John Ellis

Zierler:

Okay. This is David Zierler, Oral Historian for the American Institute of Physics. It is May 6th, 2021. I'm delighted to be here with Dr. Jonathan Richard Ellis. John, it's great to see you. Thank you for joining me today. 

Ellis:

It's a pleasure. Glad to see you.

Zierler:

John, to start, would you please tell me your title and institutional affiliation?

Ellis:

I'm Clerk Maxwell Professor of Theoretical Physics in the Physics Department of King's College, London. And I also have a visiting scientist appointment at CERN. So, that's the main affiliations. 

Zierler:

Do you spend more of your time in London or in Geneva? And this obviously would be a pre-pandemic kind of question. 

Ellis:

Yeah. So, my residence is actually just outside Geneva. So, before the pandemic I would commute to London for let's say two or three days a week during the teaching terms as often as I could when I wasn't gallivanting off to some other distant part of the world. But needless to say, that's been completely shut down since the start of the pandemic. I haven't been to England—in fact, I haven't been anywhere since the middle of March last year. 

Zierler:

What is the vaccination situation where you are right now?

Ellis:

It's improving. My wife and I were vaccinated for the second time a bit over a month ago. So, we're in the clear. We're ready to go off to the disco, but don't see many discos open yet. 

Zierler:

John, a question we're all dealing with right now, in the pandemic with remote work, how has your science been affected one way or another? In other words, in what ways have you had extra bandwidth to work on problems that you might not otherwise have, simply by not traveling all around, and alternatively, how has not seeing your colleagues and collaborators in person adversely affected what you would have done this past year?

Ellis:

I think there are two main effects. One thing that you already mentioned is that one doesn't waste as much time sitting around in airports or on airplanes. Although, I should say that airplanes are often conducive environments for getting some work done. At least, in the good old days you couldn't be interrupted on the telephone while you're on the airplane. But certainly, one doesn't spend so much time traveling. So, that's a positive outcome. Of course, the negative side of that is that you don't have the sort of random contacts that can often produce serendipitous insights or pieces of information, ideas for new projects. So, what I would say is that for the first two thirds, perhaps, of the pandemic so far, I think my overall scientific productivity was not adversely affected. But in the last few months, I noticed it's slowing down somewhat, and I think that's just because there aren't these serendipitous new ideas flying in. 

Zierler:

I'm curious about the Clerk Maxwell professorship. Can you tell me a little bit about the name or the family behind that endowment? 

Ellis:

That's named for James Clerk Maxwell, who was a professor for theoretical physics in King's back in the mid-19th century. In fact, much of his famous work, including the proposal of Maxwell’s equations, was done while he was professor at King's. So, that was roughly speaking between 1860 and 1865. So, then after that, it was decided to institute this named professorship, and it's been through various hands. I happen to be the present incumbent. 

Zierler:

A very in the moment question right now, I'm very much enjoying getting people's speculative snap judgments about the g-2 muon anomaly that everyone is very excited about at Fermilab. What's your perspective on this, and what are the prospects that this might mean new physics? 

Ellis:

I'm excited about that, too. It's one of the things that I'm working on these days. I had somehow not been anticipating that the Fermilab result would resemble so closely the Brookhaven result. Maybe out of a spirit of conservatism, I somehow expected that the experimental measurement would drift back towards the Standard Model calculation. But it hasn't. So, that's great. There are a couple of small caveats that I would make. One is that the method being used in the Fermilab experiment is essentially identical to that used in the Brookhaven experiment. The magnet is the same, although of course the individual detectors inside the magnet are new. The other thing is that there's still some dispute amongst the theoretical community about the Standard Model prediction. So, I think pretty much everybody had a consensus agreement, but there's one lattice group which reports a somewhat different result from their lattice calculation. I think people, including myself, are still trying to get their heads around that. But setting those two caveats aside, I'd say it's pretty exciting. 

Zierler:

John, sort of a broad question for where the field is right now. I know over the course of your career, you've been very well attuned to when theory has provided guidance to experimentation, and when experimental and observational advances have guided or advanced theory. Just generally, in the many fields that you're involved with right now, what's your sense of the state of play?

Ellis:

Can I just add in one remark about g-2? One piece of welcome news is that there's this experiment at J-PARC, which is currently being planned, which should come online in a few years, which uses a completely different experimental technique. So, if that comes up with a result that is the same as Brookhaven and Fermilab, then I'll be really convinced about the measurement.

So, to come back to the relationship between theory and experiment. I think everybody can agree that healthy science requires some sort of balance between the two. There's sort of a dialectic process in the advancement of science. That doesn't mean it has to be a detailed balance, that at every instant the theorists and experimentalists are in step. Sometimes theory is ahead, sometimes experiments are ahead. So, I guess, in accelerator-based particle physics, there was a feeling until a couple of months ago that theory was way ahead of experiment, in the sense that theory had successfully predicted the existence of the properties of the Higgs boson, experimentalists had discovered it, but they haven't discovered anything else new at the LHC.

So, in the last couple of months, in addition to the g-2 result that we just discussed, there's also been confirmation coming from the LHCb experiment of anomalies in B decays, which could also be harbingers of new physics. So, that's something else that I'm working on at the moment. Not to say for sure, but maybe we're moving from a period in which theory had got ahead of experiment, to one where experiment is now giving us serious challenges to work on. So, that's one field.

Another field that I'm interested in is the whole area of dark matter. So, that, I guess I would have to say, is one where experiment or observation has got way ahead of the theory, in the sense that there's gazillions of observations that tell us dark matter has to be there and, roughly speaking within 1% or so, tell us show much of it has to be in the universe today. And we theorists don't have a clue what it is. Or more accurately, we have about a thousand different ideas. Perhaps it is a personal judgment to say there is no clue pointing us which direction to go. There are a couple of theoretical ideas that I find very attractive. One is of course supersymmetry. I'm sure we're going to come back to supersymmetry later in the discussion. That's something that I've been working on for almost 40 years now. And another one is some sort of a light bosonic field for which the prototype is the axion, which again is something that I've worked on in the past. In fact, I just put out a review paper on that earlier on this week. But there are also other ideas, and I think that's an area that's really rife with theoretical speculation at the moment. 

Zierler:

John, this is as much a nomenclature question as it is a scientific question, but between dark matter or g-2, when we talk about new physics, when we talk about physics beyond the Standard Model, to what extent is this really about still understanding what fits within the Standard Model, and to what extent is it representative of the field to say that it has a full understanding of the parameters of the Standard Model, and these new discoveries would certainly be beyond that? 

Ellis:

I think dark matter, for sure, would be beyond it. There's just no way of explaining the dark matter within the Standard Model. In the case of g-2, clearly there is some debate going on with at least one lattice group contesting what would otherwise be the consensus theoretical estimate of g-2 within the Standard Model. So, I'm sort of 90% convinced that that really is physics beyond the Standard Model, but 10% remains to be convinced. 

Zierler:

This would be a very speculative question, but given your interest in supersymmetry, and perhaps some of the impatience or frustration in the field about not having facilities that operated at high enough energies to potentially see it, what are some alternative paths, including even simulation from quantum computing, that might give us a better handle on supersymmetry, and what it might portend?

Ellis:

Okay, well, the first comment that I would make is that I think supersymmetry still can provide an explanation for the g-2 anomaly. People will often say, "Ah, well, we haven't seen supersymmetric particles at the LHC. They'd have to be so heavy that they couldn't explain g-2." Well, that's not true. The LHC constrains strongly attracting supersymmetric particles directly. Also, some constraints on other supersymmetric particles, but the real strong constraints are on the squarks and gluinos. What you need in order to explain the anomaly of g-2 is a light supersymmetric partner of the muon, and some sort of spin one half supersymmetric particle type. That's not ruled out by experiment. So, one has to relax some of the theoretical assumptions that meson theorists often make, but it's perfectly possible to reconcile the absence of supersymmetric particles at the LHC with a supersymmetric explanation of g-2.

Of course, going forward, we'd like to be discovering those supersymmetric particles. I think it's perfectly correct to say that at the moment we don't really have a good idea of how heavy the other supersymmetric particles, squarks and gluinos, and so on, might be. So, if supersymmetry provides the dark matter, which is also completely consistent with what I've been saying, then squarks and gluinos should not be, quote-unquote, too heavy. But too heavy means like 10 TeV, maybe even more than that. So, if we wanted to discover them in that mass range, then we would certainly need a new higher energy collider. 

Zierler:

Using the discovery of the Higgs and the theoretical confidence that it was there as a base point, what is your level of confidence that supersymmetry will be found with the proper experiments to find it?

Ellis:

I think the degree of confidence is much less than the case of the Higgs boson. So, the Higgs boson, or something resembling it, was necessary for the renormalizability, and hence the calculability of the Standard Model. The wonderful agreement with experiment and observations, in particular, high precision measurements at LEP, and more recently the plethora of measurements at the LHC, the Higgs really had to be there in order for the Standard Model to hang together and be able to make those kinds of predictions. Of course, that's not the case for supersymmetry. Supersymmetry is an add-on as far as renormalizability is concerned, and that's one of the reasons why it's been so difficult to constrain what masses of supersymmetric particles might be.

So, if we go back a ways, there have been several instances where it's been possible to predict the masses of particles within the Standard Model just because of the key roles they played in its consistency when you calculate. I'm thinking of the prediction of the charm mass that was done by Mary Gaillard and Ben Lee and quite a number of other people back in the early 1970s. Charm had to be there and with a mass around a couple of GeV. Tick that off. Then, come the 1980s, experiments started to be sensitive to corrections induced by the top quark. So, Veltman was a pioneer in calculating those corrections, and we realized in 1988, 1989, that you could use those calculations and the emerging data on the neutral current weak interactions to constrain and eventually predict what the top quark mass will be. And of course, that prediction turned out in 1995 to be successful.

A couple of years later, in the early 1990s, we realized that you could also constrain the mass of the Higgs, although there was some argument. Constraining the mass of the Higgs was a lot more delicate because observables are only sensitive logarithmically to the mass of the Higgs. But still, they're sensitive. We can't just shove the Higgs off to infinity, and shut the Higgs down. So, in the early 1990s, we were saying that the Higgs had to be less than about 300 GeV. Those calculations got much more precise as more data came in until eventually in 2011, a prediction based on precision measurements and the absence of Higgs in direct detection scans told you that its mass should be around 125 GeV, which it is. But that was all because the top and the Higgs played crucial roles within the Standard Model in being able to make precise predictions.

Zierler:

Is this all to say that you see the discovery of the Higgs as a capstone to the Standard Model? 

Ellis:

Yeah, it's a capstone for the Standard Model as it completes the arch. Of course, once you've got the arch completed, then you can go on to build the next floor of the building. The Higgs, I think, sews up all matter of questions which can no longer be completely open. So, Higgs coupling measurements confirm some things. We understand how it is that the Englert-Brout-Higgs mechanism gives masses to fermions. But we don't have any way of calculating what the magnitudes of those masses might be. Their pattern is a complete mystery. So, that's Higgs puzzle number one.

Then we come to the mass of the Higgs itself. Again, the Higgs mechanism doesn't tell you how big it is. That, of course, is a really big problem and sits in lots of theorists’ minds. It comes back to the hierarchy problem. Why is the scale of the Higgs interactions so different than that of gravity? So, that's the second Higgs puzzle. Then, the third Higgs puzzle—in the Standard Model, you've got a quartic Higgs coupling. And that coefficient is relatively small, and you can run the theory up to high energies using renormalization group, and you find that that Higgs quartic turns negative, which indicates an instability within perturbation theory in the Standard Model. So, one possible way of resolving that is supersymmetry. (I shouldn't be advertising supersymmetry all the time!) Anyway, quartic self-coupling, that's puzzle number three.

Puzzle number four, there could be a constant in the Higgs potential. In fact, we know there is. And that, of course, is dark energy or the cosmological constant. So, my point of view is that there's absolutely no surprise at all that that should be present. The puzzle is why is it so small? Supersymmetry might want to help, but it sadly won't be the full answer. Anyway, basically, every term in the Standard Model that you right down, which involves the Higgs, is problematic.

