Notice: We are in the process of migrating Oral History Interview metadata to this new version of our website.
During this migration, the following fields associated with interviews may be incomplete: Institutions, Additional Persons, and Subjects. Our Browse Subjects feature is also affected by this migration.
Please contact [email protected] with any feedback.
This transcript may not be quoted, reproduced or redistributed in whole or in part by any means except with the written permission of the American Institute of Physics.
This transcript is based on a tape-recorded interview deposited at the Center for History of Physics of the American Institute of Physics. The AIP's interviews have generally been transcribed from tape, edited by the interviewer for clarity, and then further edited by the interviewee. If this interview is important to you, you should consult earlier versions of the transcript or listen to the original tape. For many interviews, the AIP retains substantial files with further information about the interviewee and the interview itself. Please contact us for information about accessing these materials.
Please bear in mind that: 1) This material is a transcript of the spoken word rather than a literary product; 2) An interview must be read with the awareness that different people's memories about an event will often differ, and that memories can change with time for many reasons including subsequent experiences, interactions with others, and one's feelings about an event. Disclaimer: This transcript was scanned from a typescript, introducing occasional spelling errors. The original typescript is available.
In footnotes or endnotes please cite AIP interviews like this:
Interview of Steven Weinberg by David Zierler on August 20, 2020,
Niels Bohr Library & Archives, American Institute of Physics,
College Park, MD USA,
For multiple citations, "AIP" is the preferred abbreviation for the location.
Interview with Steven Weinberg, Jack S. Josey-Welch Foundation Chair in Science and Regental Professor at the University of Texas at Austin. The focus of the interview is on how and when Weinberg became interested in cosmology, and how he defines it as a distinct discipline from astronomy and astrophysics. Weinberg explains that between the intensity of interest in particle physics in the 1950s and the speculative nature of cosmology, he had neither the interest nor the outlet to pursue cosmology in a rigorous way. He discusses some of the theoretical and experimental limitations at the time that kept cosmology in a largely “mystical” realm, and why the discovery of the microwave background by Penzias and Wilson “changed everything.” Weinberg explains what new questions can be considered as a result of evidence for a hot early universe, and he discusses when he first became interested in the formation of galaxies. He describes why the cosmological constant has bothered him for a long time, and he traces this problem back to Einstein and what Weinberg considers Einstein’s incorrect approach to his own theory. Contrasting his own experience as a graduate student, he cites John Preskill as his first student to pursue cosmology, and he explains that while his interests in particle physics and cosmology are generally separate, he always looks for intersecting research opportunities, which is well represented in the relevance of beta decay physics in the first three minutes of the universe. At the end of the interview, Weinberg surveys the value and problems associated with the term “Big Bang,” and he reflects on his career-long effort not to be dogmatic in his views on cosmology.
Okay, this is David Zierler, Oral Historian with the American Institute of Physics. It is August 20th, 2020. It is my great pleasure and honor to be here with Professor Steven Weinberg. Steve, thank you so much for joining me today.
Glad to do it.
Okay, so for our talk today, we are going to keep it on the topic of your developing interest in cosmology. And so, for my first question, I just want to get some terms on the table so that we know exactly what we’re talking about. And to historicize those terms over the course of your career. So, I want to know first where you saw, at the earlier stages of your career, distinctions between astronomy, astrophysics, and cosmology? And if those distinctions have changed over time?
Well, cosmology always seemed to me quite separate. It deals with the whole universe. Individual stars and so on come into it as pieces of evidence for something that’s happened in the whole universe. For example, the abundances of different elements in the stars tell you something about the history of the universe. But the interest is in learning about the universe, not in learning about the stars.
Actually, I never had to make this distinction very clearly in my mind until recently. My latest technical book was on astrophysics. I had already written a book about cosmology, and I didn’t want to be repeating myself. So, I had to draw the line fairly carefully. In the new book I wrote a lot about stars and somewhat less about galaxies, but I tried to avoid getting into cosmological issues, which I had already written about. Cosmology has a special interest for particle physicists like myself because we can’t get up to very high energy in our laboratories. By very high I mean at the scale where the interactions become unified and gravity becomes as strong as other interactions. More or less the Planck scale. We can’t dream of how to get up there in our laboratories, but--
And Steve, you mean by not even dreaming… Like, even beyond the SSC?
Oh, yeah, if you’re talking about 1015 or 1019 GeV. I’ve made the remark that if the whole human race would put its resources at physicists’ disposal and we could do anything we wanted with the money, we wouldn’t know how to build an accelerator that could allow us to do experiments at 1015 or 1019 GeV. On the other hand, the universe presumably once was at a temperature where typical energies were of that order of magnitude. So, we can learn something, we hope, from cosmology about really fundamental laws of physics that we can’t get at experimentally, hardly any other way.
There are other things we can do, which are not unrelated to cosmology. There are some very subtle effects that reflect physics at very high energies. One of them is neutrino masses, which have been discovered. Another is proton decay which has not been discovered. That’s another handle we have. Both of them, of course, are related to cosmology—to the issue of why there is an excess of matter over antimatter in the universe. And it is a question I keep coming back to. Well, that was a very long answer to your question.
