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.
We encourage researchers to utilize the full-text search on this page to navigate our oral histories or to use our catalog to locate oral history interviews by keyword.
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 Wallace Sargent by Patrick McCray on 2003 February 13, Niels Bohr Library & Archives, American Institute of Physics, College Park, MD USA, www.aip.org/history-programs/niels-bohr-library/oral-histories/31826-2
For multiple citations, "AIP" is the preferred abbreviation for the location.
Biographical interview with Wal Sargent. Effort made to not focus on topics covered in other interviews (ex: Alan Lightman; interview with Sargent in CHP collection re: Keck Telescope). Focuses on childhood and schooling and various professorial appointments as well as discussion of atmosphere and research at Caltech. Interest in elemental abundances and B2FH paper. Personal research at Caltech including work on QSO and research with Peter Young. Extensive discussion on the use of telescopes including learning to observe and use of Image Photon Counting System in the 1970s and 1980s. Role of serendipity in science and WS's interest in astronomical peculiarities. Also includes discussion of Sargent's personal interests including music and sports (esp. sumo wrestling).
We talked a bit yesterday about the collaboration with John Bahcall and Jesse Greenstein. I wanted to sort of talk a little bit more about your work on quasar absorption lines, and maybe a place to start would be — and perhaps this isn't entirely related to it, but in another interview that I think Alan Lightman did with you for that book on cosmology, you stated that your whole career has revolved around peculiarities, and maybe that would be a launching off point for this.
Yes, am I to apologize for this fact or explain it? [laughs]
No, please explain it. What did you mean by that and has your work revolved around odd things?
It's just that I'm inquisitive about things, and very, I think — not just science or astronomy. And what strikes me always, what interests me always, is idiosyncrasies and departures from the norm, both in people and in science. For example, the peculiar A stars, which we discussed earlier, were an ideal expression for that kind of mentality because there were lots of peculiarities, and if you dug a little more deeply you could find more peculiarities and then you could try to organize them. Then when I started work on the Zwicky compact galaxies, I was interested in finding galaxy spectra that were different than the average; that had some anomaly. And the particular kinds I found were examples of Seyfert galaxies that had not been recognized, which were more luminous and further away on average than the existing Seyfert galaxies, which were all close by. And then, as I said, some of the spectra turned out to be objects that could have been young galaxies; galaxies that have formed only in the past few maybe one billion years or so, as far as star formation goes. It's still not known. And then these galaxies turned out to have very low abundances of the heavy elements, which Searle and I studied particularly in 1972. We found that in pursuing objects that had very low heavy element abundances, which I was interested in from the point of view of finding primordial material to see if there were any gas clouds that were out in the local universe that were remnants of the primordial state. In pursuing that end, we found accidentally that the helium abundance was the same as it is in more evolved galaxies, and that indicated that the helium was already there. It was the first actual observation that indicated the presence of primordial helium from the Big Bang. But that was purely accidental. We weren't looking for primordial helium, and the general idea of trying to find something peculiar turned out to be a possible almost primordial galaxy. Then there was an interesting inference to be drawn from that. The quasar absorption lines were similar because they were a puzzle.
What was a puzzle about them?
It was not known where the absorption lines were formed. Whether they were formed in gas that was ejected from the quasar was something Burbidge and Arp probably still think they still have to, or whether it was due to intervening gas clouds or galaxies or intervening intergalactic clouds.
What's the significance of option A and option B in what you just described?
Well, if it's material that's shot off by the quasar, it only tells you about physical processes in the quasar itself, which are interesting, but of course if the other option is true, which is what we believe, then you can use the absorption lines of quasars to investigate very tenuous gas between us and the quasar; gas which may be itself primordial, which has not yet been contaminated with heavy elements which have been produced in earlier generations of stars and galaxies. So the prospect of studying the intergalactic medium and the regions around galaxies in a quite accurate way was novel at the time, 30 or 35 years ago. It's a line of investigation that I've continued to pursue. The first few years were taken up trying to determine where the absorption lines were produced. One thing you can do is look at the statistics to see whether, as you go to larger redshifts, you get more and more and more things in the line of sight.
So if an object is farther away, one could assume that there's going to be more stuff in the way?
Yes, exactly. That's a simple test. Because if the stuffs ejected then the ejection of objects earlier in the universe should be pretty much as objects later in the universe because it involves black hole physics, which should be constant in time. The observation was that as you went to higher and higher ratios you got more and more absorbers along the line of sight. Then you could study their clustering. So you could study the clustering of an object in regards to the absorption lines, even if you couldn't see what they were.
What's the significance of the cluster?
It would enable you to study the evolution of the object in structure. You could see in principal voids and the gaps in the distribution of absorption lines, and then clustering in the distribution of the absorption lines, as you looked through the rather complex structure that exists in the universe as we know it. So, being able to do that in gas, not just in visible objects, opened up a new possibility. Then continuing the business of peculiarities, a similar thing was doing a survey of about 500 nearby galaxies with Alex Filippenko, starting in the mid 1980’s. (Filippenko has two Ps and one L. You can remember this through the word “flipper”.) We did a census of the 500 brightest galaxies in the northern hemisphere to try and find oddness’s in their nuclei, which could be, for example, a burst of star formation which was peculiar to galactic nuclei, or weak Seyfert activity around the black hole, and this kind of thing. And that turned out to be very successful. It was called by somebody in Britain “a heroic project” which took 17 years or thereabouts to complete.
Did you feel it was heroic?
Yes. It took 50 clear nights on the 200-inch telescope.
So about two years’ worth of…?
Well, two years’ worth of work.
Oh, clear nights probably being the operative word.
Yes, right, and the reason it was doable was that we were observing such bright galaxies that you didn't need to use the dark of the moon, which is where the extragalactic competition normally is. We were only competing with people working on lowly things like stars at the time, so it worked out. We would observe in full moon, these galaxies were so bright. Then in the course of that work we found at least two particularly interesting things that were peculiar. One was a galaxy called NGC4395, which turned out to be the weakest, then, Seyfert galaxy. There the peculiarity was to find something at an extreme, something which Zwicky recommended in morphological astronomy. Zwicky said, “Always go to the extreme to find the biggest thing or the smallest thing or the nearest thing or the furthest thing.” In pursuing extremes, the idea is you will test existing ideas about the universe in a more extreme fashion than if you just piss around with the average object. So it's been one of my methods of working. The peculiar is often the extreme as well. Anyway, one thing was this galaxy, NGC4395, which appears now to have a black hole in it of mass only around ten to the sixth solar masses, and it's the most rapidly varying X-ray source known among extragalactic objects. The reason is that the small things have a short light travel time across them so they can vary more rapidly, and that was one of the things we were interested in. So that came out of the search for the peculiar. The other thing was a new kind of supernovae, which Alex and l found in 1986, which we mistook for the center of the galaxy. As I said, we made obtained spectra with the 200-inch of the centers of the 500 brightest galaxies. In one or two cases with a galaxy that's sort of irregular in shape, it was not possible to see which was the center. We would set on the object and make an on-the-spot decision, “This is the center.” In this one case we saw two centers and put the surface of the spectrograph across both of them. One turned out to be indeed the center. The other turned out to be a supernova, which was in a very late stage of its evolution. It had an amazing spectrum, which we couldn't identify for the first few hours, although by the next day we'd figured it out.
