General relativity

Often quoted as a theory of extraordinary beauty, with Nobel Prize winner Paul Dirac calling it “probably the greatest scientific discovery ever made”, Einstein’s general theory of relativity revolutionized our understanding of the universe. Whereas Newton’s theory attributes gravity to mass, the theory of general relativity posits that gravity is a consequence of a far more complicated relationship with space and time. According to Einstein, matter leads to curvature in space, which in turn leads to the motion of matter.

One consequence of this “warping” of spacetime is that a hugely massive object (such as the Sun) should be able to distort the path taken by light. This is a process known as gravitational lensing. If we could see a change in the apparent location of the stars depending on whether or not the Sun was in the vicinity of the sky, then the Sun had distorted the light from those stars. The extent of this distortion would be evidence for whether Einstein’s theory of general relativity was accurate or not. However, the Sun is normally bright enough to block out stars in the nearby sky. In this case, a total solar eclipse occluding most of the sunlight would give scientists the chance to see these stars and test the theory of general relativity.

In 1911, during the early years of developing the theory that he would later publish in 1915, Einstein issued a challenge to astronomers worldwide, urging them to observe total solar eclipses and search for the apparent positions of these stars with and without possible warping by the Sun. The first of these was just a year later in 1912, with no success. The total solar eclipse in 1919 would aim to succeed where earlier expeditions had failed.

Other eclipse expeditions

It wasn’t only British teams that made extensive plans to observe the eclipse. Charles Dillon Perrine, an American astronomer, at the time Director of the Argentine National Observatory, had since 1912 attempted to measure the bending of light by the Sun. He had exchanged many letters with the observatory of Rio de Janeiro, to prepare an expedition to observe the eclipse. The expedition sadly didn’t occur, as he was ultimately unable to get government authorisation.

The Carnegie Institution of Washington sent an expedition to Cape Palmas, Liberia, with the goal of investigating the magnetic effects of the Sun, eventually finding that the eclipse had a clear influence on Earth’s magnetic field. Another team went to Bolivia, one of three teams outside the zone of totality. Their goal was to measure sky and solar radiation during different stages of the eclipse, where they found that sky radiation varied proportionally to the amount of sunlight. They were also fortunate enough to observe the eclipse at an altitude of fourteen thousand feet, whilst the Sun was barely above the horizon at an altitude of 5 degrees. Professor Louis Agricola Bauer in 1920 wrote that “the great length and beauty of the coronal streamers, the splendid crimson prominence” and with “the snow covered mountains as a background, it seemed to the observers the grandest eclipse that they had ever seen”. They also sent a team of three to support the British efforts in Sobral, Brazil.

The British-led expeditions

The British-led expeditions were arranged by the Joint Permanent Eclipse Committee, with initial planning for the British expeditions beginning well in advance of the eclipse. On November 10, 1917, the Joint Permanent Eclipse Committee agreed on two observation stations: one at Sobral in Brazil, and one on the island of Principe, a Portuguese colony at the time. The decision for Sobral was largely influenced by a circular (announcement) produced by Henrique Morize of the National Observatory of Brazil. It contained detailed meteorological reports, information about modes of access, and more, to promote to international astronomers that Brazil was a home for world-class astronomy. Morize also praised the local population as a reason to visit. The Committee applied to the Government Grant Committee for £1100 to fund the expedition and to pay for equipment.  

A subcommittee was set up, consisting of Arthur Eddington, Alfred Fowler, Herbert Turner, and Frank Dyson, meeting during May and June 1918. Provisional arrangements were made for Eddington and Edwin Cottingham, a clockmaker, to travel to Principe. An assistant astronomer at the Greenwich Observatory, Charles Davidson, and Father Aloysius Cortie were to travel to Brazil. Father Cortie was unfortunately unable to take part in the expedition because he couldn’t spare the time, as an expedition of this scale would have required him to be absent from the country for many months. His place was given to another assistant astronomer, Andrew Crommelin.   

In the years leading up to the eclipse, scientific research had been hampered by the war; global politics were interfering with the scientific process and international cooperation. Not only that, but the United Kingdom had enacted conscription in January 1916, which remained in place until mid-1919. During that period, it was subject to major amendments that expanded the eligibility conditions for the male population to such extremities that conscription of the clergy was actively considered.

At the request of the University of Cambridge, Eddington had been exempted from conscription. In 1918, however, an appeal was made, and his exemption was given an expiry date of 1 August 1918. In response, he applied for exemption, supported by the Astronomer Royal, on religious grounds, since he was a devout Quaker and a pacifist. Part of their argument was that there was a “widely spread but erroneous notion that the most important scientific researches are carried out in Germany”, and that the opportunity to disprove this notion was by verifying Einstein’s theory under such rare favourable conditions, that would not arise again for a long time. He was granted a one year exemption, which would end up expiring after the war and obligations for national service had ended.

The experimentum crucis

In their report, the group suggested three possible cases for the behaviour of light in the vicinity of the Sun.