So, part of that question were these distinctions between astronomy, astrophysics, and cosmology. Over the course of your career, do you see these distinctions as being more or less stable, or are these moving targets over the decades?
Well, I’d say they’re pretty stable. Just like the distinction between nuclear physics, which is of direct concern in the study of stars, and elementary particle physics. It hasn’t really changed that much over the years. There’s certain fuzziness in the boundary. But I don’t think it’s changed very much.
And the question of exactly when you developed an interest in cosmology-- My question behind that question is when you were first forming your identity as a physicist, even in high school or undergraduate, was cosmology a field that was open to you to study? Just intellectually, technologically? Was that a field of research that was essentially available to pursue, or was particle physics really the entrée to cosmology?
I would say that until the mid ‘60s, cosmology had a flavor of something very speculative. Almost theological. Almost metaphysical. I read things about it. Popular books like James Jeans’ The Mysterious Universe. I read that pretty early, actually, in junior high school. Later, Öpik’s book The Oscillating Universe. I’m not sure that’s the title. I also read things about astrophysics, like George Gamow’s wonderful book called The Birth & Death of the Sun. These are all popular books. I didn’t know enough physics or mathematics to read any more deeply than that. But of course, I was interested. Who isn’t interested in cosmology, you know? We are in this big universe and it’s billions of years old and it contains countless galaxies. Who wouldn’t be curious about it?
So, Steve, my question there is, as a freshman at Cornell, if you were to say to your professors, you know, in 1950,1951, “Professor so and so, I’m interested in cosmology. I’d like to study cosmology.” Not in--
I wouldn’t have dreamed of saying that.
Because at that time, particle physics was red hot.
There were lots of new results coming out of laboratories. Cosmic rays had dominated elementary particle physics until the late 1940s. Now you no longer had to go up on mountains and wait for cosmic rays to do something interesting in your detector. I remember Arthur Wightman, when I took his course as a graduate student in Princeton, he almost shocked the class by saying, “The day of cosmic rays is over.” Not that they would stop hitting the earth, but particle physicists were now going to be using accelerators.
So, when I was an undergraduate at Cornell, particle physics was very interesting, exciting. Lots of data was coming out. Lots of theoretical ideas that were relevant to the data. A lot of data that was relevant to theory. Cosmology seemed fairly frozen. I knew about the Hubble program. That is, the measurement of redshift and distance of distant galaxies. I knew that that was going to tell us how the universe is expanding.
I was very excited when I saw that all that would be cleared up when the telescope at Palomar Mountain started operating—the Hale Telescope. And of course, it wasn’t. [laugh] The Hubble program went on and on. But it gradually became clearer that the universe really was undergoing a uniform expansion. And gradually the measured value of the expansion rate, the Hubble constant, settled down more or less. Back in the 1930s, Humason, Mayall and Sandage had given a value, amazingly, of 526 kilometers per second per megaparsec without bothering to estimate uncertainties. And of course, that was way off. By the time I started reading about cosmology, it had settled down to somewhere in the neighborhood of 50 to 100 kilometers per second per megaparsec. But progress was going very slowly. Cosmology was not exciting in the 1950s.
Steve, would you have been able to identify a professor on the faculty either at Cornell or Princeton who would say, “I study cosmology”? Was that even available?
That’s a good question. I’m trying to think. At Cornell… now certainly astrophysics.
Bethe and Salpeter had made huge contributions to astrophysics.
But that gets me back to my question about these delineations between astrophysics and cosmology.
Right, right. Yes. Astrophysics, meaning the structure and evolution of stars was a rich, well-respected field. It was, in fact, in a mature state. It was no longer something that was producing new insights every year. Well, as I said, Bethe and Salpeter would’ve been ideal people to go to for astrophysics. No, I don’t recall anyone at Cornell teaching or being interested in cosmology. Then at Princeton, there were people who were deeply involved in general relativity. I took a course from Valentine Bargmann, who had worked with Einstein. Well, that was a course in general relativity. I also took a course in advanced quantum mechanics with John Wheeler, who of course was very much involved with black holes and things.
But I don’t think… well, certainly, the course I took with Bargmann didn’t go into cosmology. And I don’t know of anyone who I would’ve gone to. Actually, the person who would’ve been best to go to, no one would have suggested. And I had never thought of it. And that was Bob Dicke. Bob Dicke was this great experimental physicist who had developed some of the tools that were used in the discovery of the cosmic microwave background in the 1960s. Dicke had, in fact, put an upper bound on the temperature of the cosmic microwave radiation background. I don’t remember exactly when… in the ‘50s sometime.
Has that measurement held up?
