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 Charles Misner by Alan Lightman on 1989 April 3, Niels Bohr Library & Archives, American Institute of Physics, College Park, MD USA, www.aip.org/history-programs/niels-bohr-library/oral-histories/33955
For multiple citations, "AIP" is the preferred abbreviation for the location.
Interview covers Charles Misner's family background and childhood interest in science; influential chemistry teacher in high school; education at Notre Dame and mentorship with Arnold Ross; early interest in mathematics; encouragement of parents to go into science or to become a priest; graduate education at Princeton; work with John Wheeler on relativity and topology; introduction to cosmology by Jim Peebles in 1965; attitude toward the steady state model; Wheeler's preference for a closed universe; history of the flatness problem (the "Dicke paradox"); initial attitude toward the flatness problem; motivation for looking for mechanisms to isotropize the universe; the mixmaster model; history of the horizon problem; Misner's attempt to change the goals of cosmology from describing the universe to explaining it; Russian work on mixmaster type models; reaction of the community to the mixmaster model; change in Misner's view of the flatness problem after the inflationary universe model; attitude toward missing matter; problem of reconciling theory and observations with a flat universe; Misner's attitude toward the inflationary universe model; attitude of the community toward the inflationary universe model; attitude toward recent observations of large-scale structure; nature of the inhomogeneity of the universe; importance of Freeman Dyson's discussion of the fate of an open universe over very long time scales; role of visual pictures in science; relationship of theory and observation in cosmology; outstanding problems in cosmology; inflation, particle physics, quantum cosmology; ideal design of the universe; philosophy, science, and religion; necessity of the laws of physics; question of whether the universe has a point.
I wanted to start with your childhood and ask you if you can remember any influential experiences particularly any that got you interested in science?
I must have already been interested in science at the time, but the first thing I particularly think of was getting a prize in a science fair when I was probably in seventh grade. That was because I had been playing around with a chemistry set, so I had already been using my spending money from delivering magazines or newspapers to buy extra gadgets for a chemistry set. When my mother complained about the holes in my clothes, I did an experiment on how various kinds of cloth reacted to various kinds of acids.
I'm sure that didn't give her too much comfort.
No. So that means there was some scientific interest before that. My father was an electrical engineer, and he certainly taught me a little bit [about] practical electricity and fixing things around the house. I don't know at what age.
But you think it was around the time you were in the seventh grade?
The electrical stuff? It's quite possible. I can't remember a time when I didn't know how to repair a plug. I can't remember anything very early either. That's quite possible. I went to parochial school in 7th and 8th grade, and we would get sent off across town to a public school for shop and mechanical drawing and things like that.
Did your father encourage your interest in science?
Yes. My father and my mother were both intellectuals, readers, and so forth. So there was general encouragement, but I don't remember specific things.
From your mother as well as your father?
Oh, yes. She was a school teacher. She taught me first grade before I entered regular school, so I was a year ahead of myself in school.
Did you have siblings?
Yes. I had two younger brothers and an older sister. Many years later — in fact, the year I went off to college — I had still one more brother.
In addition to your experiments with plugs and electrical apparatus, do you remember any books that you read that had an impact on you?
At the grade school age, I don't remember any books. The other thing I played with was cameras. During third or fourth grade, I bought a camera. But, no, I don't remember any books. At high school, I can remember.
Okay, how about in high school. Do you remember any books you read then?
Probably the most important one was — I guess there were two. I dug up a college chemistry text and started reading that because I was interested in chemistry, and that contained an integral sign. I ran around trying to see if anyone could tell me what that was. Roughly, none of my high school teachers could tell me what it was.
Could your father tell you?
I didn't ask him. So I don't know. Probably by the time I thought of asking him, he did. But I think the chemistry teacher was a leftover from the depression. He was actually a Ph.D. in chemistry and couldn't get a job as a Ph.D. so he was teaching high school.
Was he a good teacher?
Yes, he was a good teacher, as distinct from the physics teacher, who was a good coach. He taught shop and physics and coached the football team. A year or two later, he went off to be a college coach. I don't remember learning any physics.
You said there was another book.
Yes. When I did discover what the integral sign was, probably from the chemistry teacher, then I started talking to my father, and he dug out the old calculus book that he had used when he was in college in 1910 or so. I started studying that.
You weren't getting that in school?
You had to do that on your own?
That's right. There was no calculus taught in high schools at that time. In fact, when I went to Notre Dame as an undergraduate, calculus was not a normal freshman course. The standard was what they called analytic geometry, and you didn't get calculus until you were a sophomore. But then I got into the hands of another mentor when I went to college.
I'll ask you about that in a minute. Did you have any concept of the universe as a whole in high school? Do you remember thinking about that at all?
No. I don't remember that. I was interested in chemistry. I was interested in electricity, which I didn't realize was physics. I probably owned a radio amateur's handbook and learned about theories of complex impedances and things like that, but I didn't know it was physics.
Did you know about imaginary numbers?
Yes. That probably came in our high school algebra [course], but it certainly came when I was reading the radio amateur's handbook.
Let me go to Notre Dame. You mentioned there was another mentor there. Could you tell me about that?
That was Arnold Ross. He was then the chairman of the math department at Notre Dame. Later, I think he became the chairman of the department at Ohio State, where he is now an emeritus professor.
Math, that's right. Another friend of mine, Tom Day, who I met at Notre Dame, actually was trying to read tensor analysis and Einstein's general relativity in high school. He ran into Arnold Ross and found out that Ross would encourage people to learn what they could and not take the standard courses. So I went and talked to Ross, and he got me out of analytic geometry and into a real calculus course. I was then a chemistry major; but, at that time, I got switched to the second year physics. They had a first year survey and would begin with mechanics in the sophomore year, using calculus. So I got into a sophomore mechanics course. In chemistry, in the first semester, [they were doing] qualitative analysis, which I didn't find at all interesting — mainly, I think because you rushed through the lab hours so quickly that you never had a chance to figure out what was going on. You just had to follow instructions, get results, and be on your way. So that was the end of chemistry. I never did any chemistry again, except fifteen years ago; I read an organic chemistry book out of general cultural interest. I knew I had a lot of talent in mathematics, but I decided that I preferred something a little closer to the real world, and I started in physics.
You mentioned that you had a friend, Tom Day, who knew something about relativity theory. Did he get you interested in relativity theory?
No. I heard a little about it. We were both two bright, young freshman going through calculus and sophomore physics together. No, I didn't take any relativity until I was a senior, and then I took a graduate course in relativity at Notre Dame and got a good, solid start in it.
Had you decided that you wanted to be a scientist by this time?
Yes. It was clear I wanted to be a scientist, but I thought I wanted to be a chemistry major.
Had you decided that you wanted to be a scientist before you started at Notre Dame?
In high school?
Probably by the time I was in grade school winning these chemistry prizes. By the [time of the] science fairs, I'm sure I knew I wanted to be a scientist. I don't recall that there was a serious question.
How did your parents feel about that?
They were quite happy. They would have been happy if I had been a priest. It was high on their priorities to have some children go into religion, but they were quite happy with my being a scientist. They were certainly able to put up with a lot. My dad helped me [with my chemistry work]. There was a big workbench he had built for himself that I took over. I then built a whole set of shelves to put my chemistry things on, which was partly learned at school and a lot was with help from my dad. So there were general signs of approval that they were happy to see me get involved with what interested me.