1. The null case: gravity doesn’t influence the path of light.

2. The Newtonian case: light has weight and spacetime is Euclidean (flat), this was predicted by Einstein in 1911 when he first proposed that light bends under the influence of a gravitational field.

3. The Einstein’s case: light has weight and space-time is curved, which Einstein settled upon from 1915. Interestingly, there was a missing factor of two in the original publication, which was corrected in 1916.

In the first case, no measurable deflection of light occurs. In the second and third cases, both lead to a deflection of light, but the magnitude of deflection would be different. Eddington pre-emptively favoured Einstein’s theory. Although much literature has been written about his biases, sources dictate that Eddington had little to no influence on the final analysis or deciding which photographs to omit from the results. Dyson instead led the expeditions.

Since no other hypotheses with a complete theoretical justification existed at the time, the expeditionists were confident that if their findings gave compelling enough evidence that two of the hypotheses could be rejected, then the third hypothesis was likely the correct one.

The expedition teams each left Liverpool by ship on March 8, 1919.

The first team arrived at Para on March 23, 1919. They refrained from proceeding to Sobral before they knew what arrangements were made, so instead travelled to Manaus. They returned on April 8, 1919. The Brazilian government generously allowed them to pass through customs without inspection of their luggage, a courtesy which was repeatedly extended to eclipse observers in many different eclipses. The group then proceeded to Sobral by ship and train on April 30.

The group highlighted that the government gave extensive support to the team. Mr John Nicolau assisted the team with their baggage on the way to Sobral. The Deputy of Sobral, Colonel Vicente Saboya offered the team a house for their residence, and they were assisted to the house by representatives of the Civil and the Ecclesiastical Authorities. Dr Leocadio Araujo of the State Ministry of Agriculture, served as their interpreter. During the eclipse, he was responsible for alerting the group when totality began and calling out every tenth beat of a metronome after that. The Sobral team obtained 19 plates using their astrographic telescope, and 8 using a 4-inch camera. It was mostly clear sky for the observers during the eclipse, but there was about a minute during totality where a thin layer of cloud had obscured the view of the stars. However, this didn’t interfere with their observations.

Regrettably, the group decided to set the focus of the astrograph telescope during the night before the eclipse, so that they could test its performance against the stars. When the temperature began to rise during the day, the steel tube and mirror began to expand. As a result, all the photographs taken with the astrograph telescope were out of focus. During the analysis of Sobral photographs, all of those photographs were discarded. Consequently, the Sobral results were based only on the photographs captured using the 4-inch plates they brought as a backup.  

Development of the images took place during the night in the days following the eclipse. It was completed by 5 June 1919. The observers then travelled to Fortaleza and returned on 9 July 1919 to photograph the star field again, to get images of the stars to compare with those they had taken during the eclipse. They began taking the photos on the morning of 11 July and continued through to July 18. Once complete, they left Sobral a few days later, on 22 July 1919. Their packing was left on the island, to generously be forwarded back to England for them later.  

The expedition to Principe

The expedition to Principe was marked with misfortune. The group first sailed to Madeira which, like Principe, was at the time a Portuguese territory. They proceeded to Principe on 9 April 1919, arriving on 23 April 1919. Upon arrival they received introductions by two astronomers of the Lisbon Observatory. Like the Sobral expedition, the Portuguese government skipped examination of luggage at customs. During the expedition, they resided at a plantation owned by Sr. Carneiro, who allowed them to use his “ample resources of labour and material”. He had also generously delayed a trip of his own to Europe to host them.

Principe, at the time, was subject to allegations that African workers on the island were being subjected to slave labour, despite slavery north of the equator being banned by Portugal in 1819. The Portuguese government had been placed under heavy criticism, particularly in the United Kingdom. The Royal Astronomical Society was aware of the accusations and had made it a matter of discussion in 1917, at their meetings discussing plans for the 1919 eclipse. There are some suggestions that members of the Principe expedition, which included Eddington, withheld various details of the contributions by Portuguese and local scientists when reporting on the expedition due to this controversy. As Quakers were vocal advocates against slavery, there is some merit to the suggestion that Eddington would have wanted to distance himself from the Portuguese government.

The team were assisted by two individuals from Sierra Leone, Mr. Wright, and Mr. Lewis, who gave introductions to the observers. Some notice that in their final expedition report, the Sobral team gave extensive and long thanks to the people that assisted them along the way, including the hosting Brazilian government. In contrast, the Principe team gave no such treatment towards the Portuguese government, nor the other individuals (bar the hosts) who had assisted in other matters, such as the transport of equipment or the construction of the huts at the observation sites. Some historians believe that Mr Wright and Mr Lewis played a much greater role in the expedition than what the final expedition report indicates. Furthermore, there remain many more individuals whose contributions were overlooked.  

