went to southern Germany on a much-needed vacation with his wife. On September 15, he still took time to write a lengthy letter to his esteemed mentor back in Berlin. He confessed that he was “cut to the quick” after hearing from his editor that Einstein had suggested changes to the third edition of Freundlich’s book on general relativity. The foreword for the book had been written by Einstein himself. Freundlich wondered if he hadn’t expressed himself properly, for he could not forgive himself if it were “a fundamentally erroneous interpretation of specific things.” He promised to remedy the situation when he returned home. The letter then mentioned money, as Freundlich’s letters often did. “It is an uncomfortable thought for me that by my request for a raise in salary—although it was determined only by very exceptional financial circumstances—I burden your institute’s budget so heavily. With the best of intentions, I cannot give back more than I am dealt.”
While Freundlich’s appointment was with the Kaiser Wilhelm Institute, he nonetheless did his experimental work related to the general theory at the observatory in Potsdam, whose director was Gustav Müller. “I believe I am not mistaken,” Freundlich wrote to Einstein, “when I say that all the gentlemen at Potsdam, even Director Muller, want to cover their rear on the point of the gen. theory of relativity and not advocate its verification any more than by allowing you informally to grant me the opportunity to work independently at their institute.” Apparently Freundlich and “the gentlemen at Potsdam” had read an account of Campbell’s remarks at the July RAS meeting in London. Details of the meeting and the speeches by the Americans had been published in journals that August, a brief article in Nature and a more detailed account in the Observatory. “I noticed this shortly before my departure,” Freundlich continued, “just in little things, such as the reception of a short note by Campbell at the Lick Observatory in which he reportedly could not detect the effect of light deflection at the solar eclipse of 1918 (or ’17).… The solar eclipse of 1917 or ’18 had, to my knowledge, only a very brief totality, and the weather conditions during the same were, as far as I have heard, not favorable either. Thus C.’s result carried no great weight, especially considering that C. is not the right man for doing such a thing. I just saw once again how cowardly the majority really are.”
It’s this part of the letter that is most arresting and singular, the unbefitting remark about Campbell’s résumé aside. If one needed further proof that communication among scientists during wartime was impossibly curbed, it’s this exchange between Freundlich and Einstein. Freundlich didn’t know which year the total eclipse had taken place. In fact, there was no total eclipse in 1917. And Einstein knew even less. “Did Campbell happen to make any exposures at the time?” he wrote back to inquire. “Were these total eclipses?” He apparently mistook Freundlich’s uncertainty of the year to indicate two separate eclipses. “I did not know anything about it.” It’s extraordinary that such an expedition, undertaken by a distinguished observatory to test such an important theory, was unknown even to the theory’s creator. But the barrier was just coming down between scientists in opposing countries.
From the outside, the relationship between Einstein and his first devotee might seem at times like a game of cat and mouse. In a display of the mischievous humor he was known to possess, Einstein often had a bit of fun with his loyal employee: “I just received your letter from your vacation. I am very sorry that you cannot relax even during the summertime.” Yes, he wrote, there were problems that needed attending to in Freundlich’s third edition “if gross misunderstandings are to be avoided.” He would be happy to discuss his notes with Freundlich when he returned home from his vacation. There would be no harm done if they didn’t agree, only that “my brief foreword endorsing it would just have to be omitted.” If Freundlich had even considered fighting back over the notes Einstein had sent the editor, this thinly veiled threat would quickly crush his intentions.15
Freundlich’s letters to Einstein often mentioned reimbursements he hoped to collect for money spent on photographic plates, measuring devices, and other needed materials, not a rare problem between observatories and employees, as Charles Perrine knew so well. Freundlich and his wife’s financial difficulties, which were only getting worse, were not the most important issue to Einstein, who did offer encouragement and other help from time to time in petitioning on the younger man’s behalf. When Freundlich wrote earlier that year about the headway he was making in “determining the redshift of spectrum lines from fixed stars, as I have done for stars in the Orion Nebula,” Einstein praised him. “You have done a great service to the cause with these two papers.”
As for the teaching position Freundlich hoped to land at Potsdam with his mentor’s endorsement so that he and Käte would be more secure, Einstein agreed but thought it premature. “The Gen. Th. of Rel. must win acceptance among astronomers beforehand,” Einstein wrote to him. He ended his letter with some advice: “Don’t you get any gray hairs, now, but enjoy the rest of your vacation. Everything will straighten itself out somehow. You are still a far cry from the necessary nonchalance; your nerves lie much too bare, without any cushion of fat!” That last remark was in reference to Freundlich’s tall and very lean build.
