Tony Tyson

[Begin Session 1]

Phillips:

Today is July 16, 2024. This is Jon Phillips of the American Institute of Physics, and I am interviewing, via Zoom, Dr. Tony Tyson, so thanks again, Tony, for doing this. So, we generally like to start these long life-history interviews with just personal background, your childhood, your upbringing, before we get into your career. I know you're from California. Can you tell me a little bit about where you grew up, what that was like, what your parents did?

Tyson:

Sure. So, I grew up in North Pasadena, near Altadena, during the Second World War, and I remember the air raids [Laugh] and the blackouts. And my dad, at that time, was an auditor for the Union Oil Company in Los Angeles, and my mom was a housewife. But the smog was so bad that a doctor told my parents that they should take Tony away from this. [Laugh] And so, we left Pasadena around ‘48 or so. I went to kindergarten and also lower school at the Longfellow School in Pasadena, where my father and grandfather had both gone. Interesting place. Anyway, we struck out to Southern California, and my dad tried lima bean farming on something like 100 acres, which is now Rancho something or another, [Laugh] near Escondido. And my job was to shoot the deer, but I was small enough that I ended up picking myself off the ground every time I shot my .22. And the lima bean crop worked for a while, but in the end, the deer won. [Laugh]

Phillips:

They're persistent.

Tyson:

They are. And so, during that period of time, we built a house on the coast south of Carlsbad, a place called Terramar. It was the first house to be built there. Cliff was about fifty feet above sea level. I built that with my dad starting in ‘49, and my job was to chip the cement off of old bricks and then use that to put the pathways down. But it was an interesting time for me because it gave me the opportunity to beach comb, which was a totally new experience. And an opportunity, therefore, to wonder. I think many kids, after that generation, tend to be oversubscribed, particularly with the new cell phone era. And I had nothing. [Laugh] So, I was a beach comber and also a surfer. I took up surfing on a longboard, solid wood board, which we had in those days. Weighed more than me.

And I found all sorts of interesting stuff on the beach because the Navy, in those days, were just dumping stuff overboard. I found all sorts of really interesting radar hardware, etc., that was washing up. [Laugh] And then, I became interested in ham radio because it was really a marvelous thing. If you could build something with one or two back end tubes, battery, antenna, you could talk in Morse code to somebody all the way around the world by scattering the signal off the ionosphere. And I became hooked on, in some sense, weak signal detection, which is a leitmotif, I think, for the rest of my career. That was a lot of fun, and I'm still bitten with that bug. You sort of never lose it.

Phillips:

What was the ham radio community like at that point?

Tyson:

Well, it was basically a community formed from veterans that were coming back from the Second World War and some other kids. My friends were either tuning up jalopies or building radios. And so, we had a club, and I learned a lot there. Also, one of my first jobs actually, aside from the job I had briefly as telegrapher for the AT&SF railroad, was in a radio shop in Oceanside called Kolb’s Radio. In that era, all the radios had vacuum tubes, and my job was to help repair them. And so, I interfaced with a lot of ham friends through that as well and got a lot of spare parts from there that were being thrown out, built stuff. It was a lot of fun, built antennas, some of which worked.

Phillips:

With that early interest in the technological junk coming in, washing up on the shore, and then with the radios, did you have sort of an early inclination towards engineering?

Tyson:

I did. And in fact, there were advertisements in the newspaper back then of—a lot of surplus was being put up for auction, radio equipment. And you didn't know what was in the box. You could only just guess what was in the box. And so, I went down to San Diego to one of these depots and bid, I don't know, $15 for a giant box full of unknown radio stuff. [Laugh] And it was a lot of fun. And so, most of my equipment that I used as a youngster was the surplus radio stuff. My receiver was out of a B-17, the transmitter actually also briefly. But then, I built my own, which was better. Anyway, it was a lot of fun. Kept me off the streets.

Phillips:

Did you already have sort of an interest in science more broadly at that time, too? Or you more focused just on the radio?

Tyson:

I did, and I'm not sure I can explain it, but in my school in Carlsbad, this was the sixth grade, there was a substitute teacher for one year, and that was critical to me. She encouraged my interest in science. I did a little project in the class. And that was the first real encouragement I got. My father also encouraged me to go in that general direction. Then, for reasons that aren't clear, I got a scholarship to the local military academy because the high school there was really lousy. I think it was Oceanside. Carlsbad didn't have a high school. Oceanside High School. Anyway, I went to the Army/Navy Academy there and actually was asked to teach a physics class when the teacher got sick. And I remember attempting to teach a chemistry class, but the last thing I remember from that was the smoke was coming out of the room. Nothing is perfect.

Phillips:

At that point, obviously, the military academy was the better school, but did you have any thoughts towards perhaps continuing down that path, maybe applying at college to one of the military academies?

Tyson:

Yeah, my dad wanted me to apply to Annapolis, and I did. Just because we had no money for college, really. And it was lost in the mail. I think there was a congressman in San Diego that signed it. It was lost. And then, I got a letter from Stanford accepting me, so I went to Stanford. They gave me a scholarship.

Phillips:

So, you moved up to Palo Alto?

Tyson:

Palo Alto, right.

Phillips:

Did you start out wanting to study engineering?

Tyson:

I did. I just was on that track in my mind. And then, I took a course in three-phase motors, which was the first course that they asked us to take in addition to the usual lower-level courses, and I hated it. So, I decided to switch to philosophy, which was an interest of mine anyway, and became a philosophy major and took all the required courses. But then, later on, I got disenchanted with the philosophy of the time, or at least the philosophy that was taught at Stanford, which is called logical positivism, basically a mathematical version of the mind. And I was more interested in existential phenomenology, so I walked over to the physics department and said, “Can I switch to physics?” And the kind secretary who was in charge of the department said, “You'll never regret this.” I remember her saying that.

Phillips:

Seems to have worked out fairly well.

Tyson:

Yes. [Laugh] Then, there was a professor at Stanford, Bill Fairbank, who I think should be called a physics evangelist. He would just bend the ear of anybody he ran into on campus and talk about marvelous things that are not understood in physics and impossible experiments to try to understand them. And it was from Bill Fairbank, I think, that I got the penchant of attempting impossible experiments. Or at least difficult ones that probed physics in new ways. But I wasn't aware of that until later. I was interested in it, but it didn't affect my career immediately.

Phillips:

So, when you started in on physics, did you have any sort of direction in mind that grabbed your interest? Or were you just diving in to explore?

Tyson:

Yeah, so Bill Fairbank was a low-temperature physicist, and he had a lab in the basement. And many of the students would go there and ask questions. Even in the middle of the night, he was there. [Laugh] And so, I really liked low-temperature physics. It was fascinating to me that quantum fluids like liquid helium have a wave function that extends as far as it can. If your superfluid is 100 feet long, the wave function extends that far. And so, that fascinated me, so I decided to apply for graduate school eventually in low-temperature physics and got in at a couple of places, but chose Wisconsin because they sent me a telegram. [Laugh] This was long before email. A telegram was more impressive than a letter. [Laugh]

Phillips:

Were they recruiting you fairly hard, or was that standard procedure for them?

Tyson:

I don't know, I think probably they were. They didn't send telegrams to everybody, I guess. And I guess it was on the basis of a recommendation from Fairbank, so I went to Wisconsin, and I worked as a graduate student there for a few years on an experiment that my advisor wanted me to do that I found horribly boring. And I had an impossible experiment that I wanted to do in low-temperature physics, but nobody in the department was interested in it. But I found a professor at the University of Chicago that thought it was fascinating, so I managed to get a strange kind of fellowship. I think I may have been the first student to make use of the Big 10 Traveling Scholar Fellowship. This was initially meant to help humanities students to use a library at another one of the Big 10 universities. I was the first to actually go and do an experiment in a laboratory. And so, I had to have two advisors. That was part of the program.

And so, in addition to Dave Douglas, who was the experimentalist, I eventually ended up choosing Chandrasekhar as my advisor. And that's probably where my interest in astrophysics came from. Although as a kid, I built a telescope. Lots of kids built telescopes. And I was reading books on cosmology, actually, when I was in high school, so I guess my interest went back a little bit further than that. And Chandra had what I would call an unreasonable respect for experimentalists. [Laugh] Totally unjustified. Anything I told him, he said, “Oh, that's really great. You should do that.” Anyway, it was great knowing him. And he, I guess, influenced my career as well.

