John Thorstensen

Notice: We are in the process of migrating Oral History Interview metadata to this new version of our website.

During this migration, the following fields associated with interviews may be incomplete: Institutions, Additional Persons, and Subjects. Our Browse Subjects feature is also affected by this migration.

We encourage researchers to utilize the full-text search on this page to navigate our oral histories or to use our catalog to locate oral history interviews by keyword.

Please contact [email protected] with any feedback.

ORAL HISTORIES
John Thorstensen

Credit: Eli Burakian

Interviewed by
David Zierler
Interview date
Location
Video conference
Disclaimer text

This transcript may not be quoted, reproduced or redistributed in whole or in part by any means except with the written permission of the American Institute of Physics.

This transcript is based on a tape-recorded interview deposited at the Center for History of Physics of the American Institute of Physics. The AIP's interviews have generally been transcribed from tape, edited by the interviewer for clarity, and then further edited by the interviewee. If this interview is important to you, you should consult earlier versions of the transcript or listen to the original tape. For many interviews, the AIP retains substantial files with further information about the interviewee and the interview itself. Please contact us for information about accessing these materials.

Please bear in mind that: 1) This material is a transcript of the spoken word rather than a literary product; 2) An interview must be read with the awareness that different people's memories about an event will often differ, and that memories can change with time for many reasons including subsequent experiences, interactions with others, and one's feelings about an event. Disclaimer: This transcript was scanned from a typescript, introducing occasional spelling errors. The original typescript is available.

In footnotes or endnotes please cite AIP interviews like this:

Interview of John Thorstensen by David Zierler on June 9, 2021,
Niels Bohr Library & Archives, American Institute of Physics,
College Park, MD USA,
www.aip.org/history-programs/niels-bohr-library/oral-histories/46812

For multiple citations, "AIP" is the preferred abbreviation for the location.

Interview with John Thorstensen, Professor of Physics and Astronomy at Dartmouth and President of MDM Observatory. Thorstensen narrates the story of George Wallerstein bugging his thesis committee at Caltech and he discusses his current work on cataclysmic binary stars. He surveys all of the exciting developments in the world of observational astronomy and he explains some of the administrative considerations in the way Astronomy and Physics is divided at Dartmouth. Thorstensen describes his undergraduate education at Haverford, where he pursued his interests in astronomy, and he explains his reasons for attending Berkeley for graduate school to work with Stu Bowyer on X-ray sources. He discusses the opportunities that led to his faculty appointment at Dartmouth and he explains his increasing involvement with the MDM collaboration. Thorstensen describes his formative collaboration with Joe Patterson and the evolution of Astronomy developing into its own discrete program at Dartmouth. He explains the value of both photometric and spectroscopic applications and the value in using digital detectors. At the end of the interview, Thorstensen discusses his recent interests in white dwarf binaries, the contributions of amateur astronomers in making significant discoveries, and some of the surprises he has experienced during his career studying cataclysmic binaries. 

Transcript

Zierler:

Okay, this is David Zierler, Oral Historian for the American Institute of Physics. It is June 9th, 2021. I am delighted to be here with Professor John R. Thorstensen. John, it's great to see you. Thank you for joining me today. 

Thorstensen:

Oh, you're welcome. 

Zierler:

John, to start, would you please tell me your titles and institutional affiliations? And you'll notice that I pluralize that. 

Thorstensen:

Well, it's pretty much only one. I'm a Professor of Physics and Astronomy, which is -- we're a combine department and I'm actually an astronomer. So, I'm a Professor of Physics and Astronomy at Dartmouth College. 

Zierler:

Now, has your term as President of MDM Observatory ended?

Thorstensen:

Oh, no, no, no. I'm sorry. I guess, actually, I'd forgotten all about that. I guess, I'm also President of the MDM Observatory Corporation. 

Zierler:

I wasn't going to say anything if you didn't want to, but I figured. 

Thorstensen:

See, but the thing is, that doesn't actually pay me anything though, so I tend to not think of it as a job.

Zierler:

Very good. John, just to start, for the historical record, we should explain how we connected. So, let me just tell you, Nick Suntzeff, who I interviewed, forwarded me that hilarious story about George Wallerstein bugging his thesis defense from Alan Sandage at Caltech. So, could you just both narrate that story and explain how you came to know this story?

Thorstensen:

Well, I was there, at least when the story was told. When I was a graduate student, I was in the Bowyer Group at Berkeley, which mostly did space ultraviolet astronomy and X-ray astronomy, but also had an optical arm which I was part of. It's kind of a vestigial little optical arm. We didn't do too much with it, but it was enough so a graduate student could have projects. In any case, the Bowyer group was quite cohesive, and every Friday afternoon they would go out and have pizza and beer at La Val's just north of campus at Berkeley. We would actually have an afternoon seminar up on the hill of the Space Science Lab, and then we would all trundle down the thousand vertical feet or something, somehow, to go to La Val's for beer and pizza. And other people from the astronomy department would kind of wander through, and one Friday afternoon, George Wallerstein was visiting for some reason, and so was Allan Sandage. I don't remember if one of them had been the colloquium speaker the previous day or something like that, and was visiting for a couple of days, or what.

