by Sabrina Keating and Niya Petkova

On a chilly Tuesday morning we head off for a conversation with the new Head of School and module coordinator of Lagrangian and Hamiltonian dynamics and QFT, Prof. Jonathan Keeling. His office has changed to one larger and brighter, but we are greeted by a familiar assembly of textbooks, and a blackboard full of equations transferred from his old workplace. Following a light prelude on the superiority of blackboards over whiteboards, we sit down to discuss the duties of the Head of School, Jonathan’s path to St Andrews, his research on light-matter interactions, and much more.

We would like to start with some questions related to being Head of School. You just assumed that position?

From the first of August, yeah.

How are you finding it? I’ve heard it’s very time intensive.

I think for the first few months; I’m told always there’s quite a lot to learn about all the different things that are going on. Because unless you’ve done a huge variety of other administrative roles in the School before, there will always be some aspects that you weren’t directly involved in. I’m anticipating the first six months or so is a lot of finding out how things work, and then I’m told after that, it will get a little less full-on.

How long is the appointment? Is it fixed-term?

The appointment is for four years, extendable by another couple of years, if both the Head of School and Principal’s office agree. The previous Head of School did five years; I’m not sure what other people did before that.

Are there any changes you would like to implement to the School? Do you have a vision for it?

There’s no particular big changes that I’m intending to push. I think the situation that all UK universities find themselves in is one of increasing pressure due to finances staying effectively flat and inflation causing a lot of problems. And so, I think, the principal thing I’m trying to look at is how we make sure we continue to do well the things we want to do but being aware of reduced resources. And to make sure that the way we do that is not by asking staff to work harder, and that we try to keep track of managing staff workloads in this environment. So, I think that’s really where I’m looking at things for the first few years at least, of just trying to make sure we continue to do well the things we’re doing well but being aware that the resources we have are under constant pressure.

Can more research funding fill that gap? Or is that funding separate to what you can dedicate?

The way the UK system works is there is research funding that comes from competitively won grants, and there is a bit of funding that is awarded to the university based on the outcome of the REF [Research Excellence Framework]. A lot of that money goes into paying baseline salaries and keeping the lights on, etc, in addition to teaching income, which goes into paying salaries and keeping the buildings running, etc. When people win the research grants, that pays for postdoc salaries; there are also the stipends for PhD students. They come from a slightly different system in general, because rather than being directly on most of the grants, they’re paid by a grant that comes to a university in proportion to the total grant income awarded. There are also other grant agencies beyond the UK Government. There are charities like the Leverhulme Trust. There are things which come through the Royal Society; there are various other [sources of funding], particularly for biomedical things and for the biophotonics areas in the School; those are supported by various charities in those directions. Some of these will also pay for fractions of academics’ time. So, in those cases, if there was a significant increase in the funding that came from those, that would indeed support more, although the total amount of research funding is not going up in line with inflation and hasn’t been since 2010-2012, or so. I mean, there are opportunities there, but that’s also under pressure in the UK. It’s also a case that there’s a slightly strange system in the UK, that government funded research grants are costed at what is called 80% of full economic costing. Which means that when we write [an application], we should put in not just the cost of the postdoc and the fraction of a staff salary, but the estates’ costs over the costs of lighting, heating, maintaining buildings, and some other costs associated with the administration of grants, etc. So that ends up increasing the cost we apply for significantly, but then the Research Council gives us 80% of that, and the idea is that the university should cover the remaining 20%. So that means that actually winning research grants can cost the university money rather than bringing in extra money.

Moving on to something else now: from a very young age did you know physics was what you wanted to do? And when did you know that?

I think I knew I wanted to apply for a physics undergraduate degree around the point I started A levels, but probably not clearly before that. Before that, I thought about a variety of things. I certainly was interested in programming and thought about that. I had a brief interest in architecture, but some work experiences in architects’ office showed me that wasn’t really what I wanted to do. But yes, I think from the point I started Sixth Form, I was fairly sure I wanted to do physics, whether it was just physics or physics and maths or physics and computer science wasn’t quite clear. As an undergraduate, I was at Cambridge and did natural sciences. So, my first year I had the opportunity to do computer science as well as physics and other subjects but then specialised in physics from second year onwards.

So, in terms of your work on polaritons: did you have any inkling when you were in your undergraduate degree that that’s the kind of thing you wanted to do?