Phillips:

How were you supported during your graduate school?

Tyson:

I had part of a grant. Douglas had some funds, but not enough, so I had an AFOSR grant. That lasted a few years. Air Force Office of Scientific Research. And I don't know why the Air Force might've been interested in my work, other than the fact—I’m sure they didn't know this—that at Stanford, I couldn't pay the bills and had to work part-time at a place on campus called The Electronics Lab. Worked on what turned out to be a classified radar for the Air Force, which we never got working because the computers weren't fast enough in those days. But eventually, Over-the-horizon radar did start to work, and it’s been great in moderating the arms race.

Phillips:

Before we jump back to graduate school, I want to follow up on that a little bit. Can you tell me a little bit about the process of going through security approval to have access to classified research and all of that as a student?

Tyson:

Oh, yeah, well, that was actually fairly straightforward. I had a friend, Bob Lee. He and I both needed to work part-time, so we worked in the Applied Electronics Laboratory. They tore down that building recently, but it was near the end station of the accelerator, which I also worked on briefly one summer. And so, the University, at that time, was doing some classified research on campus, so there was a mechanism to get us students some clearance. I think it may have been secret, but it wasn't top secret, that's for sure. Anyway, it was more interesting how I lost the clearance. One day, I was leaving work late, and the guard was already on his rounds. And I went through the gate, it clicked shut behind me, and then I realized I had left my pen in my office. So, I climbed up over the gate, and as I was climbing up over the top of the gate, the guard returned, and that's how I managed to lose my security clearance.

Phillips:

Were you able to continue working on the project in any capacity?

Tyson:

Yeah, eventually, it was all straightened out. [Laugh] It was really rather curious because the project that I was working on, one of the projects, was a senior thesis in the Physics Department at Stanford. And my advisor for that particular project didn't have clearance, so he was unable to see his students’ thesis experiments. I just had to write it up and tell him about it. It utilized some radar equipment that was, I guess, still classified. Pretty basic stuff.

Phillips:

Were they able to give you a formal evaluation at the end of that if they couldn't look at your work?

Tyson:

Apparently, yeah. I got senior thesis approval. Couldn't publish it, of course.

Phillips:

Then, when you moved on to Wisconsin and then Chicago, you said you were supported by an Air Force grant.

Tyson:

AFOSR for a while, yes.

Phillips:

You mentioned you think it might've been connected to this radar project.

Tyson:

No, I don't think the Air Force even knew about that. Certainly wasn't connected to it. No, this was an experiment in low-temperature physics to attempt to understand the behavior very near the phase transition from a normal fluid to a superfluid. And this occurs around 2 degrees Kelvin. And by very near, I mean really near, within one-billionth of the temperature of the phase transition. My experiment was to measure the ratio of normal fluid to superfluid as I got very, very close to the transition coming from below from low temperatures, moving up over a range of 10 to the -9 kelvin and developed a few technologies that were required in order to be able to do that. I got a number, something called a critical index.

Phase transitions, especially higher-order phase transitions, can be described in terms of the critical index, and I got this very strange number, two-thirds. Which I didn't understand at a fundamental level, I think. But it was a great experiment, and one day, as I was writing my thesis, a guy came to give a talk, a seminar. His name was Brian Josephson, and he had a theory–he didn't know about my experiment–that predicted this number. [Laugh] And that was one of the more fun times in my career, where it was just exciting to have measured something that was apparently predicted.

Phillips:

How much of an oversight role or collaborative role did Chandra have with you on this, given that it was not in his field?

Tyson:

Zero. I suspect he didn't even read my thesis. [Laugh] He was an advisor on larger issues. And one of the best pieces of advice I think that Chandra gave me was, “Don’t accept no if you have a cool idea that you want to pursue.” [Laugh] And that's proven to be good advice, largely.

Phillips:

Did you have any teaching responsibilities while you were at Wisconsin or visiting Chicago?

Tyson:

I was a TA in a nuclear physics course at Wisconsin, which was curious because I hadn't taken the course yet. [Laugh] TA by fire, I guess. But back in those days, they were teaching us students how to build an atomic bomb, basically. They needed more people with that kind of knowledge. Not the critical classified details, but basically the process, that part of it. And then, when I ended up at Chicago in Hyde Park, I was a teaching assistant in a couple of courses. That was required also. And so, I was a teaching assistant in one of Chandra’s courses and then a statistical mechanics course.

Phillips:

As you were working through your thesis and this experiment, did you have an eye sort of towards future academic employment, or were you already considering industry? What were you thinking?

Tyson:

I wasn't considering industry at all because like many students, these days even, it wasn't a known quantity. I was never exposed to the whole idea. Although, I must admit that when I was a bored student down in Carlsbad beach combing, I had a hunger for reading scientific journals, and the director of a company in La Jolla called General Atomic let me come in and use his library. And so, I was exposed to various physics journals, Physical Review, for example, at that time. I couldn't check them out, but I would sit there reading them.

Phillips:

Did you live in La Jolla at that point?

Tyson:

No, it was a short drive south of where we lived on the ocean.

Phillips:

So, you were thinking of pursuing academia after grad school?

Tyson:

Yeah, that was basically the only thing on my horizon at the time. But AT&T Bell Laboratories, the way they recruited people was to have recruiters go out to various different universities, and various people were assigned that job. And I remember the day that Pierre Hohenberg came to Chicago, and he was asking around, of course, “Who are some good prospects? Do you have any good prospects, either theory or experiment?” And so, he came into the lab and interviewed me, and he said, “Why don't you come and visit?” And I said, “I don't want to work for the phone company. [Laugh] Give me a break” And so, he went away, and he came next year and asked me again. I said, “Okay, maybe I’ll just come for an interview.” This was ‘68, I think.

And as it turned out, I interviewed at Columbia University and at Bell Laboratories on two different days, one day after the other. And when I went to Bell Laboratories, the first guy I saw asked me what I needed, and I said, “Excuse me?” “Well, how much space do you need, how much equipment do you need?” That almost convinced me. Of course, there was no prospect of tenure, but at Columbia, that would take a long time anyway. What really convinced me was that he said, “Promise me one thing, that you will not do your thesis experiment again.” Basically, “Promise me that you will switch fields and go into a new field where you don't feel scientifically comfortable. Something that fascinates you. And come back in a few years and tell us something really cool.” That did convince me. So, after a brief hitch in Nepal, I came back to Bell Labs and decided to pursue something that interested me, mainly the claim that there were pulses of gravitational radiation coming from the center of our galaxy.

Joe Weber had built a really ingenious detector, basically a bell, a solid cylinder of aluminum suspended very cleverly around the middle to decouple the one mode that would be excited by a passing gravitational wave from vibrations of the support system in the Earth. So, I designed a better and larger version of that through a series of designs, working with Lory Miller at Bell Laboratories and eventually built this large detector. Long story short, we didn't detect anything, [Laugh] other than noise, so I wrote a number of papers basically claiming not to have detected anything, and it was a really tough time for all of us, including Joe. But in the end, before he died, I became a friend of his again, which is fun. And we were musing about the past at that point. But yeah, we didn't detect anything. There were a number of phases of that experiment. I worked with Allan Mills at Bell Labs, Ben Brown, and others.

Phillips:

I’m curious, when you were designing this apparatus at Bell Labs and working on iterations of it, what was the process for that like? Did you lay out the schematics and send it off to a machine shop, who would then fabricate it and send it back? Or were you involved in the fabrication itself?

Tyson:

A little of both. I wasn't a good enough machinist to do some of the biggest jobs, but I was allowed to use Fermi’s secret machine shop. I think it was the on the third floor of the Fermi Institute there at Chicago.

Phillips:

What was the Fermi secret machine shop?

Tyson:

Institutes have their own big machine shops, basically union shops. And they had a good one there. But it was a 9-to-5 type of outfit, and occasionally students would have a need to machine something. I never met him, but apparently he thought that it was a good enough idea to have a secret machine shop. But no, I built most of the mechanical and electrical parts of the experiment and did the analysis.

Phillips:

How long did that project take? That was sort of your initial work at Bell Labs. How long were you pursuing that?

Tyson:

Well, it took longer than I suspected it would. It was sort of a lark, and I wanted to move on. I put together the first detector in ‘71 and started making measurements and started publishing null results at that point. But it continued all the way into the ‘80s, even after I was off doing other things. Even after I came on sabbatical to Berkeley, it was still going on with other people like Ben Brown and Allan Mills. Never detected anything other than thermal noise and the noise of the electronics, sort of a mix. But along the way, we discovered that Joe Weber had been fooling himself with regard to the intensity level of the signals that he thought he was measuring. He would see this noise, and the noise he thought was purely thermal.

