The End of Time: a talk with Julian Barbour
Introduction by John Brockman
| Julian Barbour, a theoretical
physicist, has worked on foundational issues in physics for 35 years. He is
responsible for a radical notion of "time capsules which explain how the
powerful impression of the passage of time can arise in a timeless world". He lives on a farm north of Oxford
village and for the past 30 years he has made a living translating Russian while
pursuing his interests in physics.
"I've been working for myself, following my ideas," he says. I wanted to be independent because I'm not the sort of person who can produce a lot of research papers with equations, on a regular basis-I've got quite a good intuition, at least it seems to me I'm always coming up with ideas at least for myself, and some of them stand up to the test of colleagues. I just wanted to be away of all pressure to publish just for the sake of having a publication.
In a profile in The Sunday Times (October, 1998), Steve Farrar wrote: "Barbour argues that we live in a universe which has neither past nor future. A strange new world in which we are alive and dead in the same instant. In this eternal present, our sense of the passage of time is nothing more than a giant cosmic illusion. 'There is nothing modest about my aspirations,' he said. 'This could herald a revolution in the way we perceive the world.'" Cosmologist Lee Smolin notes that Barbour has presented "the most interesting and provocative new idea about time to be proposed in many years. If true, it will change the way we see reality. Barbour is one of the few people who is truly both a scientist and a philosopher."
— JBJULIAN BARBOUR is the author of Absolute or Relative Motion? and the forthcoming The End of Time, which will be published in October (Weidenfeld & Nicholson).
JULIAN BARBOUR: The question I'm always asking myself is, what is the universe and how does it work? I come at it from the point of view of fundamental physics, basic questions of quantum mechanics and its relationship to classical mechanics. Quantum mechanics was discovered in 1925-1926, and it gave a completely new picture of physics which was extraordinarily surprising, and it's still very difficult to understand. It suggests the world is not at all like we see it. That has remained a really big problem, and it's getting more and more discussion, more and more interest from people. This is what I'm really thinking about; how to reconcile the fact that the world seems to be classical, we seem to have unique past, things seem to be in definite positions, and have a definite future-that's what it seems to be like, but quantum mechanics tells us that it is different-not like that at all. The aim is to try and find a description of the entire universe that is quantum mechanical and understand how it nevertheless it can look like the classical world that we actually see and experience.
I came into it quite by chance by reading a newspaper article about the attempts that the great Paul Dirac, one of the discoverers of quantum mechanics, was making about 40 years ago to bring it together with Einstein's general theory of relativity. He'd come across a rather surprising fact and this led him to question whether the picture of space-time that was the whole basis of Einstein's theory really was as fundamental as people had thought. This prompted me to think about time itself. For nearly 36 years now, I've been thinking about time and trying to understand it at the most fundamental level. If you look at the history of physics, surprisingly few people have really thought about time and what it truly is. Even Einstein only thought about certain aspects about time; he never asked what it means to say that a second today is the same as a second tomorrow. This is a very fundamental question. Einstein somehow assumed that it is meaningful, but he never actually asked how does that come about and how can that be? He never defined the notion of duration. So there are aspects of time that haven't been fully studied, in my opinion.
JB: Can you give me another example besides duration?
BARBOUR: Certainly. One of the great questions in physics is whether there's some sort of invisible framework in which everything unfolds. Newton introduced the notions of absolute space and absolute time. Absolute space is like a translucent glass block that stretches from infinity to infinity; it's a fixed frame of reference in which everything happens. Newtonian time is like some invisible river that flows uniformly for ever. The trouble with this is that we can't see this invisible framework, all we see are things moving relative to each other. This is the relational viewpoint, as opposed to the absolute viewpoint of Newton. The challenge has been to create a theory containing genuine relationships between genuine things, and not relationships between real things and unobservable things. That's what I've spent a lot of my time working on. It's given me the ideas which I'm trying now to develop into a complete cosmology, a complete explanation of what the universe is.
JB: Did you ever get to talk to Dirac?
BARBOUR: I tried. I was studying in Munich when I read the article about him. I got so hooked on the issue of time that I went back to England try and see Dirac in Cambridge. I actually spoke to him on the phone, but he wasn't a very talkative person and he wasn't all that interested in meeting somebody who'd got half-thought-out ideas about time. I can certainly understand that.