So, we're ready to go onto the next stage. One way of addressing the next stage is the Standard Model effective field theory, where you allow yourself to consider additional terms in the Lagrangian, not contained within the Standard Model, but are made up out of Standard Model fields. These include terms with mass dimension 6, 8, and so on and so forth, which would presumably be relics of some new physics beyond the Standard Model, with some high-mass particles that can give you effective interactions between Standard Model particles, in the same way that back in the day, the exchange of the W gave you a 4-fermion interaction. You could just do the same thing all over again with effective interactions between Standard Model particles generated by other heavier particles. So, one of the other things that I've been involved with is various analyses within the Standard Model effective field theory to see what constraints they'd give us on new physics beyond the Standard Model, and if there's any indication on which direction to go. 

Zierler:

Well, John, this has been a very forward-looking discussion up to now. Let's take it all the way back to the beginning. Let's start first with your parents. Tell me about them and where they're from. 

Ellis:

Sorry, before we get to my parents, at one stage, you asked me what I think are interesting areas, and we got to talk about collider physics, and we got to talk about dark matter. I should also mention gravitational waves, which is something which I'm also working on at the moment. So, there are a couple of things that I'm involved with. One is how one can use gravitational waves as probes of possible physics beyond the Standard Model such as phase transitions in the early universe, for example, or whether the graviton has a mass. Then, the other thing that I'm involved with is promoting a project called AION in the UK, which is a project to build a series of atom interferometers to measure gravitational waves. So, it's a very exciting thing to get involved with. 

Zierler:

It's a good reminder as well that projects like LIGO are really doing some of their most exciting work right now. 

Ellis:

Right, absolutely. So, they've uncovered quite a number of puzzles which may be indicative of new physics. For example, there was this discovery of a merger of two massive black holes, the heaviest ones observed so far, which have masses in a range where the astrophysicists say black holes shouldn't exist. The so-called black hole mass gap, somewhere between 60-120 solar masses. There was this event which seemed to a merger of two black holes in that mass range. One has to be a little bit careful because there's nuclear physics and other uncertainties, but maybe that's evidence for some sort of new physics. I haven't worked on that myself. I follow that one with interest. Or it could be that those black holes indeed do lie in a forbidden range, not because of new physics, but because they were formed by earlier mergers of smaller mass black holes. Anyway, that's a very interesting puzzle that LIGO and Virgo have thrown up. And there are also some puzzles in neutron stars. But okay, anyway. 

Zierler:

There's a lot to be excited about right now in physics. There's no doubt. Alright, so let's go back. Tell me about your parents. 

Ellis:

Okay, so my parents lived in sort of a dormitory suburb of London called Potters Bar. So, they met during the Second World War, although they came from the same town. They never actually met before the war, but they met at some air force base. Then after the war ended, they went back to Potters Bar. I was born about a year after the war ended. My father was in insurance. He was initially an actuary, so there was some mathematical background there. As time went on, he moved into insurance management. And my mother was a homemaker. 

Zierler:

Where did you grow up?

Ellis:

I grew up in the same small town just north of London, Potters Bar.

Zierler:

Tell me about your early schooling. What kind of schools did you go to?

Ellis:

Okay. I went to private schools before university. So, the primary school was in the same small town. In fact, it was just a 15 minutes’ walk away from where we lived. So, the headmaster at that school had been with the British Antarctic Survey way back in the day. If I tried to construct my intellectual trajectory, I sort of suspect that might have had something subliminally to do with my interest in science later on, because it meant that I wasn't exposed just to what most people who live in dormitory suburbs do, or think about, or talk about. Excuse me, I've got a frog in my throat. So, he was the headmaster of the school. The school didn't place any particular emphasis on science, though. There was a little bit more interest in history. At some point, the school ran out of things to teach me. So, the last few months, the headmaster's wife taught me German. I guess that stimulated my interest in foreign languages, which I pursued subsequently, although my German has faded away somewhat. 

Zierler:

John, as a boy, what sticks out in your memory as some of the legacies of World War II, the suffering that Britain endured?

Ellis:

I mean, I'd have to say very little. So, my parents, after the war, their first house was what was called a prefab. It's basically like a trailer home but without wheels, which was one of many that was built for the people coming back from the war. So, I have vague memories of living in that prefab house, but then when I was about six years old, we moved into a larger, more traditional house, which is where my parents lived until they died. I was also vaguely aware of the fact that there was rationing of food, but this didn't affect me personally because I wasn't the person who had to go down to the butcher's and argue for how much I could eat. 

Zierler:

Tell me about what in America we would call high school. What kind of school did you go to?

Ellis:

Before I go to that, I just wanted to mention one thing. I used to read a lot of books taken out of the local public library, and what I remember is that when I was about 12 years old, which is more or less the time I transitioned from the primary school to the secondary school, I used to read a lot of nonfiction books because I wasn't allowed to take out books from the adult section of the library, and the children's fiction was not interesting. What I read was the children' nonfiction, and in particular, I read books on history and science. My memory may be a faulty reconstruction, but my memory is that it was at that stage that I got interested not just in science but in the most fundamental sciences. So, the science of the very small, although then I wouldn't have known to call it particle physics, and what was going on in the universe, although back then I wouldn't have known to call it cosmology. So, that's one influence that I very much feel.

Okay, so, when I was around 12 years old, I went to secondary school, which again was a private school, or what we call in England a public school. I was a boarder there, which meant that I could come home for a day at weekends, but otherwise I lived and slept at the school. So, there I was very definitely interested in not just physics, but actually theoretical physics. So, the school system in the UK has this nasty tendency to try to channel you into a specialism before you're really ready to make a reasoned choice. But it suited me just fine because I knew what I wanted to do. So, I focused on mathematics and physics in the later years of high school. I remember my headmaster being very disappointed because I'd done well in the Latin exam, and he wanted me to become a classicist. I thought, no way. 

Zierler:

John, did you always view math as an entree to physics, or did you toy with the idea of pursuing mathematics itself?

Ellis:

No, I was very definitely interested in mathematics as the language of and the pathway into theoretical physics. 

Zierler:

Now, in the British system, of course, you need to have your academic focus set at the outset of college. 

Ellis:

Right. So, I went from secondary school to Cambridge, and there I started what we call the maths tripos, the maths course. There's basically two options at Cambridge if you want to do physics, in particular, theoretical physics. One was to go through the maths course, and the other was to go through what was called natural sciences, which actually in the Cambridge system is a very interesting course. It's quite broad; you can study a range of subjects including physics, chemistry, biology, and mathematics, and then focus down your specialty later on. There are distinguished theoretical physicists who went that route. Michael Green, for example, a string theorist who is more or less a contemporary of mine, he was in the natural sciences and then in the physics department, whereas I studied mathematics and then I went on to the department of applied maths and theoretical physics for my PhD. 

Zierler:

Now, in the British system where you have to declare a major right away, to what extent does that put you on a path in physics where your focus even early on is on either theory or experiment?

Ellis:

Right. So, if you—not wanting to enter too much into details, but if you were at Cambridge in the natural science tripos, then you wouldn't have had to choose straight away. But clearly, if you're doing the maths tripos, doing the maths course, then you are a theorist. You're not an experimentalist. One of my regrets is—well, I shouldn't say a regret—one of the limitations of my undergraduate career was that I didn't really get involved in experiment, and I would also say that my education in theoretical physics was not as broad as it could have been, and maybe one might say it should have been. For example, I didn't have a course on statistical physics or thermodynamics until master's level, and the course that I went to was given by Jeffrey Goldstone, who was an absolutely wonderful guy, but his lectures were completely incomprehensible, so after a week, I gave up. I figured out that I could get my master's without learning anything about statistical physics and thermodynamics. 

Zierler:

John, was the culture at Cambridge when you were an undergrad such that you could develop meaningful relationships with your professors, or as an undergraduate, was that out of reach?

Ellis:

I would say it was out of reach. However, what I did do was develop a relatively good relationship with some of the tutors. So, the Cambridge system then, and I don't think it's changed since, was that in addition to the lectures that were provided by the university, your college would provide tutorials given to groups of two or three people, typically by research students or postdocs. I made a number of scientific contacts in that stage, some of which I've maintained since. For example, one of my first tutors was George Ellis who is a very famous cosmologist from South Africa, and still a personal friend. We still meet up every once in a while, often at a philosophy conference in Hay-on-Wye. At that time he was working very closely with Stephen Hawking. So, the first time I saw Stephen Hawking was when I was a first-year undergraduate, and there was this guy who was making his way along the corridor with some difficulty. He was already marked down as being a brilliant theoretical physicist. And there are some others that I haven't mentioned. 

Zierler:

John, looking back, how parochial was your worldview in physics as an undergraduate? In other words, was what was going on at Cambridge the be all and end all, or were you aware of what was happening in the United States, for example?

Ellis:

I would say that this mathematics tripos that I did didn't really bring you into contact with what were at that time the current developments of physics. It was providing a very solid basis: Maxwell's equations, quantum mechanics, a little bit of topology. All sorts of great stuff, but it really didn't put you directly in contact with what was happening out of the labs, or even in the research institutes in theoretical physics. However, we got a number of excellent speakers coming through, who would give invited lectures.

For example, I remember a lecture by Murray Gell-Mann about quarks. I'm trying to remember exactly when that would have been. It would have been in the mid-1960s, while I was an undergraduate. I also remember another visiting lecture was David Bohm, a famous UK theoretical physicist, one of the founding fathers of general relativity, who also had some really nutty ideas about quantum mechanics. So, there were some things like that. And then, one of my undergraduate tutors or supervisors was Jayant Narlikar, a collaborator of Fred Hoyle, and through him I got to hear about the debate between the steady state theory and the Big Bang. I guess that I was aware when Ryle, in particular, was arguing that the observations of the cosmic microwave background radiation were solid evidence against the steady state theory. 

Zierler:

Did you ever consider leaving Cambridge for your graduate degree elsewhere, or was the idea, you're already at Cambridge, this is the best place to be, why would you go anywhere else? What were the cultural considerations surrounding those decisions?

Ellis:

So, I definitely did think about going elsewhere. In particular, I thought about going to Imperial College London, where at that time there were some very excellent people. Abdus Salam, of course. The main professor there at the time was a guy called Paul Matthews. So, I actually went down to Imperial, looked around, and talked to Professor Matthews, and formed an impression whether I should go there or not. I decided not to because I realized that although Abdus Salam was doing extremely interesting stuff, he was actually relatively rarely at Imperial. He was jet setting around the world. I forget now whether his International Centre for Theoretical Physics in Trieste had already been established, and I think it had been. He wasn't around all the time. And Paul Matthews himself was not doing stuff that I found particularly interesting. So, there was that, coupled with the fact that I knew that there was a young lecturer called Bruno Renner coming to Cambridge that I had heard very good things about in terms of what he was doing scientifically, and that he was a good person to be working with. A nice guy, so I thought, okay fine. I'll stay in Cambridge. 

Zierler:

What was your sense of the most exciting areas of theory to pursue at this point? Was it only particle physics? Was anybody talking about theoretical astrophysics or cosmology at this point?

Ellis:

So, we're talking about—I finished my undergraduate degree in 1967. Then there was this one year, a sort of master's course, called Part III. So, in that, I followed a course on general relativity given by Dennis Sciama. So, he was really one of the founders of the UK School of Theoretical Astrophysics. People like Stephen Hawking, George Ellis, I'm not sure about Martin Rees, went through the Sciama school, so to speak. So, he was a very important influence, if not formative, as a supervisor or their boss. Anyway, he gave a course on GR which I took. I also took for audit a course given by Hoyle on astrophysics. Actually, it's a bit of an exaggeration to say the course was given by Hoyle. He gave about half the lectures. The other half of the lectures were given by other people because he was off traveling somewhere. Anyway, that provided quite a broad basis in astrophysics. So, I'd certainly acquired a little bit of that intellectual baggage before I started my PhD. In fact, I did consider the possibility of going into astrophysics and cosmology for my PhD, but I didn't stick with it. I had more interest in particle physics. 

Zierler:

So, you would say that at that point, cosmology at Cambridge was considered a respectable discipline. 

Ellis:

I would say so. Of course, many people would not have considered Hoyle to be respectable, but Ryle was certainly respectable on the observational side. And like I said, there was the Sciama school in the maths department which was certainly very strong—in fact, what I did when I was a PhD student was, I committed what might be regarded as a sin. Namely, during the coffee breaks, I often used to go down and sit at the general relativity table, and not at the particle physics table. That was one way in which I built and maintained my relationship with George Ellis, for example. 

Zierler:

I'm very interested in the ups and downs of general relativity as a fashionable area to study. Many of your contemporaries, particularly in the United States, would say that GR was a bit of a backwater at that point. Was that not the case at Cambridge?

