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 Jeremiah P. Ostriker by Alan Lightman on 1988 January 19, Niels Bohr Library & Archives, American Institute of Physics, College Park, MD USA, www.aip.org/history-programs/niels-bohr-library/oral-histories/33956
For multiple citations, "AIP" is the preferred abbreviation for the location.
Interview covers Jeremiah Ostriker's childhood memories of books on science; early fascination with being able to calculate things about the world; college education; influence of a Fortune magazine article on astronomy; reaction to steady state model; interest in chaotic inflationary universe models; role of belief in cosmology; education at Harvard and demoralization of students there; decision to go into astronomy; interest in a broad perspective of things; early thinking on galactic evolution and work of Beatrice Tinsley; dislike of bias and favorite models; Ostriker's style of doing science; work on explosive galaxy formation; different kinds of dark matter; history of work on dark matter; hostility of the community toward the idea of dark matter; how people destroy their scientific careers by inssiting too much on certain ideas; reaction toward the work by de Lapparent, Geller, and Huchra on large-scale structure; attitude toward the flatness problem; problem with extrapolating the universe back to close to the big bang; attitude toward the horizon problem; attitude toward the inflationary universe model and the community's reaction; importance of visual pictures in science; ideal design for the universe; importance of scientists of a broad background in the humanities and a large store of images; question of whether the universe has a point.
I wanted to start by asking you some questions about your childhood and how you got interested in science at the very beginning. Do you remember whether you talked to any particular people or read any particular books that interested you?
I actually can [remember]. But I think my interest in science was around zero. You'll see if you stop by our house this evening there's a portrait by Raphael Soyer...
I don't know Raphael Soyer.
It's a picture which was made when was I was around four. I remember it vividly because I remember the book my mother was reading to me. It was an elementary science book which had dinosaurs in it and the earth and what makes plants green and stuff like that. That was what she chose to read to me while I was sitting for a portrait being painted.
And Raphael Soyer was painting the [portrait]?
How was it that he came to paint your portrait?
It's a complicated story. Someone asked me recently because they saw the painting. I don't know. Maybe other people in my family know. It was something [to do with] a relative who had some connection with him who wanted the painting.
I think it was something like that. He wasn't that well known then.
Do you remember what the name of the book was that your mother was reading you? Ostriker No, but I can see it. I know what the cover looked like. It's a strange thing. So that just reminds me that I was always interested in [science]. Then, I think in high school, the first astronomy books I read were [James] Jeans and [Arthur] Eddington
Jeans and Eddington — their popular books?
Yes. What made a big impression on me was that from fairly simple physical principles which could be understood by a high school student, you could really calculate things which were interesting. I think it was that aspect, not so much... Most of the things that are written for students at that age now focus on personalities, because they think that's what they're interested in.
Probably a lot of kids are.
Yes, but what was interesting to me was how, with basic physics that I could understand, you could see that a star had to have the mass it had. You could calculate what the galaxy weighed. You could do all kinds of things sort of from first principles. In some cases you needed observation, but in some cases you could almost show — as Eddington, you know, attempted — that a star had to have a certain mass. That was very, very impressive to me.
You must have had a fair amount of physics background to understand why a star has a certain mass. The mass-luminosity relation, I remember, comes out pretty simply, but to actually get the mass either based on the Chandrasekhar limit or the upper limit from the radiation...
I'd have to look back. These were really popular books. There weren't lots of equations. I think there were some equations in them. But I'd had high school physics.
Were there any cosmological ideas that you came against at that age?
I don't think so. Not that I can recall now…
I don't remember the books of Eddington and Jeans myself. Do you remember the titles of the books? I know they both wrote several books.
Because cosmology at that time really meant something else. If you look at Jean's early book on cosmogony, it really talks about our galaxy and formation of the solar system, things like that I don't think [Edwin] Hubble and the expanding universe became the subject of semi-popular writing until after my formative years. So I can't think of when I first learned about all that. It was pretty late.
Other people in your generation have talked about Gamow's books and Hoyle's book, The Frontier of Astronomy.
I read Gamow. I didn't read Hoyle.
You did read Gamow. Did you read him when you were in high school, do you think, or were you in college?
Did he write one, two, three ... infinity?
Yes, he did.
I read that.
You read one, two, three... infinity?
Yes, and that made an impression on me.
Do you think you read it in high school?
Probably in high school.
You say it made an impression. Do you remember any ideas that you...
I don't remember the ideas from that book, but [I think it] made an impression on me. I wasn't sure I wanted to go into astronomy, until very late.
Was that in graduate school or college?
In college I majored first in chemistry because that was what was interesting to me in high school. Then I switched to physics. I had one astronomy course in college. If this is going to go in the archives I won't say who my instructor was. But it was so terrible that I had to petition to get out of it — it was a full year course — because whoever was teaching it didn't really understand such things as why the moon didn't fly away from the earth. This was at your esteemed institution. You may even be able to imagine...
The records will show where you went to college. [Laughs]
It was just awful. That put me off for a while. Then I went to work after college — I think partly to avoid the draft because there was a draft at that time — in solid state physics [for the U.S. Government] in Washington.
Before going to Chicago?
Yes. There was an interesting thing that was very influential to me at that point. There was a Fortune magazine article — strange to think that that was influential. I think there was a series at that time — "The Sciences." It had interviews and good pictures. There was a specific [article] on astrophysics. I remember Jesse Greenstein was in it and Chandra [Chandrasekhar] was in it. I don't remember right now the other people. I actually have it at home. I saved it.
That was between college and graduate school or somewhere around there?
About then. I have it at home so we can check the date. I know I hadn't gone to graduate school. I think it was in that one year. And Chandra, in particular, made a huge impression on me — just his presence.
His presence as revealed in the writing or in his photograph or both?
In the interview, in the photograph. I read his Stellar Structure the year between college and graduate school. It was a Dover book and it was cheap. So I thought, "Well, I'll just read it and derive everything in it." I think that was partly influential in my going to the University of Chicago. I picked that place partly on that basis.
Let me go back just a little bit, back to some earlier days before we go forward in time. Do you remember thinking at all about cosmology? You said that the kind of cosmology that Eddington and Jeans talk about was not what we mean by cosmology today. But did you ever think at all about the universe as a whole?
I probably did, but I don't think that there was anything very definite that I can recall. I had no serious preconceptions or even intense curiosities that I now recall. I may have had them. I remember when the steady state [theory] came out, I read that and I was not persuaded by the logic. I read Bondi's book.
You read that in the technical form?
I'm not sure which form, but Bondi has a little book. I can't remember. I didn't read it for any course.
Would that have been in high school?
It was later.
You were not persuaded by the logic, you say?
It might have been true, I thought But I remember his logic was that everyone postulates a big bang and that puts the breakdown in physics all at one point [in time and space], and there's nothing different about steady state; it just [has a breakdown in physics] at every point in space and time. I thought that was just a rhetorical trick. Of course, the steady state theory could have been true. It had a certain attractiveness.
