The cosmologist reflects on his fields’s recent, unexpected twists, and on why he’s come to embrace string theory
Interview by Davide Castelvecchi
August 12, 2004
|On the night of December 6, 1979, Alan Guth had the “spectacular realization” that would soon turn cosmology on its head. He imagined a mind-bogglingly brief event, at the very beginning of the big bang, during which the entire universe grew exponentially, going from microscopic to cosmic size. This explosive epoch, fueled by a yet-undetermined kind of antigravity, could solve in one stroke several of the problems that had plagued the young theory of the big bang. Twenty-five years later, I asked Guth for his thoughts on his legacy and how it fits with the discovery of dark energy and the recent excitement in string theory. Quotes from this interview have later appeared in my Symmetry magazine feature article.|
Davide Castelvecchi: What is cosmology?
Alan Guth: Cosmology is the study of the history and large-scale structure of the universe, and my own niche in cosmology is the very early universe — the first small fraction of a second of the history of the universe.
DC: How is it possible that people can understand the universe itself, as opposed to things the universe contains?
AG: We do have a number of pieces of information that we can put together to try use as a basis for constructing theories. Observations about the distributions of galaxies within the visible part of the universe, and the motions of galaxies. Also now very important are observations of the cosmic background radiation — radiation that we believe is the afterglow of the big bang’s explosion itself. And now we have very precise measurements, both of the spectrum of this radiation and also of the small ripples that exist in its intensity pattern. The radiation is almost perfectly uniform. In all different directions in the sky, the intensity we observe is the same to about one part in 100,000. But nonetheless, one does see minute differences from one direction to another. This pattern of ripples is tied directly to two things: theories about how the ripples were formed — which is where inflation comes in — and also to theories that calculate how the structures in the universe have formed from the ripples. Another important ingredient in terms of the observational basis for cosmology is the chemical abundances that we observe in the universe, Those are measured from the spectral characteristics of gas clouds and stars, and can be compared with theories about how the chemical elements were formed in the first few minutes of the history of the universe. And wonderfully, the calculations agree very, very well with the observed abundances of the lightest elements.
DC: When you first had the idea of inflation, did you anticipate that it would turn out to be so influential?
AG: I guess the answer is no. But by the time I realized that it was a plausible solution to the monopole problem and to the flatness problem, I became very excited about the fact that, if it was correct, it would be a very important change in cosmology. But at that point, it was still a big if in my mind. Then there was a gradual process of coming to actually believe that it was right.
DC: What’s the situation 25 years later?
AG: I would say that inflation is the conventional working model of cosmology. There’s still more data to be obtained, and it’s very hard to really confirm inflation in detail. For one thing, it’s not really a detailed theory, it’s a class of theories. Certainly the details of inflation we don’t know yet. I think that it’s very convincing that the basic mechanism of inflation is correct. But I don’t think people necessarily regard it as proven.
DC: You recently wrote that “the case for inflation is compelling,” which sounds like a cautious statement.
AG: It’s certainly not as well confirmed as the big bang theory itself. But I guess I’d find it hard to believe that there could be any alternatives for solving the basic problems inflation solves, like the horizon and flatness problems.
DC: Do you have your favorite version of inflation among the many that have been proposed?
AG: Not really, except that I could say that I think cosmology is moving toward describing things in terms of string theory. And there have been a number of attempts to describe inflation in that context. I think that is the future.
DC: So you think that string theory will ultimately prove to be right?
AG: Yes, I do. I think it may evolve a fair amount from the way people think of it now, but I do think string theory definitely has a lot going for it.
DC: Is string theory physics or is it just fancy mathematics so far?
AG: I consider it physics. It’s certainly speculative physics so far — unfortunately, it’s working in a regime where there’s no direct experimental test. But there are nonetheless consistency tests. If the goal of string theory is to build a quantum theory that’s consistent with general relativity, that’s a very strong constraint, and so far string theory is the only theory that seems to have convinced a lot of people that it satisfies that criterion. Just from a sociological point of view, theoretical physicists have been looking for a consistent quantum theory of gravity for at least 50 years now, and so far there’s really only one theory that has reached the mainstream — string theory.
DC: Has string theory reached the mainstream?
AG: Yes. I would say that nowadays, a theoretical particle physicist cannot ignore string theory.
DC: Speaking of sociology, in your book you describe your first attempts as a young particle theorist to describe your idea of inflation to cosmologists, and how communication would break down because people used different lexicons. Is the situation any different now?
AG: I think the situation has improved tremendously between particle physics and cosmology. Now I think that almost everybody in cosmology is reasonably fluent in the vocabulary of both fields, and I think everybody recognizes that there is a strong interface between these two fields. At the same time, now there are also important implications going the other way, with the discovery of dark energy.