What was so unique about it?
It had very strong lines of neutral oxygen and neutral calcium and ionized oxygen and magnesium, but no hydrogen; no hydrogen at all. And then it turned out to be the late stage of something that whose early stage was already known — that the supernova of Type 1b. Type 1 supernovae are now divided into three classes: Type Ia which are the type that are used for the cosmological measurements; Type Ib and Type Ie, which are different, but they all share the quality that in the spectra when they go off, they don't show any lines of hydrogen. Type IIc are defined to show lines of hydrogen.
Okay, so if you're looking at a supernova and it has lines of hydrogen, it's a Type II?
Yes. It was thought for quite a while that Type II supernovae were exponents of massive stars, and that Type I supernovae were explosions of white dwarfs which had material dumped on them from a companion in the binary system. This is the standard view still today. It was presumably thought that Type Ib and Type Ic were similar, but it is now realized that they're actually also massive stars, and they're massive stars that explode after they've expelled their envelope in winds, which are seen in the form of stars called Wolf-Rayet stars. So when a massive star goes through its evolution towards the end of its life, it ejects a quite large amount of mass. For example, a star of, say, 16 solar masses might end up with maybe 15 solar masses before it becomes a supernova. During that time it's been expelling material. If a star explodes after it has expelled a lot of material from the surface, then of course its later time spectrum will not look like that, or even its earlier time spectrum will not look like that of a Type II supernova because there's no hydrogen left.
Right, it's gotten rid of all of that.
Yes, it got rid of all the hydrogen. And we made the suggestion, having discovered this peculiar object, that it was the explosion of a star that had been a Wolf-Rayet star. This turns out not to be accepted. But again, the intellectual interest was in finding something peculiar. I was not terribly concerned about what the significance was. I was more interested in finding something peculiar, a level 1 somewhat was the idea.
The work that you were doing on quasar absorption lines. You were talking yesterday about your collaboration with Boksenberg and the instrument that he brought over, the IPCS. Was that an essential tool in doing that research?
It was on the absorption lines because it was the best detector for faint sources, and particularly for getting high-resolution spectroscopy. There was another example of peculiarities that occurred very early, which was the discovery of helium-3 in a peculiar star, which was cited in the citation when I was awarded the Russell Prize, although I didn't think it was one of my more grandiose contributions. It was an interesting peculiarity because Jesse asked me to analyze the spectra of a star that had been found to have peculiar spectrum lines at Lick Observatory, the B star. I'd only been doing spectroscopy for a few weeks, and so I didn't know anything about the spectra of B stars, or indeed about the spectra of any kind of star. So I set about measuring this thing, and Jugaku, whom we mentioned earlier, did know about the spectra of B stars, but he'd only studied measurements made by other people. He'd not actually made the measurements himself. So we set about to identify all the spectra lines, and some of the lines in B stars are hydrogen lines, which are very broad, and helium lines, which are very broad because helium is a light atom and at the given temperature helium moves around more quickly than a heavier element. So Alec's first spectroscopies would not have measured the positions of the hydrogen lines or the helium lines because they would've known what they were anyway. And we used to use microdensitometer tracings. But we didn't know anything about the actual spectra of B stars. Even Jugaku would analyze them. So we found that some of the helium lines were in the slightly wrong position for the radial velocity given by the rest of the lines. Then following that, we found that it was because this star had a very large ratio of helium-3 to helium-4. Helium-3 had never been detected in stellar spectra before. And only people who knew nothing about the details of spectroscopy would have found such a thing, I think.
What was your personal reaction to this?
Oh, that this was a great way to go. This encouraged me in my delight in finding peculiar things and studying things like the Brooklyn Dodgers without any knowledge of baseball. It's very similar intellectual activity. So, now Boksenberg. For years his detector was the best for doing the kind of work that I was doing. I tried to get the Palomar Observatory, to buy one. He was willing to sell one for about $75,000, as I recall.
This would have been 1976, around about?
Yes, mid-1970s. He made one that was put on the Anglo-Australian telescope. And I actually went with him on the first run, which lasted two weeks, to install it on the Anglo Australian telescope. We got very good data, actually.
Was it easy to use?
Yes. Well, it was if he and his team were there.
He had a team with him?
Yes, at least two people; an electronics expert and a computer expert.
Okay, so it required a small group of people to…
To keep it happy. Well, the version I used was the version that was being upgraded and changed always. That's why you needed all the people. The one that he left in the AAT worked without any particular assistance. But the curious thing was that nobody else wanted to use this damned thing.
Why?
I don't know. Not invented here, I guess. So for 15 years Boksenberg would come up two or three times a year with his little team, whom I got to know very well, of course. And we would get some time on the 200-inch and produce marvelous data, which produced most of the papers I wrote on absorption lines throughout the late 1970s and early 1980s. As I say, there was no interest among anyone else, and so in this respect and in others I was at odds with my colleagues here, I would say. It's a pattern of behavior that I want to do something different than the rest.
Were there vocal disagreements or was it just a general smoldering dissatisfaction with your colleagues?
Generally a smoldering dissatisfaction, but one of my colleagues saw us observing and said, “This looks like a flying circus to me.” You know, Monty Python was being broadcast around that time. There was an air of lunacy about the whole operation. Of course, they flew in from Britain, so it was called The Flying Circus. But it wasn't an admiring description, I would say. It was more, “I wonder how the hell these guys who appear to be pissing around can get such good results.” I was very good at identifying the astronomy and Boksenberg kept his team amused, and all along the machine was getting better. Anyway, as I said earlier, that ended around 1988 or thereabouts, when even I thought that the CCD detectors now available were better and easier to control than The Flying Circus' machine. I enjoyed the years of The Flying Circus very much.
How often would you all meet on an annual basis, do you think? A couple times a year?
Three or four times a year.
Oh, okay. As you were doing your work with that group of people, what were groups elsewhere that you viewed as the competition?
I don't remember any that were serious competition, because they didn't have — Well, at some point Wampler produced a machine called — I forget what the hell it was called now. It was a hopeless machine anyway.
The image dissecting scanner?
Yes, and it was the closest there was to competition.
This was at Lick?
At Lick, and then Wampler went to be the first director of the Anglo-Australian Observatory and made a copy of it, but the IPCS was superior. The problem with the image dissector scanner was that it used a phosphor, and then scanned the phosphor with something and then the imprint of bright stars would last on the phosphor for quite a while. So if you were observing a faint object, you didn't want to observe a bright object before it.
