Lee Smolin thinks that time is real. If that strikes you as unusual, you haven’t spent much time with theoretical physicists, who tend to think that the passing of time is either an emergent property of the universe, or, perhaps, an illusion.
“Some of my colleagues suggest that time is an approximate description of the universe,” Smolin, a theoretical physicist at the Perimeter Institute, writes in Time Reborn. “A description that is useful on large scales but dissolves when we look too closely. Temperature is like this.” The reason that some physicists have rejected time, he argues, is that they have mistaken mathematical models for reality. Smolin says that, while superstring theory has been around since the eighties, it has no experimental support—or hope of it—in the near future, because it not only describes our world, but an infinite landscape of possible worlds, without a guiding selection principle. The further scientists venture into it, the more complicated the math becomes, and the farther from our observable world it seems to be.
Smolin’s forthcoming book with philosopher Roberto Mangabiera Unger, a professor at Harvard Law School, The Singular Universe and the Reality of Time, presents their views on the importance of philosophy for physicists, why falsifiability is important for science, and why we should care about whether time is real—all of which we talked about the other day.
Reading your work, I was struck by how much more it discussed the history of thought—both science and philosophy—than I expected.
I’m a physicist, but I’ve always been interested and inspired by philosophy. I wrote about this in the Trouble with Physics: When we look at the really hard foundational issues in physics and cosmology, the philosophers can’t solve the problems for us. But knowing the history of thought is something you can learn from; scientists who think about fundamental problems are often people who are knowledgeable in the history of philosophy, at least with the problem they’re working on. But different styles dominate different times. During revolutionary periods, it becomes the dominant style to quote philosophers and historians.
Like Albert Einstein hanging out with the logical positivists of the Vienna Circle. You quote Rudolph Carnap, an important member of the circle, in Time Reborn about how uncomfortable Einstein was with the meaning of “the Now.”
I didn’t know Einstein was involved with the Vienna circle apart from his conversations with Carnap. Though he was very influenced by Mach, an early positivist. Many of that generation [of physicists] were very philosophically well-educated, and it wasn’t unusual for a scientific paper or talk to begin with mentioning Kant or Leibniz. And then my sense is that faded after World War II, when the European culture associated with those people was destroyed.
Since then, physics has been dominated by the American style, which is more pragmatic. I think Freeman Dyson described it best: We normally think that the established scientists are conservative and revolutionaries are young. But here, the revolutionaries were old, and the young generation were conservative and just wanted to apply quantum physics to understand more physics. They were frustrated by hearing so much about philosophy from the older generation of revolutionaries. And then in the late seventies, that pragmatic style had broken down.
Right, scientists have declared philosophy dead in a number of recent books. Stephen Hawking’s The Grand Design, co-written with Leonard Mlodinow, was especially funny, since the opening basically says there’s no need for metaphysics—but then the rest of the book is metaphysics.
I thought that was so embarrassing. Or Lawrence Krauss—he is a brilliant and articulate and highly effective advocate for science—and I am usually on his side, but in this one case he got in over his head and claimed that inflationary cosmology solved the problem of why there is something rather than nothing. David Albert did a review of Lawrence’s book, and I remember thinking, “This is the most cutting and acerbic review of anything I’ve ever read.” But he was right. He took Lawrence to task for being ignorant of what philosophers were trying to understand.
I did my undergrad degree at Hampshire, and it was partly physics and partly philosophy. We called it natural philosophy. I went to graduate school in physics intending to do a degree in philosophy but was very turned off by the philosophers I met. Meanwhile, the philosophers of physics of my generation, people like David Albert, Simon Saunders and Harvey Brown, are much more knowledgeable about the technical details of physics. They know physics very well, and they are very sophisticated about contemporary physics.
What you’re saying in the book with Unger is that many of the rules we’re using to create a cosmology don’t work well at that scale, right?
All of physics, except the attempt to make a theory of the whole universe as a closed system, can be understood from an operational perspective: Whether you describe it as operational or effective, basically, there’s a theory which works within specified boundaries and specified limits as to scales of distance and energy. Within those limitations, it has a methodology which generates falsifiable hypotheses.
