Jo Dunkley

Zierler:

This is David Zierler, oral historian for the American Institute of Physics. It is November 3rd, 2020. I am delighted to be here with Professor Jo Dunkley. Jo, thank you so much for joining me today.

Dunkley:

Oh, thanks for having me.

Zierler:

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

Dunkley:

Yes. I'm a professor of physics and astrophysical sciences at Princeton University.

Zierler:

Is that a joint appointment in one department, or two appointments in two different departments?

Dunkley:

It’s an appointment in two different departments, so I have a half job in each department.

Zierler:

I see. So it’s a 50/50 split?

Dunkley:

It is, yes.

Zierler:

And where are you most likely to teach classes from, to take graduate students from?

Dunkley:

Both, actually. So I do teach equally in — I teach, yeah, slightly more in physics, but really both. And I take graduate students from both.

Zierler:

And was the tenure consideration also made jointly?

Dunkley:

Ah, well, yes, except I came to Princeton with tenure. So they did that jointly, without me. [laugh]

Zierler:

That was one thing you didn't have to worry about.

Dunkley:

Yes, exactly. [laugh] Exactly.

Zierler:

And when did you come to Princeton? What year was that?

Dunkley:

2016. As faculty. I was here as a postdoc before.

Zierler:

Right. And then it was the big return in 2016. Jo, let’s take it all the way back to the beginning. I want to start first with your parents. Tell me a little bit about them and where they're from.

Dunkley:

My father is from London. He passed away a little over a decade ago. My mother is from the south, south of London. My dad left school at 15. He was a director of photography, working on films and then TV commercials. And my mum also didn't do a degree straightaway after school. She did a degree later on, when we were kids. But neither of them were academics, in any sense. [laugh]

Zierler:

Where did they meet, your parents?

Dunkley:

In London. My mother was working as an assistant stage manager at a theatre in London, and my dad was working in film. And yeah, they met in London.

Zierler:

And no scientific inklings from either of them?

Dunkley:

No, but I think my father probably had a lot of that. He didn't have the opportunities to — as I said, he left school very early on. He didn't have a good experience at school. Knowing him, I think he could have had a lot of that going on, I could tell.

Zierler:

Jo, did you grow up in London?

Dunkley:

I did, yeah. I grew up in — not central, but not way out in the suburbs, either. Kind of halfway out from the middle, in a little terraced house in North London.

Zierler:

The British, of course, are quite class-conscious in a way that Americans don’t really understand. What was your understanding of your family’s class status when you were growing up?

Dunkley:

Yeah, I think we were middle class, and we were certainly comfortable, but not excessively wealthy. My mother had grown up poor, pretty poor. And my parents — my father ended up earning enough money in his job that they then — they then basically chose to send us to private school, which is a big decision for them, given what they hadn’t had in their schooling. And so we were certainly comfortable, and I was at this school, but I was also aware — some of my friends at that same school were wealthier than us, but we were definitely very comfortable.

Zierler:

Your parents were definitely intent on giving you better opportunities than they had themselves.

Dunkley:

Yeah, absolutely. Yeah, that was really important to them.

Zierler:

What were your parents’ backgrounds from their parents during the War?

Dunkley:

My mother, her father had actually been a theatre director, and an actor, but her parents were divorced. So she grew up with her mother, and her father was sort of slightly out of the picture. And her mother I think sometimes played piano to earn some money, but didn't have professional training. And my father’s parents — I think my grandfather worked in hospital administration, but he was also — he had an interesting background, because my grandfather was adopted, and they never knew where his family had come from.

Zierler:

Jo, the private-public school divide is certainly also very different than in the United States. What did it mean that you went to private school? How do we understand that?

Dunkley:

Yeah, what that means is that we — it was a school that you had to pay fees for. So in England, or in the U.K., you either go to what’s known as a state school, which you don’t pay fees for, or a private school, that you do. Now, there’s the complication that some of the fanciest private schools end up being called public schools [laugh], right? That’s the worst. That’s more for kind of boarding schools like Eton and stuff. The school that I went to would sort of just be referred to as a private school. And so that meant you paid fees. There was an entrance exam. My school was quite interesting, because it was one of the oldest girls’ schools in London. Where it was in North-West London a lot of students were Jewish or of Indian descent, and it was quite a mixed — it was a very interesting group of girls, but also from kind of a wealthy part of London. So I went there from age seven, actually, through to 18.

Zierler:

Wow. So you were in an all-girls environment for your entire school years.

Dunkley:

Yep! [laugh]

Zierler:

Jo, when did you start to get interested in science? Was it early on?

Dunkley:

Well, I always loved maths. I kind of thought that was my thing. Like I just always loved it. My mum tells me about counting things from my high chair. And that was what I loved. So it took me a while to realize that I liked sort of thinking about science with it — probably until later in secondary school — and I started to realize that I liked to be able to like use my maths to, you know, answer questions and understand things. So it wasn’t until towards I think the end of high school that I realized that maybe physics was like something that used my maths, but in a cool way.

Zierler:

As a girl — of course it’s a counterfactual; you don’t know what would have happened — but is your sense that being in an all-girls environment was positive, in terms of allowing you to develop interests in math and science that you might not have had in a co-ed environment?

Dunkley:

I do, actually. Of course one can never do the control test [laugh] but I think there was never any question that one should try and succeed and do whatever you wanted. Our school was very feminist. It had this very strong like “Women can do everything.” And there were lots of — anyway, this was the attitude. And I had a wonderful female physics teacher, who also was just completely assuming that we would all just be able to do whatever we wanted. So I think that was very helpful. I think there was some sense — if I look to the school, a huge number of the students there went to do medicine. So there was definitely a path that if you liked science, you went to do medicine. And so thinking that you might want to do science to be a scientist was quite unusual in that environment. But this concept of just being able to do whatever you wanted — I think that was really important to me.

Zierler:

Now in the British system, at college, you declare the major right away. You have to go in knowing what it is you're going to pursue. So of course that begs the question, how did physics enter your mindset in high school?

Dunkley:

Well, that’s a good question. So I think until close to the end actually, I thought I probably wanted to study maths. I think that’s what I thought of. And then I suddenly realized that I was really enjoying — I did A-levels in maths, physics, and chemistry. Because you know, you're stuck to those limited choices even at age 16. Which never quite suited me, because I love languages and writing and I had to sort of give those up a bit, although happily sort of came back to it later. But yeah, so I liked the chemistry and the physics. And there was actually a degree in the U.K. that I did, at Cambridge, which is called Natural Sciences, where you enter and you don’t actually choose which science you're going to major in. You enter and you start with three or four sciences, and you then sort of refine them towards the end.

Zierler:

Sounds like a very Newtonian approach to things.

Dunkley:

[laugh] That’s right. But I also applied to other universities to do physics, because I thought that’s what I was leaning towards. But I was very happy to get to start this degree, because actually I wasn’t sure if I wanted to do chemistry or physics or perhaps some new area of the sciences. That there were things that you don’t study at high school. But I had realized by then that I wanted to apply myself with the maths, rather than just study it directly. I think we had done mechanics at A-level, where I suddenly realized that the bit of maths I liked was the bit that was applying it to physical problems.

Zierler:

And given your abilities and interest in math, was theoretical physics a natural path for you?

Dunkley:

Well, do you know what? Yes and no. I was really anxious about going to study it, because I had the perception, which is widely held, that the physics part of that degree was the hardest, and that you couldn't — people would go there thinking, “I'm going to do physics, and I'm going to do theoretical physics,” and then they would not do it, because there were all these other options and paths that people went down. And so I did actually start — when I got to Cambridge, actually, I found it really — I found physics really hard! It was really difficult! And I was pretty close to not continuing with it, actually, at university, because of this thing of thinking it was too hard, and that I wasn’t good enough to do it.

Zierler:

I wonder if being in a co-ed environment influenced your feelings of doubt if you were able to actually go on in this field?

Dunkley:

Yeah, I think a little bit. So I was quite lucky, because I had a very — in my Cambridge college — where you have sort of a cohort of about six of you — actually three of us were girls, which is quite unusual, actually. But nonetheless, there were also some overbearing male students. And the lectures were suddenly very full of male students. So yeah, I think that was definitely going on. But then I ended up doing quite well on my end-of-year exams, and I think I suddenly realized that maybe [laugh] — maybe I was not so bad at this after all. [laugh]

Zierler:

Jo, who were some of your professors at Cambridge that were mentors to you, or who you became close with, or who were really influential in your development?

Dunkley:

I can think of certainly a couple. So one professor who I wasn’t close to but really inspired me was Professor Malcolm Longair. He taught us relativity, and in the first year at Cambridge, it was amazing. I think this really sparked this sort of like passion of like, “This is really amazing.” Because he was just so excited and was teaching us like really cool stuff. I still remember we were working on how fast you have to run — if you’ve got a 20-foot pole, how fast do you have to run to fit it into a ten-foot garage, right? [laugh] And so that course coming in our first year was a bit of a spark of like, “This is really neat.” I also had a director of studies in my college who was really the closest academic contact I had, Professor Mike Hobson, who actually was — his research actually was in the cosmic microwave background, which I now do. And so I do actually think that that got me down my path. Because I ended up actually doing a research — in my fourth year, I did a research project with him on this topic, and that got me into what I do now. And he was always very supportive, and he kept telling me that I should keep continuing doing physics. [laugh] And that I should keep going. So he was quite a big influence.