But in any case, they were both there, and they both went to the afternoon beer fest. So, here are these two giants of astronomy, and they're just having fun talking to each other. And Wallerstein decides that at that point is the right time to relate the story of his thesis defense. So, as background, Allan Sandage was a brand-new assistant professor at Caltech at the time. He was like Hubble's protege and considered to be quite hot stuff. But he was the most junior person on the totem pole at the time. And Wallerstein was a graduate student finishing up, of course. He was defending his thesis on some stellar abundances probably. 

Zierler:

And this would have been like, what, '54 or '55?

Thorstensen:

I suppose so. I wasn't there. I was alive at the time, but only just. So, I wasn't there, but it was probably about then. So, Wallerstein gives his defense, and Allan Sandage on what may very well have been his very first thesis defense on the Caltech faculty. And the usual process here is the person stands up and describes their stuff, and they get a bunch of questions from the committee, and then they retire to another room, and the committee deliberates. It's almost like a jury. “Do we pass him? What do we say? What did you think?” Stuff like that.

So, the various eminent worthies say what they have to say and then turn around to Sandage and say, "What do you think, Professor Sandage?" Meanwhile, the thing is that it turns out that -- this is crucial -- George Wallerstein, it being Caltech, they had bugged the deliberation room. They had bugged the room so that George hurried off to whatever closet they had the listening equipment in to listen to what the committee was saying along with his friend who bugged it. So, he's listening in on this, unbeknownst, of course, to anybody. The committee turns around and says to Professor Sandage, "What did you think of this, Professor Sandage?" Sandage says, "Not very inspiring." And if you knew Sandage, he had this kind of way of pushing his voice up and projecting like a preacher or something like that. It had a kind of preacherly feel to it. Like, I am going to reveal the truth to you. It was amazing to watch him talk, because he would just get this sermon on the mount type. "Not very inspiring." So, Wallerstein relates this, "What do you think you said?" "I don't know," says Sandage. "Not very inspiring."

And the two of them just collapsed in peals of helpless laughter. And here they were, a whole career later, having conquered any heights that they would, although I don't think Sandage ever figured out the acceleration of the universe, but you know, he was a top scientist. Incidentally, anybody who wants to read about Sandage, I don't think you can do better than Dennis Overbye's Lonely Hearts of the Cosmos. It's a magnificent book, and it's such a perfect impression of Sandage. 

Zierler:

Well, John, back to, just as a snapshot in time, what are you working on these days personally, and what's more compelling in the field more broadly?

Thorstensen:

Well, this is kind of an interesting question. I've kind of wormed my way over by basically just working away, largely in isolation to some extent, into smaller and smaller subfields. But what I work on is cataclysmic binary stars. So, those are stars which have a small and enthusiastic community following them, several hundred people probably worldwide. The idea is, in these stars, you find a white dwarf in very close orbit with a normal star. The orbits are so close that the stars, oftentimes the whole system would fit inside the diameter of the sun. And the normal star basically overflows its Roche critical lobe, so matter flows from the normal star onto the white dwarf. Of course, a white dwarf is a gigantic gravitational binding energy, so the matter, as it accretes onto the white dwarf, gives off a great deal of energy. This is of interest largely as a much more accessible kind of analogy to what happens in black hole accretors, which of course are huge, especially with quasars and stuff, are a huge preoccupation of a lot of astrophysicists, and other objects like that. And neutron star accretors, of course, because again they're gravitationally powered objects, and the physics of the accretion is of interest to people.

In any case, I study those, and the things is that it turns out there's a fair number of them. They're fairly rare, but there's quite a large number of them, especially compared to, say, neutron star sources or black hole sources. So, you can study them up close. There's several thousand of them known. And it turns out that there's a lot of them where people don't even know the most basic thing about it except that there's a star there that jumps up and down. What I've taken to doing is, especially surveying them to get an accurate picture of the population, and a comprehensive picture of the population, finding especially orbital periods of these objects. How long does it take the two of them to go around?

Part of the reason for this -- I think about the way in which my career has run, I've tended to be not one of the people who is visionary, who develops a new instrument, or has a marvelous theory, or has a great idea about something that I then proceed to use every available resource to go after. I tend to be somebody who's driven by the available instrumentation. What I have at MDM is what are now considered somewhat small telescopes. 2.4-meter is now kind of small compared to the behemoths of the field. A somewhat small telescope, and also a 1.3-meter. But what I have there is a fair amount of time, and often in continuous chunks. It's ideally suited for this kind of time series astronomy on objects of this brightness. And that's why I've tended to go after this, because it's something I can do that I can generate useful data; I can generate stuff that's of interest. The thing is that in the process of doing it, these objects, most of them you look at, it's kind of like, well, this one is a lot like these other 400 objects. And it's pretty much the same kind of thing.

Now we know that, and we move on. But occasionally, you run across something that’s actually very interesting or unexpected, and of course, you can’t predict these things because they’re unexpected. But it’s easy enough to – I think the most recent example was one just before everything shut down for the pandemic. I had a new list of bright objects of this kind iwhich a Chinese group had developed out of the LAMOST telescope, which is this giant survey telescope near Beijing somewhere. So, I just picked out what looked like the most tractable of those, and just was picking away at them in December of 2019 with some Dartmouth students, and then later in January 2020.