When you took the relativity course in your senior year at Notre Dame, did it have any discussion of cosmology?
I would think that it did, but I can't remember being struck by it. It was a one semester course, and it was just one of the graduate courses that I was taking at that point. I don't remember [any] cosmology. I don't remember cosmology playing a big role until rather well into my thesis work at Princeton. Even there, it was never the prime target of what I was doing.
Let me ask you about your work at Princeton. First of all, did you have any mentors there who had a big influence on you, as you did at Notre Dame?
Yes. I should mention that there were other people at Notre Dame. Arnold Ross certainly was crucial by getting me on the right track and leading me to take all kinds of fancy mathematics. Then, in physics, I also got sort of adopted by the department and given a part-time' job. There were a handful of people that I knew very well, and this one particular professor, Walter Miller, who I had for several courses and who was involved with the van de Graff machine there where I did a lot of my work. It was certainly a very valuable physics lab. There were other marvelous teachers there, like Bernie Waldman, who I took quantum mechanics from. I had really good courses there, so that when I arrived at Princeton, I felt like I hadn't missed anything. At Princeton, I had a summer job helping with the design of the proposed Princeton Penn accelerator, with Milt White and Frank Shoemaker. I worked with them for the summer. They were doing some fancy calculations in particle orbits of the "Johnniac" [von Neumann's] computer.
Is that what you helped them with?
I helped them with that.
The theoretical [work]?
Not seriously theoretical. I was basically a handyman. Computers were too important in those days to be used for anything that could be done by hand. The machine would spit out a bunch of results, following up a couple of sets of initial conditions for that orbit. Then to get the numbers you really wanted, you had to diagonalize a four-by four matrix or something like that. So I was doing that drudge work on a Marchand (mechanical calculator) and carrying papers around. [I also learned some of the theory behind the computations].
You did that for one summer?
I did that for one summer, yes, just to get acquainted there.
Who did you start working with at first?
I spent my first year with Art Wightman. I did some projects with him on double beta decay. He was working on quantum beta decay in a variety of ways. He gave me some [projects to do].
Were these theoretical projects?
Theoretical projects, yes. They were beginning to get to be serious theoretical problems.
You were calculating things in quantum mechanics.
Right. The only course I remember taking was Valentine Bargmann's quantum field theory. I am not sure [what else I took]. I took very little in the way of course work. They didn't demand a lot. You just had to pass the qualifying exam, and I had had all the standard graduate courses at Notre Dame. So I chatted around with both math and physics students, read books and studied in the library, worked with Wightman, and tried to prepare myself for the qualifying exams in the spring, which I passed comfortably. [There were also language exams. I'd had good French in high school, and my Notre Dame Topology course used Bourbaki's Topologie Generale for a text, so French was easy. German I'd found difficult at Notre Dame, so I studied for that by reading von Neumann's Mathematische Grundlagen der Quantenmechanik.] Then I began seriously worrying about [a topic] for a thesis.
Tell me a little about your thesis work.
That really quickly came down to a question of working with Wightman or [John] Wheeler. I may have taken Wheeler's relativity course by then. Basically, the deciding point was that Wheeler was full of enthusiasm, and he thought every idea you had was wonderful. He really set goals far off to solve all the problems of the universe, and thought it shouldn't be all that difficult. Whereas, Wightman felt very morally obligated to point out that his students had not gotten their theses done very quickly. There was someone there at that time who was finishing his seventh year on his thesis. The average graduate student took five years and so forth. So, although I thought he was doing very fundamental things, and I had a lot of math background for it, I chose Wheeler because it would be more fun and I would get finished quicker. I switched over to working with Wheeler.
What was it like working with Wheeler?
Well, he had regular meetings in his office of groups of people, the students he was working with, and we would discuss everything in relativity. I spent a lot of time in the library initially and went through the literature and made myself a card file of essentially [the work which was going on] in relativity. So I got myself acquainted with what was going on there, and I continued to study mathematics. Wheeler was interested in wormhole ideas. The previous year, when I wasn't studying physics, I had learned more mathematics. I had had point set topology at Notre Dame, and my Princeton roommates were mathematicians. The Fine Hall teas were still joint math and physics. I quickly talked to people who told me how to learn algebraic topology, which I learned a bit of.
Did you apply that to the wormholes?
That was the idea. I knew enough topology that I could do the mathematical [work]. We originally set to work on these wormhole ideas and [with] geometrodynamics pushed to see whether you could build everything in the world out of the zero mass fields, which were gravity, electromagnetism and, hopefully, some type of neutrino.
So the wormhole might be an elementary particle, [that is] a small wormhole?
Yes, if you could bring the quantum in. That was certainly the long-range hope. You didn't want to say, "This is it," because it hadn't gotten that far, but that's what we were looking for, to see if by adding topology you could get by without having to have matter be some additional substance besides the fields. John told me at some later stage that he spent 20 years of his life trying to say everything is particles — scattering, action at a distance, etc. — and then he switched and decided to take the other tack and say everything is fields. Of course, ten or more years ago, he switched and again and went on to say everything is logic or something else — getting into the fundamentals of quantum mechanics and other ideas prior to particles and fields.
Did he talk to you at all about cosmology at this time?
I don't recall that there was a cosmology interest in the first years. It seems to me we got involved with, initially, neutron stars, wormholes, and eventually black holes, before cosmology came into the game. That's my recollection.
Did cosmology come into the game while you were still at Princeton?
Yes, it did. [Sam] Treiman was interested. Treiman also knew about the black hole paper of [Robert] Oppenheimer and [Hartland] Synder. He mentioned that to me once, while I was working with Wheeler. I think he also mentioned to me this peculiarity of some molecular spectra that were showing signs later recognized as the three degree microwave radiation. He was aware of that as an anomaly. This was in the 1950s, long before [the work of] [Robert] Dicke and [Arno] Penzias and [Robert] Wilson. So those ideas popped up, and I was not interested enough in cosmology to jump on them and say, "Hey, I should think about this." [My guess is that my serious interest in cosmology (as contrasted to a by-stander acquaintance with it) can be dated from an American Mathematical Society Summer Seminar at Cornell in July and August 1965. I was on the organizing committee, and during the conference heard rumors of the microwave background. I called Peebles, got the story in outline, and then arranged to get the lecture schedule changed enough so that he could come and describe his work with Dicke (and the Penzias and Wilson observations). I felt immediately that this was a turning point in the study of cosmology. His talk already asks, "What limits can we place on the uniformity of the universe if the observations establish that the radiation is isotropic?"]
At this time, had you read enough cosmology to know about the big bang model and open versus closed universes and that kind of thing?
Yes. That was floating around. I don't know at what stage. I remember listening to Willy Fowler talking about the generation of the elements at talks at the graduate college, which were general cultural talks for the whole graduate college.
He was trying to produce all the elements in the big bang, wasn't he?
That's right. He was trying to produce the elements cosmologically. I can't remember when Dicke and [James] Peebles began getting into this stuff. They were certainly in [for] many years by 1965, when it finally broke.
But I think Peebles had not started [much earlier]. At this time, did you have a preference for any particular cosmological model — steady state versus big bang, or open versus closed?
I was aware of steady state and felt that that didn't sound like physics to me. There were too many [ad hoc explanations]. They had to invent something for which there is no other application, and it didn't seem to fit comfortably into the rest of physics.