Because you'd leave a sort of ghost?
Yes. And since what you like to do in this game is calibrate on bright stars while in dusk and then do the real stuff after it gets completely dark, you couldn't adopt the optimum strategy. I've never used an image dissector scanner, but I've heard unfavorable accounts of its properties.
Okay, so you're doing this work on quasar absorption lines from the early 1970s…
The late 1960s, and then I stopped, and then in the early 1970s I went on, yes.
When do you feel the work really began to attract significant attention from outside?
Well it did at the beginning, when Bahcall, Greenstein, and I, and people like Margaret Burbidge and Roger Lynds at other places. The initial discovery of some of the physical properties attracted attention, and then there was a fairly long pause, a hiatus, I would say, because at that time Roger Lynds could get the best spectra for a few years with his Gold Spectrograph. And then Roger would carry Polaroid pictures of the spectra around with him like dirty postcards and pull them out of his pocket. There was very little publication of the data. His machine, while good, was very hard to make quantitative. It still used photographic plates as the final detector. It amplified the light and then used photographic plates. Then I think in the second half of the 1970s, when Peter Young came to be a student, it took off again, because that coincided with us getting really good data from the Anglo-Australian Telescope in 1976. As I said, I went with Boksenberg and we installed the detector and then 11 out of the 14 nights were clear, which set a record which still holds for Australia, I think. The thing worked on the first night.
That's unusual.
Yes. Well, we'd already been using this kind of machine for several years at Palomar, so it was only a question of installing it on a different spectrograph and finding problems with that particular spectrograph. Anyway, we got also superb data at Palomar, and Peter Young became my student. I take it you've heard of this person, haven't you?
No, I haven't.
Oh, well you asked me who I thought were the best observers. The best astronomer I ever met was Peter Young.
What was so great about him?
He combined amazing ability as a theorist with amazing interpretive ability and knowledge of physics. He wasn't really outstanding as an observer, but he was good enough. Okay, so this guy finished in 1975, probably, the best mathematics student in Cambridge. But he was interested in astronomy so he came to the United States to do a Ph.D. He spent a year in Texas because he didn't know what the good places were. Then he came here. The upshot was that he got his Ph.D. in 18 months and was really amazingly outstanding. As an example, you're supposed to take courses when you're a graduate student in American universities, and he didn't want to take some of the courses. In particular, he didn't take the interstellar matter course, but this was expected. So at the end of the term, the head of the department, which may have been me at the time, said, “Okay, you've got to pass the exam even if you didn't take the course.” He said, “Okay, give me three days,” and three days later he got a perfect score on the subject that he'd not studied before. So he was pretty amazing. And then for some reason he wanted to work with me, and I could never figure out why, because I think our intellectual abilities were very, very different (i.e., he was incredibly more intelligent than I). So I've often wondered, “Why the hell did Peter waste his time with me?” I think the answer was that I was doing really interesting work, and he had the sense to recognize this. And I'd had very good students, like Huchra was finishing about that time, for example; and Ed Turner; and Steve Kent, who is now with Fermi Lab; and Pat Osmer had been my student. I had very good students. They're listed on my website. But Young was by far the best of all of them; better than Steidel.
So why haven't I heard of him?
Because he committed suicide when he was 27 years old.
I see. Was he living here at the time?
Yes, in 1981. He got his Ph.D. We kept him on as a post-doc for a year and then he was made an assistant professor. What we should have done was send him someplace else. But he committed suicide on September the 5th, 1981, in the next room, actually. He was found lying, having poisoned himself, on the floor. And there'd been intimations of psychological difficulties before this, which I don't want to go into. That would be a separate story, I think. During the time he was alive and here, which was around five years, he wrote 33 papers.
In how many years?
In five years, 33 papers, which are of the highest class. 20 of them were with me, primarily on the absorption lines, although we did some other stuff on cataclysmic variables with Boksenberg. At the same time, he was working on the theory of black holes in galactic nuclei, and he worked with Boksenberg and Lynds and me on the discovery of evidence for a black hole in the center of an 87, which we published in 1978. The astronomy and the spectroscopy, and to some extent the theory, came from me, but the push came from Young.
How did that personally affect you?
I was extremely concerned — well, upset and downcast for a long time. During that time I was elected to the Royal Society and was recognized, but I didn't get any pleasure out of it because of Young's death. Not that we were so personally close. I've been much closer to most of my other students because I normally try and learn from them and also teach them something. We would do things like read novels or history or something together and then discuss it on cloudy nights. It's very easy to do that. And I would try and teach them about classical music if they didn't know about it, and they would teach me about rock music. For example, I didn't know anything about the Beatles, although I'd heard of them, until… my students told me about them. Young liked pop music, and I had very little effect on his intellectual development, I regret to say. The one area of life where we had some overlap was soccer, where he supported Leeds United as opposed to Manchester United, so we would talk about this sort of thing. For the last year that we worked together, he was having psychological problems.
You said that there were some indications.
Yes, I learned later.
Was he married?
No, and the final suicide involved a young woman. He threatened suicide for about a year before he actually did it. He told Peter Goldreich and me once, towards the end, that he'd bought a gun. We tried to take the gun away from him by a subterfuge, which was a pretty bloody stupid thing to do. We tried to get the person who'd sold it to him to take it back. We said, “This person's suicidal,” and the shopkeeper said, “Well, that's not my problem.” Gun ownership in the United States.
Yes, that's a very American attitude.
Anyway, for a whole year I was in pretty bad shape, thinking tomorrow or something, that is going to happen, and then finally it did.
How did it affect the Department, as you saw it?
I think there was an air of gloom, but nobody else, apart from Peter Goldreich, had been really close to Young. The students who knew him, the people who'd been his fellow students whom he'd passed by intellectually by finishing in a year and a half when they were struggling to finish in five years, I think really suffered. Don Schneider, who's a professor at Penn State, was a great friend of Young's. The curious thing was Don knew nothing about the psychological difficulties, whereas Peter and I knew a hell of a lot about them. Anyway, I discovered by chance on a visit to Cambridge many years later that he'd had a similar breakdown twice while he was an undergraduate, but they never told us when he came here as a student. It was a protection of privacy, I suppose, but there are ways in which you can tell people without it being official, because I knew the guy who knew who knew this, the head of the mathematics tutor at St. John's College, Cambridge, where he'd been so brilliant. I was sort of peeved that he didn't give us a warning, because we could have done things differently. Anyway, as you asked, what was the thing that we vitalized in the study of absorption lines in the 1970s? It was him coming as a graduate student, and then the excellent data from the Anglo-Australian telescope. Things would happen, like, I remember once I went down to Chile because Steve Schectman, who is over at Carnegie, saw the IPCS and built a cheaper and simpler version called the schectograph. It used a reticon, which is a two-array detector. I helped him pay for this, and so I would go down to Las Campanas and use it. One time I went down there for 15 nights with Boksenberg, actually — for a change we used somebody else's detector — and all of them were clear. I got lots of very good data on absorption lines and also on cataclysmic variables. In a cataclysmic variable, a small star like an M star is going around a white dwarf and then material is being dumped from the more-or-less normal star on the white dwarf. The period is typically a few hours, so you can follow a whole revolution of an orbit in, in this case, four hours. So what we did was to get spectra every 120 seconds for four hours on this, which had not been done before of a particular cataclysmic variable.
