The Many Worlds of Quantum Mechanics

In early spring of 1959, a young American physicist travelled to Copenhagen. He was just 28 years old. His name was Hugh Everett. He was to meet with Niels Bohr, one of the giants in the world of quantum mechanics. When Everett was sent to Bohr, he challenged every part of Bohr's understanding of quantum theory, his worldview. What Everett presented to Bohr was an extraordinary new theory that might be one of the most important scientific ideas of all time. Everett was trampling over every belief that Bohr held dear. Everett's theory came with strange and outlandish implications. It meant that there must be countless parallel universes and many worlds just like our own, populated by alternative versions of us. What's interesting is if quantum theory itself, taken realistically, tells us there are many worlds, that's why we take them seriously. To my view, the many worlds, or the Everett interpretation, if you like, is a reductio ad absurdum. Bohr rejected the idea and a disillusioned Everett left academia for good. But it was a glimpse of a new way of making sense of quantum theory. How does a physicist ask what is the physical interpretation of this quantum theory? And the intellectual battle still rages today. I don't think that there are two worlds. I prefer to have a single world with a number of internal perspectives. But today there is a new theory emerging that many believe will finally vindicate Everett. These results are sensational if you can buy the many worlds. Hugh Everett's idea that there are many worlds may have been dismissed by Niels Bohr, but it didn't go away entirely. In Oxford in the 90s, Professor Simon Saunders was plotting to bring Everett's work back into the mainstream of philosophy. The idea of many worlds was now being taken seriously. My arrival in Oxford, I had this set of ideas, some of it written, some of it not. But what I didn't have were people who got it, who, as it were, could pick up those ideas and run with it. Simon is a philosopher and physicist. Together with Harvey Brown, they brought together a team of academics that is now called the Oxford Group. I think there's a certain degree of serendipity. I think you have to have people who are creative, think outside the box. Simon belongs to this Oxford gang of philosophers who are all enamoured by this idea of the main world and are preaching it around. It was one of the most happy and fulfilling moments in my life to find young people, indeed even undergraduates, responding with real talent above all David Wallace. Simon Saunders nearly single-handedly brought the Everett interpretation to the fore in philosophy of physics. So Simon's work here was both intellectually very important and I think actually personally very courageous because that was a time where in philosophy, if not in physics, these ideas were not really taken very seriously and there was a certain amount of risk, I think, in working in these areas. Undeterred, Simon and his fellow philosophers were brave enough to suggest the unimaginable. What does it mean to have many worlds? Many worlds is difficult. It's fantastical. In the many worlds context, you're supposed to be thinking, oh, here's what's going to happen. The world's going to split into two copies. It does seem to have mind-boggling consequences that there are many individuals almost identical to oneself. There's definitely going to be one future copy of me that sees A and there's definitely going to be one future copy of me that sees B. So quantum mechanics, it begins as a theory of particles, of atoms, of molecules. So what many worlds is saying is that as Simon decides to prune the roses so he splits into two. In parallel universes, one Simon goes indoors and reads a book whilst the other Simon gets on with the gardening. So Everett had the basics. He had the basic idea. A system very much like a tree where innumerable ways of getting to different parts of the top of the tree. So we have the branching structure where each branch is a world. On each branch, we have a different physical state of affairs just like the one now around us. This does all seem rather bizarre. We are familiar with the world around us where everything behaves as we would expect. An apple falls from the tree to the ground according to classical laws. It all makes sense. But at the very small, we see classical laws break down. In the quantum world, particles behave in a weird way. They can be in different places at the same time. A cloud of possibilities. This is the strange world of quantum theory. Quantum theory has to count as the greatest and most revolutionary theory ever found in physics and the most successful. Describing this strange, uncertain behaviour of small particles first seemed impossible. But in December 1925, an Austrian physicist, Erwin Schrödinger, retreated to an isolated mountain cabin to consider the problem. His great insight was to treat the electron of a hydrogen atom not as a particle but as a kind of wave. The equation he produced was the Schrödinger equation. It tells how the state of a quantum system changes with time. Very much like how Newton's second law describes in the classical world the path of an object over time. These changes to a quantum system over time he called the wave function. When we describe an object in quantum mechanics like an electron or a photon or something like that, we describe it by a wave function which is something that's really sort of spread out in space. It's rather like a field in space. Think of the wave function, said Schrödinger, as a catalogue, a directory of all the future possible outcomes for a quantum object. It was this equation that shaped all of Everett's thinking. Just take those quantum equations and run with those. Get everything out of those. That was Everett's intent. And that is exactly what he did. With extraordinary daring, Everett said quantum mechanics should be seen as a universal theory. Just like Newton's law of gravity, Schrödinger's equation should be applied to the whole universe. Do this, Everett realised, and the mathematics perfectly described many worlds. Everett didn't put the many worlds into the theory. The many worlds aren't an extra idea that we bring into quantum mechanics to save it. The many worlds are