What about it appealed to you?
Just as we're probably all brought up to think how foolish it was to have a man-centered universe, any scientist is brought up to think that anything unique about his place in space and time is probably an artifact of either his imagination or the measurements. As, for example, the Kapteyn universe, which is a good scientific model of the universe. But it's just [due to] dust obscuration. So it wasn't something wrong with people's psyche in that case. It's just dust obscuration. So, similarly, we're bred to have a bias against anything that shows things to be man-centered. Therefore we're at a privileged point because we know, in a big bang universe; we couldn't have existed in the very distant past or the very distant future. And that seems strange. So, there's some attractiveness in the uniformity of the...
That's a fairly sophisticated argument you're giving now. It's much more indirect than the Copernican principle, it seems to me.
But I think that did...
Yes, you were thinking that way in college.
Yes. But I don't think that people are less sophisticated in college than they are at any other point My own favorite model — which I've never done any work on; but, if I understand correctly, some of the chaotic inflationary models are beginning to look like — are models in which every individual [universe] finds itself in the big bang, but globally it's a steady state. You have universes sort of budding off.
This is Linde's stuff.
Linde's, yes. I met him when he visited Princeton and it was very interesting talking to him. He started out in philosophy.
Yes, I know. I interviewed him. He's in this project.
Aesthetically, that's a model I actually had long before. Of course, I didn't know the physics for it, who did? But that is, in some sense, aesthetically the most attractive model to me. I can't tell you why.
I have not thought that cosmology should be an area where there's belief, even a very [small] component. So when people ask me, "Do you believe in this or that," I find it hard to understand the question. "Do you at Princeton" — they say you, the plural you — "think that, believe that omega equals one?" I say, "I don't know, omega's a number you measure." "Don't you have any philosophical belief about it?" I say, "Absolutely not." So I have a kind of aesthetic preference sometimes for that, but I never seriously looked into it. My own thought on cosmologies is just do the measurements and find out.
I have some particular questions about that I'd like to ask you later. Let me ask you about your early professional career. You mentioned that when you went to Harvard you didn't have much of a positive experience with astronomy. When did you decide that you wanted to go into either astronomy or physics?
I had changed effectively to a major in physics by the time I had finished Harvard. Ed Purcell was probably the best teacher I had there. He made a big impression on me, as he did on many people. The physics department there was very cold, I thought. I was pretty discouraged by the time I left. One of the things that cheered me up was in my senior year I actually designed and organized a study, a poll, of all the physics majors at Harvard. I asked them various questions about how they liked majoring in physics at Harvard. The first fact was the tremendous fall-off from the sophomore year onwards in the numbers. The second thing that came across was how they had gotten discouraged with themselves and the subject and become sort of demoralized. People were going in very optimistic about how exciting it all was. The third thing was, to a person — I think it may have been to a man, I'm not sure if there were any women — they thought it was their fault. No one questioned the system and thought there might be something wrong with the way they were being taught or the social organization of things or the fact that students were never invited to teas. When you see all the data in front of you, you realize it can't be all the separate individual things about why this one didn't study hard enough, and that one had trouble with his girlfriend, and so on and so forth. You have to look and see whether it's something in general. It actually bucked me up a little bit. [Laughs] But I was interested in the subject of physics. I think it was in my year in between college and graduate school that I decided to do astrophysics because I had had some interest in it. I remember thinking - it was one thought which was extremely definite which I had at that time — I thought, "Well, I like mathematics and physics, and I'm going to end up making mathematical models for things because I'm pretty good at it, it's fun, and there's some demand for it. Twenty years later, I might be telling Colgate Palmolive how to put the toothpaste in the toothpaste tube most efficiently by mathematically modeling this, or I might be looking at the interior of a star or a galaxy. The equations may be the same, and the work might be intellectually stimulating in exactly the same way — to solve those equations and figure out how to do these things best. But, I don't know — one's more interesting than the other." So at that level I said, if I'm going to do the same kind of work I might as well do it in an area where the application is to something which has a little glamour for me.
Do you remember why astronomy had a particular glamour for you at that time?
It's something that's so obvious I can't answer.
Was it more glamorous, say, than particle physics or...
I don't think I considered particle physics. The fact that it was sort of outside of us and bigger and away from our messy lives that you could see all of... I did a lot of reading in other areas. I probably spent more time on literature than I spent on science. I married a poet. The fact that, as an astrophysicist, you get a perspective on humankind which is, you know, here we are...
Broader than the grocery store?
Yes. Sweating on this little grain of spinning sand that appealed to me. I'd always liked writers who took a cosmic perspective. Montaigne, who liked to look at the Europeans from the point of view of the Chinese. So the idea of having a broad perspective was very attractive to me. It still is. Now that argued maybe against biology and human sciences. It didn't particularly argue against particle physics.
When you went to graduate school, do you remember when cosmology became particularly interesting? Of course you've worked in many areas of astronomy. But do you remember when cosmology in particular attracted you?
I've kind of backed into working in that area. I worked in a lot of other areas first, and still work in some other areas.
So you don't think...
There was no decisive moment. I think it was a time of opportunity, when it looked like there were a lot of interesting problems. I can give you a very specific example which makes the point clearly. All the classical cosmology was done with the point of view of getting two numbers, as you know omega and H0 [the Hubble constant]. You need standard candles and so forth. If you found the standard candles changed, you had to make corrections in the cosmological calculations. There were huge programs everywhere studying all of these things. Beatrice Tinsley made the very simple contribution of pointing out that as galaxies age they become fainter. If you don't take that into account, you get the wrong value for Ho and omega. It could be seriously wrong. The correction is significant. She gave a lecture here at the Institute [Princeton Institute of Advanced Study]. We could probably figure out the year. She demonstrated, I thought quite conclusively, that galaxies would change, would get fainter by a few parts...
This was based on observations??
No, it was really theoretical.
Just that the population in an elliptical galaxy, if there are no young stars being formed — and we don't see young stars in them — then just at the rate the old stars die, they'll change a few percent per billion years and get fainter. If you don't take this into account and you use elliptical galaxies as standard candles, which is what everybody had done...
Yes. You get a systematic effect.
You get a systematic effect and you get the wrong [result]. She got an omega — I think it was something like [a value of] two — by just taking [the result of Alan] Sandage and making this correction. I remember thinking to myself — I'm not a cosmologist, I'd worked on dynamics of galaxies, and I said this at this talk — I bet I can think of several effects, dynamical effects, that can make the galaxies become brighter by that much. Which isn't to say that your correction isn't correct, but there are probably others as well, which will have the opposite signs. If this one changes it in one way, [they] will probably change it in the other. That was sort of challenged. And then [I] worked on it, with Scott Tremaine, who was my graduate student at that time. We worked out dynamical friction of satellites spiraling into galaxies — the fact that [the galaxy] M32 was falling into [the galaxy] M31 and the Magellanic clouds into our own Galaxy. You can calculate the rate at which galaxies will get brighter by their smaller companions falling into them all the time. Typically, it's a few percent per billion years and it goes the other way [from the effect that Tinsley calculated.] That's an example of how I got into it. I don't know cosmology, but I thought, gee whiz, if this is going to change the universe from being open to closed, I could think of something else.