In terms of where I ended up doing research, I think the route I got into that is a story that’s fairly common, of a kind of sequence of accidents but then discovering which things you enjoy through those accidents. But when I was entering my final year and thinking about final year projects, my director of studies, which is kind of equivalent to the advisor, said I should go and talk to a particular member of staff because they thought I would get on with them and thought the things I was interested in matched their interest. And so, I did that. That was Peter Littlewood, who is actually now [on] a 20% appointment in St Andrews, who visits in the summer, but is mostly based in Chicago now.

So, he supervised my final year undergraduate project actually on something completely different. But then there was an opportunity for PhD project with him, and I applied for that. And that was when I started working on polaritons. I guess actually right at the beginning I wasn’t directly working on polaritons; right in 2002 when I was starting, there was some interest in polaritons, but a lot of interest in the idea of exciton condensation, so just bound electron-hole pairs. And there had been a number of experiments by the group. So, what I started my PhD in was looking at evidence of exciton condensation.

Did you find any?

Well, we worked out some signatures one might see if there was a straightforward exciton condensate. It was subsequently seen that a lot of the experimental evidence was showing something else and wasn’t really to do with the Bose-Einstein condensation of excitons. Those experiments still continue, and there’s recently been some work on cuprous oxide [1] that shows something reasonably convincing, although still not quite a clear smoking gun. But it was only in the middle or end of my first year of PhD, that I switched from focusing on exciton condensation to looking at polariton condensation.

And was that switch because you were finding it a bit iffy whether you’re doing this?

I think it was more that there were more experiments on polaritons developing. And Peter had had two previous students work on questions related to the theory of polaritons. I’d finished one project and that had been published, and then I’m looking for what to work on next. We identified something on polariton condensation and particularly understanding how to take the kind of model that he had developed for polariton condensation and ask about beyond mean field effects in that model. So that was really what most of my PhD thesis was about: how to go beyond mean field theory describing polariton condensation.

Okay, and were you still at Cambridge at that point?

That was still at Cambridge. The first project for exciton condensation I had actually started as a summer internship in MIT, working with Leonid Levitov. There was a brief period around 2002 where Gordon Brown, the then Chancellor [of the Exchequer], was very keen on encouraging models of university-industry collaboration based on things going on in the US, and as a result of this, had given a lot of money to encourage collaborations between Cambridge and MIT, and both that summer internship, and actually my PhD was funded by that scheme. So that meant that I was working on that first project jointly with people in MIT, and actually that was why I then went to MIT as a postdoc straight after my PhD.

Since you’ve been both in the UK and the US, do you find anything in the higher educational system that’s better there or better here?

There are definitely differences, actually, perhaps slightly less so in Scotland than in England, because the Scottish four-year degree with two years of pre-honours and two years of honours is in some sense similar to the structure or at least inspires the structure of a US degree. This kind of division of focusing on one subject in the last two years and having more breath in the earlier years is closer to the US system. The UK system in general, from school onwards, causes people to specialise earlier. So, this then leads to a lot of the differences, that there is I think more real breadth of subjects in undergraduate degrees, but that means that people end up with slightly less specialist knowledge at the end of a US undergraduate degree, and that then leads to there being the need for more graduate courses in a US PhD. Those graduate courses, though, I think while they start often a little bit behind where you end up at the end of your integrated masters in the UK, by the end of doing those graduate courses, you’ve picked up a lot broader specialized knowledge. So in a UK PhD typically, at least in the past, people were very focused on just their own research project and the broader thing of what is the forefront of knowledge in condensed matter physics wouldn’t necessarily be so covered by a PhD training, but that’s the thing which has been recognized for quite a while in the UK as a difference, and something that can cause some problems. And in fact, in Scotland, the Scottish universities physics alliance (SUPA) was set up partly with that in mind of sharing graduate courses across the SUPA universities, with exactly that idea that you can, in the UK context, provide some of that specialised graduate education. And I think that has made a big difference. So those differences are perhaps a little less, particularly in cases where there are those kind of graduate courses available. There will always be a difference between what’s possible in three-and-a-half-year PhD versus a five- or six-year PhD.

I think, a thing which is often said about a difference in the US is there’s slightly more tolerance for people to try things and fail. But as a culture the UK often says that if someone did something and it didn’t work out, that indicates some character flaw, and people shouldn’t be given second chances. And I think particularly in terms of translating things from university research to applications and technology, there is a culture in these places, where there are lots of startups, of people trying something a lot of times, and maybe this company didn’t work, but now you have a different idea and try that again. And I think that idea of tolerating failure is an important one and allowing people to do things and that not be the end of their career, because something once went wrong.

