Charles Misner

Smeenk:

I wanted to start off by asking you, why did you choose to work on relativity when you were a graduate student at Princeton? Why did you choose to work with Wheeler?

Misner:

Okay. I got a lot of advanced education at Notre Dame before I came to Princeton, and I’m pretty sure I’d had a relativity course there. In my first year at Princeton I worked with Arthur Wightman on the problems of double beta decay. I’d had piles of mathematics before Princeton under Arnold Ross, who was a famous teacher and also researcher. He went on to Ohio State later and ran programs for schoolchildren and high school kids to introduce them to abstract mathematics, very successfully for many years until he was in his nineties. He was quite a wonderful guy.

Smeenk:

So he was a big influence on your development?

Misner:

He was a big influence on getting me started, pushing me ahead at Notre Dame to skip a bunch of courses and that sort of thing. Had me reading Bourbaki and all kinds of stuff. Anyway, I came to Princeton, I knew all the mathematics, I knew all the physics, and I started doing some research with Wightman while taking a few courses, not much. But I was interested in this area of abstract mathematics and the kind of things that Wightman had worked on. After the first year, I passed my qualifying exams and it was time to choose a thesis director. Wightman was very up front about the fact that his students tended to take a long time and his current student was just finishing a 700-page thesis. [Laughs] John Wheeler had been bubbling around and talking to me about all kinds of things. I’m sure I had talked to him also. But then I went around and talked to him about possibilities and he was full of enthusiasm and said, “Oh yeah. This sum over histories idea for doing quantum gravity, you can probably turn that out in six months.” [Both laugh]

Smeenk:

In six months. Wow.

Misner:

I forget whether that was the first thing I started or not. But anyway, he also was drawing pictures of wormholes, and I had had enough mathematics I knew you could actually do the mathematics of the whole thing. At least I knew some of the topology and things like that. After digging in a bit, found that you could do the mathematics of it. So it was a good fit for us and we did a lot of stuff together. I was also working on another thesis project which Hugh Everett had somehow found the suggestions of someplace. I worked with that for an Annals of Physics paper "Geometrodynamics'', Charles W. Misner and J. A. Wheeler, Annals of Physics 2, 525-603 (1957). John Wheeler’s much later work on charge without charge and so forth. Geometrodynamics was the name of it. In there was a section that was supposed to have been my thesis for a year or two. It was, we discovered, actually mostly published elsewhere years before. And so at that point I relaxed back into the quantum gravity, and did turn out something in six months or so, that got published in Reviews of Modern Physics. "Feynman Quantization of General Relativity'', Charles W. Misner, Reviews of Modern Physics 29, 497-509 (1957).

Smeenk:

Right. And it’s about Feynman quantization?

Misner:

That’s right. Yes. So in all that period I was doing relativity because John Wheeler was an exciting person to work with and he had all these ideas and I had piles of mathematics that could help implement them. It was a good fit, putting these things together. He was then pushing towards his worries about the issue of the final state, what would happen when a star collapses. He was doing a lot of things with other people on the properties of matter because he was coming out of his nuclear physics phase at that point. But then I got —

Smeenk:

Did you interact much with Dicke’s group there?

Misner:

Yes. In other words, I was aware of what Dicke was doing; I would occasionally attend their seminars. Then, when I did get my degree in ’57, Dicke recruited me to be Carl Brans’ thesis advisor. Because, although Dicke had these ideas of the Brans-Dicke theory and Brans was working with him and so forth, he said he thought he was too much of an experimentalist to supervise a theoretical thesis. So I was, on paper, Brans’ thesis director. I must have contributed something to it, but not the original ideas.

Smeenk:

Right. You interacted with him as part of the committee.

Misner:

With Brans I interacted fairly regularly, I think, but not as much as Dicke did. But as for Wheeler, I got into relativity by a good match of personalities and mathematical backgrounds.

Smeenk:

This is partially based on other people’s descriptions from the thirties and fifties. Did you ever have a sense that relativity was a more isolated field than say particle physics, that it was somehow separate from —?

Misner:

Yes, very definitely. And Wheeler was beginning to break that, but it was clear that (and Wheeler said at some point to me — not during my graduate student years I don’t think, but not [much later]) that he always was interested in gravity only never put a lot of time into it until he taught a course in it, which was his way of learning it. In probably ‘52, just the year before I came. But he said that during the thirties he wouldn’t take a student to work on relativity because it would have been the end of his career — I mean the end of the student’s career. There was just absolutely nothing in the universities, nobody interested in hiring somebody who had worked on relativity. And so it was definitely out, and I think it probably continued to be out at Princeton even as John was working on it, because other people thought it was less interesting than particle physics. And it may well be that that influenced John’s eventually leaving Princeton, although I don’t know that he reached the official retirement age. He may have. But anyway, I think he was probably a bit unhappy with trying to get Princeton to hire a really good person in relativity and I would think that he had hoped they might hire Penrose or Geroch. I don’t know who else, but I mean he had a few candidates — he would like to push them through the department. So in that sense relativity was definitely not high on the priority list of either the Princeton department or most physicists. I mean, they would look at quantum gravity and say, “Well, that involves the Planck length, 10^-33 centimeters. We’re having enough trouble at 10^-14 centimeters, so that’s just pie in the sky.” It took many years until the particle physicists ran out of experiments and then found things like the renormalization group that were leading them to postulate the Grand Unified scale and so forth. They were making their own 10¹⁰ extrapolations and began to take gravity more seriously. Of course, Dicke and Peebles’ interpretation of the microwave radiation was certainly a big turning point.

Smeenk:

That leads into my next question. I was wondering what your initial reaction was to the microwave background radiations, if you recall.

Misner:

I do recall, because that was the summer of ‘65 when there was a big summer school at Cornell and I was on the organizing committee. I heard from Peebles. I don’t know in what connection, whether he called me to tell me this exciting news or whether we talked for some other reason. But anyway, I got the word directly from him that this was about to be published or had just been published and so forth, so the first thing I did was arrange for him to come up and give lectures on it at Cornell. The idea clicked immediately. I thought it was a turning point and got that whole crowd of relativists at Cornell to hear about it right away. So that was a big thing.