JB: Are the ideas full-baked now?
BARBOUR: They're certainly not as half-baked as they were. They've definitely taken quite a shape. I hope some at least will have a place in the new picture of the universe for which so many physicists are groping, one that is completely quantum-mechanical and not half quantum and half classical. What my Italian collaborator Bruno Bertotti and I managed to show is that the world is strongly relational according to the physics as we know it now, but this hasn't been properly recognized. The people like Leibniz and Ernst Mach who criticized Newton really were right. Einstein somehow or other put this into his theory without anyone, including Einstein himself, properly appreciating it. The world is relational. It is about how real things relate to real things. This is potentially important for how we try to picture the quantum universe.
JB: How do you fit into the leading edge of today's research-string theorists, quantum gravity people?
BARBOUR: My work has little direct connection to what the string theorists do. There are two main approaches to quantum gravity, and one of them is definitely much more popular than the other, that's the string theory line. I'm following a line that is at least twice as old but followed by far fewer people. It is closely related to basic questions---what is time, what is space, what is motion? Science has its fashions. String theorists are a bit like a pack of hounds following an extremely promising scent. But it is a particular scent. If they lose the trail, nothing will come of the great chase. In contrast, those basic questions will never go away. In fact, if string theory is successful, it will be very interesting to see how it does answer them.
JB: What is distinctive about your approach?
BARBOUR: My basic idea is that time as such does not exist. There is no invisible river of time. But there are things that you could call instants of time, or 'Nows'. As we live, we seem to move through a succession of Nows, and the question is, what are they? They are arrangements of everything in the universe relative to each other in any moment, for example, now.
We have the strong impression that you and I are sitting opposite each other, that there's a bunch of flowers on the table, that there's a chair there and things like that---they are there in definite positions relative to each other. I aim to abstract away everything we cannot see (directly or indirectly) and simply keep this idea of many different things coexisting at once in a definite mutual relationship. The interconnected totality becomes my basic thing, a Now. There are many such Nows, all different from each other. That's my ontology of the universe---there are Nows, nothing more, nothing less.
JB: But where does our experience of the flow of time come from?
BARBOUR: That has always proved to be difficult to attack, because if you try to get your hands on time, it's always slipping through your fingers. People are sure that it's there but they can't get hold of it. Now my feeling is that they can't get hold of it is because it isn't there at all. That what we think is the flow of time---and even seeing motion---is actually an illusion. But I come to that after seeing what the quantum mechanics of the complete universe might be like.
JB: Sounds tough. Have you got a simple picture?
BARBOUR: Let's take a simple model; suppose there were just three particles in the universe and nothing else. In some instant they would be in certain positions relative to each other and would form some triangle. Newton claimed that this triangle has in addition some position in absolute space and that it's changing in time. What I'm saying is that there isn't any of that external framework of space and time, there's just the possible triangles that the particles form. The triangles do not occur somewhere in absolute space at some instant of time, some Now. The triangles are the Nows. You are forced to some view like this if the invisible framework is denied. If we had a universe with a million particles in it there would be some relative configuration of those million particles and nothing else. That would form one Now, and all the different ways you could arrange all the million particles would make all the different possible Nows. I think the actual Nows of this universe are more sophisticated constructs involving fields, but Nows formed by arrangements of particles can get the idea across.
JB: Didn't Einstein abolish Nows?
BARBOUR: In fact no. He only showed that they do not follow one another in a unique sequence. There is no absolute simultaneity in the universe, or at least not in the classical universe. But relative simultaneity remains, and Nows as I define them form an integral part of Einstein's theory. Actually the discovery of Dirac that started all my interest in time was that Nows appeared to be far more significant in the quantum world than one might have expected coming from the normal interpretation of Einstein's relativity.