I found this one star which was just different. It had bizarre rapid flaring, among other things. I puzzled over it. I figured out what the orbital period was. I looked at some archival data and refined that. And then it struck me when I was going to write this up that this thing was really just like another object, AE Aquarii, which up to that time had been completely unique. AE Aquarii is another cataclysmic, but what it has is a white dwarf with a very strong magnetic field and a very rapid spin. And as material tries to fall in, this ionized material gets entangled in the magnetic field somewhat, and then centrifugally accelerated so that it just gets thrown back out. It basically bounces off the magnetosphere and is whipped out of the system as if it had hit a propeller. At least, that's what people figured out for AE Aquarii, was that it's the only propeller source.

The propeller was also thought to be an important phenomenon with neutron star accretors such as this, but AE Aquarii was the only white dwarf propeller source known. But this one looked just like AE Aquarii and clearly looked to be another one. So, I published that. It got picked up by Peter Garnavich's group at Notre Dame, and one of his associates, Colin Littlefield, discovered that this thing is actually eclipsing. They got on it with one of the enormous, big guns, the Large Binocular Telescope, the twin 8-meter in Arizona, and they got these beautifully detailed spectra through the orbit. They show that there's this absorption feature that drifts across the line profile in an orbital phase, which is the stream of gas spiraling out into the cosmos thrown away from the propeller. It's an absolutely beautiful system that just -- I think yesterday or the day before or something Peter did a presentation about this with a press release at the AAS. So, what I do is like -- it's not always the most -- it's not like I'm really uncovering the mysteries of the cosmos in a way of somebody who's going after a deep question by this and that.

But the thing is, the nature of the problem is incredibly gratifying, because I make progress fast. I find things out. I go to the telescope; I don't come back with some raft of statistical data that then in some years will tell me something. I find out stuff out that night. I find stuff out that nobody knew, that night, and then I can proceed with that. So, it's more much fast-moving and interesting and fun than doing things that might be considered more significant. And it is significant sometimes. Basically, at nearly 70 years of age, it's nice to be still in the game. 

Zierler:

I wonder how you might connect this research on cataclysmic binary stars with larger questions about the universe. 

Thorstensen:

Well, I've often wondered that. The answer is anything you come up with is a little bit forced. Basically, I think there's an awful lot of people in science who are just really desperately trying to show that what they do is relevant to other questions and other problems. And sometimes the actual answer is, it isn't particularly. It's just interesting. But the honest answer sounds like the copout, which is, who knows what it'll turn up? But to attempt to make that connection in perhaps, I hope, reasonably intellectually honest way, one thing is that this shows you -- it's a really unusual, pretty much unique laboratory for what happens to the hydrodynamics of this kind of flow. Why does the gas, say, for example, get thrown out by the propeller? Why doesn't it just loop on and find its way in? Accretion discs are another thing. Cataclysmic variables are one of the few places you can study accretion discs up close and personal, because in many cases, the material falls into a disc around the white dwarf and slowly accretes onto it. And there's all kinds of things that they do. There's these hiccup phenomena where they burst into -- a bit suddenly, there's an instability, and a material falls onto the white dwarf. It lights up everything, and then it fades away, and the disc reloads, and reaches critical density.

These are phenomena that you really can't study any other way, and it might be that they're so unusual that they have no applicability to anything else, but the insight that comes from just understanding a bunch of cases eventually makes the physics better. I'm not very much into the -- when I was a kid, I thought I might be a theoretical physicist or something, but I was rapidly disabused of that because it turns out, the dirty little secret is I'm actually not all that good at math. I'm pretty good at math, but I'm not all that good at math. I'm reminded a little bit of -- this is a little off topic, but Don DeLillo, the novelist, wrote a novel called White Noise, a parody of academia. His narrator was Professor of Hitler Studies, and his dirty little secret of being Professor of Hitler Studies, which is already kind of a joke, was that he actually didn't speak German. I'm not quite that bad, but it's a little like that.

Zierler:

John, beyond Kitt Peak, what are some of the observational projects that are interesting to you right now, that are either currently up or in train? 

Thorstensen:

Well, I think actually what's interesting is the thing which -- I'm actually preparing to go on an observing run next week, and what's been absolutely revolutionary is that I've finally managed to glom into the Zwicky Transient Facility. I don't know if you've heard of this thing, but it's actually possible now -- it's just revolutionary to fire a command and issue and get a light curve of one of these stars, almost all of them you're interested in, at least in the Northern Hemisphere, you get a light curve of it over the last three years, with something like 1,000 points. And you can see exactly what it's been doing. All of a sudden, this is like, I wonder if this is one of those, or that's one of those. There have been resources like this before, but they're not nearly as densely sampled, and not nearly as deep, as high quality.