DC: Is dark energy more relevant to particle physics than dark matter?
AG: I would say yes. I am not sure if everybody will agree — it depends on what your perspective is. I think dark matter is more relevant to the next age of particle physics experiments — hopefully supersymmetry and perhaps other interesting things that we may discover. On the other hand, there’s at least a good chance that dark energy is energy of the vacuum, so it seems to be telling us something about the fundamental structure of physical law, which is a big surprise. The vacuum energy has been a haunting question for particle theorists since the advent of quantum field theory in the 1930’s. As soon as we had quantum field theory we knew that the vacuum was not a simple state: It was a very complicated state with all kinds of quantum fluctuations going on. And there was no reason at all why the energy of the vacuum should turn out to be zero or small. In fact, nobody knows how to calculate the energy of the vacuum, but if particle physicists were to try to estimate it, the natural answer would be something like 120 orders of magnitude larger than the experimental bound. So it was always a big mystery, but until the advent of dark energy, the belief was that the real number was zero, because of some kind of symmetry that we didn’t understand yet — an exact cancellation between the positive and negative contributions. If dark energy is the energy of the vacuum, now you need that symmetry to make it almost zero, and then some small breaking of that symmetry to make it a small number that’s not zero. And it all gets very complicated and baroque. Nobody has the faintest idea of how it might actually work. There is also the possibility that the vacuum energy is not determined at all by the fundamental laws of physics, but instead it’s determined anthropically, using the idea of a multiverse. It’s quite possible in the context of string theory that there are many vacuum-like states, and all of them are stable enough that they could provide the underpinnings of a universe. And the one that we happen to find ourselves in is determined by random choice. One would imagine that the universe would inflate eternally through all the different possible vacua of string theory, with infinite amounts of space of every type of vacuum produced — eventually.
DC: Is this the landscape idea?
AG: Yes, that’s the catchword. If this is right, it would mean that in most regions of space the cosmological constant is enormous, and there are some rare regions of space where the cosmological constant happens to be very small. But life can only form if the cosmological constant is very small. So it’s not a surprise that we find ourselves living in one of those regions. An idea like this five years ago would have been completely anathema to particle physicists. It is still anathema to many, but people pay much more attention to this kind of idea now.
DC: Does this connect to the idea of eternal inflation, with different universes bubbling off?
AG: Yes, there are two ideas coming together here. One is the idea from string theory, that there’s a huge number of possible vacuum states. And the other is the idea of eternal inflation, that once inflation starts, it never ends, and it explores all possible vacua.
DC: Andrei Linde and his collaborators have written a paper with some string theorists that tries to reconcile the two theories — the paper now known as KKLMT.
AG: Yes. I regard that as probably the most interesting approach. I’m a big fan of that work, though I’m not one of the authors. I think it’s the starting point towards what will become a solid embedding of inflation within the context of string theory. Before them, nobody had any good idea for describing within string theory a state that would have a positive cosmological constant.
DC: Does the existence of dark energy suggest a possible connection between the “false vacuum” state that produces inflation and the “true vacuum” state of the cosmological constant?
AG: In principle, yes, although the vacuum states in string theory are really quite complicated states, with a number of degrees of freedom that describe them. Certainly, the state which drove inflation in the early part of our universe had a large, positive cosmological constant. In the end, they would all be described in the same language of string theory, and they would have many similarities. But there also are many significant differences. They are very different energy scales. So I think it’s somewhat a question in the mind of the beholder to decide whether or not there is a close relationship or a distant relationship.
DC: Could there be two different kinds of repulsive gravity then, one which acted during inflation, the other one which is acting now?
AG: What I believe, and what is the conventional belief, is that the repulsive gravity is really a feature of general relativity itself — and in fact Einstein made use of it himself in 1917 when he introduced the cosmological constant and tried to use it to describe how the universe could be static, with ordinary gravity pulling everything together and repulsive gravity — the cosmological constant — pushing everything apart. So from the very beginning general relativity incorporated the possibility of repulsive gravity. What creates repulsive gravity is negative pressures. That’s the feature of the cosmological constant and also of states of scalar fields dominated by their potential energy, which is the way conventional inflation works. Certainly the most plausible explanation for acceleration today, and for inflation early in the universe, was that the universe contains materials that have negative pressures. So at that level of description it’s the same mechanism — because it’s the only mechanism we know. But what the material is that creates the negative pressure is a more detailed question. Whether or not we believe that the KKLMT papers are on the right track, I think we don’t really know how closely related the actual state that drove inflation in the early universe was to the state the universe is in now, with this slow inflation that we attribute to dark energy.