What really intrigues me is that the totality of all possible Nows of any definite kind has a very special structure. You can think of it as a landscape, or country. Each point in the country is a Now. I call it Platonia, because it is timeless and created by perfect mathematical rules. Most strikingly, it is lopsided with a most definite end and frontiers that are there by sheer logical necessity. For example, if you consider triangles as Nows, the land of these Nows comes to an absolute end in the degenerate triangle in which all three particles coincide. This point is so special I call it Alpha. Other frontiers, like ribs, are formed by the special triangles in which two particles coincide and the third is at some distance from them. Finally, another kind of frontier is formed by collinear configurations---all the three particles are on one line. The Platonia for triangles is like a pyramid with three faces. Its apex is Alpha. All the points on its faces correspond to collinear configurations, and the faces meet in the ribs formed by the triangles with two coincident vertices.
JB: I like the sound of Platonia, but what is it good for?
BARBOUR: My conjecture is that some Platonia is the true arena of the universe and that its structure has a deep influence on whatever physics, classical or quantum, is played out in it. In particular, I believe the phenomenon that we call the Big Bang is not some violent explosion that took place in the distant past. It is simply the highly special place in Platonia that I call Alpha.
JB: I never heard or read other physicists talk like that. What do they make of Platonia?
BARBOUR: Platonia is a special case of a very basic concept in physics called a configuration space. It has been around for a long time, long before relativity. The technical name for any Platonia is a stratified manifold---the strata are what I call frontiers. Stratified manifolds are what remain if you take the potentially redundant absolute structure out of configuration spaces. Stratified manifolds have been recognized as significant for at least sixty years. But somehow configuration spaces, or the leaner stratified manifolds, have never acquired the glamor of Einstein's space-time or the Hilbert space of quantum mechanics. They are the Cinderellas of theoretical physics. I see quantum cosmology as the Prince Charming that cannot live without them. It's a hunch I have come to from thinking so much about those basic questions: What is time? What is motion?
JB: So what do you do in Platonia?
BARBOUR: There are two tasks: first of all, can you describe classical physics using that picture? That's really where my main work has been done, showing that everything Newton could do with absolute space and time can be done more economically in Platonia. That's the first thing Bertotti and I showed. Then we found that Einstein's general relativity, which was created as a theory of space-time, can be recast as a timeless theory in the appropriate Platonia. This is closely related to the discovery that Dirac made and leads on to the second task: what are the implications of the 'Platonic structure' of general relativity in the quantum universe? This is relevant because quantum theories are generally arrived at by starting from a classical picture and performing something which is called quantization.
For non-physicists it's a rather difficult thing to grasp. But you can see where the idea of a timeless universe will come from if you consider the way quantum wave mechanics was discovered by Schrodinger in 1926. In classical Newtonian physics, if you have three particles they will always be at definite positions at definite times. They will form some triangle, and the center of mass of the triangle will be somewhere and it will have some orientation. Now what quantum mechanics says is that, until observations are made, for all these quantities, there are no definite values, but only probabilities, all of which change in time.
The reason that Schrodinger could create a picture of quantum mechanics like that is because he was using the Newtonian concepts of absolute space and time. The framework that they create makes it possible to give probabilities for the triangles formed by three particles to be in different positions and for the probabilities to change in time. There's an independent time which is nothing to do with the contents of the universe. But if you are trying to construct a universe where you say there's no external framework of space and time in which the contents exist, then you can't give probabilities for the particles to be in certain overall positions in the universe and have some overall orientation, because there's no meaning to that. And nor can the probabilities change in time, because there isn't any time in which they can change. The most simple-minded attempt to reconcile quantum physics with the idea that there's no invisible framework holding up the universe---and that idea is made very plausible by the 'Platonic structure' of general relativity---leads you to a picture in which there are just probabilities given once and for all for the relative configurations of the universe. So if we had a three-particle universe the probabilities would just be for where the three particles are relative to each other, say, two close together and one further apart. That's the complete story---static probabilities for static configurations, which are what I identify with Nows.