You mean, for example, the matter creation?
Yes, the matter creation.
You didn't like that?
I didn't like that because, on the microscopic level, there didn't seem to be any way for it to [happen]. I think I was also aware that there was a cosmological constant floating around, and I would have been happy to see that be zero just on the grounds of simplicity.
What about the question of open versus closed? Did you have any inclinations there?
Not strong [inclinations], at least, I don't remember them being strong. I might have somewhat preferred the closed model. Wheeler quickly jumped on the closed model because he felt it solved the problem of boundary conditions if you required the universe to be closed. Otherwise, there might be too many solutions to the equations and too much ambiguity. Wheeler was very much committed to the closed model on aesthetic or philosophical grounds. I can't recall that, at that time, I had any preference. Dicke did come up with the "Dicke paradox." I think I remember his talking to me about that while I was still at Princeton, which means before 1963. The Dicke paradox being that the universe is too close to being spatially flat.
The flatness problem.
The flatness problem, yes.
Do you think he was talking about that as early as the early 1960s?
That's right. I have a distinct impression that I was wandering around the old Palmer lab building trying to understand whether it was fair to think about the universe in terms of initial conditions, the way you would think of throwing up a ball — because there are these constraints in the Einstein equations that play in that game somehow. So it wasn't clear to me that the energy could be variable the way it is in the analogous Newtonian differential equation.
So you thought that the energy may not be a result of initial conditions but some other [constraint]?
Yes, I just couldn't see how to play with those equations, and so I didn't come on board thinking that paradox was serious until the inflationary models came out. But, at that point, I developed a strong preference for the flat universe, feeling that the Dicke paradox suggested it.
Dicke didn't offer any explanation.
He didn't offer any explanation. The key point for me was [inflation offers an] explanation. Even if it's not the right explanation, it shows that finding an explanation is a proper challenge to physics.
Now you are talking about the inflationary model.
Now I'm talking about the inflationary model, so we're now in the 1970s.
Yes, but if you go back to when you first heard about [the flatness problem] — you said that might be the early sixties - did you take it seriously?
I took it seriously enough to worry about it for a while. I can remember it’s causing me to spend some hours chewing it back and forth. Then I guess I just dismissed it by deciding that I couldn't take it as seriously as Dicke did because I wasn't sure that the energy in the Einstein equations was a free parameter.
When you say it wasn't a free parameter, you mean that it might have been determined by physics or physical processes?
Well, somehow you seemed to have the choice of zero or plus or minus one. There didn't seem to be the continuous parameter, in my view of it at that time. I think later one sees that there is a continuous parameter, like the equation of the state of matter. You can run the universe in such a way to set these things up, but I didn't appreciate that. So then I put it aside. [My difficulty, I think, was that energy is an integration constant in Newtonian physics, while the analogous parameter here was some dynamical variable already built into the Einstein field equations.]
You mentioned a minute ago that you developed a strong preference for a flat universe. Was that during that period of time?
This was later?
This was after the inflationary universe. We can come back to that, but the two crucial features that fit together there are [Freeman] Dyson's views on the future of the universe, and the inflationary universe. Then we're up around 1980.
Yes, I want to come back to that. Let me ask you a little bit about some of your own research. Do you remember what motivated you, beginning around 1968 or maybe a little earlier, to look for physical mechanisms for producing the observed isotropy of the universe?
It was certainly the microwave background radiation and the feeling that before 1965, you had all the standard cosmological models, but their assumption of homogeneity was just because the data was good to 30 percent or something, so there was no use playing with more refined models. There was no use looking for irregularities when you can't even say anything about them. It was clear it [the universe] was not grossly irregular, but it wasn't clear that there was anything serious there. So, at that point, they [the assumptions of homogeneity and isotropy] were simply a mathematical simplification — saying that we're going to treat a homogeneous universe because [anything else] would be wasted effort in the absence of data. But, after the microwave radiation came out and began to show this [tenth of a percent] uniformity around the few arcs of the sky that were surveyed, then it became a serious problem. Things that you don't understand can be constant to 10 or 20 percent, but to one percent, it requires an explanation.
In your mind, could the explanation simply have been that the initial conditions were ones of isotropy?
No, that never seemed like an explanation to me.
I don't know. It's not clear that a postulate [of isotropy] at the beginning is better than a postulate [of isotropy] now. There was nothing to get your hands on. I guess it was like postulating the creation of matter. You can, of course, throw in anything that solves your problem, but it's got to make sense from more than one viewpoint before it's a real [explanation].
In your work at this time on looking for a physical explanation for isotropy, did you also group that problem with the homogeneity problem?
You thought of those together. So you were also looking for a physical explanation to explain the homogeneity?
Right, because homogeneity is necessary for isotropy. I knew that piece of mathematics. It was understood at that time. It could look isotropic if it was all carefully focused around us, but since Copernicus, we haven't been very happy saying we're at the center of the universe.
So it is the same problem once you accept the Copernican principle.
One of the things that I'm interested in is tracking the history of the horizon problem, if I can call it that.
Before you get on to that, there may be one other point. I really got deep into this when I went to Cambridge in 1966-67 on a NSF postdoc. When I went there, I was looking for fresh new things to do [during] that year. Peter Strittmatter and John Faulkner were playing with some kind of recent observational data that bore on this isotropy question, so I quickly ran off a calculation, which was the basis for a lot of this later work, saying, "Well, the simplest way to look at the question you're asking is to make an anisotropic universe, Bianchi type I, and see its effects for such observations." But I forget exactly what it was that they were looking at. I could probably dig out a notebook and find that.
But that was observational stuff?
That was observational stuff. It was not microwave radiation. I think it was a question of quasars or something like that. But anyway, the question was raised about whether the isotropy of the universe bore on this data, and I quickly produced a calculation using all the relativity I knew, and then I realized I had the tools to tackle this other isotropy problem. So I began to get interested in the general isotropy problem. [A fall semester seminar series on cosmology in 1966, in which Dennis Sciama's group and others participated, kept problems of cosmology continually in mind.]
Were other people at this time raising this issue as a problem — that you needed some physical explanation to explain the observed isotropy and homogeneity?
I don't know. I'm not aware of it. [The observational question, "Is the expansion isotropic and to what accuracy?" was in the air and voiced, for example, by Sciama, at seminars in the fall of 1966.]
So on the theoretical side; you weren't particularly influenced by anyone else? It was your own idea?
That is my recollection, yes.
I know that you very clearly stated the horizon problem in your 1968 paper.
No. I'm aware that there are other people that defined the concept of the horizon. [The horizon paradox is clearly stated in my 1969 Physical Review Letter. The 1968 Astrophysical Journal paper is concerned with explaining or deriving isotropy (without assuming it a priori), but does not identify the horizon paradox. Another paper prepared about the same time (June 1967) is a Paris talk, entitled "Relativistic Fluids in Cosmology," in Fluides et champs gravitationnel en relativite generale, (Colloques Internationaux du CNRS, No. 170, Paris, 1969). It shows an interest in horizons, but not a statement that those in the Robertson-Walker models are intolerable. In both the Paris paper and the Ap. J paper, my main qualitative point was to ask that physics not just find the cosmos consistent with the laws of physics, but also to (try to) show that no very different cosmos was allowed (or was plausible). I was trying to change the goals of scientific cosmology from describing the Universe to explaining it. The horizon paradox must have congealed in my mind in the spring of 1969. (The Phys Rev Letter was submitted April 14, 1969.) It does not appear in the summary of the Mixmaster solution (not so named before 1969 PRL) that won a Third prize in the 1967 Gravity Research Foundation essay competition, nor in my notebooks around November 1968, when Wheeler passed on to me some reports of related "mixmaster" work by Belinsky and Khalatnikov. Their work, like mine, was unpublished at this time. I presume that the horizon paradox arose in my mind that spring and provided enough excitement that I could publish a short description of it and the mixmaster solution (which I hoped would contribute to solving the paradox) in the short format of Physical Review Letters. The mixmaster model was later shown not to help the horizon paradox.] But no, I don't recall that there was anybody else pushing that problem as being important, an important thing for physics to tackle.