Smeenk:

Right. So when you started — so was this what really spurred your serious research interests in cosmology?

Misner:

Yes. I had done mathematical cosmology that was unrelated to the real world before… with Taub. I worried about trying to interpret the Taub universe, and the extension to the identification of the Taub and the NUT metrics. "The Flatter Regions of Newman, Unti, and Tamburino's Generalized Schwarzschild Space'', Charles W. Misner, J. math.Phys. 4, 924-937 (1963). I guess I somehow did that piece just as playing with things that looked mysterious in the literature, but then I guess when I put those two together, when I began working with Taub, he taught me how to easily find geodesics in places like that so you could put some sensible physical interpretation on what’s going on. "A Singularity-free Empty Universe'', C. W. Misner and A. H. Taub, Zh. Eksp. Teor. Fiz. 55, 233-255 (1968); English original in Soviet Physics — JETP 28, 122-133 (1969)] Although that work wasn’t published for a long time, it was done, that work with Taub on geodesics, was done quite early.

Smeenk:

There is a paper you published in ‘63?

Misner:

I don’t keep these dates in my mind.

Smeenk:

Right. Maybe I can ask you about the reference later. [Above]

Misner:

Yeah. I’d have to look at the list of publications to be able to sort them out and remember what was when, but it seems to me there are two things on that: one where I put together the Taub metrics and so on that one was an analytic continuation of the other and the global topology question, what was the global topology of Taub and so forth, which maybe even he understood, but I began to understand that. Yeah, I guess that was it, that one could appropriately interpret the NUT metric by fitting it into a S³ topology. [1963 paper above] And then I worked with him on the geodesics, and eventually that paper was published in the JETP [1968]. And I always regretted that we didn’t let it go through the standard translation back to English. Instead we sent the American translation journal [Soviet Physics — JETP] a copy of our English version.

Smeenk:

So then when you did start working on cosmology, you obviously felt that the uniformity of the early universe was one of the pressing problems.

Misner:

That’s right. Questions like that had been raised by Dicke. He had some very early papers in which he picks up what he calls the flatness problem. And I think those are in the — well in slightly obscure places like the American Philosophical Society or something like that.

Smeenk:

Right, [such as] the Jayne lectures presented in ‘69 in Philadelphia. But going back to 1965, you talked with Dicke about those problems already or was it just that…?

Misner:

I think so. I can’t be sure. I think he was posing those questions about cosmology, but I don’t know when. Sometime in the sixties. And I guess — well, certainly I realized right away that the microwave background was a real bit of physics bearing on cosmology, and the question that I don’t know where it came in, but Wheeler at some place was talking I think about cosmology. Yes, because he was always in favor of a closed universe that would recollapse. Because without the cosmological constant, which was the standard assumption, then if you have enough matter to close the universe it will recollapse later.

Smeenk:

Right.

Misner:

And he was intrigued by an analogy to something he’d seen during his wartime service or in connection with it, somehow military; he had seen descriptions of what happens when an explosion goes off underwater. And you get a big bubble with the blast from the gases and explosion. But he had seen cases where the bubble would expand but eventually it would overshoot and then it would recollapse again and expand. It would go through some oscillations as it goes through water. And the thing that intrigued him was that as it expanded it smoothed out, and as it re-collapsed it turned into a mess of all kinds of strange things. So it was that stability question of how things would smooth out during expansions and irregularities would amplify during the collapse that was intriguing. So that whole picture ties into some of these questions, but I can’t place the time. Because it suggests that people were justified, in a sense, in ignoring the Gamow discussion of the microwave background and that picture of nucleosynthesis in the early universe and the corresponding things. Justified in the sense that the observations of the universe had shown that it’s expanding and a little bit that it was homogeneous or isotropic. I mean the evidence on that was maybe the density of galaxies in different directions was uniform to within 30 percent or something. They really didn’t know much. And even if you assume that from samples off the poles of the galaxy where you could see farther away than you can in the plane that it might have been smoother than that. Even so, even if you said it’s a part in a thousand, this picture of how things smooth out as they expand, but crumple as they go back, means that if you extrapolated into the past it could have been a very messy thing.

Smeenk:

Right, right. Even if you have a good picture [of current uniformity]…

Misner:

And then it would be very hard to make all these analogies. So I mean extrapolating backwards in the universe by more than a factor of three or five was certainly a priori nonsense.

Smeenk:

Right. But then once you see the microwave background…

Misner:

The microwave radiation then comes out and says, “Hey, it’s a factor of a thousand.” Something weird is going on. And so that, somehow in connection with that, which probably just — Dicke came up with this idea that the fact that the universe is, within any range of being flat — not closed off, open like that — because he can see from the equations that that was also like these other things, a very unstable situation to extrapolate backwards. So he pointed out that mystery.

Smeenk:

I believe you went on a visit to Cambridge in 1966 and ‘67.

Misner:

Right, that’s correct.

Smeenk:

During that time in earlier interviews you have mentioned working with Faulkner and Srittmatter and how that got you going looking at anisotropic models.

Misner:

That’s right. So that was really the origin of where I first began thinking about cosmology in a physical sense. Previous [work] related to the Taub-NUT universe and so forth, was just mathematical exercises on some solutions of the Einstein equations that I didn’t see having any particular bearing on the real universe.

Smeenk:

Right. So what were the discussions you had with Faulkner and Strittmatter? I believe it was observations of quasar distributions?

Misner:

That’s right. It was just the beginnings of quasar observations. There were a handful of quasars identified, and they plotted them on [some] a blackboard globe and thought that (even though the numbers were small) they were anisotropic ally scattered over the sky. They were trying different ways where… some anisotropy would show up in the universe. And playing with these mathematical models, I knew that there were ways to describe anisotropic universes because the Taub universe is one such thing. And so I looked at two things. First I looked at the flat, anisotropic universes, things like Kasner, although that’s empty space, you can add some matter to it and still can solve the equations.

Smeenk:

Right.

Misner:

So I was looking at that and studying those geodesics and trying to see whether you could get any effects in any way related to anisotropic distributions and something observable in the sky.

Smeenk:

These are the Bianchi classification?

