The Cyclic Universe: An Interview with Princeton Cosmologist Paul Steinhardt

Interviewed by Paul Halpern on November 5, 2002

Paul Steinhardt, Wikimedia Commons

Why did you become interested in a five-dimensional cosmology, after originally advocating a four-dimensional universe?

The goal was to look for an alternative to the standard model — the inflationary/ Big Bang picture. The history is that a decade ago there were a number of different kinds of models that people were talking about and the data over the last decade has really driven out essentially all the competitors of that time other than the inflationary/ Big Bang picture. That’s both good and bad. It’s nice to see that one is able to use data to rule out things. But when you get down only to a single competitor it’s not always a healthy situation. It’s much better to have two or more competiting models, forcing you to think more carefully about your theories, your predictions and the observations. So that was really the motivation to try to look for something different. Now what had happened in the previous five years was string theory had introduced a lot of interesting new ideas into the game, thinking about fundamental physics on a microscopic scale — the idea of branes, the idea of extra dimensions. And not just extra dimensions, but an extra dimension of a second type. There’s the Kaluza-Klein type, which is kind of a curled up or rolled up dimension, but there was this interesting M-theory example which is more analogous to a line segment with two special points on either end at which lie branes or what are called orbifolds. And one had even got to the point where one had semi-phenomenological models based on this idea. One could begin to build models of the right kind of particle content, with nicer unification. There’s a nice paper by Horava and Witten in which they tried to unify the strong and electroweak interactions with gravity on a common scale using this idea. I’ve been waiting for string theory to get to a point where there was some kind of loosely speaking phenomenological model you could begin to think about more seriously. Then I asked the question, could you do something interesting with it that would be a different kind of cosmology?

Which research papers on cosmology and supersymmetry motivated your work?

I think I wrote the first one actually. It was right after the Nuffield meeting in 1982, in which the calculations of the perturbations to inflation were first computed and it was discovered good news and bad news. The good news was you got a scale-invariant spectrum, the so-called Harrison-Zeldovich spectrum. The bad news was the amplitude was too big, if you assume the inflaton was the Higgs field of grand unification. So what I immediately said is what you need is a very small coupling. The only was to get such a coupling and preserve it would be supersymmetry. I went through a series of papers with Burt Ovrut who ended up being involved again in this other issue. That’s how we first got together, trying to put together supergravity and inflation. At the time he was at Rockefeller, on his way to Penn. He came to Penn in the mid-80s, around 1985. We were collaborating during that time.

Do you still consider yourself part of the inflationary community?

Sure. I’m a cosmologist. I’m very open-minded about it. I’m just trying to find the best theory or theories to explain the observations. If there’s more than one, see how you can distinguish them, and to try to do it as unprejudiced as much as possible.

Does the rest of the inflationary community welcome your model?

Not terribly. Not as much as they should be. Not as much as I think I would be if I were in the same circumstances. I think it’s just human nature. One likes to be protective of what one knows and likes.

What has been the reaction of the string theory community?

The string theorists’ reaction is a little bit different. Their reaction is they’re trying to see if it fits in tightly with those parts of string theory that are strictly understood. There’s a problem with string theory in that string theory cannot deal with cosmology — it cannot deal with time dependence. They haven’t gotten to that point yet. So there’s a certain skepticism on their part. While they’d be very interested, they’d be reticent to say, ‘oh I think this fits in well with string theory, because quite frankly they can’t say that. All the groups applying the branes and the string ideas have a certain reticence because they cannot say, ‘yes this is consistent with string theory.’ There’s a certain level of discomfort. A lot of interest, but also a certain level of discomfort. You have two fields with a very different character, cosmology, which is putting together a story, and string theory, which is a more formal mathematical subject. It’s a cultural mismatch there.

The public reaction to the accelerating universe has been less [so far] than one would think.