Zierler:

Were there any women professors who were formative in your development? Or were there any women professors, period?

Dunkley:

There weren’t many. There were some. I remember we had a particle physics lecturer, and I've now forgotten her name, but I do remember she was — it was great to have her. But there were not — and I did actually have a — again, I forgot her name — a woman — I was looking her name up recently, actually. She was a PhD student who gave us like small classes, and she was really fantastic. And I remember we all thought she was wonderful. And so that was a nice influence, too. But in general, not that many women.

Zierler:

Jo, these terms mean different things in different times and places. Was there anyone on the faculty who would have identified themself as a cosmologist at this point? Or anybody studying early universe would have come from an astrophysics perspective?

Dunkley:

So yes, a number of them, and that included this Professor Hobson, who was my closest contact in my college. He was certainly in cosmology. Yes.

Zierler:

Were you thinking as you were considering graduate programs cosmology specifically?

Dunkley:

I didn't think of graduate programs at all when I was an undergraduate. I didn't think I was going to do graduate school. It didn't even occur to me. Did not occur to me. It was funny — so I just didn't think of being a scientist. At Cambridge, I spent a huge amount of my — I loved my degree, but my friends weren’t particularly scientists, on average. And the things that I did in my spare time were not science. And I thought I wanted to do something that was — I still had this perception back then that scientists were kind of nerdy [laugh] people who sat in a room quietly [laugh] doing boring stuff.

Zierler:

[laugh]

Dunkley:

[laugh] I didn't have any idea, you know, what it was. And I think actually in the U.K., there’s a bit less — in the U.S. now, I see more of a path towards — like an assumption that you might go to graduate school. Whereas in the U.K., my peers were not in general thinking of that. People were getting jobs. People were either doing medicine or going to law school or going to get jobs in London. Graduate school path was really unusual. And I didn't have generally that group of friends who were doing that, and it just really didn't occur to me. So I had had these like wonderful experiences during my undergrad of traveling overseas and volunteering for some NGOs, and I thought I wanted to do that. I thought I wanted to — I applied for the civil service, actually, in the U.K. Like after graduating, I thought I wanted to perhaps either work for the civil service or for an NGO, and do something like that. So actually, when I left Cambridge, that was the path I kind of — I took a gap year after my degree to figure things out —

Zierler:

So Oxford was not on your radar at all when you graduated.

Dunkley:

Oh my god, no, not at all! Not at all! [laugh] So the only thing that happened is I had realized in my last year how much I liked doing this. So I did this research project, and I hadn’t done any research projects before that. And I suddenly — I learned to computer code. And —

Zierler:

Which project was this?

Dunkley:

It was a master’s project with Professor Hobson, who had been my undergraduate advisor. And it was with, actually, cosmic microwave background data, or at least kind of simulated data. And it was my first exposure to working with computer codes and doing a research project. And I did really enjoy it. But I didn't really enjoy the environment. The academic environment, or like how I saw the kind of PhD students working in Cambridge, I didn't really feel like that was me. So anyway, in my head I realized that I had liked that, but I thought, “This still isn’t — this isn’t me.” [laugh]

Zierler:

You strike me as a very social person, and perhaps theory was a bit isolating, you thought.

Dunkley:

Yeah, that’s right. That’s right, I did. Exactly. I thought it was isolating, and then I also didn't see myself in I guess the community I saw just with that kind of brief encounter. So then I was in London for this year, kind of applying for the civil service and actually working for an NGO in London. But what I started doing to make money was tutoring maths to high school students. At the end of each day, I’d bike off and go and give an hour class in maths to teenagers. I did that on Saturday mornings as well. And I gradually realized that that was the hour of my day that I enjoyed the most [laugh] —

Zierler:

[laugh]

Dunkley:

— was doing the maths. And I realized that actually even though I kind of was passionate about a lot of these — saving the environment and trying to do various things that seemed to be more like purposeful — like my brain really missed doing the numerical stuff. So I think this kind of realization that actually the thing I thought I wanted to do was maybe more boring to me in practice than a thing that I hadn’t even thought of was something that might be interesting. I started to pursue — like looking into PhD programs. And actually my former advisor, my undergraduate advisor, when I reapproached him for a reference letter, he was like, “Of course I could see you should be doing this.” He just hadn’t managed to convince me of it earlier. So he was very supportive saying, “Yes, go and apply for PhDs.”

Zierler:

And was his advice specifically not to return to Cambridge for graduate school?

Dunkley:

No, actually. I applied to Cambridge and Oxford and London, I think. Imperial. And I did actually strongly consider staying at Cambridge. I did my undergraduate in the physics department, and there’s also an astronomy department where I could have done a PhD. So I did consider going there. But I really wanted to work with my PhD advisor, Pedro Ferreira, who I met through the application process and kind of — I decided based on wanting to work with him, to go to Oxford.

Zierler:

Why? What was compelling to you about working with Ferreira?

Dunkley:

I thought the science he was doing was really interesting.

Zierler:

Which was what? What was his research at that point?

Dunkley:

So he was still doing theoretical cosmology, where he was using, again, cosmic microwave background. In hindsight, I've had a very narrow career, right? Because I started doing this as my research project, and then for my PhD, and I've stuck with it. But there’s lots in it. So he was working on cosmic microwave background experiments, but doing really neat theoretical work from it. But to be honest, I didn't know that much about the details of it, then. I was also really going on recommendations from other senior people saying, “You should work with him,” and students as well, saying, “He’s a really great advisor.” So I think I probably based it not just on the science, but also basically recommendations of working with him.

Zierler:

What was your sense when you arrived at Oxford? Did it strike you as a very different place than Cambridge?

Dunkley:

It did, actually. Yeah, it did. It was more interactive, more social. There was a professor there in the cosmology group, Professor Joe Silk. He was the sort of senior professor in cosmology. It was basically Joe and Pedro were the two professors in cosmology. And Joe had been in the States for years. The postdocs that he and Pedro — and Pedro had been in the States, too — the postdocs that they had recruited were very international. So I was quite quickly working with a postdoc from South Africa, and a postdoc from Cyprus. There was a real — I definitely noticed this sense of like — which I hadn’t so much felt in my — this sort of brief exposure as an undergraduate — that suddenly there was this sort of interactive and international community, which I liked. It did strike me as different, yeah.

Zierler:

Was the curriculum such that you could dive right into your thesis research right away, or was there a period where you were exclusively taking courses?

Dunkley:

I could dive in right away. So it’s slightly crazy in the U.K. system — you just dive in. I think because I had done this four-year degree, that after that — you do some graduate classes in the first year, but not in a big way. Not like in the U.S. So I really started right away with research. Because I only had a three-year — my PhD was three years, so you just have to sort of start, from day one, basically.

Zierler:

What was Ferreira’s style as a graduate advisor? Was he hands-on? Did he really put together your thesis topic and then send you off to work on it?

Dunkley:

Yes and no. So he’s a really fantastic advisor, but what he did was he identified this question or project that he wanted sort of the group to work on. But he realized that the problem was too big for a graduate student to do alone, and so he and another colleague, Martin Bucher, who was at Cambridge — he was another senior person — put together a little group of five of us, I guess — so those two of them, and two postdocs, and me — to work on this project together. And so that meant I met with him regularly, but I also then worked closely with these two postdocs who were in Oxford, who I think taught me a lot about computer coding as well. I’d sit next to them and watch them [laugh] write, watch them write codes. And so yeah, I was definitely — I think it was the nature of a U.K. PhD as well. It was a project, and you sort of did it, and there were multiple papers from it, but it was all quite connected. And that really was guided by Ferreira. I didn't come up with projects or ideas of my own. I think it’s not so much the nature of the U.K. PhD.

Zierler:

Jo, I’ll ask a grandiose kind of question that you might not have thought about at the time, but maybe you've reflected upon since. What were some of the major theoretical questions in cosmology and astrophysics at that point? And how was your thesis designed to be responsive to some of them?

Dunkley:

Yeah, so this was like — I started my PhD in 2002. So it was a really interesting time when dark energy had sort of been discovered, but people weren’t completely convinced by it. Inflation was also on the table, but not lots of evidence for it.

Zierler:

Jo, on that point — let me just interject there — was your world of physics — how parochial was it? To the extent that you're coming from Cambridge, you're at Oxford now, were you aware of people like Michael Turner, like Alan Guth? Were you aware of what was going on in the United States in cosmology? Or your world in cosmology was very much a British-centric world?