Thus, quite a simple argument leads to a picture where you just have possible Nows, and the Nows are defined by how the things in the universe are arranged. That's all you get out of the theory. In fact this picture, which Dirac helped to create, crystallized something over 30 years ago. It is described by an equation called the Wheeler-DeWitt equation. (John Wheeler prodded Bryce DeWitt into its derivation. If it really turns out to be the equation of the universe, the episode will be a rerun of the way Hooke badgered Newton into his solution of Kepler's problem.) People found it very difficult to make sense of the static universe that seemed to emerge. However, I find the arguments that lead to it are strong. There is support for it in the structure of Einstein's theory and in the structure of quantum mechanics. The equation would never have been found if that were not the case. So I take the picture seriously and try and make sense of it, and ask how can we nevertheless recover from it a picture of our world; how can it be that I can sit here and see my own hands moving, yours too, if the world is completely static?
JB: Does this have anything to do with your idea of time capsules?
BARBOUR: Yes. Suppose we accept the quantum universe is static and timeless. How can we reconcile that with actually seeing motion and remembering the past? In fact, besides the direct sensing of change of one kind or another, the only direct evidence we have for time and the past comes from records, which include memories. Now records, either natural like fossils or man made, are so ubiquitous we can easily forget how remarkable their existence is according to the current understanding of classical mechanics. This is the problem of the extraordinarily low entropy of the universe. It was emphasized a century ago by Boltzmann. In the modern context of general relativity, Roger Penrose keeps on pointing out what a huge problem it is. All statistical arguments based on classical mechanics suggest the universe ought to have a vastly higher entropy and exist in a state in which records simply cannot form. Penrose wants to explain the low entropy and the arrow of time by a new physics which is explicitly time asymmetric and comes with a built-in arrow of time and forces the universe to begin in a highly uniform state. My own view is that, paradoxically, the arrow of time may be easier to explain in a theory in which there is no time at all.
I suggest that our belief in time and a past arises solely because our entire experience comes to us through the medium of static arrangements of matter, in Nows, that create the appearance of time and change. Geologists certainly deduced that the earth has an immensely long history from structures frozen in rocks. That is, evidence for time and motion in static form. Our long-term memories must also be hard-wired in the patterns of the neural network in our brains. Again, we have mutually consistent records in static form. It is even possible that when we see motion the material counterpart of the phenomenon is a pattern of neuronal connections that codes several different positions of a moving object at once, and the appearance of motion arises from their simultaneous presence in one brain configuration. As I have no expertise in neuroscience, I do not want to push this idea hard. I merely want to suggest that the appearance of time arises exclusively from very special matter configurations which we find can be interpreted as mutually consistent records of processes that unfolded in a past in accordance with definite physical laws that involve time. I call such configurations time capsules and take the perfectly conventional realist view that they do exist in an external world. However, I think they may arise in a lawful manner that does not involve time at all. If we take the Wheeler-DeWitt equation at its face value, this is what must happen.
JB: That's rather hard to believe.
BARBOUR: I'm more optimistic and I am by no means completely alone in thinking that the appearance of time can arise from an essentially timeless universe. There is a quite long-standing regular research program involving at least twenty recognized physicists devoted to the problem. One of Hawking's main papers contributed to it about 15 years ago. The aspect of the problem that most excites me is that it might abolish the dichotomy between laws of nature and the initial conditions that you have to add to them before you can make any predictions. The really interesting thing about quantum cosmology is that it should be in a position to make predictions about the universe which can't be made within classical physics.
Classical physics as it exists now has laws and it has initial conditions. If quantum cosmology really is static, there's no initial conditions. There is no time. You can't set conditions at an initial time. In my view, this means that quantum cosmology is potentially much more predictive. It should be able to predict phenomena like the arrow of time and the low entropy of the universe that in classical physics you would just have to attribute to initial conditions, and just say, well the Big Bang happened to take this form rather than a different form.
This is where I think the strongly asymmetric structure of any Platonia that we use to describe the universe could be highly significant. This possibility does not seem to have occurred to other physicists, but I think it must be relevant. If the universe really is governed by something like the Wheeler-DeWitt equation and we interpret it as determining the relative probabilities for the various different possible configurations of the universe to be realized, or experienced, then a major factor in determining how those probabilities are distributed must be the overall shape of the arena on which it acts. But that is some Platonia, in which there is always a distinguished point Alpha, from which the complete arena opens up somewhat like a flower. Surely this structure must put a rooted bias into the game. My conjecture is that this static bias in the structure of the arena funnels the higher probabilities onto the special configurations that are time capsules, which, being more probable, are the ones that we are most likely to experience. The overall structure of the arena is reflected in the experienced configurations and interpreted by us as time and a past. This may seem fantastic, but I think the arguments for a timeless universe are quite strong. If we accept them, then we must look for something really radical and powerful that does put the appearance of time into the universe. The difference between past and future is a massive asymmetry. I believe that it can only arise from some other massive asymmetry, which I find in the structure of Platonia.