But in the end, it turned out to be—and we were able to show this by reconstructing his electronics—the noise from his first amplifier feeding back into the antenna, exciting it. So, he was seeing these noisy signals that looked thermal, so his scale was off by a factor of fifty or something. But one of the innovations that Laurie Miller and I developed was to put a capacitor plate near the end of the cylinder and apply a known voltage to it, and therefore a known force, and just calibrate it that way. That allowed us to do our calibration directly rather than assume anything about the thermal noise.

But that was a null experiment. And there were a few other null experiments along the way. Again, stepping away from my area of comfort, I would say, scientifically into new areas. Tried to measure the mass of the photon, got zero, some forty decimals after it. Null experiment also. Bell Labs mixed people up intentionally. People would be next to folks in a totally different area. There was a chemist next to me, there was a mathematician across the hall.

By going to lunch, you would bump into and randomly scatter off people in different fields, and you would sit at lunch at a table where a lot of folks were talking about some challenging, crazy idea they were working on, and so there was a lot of synergy in that. People got interested in helping each other, and patents came out of lunch. Because in order to do something impossible, you have to invent something new to measure the impossible thing. [Laugh] So, there was quite a lot of that, and that, I think, in the end, was one of the justifications at the lab for encouraging at least a subset of the physical sciences division to strike out into the woods.

Phillips:

Given this directive to branch out and pursue new fields, after having worked on the gravitational detector and the photon, were you ever going to go back to low-temperature physics? Or is this where you sort of decided to stick with cosmology and astrophysics?

Tyson:

It’s a good question, and I think the answer is no. I was really interested in some of the issues—studying gravitational radiation got me interested in cosmology and astrophysics in new ways. So, I just basically got some elementary books and read up on it. [Laugh] And one of the notions that was around back then was this whole idea that there was some kind of a thing that wasn't understood called dark matter. People just gave it a name, but they didn't understand it. And as it turns out, Fritz Zwicky at Caltech had discovered that galaxies were zipping around clusters of galaxies faster than they should've been if you assigned one solar mass to their solar luminosity. That was largely forgotten. That was way back before the War, the late ‘30s.

But this whole idea of gravity intrigued me, so I, thanks to Bell Labs again, was permitted to go down to the Institute for Advanced Study and also to Princeton University, at least one day a week. They paid for that. And that was great. I worked in Johnny Wheeler’s gravity group at Princeton, and attended the Institute luncheon on Tuesdays, and got fascinated by gravitational lensing. The fact that things in the sky aren't where they appear to be, but they have been moved by the light bending that Einstein taught us all about, at least to within a factor of two, to a new place in the sky. And so, since you don't know where they should have been, the only ammunition you have in trying to reconstruct what’s going on is the telltale distortion of these objects as they move, so they have to be a resolved thing.

They have to be faint galaxies that are resolved, so then you can look at the telltale distortion. And I just made a diagram of how this would work, and it looked like it would create sort of circles of arc-like things around the center of where the intervening over-density of dark matter would be. And you could invert that theoretically and get a picture of the dark matter. So, I did that in a couple of cases. It was sort of interesting. It was a crazy enough idea that I made the tactical error of applying for telescope time to do exactly that. And it was turned down because people didn't believe it.

And so, I got together with Bill Saslaw and Phil Crane and made a proposal to basically take the same observations, but for a different scientific purpose, mainly to look at the synchrotron light in the optical from radio galaxies. It was well-known that there were radio synchrotron radiation, and maps of it, even, but we wanted to make a map of the very high-frequency tail of that. But this was using the detector of the day, which was a giant photographic plate, at least relative to the very first small CCD detectors, 8x10-inch photographic plate covering a huge area in the sky. And this radio galaxy would be a tiny, tiny thing, a millimeter in the middle.

And the data was useful, but I didn't actually make much progress on it until I had been assigned some time on a four-meter telescope at Kitt Peak National Observatory. I flew out to Tucson, and a huge storm hit. And the Observatory was closed down. There I was, sitting in a basement office at the National Observatory, wondering what to do. And I remembered then that I had all these photographic plates I had taken, and I decided on a lark to digitize them. People had been digitizing plates for a long time, but not co-adding them. I went down to the digitizer room, and this is a big machine, you put the plate down and—this is before laser digitizers. Basically, a piece of light would pass over it, very slow process on a nine-track tape, and you would get this data. [Laugh] And then, I wrote a program to co-add them.

And oddly enough, people had seen very, very faint galaxies before on these really deep plates. They were long exposures, about an hour. But had explained them away by saying, “Oh, that's just something that we like to call Kodak galaxies.” And to give them credit, there were lots of those, unfortunately, impurities in the developer, etc. But it turns out that in addition to the Kodak galaxies, there was something else. On many other plates of the same piece of the sky, those so-called Kodak galaxies will be at random different places. They won't be at the same place. They went away when you took a median average of all of the images. There remained a whole lot of very faint smudge-y things, which became obvious that they must be distant galaxies. And it was with that data that I first attempted to do gravitational weak lensing. Emphasis attempted.

Phillips:

Had the photos been taken at Kitt Peak? Or were they from a different observatory?

Tyson:

No, they were all images that I had taken or John Jarvis and I had taken from Bell Labs at Kitt Peak on the Four-Meter Telescope, pursuing this other project. And so, it became obvious that one could, in principle, detect either a rotation of the universe, and there was a paper about that, or a gravitational lensing of galaxies by the intervening dark matter. This was something that we called cosmic shear many years later, and got a three-sigma result for galaxy-galaxy lensing. And being a careful physicist, [Laugh] I decided that it would be more appropriate to call that an upper limit rather than a detection. Anyway, I reported it at various conferences, published a paper on this.

But one day, the solution of the problem arrived in my office in the form of the inventor of the CCD. He walked into my office—this is George Smith—and showed me something. I think I can quote him directly. He said, “In view of your interest in astronomy, you might be interested in this little device that we’ve just discovered makes really, really good images, or in principle, could.” And he put this little thing in my hand, and it was only three millimeters in size. And it was a charge-coupled device that they had devised to solve a company problem, which was to develop a solid-state memory that was purely solid-state, wasn't magnetic bubbles. And so, I said, “George,” and I pulled out of my desk one of these plates, “this is the kind of thing that we’ve been doing in the past.”

But it got me interested, and I had a little lab at home, and I started fooling around with it, got it to work sort of, and noticed that it was extremely sensitive. Much more so than the photographic plate, about a factor of fifty, or sixty, or eighty times as sensitive as the plates we’d been using. But of course, it had all sorts of garbage on the surface. These were surface-channel CCDs. But the important point is something that I discovered a little bit later, not long after that, that the garbage is always on the same identical place on the CCD. And if you jiggle the telescope back and forth, the galaxies will move relative to the garbage, and then you could have two images that you can solve for.

You'd have the high-fidelity image of the garbage, and then using that, you can get a high-fidelity image of the universe without the garbage. And that wasn't apparent to me until after I got– I decided to try to learn how to do this observing, shift and stare. I invented the shift and stare method to do this. There was a marvelous guy at Lowell Observatory, the director, Art Hoag, who gave me something like a month of telescope time, 42-Inch Telescope, which is a wide-field telescope, and I built a CCD camera using a somewhat larger CCD by that point, 320x512 pixels, to develop the shift and stare method through a series of trial and error. Mostly error. And from that, I was able to prove to the National Observatory that one could do this kind of thing.

And so, we were beginning to get more and more time on the bigger telescopes to pursue gravitational weak lensing. But I have to thank Art Hoag for his help in enabling this. But he also took me into his office, and he took out a giant photographic plate that he and Roger Lynds had taken. And I showed this cluster of galaxies with a big arc near the cluster of galaxies. And I think Art knew of at least some of my thoughts on this by that point, and he said, “Instead of the usual explanation of a shock wave, this could be a giant gravitational lens arc of another galaxy.” And I said, “You're probably right.” That was the state of play in those days.

Phillips:

This was the early ‘70s?

Tyson:

This was the late ‘70s by that point. 1980.