Dunkley:

I think it was quite British-centric, but also quite Oxford- and group-centric. I think as a U.K. PhD student, you don’t get that broad — to begin with anyway, you don’t really get that broad exposure. Well, that’s not quite true. I'm recalling — I think even in my first year, we had a big conference for Joe Silk’s — one of his birthdays. [laugh] A big celebratory conference. I remember lots of people coming to that, and it being sort of a — what am I saying? I did get a sense of like there being other people. I didn't know who they were at that point. But with Joe in Oxford, even though I wasn’t working with him, there was a real spirit of, “Question everything.” With both Joe and Pedro, the spirit was, “We don’t know if this Lambda-CDM model or this Flat Universe or this Dark Energy or Dark Matter — are any of them right?” And so my PhD was sort of about trying to see whether there was actually like a completely different model — the cosmological data was suddenly getting better. Pedro had been very involved in measuring the microwave background from these two balloon experiments, BOOMERanG and MAXIMA, and during my PhD, the first results from WMAP came out. And so that was 2003. So this was suddenly — basically my — the timing was like, look, suddenly there’s this overwhelming wealth of data.

Zierler:

WMAP was immediately important to you?

Dunkley:

Immediately. I remember — like, yeah, absolutely. I remember being in my graduate common room, and the papers came out I think in the evening. And that would make sense, wouldn't it? I think it was evening, U.K. time, I think. And I just — it suddenly — it was just all suddenly there, and it was really exciting. And the data was there as well. And the project that I was working on had been sort of using the pre-WMAP data, and suddenly we could use these data.

Zierler:

If you could explain a little bit more to our broader audience, beyond the field — what exactly is the big deal with WMAP and all of the new data that came out?

Dunkley:

Okay, so this thing I work on, this cosmic microwave background, it’s this image of the universe when it was 400,000 years old. So it’s the oldest picture we have of the universe. Not the whole universe; it’s a bit of the universe. And it’s kind of amazing, because we can measure this light coming from this epoch. It has been traveling for the whole age of the universe and reaches us — we can detect it now with these microwave or millimeter-wave telescopes. And this field had been around since the 1960s, and what had been happening up until my PhD was that these balloon experiments had sort of managed to survey little patches of the sky and map this image in greater detail. But suddenly, WMAP came along — it was a satellite, NASA satellite — and it took an image of the whole sky, and in much higher fidelity than before. There had been a previous image from a first satellite called COBE, which had sort of seen these first like hints of like — there were these features in that that are basically the seeds of cosmic structure that you can see popping up, at this epoch. But WMAP suddenly saw them in like amazing detail, over the whole sky. And it was a big transformation. And I think in terms of how it placed my PhD, WMAP came out and said, “Look, there is this model of the universe that fits this data, that now seems to work really well, and this seems to be our model for the universe.” And the thing of my PhD was like, “Well, are we sure? Maybe there’s something else. There’s always new data; we’d better check whether there are actually other models that fit it, too.”

Zierler:

Jo, on that point, immediately, what old questions did WMAP answer, and what new questions did it raise?

Dunkley:

Around 2000, there were still questions about — okay, so WMAP really solidified this idea that this model that fits the data is this thing called Lambda-CDM. And what it is is it’s a universe whose matter is dominated by cold, dark matter, invisible stuff. We don’t know what it is, but gravitationally it’s there, and we can observe its effects.

Zierler:

And this is the same dark matter that remains mysterious to us today.

Dunkley:

Absolutely. It’s still there. Exactly. It’s still a problem. Yeah. And then the Lambda part of it is this dark energy component that says that it appears that space itself has a vacuum energy, that if you've got nothing apparent in it, it’s still got an energy, and it’s this thing called Lambda that Einstein put in to his equations. And the space itself seems sort of geometrically flat, which means light travels in parallel lines and potentially stretches on infinitely far. And, another feature of Lambda-CDM is that we have all these cosmic structures now, today, these galaxies, thousands of galaxies. If we trace back, we say that actually — we can show that all of that cosmic structure can be seeded by really simply described initial ripples sort of in the early universe soup of particles, and that they can be described by only two numbers. Like basically all of the cosmic structures we see now, the physics that needs to describe them can come down to just these two numbers. And, that that behavior is also the kind of behavior one would expect from this kind of inflationary model for the early universe and like this rapid expansion of the early universe. So anyway, WMAP basically said, “Look, this paradigm of the contents of the universe being regular stuff, dark matter, and this Lambda — dark energy — and that the cosmic structures have been seeded by this very simple two-parameter description of the features — that fit the data.” And the amount of data it suddenly fit was extraordinary.

Zierler:

Given the trajectory, the momentum you were already on with your thesis research, did WMAP sort of turn your thesis inside out, or did it more like supercharge where you were already headed?

Dunkley:

I think it did that, because my thesis was — I was basically trying to check whether — I said to you that, you know, one could describe these seeds of cosmic structure by two numbers. My thesis was about trying to check whether there was a different model which had maybe more numbers but could have come about by very different physics of the early universe. Like the inflationary or non-inflationary expansion could have produced a different kind of balance of the initial perturbations to the early universe. I was checking — this group I was working with was checking whether this alternative was a possibility. And having this new data meant that we could then really explore it. And we did actually find that there was a different model, with like large — which was very different to the standard one — that really fit the data just as well. Because it was numerically quite difficult to make the theoretical calculations, this is what the postdocs I was working with kind of had the talent to do, was to change the theoretical models and produce the theoretical predictions. We were able to say, “Look, there is still this alternative model that has quite different early universe physics, that fits the data.” And that was exciting. But the other thing that was going on at the same time was just methodologically, which is the kind of algorithms we used that were a big part of my career too — at just the beginning of my PhD, this new method came out, and it was called Markov chain Monte Carlo, as a way of estimating parameters, estimating — again, this doesn't have to be for cosmology, but let’s say you have some data, and you have a model with like 10 numbers describing it, and you want to find out what are the best ten numbers that fit your data. It’s actually quite — it’s not trivial to do that. [laugh] You can’t search through — if you're trying to search through ten parameters, you're kind of searching through ten-dimensional space. And before my PhD, there hadn’t been a great method to do this. And then at the beginning of my PhD, this new method came out of how to do this, a statistical method. And we needed it, to do my project. And so actually a big part of my PhD was actually understanding this new method, and optimizing it, and figuring out how it worked.

Zierler:

Jo, given how cutting-edge all of this is, it sounds like Ferreira was probably as interested in your research as a fellow scholar as he was in making sure you were moving along nicely, as his student.

Dunkley:

Absolutely. Yeah, yeah. No, that’s right. That’s right. He was really — [laugh] a really good advisor. And as a group of us, we were all — it was exciting as a group. So it wasn’t — and I liked that — well, I think it was a realization that actually work does happen as a group, and that’s actually the more natural way, in my field, anyway, that we work. And I enjoyed that. And this other advisor, Martin Bucher, was also kind of a wonderful influence, because he was very, very smart. I would go and see him in Cambridge, and I would go and spend a couple of days there, and I would be asking him — trying to resolve a question, but then I would come out with something completely different, because he’d have taken us in a new sort of direction. So that was quite formative as well, because he was — yeah, the ideas and the way in which progress happened, it was nice for me to have multiple people that I was sort of learning from.

Zierler:

Jo, I'm curious if, by 2005, your lack of preparation for your next move was reminiscent of your time as an undergraduate, or your postdoc at Princeton was pretty well wrapped up even before you defended.

Dunkley:

Yes and no. I began my PhD, again, not being convinced that I would carry on with academia afterwards. I had actually thought that perhaps I was interested in doing something public science related, that a PhD would help me get towards that anyway. And so actually when I began my PhD, I didn't go into it thinking, “Okay, I will definitely do a postdoc afterwards.” But actually, once I was —

Zierler:

What do you mean by public science?

Dunkley:

Oh, sorry, I mean like sort of science education, like working for a science museum, or — like public communication of science, I guess. But actually as soon as I got to Oxford and was in this kind of community of — the research community with all these postdocs — again in this very vibrant cosmology group with Ferreira and Joe Silk —

Zierler:

The public science pursuit might have melded your family background in theatre, to some degree.

Dunkley:

[laugh] That’s right. Maybe! Although I'm a terrible — I've never liked acting. It’s funny — I like presenting, I do a lot of public presenting, but I couldn't act to save my life! [laugh]

Zierler:

[laugh] So how did Princeton come together for you? How did that all work out?

Dunkley:

It’s funny —

Zierler:

Ferreira and David Spergel and Lyman Page – were they working together at all?

Dunkley:

Do you know what? No. But Martin Bucher, who was my other advisor, he had been at Princeton, and he knew David Spergel. And so he was the Princeton connection, really. I think through Pedro, I think I soon was loving my research, and I think Pedro sort of — anyway, just assumed that I would carry on and do a postdoc in the same way that, you know, as academics, like that’s the — that’s sort of the automatic path. So this became sort of like, “Well, this should happen. I should carry on to do a postdoc.” And I think I decided I did want to. And so my final year, I did the thing that everyone does; I started applying for postdocs. I went to the States, had a hilarious time. I spent five weeks traveling around giving talks in the States. And it’s funny in hindsight, because at that time, we had very little funding in the U.K. to do that kind of thing, so I was basically traveling on a shoestring and like staying in bunk beds in youth hostels, in a way that now, my graduate students — I don’t see that happening [laugh] so much from the U.S. side.

