And if you feel like siding with Schrödinger on this then I should ask you where we are going wrong? Why isn’t the cat – or the ‘complicated collection of quantum particles’ as I might wish to refer to it – also forced into a superposition through its entanglement with the superposed state of the radioactive nucleus?

  Bohr and Heisenberg did not claim that the cat was really both dead and alive at the same time. They insisted instead that – and this has been accepted by the majority of physicists ever since – we cannot talk about the cat as even having an independent reality until we open the box to check up on it! Their reasoning was that all the time the box is closed we simply have nothing to say about the ‘real’ state of the cat. All we have to go by is the wavefunction, and this is just a set of numbers. Since we cannot describe reality in the absence of measurement then we don’t even try. Only when we are ready to open the box can we use the wavefunction to predict the likelihood of finding the cat dead or alive. And repeating such an experiment many times over would show that such predictions are correct (just as we would need to flip a coin many times to confirm the 50/50 statistical probability of heads versus tails).

  But what if we were to place a human volunteer in the box instead – ok, maybe the poison is not fatal, but simply causes the volunteer to lose consciousness.3 Surely, once we have opened the box we cannot then convince the volunteer that he was in a state of being both conscious and unconscious at the same time before we let him out. If he is conscious he would report that apart from feeling a little nervous he has been fine all along. And if we find him unconscious then he might, when revived, report how he heard the device go off ten minutes after the box was closed and immediately started to feel dizzy. The next thing he knew he was being revived with the smelling salts.

  Where is the flaw in this argument? Was Schrödinger successful in highlighting an inconsistency in the theory? Actually, quantum mechanics poses a more serious dilemma than the simple statement that, prior to opening the box, the cat’s wavefunction has a live cat part and a dead cat part. Until we look inside the box all we can do is assign probabilities to different outcomes. And because of the superposition of the different cat states, we get a certain probability for finding the cat alive, another probability for finding it dead and, strangest of all, a probability for finding it both dead and alive at the same time. This is due to the technical definition of quantum probability.4 But the quantum rules do not allow for this. They state, quite correctly thankfully, that we will only find a live cat or a dead cat and never a cat in a ghostly in-between state.

  I don’t know about you, but I can cope with only having probabilities for dead and alive outcomes; I can put this down to my ignorance and I can pretend it is no more serious than the situation of not knowing what Christmas presents I have until I unwrap them. I might even be prepared to relegate the issue of cats being both dead and alive at the same time before I resort to pointless metaphysical pondering. But what happens to that third option, the possibility of finding the cat in the somewhat embarrassing state of being dead and alive simultaneously, which the quantum formalism is so clear about? The quantum postulates get round the dilemma by stating that the only allowed results of measurements are ‘sensible’ ones. Of course this is all couched in appropriate jargon, but the fact remains that, until recently, nobody really understood what happens to these in-between ‘Schrödinger cat’ states (as they have become known).

  Many, even modern, discussions of quantum measurement do not appreciate that the paradox of Schrödinger’s cat may have finally been settled. They talk about the act of opening the box as being the moment of ‘measurement’ that collapses the superposition and leads to a definite outcome. At one time it was even fashionable to talk about a measurement requiring an intelligent, conscious observer. After all, since no one knew where to place the boundary between the quantum domain of wavefunctions and superpositions and the classical domain of definite outcomes when measurements are made, then maybe we should only place a boundary when we really have to. Since even the measuring apparatus (detector, screen, cat) is just a collection of atoms and should behave like any other quantum system, albeit a very big one, then the only place we are forced to abandon the quantum description is when it registers in our conscious minds. (And yes I know the cat is also conscious but supporters of this view argued that it didn’t have a PhD and therefore didn’t qualify as an observer!) There still exists a relatively widely held view, mostly outside of physics, that since we do not fully understand quantum mechanics and we do not understand the origin of consciousness then the two must be related.

  I have always been surprised at the number of prominent supporters this view has had. One of the basic tenets of quantum mechanics is the fact that the results of measurements are defined not only by the properties of the quantum system being studied but by the very act of measurement itself. Many also go as far as to state that it is meaningless to say anything about the quantum system without including the measuring device in the discussion as well. But to place the dividing line between measured and measurer at the level of human consciousness amounts to what philosophers refer to as solipsism – the idea that the observer is at the centre of the universe and that everything else is just a figment of his or her imagination.

  In this way, what the volunteer inside the box reports he felt before the lid was opened is not important. All that matters is that to you, outside the box, he remains in a quantum superposition until you open the box yourself, whereby your consciousness forces the wavefunction to collapse to one or other alternative.