Phillips:

I’m curious if I can ask real quick, it sounds like you had a great deal of support from your colleagues and institutional support. Did you ever get any kind of reaction from the sort of managerial or business side of Bell Labs or AT&T, why you're looking for the distortions of these galaxies unimaginable distances away, and dark matter, we don't even know what that is? I know Bell Labs pursued basic research as one of its major aims, but this seems so far removed from any kind of immediate industrial purpose.

Tyson:

Well, that was effectively our homework assignment, those of us who were selected to do this kind of crazy stuff. The farther away it was from the company mission, the better, except that they also were turned on by the idea that what you're trying to do, doing something unusual, is to invent some kind of new, disruptive technology, which the company could then license. And so, that, in the end, as far as I remember, was their view of the work that we were doing in gravitational weak lensing. So, the company isn't going to get anything from that. Maybe some publicity. But the important thing to them is that we were in the process of inventing techniques that could be of use to the company.

Phillips:

Fairly far-sighted policy.

Tyson:

Yeah. But as a result of all of that—we always had a review, everybody had a review at the end of the year. And at these reviews, we were always given this double message. “Well, it’s really great what you've done recently in this academic pursuit, but what have you done for the company recently?” [Laugh] So, a lot of us tended to work part-time on these company projects where we felt that we could help some way. And so, I worked on a couple of those along the line. One was the solid-state fingerprint detector that people use on computers, to get into cars, and everything. I was on that patent because they needed somebody who knew about charge-coupled devices. There were a number of different other company projects, including through-the-air terahertz optical communication links, which they formed a group and developed into the product that I happily did not move with when it was spun off to another company. But that was a lot of fun. And there were a lot of patents that are still referenced from those eras. But largely speaking, we had plenty to do diving off into the woods.

Phillips:

At this point, you were working on gravitational weak lensing. Who else outside of Bell Labs was working on this at the time?

Tyson:

There weren't very many people. I remember Yannick Mellier in Paris being very interested in this, and he came and visited me on various occasions. I tried to get him a job, actually, at the Labs. But they didn't want too many people working in the same crazy field. [Laugh] So, he had formed a small group in Paris to try to work on this using the Canada-France-Hawaii Telescope, and I observed with him on several occasions using that, reducing the data. But there weren't very many. There were more people working on lensed quasars. That was a big field. The first lensed quasar, actually, they asked me to bring my camera–this is back in the ‘70s–to the University of Arizona 90-Inch Telescope on Kitt Peak to image this double quasar, so that was fun.

But it’s two points of light, and you can't do weak lensing, of course, with points of light. They have to be resolved things. But I hired a postdoc, Gary Bernstein, in the ‘90s, and before that, I decided that it would be a good idea to pursue larger and larger charge-coupled device. Because what you need is to cover a huge amount of sky in order to map the dark matter in the cosmos, so you need much bigger detectors than what we had. And Morley Blouke at Tektronix and I collaborated on the development of a 2048x2048 charge-coupled device with, I think, 20 micron pixels. A huge thing, couple inches on its side. And so, I embarked on a project of building a camera using four of them. It was going to be the world’s largest CCD camera at the time.

And so, I hired Gary Bernstein, he was a graduate student at Berkeley, and he was fascinated by many of these ideas. He joined Bell Laboratories as a postdoc, and we built something that we eventually called the Big Throughput Camera, which was for these things. And we were invited to take it down to the new Four-Meter Telescope in Chile at Cerro Tololo in exchange for supporting users also using this, so we were given a free observing time, not too much, but enough, it turned out, of our own on the telescope to use it for our own purposes, and then we would stick around and support it for other user groups that were anxious to image larger pieces of the sky for some purpose, and there were a number of such groups.

But two of them were searching for supernovae, so their program was to use the Big Throughput Camera to find the supernovae, and then once they've found it, to have already approved Space Telescope time to get its spectrum. And with the spectrum, and its apparent luminosity and calibrated photometry from the BTC, they could work out its velocity and distance. And both groups, and you may know this story, were finding what they thought was an incredible wrong result. Mainly, that instead of the universe slowing its expansion due to all of its mass, which was something that was already known pretty much, it appeared to be accelerating at more recent epochs. Mike Turner famously called this dark energy, but it doesn't mean that we understand its nature. [Laugh]

Which, in retrospect, was a clever thing to have done because it’s made more mysterious by calling it dark energy. So, it was really funny, these groups would alternately come in, one right after the other. They were assigned time on sequential weeks, they would come into the control room and say, “What are those guys finding?” We had no idea, we were just steering the telescope and talking about physics. And we were also using this data to attempt–and we finally succeeded–to measure cosmic shear, mainly the correlation of galaxies at some distance with other galaxies at some other distance in terms of their orientation. It turns out this higher-order statistic encapsulates all of the information you need to map dark matter in the three-dimensional and in terms of look-back time and red shift. Powerful.

And so, we were writing up some of our results, and sitting at about 3 AM in the morning in the control room, and we said to ourselves, “We can do better than what we’re doing here. We can make, in principle, a much larger CCD array, tile a much larger area with a big CCD camera. What would be required there would be a different telescope that would have a larger, very pristine, sharp field of view that would support this larger camera, and then you would need the compute power, of course, to handle the increased amount of data. But both the CCD fabrication and the compute power relied on Moore’s Law. And so, as you were able to put more transistors on a wafer, you were able to put more pixels on a wafer or a detector. [Laugh] And so, they drove each other.

At the end of the day, we decided, “We could probably do this.” This was 1996 at that point. And I decided that I would try to do this by proposing a new project, which I called initially the Dark Matter Telescope, and I wrote a prospectus on it. Somebody reminded me in mid-1998 that the decadal survey– astronomers have this marvelous process where, every 10 years, they compete various ideas about projects that the nation might want to build or new facilities. And the 2000 decadal survey was coming up, and I had very little time because I hadn't been paying attention. On a Friday afternoon, I rushed a 50-page proposal for the Dark Matter Telescope to the reproduction office at Bell Labs and said, “Can you get me 20 copies of this by tomorrow morning?” And they said, “Sure, we can do that.”

And then, we rushed most of them to where the decadal survey was being debated. I think it was Chicago. Slipped it under the door because it was a little bit on the late side. And they loved it. They thought it was pretty neat. I did something that turned out to be, in the end, a good idea. Mainly, on the front cover, I had a computer image of rich structure in dark matter from a theorist at Princeton. And on the back cover, I had a nice color picture of an Earth-threatening asteroid. And they liked that notion. They thought that it was therefore something that probably should be renamed because the time domain isn't really reflected in the Dark Matter Telescope name.

And so, Alan Dressler, who was the head of that particular panel, decided to change the name to the Large Area Synoptic Survey Telescope. That name actually got changed very quickly to the Large Synoptic Survey Telescope because an acronym called LASST probably wouldn't play well in the competitive spirit of the times. But all of us had to rush to the dictionary to look up the word synoptic. In the end, it means the big picture. Anyway, LSST was, in some sense, born then. It was ranked number two out of many proposals, so it wasn't the top, but it wasn't the bottom either. So, we felt that it was something that was potentially real and should start a project.

So, I started hiring some people, with the help of Bell Labs, actually, have to give them credit. The company was tanking at the time. They gave us a million dollars to start the project. At that point, it was a really interesting time because I got a feeler, an offer, from University of California, Davis. I always wanted to return to California, so it looked really appealing to me compared to sticking on the East Coast at places like Princeton or whatever. And so, I decided to move. But in the process, we started the project initially through some arrangement with the National Observatory in Tucson. Sidney Wolff was just stepping down as director and authorized a certain fraction of their budget to be spent on something called the LSST project, and I became the founding director in ‘03. And we started hiring people right and left.

Phillips:

The decadal survey that recommended the LSST also recommended the project that would become the Webb Space Telescope. Were those two recommendations seen as complementary, or were you in competition?

Tyson:

No, I don't think so. Talking many years ago with Alan—we should ask him, actually—I don't think so. They're very different. The idea being that LSST is an extremely wide field of view. It covers the sky quite rapidly, so it goes wide, fast, and deep all at the same time, which is something you don't find in books on telescope optics. And then, Roger Angel reminded us that you could build a very, very fast wide-field telescope by, instead of using one or two mirrors, using three mirrors.