Zierler:

And Jo, you're single at this point? There’s no two-body problem to contend with?

Dunkley:

Yeah, that’s right, I was single. So I could sort of do — I could go there. And Pedro had done a postdoc in the States, and then Joe Silk had also been in the States. There was definitely a spirit from Oxford like, “Go try the States. Definitely consider it.” And so that’s why I went over there and did all these job talks. And then again, through Martin Bucher, I particularly applied to Princeton. I mean, I applied to many places, but I — for example, I applied to take a fellowship there, a Hubble Fellowship, which I didn't get, but it kind of put me on the radar with them as well. And I went to give a talk there, and I remember meeting Lyman Page and talking to David Spergel, and it was so exciting. I remember my visit to Princeton, actually, when I was doing job talks, and it was just really exciting. And so when they made me a job offer, I — well, actually, I still wasn’t completely convinced that I wanted to move to America and move to Princeton, which had a reputation for being quite a boring place. [laugh] But Pedro told me that I had to go. [laugh] That it was too good an opportunity not to take.

Zierler:

What were Spergel and Page working on at that point?

Dunkley:

Well, they were basically working on WMAP. This was then 2004 or 2005, when I was finishing up and when I met them. And so they were deep in WMAP. And so I think I came along and gave a seminar showing how we’d fit this new model to their data and then developed this new algorithm to do it. You know, estimate these numbers quicker and better. And so I think I presumed that they could see that the stuff that I had already learned to do during my PhD would be a good fit for joining the WMAP team.

Zierler:

In what ways did your skill set work so well with what they were doing already?

Dunkley:

What I had learned to do — kind of the final product from WMAP — there are multiple things, but one of the key things is the estimate of these numbers. What is the age of the universe with uncertainties? What is the density of the universe, with uncertainties? That’s the sort of big product at the end, that comes out. And again, that means estimating multiple numbers, maybe ten or — well, the typical number was six, but I had been doing these models that had, you know, 35 or something. Estimating these numbers from the key statistics coming from WMAP was still this new algorithmic frontier that like even with — the WMAP team was still figuring out how to do it, and I had spent my whole PhD figuring out how to do it in a robust way, and quickly. And so I had written my own numerical code to really quickly do this parameter estimation method, and also I developed this test for checking that it was really robust. So that’s the piece — that’s the kind of — the job I then took on as a postdoc. But I will say I was very grateful to Lyman Page and David Spergel, because I actually took a five-month break before I went to Princeton, to do this project in London, in high schools. So I had spent a lot of my PhD time doing like talks in schools and a lot of — again, like public science talks. I loved it. And I really wanted to spend a bit of time really focusing on that. And so I then spent the sort of first semester after my PhD in London, visiting schools in Central London and giving talks, basically, about cosmology and astronomy. And happily, both Lyman and David were very supportive that I could do that, and then I could defer the start of my postdoc until a few months later.

Zierler:

Did you spend a lot of time at the Institute during your postdoc?

Dunkley:

No. I would go out to some seminars, but it wasn’t really — no, it wasn’t a big focus of my postdoc. I did have this thing that as a postdoc — I was again in the two Princeton departments, physics and astronomy. My office was in physics and I had this joint appointment in both. So I would go to astrophysics. So those two, I already had two — [laugh] two buildings to be in. So I didn't really come to the Institute much, no.

Zierler:

Beyond David and Lyman, I'm curious who else you might have worked with at that time. For example, did you work at all with Paul Steinhardt?

Dunkley:

Not really, no. I interacted a little bit with Paul as a postdoc, but not extensively, actually. We had some discussions, but I didn't work closely with him. In Princeton, it really was the two of them. But because I then — what happened really is I joined this WMAP team, and it was a very tight-knit team. It was about only 20 people, and it was very strict, as well, where we really couldn't work on anything else. We had to just work on WMAP.

Zierler:

And where was the funding coming from for this?

Dunkley:

From NASA. I believe that Spergel and Page had a NASA grant to support me and the work there. So I had a lot of interactions with other members of that team. So for example Eiichiro Komatsu, Chuck Bennett, who was the PI, Gary Hinshaw who’s now at UBC but was at Goddard. And so most of my interactions beyond Lyman and David were with those team members. There wasn’t much scope to start new projects. I started lots of new projects within WMAP, but I didn't have lots of other interactions in Princeton.

Zierler:

Did you go to Princeton with the intent on that just being a sort of brief interlude in the States, and that you knew you were going to be going back to Oxford?

Dunkley:

I definitely didn't — not to go to Oxford, no. There was no thought of — I didn't know what I would do. I sort of presumed that I probably would maybe come back to the U.K.? That it wouldn't be forever that I would be going away. But I think I would say my whole career, I didn't have — [laugh] I didn't have a plan! [laugh] I didn't think, “Oh, I must get a faculty job.” I think I still even thought — this long-running thing of like always enjoying doing other things, too, like teaching and science communication, I think I always thought “I’ll just keep going as far as this takes me,” rather than thinking, “Oh, I must now get my faculty job.” So I definitely didn't even really during my postdoc think, “Oh, I must be getting ready for that thing.” Although opportunities came up quite early at my postdoc for these things to apply for. So I was pretty lucky in that sense.

Zierler:

Who was instrumental in pulling you back to Oxford? Was it Ferreira? Was it Silk?

Dunkley:

Yeah, it was Ferreira. Yeah, yeah, it was. Yeah. So they got this wonderful thing, which really actually in hindsight transformed my career, which was — there were these fellowships called RCUK Fellowships — Research Councils UK fellowships — that universities could apply for, that would give basically government funding for five years where you transitioned over the five years from a government-funded fellow to a full faculty member. But even from the beginning, you were treated as a part of the faculty, but you didn't have all the responsibilities. So they had applied and got one of these positions at Oxford. I remember I think it was in the — a fellowship in the dark sector. I loved this name, and actually what’s quite lovely is — so I met my current husband, actually, at Oxford, and he was a fellow in history at the college that I then joined. And he recalls hearing about this new Fellow of the Dark Sector [laugh] —

Zierler:

[laugh]

Dunkley:

— who was being interviewed for this physics fellowship. And he said he thought it sounded terribly exotic, this Fellow of the Dark Sector. So anyway, it was this Fellowship of the Dark Sector. And I actually wasn’t going to apply, because I was only a second-year postdoc at Princeton. And again, I hadn’t quite thought of going back to — it was too early. I had a three-year postdoc, and this was only —

Zierler:

Year two.

Dunkley:

Year two. And so I wasn’t — I was in the middle of very busy stuff for WMAP, in the middle of analysis. I wasn’t ready to go.

Zierler:

And they probably weren’t ready to let go of you, either.

Dunkley:

No, [laugh] exactly. But Pedro said, “Apply. Because actually you fit the job description very well.” And so I did. And actually amongst that, the timing was — that was during this time — my father had actually got very sick. He had cancer. And I had to go — I went back to the U.K. quite a bit. And this job opportunity came up when he was actually very unwell, and suddenly it seemed like an opportunity to come back to the U.K.

Zierler:

Were Spergel and Page supportive of this decision?

Dunkley:

Oh, massively, yes. I mean, they encouraged me to — at the same time, an opportunity came up in the States, too, which I pursued, and I actually had an offer in the States, as well, at the time.

Zierler:

Was there a mutual hope that you would continue the collaboration with them?

Dunkley:

I think so. So two related things. One is that in the end, I deferred going back to the U.K. for another year, to finish my direct work on WMAP. But even then after, when I moved back to the U.K., I continued collaborating with them continuously. I've never stopped collaborating with them, actually. So that worked out fine with moving countries.

Zierler:

Jo, did you have teaching responsibilities when you were a fellow at Exeter?

Dunkley:

Yeah. So that began in a gradual way. So the thing I started doing from the very beginning at Exeter was tutoring undergraduates, where in Oxford there’s this system that either you have big lectures of like 100 people, or you have these tutorials of like between two and six people. And so from the very start — I didn't do, to begin with, those big lectures, because of this special fellowship I had, but I did do those small tutorials, where I taught these small groups of Oxford undergraduates. So I did start doing that from the beginning. But again, actually, a lower level than would have been my eventual job load, which made things quite more manageable.

Zierler:

What other research projects had you taken on at this point, as you were thinking beyond WMAP?

Dunkley:

Well, I got really enthused about — one of the things that I had got — two areas. One is that one of the biggest challenges to looking at the microwave background is you're sitting in the Milky Way galaxy, and you've got to look through it, [laugh] and it’s emitting lots of millimeter light. And so a thing that I had really developed and learned during my postdoc and found fascinating was the astrophysics of the emission itself. Like, what’s producing it, and then methods for separating the signals from the Milky Way from the signals coming from far beyond. And so I sort of was developing that area with the thought of thinking of future missions, future experiments that would be trying to look for even more subtle distant light signals, and then how one could separate — how one should best understand the Milky Way emission and separate out the signals. And then I had also got involved — I was getting involved at the end of my postdoc with this new project, the Atacama Cosmology Telescope, which then has sort of been the thing I've worked on for years. It was just starting. And actually it was quite nice — so I actually — when I went to Princeton, I had the opportunity at the beginning to either work on WMAP or to work on ACT. There were these two sort of possible positions, focuses I could have taken.