It was not a letter Freundlich could take to the bank. But it wasn’t as if Einstein was himself living in the lap of luxury. The war had depleted the bank accounts of countries, not just people. The tight blockade that the British Royal Navy had wrapped around the Central Powers greatly crippled the supply of food and coal. Times were hard. “Much shivering lies ahead for the winter,” Einstein wrote to his mother in early September. “But at least, if you pay a pretty penny, you can get something to eat!” The elevator was shut down in their building, no minor thing with an apartment up five flights. He and Elsa had been forced to rent out one of their rooms for extra income. The year before, when his mother discovered she had cancer, she was visiting her daughter in Switzerland. Now Pauline Einstein was languishing in pain at a sanatorium in Luzern. Attached as Albert was to his mother—he had visited her that past summer—her illness and situation would have added stress and remorse to his daily life. He asked if Pauline’s friend, whom he had encouraged to visit her, had done so. Then he confessed he could not lend this friend money, especially given the exorbitant exchange rate on Swiss francs. “There is still no news about the solar eclipse,” he told his mother in the letter. “All the same I am frequently asked, verbally and in writing, about the result.”
Like his still-unmet colleague Arthur Eddington, Einstein had no doubt about light deflection according to his theory. Still, even though he feigned disinterest at times, the answer had to be a burden. Positive eclipse results in his favor would no doubt quickly change his financial situation and status, through lectures and book sales, not to mention future academics prizes. And a fatter bank account would certainly lighten his mother’s own dark days. He had been supporting her in Switzerland as best he could and was maintaining his and Elsa’s strained finances in Berlin. Part of his salary was also being sent to support his sons.
Then, on September 22, a cable arrived from Lorentz with tantalizing news. Lorentz had received an update from a colleague, the Dutch physicist Balthasar van der Pol, who was in England at the time doing research at Cavendish Laboratory in Cambridge. Van der Pol had attended a BAAS meeting in the coastal town of Bournemouth to hear Eddington speak on the expeditions. While Eddington needed more input from the Sobral plates—they were still being measured at the Royal Observatory by Davidson and Furner—there was nonetheless a light deflection at the sun’s rim. “Eddington found stellar shift at solar limb, tentative value between nine-tenths of a second of a degree and twice that,” the cable read. Between 0.9 arc seconds and double that, at 1.8? This provocative answer stretched all the way from Sir Isaac Newton, over the distance of two hundred years, to Albert Einstein.
Ever the proud son, Albert wrote to his mother: “Today some happy news!” He had heard from his sister that Pauline was not only in pain, but also in a dark and despondent mood. This announcement may have been the only ammunition the son had to cheer up his adoring mother, given his finances and a separation of nearly five hundred miles between Berlin and Luzerne. It was a jumble of emotions, to be sure. In what might be considered ironic, despite the dire circumstances, was the nonworking elevator in his apartment building. As he waited for a final conclusion from the British, he continued to climb the many stair steps up to his apartment. Albert Einstein, whose famous thought experiments on general relativity included an elevator that accelerated through space, now had an elevator that didn’t move at all.16
Dyson sent Eddington his conclusions after a close study of the Sobral measurements. On October 3, 1919, Eddington replied to the astronomer royal that he had been worrying over the Príncipe plates. He confided to Dyson that he wondered if he had been hasty in basing his early expectations on so few measurements. And was combining the two results the best method? But now, with the Sobral results ready, Eddington, Dyson, and Davidson began writing a paper based on their conclusions. It was titled “A Determination of the Deflection of Light by the Sun’s Gravitational Field from Observations Made at the Eclipse of May 29, 1919.” It included details of the expeditions themselves and how the astronomers had arrived at their findings. It also mentioned the Lick expedition of 1918 and that its results, yet unpublished, were likely to be inaccurate because the “probable accidental error” was so great. The paper was submitted to the Joint Permanent Eclipse Committee and received on October 30, 1919. A special meeting was called for, to be held November 6, at Burlington House, where the Royal Society and the RAS were housed.
What would the “determination” be? Would it favor Newton, one of the most brilliant scientists the world has ever known, on whose ideas the foundation of physics had rested for two centuries? Or the German newcomer, Albert Einstein?