Building on this idea of a greater tolerance to failure: I was wondering, was there anything in your studies, or even currently, that you find too hard in physics, and can’t get your head around?

I can certainly point at research projects or directions I tried which didn’t work and then I stopped and moved to other things. I am not sure of things where I didn’t do them just because I didn’t think I would be able to understand them. I think part of that is saying most things with enough time and the right resources to support you, it is possible to understand them. There are many things where, from the outside, it feels that there is a huge amount of material you have to cover to be able to do anything. And [with] most things, once you’ve got the right entry point, there is a way to get into that. I think, in terms of thinking about research, there is a there’s a different question of “Do I think that I’ll be able to do things that people working on that topic are not already doing?” And so, I think, I’m more cautious about saying, “here is a field that is very fashionable and everyone is working on, let me get into that”. And that’s not so much from worrying whether it will be possible to understand the material but worrying, “Do I already have an idea that I think is applicable in that field?” So, sometimes I can get into things in which other people are very active because I think there’s a particular idea to bring. I think there is a danger of going into a field because it’s fashionable, and what that generally means is a lot of pressure to do things first. And actually, most of what you’re doing will be done anyway in six months’ time, because other people are working on it. So, I’m keen to find places where perhaps people are not working on things, and there is something unique to contribute.

How do you find those niches? Are you reading lots of papers? Conferences?

When I was postgraduate director, a lot of this is what I said in the induction talk to postgraduate students; there is definitely a need to do a lot actively to know what is going on and what are interesting directions: reading papers, or skimming many papers, and then reading a few is definitely important. As an academic, you also end up reading papers quite often as referee, so you see things that way, but also making sure there’s some time to be able to look at papers you’re not refereeing but are just interesting. Going to conferences is very important. And what is particularly important there is it’s not just listening to the talks. Actually, quite often the talks are things you already have seen a version of or have read the paper or something similar. What is very valuable is talking to people and finding out what we’re working on; finding out what questions they think are interesting; saying you’re thinking about something and seeing what ways people react. And so having the ability to talk to people at conferences, and then also having a network of collaborators you can talk to and say, “I have this idea. What do you think?” is often useful.

Just circling back. You were an undergraduate at Cambridge, did an internship at MIT, then PhD at Cambridge. So how did you end up going to MIT after your PhD?

So, at the end of PhD, I was applying for all kinds of next positions. One of the things I applied to is a thing called Lindemann Fellowship, which, at that time, funded one year of research in the US. These days it still exists and funds two years of research. But yeah, I got that, specifically to go to MIT to work with Leonid Levitov, who was the person I had done the internship with. [I] worked on a quite different project there about how to create idealised or ideal pulses of electrons in one dimensional wires. Specifically, how to put an electron in without messing up the Fermi sea and creating electron-hole pairs, because most things you do with applying voltages will create electron-hole pairs. I also, at that point, was awarded a Junior Research Fellowship in Cambridge, so I came back at the end of one year in US to take that up. And then while there, I got an EPSRC Fellowship, which I could then move to a permanent position here.

Was that the first time you came to St Andrews?

I’d been to St Andrews to visit to give a seminar once before I came for interviews.

Was there a particular reason you chose St Andrews, or was it more of a happenstance?

Well, so while the areas I work in actually could conceivably be thought of as quantum optics, or light-matter interactions behind pretty much everything, I’ve always felt that I fitted well in condensed matter theory groups, because often what I’m interested in is many-body physics, and there’s a fairly small number of places in the UK that have an active condensed matter theory group. And so, when I was applying for jobs, I was looking at those places, and this was one of them. But of all places where there’s more than, say, two people working in that direction, there’s probably about five or six [such] places in the UK, and so this was clearly one of them.

Was this the one you most wanted to go to? Or was there anywhere else you kept your eye on?

I guess some people take the approach of being absolutely confident they will get a job, and they will just hold off until the offer they want. I took the approach that I applied to places that I was happy to work in and accepted the position when I got it.

Taking the statistically advantageous route. And when did you start here?

2010. It’s been 14 years.

Have you seen many big changes in the department? Good, bad, neutral, anything interesting?

