Notice: We are in the process of migrating Oral History Interview metadata to this new version of our website.
During this migration, the following fields associated with interviews may be incomplete: Institutions, Additional Persons, and Subjects. Our Browse Subjects feature is also affected by this migration.
Please contact [email protected] with any feedback.
This transcript may not be quoted, reproduced or redistributed in whole or in part by any means except with the written permission of the American Institute of Physics.
This transcript is based on a tape-recorded interview deposited at the Center for History of Physics of the American Institute of Physics. The AIP's interviews have generally been transcribed from tape, edited by the interviewer for clarity, and then further edited by the interviewee. If this interview is important to you, you should consult earlier versions of the transcript or listen to the original tape. For many interviews, the AIP retains substantial files with further information about the interviewee and the interview itself. Please contact us for information about accessing these materials.
Please bear in mind that: 1) This material is a transcript of the spoken word rather than a literary product; 2) An interview must be read with the awareness that different people's memories about an event will often differ, and that memories can change with time for many reasons including subsequent experiences, interactions with others, and one's feelings about an event. Disclaimer: This transcript was scanned from a typescript, introducing occasional spelling errors. The original typescript is available.
In footnotes or endnotes please cite AIP interviews like this:
Interview of Joshua Frieman by David Zierler on October 6, 2020,
Niels Bohr Library & Archives, American Institute of Physics,
College Park, MD USA,
For multiple citations, "AIP" is the preferred abbreviation for the location.
Interview with Joshua Frieman, head of the Particle Physics Division at Fermilab, and professor of astronomy and astrophysics at the Kavli Institute for Cosmological Physics at the University of Chicago. He recounts his childhood in Princeton as the son of a physicist and his decision to attend Stanford as an undergraduate, where his interests in cosmology developed. Frieman explains that his options for graduate research in cosmology were narrow and his reasons for going to the University of Washington to work with Jim Bardeen before moving to Chicago to be Michael Turner’s first graduate student. He discusses his interest in approaching cosmology from the perspective of particle theory and his thesis focus on curved space time within a cosmological context. Frieman describes his postdoctoral work at SLAC and his first position at Fermilab in the theory group that Dave Schramm had started. He discusses his work on the Sloan Digital Sky Survey and then the Dark Energy Survey. Frieman explains what might be needed to understand dark energy, he describes his appointment at Chicago, and he explains the origins of the Magellan Telescopes project. He discusses the value of the Aspen summer sessions and his involvement with P5, and explains the value of the 2010 Decadal Survey. At the end of the interview, Frieman surveys the current slate of project at Fermilab and emphasizes the value of incorporating cosmological perspectives to high-energy and particle physics.
Okay, this is David Zierler, oral historian for the American Institute of Physics. It is October 6, 2020. I'm so happy to be here with Professor Joshua A. Frieman. Josh, thanks so much for joining me today.
It's my pleasure.
Okay, so to start, would you please tell me your title and institutional affiliation or affiliations as it were?
Sure. I'm currently head of the Particle Physics Division at Fermilab, and also a professor of astronomy and astrophysics at the Kavli Institute for Cosmological Physics at the University of Chicago. It's a mouthful.
Yeah. Now, did, did those two positions sort of automatically work with each other in terms of your appointments? When you took on one did you take on the other at the same time?
No, they're separate institutions. Fermilab and the University of Chicago obviously have strong connections, but they're really separate positions. There are a number of people who have this kind of arrangement, like I do, but it's not an automatic thing.
Generally, how do they work in mutually beneficial ways, both in terms of the research, in terms of the kind of people you interact with, in terms of your teaching?
I can only speak for myself on this. What the university provides is the academic environment, access to students and postdocs, and a much broader intellectual environment across astronomy and astrophysics and physics than Fermilab. Plus, all the benefits, I would say, of a private research institution. Fermilab is more narrowly focused on high energy physics, particle physics and cosmology. And it has the tremendous resources of a DOE national laboratory, the ability to do large scale projects, an incredibly dedicated, and really first rate, scientific and technical workforce. So they're really very complimentary.
Now, when you have graduate students, is it assumed that their graduate work is going to be relevant to what's going on at Fermilab or not necessarily?
Not necessarily, but ideally, yes, because my work spans both institutions. With students, generally you start off by suggesting a research problem and seeing how it goes. If it works out, then by mutual agreement you continue guiding their PhD research. In that phase, I will continue to suggest various research topics, but I also encourage them to find research projects that excite them—if that moves them away from the research at Fermilab, so be it.
Because I think that in the long run to be successful students need to find their own paths, and if sometimes that may take them away from the core of my research, that's fine. So there are no limitations like that. On the other hand, we have had some students who have worked more closely along the primary lines of research at Fermilab. And yes, that tends to be a narrower scope within particle physics, high-energy physics, cosmology.
Well Josh, let's take it all the way back to the beginning. Let's start first with your parents. Tell me a little bit about them and where they're from.
Sure. My father grew up in New York, in Hell’s Kitchen and in Brooklyn. He became a physicist, a plasma physicist, and worked at the Princeton Plasma Physics Lab while I was growing up. My mother was originally from Washington, DC. They met shortly after World War II. My father had been in the Navy toward the end of the war. He had been a deep-sea diver and witnessed some of the early post-war atomic bomb tests at Bikini Atoll. When he came back from the war, he went to grad school at Brooklyn Polytechnic to get his PhD. Around the time he was finishing, the Korean War was starting up or about to, and he was still in the Navy Reserves.
So the story told when I was a boy was that he went to some military office and said, “Look, I'm getting a PhD in physics, what can I do?” I know this must be embellished, but the story was that someone in this office held up two phones and said, this phone connects to Los Alamos (where they had done the Manhattan Project) and this other phone connects to Princeton; both places were involved in research related to the hydrogen bomb, but I don’t think he knew that at the time. He said, “Well, New Jersey, that's much closer to New York.”
So that's the story of how we ended up in Princeton. I don't know if that's completely true. But he did certainly go to Princeton to work with Lyman Spitzer and John Wheeler. Although the group there was initially doing work related to the hydrogen bomb, it quickly spun off a research effort on controlled fusion, which we're still, 70 years later, trying to have as an energy source. Lyman Spitzer was a real visionary in many ways. And so my father got rapidly involved in that effort, and as a result plasma physics became his primary academic work.
Josh, I love Brooklyn geography and locating where all of these famous physicists from Brooklyn came from – where did your father grow up? What neighborhood?
Oh, gosh, I don't remember.
Did he go to Brooklyn Tech?
You know, I don't even know where he went to high school. For college he went to Columbia, they had a sort of forerunner of ROTC during the war. He went there and studied engineering. And as soon as he graduated, he went into the Navy as an officer, briefly went to the South Pacific toward the end of the war. But that's a good question. In fact, there's an AIP oral history with him somewhere. So...
Is there really? Oh, wow, that, you know--
I believe so.
--having the, the parent child dual oral histories is something very exciting. We have a few of them. So that will be fantastic to be able to add you to that. Where, where did your parents meet? Did they meet in Washington?
They met on Fire Island, which was, at that time, a popular summer vacation spot for those on the East Coast. The story I heard was, they got romantically involved, but my maternal grandfather felt that my mother was too young to get married—she was still in high school or had just finished—he insisted she go off to college and was convinced that when she did that they would just part ways, and he wouldn't have to worry about this for a while. But they stayed together and eventually got married.
And did you grow up in Princeton?
Yeah, I grew up in Princeton.
And your father's whole career was at the laboratory there?
I would say the main part of his career was at the plasma physics lab there. And then he went for three years to Washington, DC. He was at what’s now the Department of Energy. I'm not sure, I guess it was already the Department of Energy then. In the Carter administration.
Right. That's when it would have changed over from AEC. Yeah.
Yeah. And then he decided not to go back to Princeton. They moved out to San Diego. He worked for a private research and development company for a few years, called SAIC. And then the last part of his career, he was director of the Scripps Institution of Oceanography in San Diego.
Now in Princeton, did he have affiliations with the Department of Physics, or did he spend time at the Institute at all?
He was mainly at the Princeton Plasma Physics Lab, a DOE lab, but he also taught in the Department of Astronomical Sciences at the University.
Now growing up was his style as a father, would he involve you in his career? I mean, did you grow up sort of understanding the professional world of physics and what he did?
Well, I'd say yes and no. I think, until I learned a little bit of mathematics and physics in high school, I didn't really have a connection or an understanding of what he did. I think I remember going to the Plasma Physics Lab once or twice, but it wasn't like we hung out there. We lived relatively far away from it. It wasn't a place where we would go on the weekend.
