skip to main content
Home  /  Interviews  /  David Hitlin

David Hitlin

David Hitlin

Professor of Physics

By David Zierler, Director of the Caltech Heritage Project

July 8, 2022

DAVID ZIERLER: This is David Zierler, Director of the Caltech Heritage Project. It is Friday, July 8th, 2022. I am very happy to be back once again with Professor David G. Hitlin. Dave, it's great to see you. Thank you for joining me again.

DAVID HITLIN: You're welcome. Last time was fun, and let's see what we can do this time.

ZIERLER: What I want to do, just as a guide to researchers in order to orient them as to the nature of our discussion today, let me just state for the record that the oral history we're doing now should be considered really as an addendum to our earlier conversations from May 2021, when I directed the Oral History Program at the American Institute of Physics. Those discussions, of course, constituted a full coverage of your career and life, but really, they did not adequately cover certain topics that were specifically germane to your work at Caltech and more generally Caltech history. Of course, as it happens, I became Director of the Caltech Heritage Project, and we had the idea to reconnect for a discussion on several issues of specific relevance to Caltech history. That's the aim for today's discussion. Dave, before we get to those topics, just a few questions to update your work on where it was since our last conversation. As we discussed then, the muon g-2 experiment at Fermilab was causing quite a stir. What has transpired since?

HITLIN: On an experimental level, basically nothing has changed, but the discrepancy between that measurement and the Standard Model calculation of the value of g-2 has been a subject of very detailed worldwide scrutiny over many years, and that has only intensified. People are clearing up minor discrepancies among different approaches to various components of the g-2 calculation. It's very complex. Some aspects are extremely well-understood, and there are a few that have some theoretical uncertainties as to what treatment is appropriate to derive a number from the calculations. Many people have been looking at this for years and years; this new experimental result has stimulated them to revisit the calculations and make improvements. There are little nips and tucks around the edges, but nothing major has changed at this point.

ZIERLER: Of course, it would be very preliminary but the reason why this was headline news in 2021 was it was potentially suggestive of new physics beyond the Standard Model. In the year-plus since then, are we in a position to be more or less optimistic about whether this is possibly new physics?

HITLIN: I don't know that we can draw any conclusion other than what we had before. There is a statistically significant discrepancy. The first thing of course is to reexamine the experiment. The second is to reexamine the theoretical calculations. The experiment looks like it was very thorough, and nobody has turned up any issues with it. They are going to be taking more data; they will be able to improve the experimental errors by a significant amount, more than a factor of two, over the next several years, with more running. They're not yet at the level of being limited by systematic errors, although they will be starting to come close. g-2 one of several small discrepancies that exist with Standard Model predictions. It's probably the one with the solidest theoretical prediction, the most scrutiny and the largest statistical significance, so it's very important, but I don't think it's possible to predict whether it is the crack in the Standard Model that you'd like to see.

As I said, there are other discrepancies floating around as well, and the nature of these things is that sometimes with improved measurements the discrepancy gets smaller or goes away; sometimes, it stands. None of these things have yet stood the test of time, but there are a few still unresolved results out there, and we'll see. The classical way you would like to find a phenomenon beyond the Standard Model is to find a new particle, perhaps a supersymmetric particle at the LHC. That would be unambiguous, and everybody would say, "Okay, that's the answer." But nobody has found that, and the limits they have placed on SUSY at the LHC are becoming very constraining. It was really quite interesting—this last week, they had a celebration, a tenth anniversary of the discovery of the Higgs back in the summer of 2012. That was really an interesting event, and of course it also coincides with the start of the new long run of the LHC after a long stoppage. People were also talking about what are they going to find at the LHC? What are their goals? What are they looking for? Nobody was talking about supersymmetry. It has largely vanished from the map, although it was touted for decades to be the main objective of the LHC. I found that interesting.

ZIERLER: The other major project of current import to you when we last talked of course was the LDMX experiment at SLAC relating to dark photon and dark matter searches. Has there been progress on securing approval for that project?

HITLIN: I don't know. We just last week had a review of our progress from DOE together with a total of six others — they have something called the Dark Matter New Initiative, which is evaluating new proposals for experiments. There were six of them in various stages. They reviewed them all together in three days, which is not a recipe for going deeply into anything. But they had a good crew of reviewers, and we did a good job, I think, and got very good marks. That has not yet turned into DOE writing a check, so we are pressing them to try to understand on what timescale this could turn into a funded project. They don't know, and we don't know. We were hoping to get a start in fiscal 2023 which starts this October. They didn't say "no" to that, but it certainly doesn't seem like it. The hope is that we'll get funding to carry on with our development work. We've done a lot of work; we had a couple of beam tests at CERN. I've been to CERN four times in the last six months, actually, for these tests, which went quite well. That and preliminary engineering work that we're doing needs funding. They have been giving us funding on a scale to allow us to move forward, and so the hope is if they don't fund the project in the coming fiscal year, they'll fund the development work that we need, to be ready. It looks encouraging, but it is not a done deal.

ZIERLER: Are you increasingly optimistic that this is one of the most promising avenues for dark matter research?

HITLIN: Yes, absolutely. This is a bit different from the bulk of the experiments in which they've looked for the interaction of dark matter with sensitive materials like germanium or liquid xenon, looking for very small signals due to a collision of a dark matter particle with the detector. That has been the dominant approach that has produced most of the limits on dark matter of a certain mass and a certain coupling strength. The approach of LDMX and a couple of other experiments now is to attempt to produce the dark matter in an accelerator. As the limits on the masses of heavy dark matter—what used to be called WIMPs, which stood for weakly interacting massive particles—seems to be getting quite constraining, so that WIMPs may not be a viable model of dark matter anymore, you therefore want to explore lower mass regimes. That covers a very wide range down to millielectron volts up to perhaps as high as a GeV. The upper regime from say an MeV to a GeV looks like it's probably best explored at accelerators. You produce the dark matter in a collision and look for a signature. That's what LDMX is about. Our calculations show that the experiment has higher sensitivity than its competitors over most of the range, so it appears to be something that you want to do. Everybody agrees with that; it's only a question of "Where is the money?" We are optimistic that it will happen.

ZIERLER: It's amazing that between these two projects, you're really operating at two frontiers in physics: what is dark matter, and what might physics beyond the Standard Model look like. My question is, where if at all is the complementarity in these projects? In other words, will understanding dark matter yield insight into new physics beyond the Standard Model? Conversely, will a discovery of new physics yield better understanding as to what dark matter might be?

HITLIN: That's an interesting and rather subtle question, actually. Dark matter is not part of the Standard Model, so if you found it, you would have certainly found something beyond the Standard Model. On the other hand, dark matter is very unlikely to be, for example, the cause of the g-2 discrepancy. That would be something entirely different. Or the answer could come from the main thing that we're working on, Mu2e at Fermilab. That is again a search for physics beyond the Standard Model, but a very different approach that would have nothing whatsoever to do with dark matter.

ZIERLER: As you mentioned, of course at LHC there's a lot of excitement this week with the third run. For you and your research, what are you paying most attention to right now? What's the most tantalizing prospect as LHC is operating at ever-higher energies now?

HITLIN: They've been running for years and years, and so the nature of the game is that whatever they've done before, they will do again with a slightly higher center of mass energy and a lot more data. They've published hundreds and hundreds of papers. Those results will of course get better with improved data. Their sensitivity to potential physics beyond the Standard Model will get better as well. But more or less, they're going to do the same sort of analyses, but better. That is of course interesting. It's not a criticism; it's just the nature of the game.

ZIERLER: Let's be super optimistic and pie in the sky and say that the energies now might yield supersymmetry. To go back to my previous question, is it possible that that's a linchpin linking new physics in the Standard Model and dark matter? Is that something that could put it all together?

HITLIN: Absolutely. Yes. If there were evidence of supersymmetry, then one of the prime candidates for dark matter is what is called the LSP, the lightest supersymmetric particle. It would be the lowest mass in the hierarchy of many different new supersymmetric particles, and it would have nowhere to decay to, so it would be a stable object and it could constitute the dark matter. That linkage has been there for decades, and it was a real motivation for thinking about supersymmetry. But you've got to find some evidence of the other supersymmetric particles before you ascribe the dark matter to the lowest-mass one.