Ellis:

I would not have said so. I mean, I would have said that it was a very active area at Cambridge at that time. I'd have to look back and figure out when the famous Hawking had his book on large-scale structure and spacetime, when that came out, so I can remind myself when the Hawking-Penrose singularity theorems were proved, and stuff like that. But I would say that GR was certainly big in Cambridge. In addition to the people that I mentioned, I might also mention Brandon Carter, who was in the math department at that time. In fact, in my second year of my PhD, I went to my supervisor and said, "Well, why don't we organize exchange lectures between the particle physics group and the GR and cosmology group?" So, for one term, maybe even two terms, we would alternate one of them giving a lecture, and then one of us giving a lecture, and I remember one excellent lecture given by Brandon Carter. 

Zierler:

John, later on, of course, so many particle theorists would go into astrophysics and cosmology. Was your sense that your intellectual ideas at this point were a bit ahead of the curve? Were many other people thinking about the potential applications of particle theory to astrophysics?

Ellis:

I think I'm probably going to have to say no. I might mention, by the way, just one other chance contact, which was that I went to school in Schladming, in Austria, when I was a first year PhD student. And there was this guy there whom I recognized from Cambridge. I had seen his face. So, we introduced ourselves, and he was Gary Steigman, who of course made many of the pioneering calculations on Big Bang nuclear synthesis. We collaborated later on, but at that time he was interested in the cosmological matter-antimatter asymmetry. Still a big subject. So, yeah, if I think about my contemporaries, I know a number of contemporaries who went on to become extremely well-known. I already mentioned Mike Green. Peter Goddard is another one, but they were much more into mathematical physics. 

Zierler:

John, what was the intellectual process of developing your thesis research?

Ellis:

Okay, so I mentioned that one of the reasons I stayed in Cambridge was because I knew there was this guy Bruno Renner who had been a student at Cambridge, and now had been in the US. He was working with Gell-Mann, and he was coming back to Cambridge as a lecturer. Like I said, he was a good physicist, and also a nice guy. So, when the time came for me to choose who I would like to work with as a supervisor for my PhD, I remember the professor asking, “Well, do you want to do group theory, or would you rather work on analyticity?” I should say, Cambridge at that point was very big on the analytic S-matrix, and all that sort of Chew-Mandelstam sort of stuff. I said, “Group theory.” So, then he said, “Well, do you want to do the more mathematical side of group theory, or are you more interested in the applications of group theory and symmetries of particles?” I knew what the answer should be in order to get the supervisor that I wanted. So, I got him.

So, he gave a couple of startup problems to work on, which I can't say I did a particularly good job on. One of them was an extension of what he had been doing with Gell-Mann and Oakes, which was to do with chiral symmetries, specifically, chiral SU(3) x SU(3). There was interest in the role of strangeness in this whole chiral symmetry story. So, that was one project that he suggested that I work on. So, I worked on that, but it wasn't particularly exciting. A paper came out of it, but I think it was one of my more forgettable papers. And the other project that he proposed was to look at the decay of the rho meson into four pions using chiral symmetry. So, this I didn't do a particularly good job on it, and though Renner thought it was going to be solo project by me by myself, it eventually turned out to be a joint project. So, those two projects took up the first few months of my PhD. I would say that thereafter I was pretty much self-directed.

So, one of the topics that interested me was there was some work in particular—so, the dual resonance model had been proposed by Veneziano and people like Lovelace were trying to apply it to the scattering of pions. So, pions of course, because of PCAC exhibit Adler zeros in their scattering amplitudes. So, what happens if you impose that Adler zero condition on the dual resonance model of scattering. Well, you get a result which looked very much like the results in chiral symmetry. So, you're putting in one aspect of chiral symmetry, namely the Adler zero, and in some sense, you're getting the whole package. This was somewhat puzzling to people, but I was able to show that this was just basically a simple consequence of the fact which if you do a series expansion in the terms of the Lagrangian, as in an effective field theory, then the Adler zero condition is enough to basically fix, for example, pion scattering, and the dual resonance model is just one example of a model that does it. So, I wrote a few papers about that including one with my supervisor.

Interestingly, one or two of those papers got citations again in the last two or three years. It went through the doldrums for about 50 years, but it's back in the citation rotation because of all the interest in low-energy theorems, and to what extent they determine scattering amplitudes. 

Zierler:

John, what was going on in the world of experimentation when you were doing your thesis research that may have been relevant for your work?

Ellis:

I think the thing that was most relevant was the discovery of scaling in deep inelastic scattering. 

Zierler:

At SLAC.

Ellis:

At SLAC, yes. So, that was announced I guess at a Vienna conference in 1968. Some of the senior people, faculty, in the Cambridge group started working on a model of partons, which had just been proposed by Feynman. Bjorken also had some new ideas. So, that set of ideas was coming into the mainstream. I think it was probably the most relevant experimental development. 

Zierler:

John, given how wide-ranging your thesis research was, what would you say is the connecting thread, intellectually, scientifically, that brought all of these interests together for you as a graduate student?

Ellis:

Approximate symmetries of hadrons and their origins.

Zierler:

There's more. 

Ellis:

So, the next thing I got interested in was the possibility that scale invariance might be an approximate symmetry of hadrons. This is an idea that Gell-Mann amongst others was thinking about at the time. So, I got interested in formulating the effective Lagrangian, like chiral effective Lagrangians, but for this scale symmetry. So, I wrote out a Lagrangian, and the next week there was a paper, a preprint, by Abdus Salam on the same subject. So, I thought, “Oh shit. I've been scooped.” I looked at this paper, and then I saw that, actually, Salam had made a simplifying assumption which was not necessary, and also, he'd actually missed out one of the most interesting aspects of combining scale invariance and chiral symmetry.

So, I told my supervisor, and he said, “Well, Salam is coming to give a lecture next week in Cambridge. Why don't you go to his lecture and talk to him afterwards and see what he says about your work?” So, I did. And Abdus Salam wasn't at all interested. So, I went ahead and wrote up my paper. So, his reaction to my work was completely different than that of Bruno Zumino. You've heard about Bruno Zumino. 

Zierler:

Of course. 

Ellis:

So, in the summer of 1970, I visited the Brandeis summer school, which had this fantastic array of mind-boggling lecturers, including Bruno Zumino. Bruno Zumino was talking about effective Lagrangians, and lecture two, he’s talking about effective Lagrangians with scale invariance, very similar to Salam’s work. And I went up to him afterwards and said, “Actually, I think you've missed a couple of points there.” So, he immediately got the point. He paid attention, first of all, he got the point. Lecture three, he said, “Well, I'm going to give lecture two again because it's been pointed out that I missed a couple of important points.”

Zierler:

To go back to that question about, what connected it all for you in your thesis research?

Ellis:

It was like a little bit to what I said from how I feel at my PhD interview. What do I want to work on? I didn't want to work on analytic properties of scattering. I wanted to work on something which seemed to me to be more physical, and specifically on symmetries. Strong interactions at that stage was a complete puzzle. Analyticity was one way of trying to understand what was going on, but symmetries was another way of attacking the problem. So, I was interested in chiral symmetry, I was interested in scale invariance, and the title of my thesis was Approximate Symmetries in Hadrons, where I tried to tie those interesting things together. 

Zierler:

Now, at Cambridge, at that time, was there a thesis defense with a thesis committee?

Ellis:

So, you had a thesis committee consisting of one person from inside the department, and one theoretical physicist from somewhere else. Now, after two years of my PhD, my advisor, Bruno Renner, was German and he left to take a up a position at DESY in Hamburg. So, at that time, he arranged for me to go to spend the last academic year of my PhD at CERN where I was nominally under the supervision of John Bell, although we'd only discuss from time to time. I wasn't working with him, or even on things very close to what he was interested in.

Anyway, so the last year of my PhD I was in Geneva. So, how to organize my oral, and who should be on my committee? It was figured out that I, my advisor, who at that time was employed in Hamburg, so he could count as an external examiner, and my professor, John Polkinghorne, who was my internal guy, we all three were going to be at the conference in Trieste at the same time. So, it was arranged that we should do the oral exam in an office in the International Centre for Theoretical Physics there, in one of the conference lunch breaks. 

Zierler:

What postdoc opportunities were most compelling to you at this point?

Ellis:

So, I definitely wanted to go to the US, so I applied to a whole bunch of places. Harvard, MIT, Princeton, probably Berkeley, I don't remember, and SLAC. 

Zierler:

Mostly by reputation, or because there were particular people you wanted to work with?

Ellis:

Mostly because these were places where people were doing the sort of physics that I was interested in. I would say, particularly, the structure of hadrons that were being studied at SLAC, for example, with Bjorken and the deep inelastic scattering experiments. Let's see, where was Weinberg at that point? He'd gone from Berkeley—I think he was at Harvard at that time. Anyway, that was basically the connecting thing. 

Zierler:

What did you ultimately choose?

Ellis:

SLAC. So, actually, that was largely on the advice of Bruno Zumino rather than anybody else. In fact, I did hesitate whether to take up a postdoc appointment in the States at all, because after I'd made my applications, and after I started getting offers, I discovered that Murray Gell-Mann was coming to CERN on sabbatical the following academic year. I did inquire whether it would be possible or a good idea for me to stay at CERN as a postdoc. But Bruno Zumino was clearly pushing me out of the nest.

Zierler:

Had you ever been to the States before?

Ellis:

Yes. I went to the States for the first time in the summer of 1967. Stop me if I digress too much, but back in those days, when I was an undergraduate, I hadn't yet really focused on particle physics. My mental choice between particle physics and cosmology really came a year or so later. So, I had worked the previous summer, gotten an internship with a British aerospace company working on the optimization of rocket trajectories for satellite launches. So, in the following summer of 1967, I thought I would like to go work for some American aerospace company, and I got an offer from Boeing to go work in Huntsville, Alabama. So, I got ready to go and then they wrote to me and said, “Oops, sorry. We just realized we need to get security clearance and there's no way we're going to be able to do that.” So, there I was, two weeks before going to US and no work. 

Zierler:

Were they surprised that you were a Brit? Was that the issue?

Ellis:

I think it was just an unusual case. They hadn't been through all the steps, and at some point, some H.R. guy said, “What? You can't employ this guy.” Anyway, so through nepotism, my father arranged for me to work in a branch of the insurance company that he worked for in Boston. So, my introduction to the US was going to stay with some friends of the family in New York, and then going to Boston. It was a rather informative experience. The first evening I was in Boston, I went for a walk by the Charles River, and there was a concert by the Boston Pops. I come back to the hostel where I was staying, and there was a guy there drinking beers, watching the TV, baseball. He said, “Where'd you go?” And I said, “Oh, I went to the Boston Pops concert.” And he said, “Oh, I went by there but there were mixed race couples. I don't know about you, but I don't follow that sort of immorality.”

Zierler:

Welcome to Boston.

Ellis:

Well, I think this guy had been from somewhere else in the US. But anyway, welcome to the US in any case. So, while I had been in New York, the guy in whose house I was staying explained that he had a daughter who was working for Volunteers in Service to America as a teacher in West Virginia. 1970, when I went back to US for the Brandeis summer institute, I was going to stay at their place again. But a few months before I was scheduled to go, I read in the newspaper about how some house in Greenwich Village had blown up. It had a Weatherman bomb factory in the basement, and something had gone wrong. A few days later they gave the address, and it was the house that I had stayed in three years previously. So, the daughter had come back. She was radicalized into setting up the Weatherman bomb factory, she survived the blast, and then she was on the FBI's most wanted list for many years until she gave herself up. 

Zierler:

Back to your postdoc, what were your initial impressions of SLAC when you arrived?

Ellis:

It was great. Bjorken was definitely at the height of his powers. Of course, this is after the heyday of scaling and deep inelastic scattering. That was a couple of years old by that time. And the e+ e- experiments had not yet gotten going. That came a little bit later. So, in some sense, it was a sort of lull between two peaks. But still, it was a great place to be. Sid Drell was very friendly, Stan Brodsky was very friendly, also Fred Gilman. So, it was a good place to be. 

Zierler:

Did you meet Feynman on one of his periodic visits to SLAC at that point?

Ellis:

I don't remember meeting him. I think his legendary visits to SLAC were a few years previously. I came into contact with Feynman later on. I just spent one year as a postdoc at SLAC, and I spent one year at Caltech before returning to Europe.

Zierler:

Did you interact at all with Dick Taylor at SLAC?