So, he wouldn't take you to the lab on Saturday morning? So that wasn't sort of the kind of thing he would do?
Not really, he was a theorist, he wasn’t tinkering in the lab. But there were other ways I was exposed to physics growing up. It's sort of an odd story. But...we can decide whether we want to include this in the transcript or not?
OK, this is a personal anecdote about my father, but he's no longer with us so he can't object. After dinner every night, he would go into the bathroom to do physics. There's no other way to put it. At that time, he was a smoker. I think he felt that was where he could smoke, with the door closed and the fan on. He had these long yellow legal pads and a felt tip pen, and he would do calculations, write equations. And then he would leave half of the sheets in the bathroom trash can, because he didn't need them anymore. I remember occasionally, when I was in high school---you go into the bathroom, there's nothing to read. It was like, oh, here's something. I would fish out these yellow sheets of paper. And I remember it looked like hieroglyphics---it was completely incomprehensible. It was really only much later that I realized he had been solving the equations of magnetohydrodynamics related to plasma physics. I certainly didn't have an understanding then.
But I knew that mathematics was the language of his work. And clearly, it was a very powerful language. So that was one exposure. Another one was later in high school. I was taking physics, and we had to do a project and write a report about some research topic in physics. So I decided I was going to interview him, and write a report about fusion. So that was the first time that I learned at a superficial level a bit more about what he was doing. I tried to read some textbooks about plasma physics and fusion, what they were trying to do. And I remember very clearly him saying that we would have fusion energy in 20 years. This was in the mid-1970s. And the joke was that whenever you would ask someone, when are we going to have fusion, it was always 20 years.
To put a coda on that, about five years ago, I was invited to give a colloquium at the Princeton Plasma Physics Lab. I hadn't been there since I was a child, and my father had since passed away. It was a very nice visit. They gave me a tour, probably of some places I may have seen as a child, but I didn't remember. I remember telling them, when my father worked here, he said we would have fusion in 20 years. And my host said, “Yeah, now it's fifty years.”
They told me that the fundamental physics issues of plasma instabilities, which my father had worked on, were largely understood, but that it's really become an incredibly complex engineering problem, which is going to take decades to solve. So that was the extent of my direct connection with what my father did. But yes, undeniably it had an influence on me.
Josh, growing up, did you go to public or private schools in Princeton?
I went to public school for elementary and middle school, and then I went to private school for high school.
And in high school, were you a standout student in math and science yourself?
Stand out...no, I think I was a good student in math and science. I went through AP in math and physics. So, I would say I was a good student--good enough.
When you were thinking about college, were you thinking specifically about physics programs?
No, not at all. As we talked about earlier, physics was always something in the background, because I knew my father did it. But it wasn't something I was focusing on. I was interested in history and philosophy. I certainly didn't go to college thinking that I was going to become a physicist. I knew it was a possibility. But at that time, I was actually much more interested in things like history and philosophy.
What schools did you apply to as an undergraduate?
What schools did I apply to? Wow, okay. I’ve got to dig into the memory. Let's see. Well, I know I applied to Stanford, because that's where I went. I applied to Harvard, didn't get in there. Let's see, I think Williams, Cornell, Michigan...that’s all I remember.
So it sounds like specifically not an MIT or a Caltech where you would have had a more limited technical education.
Yeah. I wasn't interested in that at all.
I definitely wanted a broad liberal education.
And what year did you start at Stanford?
Okay, and so at what point did you get more involved in, in physics? When did you sort of start to look at that as a possible profession or career path?
Well, I know it wasn't during my first year. I had enough AP credits to place out of freshman physics anyway. So my freshman year was really all about liberal arts. I think it was toward the end of sophomore year, when I was like, “Okay, I've got to start thinking about a major.” I had taken a lot of philosophy courses, and it was either physics or philosophy, or do I try to do both? And I honestly don't remember the thought process that led me to physics. It wasn't my father saying “You have to follow in my footsteps and become a physicist,” there wasn't pressure like that. On the other hand, there were certainly discussions about employability and that having a scientific background could be a launching pad for a number of different possible careers. Whereas, at that time, my perception was that there were quite a lot of unemployed philosophers.
Also, during my sophomore or junior year, there was a physics colloquium given by Dennis Sciama, a cosmologist visiting from Oxford. He had been Stephen Hawking's thesis advisor. When he gave this colloquium in the late 1970s, it was the period of time when the modern rebirth of cosmology was just starting: particle physicists were realizing that they could understand the very early universe through the application of particle physics theory. In Sciama’s talk, he started at the present time in the universe and worked backward in time; by the end of the colloquium, he had gone back to a tiny fraction of a second after the Big Bang. And the fact that someone could talk about what happened 10 to the minus 35 seconds after the Big Bang was just really captivating to me. I think it also appealed to my earlier interest in history, this idea that you could treat cosmology as like archaeology on the grand scale, that you could use cosmological observations like pottery shards to reconstruct the beginning of the universe. As an idea, that really just grabbed me.
That sparked my interest in cosmology. At that time, cosmology was not a big field and typically wasn’t taught to undergraduates. So at the beginning of my senior year—this must have been the Fall of 1980—I went to a professor, Clifford Will, one of the leading experts on tests of Einstein’s general relativity, and said I'd like to do an independent study to learn general relativity and cosmology. And he said, “Great,” and I basically spent the year going through what at that time was the big book in gravitation and cosmology, by Steven Weinberg.
And by the end of the year, we ended up doing a little research problem together. So that was just following my nose into something. But the actual decision to go into physics, I'm not quite clear how I arrived at that.
So nowadays, cosmology, I mean, there's so many people who are interested in cosmology, the field is just exploding. It sounds like what you're saying is in the late 1970s, in some ways, your interests were ahead of the curve. And really, even in terms of the faculty, there weren't so many people, even at a place like Stanford, who were specifically focusing on that.
Yeah, I think that's right. Bob Wagoner was there at the time, and he was, I think, the only cosmologist on the faculty, but because of the connection to general relativity I ended up going to talk to Cliff. I don't remember how or why when I approached him, but he was interested or at least willing to supervise my independent study. But yes, it was certainly the case that it was not really established as a major field of study at that time, it was more of a startup field I would say.
Now, it turns out that, just up the road from Stanford at SLAC, was Alan Guth, who was a postdoc.
And he was inventing inflation. So I remember toward the end of my senior year of doing this study of cosmology, Cliff Will gave me Alan Guth's paper to read and said, “This looks like it might be interesting.” We had been studying a bit of cosmological perturbation theory, reading a paper by Bill Press and Ethan Vishniac, to understand how density perturbations evolve. And he said, “Let's look at perturbations in this inflation model.” And I was like, “Okay, fine,” I had no context for this. These were probably the first scientific papers I'd ever read. I didn't realize how epoch-making Guth’s paper was for the field.
Of course, I didn't understand most of it, because I hadn't taken quantum field theory or studied grand unified theories (GUTs). But with Cliff’s help I gleaned enough of it that we could do this little project, which we eventually wrote up for publication. So for a short while after that, I assumed that every paper in physics was like Alan Guth's paper on inflation.
It's a high bar.
Yes, so I thought cosmology is pretty nice. I didn’t realize that his was a once in a generation paper. So--
So Alan was a postdoc-- Alan was a postdoc at SLAC? Not in the department?
Yeah, that's right. He was in the theory group at SLAC.
That's interesting. I didn't know that. And that actually, I was going to ask you—so did you spend a lot of time, did you hang out at SLAC? Was that a good place for an undergraduate interested in cosmology to be at that point?
No, I worked there one summer in Burton Richter's high-energy physics lab, just as a summer job, working on some equipment, trying not to electrocute myself. But I wasn't involved with the theoretical stuff going on at SLAC. I think Alan was there as a particle theorist who got interested in cosmology when he was there. But I was a postdoc in that same group several years later. I was kind of the lone cosmologist. I think it was sort of like this exotic, rare species that you--
--you bring them in and see if they'll survive.
And Josh, just to be clear, I mean, nowadays, so many people in particle physics use that as an entree to cosmology. But for you that was, that, that, it did not happen that way. You didn't start with particle physics and get to cosmology, that route, you sort of started with cosmology.
Yeah, I started with cosmology as an undergraduate.
Then filled in particle physics in graduate school; so I really began from this interest in the very early universe, in trying to understand the origins of the universe. That's where I started, not from particle physics.
Right, right. So by the time you finished your undergraduate, you were pretty confident for graduate school, you wanted to stay on the cosmology track.
