Edwin Turner

Lightman:

Let me begin by asking you some questions about your childhood and anything that you remember that first got you interested in astronomy, or any people or books that had a particularly strong impression.

Turner:

Not unlike lots of astronomers, astronomy’s always been something I’ve been interested in as long as I can remember, at least since I was six or seven years old. I remember people saying that I wanted to be an astronomer or a garbage man when I was six. When I was grade school age, I read a fair number of books about it. I don’t remember any specific [ones]. I don’t have a good story like I read this book and was really turned on to the subject, but I was very interested in it and read lots of books. There’s a planetarium near where we lived at the University of North Carolina — the Morehead Planetarium — and my parents used to regularly take me to the shows there. I remember that as always being a very exciting expedition. I’m sure a lot of the lectures and exhibits there had a serious impact. I guess the earliest connection that I can make to astronomy — I remember it from a young age [so it] must have made some sort of impression on me — was when I was about three or so. I had polio and [for] a number years after that I was fairly sick. I often had to wear a lot of orthopedic contraptions and things while I [slept], and during the summer particularly it would be very hard for me to sleep, with all this junk. So I can remember on a number of occasions, on hot summer nights, my mother would take me outside and lay out a blanket on the lawn. We’d lay down and look up at the sky. I guess the idea was just to get me distracted enough to go to sleep. I don’t remember if it worked, but in any case we would lay down and look at the sky and talk about it at some preschool level. I have no idea what we said but I have a clear memory of that and I suppose that’s when I first became aware of the sky as a kind of particular object or place. It may have had nothing to do with me later being interested in astronomy or maybe it did, I don’t really know.

Lightman:

Could you tell me a little bit about your parents, before we go on?

Turner:

Sure. I’m not sure exactly what sort of thing you want to know. Neither of them is in science or any technical field. Neither or them, in fact, has a college education. My mother graduated from high school. My father didn’t because of World War II. He went into the service instead of completing high school. They both, on the other hand, have always been very supportive of and interested in what you might, in some broad sense, call intellectual kinds of things. My mother particularly has always been quite interested in science. She in fact had an ambition to be a chemist when she was in high school. As far as their effect on me — there are many deep ones, of course, as with any parent and child — but as far as astronomy goes, they were always very supportive of it. They never — as I can well imagine a parent doing — regarded [astronomy] as an unwise, eccentric, or crazy ambition. Certainly they were very supportive. For instance, when I was admitted to MIT and was trying to decide whether I should go there, there were actually a number of people, including the minister of our church, who advised them that they shouldn’t let me go up to the North to go to school. This was a small-town, Southern mentality. It would be just as good to go to the universities in North Carolina. And there are very good ones there, of course. But they were very supportive about that and fended off all that pressure. Perhaps I’m not even aware how much there was. And they made significant financial sacrifices as middle-class parents generally do to send people to college. So I think that they’ve certainly made it possible. I don’t think they ever particularly pushed me that way, at least not that I was [aware of]. If so, it was so subtle that I wasn’t aware of it.

Lightman:

What occupations did your parents have?

Turner:

My mother was mostly a homemaker, as they say today. She did work at various times at different jobs. She was a librarian for a time, [and she] worked for Xerox. By the time my brother and I were older, she usually worked in part-time jobs, in a variety of different things. My father was in the restaurant business and was essentially the manager of a restaurant in a chain. He had worked in that ever since the 1940’s when he came back from World War II. Essentially he started out as a — what do you call it — a storeroom boy and worked his way up so to speak. But for all the time I remember, he was the manager of a restaurant.

Lightman:

I want to go ahead in a moment with your general educational experience, but let me first ask a background question. Could you tell me what some of your favorite novels and books are — not necessarily science, but just some of your favorite books at any age?

Turner:

One theme that I was interested in for a very long time through most of my childhood and even now read some — although in the last ten years it’s been a decreasing amount — is science fiction. When I was younger I was heavily into all of the classical science fiction — Asimov and Heinlein and the whole standard gamut of well-known science-fiction writers. Outside of that, I would not say I’ve had any particular zeroing in on things in fiction. I’ve read, and still do [read], a fair amount of history. Things like Bruce Catton’s history or Winston Churchill’s histories of the English-speaking people are wonderful books. Churchill also has a very nice history of the Second World War. I remember those as high points in my reading. I like mysteries some, but not in any compulsive sort of way. Pretty catholic tastes as far as reading goes. I have read a pretty wide range of things. These days I find myself mostly reading poetry and essays, and things like that I particularly enjoy George Will, with whom I don’t always agree but he writes so nicely. For one thing, I think that I look for things now that I can read in five minutes because that’s about all the uninterrupted time I ever have.

Lightman:

Before I go on, I remember the last time that we talked about your childhood, you said that you were very impressed by questions of scale and the distances to the stars. Could you reiterate that?

Turner:

Certainly. One of the things that I remember being most grabbed by intellectually when I was a kid — and it still is in a lot of ways I think what I like in astronomy — I really like to work on the far-away, on the quasars mostly these days. But when I was a kid, in particular, I was quite gripped by just the scale of the universe. All these things like how big things [are] — like if the sun was a watermelon or a grapefruit, how far away is something else. Generally [I was interested in] the idea of big numbers and large scales. I used to spend a lot of time thinking about that kind of thing. I remember particularly once [that] I found some popular book or the other that had a little table in it of fairly nearby stars, with their names and the distance in light-years and probably some other information. I made a little table to go along with that in which I found out what historical event was going on when the light that we were seeing coming from that star [first left]. I was also interested in history then, so that was the connection between the two. The bright stars of course — because of the broad-luminosity function of stars — have a wide range of distances among the stars that you can see. So just with naked-eye stars, [these tables ranged] over modern history.

Lightman:

You were what age? Were you a teenager at this time?

Turner:

I was probably in junior high, early adolescence, twelve or thirteen would be my guess roughly. I’m not sure, but probably something like that.