I think of the Catalina surveys of the northern sky, but that didn't work in the galactic plane at all. It was too crowded for them. They were also largely interested in asteroids and such. That was great, but this was a little difficult to use. It was harder to query it automatically, which I'm highly into automation. I've got this elaborate system of lists and catalogs and software to enquire about things and correlate things and stuff like that. Because there are thousands of objects, so I can't just be there with a bunch of dusty logbooks, going, "What about that?" If I want to find the interesting objects when I'm about to go observing, I have to have everything automated so that I can consider targets one after the other pretty quickly. So anyway, the Zwicky Transient Facility has been a big one. I think the LSST is going to be an embarrassment of riches, the Vera Rubin Telescope, when that gets going. That's going to be absurd. 

Zierler:

An embarrassment because maybe it will yield so much data that we won't know what to do with it all?

Thorstensen:

Yeah, I think so. But believe me, the community has considered that a problem, and there's an enormous amount of effort on many different parts to come up with these machine learning things, and data brokers, and this and that and the other thing. It's remarkable because it's all coming at the same -- it's all building upon this giant infrastructure of digital stuff out there in the economy. You're not going to get like 8 gazillion gigabytes per night until you have some capability of handling it. But it's going to be amazing. The LSST is an amazing project. Another one that I've been kind of seeing up close and personal a little bit is the DESI, Dark Energy Survey Instrument, which is right there on Kitt Peak.

In the before times, before COVID, MDM is two miles down the hill, so I'd wheel myself up to the top of the hill to get dinner at the dining hall each night. At least, that's what I would do. So, I'd be sitting there in my bike clothes and talking to people. But there would be these teams of engineers and scientists from DESI who were busy trying to get the thing onto the telescope and working and doing each bit in turn. It's an amazing effort. Hundreds and hundreds of people. I had a personal connection, Parker Fagrelius, who is one of the managers of DESI, one of the people who was most instrumental in actually coordinating all the teams and such. I taught her first astronomy course at Dartmouth, so that was nice to see her. So, in any case, that's going to be amazing. They just passed their millionth spectrum, and they've been at it only for a few months. 

Zierler:

And just to be clear, a million is big. That's impressive. 

Thorstensen:

A million spectra, because they can do 5,000 at a chunk. So, that's only like 200 fields or something like that. It's a million spectra. It's unbelievable. It's actually very believable because the amount of talent and money and engineering acumen that went into it is astonishing. I was kind of amazed because it turned out that they apparently really did figure it out, and they didn't get the thing on the telescope and have something go wrong so badly that it's like, oh God, that's going to cost a year. We have to redo this whole thing and rethink this whole thing. Everything actually worked correctly, apparently, which is amazing. So, DESI is a big thing. There are some other things going on. There's new X-ray satellite -- the Germans have a satellite called eROSITA that's apparently going to generate lots and lots and lots of sources they need followed up. In fact, I put in a bid through a weird connection to be one of their collaborators so they can whisper sources to me. Go look at this and tell us what it is. I love that kind of stuff. 

Zierler:

John, administratively, I'm always interested in the way various universities separate or not departments of physics and astronomy. How does that work, for better or worse, at Dartmouth?

Thorstensen:

I think in the old days, astronomy was like this growth on the side of physics, if you will. Starting in the mid or late '70s, they hired a plasma physicist who decided he wanted to do astronomy instead, and then he hired another astronomer. That was Delo Mook who got us involved with MDM. Then, they had MDM, and they have these two astronomers, and Boley who was a plasma physicist, and Mook, neither of them really were making a go of it when I was hired. It was sort of like, let's -- I was hired, and then another theorist to sort of repair the astronomy effort, or at least get some action going. So, I was hired in 1980, and what happened then was sort of another couple people have come on board over time. Gary Wegner was hired shortly after that, in '83 or something like that. So, then we had four nominal astronomers.

So, the way the physics department works, or own department, is it's a little bit siloed into a few different major areas of interest. Dartmouth is not a huge place, and it can't cover the whole waterfront of physics. We don't have a whole bunch of particle physicists, or something like that. We got basically plasma and space physics, condensed matter physics with a particular kind of exotic quantum thing, as being quantum information, and some other things as a focus of that. And then astronomy and astrophysics, and sort of floating in among there is cosmology people who are primarily Robert Caldwell and Marcelo Gleiser. And probably somebody who I've left out and will annoy. But in any case, it has these areas of strength, and astronomy has kind of grown up to this. And then what happened is astronomy just managed to kind of drift away a little bit in ways that are important. For example, we now have what amounts to a semi-independent graduate program.

So, graduate students don't have to take the physics qual. In fact, they have a whole different sequence of hoops to jump over that are much more research based than just, you know, you're going to take a bazillion classes, and then eventually we'll figure out if you're any good at actually doing the work. It's much more like, let's get you in from the beginning, keep you engaged, and so on like that. So, I think we've managed to establish a quasi-independent thing, despite the fact we're in the same department. Face it, having a whole separate department administration is kind of an administrative nightmare, so it's much more efficient for a moderately small place not to have a separate astronomy department. 

Zierler:

Well, John, let's take it all the way back to the beginning. Let's start first with your parents. Tell me a little bit about them. 