And so, did that influence at times the programs you applied to? Where would be a good place to do cosmology?
Yeah, there were very few places.
I didn't know anything; I didn't really research it much. I remember asking Cliff Will, where are there places with people doing cosmology, and it was a pretty small list. So I ended up going to the University of Washington in Seattle, because there was a physicist there, Jim Bardeen of the famous Bardeen family, who had written what became a very influential paper on cosmological perturbation theory. I think that shaped my interest in going there to work with him. I went to Seattle for a year, but I think I quickly realized that my interests weren't exactly aligned with his interests.
While I was at UW, Michael Turner visited from Chicago, he had recently joined the faculty there, and he came to give a colloquium. And for me, it was like Sciama's earlier colloquium, going back in time to the very beginning of the universe, the thing that had sparked my initial interest in cosmology. Michael had a very engaging style of presenting things, kind of theatrical, I would say. And I was like, wow, there are people in the US doing this kind of thing, not just in Britain. Michael and David Schramm were both at Chicago: a place with two cosmologists, that's where I should be if I want to combine particle physics with cosmology to study the early universe. Jim Bardeen was less connected to the particle physics side, his background was more on the general relativity end. So I decided to move to Chicago for my second year and the remainder of my PhD and ended up as Michael Turner's first thesis student.
Now, so your motivation to go to Chicago was because of Michael, or you really developed that relationship after you had settled on Chicago?
No, it was definitely because of that. Chicago was the only place where there were two people doing that. Most places had zero.
So in my view it was the center of the cosmological universe--
--so to speak. And I like to think it still is. So my decision was very specifically to go work with that group.
And intellectually, Josh, what was compelling to you about needing that foundation in particle physics as a route to, you know, developing further your expertise and appreciation for cosmology?
I think it was just the fact that, to make models and theories of what happened in the first few seconds or minutes of the universe, you have to use the language of particle physics. That's the theoretical framework for understanding how particles interact at those early times, and therefore, to understand how the universe evolved at those early times, you have to have that marriage of those two disciplines. I think I realized that, not having yet learned particle physics, it would prevent me from exploring that. I really liken it to, as I talked about earlier, learning a language. Mathematics is the language of physics, but particle physics is really the language that you need to speak to understand what happened in the early moments of the universe.
Now in emphasizing the unique approach of a Dave Schramm and a Michael Turner, it's, it's interesting-- What about somebody like Steven Weinberg? Did you not see somebody like Weinberg having that same approach in terms of applying particle physics to cosmology?
Yes, he certainly did. But also, I think he was less centrally focused on cosmology. He certainly wrote that great book-- it was fantastic. But he is primarily a particle physicist. And at that period of time, he was getting into some of the more theoretical ends of the discipline. Later on, I did consider going to Texas for a postdoc, where Steven Weinberg had moved, but ended up going to SLAC instead.
And Chicago was exciting because it was a young, growing group. It was Michael and David, and soon after I arrived Rocky Kolb moved to Fermilab. And they brought in a bunch of postdocs, so there was just a vibrancy there from that collective that you don't get working just one on one with a single faculty member.
And so what, what were some of the, the most exciting developments in particle physics when you were a graduate student? What were some of the big ideas in that field at the time?
In particle physics? Well, I think the main one was the triumph of the standard model of electroweak interactions. And then the discovery of the W and Z bosons at CERN, which solidified that. That gave people confidence that gauge theories of elementary particle physics were a key foundation of particle physics. And also coupling that with the notion of spontaneously broken symmetries: you have this theory, which has symmetry in it, but the world doesn't. So, something happened. That symmetry that's in the theory was broken in nature, the theory of how that happened involved some kind of phase transition in the early universe. That got people quite excited about what happens in a phase transition in the early universe.
And that's what led to Guth's theory of inflation: he reasoned that the GUT phase transition could be a first order phase transition, the universe could get held up in the symmetric phase and take a cosmologically long time to reach the broken phase. When it’s held up in that symmetric phase, the universe will expand exponentially, and that could solve these cosmological problems. Alan came at it very much from the particle physics side. So this notion of the Standard Model, spontaneously broken gauge symmetries, and then extending them to larger symmetries---the whole notion of grand unification had come up a few years earlier.
The first Grand Unified Theory of Georgi and Glashow gave us a way of thinking about very early cosmic epochs. That was the framework for Guth's thinking about inflation, but more broadly you had this developing theoretical landscape of models that could allow you to meaningfully speculate about the first tiny fraction of a second after the Big Bang. Another theme was evolution of the inflation model. There was Guth's original model, in 1980 or 81. And then in the early 80s, Paul Steinhardt and Andy Albrecht, and Andrei Linde who was still in Russia at the time, developed what became new inflation, which solved some of the problems of Guth's original model. And that started this explosion of model building. Those ideas were very exciting, and other sorts of things would pop out of these phase transitions, for example, magnetic monopoles, cosmic strings, all these things that could be produced in phase transitions. It was very rich, and it seemed like every week there were new ideas coming out.
Also, the notion that you could use these models to explain why there's matter and not antimatter in the universe, the baryon asymmetry, that was all percolating in the late ‘70s and early ‘80s as well. So it just seemed that there was this tremendous ferment that had been released by this notion that you could apply these particle physics theories to the early universe. And you're right, that really brought in a lot of people from particle physics theory into the field, which I think was a good thing. Frankly, cosmology had been a bit of a lonely backwater for decades before then.
Josh, in terms of your own development, what, what did you see as your skills or abilities that, that you wanted to draw on? In terms of your own contributions to the field? What did you see that you were good at, that you could add at, you know, at this very exciting moment as the field develops?
That's a good question. I never thought of that.
I mean, because there's, these are personalities we're talking about as well, people that have different skill sets, there's creativity, there's abstract thinking, there's just raw mathematical ability. What, what did you discover as a graduate student, as you were developing your professional identity, you know, the kinds of things that came naturally to you or that you were good at?
Hmm, that's a good question. To be honest, I'm not sure. I certainly wasn't the best mathematician, by far. I could solve problems, but certainly there were people I knew who were much more technically adept at mathematical equation solving.... The best answer I can come up with is that I tried to develop a way of thinking about problems, it's almost a sort of analogical thinking. I don't know if this makes sense. But, I think part of what I've tried to do is to try to apply lessons that we've learned, maybe in other problems, or in other areas of physics. I think a lot of progress gets made when you realize that this problem looks or smells like this other problem. And people have developed fruitful approaches for how to solve these other problems—let's try to apply that here. It’s not necessarily thinking by analogy but using a sort of “lessons learned” approach.
This comes up all the time in physics. We tend to think of physics as these separate sub-disciplines. But I really am a believer in a kind of unified view of physics. Techniques and methods and ideas in one area of physics often catalyze developments in other areas, for example, there have been a lot of people who go back and forth between condensed matter physics and particle physics. I think part of what I've tried to do, at least in some of my work, is apply that lateral, analogical thinking. It doesn't always wor-- it's not always applicable. But I think everyone develops a certain methodological toolkit. And when you're holding a hammer, everything looks like a nail. So at least for a certain phase of my career I felt that was a fruitful way to approach things.
Now, what was, what was Michael's style as a mentor? Would he encourage you to develop ideas on your own or in terms of how you put your dissertation topic together? Did he essentially hand you a problem that was relevant to what he was working on at the time?
Yeah, good question. I think early on, the problems that I worked on were ones that he had suggested. And so we worked together on those. That was useful, learning how to do physics by working with someone who's done it. For my thesis, I decided to go off on my own and got interested in a problem that wasn't connected to what Michael was doing. I bounced it off him, made sure he was okay with it and that he thought, if I made progress, I would get a PhD out of it. There was this sub-area of marrying quantum field theory with general relativity called quantum field theory and curved space time, there was a famous book on it by Birrell and Davies, where you would take different quantum field theories, put them into a curved spacetime of general relativity, and then look at what happened. And you get interesting phenomena that can occur because of the fact that this model is living, not in flat Minkowski space time, but in a curved space time. I was interested in pursuing that in a cosmological context.
So, for example, there are situations where you can get spontaneous creation of particles in the very early universe, because the universe has high curvature. As you get closer to the Big Bang, those effects tend to get more significant. That was different from what Michael was working on, but he basically said, “Yeah, okay, sounds interesting—just make sure you solve some problem that you can do.” It was a little bit harrowing. I didn't really start the project until the beginning of the summer when I was supposed to graduate that fall. And I had a postdoc lined up at SLAC. I definitely would not let a student do that these days. I don't know if it was because I was Michael's first student, or he just had enough trust in me that I would get something useful out of it. But it was interesting. It was fun.