Lightman:

So you were well aware of the concept that when you looked further into space you’re looking back in time?

Turner:

Yes, which is a neat [concept] and one that grabbed me. I think, in a way, one of the things that particularly attracted me about astronomy then was the feeling that we, stuck on this little, tiny and obscure and very limited planet, living also only for this short little time, embedded in both a universe big in space and a history long in time, can learn about all of these vast reaches of space and learn about the stars and learn about the past by looking far into it. I’m sure I went around explaining to people how you could look back in time. I probably abused many of my peers in those days with little mini-lectures on the subject, but I thought that was particularly [interesting].

Lightman:

At this age, before going to college, do you remember having a preference for any particular cosmological models?

Turner:

No, I don’t think so. I was aware, just from reading and so on, that there were different cosmological models. Certainly I was aware of the contest between the big bang and the steady-state theory which, in fact, was pretty well dying off in professional circles at the time I was in high school.

Lightman:

The steady-state was dying off?

Turner:

Yes, but it was still very hot in the popular astronomy literature and textbooks and popular books and so on. I was certainly aware of those two, and I think that if you had asked me I would have probably said that the big bang was right, but most likely that’s just because that’s what the books that I read told me.

Lightman:

Within the big bang models did you have any particular preference at this time for an open universe versus a closed universe?

Turner:

Before I went to college, I think I would say probably not, or at least I don’t remember any one. Fairly early, I developed an attachment to the open model, but I would guess that that was after I went to MIT. I’m not particularly sure. As best I can remember, the big cosmological issue in my mind in high school was the steady state versus big bang. But at that time I think I was more focused on non-cosmological parts of astronomy, partly because in high school I was an amateur astronomer. I did the usual rite of passage of grinding your own mirror and making a little reflecting telescope and being in an amateur astronomy club and all that sort of stuff. Of course, with instruments like that you tend to get focused on things that you can see with instruments like that. [Cosmology] did not seem like a big part of astronomy to me at that time, I think.

Lightman:

Tell me a little bit about your undergraduate education at MIT.

Turner:

I took a physics degree there. MIT doesn’t have an [undergraduate] astronomy program. While I was at MIT, I was pretty clear in my mind that I wanted to do astronomy. By that time — or soon after arriving — I understood that that meant getting a Ph.D in physics or astrophysics, and so that was more or less my goal. I think I never seriously doubted that while I was at MIT. I took a physics degree eventually because they didn’t offer an astronomy one, and I got as much astronomy as I could. There was actually rather little offered. There was one survey course which I never thought of as being a particularly wonderful course. There were some very good seminars that were a side offering. I remember particularly a seminar that Phil Morrison and some other faculty members gave which was all centered around various aspects of the Crab Nebula. So it was kind of focused on that but you could of course tie in lots of things [such as] supernovae and this and that. I remember that as being a particularly exciting course. I took some course in planetary or solar system astrophysics from Irwin Shapiro which I remember enjoying. But most of the coursework was physics oriented. When I was at MIT, I also sought out opportunities to get involved with astronomers and got jobs working with the X-ray group at the Center for Space Research. I actually had some summer job at AS&E [American Science and Engineering], which used to be one of the big centers of X-ray astronomy in Cambridge. I did my senior thesis with Phil Morrison on a theoretical topic that had to do with cosmic rays. So I was kind of looking for [opportunities]. I began at that time to go to colloquia that were in the astronomy field. As you know, they have a Tuesday afternoon astronomy colloquium series there which I started to go to in my senior year. I think that was probably the first time I was really part of a group of professional — I don’t know if it was fair to say I was part of the group — but at a place where a group of professional astronomers were together carrying out a professional function. In my later years [at MIT], I began to read in the Astrophysical Journal and other astronomical journals. So I was definitely headed that way.

Lightman:

Did you know at this time that you were going to go into cosmology or were you that specialized yet?

Turner:

No, I don’t think so. I became more interested in cosmology. I remember reading Dennis Sciama’s book.^[1]

Lightman:

As an undergraduate?

Turner:

As an undergraduate. I believe it was out then. I read that and got a lot more interested in the open versus closed issue and so on.

Lightman:

Let me ask you about that — the open versus closed [models]. You mentioned earlier that sometime during this period as an undergraduate you had a preference for the open universe. I remember you mentioned that to me the last time we talked. Could you tell me why you had that preference?

Turner:

I think there are two reasons — by which I still feel moved in that direction. On the one hand, I’ve always thought that a sort of one-performance-only universe was aesthetically pleasing: a kind of a universe that is born and then goes through all of this evolution and then just sort of fades away. I don’t know [why but] for some reason I always thought that was an appealing image rather than the ones which most people find more aesthetically pleasing — the time-symmetric. I think it was less a mathematical aesthetic and more [an] emotional one. So I always liked that. The other thing I liked is that by the time I was an undergraduate I was aware that this was an experimental question and that if you looked out and added up the matter that we could “see,” so to speak, in the stars and galaxies and interstellar medium that you didn’t get nearly enough.

Lightman:

Enough to close the universe.

Turner:

Enough to close the universe. So the evidence looked like it was open. And I was — I have to be careful not to project back [my] current vocabulary and ways of thinking onto that time — but I think it was clear to me even then that if the universe were closed, that implied that we were kind of somehow missing most of the game. Somehow it was hidden or there was something [mass] out there that was very much different than what we could directly observe, and I thought that was unappealing. Also I think it’s a little bit scientifically improper to make the argument that is sometimes made that a closed universe would be nicer, therefore the stuff must be there. It seems to me that if the evidence was that it wasn’t there, it wasn’t there. It should just be an empirical question. But also I don’t think I liked the idea that “everything you know is wrong,” which is the title of some Firesign Theater album.

Lightman:

Firesign?

Turner:

Yes, I think it was called Firesign Theater. It was a San Francisco kind of 1960’s comedy group. They once had an album called Everything You Know Is Wrong. I think that’s a worry that astronomers have to have all the time, or cosmologists particularly, that everything we know is wrong.