The department has grown over that time, particularly, the number of undergraduate students has grown significantly. The number of staff has grown a little bit. I’m not sure there’ve been any radical changes; while 14 years, I guess, is a long time on the time scale of an undergraduate degree, it’s not a huge time scale in the history of St Andrews. So, I think many things have remained fairly similar.

What modules did you teach in the beginning?

So, I’ve taught relatively few modules as coordinator, really only 5004 and 4038, so only QFT and Lagrangian and Hamiltonian [Dynamics]. For the first few years I was here, I was also doing a lot of other tutorials on things like electromagnetism and trying to remember what else, there were various other things.

Could you describe your research in a very simple, easy to understand [way]?

So, I’d say that the two broad areas that I look at are open quantum systems, which means quantum systems that are interacting with their environment, so not perfectly isolated, and understanding how to model that without having to explicitly model all of the environment, because that’s a difficult problem in quantum mechanics. And the other broad area is systems involving interaction between light and matter, hybrid light-matter states, and those are very closely linked, because almost anything that is trying to use light interacting with matter will be an open system, because it’s very difficult to perfectly trap light, so there’ll always be light leaking out of systems. And therefore, those two things link together. Within that, there are three, perhaps four directions that I’m interested in at the moment. There’s a lot of work I do with a collaborator in Stanford, who works on experiments with cold atoms in optical cavities and trying to use this for all kinds of interesting, many-body quantum phenomena. So, things we’re doing right at the moment is building a realisation of the Hopfield associative memory, so the thing that last week’s Nobel Prize [2] was on. But a physical realisation of this using cold atoms and optical cavities and exploring spin glasses. So that’s one broad area.

The second broad area is organic polaritons. These are hybrid light-matter particles, and the organic bit is built on organic molecules, so carbon-based molecules. And the interesting thing about that, from an experimental point of view, is that these can operate at very high temperatures. So, you can realize Bose-Einstein condensation in a room temperature experiment, not needing ultra-low temperatures. The interesting thing from the theory point of view is organic molecules are quite complicated in terms of their interaction with light. When you look at the absorption spectrum, they show all kinds of peaks, and this is because the electronic transitions are coupled to vibrational modes. And this leads to an interesting challenge of the competition between light-matter coupling and the coupling to vibrational modes. So that’s the second broad area.

And then the third broad area is there’s a lot of stuff I’ve done in collaboration with Brendon Lovett [who is] in the office next door about dealing with systems coupled to an environment where the environment is complicated, and particularly what is called non-Markovian environments, which means things where you can’t ignore the fact that the environment has a memory: that things which escape from a system might come back again at some point. And most treatments of open systems are based on the Markovian approximation of saying the environment just forgets things immediately. And so, you can always think of it as just replacing it with a new copy of the environment and not worrying about the information stored there. What we have done over the last seven or eight years is to say there’s a whole new approach to non-Markovian systems based on the theoretical concept of matrix product states, which allows you to do calculations very efficiently and [we are] trying to now work out what can we really do with that now, narrowing over those methods that are possible.

As a theorist, is it important for you that you do something that can be tested experimentally?

Both that way around and also things which are driven by saying there is this weird observation and trying to explain it. Condensed matter theory has had a long history of kind of starting from observations of things like superconductivity, like different problems of magnetism and asking, “How do we build models? Let us understand what’s going on.” Those might be simplified models, but they at least give you the essence of what’s going on. And then, in some cases, allow you to really build more complicated models to really predict things. I think this was exceptionally productive in the first half of the 20th century, up to about the 50s and 60s. But then after that, there was a problem that a lot of the easy problems got solved, and a lot of the remaining things with condensed matter were things where none of  the easy theories seemed to work. And so, it led to a period where there was a lot of people developing simplified models or complex phenomena and arguing amongst themselves about which simplified model works, because none of them were really able to fully reconcile with the experimental data, but also the experiments were not particularly clear on what was happening. And I think that then led to a whole range of theory which is built on a kind of pyramid, building on all other bits of theory, but there are still bits of experiment where one can, I think, make progress by talking to experimentalists and finding questions that people are interested in working on with both theory and experiment.

What’s your opinion on string theory?

So, I’ve never worked in anything connected to string theory. I think this is a thing where there were some clearly very elegant theoretical ideas. In terms of a topic to go into working on, I think the difficulty is to really be clear, what is the new thing that one can do? Why is it that there is a possibility to do something new here that hasn’t been done in the last 20-30 years of research in that direction? What is perhaps worth also noting is with some of these ideas one of the more positive aspects in condensed matter theory is realising there are ideas from high energy physics that might be applicable in condensed matter physics. That you might be able to take what are elegant theoretical ideas and think of ways of realizing those, or of using those to understand aspects of condensed matter physics. So, there are some very nice examples of that. Whether they have really yet solved problems, I think is still an open question.