I agree. I think people are really missing the boat on this. This is truly a revolution of Copernican nature; this is not just another addition. What the cosmology community has done for the most part is ‘oops we’ve missing an ingredient, let’s add that ingredient, everything fits beautifully, we have a wonderful model.’ My reaction is: time to step back and re-evaluate. Copernicus taught us that we don’t live in a special place in space. This led to the very important Copernican principle, which guides us. And it was translated into time. And that was part of the shock of Hubble’s discovery that the universe is expanding; it was evolving and not all times are the same. That became a modification, but one that could be accommodated. So you don’t live in a special place in time, but are just part of a steady evolution. There is the biological analogy, which played on people’s minds. Now we discovered that the universe is accelerating, so what does that mean: a) We do live in a special place in time. We’re right near the transition point between deceleration and acceleration. It had to be decelerating when galaxies formed or they couldn’t have formed. So its not true that all points in time are the same. We’re not part of some continuum; we’re really at a genuine transition point and I think that is something that has to have profound meaning for science. So that’s part of the motivation. Certainly my thinking is you can’t just pack it in there, you have to think of something more profound.

When do you think the public will have greater awareness of these ideas?

What is means for people’s psyche, that will take longer. I think the full extent of the implications haven’t been worked out yet. If you were around at the time of Copernicus you might have said, he wants to make the Sun the center, you want to make the Earth the center; it doesn’t mean too much. But then by the time you get to Kepler and Newton it means a lot. So it wasn’t just another detail. It turned out to be a very profound thing. I can imagine this will be a very profound thing by the time it’s through.

What is the relationship between religious and scientific ideas in cosmology?

I find it interesting the degree to which one paradigm reflects ideas that people had before science. It’s interesting that there is any kind of connection. So if you look at various cosmological paradigms: the idea of a created universe, the idea of a cyclic universe, almost every idea that we talk about today you can find. People had that general notion, they just didn’t know what to do with it at the time. So its both good news and bad news. Its interesting that people had the idea before. You also wonder how original you’re being; if your not just taking the old idea and dressing it up with science. I just find it amusing. It doesn’t add or subtract from what I’m doing. When it comes to the science part, I have a very narrow, pragmatic point of view about things. I just want to know if its more powerfully predictive or not as predictive as a competitor? If it is, that’s good; if not, that’s bad. It’s as simple as that. And I don’t have a metaphysical point of view about it.

In your theory, time has no beginning. What are the implications?

I actually think its an advantage to have no beginning of time, because I think its kind of a disturbing idea to go from no time to time. I have colleagues who talk about time being an emergent thing. Maybe time doesn’t have a beginning but it kind of emerges in some way. I don’t know what that idea means so I thought it would be much easier to imagine at least indefinitely far back — I can’t think about infinitely — and not just 15 billion years old.

Would history be repeated, like the concept of eternal return?

Only in a statistical sense. Unfortunately we would not come back again.

How do you distinguish your model from Tolman’s oscillatory model?

Entropy density does not accumulate. In a cyclic model that Tolman was considering the three dimensions expand, but themselves recontract at the crunch. So what entropy you produce on the way out would add to the new entropy. The entropy per unit volume, which is the important thing, would be greater. If the entropy per unit volume is greater that means the next cycle is longer. So what’s happening in ours is that you are producing entropy but our three dimensions don’t recollapse, so it just remains thinned out after the period of expansion. Now when you collide you produce new entropy at very high density. The old stuff is of such low density that its just irrelevant at the end of the next cycle. So you can go right into the next cycle with essentially the same density you’ve had before. The total entropy has increased, but the entropy density goes to negligibly small each cycle.

Image from Diana Battefeld and Patrick Peter, “A critical review of classical bouncing cosmologies,”

What are the dynamics of your model?

There are two different ways of describing this model, one of which is five dimensional and one of which is four dimensional. So I was beginning with the five dimensional description which I find geometrically easier to picture, in which the universe goes through phases, when its dominated by radiation or dominated by matter or positive dark energy. But the branes are stretching and more or less remaining at a fixed distance. And then the second phase, in which the stretching more or less stops and the branes and essentially colliding. In this picture the contraction that’s occurring is only the contraction of the extra dimension. Our three dimensions aren’t contracting.

What happens during contraction?

There’s two branes and they collide together and bounce off one another. That would be what we call the crunch. The extra dimension is collapsed at that point, its momentarily zero. It’s when matter and radiation is created in the collision. All kinds of couplings and masses change. When the branes come apart, you’re seeing the universe just after this event. You see the universe full of matter and radiation, and the matter and radiation, through its gravity, causing the branes to stretch, just after the Big Bang. It wasn’t actually a “Big Bang” it was actually a more modest bang: a collision between orbifolds. You wouldn’t know that; there’s no way of detecting that once the collision has taken place.