Misner:

Yeah, Bianchi classification [a group theoretic classification of homogeneous cosmological models.]That would be Bianchi I. But then the Taub universe was the Bianchi IX class. So I also at essentially that same time developed the mixmaster universe. But when I came to publish results in the Astrophysical Journal, I concentrated purely on the Bianchi type I. "The Isotropy of the Universe'', C. W. Misner, Astrophysical Journal 151, 431-457 (1968). Although I had at that same time [received] a Gravity [Research Foundation] second prize, I guess, actually third prize, 1967 for an essay that described the mixmaster universe.^[1] And this led to some problems with the Russians, because that Bianchi thing was — I mean, the Gravity [Research foundation] essay was not so to speak in the standard literature, and they had published the first in the standard literature and I think are sometimes annoyed that I felt I had also invented the thing.

Smeenk:

So when you say developed the mixmaster universe, you mean developed the particular Hamiltonian techniques for studying the evolution of this as you approach the singularity?

Misner:

That’s right. Or just having an idea that there would be these repeated changes of anisotropic expansion going through an infinite number of epochs and so forth. Now the neat description of that in terms of counting and so forth was definitely the Landau — no —

Smeenk:

Lifshitz.

Misner:

Lifshitz’ way of —

Smeenk:

— and Khalatnikov.

Misner:

Yes, yes. But the same behaviors I had described in this gravity prize essay.

Smeenk:

Right. Okay, I want to go back actually to your 1968 paper on the “Isotropy of the Universe.”

Misner:

Okay, yes. Because that was definitely trying to talk about the real universe, whereas the mixmaster stuff was talking about, again, more or less mathematics. I mean, it was related to this same kind of stuff, but it was, from the point of view of describing the universe, it was a little extra, fit into a closed universe instead of an open one — instead of a flat one I guess. Well, no. I was interested in the hopes that it would bear on the horizon problem, and eventually it turned out not to.

Smeenk:

Okay. So, just to backtrack a little, one of the things that I was really interested in this paper is that you urge a new approach to cosmology and you emphasize that you should try to predict as much as you can of the presently observable universe just by looking at what properties would be independent of initial conditions. And I was wondering what motivated this approach. Was this —?

Misner:

Oh. Well, that was some kind of philosophical prejudice. I don’t know. When I started making this Zeno paradox or anti-Zeno paradox discussion of the early universe that was —

Smeenk:

That was in Misner Thorne and Wheeler, I’m sure.

Misner:

I know it’s in there, but there was a paper called “Quantum Cosmology” which was in the same issue as another paper that may be called “The Origin of Time.” [Charles W. Misner, Physical Review 186, 1328-1333 (1969)]

Smeenk:

I think it’s called “The Absolute Zero of Time.”

Misner:

“Absolute Zero of Time.” Okay.

Smeenk:

Yes. They’re right next to each other.

Misner:

Okay. And when was that? I don’t remember.

Smeenk:

That was ‘69.

Misner:

That was ‘69. Okay. So the ‘66-‘67 work, those ideas must have been developing together at about the same time. But it had somehow been a question in the air that the physics equations are deterministic so initial conditions determine the future —

Smeenk:

Right.

Misner:

The ideas of deterministic chaos hadn’t been coming to the open, although of course Poincaré understood this. But they weren’t general knowledge at that time. So the question of having a theory of cosmology, it seemed to me sort of unlike the rest of physics to be satisfied with just saying, “Okay, there are a million possibilities and we’ve observed which one is happening after measuring three numbers” — you know, cosmological constant, Hubble’s constant, and the acceleration parameter, which was the game among the astronomers for several decades, which was challenging. In fact it’s just now being done, although they’ve got some other numbers that are being done, but getting those three was amazingly difficult. So but it nevertheless seems that how are you going to have a theory of — how can you say you’ve explained the present universe if you can just say, “Well, there are millions of possibilities and we’re one of them”?

Smeenk:

So you thought in this case it was something that was different in cosmology than in the rest of physics, where in cosmology you were happy just saying, “We have a bunch of models and we find one that matches observations.”

Misner:

Right. Yes. I didn’t think that was an adequate challenge to cosmology. In line with this idea that there could have been an infinite past in the sense that an infinite number of important phenomena took place in the past, on whatever renormalizing time scale, that you might be able to get something closer to thermodynamics where you talk about how things are going to be when you let them sit for a while, so to speak, without having to ask how it got started.

Smeenk:

Right.

Misner:

And so I wanted more of an analogy to that, to say, “Okay, if the universe is old it doesn’t matter how it started.” There ought to be some way — obviously some detail matters, but the general statistical picture might be able to come out like statistical mechanics. There should be some viewpoint that says, “We can really say that this is the natural kind of universe to have” rather than saying, “Well, the universe might have lasted a twentieth of a second, or it might have lasted forever” — all these possibilities and we find out that one of them was somehow favored.

Smeenk:

It seems like there could be two different worries about having an explanation where you have special initial conditions, and one worry could just be that you think that it’s a false assumption that, say in this case, that the early universe was uniform prior to the background radiation. You think that that’s just a false assumption, so the explanation is wrong because you’ve made a false assumption. And the other, it sounds more like what you’re saying, is that no, it’s not just that it could be a false assumption; it’s that you want an explanation that’s robust in the sense that it doesn’t depend upon initial conditions.

Misner:

Yes. It’s more like the second one. I felt the only alternative for that to say that we somehow understood the universe would be to have a theory of initial conditions and essentially physics has no experience with that. We have theories of dynamics, how things change, but nobody has a theory of why an experiment is set up. [Laughs] And I didn’t outright exclude the possibility that someone might invent an attractive theory that says initial conditions have to be so-and-so, but it seems to be so different from all the rest of physics, I thought that showing such an idea would require a tremendously attractive viewpoint which in some way supported the thing. I didn’t see any clues in that direction at all. [Much later the Hartle-Hawking proposal for a no-past-boundary condition on the Feynman integral wave function of the Universe provided a model for viewpoints that propose a theory for (avoiding) initial conditions, but even now (2002) these remain unsubstantiated speculations.