So you can see changes in...?
You can see the actual plume of material from one star as it goes on. You see it from different aspects as the orbit goes around. It's really beautiful stuff, actually, from a spectroscopist's point of view. We got into that because the IPCS was ideal. In any case, I came back with 15 nights' worth of data, put the tapes on Young's desk, and then I collapsed for about two days. When I came in, on my desk there were all the reduced data, and reductions are not easy, actually. They required a very good knowledge of the primitive computers at the time. So he was a tremendous data analyst as well as a tremendous theorist. The papers I wrote with Peter were about the best that I have written, I think.
Do you have a particular favorite?
Well, two. There's a paper on the Lyman-Alpha Forest of 1980, [1] which really worked him, actually. I once went to China and somebody — it was a group of astronomers from some place in interior China said, “It took us six months to understand this paper,” and I said, “Well, can you imagine how much; how long it took us to understand it?” If you read it carefully, you can see the hand of different authors. That is, there's one calculation that is given twice in the paper with different answers. I actually put the whole thing together in the end, and I was out of it for, well, maybe even years afterwards. The intellectual effort was so extreme.
What was the reaction to the paper?
Oh, it went down very well. It's been cited more than 500 times, I think, which is a lot. Yes, it's the paper which I think, for most people, brought the Lyman-Alpha Forest into existence, and people like Ostriker and such started to think seriously about theoretical work on it.
Why is the Lyman-Alpha Forest important?
Because it gives you a view of gas in the intergalactic media and in the halos of galaxies over the whole history of the universe, as long as there have been galaxies and as long as there's been intergalactic matter. You can study it through the absorption lines. My other favorite paper was the one on the black hole in M-87 written completely in 1978, where Boksenberg and I got the data at Kitt Peak with Roger Lynds. As I said, I had worked out how we would handle such data using stellar dynamics, because I'd worked on the motion of the globular clusters in our galaxy to try and get the mass of our galaxy. So I knew about the Boltzmann Equation and all that kind of thing that you have to use. The moments of the Boltzmann Equation. You derive the equation in spherical polar coordinates, which you need to use to get masses. Actually, my theoretical ability is not all that bad compared with most of my colleagues, I would say, but it was weak compared with Young. In any case, I had the equations and then Young reduced the data and fitted them to the equations that we'd developed. We found there was evidence for a black hole in mass three times ten to the ninth solar masses, which the space telescope confirmed around 1991, more than 10 years after the original start.
So a good vindication of what you suggested?
Yes, it's recognized.
One of the things that I've noticed in what you've been describing, observing runs with a large number of clear nights and things like that, how important has serendipity been to you as you've been doing your research?
Well, extremely important, as I've indicated with respect to the peculiar things that pop out. I tried always to make my students flexible; that is, not to be too rigid about what they were trying to find out and to be prepared to be diverted from their initial goals if they find out something more interesting. So I think in astronomy, in this phase of astronomy, which I predict will last for at least another 100 years. We don't know enough to specify all of the questions very precisely and so we have to rely on accidental explorations, exploratory projects, which result in serendipitous discoveries.
That opens up a whole line of questions about how research should be done, because with observing proposals and requests for time with the space telescope, it seems that there is a certain amount of management and preplanning that goes into it that runs counter to what you're saying. I'm wondering what your reaction to that is.
I think it's bad, and I think the existence of the great private observatories in the United States is a counter to that trend, which is always being eroded, but I try to make the erosion as slow as possible. Of course, sometimes when you set out on an exploratory project it can fail, and so you don't discover anything, except you've got spectra of 200 more galaxies or something. In general, I think that people at a place like this have a duty to have a grand vision; not necessarily a grand vision of a particular problem, although I think that's fine, but also to embark on projects which are large in scope, which if you're lucky will result in something new and interesting. It's making the choice of those that's very difficult.
How does this compare, then, with the type of research project that would be undertaken at a national facility, for example?
Well, you have to keep applying for time. Here you can be pretty sure, or at least in the good old days you could be sure you'd get your 25 to 30 nights a year, and so you could recon on that and work accordingly. Nowadays there's slightly more of an attempt to control, but it's not terrible. In the national observatories you pretty well have to specify the project in such detail that you almost know the answer before set off. I've been on the time-assignment committees at Kitt Peak and Cerro Tololo, for example, and also on the space telescope time-assignment committee. And getting through exploratory projects is very difficult because there are many very well written proposals which promise a result. So you think, or even I think, “Well, hell, this other thing might not pan out so let's do the thing that promises a result.” The temptation to do that is always there, but I think it's a bad system. I also criticize some of my colleagues here for doing things that other people have done, but slightly better, or doing things that other people are doing and trying to scoop them because we have more telescope time. I don't approve of that kind of science. I think we should be aristocratic socialists and try to do something grander and more ambitious. Of course, when push comes to shove, the only way you can select such projects is on the track record of the proposal. I think the track record of the proposer should count for a lot, but then, of course, you run into the possibility of favoritism. “Oh, boy, that works,” etc. “This guy did something important 50 years ago so he needs a lot of telescope time to try and do it again,” which of course is not a good idea.
Right. The history of astronomy in the US versus a place like Great Britain is so different because you have this strong private and public tradition in the last 50 years, for example; the public observatory system and the private observatory system. I'm wondering what your thoughts are in terms of that being an advantage and disadvantage versus a place like Canada or the UK, where you really only have a public system.
I think both Canada and the UK have done very well, and they've managed to somehow reap the benefits, particularly in Canada, of the public system while operating it a bit like a private system. For example, the Canada-France-Hawaii telescope does some very good stuff, because it now does, as far as I can tell, large projects, and usually these are well chosen. Similarly, the Anglo-Australian telescope, even though it's in a shitty site where it's cloudy all the time and the seeing is poor, has done really well by comparison with U.S. observatories, by concentrating and by having clever people, since the Brits are intrinsically more clever than people in other countries, by having clever people and concentrating. The Continental Europe, I think, has done less well.
In terms of the European Southern Observatory's record?
Yes.
How come?