The day of the eclipse itself was a largely unsuccessful one. The observers were at the mercy of nature, which had cursed them with thick cloud cover. Most of the images taken successfully captured the light of the solar corona but failed to capture the stars. Fortunately, they were able to capture a few photographs which showed enough stars that some measurements could be made, but in the end, only 5 stars would be used in the analysis, coming from only two photographic plates. In comparison, a repeat of the experiment by an American team in 1922, under the direction of William Campbell, which calculated measurements on over 100 stars.  

After a disappointing eclipse day, the observers were then burdened with the matter of an incoming strike of the local steamership company. This forced the expedition party to leave the island significantly earlier than planned to avoid being stranded. Leaving Principe on June 12, they transhipped at Lisbon and reached Liverpool on July 14, 1919. This early departure meant that the observers were unable to take photos of the star field at the observation site. Consequently, for analysis of the Principe photographs, they had to rely on the Sobral plates and ones taken afterwards in Oxford, England. This is problematic because it opened the risk for the photographic plates to expand or contract due to variations in temperature before the measurements were made. The accumulation of these issues significantly increased the scope for error, which many scientists and historians ultimately consider too significant for them to place their confidence in the final measurements.

Dyson announces the results

Whilst work was underway to analyse the plates in the months after the eclipse, skepticism surrounding Einstein’s theory continued to persist and would continue still after the results of the eclipse were announced. Gravitational red-shift tests in America were failing to make observations which aligned with Einstein’s theory.  

On November 6, 1919, in a room packed with people, Dyson announced to the world the results of their expedition at a joint meeting of the Royal Society and Royal Astronomical Society. Einstein had already been informed during the early stages of the analysis that a deflection of light had been detected, news which he accepted joyously. However, at that stage of the analysis, it was thought that the deflection of light would be in an interval of between 0.9 and 1.8 arcseconds. Einstein’s theory asserted that the deflection should be 1.74 arcseconds. On the other hand, Newton’s theory asserted that the deflection should be 0.87 arcseconds. Both values still seemed reasonably possible, but Einstein remained confident in his work.

At the meeting, Dyson revealed that the analysis of the (unrejected) Sobral plates gave a deflection of 1.98 arcseconds, with a probable error of 0.12 and a standard deviation of 0.18. The Principe results, on account of the expedition being subject to lots of misfortune, gave a deflection of 1.61 arcseconds with a probable error of 0.30 (or a standard deviation of 0.45). On account of their error, the discarded Sobral astrograph plates had two separate results: 0.93 arcsecs and 1.52 arcsecs. Scale parameters on those plates had to be estimated using the comparison plates because the eclipse ones didn’t show enough stars. The authors of the expedition report were left with “little doubt that a deflection of light takes place in the neighbourhood of the Sun and that it is of the amount demanded by Einstein’s generalised theory of relativity”.  

At the time, many were satisfied with the validity of the results but, over time, more literature would arise both for and against the results, particularly later in the century. During the announcement meeting, Ludwik Silberstein, an accomplished physicist, pointed to a portrait of Newton in the room and announced, “We owe it to that great man to proceed carefully in modifying or retouching his Law of Gravitation.” Later, criticism was directed at the precision of the results, with some detractors going further and suggesting that the findings were politically motivated. These critics asserted that the data were fit to match the expedition team’s pre-emptive beliefs and an eagerness to support Einstein who, much like Eddington, was a pacifist and scientific internationalist. Additionally, some suggest that a more cautious scientist would go not much farther than to conclude that a deflection of light does occur, but the precise value is unknown. Stephen Hawking would suggest in 1988 that the error was “as great as the effect they were trying to measure” and that the confirmation of Einstein’s prediction was either pure luck or, worse, a result of fitting the data. Although such criticism became more prominent in the later 20th century, re-analysing the photographs using more contemporary methods would reach similar findings and support the team’s justification to discard the plates they did. 

The expedition team were aware of the limitations of their results, and it was general scientific consensus that further evidence is essential. In stark contrast, the press and the wider public at the time didn’t share their caution, and had little doubt about the validity of the results. The morning after the announcement, the Times newspaper opened with the headline:

“REVOLUTION IN SCIENCE

NEW THEORY OF THE UNIVERSE

NEWTONIAN IDEAS OVERTHROWN”

The 1919 expedition has been the subject of plenty of literature, with commentary from scientific, theoretical, political, and historical perspectives. It was this expedition that led to general relativity becoming the most widely accepted theory for gravitation, and gave Einstein international fame. The theory provided solutions to the questions which, in previous decades, had remained unanswered. For instance, scientists had previously considered that an unexplained shift in Mercury’s orbit could be due to an inter-Mercurial planet, Vulcan. This idea, among other conjectures, were all discounted in favour of the explanation by general relativity.

In the coming years, scientists did place more faith in the results, so general relativity became an essential part of an education in theoretical physics. This was likely due to evidence provided from other expeditions and experiments later on. An American team repeated the experiment in Australia in 1922, considered by many to be the true confirmation of general relativity for its more comprehensive dataset. Nonetheless, repeats of the experiment occurred beyond 1922, including recently in 2017, with the benefit of much better technology than the original experiments a century before.