Throughout your career, what are the biggest challenges, if you’re comfortable sharing, that you’ve encountered, and how did overcome them?

I think one of the biggest things is that a career in academia involves an awful lot of rejection. The success rates on grants are low. So, you are writing a lot of grants knowing probably this is going nowhere, probably this is not going to be funded, and then you’re going to have to come up with new ideas and write a new grant. It’s also true with writing papers, there are very few papers which you send in, and the referees say, this is brilliant; it should just be published. There is almost always an argument to be had with the referees. Sometimes you lose that argument, and the paper’s rejected from that journal, and you have to think, was there a fundamental issue or was it just that the referees didn’t think it was important enough for that journal? So, there’s an awful lot of time spent having criticism of your work, having ideas rejected. And so, there is definitely a need for resilience and trying to separate the criticism from feeling like [it’s a] criticism of yourself.  And the resilience to just try again and again because a lot of these things, particularly with grants, I think a lot is down to chance. I think there is some element one can do, but even with the best written grant, you can end up with a referee who was having a bad day and just didn’t like it, and with low success rates, if there is one referee who didn’t like a grant, then it’s unlikely to be funded. So, there is always that chance, and one has to get used to living with that. I think that’s probably the biggest thing I think of as a difficulty in an academic career; that constant rejection.

Have you ever had imposter syndrome? Or are you one of those people just never gets it?

There are elements of it. So, I think as Head of School, there’s a lot of points where I think, what, “why am I supposed to know the answer to this question, how? How is it that my career is supposed to have prepared me for this?” I think I was never confident I was going to end up in an academic career. Always along the route, I was thinking, “I have to apply for all the things I can. I have to apply for lots of postdoc fellowships. I have to apply for many jobs in places where I’d be happy”. So, I wasn’t taking the approach of “It’ll all be fine. I’ll just apply for this one thing and see what happens.”

So, I think, certainly most people feel some elements of imposter phenomena. It differs depending on what kinds of background people have, what messages they’ve had in school and so I think it’s not been a major problem in my career, although I think probably with most Heads of School, there are very few people who would say they felt perfectly prepared for the role as they stepped into it.

You mentioned way earlier on, when you were in school, you were looking at architecture. Have you got like a secret, artsy kind of background, I mean, [points to the top of a shelf] I don’t know what that is, like, that little sculpture thing up there?

This is part of a public engagement. It’s a Kapitza pendulum. So, like the one in the lab, except you can hand it to children.

Not particularly, I guess, like many physicists and mathematicians, I have quite a keen interest in music, not playing or doing anything myself anymore, but yes, I listen to quite a lot of music.

Favourite music style of artist?

So, Bach is clearly my favorite composer, which is a fairly standard answer.

Did you play any instruments before?

At school, I played clarinet, piano.

What other stuff do you enjoy doing outside Head of School, physics; do you have time for anything else?

Right at the moment, there’s no time for much else; when I get time, walking in Fife, and particularly the Fife coastal trail is a great thing to explore.

What part of physics did you struggle with? Is there any physics where you were just, like “Ah, that’s hard.”

Well, I’m not sure in terms of hard, but things that as an undergraduate I didn’t enjoy the courses. Somewhat ironically, it was courses about light-matter interaction. I did chemistry as A level, and I didn’t really enjoy chemistry as A level. And actually, that was why I chose to do computer science as one of my options in natural sciences in my first year as an undergraduate, to avoid doing chemistry. And then, ironically, I now do things about organic polaritons, which involve a lot of understanding of chemistry. I think in both cases the point was I didn’t get at the time what the interesting topic was, and it felt a bit boring. So, there were definitely things like that. But actually, when looking at them in terms of research, they are much more interesting.

References

[1] Morita, Y., Yoshioka, K. & Kuwata-Gonokami, M. Observation of Bose-Einstein condensates of excitons in a bulk semiconductor. Nat Commun 13, 5388 (2022). https://doi.org/10.1038/s41467-022-33103-4

[2] Awarded to John J. Hopfield and Geoffrey Hinton. For more details: https://www.nobelprize.org/prizes/physics/2024/press-release/