And so, the design that we had of the telescope used Roger’s three-mirror design. I gave a talk at Stanford, at SLAC, in ‘98 on this, showing a picture of all this and how one might go about making observations of it, showing maps that we’d already made by that point of dark matter. But at that point, I had called it the Dark Matter Telescope. So, the project initially just had a few people. Chuck Claver was one of the first people that we hired on the project. And he was at the time, still is, in Tucson working at NOAO and allowed to spend 50% of his time working on the project. Then, we had a number of other engineers join from there. Hired a project manager.

I've developed this process, which in retrospect, has had mixed results, of hiring two managers in the same box at all levels in the organization chart, a project manager with a clipboard and a whip, and a project director with more of a scientific outward focus. And that works fine, if they get along. And we spent a whole lot of time in the initial periods– I spent a lot of time worrying about funding because it wasn't real enough yet for us to submit a construction proposal. We had so-called design and development proposals that were funded initially by the NSF. Later, the DOE through SLAC funded some of the camera work we were doing.

Phillips:

At this point, had you already spent down the million from Bell Labs?

Tyson:

Yes, we had by that point. And so, we got on the order of $100 million in these design and development grants. But I was still, as director, worried about the fact that we really had to have a credible source of private support to jumpstart any federal investment. Back when I was at Bell Laboratories, I was on a committee, which was called the General Research Colloquium Committee, and we had a speaker that we would fly in about once a month telling us something totally new. Not something the company was doing, but something somebody else was doing. It was fascinating.

And so, there were 1,000 people in the audience, all the research people at Murray Hill. And one of the speakers that I invited as chair is named Nathan Myhrvold. At the time, he was head of Microsoft Labs. And I remember his very humorous talk, which he entitled “Road Kill on the Information Superhighway.” And afterwards, we had lunch, and he came and chatted in my lab and office about this so-called Dark Matter Telescope idea that I had. He was fascinated by all of the technical arguments, but mostly by the fact that the data would be huge and would be manageable, in principal. But then he disappeared. I emailed him a few times. He’s a busy guy. And then, one day, I got an email. It was just right in the middle of a project review. We were finally allowed to write a proposal to the NSF for the construction of LSST, but one of the very first things you have in any agency review is sort of a pre-review.

“Does this make sense? Should we even consider doing this?” And it was during this review around lunchtime that I got an email saying, “Where do we send the $30 million?” I had wind of this because Nathan had sent me an email saying that, “You might hear about something interesting in terms of the private funding.” But just a hint, not who it was. Anyway, this was $20 million from Charles Simonyi and $10 million from Bill Gates. And we proceeded to write up the agreements. That made us real in the eyes of both the astronomy community, but also, more importantly at that time, the NSF and the DOE.

Phillips:

And that money was from Simonyi and Gates personally?

Tyson:

Yes. And so, we have since named the telescope after Charles Simonyi’s family. It’s called the Simonyi Survey Telescope. So, we started getting much more interest on the part of the agencies, both the NSF and the DOE. Because the DOE was already, by that point, invested in building the camera. I decided that there were several places where we could build the camera, and that's sort of my thing. But we ended up choosing SLAC. Good track record. And so, Steve Kahn and I had begun to hire people at SLAC, Kavli Institute, for this purpose, in 2004. There was a lot of proof that we had to offer to the agencies before they would advance us to the next step. Multiple steps. And there was a lot of engineering, ran through lots of designs for software, for hardware. And then, it became obvious to all of us that we could actually start bending metal and building the thing.

However, this was only a few years before the next decadal survey, the 2010 Astronomy & Astrophysics Decadal Survey. And so, the agencies, in their wisdom, said, “Okay, great. However, why don't you just hold off, and we’ll see what the decadal survey has to say?” And so, I started spending almost all my time interacting with the decadal survey panels. They were asking us questions. “Can you prove this? Can you prove that?” etc. And so, we would exchange these documents with the panels, which was very constructive, and there was this tension between trying to do something like this from space–and the project at that time was called SNAP–or from the ground. But the space-based project, and this is still a true story, suffered from the fact that you can do the wide and the deep, but you can't do the fast.

They have to take longer exposures, they have to somehow send all of this data down. The advantage of doing it on the ground is, you can zip around in the sky, take a whole lot of short exposures, and do it wide, fast, and deep. Anyway, long story short, the decadal survey ranked us number one. I had no knowledge this was going to happen. There was a leak by somebody at the Stanford PR department that claimed to have known that we were going to be one. But this sort of stuff happens all the time. None of us believed it. Interestingly enough, just by coincidence, this was during a yearly all-hands meeting that the project holds. This was in Tucson.

That Friday, on a big screen up in front of the room was projected a video feed from Washington in which Roger Blandford gave the results. For all of this, of course, I, as director, had two press releases prepared. [Laugh] And it was delightful, there was a roar that occurred in the room when the top choice was announced LSST. But we all expected the large telescopes, the 30-meter Very Large Telescopes, would be number two, and they weren't even number three. At least I was in shock, and I delayed sending out the press release because of this. They decided that it wasn't ready for primetime.

Phillips:

Did you have a sense from your discussions with the decadal survey panels as to… obviously the feasibility and viability were a big component of that. But as far as the underpinning sort of motivation for it, do you have a sense of how much the Dark Matter Survey was actually a part of that and how much of it was the near-Earth asteroid detection, the Kuiper Belt Survey, those sorts of other projects?

Tyson:

It was all of the above, but some of the hardest things to prove had to do with cosmic shear from the ground. And so, some of the toughest questions that I got were, “Prove that you can have a residual error in your shear measurement of 10 to the -6,” or whatever was needed by that point. And so, we had done simulations of the LSST observation happily by that point, including all of the bad things that could happen not only with the detector but also with the atmosphere and were able to show that we could do it. That was tougher to prove, and it occupied more of my time.

But there was a huge amount of interest in the time domain. In retrospect, I think that that's probably something we’re going to be remembered for 100 years from now is something new in the time domain, which we don't expect, and we can't tell you about, because we don't know about it right now. Something unusual. But then, it took a couple of years for the National Science Board to get around to approving the project. It’s a very long chain of events. And in the middle of all these years, many of us went to Washington every April to interact with Congress. It’s a traditional thing to do if you're building a big project. Every couple of years, it would be a new staffer that you would have to bring up to speed. But Bill Smith, who was, at the time, the head of the Association of Universities for Research in Astronomy, took us around.

He was an old congressional hand. He himself had been a staffer many years previously, and so Bill introduced us to all the right people, and I started to get to know a number of folks in the various other agencies in town, which turned out to be helpful to the project along the way. But most of my time was spent on proving the science viability. At that particular time, our design was sufficiently far advanced that we felt sure that we could start building stuff. And we, in retrospect, were almost ready. [Laugh] When you start putting things together as opposed to building the thing itself, you suddenly discover, “Wait a second, I thought you said inches, not centimeters,” etc. And there have been moments like that in the project. There have been errors. I think almost everything in the project has been broken once, even the optics.

Phillips:

At what point during the development project did you settle on a location?

Tyson:

That was early on. In terms of proving the plausibility even of this, we needed to have a place to put it. Already starting in 2005, we formed a committee called the Site Selection Committee, and Victor Krabbendam, who had a track record already of building telescopes various places in Chile, was the manager that we put in charge of that process. And he then selected a committee of people that he knew in the field. There were a number of possibilities. I think we started with something like seven or eight sites, eventually ended up with one, Cerro Pachón, near Gemini South, actually, on a mountain in sight of Cerro Tololo, where all of this began. And there were some humorous moments in that whole process, but in the end, we ended up with Cerro Pachón. It was a developed site already, it had power, it was managed already by an international agreement with the National Science Foundation.

Phillips:

Did you get to personally survey the sites that were under consideration as part of this?

Tyson:

A couple of them. But my focus was really on fundraising at that point, sorry to say. But I, of course, was very familiar with the Chilean site.

Phillips:

Speaking of fundraising, I know you had the $30 million from Simonyi and Gates. And I believe you've mentioned that you had a corporation you set up, the LSST Corporation.

Tyson:

Yeah, that's an interesting story. So, in order to mix federal investment with private investment, so-called public-private partnership, you have to have a 501(c)(3) corporation in order to do that. Very early on, in fact when I was still at Bell Labs, we decided to set up such a corporation in Tucson. John Schaefer, ex-president of the University of Arizona, was the head of that. And so, that's where Simonyi’s and Gates’s funding was sent. And it was used mostly to fund the mirror development, this very strange kind of mirror that has two very different surfaces on it. And that was the majority of… because at that time, that was our biggest technical risk. It’s no longer and hasn't been the biggest technical risk for years, but at that time, that was the tallest pole. And so, the LSST Corporation was formed. There was a board that I, as director, reported to. And there were tensions between my management style and John Schaefer’s management style. I like to build consensus and get people behind the project. John likes to give orders. And eventually, he stepped down. But there were some difficult times for all of us in those years.