Lightman:

So you preferred to think that what you see is what you get.

Turner:

Right.

Lightman:

Something like that.

Turner:

But to finish answering your earlier question. I think when I was at MIT as an undergraduate I was more interested in X-ray astronomy and high-energy astrophysics — collapsed objects and neutron stars and all of that. Obviously, just again as a result of that being the focus of a large part of MIT’s astronomy community. I thought astronomy was all KeV’s and proportional counters.

Lightman:

Tell me something about your graduate school career at Caltech.

Turner:

As you say, I went to the astronomy department at Caltech, thinking I wanted to be a theorist actually, but was soon disabused of that idea. I became interested in optical observational astronomy. Wal Sargeant was my thesis advisor, and I also worked with him before I started my thesis. He was probably the faculty member who had the most direct influence. Also, to a considerable extent, I think of Jim Gunn as a kind of informal secondary advisor and I saw a lot of him. [During] my first year or so at Caltech, I think I really wasn’t very happy. I didn’t like it very much and considered leaving.

Lightman:

Why was that?

Turner:

A wide variety of reasons. Partly, I think that the coursework load was very heavy. By the time I was a senior and the summer after my senior year at MIT, I’d been spending a large fraction of my time on research — or programming at least, working for somebody doing research. Going back to the one-problem-set-a-week- in-three-different courses grind reminded me of the early MIT years, which I didn’t like very well. I was more or less a hippie at MIT, I think it would be fair to say. Caltech was not very conducive to that sort of thing. So there were a lot of changes. I think it was just a variety… Also I’d been married just before I came out to Caltech and my wife couldn’t find a suitable job for the first year so she wasn’t very happy. So all of those things just kind of made it a grim year. But by the second year, she’d found a job and I got to start doing some research and after that I enjoyed it a great deal.

Lightman:

What was some of the early research that you did, leading to your thesis?

Turner:

I think the earliest thing I did there that resulted in any publications was work with^[2] Wal Sargeant on compact groups of galaxies and alignments. There were issues then about so-called “chains” of galaxies, which are frequently seen in these very compact groups. There were questions of whether this was because they were disks — systems of galaxies that were organized in disks that we were seeing edge on or whether they were just chance projection effects. So I did some work on that.

Lightman:

What about your thesis?

Turner:

My thesis^[3] was on dynamics of binary galaxies and groups of galaxies, which I got into essentially after having been at Caltech for a couple years. I’d been much more focussed on extragalactic astronomy and I looked around and tried to see [if] there was there some topic which seemed like a fundamental topic in extragalactic astronomy which had been neglected for a while. I decided binary galaxies were. There had been some classic work by Page^[4], mostly in the 1950’s, using just photographic spectroscopy. It seemed to me that it was now possible to do a much better job on that. It attracted me because it was what I thought of as a fairly fundamental problem in extragalactic astronomy, sort of a statistical problem, and I rather liked statistics. It seemed like something where a new observational initiative was really called for. So most of my thesis was doing a big observational program on binary galaxies and a statistical analysis of the data. I also got involved with groups of galaxies as a related topic. There, I didn’t get any new data, but I analyzed a lot of existing data, mostly in collaboration with Rich Gott. One of the issues that I was interested in early was that it seemed to me that a lot of the discussion of dynamical masses was tied to rich clusters, which are actually a rather rare thing. Only a few percent of galaxies are actually in rich clusters. So it seemed to me that something on the masses of galaxies in more common systems would be valuable.

Lightman:

So you were measuring the masses of these binary galaxies and groups using a virial method?

Turner:

In the case of the groups, it was a virial method. In the case of the binary galaxies, it amounted I suppose to a virial theorem. There, it was actually comparing the data with simulations and more explicit models including distributions of eccentricities, true spatial separations, that sort of thing. Of course the thing that came up — I didn’t know about it when I started the thesis, but [it] soon came up — was this issue of massive halos of galaxies. At that time in Princeton, a pair of papers, one by [Jerry] Ostriker and [Jim] Peebles^[5] and the other by Ostriker, Peebles, and [Amos] Yahil^[6], suggested that galaxies might be just some bright tip-of-the-iceberg, some luminous bit of stuff embedded in some much larger and more extended mass distribution — a dark halo. One of their key bits of evidence for this was: they plotted various mass determinations of galaxies as a function the distance over which this mass had been determined, and they had this nice linear relationship which showed the mass going up and up as you measured it on a larger and larger scale and they had a variety of points using different techniques. There was one point that was way off, which was Page’s binary galaxy point. It was like an order of magnitude below the line. And, of course, suddenly, when they published these papers and when Jerry Ostriker began going around and promoting this idea of the massive halos and insisting that people take it seriously, this discrepant point suddenly became the focus of a lot of attention and interest, which made my thesis seem much more fashionable. That was very welcome as far as I was concerned, and in the end, in fact, I focused the work on being a test of the massive halo hypothesis.

Lightman:

Did you have any speculation as to how this was going to tum out before you got your results?

Turner:

Yes, that’s interesting. Alan Sandage asked me that, I think, on my first observing run on the mountain. He said, “Well what answer are you going to get? You should always have some answer in mind before you start.” I was relieved to hear him say that, since I did think that I would find low masses [for the galaxies in binary systems]. I thought it would contradict the [massive halo claim]. I thought this dark massive halo stuff was probably wrong…

Lightman:

Did you have reason to think it was wrong? Does it go back to that earlier sort of philosophical feeling that what you see is what you should get or were there other reasons?