Ellis:

No. I didn't at that stage interact very much with experimentalists. I became friends later on with Dick Taylor, his wife got me to act in a Gilbert and Sullivan production, but not at that time. Name dropping Nobel Prize winners. 

Zierler:

That's enough of the senior people, John. What about your peers? Who were some of the other postdocs who were with you at that time?

Ellis:

The guy that I was most friendly with, and we collaborated, was Mike Chanowitz, who went on to become—is now a scientist at Berkeley. We found common interest in scale invariance. One of the things that I did back then was, together with Mike, we realized that in a theory with scale invariance, you would have an anomaly from an analogue of the axial triangle anomaly of Adler, Bell and Jackiw.

So, this is another anomaly, which if there was a scalar boson, would give decays of that scalar boson into two photons, which was actually something that Schwinger had calculated, I think it was in 1951. But later on, we realized it was an anomaly. In some sense, it is the prototype of the major scale anomaly that is proportional to the beta function of QCD. But we didn't know at that stage because QCD wasn't on the market yet. One of my regrets is that having discovered this anomaly, somehow, we didn't push it through as much as we might have. I’d also like to mention Bob Jaffe (we puzzled together over strange quarks in the proton) and Inga Karliner (we later proposed how to determine the gluon spin experimentally in e+e- annihilation), who were students when I was a postdoc at SLAC.

Zierler:

Another thing that was not yet on the market, but I wonder, looking back, if you saw any evidence to foreshadow the discovery of the J/Psi. 

Ellis:

No. No.

Zierler:

Total surprise. Nobody was thinking about this at SLAC at this point. 

Ellis:

Right. So, the year that I was at SLAC was '71-'72. The discovery of the J/Psi was of course in the autumn of '74. So, I'm trying to remember when e+ e- collisions started at SLAC, but let's say 1973, pretty much. So, it really wasn't getting going. There was some sort of awareness growing from '73 that the rate of the annihilations of e+ e- to hadrons was larger than we expected from the parton model. So, there'd been experiments at Harvard that I guess were the first to discover some excess. In fact, together with Bob Cahn I wrote a review of that in 1973. But there was no hint of any structure. And even Burt Richter was talking about the idea that maybe electrons and positrons were strongly interacting so that the cross section, so instead of going down like 1/E-squared it'd actually be more constant, like a hadronic cross section. I'm not sure whether he ever wrote a paper about it, but he was certainly talking about it. 

Zierler:

Now, you were at SLAC for a relatively short period of time. Did you cut a longer appointment short to go to Caltech, or it was built to be a shorter appointment?

Ellis:

No, no. I cut it short because, as I mentioned, Gell-Mann was still my hero at that point. He's still my hero, but anyway, he was my hero, or one of my heroes at that point. And after his sabbatical at CERN in the academic year '71-'72, he was coming back to Caltech for '72-'73, so I figured I'd go there.

Zierler:

Did you first meet Gell-Mann at Caltech, or had you known him previously?

Ellis:

I had met him previously. In particular, at a conference at Coral Gables in 1971. So, I got an invitation to speak at this Coral Gables conference, which was kind of exclusive. Sidney Cohen, Ken Wilson, Murray Gell-Mann, etc. And I got an invitation to speak while I was a research student, and I didn't have a travel budget. There was a little bit of bickering and backwards and forwards at CERN as to whether they could stump up the money for me to go to this conference in Coral Gables. Anyway, finally, they allowed me to go and talk about my work on scale invariance, and that was when I met them. 

Zierler:

I wonder what ideas he might have come fresh from CERN back to Caltech with. 

Ellis:

So, at that point, he was interested in a couple of things. One of them was the so-called light cone algebra, which was a generalization or extension of current algebra, which was one formulation of chiral symmetry. And the light cone algebra, there were applications in deep inelastic scattering. So, it was a way of linking together low-energy chiral symmetry and high-energy deep inelastic scattering. He was particularly interested in the possibility that you might extract light cone algebra from the properties of quarks. So, Gell-Mann, you should remember, at that stage, was with the schtick of extracting properties of the strong interactions from quarks, but not necessarily, in fact, perhaps not, in the physical reality of quarks as such. So, that was one of the things that he was interested in.

And the other thing which he was interested in, and I'm not sure if I got the time sequence exactly straight, was QCD. And in fact, he has claimed to be one of the founding fathers of QCD. It's certainly true that he had been doing a lot of thinking about it, but a lot of elements were out there. On the other hand, for me, there's no doubt that the crucial breakthrough in terms of establishing QCD as a means of strong interactions was done by the people who discovered asymptotic freedom. 

Zierler:

When asymptotic freedom hit, do you remember, were you at Caltech, or were you at CERN at that point?

Ellis:

Let's see, that was early 1973, so I was at Caltech at that point.

Zierler:

You were at Caltech. And did asymptotic freedom immediately resonate? Was it understood to be revolutionary right from the beginning?

Ellis:

I would say, yes. Although, I don't think that I really understood it as well or as deeply as I could have or should have done. I think that if I look back at my early career, one of the things that I regret is that I did not do enough field theory. Specifically, I think it was a little bit of a mystery to me still, at that point, asymptotic freedom. I wasn't in a position where I could sit down and go away and calculate the diagrams and verify the beta function of the coupling. 

Zierler:

Besides Murray, who else did you interact with on a sustained basis at Caltech?

Ellis:

So, there was one young guy who was also at Caltech at that time called Rich Brower, whom I knew from CERN, as a student. Other than that, not so many people. I continued working with Mike Chanowitz at SLAC, and also, I did a research project with Yitzhak Frishman, who was visiting at SLAC. In general, I collaborated with people at SLAC more than I did at Caltech. 

Zierler:

John, just as with SLAC, your time at Caltech was relatively short. Had you planned to stay there longer than you did, or was there an opportunity at CERN that you wanted to take as soon as possible?

Ellis:

I had vowed to myself that I would not spend more than two years in the United States. So, I packed as much as I could into those two years and then my plan was to go back to Europe.

Zierler:

Why limit yourself to only two years at the outset?

Ellis:

Well, again, I have to say, I'm European. The US is a great place. There are lots of wonderful people. But it is not my place. I think that I had this idea that I wanted to contribute to the extent that I could to the building up of particle physics in Europe. 

Zierler:

And that, of course, meant CERN. 

Ellis:

Well, initially, the option would have been to return to Cambridge because they had given me a fellowship. I could have gone back to Cambridge for a few years. They deferred it for two years when I went to the US, and then I wrote back to them after I decided that I should go to CERN instead. So, I wrote to them that I resigned my fellowship. And they said, “That's a shame. But actually, according to our records, you never formally became a fellow because you never actually came to Cambridge for the kissing of hands ceremony in the College chapel.” So, they asked me whether I could please drop by Cambridge on my way from the US to CERN and go through the ceremony so that I could become a fellow.

Zierler:

Was your initial appointment at CERN as a postdoc, or was it a full staff position?

Ellis:

Postdoc. 

Zierler:

What group did you join? What was available for postdocs at that point?

Ellis:

So, the theory division, as it was called then—the name changed within a few years. Anyway, the theory group was not subdivided. It was just a big group with typically 100 names on the telephone list, and 15-20 staff members. Maybe half of them were long-term people with indefinite contracts, and half of them were on fixed term contracts. 

Zierler:

Now, as you say, at SLAC, you didn't have much contact with the experimentalists. Was that also true, at least initially, at CERN?

Ellis:

Yeah, I'm afraid so. I'm afraid so. I'm afraid so. 

Zierler:

To the extent you were aware, what were some of the most exciting experiments that were happening at CERN at that point?

Ellis:

There was the Gargamelle experiment that had discovered neutral currents in 1973, which also found evidence for the quark quantum numbers of partons by comparing the scattered cross sections of neutrinos with those of electrons. That was certainly the main exciting experiment. 

Zierler:

When did you first meet Mary K. Gaillard? Was it at CERN at this point? 

Ellis:

So, she was not at CERN in the academic year '73-'74, unless I've got my dates wrong. That was the year that she was at Fermilab working with Ben Lee. When she came back in mid '74, that's when we met. 

Zierler:

When did you start working with her?

Ellis:

Well, pretty much immediately. 

Zierler:

What was she working on at that point? What was the starting point of your collaboration?

Ellis:

She had just written this big paper with Ben Lee on charm. I'm sorry, I'm feeling a little confused. So, I came in '73-'74. Maybe during that year, she was at CERN, but I didn't have any contact with her. Then, '74-'75, I think that was probably when she was at Fermilab. Let me just check and continue with the story. No, I was right. So, I came in '73. In academic year '73-'74, she was at Fermilab. So, it would have been summer of 1974, I was actually asked to give some sort of review talk on e+ e- based on hadrons, and in preparing that talk, I got myself convinced that charm was the only sensible explanation. Then in the autumn of 1974, the J/Psi was discovered.

So, I guess I must have had first contact with Mary K. that autumn when we were discussing the possible interpretation of the J/Psi particle. I forget now, it was probably 1975 that we had a big debate in the main CERN auditorium about the interpretation of the new particles, and Paul Matthews was invited to talk about the possibility it might be color particles. I was asked to talk about charm, and I remember Paul Matthews characterizing my point of view as the establishment point of view, which I found a little strange.

Anyway, so in the first year that I was there, Mary K. was not, and the second year, we had interactions through charm particles, and then it was sort of mid 1975 when we started working together. We worked, in particular, on this paper on the semileptonic decays of charm. As she commented in her interview, I suggested, “Well, couldn't we do something in relation to the charm, because charm particles are heavier, and because asymptotic freedom and the strong interactions are weaker, so you get some sort of enhancement, but it will not be as strong as it is for strange particles.” That could help explain why it was that experiments at SLAC were not seeing lots of strange particles in hadronic final states above the charm threshold, and more leptons. 

Zierler:

John, how closely were you following the work at Harvard with Georgi and Glashow on grand unified theories?

Ellis:

Not so closely at that stage. We got interested in a little bit later, and that was about 1977 when I first got interested in grand unified theories. 

Zierler:

And you were still in contact with Mike Chanowitz during this time?

Ellis:

So, Mike came to CERN for a few months' sabbatical, and we were looking around for something to do. We decided to get interested in what other group structures besides Glashow's SU(2) x U(1) would be compatible with the natural conservation of flavor in a neutral current weak interaction. So, we wrote a paper with Mary Gaillard on that subject. While we were doing that, we found that the gauge groups that we were led to were the ones that fit inside grand unified theories. In particular, in the simplest grand unified theories, we realized that the bottom quark and the tau lepton should have the same mass. I think that it was me who first observed that their ratio happens to be 1 in the simplest grand unified theory.

So, I remember thinking, ah shit, because the bottom quark had not been discovered. It had to be heavier than the tau lepton. And about five minutes later, I thought, ah un-shit, because there was a calculable renormalization of the bottom quark mass through the strong interactions. The quick estimate was that would increase the bottom quark mass by about 2-5, compared to the tau lepton. So, we put out our preprint, we sent it to the journal, it was accepted, and then the bottom quark was discovered. The bottom quark was discovered with a mass about 3 times that of the tau lepton. So, we got back the proofs, and we looked at the proofs, and we had written that our prediction for the bottom quark mass was about 2-5 times the tau lepton mass somewhere in the text, but it wasn't in the abstract. So, I handwrote in the abstract our prediction for the bottom quark mass. Number 2, word "to," 5 times the tau lepton mass. And this was transcribed in the final version of the paper as 2,605 times the tau lepton mass. The typesetter couldn't read my handwriting, so that paper has what must be the least successful prediction for the mass. 

Zierler:

John, where does Melissa Franklin enter into all of this?

Ellis:

Let's see. At the moment, we're in 1977, right? Right, so she was at CERN. I'm not sure in what capacity, whether she was on an internship or whether she was a PhD student. Probably an internship. So, there's a group of us who got friendly, friends and collaborators at CERN. And after the bottom quark was discovered, then Serge Rudaz, Mary K., Dimitri Nanopoulos, and I got together writing the paper about the phenomenological properties of the bottom quark and the top quark. Phenomenology of the next left-handed quarks, or something like that. So, we were working on this paper, and one evening I was on my way home from CERN to my flat in central Geneva, and I stopped off to visit some friends. There, I gained inspiration in a way that my wife refuses to allow me to describe, but you can find it if you Google “Penguin Diagram”, and I went back to my bachelor pad, and I continued working on the bottom quark paper, and I had this inspiration.