Smeenk:

Let me ask you about someone who I think has the opposite intuition. Roger Penrose has argued, based again on thermodynamics, which actually the state of the universe should be really improbable in order for the second law of thermodynamics to work. I mean, I’m not sure whether I’m catching all the subtleties of his argument, but he seems to be pushed in sort of the opposite direction. Have you thought much about Penrose’s position?

Misner:

I haven’t thought about it very hard, no, but generally I have been suspicious about thermodynamics in general relativity simply because I have never seen a good definition for entropy that includes a gravitational contribution. You can get it for isolated systems. And so the thing I resonate best with is arguments that I first saw in Fang and Li this little Chinese high school textbook. [Creation of the Universe by Fang Lizhi, Li Shu Xian, T. Kiang, Li-Chih Fang, World Scientific Pub Co, 1989; ISBN: 9971506009; Paperback (September 1988) ISBN: 9971506017].

Smeenk:

I haven’t heard of that.

Misner:

It’s written by Fang and Li, husband and wife. Fang Lizhi was this scientist now in Arizona who was famous for a while because he was holed up in the American Embassy in China as a, you know, whatever they call them, unwanted person.

Smeenk:

Political refugee?

Misner:

Yes. And eventually he was allowed to come to the United States, with his wife, who was also a scientist. And so they wrote this little textbook, and in it he’s got some real gems of understanding cosmology and its questions and points out that —

Misner:

Something like a globular cluster, you can treat because you can do the energy of the system unambiguously. Then you would come to the conclusion that gravitational systems have a negative specific heat. Now that’s — he’s not unique in making that statement, in the sense that as they lose energy they get hotter and then become faster. Right?

Smeenk:

Right. I’ve seen that argument for black holes.

Misner:

Yes, right. And it’s certainly evident that as stuff falls into black holes you get X-rays going forth at very high temperatures and so forth. Yes. So but he makes a story out of that to suggest that this feature of gravity is what is responsible for — because it does something to entropy. After all, you calculate entropy as an integral over specific heat with respect to temperature. So he says, “Well, this is one of the clues for how you can start with a relatively uniform universe and get to things like us,” that the subsystems begin organizing and that the gravity is really the clue to making the universe interesting in that way. So I have always tended to that side of the argument that says entropy is a lot more mysterious in general relativity. And in closed universes, both entropy and energy are sort of hard to define, although lots of people suggested (including MTW) that a closed universe has net zero energy. It’s still hard to turn that into believable mathematics. And the best hope for that — which seems like it may be a workable system — was put out by Lynden-Bell about five years ago, a paper; I think it’s in Monthly Notices, but its Lynden-Bell and somebody he worked with. “Mach’s principle from the relativistic constraint equations”, D. Lynden-Bell, J. Katz and J. Bicak, Mon. Not. R. Astron. Soc. 272, 150-160 (1995), with errata in Mon. Not. 277 1600 (1995). They managed to define a version of the stress energy pseudo-tensor for gravity, which would include not only gravity but all other parts of energy (or all forms of energy could be included), in a form that allows this argument. Namely, that since the total energy can be evaluated as a surface integral by sort of a Gauss theorem, just push that surface out and if the universe is closed it eventually becomes a small surface or a zero size surface, and so you end up with zero. Well, that argument a priori never worked because the Gauss theorem, unless it’s applied in a proper tensor way only works in a single coordinate patch.

Smeenk:

Right. And then you need Killing vectors?

Misner:

And you need at least two coordinate patches. What?

Smeenk:

Don’t you usually need a symmetry to go from the differential form to the integral form of the —?

Misner:

Well, the pseudo-tensor idea, if you’re working in a single coordinate patch you don’t need any special symmetries. But you do typically need an asymptotically Minkowski boundary condition to be able to turn that into something that has an unambiguous meaning.

Smeenk:

Right.

Misner:

And Lynden-Bell came across a way in which you can get around that particular difficulty. So that may be a clue to making progress on this problem. But I don’t believe the question of the total entropy of the universe, has been properly defined, and therefore I am unimpressed by arguments of that sort. Entropy normally done by cosmologists is the locally measured entropy, you know, entropy density that doesn’t simply add up.

Smeenk:

Right. I want to go back to your work on the horizon problem. How did you come to recognize the horizon problem? Was it while you were working on the neutrino viscosity and other means of isotropizing the early universe?

Misner:

Well, the neutrino viscosity was basically I think an attempt to solve the horizon problem, is that right? CWM 2002: No, I think it was an attempt to solve the problem of why the Universe should be approximately isotropic.

Smeenk:

Chronologically the paper “Isotropy of the Universe” comes just a year before the mixmaster paper.

Misner:

Right. But those were — But there is a Phys. Rev. letter in which I specifically talk about the horizon problem. When was that?

Smeenk:

That was ‘69.

Misner:

That was ‘69. I see. That was much later. Okay. I can’t actually remember when I began worrying about the horizon problem. I think I’ve heard Martin Rees say that such things were part of the discussions at DAMTP in Cambridge during my visit there, but I can’t specifically remember when I began worrying about it. Although since the microwave radiation was clearly in the background and in my worries about those times, so it’s possible I calculated, that is realized that there was a horizon problem, but by the time I was publishing, I was publishing ways of attacking it.

Smeenk:

Yes, you published the problem along with the mixmaster.

Misner:

Right. Yes.

Smeenk:

So one question I had about the horizon problem is, does the presence of particle horizons imply that you can’t have a dynamical explanation of the uniformity of the early universe? This is the sort of thing I have in mind: if you were able to show the neutrino viscosity or some other effect led to isotropization, even if there were particle horizons present would that indicate there is a problem with the model?

Misner:

Well, I would think the neutrino viscosity could smooth things out; that is, bring isotropy within a horizon size, but if you were actually limited by the horizons I don’t see that it would bring the two different pieces. I mean, if we look out to the microwave background and two pieces of it have never contacted each other, each of them within its own lump might kill all the short wavelengths but there is no reason the two should be equivalent unless they had somehow interacted.

Smeenk:

So it’s sort of a range of initial conditions. Neutrino viscosity might make it a larger range of initial conditions which could lead to isotropy, but you don’t see how it could smooth out all of the sky to the same temperature.

Misner:

Right.