what quantum mechanics was saying all along. Everett just said, take the theory seriously, take it literally. Everett would have been first introduced to Schrödinger's equation when he was an undergraduate at Princeton. It would have been demonstrated by the two-slit experiment. Every student of quantum mechanics comes across the two-slit experiment with a single photon. It is, as Richard Feynman memorably put it, the central mystery of quantum mechanics. So what we're looking at here is the laser source. Now, this is a laser that can be attenuated, turned down to such a weak intensity that a single photon is passing through this apparatus at a time. We have the two slits. Iconic example of the strangeness of quantum mechanics is a particle that passes through two holes. The photon, the individual photon goes through one slit, some interaction here, arrives at the screen, at the photomultiplier. It's recorded as a blip. What would you expect to see given only one particle at a time passing through the apparatus? Well, if the photons were behaving like, say, tennis balls, you would expect that the balls would hit the screen in just two places behind each slit. But when single photons are fired at the slits, something strange happens. The photons are hitting the screen in a place that would be impossible if they were travelling in a straight line through the two slits. So, instead of two clear bands in line with the slits, the photons create a series of smudges. It looks as if each single photon somehow goes through both slits at the same time and there is interference between the two, as if it were a wave. We have to consider that it passes through both holes because if we think that it passes through one, we don't see interference. We don't see the fringes of interference. Everett was entirely familiar with this device, this apparatus. What he made of it is that if, at the very small, quantum mechanics tells us two contradictory states of affairs are obtained at the same time and that is boosted up to the macroscopic level, then big, as well as small, can be in two places at the same time. We can now do a two-slit experiment, for example, with organic molecules. So, all of the evidence in favour of the microscopic validity of quantum mechanics seems to me to spread all the way up to the macroscopic level. If this sounds a very simple conclusion, then that is because simplicity is at the heart of Everett's many worlds. Thinking in terms of simplicity, the key element is just to be very simple-minded about things. One has a rule, a Schrodinger equation, which tells you how a something, the wave function, evolves over time. The Everettian story is just to say, OK, let me just take that straightforwardly as a description of the world. Things, even people, in two places at once. This is the central idea that nobody had understood prior to Everett. To suggest that quantum mechanics applied to both big and small was certainly a revolutionary idea. Before Everett, physicists had divided the world into two, the large familiar classical world and the small, weird quantum world. And one of the greatest physicists of them all was Niels Bohr. Whenever it was sent to Bohr, he challenged Bohr's views. He challenged every part of Bohr's understanding of quantum theory, his worldview. For Bohr, quantum theory only kicked in at the macroscopic level and only in a well-defined way, given an explicit experimental context. Bohr's philosophy was called the Copenhagen Interpretation of Quantum Mechanics. The interpretation was developed by Bohr with help from his colleague Werner Heisenberg in 1927. It was named after the institute he directed in Copenhagen. However, Bohr's theory may have divided the world into big classical things like tables and chairs and small quantum particles such as atoms, but he also said something else. He said that when you look at a particle in its fuzzy quantum world, it stops having an infinite number of different possible positions and immediately appears as a single particle in one particular place. So in the case of the two-slit experiment, for example, the electron goes through the slit or the photon goes through the double slit, it spreads out, but when we look at the screen, we always find the electron or the photon in one definite place. Bohr's Copenhagen Interpretation provided an explanation for this strange behaviour, and it was all to do with the particle's wave function, its catalogue of quantum possibilities. Bohr said that it was the act of measurement, simply looking at something that collapsed the wave function with all its many possibilities into just the one single outcome, the reality around us that we actually see. But to Everett, this made no sense. Why, for something to exist, does someone need to be looking at it? And Everett was not alone in this thought. So Einstein famously asked, is the moon there when nobody looks? And of course he meant it rhetorically. He was giving an example of something where any fool would concede that the moon is simply there. I think that the moon is there. There's no doubt that the moon is there when we are not looking at it, right? We are not a special object in the universe. We are just natural creatures in a natural world. We're just a piece of the universe, a piece of physics, like everything else. And like everything else, we interact with the rest of the world. So things get a reality for us when they interact with us. Schrodinger, in particular, Schrodinger did not agree with Bohr. The idea that on performing a measurement, something radical and dramatic and discontinuous and indeterministic takes place on measurement, he found absurd. What counts as a measurement? Bring out this absurdity. He gave his experiment involving a cat. What Schrodinger did to the poor cat was use it to sort of make clear how crazy sounding the way people were talking about quantum mechanics is because the cat is just a very vivid way of measuring a system. A cat is locked in a box. In the box is a radioactive substance, a flask of poisonous gas, and hammer triggered by radiation from the decaying uranium. So Schrodinger said, right, you couldn't do this these days. Wire up the possibly decayed atom to some poor cat and to some gadget that kills cats. And so if the atom decays, the cat