Thorstensen:

Oh, my parents were just extremely interesting people, I think. My mother was born into a lower-middle class, very much working class Irish Catholic family in the '20s. Her mother was actually German American, enough so that her mom actually could speak some German. Her dad had emigrated. In any case, her father was a wonderfully, incredibly smart guy, but he had a drinking problem. He was also a really gentle person. She said he was a drunk, but he never ever was violent or hit anybody. He wasn't even hardly ever mean to anybody. He was also a passionate social justice warrior. He was a union guy, and he had a rough and tumble life. He didn't live past his mid '60s, I think like that. But he had that kind of Irish gift of the gab and everything. My mother actually grew up in a kind of strait-laced Catholic school background, although their parents didn't really care for the church that much. They did lip service, but she grew up in that, and she was one of these people who just takes to the academic side and the reading and all of that. She went to the University of Minnesota.

Meanwhile, my dad -- because there has to be another person involved -- my dad was the child of Norwegian immigrants, or Norwegian first or second generation born in the country. His dad was an incredibly smart guy also, who in the family grocery store in Milwaukee, had decided he wanted to do something else, and had gotten some kind of a free ride for a leather technology program at Pratt Institute in Brooklyn in 1912, or thereabouts. So, he took off to Brooklyn in 1912 to learn how to be a tanner. He said they studied like crazy, and a year or two later he returned to a job in Milwaukee where he could put his skills to work. He was a tannery guy there. My father was born in Milwaukee, and then they moved to Red Wing, Minnesota, where my grandfather got a job as a foreman of a large tannery.

So, he had a very responsible job there. My grandfather went on to become president of the American Leather Chemists Association. He was a very forbidding man. There were two professions that a child could go into, as far as he was concerned, which was a leather person, or a doctor. Those were the only ones that anybody could do. So, my father, of course, attempted -- he was the first born, and it was kind of a pressure cooker atmosphere. He totally failed at pre-med. Even though my father was incredibly smart, he totally failed at pre-med. He just hated it. He also had no motivation. So, he ended up going to University of Minnesota just sort of as a fallback after two years at Northwestern, and he met my mother.

At Minnesota, he studied speech. He'd been a lifelong stutterer, and he kind of figured out how to deal with that. He studied speech and also English. My mother and he discovered a common love of drama. In fact, he said the first thing that attracted -- they were in a French class together -- the first thing that attracted his attention about my mother was her voice. My mother had an absolutely magnificent speaking voice, just beautifully -- she and my father did dramatics at the University of Minnesota. My father had an amazing voice as well. So, both of them had these incredible voices. And they became -- my father went off to war, my mother married him during the war, and had a child a full gestational period later, but not much more. 

Zierler:

Strategic timing. 

Thorstensen:

Strategic timing, yeah, and they went on from there. My father, afterward, attempted to complete graduate school in English but wasn't able to do it because he was so neurotic about it. But in the meantime, he got a full-time job teaching at SUNY Albany, before it was SUNY Albany. 

Zierler:

Amazing. 

Thorstensen:

And he got tenure. So, he was a tenured associate professor of English until the day he died. He never became a professor, and he never got a PhD. But he was a person who read enormous amounts. He was incredibly curious about things. He had a penetrating insight into things, and he was also extremely good at understanding science. You wouldn't think so, but he was just very, very astute about science. So, those were my parents. My mother became an English teacher not long after I was born and became a beloved teacher at Albany Academy for Girls, Mrs. T. Kind of legendary. It was kind of amusing because I had a student, Kevin Hand. I don't know if you ever met this guy, but he's kind of a prominent astrobiologist at, I think, NASA Ames. But I was his undergraduate thesis advisor at Dartmouth. He's a really interesting guy. Very, very interesting guy. Anyhow, Kevin returned to Dartmouth for something or other, and one of the Dartmouth publicists was interviewing him, and she asked, "Who were some of your favorite teachers at Dartmouth?" And he said, "Oh, John Thorstensen. He was my thesis advisor." And she said, "Oh, his mother was my favorite teacher."

Zierler:

Family business.

Thorstensen:

It's the family business, teaching. I'm the black sheep because I didn't go into English. Everybody else went into English, but I went into the sciences because they always just fascinated me. So, the background that produced me was this first-generation intellectuals, passionate about learning, vast amounts -- the house I grew up in was just rich, rich, rich with a wide variety of reading materials. I'd just pick stuff up and read it. So, it was an incredibly rich environment. I was also absolutely expected to do well. There was never any explicit pressure, but boy, I knew. 

Zierler:

Now, I'm always amazed, from Joe Taylor to Stephon Alexander, Haverford College is just tiny powerhouse in physics and astronomy. Did you know that going in? Was that part of your motivation?

Thorstensen:

I didn't quite know it. What happened was I remember going and interviewing at Haverford. I hadn't even -- the only reason I went to Haverford was because, basically -- it was so casual back then. None of this stuff of sort of from the time you're a high school sophomore plotting your attack on whatever school you want. It was like my senior year, and my mom said, "Oh, you know, the librarian at my school, her son goes to Haverford and really likes it. That's a place you might want to go." So, I was just looking around for places to go to in the fall of my senior year. So, it was really late in the game. The fall of my senior year, I took an airplane down to Philadelphia, and I went to Haverford and Swarthmore. Just unannounced -- I was going to stay with this kid. The student who was there at Haverford, I could stay in his dorm room. That was okay.

