Eddington's Eclipse Expedition (1919): Proving Einstein Right

Eddington’s eclipse expedition was a British scientific project that tested Albert Einstein’s general theory of relativity during the total solar eclipse of 29 May 1919. Teams on the island of Príncipe, off West Africa, and at Sobral in north-eastern Brazil photographed stars near the darkened Sun and found that their light was bent by the Sun’s gravity by the amount Einstein predicted. When the results were announced in London in November 1919, Einstein became a world celebrity.

Background: a prediction to test

In November 1915 Einstein completed general relativity, a theory describing gravity as the curving of spacetime by mass. One of its clearest predictions was that starlight passing close to the Sun would be deflected by about 1.75 arcseconds, a tiny angle roughly equal to the width of a coin seen from about 2 kilometres away.

Newton’s theory, if light was treated as a stream of particles, suggested a deflection of only about half that, around 0.87 arcseconds. Measuring the difference would therefore show which theory was correct. The only practical way to do it was during a total solar eclipse, when the Moon blocks the Sun’s glare and nearby stars become visible.

Earlier attempts had failed. A German expedition to Crimea for the eclipse of August 1914 was caught up in the outbreak of the First World War, and its members were briefly interned in Russia.

Planning in wartime

News of Einstein’s theory reached Britain during the war through the Dutch astronomer Willem de Sitter. Arthur Eddington (1882–1944), Plumian Professor of Astronomy at Cambridge and secretary of the Royal Astronomical Society, became its leading champion in the English-speaking world.

The Astronomer Royal, Frank Dyson, pointed out that the eclipse of 29 May 1919 would be ideal, because the Sun would then lie in front of the Hyades, a cluster of bright stars. Planning began in 1917. Eddington, a Quaker, was a conscientious objector, and Dyson helped arrange for his exemption from military service to be tied to his leading the expedition. The Armistice of 11 November 1918 came just in time for the teams to sail.

To reduce the risk of clouds ruining the observations, two expeditions were sent. Eddington and Edwin Cottingham, a clockmaker, went to Príncipe, then a Portuguese colony in the Gulf of Guinea. Andrew Crommelin and Charles Davidson of the Royal Greenwich Observatory went to Sobral, in the state of Ceará, Brazil.

The eclipse of 29 May 1919

At Príncipe, the team set up at the Roça Sundy plantation. On the morning of the eclipse it rained heavily, and clouds covered the Sun for most of totality. Eddington exposed 16 photographic plates; only a few showed enough stars to be useful. The clouds thinned during the last part of totality, and Eddington later wrote in his notebook that one plate showed several stars that gave a result agreeing with Einstein. Strikes by shipping companies then threatened to strand the team, and they left the island earlier than planned, so they could not take comparison plates there.

At Sobral the sky cleared and the observers had good conditions. They used two instruments: an astrographic telescope borrowed from Greenwich and a smaller telescope with a 4-inch lens. A Brazilian team led by Henrique Morize, director of the National Observatory in Rio de Janeiro, also observed the eclipse from Sobral.

Totality lasted about five minutes at both sites. The 1919 eclipse was one of the longest of the 20th century, with a maximum duration of almost seven minutes over the Atlantic Ocean, which gave the observers unusually long to take their photographs. After the eclipse, the Sobral team stayed on to photograph the same star field at night in July, without the Sun, so the positions could be compared.

Measuring the results

Back in England, the plates were measured over the summer. The results from the three sets of plates were:

The images from the large Sobral astrograph were blurred, probably because the Sun’s heat had distorted its mirror, and Dyson and Eddington judged them unreliable. The other two sets favoured Einstein.

The measurements themselves were painstaking. Each eclipse plate was compared with a reference plate of the same stars, and the tiny shifts in each star’s position, often fractions of the width of a hair on the plate, had to be separated from errors caused by temperature changes, the telescope and the photographic process. Eddington was well known as a supporter of relativity, which is one reason his role was later questioned.

The announcement

The results were presented at a joint meeting of the Royal Society and the Royal Astronomical Society in London on 6 November 1919, under a portrait of Isaac Newton. The meeting was chaired by J. J. Thomson, President of the Royal Society, who described the result as one of the greatest achievements in the history of human thought.

The next day The Times reported a “Revolution in Science”, and newspapers around the world followed. Einstein, then living in Berlin, became instantly famous. Journalists sought interviews, and popular books and lectures tried to explain relativity to the public, often claiming that only a handful of people in the world could understand it. Eddington himself wrote widely read books that presented the theory to general readers. Coming a year after the war, the story of British scientists confirming the theory of a German-based physicist also appealed to those hoping to rebuild international science.

Controversy and later confirmation

The result was not universally accepted at once, because the measurements were difficult and the uncertainties were large. In 1980 the philosophers John Earman and Clark Glymour argued that Dyson and Eddington had chosen which plates to trust in a way that favoured Einstein.

Later work has largely defended the expedition. In 1979 astronomers at the Royal Greenwich Observatory re-measured the Sobral plates with modern equipment and obtained results consistent with Einstein. Historians such as Daniel Kennefick have argued that the decision to discard the blurred plates was reasonable and made mainly by Dyson, not Eddington.

Further eclipse expeditions, especially the 1922 observations at Wallal in Western Australia, confirmed the deflection. Since the 1960s and 1970s, radio astronomers have measured the bending of radio waves from distant quasars with far greater precision, in close agreement with general relativity.

Key people

• Arthur Eddington led the Príncipe expedition and became the main interpreter of relativity for English-speaking readers.
• Frank Dyson, the Astronomer Royal, conceived and organised the project.
• Andrew Crommelin and Charles Davidson made the successful observations at Sobral.
• Edwin Cottingham assisted Eddington with the instruments on Príncipe.
• Albert Einstein made the prediction; he learned of the preliminary results by telegram from Hendrik Lorentz in September 1919.

Timeline

• 1915, 25 November: Einstein presents the final equations of general relativity.
• 1917: Dyson identifies the 1919 eclipse as ideal for testing the theory.
• 1919, March: The expeditions sail from Britain.
• 1919, 29 May: Total solar eclipse observed from Príncipe and Sobral.
• 1919, September: Einstein hears of the preliminary results from Lorentz.
• 1919, 6 November: Results announced at a joint meeting in London.
• 1922, 21 September: Eclipse observations at Wallal, Australia, confirm the deflection.

Why it matters

The 1919 eclipse expeditions turned general relativity from an abstract theory into an accepted part of physics and made Einstein a household name. Like ancient eclipse records such as Thales’ eclipse of 585 BC, they show how a few minutes of darkness can change how people understand the heavens.