Turner:

I think partly that, and partly some more technical reasons that just had to do with the fact that I could see various weaknesses with some of the other points that had been plotted. I was quite reasonably heavily influenced at that time by the [Richard] Gott, [James] Gunn, [David] Schramm, and [Beatrice] Tinsley work^[7], which argued that the universe was open by a wide margin. At that time, people tended to — improperly, I would say now — equivalence the idea of dark, massive halos with a closed universe. The idea seemed to be “Well if there’s all this dark matter, then that was probably just what was needed to close the universe.” Of course it doesn’t follow. There’s plenty of room for dark, massive halos without closing the universe, but the Gott, Gunn, Schramm, and Tinsley stuff influenced me. Also the fact that I thought Page had already done it — I thought I could do [it] better, but I didn't think that it would change the result by an order of magnitude. So I was quite surprised when I did the models. I did the analysis and the point moved up, bang on to the Ostriker and Peebles line. Also I ran models which were intended to distinguish between whether the mass distribution in the galaxies was extended or compact and it seemed to favor extended. So it was not the result I expected but I was happy enough with it I mean my expectation wasn’t so strong that I just published it and forgot about it. I went around telling people it was all right. I was converted to a…

Lightman:

To a dark matter…

Turner:

Well, at least, a dark, massive halo person. By the time I finished my thesis in 1975, there was still at that time lots of [debate] — there’s still a little of course — but at that time at meetings and things, the whole hypothesis was still debated pretty hotly. I remember defending it and supporting it.

Lightman:

Did your own result here alter your belief in an open universe at all?

Turner:

No, I don’t think so — which maybe just reflects how illogical or in some sense irrational one’s beliefs on these things can be. In some sense I had just become convinced that there was ten times more matter than we could see, but it didn't move me much on the question of whether there might be a hundred times more than we see — enough to close [the universe.] I think the defense I made, the scientific reasons, were mostly of the Gott-Gunn- Schramm-and-Tinsley type: the deuterium and so on. So I thought it was at most ten times more dark — I thought it was probably baryons in dark stars of some sort and you could put a pretty consistent picture together at that time. An omega [equal to a] tenth basically fit pretty well everything from clusters of galaxies to binary galaxies to deuterium. The only real fly in that ointment was the covariance function work of Jim Peebles and so on which claimed to give evidence for a large omega, an omega equal to one or greater. I remember explicitly conversations with various people at that time in which the opinion was expressed — and it was certainly mine — that other than that covariance function result, everything seemed to be consistent with an omega of a tenth. I think it was by this time that I’d become relatively obsessed with the open/closed issue — or obsessed isn’t exactly the [right] word, but that had become a sort of a major focus or organizing theme, and that really led into the work^[8] that I did right after I got my Ph.D on N-body simulations of galaxy clustering. [That work] was an attempt to see if the covariance function argument was really a strong one.

Lightman:

Do you remember when you first decided that you might want to do that work? The N-body simulations?

Turner:

Not in the sense of when we definitely decided to do it, but I know where it started, which was with a colloquium that Jim Peebles came and gave. I think it was probably in the physics department, might have been the astronomy department —

Lightman:

At Caltech?

Turner:

At Caltech, in around my last year there as a graduate student, certainly towards the end. In this he showed a movie of an N-body simulation that he and Ed Groth^[9] and perhaps other people had made. [It] was just an early — perhaps the first — example of these now standard kind of simulations. This was part of a talk on the covariance function and galaxy clustering. All he did really was show the movie and put up the pictures and say, “Now this really has [just] gravity in it. Doesn’t it more or less look like the galaxy distributions we see in the sky?” It was just a kind of audio-visual enhancement of his show. But the group of people that heard it had lots of questions, like “Have you calculated this for the N-body simulation? Have you calculated its covariance function? Have you looked at how it evolved?” and a whole bunch of questions. Jim had not. I think he has never trusted N-body simulations enough to really get into that game, but it seemed to Rich Gott and me talking about it afterwards that this was a potentially very powerful tool for studying large-scale structure because you could put in whatever physics you thought might be going on and you could simulate observations of the universe and compare them to the actual observations and test models in a fairly direct way. So we thought that Peebles and Groth had a great idea and they weren’t exploiting it as much as they should. That was the germ of it. Rich and I had had some discussions that maybe we could write a big N-body [program], but then Rich went off to Cambridge, England, as a postdoc for a year and met Sverre Aarseth. Sverre, of course, is one of the world’s greatest experts on N-body codes and already had some big ones. So that started a collaboration in which we did many N-body simulations and calculated what we thought Peebles and Groth should have calculated. We tried in particular to see whether you could tell the difference between an open and closed universe from a covariance function. Our conclusion was that you couldn’t, which was reassuring. I remember when we got the first models back that had large and small omegas and ran off the covariance function. We felt sure that the small omega covariance function would have a big break in it the way that the hand-waving theoretical argument said that it should.

Lightman:

Yes, it’s the back-of-the-envelope argument.

Turner:

We pulled the plots off the plotter and we laid the graphs down on top of each other [and] you couldn’t see a difference.

Lightman:

So what did you conclude from that? I mean, did you conclude that Peebles’s argument for an open universe using the covariance function was then spurious?

Turner:

Yes, there were about four or so thick papers that we did on those simulations that have all sorts of detailed stuff in them. But the basic conclusion of that work, in the context of the open versus closed thing, and the result that was most important to me [was that] the “cosmic astrology” notion that you can read off the future of the universe by looking at the positions of the galaxies in the sky just turned out to be wrong, as far as these simulations went. Later, other people did simulations that were better in various technical ways and involved more points, and I think they did find rather small differences between open and closed universes in the sense predicted by the back-of-the-envelope calculation. But very much smaller effects than had been predicted and ones in fact which I don’t think one could ever have hoped to recover from at least the angular covariance function on the sky, because of all the projection effects and so on which tend to wash out little details anyway. So I think our basic conclusion — which we may have overstated [in saying] that there was no differences — that you couldn’t tell the difference is probably correct.

Lightman:

Were you disappointed with this? How did you feel about this conclusion?