Who else was on your committee?
My thesis committee? Oh, gosh, good question...
And the broader question there was, I assume, besides Dave and Michael, did you self-consciously want to have a good representation from both cosmology and particle physics?
I don't think I thought about that. And I'm trying to remember who was on my thesis committee. Yeah, I don't remember honestly, this is embarrassing. I think Bob Wald was on it, he had done a lot of important work on quantum field theory and curved space time. But I honestly don't remember who else, I don't remember the thesis defense, any of that.
What were you--
I think I was probably moving to Stanford in a week. So it was like, OK, let's just get this done.
Get it submitted. Get the defense and then move on. I don't think I was... Yeah, there wasn't a lot of thought that went into it.
Josh, you mentioned you were considering Texas for a postdoc. Where else were you considering? What were some of the intriguing possibilities for you, after you defended?
Um... Well, again, after I defended, I'd already accepted months earlier a postdoc at SLAC. I'm trying to remember where else... Texas, Maryland, I can’t remember where else. But I think SLAC was attractive for a number of reasons—I had enjoyed my time as an undergraduate there, I liked the West Coast, it had a large theory group, which I had found stimulating as a graduate student. So, I think all that attracted me.
Mm hmm. So what, what was going on at SLAC? By the time you got there, what were some of the exciting developments?
String theory. This was now 1985 and there had been this explosion in string theory in 1984, 85.
John Schwarz, Michael Green, Ed Witten, many others. And then it just took off and was like a wildfire racing through particle physics.
Yeah. Did you work with Leonard Susskind at all?
I didn't. I decided that I wanted to really focus more on cosmology. There were some people thinking about string theory and cosmology. As a graduate student, I had written a paper with Rocky Kolb on Kaluza-Klein theories and cosmology. While not string theory, Kaluza-Klein theory does involve extra dimensions. While that was fun, I decided not to continue in that direction.
I wonder if, you know, coming back to Palo Alto after your time at Chicago, if your sense was even in that relatively short interim cosmology was sort of less of an exotic endeavor at that point. More people were getting interested in it. There was, you know, just more work being done.
Having been in Chicago, that was certainly the feeling. But it took a number of years for other institutions to really grab onto it. I think part of the reason they offered me the postdoc at SLAC was that it was a low-risk investment compared to hiring someone on the tenure track. A postdoc will only be here a few years, if it doesn't work out, that's fine.
And it's just theory anyway. Just give him a blackboard, he's all set.
Perhaps they also felt that Alan Guth had done a good job there and wanted to try to reproduce that. I actually ended up working more with people from the physics department at Stanford. Savas Dimopoulos had gotten interested in cosmology, and he had a couple of students working on it. And there were a couple of other students interested who were mostly working on their own, so I got connected with some of them and collaborated with them as well. I also worked a bit with Brian Lynn, who was there. But I think it took quite a number of years for Stanford and SLAC to decide that cosmology was a growth direction for them. I don't think I changed their minds on that. And, of course, now both SLAC and Stanford have major efforts in cosmology.
Right. But even, as you're saying, even as a postdoc, you were still sort of ahead of the curve in terms of cosmology developing.
I think they would have viewed it not that I was ahead of the curve, but that it was a curve they weren't yet interested in. They didn't want to be on the curve.
Right. Yeah, sure, sure. But obviously, in retrospect, it was a curve that they should have been on, given the, the explosive growth of the field not that long after.
I suppose that's true.
Josh, when did you go on the job market? It was a two-year postdoc? At SLAC?
It was a three-year postdoc. After one year, I kind of tested the faculty waters a bit but decided not to take an offer. And then after my second year was when I applied more seriously.
And were there open positions specifically in cosmology at that point? It's an important question, because open faculty positions is a great shorthand for the status of a subfield at any given history and time. And so, if there were top physics programs that were specifically advertising for a cosmology position, you know, that really says a lot.
Yeah, that's a good question. In retrospect, I think particularly the first year I applied, it was mostly not cosmology focused positions, it was positions in physics departments where they were looking for someone interested in theoretical astrophysics, or it was positions in astronomy departments where they were looking for someone doing theoretical astrophysics but not necessarily cosmology. And I remember going to a couple of places and feeling that we were not really speaking the same language. Which was fine. It was interesting to learn about different areas of physics and astronomy research.
As a graduate student, I was in the physics department, and I picked up some knowledge of astronomy and astrophysics, but was mostly self-taught. And at SLAC I was mainly interacting with particle physicists. So it was interesting to interact with people who were professional astronomers, but I think they viewed what I was doing as a bit on the fringe of astrophysics. But by the next year, there were more places that were interested in building cosmology groups. For example, Gary Steigman had moved to Ohio State and was starting a group there. But it was still a relatively small number of places.
So, so what was available to you? Where were you able to apply where you thought that you had a reasonable shot given your background?
Again, I don't remember everywhere I applied. I think I had faculty offers from Fermilab, Michigan, Ohio State, University of California, San Diego, plus a second postdoc offer at LBL, and an informal postdoc offer from Bill Press at Harvard CfA. I don't remember all the places I applied to and interviewed.
But to get back to that first question with Fermilab, that was separate from University of Chicago? That was a full-time appointment at, at Fermilab?
Yeah, that's right. And again, the group at Fermilab had been started by David Schramm.
David convinced Leon Lederman, who was the director of Fermilab at the time, to start a cosmology group as kind of a subset of theoretical physics at Fermilab. In the early 1980s, they really developed and built that group up. They brought in Rocky Kolb, Mike Turner was spending a lot of his time at Fermilab and living near the lab, David was spending part of his time there, plus they brought in Andy Albrecht and Neil Turok. When Neil moved to Princeton, I was offered the job at Fermilab.
When I had been a graduate student at Chicago, every Monday a few of us would drive out to Fermilab for the weekly seminar. And then on Monday evenings, in Mike Turner's basement was primordial pizza, where everyone would get together as part of a journal club to discuss the latest developments. So, there was a lot of cross fertilization between Fermilab and U Chicago. When I went to Fermilab as a tenure-track scientist, I told them I was interested in teaching. And so, within a year or so, I was teaching at Chicago as well. And through that, I started working with graduate students, etc.
It's such a great base of comparison you had coming from SLAC to Fermilab in this sort of rarefied world of, of theoretical cosmology. What were some of the differences in terms of the approach at Fermilab?
Hmm. Compared to SLAC?
Yeah, I mean, just because it's such a, it's such a perfect comparison. It's really an apples to apples [comparison] because it's theoretical cosmology within the larger context of a major national laboratory and not, you know, a smaller sort of physics department.
Yeah, that's a good question. I think it was very different because at SLAC at that time cosmology wasn't seen as a central activity, it was still a little bit of a fringe activity. I think the real focus there was on developing string theory and other areas of high energy physics. Whereas at Fermilab it was becoming a major activity. So it was just a very different environment. It was also I think, a younger environment. Michael Turner and Rocky Kolb were still in the early phases of their careers then. And David Schramm brought a lot of excitement. At SLAC, I was the postdoc doing cosmology.
Whereas Fermilab had five postdocs doing cosmology. So, the level of activity was quite different, I would say. Through David Schramm, Fermilab had really embraced this, while other labs were just testing the waters at that time, but hadn't yet jumped in. Yet even with the large group that Fermilab had, there was a long period where there was skepticism among the particle physics community that this was a thing that should be happening at the national labs. I expect there was some of that resistance at SLAC and elsewhere as well—a sense that the national labs should stick to hardcore particle physics that has a direct connection to the experiments they are doing at the labs.
And overall, cosmology was still seen as a bit on the fringe. The joke was that in cosmology, you had order of magnitude uncertainties in the exponents of the cosmological parameters. This was the period where there was great theoretical ferment in cosmology, but still almost no data. So, it was much more of a purely theory-driven field than particle physics was. Particle physics had the triumph of the W, the Z, all the quarks, every piece of the standard model was falling into place, there was this well-established standard model. And I think, frankly, there were people looking down their noses at cosmology as like the Wild West: they can say anything they want, and you can't test it.
And that resistance was there for quite a number of years, even after the group became established at Fermilab. It took quite a bit of time for cosmology at Fermilab and at the other national labs to really become accepted and embraced.
Josh, in that vein in terms of understanding where cosmology fits into things, looking ahead, when you are named head of the theoretical astrophysics group, what might we read into that in terms of the things that you were really interested in, but administratively, you know, where, where the research fit into the overall structure of, of the, of what was happening at Fermilab in the mid-1990s? Sort of like, like a, like a Venn diagram, like where, where's the overlap with astrophysics and cosmology where, you know, as you were saying, like cosmology might have been the Wild West, astrophysics might have been sort of more comfortably ensconced in the, you know, the hard physics, you know, program at Fermilab.