Smeenk:

Okay. So —

Misner:

So the first thing that did that was inflation. And you will find recent papers by Peebles in which he says inflation is still not established but clearly provides a kind of explanation for what’s going on, and there are other people who say, look, “Unless you get a natural measure on initial conditions to tell you what probability means at the beginnings, anything you want, you can run the equations backwards and find some weird initial conditions that give rise to it.” So there is this contrast between the intuition of many cosmologists who see the effects of inflation in damping inhomogeneities, also washing out horizons. So there is this tendency to say, “Okay, that’s an explanation,” and mathematical types say, “Well, unless you can define what probability measure we are using we can always look for some initial condition that gives rise to a completely different universe.” Intuition says, well, those are measure zero but nobody ends up measuring these, so the argument doesn’t convince mathematicians.

Smeenk:

So are you more sympathetic with the intuitive argument?

Misner:

I’m more sympathetic with the intuitive argument. Specifically I would say that the horizon problem and Dicke’s flatness conundrum, and the fact that the universe does look the way it does now, very uniform, [indicate] that if it wasn’t inflation it had to be… That’s an example to say that there should be some theoretically understandable, plausible explanation for why the universe is the way it is. There should be a mechanism that produces the uniformity rather than just saying, well, that’s the way God made the universe.

Smeenk:

Right. So once we have an explanation it would be very hard to go back to assuming that it was some part of the initial conditions.

Misner:

That’s right. I mean, it may remain a mystery for a long time, but that’s somehow the beauty of mysteries. They usually end up being solved while posing some deeper mystery! [Laughs]

Smeenk:

Let me go back just a second to another approach to try to explain isotropization, and this was I think initiated by Zel’dovich. Leonard Parker also worked on this, and this was the idea that particle creation in the early universe could lead to isotropization. And you also worked on this using I guess mini-superspace quantization method.

Misner:

Yes, well I’ve worked on the quantization methods. I never really worked on the particle creation side of it. But yes, I would regard that as better physics than neutrino viscosity. It’s a dissipation mechanism which is well founded in theory, I think, and it would be a very acceptable part of anybody’s picture of how the universe got where it is.

Smeenk:

Okay.

Misner:

I haven’t followed these things closely enough to know whether it solves the problem.

Smeenk:

I don’t know what the sort of current literature is. I was just wondering. So your initial assessment was very positive. It struck you as something that could solve the problem.

Misner:

Yes, right.

Smeenk:

Okay. So another general question, who were you working with in a lot of this work in the late sixties? Who were your collaborators or colleagues that you interacted with a lot in the late 60s early 70s?

Misner:

Oh, I think the most important person was Sciama. He was an amazing leader. He could pull together groups of people, inspire them to work on interesting problems, he knew the literature and problems over a very broad scope and could bring up questions that other people would try to solve. I think he was inadequately recognized in Britain for the really outstanding kind of leadership he provided the relativistic astrophysics community in the world. Obviously the people he brought together were also important too, but working there in Cambridge that year and later visiting at Oxford, both of those were very valuable experiences. Next to Wheeler I’d say he’s probably the most important influence on my professional development.

Smeenk:

Is there anything that I’ve left out in asking about some of the research you were doing in the late sixties, early seventies, regarding cosmology in particular or other interests as well?

Misner:

I don’t know. I mean, I have off and on dabbled in the philosophy of science and sometimes in the interactions between science and theology, so I certainly did some of that in the seventies. I don’t remember whether I was doing anything [at that time]. Well, sometime in the sixties there must have been, with this pre-thesis work with Wheeler on geometrodynamics, there was this theory of Einstein equations combined with Maxwell equations. I always took that as a prime example that somehow the laws of physics as we know them don’t seem to be necessary. In other words — Wheeler likes to quote Einstein as saying one thing he would really like to know is whether God had any choice when he wrote the laws of physics. There is a sense in which you have to know what the ground rules are. And so I think this example bears more on the ground rules, but Maxwell’s equations plus Einstein’s equations, this concept of Wheeler’s geometrodynamics that is things are nothing but fields (and of course you never expected them to be pure classical fields). The point is one eventually found that there wasn’t a lot you could do with pure classical fields. You can get charged massive particles and the particles would be black holes. They could be charged or uncharged. Their interactions seemed to be completely determined by the Einstein-Maxwell equations. You could have bound systems. They would of course eventually radiate classically; you don’t have quantum mechanics to make them infinitely stable. That’s a fairly rich universe and it always seemed a challenge to scientists or metaphysicians in the sense that anyone who wanted to admit the ground rules under which physics had to be done to say, “Why is this apparently completely internally consistent theory not implemented?”

Smeenk:

So if you were trying to look at it from the approach of just formulating a completely consistent physical theory. Why is that not necessarily the case, why is something else?

Misner:

Yes. Right. Right. So this still appeals to me as an example to say unless you can refine the ground rules there seems to be an alternative set of rules that make an interesting universe that’s not ours. And that of course is very much not ours, because there is no quantum mechanics. And of course there is no theory of initial conditions for this universe. There exist solutions of the Einstein equations that essentially have galaxies made out of black holes and so forth and they should last for quite a long time. Why not?

Smeenk:

Right. Actually, there is one thing that I forgot to ask you about which I had in my notes, which is, in this paper on the absolute zero of time you adopted a position of tolerance towards singularities.

Misner:

Right.

Smeenk:

And I was wondering first of all what motivated that position, was that very different from a lot of the people who were working in the field at the time and their position toward singularities?

Misner:

Yes, I think so. I think most people felt that singularities were absolutely abhorrent. And I don’t completely disagree with that in the sense that you can say that the prediction that a singularity must occur inside a black hole or if the universe were to recollapse in our future. If you could show it leads to the Planck length… everyone expects that (as in the simple examples) it leads to infinite curvatures and infinite densities and things like that. Assuming that’s the case it brings you down to the Planck length and it’s clear that nobody would defend the present theories as adequate to deal with physics at the Planck length. So in that sense, yes. I mean, I think Wheeler was hoping there might be some out before the Planck length, but never found that. But the other attitude on the universe was simply, since Hoyle and others were, and still are, willing to say that a steady state universe is an acceptable situation, I was just in some sense introducing something like the renormalization group to say that if you just keep re-scaling time according to what’s needed for the current situation it seems quite possible to have a — not quite a steady state universe, but the renormalization of time thrown in, something that lasted forever — lasting forever in the effective sense. And it would be perfectly adequate if you could find something like statistical mechanics or thermodynamics explanation of the present state to say okay, I don’t care that the past proper time is a finite length of time as long as there are an adequate number of interactions to make this statistically required standard state arise. So it was more in that sense that I didn’t think the proper time being finite or densities becoming infinite was a disaster; it just meant that things go on at a different scale. And I’m still interested in playing that game for the future.