Let's go back the other way.
Not that I wanted to make the universe either open or closed.
I think we titled the paper "Another Evolutionary Correction" because the presumption is that there are still others. But she had made the point. This wasn't criticism of her. She had made the point that the standard candles did change and you better pay attention. So this was more along the same line.
You just followed along the same line of thought.
Yes. But that's really an anecdotal example of how I got into cosmology. The problems were there, they presented themselves and then...
That was around 1975 or 1976?
Probably. That's probably the first thing I did which was cosmological per se.
Going back to the early 1970's, do you remember having a preference for any particular cosmological model, either open or closed or homogeneous or inhomogeneous, or any large feature such as that?
No. You know scientists have followed their own biases and my principle bias at the time was being contemptuous and intolerant of all of these people who had specific models. How could they be so certain when the evidence was as confusing and inconclusive?
Theoretical as well as observational?
Yes. Only a small class of models was being examined. If you look historically, almost all of the models at any given time that people have are wrong. So there's no particular reason why they shouldn't be at this time, and why should scientists be so stupid as to not realize this? I had made myself obnoxious in a small way by making these remarks. I still do that to some extent. There was a meeting on large scale structure, and in my summary talk, I said: "Well, on the basis of history, I think it's extremely probable that most of the models we're currently discussing are wrong. What one wants is discriminants which talk about broad categories of models, rather than attacking or defending this particular little one."
You think that's a mistake we're making now, focusing on narrow models rather than talking about broad categories? Or do you think things have changed in the last...
Well, if I can be slightly arrogant on this, I think science progresses because you focus on definite issues. You take the most predictable models and you work out all the details. You either falsify them or they're satisfactory and then you go on. So it does pay to do this. But, to some extent, it's something the troops do. It troubles me when I see people who I think should be slightly above the fray and [should be] thinking about the broader issues...
Going back in the trenches.
Yes. At least some people who have more perspective should be spending their time thinking about tests which will discriminate between Gaussian and non-Gaussian perturbations, as an example. Or between models in which initial conditions are important and models where it becomes a self-sustaining situation, where it is as star formation in the Galaxy. Just to take extremes. And to find out ways of classifying, in a broad way, the types [of models] and see if you can make predictions which will make the cuts and rule out large pieces. There tends to be a certain amount of "groupie" inclination in this area, [in which a large number] of people will be working on hot dark matter, or cold dark matter.
Yes, and omega equal one.
Yes. The belief — sometimes it seems to be the simplest way — it's the Manichaean logical fallacy of dividing things into twos. [It] seems to be easier for us and so you get to the point where you think if you can think of an argument against someone else's, then it's an argument for yours. Whereas, in fact, the truth is probably off in some other dimension [laughs] and neither one of you are in...
In good shape.
Yes. So I don't think I had any particular [preference] for any particular models early on. And I don't know. I've thought then and I think now that there are missing ingredients, essentially. You can't solve problems if you don't have all the pieces together right. There's no way of making the puzzle if you don't have all the pieces. Just to take an example of a current fad which, even though I'm working on it, I'm willing to admit it could be quite likely wrong — strings. I thought it was worth investigating because it was a new ingredient. If it turns out that models with cosmic strings are correct in some sense, i.e. that they're essential to understanding large scale structure, then no matter how hard you work without that ingredient — you'd have the right overall topology of the universe presumably, or could have gotten it — but you wouldn't understand the large scale structure. It may turn out that that's wrong, but some kind of decaying dark matter is a critical ingredient. So insofar as there's a real chance that some previously unexamined possibility from particle physics — or something which we haven't even thought of yet which will come up in a few years from now — is the critical element, there's no chance that any of what we're doing now is right. Of course, it could be that we now have assembled all the pieces. The opposite fallacy is the way the Greeks did it, where everything is always up in the air, and you're always at liberty to invent new hypotheses, throw out old ones, [and] not work out anything in detail really. So that's the argument for focusing on fairly narrow models. It's the way science has progressed in the past.
It allows you to rule them out.
Yes. To really understand them well enough to rule them out rather than always throwing in new ingredients. By and large, I think that's the proper way of proceeding rather than just invent new physics all the time — to see where the old physics gets you. But I just think it's gone almost too far in some cases of cosmology, where people are agonizing over the minutia of a very small fraction of the possible models where they have been casting their net. Whereas, even given current physics, a larger category of models [should be considered.]
Is this one of the motivating factors for your explosive galaxy model — your hydrodynamic model of galaxy formation?
Exactly. I was really thinking how stars form. They are real self-gravitating systems [like galaxies]. We don't lay it all on initial conditions. We don't say, "Well, when the galaxy formed there was a little blob that you could have said was going to be Sirius and another one which was going to be Procyon." The process of star formation involves energy output due to supernovae, which changed the circumstances enough so that basically things to a large extent forget their initial conditions. I thought of this argument a long time ago. I wrote a paper — I think it was 1975 or 1977 — to see if plausible explosions from one galaxy could reach out to the next one. There's a paper with Joe Schwartz, who's now at Harvard. If you couldn't fill the universe with these overlapping [shells], then there's not time enough, because any characteristic time you make for star formation is only one hundredth the age of the galaxy. You don't have a lot of time to affect the conditions. It [the interstellar medium] needs to cook. Whereas it's not obvious that that's the case in cosmology. At least I concluded [in that paper] that there is sufficient time, but not by a giant factor. It's by a factor of several to ten rather than a factor of a hundred. So it seemed to me therefore that some galaxy formation, however you start it, would affect subsequent galaxy formation and that one really should take this into account.
Was your motivation and thinking about this idea a natural consequence of your thinking about star formation, or was it a result of saying, "Let me think about this problem of galaxy formation and large-scale structure, and let me imagine all different other ways where this could have happened," as in the anecdote with Beatrice Tinsley. Do you remember which of those two, or was it a third, or a combination?
It was a combination of those two, plus one other element. The work that I did on pulsars led me to think about supernovae, when I worked [with Jim Gunn] on the Crab Nebula. Later I worked with Chris McKee on the interstellar medium in our galaxy. I'd learned how much explosions from stars really dominate the medium — determine its pressure, temperature, density, motions — to such an extent that past star formation in the form of supernovae tends to determine most of the properties. Not entirely, because UV [ultraviolet radiation] comes from early-type stars. But to a large extent it can determine the properties, of the medium from which new stars are made, and it seemed obvious that this would be worth exploring in the cosmological context. So it was that added ingredient which was sort of a straight line from other work I'd done.