This diagram that we were writing down, which is important in bottom quark decays, looked like a penguin, if you distort the lines in the Feynman diagram sufficiently, or maybe it doesn't. But, of course, I have missed out the crucial step, which was that on a previous evening, Serge, Melissa, and I had gone downtown to a pub where we had embarked on a game of darts. Melissa had bet me that if I lost the game of darts, then I should put the word penguin into my next paper. So, about halfway through the game, she actually left it and Serge finished off her side of the match. And Serge won – but the bet was not if she won. The bet was if I lost, I should put the word penguin into my next paper. So, I got the wrong sequence, but you can put it together in the right order.

Zierler:

How did this famous story actually move physics forward, do you think? What was the science behind this?

Ellis:

Let us think back when we were talking earlier on about the enhancement of hadronic kaon decays and how the analog of that is weaker for heavy-quark hadrons in general. That discussion was in the context of, if you like, the strong interaction modification of Fermi's four-fermion theory, where you write down the basic interaction which looks like four fermions interacting at a point, and then you dress the diagrams to include the influence of gluons. Well, in fact, some Russians, several years previously, argued that hadronic loop diagrams would also be important. And then when the bottom quark was discovered, then we argued that similar diagrams would also be important for bottom quarks. Although, the main impact, as it turns out, would be different. For example, all this excitement about flavor-changing bottom decays, anomalies in B goes to S plus di-leptons and so on, it all has to do with penguin diagrams. Anyway, so the diagrams had been highlighted by the Russians, but they didn't have a catchy name for them. They just came along with a conventional name for them. 

Zierler:

You said initially, just like at SLAC, that you were not so connected to the world of experiment at CERN. Did that change at any point, or was your world even at CERN mostly in the theoretical realm?

Ellis:

No, I would say, very quickly it got entangled with the experimental side. For example, relatively early during my career there, I was put on one of the committees reviewing experiments on the SPS accelerator at CERN, to the extent of proposing an experiment which was subsequently done. That began in '74-'75. Then in ‘75-’76 Burt Richter spent his sabbatical at CERN. That was when he wrote a paper on scaling laws for large e+ e- colliders, which then provided the stimulus for the LEP project at CERN. And Mary K. and I were commissioned to write a summary of theoretical physics for this accelerator. So, things like electroweak measurements at the Z peak, e+ e- to W pairs, searches for new particles, and of course, the search for the Higgs boson. 

Zierler:

We talked about it in sort of broad terms at the beginning of our conversation, but historically, where in the chronology do you start thinking about supersymmetry, and particularly about supersymmetry and its relevance in the search for dark matter?

Ellis:

Do you mind if we hold on to that for a little while?

Zierler:

Sure.

Ellis:

I just want to mention some of the other papers from that period, which I am particularly proud of. 

Zierler:

Please.

Ellis:

So, one was a paper that I wrote with Mary K. and Graham Ross called “Search for gluons in e+ e- Annihilation.” That was where we proposed looking for three jet events. That was the first search directly for gluons. There was indirect evidence for gluons from deep inelastic scattering, but there was no direct evidence, and this was the simplest process that we or anybody else could think of that would give direct evidence of the existence of gluons as physical objects. That was an idea which I had one day when I was walking back to my office from the CERN cafeteria. I can tell you, within five meters, exactly where the idea struck me.

Zierler:

Please.

Ellis:

I don't know if you know the layout of CERN.

Zierler:

A little bit. I've heard. I've never visited though. 

Ellis:

Well, when you come, I'll show you. So, between the theory department and the cafeteria, there is a bridge that one has to cross. I was just walking back from the cafeteria, and I had just come to the end of that bridge, and I was just about to turn into the theory department, and I thought, ah!

Zierler:

That's it. 

Ellis:

That's it. So, that's one thing that I'm proud of. Another one, of course, is the “Phenomenological Profile of the Higgs Boson” that Mary K. and I wrote with Dimitri Nanopoulos, which as Mary K. has mentioned pioneered amongst other things the calculation of Higgs decay into two photons, which was one of the discovery modes at the LHC many, many years later. We also wrote a paper on CP violation where we pointed out that the Kobayashi-Maskawa model gives predictions which deviate from the super weak model, which was at that time the sort of go-to model for CP violation and interactions. We gave an estimate of how large the direct CP violation in K to 2 pion decays could have been. That bound was saturated by experiments 20 years later. 

Zierler:

John, a question, given how foundational your collaboration with Mary K. was, what was the source of the magic? What was the collaboration that made your work with Mary so productive, in terms of your sensibilities and what you brought to it, and what she brought to it?

Ellis:

So, here again, I think it's kind of similar to what Mary K. said in her interview with you. But I was coming with ideas and some of them were crap and some of them weren't so crap. It's certainly true that she was the heavy-duty calculator, although I made sure that the important ones I did independently. Calculating e+ e- into quark-antiquark + gluon for example. But I think there was this combination of ideas, and the meat to put on the skeleton. 

Zierler:

Did you pay much mind to the fact that Mary K., as a woman, was such a pathbreaker in this environment, that there really weren't many women operating at her level?

Ellis:

I guess, I was aware of that, but to me, it seemed perfectly normal that a woman should be doing top rate theoretical physics. It didn't seem strange. I did see something strange in the fact that my colleagues at CERN didn't seem to give her the credit and the support that she deserved. Together with Bruno Zumino, we did push, as strong as we knew how, to get her a permanent position at CERN but that did not work out, and she left for the US.

Zierler:

Yeah, yeah. Are we ready to move on to supersymmetry, or were there other papers that you wanted to discuss beforehand?

Ellis:

We sort of touched on the subject of grand unification, so probably we shouldn't entirely miss the second paper that we wrote on grand unification. It's got over a thousand citations so some people must have found it interesting. 

Zierler:

Please. 

Ellis:

I'm not going to discuss the paper in detail. It's just basically building on what had been done before, but it got a lot of attention. In some sense, the paper with Mike Chanowitz broke more new ground. It was also around the end of the 1970s that we first wrote papers on particle cosmology. So, Mary K., and Dimitri, and I wrote a paper on the baryon asymmetry of the universe, and I wrote a paper with Gary Steigman on supermassive particle physics, top, and constraints from cosmic rays. 

Zierler:

On the topic of grand unified theories, and later, what you would call the theory of everything, I wonder if you might address some of the misperceptions around both of these terms and the ways that grand unified theories and the so-called theory of everything are and are not related. 

Ellis:

Okay. So, I think that we actually coined the term grand unified theories. We, meaning either myself or Dimitri, but we didn't have the guts to actually—call them GUTs—I think they do more or less as it says on the tin, right? To unify all the fundamental interactions in the simple gauge group. So, I think it's probably fine to call them grand unified theories. 

Zierler:

However, where is gravity in this? Without being able to incorporate gravity into the Standard Model, how grand of a unified theory could it be?

Ellis:

Oh, well, that's when you get the theory of everything, right? So, back in the late 1970s when we coined the term GUT, people were not really thinking about combining gravity with the other fundamental particle interactions. 

Zierler:

Why not? What was holding them back from that—putting it all together?

Ellis:

I think it was just a step too far. People were still digesting the Standard Model, and SU(2) x U(1). SU(5) and SO(10) were kind of out there in terms of being quite speculative. And gravity just seemed to be completely disconnected. So, the whole program of combining gravity with the other fundamental particle interactions really only came onto the scene in the 1980s, in particular, with Green, Schwarz and Witten in 1984. 

Zierler:

So, string theory, of course, is central to these developments, as you say.

Ellis:

Yeah. So, there have been other proposals that are possible theory of everything, including one that we built together with Mary K., and Bruno Zumino, and to some extent, Luciano Maiani, based on N=8 supergravity. But the only serious attempt is string theory. By the way, a small comment. I actually think the term theory of everything was invented by journalists, and I'm perhaps responsible for popularizing it and bringing it into the physics jargon. In particular, in a news article that I wrote for Nature, where, if I'm not mistaken, the title was “String Theory: Theory of Everything or of Nothing?” 

Zierler:

It could go both ways. 

Ellis:

It could go both ways. For me, theory of everything, right from the get-go was sort of an ironic title. People criticized it because it was not a theory of everything. It was not explaining superconductivity. That, of course, was not a claim we were making. We didn't really understand the ways the term would be understood. 

Zierler:

So, as you say, at least in the late 1970s, merging general relativity and quantum mechanics was a step too far, but what were some of the developments moving forward into the 1980s that may have seemed to make this more realistic?

Ellis:

So, I might be just modifying what you just said. So, string theory provided a possible framework for unifying gravity with the other fundamental particle interactions. But you said, gravity and quantum mechanics. Now, that's a somewhat different story, and it was Stephen Hawking, with his work on the quantum properties of black holes, who developed very much the idea that there might be some sort of fundamental contradiction between quantum mechanics as we understand it, and general relativity. So, this comes out with the name of the great information paradox, and so on. There's also something I have speculated about on occasion—I think for the first time, that was in 1983. But if you want to stick to chronological order, we should go back to supersymmetry. 

Zierler:

Yes. So, my question there, of course, is to come back to experimentation, in the early 1980s, what ideas were there that supersymmetry would be detected?

Ellis:

So, of course, people thought that supersymmetric particles might be light enough to be produced in particle collisions at accessible energies in e+ e- collisions, or hadron-hadron collisions. I guess it must have been in 1984 that the UA1 collaboration discovered some events with missing energy, which was a possible signature of supersymmetry. In fact, we wrote a paper proposing that interpretation, which turned out to be bullshit. But that was certainly an idea which was out there in the market starting around 1982, or 1983, when we first proposed the missing-energy collider signatures in supersymmetry was probably in '82 or '83.

Zierler:

At this time, were you following the developments that would lead rather quickly to planning for the SSC?

Ellis:

So, the SSC of course was formally proposed in 1983. I was actually at SLAC on sabbatical in the academic year of '82-'83. So, during that time, I was working on various aspects of supersymmetry phenomenology. For example, that's when I wrote my first paper on the supersymmetric contribution to g-2. And it was also during that period that I wrote my first paper about supersymmetric dark matter. Anyway, so there was this committee which proposed to drop the ISABELLE collider at Brookhaven, and instead, go with the SSC. During its deliberations, the committee made a tour of various national labs. In particular, SLAC. And then Burt Richter asked me to make a presentation about the physics you can do with high energy linear e+ e- collisions. That was in 1983. My transparencies are probably in the basement somewhere, and I'm not sure whether I talked about supersymmetry. I can't imagine that I didn't talk about it. 

Zierler:

When did you start to think about supersymmetry and its value for the search for dark matter? Was that right at the beginning, or did that come later on?

Ellis:

Very good question. Very good question. When was the first time I wrote a paper about detection of supersymmetric dark matter? So, the first paper we wrote in '83, and published in '84, we gave calculations of the cosmological relic density. Already in that paper, we had some discussion of the scattering cross section on regular matter. No, maybe not. I think we were making a non-relativistic expansion of the annihilation cross section, not for the scattering cross section. I'm afraid I may have to plead the fifth on that one. The first paper on direct detection of dark matter in scattering was, of course, written by Goodman and Witten. I think that was in 1984. 

Zierler:

Yeah. And while we're in that time, I wonder if you could provide your perspective on the so-called superstring revolution of 1984. 

Ellis:

Okay. Before I answer that, I've come up with the answer to your question, which was, if I'm not mistaken, 1987.

Zierler:

Dark matter. 

Ellis:

Yeah. Yeah, that was '87. 

Zierler:

So, there is a bit of a lag between connecting supersymmetry to dark matter. 

Ellis:

Well, the connection between supersymmetry and dark matter was there from the beginning. That dates back to '83. But the direct searches for supersymmetric dark matter particles, that I really got interested in, like I said, in '87. 

Zierler:

So, back to the string revolution of 1984.

Ellis:

Yeah. So, that was something that really got into my head in Argentina. In Argentina, it was a summer school, because it was in January of 1985. I remember Sergio Ferrara was there also. I guess it was talking with him that I realized the significance of these papers by Green and Schwarz. Then, when I went back to CERN --

Zierler:

What was the significance? What did you see that was so special right from the beginning?

Ellis:

Well, because they showed how you could incorporate the violation of parity, a crucial aspect of the Standard Model, in string theory. There didn't have to be left-right symmetry. 

Zierler:

Did you think that you would be a string theorist at this point? Was there excitement to the extent that this was where everything was headed and should be headed?