dies. If the atom doesn't decay, the cat lives. If you take Schrodinger's cat, that would be something where you initially have a cat and then you do an experiment which, according to the evolution of Schrodinger's equation, would provide what we call a superposition of a cat being alive and dead at the same time. And that's what the Schrodinger equation evolves to. But then you say you do a measurement on the cat or look at it and what you see is either a dead cat or a live cat. The idea of superposition is at the heart of the Schrodinger equation. The wave function, remember, describes each of the states of the superposition. It's what makes it like a wave equation. States superpose rather in the way that waves of water can be added to one another. Now quite why looking in the box suddenly makes just one of the states of the superposition, say a live cat, appear, whilst another, the dead cat, just brings out of existence, even Bohr could not explain. It did seem to be all rather mystical. Niels Bohr and his Copenhagen interpretation sort of admitted it was a cheat, in a sense, by saying you've got to treat the measuring apparatus as a classical device. So you don't treat it according to quantum mechanics. If you did treat it according to quantum mechanics, then the measurement would say cat dead and cat alive at the same time. And the thing is that the Schrodinger equation preserves this superposition forever. And so to Everett, it was clear that there was no need to cheat. He said just treat the measurement apparatus as part of the quantum system. This was Everett's big central idea. Simply let the Schrodinger equation apply to everything, the whole universe, including the experimental apparatus. For Everett, quantum theory was to be taken as a universal theory, a fundamental theory that applies to everything, big and small. So apply it to the experimental apparatus as well. Apply it to the experimentalist as well. These ideas were inathema to Bohr. To many physicists, Everett's ideas were also profoundly disturbing. If the wave function didn't really collapse, then this could only mean one thing. Every possible state of the superposition of the cat would be preserved. The Everett interpretation had arrived at the fantastical idea of many worlds. Two cats, each cat in its own separate parallel world. And imagine people also involved in this experiment. We have the superposition of people seeing the cat alive, the superposition with people seeing the cat dead. So instead of just being cat alive and dead at the same time, there's now cat alive, David sees living cat, and cat dead, David sees dead cat at the same time. It's the inevitable consequence of the equations. Everett was prepared to run with that. Nobody else before him had the courage to suggest this. But the measurement paradox afflicts the recovery of anything like the world that we see in the mathematical formalism. But for Simon and the Oxford group, their support of the many worlds interpretation means facing the criticism of leading physicists who simply do not buy Everett's ideas. For Roger Penrose, the problem with the Everett interpretation is Schrödinger's equation itself. If you believe that the Schrödinger equation, or unitary evolution as it's sometimes more correctly called if you like, that unitary evolution is an exact law of nature, then you are led to universes in which all these different alternatives somehow coexist. Now the question is, why don't we see that? And this is where the problem for this Everett type interpretation comes in, because we don't see that kind of world. Simon is following the steps of Schrödinger, and so what exists is this wave function. This is a realistic position, that's the thing which is out there. It creates a number of problems, like how do you go from these wave functions to the actual things of the world that we see, which are chairs, shoes, trousers, air, particles, stuff. To my view, the many worlds, or the Everett interpretation if you like, is a reductio ad absurdum. It's telling you what, if you simply believe that the Schrödinger equation, or unitary evolution, is exactly true of the world, you get nonsense. Well, let me use a metaphor. We are used to the fact that each of us has a different perspective on the world. I see this room from a different perspective. I see a chair from here. You see a chair from there. Another person sees a chair from there. So the complexity of reality can be seen as an intersection of perspective that match or don't match. Simon wanted to break it up into a multiplicity of worlds. I prefer to have a single world. With a number of internal perspectives. Carlo is not prepared to apply quantum theory to everything. I've talked about big and small. Even more important is that it's everything. For Carlo, there's always the observer who is outside the system. Many worlds is difficult. It's fantastical. And in a way, there's concern that the ideas can't be taken seriously. This has been extremely difficult for me in those early days. The trouble was Simon knew all too well that Everett's theory was not without its problems. Everett may have described the development of the superposition of worlds, but he could only do so through experiment. But there was the problem. The only way that he had of getting out these macro superpositions was in the context of an experiment. It seemed it was no different from Bohr. You needed the experiment. You needed actually the dynamics to be cooked up for the experiment to give you the superposition of worlds. That leads you to the many worlds theory, which is my... Everett's insight, it didn't really take measurement itself out of the story. So it didn't fully fulfill Einstein's hope of a physics with no reference to measurement or observation. I mean, what about a world, a universe without any people, without any experiments? Would there not be many worlds in that universe? Is it just somehow with human beings that we create these many worlds? This made nonsense of the theory for many people. Try to reformulate quantum mechanics to avoid troublesome mention of measurement or observation. But Simon had arrived to Oxford with a set of ideas that he believed would solve this problem and vindicate Everett. The 1980s had seen the emergence of a new mathematical tool