There were not experienced people there, to start with, using big telescopes. Even in Britain there'd been this tradition of going to Australia or South Africa if you really wanted to be an observer. There was no similar European tradition. Even though astronomy was very well developed in Holland, it tended to be either in recent-after the war, radio astronomy, on the one hand, or analyzing other people's data in an intelligent way, on the other hand. This is how Kapteyn worked, for example, and to some extent Oort. There was no tradition of actually using telescopes in Europe in the “big time.” That is, in the very good climate and big telescopes and modem detectors and all that kind of thing. That's been very slow to develop. And the present system is far too bureaucratic. It's much more bureaucratic than the national observatories here. I've been to the old ESO, but I've not been to Cerro Paranal or seen the new telescope. As I understand it, you put in a proposal which specifies the observation in great detail; exposure time, the position angle of the slit across the object, everything. Then the observation is executed by some professional. You can be there, but you're not allowed to interfere very much, as I understand it. In real astronomy, when you set on an object, particularly a galaxy, you're getting the first good look at it, particularly the central regions because usually the central regions are overexposed in pictures, particularly the old photographic pictures, so you look at it in the first look. So if you've decided to have the position angle across this direction, you might see something over here and you want to have the position angle over in the other direction. You can make that decision on the spot at what I would call reasonable observatories, but I think you have to fill out a form and have it approved at ESO.
Which brings us back to that serendipity question that we started to talk about? I have some questions about Keck, and no so much in terms of its planning and construction, because that's been covered elsewhere, but I wanted to ask you what you see Keck's most important contributions as being.
I once listed them in connection with a talk I was giving on CELT, from which I have now been expelled, so it's not been close to the top of my stack. I think I was in a bad mood and threw away a lot of the stuff.
You've actually been expelled?
Yes.
What do you mean?
I was director of Palomar for three years and I was asked to stay on, and then after about three months I fell out with my boss, which was the division chairman of the —
Who was that?
It was Thomas Tombrello. He's a physicist. I quit a few hours before I would have been thrown out anyway — I decided to make the first move — in disputes about our relations with the University of California, which I don't want to go into. It's too recent; the year 2000 although I was one of the people who started CELT, and indeed invented the name…
It's a good name.
Yes, it's particularly good because Anneila is, of course, of Celtic origin-Scotland and Ireland. Some of my ancestors came from Ireland, and so I thought CELT was pretty good. It went with Caltech Extremely Large Telescope, and then California Extremely Large Telescope. Carnegie and stuff come in; they were also in California and began with a C, so it was all worked out beforehand. Okay, so what were the most important developments? I think they were mostly serendipitous. In 1984, the people who were designing Keck produced a thing called a blue book which is design requirements for a ten-ten meter. Let me just see if I've got it here.
Sure, I'll pause this for a second. Okay, so this is the blue book?
The blue book, the design of the Keck Observatory and Telescope, and then some place in it there's science; a list of things to do. I looked at this at the time when we were thinking about CELT, and I found that in some respects they've been pretty prescient, and Sandy Faber was one of the authors, I think. They'd followed, which I guess you have to do in this business, they'd extrapolated from what was already known, which Zwicky would not have done and in a similar situation I wouldn't have done either; not to the same extent. To please politicians, the people who give money, you've got to say you're going to do particular things, and these particular things normally are embellishments of things that you've already done. In this case, one of the things, for example, was studying the evolution of galaxies out to a redshift one, which is about halfway back in time to the Big Bang. It's a fine thing to do, but I would never have written that down because it's too well defined. In fact, before much had been done on that project with Keck, Steidel came up with a way to find galaxies in large numbers at a redshift of three, so why piss around with galaxies at a redshift of one when you can go much further? Indeed his technique would be adapted, in ways that had not been foreseen, to find galaxies at lower redshifts. They are plodding along with that proposal, that piece of work. It now takes the form of the DEEP project, the Deep Extragalactic Evolutionary Probe, which is headed by Sandy and Marc Davis. I've been, from time to time, interested in joining it, but I don't like large teams where I cannot be the dominant figure. I don't like doing science which is so well prescribed. No doubt they'll find out things which they had not expected, but it didn't seem to me daring enough to justify the construction of a telescope; as it turned out, two telescopes for $200,000,000. Okay, that was one example. There were others, all of which were sensible, good things to do. In fact, I would say the main things that have been interesting were David Tytler's work on the deuterium-to-hydrogen ratio in the Lyman-Alpha clouds, which gives you the primordial value because the material has not been through stars, as we understand, which destroy deuterium. This gives you the barium density, which agrees, it turns out, with the value determined by completely different methods from the cosmic microwave background radiation. It's the peaks in the power spectrum. That was one very important thing, which was not mentioned in that book, although people including me had already tried to find deuterium in quasar absorption lines before this. The work on the first brown dwarf which had been discovered at Palomar, but the work on the spectrum was done with Keck (very fine work that Kulkarni did, that wasn't mentioned in the blue book despite the fact that theoretical possibility, the existence of brown dwarfs, had been known at the time.) So that's number two. Number three was Steidel's spectroscopy of galaxies at a redshift of three. The spectroscopy couldn't be done with any other telescope because the objects were too faint. As I've said, that wasn't anticipated, that you could go to a ratio to three. There was another, the discovery of the extragalactic nature of gamma ray bursts, which was done with a Keck spectrum of a quite faint object. Gamma ray bursts were known in 1984. They've been known since 1970. It had been speculated from time to time that they were extragalactic, not galactic, objects.
That would be Pacynski's work?
Yes, Pacynski's work, but nobody thought, “Well, will the optical flashes be sufficiently bright to photo-detect with a big telescope?” Nobody suggested that the galaxies that they might be in would be detectable. There are several examples where I think Keck has made very important contributions which were not anticipated. The work on the stars around the galactic center, which Andrea Ghez has done and similar work is being done by Genzel at the VLT, was not anticipated. It has been extremely important. If anybody had told you in 1984 that you would not only detect stars near the galactic center but also see them moving in the sky at almost 10,000 kilometers per second, you would have been laughed at.
Right. Is that the most surprising result for you that has come out of the two Keck telescopes?
I would put the gamma ray burst business, Steidel's business on galaxies at a redshift of three, and the motions of stars near the galactic center as being the most surprising.
Have you been pleased with how the Keck telescopes have functioned?
At one level, yes, and at another level, no. The fact that this crazy scheme worked at all was very pleasing because, until Keck actually came on-line, I think it's pretty well true that nobody outside the small community of people who worked on it were confident that it would work.
In terms of the segmented mirror?
Yes. My friend Boksenberg, who is no slouch when it comes to instrumental things, didn't believe that the system would work. He thought, as did others, that if there was wind or gravitational disturbances, that waves would cross the mirrors and they would move relative to one another in a way that was uncontrollable. Those of us who were really familiar with the details of the engineering and how the sensors worked and the actuators and all of that were confident that would not happen.
Did you have any reservations at the outset that this was a crazy idea?
No, because when we joined the University of California in what turned out to be Keck in 1983 or 1984 or thereabouts (1983 we had the first discussions), we looked very carefully at what Nelson had done. We had engineers like Bob Leighton, who built the various dishes and whatnot, look at the scheme in detail. We went to Lawrence Berkeley Lab and saw the lab tests of the various prototypes of the components and all of that, so we were —
Jerry Nelson was a Caltech graduate, too?