Yes, that is what I wanted to ask you.
As I say, I think I had heard of the Dicke problem before that, but I didn't take it seriously. After stewing over it for a few hours, I then pushed it aside and didn't worry about it at all.
Let me ask you for historical purposes whether you remember the reaction of the community to the mixmaster model? I guess that was 1969 or thereabouts.
Well, there are two models. The mixmaster is — the closed universe, and that essentially came up because I had done the fiat, Bianchi type universes first. Well, actually in my notes I probably did them more or less at the same time, but I wrote up the closed one much later. I probably worked it out originally just [because] once you've got a bunch of tools, you try to apply them. I had done [work] on the isotropy of the microwave radiation and how the Bianchi type I universes would affect that. I had all these tools, which I developed working with Abe Taub. He essentially taught me that you really could solve geodesic equations and things like that in these various universes. He had years before done studies of homogeneous universes, and we had a paper together on geodesics in the Taub-NUT model. That I did at Princeton while he was there (1962-63). I would have to look to make sure. But anyway all this differential form stuff, which I learned in mathematics, working with Wheeler on wormholes, is ideally set up to do homogeneous universes, and therefore, I did anisotropy. That got me into the mixmaster [models]. A little later came the question of whether the mixmaster would do some good on horizons. Eventually it did not solve the horizon problem. [The "Mixmaster" was a trademarked brand name for a widely used kitchen appliance (mixer). I applied it to the solution of the Einstein equations for the Bianchi Type IX homogeneous cosmologies. The 1969 Phys. Rev. Letter that stated the horizon problem also conjectured that the mixmaster solution would help solve it. That conjecture was later shown to be mistaken. The mixmaster label continues to refer to the Bianchi Type IX solution, whose motions are more like a bread kneading machine (not popular them) than a mixer.]
Well, did you initially think that it might help?
Yes. In a piece of that work — not the first time I ever did that particular solution of the Einstein equations, but by the time I called it mixmaster and was looking at it in more detail - my hope was that this kind of pancaking of the universe would wash out the horizons, which I considered a serious defect [in the standard model].
How did other people regard this work? Were other people also hopeful that this would solve a serious problem?
Well, I think Wheeler was. Wheeler was very excited about it and very happy about it. I don't remember other particular interactions, except with the Russians, because essentially they had done the mixmaster at the same time, and I just couldn't understand their work. It didn't come out that they were dealing with a closed universe. They were dealing with a special case of these equations that they had been studying, looking at the initial singularity. They really had a different way of parametizing the solution — which [was better for looking at] some of the interesting aspects of certain chaotic behavior.
I gather that they weren't specifically trying to solve the horizon problem with their work. Is that right?
Right. I don't think they had that in mind at all. They were trying to solve the singularity problem. There was a time when they were claiming that the initial singularity was not generic. That [claim] was [based on] imperfect logic, which said that we can find a solution with singularities, but since it is a very special solution, there is no reason that the general solution would have such singularities.
And their solutions oscillated rather than had a singularity, is that right?
That's right. No, their solutions all had singularities, but in the simplest cases they were simply one pancake, like the Bianchi type I. Then they generalized it to get one more arbitrary parameter into their solution and had essentially the mixmaster, where it [the universe] was doing this whole series of bounces that never actually became singular, although the sequence extrapolated to a singularity.
But they still had some initial singularity?
Yes. But they thought that the initial singularity was only because they had not got the last arbitrary function into the solutions, and that if you got the general case, rather than the one they were able to deal with, it would be nonsingular. Of course, at about that time the singularity theorems came out and proved that that was not correct.
You have already said that when you first heard about the flatness problem from Dicke, you didn't take it seriously because you thought the energy of the universe, and therefore the value of k, may not be a free parameter.
Did your view of the flatness problem change any as a result of the inflationary universe model?
Yes, I would say the inflationary universe model was crucial. It was not crucial that the inflationary universe be right. What was crucial was that the inflationary universe provided an example that turned the Dicke paradox into a standard physics problem. Here, by proposing a certain dynamics, you could solve that problem, explain it. Dynamics is an explanation to me, whereas a fiat that the universe starts out homogeneous [and flat] is not an explanation.
If I am hearing you right, you are saying that having an explanation of the problem made it a real problem for you?
That's right. It could be solved correctly or incorrectly, [but] once you have seen one example of an attempt to solve it, you could feel that, Okay, this is real physics, and if that attempt doesn't work, some other one could.
And once you decided it was real physics, you took it seriously as a problem that needed a solution?
That's right. Then I took it seriously as a problem that needed a solution. I would say even more than that. First, it allowed me to understand this question of whether energy was a free variable in the initial conditions, in the sense that it became free by changing the dynamical laws through the equation of state in early times. You can [therefore] arrive at some given time with different energies, either open or closed or whatever. Inflation seemed to show that there were "dials" you could turn in the equation of state that would lead to consequences that made this thing intellectually controllable, as distinct from being somehow built into the Einstein equations that the universe has to be closed, and therefore the total energy is zero. [Also, further confusion exists because' "total energy" is not generically defined for cosmological models.]
But you consider this dial that you turn in the equations to be more natural than the specification of initial conditions?
Of course, it could have been that the dial you had to turn was a very peculiar and special and fine-tuned dial?
Right. I understand people's concerns about fine-tuning, which is also a relatively recent [development] of the early 1970s. But fine-tuning somehow seems curable, in the sense that it is in the middle of a large and detailed theory, which has lots of places to modify it. [Something] like strings can come up. You prove theorems that say such and such things are impossible, and then someone discovers there is a new way around the hypothesis, and these things are possible. So, once you saw it imbedded in conventional theory, there seemed to be a lot of hope, whereas the simple fiat, not imbedded in theory, didn't offer you a chance to try it a different way.
You mentioned a little earlier that once the inflationary universe model came out, you became strongly attracted to a flat universe.
Yes, because now I take the Dicke paradox more seriously than the inflationary models. The inflationary models show that good physics probably is capable of producing an answer to this problem. The present models may not [explain] all observations and may need some changes. The elementary particle physics may be sufficiently different that the explanation will be sought elsewhere. But my feeling is now that I take the Dicke paradox seriously. What it says is: since this number, that is, the ratio of the curvature of space to the curvature of space-time, the ratio of three-dimensional to the four-dimensional curvature, evolves rapidly in the standard model, which looks good back to at least [the epoch of] nucleosynthesis — Dicke has got this very strong argument, which says something set that number very close to [one]. You know [that setting of the number] has to be a long way back, at least before nucleosynthesis and maybe way back to the grand unified [epoch]. Something had to set it. Therefore, you argue that whatever mechanism set this ratio to one, out to the eighteenth or maybe the forty third decimal place, had to be an exponential mechanism. We don't know of any number that says we're exactly that many decades away now from the critical time in the universe.