Yeah, but this was a high upper bound. [laugh] The actual value was well under that upper bound. So, yes, it held up. It’s described in my book, The First Three Minutes. Dicke-- There was a peculiar information gap between the experimentalists and theorists. And I didn’t know anything about it. I didn’t know anything about a cosmic microwave background when I was a graduate student. To me, cosmology still meant mapping out the large-scale structure of the universe. Its expansion and curvature. I hardly even knew that there was an early universe. There must have been, but I never thought of it. And neither did most people then. No, I don’t think there’s anyone I could’ve gone to. I never thought of going to anyone to study cosmology when I was a graduate student.
And so, Steve--
Particle physics was so exciting that I didn’t miss it.
Right. But that’s two separate issues. No matter how exciting particle physics would’ve been, that still doesn’t answer the relative nonexistence of cosmology. It could still be there regardless of how interesting particle physics was.
I think a course on general relativity might have included the derivation of the Roberston-Walker metric for the large-scale expansion of the universe. I don’t think it did. The books on general relativity, as I remember them from that period, like Peter Bergmann’s book, didn’t get into that. There was a separate literature. There was a book by Milne on the large-scale structure of space and time. Very mathematical. Very unphysical. The kind of book that fit in with the Cambridge style in mathematics. But people who wrote books on general relativity didn’t cover cosmology, generally speaking.
Steve, to foreshadow to the mid-1960s, this does raise the question, looking into the 1950s, what were both the theoretical and experimental limitations that made cosmology, as you said earlier, essentially a mystical enterprise that really was not yet ready to be pursued?
Oh. Well. That’s an interesting question. I’m not sure. There was a kind of dogmatic flavor to a lot of cosmological theorizing in those days. I remember Fred Hoyle with some colleagues had what I thought was a very attractive theory (now I’m going back to the early ‘60s or late ‘50s. I’m not sure): the continuous creation theory. I thought that was lovely because if it really is true that the universe is always producing new matter to fill up the gaps as it expands, then there was at least the hope that you could understand what you see in the universe as a mathematical consequence of the stability condition—that the universe would always look the same. And just like you don’t try to find out the energy levels of the hydrogen atom by thinking of how the hydrogen atom is formed, you find them by asking what kind of states are stable, have wave functions that only oscillate in time, and don’t change in any other way.
The continuous creation theory seemed to me very attractive if it were true. But then there was evidence that it wasn’t true. It didn’t come from the microwave background at first, it came from counts of radio sources. Because if the universe isn’t evolving on the average, so that the probability distribution of radio sources with given strengths doesn’t change with time, then you can calculate the number density of radio sources as a function of source strength. And it goes like some kind of power, I think strength to the minus three halves power or something. I’m not sure. It’s a very specific, definite prediction. And the data from the 3C survey of radio sources didn’t bear it out. I remember, Hoyle gave a talk in Boston advocating the steady-state theory. I raised the question, what about these radio source counts? (It wasn’t my idea that they rule out the steady-state; a lot of people were saying that.) Hoyle wouldn’t really confront the question.
I once got on a bus at some conference with Hannes Alfvén, the notable Swedish physicist who had introduced the idea of Alfvén waves in magnetohydrodynamics. This was later, after the discovery of the microwave background. I asked him if some kind of magnetohydrodynamic phenomenon would play any role in the early universe. He said, as well as I can remember, “Is your question posed in the context of the Big Bang theory?” And I said, “Yes. That’s what I was thinking of.” And he said, “Well, then I can’t answer it.” He had his own theory which had to do with matter-antimatter annihilation creating the microwave background—in fact creating everything. He thought that there were islands of antimatter out there in space, I think. I hope I’m not doing an injustice to him. So, there was a dogmatic flavor to cosmology.
In the late ‘50s and very early ‘60s, before the discovery of the microwave background, I gave myself the assignment of learning something about cosmology. The way I learn is to think of a research problem that forces me into the literature, because just reading the literature without a goal in mind is much more difficult. So, I gave myself the research problem: What happens to the neutrinos in the universe? I had read in Bondi’s book about cosmology about Olbers’ paradox. According to the most naïve picture of the universe, that it’s static and eternal, every line of sight would end on the surface of some star. You may have to go out very far, but eventually if the universe is static and infinite, every line of sight hits a star. That means that the whole sky should glow with a temperature of thousands of degrees, like the surface of a star.
Well the answer to that is, of course, that the universe is expanding, and in its present phase, it has only a finite age. So, there really isn’t an Olbers’ paradox. Or if you like, this is the solution. The expanding universe provides a solution to Olbers’ paradox. And it occurred to me, part of that solution is that the expansion means that the light from very distant sources is redshifted. You don’t see very distant stars with an apparent surface temperature of 3,000 degrees Kelvin. If they’re far enough away, they’re moving away so fast that the light is redshifted and they look much cooler. But then it occurred to me that the redshift lowers the energy of neutrinos, but not their numbers. The lepton number is conserved, or at least we thought so then.