Yes, he'd worked for Neugebauer as an undergraduate on infrared stuff as an undergraduate, although he did his PhD in high energy physics and realized that he didn't like working in large teams on boring subjects.
You said, I guess, if I understand right, at the technical level, you're happy at how it's functioned. What's the other one?
We've not done well with the instrumentation. The instrumentation was managed by a committee that I co-chaired for many years, and we were not used to managing projects of that magnitude. Instruments cost around $4,000,000, and sometimes a lot more; sometimes a bit less. But astronomers were just not used to managing projects on that scale. Gerry Smith, the Keck project manager, refused to manage the instruments and the telescope construction. He wanted to concentrate on the telescope. Let the scientists piss around with the instruments. He would have managed things better but his plate was full. We should have had professional managers. The instruments, while not completely satisfactory, have produced amazing results, without question, partly because the astronomical community that uses Keck is more experienced than the astronomical community in some other places. The guys who go down to ESO or who are going to use Gemini are often not able to improvise with their instruments.
You were saying that you're more used to improvising.
Yes, the people who were brought up around Lick, Palomar, and Mt. Wilson are used to improvising and dealing with equipment that's not meant for the hoi polloi. So we compensated by having instruments that were sometimes not engineered pretty well, but by having a skilled cadre of users who passed on their skills to other people as time evolved. But then the actual telescopes have not proven reliable in the actual operation, I think.
Which parts of them?
Particularly the domes. The sad thing is that the high tech stuff, the positioning of the mirror relative to one another and all of that, works pretty damn well, particularly now. It doesn't break down. To start with, there were problems with the software, but not so much with the hardware. Now there are not even problems with the software, but the dome shutters, for example, have been a consistent source of failure.
Which is odd because…
It's an old art. So my conclusion is that old-fashioned heavy engineering is going out of style. People use computers to design things, whereas what they used to do was guess what the loads would be and then double their estimate.
Yes, sort of the eyeball engineering that went into building Palomar, for example; that type of thing.
Yes, that really works. Palomar works night after night after night without problems. The dome shutters never get stuck. The same design couldn't have been used on Keck, although with Keck domes they were about the same size because the dome shutters stick out a long way, and in heavy winds they would be cause of problems. The winds are much higher on top of Mount Molokai, and so the up-and-over system, which is used on other observatories, was the sensible one to try and use, but it turns out not to work very well. So the upshot is that we don't have enough money for maintenance and fixing things, and we don't have enough money for instrumentation, but it works very well when it works.
Do you use Hubble much?
No; some, but just a little. I was on the Key project on quasar absorption lines with John Bahcall and returned to Bahcall's company after 30 years and found that I still did not like it, just at the time when we were getting the first Hubble Key project data, which was delayed because of the Challenger accident, and was delayed — Well, the data were poorer than we had anticipated because of the problem with the telescope. Just at that time I started getting really good data from Keck with tremendous resolution with signal to noise ratio that would have been unequal, and so I didn't really put my heart and soul into the Key project.
What do you think the effect of Hubble has been on the astronomy community as a whole?
Well, probably beneficial as a whole. It has a fundamental problem that it's a small telescope. It's only 94 inches. To do spectroscopy, which, as you may have gathered, is what I like to do — I practically never take pictures — is not all that great. Of course, it can take spectra in wavelength regions which are not accessible, but the spectra of quasars that come out of the key project are not great. More recently, the spectra obtained with the STIS (Space Telescope Imaging Spectrograph — you can take pictures with it or you can get spectra) of resolution which approaches that that you can get with big ground-based telescopes, but then are so few bright sources in the sky that it hasn't made much of an impact, at least on things that I do. The pictures that you see in books and newspapers that space telescope has produced are often quite amazing as pictures, but I regard them as art rather than science. They show you the complexity of regions of the universe that I would not wish to enter into because they're just too bloody complicated. The one exception is the Hubble Deep Field, which I think has been a great benefit to extragalactic astronomy.
Why?
For one thing, it provided, as you know, better pictures of objects in the distant universe, like Steidel's studies, for example, but also the pictures could be taken in several wavebands, including those on the ground. It led to the first really serious attempt to get photometric redshifts, where you estimate the redshift of a galaxy by using its colors, but in many bands. The initial claims on doing this, by some people, were pretty hysterical. For example, it was suggested that objects at redshift nine appeared on some of the pictures because they were only seen in the reddest band and not in the other bands. Anyway, silly things were said which many of us recognized. After the hoopla had died down, people did begin to seriously estimate the redshifts of galaxies just by measuring them in these broad bands where you could go very, very faint; fainter than you can do in spectroscopy, because the losses in the spectrograph are greater. So that was a great benefit. It's possible to estimate the distribution in redshift of galaxies at the Hubble Deep Field without getting individual redshifts for each object.
I have a couple of general questions about Caltech because you've been here for a while. How have you seen the institution change under the various presidents? When you first came here, DuBridge was president, and then, I guess, Harold Brown. I'm probably skipping a few people, but Goldberger and then Baltimore — maybe you can't comment because Baltimore's still here.
Everhart and Baltimore. I think it was ideal under DuBridge. As you know, the Institute works on a division system where there are six divisions, of which physics, mathematics, and astronomy are one. The way it was set up and initially operated, great men were in charge of each division and they reported directly to the president; DuBridge, notably. When I came here there really were great men in charge of each division: Pauling was in charge of chemistry; Beadle, who got a Nobel Prize, was in charge of biology; Carl Anderson, who got a Nobel Prize, was in charge of physics, mathematics, and astronomy. The engineers were always slightly below the level of their peer scientists. Humanities had a good guy in charge. The geologists had very eminent geophysicists in charge. So there was a lot of paternalism as a result of this, which was entirely good in my opinion. There was no need for democracy. People could concentrate on doing their work. As far as I can see, there was very little friction because the great men sorted it out. Then over the intervening time period, sciences got more bureaucratic, generally. The institute has gotten more bureaucratic because there are more government regulations to observe. There are fewer sources of funding. The good old air force no longer supports astronomy, except maybe in some very small way. The institute has hardly gotten any bigger as far as the faculty is concerned, and the student numbers are about the same, but there are many, many more post-docs. The number of post-docs has probably doubled or tripled between when I came here and now. That's because various subjects, particularly biology and chemistry, require a lot of cannon fodder. That is, the work involves intense specialization. Person A knows how to do this and passes on the result to person B, who knows how to do that.
Very much an assembly line type thing?