So, I had a place to stay, but it was just like whatever. I didn't arrange for interviews or anything. And every place I went, I went over to the astronomy department, and just said, "What's up here? I'm a prospective student." I remember wandering into Louis Green's office at Haverford, in the evening, and he was still there working away on stuff. And I said, "I'm a prospective student. What can you tell me?" And here's this old guy in this beautiful old observatory building, this kind of big stone building. He says, "Well, our students try to do research and such. Larry Auer published this thing, and Bill Foreman is --" I think Bill Foreman was there working on something, actually. He had come back to scan a plate. I think he just graduated the previous year or was just finishing. He later became this big guy at Harvard. So, essentially, he was saying, we have all of this published student research. Students work on real stuff. And I just thought, oh, this is for me.

I went to Swarthmore, and they have this beautiful, giant telescope, this Sproul 24-inch refractor, which was in this big dome -- it looks like from central casting and stuff. And the guy was explaining the work they do there, and he said, "We take the plate, we put it in here, we do this, we measure it, and we look at the stars." I wasn't -- boy, this really just seems like the science of many years ago, which it was. It's ironic I think that I was so dismissive of this, because I later spent a good part of my career as an astrometrist, actually measuring parallaxes of cataclysmics. I realized it would be possible to do this with modern instrumentation much better than they could do back then. I actually did a program of it and had some impactful papers from it. But it was odd that I eventually ended up circling back to astrometry.

But for Haverford, that was it. That, plus there was this other guy there. One of the students said, "Oh, yeah, I know this guy. He's an astronomy major." So, I went to talk to him. He was this wonderful fellow, Gene Hodges, just very gifted guy from Tucson originally. Also, interestingly, very much working-class origins, I think. But Hodges said -- he's kind of a big talker. He said, "Astronomy has the reputation of being the hardest major on campus." I don't know if that's true. I think it probably didn't, and he was just -- but it did have a reputation of being pretty rigorous. And I thought, yeah, that's for me. I want the hardest major on campus. 

Zierler:

Now, was the plan the double major in physics and astronomy, or that was just what was most sensible?

Thorstensen:

I was an astronomy major from the get-go, and I tacked on a physics major at the end because I had enough courses. Back in the day, at Haverford, you could double count courses. It would count both towards physics and astronomy, so it was very easy. At Dartmouth, you can't do that. So, you have to pretty much stand on your head and whistle Dixie to double major because you have to carefully parse -- okay, I'm going to count this course toward this major, and that course -- and then they have to get the right requisite count for both. So, Dartmouth is much harder, but at Haverford it was pretty easy to double major. 

Zierler:

How well did you take advantage of Haverford's reputation for encouraging students to do their own research?

Thorstensen:

Well, as it turned out, the year that I arrived at Haverford, I had no inkling that this was going to occur until freshman week, but there was a brand-new second astronomy professor, Bruce Partridge. He had just come from Princeton. He had actually just done the very first microwave background anisotropy experiment. And much to his credit, he found nothing. He found upper limits. If he'd seen an anisotropy, or claimed to, his name would be mud because his sensitivity was nowhere near where you could see any anisotropy, even the dipole. The dipole wasn't even found until the late '70s, I think. 

Zierler:

Sometimes in science, finding nothing is the best way forward. 

Thorstensen:

Yeah, well, he found nothing, and basically said it just looks very symmetric -- and to first order, and maybe even higher orders, the microwave background is really isotropic. He measured and said, "It's isotropic better than 1%." It was a good conservative, solid bound. In any case, he was fresh off that, and he was young and charismatic and had a kind of -- it's sort like, he was one of these people who had just incredibly strong personality that sort of says, "Maybe I should be your role model." One of those. He was also strict, and acerbic and stuff like that. He was also -- as it turns out, from many experiences over the years, he was an extremely kind man. But he was clearly tough as nails. It was great.

So, I fell in -- I took a course with him my freshman year as an extra course. It was just sort of like his liberal arts course for astronomy of the last decade, and it was just fabulous. And I went on from there. Later, I did a project course with him on basically the straw man argument of a sort. At the time, nobody had any idea what the missing mass was, what also dark matter. We had a very poor idea of how much there was, and so on, and the thought was that perhaps the missing mass would be enough to close the universe. In other words, it would take it to the flat geometry. They had no idea about the dark energy and acceleration at that point. That was many years in the future. So, then the question is, you start going through this laundry list of, well, could it be this? Could be that? Could it be the other thing? The answer, could it be cold neutral hydrogen in space? Well, no. You would see gigantic Gunn-Peterson absorption and such. And one of them is, well, maybe it's just a bunch of black holes, because after all, they're black. And if you have black holes, perhaps they would be the remnants of an early generation of really massive stars.

So, I wrote a paper with Bruce called “Can Collapsed Stars Close the Universe?” The handle to get on this is if you have an early generation of massive stars, they generate a whole lot of light. They generate it at high redshift perhaps, but how long -- so, what I did was a detailed calculation which started with a straw man thing, which was basically, if you had a flat universe with a whole critical density full of just massive stars, and they all light off at some epic, and they live as long as massive stars do, and they produce energy in the wavelength that they do. What, in the end, do you see for the sky brightness today? The light is still there. It's just redshifting and propagating.