Theoretical astrophysics, at Fermilab, was essentially cosmology.
Okay, so yeah, so those are interchangeable terms, as far as Fermilab is concerned?
We called it theoretical astrophysics, but it was basically cosmology.
If, I mean, what, what might we read into the fact that you called it theoretical astrophysics and not cosmology? Because there are, of course, meaningful distinctions there, even, even if Fermilab itself might not have appreciated that.
I'm not sure. I don't know the origin of the name. My guess is, there was a theoretical physics department at Fermilab, so there would be a certain symmetry to having a theoretical astrophysics group; probably this was also recognition that not everything we did would be neatly classified as cosmology.
Most of what people in the group did was what you would call cosmology. But, over the years, increasingly people branched out into adjacent areas of astrophysics. In terms of my becoming head of the group, by 1994 Rocky Kolb had been head of the group for 11 years, and I think he wanted a break. So I was next in line to shoulder the burden for a few years.
Josh, I want to ask about some of the major scientific projects and research collaborations that I know that they overlap chronologically, so we can sort of toggle back and forth. So let's just start with the Sloan Digital Sky Survey. What, when did that begin? And were you sort of, you know, present at the creation?
I wasn't present at the creation. I think it was first discussed in the late 1980s, and I became peripherally involved in the mid-1990s. And that was because my interests had evolved toward large-scale structure. As we discussed earlier, from the late 1970s through the end of the 1980s, the renaissance of cosmology was really driven by theory. In the late 1980s and early 1990s was this recognition that development of new technologies would enable major advances in observational cosmology, to test those models. This involved development of bigger telescopes, new detectors, and the advancing technology of Cosmic Microwave Background experiments.
The Sloan Digital Sky Survey was designed as a cosmology probe from the start, building on advances in CCD detectors, and that was something new. I was getting increasingly interested in the idea of testing these models that we had for the early universe. After 10 or so years of theory, there was a sense that we have a number of theoretical ideas here, but to really make progress, we're going to need some way of testing so I can tell, is this model okay or is it ruled out? And the only way to do that was to have data.
The Sloan Digital Sky Survey (SDSS) was going to be the first really large-scale survey that could probe those ideas. Before Sloan, in the early to mid-1980s, were some of the first galaxy redshift surveys. In 1986, the Center for Astrophysics published the first redshift survey of 1000 galaxies, and it showed this very clumpy large-scale distribution of galaxies, there was a famous picture that looked like a skier or a stick man. People realized that these maps contained a lot of statistical information that could inform our understanding of how large-scale structure formed. I think the initial SDSS discussions were among astronomers at Princeton and Chicago and elsewhere.
But then at some point, they made a pitch to Fermilab that this is something that a national lab could contribute to, because it was going to be a large-scale project, which national labs are good at doing—they know how to build and manage large facilities. I really got interested in it through the 1990’s as the project was being developed and eventually found myself spending more and more time on it-- it was aligning better and better with my own research interests in large-scale structure.
Josh, were you working at all with, with Michael Turner in the late 1990s, when you know, dark energy really hits the scene?
Not in the late 1990s, no. My interest in dark energy stemmed from the mid-1990s. Going back to the early 1990s, there were starting to be hints about what we now call dark energy. There were results from a galaxy survey in the early 1990s called the APM survey, and there were measurements of the Hubble constant in the mid-1990s that were getting very large values, a Hubble constant of 80 km/sec/Mpc. The only way to reconcile that with other data was to have something like dark energy. But there were also limits from strong gravitational lensing statistics that appeared to be in contradiction with the idea. I did some theoretical work in the mid-1990s trying to explain these different observations with a model of what later came to be called Quintessence. This came out of some earlier ideas that we had had about inflation going back to the early 1990s. So that was what sparked my interest in dark energy.
In 1997, we decided to organize a workshop at Fermilab called The Missing Energy in the Universe. By that time, there were enough observational hints that there was growing theoretical interest in this. We wanted to get mostly a bunch of theorists together to discuss different ideas. The workshop didn't happen until spring of 1998. Meanwhile, in January of 1998, came the exciting announcement of the supernova results, the discovery of cosmic acceleration. So, suddenly there was tremendous interest, and this little workshop we had been organizing suddenly had a lot of people wanting to come and discuss this. It was starting to look like cosmic acceleration was a real thing. So, I got into dark energy starting, theoretically, a few years before the supernova discovery. And then, we soon realized that to test models of cosmic acceleration we were going to need a lot more data, including supernovae and observations of large-scale structure.
And the Dark Energy Survey is specifically responsive to that need for more data?
Oh, yes, we started that project in the early 2000s specifically with that goal in mind.
Yeah, and in terms of instrumentation and computational power, what have been some of the most fruitful avenues for, for gathering that needed data?
Well, I think the main one is the ability to put together large cameras comprising mosaics of CCDs that can cover a very large focal plane, as the SDSS had shown. That's what the Dark Energy Camera was designed to be. And putting it on a good-sized telescope, the Blanco four-meter telescope at Cerro Tololo in Chile, which is an excellent astronomical site. Also, at Lawrence Berkeley Lab, they had developed thick CCD devices. Because they were thicker, they were much more sensitive to red light than conventional thin CCDs that had been used in astronomy up until then. We knew we wanted to cover both a large swath of the sky and to do a deep survey, sensitive to distant galaxies, which appear red, since their light is redshifted.
So, we wanted to have CCD detectors that are especially sensitive to red light, otherwise, the survey was going to take too long---we could do cumulative exposures of minutes rather than hours. That meant we could survey a very large patch of the sky, about one eighth of the entire sky, to sufficient depth in five years rather than decades. So, the Dark Energy Survey benefited from this combination of things.
Also, as you I think you're hinting at, when you get these really big cameras, that means you're getting a lot of data per unit time. So the fact that computing power kept growing as Moore's Law meant that, by the time we started the survey, the computing requirements for transmitting and processing the data were large, but they weren't, so to speak, astronomical. 10 years earlier, it would have been much harder, but we knew that compute power was going to keep on marching--
--ahead. And that by the time we got on the sky, it wasn't going to be an unprecedented challenge to process the data.
Sure. Sure. Josh, it's kind of a retrospective question, but it seems like a good place to ask it now. So, in the past 20 to 25 years, are you surprised that there's still so much we don't know about what dark energy is? Was there more optimism at the beginning of this quest, that, you know, sort of more fundamental understanding would have been achieved at this point? Or sort of looking at where we are now, does it sort of make sense? You know, how well developed the field is?
That's a good question. I would answer that in two ways. There are people who have expressed skepticism of doing these big surveys aimed at dark energy, such as the Dark Energy Survey, the Vera Rubin Observatory Legacy Survey of Space and Time, the Nancy Grace Roman Space Telescope, Euclid, etc. They would say we know what dark energy is, it's just Einstein's cosmological constant—it's just the energy of empty space. We don't know why it has that value, but that's the simplest explanation. And if you look at all the data to this point, it's all consistent with that hypothesis. So a skeptic would say, “Why are you bothering? We know what the answer is.” My view on that is, just because you know the answer doesn't mean that that's the answer.
An argument that I like to use for continuing to do these surveys is that we think this is the second time that the universe has been accelerating. We think the first time was inflation, a tiny fraction of a second after the Big Bang. When inflation ended, the universe stopped speeding up and slowed down for billions of years. And then 7 billion years after the Big Bang, it started speeding up again, a second phase of cosmic acceleration. While a number of particle theorists would say it must be the cosmological constant speeding up the universe today, we know it couldn't have been the cosmological constant that drove that first epoch of acceleration, because inflation ended and a constant is, by definition, a constant.
If it had been the cosmological constant then, inflation would have gone on forever, and we wouldn't be here. I like to imagine a roomful of particle theorists sitting around 10 to the minus 35 seconds after the Big Bang, when the universe started inflating. They would have looked around and said, it's just the cosmological constant, this is going to go on forever. But it didn't. So, the fact that they would have been wrong at that time to me is not a good argument for them being right this time.
That's a very important point.
To my mind, it's an empirical question.