I want to come back to that in a moment. I'd like to talk to you a little bit about how you reacted to some of the new discoveries in cosmology in the last ten years. Before I mention specific items, let me ask you just generally whether your own ideas about cosmology have changed any in the last ten years?
I wrote a paper with [James] Peebles on the size and mass of the universe, where we said omega was 0.1. That still seems to be all right. At that time it was considered very high because we were using massive halos to get it up to 0.1, which was about a factor of five higher than people had gotten previous to that time. Still I haven't seen any numbers for omega larger than that which are at all convincing. So in the question of how much mass there is in the universe, my ideas haven't changed very much. So it depends on how far back you want to go.
Would you say that, based on this, you think that you have an open universe, or do you not go that far?
I don't really go that far, because I can think of ways of separating out the baryons from the dark matter.
When you mean omega, do you mean just the baryonic omega [the contribution to omega from baryons only], or are you talking about the total omega?
Let me put it this way. This may be too technical for this purpose — but dark matter is used in two opposite senses, which is extremely confusing. It's used for matter which we know is there, because the only way we know matter is there is by measuring it [gravitationally] with G [the gravitational constant], dynamically.
But which we can't see.
But which we can't see. So [in this case], there's more, we know, than we can see. [Dark matter is also used] in the opposite sense, [in which] they say omega is one and the difference between omega being 0.1 and 1 is dark matter and that's matter which we've no evidence for.
Yes, of course.
So my conclusion is that the matter associated with galaxies, which has the same distribution as the galaxies — that corresponds to an omega of 0.1. I haven't really changed my mind on that for a long time. Now, is that baryonic or not? It's dark; a lot of it, but it could be dark baryons. I'm agnostic on that, because it is interesting that the best estimates for omega baryon are considerably larger than the best estimates of the stellar matter in galaxies. So that indicates to me that there have to be some dark baryons, a fair amount of dark baryons, either the low mass· stars or black holes or something. It's also true that omega baryon from light element nucleosynthesis is not that far from Omega dynamical as measured near galaxies. So an economy of hypotheses would be to say you've only one kind of dark matter and that's baryonic or former baryons.
And that would be consistent with all the observations.
You might have to stretch some of them a little bit, but that would be the most economical hypothesis. But now, if you take an explosive picture, then you have hydrodynamic forces which could push matter — baryonic matter — with respect to dark matter. Shocks will push it Electromagnetic pressure will push it So if they have been important in galaxy formation, then you've segregated to some extent baryonic matter from other kinds of dark [matter] — if there really is a non-baryonic kind of dark matter. Then you could have segregated [the various kinds of dark matter], and therefore the dynamical measures in the vicinity of galaxies would not be a representative sample. So, for example, the work I'm doing on superconducting strings. It's quite satisfactory in that case, if you want to have omega equal one, because the dark matter would still be quite smooth and the baryons blown into shells around voids. That works out all right I don't say it's correct On the other hand, there could be no dark matter in that picture. The theoretical physicists looking at this always seem to have a bias for omega equals one, [because] of inflation [the inflationary universe model]. But I know there have been other inflationary models, which said that omega must be less than 1. [Richard] Gott had one. And new inflation [modifications of the original inflationary universe model] — I would not be surprised if people did other things on that. So I'm just open-minded on the value of omega. Global omega. On the missing ingredient, though, there are enough inconsistencies, even without talking about large scale structures, to make me think that we're missing something in the global cosmology. For example, the ages of things really don't square very well. You can patch that by a cosmological constant.
Couldn't you also patch it by just saying that we don't understand the ages of globular clusters to within 50 percent or something like that?
Mixing length theory and all that business?
There are a number of different things there. It could be. That's one way out. But I once went through an exercise which I didn't publish. I used every argument I could think of for lambda [the cosmological constant] — observational not theoretical and tried to treat lambda as a parameter which is to be determined by observations. [For]every different observation I could think of which was a bearing on lambda, I calculated a value. Then [I] just looked at the array of these values. Are they all on one side of zero, are they all on the other side of zero, do they make a Gaussian [distribution about zero]? They all tended to be of the same sign. For example, they're all in the direction of giving you more space — more distance up to a given redshift. That's the way I think about it. And more time, up to a given redshift. There are many things that would work out well, if that were true. For example, the gravitational lensing gets better. So if you just treat [lamda] as an observational parameter, maybe there's a non-zero lambda. It strikes me as a large possibility, at least, which isn't very much examined.
There are certainly upper limits to it, based on dynamics. It sounds from what you've said that in the last ten years you've been sufficiently flexible and open that you haven't really felt seriously challenged, you haven't had to re-adjust your thinking dramatically. Am I misinterpreting...
That sounds as if I'm closed-minded, that nothing can make me change my thinking. But I'm trying to think of what have been the most important observations...
Let ask you about some particular observations.
Let's start with some of your own work. Do you remember the initial reactions of other astronomers to your hydrodynamic model of galaxy formation? When you first wrote the paper and talked about it, do you remember how people in the community reacted to it?
Well, it's probably best not to quote.
This is kind of a sociological question. You don't have to mention particular names.
I'll mention a positive one. The first and most significant thing that I think I've done [in this field] is the whole dark matter business and massive halos.
O.K., let's go back to that.
Or were we going to cover that later? Because that to me was more shocking.
Then I'd rather talk about that.
On the massive halos, there were two papers which, first of all, were confused in people's minds, although I thought they were fairly clear. One with Peebles said that the picture we have of the galaxy — inside the galaxy — must be wrong, and the spherical component must be a larger fraction of the total. It didn't change the mass of the galaxy at all. It just said that, if the galaxy's going to be stable, it can't be the flat disk that we "see" in other galaxies like [NGC] 4565. You have to have - whether it's a third or half - some significant fraction of the matter hot [in a halo]. As far as I know, I don't think anything major has happened to change that overall conclusion. All the models now have that. The other thing was to just look with what seemed to be a cool eye, in a non-theoretical way, at the accumulated observations of masses of galaxies. By various determinations, they really seemed to indicate that there are a lot of mass at large distances from the galaxies. As you know, [Fritz] Zwicky had said the same thing for clusters of galaxies. But the idea seemed to be at that point true on the larger scale and then [on the smaller scale] galaxies had a well-defined mass and a Keplerian rotation curve. And there might have been something wrong with Zwicky's [claim]. What we did in our paper was just to show that as you go to larger and larger scales in between these two scales, the mass just keeps on increasing more or less linearly proportional to the radius. At least for some distance, until you get up to mass-light ratios for galaxies which are comparable to what we find in the clusters — which indicated that, the total [extent] of radius and mass of galaxies was probably ten times the visible one.
And that surprised you?