Ellis:

Now I wasn't about to become a string theorist. There were a thousand much smarter, technically, more adept people than I. It was for them to develop string theory. What I was interested in was string phenomenology, and what you could do to derive something like the Standard Model on the basis of string theory, and whether there might be some specific phenomenological signatures of string. So, that was something that I worked in the 1980s. In particular, well, there were various different aspects of it, for example we derived this modified SU(5) grand unified theory which was a rather nice framework for unifying all the other particle interactions from the string theory that would actually be part of the theory of everything. 

Zierler:

To return briefly to dark matter, with the excitement potentially of supersymmetry as a dark matter candidate, at the time, circa the mid-1980s, what were some of the other discussions? What were some other possible candidates for dark matter at this time?

Ellis:

Well, of course, axions were around. I got interested in the constraints that you could impose on axions from supernova explosions, in particular, Supernova 1987a. That was the origin of my interest in supernovae which is another story which I would like not to forget while we're moving along. So, of course, there were ideas for looking for dark matter through annihilations in the galactic halo to produce antiprotons and gamma rays, and the capture of dark matter particles by the sun, which might produce energetic solar neutrinos. These are things that I, amongst others, was interested in in the latter part of the 1980s.

Zierler:

At the time, I wonder if you can convey how much optimism there was, given that—we're talking about 1983. We're coming on nearly 40 years still not understanding dark matter. Was there optimism at the time that dark matter would be understood in the short term, or was there a recognition that this would be a long-scale problem?

Ellis:

I'm not sure when people became 100% convinced that dark matter was a reality. Many people might say that conviction only really came in the 1990s. Although, obviously, the key evidence had already been produced in the 1970s by Vera Rubin and Ford, and there were many other indications pointing in the direction of dark matter. I remember that there was this observation of X-rays from gas in clusters which was one of the key pieces of evidence, which came in the 1990s. Then, of course, there were the cosmic microwave background radiation observations that followed on. So, I would say that I only became 100% convinced sometimes in the 1990s. In the 1980s, I was probably only two thirds convinced. 

Zierler:

On the administrative side, what were some of the things that you were doing at CERN that were not purely in a scientific capacity, as a leader of the theory group, for example?

Ellis:

Okay. Well, perhaps before we do the theory group, I mentioned earlier on that I was on this committee selecting experiments for the SPS 400 GeV accelerator at CERN. And then, in the early 1980s, I was put on the committee selecting experiments for LEP, and I was involved in that for quite a number of years, editing reports on physics that they could be doing at LEP, and so on. Mary K. and I had written a report in 1976. So, what's the job of the head of the theory division? I think, my interpretation is actually largely to be invisible. Perhaps I've said that wrong. I think what you have to do is to shield theorists from the sort of managerial, administrative stuff. They should be left free to get on with their theorizing. So, that's axiom number one. Axiom number two, at least in my mind, was to foster the relationship between theory and experiment at the, if you like, the shop floor level. Not just reviewing experimental proposals, but actually talking with them on a regular basis with experimentalists participating in discussions. Things like that.

I remember, during the 1980s, there was some criticism of the theory division that it was not enough engaged with the experimental program. My predecessor was personally interested in experiments, but not many of our colleagues were. To me, that was the most important, to get involved in the experimental core business in the laboratory as far as possible at all levels. And of course, there was the aspect that we should try to help in setting the scientific direction for the laboratory. So, the director general has got his advisory committees. In addition to selecting experiments, they obviously have to prioritize between different areas of research, and so on. So, I would sit on that type of committee and try to do the best that I could to provide advice to the director general, who for most of my time was Carlo Rubbia. Of course, it was during this stage that the idea of building the LHC came to the fore. So, I was involved in a lot of internal discussions about that, helping to write documents for council, and trying to convince them to support the LHC, which of course was a hard sell until the SSC was canceled. Then it was easier.

Zierler:

Specifically on that question, John, I'm curious how closely you and CERN generally was watching the developments, both at the moments of greatest optimism in the late 1980s that the SSC would go through, and what it meant for CERN when the SSC collapsed in '93-'94. 

Ellis:

Well, I think we were all following what was going on very closely. Quite a number of CERN people, not me, but quite a number of other CERN people were sitting on advisory committees for the SSC, for example. Personally, I was never convinced that the SSC would ever be built. I could just see all these problems of the choice to build it on basically a green field, though I'm not sure it was actually a green field, but it was a field anyway, a green field site, which was driven by political considerations. I still believe if it had been built at Fermilab, it might have been built. Much more likely to have been built. There was also the fact that the project was so large that the particle physics community in the US, I don't think, had the managerial and administrative skills and experience to bring it off.

So, they brought in people from outside, in particular, people with military and industrial experience. That was maybe not a good way to go. There was the controversy within the U.S. physics community. People like Phil Anderson criticizing the SSC. So, I was, like I said, never convinced it would have been built. I never argued against it. Not something that I would ever do. But I was not surprised, I have to say, when after the new president was elected who also didn't give a shit about Texas and the SSC that the thing went down. 

Zierler:

Now, even if you weren't surprised when the collapse was a reality, did that increase pressure at CERN that the LHC should be conceived to operate at higher energies than it otherwise would have, because absent the SSC, those energies simply would not have been available anywhere else?

Ellis:

I'm not too sure. I think, in some sense, the pressure might have been growing in the opposite direction. When the SSC was proposed, it was proposed with the circumference of, whatever it was, 89 miles, or kilometers. Kilometers probably. Anyway, much bigger than the LEP tunnel in which the LHC was going to be sited. To my mind, it was clear the idea was to build something that was so big and capable of getting to such high energies that CERN could never duplicate it. So, it was to remove CERN competition from the political equation. So, then, the idea at CERN was, okay, we've got limited real estate. We have to go with the strongest field magnets possible in order to get to the highest energy possible.

So, I actually participated in the first physics studies for the LHC back in 1984. I remember thinking about what to do with 20 TeV in the center of mass. Now, you could say, once the SSC moved out of the way, then you no longer have to bust a gut to get to the highest possible energy. Maybe just back off a little bit and be happy with a somewhat lower center of mass energy, but anyway it would be a lot higher than what you have with the Tevatron proton-antiproton collider. Maybe from a technological point of view that's somewhat of a conservative choice.

But in fact, basically, LHC was on the same trajectory as before the cancelation of the SSC. I should say, but the way, that I was quietly confident that whatever happened, CERN council would approve the LHC. Not 100% confident, but certainly more than 50% confident. Even before the SSC was canceled.

Zierler:

And to what extent did the theoretical guidance of where the Higgs boson would be, the mass of the Higgs, to what extent did those theories influence technical discussions about the energies that the LHC would be operating at?

Ellis:

I don't think we theorists really influenced it. I think the place where there was scope for influence was in the maximum energy of LEP. So, for the first days of LEP, went up to 90-95 GeV, and then there was this plan to put in superconducting RF to increase the central breadth mass energy to above the WW threshold. Technically, it would have been possible to push it to a center mass energy of 220 GeV. If that had been done, it would have found the Higgs boson. But that option was not pursued because people wanted to go ahead and build the LHC. I was not personally involved in those debates about pushing the center of mass energy at LEP. Maybe I should have been, but I wasn't. When I say, should have been, maybe I should have gotten myself involved, but I didn't.

Zierler:

And to return yet again to supersymmetry, post-LEP, what were some of your ideas with regards to benchmarks to supersymmetry?

Ellis:

So, as you know, I wrote a series of papers on the subject. In particular, with our current director general, amongst others.

Zierler:

That would be Fabiola Gianotti, of course. 

Ellis:

Yeah, right. So, we just thought it would be a good idea to help ATLAS and CMS experimentalists fix their ideas if we proposed some specific supersymmetric scenarios which were consistent with what we knew and looked at what would be the phenomenology of the signatures and discuss why we're not finding them. 

Zierler:

On the question of ATLAS and CMS, in terms of the scientific value, where did you see the opportunities with regard to competition, collaboration, and even redundancy? 

Ellis:

So, I think it's very important if you possibly can to have these two different sets of eyes looking at the same science. So, at LEP, we had four sets of eyes, because five would be too many. In fact, I mentioned that I was on the LEP experiment selection committee, and we said that three sets of eyes were enough, but the director general wanted four. So, there were four experiments. In the case of the LHC, I was on the LHC experiment selection committee, and basically, by the time the committee had to decide—no, that's not true. There were three ideas, and we opted for CMS and ATLAS because they were very different experimental general-purpose concepts using different technologies, different types of magnets. So, you could be pretty damn sure that even if there was some fatal flaw in one of the designs, or one of the technical realizations, that that would not kill the program.

Zierler:

It's a counterfactual, because there's no way of knowing in an alternate history, but is your sense that because there was this redundancy, because there was this competition, the Higgs was found sooner or earlier than it otherwise would have been?

Ellis:

I have no idea, but I sort of suspect, yes. I think it was kind of found as quickly as it could conceivably have been found. I think that if you had half as many people working on the problem from half as many different points of view without any competition, I'm sure it would have taken longer. 

Zierler:

John, what is a rough chronological sketch from when the excitement about the Higgs discovery really starts to build? When is it about really analyzing the data to make sure that this is what it says it is, and when is it when the collaboration says it's time to announce this? What years are we roughly talking about, and when is the excitement minimal, and when does it really start to build?

Ellis:

I would say it took about a year, from mid-2011 until mid-2012. Unless my memory is playing tricks on me. Mid-2011, at least on the corridors one started getting the impression that there's some indication that there's something there. And in late 2011, there definitely were some indications. But at that time, it wasn't so obvious what the mass was. For example, at that stage, it still seemed possible that the famous LEP Higgs candidate that had been discovered at 115 GeV might actually be at the right mass.

But then in the first half of 2012, the experiments clammed up. No more news came out, and then July the 4th, 125 GeV. So, I personally wasn't aware, wasn't following so much what the rumors might have been during those few months in 2012. I was never called into consultations for ATLAS or CMS, for example. Maybe some people were, but for me, it was just a great experience when we heard that there was going to be something special announced on July the 4th, because we knew there wasn't going to be a special announcement of nothing, so it was something to look forward to. 

Zierler:

Given the level of theoretical guidance prior to the discovery, what do you see in broad historical terms as the scientific value of discovering the Higgs, and where it actually was? 

Ellis:

Well, I think there were plenty of indication going back to the early 1990s in a way. We analyzed radiative corrections within the Standard Model, using precision electroweak data from LEP. They told you there had to be something like the Higgs. May not have been exactly the Higgs. It may have been a placeholder for something else. But it had to be something, and it had to be less than about 300 GeV. Now, if you actually think about that information, I think that told you something about the coupling of the Higgs to the W boson and the Z. It didn't tell you about couplings to quarks, for example.

So, even if you accepted those indications, and that confirmation in 2012 of the predictions from the early '90s, the discovery of the Higgs went beyond because it gave us a lot of additional information about the couplings of the Higgs boson confirming it wasn't just the couplings to the W and the Z, but not all the Standard Model couplings. But all the other couplings that have been measured so far at the LHC to fundamental fermions also look like the Standard Model. And the precise value of the Higgs mass is also interesting. So, I think that a lot of extra information has been provided. 

Zierler:

What new questions immediately were able to be raised post-discovery that may not have been prior to the discovery of the Higgs?

Ellis:

Okay. So, there's a couple of things which certainly I focused on after the discovery. So, July the 5th, 2012, some new particle has been discovered. No dispute in that. And it was discovered by searches that were targeting the Higgs boson. But was it the real Higgs boson or was it an imposter? So, a couple of things that you can do are to verify with its couplings to particles, bosons, but also fermions, are just as predicted in the Standard Model. One characteristic is that the couplings of fundamental fermions, for example, should be proportional to their masses, and there's a specific relationship between the coupling to fermions and the coupling to the W and the Z. So, these are two things that you want to test.

So, that was something that I set out to do with my then new PhD student, Tevong You. We wrote a few papers and proposed types of analysis that the experiments then took up much more completely and much more professionally. The question basically is, are the couplings of the Higgs boson to other particles proportional to their masses? The answer could have been no, but the answer is, as far as we can tell so far, yes.

So, another thing which you can check is, what is the spin and parity of this new particle? So, you cannot have spin 1 because of the Landau-Yang theorem. Could have spin 0, could have spin 2, could have spin bigger than 2, which people often forget. But let's focus on spin 0 and spin 2. So, then there are a whole bunch of statements you can make about the angular distribution of Higgs production and Higgs decays which are characteristic for a spin 0 particle with positive parity. And these were things that we and others developed. And when the data came in, we confirmed that the other hypotheses were strongly disfavored, and basically the only hypothesis left standing was that this new particle was a scalar.