So, what do you see for the sky brightness today and can you set a limit where you say there can't be more than this number of massive stars? The answer was, it was possible to make it, but the stars would all have to be formed at a very high redshift and burn out very quickly. You couldn't have a tail of like low-mass stars that were still around. Any sizable chunk of the critical density surviving as low-mass stars into a less redshifted region would exceed even the crappy sky brightness limits they had then. Nowadays, you look back on this 40 years later or something, and it kind of looks ridiculous because who would ever think there was this? It was almost 50 years later. Why would you think that knowing what we do today? Well, we didn't know what you do today, and that's why you start running down these things. 

Zierler:

In this environment, are black holes real at this point? This is very much a generational question as it is a scientific question. 

Thorstensen:

Oh, black holes were very much real. I never actually knew a time in which they weren't. For that course I took as a freshman, I wrote a paper about active galactic nuclei, essentially looking at the big question of, what are they and how do they produce this prodigious energy? What I came up with was, after surveying the ideas, it really does look like they're just massive accreting black holes. I didn't think that up, but looking at the literature, I sort of thought that's by far the most plausible idea for what's going on with these things. So, even then, it was considered that way. The other thing was that Uhuru had launched in '72, I think. They were starting to see things like Cyg X-1, which had an apparent mass which was above the neutron star upper limit. So, that looked like a black hole. 

Zierler:

John, socially, was Haverford active in campus protest and the antiwar movement, civil rights, things like that?

Thorstensen:

Yes, but it had a different flavor. Actually, it's very interesting. After I had applied and been accepted to Haverford, that was the spring of '70, which was when Kent State happened -- the invasion of Cambodia was what triggered that. So, of course, the campus has erupted in the spring of 1970. Now, Haverford is a Quaker school. It's not run by the Quakers, but it was founded by the Quakers, and has a lot of Quakerly influence in it. And there wasn't any military -- no Defense Department funded research on campus by decree. You can't take money from the Defense Department.

So, the students realized that Haverford was not the enemy. And they went to Jack Coleman, who was the president at the time, and was an absolutely remarkable man. So, a delegation of students when Kent State happened went to Coleman and said, "What the hell are we going to do?" And Coleman said, "I have an idea. Let's shut down classes for a couple of days, rent a bunch of buses, and run any student who wants to down to Washington to petition their representatives in Washington." Hundreds and hundreds of students -- what this did was it instantly diffused and residual thought that Haverford was part of the establishment, or something like that. Let's do something constructive and thoughtful and meaningful this way and communicate our concerns. It worked brilliantly. He was a hero. The students just considered -- they would follow him off a cliff at that point. He was a hero.

So, basically, Haverford was not the kind of place where there was a lot of resentment toward the school at all. We never occupied the administration building like they did at Dartmouth, or anything like that. Meanwhile, the school -- Steve Carey, who was the vice president of the school, kept getting arrested for doing things like blockading troop ships and canoes -- or munition ships and canoes. So, he was doing that, but the most remarkable thing that was going on at Haverford, which I had no inkling of, and nobody had any inkling of, wasn't revealed until 40 years later. That was, Bill Davidon -- I don't know if you've heard this story.

Zierler:

No.

Thorstensen:

Oh, this is an absolutely remarkable story. Bill Davidon was this brilliant professor of theoretical physics at Haverford. Mathematical physics, really. He was a mathematician by trade. He was incredibly facile. By the time I had him for classes, he would just walk in and say whatever he felt like. He was not somebody who we would find useful for learning from in a systematic manner at all, at that point. But I learned a lot from him, just because somebody that smart gets up and talks, you learn a lot. And everybody knew that he was like really, really, really into the antiwar movement, and was it was probably using most of his time at that point. But, you know, so what?

Well, it turns out later, that there was this burglary in Media, Pennsylvania, at the FBI field office there, which was at the time unguarded at night. A bunch of people broke into that in this well-planned and rapidly executed burglary, went through the files and stole a bunch of them, and drove off into the night and were never seen or heard again. A month or two later -- I don't know how long it took -- they had rifled through the files and found the stuff that they were after. They sent copies of these to several newspapers. Here are secret FBI files about what they do to suppress dissent. There's this mania in bureaucracy for everything -- so, all of this stuff -- it actually took a year of some reporter rifling through these to find COINTELPRO, which was the big FBI -- let's harass all antiwar and leftwing demonstrators. And this led to the Church Committee hearings later in the '70s, and stuff like that. It was a huge thing. It completely destroyed the FBI's reputation for fairness and probity when it came to dealing with political dissent. And it turned out that Bill Davidon masterminded that whole thing. This was later revealed.