ZIERLER: Let's now go back in history and talk a little bit about the origins of your career at Caltech. Tell me about that first visit to Pasadena. What year was it? What were the circumstances? What were your first impressions?

HITLIN: It was 1979. I was looking for a job. I had a job offer, a tenure offer, from the University of Washington. They were collaborators on the Mark III, the new experiment I had started about a year before at SLAC. I was the founding spokesman for Mark III. They saw an opportunity, if you like, for the University of Washington to play an important role in the experiment, so they offered me this job, a good job. I came very close to taking it, and then I was contacted by people from Caltech. What had happened there was that Alvin Tollestrup, who was a major figure at Caltech and a very important person in the construction of the synchrotron—the machine that they had on campus here—left Caltech to go to Fermilab to work on the Tevatron, the new superconducting collider at Fermilab. He played an enormous role in the construction of the Tevatron. So, there was a job opening at Caltech. I don't know any details of who they were interviewing or looking at, but they asked me to come down and give a talk and look around.

ZIERLER: Do you remember what your talk was on?

HITLIN: It would have been some charm meson production and decay results from the Mark II, which was just starting to get results, with a coda on Mark III, which at that time was just being designed, but not yet constructed. I didn't really know anybody at Caltech very well. I knew Frank Sciulli, who had some years earlier done an experiment on form factors in semileptonic kaon decay, as I had done. That was in 1973, several years before, but we had had some contact about that. The experiment I wound up doing was a substantial advance in that area and resolved a big problem with agreement with current algebra predictions. We had had a little contact about that. Frank was the only person I knew at all.

I went through the usual one-on-one discussions with many on the faculty, which seemed to go pretty well. That evening there was a dinner. It's usual to take the take the speaker out for dinner. I went to the restaurant with Bob Walker, the PI of the group, and the other people were to meet us there. As we were waiting in the restaurant, and I asked Bob, "Who's coming?" He said, "Barry Barish. Frank Sciulli, Jerry Pine. Ricardo Gomez is coming; I hope he won't get into a fight."

ZIERLER: What would that have been about?

HITLIN: That's a quotation. I was a little surprised about that. One of the other faculty members was Ricardo Gomez, who was a very interesting person. I won't go into enormous detail. The reason Bob had said that was, a week or so before they had had another dinner —I don't know if it was the same restaurant or another one—but Ricardo Gomez had been there, and he had gotten into a fight. The problem was that Ricardo drank and got into all kinds of scrapes. He was thrown out of his Athenaeum membership. There were stories of him being thrown off airplanes. At any rate, he was an unusual guy. For some years, he had get-togethers on Sunday afternoon at his house which were a lot of fun. He had a big lawn, and he had a croquet set; so we would play croquet. Then we would play cards later in the evening. He had a favorite card game called Buenas y Malas, which was sort of like Gin, that we would play late into the evening.

After the dinner at the Mexican restaurant. Frank Sciulli had a small get-together at his condo. I remember that very well, because there was wine and beer, and there was a bottle of scotch on the table; just one bottle of scotch, not a full bar. I don't remember seeing anybody drinking the scotch, but by the end of the evening, there was almost nothing left in the bottle. At the end of the party, Ricardo said, "Can I drive you back to the Athenaeum?" I thought I knew where the scotch had gone, but nonetheless, I accepted. He drove back to the Athenaeum, perfectly well, actually, and it was all fine. They offered me a job soon after, as associate professor without tenure. Had it been with tenure, it would have been a no-brainer. Without tenure, that was taking a big risk, but I decided that it looked to me like Caltech was my kind of place, so I decided to take a chance.

ZIERLER: Would it have been on an accelerated tenure clock, though, at that stage in your career?

HITLIN: No, I didn't get tenure for six years. The idea was to be able to demonstrate some real accomplishments from Mark III, which I was able to do on that time scale.

ZIERLER: Did it feel still like the department of Gell-Mann and Feynman? Obviously, they were past their prime at that point, but how large did they loom in the department and at Caltech when you joined?

HITLIN: Oh, they were still around and still engaged and major figures. I agree with you that their prime accomplishments were behind them, with the exception, I think, that Feynman basically invented quantum computing in the early 1980s. No, they were both around. Murray was starting to pull away, in the sense of he was getting very interested in complexity and things of that sort, things that went beyond conventional field theory. He started to get involved with the Santa Fe Institute and then moved there for the rest of his career. Feynman went to all the seminars. He was very active. He did the space shuttle disaster committee, an important contribution. He died in 1987, if I remember correctly.

ZIERLER: Did you interact with them? Did they show interest in your work?

HITLIN: Feynman, a bit. If there was a seminar about things we were doing, he would actively participate. Only on very rare occasions did I ask him a question about something. There were new young people around. There was Mark Wise, David Politzer and Stephen Wolfram. They were more the kind of people who I interacted with. Wolfram, who was actually a graduate student when I started, was an interesting person. He never wrote a thesis, but he had written a bunch of papers, so they stapled them together and gave him a PhD. He then started working on SMP (Symbolic Manipulation Program) which eventually evolved into Mathematica, a way of doing symbolic mathematics.

There then was a controversy, which I only saw peripherally. Stephen needed a bigger computer. The shared HEP computer was a Digital Equipment VAX 11-780. Stephen had exhausted its capabilities, so several of us went to Murph Goldberger, who was the president at the time, and got Caltech funds to buy him an identical computer for his sole use. He got a group of people together and did a lot of development work on SMP. Then, he wanted to commercialize it, and that's where the controversy developed, because he had done this work on a Caltech computer, and so there was an Institute interest in patents, copyrights, or whatever it was—which he didn't take kindly to. Eventually he left Caltech and went to the Institute for Advanced Study before founding his company. It was a controversy that went on for quite some time. I don't know in detail how it resolved, but Mathematica did become a very successful product.

ZIERLER: At this stage of your career, was most of your research agenda centered at Fermilab? Was that the place for you to be?

HITLIN: No, I was working at SLAC.

ZIERLER: When did Fermilab enter the picture? When you were thinking about Mark III versus neutrinos?

HITLIN: That's somewhat different. When I came to Caltech, the obvious thing for me to have done was to continue with the Mark III. But Caltech had a major involvement at Fermilab. Jerry Pine and Ricardo Gomez were working on an experiment at Fermilab. Bob Walker, and Tollestrup, before he left, and Geoffrey Fox, who was really a phenomenologist, were working on another experiment at Fermilab. The big Caltech experiment at Fermilab was an important neutrino experiment by Barry Barish and Frank Sciulli. When I arrived, there was some interest in getting me involved in the neutrino experiment. I went out to Fermilab and talked to people and looked around, and thought about it pretty hard, but I decided it would be better for me to stay with the Mark III, to get it off the ground, get it built, and do that physics.

That was okay with the group; it was not a problem, but you have to get support for it. I initially had no explicit funding from DOE. There was an overall grant, of course, but how things were apportioned was still to be determined. I was actually pretty naïve about how funding worked. Having been at SLAC and Stanford and worked in other people's groups, there was always funding. You just did what you did, and the money was there. Now, it was going to be my group, and so I had to figure out how to do that. One of the things one does in those instances is to negotiate a startup package with the Division Chairman. I never did that; I never thought about it. Nobody ever said anything. I just walked in and started working. Bob Walker was the PI of the DOE grant for the first few years. Then Bob retired and Barish became PI for quite some time until he went to LIGO. Then I became PI for about fifteen years. At the start I was given some grant funds, so I started hiring postdocs and taking on students. Many years later—I was working on SLD, actually—I remember getting a phone call from the Division Chair, Robbie Vogt, saying, "We never gave you a startup package. There is some money available, would you like some?" He provided funds that allowed me to hire people to work on developing a new kind of calorimeter for SLD. So, I got my startup package eventually, in this rather unorthodox manner.