Zierler:

And how did Atacama become possible for you? How did you become aware of it?

Dunkley:

Well, so, Lyman Page was leading it. That was his other big project in Princeton. And David Spergel as well, actually. They were beginning that. And so when I was in Princeton initially, I think it soon became clear that most of my time would be working on WMAP. But by the end of it, it became clear that I could start getting — that this was an interesting project to start being involved with. And so I actually went to Chile in my final year, or in 2007 — I guess my second year of Princeton — as it was being constructed, actually. We had a summer school — I was again useless as a — I was a theorist; I couldn't really go and be a useful experimentalist. But we had a summer school down in Santiago, that a number of us went to. And I really wanted to go and see the telescope, so I arranged to actually give some talks — again, to the local schools in the town, where the telescope is, to tell them a bit about the project. Because I partly wanted an excuse to go and see it, and also, again, this kind of was connecting to my interest in science communication.

Zierler:

Jo, in what ways was Atacama a return to cosmic microwave background for you, and in what ways was it a new endeavor? New questions and new theories to consider.

Dunkley:

So the Atacama Cosmology Telescope was still a cosmic microwave background experiment, but was trying to tackle different problems. So one of the things it was trying to do was to look for — so this light that we measure comes from billions of years back, but it actually travels through — travels past a lot of stuff in the universe on the way. And one of the things it does — it scatters through galaxy clusters. So big, huge clusters of hundreds of galaxies, the hot gas in them scatters the microwave background light and shifts its overall intensity spectrum, so that actually you can — they show up as these little spots in the map, in the map of the CMB. So one of the big goals of ACT was to use this CMB now as backlight to look for these galaxy clusters, as well as also trying to look at finer details to do things like understand neutrino physics and other properties of the early universe that you just couldn't see with WMAP. So these kind of new things were appearing that we had to now understand. And so the fact that it has this — the key difference of this new experiment was it was high resolution. It’s a six-meter telescope, and so you can now see the sky at greater resolution.

Zierler:

Jo, what are some of the advantages and limitations of a land-based telescope versus a space-based telescope, as it relates to the specific research questions you're after?

Dunkley:

There’s a couple of things. Just by being on the ground, you've got water vapor above you — even in Chile, even in the desert, the driest place on earth, you've still got some little bits of water vapor, and that masks some of the background signal that you'd like to see, and it does it typically on very large angular scales in the sky, because it’s kind of coming from big coherent not quite clouds, but equivalent of clouds in water vapor. And so you can’t really see very large-scale features in the sky. But then again, that wasn’t so bothersome, because we had measured those with WMAP, and the big advance is you can see with fine detail. The two other limitations are there are some wavelengths that you just can’t measure from the ground because of the water — because of the atmosphere. And so you’re a little bit more limited in how many wavelengths you can survey the sky at, which makes it a little bit harder to separate out like — again, light from our galaxy, also light from many other galaxies. This telescope is at high enough resolution that other galaxies throughout the universe, if they're emitting in the millimeter, you can now see them as well, and they pop up as little kind of dots or little subtle signals. So anyway, there is a great challenge of separating out all those signals because of the limited wavelengths you can actually observe from the ground. You also can’t just see the whole sky from Chile. You can see kind of — you can see half of it, but even then, we were kind of focused on just looking at a little patch of it, rather than surveying the whole sky.

Zierler:

What were the origins of your involvement with LSST? Was this happening right at the same time?

Dunkley:

No, this was a little bit later. Yeah. I'm trying to remember now. [laugh] Actually, my great enthusiasm for LSST — I was increasingly getting — so during my time at Oxford, I was increasingly interested in large-scale structure probes, so probes of the late-time universe, the galaxies that you can see with optical telescopes, rather than just stuff you could see with a CMB telescope. Because I realized that to answer some of the big questions, particularly about dark energy, and also things like neutrino mass, you just have to go and look at the universe later on — you have to see how it evolves. You can’t just look at it early on, or even these subtle effects as a backlight with the CMB. So it was clear to me that one needs those probes, too. And actually, again in Oxford, Pedro Ferreira, who was now my close colleague — he had gone from being advisor to my close colleague — he was very interested in testing modified gravity and really using large-scale structure probes in a big way. So I think the group in Oxford, we were really interested as a group in using these galaxy probes. So even though I was keeping up my sort of CMB program, as a group, we were thinking a lot about large- scale structure. Now, many people were also working on the Euclid satellite, which will be flying soon from Europe. But I ended up getting most enthused in LSST through Ian Shipsey, who’s now the head of Physics at Oxford, who moved to Oxford from Perdue and really wanted to bring LSST activity to Oxford. It became quite exciting, thinking about — and he was moving to Oxford with an experimental background from LSST. His group had worked on the detector side of the LSST camera. And again, with my enthusiasm for kind of trying to — enjoying — I liked the connection of the experiment to — like building an experiment to then understand the data. So this sort of idea of having local expertise on the people building the detectors for LSST and then us — and then thinking about how to do the cosmology seemed exciting. So we sort of built — yeah, we decided to increase our involvement in the Dark Energy Science Collaboration that’s part of LSST in Oxford. And I think that was largely inspired by Ian Shipsey coming to Oxford. And again, but it fitted in very well with our group in Oxford at the time wanting to combine our work on CMB with galaxy surveys. It’s quite nice to see now that lots of the postdocs who were in Oxford at the time have now gone on to take on real leadership roles in that collaboration. And some of them, I advised. Many of them, Pedro advised. But that was something that I — it’s nice to look back on, and see all these young, junior people who are now kind of playing leading roles in that project.

Zierler:

Jo, at Oxford, is tenure conferred at the associate or the full professor level?

Dunkley:

Oh, it’s weird. Sort of neither. So you basically have a review — this is again — in hindsight, I had a very lucky career. You have a review after five years, but really the spirit is that from when you get a job, there’s not a sense that you're going to lose that job.

Zierler:

So it’s the opposite of a place like a Harvard or a Stanford, where there’s not a culture of promoting from within.

Dunkley:

Yes, yes. The sense is, you've got to do something — maybe this is not necessarily still the case — but there was a sense that you've got to do something really wrong [laugh] to not keep your job.

Zierler:

And so that would foster a culture for junior faculty that they were really supported by the department, by the senior faculty.

Dunkley:

Yeah.

Zierler:

Because there was the assumption that, “We want you to succeed. We want you to stay here and grow your career.”

Dunkley:

Absolutely. So in hindsight, I was so lucky, because I got my job at Oxford in 2007, when I was — I can’t do the maths now, but I was 27, I think? And that was really — even though it wasn’t like guaranteed permanent, it was kind of a permanent job. And so that reduced — removed a huge amount of stress from my life.

Zierler:

Were there any difficulties in managing your relationship with Ferreira, now that you're grown up, so to speak?

Dunkley:

Do you know what? I think we didn't really sort of decide upon it, but we basically became very close colleagues, but working on different projects. So actually just recently we published a paper together with our collaborators, but we didn't work on the same things. I had really developed into working on, at that time, really the CMB analysis projects, and he was working on more modified gravity and other theory projects. With data, but different ones. And so it worked out well. I think we sort of knew that we needed to have our own things. He probably knew that I needed to have my own thing. [laugh] And so it ended up working very well, because I had sort of my students and postdocs, and he had his. But we were a very tightly coupled group. We had joint group meetings every week, and all the kind of administration we did together. We’d jointly search for postdocs and recruited students and stuff. So we did that together, but the actual research papers and projects, we did separately.

Zierler:

Jo, had you kept up with your public science interest at all, in England?

Dunkley:

Yeah. I did keep that up. So a big thing I had done in the States had been to teach this course for middle school teachers. Again, actually funded by — it was actually the public outreach project for WMAP, funded by NASA, that I then did — I sort of did full-time for a few weeks — to teach this course. And that material gave me loads [laugh] of material to then sort of go back into schools and give talks, when I was back in the U.K. So back in the U.K., I started doing more media work as well, like doing radio interviews, TV things. And that was something I quite enjoyed. I did a wonderful training session with the Royal Society on how to do public science. I can’t think of — I'm not saying it right — media training. It was media training by the Royal Society. I found it incredibly valuable, and I then took up more opportunities to do things. They basically said to us, “If someone gives you a chance to talk about your research for two minutes on the radio, you take it. Don’t feel like you're not the expert.” So I started also doing more interviews with journalists, talking about — commenting on results.

Zierler:

What have been some of the most important things that you've wanted to convey to a lay audience about your research and about how we understand how the universe works, generally?

Dunkley:

I think a big message is that we have this story that connects from a second after the Big Bang to today. Like we basically know how we came to have a universe filled with galaxies and stars today.

Zierler:

Meaning these are not just theoretical abstractions. You want to convey that point.