So, there’s a question for neutrinos which doesn’t arise for photons. What happens to the neutrinos from the past? That was a good research problem. I wrote a couple of papers about it, and actually suggested an experiment to look for a Fermi sea of very low energy neutrinos. Which of course, isn’t there. But in that work, it never occurred to me that the very early universe would have been hot. [laugh] I mean, of course, it would’ve been hot. In those papers, I was thinking of a cold, early universe. And so, the Fermi sea that I was talking about was a Fermi distribution with a nonzero chemical potential, but zero temperature, like you have for electrons in a white dwarf star. It was dumb.
At that time my wife and I were going around the world. We were living in San Francisco while I was teaching at Berkeley. I was going to spend a year at Imperial College in London. I had the bright idea of going around to London the wrong way around, going west, and checking the theory that the world is round. In Tokyo, Kyoto, Hong Kong, and Bombay, I kept giving talks about this work on cosmic neutrinos. So, [laugh] it got me around the world. Although, some of it was based on a misunderstanding about the early universe. Everything changed with the discovery of the microwave background by Penzias and Wilson.
And what year was that?
That was ‘65. It should have been discovered long before. Dicke could’ve discovered it if he had known to look for it. Gamow’s collaborators, Alpher and Herman, had actually predicted a radiation temperature that redshifted to the present would be about 5 degrees Kelvin. Yet, they never succeeded or perhaps even tried terribly hard to get radio astronomers to look for it. Dicke could have found it. In the end, finally—largely through the theoretical work of Jim Peebles—a group at Princeton, Dicke, Peebles, Roll and Wilkinson, actually did set out to look for the microwave background. They were scooped by Penzias and Wilson who discovered it more or less by accident. But that changed everything. Because now we had not only evidence that there was a hot early universe, but we had details that could be measured experimentally that would tell us about the early universe.
How did this discovery confirm that the early universe was hot?
Well, it was a thermal distribution. And furthermore, it was in the range of temperatures that was expected from the helium abundance. In the first three minutes, neutrons gradually turn into protons. If the universe was cold—that is, a relatively low ratio of photons to baryons—nucleosynthesis would have started early, when there were nearly as many neutrons as protons, and most of the matter of the universe would have become complex nuclei, especially helium. On the other hand, if the universe were very hot, nucleosynthesis would be delayed until virtually all neutrons would have decayed into protons. The stars would start as pure hydrogen. In fact, the stars started their lives as mostly hydrogen, but with about 25% helium. And that dictates a range of values for the ratio of photons to baryons. That ratio does not change much as the universe has expanded since then, so if you know the baryon density now you can figure out the photon number density and its temperature now.
This is a calculation that was done by Alpher and Herman. I’m not sure, I think around 1950. Maybe even as early as 1948. They didn’t do it exactly right. They assumed that the universe started just with neutrons. In fact, as a Japanese theorist, Hayashi, later pointed out it would have been in thermal equilibrium with equal numbers of neutrons and protons at very high temperature. But that doesn’t make that much difference, actually. They had a pretty well-founded prediction of five degrees Kelvin for the microwave background. Which Dicke could easily have found. Dicke and Roll and Wilkinson. And yet, they didn’t. But, as I was saying, the temperature came out right. It wasn’t five degrees; it was 2.7 degrees—good enough for government work. The evidence got increasingly strong that it was a Planck black body distribution. So, we were really looking back to a time when the universe is in thermal equilibrium. Of course, the cosmic temperature is anything you like if you scale back to an early enough time. But at end of the first three minutes, the temperature was in the neighborhood of 109 degrees Kelvin. And that has become more precise from measurements of the deuterium abundance.
What was so significant--
Well, you see, if I can interrupt, this is real physics. This is no longer cosmological dogmas or metaphysics; it is taking data and making plausible calculations and getting results. It was wonderful. I mean, this really turned me on to cosmology. I was teaching a course in general relativity at Berkeley and then at MIT, and I was impelled to include this stuff in my course. Well, this is something that Peebles later called physical cosmology, as opposed to the old cosmology based on general relativity, which was just geometry. I brought this into the course first in in the late ‘60s. My lecture notes got more and more detailed, and finally turned into a book.
As a result of this discovery in 1965, what immediate new questions were raised for you?
Hmm. What immediate new questions? Um.
In other words, we have a hot, early universe. This as you say, really makes cosmology a significant field. Sort of, right out of the box. And of course, as a result, this raises new questions that might not have been able to have been asked prior. So, what might those questions have been?
I wish I had been smart enough to ask the right questions. I did eventually, like many of my colleagues, ask questions about the origin of the tiny excess of matter over antimatter. If you go back to a time when the temperature in energy units was more than a GeV, you would’ve had huge numbers of nucleon-antinucleon or quark-antiquark pairs. But you’d have about one excess quark per 109 pairs. And we know that because the number of pairs would have been comparable then to the number of photons.
In thermal equilibrium, as long as the temperature’s high enough, the number of anything is about the same. The number density of photons or quark-antiquark pairs goes as temperature cubed. And when all the quarks and antiquarks annihilated, you would still have that little bit of excess left over. From then on you would have a constant number of photons and baryons in every comoving volume as the volume expands. So, the ratio of the excess of quarks over pairs in the early universe would have been roughly the same as the ratio of baryons to photons that we observe now. That’s about 10-9. There are about 10-9 baryons in the universe today, and no antibaryons to speak of, for every photon. Well, if I had been smart, I would have started thinking right away about where that number, 10-9 comes from. I didn’t. I did later, at a time when other people were thinking about it too. That was certainly an outstanding question I should’ve asked.