Smeenk:

You mentioned that yesterday, Dyson’s…^[2]

Misner:

Right. Dyson’s paper. Yeah. And to modify that to take into account what is presently known about cosmology instead of what was known then. Well the other point on Dyson, though, I mentioned this business of Fang Lizhi explaining that excitement in the universe is essentially due to gravity and its specific heat and so forth. Dyson had developed that whole line of thought independently. He gave a lecture at Maryland and has probably given it elsewhere. But he added more. He had a bunch of biological metaphors that were also fitting into his description of the universe. And I don’t think it’s been published in a very accessible way. I may be responsible for that, because when he gave a talk at Maryland I pointed out Fang Lizhi and so he recognized that that was the same as part of his ideas. But the other part was very interesting. It was being done with Lynn Margulis. [I’m bad on names, a woman biologist [Lynn Margulis ??] who was famous for pushing the — well, she was Carl Sagan’s first wife. I’ve forgotten her name, but.] She’s a very excellent and well-known biologist. [cf. http://www.umass.edu/newsoffice/archive/2000/013100margulis.html .] She had been working with Dyson on this. They had a whole bunch of wonderful biological analogies that were also playing a role in the big picture about how the universe is developing, a number of examples to bring in. So I would still like to see them publish those ideas.^[3]

Smeenk:

That sounds very interesting. Let’s see. Do you recall working on any of the other fine tuning problems in the sixties or seventies or thinking about them instead of — what I have in mind are things like the matter-antimatter asymmetry. You have mentioned the flatness problem, and how that became important later.

Misner:

Right.

Smeenk:

Another one that’s often mentioned is the entropy measure in terms of number of photons per baryon.

Misner:

Yeah. No. I heard about those problems, but I never worked on them.

Smeenk:

Okay. And let’s go on then to discuss your assessment of inflation. So you mentioned earlier — and I think you discussed this as well with Lightman in an interview a few years back, that your assessment of the flatness problem changed based on inflation.

Misner:

Yes.

Smeenk:

And is it correct to say that in some sense it seemed much more like a legitimate problem once you had a solution to it with inflation?

Misner:

Yes. I had problems understanding what the problem really was when Dicke first presented it, and that was basically because I had too narrow a picture of the determinism in the Einstein equations. And inflation made me realize that you can affect how things behave if you can tweak the state of matter, so to speak. And so inflation was a major introduction of new ideas into possible equations of states of matter and had this very qualitative influence on the postulated past history of the universe. So that sort of opened up the question to me as something that you could make alternative theories about.

Smeenk:

What do you think is most important in assessing new ideas such as inflation? If you remember when you were first introduced to the theory, what was the most important thing in assessing this for you?

Misner:

I guess the main things are: is it a clear proposal so that you know what the proposed theory is, can people agree on what the consequences of the theory are, and do those consequences fit some observations. So that’s the critical rules.

Smeenk:

One thing that critics of inflation have often complained about, and even proponents like Mike Turner, have said that inflation is a paradigm without a theory in the sense that the actual mechanism for generating inflation is still somewhat mysterious twenty years down the road.

Misner:

Right.

Smeenk:

Is that then a problem for assessing the theory?

Misner:

No. It’s a challenge in the sense that one expects these things to be clarified, but most every theory has got some deeper level of detail that you would like to work out, and obviously this kind of thing interacts with whatever remains to be understood, which is a lot about the structure of particle physics at those wild energies.

Smeenk:

Would you be happy with a theory of inflation that doesn’t necessarily have a tight connection with high-energy particle physics, so say Linde’s ideas that you don’t necessarily need to have the properties at the scalar field determined by particle physics. But you could still have inflation.

Misner:

Well, I would regard such theories as stepping stones. In other words, if you can’t solve all the problems at once why you make some conjectures on one phase of the physics but if they became flatly contradictory to what was known from some other phase, why the whole system would have to be reworked. The fact that two people or two groups, or themes in the development of the physics going off in slightly different directions wherein they clearly all have a large component of speculation that they make different choices to best fit the data that fits in that field doesn’t bother me immediately, but if things became not so speculative and people thought they really had firmed them up, why then I would think they should agree. In a sense I’ve claimed that was the strongest argument for Einstein’s theory until about 1960, in the sense that people always said, “Well, there’s only three experiments,” you know, the bending of light, Mercury, and the red shift.

Smeenk:

Right.

Misner:

And my feeling was that that wasn’t really true, so in a talk at the Institute for Advanced Study on Einstein’s birthday I pointed out that there is a sense in which you can’t build an accelerator without general relativity. And the reason for that is that if you build an accelerator you have to align the magnets, and what keeps the magnets in place is primarily friction as they sit on the floor. If there were no gravity the whole thing would float around and couldn’t be aligned. So by necessity engineers have to believe in both gravity and electromagnetism — special relativity. Well, you can’t believe in both of those in the same brain in any means one knows except general relativity. Otherwise you are lobotomized; you’d have to do one part of it with one view of the world and another part with another view of the world, and so they never talk to each other. So if you really believe there is a way of thinking about things which fits together, at that time and still the only way to put these things together that was known is general relativity, or Brans Dicke or something else like that. But you’ve got to have such a theory. And so I would say the same thing for the eventual understanding of the early universe. If you really think that you’re claiming to have nailed it down in some way, why then all parts have to fit together.

Smeenk:

Right. So you’d say at this point inflation is still much more speculative than say general relativity because…

Misner:

Oh, [laughs], yes. I mean, you know, they are talking about phase of particle physics where people are spewing out theories wholesale, more or less. They have been trying to cut down the thing with, by saying well a lot of these theories are equivalent different versions of M-theory, which doesn’t quite exist yet, and all that kind of thing. So it’s clearly on a totally different speculative level.