It surprised me. But it just seemed to me what the data said. It wasn't a theoretical argument I had no axe to grind. The first thing was that people reacted to that with very strong hostility. I couldn't see particularly why. It was just a fact. The other thing is that there was what seemed to me willful confusion. For example, to mix up the two papers and say, "But it doesn't show anything. It just shows that more of the mass is in the spherical part rather than in the flat part." Or, for people on the other part to say that Ostriker and Peebles have shown that the stability of galaxies requires that they be much more massive, which we didn't show at all. So both the proponents and the opponents seemed to just not understand it. And the level of hostility was, I thought, fairly extreme. Not in my department. In fact, Martin Schwarzschild showed me that he had a rotation curve for M31 [the Andromeda galaxy] twenty years before and argued the same way that the mass is much more extended than the light, which was really a pioneering paper on that. But in some sectors of the astronomical community, I think: it's fair to say that it was greeted with hostility. Your memory probably goes far enough back so you can remember that.
You didn't have any ideas or any opinion about why people were hostile to this? Was it just any new thing or...
I think: it was partly [the reaction to] new things. In some cases it was people who were sort of sorry they hadn't done it themselves, I think…
When was this in time relative to Vera Rubin's work on the...
Oh, it was way before.
That's what I thought.
Yes. It was at least five years before, I think. Vera, in her early paper, I said she was confirming what we had done. For a long time — and I still think probably this was the best — Vera's [work provides] best evidence for it. But what we had used were binary galaxies, rotation curves of galaxies — a whole slew of different [things] –- satellites of galaxies. I think it was just the idea that if you're told that everything you've been doing is only...
Ten percent of what's there?
Ten percent of what's there. Then I found the reaction changed so it was a kind of a sea change from reaction with hostility to having forgotten and denying that we did it. I kept on running into people who would explain to me that I was probably too old to realize that things had changed and that galaxies had massive halos and things like that. Students would tell me this. [Laughs]
So it became accepted and they forgot that you were the one who…
Yes. Of course, obviously there were other people involved I think that was my first subject in cosmology where I ran into people's very strong feelings.
You think those feelings were based just on the body of work, people's own investment of their own work?
I think so...
Because it seems to me that [your] finding doesn't immediately challenge cosmological theories, although it certainly implies that omega might be larger than we thought. Even with your dark matter, you still didn't get up to an omega [of] one.
No. But it was very challenging for people who had been doing things. I remember there was one meeting where [Agris] Kainajs showed some rotation [curves] for galaxies and showed that you could fit them with just the matter that you could see. I think he had found four galaxies. But everyone came back from that meeting and said "See. We don't need massive halos. There's no more dark matter. At last we can get rid of it. It's all wrong." And there was a big sigh of relief. I think: he never published it, but I remember there was a mini-flap, so that if you follow the sociology of these things...
On the basis of one report, unpublished, people were willing to…
Yes. For a period. It astonished me. I said, "But this is just four galaxies. He hasn't published the data on which it's based. What did he really say?" But apparently he gave a very convincing talk.
It hit the spot.
Yes. But clearly people wanted to hear it, at that time. So that was one example. I think now it's fairly much accepted that there's a good deal of dark matter in the outer parts of galaxies. In fact, I think from. Scott Tremaine's latest work, it's probably true that we've overdone it slightly — at least for our own Galactic halo, it [the dark matter] doesn't quite go as far [out] as I thought. On the explosions, the reaction varied [Yakov] Zeldovich was extremely positive. Or rather, it was mixed. I heard from — I can't remember if it was [Rashid] Sunyaev or someone else who visited and said that Ostriker has killed our…
Pancake model. I was pleased that he had taken it so seriously. I didn't think that I'd killed anything because I had and still do see it [explosive galaxy formation] as an amplifier, essentially, even if things started some other way. And the origin of those perturbations could have been dark matter pancakes. But he at least took it very, very seriously. I think the prevailing view in the United States, in the West, has been benign neglect. "Oh, he's just carrying on."
Yes. I remember a colloquium that you gave at the Center for Astrophysics, where that seemed to be the tone. Of course, I think people take it much more seriously now.
Yes. So to some extent I've worked in the wilderness on this subject for some time. Fortunately, I do enough other things as well. I thought, "Well, it's either right or wrong, and time will tell." I'm not in an insecure position so I can afford to do it. I tried to be wary of a phenomenon I notice in other people's careers. There are many people who do a number of right things. They do something, it’s right and people recognize it and they're applauded and they applaud themselves, and they go on and do something else right. They go on until they do something wrong. Then, when they do something that's wrong — and, of course, in any career people do things that are right and wrong — people object to it, as they objected to things that other people had done that were right. Being so clever, they can defend themselves against these objections. But they're now defending something that is wrong. Then they get trapped into spending the rest of their career defending something [incorrect]. You forget everything else they've done because they've spent the last third of their career defending an increasingly untenable position. It's a scientific equivalent of the Peter Principle, but it's not their incompetence but their competence that does them in because nobody can find the inconsistencies and holes in the attacking arguments. I think with [Robert] Dicke and [the] Brans-Dicke [theory] and [Fred] Hoyle and [the] steady state [theory], which really happened. In the eyes of many people, their previous accomplishments were overshadowed, because they spent such a large fraction of their career on things which could have been right and were clever ideas, but which just didn't happen to be right. So with these cautionary tales in front of me, I've always made a policy of not defending against attacks. When, for example, [Geoffrey] Burbridge wrote what I thought was a badly argued, ill-intentioned attack on the massive halos of galaxies...
You just let it go?
Yes. For example, he said it's a mistake using the satellites of galaxies, because they're not where we see them now. They've moved in further, they're tidally limited. We had, in fact, explicitly included that factor in our calculation. I just thought, "Let it go. Don't defend it." Because otherwise I'm spending my time defending against attacks. Just go on, do something else. So I've wondered on this explosive thing whether I'm doing the same thing. I think as long as I continue working out the aspects of it, rather than defending against attacks... And if I see something wrong with it, I'll just drop it because I don't feel that I have that much invested in it for the present point, it keeps on showing up surprising things that are attractive. We certainly did predict that things would be on bubbles.
Let me ask you about the bubbles. Do you remember when you first heard about the work of de Lapparent, Huchra, and Geller? Do you remember how you personally reacted to that?
I thought, "Oh, that's nice." That was it. It was not conclusive because in fact you can get things like [what they observed] in other ways. But it's certainly good news rather than bad news.
Good news for your general view of things?
Yes. At least it's consistent with the type of things [I had been working on]. I have noticed that the approach to the explosive amplification picture has changed from completely ignoring it to putting in a footnote in papers — [for example, saying] whereas we've disproved this, it might be consistent with the explosive origin or things — to people considering it among the possibly viable alternatives. And that's as much as I would say. So I think it really should be looked at. My guess is that what will really turn out is more complicated I can, incidentally, see a shift now that people are beginning to look at hydrodynamic effects in galaxy formation. Lots of people. Once they start doing this, they're going to rediscover all of these things. They'll come in a slightly different context. But I guess that the parade will move on and much of what I've done will be essentially incorporated. Hopefully the true parts, not the false parts.