I was also interested in the formation of galaxies. I wrote a paper (I think this was the very early ‘70s) about the formation of clumps of matter, clumps that would eventually become galaxies. This happens not in the very early universe when deuterium is formed, or even earlier when inflation occurs, but certainly many billions of years ago, when the universe was maybe, oh, what should I say? A hundred times smaller than it is now. For some reason I had been studying thermodynamics and hydrodynamics. It may have had to do with defense consulting. I had learned about a phenomenon called bulk viscosity. Most people know of viscosity in terms of shear viscosity, which is the force that is exerted when you have a shear flow of fluid that is flowing with a speed that varies from one layer to another. There’s also a bulk viscosity that occurs when fluids just expand or contract homogenously, with no shear. I thought bulk viscosity might play a role in the formation of galaxies. But I knew very little about it.
The formation of galaxies is a huge subject. There are people who spent their whole working lives on it. I recently learned that we still don’t have a really good theory of where the spiral arms come from. But I was interested in the issue of galaxy formation and hoping to use the little bit I knew about hydrodynamics to have something useful to say about it. I don’t remember the work very clearly. The fact that there was a hot, early universe didn’t rule out the possibility of an oscillating universe, the possibility that the universe goes through a sequence of expansions and contractions. In fact, that idea is still alive. It occurred to me that as the universe expands and contracts, even if it does so in a spherically symmetric way, bulk viscosity would keep increasing the amount of entropy per baryon in the universe. I was assuming baryon number was conserved, so that as you increase the entropy, you would increase the entropy per baryon. I think I wrote about that at the time. None of these led to much.
Steve, could you have written The First Three Minutes any earlier than you did?
Well, before I wrote The First Three Minutes, which was in ‘77, I wrote my treatise, Gravitation and Cosmology. It was the fruit of my teaching the subject, bringing cosmology into a course in general relativity, not just the Robertson-Walker metric, but physical cosmology, nuclear synthesis and the microwave background, and so on. I had been teaching that at MIT on and off. The lecture notes gradually got better and better. Eventually it became a book. That was the first book of any kind that I wrote. It was published by Wiley. It got tremendous reviews. That was very important in my life because before that I had been spending my summers doing defense work with the JASON Group. It was somewhat glamourous work because I was learning secrets, things that my neighbors didn’t know. It was also interesting. I was learning things that I never learned as a graduate student, like hydrodynamics, which I didn’t know until I learned it in connection with defense work. And I needed the money. Looking back, it wasn’t much, but in those days it seemed like a lot of money.
Steve, I guess my question is because the first--
Let me finish my thought.
So, when the book came out, I realized I was gonna be writing books rather than going off in the summer to do defense work. I could have written The First Three Minutes immediately after that, in the early ‘70s, but it didn’t occur to me to do it. I had never written a book for the general public. The trigger for The First Three Minutes was accidental. In 1976 Harvard inaugurated its undergraduate science center, and I was asked to give a talk. So, I gave a talk about the early universe based on my treatise. I gave a sort of cinematic description of how the early universe expanded. First frame, second frame, and so on. A friend of mine, the sociologist Dan Bell, was in the audience. He was taken by a remark I made, that “after the first three minutes nothing of interest would ever happen in the universe again.” [laugh] It was a joke. It was an exaggeration. There was a sense in which it was true. The chemical abundances were fixed with which the stars start their lives. At the end of the first three minutes.
That was obviously a joke. But Dan thought that was great fun. So, he convinced a publisher, Erwin Glikes of Basic Books, to propose that I write a popular book about it. And I wrote it during the year of ‘76 to ‘77. I could’ve done it two or three years earlier. It was just that I got instigated by Bell and Glikes. When The First Three Minutes came out it made a success. It was translated into a lot of languages, got very good reviews. I asked Glikes, “How did you know that this would work out so well?” And he said, “Well, with your title, The First Three Minutes, and my ability to sell foreign rights, I knew we would do okay.”
[laugh] Steve, I guess my question about if you could’ve written it earlier-- as a book intended for a public audience, right? There’s a certain duty, I suppose, you have that what you’re presenting is, you know, settled science. What we really understand. Whereas, in academic papers when you’re throwing out ideas at conferences, there are disagreements and things like that. So, I guess my question is, was there anything that was provisional in your thinking prior to The First Three Minutes that you only felt by 1976, 1977, that you were truly confident that, you know, you can get this word out to a broad audience? This is how things really worked?