  Don’t tell me the score

  While chatting to my brother recently about quantum mechanics and football (what else is there?) he relayed what must be a familiar feeling many of us have that is quite similar to the role of the observer in quantum mechanics. You see, we are both lifelong supporters of Leeds United,5 and he had often wondered whether, if he recorded a Leeds match on video and then sat down to watch the recording without knowing the score then, for him, the match does not yet even have an outcome. But if we push the standard interpretation of quantum mechanics to its logical conclusion then it seems to be saying that this is not just a matter of his ignorance of the final score; information that is ‘out there’ and known to millions of people. It is only after he has seen the end of the match that the superposition of all possible results collapses, for him, to the one outcome he measures: the result he sees.

  This simple optical illusion shows in a rough way how the act of measurement affects a quantum system. If you stare at any one of the white circles you will notice black dots appearing and disappearing in some of the surrounding circles. But if you switch your gaze to one of these black dots it immediately turns white again and stays white. You can never catch a black dot in the act. Naturally, the way this trick works has nothing to do with quantum mechanics, but I think the analogy is quite neat.

  However, until he calls me to tell me the score, assuming I do not know it yet, then my brother will be, to me, in a superposition of having knowledge of all possible scores. In hearing the news from him, I have in effect carried out a measurement on his quantum state and forced the superposition of the various happy and disappointed versions of my brother to collapse to just one.

  Julie’s words again ring in my ear as I write this. But I would have to agree with her that this really is a ‘load of nonsense’! I prefer to think that there is an ‘objective reality’ out there that exists whether or not I look. To give you a more solid example: a radioactive uranium nucleus buried in the earth will emit an alpha particle that can leave a visible track of crystal defects in the rock. It does not matter whether we look at the rock today, in a hundred years time or never. The track will be there. What if the rock is on Mars and is never discovered by a conscious observer? Does it sit in a state of limbo, both having and not having a track etched into its structure? Clearly measurements must somehow be taking place all the time, and conscious observers, whether they wear white lab coats or not, cannot play a role. The correct definition is that a measurement is deemed to have taken place when an ‘event’ or ‘phenomenon’ is recorded. This would be something we could perceive, if we wish to do so, at a later time.

  The above statement might sound so sensible and obvious that you could be forgiven for wondering how any quantum physicists could be stupid enough to hold an opposing view. But then again, if we have learnt anything about quantum mechanics it is that searching for rational explanations is a futile exercise. Many deep thinkers have struggled with the issues surrounding quantum measurement, and most of the differing views they end up holding cannot be refuted so easily.

  Over the years, I have had many discussions with a theoretical physicist colleague of mine, Ray Mackintosh, with whom I do not always agree, but almost always concede that the points he makes are valid. Our most entertaining ‘debates’ tend to be in bars during evenings while we are away at nuclear physics conferences together. (This is the usual setting for most such deep discussions among theoretical physicists who do not have the ‘pleasure’ of the experimentalists’ night shifts in the lab.) Mackintosh raises the following objection when he hears criticism of the view that nothing exists until it is measured (and its wavefunction collapses). Of course the electron, or atom, or Moon, exists prior to measurement. All objects, whether microscopic or macroscopic, have many well-defined properties (i.e. quantities that are not in superpositions of more than one value), such as their mass or electric charge. These things are not subject to any uncertainty. So just because we must rely on the wavefunction prior to measurement does not imply that object it describes is not real. Of course, certain of its properties, such as exactly where it is or what energy it has, cannot be said to have definite values, but then that is just the nature of quantum objects. We should not dismiss their existence just because we cannot have a mental picture of what they are really like.

  Two stages of the measurement problem

  So what constitutes a measurement? What do we mean by an ‘observation’? And where is that magical transition from wavefunction to actuality? We do know this much: a measurement must first involve an interaction of some kind between a measuring device, which I have not yet specified, and a quantum system. The latter could have until that moment been in a superposition of different states corresponding to different possible outcomes of the quantity being measured. This interaction will lead inevitably to an entangled state of a system plus external measuring device, forcing the latter into a superposition of having measured all the different possible values of the quantity at once. Then what?

  Let us analyse a little more carefully the particle interferometer set-up of the last chapter. Remember that the wavefunction of a single atom will, on entering the device, split into two components, each travelling along one of the arms. And whilst I must describe the atom by a wavefunction, I know something strange must be happening since I see a real interference signal on the other side. I also know that if I look through a porthole in one of the interferometer’s arms to see whether an atom is travelling along that route then I will detect half of the atoms. But now of course the interference pattern disappears.