Phillips:

Is the LSST Corporation still in operation?

Tyson:

Well, it’s interesting. It has been until recently, but it’s changed its focus and is now called the LSST Discovery Alliance. Because they don't anymore have anything to do with the construction project. One of the things that we did as LSST Corporation, starting fairly early, was to fund clever young people to join the project or do something really interesting. And they've focused in on that. So, educational public outreach, funding scholarships for undergraduates, graduate students, and fellowships for postdocs through philanthropy. And so, that's been recently a very great success. That's a nice success story.

But at one point, part of this was motivated by the fact that there was a catch-22 in this whole process. [Laugh] And that is, we submitted a proposal to the National Science Foundation under the auspices of the LSST Corporation, right? They liked the proposal, but they didn't like the fact that it came from the LSST Corporation because it had no track record at the NSF. That's the catch-22. Long story short, the Association of Universities for Research in Astronomy, AURA, has a track record. They've managed the Space Telescope. They manage all sorts of things, including all the optical observatories. And so, we decided to resubmit it later through AURA, and that proposal was reviewed. That sort of precipitated this decline of the LSST Corporation.

Phillips:

Obviously, the LSST Corporation was named for the LSST. You mentioned that the telescope is named for the Simonyi family. The observatory itself, the name has changed to the Vera Rubin Observatory.

Tyson:

Oh, yes, indeed. This whole idea of naming the observatory after our friend Vera Rubin came to many of us sort of simultaneously. It’s rather interesting. But there was a staffer who thought the idea would be pretty cool. There was somebody at the NSF. And then, Steve Kahn and I decided that it would be a really good idea all at the same time. And I knew Vera quite well because it turns out that way back when we were at Lowell Observatory, we were there together. She was observing on the 72-Inch Telescope at Lowell, and we were on the 42-Inch Telescope. And then, starting somewhere in the mid-‘90s at an annual meeting in Washington, Vera and I would meet and chat about all sorts of things, and she was really fascinated by the fact that you could non-dynamically investigate dark matter. Her whole understanding and contribution to the field actually was a dynamical understanding, rotations, curves of galaxies. So, she was fascinated by weak lensing because of that. And during our last meeting, where I met her last, she said that she was very happy that LSST was going to happen.

Phillips:

Seems very fitting.

Tyson:

Yeah, absolutely. And the other interesting thing is that the notion that—LSST, all the legal agreements, all the paper, all the documents, all the publications, what do you do about that? It turns out that we thought up a suitable substitute. The Legacy Survey of Space and Time. We still have LSST, but it’s spelled out differently. I have t-shirts with both of them, a separate drawer. One for Large Synoptic Survey Telescope, and one for the other. [Laugh]

Phillips:

I know that the observations are supposed to begin next year. What has been sort of the process of getting optics built, getting the site constructed, up til now to actually getting the Observatory up and running?

Tyson:

Well, we’ve had the usual problems in any large construction project with vendors. COVID set us back actually more than two years. If it hadn't been for COVID, we would've come in, oddly enough, on schedule and on budget, which is unusual in the field of large astronomy facilities. And then, there have been various, obviously unplanned, events, accidents. The camera was very recently shipped down from SLAC to the mountain, and it’s been sitting there for a month, and we still have not turned it on.

Because we’ve had some issues with the Dynaline coolant system that is required to cool everything off so that we can then turn it on. But that might happen only a week from now, so we’re holding our breath to see if it still works. It was working when we shipped it. But we’ll see. So, the schedule is very much up in the air. It keeps moving a little bit. Turns out that we knew that we were going to be in a situation where we needed to test everything all the way to the software and observatory operations.

In advance of having the camera there, we knew that we’d eventually be in this situation, so we built something called a commissioning camera, which is nine charge-coupled devices, a 3x3 square matrix of these big charge-coupled devices rather than the 189 charge-coupled devices that are in the camera. And that's been down there for a while, and we’re about to put it onto the telescope, so that'll be fun. And it’s going to have a run which is probably going to be on the order of two months, maybe a bit longer, and we’ll learn a lot from that, be able to test a lot of things.

Phillips:

Once operations do commence, and you have the telescope and the camera up and running, in your role as chief scientist, are you going to be in a position where you can still pursue your own research agenda there? Or are you going to be primarily overseeing everyone else?

Tyson:

No, I’m just fascinated, as I always have been, with making things work. And I also have a science agenda, but it’s rather broad. My students and I have been working on simulating some of those scientific use cases for LSST, coming up with potential gotchas, which we then go back to the project and say, “Show me that you can do this.” [Laugh] Yeah, there are just a number of really cool things that we can do with LSST, but I have to admit that what I'm very much most interested in, in terms of the science, is something that we cannot discuss because we don't know it. Mainly, something totally unknown that is not in our 600-page science book that we wrote back in 2009. But something novel, and interesting, and new. And that's going to be a lot of fun.

And I’ll bet you that that's going to be in the time domain rather than some cosmology result, which is really my own field. But I think there's more discovery space for the unknown in the faint time domain. Because that's never been done, sort of unexplored. I’ll bet you that the universe has some secrets. One of the concerns that we've run up against is the effect of the low-Earth-orbiting satellites, and their glints, and flares, and trails, and the effects that they cause on our camera. So, we've been working here in the lab, testing this—I set up a lab here shortly after moving here from Bell Labs—to test and characterize the charge-coupled devices that we use on the LSSTCam. And we’re now optimizing their clocking, and voltages, and everything for the camera once we turn it on. We’re very deep into that rabbit hole.

Phillips:

It’s an important one. You did move down to Davis in the early 2000s, right? What prompted your departure from Bell Labs?

Tyson:

Oh, it was just that I was always looking to return to California. It’s where I grew up. The company was tanking. My last, I would say, 10 years there were very stressful because of that.

Phillips:

Was that a consequence of the AT&T breakup?

Tyson:

Yeah. If I had been cleverer, in 1986, when the company was broken up, I would've seen the end game, but I was too busy playing with my toys. And so, I ended up having to defend a lot of people that were doing very complicated, ambitious things to a board that wasn't interested in that anymore. So, my wife and I got simultaneous offers at the University in physics, and it seemed like a really logical move to make.

Phillips:

Would you say that was a fairly sudden shift after the monopoly breakup that the culture changed in the administration of the lab?

Tyson:

Oh, I don't think so. It was really gradual. I really didn't detect much until the early ‘90s. Then, you're probably aware that the company was bought out several times by others, larger fish eating the previous large fish. And with each one of those, there was a trimming of what they called the blue-sky research part of it. And this has happened not just with that one company, but with all industries worldwide. It used to be that it wasn't the universities, but it was really the private think tanks—this is back in the ‘30s—that did all of the out-of-the-box thinking. And that was certainly the justification for this at Bell Laboratories in the early days, continuing on through my career there. But the horizon shrunk from 20 years down to a few weeks. Basically, it came to a point where, “Can you help us slap together a box and sell it in the next couple of weeks?” to compete with what they envisioned to be their competition. Bell Labs still exists. It’s basically changed its focus into software and things of that sort. There's a little optical work still being done, optical fiber.

Phillips:

All the big industrial 20th century concerns, Bell Labs, IBM, seem to have pursued that path, which is unfortunate.

Tyson:

I have no regrets whatsoever. I was really lucky to have been snarfed up by Bell Labs.

Phillips:

How was the return to academia after spending so long in industry?

Tyson:

Well, it wasn't that abrupt because I was part-time at Princeton for about 15 years, so I was involved in that academic climate. I had students from Princeton, for example, at Bell Labs. But I'm struck by the fact that there is, in universities—and I knew this, but it becomes much more apparent when you actually are swimming in the middle of this pond. Universities still have silos of excellence, and if you have an out-of-the-box idea that involves another silo, or God forbid, another college somewhere else in the University, to form viable collaborations, and to write proposals, and whatnot, and that's been an uphill fight for many people at universities. I have been popularizing the idea here that what you need to do is to have more cross-silo couplings and inventing disruptive technologies in the process. So, some people are interested in that. But there's a huge amount of momentum and a sense of ownership on these silos. [Laugh] The same arguments get repeated in faculty meetings worldwide.