Dunkley:

Yeah, we have the data, and we can trace it through. So I wanted to make that point. Exactly. Because I think there’s a sense that cosmology is all just philosophical ideas. So this idea that we actually have measurements that we can answer — but also the simplicity of it. Like that we've discovered that one doesn't — of course there’s a lot of complexity [laugh] in the universe, too, but the overall idea of it being quite a simple — there is this simple model that works — I try to convey that. But I think to me also, it has been also just as important to convey the idea of science being exciting. That to me is just as important as the content.

Zierler:

Jo, I want to ask a question that bridges the transition from Oxford to Princeton. I love how on your CV you have proudly listed that you took two maternity leaves. It’s sort of like you're implying there, “I was a mom during all of this, too. I have a personal life. There’s other things that are happening. And taking time away from the sciences is important.” It’s just a great implicit message on your CV that that’s part of the story, as well. So I want to ask, in comparing Oxford and Princeton, since you took maternity leaves at both institutions — in very different ways, Oxford and Princeton are very, to state the obvious, old-school kinds of places. I'm not sure about Oxford, but legendarily, the Princeton Department of Physics didn't even have a women’s restroom until, you know, not so long ago. So just in terms of painting a broader picture about the commitment to inclusivity and diversity at both institutions, what was your experience in terms of taking maternity leave, in terms of the kind of support that you got administratively, and from your colleagues in saying, “You know what? I'm a woman. I'm a scholar. But I'm also taking a pause from everything, because that’s personally important to me” in a way that most men — that would never even occur to them, when they have children. So I wonder if you could compare and contrast your maternity experiences at both institutions.

Dunkley:

They were different. So the first — so at Oxford, I felt very comfortable taking the leave, and that it was accepted. I had actually already gone through the process of supporting a postdoc of mine to take her maternity leave, so I knew all the policies pretty well. And I knew the policy was that you can take six months paid leave, no question. You can take longer than that, too. And there was really an acceptance of leave — both maternity, and thankfully increasingly paternity leave — is important. And actually again, I had a good model from Pedro Ferreira, because he had been a single dad, actually, when I was a PhD student, and fitting in parenting with research had been just integral to his career, and so I saw that firsthand. And so he and other colleagues were very supportive of me taking time off, taking the leave. But the nice thing there was the leave was standard. It wasn’t weird that I’d be taking at least six months, if not more, off. That was just normal. And my colleagues were very supportive about helping — again, at the time, I had this grant, this European Research Council grant, which also made a huge difference to my life and career, because it was five years of funding. And again, during that five years, I had much less teaching and administration than usual. And I had my first baby during that grant period, so that again made it even easier, because I sort of wasn’t actually having to sort of let — not take on so much — what am I saying? Because I didn't have that much teaching and administration anyway, then taking this break wasn’t in a way such a big deal. But my colleagues and including David Spergel was very supportive. He and Pedro and other colleagues really helped me figure out how to manage my group at that time. Because one of my postdocs went to Princeton for three months to work with Spergel instead of me, right? So he was very supportive in terms of making sure I could then not have to worry about all the members of my group not having direct supervision. And the same with Ferreira and others at Oxford. So in the end, I didn't have too much of worry about having to do stuff right away. And everyone was very supportive about not getting back into it right away. Like the one thing I remember I did during my first few months was to read postdoc applications. But that’s because I wanted to be involved in that. No one sort of said, “You must do anything.” So I found that process quite good and straightforward, and I gradually starting seeing some of my — for example, one of my graduate students started to come round to my house for our weekly meeting, after a few months. I’d had the baby, and he’d come, and we’d have our weekly research meeting in my house, with the baby. And I think that worked out fine. But apart from that, I didn't do — I didn't meet other people. I met one postdoc occasionally as well. But I also felt like it was — and I will say, I felt it even more at Princeton, but I did feel that it was part of my job to show that one could take leave and it would be fine. I did really feel like that was — as a role model to other people, it better work. So I think that really influenced as well this thing of like, “I want to do this properly.” And I set an autoreply. I had an autoreply on my email saying, “I'm on maternity leave. I'm not available.” [laugh] In Princeton, it was harder. I was militant that I was taking leave.

Zierler:

[laugh]

Dunkley:

And actually, I got pregnant very soon after moving to Princeton, and I — we were planning a second child, so in negotiating, in discussions, in terms of moving, it was clear that I was going to take a leave, or I would like to take a leave. But I found the policies much less welcoming than in Oxford. The official policy in Princeton is you get teaching relief for a semester, as a new parent, but formally, you're supposed to continue at your other duties. Somehow. Now, no one made me do that —

Zierler:

Jo, to what extent do you read into these differences as being specific to the institution, and how much of it is broader cultural approaches in England and the United States?

Dunkley:

I think it’s mainly — I think the latter. I think it really is cultural. I think Princeton was very — I don’t think Princeton is unusual in its approaches, and certainly no less supportive than others in the States. I just think it is a cultural thing. In the States, people go back to work. I saw a lot of academic parents back at work after three months, which to me just seemed — it just seemed alien. And that’s fine, and many of these people wanted to be, and I saw lots of colleagues — I had a colleague — colleagues who were having children about the same time, and they wanted to be back at work. And I just didn't. I just didn't. So I had to be much more militant to myself in Princeton that I was taking leave, and I would do it. Because the default would have been not to take that much leave. But I never felt unsupported in that. I just was doing something — but I also found — I did actually find it quite difficult, that like I knew that while I was taking all this lovely leave, that this wasn’t available to all my junior colleagues, because they didn't all have the opportunity to take eight months off. So I definitely — it’s something that I'm very aware of, and would love to see changed, is those policies, in general. But for my own experience, having done it the first time, I made sure to take the proper time again. But I think if I had done my whole career in the States, I probably would have had a different experience.

Zierler:

Jo, circa 2015, 2016, are you thinking about moving on from Oxford? Did Princeton sort of recruit you out of the blue? How did that all come together?

Dunkley:

There were some things about my job in Oxford that I found challenging. And I will say, the benefit of taking maternity leave, particularly that first one, is it made me really evaluate which parts of my job I enjoyed and wanted to do. Because I suddenly was in this situation where I was like, “I can either spend time with my lovely baby, or I can do my job. And if I'm going to not spend time with my lovely baby, I’d better really like my job.” [laugh]

Zierler:

[laugh]

Dunkley:

So it really made me make choices about my job. I actually stopped working on one project, because I realized I wasn’t willing to do that — that wasn’t a good reason to go off to work, to work on this project. But it also made me reevaluate, again, the balance of my work. And so I was coming up to applying for a second European Research Council grant, but the rules in my college at Oxford were quite strict about the fact that you couldn't — I basically like spent a lot of years not doing my full amount of teaching, in my college. I had used research grants to sort of — it’s called buying out your teaching, where you use the grants to fund someone else to do some of your teaching. It soon became clear that that wasn’t going to be an option for much longer, and I simply couldn't see a path where I could have a young family, do my research, and also have this responsibility in that college in Oxford. I just couldn't do it. And so I started to explore options of adjusting my job a little bit. But in doing that, I also started looking around for other opportunities.

Zierler:

I don’t want to burden you with a discussion about all of the amazing awards that you've been recognized with —

Dunkley:

[laugh]

Zierler:

— but one juxtaposition that’s irresistible is the Breakthrough Prize in 2017, and then you being named an officer of the Order of the British Empire, which is like —

Dunkley:

[laugh]

Zierler:

— it’s the best old-world, new-world juxtaposition that I can think of, right?

Dunkley:

[laugh]

Zierler:

So I wonder if you can reflect a little broadly on what the significance of being recognized with one of the oldest institutions in human civilization, versus, like, the epitome of startup culture — "This isn’t the 19th century anymore, this is big tech, we're thinking about the future, we can do what we want, we don’t care what the Nobel Prize Committee says.” Right? I wonder if you could reflect broadly on what it’s like to be recognized in both of those very different worlds. And beyond how nice it is to be recognized, how that’s useful for your own research agenda and your work as a science communicator and as a teacher.

Dunkley:

It’s true; it is quite a contrast. I think that — well, let’s see. So the Breakthrough Prize — I do see them as — they're very different things. One thing I think — however old the British institution is, the sense of that is that — I was awarded it for science, but it’s coupled with service. Of course I don’t know all the details, but I'm pretty sure that that award was connected to my work in public science, coupled with my research. It’s not disconnected. Whereas with the Breakthrough Prize, the New Horizons Prize, that’s pure research. I don’t think there the issue is at all the larger [laugh] — it doesn't really matter what else you're doing with your life and time; they just care about cutting-edge research. So I do feel really — it’s nice to feel that like both things are important. And actually I had that a little bit before with this Rosalind Franklin Award that I got from the U.K., where again it was supposed to be for both things, like doing public science and research. It’s very nice to think that both things are valued. But yeah, the new — I do love the fact that the Breakthrough Prize is trying to publicize — it’s trying to make science cool, you know? [laugh] — it’s trying to change the image of what is a scientist, of what being a scientist is. And I do think that’s really valuable, because again it’s pushing — it’s trying to really, I think, change attitudes and get more people, younger people, doing science. So in a way they're maybe not completely disconnected, because again, the awards from the U.K., even though they're ancient, they're also trying to say, “Look, it’s important to share science with the community.” So in different ways, I think they're each trying to say, “Look, let’s try and encourage more people to get into science.” So even though they're quite different institutions, maybe in the end they're sort of after the same thing.