No, I don’t think-- Obviously, a lot had happened from ‘65 to ‘72. The properties of the microwave background had gotten pretty well nailed down. That gave me confidence to write the cosmological parts of my treatise in ‘72. From ‘72 to ‘77, no, I can’t recall that anything changed very much. In my treatise, I did have a final chapter about speculative ideas. Every chapter in that book started with an epigraph, a quote from someone who had said something that I could interpret as relevant to the chapter. And in that chapter I had a quote from the pre-Socratic philosopher Xenophanes who says something… I can’t quote it exactly. Well, in fact, it’s only a translation anyway. “Of certain truth, no man can know. And of the gods and the other things I write about. Even if we said the truth, we would not know it.” Wow. I wish I remembered it more precisely. So, Chapter 16 in my treatise that had to do with speculative ideas. But they were still speculative in 1977. I could’ve written The First Three Minutes in 1973. It wouldn’t have made any difference.
Now those ideas that were speculative through 1977… just fast forwarding to today. What remains speculative? And what has been settled in the intervening years?
You know, I don’t really remember what I regarded as speculative. I think I was writing about the oscillating universe which very few people take seriously now. But some people do. So, it’s still an open possibility. You know, we really don’t have a handle on t=0. If we date things from a nominal singularity, the singularity in our equations, we’re very confident in what happened in the first three minutes going back to maybe the first second. Earlier than that, there’s a very well supported idea which we’re not 100% certain of, the idea of inflation, due originally to Alan Guth, which didn’t exist in ‘77. That idea came along in the ‘80s if I remember correctly. I think it’s probably right, but I can’t say it’s been nailed down. That has to do with physics at a much higher temperature, much earlier time. But even if you took inflation as something certain, something you could take to the bank, we still wouldn’t know the conditions before that. We wouldn’t know whether there was a singularity earlier than the time of inflation. Or whether there was an earlier period of contraction. The big uncertainties are still there.
Now we have other uncertainties. I’ve been bothered by the cosmological constant for a long time, wondering why it wasn’t huge. After I came to Austin, I wrote about it, and I gave a series of lectures at Harvard called the Cosmological Constant Problem, in which I surveyed a number of suggestions. I wasn’t the only one who was worried about this issue. Why isn’t the cosmological constant-- Or to put it another way, the energy density of empty space, why isn’t that enormous compared to the empirical upper limit set by the rate of expansion of the universe? We knew it was many, many, many orders of magnitude less than you would guess from a back of the envelope calculation. That was a great mystery. In my lectures at Harvard, I went through all the proposals that had been made that I knew about. I had negative things to say about them all, except for the possibility of an anthropic explanation. That is that we’re in only one sub-universe of a multiverse and in most of them the cosmological constant is very large. And it’s only in the ones where it’s small that life could arise because it’s only in those that galaxies and stars could form.
So, I wrote about that. With two collaborators, Hugo Martel and Paul Shapiro, we wrote a paper called Likely Values of the Cosmological Constant. We said that it’s probably a good fraction of the total energy density of the universe now. And then that was confirmed with the discovery of the accelerated expansion of the universe by two big groups who were justly honored for the discovery. They showed that the vacuum energy density is, in fact, something like three quarters of the total energy density of the universe, which is in the ballpark that Martel and Shapiro and I had anticipated. But this kind of anthropic argument is never precise enough to make a numerical prediction so well defined that you can actually be confident that you’ve got the right reasoning. Anyhow, that was a preoccupation of mine in the ‘80s, but not in the ‘70s. I don’t recall. Maybe it was. I don’t remember when I first started worrying about the cosmological constant.
I always thought Einstein was wrong in his approach to his own theory. I think Einstein took the view that you should make the theory as simple as possible and only put into the equations things that actually have to be there. He didn’t want to put in a cosmological constant except that he had this idea that he wanted to have a static universe where the cosmological constant would balance the gravitational attraction of matter. When he realized the universe wasn’t static, he said it was a great mistake to have introduced the cosmological constant. And he also seemed to have been happy with the idea that his field equations were second order partial differential equations. Why not higher terms? Why not terms involving more derivatives or more powers of the curvature?
I guess Einstein would have said, “Well, there’s no need for them. We’re supposed to make simple theories.” I don’t buy that. I think simplicity, like everything else, has to be explained. I think there is a good explanation for why general relativity with only the terms in the field equations that Einstein included worked so well. It’s because the other terms at the scales that we’re capable of studying would be highly suppressed. So, we don’t see them. The cosmological constant doesn’t fit that. The kind of order of magnitude guess you would make for the cosmological constant, as I said before, is many, many, perhaps 120 orders of magnitude, larger than allowed by observation. I think Einstein, instead of being sorry that he had introduced the cosmological constant—he should have worried why it wasn’t there.
If he had worried, what would that have meant?
I don’t know. Einstein being Einstein, he would have gotten a lot of other people worried a lot earlier. I don’t know that he would’ve been able to solve the problem. We haven’t really solved the problem even now.
The question of the initial promise with inflation from Guth. Was there an initial promise that it would have gotten close to t=0? Was that part of the attraction and the excitement early on?