Smeenk:

Right. Do you think that there will be a role of quantum gravity in trying to understand initial singularity? And if inflation turns out to be right, it will probably be fairly minimal, but have you been interested at all in quantum gravity and the possible implications it might have?

Misner:

Not especially, no. I haven’t been able to…

Misner:

I’ve never been technically strong with quantum field theory, and so I haven’t put a lot of effort into things in that direction. That means I don’t have well-grounded opinions of what might be going on. But I’ve been more intrigued, as I mentioned yesterday, by the possibility that things would change dramatically at the GUT scale, that inside black holes you might get false vacuum that would change the nature of the singularity or avoid the singularity completely or something so that you might be able to understand the whole history of a collapse of a star into a black hole, something going on inside possibly never at densities higher than the GUT scale, with a curious enough equation of state that you didn’t have to get to a singularity and eventually Hawking radiation releasing all that inside back into the universe billions of years later. So that’s a very wild speculation at the present time. I mean, it’s not on the scale of what we’ve been talking about before, but it’s something I’ve thought about for a few months, [laughs] not something I’ve worried about on scales of years. But it’s just to say I’m quite open to the possibility that you could find something ingenious in particle physics. I mean, it’s primarily brought up by inflation and the present cosmological constant as they call it, which I think might be just the beginnings of another inflation that takes place on a much larger time scale.

Smeenk:

Right.

Misner:

And so all that suggests is there really are things in the universe that are inconsistent with the energy conditions that are assumptions in the singularity theorem. So the real universe isn’t everywhere at all times going to satisfy these energy conditions; maybe the singularities are not as likely to occur as it appeared twenty years ago.

Smeenk:

Since we’re discussing the cosmological constant, this has sometimes been advertised as one of the largest problems in reconciling quantum field theory and general relativity, but if you try and calculate the cosmological constant in terms of the vacuum energy density in field theory you get something that’s several orders of magnitude — I mean far, far too large.

Misner:

Right.

Smeenk:

And I was wondering what your opinions are about that. Have you worked much on that? Or what’s your impression?

Misner:

No, I haven’t worked on that discrepancy between — not calculations but hand waving dimensional analysis in quantum field theories; and, yes, there is this dramatic difference. I think it’s probably much harder to solve that problem when there is a present day cosmological constant though I don’t regard it as a constant. I regard it as like the early inflation —

Smeenk:

Right.

Misner:

And false vacuum, some kind of strange equation of state or a weird form of matter that we haven’t noticed around, although it is supposedly 70 percent of the universe now [laughs]. The field theorists used to say well the problem is to explain why [the cosmological constant is] zero. It seems to me, a priori, that it would be easier to build a theory that would explain a zero than a theory to explain whatever it is right now.

Smeenk:

Right. And something that is transient, something that went through an early stage and now is important again.

Misner:

Right, yes. That’s also a big number. It’s not 10⁹⁴ or whatever it is versus zero, but it’s 10⁸⁰ versus 10^-10.

Smeenk:

So another question I wanted to ask you is just — and this may be related. You were referring to the Dicke number coincidences, and that is the question of is there a proper role for anthropic reasoning in cosmology? This is one of those things that causes perennial debates.

Misner:

Yes. I would think — we haven’t got anything better, so that I think the simpler anthropic arguments that say a variety of things would be inconsistent with our being here are reasonable arguments. They of course open up the question of whatever it means there being an infinite number of universes and only some of them get analyzed, because they are rich enough to produce analysts. I think the observation that certain conjectured initial conditions or laws of physics or something like that are inconsistent with human life evolving in the universe are useful remarks, but their deeper significance I don’t claim we have any insight into.

Smeenk:

Okay. Just a very general question again. I really liked the way you put the emphasis on being able to predict properties of the observed universe based on other features in the field equations or introducing a specific equation of state [in the 1968 paper]. So now its thirty-three years after that, and do you think we have come a lot closer to being able to predict the observed properties of the universe based on inflation and other developments in cosmology?

Misner:

Yes, I think there has been a lot of progress. I think so. I would like to actually see more speculation that could be in a form that was accessible to the general public about the future universe. Because we’ve got data on the past, of course, people concentrating on fitting their theories to the past, but I think part of it that sells it to the general populace is just the general feeling of awe and how complicated and beautiful the universe is and how unimaginable — previously unimaginable. Now people get these pictures drawn for them and they see what’s happened in the past, but I think there is not too great a difference between an extrapolation by 10¹⁰ in the future and 10¹⁰ in the past. And you get a little more feedback on the past from observations, but I think it would be a lot of fun to translate what limits there are on the future — especially in the kind of imaginative ways that Dyson started back in his paper on biology in the universe. And so I would be happy to see more of those kind of things thought about and analyzed.

Smeenk:

I also wanted to ask you a general historical question of how the field of cosmology has changed. You’ve been working in it for a time period that spans a lot of changes and developments and I was wondering how you feel the field has changed, maybe in terms of people working in the field, or in terms of the institutional support, or the amount of publications.

Misner:

Well, I think it’s an amazingly bigger field now. That is both in people and in data and in validly descriptive theories. One sign of that I particularly noticed is that the sales of MTW have increased in the last ten years. [Laughs] And I presume that’s primarily due to the fact that large numbers of astronomers who in the past could afford to have very minor or no knowledge of general relativity now feel that it’s an important part of their work is to have some insight for it. The change hasn’t been dramatic, but I think it’s a factor of two bigger now than it was ten years ago. For an old out-of-date book that’s quite something.

Smeenk:

That is interesting.

Misner:

Yes. But clearly there are huge numbers of — the Texas Conference of Relativistic Astrophysics attracts, of course, not only cosmologists but also [people working on] active galaxies, black holes and all kinds of things. But the number of people working there has changed dramatically, certainly since the sixties. And the corresponding amounts of data that are coming from things like various satellites, the MAP satellite going up next month, and all of this X-ray stuff about what’s going on close to black holes and so forth, that’s very expensive and involves large numbers of people and it’s producing very exciting information about the universe.

Smeenk:

So would you also say the status of cosmology has changed? Earlier you said that when you started working in the field that relativity was relatively isolated.