Turner:

Well, mixed. The idea is very pretty. Not only were you supposed to be able to read off omega but also, as you know, the power spectrum of the initial density fluctuations. So it looked like it was potentially a very nice, direct probe for going back into the early universe, and we just found that it was very insensitive to everything. In some sense, I think the covariance function is very sensitive to gravitational clustering. If it’s gravitational clustering it will have the right kind of covariance function and so the fact that we see that in the sky strongly suggests the perhaps not-too-surprising conclusion that gravity plays an important role in distributing the galaxies, but beyond that it didn’t seem to have a lot to say.

Lightman:

I seem to remember that the three of you had nicknames for your collaboration.

Turner:

Right. We had a standing joke I guess. Since we were doing these simulations, we used to say that the collaboration was a simulation of God, Mother Nature, and an astronomer. Rich Gott — which was also a pun on his name — was God because he generated or prescribed the initial conditions — he came up with various prescriptions for laying those down. Sverre Aarseth, who had the big numerical code, executed the laws of nature. He worked out the results. [He was Mother Nature.] And I usually then took those tapes that Sverre had generated and calculated simulated observations — calculated the covariance function or found binary galaxies in them or whatever. I did things to the simulations that were analogous to what astronomers do to the sky.

Lightman:

So you were the astronomer. I wanted to talk to you a little bit about your reactions to some of the new discoveries both theoretical and observational in the last ten years. Let me ask you first a general question, then I’ll get more specific. Generally, have your views of cosmology undergone any major conceptual changes in the last ten or fifteen years?

Turner:

[Pause] I guess not. In some ways I am influenced by the inflation stuff to take the larger omegas a little more seriously, but I don’t think I’m as moved by that as most people are. When I do papers, I tend to use omega of one as a convention, just because I think that seems to have become the accepted convention and I think it’s very useful. I know from my binary galaxy work it can be very confusing if everybody uses different H₀'s [the Hubble constant], different measures of mass to light ratios, different photometric bands and makes different dust obscuration corrections. [With] these parameters that we don’t really know, I think it’s — very useful if everyone settles in to some convention, so I’m willing to go along. My own work at least has not been particularly focused on the omega question in the last ten years, or not nearly as much. I think I’m a little more — how shall I say — cynical about the probability that we will really have any clear-cut answer soon or inside my career.

Lightman:

You mentioned the inflationary universe model^[10]. Let me ask you specifically about that. Do you remember what your reaction to it was when you first heard about it?

Turner:

I thought it was extremely clever and a big surprise. I was not expecting that someone would come up with such a clever new idea about the early history of the universe that would help so much with the causality problem and all these other things that one was aware of — these little uglinesses of the early big bang. So I thought it was very clever. I also thought it was very speculative. I guess when I first heard about it I thought [that what] would have happened with the inflationary universe — nothing like the influence it actually had — was [that it] would have been put aside, so to speak, as a clever little speculation or a clever idea that would always be mentioned when one discussed the early history of the universe, as a possibility that might well have happened. I would not have expected what did happen, which is that it became the foundation of a whole renaissance in the study of the early universe. The whole thrust of cosmology in the last ten years has in many ways been exploring the ramifications of that idea. I think I would have thought that it was too speculative and too hard to check for that to have happened.

Lightman:

Do you have any personal opinion as to why the inflationary universe model caught on so dramatically and universally?

Turner:

First of all, I think it just does have the property that it offers explanations for things that are otherwise hard to explain and it is very clever. But the other reason I think is that it is a paradigm, if I may use that word, or a model which allows one to do lots of cute calculations. It’s a sort of theorists’ gymnasium, so to speak, where one can go and there’s lots of nice problems to do and to be worked out. One reason I wouldn’t have anticipated that was [that] I didn’t understand the theory well enough — maybe I still don’t. But it wasn’t obvious to me when the theory was originally described to me that that would be the case. But I think that does tend to happen when some new idea comes up that has lots of problems or aspects of it that can be worked out. Naturally people do. There’s nothing wrong with that. But I think that it often gives the impression that the theory is likely to be right, or that the field is more important than maybe it is. It’s just the fact that there are nice calculations that you can do. In other words, the theory of star formation is obviously incredibly important to astrophysics but you don’t really have theorists flocking on to that, just because there’s not all that many well-defined calculations and problems you can set yourself and solve. I can think of analogies. The field I’ve mostly been working in — gravitational lenses — is a bit like that. There’s been a huge number of theoretical papers on that. There are probably twenty theoretical papers on it for every gravitational lens candidate that’s been found in the sky and the reason is that there are a lot of nice problems that no one had thought of before that you can work out.

Lightman:

Theoretical problems.

Turner:

Right. Theoretical problems. Black hole theory in the 1960’s was a bit like that. It involved techniques and discoveries that allowed one to calculate out all sorts of detailed properties of black holes even though at best maybe we had a little indirect evidence for one. There’s an observational analogy to that, which is [that] if we get some new instrument or technique or detector that lets us study something that we couldn’t study before, everyone goes wild working on it just because it’s something new.

Lightman:

Because it’s there.

Turner:

Yes, it’s something you could do. But I think that often misleads people in their assessment of how likely something is to be true. If you made a great pile of all the papers written on inflationary cosmology, [it] would be very impressive, and there’s some kind of subconscious impression that forms in the community that this must be right because there’s so much work on it or it must be in the right direction. Really, I think you would have to say — a more external rationalistic approach would say — “Well, what are the hard pieces of evidence that this actually occurred?” I wouldn’t say there’s any evidence against it, but there’s not a whole lot for it. I still regard it as something that could easily be right or could easily be wrong. You shouldn’t be surprised either way.

Lightman:

Let me go to another theoretical idea that’s been put forth, which actually was a motivation for the inflationary universe model. That is the flatness problem. I think it first came to the community in a big way in the paper by Peebles and Dicke in the 1979 Einstein Centennial Volume,^[11] although it had been discussed for a decade before then. Do you remember how you first reacted to hearing about this flatness problem or flatness puzzle?