We have a theoretical prejudice, and that theoretical prejudice may even be right, but ultimately, it's an empirical question that we have to answer. We've gotten much better precision over the last 20 years. And the data look consistent with the cosmological constant. But we need to get more and more precise measurements. I think with the Dark Energy Survey, we're going to reach the precision we expected to reach. And then the next generation of surveys, Rubin, Roman, Euclid, etc, will do even better. Personally, I would have hoped that we would have seen a departure from the cosmological constant by now, that would have been much more interesting. We haven't seen that yet.
But it's still possible that we will, and we have a lot more measurements to do with the Dark Energy Survey. We've only published analysis of a small fraction of the data. There are still potential surprises there. I'm still hoping for surprises, because we would learn something if we saw that dark energy was not the cosmological constant—if it was something else, that would be revolutionary. But even if we continue to minimize the errors, and it looks like the cosmological constant, that will tell us something, too.
Josh, what are we waiting for in the world of experimentation that might make for that sort of dramatic moment where dark energy reaches maturity in the field? Is there a, is there a LIGO for dark energy that, that we're waiting for?
I don't think it's quite like that; for gravitational waves, there was basically nothing that had the sensitivity to see gravitational wave sources, and then there was LIGO.
For dark energy and surveys, it is much more of a gradual progression. There were the early surveys, the Sloan Digital Sky Survey among them, in the early 2000s, and we have the Dark Energy Survey and other surveys now. And in the decade of the 2020s, we will see this next stage. Fifteen years ago, Rocky Kolb chaired a committee, the Dark Energy Task Force, that classified different stages of dark energy experiments. At that time, they were in stage two; the Dark Energy Survey is the quintessential stage three experiments.
The stage four experiments are the DESI Survey—which is just about to start—Rubin, which will start in two or three years, and Roman and Euclid in space. I think that those four projects together will be the next major step in the field. It will be just a tremendous amount of high-quality data. We'll learn a lot, and the error bars will continue to shrink. And the question is, will they continue to shrink around the theoretical prejudice point? Or will they start to exclude the theoretical prejudice point, which would be exciting. I think that's still a possibility. We don't know. That's where I think the field is headed over this coming decade of the 2020s.
Josh, at what point are you really involved with the department at Chicago in terms of teaching and taking on graduate students? Does that sort of happen right at the beginning a year in from Fermilab, or that's, that's a relationship and ongoing responsibilities that really grows over time?
I would say early on, it was primarily teaching at the graduate level. Over time, I became more involved in the Astronomy and Astrophysics department, serving on faculty committees, etc. Then I started taking on graduate students in the early 1990s. I've had a pretty steady flow of students since then. About twenty years ago, we started the Center for Cosmological Physics at Chicago, which is now the Kavli Institute for Cosmological Physics, and that elevated the level of activity substantially at the university. I’ve been teaching one quarter per year, a reduced teaching load, since my primary appointment has been at Fermilab. I haven't taught for the last couple of years because I've been in management as a Division Head at Fermilab. In 2021, I’ll step down from Fermilab management, and my primary appointment will shift from Fermilab to the university, so I expect to become substantially more involved in the department then.
When did the Magellan Telescopes project get going?
I’m not sure, but the project to build the Magellan Telescopes must have started around the early 1990s, and they saw first light in the early 2000s. The University of Chicago only got involved much later, well after the telescopes were built and operating. Chicago had a long history of forefront optical astronomy, going way back, but in the 2000s it did not have guaranteed time access to a large, private telescope. There was an interest in rebuilding that effort. We were part of the Sloan Digital Sky Survey, but that's a very different kind of thing: in a survey, you're part of a large collaboration, everyone gets the data. And there are certain projects that it's really focused around and limited to, whereas with telescopes like Magellan, individual investigators can study particular astronomical objects in great detail.
So, it's a very different way of doing astronomy. And that kind of principle investigator-driven astronomy had languished a bit at Chicago. Chicago was part of a consortium that owned a three-and-a-half-meter telescope, next to the Sloan Survey telescope, but by the mid-2000s that was a relatively small telescope, compared to the 6, 8, and10 meter telescopes, where you could do much more. Moreover, Chicago was hiring new faculty who use large telescopes, but didn't have a dedicated facility that they could use. So the university was able to allocate funds to purchase time and gain access to the Magellan telescopes. And we're also part of the next-generation Giant Magellan Telescope project as well.
Josh, when did you get involved with the Aspen Center for Physics? And what, what goes on there that's specifically useful for your research agenda, both in terms of the collaboration, your exposure to, you know, what other people are doing in the field?
I started going out to Aspen for workshops, I think 1986 was probably the first time I was there officially. For over 55 years, the Aspen Center has hosted workshops in a variety of areas of theoretical physics. Cosmology and astrophysics have been important topics there from before I started going there. Unlike a standard scientific conference, where you sit in an enclosed room with 200 other people, listening to people give talks from nine to five every day for a week, Aspen workshops are completely different. There are relatively few formal talks, typically outside at a blackboard, and they are much more focused on informal interaction in this beautiful setting in the mountains. The goal is for people to discuss and make progress on the cutting-edge problems in each subfield.
So rather than spending a few days in a darkened room listening to people, you could spend three weeks talking to people. That’s enough time for brainstorming, in-depth conversations, and developing new collaborations. For me, it was tremendously important in terms of meeting other people in my field, learning what problems they were working on, and often realizing that they were interested in or working on something related to what I was doing or interested in. From there you talk at a blackboard and you may develop a new collaboration. That’s the essence of the Aspen Center. Every summer 500 physicists come together over the course of the summer. There are about 80 or 90 there at any one time to discuss cutting edge problems, and it's all about these informal interactions. A lot of important theoretical work has come out of it. String theory was an example. It's a place I fell in love with, a place to combine physics with being in the mountains, being outdoors, hiking, biking, there’s a classical music festival, a lot of--
It beats Illinois in the summers, right?
Yes, but beyond the climate, what’s special about Aspen is that it's a place where physicists come together and forge new connections. A lot of physics conversations start on a hiking trail; you hike for a few hours with someone, you get to know them. You can talk physics, even though you don't have a blackboard. And it's just a unique atmosphere. And so, as I said, I started going there in the mid-1980s and just found it very enjoyable and stimulating. It's where I learned a lot of what's going on in the field, developed friendships and collaborations with people, with cosmologists from around the world.
The other interesting thing about the Aspen Center for Physics is the way it operates. It's a collective of 75 physicist and several non-physicist members who determine and execute the scientific program, with a paid staff of only 3 who actually operate the Center itself. New members are elected by the membership and are all volunteers; in some ways it resembles a large academic department at a university: the members serve on committees and rotate from committee to committee each year. Because I had benefited so much intellectually from being there for workshops, when asked to serve as a member, I was happy to do it. People are typically members for 10 years, and then rotate off, so we get new members cycling in each year. It was started in the early 1960s by three physicists, because they liked being in the mountains, I think.
For many years it was quite informal, but it also developed a reputation as an old boys’ network. There was a realization that it needed to be more diverse and dynamic. Especially over the last 20 years, we've worked hard to make it more inclusive and diverse along a number of axes, but that’s still very much a work in progress. I served two terms as vice president of the Center and since a year ago, I've been the president. For me, it's a way of giving back to this institution that has been important for my professional development. As you can imagine, COVID-19 has been quite challenging for the Center, because the essence of Aspen is people getting together informally. You've probably seen a number of large conferences have gone online this year.
That works for conferences, because people can give talks over Zoom. But for a place like Aspen, which is all about informal, extended interaction over weeks, you can't really replicate that over Zoom. We've thought about trying some virtual reality software where people can gather online, but it's not the same as sitting under the aspen trees with a few people over an extended period of time. So last spring I had to take the difficult decision to cancel the 2020 summer workshops. And we're not going to have people there this winter of 2021 either. We're now really focused on finding a way to have workshops there in the summer of 2021, with appropriate COVID safety precautions and protocols, etc. To my mind, it's a special place that's highly valued by the physics community, and we don't want to lose it. So, we're working hard to bring it back. I was out there a couple of weeks ago (in Sept. 2020), with only one other physicist around, but we realized that we could structure things in a way that, if the COVID situation next summer is like it was this summer, we could find a way to get people together safely, with low enough density and appropriate precautions, and that it would be a valuable experience.
Josh, I'm intrigued by your involvement with, it's a mouthful, Particle Physics Project Prioritization Panel, obviously that's why we know it as P5. How did you get involved in that? And is that indicative of the fact that, you know, even from your graduate school days, you've always been involved in one way or another with the world of particle physics?