Zierler:

Jo, as a sign for the next generation in cosmology and astrophysics, what are some of the most exciting projects that your graduate students have done in recent years?

Dunkley:

They're doing wonderful things, and it is one of the things I love most about my job, is working with the graduate students. So that includes coming up with — well, okay, so one of the areas is trying to figure out how to take full advantage of computer simulations. So right now, we have data, we have images of the sky, and we often extract kind of statistics from them and then compare those statistics to theory. And really the field is moving maybe more in a direction of just numerically simulating in our computers many versions of the universe and extracting far more information from those simulations than we would have done originally with just old-fashioned statistics. So I think one of the things I do see as an exciting frontier is more use of large-scale computer simulations and moving towards machine learning techniques to like pull out more information from surveys than we have before. So that’s one area that I'm excited about seeing my students work on. Then another thing is — even just looking at — I love to see the students see new data for the first time, and be trying to — for example, like looking at our data from the Atacama Cosmology Telescope, which is what I'm working on now, some of my students — seeing the students being able to turn like — not meaningless data, but raw data — into a new statistic that then gives us a new answer about — for example, we've been trying to resolve this problem of how fast the universe is expanding. It has been a big sort of controversy in astronomy about what the Hubble constant is, like how fast the universe is growing. And so seeing students being able to actually weigh in on that answer, and actually be able to contribute to these big questions with our results, to me that’s — I love seeing that in the students.

Zierler:

And on the undergraduate side, what are the favorite classes that you like to teach undergraduates?

Dunkley:

That’s a good question. And it has been quite different at Oxford and Princeton. I have to say, I do actually miss quite a lot doing the tutorials that I used to do in Oxford.

Zierler:

And even a small undergraduate seminar here does not quite approximate that experience?

Dunkley:

It doesn't. And one of the differences is also the — well, it could do, but it just so happens that in Oxford, I would be teaching undergraduates who are purely doing physics. We mentioned this before — they would be majoring only in physics, and that’s all they would be doing with their whole week. So I would see them after a week where they had spent like their whole week working on quite challenging problems that we would then go through together. They’d have spent a few days getting into the thick of these problems, and then we’d go through the hardest bits together, and show each other how to do the problems — have some students show others. It was a style of teaching and working that I liked. Now, I think that actually may be — I'm sure there is plenty of that in Princeton, but since I've come to Princeton, I've mostly been teaching non-majors. So I've been teaching physics to non-majors, mostly electromagnetism to typically engineering-level majors. And also astronomy and cosmology to real non-majors, so people who would be typically majoring in the humanities. Now, I've enjoyed that, because it sort of tapped in again to more, I would say, my public science work. I've been using material from my book to teach the undergraduates, and I like that. But, the reality is that since I have moved to Princeton, I've just happened to have taught less to the majors, and so I sort of — I miss that level of working through the gritty physics problems.

Zierler:

Jo, besides simply leading by example, I wonder if you specifically see yourself in a mentor capacity and a role model capacity to undergraduate women, who may or may not be thinking about a career in science, but who you might have opportunities to exert positive influence and encouragement.

Dunkley:

Yeah, I hope so. I hope so. And I've seen that increasingly as my role since moving to Princeton. I think before I came here, it was less sort of something that I felt was a part of my job. And I feel it is a part of my job now, is to be that person. Not the only person, but you know, there aren’t — particularly in the Physics Department in Princeton, there are very few women. There aren’t that many in astrophysics, either. So I think that just being visible [laugh] is really important. It’s also connected to wanting to take proper maternity leave with my second child. It’s like I want to show that I'm doing it. I want to visibly be — I was glad to teach a class, heavily pregnant. I think it’s good for undergraduates to see that you could be taught a physics class by someone who’s nine months pregnant. So yeah, I do think that. And actually I noticed — I had a wonderful — I was just teaching this past year the intro physics class, and I did have a lot of women in my precept. And yeah, I hope that I at least encouraged them that they can be taught by women as well as men. So yeah, I see that as part of my job. The convenient thing for me is that I can do that just by existing, which [laugh] — but I also do take efforts to mentor people. I would say I probably mentor more at the graduate level. I think for the undergraduates, again, I try and be visible, but I think I more proactively take a role in graduate students and postdocs. I've really enjoyed trying to help postdoc women particularly navigate having children of their own, and balancing careers and their lives. I've taken a pleasure in that.

Zierler:

Besides just the administrative distinction or decision to be dual-hatted between physics and astrophysical sciences, what does that say both about your research and about the division between these two departments? What’s the takeaway there, just in terms of understanding the science and how it plays out at an administrative level?

Dunkley:

Well, it’s funny; I can see how it has ended up happening. It’s like those who build the experiments that I work with are almost always in physics departments. The knowledge and expertise to even build them and to understand the technology, to work on the technology, fits better in physics.

Zierler:

So the world of experimentation is located largely in the physics department, is the idea.

Dunkley:

Largely. Certainly at Princeton, in my field, like all of my experimental colleagues are in the physics department. Now, the science gets interesting, because I don’t think of myself as interdisciplinary, right? [laugh] I think of myself as doing something that’s very well-defined. Without the historical background, I wouldn't expect it to be in two different departments. And actually then my background is that I came through Oxford where astrophysics was a branch of physics. And to me, that’s actually how I see it. I think of it as a branch of physics. I think of myself as a physicist who specializes in astrophysics. That’s how I see myself. But when it comes to science, certainly there is a sort of division of like, if your questions are really what is the physics of neutrino particles, or what is the physical nature of dark matter, or what is the physical model for how the early universe evolved, those are traditionally more physics questions. Whereas if I then think about perhaps questions more related to galaxy evolution, like how did the — you know, yes, we see the evolution of the universe, and we've got these fundamental physical laws, but how did that evolve, and how do stars and gas and dust play a role — then we kind of veer into more astrophysics. And certainly an area where it certainly would be astrophysics is — I told you I think a lot about how to model and think about the emission from our Milky Way galaxy as a contaminant for this light that I look at. Well, understanding the emission of dust grains in the Milky Way and how observations tell us about that, that falls more squarely into astrophysics. I have to say, in Princeton, I'm happy to be in both departments, because if I weren’t, I wouldn't get the benefit of both of them, and what I do really does fall into both.

Zierler:

Although a 50/50 appointment probably means 150% responsibility.

Dunkley:

Yeah. Administratively, I wish it was one department.

Zierler:

Jo, just to bring the narrative up to the present, what have you been working on in the last few years? What have been the most exciting projects for you and your collaborators?

Dunkley:

I’d say that one of the things that I really got obsessed by, and we have been working on lately, is this question of the Hubble constant. Now, the Hubble constant is a measure of how fast space is growing, and of course I would like to know how fast space is growing. But the real interesting thing is, these two different measurements were giving different answers. One was coming from the early universe, and one was coming from looking directly at galaxies moving away from us. And they were getting different answers. And that was a sign that perhaps our underlying physical model for the universe was wrong. So we've been talking about — we have this standard cosmological model, Lambda-CDM, and we know there are problems with it, because we know we don’t understand dark matter, dark energy, the early universe model. But we haven't been able to shake it hard enough to see a real problem emerge. And this was the first — or this was a hint — that a problem had emerged. And so we knew we couldn't resolve the problem completely, but one of the things we set about to do was to basically check if this one measurement from the other universe was right. You know, you always worry about subtle experimental systematic effects that could be subtly giving you a different answer. And so we sort of set about trying to do this with our experiment in Chile. And it took a long time! It took longer than we thought, because the analysis of the data was really hard. But we got there, but it has really occupied a lot of my time over the last few years, coupled with thinking about theoretical models that could actually be explaining the difference. And so, in this past year, that has sort of dominated what we were doing. We had a great excitement earlier this year, in February, when we sort of unblinded our results, and we got our number, and it was like bang-on the number that had come from the Planck satellite, that people were wondering whether it was really right. And we're like, “Okay, this number is [laugh] — this number is right.” So that was exciting. But then coupled with this opportunity to then really try and test these possible alternative models even better. So, yeah, that work has really dominated my past couple of years.

Zierler:

Jo, where is gravity in the standard model of cosmology? I'm thinking of the problem of gravity in the standard model of particle physics. Where is it in cosmology?

