halfacanuck

First, imagine you have a sinkful of water, the surface of which is glassy smooth and undisurbed. Now drop a frozen pea into the water. What happens? You get ripples, waves of water, moving out from the point of impact across the surface of the water in concentric circles. What happens if instead of one pea, you drop two, a few inches apart from each other? They each create waves, as before. But what's interesting is what happens when the waves from each impact meet. Where a peak meets a peak, a higher peak is created. Where a trough meets a trough, a lower trough is created. Where a peak meets a trough, they cancel each other out and the water is flat, with neither a peak or a trough.

This is called "interference": the waves from each impact are interfering with each other. This phenomenon is not limited to water waves, but happens any time a wave appears in whatever medium, be it water or air or otherwise. It happens with sound waves when one speaker of a pair is wired incorrectly (the wires on the back switched). If the speakers are then oriented to face each other the music coming from them will reduce in volume, because a "sound peak" from one speaker will meet a "sound trough" from the other (incorrectly wired) speaker, and will cancel each other out.

The same thing happens with light. Light, too, can behave like a wave. This was proved in the 19th century by a guy called Thomas Young, who set up an experiment roughly along the lines of the crappy illustration to the right. He arranged a light source (in his case, sunlight) so that it shone through a screen with two slits in it, and from there onto a wall. Common sense would suggest that on the wall would appear two stripes of light, since it's shining through two slits. What actually appeared, however, was not two but a series of stripes, with the middle stripe being the brighest and the others diminishing in brightness as the distance from the center increased.

What was happening here? Well, the light shining through one slit was interfering with the light shining through the other. The peaks and troughs of the light waves were mingling in the same way water waves do. The bright parts of the pattern were where peak met peak or trough met trough. The dark parts of the pattern were where peak met trough and cancelled each other out. Hence the light-dark-light-dark pattern repeated across the wall. If one of the slits were covered up the interference pattern disappeared, because there were no longer two beams of light interfering with each other. One beam of light cannot interfere with itself, and caused what we would expect: a single stripe of light on the wall. This could only be explained, decided Young, if light were a wave, and behaved as waves do.

For a long time this is how things stood. Light is a wave, behaves as all waves do, and one beam of light is able to interfere with another to create interference patterns, just like ripples on the surface of water. Then Einstein came along and turned everything on its head.

I'm not going to go into details of how he did that, not because the general principles are massively complicated but because it's not important to the weirdness I was talking about at the beginning of this post. Suffice it to say that Einstein, never one to take any notice of prevailing wisdom, went ahead and proved that light is not a wave but is in fact made up of particles. The particles in question are called photons, and Einstein proved that a beam of light is simply a stream of these particles rushing from one place (the light source) to another. Because particles are discrete objects, distinct things, they are not waves and do not behave like them. Particles no more interfere with each other to create interference patterns than do pool balls or slices of pizza. Photons represent the smallest unit of light possible: they cannot be subdivided.

This caused, to put it mildly, a bit of a conundrum for scientists. On the one hand Young's double slit experiment proved, beyond all doubt, that light is a wave. On the other hand Einstein proved, beyond all doubt, that light is made up of particles. It is here the weirdness begins.

The conundrum has never really been solved, at least not in any way that makes sense on a human level, in any way that corresponds to reality as we perceive it. The solution is called the "wave-particle duality" and says, in simple terms, that light is both a wave and made up of particles at the same time, depending on how we measure it. Light changes what it is, and how it behaves, depending on the nature of the experiment we set up to examine it.

That's pretty weird, but it gets way weirder. What would happen if we repeat Young's double slit experiment but this time sending only one photon at a time? Since a single photon obviously can't interfere with itself, surely we would see two stripes of light, since there are two slits in the screen and the photon would have to pass through either one or the other?

What was seen, to the astonishment of all, was an interference pattern. This made absolutely no sense. Only one photon was going through one slit or the other at any given time, so where did the interference pattern come from? What were the photons interfering with? In effect, the light was acting like a wave in making an interference pattern, even though no wave existed, because only one photon at a time was sent through.

As with Young's initial experiment, if one of the holes were covered the interference pattern disappeared. The implication is staggering: the photon passed through one of the slits, but somehow "knew" that the other slit was there. It also "knew" whether the other slit was open or closed: when the other slit was open, the photon landed only where it "should" land to form the interference pattern (i.e. no photon landed in the dark gaps between the light stripes); when the other slit was closed, there was no such restriction.

How does a photon "know" how the apparatus is set up? How does a photon, an elementary particle, "know" anything?

The mystery deepens further with an experiment in 1999 dubbed the "quantum eraser." The scientists very cleverly devised a repeat of the one-photon-at-a-time double slit experiment in which they could detect which slit each photon passed through without affecting either the speed or direction of it. They did this using filters that polarized ("changed the orientation of") the photons going through one slit to a certain direction, and the photons going through the other slit to another direction. Thus when the photons hit the sensor they were able, by inspecting the polarization of each photon, to determine through which slit it had passed.

In all other respects it was identical to the previous experiment, with the only difference being they would be able to tell through which slit each photon passed. But when they ran the experiment something extraordinary happened: the interference pattern was not there.

How is this possible? How is it possible that simply by being able to know through which slit each photon passed they caused the light to stop creating an interference pattern? How could knowledge, or the potential to know, cause the light to behave differently?

We come at last to the weirdest, and most important, part of all. The scientists repeated the experiment once more, but this time they placed another filter on the other side of the slits, between the slits and the screen onto which the light was projected (their equivalent of Young's wall). This filter polarized the photons again, so that once they passed through it all the photons were oriented in the same direction, and thus the scientists could no longer discover through which slit the photons passed (they "erased" the potential for knowledge). Result? The interference pattern returned. The photons started behaving like a wave again.

Let's go over that again: the photons were still being "labeled" as having passed through one slit or the other, by the first set of filters, between the light source and the slit. Between the slit and the second filter the photons were still, at that point, potentially identifiable as having passed through either the first or second slit. Had the scientists observed them at this point, as in the previous experiment, no interference pattern would emerge. But because the photons passed through the second filter, which stripped them of the identifying "labels" they'd been given, the interference pattern came back.

The only difference -- the only difference -- is whether or not the scientists were able to tell through which slit the photons passed. It is the very possibility of knowledge that caused the light to behave differently. It is as if the light somehow knew what the scientists were trying to do (namely, treat light as both a particle and a wave simultaneously), and prevented them from doing it. Either we consider light a wave (for which the concept of "which slit it passed through" is meaningless, since a wave would pass through both) and get a wave-like interference pattern, or we consider it a stream of particles (each particle of which must go through either one slit or the other) and not get a wave-like interference pattern, since particles can't interfere with themselves. We cannot do both. We must pick. We literally decide what light is.

This is not me indulging in poetic license or hyperbole. These are not the results of some one-off never-to-be-repeated experiment by a crackpot inventor in his garage. These results have been repeated over and over, by many reputable scientists in different laboratories around the world, and form part of the canon of modern physics. There is, as yet, no explanation for why any of this happens. Perhaps there never will be. But it is clear that, at the subatomic level at least, abstract non-physical concepts such as knowledge and potential for knowledge have an actual, scientifically verifiable effect on real physical systems. We literally change the world just by observing it.

This is, as far as I'm concerned, spookier and more exciting than any ghost story. I hope you feel the same way (it would be a shame to come this far and get nothing from it). Suddenly things like psychokinesis and telepathy seem a lot less far-fetched...