Phillips:

I feel like we’ve covered most of the scope I wanted to cover. Is there anything that we should've gone into more detail in or any topics you can think of that we didn't get to touch on?

Tyson:

Aside from the fact that I find science extremely exciting? It’s something that I would do if nobody paid me. And actually, I don't get paid anymore since I stepped down from teaching just recently. But it’s fun. A lot of fun. And just being surrounded by all of the excellence of the people on this project just reminds me of the golden days at Bell Laboratories.

Phillips:

It’s very exciting, and I'm personally very excited to see what comes of the Rubin Observatory. I’m hoping to read more in the near future.

Tyson:

You and me both.

Phillips:

Well, thank you. I'm going to pause the recording here.

[End Session 1]

[Begin Session 2]

Phillips:

All right, today is August 20, 2024. I'm here again with Dr. Tony Tyson for part two of our interview. And again, thanks for doing this. So, to start off, I know you wanted to bring up some thoughts on, again, the young scientists working at LSST coming to your meetings and sort of the future of the field. So, if you wanted to take that and run with it…

Tyson:

Right. For decades, most of the people that would come to our yearly all-hands meeting were actually those of us that were working on the project. There would be an occasional group of other people that would drop in. And it was great to see new people and new faces. But what has happened recently is that we started– we changed the name from the all-hands meeting to the Project and Community Workshop. Anyway, we invite the community in, and this last meeting that was held just a few months ago was amazing because I looked around, and I didn't recognize anybody. And that is a good sign.

They are young people that have shown so much interest in the project that they've come across the country to go to a workshop in Palo Alto at SLAC. And it was really great to see all these young people who were, probably in most cases, not even born at the time this project started. And they are the future. They will inherit whatever giant– they will make whatever wonderful discoveries LSST is destined to make. And so, that was a great feeling. And it was fun talking with them.

Phillips:

As kind of a follow-up to that, last time we talked, you had mentioned that you saw what you thought would be the future of sort of the big discoveries that LSST is going to produce, being in the time domain. I know that's not your primary area of focus, but I wonder if you could say a bit about the kinds of observations they're going to be making in the time domain and the kind of work that– you mentioned that you don't know what’s coming, and that's what makes it so exciting. But from what you can see now, what directions that's going in.

Tyson:

Yeah, so, we all noticed, really, decades ago, that there are lots of telescopes, even some wide-field telescopes, that have investigated things that explode in the middle of the night. [Laugh] Like supernovae, etc. There's a whole class of objects that change their brightness very fast or that move, like comets and asteroids. So, those are known things. But they're all bright enough that they would've been noticed in those surveys. Smaller telescopes. Usually, sky surveys actually end up using small telescopes because it’s easier to build a wide-field telescope small.

You can make a plot where the X axis is the time, in some units, like days or hours, and the Y axis is the luminosity of the thing that's exploding, or moving, or whatever. So, this plot is populated– the righthand side of this plot, where you look at sort of long-lived phenomena, like supernovae that live for months, and even longer in some cases, novae even longer, that's populated. But the really interesting thing is that the lefthand side of that plot is unexplored. So, the faint time domain is unexplored because we haven't yet done a wide-field sky survey that used a big enough telescope to capture very faint light from things at very high red shift or that are intrinsically very faint. And so, this is an area where LSST is just totally new. No one else is doing this.

And so, I suspect since it’s unexplored, and it’s such a very large range, even in log space, it’s such a very large area of this plot that we are just bound to discover new physical phenomena that we haven't even thought of. And so, I think it’s in this area—and you need something like the LSST because it goes wide, fast, and deep all at the same time. If it’s wide and fast, you're not going to see very faint things. Anyway, that's why I think that we will be known 100 years from now for a new phenomenon that is discovered in the faint time domain.

Phillips:

It sounds very plausible to me, and I'm excited to see what happens. [Laugh]

Tyson:

You and the rest of us.

Phillips:

The second thing I wanted to sort of do a follow-up on was, one of the sort of crucial components, obviously, for all of this going back several decades now is the CCD cameras that you've been able to build, and grow, and apply to larger and larger telescopes. I was wondering if we could get back into a little more detail from the point that your colleague at Bell Labs approached you with his first tiny, little CCD device to the process of building them out and expanding them.

Tyson:

Yeah, so, George Smith came into my office and showed me this very small, several-millimeter-sized charge-coupled device that they had built, which was actually destined to be used in something that the company called a picture phone, which was an unmitigated market disaster because the company did not do a reasonable market survey. In these days, people would love to talk to one another on the telephone using audio, and some of them thought theoretically that it would be sort of cool to be able to see who you're talking to. But they didn't ask the question whether they would be comfortable with the person that they were talking to would be able to see them.

And it turned out, for technical reasons also because at that point, we didn't have optical fibers, data was all analog, no digital data at that point—that came later–so it was a bunch of copper wires. It was going to cost a couple hundred dollars per month for you to have a picture phone. In the end, it was just managers at Bell Laboratories that talked to one another on a picture phone and a few other companies, and it fizzled quite rapidly. And so, the idea was to replace the vacuum television tube, called an image orthicon, in the then picture phone with the charge-coupled device. And so, it had no home in terms of where it was going to go in the development group.

At Bell Labs, we had basically simultaneous research division, where most of us– I resided there in the search division, where at that time, it was sort of blue skies, just discover something new. And then, there was a development group over in building two, and we worked together, and that was one of the great things at Bell Laboratories, getting stuff done for the parent company. But anyway, he came into my office with this little thing, and it evolved gradually—we got bigger ones. After the picture phone disaster, people left that development group and went off and started places like Fairchild Semiconductor. [Laugh] Heard of that? Texas Instruments. And several other companies.

And they realized that there was going to be a market for CCDs somewhere. It wasn't clear. Anyway, they started making bigger ones. So, I built a camera based on one of those chips also and took it to Lowell Observatory. And so, the cameras that we built started being bigger and bigger because the CCDs were bigger. That's how it evolved gradually. Now, we have the LSSTCam, which is 189 charge-coupled devices, each of which is 40 millimeters by 40 millimeters in size.

Phillips:

And I know from our last conversation that Moore’s Law and the sort of increasingly compact fit of ever-smaller transistors on a chip was the key thing that allowed the cameras to get bigger and bigger, as you were saying, or higher resolution. But were there any other technical hurdles to overcome or developments there?

Tyson:

Well, I may have mentioned it before, but one of the big technical hurdles–I didn't initially realize this—turned out to be the telescope itself. How do you build, and with how many mirrors or lenses, an extremely wide-field telescope that could cover ten square degrees of the sky, for example? At the time, we only had cameras that would go up to one square degree, and even that was difficult, and the images were sort of distorted anyway. And so, Roger Angel came up with a solution to that, which was, instead of using one or two mirrors, using three mirrors. And it was his three-mirror design that enabled us to do this.

It was clear at the time that because of the silicon fabrication process, larger groups of CCDs were going to be possible because you could fabricate them in quantity, and the increased amount of data would also be processed in real time because of Moore’s Law. Also, because Moore’s Law drives not only the fabrication of charge-coupled devices, but also of transistors and computers. We understood that pretty early. There was this synergy, because of silicon fabrication technology that allowed us to scale what we had built already, which we called the Big Throughput Camera, up to much larger things. That all occurred in the late ‘90s.

Phillips:

Great. We can chat about anything else that comes to mind, but the last question I really wanted to make sure I got on here is, we talked very briefly last time about your conversations and friendship with Vera Rubin. I was wondering if you could say a bit more about your relationship, and then maybe also, in particular, the contrast between her approach and the dynamic approach to dark matter observation and your own using gravitational weak lensing, and how those interacted.

Tyson:

Yeah. Well, up until we started meeting yearly in Washington in April, I didn't get to know her all that well. Turns out, my wife, Pat Boeshaar, knew her very well because they had worked together on telescopes at Lowell Observatory. But Vera and I ended up on the mountain at Lowell Observatory and chatted briefly during that period of time. She, using a spectrograph on the 72-Inch Telescope, was measuring rotation curves of galaxies and looking at how the stellar spectra looked as a function of distance from the center of the galaxy, and it just didn't make sense, of course, initially, when people looked at it. It seemed to violate Kepler’s Law because the mass was not where the light was.

