Science: A History in 100 Experiments

Common sense tells us that if I hit a cricket ball on a playing field in England, this has no effect on a cricket ball in Australia, even if the two balls were manufactured in the same batch in the same factory and once nestled together in the same box. But does the same common sense apply to things in the quantum world, such as photons and electrons? Bizarre though it may seem, in the twentieth century, quantum physics raised the real possibility that the answer might be ‘no’. This was eventually proved in an experiment carried out in 1982.

It all started in 1935, when Albert Einstein and his colleagues Boris Podolsky and Nathan Rosen presented a puzzle (sometimes known as the ‘EPR Paradox’) in the form of a thought experiment. This was later refined by David Bohm, and later still by John Bell. In its later form, the puzzle concerns the behaviour of two photons (particles of light) ejected from an atom in opposite directions. The photons have a property called polarization, which can be thought of as like carrying a spear pointing either up, down, or at any angle across the direction of travel. The key feature of the puzzle is that the photons must have different polarization, but correlated in a certain way. For simplicity, imagine that if one photon is vertically polarized the other must be horizontally polarized.

Now comes the twist. Quantum physics tells us that the polarization of the photon is not determined – it does not become ‘real’ – until it is measured. The act of measurement forces it to ‘choose’ a particular polarization, and it is possible (indeed, straightforward) to set up an experiment which forces a photon to be vertically polarized, or forces it to be horizontally polarized, whichever you wish. This is scarcely any more sophisticated than letting the light shine through a lens of a pair of polarizing sunglasses. The essence of the EPR ‘paradox’ is that, according to all this, measuring one of the pair of photons and forcing it to become, say, vertically polarized instantaneously forces the other photon, far away and untouched, to become horizontally polarized. Einstein and his colleagues said that this is ridiculous, defying common sense, so quantum mechanics must be wrong.

After John Bell presented the puzzle in a particularly clear form in the 1960s, the challenge of testing the prediction was taken up by several teams of experimenters, leading up to a comprehensive and complete experiment carried out by Alain Aspect and his colleagues in Paris in the early 1980s. Although such experiments have since been refined and improved, they always give the same results that emerged from the Aspect experiment itself.

The key feature of the experiment is that the choice of which polarization will be measured is made automatically and at random by the experiment, after the photons have left the atom. At the time the ‘forced’ photon arrives at the polarizer, there has not been long enough for any signal, even travelling at the speed of light, to have reached the other side of the experiment. So there is no way that the detector used to measure the second photon ‘knows’ what the first measurement is.

It would be very difficult (just about impossible, even with present technology) to do the experiment literally with pairs of photons, two at a time; but in the Aspect experiment and its successors very many pairs of photons are studied, with more than two angles of polarization being investigated, and the results analysed statistically. John Bell’s great contribution was to show that in this kind of analysis if one particular number that emerges from the statistics is bigger than another specific number, common sense prevails and there is no trace of what Einstein used to call ‘spooky action at a distance’. This is what Bell expected to happen, and it is known as Bell’s Inequality. But the experiments show that Bell’s Inequality is violated. The first number is smaller than the second number. Experiments are somehow particularly convincing when they prove the opposite of what the experimenters set out to find – it certainly shows that they were not cheating, or unconsciously biased by their preconceptions! But what does it mean?

The pairs of photons really are linked by spooky action at a distance, confounding ‘common sense’, in a state that quantum physicists call entanglement. What happens to photon A really does affect photon B, instantaneously, no matter how far apart they are. This is called ‘non-locality’, because the effect is not ‘local’ (specifically, it occurs faster than light, although it turns out that no useful information, such as the result of the 3.30 race at Newmarket, can be transmitted faster than light by this or any other means). The Aspect experiment and its successors show that the world is non-local. And this strange property even has practical implications, as in the world of quantum computing (see here).