Right. Maybe some of the more detailed calculations will carry it forward...
Yes. I'm pursuing a lot of them now at Princeton.
Let me ask you about a couple of theoretical ideas. Do you remember when you first heard about the flatness problem, which I think sort of electrified the community first in 1979 in the article by Peebles and Dicke in the Einstein Centenary volume. At least that's when many people claimed that they first heard about it. Do you remember when you first either heard or thought about this problem yourself, and how you thought about it?
Now, let's see if I understand the problem. That's the [problem] that if omega is 0.1 [now], then it was very, very close to one for a very long time.
Yes. A very, very long time ago. I think in Peebles and Dicke's article, they say that for omega to be as close to 1 as it is now, when the universe was one second old — and they only went back to one second, not to the Planck time — that the kinetic and potential energy had to be balanced to one part in 1030 or whatever —
I guess it's one of those things which I thought I always knew, but I couldn't have always known. It must have been true — since I'm here and Peebles and Dicke are here — that probably they mentioned this to me long before they wrote it up in 1979. It's conceivable that Jim Gunn told it to me as well, or that I thought it up for myself. As far back as I can remember, I've known this argument. The most likely thing is Peebles mentioned it to me sometime, and I said, " Oh yes, I guess so."
It may have originated sometime in the early 1970's. It's not as old as the horizon problem that Rindler discussed first in the 1950's. Anyway, the most important thing I want to ask you is: when you first began thinking about [the flatness problem],did you take it as a serious argument, in the sense that the universe requires a physical explanation as opposed to random initial conditions that produce an omega close to one. How did you think of it? Did you take it seriously as something that demanded an explanation?
It didn't demand an explanation. I thought it was a good argument, but the idea of extrapolating our present universe back to one second is not all that convincing to me. I thought then, and I think now, that we can extrapolate it back to a few minutes because the light element nucleosynthesis in the big bang works out all right.
Even if you go back to a few minutes you're going to find a very delicate balance between...
Sure, but it's not a smaller parameter than lots of other small parameters that you see around. It didn't force me to — I thought it was an interesting argument. But it's not like a mathematical demonstration. You can always think: of so many things... Does omega go as one over one plus z [redshift parameter] in the radiation dominated era, or is it — Do you remember?
It's omega minus omega-critical over omega. I'm not sure, but I think it's one over one plus z. It's a power of time and —
Yes, but I'm just trying to remember which power. But what I remember thinking — I once worked it out — is back to nucleosynthesis, which is as far back as I think our extrapolations are tested, it's a small number but it's not unbelievably small. Let me give you another example from cosmology just to make the point. The fundamental thing people believe — and people have tried since Einstein — [is] that they can calculate everything from G [the gravitational constant], h-bar [Planck’s constant], and c [speed of light]. The way I think about that is this. On one side of the equation you have the electron mass, which you're going calculate, but it's not a primitive number. On the other side you have the mass unit that comes out of the Planck mass. In between you have, what is it, [a factor of] 1019?
It's 1019 GeV versuso striker: An MeV.
An MeV. So 1022.
1022. So you have to calculate then, by pure thought, a dimensionless number, which is 10-22. It's not zero. It's 10-22. And we say that anybody who has ever thought of making elementary particles out of only those things, believes that nature contrived to make a dimensionless number which is 10-22.
So you think that in the same way nature could contrive to produce this very striker I'm saying we absolutely assert that it has produced this. I think if you're going to have a fundamental theory which is only h-bar, G, and c and are going to calculate the mass of the electron, then we absolutely assert that nature can contrive a dimensionless number of 10-22. I've talked to Ed Witten about this. He says that kind of thing happens you have e [the exponential constant] to minus a big number. So if that's the case, on something as fundamental as that —
You see no reason why nature couldn't have contrived [to form the universe with a very fine balance between kinetic and gravitational energy]...
Yes. I think you can invent inflationary models which do exactly this.
Oh yes, of course. But the flatness problem was before inflation. It was one of the things that motivated [Alan] Guth to think of the inflation model in the first place because he wanted to find some physical way of explaining this. I guess the argument is the most compelling, at least when I've heard it, if you are willing to accept the fact that the universe could have been made with lots of different initial conditions, and that the particular initial conditions that made our universe are a set of measure zero, and that seems improbable.
But unless there's a good physical theory... in other words, the probability of an electron having a mass of 10-22 of the Planck mass is off-hand pretty small.
Unless there's a physical theory.
Unless there's a physical theory. Maybe there's a physical theory [in which] — even if I don't happen to know it this time — I actually just compute it In Linde's theory, or somebody's theory and that's the way it turns out to be.
Yes. Right. I think this argument that there should be a physical theory is what motivated Guth.
OK. I didn't find the fact that a number is small, therefore it has to be zero, an absolutely compelling argument. That's my point. Because I didn't see why we couldn't have a small number. It may be difficult for me to imagine calculating it, but since I can see that we're trying to do that in other domains, this didn't seem to be any worse.
I guess the question that I'm asking is that some people react to this argument as saying that it didn't require a calculation at all, that the universe is as it is. We go out and measure it and don't ask questions about why it is as it is. A lot of people took that to striker… I didn't take that view. For example, I argued with Peebles for years on the horizon problem on that.
Did the horizon problem bother you?
It always bothered me.
OK. So this bothered you in the same way then?
Yes, but I assumed that a physical theory would provide an answer to it.
OK. That's the thing that distinguishes the different points of view on this.
If you have some very special conditions, I would presume that it's determined by a well-defined physical theory, rather than just being a coincidence that things on the left and the right are the same. That [coincidence] strikes me as being very unaesthetic.
Yes. That's what I was trying to get at although I didn't ask the question very well.
OK. I hadn't understood the question.
I didn't ask it very well.
Let me put it this way. I've always assumed that omega is determined by a physical theory, not by "initial" conditions.
How did you react to the inflationary universe model when you first heard about it?
I thought it was a clever idea.
Why do you think it caught on so widely and so quickly?
I'm not competent to say. I think people like things that come from particle physics. And there were real problems. I could see there was a problem that it addressed. The other thing, I think philosophically, is we don't like singularities. In physics we're taught not to like singularities. If you get a solution which has a singularity, you throw it out. You're taught that in college physics. You keep the one without the singularity. So therefore, insofar as it addressed the singularity that was built into Big Bang theories, [and] made a little more complicated and plausible picture of it, it was attractive. So I think that aspect of it — the flatness and the horizon [problems] — it was addressing questions that in the standard cosmologies were just left hanging there in a little bit of an embarrassing way. It seemed to address them all at the same time.
One of the things I'm interested in is how astrophysicists or physicists use visual images in their work. Some people seem to use visual images or metaphors a lot and other people don't use them hardly at all. Do you use visual images a lot?
Yes. There was a volume in honor of Stan Ulam, who's really a mathematician. He commented — he'd spent time with Fermi on this — that physicists always think in just pictures first I think there's nothing I've ever done where I don't think that way.