JB: Is there any way of testing your idea observationally?
BARBOUR: I cannot as yet see any direct experimental way of testing this particular idea. What is needed above all is development of the mathematics. In my view, quantum cosmology is rather like the quantum physics of the stationary states of huge molecules, and the development of ideas used in atomic and molecular physics might help. Quite a lot of work has already been done in this direction in the program I mentioned earlier. But in principle predictions can be made in the context of the Wheeler-DeWitt equation. However, nearly all of them at this stage are rather difficult because of the mathematics. You're dealing with complicated systems, you don't know how to find the solutions; there's a whole lot of issues there.
Another issue is that cosmology is a special subject because it's dealing with a unique thing. There is only one universe. There's a philosophical question: how can you do real science on a unique object?
JB: Have you ever considered that the world is an invention, and there's not an a priori existence before discoveries of people like yourself?
BARBOUR: I believe in the world; I'm a realist, but it remains a conjecture.
JB: Isn't this a kind of naive materialism? Wallace Stevens addressed the idea that the words of the world are the life of the world. Norman O. Brown noted that nature isn't created ... it is said. But I don't know of any notable physicists today who are seriously concerned with the role that language plays in the creation of reality.
BARBOUR: The words are meaning something. What impresses me is that, despite what you say, the rules of the game of science have stayed amazingly constant, despite the fantastic changes in how we see the world. Basically, the assumption has always been that that there are material things that move around subject to the constraints of geometry. There is change subject to order. Science---or at least physics---has been about establishing how those changes take place and describing them mathematically. Every now and then they lead to a dramatic new way in looking at the world, but the rules of the game have always been the same really, going right back to geometry and ancient astronomy.
I personally believe the world is
still probably very much richer than we imagine, and that we still may well be
only just scratching the surface of it. If you climb a mountain range you get
different views as you go up. When you've got to the top you can understand what
you could see lower down, but you couldn't understand it properly when you were
lower down. I see the progress of science being like that, that suddenly
completely new vistas are opened up, and you find new ways to think about it. I
do think we are discovering the world, not inventing it. But John Wheeler
sometimes seems to suggest that we create the universe. He thinks that by
insisting on finding a consistent description of it we conjure it up by a kind
of conspiracy. He illustrates the idea by a variation of the game of twenty
questions in which there is no object at the start of the questioning. Instead,
each answer that is given must be consistent with all those already given.
Eventually, there emerges some object that matches the answers.
All physics one tautology;
BARBOUR: As it happens, this is about an issue very closely related to quantum cosmology. What is the meaning of general covariance? Empson is arguing that it is a tautology. I think he is right in that but not in the claim that all law becomes the assumption of the description. Tensors relate different things and bring them into lawful connection.
Let's consider a great experience I just had. I witnessed the total eclipse of the sun in France on August 11th. I remember vividly reading as a boy that there would be this eclipse in 1999, the first one visible in England in a long time---actually I went over to France to see it. Now there isn't any doubt about an eclipse; either you see an eclipse or you don't. The astronomers predicted that it would be seen as total in the medieval town of Senlis just north of Paris, and sure enough we did see it total. That's pretty impressive. I don't think there's anything to do with inventing there---the impression of actually seeing the sun totally eclipsed is quite unambiguous. My memories of reading about the eclipse as a boy and my memory of the actual eclipse are not tensors but they are real different things that match up. That's why I believe there's something real out there in the world and that we are getting our hands on it.
JB: Your theories appear to be bumping up against the ideas of many of your colleagues? Who might be sympathetic?