P5 is a committee constituted by the High Energy Physics Advisory Panel to recommend priorities to the US funding agencies across all areas of particle physics every five or ten years or so. I was on P5 10 or 11 years ago. Earlier in this interview, we were talking about cosmology at the national labs and how, initially, it flew somewhat under the radar. But by the mid-2000s, cosmology was becoming more established as a bona fide branch of particle physics within the Department of Energy’s Office of High Energy Physics. That P5 panel laid out the notion that particle physics encompass three frontiers. One was the energy frontier, collider experiments that are now based primarily at CERN with the LHC.
The second was the intensity frontier, which encompasses precision experiments mainly with muons and neutrinos. And the third was the cosmic frontier, I think that was one of the first instances where cosmology was given that degree of legitimacy and prominence within the US particle physics program. I assume I was asked to be on that committee in part because of this recognition that cosmology was a vibrant and growing part of the field and because I was working at a national lab at the nexus of particle physics and cosmology. I was on the Astronomy and Astrophysics Decadal Survey a couple of years after that.
And again, that was for the National Academy, the Decadal Survey, right?
Right, that's every 10 years. There's another one going on now.
And are you involved with the current one?
No, I did my penance. I think that P5 was in a way a recognition that cosmology is a part of particle physics.
And I think the Decadal of 2010 was in some ways a recognition that particle cosmology, if you like, was a part of astronomy and astrophysics, as well. I think that both fields, astronomy and particle physics, realized that it was beneficial to claim some ownership of this interface area. Large projects such as the Vera Rubin Observatory were initially driven by cosmological concerns, but its mission has become broader within astronomy and astrophysics. The same thing happened with what’s now the Nancy Grace Roman Telescope. That's been a big change over my career—cosmology started as this fringe activity with only a few people, with few postdocs or faculty. Then it grew, but it was mostly theoretical.
And then it really expanded when it started becoming observational with the Sloan Survey, COBE, WMAP, Planck and other surveys. And then with even larger surveys, it had matured, and it was officially adopted by the powers that be in particle physics and astronomy, as well as the funding agencies. That was very important because, when it was small and under the radar, it was not always on a stable path for funding for research. It’s now become a mature field. It's been interesting to see that evolution over the course of my professional career.
Josh, besides being sort of an institution in and of itself, what, what are the values of the Decadal Surveys? What is it, how does it allow, besides just sort of a major self-assessment, how is it good for the field itself? In what ways, you know, looking back, you're not involved in the current one, but what's advanced or has been more fruitful as a result of that 2010 Decadal Survey?
The Decadal Surveys are best known for prioritizing large- and medium-scale projects as recommendations to the funding agencies: NASA, NSF, and DOE. Projects are typically divided into classes by size and by whether they are ground- or space-based, and then the top few projects in each class are prioritized. So, for example, in 2010 we said the number one priority for large scale projects on the ground is LSST, which is now the Vera Rubin Observatory. For space, we said the number one large-project priority for NASA over the next decade should be the WFIRST mission, now known as the Nancy Grace Roman Telescope.
These recommendations are extremely important for the community because the process of arriving at them is really meant to end up reflecting a community consensus. We take input from the broad community in formulating these priorities, and then the funding agencies, in principle, use this prioritization to make their spending plans for the next. So, for example, LSST was approved to start construction pretty soon after the 2010 Decadal report came out. And I think it was really that push from the Decadal Survey that got the National Science Foundation and the Department of Energy to come together and decide to build it. It’s an $800 million project, so it's a big thing.
The Decadal Surveys are also important because they identify certain key questions or areas of research that they think are ripe for advancement over the next decade. And that also affects the kinds of research that the funding agencies may decide to support, because it's in an area that's been strategically identified as important for the field by the Decadal. The Decadal is central for NASA, for the space-based missions, and for the National Science Foundation, for ground-based astronomy. It is also useful input to the Department of Energy, but the DOE in addition has the P5 process, which is seen as more central to developing its priorities in high-energy physics. P5 is focused on particle physics, including cosmology, whereas the Decadal is really focused on astronomy, including cosmology.
Right. Right. What's the current state of play with the DES-SPT crew?
We have an active collaboration between the Dark Energy Survey and the South Pole Telescope. In fact, that's how DES started: the South Pole Telescope was going to be built, they were going to do a big CMB survey of the Southern Hemisphere to study clusters of galaxies with the Sunyaev-Zel’dovich effect. And there was recognition that you needed a large optical survey to measure approximate redshifts of these SPT clusters in order to use them to do cosmology. So that was the original mission of DES. The DES science program quickly expanded beyond that to include galaxy large-scale structure, weak lensing, and supernovae. But that collaboration with SPT, and now also with another CMB experiment called ACT, is important: having optical data from DES combined with microwave background data from SPT and also ACT will give us stronger and more robust constraints on dark energy. Those are both active collaborations, and a number of joint papers have come out, but the major results are still to come.
Josh, to get back to your, your work as a mentor to graduate students, it's always a great opportunity to see where the field is headed based on some of the most promising work that graduate students are doing. So over the course of your career advising, you know, these, these, this wonderful, impressive list of graduate students that you've had, what are some of these sort of big takeaways in terms of where the field is headed for, you know, people who are just starting in their careers and are thinking about, you know, where, where might things be heading in the next, you know, not just the next 5, 10 years, but, you know, the next 10 or 20 years?
That's another good question. I wish I knew, because then I would know which problems to give to my graduate students. My work has evolved from mostly theoretical to mostly observational work involving analysis of ever larger data sets. Thinking ahead over the next decade, all four major next generation dark energy projects that I mentioned, they're all going to produce data sets that will very rapidly dwarf what we have already. It's like when people do these comparisons of how much information you have on your iPhone versus what was in all the libraries of the world up through some date.
It's going to be the same kind of thing. Whereas we feel now that we're drowning in an ocean of data, we're actually just in this little, tiny bathtub of data. And we're about to jump into the ocean. I think things qualitatively change when you make those large, order of magnitude transitions in the amount of data you have. For example, everything needs to be controlled much more precisely. Working with simulations of the universe becomes much more important. To analyze your data, to figure out how precisely you've measured something, you need to have a whole suite of simulations of the evolution of structure in the universe that you compare to your data set.
So, our students now need to be broader in some ways, because they need to be able to understand how to analyze data, but they also need to understand the physics of simulations and what are their pitfalls or shortcomings. So that's one aspect. The other aspect is just because the computational challenges keep growing, the sophistication that we bring to the data itself has gotten much greater. So, I think, to finish as a PhD in astronomy, or at least in survey cosmology, you're almost also getting a PhD in statistical methods.
Because by the time you’ve finished, you've had to master this whole panoply of tools, simply because the data have gotten much larger, more complex. And you need to optimize every method you have of analyzing the data. That’s critical, because, since the Vera Rubin Observatory costs $800 million, if you find some way of eking out 5% smaller error bar in your measurement using that data set, you've saved the equivalent of millions of dollars. So these things become actually quite important.
Also, I think there's going to be an explosion in using artificial intelligence and machine learning tools to analyze cosmic surveys, we're really just at the tip of that iceberg in applying those methods to analysis of data. I think those kinds of new numerical methods and computational approaches are really going to be more and more of what's happening, at least in this area of cosmology that deals with large data sets. I think it's great, and it's exciting.
But I just want to make sure that we train students in a way that they can still see the big picture as well, because these analyses get so arcane and abstruse and detailed that one can lose the forest for the trees. On the Dark Energy Survey, I think we have over 100 graduate students working on different aspects of these analyses. And they become great experts at doing this one kind of analysis, but I want to make sure that they understand all the other things that go into this as well. So I think that's going to be a challenge in training of graduate students—having them learn these important, increasingly complex tools, but having them still be able to maintain a broader picture. I was reminded of this because I just saw this documentary called “The Social Dilemma” on Netflix.
And one thing that was interesting about it was that there were people working for these companies in Silicon Valley who had been, say, optimizing one particular algorithm, but they didn't realize they were part of this whole empire of algorithms that was potentially going to take over the world; at least, that was my perception of it. I don't want that to happen to our graduate students. While developing that expertise and going deep, they should develop that broader vision as well.
So, I think it's an opportunity, but it's also a challenge going forward. I have a colleague at Fermilab, Brian Nord, an expert in machine learning, who wants to develop a next generation of telescopes and surveys that will be designed and analyzed completely with machine learning. Does that mean that at some point we'll just take humans out of the whole process? I think that will be more benign than what happens now in many ways with Facebook and Google, where things are happening that perhaps humans didn't actually intend to happen. But I want to make sure we keep enough engagement with science and human intention with it, because I think that human element is important.