Misner:

That’s right, yes. No, by the time inflation was here, but somewhat earlier, it became clear that the mainstream particle physicists could not dismiss general relativity and gravity as being not of any relevance. And so it’s become mainstream. I mean, string theory every time it’s in the doldrums, which happens apparently periodically, it gets a big burst of energy every five or ten years, but in between times why the one thing they hang onto is they say nobody else can talk about quantum gravity except us. [...Short interruption by David Fiske regarding lunch...]

Smeenk:

I was just asking you about the status of cosmology. Are there any…?

Misner:

Yes. So the statement was that yes, now general relativity is an accepted standard part of physics, whereas in 1955 it wasn’t. And then probably only one physics Ph.D. out of twenty or fifty or something had a general relativity course by the time they got a Ph.D. Nowadays I think most physicists want to have studied some general relativity before they get out, and most astronomers. So it’s very definitely part of the mainstream now, and very definitely it wasn’t in 1955.

Smeenk:

And I also wanted to ask you if there were any — I mean there’s been a load of interesting observational and theoretical results in cosmology more recently. I’ve already asked you about inflation, and I was wondering if there was anything else that you would like to comment on. Or maybe your own work in numerical relativity?

Misner:

No, well that’s part of the preparations for LIGO and LISA. Those are also wonderful things. Not really cosmology, because it’s related mostly to black holes, and related to things like neutron stars and so forth… Of course, that’s what brought me to Maryland basically was the fact that Joe Weber was there, starting to think about gravitational waves.

Smeenk:

I didn’t realize that.

Misner:

Yes. The whole gravitational wave community realizes that he’s really the father of gravitational wave research. And I think the general feeling is that they regret that they didn’t give him more honors towards the end of his life, because he was so convinced that he had already seen gravitational waves that every opportunity to honor him would be turned into some kind of springboard from which he would preach this gospel of “we’ve already seen it,” which was widely rejected. Even the people who knew that they couldn’t produce LIGO and other things if he were given too big a platform to say “it’s not necessary because it’s already been done” recognize that the whole effort would never have been started if he hadn’t shown the world that you could take gravitational waves seriously. Before him nobody did. Einstein looked at them and dismissed them. So did other people. Said yes, they should be there but they can’t be measured, so stop thinking about it.

Smeenk:

So how closely did you work with Weber when you were [at Maryland]?

Misner:

Not terribly closely. No. I mean I saw him frequently and so forth, but I didn’t have I guess enough of a talent for real astrophysics to start… I mean, I should have when I first went to Maryland if I had more of an astrophysics talent and started doing theoretical surveys — what possible sources of gravitational waves there are and so forth, but that wasn’t my talent, so it was more Kip Thorne who did that kind of thing.

Smeenk:

So you’ve mentioned Dennis Sciama as one of your most important influences and then Wheeler as well.

Misner:

Right.

Smeenk:

Have there been any collaborators or colleagues that you’ve been working with more recently that you have either been really influenced by or that you have influenced them or just —?

Misner:

Of course another person in the past was this Arnold Ross, mathematician. And Abe Taub was a significant influence. He taught me a lot of things as a collaborator in a period of a couple years. And he also was a wonderful mentor to the Maryland student Vince Moncrief. Moncrief came from Maryland and he was a not very spectacular student in his first tries at research, and then he came into general relativity and did — pretty much on his own, but he did work that showed he really had a good brain, and so he was able to get a postdoc with Taub, and Taub brought out the best in him, he did a great job. But I mean, I got into numerical relativity because I wanted to, but the openings were to work with — because I thought it was important and it was a piece of classical physics that I figured I could do things with. I had worried about numerical relativity at Cambridge in 1966, I had started studying numerical analysis and things like that because Wheeler had seen computers being used in bomb work and every once in a while suggests when you’re attacking a problem, “Think what you would have to do to let a computer solve the equations, and then by the time you have formulated the equations well enough you will find you can solve them yourself,” that was the idea then. [Laughs.] So computers were just being used as… a discipline on your thinking. If you could explain it to a computer, you could probably understand it.

Smeenk:

Right.

Misner:

But it left me with the impression that computers would come into relativity one day and I did a lot of work. So I was happy to get back into it then. Matzner and Joan Centrella were very helpful and helped me to get started in that.

Smeenk:

You did some — was it computational work with Matzner on viscosity and dissipation, or was that all analytic?

Misner:

Well, my parts of it were analytic. He did some computational work. Yeah.

Smeenk:

Okay. Because that was in the early seventies.

Misner:

That’s right.

Smeenk:

So the more recent work…

Misner:

And then he did try some — Well, I guess he tried some computer algebra, and he was doing things at Maryland. And that was with a very rudimentary program. I’ve forgotten what it’s called even now. But anyway, it was a longtime predecessor of Mathematica, and Mathcad, and all these other things. So he had been introduced to computers but the numerical work he did was not with me. I don’t think there were any — I mean, if there were any computations in this paper he probably did them but I can’t remember if there were.

Smeenk:

Right.

Misner:

So and of course in this past year I’ve found a very beautiful set of people to talk to and work with at Albert Einstein Institute in Potsdam.

Smeenk:

And that’s where you’ve been for the past six months?

Misner:

Six months. Yes. And at Maryland of course there has been Bei-Lok Hu and we have had a very close personal relationship talking to each other about physics and things like that, but we’ve never actually published a joint paper or done a close collaboration in physics in that sense.

Smeenk:

Right.

Misner:

So I enjoyed working with Ted Jacobson and Dieter Brill at Maryland. A number of postdocs, Bill Stoeger was one that worked with me very early and has done things, Yavuz Nutku who was a remarkable, lively character, got off into things related to harmonic maps that I got interested in for a certain period.

Smeenk:

And also education in physics?

Misner:

Education in physics? Yes. That I put in ten years or something working on education in physics, some of them trying to think about research in the area and some of them just administration in the department. Influences there were Joe Redish in our department [Maryland], which now supports work in education research, and Bill MacDonald. So that was — so there were a group of people there I enjoyed working with.

Smeenk:

Okay, well I think that pretty much wraps it up for the questions I have at the moment.

Misner:

Great.