Well, it got a lot closer. I think Guth and everyone else would realize that the story of inflation starts at a definite time, a time that’s much closer to t=0, but it doesn’t illuminate whether, for example, there was an initial singularity or not. I don’t think that was part of its appeal.
Is t=0 an answerable question, in your view?
I don’t know. You know, it’s like questions about the multiverse or any of the really big questions. They get you far outside your experimental grasp. But that doesn’t mean that they’re hopeless. If you had a theory that worked very well and that as a mathematical consequence, for example, predicted that there was a singularity at a definite time in the past, or that there wasn’t, either way, if the theory worked well enough in other contexts. For example, if it explained the excess of baryons over antibaryons and explained the strength of the initial fluctuations out of which galaxies grew, and explained this, that, and the other thing—if it worked well enough in other contexts, then you would believe it.
For many thousands of years, no one saw the other side of the moon. But the theory that the moon was a sphere worked very well in a lot of contexts. So, it wasn’t a surprise finally, when people managed to fly around the moon and see it really did have another side. It really was a sphere. There’s a kind of naïve positivism that in science, I think, does more harm than good. That says you shouldn’t believe anything unless you can observe it directly. It led some scientists at the beginning of the twentieth century, like Ernst Mach, to doubt the existence of atoms because no one had really directly experienced the atom. I mean, it was all theory. Well, when the theory works well enough in a lot of different ways, and you can see it really hinges on the existence of atoms, then you’ve observed atoms. That’s what observation is.
Steve, to go back to an earlier point in our conversation. I asked you about, as an undergraduate or a graduate student, if pursuing cosmology was even available to you. And I asked that, sort of, as a stand in for gauging the maturation of the field for which of course, it was immature during the 1950s. And so, looking back at your students, when would you locate historically, the first students you would’ve had who would have expressed that interest in cosmology and you were in the position to say, “Yeah. This is a real field. This is something that you can pursue and do a dissertation on”?
I don’t remember specifically. I’ve had students who worked on cosmology, but they’ve been fairly recent, here in Texas. John Preskill was a student of mine at Harvard who entirely on his own initiative did research (I think this was his thesis) on the abundance of magnetic monopoles that would have survived from the early universe. So, Preskill was someone who was self-propelled toward cosmology. Actually, he left the field and went into quantum information theory and was awarded an endowed chair at Caltech. I’m very proud of him. I think he was the first student I ever had who worked on cosmology.
I don’t collaborate very often, but I did have a nice collaboration with the Korean-American theorist, Ben Lee. In our conversations he got interested in cosmology, which he hadn’t been before. We wrote what I think was an important paper. I think it was the first paper about the so-called WIMP miracle. That if you imagine that the dark matter particles are weakly interacting massive particles which are stable but can annihilate in pairs, and that stop annihilating as the universe expands because they become sufficiently rare so that they can’t find each other anymore, you can calculate the abundance, and from that calculate the dark matter density in grams per cubic centimeter. And if you take the observed dark matter density and run the calculation backward, you can predict that these particles would have a mass in a familiar range, like 10 to 100 GeV.
Unfortunately, I think we foolishly referred to these particles as heavy neutrinos, although what we were doing would apply to any kind of WIMP. So, I think it would have been better, it would’ve had more of an influence, if when we wrote our paper we had just everywhere said “wimp” or “weakly interacting massive particle.” where we had said “heavy neutrino.” But that was an example of someone, Ben Lee, who had done very good work in particle theory and suddenly getting converted into an interest in cosmology. Which is what I took your question to be about. That was in 1977. Ben tragically died in an automobile accident a little bit after that. He was the best collaborator I ever had.
In the past 20 years, it’s been quite common for particle physicists to go into cosmology. Do you see your interests 20, 30 years before that as being ahead of the curve in any way?
A little bit. I was teaching general relativity and cosmology at Berkeley in the mid ‘60s and then at MIT in the late ‘60s. And then at Harvard in the ‘70s. I think I was a little ahead of the curve. But not that it mattered very much. Most of my work was not in cosmology. Most of my work was particle physics having nothing to do with cosmology. I think about 10% of my work has been in cosmology, if I just count papers.
And intellectually, do you tend to separate, in your mind, those fields? When you work on particle physics are you not thinking about cosmology and vice versa?
Oh, I’m always thinking about how one can help the other. In the late 1970s I suggested a correction to the Standard Model that would lead to small neutrino masses. When I say small, I mean less than an electron volt. By then already there was some evidence of such neutrino masses from the solar neutrino problem. It’s been pinned down since then, that there really are neutrino masses like that. I had introduced what’s called a dimension 5 operator. And lately I’ve been thinking about whether or not the existence of that operator would have other cosmological implications. So far the answer is no. I haven’t been able to squeeze anything interesting out of it. But, it’s the sort of thing you think of. If you think of some wrinkle in elementary particle physics, whatever it is, you’ll immediately ask yourself, “Are there any cosmological implications?”