You think that helps you work on problems?
I wouldn't even make it as weak as that. It's essential. That's what I do. In fact, [in] every single thing I think I've done I can first just close my eyes and think about it until I can sort of see what happens, and then I can work out the equations later. I mean it won't always prove that I was right. But then I work out the equations and see whether it comes out. I remember the first thing I did that was of any interest was on this question of a mass limit for white dwarfs, which was in the early 1960's. I argued that if they have finite angular momentum — not zero — then there's no Chandrasekhar mass limit. If they have finite angular momentum. As long as it's not zero. Because they're always contracting, neglecting general relativity for the moment. They will just contract until rotation balances gravity. It took me a long time to make all the computer programs work, to get the [models for] rotating white dwarfs. But it seemed obvious to me that that had to be the case. Obviously, there are many other complications like stability and this and that, which I wasn't aware of when I was first working on it. I think that that's just the way my mind works. I visualize things first.
In the last few minutes, let me ask you to take a big step back and maybe think a little more speculatively than you have, and ask you a couple of questions along those lines. If you could design the universe any way that you wanted to, how would you do it?
Well I like Linde's ideas, insofar as I understand them — where you have lots of firecrackers going off. It's globally steady state, but locally everyone finds themselves in a big bang universe. I don't know why that is aesthetically appealing, but it always is. If you can construct such a thing...
So you would make something like that?
Yes. I like things which are hierarchical also.
Things with structure...
At all levels.
At all levels.
It's less boring. The steady state always seemed to me an extremely boring universe…
So [Linde's model] has some of the attractive features of the steady state but…
But doesn't have the boring features.
[It has] the excitement at the same time.
It's like, would you rather live forever or go through cycles and be different people at different times and have many different lives? I think most people would say that they'd rather have many lives than live as the same person forever.
I was talking to Carol Rigolot yesterday and she mentioned to me that you had said in your course that you thought that theorists — maybe I'm mis-stating what she said — but she said that you said something along the lines that theorists who have a broad background in the humanities and literature and so forth have a deeper reservoir from which to produce new ideas. Could you elaborate just a minute on that?
I didn't say [exactly] that. I said something that's close to it and that has struck me as something very interesting. [Where do scientific ideas come from]? This is probably as good an excuse to talk about that point as any. It struck me as uncanny how many ideas which we've had on a mythological literary basis [that] have turned out to be right. In some crude sense, right. An example is the Lucretian idea of atoms. [It] is the idea that we're still with to this day. You can ask what evidence did he have for it. None. I can't think of any evidence. It's turned out to be an extremely useful idea just way beyond what anyone would have thought, even though we've had to go on [to further] levels and levels and levels and we've never found the ultimate particle. Another example could be that Judeo-Christians — and it goes back to the Babylonians –- have a big bang cosmology. We can think of many examples, including literary ideas, which existed for black holes in the past. I think you could find a number of examples of such phenomena. Now there are many different possible explanations for [this unexpected scientific correctness of mythological thought]. One is the Nostradamus effect. When things come out right; you notice it, when things come out wrong —
Yes. I was thinking, when you mentioned the Lucretian [view,] that the Aristotelian view - which lived side by side with that — was very different. Even though he had his four elements, he had a very different idea about vacuums and non-vacuums — very different from the atomistic view of Lucretius.
And you could say that one failed, so you pick out the ones that are right [and remember them]. But my argument against that is that the a priori probability of any completely ad hoc idea being at all correct is so tiny — it's like [how] mutations work — that even if we notice them, it's amazing to me that anything the Greeks said on the basis of no evidence would be such a powerful construction of an idea.
Maybe there are only certain ways that the human mind thinks…
Well, let me come back to that. So on the Nostradamus fallacy; I think that's one explanation. But I don't think that it is sufficient I'm just trying to think of all possible interpretations. Another possible interpretation is Fred Hoyle's panspermia [idea] that there was a race of very intelligent beings and they understood all of these things and we come from them. So it's built into our genes — all the answers to these things — because they were previously known. I think in the past you could have said that was probably absurd, because — you know, Jungian memory is impossible — but we now know that animals do have genes and DNA can control really a lot. I just don't think that this is too plausible. So let me ignore that one. Another one — which the sort of deconstructionist approach would take — is it's all just metaphor. There's no "reality" there anyway. We just pick and choose among these things on a basis that's more or less fashion. Lucretius isn't any more correct than Aristotle. The atomic theory of matter isn't any more correct than the phlegm theory. But the current state of science adopts this model or that model. [This picture is], I think, a false reading of Kuhn, which is very common and [consequences] that have followed from that [false reading]. I reject that. I'm a realist I think, you do the experiment and the needle points.
And that it's real.
OK. But that's a possible interpretation. Now, having rejected all of those, I may not come up with the right one, but here's the one which I've thought of. It's an appealing one and it's very open-ended to me. We have a store of images, which I don't think are unique to a civilized person, the literate person. Probably the literate person has more of them than the less literate person. But we absorb them somehow or other and they provide the basis very often for the models which we're willing to explore and entertain. Some of them we can falsify and some of them survive. Now what that means implicitly is that, had we come from another cultural tradition which had other models in it — not that we would disprove things that we now know to be true — but, rather, there are other phenomena now which we're not even investigating but which are quite important, which we would then be able to address. Because you just don't even address things if you don't have the mental pictures for them. And as our cultural experience becomes broader, as more different cultures are brought essentially into the world of science —
You get a larger store of these pictures.
Right. And we may be able to address problems which we've just ignored. Because I'm continually struck in the history of science by things which are perfectly obvious, but which couldn't be seen even though they were perfectly obvious. You know the famous example of the Crab Nebula. It wasn't seen anywhere in Europe. They have no records of it. The monks have no records of it. Many other cultures, including American Indians, Chinese etc. have them. That was because for the Europeans, there were "fixed stars." My guess is that if you have an idea of fixed stars and you see something which isn't — you just assume, "Well that's some atmospheric phenomenon or it's just..." You ignore it. The monk doesn't put it in his day book. And so you can't study things because in some way you reject the reality.
And you're almost never even aware that you're rejecting the reality.
Right. So when people ask about extra-sensory perception, things like that, I always think that there are things that we think are part of our senses now which would have been extra-sensory perception fifty years ago. One example is that they now know that people can tell when other people are looking at them, in part. I put a screen between you and me, and I turn my head this way and this way and this way, and someone tests whether you can tell. First of all, it's statistically significant that you often can, more than random. Second of all, it depends on how far away we are, which is good because most of the false effects don't [scale with distance properly]. Now I think there's a simple explanation [known]. We all give off a lot more infrared from the front of our face than from the sides, and we have infrared detectors because we can tell if we turn a light bulb on [with our eyes closed]. When the person turns their head, then they think that [others] are looking at them and if we use aluminum foil rather than silk...