DC: Could there be a particle physics experiment to probe dark energy?
AG: I guess I do not see the dark energy influencing or being influenced by particle physics experiments in the foreseeable future. It certainly is highly relevant for astrophysical observations. One important thing we’d love to know about dark energy is whether or not the energy density is constant over time, as it would be if it were a cosmological constant. Or, it could vary with time — in which case, our best explanation would be that it’s energy of a slowly evolving scalar field that fills all of space. That’s usually called the quintessence. There is some hope of answering that question by more detailed astronomical observations. And the best handle of that is probably still the distant supernovae, with experiments such as SNAP.
DC: So is dark energy relevant to particle physics not so much on the experimental side, but because it points to an open problem in its theoretical foundations, i.e., the quantum field theory?
AG: Yes, in terms of trying to understand the foundations of theoretical particle physics, I think it’s very important. In particular, it seems to be suggesting that there may be no physical principles that determine what the vacuum of string theory is. Maybe it is just all possible vacua happening in all different places. Now, I really hope that that turns out not to be the case, because I like to think that physics is more predictive than that. But that is certainly the direction that the dark energy is pointing towards — and it may turn out to be the right direction.
DC: In either case, will a better understanding of dark energy shed light on inflationary cosmology?
AG: Yes, I think so. If it turns out that the only explanation for the dark energy is this landscape idea, that says that if we want to understand how inflation really works, we have to understand it in the context of the landscape of string theory.
DC: Does inflation rule out the possibility that the universe might be spatially closed — what mathematicians call topologically compact?
AG: Not completely. The statement that the universe is flat is only an approximation. Inflation drives the universe towards flatness — in fact, if enough inflation happens, it drives it incredibly close to being flat. But you could still imagine a universe that started out closed, and at the end it would be very large, but still closed. It would look flat, because the radius of curvature would be huge. On the other hand, it does all become much more complicated, because remember that we’re talking about spacetime, and not just space. And inflation tends to make the spacetime structure of the universe very complicated, with inflation continuing in some regions and stopping in others. Imagining the kind of complicated things that can evolve, I think the right conclusion is that the words open and closed don’t really apply anymore. On a very large scale, the universe is really neither of those.
DC: Correct me if I’m wrong: The onset of inflation being a very local phenomenon, the universe to which our physical laws apply isn’t likely to have interesting topology, because it arose from a local fluctuation.
AG: That’s right. On scales much larger than we can observe there might be an interesting topology. But inflation would suggest that in the scales that we can observe, the topology would be locally R 3. But this has not stopped cosmologists from exploring other possibilities. One of the anomalies that people are concerned about currently is the observation by WMAP of the very low values of L — the low multiples. Those fluctuations are significantly smaller than what was expected from inflationary models. It could just be a fluke, but people have suggested other possibilities, such as a universe that is periodic in space, with periodicity of the order of the current horizon distance. But so far people have not found anything along those lines that’s consistent with the data that’s observed.
DC: A mathematician called Jeff Weeks, together with a group of physicists, have published a controversial paper in Nature last fall. They searched the WMAP data and claimed it revealed a periodicity, with the topology of a Poincaré dodecahedral space. If that evidence were to be confirmed, would it pose a problem for inflation?
AG: Yes, I think it would be very hard to reconcile with inflation.
DC: Virtually all the cosmologists and astronomers I have talked to seem to think that the next big thing will be something called the B-mode polarization of the cosmic microwave background.
AG: Yes, that is very exciting. The B-mode, if present, would be the sign that we have found the effect of gravity waves, and not just of density perturbations. Gravity waves would give us a handle on the energy scale at which inflation occurred. One of the big uncertainties in the wide class of inflation theories is that inflation may have at happened at any of a tremendously broad range of possible energies. The kind of physics that you want to think about, to understand how it happened, depend very much on that. So it would be very important to get some observational information.
DC: Is this going to be an exciting time for you, to see how things evolve?
AG: Certainly, yes. It’s been incredibly exciting, ever since COBE [NASA’s Cosmic Background Explorer, whose results earned its scientists the Physics Nobel Prize in 2006]. In the early days of inflation, when I and a number of other people tried to calculate the density perturbations that would arise from inflationary models, I really never thought that anybody would ever actually measure these things. I thought we were just calculating for the fun of it. So I was kind of shocked when the COBE people made the first measurements of the non-uniformities of the CMB. And now they’re measuring them with such high precision — it really is just fantastic.
DC: And that could happen again — experiments that were considered beyond the realm of possibility will become reality?
AG: Yes, that seems to happen almost every year now.
(Posted on sciencewriter.org January 29, 2007)