The first postdoc I hired was Rosemary Baltrusaitis. The second was J.J. Russell. My first students were Jay Hauser, Jeff Richman and Dan Coffman. I've had lots of students since, but those were the first. They all went on in HEP and did really well. Jay is a professor at UCLA, and Jeff is at Santa Barbara. People did well. I have tracked the trajectories of the postdocs and students after they left Caltech, not only in my group but for all of the high-energy physics postdocs and students, including the theorists, to the extent you can—where they went, are they still in the field, et cetera. The list is quite impressive.

ZIERLER: Given that you got started with your research before the startup package, what were you able to do as a result, between the DOE grant, the package, the Presidential Young Investigator Award? What did all of that allow you to do?

HITLIN: You need salary support, travel funds, and the wherewithal to do R&D. A couple of months' salary is typically charged to the grant, if you have a grant. Caltech is unusual in that respect, in that the salary you get is a yearly salary, and if you don't have research support, you still get a yearly salary. In most universities, you get a nine-month salary, and then if you have a grant, you can supplement it for the summer, which covers typically one or two months of support. If you don't have a grant, you don't get paid during the summer. Caltech is different in that respect, in that you get paid year-round, and then you reimburse the Institute from your grant.

Then you have to assemble a group of postdocs and students to get things done. Supporting the postdocs and students requires financial support. I had enough initially from the grant, a small amount, and this Presidential Young Investigator Award. That's actually an important program; it has now got a different name, and it has been expanded, but it is a program that had started a few years before I got one. It's a highly competitive process, intended to give people the wherewithal to get started in a new place and it's very competitive. I've been on the committee that makes the choice of these things several times. Typically, if you look at who got the awards and who actually went on to make an impact in the field, it's very impressive. Whether it's because the committee that makes these selections was very wise and made good choices, or whether it's a self-fulfilling prophecy because those people then had resources, I don't know, but the results are quite impressive. It's been a very successful program.

ZIERLER: Tell me about the kinds of students who were attracted to your group. What were you looking for, and what were they interested in at that time?

HITLIN: I don't think there's any particular pattern there. We have students that are very interested in doing the physics. There are people who are more interested in the instrumentation. There are people who are interested in the details of analysis. These days that involves machine learning, or other neural network-based analyses in which there's a lot of real computer science. What students wind up doing after they get their degree depends very much on what their interests were. Many of them go on in physics and make a career in academic physics. Some do physics in industry. It's very rare that they can't find a job in physics if they want one. Some fraction of them wants to do something else, either something with their physics training directly but not high-energy physics, or something in computer science, because they become really quite expert in that. Over the last few years, my last three or four students have actually gone off to big data, which I found a bit disappointing, but I can understand it, because they have the skill and they can make three times the amount of money they make as a postdoc, so they do that. On occasion they go into finance, using another set of computer skills they have. Quite a number of my students have actually gone on to be quite successful in the HEP field.

ZIERLER: As you were building up your research group, what were the teaching expectations as an associate untenured professor?

HITLIN: We all have similar teaching responsibilities. Most people teach two quarters out of three, or three quarters out of three. Two, I think, is more typical. The teaching load is not enormously burdensome. Teaching is fun because the students are incredible, and it's always a challenge. Over the years I taught mainly quantum mechanics and the undergraduate and graduate particle physics courses, basically phenomenology. When I got really busy with BABAR in the early 1990s, I actually had a two-year leave. Caltech doesn't have a sabbatical program, a time off every seven years kind of thing, but in a less formal way, it basically allows you to do the same thing. I had never done that since 1979. I had never availed myself of a leave, so when BABAR was getting going, I asked for two years of leave, which was granted. I was given a formal appointment at SLAC, as I had project responsibilities that required signature authority. I lived here, going to SLAC four or five days a week.

Then when the leave was over, I taught a new course. In a curriculum revision, a new required course, was mandated in all the divisions. Actually, there were two. One was called Oral Communication and the other was called Written Communication. Four or five years later they were combined into one. Steve Frautschi, who was the executive officer at the time, asked me if I would develop the Oral Communications, and I agreed, with a proviso. The starting assumption was these courses were like freshman composition and its oral analog, and I didn't think that was particularly interesting. I said I would be willing to do it if it were a senior course. I was doing the oral part only; I don't remember who was doing the written part. The idea was it would be a kind of a senior seminar involving contemporary research. We would discuss contemporary physics but would pay particular attention to the organization of presentations, how to give effective presentations of different audiences. I have been doing that ever since, more than 25 years. I typically do it two quarters a year. Because the students are making these presentations, it has to have a limited enrollment so they all can take their turn. Typically, the most you can fit in for the things we do is nine students. There was a crunch last year because of the pandemic, and things were a bit confused, and so I gave the course all three terms instead of just two. Typically, I've done two, and this next year, I'm going to do two.

The students make three presentations. One is a regular technical seminar. Seminars around the Institute are typically an hour. Seminars given at a conference and a plenary session are typically a half hour. In order to fit things in, we define it as a half hour. You have to make some assumptions; that is, who is your audience.? The assumption I ask them to make is that your audience consists of people who—for example, if your topic is neutrino physics, then your audience is the high-energy physics group at Caltech, not only the people who do neutrino physics; all of the people who do high-energy physics. You can't assume you're talking directly to a complete expert, but you're talking to high-energy physicists who understand what you're talking about, and you have to tailor the presentation accordingly. Of course, the audience isn't that, so you have to compromise a bit and provide some introductory material, so the class knows where you're going. The next talk they give is what you would do in a parallel session at, say, an American Physical Society meeting, where there are typically 15-minute talks, 12 minutes of presentation and three minutes of questions. There, the audience is much more expert. You're broken up into small groups, so if you're talking about neutrino physics, everybody in the audience can be assumed to be a neutrino physicist; you don't need any introductory material. You don't have to put anything in context. You just talk about the latest, hottest thing you did, directly. That's all you have time for in any case. Then the third talk, another half-hour presentation, is aimed at a popular audience; that is, the kind of talks that are given in Beckman Auditorium in the evening, those sorts of lectures, where the audience is a group of interesting, intelligent non-scientists. That's the one they generally find the hardest to understand how to do. They're used to talking to physicists and their friends who are scientists, and this is a different game, so that's the most difficult one.

When the oral and written coursed were merged, I added on a requirement that they write a paper, at a Scientific American or Discovery magazine level, to a popular audience. Rather than have them just write a paper, hand it in, and get a grade, we make it a process wherein they write a draft, we circulate the draft to two other students, a little round -robin. I provide some guidance of topics that they should address, to make comments on clarity, organization, et cetera. The TA and I do that, as well. So, they get real feedback on their draft. That's the way the real-world works, so I wanted to give them a taste of how that actually goes. First, they make an outline, then we make sure the outline makes sense, which it almost always does. Then they make the draft, they receive the comments, and they make a revised draft, again as with real-world paper writing. That's what this course is about. They seem to like it and get something useful out of it. Their topics vary. If they've done a SURF, they can use that topic; they have to treat it differently than what they've done for a SURF, but at least they have material, and they have to treat it in these three different aspects. Or if they are interested in learning something new about a subject, they can review a subject in different aspects, and I help them in finding references and things of that sort to get them going.

Once a student got very interested in violins. He didn't go so far as to make his own violins, but he did, put together a bunch of cheap violin kits with variations, and did various measurements on them, just for fun, and talked about the physics of what determines the sound of the violin. There's just a variety of topics that get covered. I ask them to run the topics by me so that we make sure they don't go off the deep end with something inappropriate, but I give them wide latitude. It works pretty well, I would say. My criterion is that they give three talks, and it's almost always true that the third one is better than the second, and the second is better than the first one. So, they are getting something very specific out of this. I tell them, "There's nothing very complicated about doing this; the way you get better at it is by doing it." That's in fact what you see, time after time.