Zierler:

To what extent did you envision these new questions being answered within the framework, the existing infrastructure of LHC, and to what extent would this require new particle accelerator projects within and beyond CERN?

Ellis:

So, I think the first round of answers to these questions could be provided by the LHC, I mean, were provided by the LHC basically within a year of the Higgs discovery. So, I think with the data taken by November 2012, we already knew that this is not a spin 2 particle, and we already had strong indication that the couplings were proportional to the mass. So, certainly, by mid-2013, that was established, and that was why the Nobel Prize committee, when they gave the prize in October 2013, said “without any reasonable doubt, it is a Higgs boson”. I should remember the precise quotation because it was taken from one of our papers. Beyond any reasonable doubt. Beyond any reasonable doubt.

So, there's another little anecdote about that because we wrote that in our paper, sent our paper off to the journal, and then the referee wrote back saying, "Beyond any reasonable doubt is not a scientific judgment. A legal statement is not a scientific judgment.” We had to change that phrase. So, that phrase does not appear in the published version of the paper. Somewhat ironic, it wasn't good enough for the referee, but it was good enough for the Nobel Prize committee.

Zierler:

That's great, that's great. John, while we're in the neighborhood of fundamental discovery post-2013, are you following LIGO closely at this point? Are you interested in gravitational waves specifically?

Ellis:

Yeah, sure. 

Zierler:

In what ways? What's fascinating to you about this?

Ellis:

Well, as I already mentioned, I'm certainly interested in the new astrophysical phenomena that they reveal. But I would say that my principal interest is in how we can use gravitational wave observations to probe fundamental physics. Either modifications of gravity, or possible physics beyond the Standard Model. In fact, very soon after the first LIGO paper appeared, we wrote a little paper where we pointed out how you could use the wave form of the first event, which I remind you, gave gravitational waves over a range of frequencies. Now, if the graviton had a mass, then that waveform would be modified because the velocities of waves of different frequencies would be affected in different ways by the graviton mass.

So, we wrote a paper pointing out that you could constrain the graviton mass in that way. We were certainly not the only people who were aware of that, and the LIGO collaboration also pointed out that you could constrain graviton mass. What you could also do, which at least initially the LIGO collaboration did not do, was to use that also to constrain the possible violation of Lorentz invariance. So, Lorentz invariance tells you that E squared equals p squared. Energy is equal to momentum. But you could imagine a theory in which there's an additional term in the dispersion relation so that E squared is equal to p square plus some additional piece depending differently on p. That would violate Lorentz invariance and one of my long-running interests has been probing Lorentz invariance using different distant astrophysical sources. We used the same idea to analyze LIGO gravitational wave data. 

Zierler:

We should go back in the chronology a little bit and return to supernovae, in the 1990s specifically. When did you get involved in this? 

Ellis:

Sorry, can I just continue on this LIGO train of thought, and then we can come back to the 1990s? 

Zierler:

Please.

Ellis:

So, other things that you can look for are possible stochastic gravitational wave backgrounds coming from processes in the very early universe like first order phase transitions. So, I've in the last few years been doing some work with my postdocs on that. And another intriguing thing was the report by the NANOGrav experiment, I guess it was in September of last year, of possible signature of gravitational waves. Do you know about this?

Zierler:

No. 

Ellis:

So, pulsars, as you know, emit a beam like a lighthouse, which is extremely collimated and regular. If a gravitational wave comes by, it can perturb those beams, or the timing of those beams. And you can use this effect—NANOGrav looked at 47 pulsars, and you can look to see whether there was any hint of some background gravitational waves which mess up the pulsar timings, and they said, "Yes, we have some evidence for some effect. Not strong evidence, but some evidence." So, the default interpretation of that would be astrophysical processes involving pairs of supermassive black holes.

However, an alternative possibility is cosmic strings. So, you've got these string-like defects wandering around in space. Every once in a while, they meet, and cross, and form loops. The loops then radiate gravitational waves until they disappear. And maybe that's what we were seeing. So, Marek Lewicki and I wrote a paper about that. Those are a couple more examples of fundamental physics that you can do with gravitational waves.

Okay, back to supernovae. So, I mentioned that back in the 1980s, I got interested in using supernova 1987a to constrain the properties of axions, and also to constrain the number of neutrinos. And somehow or other, I'm not sure exactly how it happened, but one of my principal axion supernova collaborators was Dave Schramm, a cosmologist who died unfortunately in a plane accident. So, we somehow got interested in the possibility that a nearby supernova explosion could have caused a mass extinction on Earth. We weren't the first people, but we had some new ideas how it might have happened.

And then, we got to thinking, together with a student of Dave Schramm's, Brian Fields, what happens if the supernova is not close enough to kill us off, but a little bit further away? In which case, an explosion would not cause a mass extinction on Earth, but it would potentially deposit radioactive isotopes on Earth. We did a detailed survey of different isotopes that might be produced in supernova explosions. So, we put out our paper. That was in '95-'96. Then, in '99, an experimental group from the technical university in Munich made the first report of seeing the radioactive iron-60 isotope in deep ocean deposits. And that has subsequently been confirmed, now there are lots of reports of iron-60 in ocean deposits, in Antarctic ice, in lunar rock, cosmic rays. There's radioactive iron-60 all over the place. The point about this is that the half-life of iron-60 is about 2.5 million years. So, if you see any, it's not a relic leftover from before the solar system was formed. It's something that could only come from a relatively recent explosion, and a supernova of the order of 100 parsecs away would produce these sorts of iron-60 deposits which had been seen.

So, that's been one of my hobbies over the last, I guess it's now 25 years. How time flies when you're having fun. Anyway, so recent developments are that there are some reports of measurements also of plutonium-244, which has a somewhat longer life of about 80 million years. And there is some observation—there was one observational claim of manganese-53. Now, the interesting thing about plutonium-244 is it would have a different astrophysical origin, probably. Supernovae are probably not very good at making plutonium-244, but a kilonova due to a merger of two neutron stars would be. So, this is maybe evidence for a kilonova within the last 50 million years or so. So, we're just in the final throes of writing a paper about that. In fact, I have a Zoom meeting to discuss that in 54 minutes' time. 

Zierler:

In what ways are you more or less confident about understanding dark energy than dark matter?

Ellis:

I guess, I'm not really very optimistic. It's just a number. As I said, the puzzle is not so much that it's non-zero. The puzzle is that it's so small. So, the example that we talked about, in the Lagrangian for the Higgs, earlier on, one would expect a contribution that is like 10⁶⁰ times bigger than what has actually been observed. So, there was a lot of surprise when supernova people reported evidence for dark energy back in the 1990s. I guess, in retrospect we shouldn't be surprised that there is dark energy, but we should wonder why it's so small. I have no good ideas about that. Supersymmetry would help.

So, the mass of the Higgs gets quadratically divergent corrections, which would be cancelled out by supersymmetry. Good. Dark energy gets quartically divergent corrections, lambda to the fourth power. Supersymmetry gets rid of that. That's good. If you choose a particular type of supersymmetry, then you can also get rid of the next corrections, which are lambda squared. That's also good, but you still don't get the thing down to the magnitude of milli-electron-volts to the fourth power, which is what we need. 

Zierler:

I wonder if you could talk about, more recently, the origins of the AION experiment in 2019. 

Ellis:

Oh, yeah. We still can't quite agree on the pronunciation. I call it AION [“A-on"]. So, my involvement in that originates from a Stammmitarbeiter of mine, Oliver Buchmueller. We did a lot of work on supersymmetry searches at the LHC back in the day. So, he got interested in atom interferometers as a way of looking for gravitational waves in the mid frequency band a little between maximum sensitivities of LIGO/VIRGO at higher frequencies that go to 100 hertz, and LISA which has a frequency of about 10^-2 hertz. So, there's a gap around 1 hertz.

So, there's a concept which to my knowledge was first proposed by people at Stanford, in particular, Peter Graham, Savas Dimopoulos, Jason Hogan, combining a couple of atom interferometers, making an atom gradiometer. So, a gravitational wave comes by, and you can imagine that the paths of the atoms in the interferometers is looking a bit like a hysteresis loop. So, a gravitational wave comes through, and it changes the relative phases of the guys going around different sides of the loop. But there is laser noise, and if you have two of them, and you subtract the signals of the two of them, you get rid of the common laser noise, and you're just left hopefully with the gravitational wave signal. That's the basic idea.

So, anyway, Oliver has been proposing that we do a similar experiment in the UK, and I'm providing theoretical advice. We've got some initial money to build the first 10m version of this thing, and we hope to have it built in, let's say, three years from now. In the meantime, the Stanford guys are building a 100-meter device at Fermilab. We think we're going to be able to build one which is technically somewhat superior, so it's a little bit complicated to compare their detector with our own experiment, but we hope to combine the two in a network much like LIGO/VIRGO. So, the “N” in AION stands for the network. So, the initial 10m AION device, I don't think we’ll have any interesting sensitivity to gravitational waves, although it might be able to detect some forms of ultralight dark matter. But at 100 meters, we may be able to test a gravitational wave signal. That's my principal interest in the project. 

Zierler:

Another question more on the administrative side, what were your decisions about going emeritus at CERN and taking the position at King's College?

Ellis:

Well, the choice of going emeritus at CERN was no choice. CERN has a fixed retirement age of 65. Yeah. And I really hadn't formulated any very specific plans. I wasn't going around looking for other positions. Then, my friend and collaborator, Nick Mavromatos, who was a professor at King's, said, “How about becoming Clerk Maxwell Professor at King’s?” So, I had to consult my wife, and that didn't take very long. Clerk Maxwell Professor, how could you possibly say no? Anyway, the other option was to go cold turkey into retirement. 

Zierler:

Was it also an opportunity to more teaching and supervising of graduate students that you may not have had for all of your time at CERN?

Ellis:

Yes, although actually I've only had one PhD student while I was at King’s. I think that I spent so long not working with graduate students that I don't think I'm very good at it. I think I'm much better with postdocs. So, I've had some good postdocs. Still have some good postdocs. And I like supervising undergraduate students for their project work. But graduate students are kind of a commitment, and at my stage in life, I cannot guarantee that I'm going to be around for the next four years, and able and willing to supervise a PhD student. 

Zierler:

What about undergraduate courses? What have been some of your favorite courses to teach?

Ellis:

I haven't given any. 

Zierler:

You haven't?

Ellis:

No, no. Mine is a pure research position. Apart from project work with third and fourth year students, but I've never had to do an undergraduate course. 

Zierler:

Of all of your visiting appointments and affiliations, I'm particularly interested in your affiliation with the T.D. Lee Institute in Shanghai, and what you see, perhaps, as China's long-term contributions in high energy physics. 

Ellis:

I think they have absolutely tremendous potential. What can I say? Lots and lots of people and very hardworking.

Zierler:

What about the appetite of the Chinese government to support basic science? Do you think that's there?

Ellis:

I believe they're shifting in that direction. I think the Chinese leadership is probably much more in tune to science and technology than the leadership in western countries. The Chinese leaders are not lawyers or buffoons in the way that they often are in the west. They often have training in engineering. So, in the past they've certainly demonstrated strong support for big engineering projects, but I think they're now moving over into big science projects. So, there's the FAST radio telescope, for example, synchrotron radiation facilities, a neutron spallation source, etc. So, so far in high energy physics, they haven't built a really big project, but I think there's a fighting chance that the high-energy physicists in China will get approval for their big CEPC electron-positron collider project.

Zierler:

I wonder how you might compare some of these projects in China with next generation projects that are being considered for CERN.

Ellis:

Right. So, obviously, CEPC is very much like the FCC-ee project that we have at CERN, which I've also been one of the promoters in the last few years. And in the long term, the Chinese would like also to put a hadron-hadron collider in the same tunnel, just as FCC-ee would be followed by a hadron-hadron collider in the same tunnel. So, yes, what they have in mind is very similar. But I think CERN, in principle, has greater expertise because we've built many large projects before. We've got a strong industrial base, and what I think is very important, we have a strong international collaboration network. If I were a Chinese physicist, one of the things that I would be concerned about is that the China government attitude is not friendly towards international collaboration.

