Light and waves
29 May 1919
Starlight bends
Arthur Eddington, Frank Dyson and the eclipse expeditions
Can the Sun bend starlight?
During a total eclipse, expeditions to Príncipe and Sobral photographed stars near the darkened Sun. Their positions had shifted, as if the Sun’s gravity bent their light by about the amount Einstein predicted. Newspapers made Einstein famous overnight.
How they figured it out
Arthur Eddington, a Quaker who refused to fight in the First World War, championed the work of Einstein, a scientist from an enemy country, while the war was still on. With Astronomer Royal Frank Dyson he planned expeditions for the total eclipse of 29 May 1919: one to the island of Príncipe, one to Sobral in Brazil.
It was cloudy at Príncipe until the last moments. When the results were announced in London in November, newspapers declared a revolution in science, and Einstein became world-famous overnight. Historians have since debated how precise those plates really were; later measurements, especially with radio telescopes, confirmed the bending firmly.
The equation
Light grazing the Sun is bent by about 1.75 arcseconds, twice the simple Newtonian guess.
See it in real life
The bottom of a wine glass
Look at a candle through the stem and base of a wine glass and you see it smeared into an arc or a ring. Massive galaxies bend the light of galaxies behind them the same way, making rings and arcs astronomers use as natural telescopes.
Where engineers use it
Testing predictions
A clear, risky prediction tested by measurement is how every engineering model earns trust.
Where it came from
- General relativity November 1915
Led to
General relativity predicted how much the Sun bends starlight; the 1919 eclipse tested it.