Somehow, galaxies, at least the ones that she chose that enabled her to do this with sort of semi edge-on spirals, galaxies were embedded in a huge envelope of massive invisible stuff that stabilized their rotation. And the rotation curves showed that dynamically. But she wasn't a physicist, she was an astronomer, and she never– I don't know if she really believed in dark matter as a physics concept. She believed in it as an effect, an apparent effect that had a name that Fritz Zwicky, as it turned out, had given it way back in the late ‘30s. And she was interested, really, when we talked, in the fact that the gravitational weak lensing and the gravitational lensing techniques that I was working on were completely independent of any kind of dynamical measurement.

Weak lensing basically showed you can make images of the dark matter, it showed it that was there because it caused the background galaxies behind the mass concentration to move to new places in the sky and thus be distorted, and we could look at the distortion and invert that to get the picture. She was fascinated by that because it was the only piece of evidence that was other than dynamical. But in the end, I think that she knew that LSST would be something that would settle this issue once and for all. And simultaneously, I should mention that others had jumped on this bandwagon using radio telescopes, and they had looked at the gas that emits in the radio that has lines.

I'm thinking of Mort Roberts primarily here, who recently passed away. And they also found these so-called flat rotation curves, just as Vera and Kent Ford had been finding using their optical spectrograph. And so, by the late ‘60s, I think the astronomy community truly believed that there was this effect, and we were all puzzled by what this dark matter was, the physics of dark matter. That's sort of been a theme for my own research, after that period of time, has been to try to do experiments and to work on LSST, also, but actually do other laboratory experiments to probe the physics of dark matter.

Phillips:

You said Vera Rubin and the astronomy community more broadly saw dark matter as an observable effect but not necessarily the physical entity that could be observed. Was there a larger sort of debate over that particular aspect of it at the time?

Tyson:

Oh, of course. That's a great question, actually. There were countless theories, of course, of what this might be. Is this a new particle? It was evident to many people starting at that time, particularly came to a crescendo, I would say, in the mid-’80s, but there was a recognition on the part of the physics community that this was beyond the Standard Model of particle physics. That is to say, just as gravity itself is, of course. The physics of gravity is not understood.

And so, the fact that dark matter existed was evidence for new physics, which is exciting, and what excites all of us is that it’s new physics. So, there have been and continue to be countless marvelous ideas on the part of theorists of what you could introduce as an extension to the Standard Model in the Lagrangian, for example, that would couple to a dark sector. And hopefully, also try to explain something else. You don't want to have a new tooth fairy for each one of these new kinds of things that people discover. And so, the idea was, on the part of the theoretical community, to try to see if some known symmetry within the Standard Model, which was a puzzle, would also be explained by this extension to the dark sector.

For a long time, the physics community thought that if the dark matter was thermally created in the Big Bang, that it might be a heavier particle, a so-called WIMP, weakly interacting massive particle. And there have been marvelous laboratory searches for WIMPs going to incredible sensitivities, to the point where now, the neutrino background is setting a floor. And many people pointed out that there's actually “more room at the bottom.”

Namely, if you go to much lower energies and go away from thermally generated dark matter to something where it was introduced by some kind of offset in the original symmetry of the Big Bang, or some odd kind of fluctuation that occurred during inflation at very, very low masses, microelectron volt masses, for example, that there would be some– this is a much larger region, where you could explain at least some, if not most of the known dark matter in our own galaxy by this kind of “particle.” It’s not really a particle because if you go down to such low energies, it really has a wavelike character more than particle-like character. Anyway, we’ve done some experiments there very recently. Of course, found nothing, otherwise you'd have heard about it, setting a new limit down in the microelectron volt range.

Phillips:

Are you able to publish the results when you have these null experiments?

Tyson:

Yes, they're published in the Astrophysical Journal, and also Physical Review D is one of the favorite places to publish experimental results. And so, one of our experiments just was published last month there, where we set a new limit over a range of masses down in that wavelike. We can do better, and we’re hoping to do follow-on experiments.

Phillips:

Those were sort of the main topics I wanted to make sure that we circled back to from last time. Is there anything else that you wanted to make sure we added in beyond your thoughts on the future of the field and the young scientists up and coming?

Tyson:

That's the most exciting thing that I see on my current radar. I should tell you that just today, I got some good news. I mentioned to you earlier that I was worried that our camera might not work after we shipped it down there. [Laugh] Just this morning, we got some good news, that several of our colleagues turned on the electronics on the camera, and they all work. At least the readout electronics. After we cool the focal plane down in a week or two, we’ll see if the CCDs are behaving themselves. But we were worried about the power supplies in particular. That's good news. The other aspect, I guess we discussed it earlier, is, back in the grand days of Bell Laboratories, where at least some people were encouraged to go off into the woods– this was a very selective thing.

Not everybody was told this. Bell Labs was a big place. AT&T was, like, 300,000 employees. Bell Labs, perhaps 15,000 employees, and there were a few hundred of us that were in the Physical Research Lab, and it was mostly those that were in the research division and in the Physical Research Lab that were given this kind of freedom. It was done on a piecemeal basis. But most of the folks that I talked to that are much, much older than I am, that are still around, [Laugh] when they were hired back in the late ‘50s and ‘60s, they were also told to do something unusual and see what happens. Of course, the idea being that if you want to do something that's out of the box, you have to invent some kind of new device to measure it. That technology is going to be useful, maybe.

Phillips:

You mentioned that this was a selective process, but it also sounds like, when we talked last time, that you were sort of brought in, starting off with the low-temperature physics, but sort of right off the bat with the idea that you would be doing this more sort of out of– so, was it selective in that people were sort of brought on to be tracked towards that, or was it more of an internal competitive thing?

Tyson:

I don't think it was competitive, no. The nice thing about the Labs at that point was that there was really—within the research division, anyway — no competition. We were encouraged, in fact, to work with everybody in sort of a blue-sky attitude. And we were all funded just enough to do one experiment, not a bunch of them. Just enough to do one crazy experiment. But there was competition in the development areas, fierce competition. Not only between managers there, because everybody had their own idea of what direction AT&T should go in, but also competition with other companies. And during all of this period of time, AT&T was being deregulated gradually, by steps.

Nobody knew exactly what would happen. Bill Baker, the president of the company when I was there in the ‘70s, told us once in private that if AT&T were fully deregulated, that Bell Labs would cease to exist. Because we’d have to give away our technology immediately because that was part of the deal. And so, there would be no reason to let people discover new effects, so there was no business purpose for it, and there would be no funding. We got our funding mainly from the operating companies, by the way, because that's where the money was being made. Also, a little bit from AT&T long lines. But when the operating companies were spun off by the Justice Department, the whole reason for the funding for Bell Laboratories– it was very clear that it was going to vanish. And so, in the research division, our own customer was Western Electric, and that wasn't a very large…

Phillips:

Was there an effort at first after the initial breakup to keep Bell South, and Bell Atlantic, and everything funding Bell Labs, even if they weren't part of the overarching parent company anymore?

Tyson:

There was, but the Justice Department pushed back and forced them to complete the spinoff. And so, some of them decided to create their own research outfit, which further decreased the need for the research division at Bell Labs. This came to a head, of course, in the late ‘90s, when we became Lucent, and then Alcatel-Lucent, and of course, recently, after I left, it’s changed once again to Nokia Bell Laboratories. Anyway, it’s been a marvelous ride, and I'm terribly fortunate to have had a career there. And also fortunate to be at a university where I've begun to realize the importance of cross-disciplinary collaboration of the sort that we were encouraged to do at Bell Labs, hiring clever young new experimentalists to do impossible experiments to measure small quantum effects. And we have done so recently with the hiring of Nancy Aggarwal, who is collaborating across colleges. I've decided to promote something called X Lab, which is basically a collaborative group at the University across colleges that is reminiscent of the kind of freedom for exploration that we had at Bell Labs. And I've gotten a chancellor interested in that, in X Labs.

Phillips:

What sort of funding sources are you looking for, for that?

Tyson:

That's the problem. I think the reason something like Bell Labs succeeded was that there was a funding source. We were a monopoly, which is the main reason that whole thing worked. Yes, there has to be another source of funding. And so, that's what we’re looking at.

Phillips:

Excellent. Well, thank you so much again for taking the time. I'm hoping that perhaps, maybe a ways down the road, once LSST has been up and running for a while, we could talk again.

Tyson:

That would be fun. It’s beginning to look more and more likely to me, so we’ll see.

Phillips:

I will pause the recording here. Thank you, again.

[End Session 2]