The mechanism had to set [omega] really very close to one.
Yes. And if it was going to set [omega] to forty decimal places, why not to a hundred? So, there's no particular reason to think that just now at the present age of the universe when man was evolved, this number is about to deteriorate, and we're going to fade away from being on this borderline. So, we're on the borderline for dozens or hundreds of decades of expansion into the future. That doesn't say [that omega is] one exactly, but it says it is one to a precision that we are not normally prepared to deal with. Either that, or there is something very seriously wrong with our cosmology, and something is dramatically overlooked if that number is not one right now.
If you think it is very close to one, then how do you reconcile that [belief] with the observational evidence that it is 0.1?
Well, I haven't followed the observations [closely] enough to criticize them, but roughly I have a feeling that the luminous matter is, of course, very small, and the [quantity of] non- luminous matter that is detected gravitationally keeps changing, depending on the scale on which you try to find it. That's just my impression.
If I were to tell you that most observers who work on this believe that the observed value of omega, including the non-luminous matter, is about 0.1 or 0.2 at the most, what would you say to that?
I would say, "Keep looking for ingenious ways to hide matter, maybe in exotic particles." I'm not sure what they include in their 0.1 — heavy neutrinos and all.
The [observers] include all matter that they have gravitational evidence for, whatever form it's in.
Yes, but it has to be clustered for them to see it, and so you just need some kind of matter that spreads out wider than the clusters they're looking at.
So you would tell them to find more ingenious ways to look for more dark matter?
That's right. Of course, I would also say to the theoreticians - the experimentalists sometimes have qualitative ideas that are more imaginative than [those of] the theoreticians — "Look for ways in which our picture of cosmology could be so dramatically wrong." That means whatever set this omega back in the past to very high precision, there must be a number in the universe that correlates the present age of the universe with the mechanism that set that.
Yes, if omega is not one.
If omega is not one. So, the mechanism that set it close to one in the old times must have known some number that correlates with the evolution of man.
If omega is indeed .1 today, rather than 0.99999.
Right. So, you either have to find that physical constant, that mechanism that, either by chance or non-chance, had some relationship to the present age of the universe, or we've got to find the missing matter.
What is your personal view as to which of these is more likely?
More likely, there is some invisible matter spread out wider than our present gravitational sensitivity to it.
Is that because you have a difficult time imagining the physical process in the early universe that could have known the age of the universe now, the epoch of man?
That's right. Yes.
So [your preference is based on] physical processes that you can imagine.
That's right. There are occasional coincidences in numbers that aren't all that coincidental — [for example], the size of a star, from G, h, c, and the proton mass.
But there is physics behind that.
There is physics behind that, and maybe such a simple bit of physics will predict the present age of the universe is [the point] when omega starts differing from one. But we haven't seen any hint of that. [There are] so many orders of magnitudes to play with. We are talking about mechanisms that may have operated at 1017 Ge V, and now we're at 10-13 GeV [i.e. cosmic background temperature of 1030 degrees K, now 3 degrees K].
So you think that the least radical explanation of this is some dark matter that is not seen, rather than that the early universe knew about the epoch of man?
And you base that statement on conceivable physical processes?
We've talked a lot about the inflationary universe model, and you have said a fair amount about it already. How speculative do you think it is? Is it something that you think is likely to be true in some version? You mentioned that it was very influential in allowing you to see that physical processes could determine some of these quantities. Do you remember how you first reacted to [the inflationary universe model]?
No, not specifically. I can't remember, [but] my guess is that I liked it. It sounds great. Then and now, it has this slight drawback that it relies on physics at unreasonably high temperatures. It used to be that particle physicists in the 1950s laughed at relativists, saying "What are you doing playing with numbers that don't make any difference?" In quantum gravity, [the distance scale is] 10-33 centimeters. Nothing is known below 10-13 or 10-14 centimeters, so it's nonsense to talk about those extrapolations. Of course, now the particle physicists have their own motivations to extrapolate back that far and are doing it almost as wildly as the relativists would, or more so. So, [I have] a little hesitancy to believe that you could sort of skip 10 orders of magnitude in energy with nothing interesting happening in particle physics and still get the right answer. Not that it's never happened in physics. Maxwell's equations extrapolate the thirty orders of magnitude from the proton to the galaxy with no need for modifications. So, it's not impossible, but one's a little cautious.
So you were interested, but cautious.
Yes, interested, but cautious. However, [I was] roughly convinced that something like [inflation] would probably work. If that's not the answer, then the real answer is probably not too different. It wouldn't surprise me if the real answer was inflationary, but that the actual mechanisms leading to the inflation might be different.
Why do you think the community has been so enthusiastic about the model? Do you think it's for the same reasons you were?
Yes. And the fact that the particle physicists began to realize — I think that [Alan] Guth has said this — that there is not that much flexibility in building models of the universe. If (people) were not in the game, they would look at [cosmology] and think that you can do anything. But then when they tried to start making the models themselves, they saw there is a lot more limitation. You don't have that much freedom. It is going to be a hard job to get [a model] going that satisfies [all the required conditions]. So there was that. [Cosmology] was not just pure invention, where you could throw out models everywhere and take them seriously. The other thing was that there is a nice long list of observational problems that inflation solved. It's not just that it [inflation] solved one thing, but that it solves 6 or 8 things. Some of them are statable in terms of the standard model, like the Dicke flatness problem and the horizon problem. Some of them are only statable in terms of higher levels of elementary particle speculation, like the monopole problem, and things of that sort. But there began to be a list of [solved problems]. I think [Andrei] Linde once made a list of a dozen [previously unsolved cosmological] problems, only two [of which] were untouched by inflation — namely the initial singularity and the cosmological constant (why it's zero rather than very large). So, there actually are data, and people began to appreciate that. I think that relativists knew of this kind of data. It's not the same as experiments. Its gross features, like the Olber's paradox of the dark night sky. It may not be hard to do that experiment, but it was remarkably insightful to say this means something, which we have to fit in with our theories.
You mentioned to me a few months ago that you consider the flatness problem to be in the same category of importance as Olber's paradox?
That's right. I think this time, people are taking it seriously, whereas the Olber's paradox was never taken seriously until [Hermann] Bondi and [Thomas] Gold used it as a pedagogical device to discuss cosmology. It 'was never taken seriously [before] that, because there was always an out. At the particular times when people [early scientists] thought about it, they didn't know enough thermodynamics, for instance, to realize that interstellar opacity would not solve the problem.
The universe would still heat up and become infinitely bright.
Let me ask you about an observational discovery. Do you remember when you first heard about the work of [Valerie] de Lapparent, [Margaret] Geller and [John] Huchra on the large-scale structure, this bubble-like structure?
How did you react to that when you heard about it?
I think by that time I was not active in cosmology anymore. So I was intrigued, but it was not something I was going to jump on professionally because I was not working in that area at the time. I had a general tendency that once things got away from the simple mathematics of general relativity to real astrophysics — like with black holes, once [the research] got away from idealized black holes to the accretion disks, X-ray sources, the plasma, and a lot of the detail — it just wasn't the kind of physics I was more efficient at than other people, and so I just left it to the experts. It's getting that way in cosmology. There are a lot of relevant details, like these large surveys of [galaxies]. Essentially because I hadn't played the problem of the early perturbations to produce galaxies in the inflationary models...