  To simplify matters I will assume that the way I detect an atom is by opening a small porthole in the arm and seeing a tiny flash as light bounces off the passing atom. So what happens if I now decide to close my eyes? I now have no knowledge of which way the atom has gone; does this mean the interference pattern comes back again? This would imply that the physical act of moving my eyelids is controlling the outcome of the experiment. Not a very convincing argument. What now becomes obvious is that the measurement must somehow be taking place because I have opened the porthole and allowed light to come in and interact with the atoms. The interference disappears whether or not my eyes are open. Of course I could argue that the interference signal at the end is still a possibility (one of three actually: either the atom was detected in the arm, or it wasn’t and must have gone through the other arm, or it was both detected – by the light coming in through the porthole – and not detected, at the same time). The third of these options is the Schrödinger cat state (the dead-and-alive-at-the-same-time option, that is) and leads to the interference signal in the interferometer. Of course I never see this option when I open my eyes, and the conscious observer supporters would still argue that the act of my seeing, or not seeing, the atom is what ultimately eliminates the option that gives rise to the interference.

  The problem is made clearer if we make sure that the flash of light due to an atom going past the porthole leaves a permanent record in an instrument placed outside the interferometer. This now plays the same role as Schrödinger’s cat in the box. Is the instrument capable of existing in a superposition of recording and not recording a flash?

  All in all, we can identify two distinct issues relating to the measurement problem:

  (i) Why is it that we never see superpositions of such macroscopically distinguishable states (such as cats being dead and alive at the same time, and instruments both recording and not recording signals)? After all, we do see the effects of such superposition in the interferometer when the porthole is closed, and of course in the two-slit experiment. But there the states involved were microscopic ones involving individual atoms.

  (ii) Even if there was a way of getting these Schrödinger cat states out of the way before we look, is there not still further wavefunction collapse necessary to remove all the remaining possible options bar the one we actually end up observing? Thus, Schrödinger’s cat could be dead, alive or both. Removing the ‘both’ option still does not tell us how one of the other two possibilities is discarded when we open the box.

  In an atom interferometer device, the atom’s wavefunction is forced into a superposition of travelling along both paths at once. But if we open a porthole to check which way it has gone, we force it to make up its mind. If we observe the atom then we instantly collapse the component of the wavefunction in the other arm to zero. Even if we don’t see it, we still force it out of the quantum realm to become a definite particle, travelling now via the other arm.

  Decoherence

  During the 1980s and 1990s, the first part of the problem was cleared up. Physicists came to appreciate what must be happening when an originally isolated quantum system such as a single atom, existing happily in its superposition, becomes entangled with a macroscopic object. It turns out that the superposition of different states forced upon such a complex system involving a trillion trillion atoms simply cannot be maintained and very quickly disappears, or decoheres. One way to think about it is to say that the delicate superposition gets irretrievably lost among the stupendously large number of other possible superpositions due to the different possible combinations of interactions between all the atoms in the macroscopic system. Recovering the original superposition is a little like trying to un-shuffle a pack of cards so that all four suits are separated, only incredibly more difficult.

  Decoherence is a real physical process that is going on everywhere all the time. It takes place whenever a quantum system is no longer isolated from its surrounding macroscopic environment and its wavefunction becomes entangled with the complicated state of this environment. This might be anything from a photosensitive screen or electronic device to the surrounding air molecules. If the coupling to this external ‘environment’ is strong enough then the original delicate superposition very quickly gets lost. In fact, decoherence is one of the fastest and most efficient processes in the whole of physics. And it is this remarkable efficiency that is responsible for decoherence having evaded discovery for so long. It is only now that physicists are learning how to control and study it. I will describe in Chapter 10 a few of the remarkable experiments that have been carried out over the past few years in which we can actually see decoherence in action.

  Technically, physicists talk about the environment wavefunction losing all remnants of the initial phase correlations between its two entangled pieces. A more colourful phrase is simply to say that once a quantum superposition gets entangled with the outside world, all the weirdness leaks out and gets lost so quickly that we never see it in action.

  The process of decoherence is still an area of active research and is not yet fully understood. But we can at least begin to make sense of the first part of the measurement problem. The reason we never see Schrödinger’s cat both dead and alive at the same time is because decoherence takes place within the box long before we open it. This is not even down to the cat but would take place much earlier, due to the device housing the radioactive nucleus and the poison. This is what forms the macroscopic environment immediately surrounding the radioactive nucleus.

  Another example in which we can apply the idea of decoherence is the case where a detector registers whether an atom is in one arm of an interferometer or going through just one of two slits. Here, the explanation is easier. The detector must, in order to gain information about the atom’s position, couple to the atom’s wavefunction and the entanglement quickly leads to decoherence. Note that I said ‘the atom’s wavefunction’ and not ‘the atom’ since the detector not registering would imply that the atom has gone via the other route. This is also a ‘measurement’ that will destroy the interference pattern. So, even though the detector and the atom need never physically come into contact in a classical sense, there is still entanglement with the wavefunction.

 

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27