The other thing I do is an innovation to improve attendance. Caltech students are often known for not bothering to go to class, and we started seeing that problem. My solution was to have the students grade each other's presentations in real time. Initially that was done using a paper form; now I use a Google form. I tell them to bring their laptops to class. Then, in real time, as a presentation ends, they fill out this form, giving grades in several categories on the presentation. I give them a few days, because some of them take a while and don't do it quite in real time. We video the presentations and we upload them to the course webpage. I then ask the students who made the presentation to view their own presentations and grade themselves. I circulate the resulting grades to the presenters, anonymously, with the graders' names removed, to give them feedback on how their presentation was received.

At the end of the term, I produce a graph which shows on one axis the class grade for presentations, everybody together, and the other axis, the self-grade, the grade the presenter gave themselves. If everything were completely in accord, the points would be on a 45-degree line. Of course, they tend to be, but some are above, and some are below. They know what grade they gave themselves, so they can compare and see whether they were more optimistic about how well they did than the class was, or less optimistic. I know the personality of the students, and I can usually guess pretty accurately which ones are going to have a higher or lower opinion of themselves than the class does, so it's kind of amusing. That's what I've been teaching for many, many years.

ZIERLER: Of all of your committee work at Caltech, what stands out in your memory as enjoyable or meaningful to you, or even possibly useful in your research agenda?

HITLIN: There are different kinds of committees. I have been chairman of the Graduate Admissions Committee several times for long stretches. The last was until about three years ago; I had done it for seven or eight years, and I had done it previously for other long stretches. You really get to view and shape the incoming class. It's a lot of work for the committee, because the number of applicants has been going up and up. Last year we had 800 applicants or so. We typically admit—it varies from year to year, depending on what the matriculation rate was the year before, to try to keep a balance, but we typically admit 60, 75, something like that. We get an acceptance rate of 30, 35%, so we typically aim for a class of around 25, of that order. It's a highly selective process and a great deal of work for the committee to evaluate all these applications.

The way it had traditionally been done is each application was read by three people and then the scores are averaged and then sorted. You survey the individual areas, how many condensed matter openings for students are there, theory and experiment separately, how many high-energy physics openings are there for the next year, et cetera. Then you sort these things by rating and then by category and you try to make a match. You have discussions about them and whether somebody happens to know a student well for some reason or knows who the people who wrote their letters were very well, or things like that. It's a bit of a negotiation, but it's an interesting process. They have now decided that it's too much work to have three people do the evaluation, so when I stepped down, they now have two people do it, which I have a little bit of trouble with, because there is a variation between different raters, and some more variety seems to me to be better, but it's a lot of work. We also then have the students come in March, around Easter vacation. All the major schools do that. They make a tour around the West Coast or a tour around the East Coast. We make presentations and we talk to them in some detail. It's an interesting process which takes a lot of time but yields really impressive students.

Other committees I've been on—typically there are these qualifying exams which we give them after the first or second year, on classical physics and quantum mechanics and things of that sort. That is taken pretty seriously. We look at the questions, we revise the questions, we look at the solutions, make sure they're right, and do the problems. We just gave the exam last week, so I have a bunch of them on my desk here. Another kind of committee that we have of course involves promotions and staffing. The interesting thing there is I did a bunch of those early on; I haven't been on one of those committees since the late 1980s. I don't know why, but nobody has asked me since then. it is literally more than 30 years since I have been on one of those committees.

ZIERLER: We talked in some detail in our original conversations about the SLD liquid argon calorimeters, but what might be a specific Caltech angle to building these instruments that we might not have covered?

HITLIN: I don't recall precisely what we covered, but that makes for an interesting story, which involves work we did at Caltech and at SLAC. I think we did talk about the idea of building this calorimeter, which was a major component of SLD, which was a very large detector. I was in charge of building the liquid argon electromagnetic calorimeter for the Mark II while I was at SLAC. When it came to build this new detector for the SLC, which we called SLD, we decided that that would be a good technology to use there as well, and it was something I had some expertise in, so I led the design and construction of that component of the detector, together with other groups. In particular we had the University of Washington—not the same group that worked on Mark III, but another group from Washington - people from the TRIUMF lab in Canada and the University of Victoria, and Columbia. It was a good collaboration. In the end components were built at all four places; it was a rather big construction project—but we did the original development and built the prototypes here at Caltech.

The idea was to use uranium as the radiator. You have a heavy material that causes the particles entering it interact and to make what are called showers. The showers consist of charged particles, largely pions, together with electrons and photons. You then have a sensitive medium, in this case liquid argon, which is interspersed between these multiple plates of the heavy material. You apply an electric field to the plates to collect the electrons liberated in the liquid argon through ionization. That electrical signal is proportional to the energy that's deposited by the incoming particle, hence the name calorimeter. There was an idea that originated at CERN that if you used a fissionable material as the heavy absorber, particularly uranium-238, that it would improve the energy resolution, because of a phenomenon called compensation. It so happened that at that time there were a lot of experiments that wanted to make these types of calorimeters. They wanted one for the DØ experiment at Fermilab. There were two that were to be built at the DESY Laboratory in Germany. Carlo Rubbia, who had discovered the W and Z particles with the UA1 detector at CERN, wanted to build an upgraded version of that experiment using this technique. So there were five requests to get sufficient uranium to do this. Who has uranium? The Department of Energy! In fact, they have lots of depleted uranium; that is, uranium from which the fissionable U-235 has been removed. What's left over is then largely U-238. They had a huge amount of it; they were happy to get rid of it. But it was not in metallic form; it was in some other chemical form—I think it was a sulfate powder—packed in barrels.

When they got all these requests, they formed a committee with one representative from each experiment. I was the SLD representative. We met several times, usually in Germantown, Maryland, at DOE. They did a prioritization. The highest priority was of course Carlo Rubbia who had just done this really remarkable experiment at CERN. I don't remember if we were second or third, but we got what we needed. An amusing thing about the meeting was this little anecdote, actually. I knew Carlo for years, because he had done kaon experiments related to those we did at SLAC. He was and is an interesting guy. At a coffee break, he took me aside and said "The SLD proposal for the calorimeter is plagiarized from the UA1 proposal." I said, "I was the editor of that section, and I know who did what, and it's not plagiarized. I'll bet you a bottle of champagne." He reached into his briefcase, and he pulled out two pages, the front page of each proposal, and they were absolutely identical, word-for-word identical. He knew he had me; he already had this evidence in his briefcase. I said, "Okay, I owe you a bottle of champagne". I was pretty chagrined, so I went back home to try to understand what happened. It didn't take very long to get to the bottom of it. I told you that one of the groups that was working with us was from the TRIUMF Laboratory. The new director of the TRIUMF Lab was named Alan Astbury, from the Rutherford Lab. He had worked on UA1 and the UA1 upgrade with Carlo and had written the calorimeter section of the UA1 upgrade proposal. Then he left Rutherford Lab and UA1, came to TRIUMF to be Director and joined SLD. When we decided to write our proposal and divided up writing assignments, Alan said, "I'll do the introduction." So, he took his UA1 upgrade introduction -- which is a general discussion on the properties of uranium, the properties of argon, things of this sort--and used it verbatim as the SLD introduction; it was indeed identical. That's not plagiarism. The same person who was involved individually on each enterprise used his own work. So, that was fine; it was not plagiarism, but it was funny. Carlo clearly knew all this. The next meeting of this committee at Fermilab a couple months later, and in the intervening period, Carlo won the Nobel Prize, so he didn't come to the meeting. On the way from the airport, I bought a bottle of champagne and some plastic glasses. Since Carlo wasn't there, rather than have the bottle of champagne taken home to him, we just drank the champagne at the committee meeting. He never actually got it.

In the end, we got our uranium for a beam test and built a prototype device. Depleted uranium is a very interesting material in that it's pyrophoric; it catches fire rather easily. Not as a big chunk, but if you have a very fine sliver of uranium, it can catch fire when it gets hot, kind of like magnesium ribbon. It also oxidizes very readily. So, if you want to make an electrical connection to it, you have to use a grinder to remove the oxide and then solder to it right away. The material that comes off the grinder catches fire instantly.