Yes. They've even lengthened the time that you can be here as a post-doc from three years to six years. Even people whose opinions I respect say that you can't learn enough in three years to go on to be a professor someplace. I think what they want is more cannon fodder — cheap labor. I think the general fame and prestige of the division chairman has gone down; not uniformly, but now it's much more spotty. As regards the individuals — When Harold Brown came, people were very apprehensive because he'd been Secretary of the Air Force. He'd spent all of his life in military science, but he turned out to be fine, actually. He was a very cold sort of person. He didn't have very close relations with members of the faculty that I knew about. I don't think he had close relations with anybody, actually; probably not even with his wife. He was a very good manager and in fact, while he was here the number of bureaucrats was reduced. When the Institute found itself, as it always is from time to time, in financial difficulties, for example, after they stopped the military supporting pure science, there was…
The Mansfield Amendment.
The Mansfield Amendment. As I recall, that was in 1968. I think it was about the time that Brown came here. His response was to cut down on the number of bureaucrats, whereas what's happened more recently with Everhart and Baltimore is that the number of bureaucrats is increasing exponentially, they claim because of government regulation. For example, I think that at any time on the campus there are five resident government auditors who have to be supplied with paper and whatever else they need every day, etc. Up at JPL there are even more government auditors, of course. That sort of atmosphere is different to what it was, to other systems. I still think that DuBridge had the most profound effect on the place.
How come?
After the war he made very wise choices in hiring, at least in fields that I know about; for example, getting Feynman to move from Cornell, getting Gell-Mann to come here as a young man, and getting Jesse Greenstein. New subjects were started. Generally the choices were very good, and even the choices of things that were being continued were good. A place like this has to choose what it does; it can't cover all fields of science, of course. For example, in physics, Caltech is a very well-known physics department, but its reputation in solid state physics is really its weakest point, I think. From time to time there are attempts made to redress this situation, particularly now that Gell-Mann has left. He used to call it “squalid state physics”. He had a low opinion of it. I think in DuBridge's time, generally wise choices were made, like, for example, starting ratio astronomy, which Jesse pushed, but of course he needed the support of the upper end administration as well.
Do you have a favorite DuBridge story?
No, I don't have any. Well, I only have one thing that I remember about DuBridge, which showed me his powers of leadership in a curious way. There was a student concert over in the Beckman Auditorium, I think. It wasn't very good, but we'd gone along. At some point a piece stopped but the audience didn't know whether to applaud or not, so DuBridge started the applause and everybody then applauded because DuBridge was. That was a very small thing, but he showed that even though he was in a situation that he didn't really understand (I think he was not heavily into classical music) at least he knew the right thing to do. It was a very tiny thing, but he was like that, actually; very good at personal interaction.
How did the university change when women were admitted as students?
Well I think it got better. I think the first women undergraduates were admitted around 1970, but there were one or two women graduate students when I came here.
I'm trying to think, in this program, Virginia Trimble was here.
Yes, she's the first one you think of. She came in 1964 and graduated in 1968. She was here when I first came on the faculty. Judy Cohen was a student around that time, and my wife was a student around that time. Jesse was very good at promoting the cause of women, actually, although I don't think he ever intended to. I don't think it was ever a conscious thing, but maybe I'm wrong.
Was he a liberal person?
Yes.
I mean politically liberal.
Yes, very. Well, by American standards, very liberal. He came from a rich family and he had a fair amount of money, but he was definitely a liberal democrat in political views, and he was socially tolerant as well.
Okay, so he wasn't, to use the current popular phrase, a “knuckle-dragging ape” then, when it came to trying to keep women off the campus and all of that?
Oh, no.
Did you see the integration of women into campus life — How did that go? Was it a smooth or turbulent transition?
I think it was, for them, probably difficult because there were so few of them. The first year of women undergraduates there were four, and I know two of them still. They, to my surprise, look upon those days with great affection, actually. It's surprising. But I'm sure it was difficult for them because there were so many men around, and the men are selected to be socially inept.
I guess there's a special test for that.
Yes.
Okay, I just have a few very general questions. These relate more to your views about major changes in physics and astronomy. However or narrow or broad you wish to construe the question, what do you see the major changes being since you entered the field?
Of course, there's a big change in the style of doing science, which involves far more people, which is necessary, although sometimes it's taken to excess. I think a good think has been the ease of propagation of scientific results which exists now as compared with the old days. The journals used to come out every few weeks, maybe once a month. You can see by looking at that.
Yes, I was looking behind you. There's a wall about ten feet high and 30 feet long of journals.
The great thing then was to get preprints, because the preprints would arrive long before the journals. So there was a sort of elitism of preprints, where prominent scientists would be on the preprint lists, but if you were off in the boondocks someplace you wouldn't get the preprints.
I wasn't aware of that. I thought it was a more democratic distribution.
No, and so I would get letters, say, 30 years ago, “Dear Professor Sargent, will you put me on your preprint list?” As an aside, the cost of sending out preprints was quite large, and particularly when money became tight it was no longer possible to send them everywhere to everybody, but if somebody sent me a letter I would send a preprint to India or someplace like that. Therefore, the advantage of being in a place like this was doubled because you were in touch with all the latest stuff. Somebody from Princeton would come and give a talk or something and you would get the preprints, because they wanted to impress you. Now, of course, the web is accessible to everybody. As a result, everybody gets the same information. Sometimes something is hidden from the general world, but in a particularly small community there's not much of that, actually. So I think the democratization of the data and the knowledge has been a really important feature. I'm surprised you're not familiar with this.
I am, but yours is an unusual perspective.
Perspective.
Right, but more frequently people, depending on their creativity, will say, “Well, computers or things like that.” The most interesting response I've gotten to that question has been jet travel, from a more elderly person, who thought that the idea that you could now travel from Harvard to Hawaii in no time had really changed astronomy.
There, I disagree. Again, because when Zwicky died, he left me some of his precious possessions, one of which was George Ellery Hale's diary for the year 1901. I don't know how the hell Zwicky got it, but he didn't even know that it was Hale's diary.
Do you still have it?
No, I gave it to the Caltech archives. The point is that I read this, and Hale's handwriting was not all that good so I had to spend a fair amount of time on it. He was living in Yerkes at the time, and what I found was he had grand plans already for building observatories. He would travel all the time. He would take the train from Yerkes to Chicago. He would take the train from Chicago, then to Cambridge, Massachusetts, or to Washington. Occasionally he would travel all the way to California. The upshot was that Hale spent about as much time traveling as astronomers do now. It was in trains, and you didn't get there as fast and you didn't go across the world. Then he records that some astronomer came back to Yerkes after the eclipse in Indonesia, which had been six months before. It took longer but you took longer over it, and so I don't think it's made all that much difference. People are slightly more fanatic. A train, I think, is a much more peaceful form of travel.
No doubt, yes.
Airplanes could be made equally peaceful if they gave you more space, but in airplanes you're not distracted by the view out of the window. So I don't think that's made all that much difference, based on that insight to do with Hale. The computing business, of course, has changed things.
Again, the reason, I guess, my reaction to that is being rather a banal answer was it's changed every aspect of life, and I don't really know if it's had any particularly unique impact on the sciences that it hasn't had, say, for example, in making airline reservations.