Turner:

I think I probably first heard of it when I was still at Caltech, probably from Jim Gunn. I remember somewhere or the other hearing [John] Wheeler mention it in a lecture as well. I think I thought of it as one of those things that’s a little hard to know what to do with. It seems curious but, maybe it’s just the way things are — like the sun and the moon being the same angular size in the sky, which they are incredibly precisely. That’s something that just doesn’t [have any explanation] — that seems like it must be an accident in our understanding of things and you’re a little uncomfortable to see such a strange fluke, but since it’s only one thing and there [are] many things you can notice in the world, [you’re] not uncomfortable [enough] to really worry about. The flatness problem did not bother me a lot. It was something that, if someone could come up with an explanation for it, great, but otherwise I don’t think I would have lost any sleep over it. I guess if someone had insisted — at that time, before I heard about inflation — that I say something about that or take it seriously, I would have said, “Well it’s something that we may understand some time in the far future when we understand quantum gravity or God knows what. It may be that it’s some result of some early physics that we don’t know.” It did seem like sort of a coincidence, but I think I would have thought of it in a rather remote and vague sort of way. I did not think that that was a problem that…

Lightman:

That required an explanation?

Turner:

I thought of it as either something that didn’t require an explanation, or, if there was going to be an explanation, it seemed to me off on the horizon. It was a problem for some future generation or something, something far away.

Lightman:

Has your point of view about the significance of that problem changed any? Do you have the same view now, that is, do you view it as something that’s accidental or something that requires a deep physical explanation?

Turner:

Having had an explanation for a while, it now seems more important to have it explained. As merely a practical matter, I don’t think that inflation is likely to be replaced as our best consensus, or best guess, at the early history of the universe unless it is replaced by another theory that can also explain [the flatness problem] in some way. I think that we won’t give up inflation for some other theory unless the other theory can also cover that base.

Lightman:

Can also explain the flatness problem.

Turner:

Right. That is unless… One could imagine, although only barely, I think, that some experimental test of inflation would prove that the universe had never inflated. I can’t quite imagine how that is. If you had some empirical evidence of that I guess you might be forced to go back to throwing up your hands. But I have a feeling that in a search for some replacement theory, this [the flatness problem] would be taken seriously. I don’t really quite think that it has to be, in the sense that if one wanted to, [one could] go back to the kind of pre-inflationary idea where you blamed a lot of things on just the initial conditions. Initial conditions are really, in this context, a very special kind of idea or thing, and it’s not very aesthetic to imagine that the universe had some [special initial conditions]. It’s certainly much more pleasing to have some theory to explain it, but it doesn’t strike me as intolerably ugly to just have to say, “Well, the fact that anything exists at all is so strange in a way, that maybe the initial conditions are just very special in some way.” I don’t know. As far as the practical evolution of the field, I think it will continue to play a very big role.

Lightman:

How did you react to the results of De Lapparent, Geller, and Huchra^[12] on the large-scale structure of the universe?

Turner:

The main reaction I had was that it seemed like a fairly convincing indication that the old gravitational hierarchy — the amplification of small, initial fluctuations by gravity — was not the whole story. I guess I don’t really think that even now that result has been fully appreciated. I don’t know. It seemed to me to that it was such an unexpected result.

Lightman:

Were you expecting to see things much more homogeneous?

Turner:

Yes. I just wasn’t expecting to see anything with such a large amplitude and so regular, if you like, on large scales where, on the basis of the covariance function work, one would have thought the fluctuations should be small. And I guess I was heavily influenced by lots of N-body simulations, including our [Aarseth, Gott, and Turner] own, and never having seen anything like that. Of course, in the year since, people have wiggled around in various ways to try to produce something that looks a bit like that [the observed bubble-like structure]. But I think my reaction, which was not much different from that of most people’s, was one of considerable surprise. And I think that probably means that at least on large scales there is some process going on that we don’t [understand] yet, or at least there is no single good candidate. Jerry’s [Ostriker’s] explosions^[13] hint in such a direction, but it just seems to me that [the observed bubble-like structure] makes the whole vocabulary with which we used to discuss galaxy clustering — where we’re talking about small groups and binaries and hierarchical clustering and so on — to have missed the boat. Clearly, there must be some mapping of those old concepts onto the new picture, but I don’t think we really know what it is. [The new observations] illustrate the impotence of the covariance function, if you will, when applied just to galaxies distributed on the sky, because this [structure] is clearly not a small perturbation or some slight effect, but some dominant kind of phenomenon going on that we just missed entirely. I don’t think it’s clear yet what it will mean, but I do think it means that all that work that Peebles and Groth started and [the N-body simulations] that Arseth, Gott and I did and other stuff that a lot of other people did better and more probably is not [relevant.] We may have learned some physics from it, but I don’t think we learned anything [much] about the [real] universe from it.

Lightman:

Does that bother you at all, that maybe a lot of work that has been done has been irrelevant?

Turner:

Well, in a sense, yes. I guess I would say that it has given me a considerably more modest or pessimistic view of what we can hope to accomplish in cosmology and astronomy. Especially on these very big questions. It seemed to me [to be] the antidote to the cosmic astrology dream of reading off the fate of the universe from the positions of galaxies because [the new observations of large-scale structure] showed how far one could go wrong with a few simple, pretty ideas and some data. Again, it’s along these lines we’ve discussed that everything was fine and very nice in a theoretical way and then the data reared its ugly head. Another discovery, which may seem totally unrelated but about which I felt much the same way, was the Voyager pictures of Saturn’s rings. Here is an object which has been known since antiquity, which has been studied by astronomers forever, which is nearby, which we can resolve with many resolution elements across, which we can study [by] starlight shining through and sunlight reflected from, which involved only simple physics — basically a point mass [for the planet] and gravel [for the rings]. If there is any astronomical object of any significant complexity which you could argue that we ought to be able to understand, you might have thought we could understand the dynamics of Saturn’s rings. People have reflected radar off the rings, you name it. We’ve got all sorts of data. It’s a reasonably simple situation, and we’ve done centuries of work on it. And the pictures were just on their face full of surprises, just totally unanticipated — a major phenomenon, again not these tiny little effects. How many artist’s conceptions have you seen of Saturn’s rings? I don’t know how many I’ve seen, various models and things. None of them had any hint of [what we saw in the photographs]. It’s all perfectly obvious in hindsight. Some of it actually isn’t so easy to explain in hindsight, but much of it is. Then you ask yourself how well do we understand something with the complexity of the universe as a whole, based on a handful of observational facts. How well do we understand a quasar, when we can’t even resolve it? It’s just a point of light.