So there's a physical explanation for this.
Which is infrared. But people didn't know about infrared. So they could have done the same experiment, proved it the same way, and it would have been “extra-sensory perception" detection fifty years ago. My guess is that there are a lot of phenomena which if we don't have the mental apparatus to notice — like new stars or the like — [we don't see]. I'm astonished looking back over the history of astronomy, because this happened in my time. Probably the most significant thing in cosmology that happened in my time, you didn't mention. That is, if you look at cosmology from the 1920's to the 1950's, what everyone studied was the large-scale structure of the universe — omega and Ho [the value of the Hubble constant], Friedmann models. And the structure within it — galaxies, clusters of galaxies — was only interesting insofar as the objects could be [used as] standard candles or as standard meter sticks. And no one asked, where did these things come from? You can't find papers on it.
Maybe it was a harder question.
No, but you'd at least· think that someone would address it. Maybe Gamow did —
Well, Einstein's papers and the early papers on relativity — [Alexander] Friedmann's paper and [Wilhelm] De Sitter's paper — don't address where the galaxies came from either.
I'm saying no one did. It only came up really — because this really did happen in my scientific lifetime — when people were using these as standard candles.
And they realized that they weren't good standard candles.
They say, "Look, they weren't there at a z of 1000, and they are there now. So therefore they were made in between now and then. If they were made in between now and then, they were changing during this interval. If they are changing within this interval, then they're not good standard candles." But then you say, "My God, forget about the standard candles. Where did these things come from?" And, thus, the whole thrust of cosmology has changed to understand the origin of stars, of galaxies, clusters of galaxies and large scale structure. There's much more work done on that now than in learning Ho and go.
I think partly that might be because it's a harder problem. With Ho and go, at least you could set up the problem, as Einstein and other people did.
Well it's not as if we've solved the one before...
No, you haven't solved it, but you've set it up — you know you have to measure both quantities.
I think it's amazing that for 40 years, no one asked those questions, which have got to be obvious questions. That if things were uniform early on, how do we develop structure?
[Georges] Lemaitre worried a little about galaxy formation.
Well, I don't say that there was none.
Yes, he did worry about that.
I'm sure you could find smart people who in the past did, but there was very little work. So that's an example of a blind spot. My guess is that there are many glaring blind spots now, which will be glaring in the future — things in astrophysics and other areas where people will [ask] why didn't they examine so and so, when the data was all around them and they weren't looking at it. I think in some cases at least it's because we don't have the equivalent of a Lucretius. We haven't been set up for it. Where we have, the ideas have conditioned us for the kinds of questions we're going to ask and the kinds of models we're going to apply.
Let me ask you one final question.
There's some place in Weinberg's book on gravitation and cosmology where he makes the interesting statement that the more that we learn about the universe, the less it seems that the universe has a purpose. Have you ever thought at all about this question of whether the universe has a purpose?
Yes, and I find myself mystified by the question. If we go back to what Lucretius said the purpose of the universe was...
He didn't talk about a purpose.
That's the point.
In fact the whole point of the atomistic picture was to rid us of the vagaries of the gods.
Right. So there has been always more than one stream. It's not something new. That's my point on that. I think those who want to make much of us — the forked radish — that we are, you know bipedal ape etc., and want just to find us writ large, and somehow or other that there is a significance in human terms, are always struggling to find...
To find a purpose?
Yes. Wheeler probably is one who is most convincing in doing that. But I've always thought the enterprise was essentially quixotic…
 e.g. J. Jeans Astronomy and Cosmogony (Cambridge, 1928); The Universe Around Us (New York: McMillan, 1929)
 e.g. A.S. Eddington The Expanding Universe (New York: McMillan, 1933); The Internal Constitution of the Stars (Cambridge, 1926)
 F. Hoyle, Frontiers in Astronomy (London: Heinemann, 1955)
 George Gamow, One, two, three... infinity (New York: Viking, 1947)
S. Chandrasekhar, An Introduction to Stellar Structure, (Chicago: University of Chicago, 1939)
H. Bondi Cosmology (Cambridge, 1952)
 J.P. Ostriker and L. L. Cowie, "Galaxy Formation in an Intergalactic Medium Dominated by Explosions," Astrophysical Journal, vol. 243, L127 (1981)
 J. Schwarz, J.P. Ostriker, and A. Yahil "Explosive Events in the Early Universe," Astrophysical Journal, vol. 202, pg. 1 (1975)
 J.P. Ostriker, P.J.E Peebles, and A. Yahil, "The Size and Mass of Galaxies and the Mass of the Universe," ·Astrophysical Journal, vol. 193, L (1974). Editor's note: Ostriker also wrote a paper on massive halos of individual galaxies with Peebles: J.P. Ostriker and P.I.E. Peebles, '''A Numerical Study of the Stability of Flattened Galaxies: Or, Can Cold Galaxies Survive?" Astrophysical Journal, vol 186, pg. 467 (1973)
 J.P. Ostriker and P.J.E. Peebles, Ope cit.
 J.P. Ostriker, P.J.E. Peebles, and A Yahil, Ope cit.
 F. Zwicky Helv. Phys. Acta, vol. 6, pg. 110, (1933)
 V.C. Rubin, W.K. Ford Jr., and N. Thonnard, "Extended Rotation Curves of High Luminosity Spiral Galaxies. IV. Systematic Dynamical Properties," Astrophysical Journal Letters, vol. 225, pg. L107 (1978)
 International Astronomical Union Symposium #100, Besancon, France, August, 1982
 V. de Lapparent, M. J. Geller, and J. P. Huchra, "A Slice of the Universe," Astrophysical Journal Letters, vol. 302, pg. L1 (1986)
 R. H. Dicke and P. J. E. Peebles, "The Big Bang Cosmology -- Enigmas adn Nustrums," in General Relativity: An Einstein Centenary Survey, ed. S. W. Hawking and W. Israel (Cambridge University Press, 1979); the flatness problem was actually stated earlier in R. H. Dicke, Graviation and the Universe, The Jayne Lectures for 1969 (American Philosophical Society, 1969), pg. 62.
 Editor's Note: W. Rindler indeed wrote a paper on the cosmological horizon, Monthly Notices of the Royal Astronomical Society, vol. 116, pg. 668 (1956), but this paper did not make reference to the observed uniformity of the universe or state the "horizon problem." The horizon problem was probably first stated after the discovery of the cosmic background radiation, possibly first by C. W. Misner "The Isotropy of the Universe," Astrophysical Journal, vol. 151, pg. 431 (1968).
 A. Guth, "Inflationary Universe: A Possible Solution to the Horizon and Flatness Problems," Physical Review D, vol. 23, pg. 347 (1981)
 supervises interdisciplinary courses in Princeton's Council of the Humanities
 Editors's note: This statement actually occurs in S. Weinberg, The First Three Minutes (Basic Books: New York, 1977), pg. 154.