Or if you hear of something cosmological as, for instance, a while ago there was some evidence that a kind of polarization of the microwave background was discovered that showed that there are gravitational waves left over from the Big Bang. This conclusion actually was wrong. The incorrectness was pointed out by a former student of mine, Raphael Flauger, of whom I’m very proud. But when we thought that was true, we thought there were gravitational waves left over from the Big Bang, which told us something about the physics of inflation. And some of us were thinking, “Well, what kind of elementary particle physics model could lead to that kind of physics?” Well, the thing went away. I mean, there may be gravitational waves from the Big Bang, but the evidence for it went away. But that’s the kind of thing that happens. It’s certainly nothing special to me. I think generally speaking, particle theorists always have in the back of their mind possible, cosmological implications. And cosmologists have in the back of their mind possible lessons for particle physics. Um. That’s standard now.
Steve, sort of a broad question. Given how intensively you were focused on theoretical particle physics for the course of your undergraduate and graduate degree, looking back, in what ways was that the perfect intellectual foundation for your oncoming interests in cosmology? And in what ways did it not prepare you for thinking about cosmology?
Well, as far as the cosmology of the first three minutes, the period of nuclear synthesis, that’s very much involved with beta decay physics. And beta decay physics in the ‘50s was at the center of our trying to understand the weak interactions. So, yeah, people who had grappled with nuclear beta decay, as I had, and many other people had, were primed to understand the reactions that went on in the first three minutes where neutrons and protons were turning into each other through the beta decay interaction. If you go into more modern particle physics having to do with gauge theories, the Standard Model and all that, it is used by people who study inflation. But it’s not an essential prerequisite.
And to reverse the question, in what ways has your focus on cosmology enhanced your understanding and research in theoretical particle physics?
Oh, not much. Cosmology provides a challenge. The biggest challenge now is to understand the dark energy. The next biggest is to understand dark matter, the WIMPs or whatever they are. And that directs you into certain fields. For instance, worrying about dark matter strongly propels you toward thinking about supersymmetry. But it doesn’t help particularly in thinking about it.
Yeah. I think we’ve pretty well covered-- You know, I think one or two questions would be fine. I’ll say my penultimate question is because you have been concerned with conveying cosmological concepts to a broad audience. I wonder if you can reflect on the term “Big Bang”? And the way that it serves as a useful metaphor for a broad-based audience to understand the origins of the universe, and the way that it really sows more confusion than might be necessary?
Ah. Well, I’ve encountered that. Originally, as I’m sure you know, the Big Bang was a pejorative term used by Hoyle and other believers in the steady-state theory. They thought the idea of a cosmic explosion was silly. And they called it the Big Bang. It’s become a shorthand for the whole expanding universe picture. It is profoundly misleading in the sense that people think of the Big Bang the way you would think of an ordinary explosion on Earth. Something that occurs when, say a warehouse full of ammonium nitrate blows up, an explosion with a definite center that expands and does all kinds of mischief in the neighborhood. It’s very hard to get people to get into the idea that it’s not a localized explosion. It is something that happened everywhere. It is ongoing. We are in it. And it’s everywhere. So, I think “expanding universe” would be a much better term. But, what can you do?
[laugh] Steve, for my last question. Of course, so many issues in physics are a matter of debate among the most eminent scientists. And so, I’d like to ask you if you see yourself as part of a particular intellectual tradition or scientific tradition where your views are representative of a particular argument or tradition? And what you might see as some future prospects in the kinds of things that you’ve promoted in cosmology over the years?
Well, earlier in our conversation I talked about the dogmatic flavor of cosmology. Of course, no one ever regards their own views as dogmatic, but some people regard everyone else’s views as dogmatic.
And I wanted to avoid that. And at the same time when I wrote my book Gravitation and Cosmology, I wanted to give the feeling that this has really become normal science. We now have a body of physics about which we’re not certain, but it’s well enough established so that we can take it seriously and use it as a basis for further work. So, I introduced the term Standard Model. That’s the heading of a chapter in my book. The whole point of the term Standard Model was to emphasize that it’s not a dogma. We’re not certain of it, but it’s worth having at the center of our thinking. It’s well enough established for that. If we’re going to learn something really new, it will be as a departure from the Standard Model. But you have to have something to depart from. And then I brought that term into particle physics later. With the same spirit we now have a Standard Model of elementary particle physics. It’s not the final answer. It certainly isn’t.
In fact, I’ve done my part in trying to think of modifications that would not do violence to the underlying spirit. But it’s worth taking seriously. Some of the philosophy of science has been horrified at the idea of science becoming like religion. Mach, who I mentioned earlier, when he wanted to criticize the idea of atoms, referred to it as a religious dogma. It was not a religious dogma. It was a Standard Model of viscosity and Brownian motion and various other things that works very well. Physicists were right to take it seriously and see where it led them. It was not a religious dogma, but it was not empty vaporizing either. That’s the way I feel about Big Bang cosmology and that’s the way I feel about the Standard Model in elementary particle physics.
Well Steve, I want to thank you so much for spending this time with me today. It’s been an honor doing this and I really appreciate it.
Well, thank you for your good questions. I enjoyed the conversation.