Eclipse week 5 – 1919

[tweetmeme only_single=false service=”wp.me” source=”allinthegutter”]

It was nearly 10 years ago that I had the chance to witness my first – and only – total solar eclipse. It was every bit what is cracked up to be, and the feeling that you’re witnessing something special comes effortlessly. You don’t have to be an Astronomer to enjoy the uniqueness and transiency of such a brief moment, and millions of people in Asia had the same pleasure this week that I most certainly hope to have again in the near future.

But you probably do have to be an Astronomer to turn an eclipse into an event which will change the face of Physics. So today we’re not here to talk about the 1999 or the 2009 eclipses, but rather the total eclipse of the 29th of May, 1919. This puts us four years after Albert Einstein published his theory of General Relativity, in 1915.

General Relativity, at its most basic, is a theory of gravitation. It provides a theoretical framework which we can use to explain the behaviour of gravitationally interacting systems that we observe – like the solar system, or the Earth-Moon system, the Milky Way, etc. Most importantly, like any other scientific theory, it also allows us to make predictions about how gravity should affect these, and other, systems.

Prior to Einstein and General Relativity, our understanding of gravity was motivated by Newton’s theory of gravitation, often also referred to as the classical theory. In what we may call “everyday situations”, both theories make the same predictions, but General Relativity makes significantly different predictions, or presents very different explanations, for things like the geometry of space, the passage of time and how light propagates. Crucially, General Relativity predicts that the mass of an object affects space(time – the merging of space and time into a single mathematical object is also a consequence of General Relativity), and that the shape of spacetime affects the way light travels. These are stunningly anti-intuitive ideas, because they are only noticeable in regimes far detached from our everyday experiences. For example, we need a very large amount of mass to notice a very small deflection in light’s path.

But Einstein was not the first to suggest that mass affects the path of light. In 1801, Johann Georg von Soldner used Newton’s corpuscular theory of light – which states that beams of light are streams of particles of tiny mass – to calculate how mass would affect the path of a beam of light. He arrived to a result which is known as the Newtonian result for the bending of light. However, because we know now that light is in fact made of massless photons, there is no formal way to correctly treat the bending of light in Newtonian gravity. This was not known at the time, so the result stuck. Unaware of Soldner’s calculations, Einstein in 1911 calculated what the bending of light should be in his new theory of General Relativity, which at the time was work in progress. He reached the same value that Soldner had done, over 100 years previously. Crucially, once his theory was finalised in 1919, Einstein revisited this problem and realised he had made an error – the bending of light, once one takes into account the curving of space, should be twice that of the Newtonian result. Such a clear distinction between the predictions of two competing theories is a blessing – it gives scientists the chance to design an experiment which can show which one, if either, is correct.

We couldn’t look in the Earth for a suitable system to measure this effect in, but in 1919 Astronomer Royal Sir Frank Dyson and Plumian Professor of Astronomy in Cambridge Arthur Eddington decided to look elsewhere – they turned the Sun, the Moon, and a distant cluster of stars called the Hyades into their own laboratory. The idea is that the mass of the Sun is large enough to bend the light which passes nearby, as that of distant stars which are sitting behind or very near the Sun from the Earth’s perspective. This is of course happening all the time, but we simply can’t see the light of distance stars near the Sun because the Sun is so much brighter than the stars we are trying to observe. Unless something really big gets in the way and blocks the light from the Sun – say the Moon (which is actually much smaller than the Sun, but sits just at the right distance – see Emma’s post). A total eclipse makes therefore the perfect opportunity to see how the position of stars in the sky changes when their light has to travel close to the Sun on their way to us.

The eclipse of 1919 was a good and timely opportunity to measure this effect, and expeditions in the island of Principe (off the west coast of Africa) and Sobral in Brazil were planned to do precisely so. The Moon would block the light from the Sun for almost 7 minutes – an exceptionally long period of totality! What no Astronomer can ever do, however, is plan to the weather. And so, for over 6 minutes, Eddington waited and stared at a cloud.. and prayed. The cloud did disappear, giving Eddington and his team 10 seconds – 10 precious seconds! – to take the needed photo. You can see the negative of this photo on the left, and if you look carefully you can see the horizontal lines which mark the positions of the stars. The expedition in Brazil got perfect weather, but later a flaw in the telescope setup meant that the results could not be used. So Eddington and Dyson went home to analyse the data, while the community – and Einstein – waited.

Within the margin of error, the shifts observed in the positions of the stars were more in agreement with Einstein’s predictions that Newton’s, and Einstein was thrown into stardom once the results were announced. It is worth mentioning that there was healthy controversy at the time, and the data analysis of Eddington was challenged – as it should have been – by the scientific community. Such a leap in scientific thinking never gets an easy ride! But results in following expeditions confirmed the 1919 results, as did other experiments which measured slight departures from Newtonian predictions in different systems (the orbit of Mercury being one example). General Relativity has been proven time and time again to be accurate within our measurement errors, and it’s a deep and beautiful theory. Given another chance I will tell you why and how some scientists feel the need to adapt General Relativity to explain some recent observations of the distant Universe, and how controversial and interesting a topic that is. But not now..

Personally, I find it rather interesting that this event really catapulted Einstein into the public eye. You can see on the right one of the headlines at the time, and in fact the story was picked up by newspapers and magazines all over the world. It is particularly interesting given how complex General Relativity is, which doesn’t make it a readily accessible theory to the public. This didn’t stop the world from taking an interest, and hopefully encouraged many people to try and understand it, even if only a small part.

I also finding it amusing that two other expeditions, planned for 1912 and 1914, failed due to bad weather and the war. But had they happened before Einstein corrected his predictions (in 1915), his theory would have been proven wrong – before he had a chance to correct it!

I’ve stayed too long, but for those who want to know more let me recommend Peter Coles’ excellent exposition of this subject here.

I believe this concludes our first mini-series of posts! I hope you enjoyed it, and learnt a thing or two. If you have any ideas for more mini-series just leave us a comment – we’ll be happy to consider it.