For example, I'm on a review committee for a Chinese lab with somebody from Canada, and he said, “Under current circumstances, I am not prepared to go to China,” because the Chinese have basically kidnapped Canadian businessmen in a diplomatic spat, and I guess they're going to hang on to those businessmen until this dispute goes away, at least. And if I was Australian, I would also be a little bit nervous, because obviously relations between China and Australia are strained. I was going to say they haven't degenerated to the same stage as relations with Canada, but in fact they have, because China has arrested a Chinese-Australian news anchor. So, they've done a similar thing. And now, they are apparently insisting that any foreigner that comes to China be vaccinated with the Chinese vaccine, which is basically an impossible criterion to meet because you cannot get the Chinese vaccine in a western country. So, I think this is to my mind potentially a not-so-subtle way of keeping out foreigners, and that is not conducive to scientific progress. 

Zierler:

John, what about some of the experiments or observational projects in astrophysics right now? What's most compelling currently, and what's slated in the near future that you've been paying attention to?

Ellis:

So, obviously I'm interested in LISA, and there is also a follow-on project to AION that is space-based, called AEDGE that I'm also promoting. So, that's with gravitational waves. In gamma ray astronomy there is the CTA project, and I'm actually on the international technical review committee for CTA, and I’m very much interested in that kind of thing. Then there is the SKA project in radio astronomy. That would be great for probing this NANOGrav PTA signal for gravitational waves. SKA would have a much greater sensitivity, both in terms of the magnitude of the signal produced, and in terms of the frequency range. There's a whole bunch of other projects in astronomy, the LSST for example. I think observing light fluctuations is potentially a very interesting development. There may be a whole new universe of transit phenomena waiting to be seen. 

Zierler:

To bring our conversation from the 1980s all the way up to the present on string theory and string phenomenology, what's your sense of the state of play right now? Is string theory still doing exciting things? Are you still interested in where it's headed from here?

Ellis:

So, I have to confess that I haven't been following the details of string theory in recent years. I have been very impressed by some of the spin-offs of string theory, like holography for example, and its applications to condensed matter physics, heavy ion collisions, that all seem to be interesting. Although I've only got superficial knowledge of that. So, we've still got this tremendous problem which is a tremendous hierarchy between the mass values where we can do experiments and where string theory applies for sure.

Zierler:

And where does compactification play in all of this?

Ellis:

String compactification may be on a scale close to the Planck length, which was the original, default proposal. Then, we began working on effective field theory derived from that as it appears at much lower energies, and you're going to be looking for some trace of strings there. So, one thing is there may be some relationship between the parameters of the Standard Model particles, or maybe there are some characteristic additional particles that might appear, a characteristic prediction of string theory. So, that program is alive and worth pursuing. One of my current projects is related to some string inspired model of grand unification, as I mentioned earlier. But I'm not sure how promising that is. There's this question of reconciling gravity with quantum mechanics. So, it's something which I also committed myself to speculating on in the past. Haven't done so much recently.

Perhaps in the future, observations with gravitational waves could give us some indications of quantum effects. I haven't seen any convincing proposals of that yet, but that's something which is worth pursuing. Something else which I've been interested in in the past, something which you'd not expect in conventional quantum field theory, but maybe in quantum gravity theory you might have some sort of violation of Lorentz invariance, which is somehow a sacred symmetry of quantum field theory. So, there's another paper I wrote in the 1990s that has gathered over a thousand citations proposing tests of this possibility with astrophysical data. So, I have to say, that it's still very talked about. It’s quite speculative, but that doesn't mean to say it's not worth doing. It's definitely worth doing, but I'm not doing so much of it right now. 

Zierler:

And relatedly, what about phenomenology as it relates to quantum gravity? What's been compelling to you in recent years on that front?

Ellis:

I think one of the most fascinating avenues is the cosmic microwave background radiation. There was this excitement a few years ago with some sort of forgotten piece of mistaken evidence, the polarization in the cosmic microwave background radiation, which would actually come from quantum gravity effects. So, if that is something that you could actually observe, then I think maybe we can do something with that. It would also be nice if one could see evidence of some quantum phenomenon close to the horizon of the black hole. Steve Giddings is somebody who has been thinking about that, in particular. I had thought that maybe one could do something with observations by the Event Horizon Telescope, which would give us some sensitivity there, but it looks very difficult. I was very excited about the EHT observations when they first came out, but we haven't learned very much in the last couple of years, so I'm still waiting for some more results from that class of observation. Maybe with gravitational waves there's some effect we can look for, but we haven't seen anything convincing so far. 

Zierler:

John, just to bring our conversation right up to the present, in recent months, recent weeks, what are the things you're working on right now?

Ellis:

Oh my gosh --

Zierler:

Too much.

Ellis:

So, we're trying to finish off this paper on radioactive nuclei from a nearby supernova or kilonova explosion. I'm also working on some possible interpretations of the g-2 result. Also got a project on B decay anomalies that I’ve taken up with some colleagues at King's. We are continuing to work on gravitational wave phenomenology, in particular for AION and AEDGE, that's something we're also working on. Why are there so few hours in the day? What else can I say? I'm interested in searching for magnetic monopoles. I'm actually in the MoEDAL collaboration at CERN that's working on that, and just editing the paper on that later on today. 

Zierler:

More broadly, what you're saying is that for a physicist, emeritus does not mean retired. 

Ellis:

No, no. You've got some more freedom, right? You can, to some extent, choose the amount of administrative work that you choose to do, and pick and choose to some extent what fields you work on. Gravity, high energy astrophysics, particle physics, etc.

Zierler:

Well, John, for the last part of our talk, I'd like to ask a few broadly retrospective questions about your career, and then we'll end looking to the future. So, the first is, I know that science communication and public outreach is very important to you and it has been very important to you. What are some of your key motivations in reaching out to a broad audience to convey your science, and why what you and your colleagues do is important?

Ellis:

I think it's very important to have a scientifically literate general public. But there’s an even deeper question than that, given the current socio-political context. The public should base their opinions, their ideas, more on evidence. And they should question the sources of things that they read on the internet, for example. They shouldn't just believe somebody because that person shouts loud, and so on. And I think everybody has become much more conscious of the importance of that over the last few years. That's something that I certainly believe. So, I regard that somehow as being the bedrock of why I think scientific outreach is important. Scientific literacy in the next age, and familiarity, comfort with scientific ideas and the scientific approach in general.

Zierler:

Your literary output is legendary, with over 1,000 papers, and your literary impact is also legendary with your h-index. I'm curious what your feelings are on the scientific value of the h-index, or other rankings, as a measure not just in a horserace of who's making the most impact, but the value in this is the standard that actually moves the science forward. 

Ellis:

I think I'm going to have to say that I think it's essentially worthless. Great scientists' contributions are not always measured by the number of publications, or even by the number of citations that they get. Ken Wilson famously got tenure with basically no publications, but he's been enormously influential in a fantastic range of theoretical physics. So, I guess I can afford to be deeply cynical about the whole citation thing, and particularly about the h-index.

For example, if I'm looking at a possible recruitment of a postdoc or a prospective junior faculty member, of course I would look at the citations, of course I would look at the number of publications, but I will be looking for some sort of spark. Conversely, if some colleagues tells me, “Oh, this guy is really brilliant,” I would say “Just hang on a second. Is that brilliance reflected in the way the community reacts to what this person is producing?” So, I guess I shouldn't say that citations and the h-index are worthless, although the h-index is probably pretty worthless. But I would say they're just one element in how you think about the importance and the relevance of the influence of the scientist. 

Zierler:

To punctuate the point, is there a paper that stands out in your memory that's not as highly cited as some of your others, but that you personally think is significant?

Ellis:

Well, yeah. I would say this paper on how to discover the gluon. It's very strange. It's a fundamental part of QCD. It's a very important fundamental part of it. But I don't think that it's got the same degree of recognition as many other fundamental parts. The W and Z, the charm quark, or the Higgs boson. It just hasn't got the same degree of recognition. Our paper in particular hasn't got as many citations as I would have expected or, dare I say, as I would have hoped. And I think to a large extent that's because after we proposed it, when the thing was actually discovered, there was a big dogfight between experimental collaborations as to who should get the credit. Somehow everybody focused on that dogfight rather than the theoretical paper that actually proposed the thing in the first place. 

Zierler:

John, of all of the honors and awards that have been bestowed on you, is there any one in particular that's either most personally meaningful, or has been most scientifically useful in terms of the connections it has created, or the doors that it has opened?

Ellis:

I don't know. I think I would rather pass on that question. Of course, it's nice if somebody honors you in some way. My head of department liked the awards from the Institute of Physics and the Royal Society which many people regard as being most significant. That's nice. Then I got this prestigious thing called Commander of the British Empire, which is nice, too, but it's not in itself very significant. I'm not in it for the honors. I'm in it for the personal satisfaction, and the fact that I think that I can help move science along, and hopefully in addition to science in particular, humanity in general. I hope.

And it also motivates a lot of the sort of what you might call evangelical work that I've done in various different countries around the world. But that's a whole different story. For over a decade, I advised several CERN Directors-General on relations with countries around the world, as a sort of roving ambassador. I believe strongly in the universality of science, and its role as a bridge between different cultures. I have been particularly interested in building relations in the Middle East and Latin America, and am particularly glad that several countries where I initiated relations are now members of CERN.

Zierler:

So, on that question of personal satisfaction, perhaps this would be an easier question for you to answer. Of all of the things that you have worked on, all of your collaborations, is there any particular one that stands out as being most personally satisfying to you as either a Eureka moment, or something that wasn't understood but you were part of making it understood?

Ellis:

I think I'm going to have to say the gluon and the Higgs boson.

Zierler:

Why?

Ellis:

Even though they're now 45 years in the past. They opened up new areas of research, as well as helping confirm the existence of some key particles in the Standard Model. But there are other things which I like as well. For example, this supernova stuff. We actually proposed what turned out to be the first experimental evidence of some impact upon Earth of a nearby supernova explosion, or any astrophysical event outside the solar system. I think that's pretty cool.

Zierler:

Conversely, what has gnawed at you over the course of your career? In what ways have you theoretically always hit up against a wall and don't seem to find a way around it?

Ellis:

Well, I guess this whole issue that we've been talking about of string phenomenology, of quantum physics, and quantum gravity. Those are two things that still gnaw away at me. If I can see some possible breakthrough in that, then I would drop all these Overleaf documents that I've got on the screen behind you, and work on that. 

Zierler:

Do you see quantum computing specifically as providing breakthroughs that have not yet been available?

Ellis:

I can see that it has certain applications in experimental physics and in analysis. Already, some of my experimental colleagues are using quantum computing ideas. For example, in analyzing LHC data. I think more generally, not just in experimental physics, but potentially also in theoretical problems, quantum computers could speed things up. I don't know that it's going to revolutionize the field, or whether it's going to turn out to be a technique for accelerator analysis tools. I don't think it's going to help us theoretical physics discover the theory of everything. I don't think that we theorists are going to be put out of a job by quantum computers any time soon. 

Zierler:

John, for my last question, I'll ask something that will bridge past, present, and future, and that is, chronologically and generationally, of course, a theme of your career is you've been present at the creation of so many things, so many fundamental areas in theoretical physics. I think that it has encouraged you to have really an unmatched breadth in the kinds of physics that you've pursued.

To the extent that you interact with graduate students and postdocs, do you see your approach in a general study in theoretical physics that spans all kinds of disciplines? Is that the career model for the future, and if not, what might you offer in terms of advice on the most promising areas to specifically focus on?

Ellis:

I think that I would encourage students to try to go for as broad a background as possible. That doesn't necessarily mean to say work on a wide range of topics, because there's a danger of spreading yourself too thin. But certainly, in our group at King's, we work over a wide range of topics in theoretical physics, ranging from quantum field theory through to cosmology and astrophysics. But we are one group. We meet twice a week. Once the Journal Club, once the seminar. There are other more informal things with subgroup meetings, but those are things where everyone is expected to show up, and it's a mishmash of different topics. I think that's very beneficial for the students. They're not just exposed to just this type of calculation, this type of calculation, this type of calculation, over and over and over again.

Of course, the student has to find their own way, the things that turn them on, guided by their supervisor, within the sort of general framework. But I think this general framework is very important for flexibility in their subsequent careers. Maybe they're not going to stay in particle physics. Very likely, they're not. But if they do, then particle physics, or whatever the fundamental physics in cosmology, is going to change in the coming decades. You don't want to be left high and dry because of over-specialization. 

Zierler:

John, it's been a great pleasure spending this time with you. I'm deeply appreciative that we were able to do this, and that we captured your thoughts and perspectives for the historical record. So, thank you so much. 

Ellis:

Okay, well, thanks very much.