So you hadn't been in the theoretical counterpart of that [observational discovery].
Yes, I hadn't been in the theoretical counterpart of that, and so I read it in Physics Today and publications like that, rather than diving in and looking at it very hard.
I guess there is a whole group of observations of a similar character — the large-scale streaming motions, and so forth — that have showed that the universe seems to be more inhomogeneous and anisotropic than we thought, in terms of the galaxies.
Has that kind of idea altered your thinking about the big bang model at all?
Well, just in the sense that it leaves a big mystery as to how the microwave radiation can, on every scale looked at so far, be very homogeneous. I guess I haven't put the numbers down to say whether there is a flat contradiction, but that is one of the things that one would worry about. The other is simply that the inflationary universes do lead over to all these wild ideas of parallel universes and things like that, which now become mathematically not so wild. People using anthropic explanations or [ideas of] organizations of information throw around ideas of ensembles of universes and so forth. Whereas, the inflationary models make a mathematical model of our universe where our universe is embedded in, presumably, a bunch of parallel universes in a way that could be made somewhat clear mathematically and would have significance on a Dyson type of long future. If there is intelligence worrying about the universe 101010 years from now, they may be seeing what we now think of as parallel universes.
What does that have to do with this inhomogeneity?
It is a question of whether there is a simple homogeneity within which there are perturbations, or whether these is a sort of fractal organization of irregularities, which keep showing up on all possible scales, including those much larger than the [observable] universe. That sort of picture is, to me, an a priori plausible picture.
So the universe just could be a lot larger than the observable universe?
That's right. But then there is a serious technical problem of fitting that kind of over-broad picture of what things are like on scales larger than we can see and [the] spectrum of irregularities that extend beyond the observable universe, with the observational uniformity of the microwave and the irregularities [we see]. That's a technical problem that will be very challenging and very important. It's not at all clear that it will be solvable in anyone of these very broad frameworks that we just sketched.
So, other than reconciling the [homogeneity implied by the] microwave observations] — if I understand what you said — you feel that the apparently local inhomogeneities and anisotropies might be accommodated by recent thinking about either much larger universes, or parallel universes.
That's right. It suggests that you would want to make whatever spectrum of irregularities you are finding either cut off for well understood physical regions above a given size, or be part of a spectrum that goes wider than the scale of our present [observable] universe.
Besides these few examples we have talked about, have there been any other developments either in observations or in theory, in the last 10 or 15 years, which have had a major impact on your thinking?
I would say one is the Dyson article in The Reviews of Modem Physics on the distant future. I think that is a tremendously interesting and insightful idea, which is made more so by the inflationary universes that came out a couple years afterward. [Being] convinced that the Dicke paradox is solid and significant, I expect a very fiat universe, which means that I expect a very long future.
But you could also have a very long future in any open universe, without having omega so close to one.
That's right, but there was no compulsion to be in an open universe. Now there is a compulsion.
I understand. So calculations of the distant future have more relevance.
Yes, they have more relevance because I am more convinced that omega equals one now than I was about any definite value of omega prior to [the inflationary universe models]. So, now it looks like we should deal with that long future. I had some glimmering myself of these ideas before the Dyson paper — the idea that information doesn't have a strict energy requirement. [The quantum formula for energy], hv, says that you can have one quantum, one bit of information stored with very little energy if you're willing to do it at a low frequency.
Or over a long time, whatever that corresponds to.
Yes. Dyson put all these ideas together with calculations and imaginative ways of making estimates. I summarize his conclusion by saying that the universe will last long enough for civilization to rewrite its literature infinitely many times. I think he says in his paper that with information processing not only is the universe infinite in the simple sense of proper time — which I don't like as a good measure of things. Somewhere I've said I don't believe the universe is of a finite age just because the proper time age is finite. That's not the correct way to talk to the layman or the philosopher about the age of the universe. And the same thing for the future. Even though it's infinite in proper time, does that mean that infinitely many things can happen? Dyson answered that. He said, "Yes, if you play your cards right," which means evolving your species into a civilization that can exist at 10-10 degrees Kelvin.
So even though it gets colder...
Even though it gets colder and slower. You see, if it gets slow too fast, it won't be able to do much. Dyson precisely says, even though [physical processes are] getting slower, time is expanding in front of us so rapidly that, as I say, you can write your literature infinitely many times. I think that is a vision of our place in the universe that is very inspiring and very dramatic. I'm surprised nobody is paying much attention to it. It's not available to the general public at all.
I don't know whether the general public cares about what is going to happen in the very, very distant future.
They care about what was in the past. The big bang is on everyone's lips. If you have a talk or a public lecture on the big bang universe, you will get crowds. People are very intrigued by the beginnings of the universe. I'm surprised that this glorious picture of the future doesn't even attract physicists. Nobody has even bothered to rewrite [Dyson's] calculations since the inflationary universe has come in. No one has picked up on it and said, "Look I can play a little game here and there, filling in more details of what we might conjecture." I'm surprised there is so little interest about this picture of the future. I find it very intriguing.
Let me take just a slight detour here. One of the things that I am interested in is whether physicists use visualization in their work — whether you have to have a mental picture of what you are working on in order to do it. Do you use mental images and pictures at all in your own work?
I think so. I am much happier with things that are geometrical. There is a question of whether geometrical means pictures, but it certainly starts from pictures. I suppose when I think of a closed universe, there is a picture of a sphere in my mind. But there is also so much backlog of mathematics that one has worked through and can call up snippets of, that may be an important part of the picture also. In mathematics, I never got terribly comfortable and easy with pure algebra.
That really has no visual properties.
Yes, because somehow it doesn't have a visual lead-in. You get to these parts of algebra where they draw these commutative diagrams — half pages full of arrows going back and forth and somehow that...
That's pretty abstract.
It's very abstract, and even though it was a picture, it wasn't one that appealed to me as geometry. I've found that there is a kind of pleasure I get from working in geometrical terms — which are not always pictorial, but they somehow have a sense of pictures behind them — that is not the same as I would get in other areas.
Let me bring this back from the digression. Do you think that theory and observation have worked harmoniously in cosmology in the last 10 or 15 years?
Yes, it seems to me that observations provoke lots of theorizing. Of course, it's a different game from tabletop physics. The ratio of observation to theory is very different. As we mentioned before, people are beginning to appreciate some of these very qualitative things, like the absence of monopoles as important observations. So, yes. You have to be imaginative about what constitutes an observation to get anywhere in cosmology, because it's not as though we really want to wait until Dyson's 101010 years before we settle the questions. We would rather see if there aren't ways to be imaginative and bring some weight of evidence to bear [on these questions] in our lifetime. That evidence will not always be like the tabletop experiments. Some of it, I think, will be consistency relationships. It will be that all of physics ought to eventually hang together in one solid picture, so that you have one starting point and a bunch of correspondence principles that lead you down to more practical theories that fill in the details in understandable ways in one or another domain. But one would like to see it all put together, and I think some of that - how it fits together — will also be part of the argument. In fact, that sustained a lot of people who were working in relativity for years when there were no experiments. I think when you look back on it; you would say it was justified, that there were experiments. I guess I said this at the Einstein Centennial. The experiments were basically that Maxwell's theories and quantum mechanics could explain everything that goes on at the tabletop except why the apparatus was sitting on the table. Special relativity was in there, but if you wanted a gravitational field to stabilize things so that your experiment didn't [drift] away; you needed Newtonian gravity, if you refused general relativity. There really wasn't any alternative. And Newtonian gravity couldn't tolerate special relativity. So there really were experiments every day. The fact that you design a cyclotron without tying it down to the floor.