Dunkley:

Well, do you know what? Cosmology comfortably includes general relativity. So everything we do in cosmology starts with general relativity as the description. Now, we do run into trouble right at the beginning where we put in fluctuations that we think connect to quantum mechanics, and we don’t yet have a good way of connecting those two up. But ultimately, when we can evolve — when we track how things should behave, we're assuming general relativity, and that sort of tracks through to then figuring out how things should evolve together, and yeah, how gravity works. So that is kind of embedded in what we do. And again, only when we push up against things like black holes or the early fluctuations in the early universe do we start having some trouble reconciling quantum mechanics and relativity. So that’s definitely in there. But yeah, the other thing I've really enjoyed that has taken up quite a lot of my time is that we've been designing this new observatory, the Simons Observatory, which is basically the next step between the Atacama Cosmology Telescope, that will be actually just next door to it in Chile. And I, I guess back in 2016, when I moved to Princeton, took on the role of — I was chairing the sort of theory and analysis committee, the Science Committee, to really kind of try and identify what we should build to answer our most interesting questions, science questions. And so one of the things I'm really excited about is using cosmic universe measurements to actually try and weigh the neutrino particles. I think it’s really cool that you can actually use the universe as a lab, and actually try and learn fundamental particle physics from these measurements. But that’s among many other things. So that really dominated my work for a couple of years as well, was trying to pull together different things — figure out what science we wanted to do, and what we needed to build to get that to happen. And it’s neat, because now, we're building it, and it will be on the sky in 2023 or so. So that was the big part of my last year, certainly.

Zierler:

Now that we've worked right up to the present, for the last part of our talk, I want to ask sort of one really big science question, and then a sort of forward-looking into the future kind of question. Thinking about CMB and WMAP, your research wheelhouse is in this very specific place in the early universe. I want to ask both — to the extent that we can understand the universe as a historical chronology, if I can ask you to explain how your research helps us understand before that period — you know, all the way back to T=0 — and then to take it all the way forward to cosmic eschatology. So I wonder if you could reflect broadly on what your involvement with CMB and with WMAP tells us about both the very beginnings and the very ends of the universe.

Dunkley:

It’s sort of easier to go backwards than forwards. So this image that we see of the universe at 400,000 years, we can actually work backwards quite easily, because the physics is linear at that point. Like we're looking at subtle features that are like one part in a hundred thousandth of the sort of mean density. And so, the universe was actually so simple back then, and so relatively unperturbed, that we can actually numerically then trace backwards to say what must have been the physical scenarios very early. Like, you know, in the first fraction of a second. Now, it doesn't mean that we have completely resolved that. We're still looking for new features and all the new scalar fields that could have been coming at the beginning, or new subtle effects that we haven’t noticed, But ultimately, it’s quite a big part of what we do is exactly to — we are numerical — our theories and our codes start at almost time zero, and then they evolve forwards, and we say, “Okay, what would I see at this particular epoch?” So evolving back to the beginning is definitely part of what we regularly do. Evolving forwards — I guess two things. One is that if you even just know what’s going on at 400,000 years and you just say, “I'm just going to let gravity do its job,” you do end up [laugh] forming cosmic structures, and you end up seeing how things will evolve through to today. Because of course what happens later is it gets non-linear, and it gets much harder to interpret, as big cosmic structures form. And that has partly influenced my interest in then not only working on the early universe images, but then also increasingly being involved in the large-scale structure surveys themselves, because they have so much more information about the non-linear behavior and how gas traces the dark matter in a way that you can’t predict from just looking at that early epoch and evolving things forward. So tracing the early epoch forward can tell you big things, like it can tell you how fast the universe might be growing today, whether it should continue expanding forever, whether it should collapse down or keep going. It can answer those big questions, but it can’t completely answer all the questions about how have galaxies evolved, and how has our Milky Way evolved. So yeah, there are things that they can’t answer, and that’s why you need the actual optical images, to get those answers.

Zierler:

I'm just thinking about where you are right now — is string theory something that’s compelling to you? Is this something that you think can be relevant at some point in the future to the kinds of questions you're asking?

Dunkley:

Yes and no. One thing that I keep in the back of my mind is that we have a major problem in cosmology, that there’s these huge questions we haven't answered. And so, okay, it might be that there are relatively simple answers to them, that there is a dark matter particle that we're going to find, and that we'll understand why the vacuum energy is like it is. But I do sort of feel there is a bigger paradigm shift that we have to get to. There’s something new and bigger that’s behind why there are these such big questions remaining. And so that tells me that we need — we're not done yet. Like we do need something beyond the physics that we have, even if it maybe doesn't impact — even if maybe I can’t find all those hints or signals, test it with my data, I know that in the back — in this bigger program of trying to figure out what’s going on, we need something beyond what we've got already. I'm not convinced that string theory is it, but we better be working on that, and working on other possibilities. Because we're certainly not done. It would be kind of naïve to think that we have our kind of fundamental physics sorted. So yeah, I'm not desperately enamored — although I will say string theory — Brian Greene’s book about string theory is one of the things that got me back into doing a PhD. So I read his Elegant Universe in a desert in South America, backpacking, at the end of my degree, and it really — it got me thinking. [laugh] So that’s my main connection [laugh] to string theory. Now, I don’t — I have less of a connection to it.

Zierler:

Jo, I want to ask, for my last question — the theme of our talk has been — I mean, there’s just so much that has been exciting throughout your career, and you've been — whether it’s serendipity, it’s talent, it’s being in the right place at the right time, you've been so central to all of these exciting developments. I want to ask specifically, because there’s so much fundamental work, so much exciting discovery to be had in the future — we don’t even barely understand these things, but we see that there’s a map to get to what we need to know. I wonder if you could think specifically about the interplay of experimentation and theory, as it pertains to cosmology and the kinds of questions you're after. Historically, sometimes the theorists are leading, and the experimentalists come in, and they make the data, and there’s that interplay in that way. Where do you see yourself as you work with your colleagues in experimentation, in terms of broadly pushing the field forward?

Dunkley:

That’s a good question. Let me think. I do think that right now in cosmology, we are lacking a little bit maybe the theoretical inspiration to come up with really new ideas. It is happening; I'm just not sure it’s happening as fast as our data are improving. I think that we're making significant progress in testing. There are things that we do now — there are theories or models that we do know that we can test with the data, but I do feel like the wealth of data that has appeared through experiments in the last ten years, 20 years, and that’s going to come in the next ten years, I do worry that we won’t have — that we don’t have enough theories to keep up with it. Although I feel that that’s maybe changing with dark matter. I'm seeing interesting ideas come into — anyway, since we haven't seen dark matter produced at the Large Hadron Collider, we're now seeing what interesting novel things come up, like very light particles, fuzzy dark matter, things that we didn't think of before. So I'm seeing that happening. But we maybe need to have some bigger theoretical ideas, to almost keep up with the advances in the massive data. But I do think we're also — what’s going well, I think, is that there is a good, tight coupling between the people designing the experiments and doing surveys, and then how to interpret it. And a lot of that does come down to just — the theoretical work comes both into like analyzing the data, thinking of ways to do it, but then also coming up with the fundamental theories. So I think there’s a good coupling between the ways that we pull out information from the experiments. But yeah, maybe we are still missing some big inspiration about the ultimate theories.

Zierler:

And is there a single question mark for you that serves as the greatest motivator, in terms of the things that you're most curious about? The things that you feel in the course of the future of your career are really attainable?

Dunkley:

Well, okay, there’s one thing that I think is attainable that I think will be really fascinating, and we're going to see. We're going to figure out the mass of neutrinos. Neutrinos are such a mysterious sector. We don’t know why they have mass. We don’t know what mechanism gives them mass. There’s much we don’t know about this whole sector. And we really are, in cosmology, going to detect their mass. Unless something weird is happening, which is also interesting, that’s going to happen in the next ten years. So I often turn — to me, like there are many things that I don’t know if we'll be able to measure, but that is actually — that’s something that we will get. So really to me that’s exciting, and I'm fascinated to see how it will play out between our cosmology field and the particle physics world, where we'll have to convince our particle physics colleagues that we've really done this measurement right. But I feel on a bigger — that will be so interesting to me. And particularly if it reveals unusual things about this sector of particles. But I feel — the other area — I just — those are the neutrinos. They're part of the dark matter. I want to know what the dark matter is. [laugh] Like it’s so — it’s there, and it’s just like —

Zierler:

That’s it.

Dunkley:

You know, it has been years now [laugh] it has been there, and it has been a long time. And I just — I worry that it will be 100 years from now, and we still won’t know what it is. I'm so fascinated to see if ten years from now, we'll understand it. And I love reading about, you know, Fritz Zwicky naming it, so many years ago, and here we still are. I'm sort of amazed; I look back on some of my research talks and proposals, and I gave — the job talk I gave to get my job in Oxford — and I was like, “In the next five years, I'm sure we'll have a hint of what dark matter is.” [laugh]

Zierler:

[laugh]

Dunkley:

You know?

Zierler:

Well, Jo, on that note, we'll have to stay tuned and see if ten years is feasible, and if not, we'll see what it takes beyond that. Jo, it has been so fun spending this time with you. I want to thank you so much for doing this. It has been really tremendous to hear your perspective and involvement on all of these things. And on behalf of everybody in physics, good luck in accomplishing all of these things. Because it’s quite exciting indeed.

Dunkley:

[laugh] Thank you so much. Thanks for having me.