We built several large prototype modules. Because uranium oxidizes so readily, when it's hot rolled into plates often what happens is that tiny slivers of the material are pressed into the surface. You cannot easily see the slivers, but when you put an electric field on the plates so there's a force on any material on the surface, these slivers stand up and make a short circuit. As you're assembling the structure it's very hard to see this, but when you put it together and it's got all these little narrow two-millimeter gaps and you apply a couple of kilovolts, you get short circuits as these slivers stick up. That's really a pain in the neck as you're building the device.

It turns out that the wife of one of my students, Fritz Dejongh, was a large-animal veterinarian. One of the things they do sometimes is work on the legs of horses. They have special tools. One of these is a borescope, a fiberoptic with a lens on the end that you can stick down into the leg and look around. We got one of those so we could stick into the narrow space between planes and look around and find the little slivers that were sticking up. These plates were about one meter by two meters long. You had to reach in quite far to see the slivers. The veterinarians also have a very long forceps that can reach into a horse's leg. We got one of those, and then could reach in to grab the slivers and remove them. So, we developed a debugging process to find a remove the short-circuiting slivers. It took as much as a day or two to clean up a module, but we were able to do it because we had a student who married a horse veterinarian.

ZIERLER: That's an amazing story! [laughs]

HITLIN: We then put these modules into the test beam at SLAC, and we had a lot of excellent data, and we didn't see this compensation effect. I think I mentioned this in the other interview. We did not see it. It wasn't there, in fact. The CERN group that had claimed compensation had made a mistake, which we found. So, when we went to actually build the real device, we used lead and steel instead of uranium, avoiding all the problems that came with uranium, since there was no real performance benefit.

What happened in the other places was rather interesting. They all used uranium. Carlo's UA1 upgrade of course got the uranium first and built their experiment, and it never worked, at all. The reason there was they were not using liquid argon; they were using TMP, a room-temperature liquid, which almost everybody thought was a really bad idea and wasn't going to work, because it's almost impossible to make a liquid of that type pure enough to collect charge. They went ahead and did it anyway, and then it actually didn't work, so that experiment never went back on the beam line. DØ at Fermilab was built using uranium. It used liquid argon, and it worked, but it didn't see compensation either, because there was no compensation. Then there were two experiments at DESY, one of which, HERA, used liquid argon, which did not see compensation. The other one, ZEUS, used plastic scintillator as the sensitive medium, which has a lot of hydrogen in it, and they did see compensation, as expected. So, only one of the experiments actually made real use of the uranium, and SLD just skipped it altogether. We didn't bother because we knew it wasn't going to do anything for us.

DOE, having had this formed this committee and put in a large effort, asked me to organize a conference on the subject of compensated calorimetry. In 1987 we ran a conference here, and had each group come and show what they knew. At that point, we had a very clear theoretical explanation as to what was going on and why one technique would work, and another would not work. We got everybody together, including the people who had made the initial mistake, and cleared up the situation once and for all.

ZIERLER: We talked in some detail about the CITAR collider at Caltech, but we really didn't get into why it was designed and built at Caltech. What's the significance there?

HITLIN: It wasn't built; the idea was born here, but CITAR was never built. We knew that to measure CP violation in B meson decay, we needed an asymmetric collider. The question was, where do you do it? The obvious place for us—this was Frank Porter and I who were working together on all of this—was SLAC. We had long experience at SLAC. There was a tunnel. There was a machine that could be adapted, et cetera. But SLAC was not in the slightest interested.

ZIERLER: Just to orient ourselves, what were the science objectives here?

HITLIN: The science objective was to measure CP violation in B meson decay. There had been measurements of CP violation in K meson decay starting in 1964 by Fitch and Cronin who got the Nobel Prize for that in the 1980s. But trying to interpret what you measure in the
K meson system in terms of the properties of the underlying quarks turns out to be very difficult. It was realized in the early 1980s that if you were to measure the same thing—CP violation in B meson, neutral B meson decay—you would be able to relate the measurements directly to the properties of b quark decay, as opposed to B meson decay, which means you'd really be learning, without theoretical uncertainty, what you really want to know.

This was facilitated by several theoretical advances. One was the model of Kobayashi and Maskawa, which showed that if you wanted to be able to incorporate CP violation into the Standard Model there had to be three quark generations. They got the Nobel Prize for that in 2008. That was one thing. The second was the realization by several people—Bigi and Sanda and a couple of others—that you could in fact make this measurement in a particular type of B meson decay, that is, the neutral B going to a J/psi particle and a K short meson, and actually it would be large enough to be potentially measurable. That was the motivation. It was a very clear physics objective, a very important physics objective, because this had never been measured, and there was the possibility that when you made this measurement—there was a very specific Standard Model prediction, but perhaps when you made the measurement, you would get a different value. It was shown early on, in fact, by Sakharov, that you could potentially explain the baryon asymmetry of the universe, why we have baryons and not anti-baryons, if there were a set of conditions, one of which was CP violation in these decays, which had not been measured. If we measured it, there was a chance that you would see a value other than the Standard Model and that would go some distance toward explaining the baryon asymmetry of the universe. In the end, when we measured this, we got the Standard Model value, so that was not the answer to the baryon asymmetry of the universe. It was, however, a very important component of the completion of the Standard Model, if you like, and that's why Kobayashi and Maskawa got the Nobel Prize. As I think I told you, they invited Jonathan Dorfan and I from BABAR, and our opposite members from Belle to the ceremony, since we provided the experimental support for their theory. This was very nice of them and made for a very interesting week in Stockholm that I documented in the SLAC Symmetry magazine:

That was the motivation. The obvious place for us to think about doing this was SLAC, but they were just not interested. They had bigger fish to fry. Burt Richter, who was by that time the Director, wanted to build a large linear collider, an electron-positron collider; this has still not happened more than thirty years later, but that was SLAC's objective at that time. Since there was no interest, we decided to figure out how to do it ourselves. There were other people interested as well, in particular David Cline of UCLA, who had a different approach. We joined forces to form a small group called the Southern California Consortium for a B factory, which involved people from Riverside, Santa Barbara, Irvine, UCLA, and Caltech. We had monthly meetings. The idea was that UCLA would explore their linear collider approach, and we would explore our approach, which was asymmetric colliding storage rings. We worked this way for quite some time.

There were ideas about an expansion of the UCLA campus across Veteran Avenue. The idea was that this would be a good place to put the linear collider that Cline was exploring. We started looking at what we could do on the Caltech campus, around the periphery of the Caltech campus, or under it. In fact, the optimal size if you were starting from scratch would be even smaller than the campus, but that was the size of the PEP ring we familiar with. It would be on the campus, but underground. There was precedent for that. The symmetric B factory at Cornell was right on the Cornell campus under the football field, so that kind of thing could be done. That's why we called it CITAR, for California Institute of Technology Asymmetric Rings.

Another site could have been at Edwards Air Force Base. JPL had a site at Edwards that dated back to its origins, when they were doing rocketry and developing rocket fuels and setting off explosions. In those days, there was a helicopter shuttle between JPL and Edwards Air Force Base. I don't think that exists anymore. The base provided a fenced, controlled area. You wouldn't have done it underground there; it would be above ground with surface shielding. The SPEAR accelerator, the first accelerator at SLAC that we talked about, the one that discovered the J/psi, and the original magnetic detector, and the Mark II and the Mark III, was not subterranean. It was all built in a parking lot, with concrete blocks stacked on the sides and top as the shielding. For higher energy, that would have to be a little thicker, but still a perfectly practical approach.

Another thing that we haven't talked about is exploiting the synchrotron radiation emitted by the beams in a storage ring. As you bend the electrons in a circle they emit synchrotron radiation, typically x-rays, which are valuable for all kinds of biological and solid-state studies. There are many of these rings now around the world that do that. This was initially developed as a parasitic technique at SPEAR, before they eventually took over the entire ring for their own purposes. It was done at Cornell. Special-purpose rings were built at Berkeley, at Brookhaven, at Argonne, everywhere, but you could also do it parasitically. There were lots of people at Caltech who do that kind of physics, using synchrotron radiation, for biology, for condensed matter studies, or whatever. They would go to SLAC or Argonne or other places to do this work.