What is the 4D picture?

Now the separate picture, which is much more subtle. You could describe it in the four dimensional theory with a scale factor, which is, in fact, going to zero. But even in that case you would see that the density and entropy do not diverge the way you normally think of them as doing. Normally, in the Tolman models, everything comes back together so that the density of matter and radiation, the temperature all go to infinity at that point. Whereas in my brane picture, the density remains small, it remains thinned out.

What do extra dimensions bring to cosmology?

This is in a sense what extra dimensions have brought to things like cosmology. It’s not that there’s nothing you can’t do in the field theory; there’s things that you wouldn’t do in the field theory. It’s giving you a pretext. Why do things look so peculiar in the field theory — it makes sense in the higher dimensional picture. So its one reason you could argue: is it really an extra dimension or is it something that behaves exactly like an extra dimension. You could write the theory in either language to describe most of the properties of the theory.

What happens during the turning point of your model?

As the branes are approaching one another they more or less stop. If you saw that the density was remaining constant you would conclude that the universe has stopped expanding. Now if you went in the field theory, “a” is contracting, yes but beta [another scale parameter] is expanding. So you wouldn’t be able to tell. The important thing is during the expansion phase we see ourselves receding, but then we see ourselves stagnating, staying more or less at the same distance. If we don’t see our density increase, we conclude that we’re not expanding or contracting. We’re in the expanding phase, and we’re in the midst of the accelerating expanding phase, then the branes are far enough apart that there’s a positive energy. When they move close enough together, the potential energy goes from positive to negative, then the stretching will slow down. The matter density would stay the same. There would be no more Hubble red shift, but we wouldn’t see a blue shift either. It’s really stagnation.

In what sense is your theory like the Big Crunch notion?

If you could see the extra dimension, you would see that the extra dimension is shrinking. As this dimension shrinks, that changes the values of all the fundamental couplings. They are all fixed by the distance between the extra dimensions. Field theory would change, forces would change, but it would be a very slow effect until the very end. It’s a very slow process. Once it turns to the decelerating phase you still have roughly another 10 billion years to go. But finally, in the last few seconds you’d see some significant changes in the fundamental constants and that would be the hint that something is about to happen. The way you would view that is the something really strange is happening in the universe. Some tremendous form of energy is building up all of a sudden. It would reach a crescendo, and then, bam, the universe would fill with matter and radiation. That would be the collision. You and I would be vaporized unless we were otherwise protected. Black holes would survive, but most things would be vaporized. So then that universe is full of matter and radiation again.

Could you communicate with later periods?

You might be able to send messages, although whom you’d be able to communicate with would be pretty rare. A black hole survives, so you could make an arrangement of black holes that spell out ‘Hello.’ The problem is that during this period of accelerated expansion, the universe has expanded exponentially. So the only people able to read that message are people right near where you were. That’s a very small fraction of the total population of the universe. We can only see 15 billion light years today, so the chances of communicating with civilizations spread out maybe once every hundreds of trillions is negligible. So you can only send messages to your local neighborhood at best.

How would the CMB fluctuations be produced? Any other implications?

A fluctuation on the brane imprints itself as a fluctuation in the temperature. There are the same predictions for the density, but very different predictions for the gravitational waves.

Where did you grow up originally?

I was an Air Force brat, so that means I moved every three years. But beginning in fourth grade I grew up in Miami.

What were some of your early interests?

Astronomy was my first interest, then I dropped it for many years. Many of the first books I read were on astronomy, that was really fascinating to me. But then I got interested in other things. I always liked science in general, as far back as I can remember. I had a telescope, a chemistry set, a biology lab and did physics experiments. Anything that was scientific I was interested in. Doing astronomy in Miami was difficult, because you either had to go where the lights were or where the mosquitos were. I remember going out to the Everglades, literally running out of the car, setting up the telescope, running back to the car, putting all kinds of stuff on you trying to fight the mosquitoes.

Paul Halpern is the author of fifteen popular science books, including The Quantum Labyrinth: How Richard Feynman and John Wheeler Revolutionized Time and Reality.

Physicist and science writer. Author of Synchronicity: The Epic Quest to Understand the Quantum Nature of Cause and Effect