not available to Everett. It was called decoherence theory. The Oxford group believed that decoherence theory could describe how the branching of worlds occurs regardless of whether measurements are performed. And why it is that the world in which we exist never recombines with all the other branching worlds. And why each world is oblivious to all others. What decoherence theory taught us was that this constant branching is something that's happening naturally just as a result of physical processes all the time. The University of Oxford, today at the forefront of quantum computing, is no stranger to decoherence theory. Quantum computers work by maintaining the quantum state of a system in a superposition of states, each computation delicately interacting with each other. If there is any contact between the system and the surrounding environment, then just as the ripples from a stone dropped in a lake disappear into the environment, so the interference effects, so typical of the quantum, are washed out. This is decoherence. And when it happens, all quantum behaviour is lost. I took over these methods developed by others, decoherence theory, and very specifically the decoherent histories formalism, because that is the one that really makes best sense of Everett's ideas. And I showed how the standard problems that beset the theory could be solved. Decoherence theory turns out to be the answer of why it is that one actually gets a plurality of macroscopic determinate worlds out of a fundamentally, if you will, formless evolving wave function. The whole language of worlds and branches becomes okay when those interference effects get washed out, and that's what decoherence does. First and foremost, you no longer have to make reference to experiments. That was the most important of all. These ideas work for any kind of macroscopic process. There doesn't have to be a measurement process. In fact, all it needs is the universe itself. It is the leaking of the quantum interference effects into the environment that makes it appear there is wave function collapse. Interference, which is the quantum phenomenon that leads to all the paradoxical results of the two-slit experiments and the like in quantum theory, interference is a delicate flower. And when you're trying to create quantum mechanical phenomena on large scales, then very quickly that interference disappears. And that disappearance of interference is what we mean when we say that the worlds, the branches in the quantum wave function stop interacting with each other. When decoherence happens, that's when somebody in one branch can no longer interact with the going on in another branch. It all sounds rather fantastical that we are surrounded by numerous other universes that don't interfere with each other and that we can only see the one we ourselves are in. But is it really so strange? The collapse of the wave function is this idea of signals and radio waves being made at the same time. Each of them is oblivious to all of the others that are being made. But we think, hang on, there can only be one signal. So all of the others somehow have to go away. Why do the others have to go away? If they don't interfere with one another, why couldn't they all be present at the same time? And of course they are. We tune into different radio stations. We only hear one at a time. Each frequency has decohered from all the others. Just as we are surrounded by different radio channels, so we are surrounded by entire universes, all possible versions of the superposition, all independent from each other. The process of decoherence provided the basis of a unique way of looking at the superposition principle. In other words, it provided a unique account of how these branches were described within the theory. If decoherence has happened, then that's when the many worlds interpretation will say branching has occurred. And then its description of a situation will be, no, there are really two cats here, one on one branch of the universe and one on the other branch of the universe. One of them's alive, one of them's dead. It's all quite classical. And it's how this classical world emerges out of the wave function, where decoherence finally answers Everett's critics. Just like radio stations, decoherence theory has a unique way of showing the superposition of worlds. The Schrödinger equation dictates how the sequence of superpositions unfold. But track an individual term in the superposition from one time to the next, and what emerges is a classical world. Do the same for each term, and you have, like radio stations, many parallel worlds, each obeying approximately classical equations, each story slightly different from each other. And it's not just Newtonian laws that reveal themselves, but also equations for gases, for fluids, equations for crystals, for chemical reactions, for air particles, for materials that chairs, tables, and books are made of, and even cats. Decoherence theory gave us macroscopic superpositions without measurements. They're happening all the time. It gave us the full vividness and richness and variety of macroscopic world and matter in all its forms. It took those branches and it showed why branching structure is there, why you don't get recombination of branches. The contribution of Simon and his colleagues to the idea of many worlds is nearly done. Bohr relied on the idea of classical experiments, Everett replaced them by quantum experiments, the Oxford group got rid of the dependence on experiments altogether, and they showed how the classical reality of our own world arises out of the wave function. Whilst the debate about many worlds will continue on, the Oxford group is now breaking up. After two decades, they are now going their separate ways, some to new posts in America, others to retirement. But their legacy is clear. You basically only need things like the principle of superposition and the basic mechanisms of decoherence for the many worlds story to emerge. These results are sensational from the point of view of realism, if you can buy the many worlds. What I believe is that the only way to interpret quantum mechanics realistically is in terms of parallel universes. There are parallel universes insofar as quantum mechanics is true, and I don't know if quantum mechanics is true.