Lightman:

This gets back to your earlier suggestion that there’s a chance that we’re going off in totally the wrong direction.

Turner:

Something like that makes me think there’s a reasonable chance we’re [all wrong]. As far as the large-scale structure is concerned, it quite discourages me. I haven’t worked on this since. If I had an idea, I would, but it seems clear to me that all of the lines of thought I was pursuing are not on the right track and I have not been particularly persuaded by any ideas that have been put forward since then. At least I don’t have any specific ideas for anything to do, so I’ve more or less stopped working in the field on the basis of the surprising results. ^[14] [** In terms of our general cosmological picture and the ideas about the early universe in particular, I also think there’s a fair chance of really missing the boat, of being in serious error. I think that there is a particularly good chance that our theoretical ideas grossly, qualitatively underestimate the complexity of reality. That is what usually happens when there is little data. We make a model that is no more complicated than it needs to be to explain the data, failing to realize that nature most often realizes complex situations even in apparently simple situations, as in the case of Saturn’s rings. People with training in physics, like nearly all astronomers and cosmologists, are particularly likely to fall into this trap. The fundamental physical laws have a subtle and beautiful simplicity about them, symmetries and so on. Physicists learn, or are taught really, to respect or even venerate this quality of elegant simplicity. Looking for it is even a tactic which has paid off in the search for new fundamental laws. However, I feel that this approach may backfire dreadfully on us in our attempts to guess at the actual realizations of the simple laws in the physical universe, to guess at the early history of the universe and so on. In fact, I think that the ability of the universe to manifest such astonishing complexity as we see around us on the basis of profoundly simple fundamental laws has a kind of beauty of it own which may just as well apply to cosmology as to planetary ring systems and so forth. **]

Lightman:

One of the things that I’m interested in is how scientists use metaphors and visual images in their work, to whatever extent they do. For example, in cosmology, one very powerful metaphor has been the expanding balloon model for the expanding universe. Do you use metaphors and visual imagery in your own work?

Turner:

[Pause] Yes, I think so, although… Certainly imagery. I think that astronomers or scientists are of the type that [either] think of the equations or they think of a picture [of the physical situation] when they’re discussing something. And you can sort of distinguish the two: the ones that think of the equations talk about the equations. They talk about the second term being big or that kind of thing. And the ones that think in pictures talk about the thing in a more physical way. I certainly tend to think of things in a picture way as opposed to an equation way. I don’t know about metaphors, exactly. I mean, I don’t generally think of the balloon and that sort of thing.

Lightman:

Have you ever tried to picture the universe during the time around the big bang, around the beginning? Do you have any image of that or do you just never try to visualize that?

Turner:

No, I think about that some. The trick there is to not think about it as something you’re outside of looking at — you know a little thing going “bang” — but to think of something you’re inside of.

Lightman:

Are you able to do that?

Turner:

A bit, yes. I imagine the atoms or the nuclei or the quarks or whatever it is, depending on the epoch, as like galaxies. You can imagine yourself sort of sitting there on some quark or on some point in this early fluid and having the density dropping incredibly fast all around you and things shooting off. Or, alternately, during inflation you think of everything shooting off but the density not dropping somehow. It seems peculiar. I’m not sure if this is the kind of thing you’re looking for, but when I think about the early universe, I generally think of it with the same picture in my head that I use for the universe now except to replace the galaxies with — Oh, I don’t know — hydrogen atoms if you’re talking about the recombination epoch, or nucleons if you’re talking about nucleosynthesis, or I’m not exactly sure what but some particle whose name I don’t know if we’re talking about some earlier epoch. And you think about forces other than gravity playing a role but it’s a kind of scaling metaphor or something like that. So I’m much more inclined to think of it that way — and to try to get students to think of it that way — than to think of it in the more obvious way of: here sits the universe a centimeter in diameter and I’m over here somewhere.

Lightman:

Yes. As both an observer and a theorist, how do you think that theory and observation have worked together in cosmology in the last decade?

Turner:

I don’t know if I’m a theorist. I often say that the secret to my career has been that the observers think I’m a good theorist and the theorists think I’m a good observer, but really I think I’m an observer in my own mind and not very much a theorist, but I like the theory. I’ve done some work in some kind of borderline areas of theory, but always aimed at interpreting the observations. But I think that there’s a real problem — to get to the heart of your question — in that the theoretical work and observational work in cosmology are diverging to a considerable extent. We’re getting theories which are tied by more and more extended and tenuous chains of argument or calculations to things that we can directly observe. I think there’s this worry, or nightmare, that we could end up conceivably at some future time with several different scenarios for the evolution of the early universe which perhaps are very different physically in terms of what we imagined happening or why [it happened] but [with] no testable consequences. In particular, maybe no testable astronomical consequences. Maybe you’ll be able to test a little bit, if the basic physics is right, in an accelerator somewhere, but that’s not quite the same as knowing that it really happened in the universe. I mean, nuclear physics was well understood and you certainly had nuclear reactions — the physics of, say, the nucleosynthesis epoch is under good control, but getting the direct astronomical evidence that in the big bang there really was an epoch of nucleosynthesis is a separate thing. I’m quite worried that we could get into a situation in cosmology where we were using physics, which maybe was tested in the laboratory or maybe not — I’m just not sure even how easy that is — in which the astronomical predictions are so vague that we won’t really have any contact between the most fundamental part of the theory and observations. Now there’s always some theory that’s aimed directly at interpreting the observations, but in some sense the deepest part of our work on cosmological theory, I think, is receding beyond the reach of the observations, and I fear that more and more the connections may be being cut. I’m reasonably persuaded — perhaps just by being close to it — that some of Jerry Ostriker’s ideas of explosive galaxy formation and so on may be correct, and that such things may have played a role in the large-scale structure. That will disconnect yet another set of observed phenomena of the galaxies, the sizes and masses of galaxies and their distribution and so on, from the really early epochs of the universe.