Yes, you can do it more easily. It changed the nature of the division of labor in a way which I think is quite interesting. When I came here, the observatories employed computing assistants to work the old-fashioned hand calculators.
Back when women were computers?
Yes. Santa Barbara Street (i.e. Carnegie) had some women, and Shapley at Harvard, etcetera, and we had two or three people in this building. Then, of course, when real electronic computing came in — even when I came here it was used a little — gradually these skills became less necessary; the old-fashioned calculators. The people who learned computing were the young scientists, of course, and so they would do it for themselves. But now we're back at a stage where what I need most of all is somebody who is familiar with computers, has no ideas, and will do what I tell him to do. So that circuit we've not quite gotten around. There isn't a recognized role, I think, for the educated but un-ambitious computer assistant.
I have to ask about the sumo wrestling. When did this come about, and why?
Well, when I worked with Jugaku in the years 1959 to 1962, through him I got interested in Japanese life, and he taught me about the tea ceremony and haiku. I actually like poetry. I actually read poetry quite a lot, although I have a restricted group of poets whose work I read. Anyway, haiku were an interest to me, as was generally the nature of Japanese society, which is very different than Western society, as I discovered, even from Jugaku. Jun never mentioned sumo. I knew it existed, and I later learned that he didn't know anything about it at the time. He thought it was rather beneath him. So Japanese interest was planted in me. I never went to Japan, and then around 1975 I was tuning through the television channels one evening and I found sumo. At that time there was a program called Sumo Digest, which gave you a summary of the bouts of that day in a tournament. It came on around 11:00 at night, and there was a professor from USC who taught Japanese at USC, who provided a voiceover. That is, there was a Japanese commentary in the background, but it was this guy in the foreground. He explained it as the bout proceeded, and I was completely hooked because it was a bit like my earlier introduction to baseball, where you try and understand a certain — Well, the infield fly rule caused me great difficulties when I was listening to commentaries and didn't know exactly how the field was shaped. The announcer said, “The second base dropped the ball and the guy's out,” etc. After a while I figured it out. I figured out the reason for the infield fly rule, which has to do with getting one person out rather than two, which you can do by deliberately dropping the ball. Anyway, sumo was a bit like that, but then the ritual impressed me. It's extraordinarily ritualized. Just to give you a few examples, there were different grades of sumo. In the top grade, the Makuuchi division, the umpire wears slippers, but in the lower divisions he doesn't. They wear similar outfits, but only in the top division wear the gi. After a bout, the wrestlers remain in their comers until the two new wrestlers step into the ring, and as they step into the ring they're offered a scoop of water out of some sort of receptacle, but only the wrestler who has won gives the water to the person in his comer. In the other comer, the one who has lost doesn't give water because he has lost and therefore it would be bad luck. There are all kinds of little rituals like that that are a microcosm of Japanese society, I think, where ritual behavior is more important than it is in our society.
Are there any parallels between astronomy and sumo wrestling?
Not as far as I can tell. So I followed it for all these years, initially with Todd Boroson, who is now director of Kitt Peak. He was an undergraduate here. We wrote some papers together while I was an undergraduate. We read novels together, particularly Anthony Powell's A Dance to the Music of Time, which is 12 volumes. We read these together during the time he was an undergraduate, and we would watch the sumo at night, him in the student houses and me at home, and then discuss it the next day. I follow it now. I can now follow it much better on the Internet, because I can actually see little recordings of the bouts on the Internet.
Have you been to a match?
Yes, twice. There was an exhibition at UCLA around the late 1970s, actually, and Todd and I went over there and we found ourselves sitting next to Hal Zirkin, who is a professor of astronomy here who works on the sun. He'd been to Japan several times and he also got hooked. The other one who's hooked is Maarten Schmidt.
I never would have guessed. That's surprising, because Maarten has this very — not that you don't, but he has a demeanor that's slightly different from yours.
Yes. Well I introduced sumo to Maarten and he's very interested.
That's really interesting.
We often discuss sumo when we're sitting with our coffee outside. So I saw this exhibition at UCLA, and then about four years ago I was asked to go and give a lecture in Japan for $1,000 plus my fare, etc., to the best science graduate students in Japan; biologists, chemists, not engineers, but pure science. Anyway, I said I would go and give the lecture. There were three Western lecturers and three Japanese, so they had six lecturers. I said I would go and do this if the guy would take me to a sumo stable. It's called a beya, where the sumo wrestlers practice. Particular sumo wrestlers are attached to a particular stable, and there are several of these stables around Tokyo. The younger ones live there and they all practice there every day, and they all eat together. Their training consists of wrestling. They do hardly any lifting. The upshot was that my Japanese host arranged for me to go to the Azumazeki Beya, which was run by the first Hawaiian to get into sumo, Takanohana.
When I send the transcript I'll put little question marks next to these names.
Okay. Takanohana is now retired, but he was running this show. The star wrestler in Azumazeki Beya at that time was Akebono, also a Hawaiian, who reached the top rank Yokozuna and retired about a year ago. So I actually saw the great Akebono in a room that was not all much bigger than this one. I sat on a wooden platform at one end and the guys would wrestle.
What did you think of it?
Well, I thought it was far more fearsome than I'd thought from watching it on the television or from watching it at a distance on the [???].
The way it's portrayed in American, or Western society is a sort of a comical, joking sort of thing.
It has that aspect to it, but the guys are amazingly quick and they really — One of the conventions of sumo is you don't try and hurt the other person. If accidentally a drop of blood is shed during the ring, they stop the bout and they clean it up and clean up the wrestler.
So it's not like boxing at all?
No, the idea is to overturn the other guy, but without hurting him. In fact, there are rules having to do with if any part of the body other than the feet touches the ring before the other one, then you've lost. It's the one who is first to be pushed out of the ring. Now, there was an exception to the rule. If you tipple somebody over and you're going to land on top of him, and to stop that you put your hand down so that in fact your hand gets there first, the other guy still lost because you've done this in order to avoid hurting him. So there are all these kind of rules, which are very interesting. When they meet, and the charge at the beginning, is really fearsome, actually; the sound of the flash meeting is quite amazing. But as I say, the ritual, the politeness, is very different from American sports. In Japan, boasting is just not allowed at all. At the end of the tournament, the guy who has won may be interviewed on the television, and the typical interview goes, “So, what did you think about winning this tournament?” and the guy blushes, literally, and says, “I tried to do good sumo.”
Yeah, I don't think you'd hear a professional basketball player saying that.
No.
Well this seems like a good place to stop. I definitely had to get in the questions about that. I remember from your Henry Norris Russell talk there were some joking jabs at you about your fondness for sumo, so I had to get that in.
Yes, at Manchester United I think.
Yes. Thank you for the interview.
[1]The distribution of Lyman-alpha absorption lines in the spectra of 6 ASOs ApJ Supplement (1980) p. 41-81.