Well, Josh, now that we've worked our way sort of right up to the present in terms of the narrative, I want to ask you for the last part of our talk, sort of one broadly retrospective question, and then a forward looking question. And I want to come back to this idea about, you know, how your research agenda has shifted to some degree, you know, away from the theory and more towards the analysis of the data. And, of course, that's very intriguing, because it says, it says much bigger things, you know, beyond your own career, and it speaks really to the field.
And so how much is the-- how much can we move into, can we read into your move away from the theoretical work, at least on the day to day, just in terms of, you know, there's just more fundamental work that can be done in terms of analyzing the data versus ongoing work in the theory? I mean, is the data analysis, is it simply a matter of, you know, bigger and more interesting science is going to come out of understanding this data better, versus a self-conscious decision on your part, you know, to stay really engaged with the theory as you were earlier in your career?
That's a good question. My increasing foray into cosmic survey data really started when I got involved with the Sloan Digital Sky Survey, after it was already in construction. But with the Dark Energy Survey, I was there at the birth of the project and helped push it forward. But in all cases I've tried to follow the path from my theoretical questions into building surveys and analyzing data to understand those theoretical questions. That's where I'm still coming from, those are the questions I still want to answer. When giving talks, for some audiences I will describe myself as a lapsed theorist who has fallen off the wagon into the vice of observational cosmology.
With other audiences, I will describe myself as a recovering theorist who has seen the light of observational data. Even though that's tongue in cheek, I do feel elements of both of those professional identities. And I try to keep enough contact with theoretical developments so that I don't feel completely divorced from that. I think that's important. I think it's possible to do that in cosmology, to have a foot in both camps, compared to other areas of physics, even though I've really done very little on the theoretical side in a number of years.
I contrast that with experimental particle physics, where for the most part you're either a theorist or an experimenter. Yes, the high-energy experimenters talk to the theorists and vice versa. But there's much more of a dividing line between experimental activity and theoretical activity than there is in cosmology. I think a number of cosmologists have managed to maintain a more porous divide between theory and observation, and people find it easier to flow back and forth across that divide than in particle physics. So I think one of the things that's remained attractive to me is that in cosmology it's somewhat easier to straddle that divide.
And I think that's healthy. I mean, if you would have asked Enrico Fermi, “Are you a theorist or an experimentalist?” he probably would have looked befuddled.
He would have said, I'm a physicist.
Whereas now, it's just an accepted thing that you're divided into one or the other.
Not only accepted, but it's baked into the whole way that graduate physics is built.
Yeah, it's used in graduate school. When universities or labs do faculty searches, they generally specify whether it’s for a theorist or an experimentalist. So it's become a self-perpetuating thing.
For certain fields, it's a necessity: just the nature of the work is such that it's almost impossible to do both.
Whereas in cosmology, at least up until this point, there are a number of people who are able to straddle it, the field is such that you can still do both, and I think that's healthy. Going forward, I would like to do more of that with my graduate students, try to expose them to both sides.
Some students will just be more naturally inclined to do theoretical work, others more experimental work. I'm not someone who's in there building things in a lab, anyway. So, for me, the divide is just doing theoretical work versus analyzing data.
So, to finally answer your question, I think my interests have just evolved as the field has evolved. It was not done with a lot of intention or strategic planning. I was just following what I thought would be interesting.
Well, it's worked well. It's worked well for you, that's for sure.
Well it’s given me satisfaction.
Well, Josh, for my last question, I want to look ahead at Fermilab and I want to ask specifically about the future of both theoretical cosmology and observational cosmology within the broader context of where the lab is headed. So in high-energy physics, right, when you look at sort of historical moments at places like SLAC and Fermilab and Brookhaven, there are always these existential moments where a high-energy physics project or collider sort of literally, you know, it was, figuratively it hits a wall, but then it throws into doubt, like, the whole future of the lab, you know, like to think about the whole story about what does Fermilab look like post-Tevatron.
And so, my question—to what extent does cosmology, is cosmology sort of future proofing itself, where there isn't, you know, sort of diminishing returns in terms of, well, CERN found the Higgs and what are we going to do next, and we're never going to have an SSC in our generation? To what extent are the, the projects that you're involved with at Fermilab, are they sort of immune from both those scientific limitations and also budgetary limitations because of the extreme costs that are associated with, you know, next level particle physics?
That's a really good question. The major next-generation thrust for Fermilab in experimental physics is the long-baseline neutrino program project called LBNF-DUNE. They're going to build a huge underground detector in South Dakota, we're going to shoot neutrinos at it from Fermilab. And it has very specific goals of understanding fundamental properties of neutrinos. It's a multi-billion dollar project, the next big thing for particle physics in the United States in terms of scale. We're not doing the SSC, and the Linear Collider, if it happens, isn't going to be in the US, so the US community and funding agencies have decided that's the next big thing, and it has broad support. The other national labs, SLAC, Brookhaven, LBL, Argonne, plus a few others, they do high-energy physics, but unlike Fermilab they are not primarily high-energy physics labs.
Fermilab is the last high-energy physics lab left that does only high-energy physics, and it is half of the Department of Energy's high-energy physics budget. So it occupies a unique place because of that. And while LBNF/DUNE is the core of its future program, having just a single activity is not viable intellectually. You need to have a multiplicity of activities, a healthy mix of small and large-scale things going on. Fermilab is the primary particle physics lab of the United States, and as we discussed before the cosmic frontier is recognized as an integral element of the US high-energy physics program. So, I think cosmology should and will be part of Fermilab’s future. It complements the neutrino experiments, the muon experiments, and the LHC experiments. There are people at the labs who migrate between these activities, spreading technical and human synergies between them. So, I think having cosmology as part of that intellectual mix really adds to the overall intellectual ferment of the institution.
Other labs, particularly SLAC and LBL, now have less of their mission focused on high-energy physics. As a result, they have actually increased their efforts in cosmology. They're not going to be the host lab for the next big multi-billion-dollar high-energy physics experiment. So, they've developed cosmology efforts even more than we have at Fermilab, which I think is appropriate. SLAC is the lead lab for the DOE efforts on the Vera Rubin Observatory, LBL is the lead lab for the DESI project as well as for the next generation Cosmic Microwave Background experiment, CMB-S4.
So, I think what's happened is actually quite healthy. We have a number of labs that really have significant cosmic frontier programs that I think complement each other quite well. With regard to future proofing, the cost scale of these cosmic projects is growing. The Sloan Digital Sky Survey at the end cost about $80 million. The Dark Energy Survey cost about 40 million to build, the DESI project is probably 80 million. As I mentioned, the Vera Rubin Observatory is of order 800 million. They're getting to the scale where it's not just a little side thing you can do while you're doing your multi-billion-dollar project. The next generation Cosmic Microwave Background experiment is projected to be over 600 million.
So, these projects are now getting to the point where they are competing for resources with other parts of the field. The Snowmass community process is going on now, that'll lead into a next P5 process for high-energy physics. That's why we need things like P5 and the Decadal. Because all of these things are growing in scale and becoming more expensive. Trying to figure out how to fit them all in is becoming more and more of a challenge. I don't know if I really answered your question.
Absolutely, absolutely. It's obvious that there's lots of exciting things to look forward to in the future, but that nothing is a foregone conclusion in terms of the long term prospects, that these things are going to continue to be supported, you know, by society at large. So that's an ongoing, it's, it's, it's, it's incumbent on everybody in the field to continue to emphasize that this is still important science worth supporting, for sure.
You asked earlier about where dark energy is going, and I laid out these projects over the next decade that collectively will be several billion dollars. And one question people are going to be asking now is, okay, what's next after that.
After we've done Rubin, Roman, Euclid, DESI. Do we keep going? Is there going to be a stage five dark energy experiment? Is it going to cost billions of dollars? Do we know enough whether we should even do that? At what point do you decide, this science question is just becoming too expensive to answer?
I think that's an interesting question that the field now has to face because cosmology has become big science.
Right, right. Josh, which is a great narrative loop to, you know, getting back all the way to when you were an undergraduate where it was the perfect, not only was it not big science, you could barely find someone to, to teach it to you.
That's right, it's evolved considerably. And so it does raise the question you raised about high-energy physics: when things get to the multi-billion dollar scale, how does the field keep going? And the field starts to reach these existential questions.
I think cosmology is now entering the era when those same questions are coming up, so it will be interesting to see what happens.
Stay tuned. Well, Josh, it's been an absolute delight speaking with you today. It's, you really shared a wealth of insight from so many areas in science and, and policy. Thank you so much for spending this time with me. I really appreciate it.
Thank you. It was a lot of fun, great questions. And, you know, I really appreciate it.