BARBOUR: The string theorists will probably not take it very seriously, because I'm tied to a certain approach to attacking the problem and they will perhaps just say well he's doing the wrong thing. But I don't worry too much about that, because certain approaches can become unpopular for a long time and then come back in again. There are certainly people who do think about the fundamental issues of how you describe the world and these very basic questions of what is motion and so forth; they should be sympathetic to my approach. The people who will not like me are the people who don't want to take the many-worlds interpretation of quantum mechanics seriously, and who are trying to modify quantum mechanics so as to banish that specter. I don't think Roger Penrose would take it very seriously, because he doesn't like any form of the many-worlds interpretation of quantum mechanics. Stephen Hawking might be quite sympathetic, because he's been basically working along those lines for many years now. Other people who I would not expect to be sympathetic are Murray Gell-Mann, Jim Hartle, and several others who have developed the so-called consistent histories approach to the interpretation of quantum mechanics. They start out by taking a complete history as being a fundamental concept, whereas for me the Now is the starting point. They start with a string of Nows.
JB: Lee Smolin certainly talks about your ideas.
BARBOUR: Lee and I are great friends, and we've talked a lot and we've developed ideas together---we don't follow exactly the same line now, but we share a lot in common. We are both very interested in structure, and in thinking that the way things are structured is the really important thing. That relates to this basic issue of whether the world exists in some all-powerful background or whether the world is really just an interaction between the things that are in it. We share that view very much.
JB: What did you think about the discussion of your ideas on Edge?
BARBOUR: The people who commented tended to be the people who know about the so-called problem of time, that time seems to disappear altogether when you try to make a quantum universe in this particular approach. Basically they were fairly sympathetic to my viewpoint.
JB: What are you working on at the moment?
BARBOUR: First of all, I'm working on some ideas which try and take scale out of physics. This is analogous to the issue of duration: how can you say a second today is the same as a second tomorrow? You can also ask, how can you say that an inch here is the same as an inch on the Andromeda nebula? That's also a very deep question, and I think I've made some progress with an Irish colleague in working on that.
What I really hope for is to build up more and more evidence for my overall picture of what the universe is like. You never can tell whether you're on the right line, but I just can't stop thinking about these things. If you look in the history of physics, you very often see that before a definite breakthrough is made, there's some sort of qualitative idea appears, and then gets made more precise, more mathematical. I would hope that I'm making some contribution to such a development, not only with qualitative ideas like Platonia and time capsules but also in simplified models as in my work with Bertotti.
JB: Some have said if your theories are correct our perception of the world will change dramatically.
BARBOUR: Certainly if I, or other researchers, are on the right track, there's definitely got to be a different overall view of the world. But it's very hard to predict exactly how it would change things. How could Copernicus know what would come out of his proposal? What I feel for myself is that by concentrating on the things that we know are in the world, it makes one think about the actual world more, and I would say cherish it and value it more, and perhaps take a more relaxed attitude toward life and sit back and enjoy it more. This is actually happening to me personally---maybe it's just because I'm getting older and don't want to miss things, but certainly I'm aware of savoring the moment more than I when I was young. And it's partly influenced by my idea that really the universe is static, and the only things that are real are Nows, in one of which we now are.
Some years ago, I heard Dame Janet Baker interviewed on radio. She was asked if she ever listened to her recordings and, if so, what were her favourites. She said she hardly ever listened to them. For her, every Now was so exciting and new, it was a great mistake to try to repeat one. In her singing, she made no attempt at all to recreate earlier performances and do the high points in the same way as the night before. Again and again she spoke with the deepest reverence of the Now and how it should be new and happen spontaneously. "The Now is what is real", she said. I thought it was the perfect artistic expression of how I see timeless quantum cosmology.
JB: Well, you are clearly into the big questions. Does anything less than everything mean nothing to you?
BARBOUR: I'm nothing if not ambitious in that way. How well I succeed is another matter, but it keep's me happy trying.
JB: Are you so in love with the universe that you have to pretend human emotion?
BARBOUR: Absolutely not. No, I'm human emotion through and through, but I love the universe too. I would love to show that the universe is as rich as it appears to us through our senses and not rather drab as it appears in science. For me one of the great miracles is colors, and different sounds and sensations. This is the problem of the qualia. I would love to create a physics in which they exist and form a fundamental part of the physics just like mass and electric charge do. But that really is a distant dream.