I thought that one way to get support for CITAR would be to talk to these people here on the campus about working together, to provide an additional motivation for building this accelerator on campus. We would do high-energy physics, and they would parasitically do synchrotron radiation studies. I found the relevant individuals, half a dozen or so, who did that kind of work, and we had a meeting. The idea was, let's pitch in and work together on this, and let's see what comes of it. I had done some preliminary work, so I knew what the capabilities could be. It actually would have been quite impressive, in fact. However, the reaction I got uniformly from all these people was, "Oh, that's extremely interesting. When it's ready, call us, and we'll bring our experiment down." There was absolutely no willingness to go the route of "Let's design this, let's figure out what the best way to do it is, let's figure out how the two operations will work together," all the kind of things you would need to have something that was beneficial and viable and helped move the project forward. Their attitude was just, "Well, yeah, that's great, I'll be very happy to use it." That just didn't help; it didn't get us anywhere at all.

So, we just pursued it as we talked about before, doing the initial accelerator physics studies. We raised some money for this work. I think I had mentioned that I had some involvement with the Caltech Associates to try to raise money for this initiative. We did get quite a nice amount of money— $100,000—from Millard Jacobs, who was a member of the Caltech Associates. We used it for two things. One was to bring an accelerator physicist named Jeff Tennyson to the campus for six months. The other was—one of the things we were worried about is we had this very intense beam and you have to make a region where the particles collide which is very thin so that you don't scatter the particles as they emerged from the beam pipe. At the same time, you have to worry about having to cool the pipe so that the circulating beam doesn't overheat the beryllium beam pipe. That was a hard problem which we didn't really know how to do at a professional engineering level. So we used some of the money to fund a study at JPL who knew how to do this kind of stuff, and they wrote a very nice report showing that you could do this, and what the best way to provide the cooling was to keep it thin, what the temperature rises would be, and all that. This was a rather important engineering concept; we could show people that we actually knew how to solve that difficult problem. We worked on this using Jacobs' funding and made a lot of progress.

What then happened was that one day people from SLAC appeared at one of our meetings, which were open. Other people at SLAC—Elliott Bloom, Gary Feldman—had made common cause with Pier Oddone from LBL to try to do something of this kind at SLAC. They had taken the wrong technical tack and so what they had done actually was not viable, but that wasn't obvious at the beginning. They just were unable to arouse sufficient interest at SLAC either, but some of them still actually wanted to pursue it. Then Jonathan Dorfan came to a meeting, Jonathan was an old friend. He did his thesis on my E92 experiment, and then we worked together on the Mark II. He was then a group leader at SLAC, the inheritor of Richter's Group C. We decided to make common cause to try to get this done at SLAC. We already had a pretty good initial technical take on what needed. With Jonathan's involvement, we started getting accelerator physicists from SLAC and LBL to participate. They initially did not like our concept at all.

ZIERLER: What was the issue?

HITLIN: The issue was that you had to do two things that had not been done before. One was you needed a much higher luminosity than any accelerator of this type had ever had, by at least an order of magnitude, maybe more. The other was that these were asymmetric collisions. You didn't have two beams at the same energy; you had two very different energies by about a factor of three. That had never been done before, so you had to figure out a way to make that happen and get the parameters to all work out. There are different ways of approaching it. The way we chose was to have a conventional charge in the individual bunches of electrons and positrons that you collide in the rings. You can either have very intense bunches and fewer of them, or less intense bunches and more of them. There are a lot of technical things that follow from that initial choice. The conventional way to do it was to have very intense bunches and fewer of them. We looked at that and we decided that would make for a lot of difficulties, so we hit upon this idea of a very large number of bunches, each of which was not that intense, which solves some problems, but creates others. When we presented that idea, the people who had been working on some of these earlier approaches before at LBL didn't like it. But we started working together on it, and we found that in fact it was the right way to go. The engineering design then coalesced around this concept, and obviously it proved to be the right answer. But it made for quite an interesting initial set of discussions.

One thing we did do, which I think was actually very important, is that not all, but many of our meetings in these early phases where we were trying to find a conceptual framework, involved both the detector physicists, the experimental physicists, and the accelerator physicists, together. There wasn't one camp and another camp, and people doing incommensurate things; we worked on it together. We all made sure that we were on the same page as we the project developed. That continued all the way through the design phrase and through the operations phase It was a very successful model. I'm not sure that extrapolates to places like the LHC where the scale is bigger, but it did work very well at our scale. We continued that when we were working on SuperB, with the experimentalists and the accelerator physicists working very closely together.

ZIERLER: A totally different topic, one I've come to appreciate the importance of, and that is the Caltech Associates. Tell me about some of your interactions with them and the significance of that.

HITLIN: I hadn't had an enormous amount of interaction with them. It was all centered around this question of getting the B factory started, so it was in a very well-defined period of time, about five or six years. The Associates are interested in what goes on at Caltech, and they provide financial support. In turn, Caltech puts on activities for these people and inform them of the interesting science going on at the Institute. It sometimes involves field trips with geologists or visits to an accelerator or things of that sort.

There are regular Associates dinners at the Ath. Back in those days, which was late 1980s, early 1990s, they were black-tie affairs. I don't know if they still are; I haven't been to one in years. I had a tuxedo from when I got married that I hadn't worn since, but I dug it out and it still fit. I don't know if it would fit me now! One of the things I remember very well—Herman Wouk, the author, was a member of the Caltech Associates, and I had dinner with him on one occasion. That was interesting for two reasons. My former wife, who was principal of various Jewish religious schools, got into a furious argument with Wouk over something to do with religious education. The other thing was that Wouk kept kosher. We know the food at the Ath is pretty good, but it isn't kosher.

ZIERLER: That's right. [laughs]

HITLIN: His meal was therefore one of these foil-wrapped airline meals—everybody else was eating really nice food, and his meal was an airline tray. But that was his choice.

ZIERLER: When you were talking about synchrotron radiation, what were the other options that you were considering?

HITLIN: I'm not sure what you mean.

ZIERLER: One of the things we were going to talk about was the synchrotron radiation option.

HITLIN: Yes, right.

ZIERLER: The question is, what else was there to choose from?

HITLIN: You either have that option or you don't.


HITLIN: The point is, this is something that comes not for free, but for a relatively small additional investment, if you want to do it. If you want to make use of the synchrotron radiation, you can. When we built PEP-II at SLAC, we did not do it. The reason was that we used an existing underground tunnel, the original 2.2-kilometer PEP tunnel. The synchrotron radiation comes off at a tangential angle to the ring, and you need to go some distance out to place experiments. If you have it in a tunnel, there's no practical way to do that, because you hit the wall of the tunnel too soon. It was done at SPEAR because, as I said, SPEAR was in a parking lot and the shielding was just concrete blocks, so all you had to do was move one of the blocks to let the synchrotron radiation out. It was very easy to add it. We could have designed a new machine. Even if it were in a tunnel, you could easily, ab initio, make tangential ports through the tunnel to allow you to do synchrotron radiation. But you have to design that in from the beginning.

ZIERLER: The last major topic for us to touch on in this addendum talk of course is BABAR. Just to orient ourselves chronologically, when was the two-year leave that you took?

HITLIN: It was 1994 and 1995.

ZIERLER: The idea was this was just so intensive you couldn't do this as part of your general responsibilities?

HITLIN: Basically, yes. We had been working on this idea since 1987. We first took it to one of the Snowmass activities in 1987 and 1989, where I ran working groups on CP violation experiments. We worked hard on the design. We worked to get it to become part of the SLAC program and all of that. This was all done by a relatively small group of 15 or 20 people working on the various aspects. Then at the same time we were doing the political work to try to get approval, which involved things I was doing, and SLAC people were doing. SLAC had a political consultant named Pat Fulton, who was also then hired by Caltech. All of the labs and universities hire lobbyists to be clued into what's happening on Capitol Hill and to try to move projects forward in the congressional sphere, in the OMB, and all that sort of business. We had visits with various representatives and senators. I did some of that. Burt worked with the governor's office; he worked with the local representative from San Jose, Vic Fazio, and later with Anna Eshoo from San Mateo, who is still in office, to try to get PEP-II moving. The Pasadena congressman was Carlos Moorhead, who was Dean of the House; he was helpful in producing a letter of support from the California delegation. All of that was done by a small group of people trying to move the project forward.