Lightman:

How do you feel about using our general theoretical models to extrapolate backwards ten or fifteen billion years to the first nanosecond? Do you think that we’re more or less justified in doing that? Do you think that's a good thing to do?

Turner:

Oh, yes. I think that it is. [What I was saying] almost sounds like a criticism of the theorists, but obviously any opening that we have, we should pursue in the calculations we can do, because there’s no telling what you’ll find at the end. Conceivably, there’ll be some very elegant way of tying the theory back to the observations or even if there isn’t, it’s still valuable to do the theory and I would still represent it as our best guess. But I’m very distrustful of our ideas getting too far removed from being checked empirically. I just think the chances of going wrong are very large. I think that’s unfortunate. There’s nothing to do about it, I guess. The universe may have “forgotten” how it formed, so to speak, at least within the limits of what we can find out using our current abilities to tell what’s out there. Then it becomes a little like history. This is this other subject I’m interested in. History is very frustrating. There’s just a finite amount of information about what Alexander the Great did and thought, and there are no doubt very important facts about great events that we’re interested in that just were not recorded and they’re not passed down to us, and we’re not going to find out. We’re just not. In a certain style of history, one can imagine that Alexander did this because of that reason that he was thinking in his head, but you’ll never know if you’re right. I don’t think that makes it intellectually vacuous or uninteresting. I still like history, but it is not as nice as if you can actually find out if you’re really right or not.

Lightman:

Let me ask you a question that will require you to take a big step backwards. If you could design the universe any way that you wanted to, how would you do it?

Turner:

I may take longer to think about that than [the time we have.] The changes I would be most inclined to make are not of interest to astronomers and scientists I suspect. Well, they are not of an astronomical or scientific nature. One could make lots of improvements in the world on how people behave.

Lightman:

You mentioned earlier that you would prefer that there not be a lot of dark matter, because that would be a malicious thing.

Turner:

Certainly, I would have the universe be rich and complex and accessible. I guess those are the way I would describe it. Accessible to observations and, even better, accessible to interaction. It would be nice if the speed of light were a lot higher relative to the size of the universe. Another way of saying that I guess is [that it would be nice] if the universe were younger or the energy requirements were such that — Obviously, if you start fiddling around with these things, you’ll soon have no life left, but it would be nice if chemical rockets were practical vehicles for traveling to the stars and things like that. The astronomical universe has this annoying property of being like a bakery filled with delicious goods but you’re outside with your nose pressed against the glass. You can see it but you can’t touch it. ^[15] [** In addition to accessibility, I would like a universe with richness and complexity, which is clearly what we’ve got to a considerable degree. Nevertheless, the more the better. For instance, I much prefer theories in which the dark matter is some sort of population of dead stars, a Population III in current jargon, which had to form early in some unknown way and environment, had to influence other components of the universe, have some evolutionary history, have some current set of complex properties, and so forth. Theories in which the dark matter is taken to be merely a sea of some sort of exotic identical elementary particles, weaking interacting, hardly participating, with properties so simple that they can be derived from first principles in a few lines, strike me as very dull, a real waste of potential richness for the vast majority of the stuff in the universe. Perhaps this example is just a sort of turf war. I’d rather see the dark matter in the astronomer’s domain than the physicist’s, but I’d still stand by the general point of preferring richness and complexity. **]

Lightman:

So you’d like to be able to get around to the stars.

Turner:

Sure, you’d like to be able to get that barrier down. You’d like to be able to look at a quasar as closely as we looked at Saturn or better. You’d like to be able to find things out. Maybe this would make science too easy. But I am inclined to think that astronomy is a little too hard. One can take various philosophical attitudes towards the nature of the truth and so on, but if we take the most simplistic view of an external reality which has some truth value, and our job as scientists is to try to get hold of that, I think that a hard-nosed rationalist might conclude that the job is too hard for us. And that, in fact, there is much interesting and true stuff about the universe that we will never learn, or not in a foreseeable time. [As I said], much of what we currently believe may well be wrong, or at least we won’t be able to find out whether it’s true or not. And that’s I think very disappointing. We talked in the beginning about when I was a kid, and one of the things that attracted me was the idea that we could perhaps know things that were far beyond the spatial and temporal extent of our lives and you could explore the huge reaches of space in some detail. Maybe it’s just a middle-life crisis, but [now] the whole subject [seems] a little childish in that [sense].

Lightman:

So you’d like to see a universe that was accessible.

Turner:

I don’t know if that’s really how you change the universe but in the context of being a professional astronomer, that would be an attractive change.

Lightman:

Let me end with one final question. Somewhere in Steve Weinberg’s book The First Three Minutes, Weinberg says something like that the more that the universe seems comprehensible, the more it also seems pointless. Have you ever thought about this question of whether the universe has a point or a purpose?

Turner:

Yes, I guess a little bit. I remember when I read that, I was very attracted by it. It was intended, I think, to annoy the reader, but in at least my case it failed. I think that my understanding of the point of the universe is probably — if I had to kind of give it a name — close to the kind of Buddhist concept of something that just [is]. Does the water mean to catch the reflection of the moon? It just is. I think trying to impose reasons on it, or points, is too anthropomorphic and just a dream.