That showed you that there were other forces.
It showed you that there were other forces, and there really was no theory that had both the relativistic particles running around and the magnets maintaining their positions rather than drifting away when someone leans on them. There was no other theory except general relativity that explained that fact. Of course, it explained it very quickly and said this correspondence principle says use this set of equations for that calculation and the other set for the other. So people never dealt with the theory, but conceptually, it was the only theory that allowed both those things in the same lab. That kind of a thing will probably be better appreciated and, when available, will have to play [a part] if we are going to make of the decisions, based on observations, as to what's going on in cosmology.
Let me ask you what you think the outstanding problems are in cosmology.
The outstanding problems are certainly the large-scale problems. As I said. I haven't been close to the day-to-day problems for ten years, so I don't know those, and they are, of course, where the answers will ultimately come — just by checking the fine detail. But the broad-scale problems are to situate something like inflation within a more solid framework of elementary particle physics. So there are questions of really understanding elementary particle physics and not being quite so speculative when we make models of the very early universe. That includes, ultimately, all the questions of the Planck length and the quantum aspects of the beginning of the universe. The Hartle-Hawking idea that maybe quantum mechanics provides the initial conditions of the universe, allowing you to sidestep the singularity problem, is a very intriguing kind of thing. That should continue to receive attention.
When you say sidestep the singularity problem, you mean sidestep the necessity to specify initial conditions? Is that what you mean?
Well, they specify initial conditions. They say that the wave function of the universe should be non-singular in the Euclidian domain, where the path integral is being evaluated. It essentially provides a film of the universe run backwards, collapsing toward the initial singularity, and dissolving into something else, by essentially going off in another dimension, in imaginary time.
It is a boundary condition, isn't it?
Yes. But it is somehow a nice, clean mathematical boundary condition looked at within the theoretical framework from all sides, and it fits nicely. And there are all these questions of four dimensions. Are we going to make a theory of a higher dimensional universe which condenses into four dimensions of space and time plus a bunch of other dimensions that are essentially the graph paper on which gauge fields are plotted. As I said, I think extrapolating speculatively into the future is just as good a game to play as extrapolating to the past, although the past does have this advantage that there may be some fossils that provide you with a way of checking your speculations. It just seems to me that if you're going to run 20 or 30 orders of magnitude in your theory, because you believe that theory, you should run in both directions and see what it says.
Let me end with a couple of philosophical questions. You may have to put your natural scientific caution aside a bit. If you could design the universe in any way you wanted to, how would you do it?
I never have thought about designing the universe. I am interested in the question of the design of the universe. I have published papers on philosophy and cosmology and theology. I do see the design of the universe as essentially a religious question. That is, one should have some kind of respect and awe for the whole business, it seems to me. It's very magnificent and shouldn't be taken for granted. In fact, my tendency is to believe that is why Einstein had so little use for organized religion, although he strikes me as a basically very religious man. My feeling is that he must have looked at what the preachers he had known said about God and felt that they were blaspheming. He had seen much more majesty than they had ever imagined, and they were just not talking about the real [thing]. My guess is that he simply felt that the religions he'd run across were blasphemous and did not have proper respect, or proper dignity, for the Author of the universe that he seemed to be seeing. But then there is this other question of more straight philosophy. Is the design of the universe distinct from its enactment? Are the laws of physics necessary? I think in the time of the Enlightenment, people felt that if you think hard enough, it would be clear what the laws have to be — Descartes certainly [felt that way]. And the shock of quantum mechanics and relativity eventually, I believe, convinced people that even though Einstein did a lot with pure thought, that pure thought was not likely to be adequate.
Because we hadn't anticipated quantum theory and relativity?
Yes, right. [Physics] probably did need a nudge from experiment, however philosophical or theoretical you wanted to be, to put you on the right track. It might be basically because quantum mechanics was inconceivable in the 19th century. That is to say, all of human intelligence and culture had not had the necessary experience that these thoughts [could] actually appear in anyone's head. The whole evolution and the experience of a civilized culture was necessary before you could develop these ideas, and now we can think them. But before, I would say they were inconceivable. So we need these nudges. But there still is the question: could there be a different set of laws that was as good a theory of physics or was an acceptable theory? And is there a distinction between being able to conceive of the laws and having them actually apply to something? That ties into another little corner of philosophy, which is: Is there a distinction between the conceptual parts of our physical theory, which is things we can turn into mathematics like electric fields and wave functions. If we had enough of that stuff — if we had enough of the conceptual side - would it turn into matter? When we look at matter, we always keep taking it apart and finding out the inner pieces, but the inner pieces are never the explanation. The atom is not explained by the nucleus and the electron; it's explained by the Schrodinger equation and the electric fields. The nucleus and the electron are just little question marks and all the explanation is in the relationships among them that we put into our equations. Every time you resolve something new, the nucleus or the proton is no longer a single question mark; it's a nice, organized arrangement of parts. So, again, what we are able to deal with is the organization, the arrangement, the motions, not the parts. This all leads to the other extreme. Wheeler's field explanation, to say that everything is fields, can be done in geometrodynamics. You have pure Einstein-Maxwell equations. It's not clear to me that we can say anything is wrong with the Einstein-Maxwell equations, the classical Einstein-Maxwell equations except agreement with observation. They don't agree with observations, but is there some philosophical viewpoint in which they are improper physical theory? They have an initial value problem, they have lots of solutions. You could make models of the solar system by replacing each sun and planet by a black hole; they would run around and do Keplerian motions, so there is a lot of physics in this model. There are partial differential equations with existence theorems. As far as we know — and it remains to be proven — all singularities are hidden, so you could just cut off the black holes at the horizon and not worry about what goes on inside, and there's a perfectly good physical prediction for the subsequent future. So I don't know, on philosophical grounds, a check-off list for what constitutes a good physical theory. [Other than] comparison with experiment, is there any internal property that this theory doesn't have that a real theory ought to have? Maybe there is a distinction between a conceivable theory and an actual theory. If so, what can you say about the actualization? Do we just blame that on God or can we get more insight? My guess is we blame it on God and say there are things that we can think about but can't do.
If you were allowed to conceive of a theory yourself, or if you were allowed to build certain properties into the universe, what would you do?
I find the universe I see is always more beautiful and preferable to any I could have previously imagined — the more details I see of it. So in that sense I like" the present universe. If I wanted to put that into a phrase, I would say "a universe which is inexhaustibly intelligible," where you could keep understanding things and the game never gets boring.
That's a beautiful way of stating it. Let me ask you one last question. There is a place in Steven Weinberg's book The First Three Minutes where he says that the more the universe seems comprehensible, the more it also seems pointless.