PEP-II/BABAR may or may not have ever happened, except that in early 1993 the SSC was cancelled. PEP-II and BABAR were something of a consolation prize to the field for the cancellation of the SSC. I'm not sure that was a fair trade. If the SSC had gone forward, it would have been a much higher energy and much better machine than the LHC. It had 40 TeV in the center of mass, as opposed to the LHC's 13.6. It would have been ready earlier. It would have had higher cross-sections at higher energies. We would know a lot more now than we know from the LHC. Anyway, the SSC was cancelled, so all of a sudden, we were approved, but we were this hardy little band of people that was nowhere near capable of building an accelerator and building an accompanying detector and experiment.

We had to very quickly form two groups, an accelerator group and an experimental group. We had done some spadework on this. David Leith, who was the research director at SLAC, and I had worked together making the rounds of all the funding agencies in Europe the year before and seeing identifying people in the various countries who had some interest in this. There were French groups, Italian groups, German groups, U.K. groups and Canadian groups interested, and this involvement was backed by the funding agencies in these countries. What was missing of course was approval from the DOE. But when we got the approval in 1993, there was a ready-made group of people standing by and ready to join.

We issued a broad invitation for a workshop in December of 1993. I had held a bunch of workshops over the previous few years. The workshops that started out with perhaps 25 people and had grown to 75 or 100 people over time. When we put out this workshop call, and now it was known that there was actually funding, that was a completely different situation. Hundreds of people showed up, ready to go. We needed to get organized quickly. We had a very aggressive schedule, actually quite well-worked out, but very aggressive, and we had competition, because the Japanese were doing something rather similar. We needed to stand up a functioning organization rather promptly, that could design the detector and start to get it built. It was the same for the accelerator, to start finalizing the concepts, gather the troops and make it happen.

This became a full-time activity, to answer your question. It really seemed like the right way to do this was to just take a leave and do it, and there was plenty of precedent for that sort of thing. It worked out well. I was still living here; I never moved up to SLAC. I would typically go up on Monday morning on a 6 AM plane and come home late that evening. I would spend Tuesday on campus. Then Wednesday through Friday, or Saturday if something was happening, I would be at SLAC. That wasn't particularly good for home life, but it was sufficient presence to make things happen. We had email by that time, of course. We didn't have Zoom, but we had email. We did have video conferencing. We had a microwave link between SLAC and LBL that had been used back in the 1970s that worked well. It really was a full-time activity. Jonathan ran the accelerator project, and I ran the detector. We had all of these countries that had to get organized. We had to have a governing organization. We had to write a Collaboration constitution. We had to stand up a Collaboration Council that had one representative from each institution, that dealt with certain activities of Collaboration. We wound up in the end having 70-odd institutions from a dozen countries on BABAR. The Council is the organization that formally hires and fires the Spokesman, things of that sort; sets policies about publishing; a lot of important activities. Then there's another group, an Executive Committee, which would have one representative from each country. Because the U.S. was half the collaboration, we had several—one from SLAC, one from LBL, a couple from the universities. Then one representative from each of the major countries—Germany, Italy. France had two because it had two funding agencies, so we had to have two. It was roughly 50/50 US/non-US. That organization then became the body that made the technical decisions, decisions about appointments of managers, once in a while the difficult decision about removing and replacing a manager.

We had to stand up these organizations and figure out how to work together. We originally said we would have monthly Executive Board meetings in person. We did that for five or six months, and then people got really tired of that. Transatlantic travel once a month was too onerous for a lot of people, so we eventually cut that back to quarterly with phone meetings in between. Getting all of that going while trying to finish the design of the detector, develop a budget and schedule, get various approvals, start building and testing things on a very aggressive schedule, was a full-time job.

ZIERLER: Administratively, on a two-year leave and bring principal investigator on the DOE grant, is that to say that the grant paid your salary for those two years?

HITLIN: There was an arrangement made with SLAC, so SLAC paid part of my salary and Caltech paid part of my salary. I don't remember the division. I had to have a formal appointment at SLAC, because I had to sign requisitions and make personnel evaluations, all kinds of things that needed a formal appointment at the Laboratory. That continued the whole six years I was Spokesman. You needed that sort of formality to make things work. We built both PEP-II and BABAR in record time. We went on the beam in May 1999, about a month ahead of schedule. There were a couple odd little pieces that weren't quite done, but basically, we went on the beam line and took initial data a month before we were scheduled to, which just doesn't happen. That was the only example in my career where we actually beat the schedule. But we did. We won a DOE project management award the following year, beating out the re-plumbing of the National Petroleum Reserve.

ZIERLER: Tell me about the element of competition and where HEPAP came in, for example.

HITLIN: HEPAP is the advisory panel to the DOE and NSF. I was on HEPAP starting around 2000, which was quite interesting. The important thing I think to get across was that we (PEP-II and BABAR management) worked very intensively to have a competitive but respectful and harmonious relationship with our Japanese competitors. We had a lot of exchange of ideas, joint conferences, all kinds of things. We worked very hard on that. When they finished their detector and we finished ours and the accelerator and we started taking data, we had a much easier startup than they did. We came up to speed more quickly. Eventually, they wound up with a higher luminosity and got more data than we did, largely because we got prematurely shut down during the 2008 financial crisis. In the startup phase, we took off faster than they did. That sort of competitive situation is the kind of thing that interests the reporters from Science magazine and Nature and things of that sort. They were a little bit aggressive about it. There was an article that appeared which quoted Helen Quinn, who was a theorist at SLAC and a member of BABAR, with—I don't remember the exact wording, but it was some sort of a disparaging remark about the Japanese competition, in this article.

ZIERLER: Meaning that their science wasn't up to snuff?

HITLIN: No, that they were not doing very well in their commissioning. I know Helen very well, and she would never do that. She's an extremely judicious individual. She has been president of the APS. She told us she didn't say that, and we believed her. Jonathan and I then got on a plane and went to the KEK Laboratory and apologized on behalf of SLAC and the collaboration, to preserve the relationship. That was the way we were handling the relationship between BABAR and Belle.

When it came to HEPAP, the head of the Office of the Science at the time was Ray Orbach who had been vice chancellor at UCLA. I had met him during the Southern California Consortium activities. He came to the HEPAP, addressed the group, and gave an introductory rah-rah talk. Two of the things he said were — one, we're going to find the Higgs at Fermilab, and we're going to beat the hell out of the KEK B factory. I raised my hand and I said, "I can't comment on the Higgs, but I can comment on the B factory business, and yes, we are being very aggressive, and I hope we are going to beat the hell out of them, but we're not going to be overtly aggressive in our stance, or we shouldn't be, at the Department of Energy level, that aggressive about the relationship between these activities." Nobody said a word. Nobody on the committee said a word, so I was just sort of left hanging out there.

At the end of the meeting, the committee writes a letter to the Office of Science about what was discussed. As these letters are, it's full of boilerplate and platitudes and whatever, but it also addresses substantive issues. The chairman, Fred Gilman, went down the list "What do we say about the B factory situation?" Both PEP-II and KEK-B were starting to run, and it was a major activity. Half a dozen hands go up: "Well, we need to be very careful"—basically echoing exactly what I had been saying, but they wouldn't say it in front of the boss. So, we did get something in the letter, of a cautious and calming nature, to preserve the relationship.

ZIERLER: What has come as a result? What are some of the benefits we see in repairing that relationship?