I give a bunch of examples in the book, but one way to argue is that the application of a theory through an experiment is always approximate. You’ve got the interactions that are too weak to influence things; interactions that are too weak to measure; interactions between the system itself; and what is left outside of the boundaries of the system we are studying.
You might think that to describe the universe as a whole, you just describe part of it, but you can’t do that. One of the problems is that once you describe the whole universe, you have to account for not just its survival, but also, why these laws. That’s what you need to think about when you’re trying to make a full theory.
I’ve been thinking about this for a long time. You want to account for the choice of the laws, which have empirical consequences and are testable. Something has to happen. There has to be something which allows laws to change that is empirically testable, and that’s only possible if the change is something real that happened in our past.
When did you start thinking seriously about time?
Oh, I’ve always thought about time! It’s at the core of the physics of quantum gravity, which is the main thing I’ve worked on all my life. But for these ideas, there were three parallel movements that came together.
First, the laws of nature have to evolve to be explained. I started thinking about this in the late nineteen eighties, because of string theory. Instead of a unique solution, it was very clear there would be a landscape of string theories—a vast proliferation of different versions of the theory. So I proposed a model by which the laws of nature could evolve through the history of the universe, analogous to natural selection. It’s called “cosmological natural selection,” and I first published on it in 1992. That was the first step, and I’ve worked on it on and off since then. The implications took a long time to sink in.
I didn’t see until I started to talk to philosophers that there was a contradiction between believing time was emergent, and believing laws in nature evolved in time. I had an interaction with Roberto Unger and this made me realize there was a contradiction in my work—believing that laws evolve in time, on the one hand, and believing that time emerges from timeless law, on the other.
The other thing happening about that time was that some of my friends and colleagues working in quantum gravity began to treat time as a fundamental quality, a fundamental part of the picture. I had to take seriously what they were doing, even though it was heterodox from a mainstream point of view. And there were some internal technical issues to making time emerging work, and those technical issues began to bother me.
So with those things, my view began to shift toward time being real.
String theory has been the dominant force in physics for a long time, but this isn’t optimal to you, right?
People should be examining hypotheses which are testable, and abandoning hypotheses for things that aren’t testable, instead of chasing ideas that they love that aren’t testable. Anybody who looks at the recent history of physics has to see [the lack of progress] as a crisis of explanation.
The results from the Large Hadron Collider are extremely disappointing to people working on particle physics. Many thousands of people fear that forty years of work are lost—their whole careers they’ve spent on very intricate and elaborate models with motivating ideas, they have now to reconsider in the light of the evidence. The question is how people think when what they thought was contradicted.
One of the issues is that the standard model is extremely fine-tuned on 29 parameters to take either very large or very small values. Once you accept the idea that laws are fine-tuned—and that this requires explanation—you have to believe there must be some dynamical process acting in history that did the fine tuning. Then you’re looking at something like cosmological natural selection.
If you were given an ideal experimental set-up to test your theory, what would you look for? What’s the best way to see if cosmological naturalism works?
There aren’t many predictions, but there are a few. One is holding up remarkably: When I first published [in 1992], I predicted no neutron star should have a mass of more than 1.6 times the sun. That wasn’t quite right; the prediction was based on a theoretical calculation done by nuclear astrophysicists, and when that was done more carefully the upper limit predicted ended up being two solar masses.
So if a more massive neutron star is found, our universe’s natural laws don’t produce the maximal amount of black holes and cosmological natural selection is falsified. Presently the heaviest neutron star whose mass is precisely measured is just within the prediction with 1.97 solar masses, but there are other neutron stars whose masses are less accurately measured, but whose central values are above two solar masses. There could easily be a new measurement that would precisely measure a neutron star mass of more than two solar masses and this would contradict the theory, so the theory is quite vulnerable to falsification, which means it is real science.
This interview has been edited and condensed for clarity.