Yes, I come down on just the opposite side of that. I'm saying how impressed I am with the beauty and intelligibility of the universe. We would have to get into a whole other thing about the meaning of truth, which I have written a little bit about, in the sense of saying that the only good solid things you know are the theories that are already proven wrong, like Newton's mechanics. We will probably keep using them forever; we know exactly what's wrong with it, so we know where we can't use it. Therefore, it's reached its archival, permanent stage. We can increase our knowledge that way. [But to come back to] Steve Weinberg's comment, I don't see it [the universe] as pointless. You might call Newtonian theory a myth in that we know what it's good for and we know its limitations. Well, it's not so much of a myth now as it was in Newton's time, when people were unaware of the limitations. In that same sense I think there are truths of religion which are real truths, but which are also myths — myths in the sense that we will not want to change them when we understand things more deeply, but we will understand things more deeply. [For example], Newton's theory was once understood and believed totally, and now it's understood and used and provides us with a grasp of nature, but we have some feeling that there are other things beyond it. My feeling is that in religion there are very serious things like the existence of God and the brotherhood of man and a lot of other stuff, which are serious truths that we will one day learn to appreciate in perhaps a different language on a different scale. We will probably always continue to teach them in the traditional ways — and think of them like Newtonian mechanics: you don't want to play baseball with quantum mechanics. There are a lot of things where you use the applicable correspondence principle theory rather than the deepest insight available. So I think there are real truths there, and in that sense the majesty of the universe is meaningful, and we do owe honor and awe to its Creator. With this Dyson future, I don't see anything wrong with imagining that civilization will succeed and evolve so that intelligent, responsible beings discuss physics or what comes after, long after the temperature has gone down and the heartbeat is once per ten billion years. The activity will continue apace and be more glorious, and we're part of it, helping to produce it. I think there is a lot of meaning in the whole operation.
That may be a good place to end.
 J .R. Oppenheimer and H. Synder, "On Continued Gravitational Collapse," Physical Review D, vol. 56, pg. 455 (1939)
 P.J .E. Peebles, "Microwave Radiation from the Big Bang," in Relativity Theory and Astrophysics 1. Relativity and Cosmology, ed. by J. Ehlers. (Vol. 8 Lectures in Applied Mathematics, American Mathematics Society, 1967).
 Note added by Misner: My memory is probably faulty here. In the 1954- 56 period, I most likely heard Fred Hoyle, defending the steady state theory by producing elements in stars and in supernovae.
 Note added by Misner: This probably was later, in the Fall of 1969, when I was a visiting professor at Princeton and attended the Dicke/Peebles seminars in Palmer Lab.
 The flatness problem was first stated in print by R.H. Dicke in Gravitation and the Universe, The Jayne Lectures for 1969 (American Philosophical Society, 1969), pg. 62
 A. Guth, "Inflationary Universe: A possible solution to the horizon and flatness problems," Physical Review D, vol. 23, pg. 347 (1981)
 F.J. Dyson, "Time without End: Physics and Biology in an Open Universe," Reviews of Modern Physics, vol. 51, pg. 447 (1979)
 C.W. Misner, "Neutrino Viscosity and the Isotropy of Primordial Blackbody Radiation," Physical Review Letters vol. 19, pg. 533 (1967); "Transport Processes in the Primordial Fireball," Nature vol. 214, pg. 40 (1967); "The Isotropy of the Universe," The Astrophysical Journal, vol. 151, pg. 431 (1968); "The MixMaster Universe," Physical Review Letters, vol. 22, pg. 1071 (1969)
 Note added by Misner: Faulkner and Strittmatter had plotted the 13 quasars (Nov. 1966) with a z > 1.5 on a blackboard and found them concentrated in two small groups near the galactic poles.
 See ref. 8.
 See ref. 8.
 V.A. Belinskii and LM. Khalatnikov, "On the Nature of the Singularties in the General Solution of the Gravitational Equations," Zh. Eksp. Teor. Fiz., vol. 56, pg. 1701 (1969); English translation in Soviet Physics-JETP, vol. 29, pg. 911 (1969)
 A.G. Doroshkevich and I.D. Novikov, "Mixmaster Universes and the Cosmological Problem," Astron. Zh., vol. 47, pg. 948 (1970), English trans. in Soviet Astronomy - AJ, vol. 14, pg. 763 (1971); A. G. Doroshkevich, V. N. Lukash, and I. D. Novikov, "Impossibility of Mixing in the Bianchi Type IX Cosmological Model," Zh. Eksp. Theor. Fiz., vol. 60, pg. 1201 (1971), English trans. in Soviet Physics - JETP, vol. 33, pg. 649 (1971).
 Note added by Misner: I now date hearing about the flatness problem as the Fall of 1969 in Princeton.
 See ref. 8
 C.W. Misner and A.H. Taub, "A Singularity-Free Empty Universe," Zh. Eksp. Teor.Fiz., vol. 55, pg. 233 (1968); English original in Soviet Physics JETP, vol. 28, pg. 122 (1969)
 V.A. Belinsky, I.M. Khalatnikov, and E.M. Lifshitz, "Oscillatory Approach to a Singular Point in the Relativistic Cosmology," Usp. Fiz. Nauk, vol. 102, pg. 463; English translation in Advances in Physics, vol. 19, pg. 525 (1970); See also ref. 12.
 E.M.Lifshitz and I.M. Khalatnikov, "Problems of Relativistic Cosmology," Usp. Fiz. N auk, vol. 80, pg. 391 (1963)
 See ref. 12.
 R. Penrose, "Gravitational Collapse and Space-time Singularities," Physical Review Letters, vol. 14, pg. 57 (1965); S.W. Hawking and R. Penrose, "The Singularities of Gravitational Collapse and Cosmology," Publications of the Royal Society of London, vol. 314, pg. 529 (1969)
 V. de Lapparent, M.J. Geller, and J.P. Huchra, "A Slice of the Universe," Astrophysical Journal Letters, vol. 302, pg. L1 (1986)
 See Ref. 7.
 C. Misner, K. Thorne and J. Wheeler, Gravitation, (San Francisco:. Freeman, 1973), pp. 813-4; first ref. 26, pg. 92.
 C.W. Misner, "Symmetry Paradoxes and Other Cosmological Comments," in Some Strangeness in the Proportion: A Centennial Symposium, ed. H. Woolf (Reading PA: Addison Wesley, 1980)
 S.W. Hawking, Nuclear Physics vol. B239, pg. 257 (1984); J. Hartle and S.W. Hawking, Physical Review D, vol. D28, pg. 2960 (1983)
 C.W. Misner, "Cosmology and Theology," in Cosmology, History, and Theology, ed. W. Yourgrau and A.D. Breck (New York: Plenum, 1977); "Infinity in Physics and Cosmology," in Infinity, ed. D.O. -Dahlstrom, D.T. Ozar, and L. Sweeney, Proceedings of the American Catholic Philosophical Association, vol. 55, Washington D.C. (1981), pp. 59-72.
 C.W. Misner, "Some Topics for Philosophical Inquiry Concerning the Theories of Mathematical Geometrodynamics and of Physical Geometrodynamics," in P SA 1972, Proceedings of the 1972 Biennial Meeting of the Philosophy of Science Association, (Reidel, 1974)
 S. Weinberg, The First Three Minutes (New York: Basic Books, 1977), pg. 154
 C.W. Misner, "The Only Reliable Theories are Fallen Theories," Conference on Foundation Problems of Physics, University of Texas, Austin (1984); see also first ref. 26, pp. 97-99; second ref. 26, pp. 66-67.