HITLIN: We learned things from each other. One of the reasons they did not have a good startup is they had a problem with secondary electron emission from their beam pipe. It's something called multipactoring. We had paid more attention to that than the Japanese accelerator had. We had a technique for dealing with it. We actually sent someone to Japan who worked with them to solve that problem. These things are really very collegial by nature. We kind of do it by second nature. The problem was that Orbach didn't seem to understand that and could easily have made a mess of that kind of relationship. We just kept that up. We had lots of interactions. When we were working, for example, toward the next-generation B factory, the one we started doing in Italy and the one that they actually got built and are working on now in Japan, we had many joint workshops, in which we worked together on solving various problems and comparing notes, even though we were in intense competition. It's just the way you do it. It's mutually beneficial for everybody. You don't have trade secrets; it's not a concept. It was important that this approach be preserved.

ZIERLER: The last topic for us to cover, and it's one that's a bit of a retrospective look, and that is, CP violation over five decades in your career. The first question is, where are we now relative to the very first ideas in CP violation? What has been resolved, and what is still a big mystery?

HITLIN: From an experimental point of view, CP violation has been seen, as I've said, in the decays of K mesons and the decays of B mesons. The advantage of B mesons is there are lots of measurements you can make which are directly interpretable in terms of fundamental quantities. We have made an enormous number of measurements and we've really made very stringent tests of the Standard Model. One thing to understand is that all of the decays of B mesons that exhibit CP violation are very rare processes. They typically are a part in 104 to a part in 105 or 106 of all the decays of B mesons. That's why you need to have such an intense accelerator, because you have to produce a lot of B mesons to get a good statistical sample of these very rare processes. Many measurements have been made on these, and every one of these has basically reinforced the Standard Model. Your best shot at finding CP-violating decays that don't validate the Standard Model are now even rarer B meson decays that involve what's called penguin diagrams, and they are 5% or 1% of these other rare decays. That's why in a super-B factory you need even higher luminosity to get statistically viable samples. These decays provide another potential avenue to probe deeper CP violation. There are as yet no measurements of those decays which are of sufficient precision to tell you very much. That can be addressed, in principle, when the full sample of B's are obtained at the KEK super-B factory. Their time schedule has slipped rather dramatically, so they're now talking about having that sample in the year 2030-something. That's many years longer than what they initially planned. They have not done very well in their startup. They're currently shut down for two years to make some modifications, so it's going to be a long time before those explorations make progress. There is an experiment at CERN called LHCb which makes its B mesons in a hadron machine, which can in principle do many of the same things. They are doing well, but it's even harder to make those measurements there than it is at a B factory, so, they haven't done it yet. They have a shot, and so that may happen as well in the next five years or so. They're starting up again too, with modifications, with the upgraded LHC, and they will take a lot more data, so that could potentially be interesting.

So, the answer is we've learned a lot with what we've done with the data from Belle and BABAR. The next-generation experiments are in progress but haven't delivered yet. Frank and I placed our bets, in fact, that that would not necessarily be a productive avenue on a reasonable time scale to actually learn what you would like to learn about going beyond the Standard Model. That is why we did not get involved in the Japanese super-B factory. We placed our bet-on muon to electron conversion as a route to new physics. The other side of the CP coin is K decay. As I said, the K decay measurements that have been made thus far in that respect are not—the CP violation measurements—started the whole field but were not easily interpretable in terms of the most fundamental quantities. There are a couple of decays of kaons, which are extremely rare and extremely difficult to detect, which do have that sensitivity and that connection to the fundamental quarks. There are experiments that have been designed and even one that has been launched at CERN, to attempt to do this. These decays are even rarer. We're talking about few parts in 1011 of all K decays. You can make more K decays than you can make B's, so you can actually attempt to get large enough decay samples, but it's really hard. That work has been ongoing for years and years. It doesn't have results yet but may in the next few years. In both of those fronts, it's still alive, they're still doing interesting things, and they might have something to tell us about these other aspects of CP violation.

In 2014 there was a commemorative conference held in the U.K. at Queen Mary University called the 50th Anniversary of CP Violation. Most of the discussion and presentations were about the things we had done at the B factories, but the main participant and the reason the conference was held was Jim Cronin from Chicago, one of the pioneers of the first experiment that saw CP violation. He gave an extremely interesting talk that I remember very well, in which he went back to their old lab notebooks from the experiment and showed all the documents that they had used to get approved, which was two pages and one committee meeting, and then how they took the data, how they understood the data. Here was this enormously important experiment; it involved four people and a time scale of a couple of months, and they did a very important thing. That's what Jim was emphasizing, how much the world had changed over time, where now to even aspire to do something of similar impact and importance takes two decades, a thousand people, and hundreds of millions of dollars. It was interesting to juxtapose these two things in this conference.

ZIERLER: Perhaps the most accurate gauge of where CP violations is looking to the future—is this an area where you would advise graduate students to focus on?

HITLIN: If you want to do high-energy physics, and you want to be part of these now inescapably large decade-long activities, sure! CP violation in the quark sector is still one of the most interesting things there is, but the game has changed quite a bit. We haven't found anything beyond the Standard Model, although as I've just told you, we're still looking. The new emphasis is on CP violation in the lepton sector, and there are different ways to approach that problem. The most straightforward one is to look for CP violation in neutrino scattering, the difference between neutrino and anti-neutrino scattering. If those cross sections are not identical, that's CP violation. But this is entirely in the lepton sector, and to be relevant to the baryon asymmetry requires an additional feature, which is not a part of the Standard Model; the neutrinos themselves need to be Majorana particles. That is, the neutrinos need to be their own anti-particles. This is possible, but there is no evidence as yet. This is also the subject of major searches, in particular in double beta decay, which we mentioned in our previous discussion. In order for CP violation in the leptonic sector to be relevant to the baryon asymmetry, you need to have Majorana neutrinos. There's a lot of activity on double beta decay going on, and of course there is DUNE, the major experiment at Fermilab, sending neutrinos to South Dakota, which has as its main objective, as does its rival experiment in Japan, to measure CP violation in the neutrino sector. That is a major thrust of the field.

ZIERLER: For my last question, with over 40 years in the Physics Department at Caltech, what has changed and what has remained the same? That's an open-ended question, either for better or worse.

HITLIN: The people have changed, of course. It's all the relationship between individuals, so that is a bit different than it used to be. The activities have become more far-flung. A lot of people spend a lot more time at CERN or Fermilab, things of that sort. I think the interaction between the individual groups in high-energy physics are not really as close as they used to be, not really as interactive as they used to be. That's just a fact of life.

ZIERLER: What about the interplay between experimentalists and theorists? The number of theorists versus the number of experimentalists?

HITLIN: I don't think that has changed very much. It's all very small numbers, of course, at a place like Caltech, but that hasn't changed very much. There have always been those theorists who don't make very detailed contact with experiment, and then those that are more of this phenomenological bent who do. Typically, the string theorists are much more mathematical and don't need or make a lot of contact with the experimentalists. John Schwarz, who is the original string theorist, in fact knows a lot about phenomenology and has been very interested and interactive. The younger string theorists are quite mathematical. People like Mark Wise, are interactive, can interact and answer questions. Mark in particular is a treasure and very interested and very helpful.

ZIERLER: Finally, last question—a counterfactual one, just for fun. With all of the twists and turns in your career, had you not ended up at Caltech, do you think your research would have been significantly different? In other words, by dint of being at Caltech, how central was that to the kinds of research you've pursued over the decades?

HITLIN: I have no idea how to answer that question. Resources are important, and we've had adequate resources here to do what needs to be done. The group has never been huge, but it has been reasonably well supported, and we've had enough capability to take on big things and originate and have substantial impacts on experiments. Whether that could have been done at other places, I would think it probably could have. Opportunities may have been different, or style may have been different, I don't know, but I think it could have been done. Caltech is of course unique in its approach to certain things, which as far as I'm concerned has been all to the good. But I really can't answer the question; it's totally speculative.

ZIERLER: On that note, I want to thank you for having the bright idea to reengage and to have a more specific Caltech-oriented discussion. It's going to be a great addendum to the original one and fantastic for the history of physics and for Caltech. Thank you so much.

HITLIN: You're welcome.