The amazing thing is there were seven people in the conspiracy, and not one of them said a word. They were never caught. The only time it was revealed was at a dinner party, apparently. There's a great New York Times video on this. At a dinner party, where this woman who had been the young girl who claimed to be a Swarthmore student checking the place out, when she was actually looking for alarms and things. Some reporter was saying, "I wonder whoever did that. Nobody ever figured it out." And she said, "I know who did it," and then revealed the whole thing because it was so much later that nobody was ever going to be prosecuted for it or anything. But it was a huge thing. I mean, my physics professor brought down the FBI when nobody know. So, that's what happened with Haverford and the antiwar stuff. Being a Quaker school, there was no question that they were on the good side, and nobody was trying to -- 

Zierler:

That's -- wow, I am glad I asked. I did not know any of that. That's great. John, what kind of advice did you get for graduate school? Programs that would be best suited for you, even professors to work with.

Thorstensen:

Um, no. I got almost no advice for graduate school. I think I just sort of said, what are good places to go? I really hadn't given it that much thought, but I think what I did was I said, what are the top places? Here's what was going on at the time, which is kind of interesting. I heard -- there was an awful lot of rhetoric just following -- this is just at the end of the postwar boom, and all of a sudden, the curves weren't going up fast anymore. In fact, they were flattening out. People were starting to realize that a lot of people are going to train in astronomy, especially, and maybe physics, and they're not going to get jobs. So, if anybody is banking on the idea that they're going to go out and get their astronomy PhD and then get a cushy job in the astronomy industry forever, that's not going to happen necessarily. This has remained the case, although it didn't turn out quite as dire as quickly as people had thought, especially I think because of things like Hubble Space Telescope, which absorbed a large, large number of PhDs, and some other projects which absorbed a large number of PhDs in non-academic positions, but stuff where they're really doing the work.

In any case, I was surrounded by all of this gloom and doom rhetoric as a late undergraduate, and I sort said, “Well, how can I negotiate this? Okay, I'm doing pretty well here. There's a paper that I'm working on and whatever.” And I thought, well, why don't I just apply to just the best schools I can think of, and if I don't get in, I'll take it as a career signal and just spend a year doing something else. Who knows? Maybe I'll decide to go to med school or something like that. So, I just applied to the best schools I could think of, which was Harvard, Berkeley, and Caltech. Meanwhile, it turns out that -- I think in part because of this background with teachers as parents, and soaking in academia all this time, and also because of sort of broad ranging education at Haverford, Louis Green threw us in the deep end of the pool with a book, saying, "Here, read Principles of Stellar Evolution in Nucleosynthesis. You won't understand much of it, but you'll be able to figure out some of it." Whatever. It was kind of an unusual form of education, but because of all that, when I went to take the GREs -- I've always been a demon test taker -- I did superbly well on the GREs. 95 percentile in physics, and so on like that.

So, the grad schools all looked at me and went, whoa, here he is. And I think I had good recommendations too, because I had done good work as an undergrad. So, I got in everywhere. I got into all the schools I applied to, and I thought, okay, three lemons just came up. Ding, ding, ding. Maybe I should go on and see if it works, which it's incredibly lucky to find that it did. And if I hadn't gotten into any of the top grad schools, I would have just said I'll think about something else. Maybe I'll come back and apply to second tier schools, or maybe I'll do something completely different, because the rhetoric at the time was so bad. 

Zierler:

You have some pretty good options. Why Berkeley?

Thorstensen:

Well, that' very interesting and fun. As it turned out, my brother-in-law was a grad student finishing up at Harvard in American Studies. He went on to become an institution at Amherst College, Barry O'Connell. In any case, he was just finishing up, and he and my sister lived right down the street from Harvard Smithsonian Center for Astrophysics. I lived in Albany, so I took a bus over and visited them, and just walked into Harvard Smithsonian Center for Astrophysics unannounced and just absorbed some of the atmosphere. There were a bunch of grad students -- and this is the kind of thing, which, what are these decisions based on? Who the hell knows? But a bunch of grad students in the hallway chitchatting about stuff, and they seemed to be primarily interested in which faculty member was climbing higher on the status totem pole, and influence totem pole than others. And they seemed bitter and cynical, and it was just like, uh --

Zierler:

This is not your scene. 

Thorstensen:

This is not my scene. I bet if I went there, it would have been fine, but it didn't leave a good impression. Then, I went to Caltech. I flew out to the west coast over break, I guess, and I went to Caltech. I arrive at Caltech -- I had a friend from undergraduate school who was going to Caltech in neurosciences, so I stayed in her apartment for a little bit. Again, I wander over to the astronomy department, and I think that it had something to do with the day. It was pouring in Pasadena. It does so in the winter sometimes. So, it was a really dreary day, and the only person I could find was Jim Gunn, who's wonderful of course. But Jim Gunn -- I don't know if you've ever met him, but he struck me as being wound up about as tight as he could get. So, he made a pitch, which was good -- and the other thing was that all the Caltech grad students I ran into, all largely in other departments, appeared to be borderline suicidal.

Zierler:

Because it was such a high-pressure environment?

Thorstensen:

Yeah, I think so. It was a very high-pressure environment, and they just seemed very unhappy. And then I went to Berkeley, and it was a much nicer day, and of course the scenery in the Bay Area on a beautiful day when you look out over the bay, it's just gorgeous. And I, once again, showed up, and Len Kuhi talked to me and said, "We've got this. We've got that. It's great. Let me see if I can find you a grad student to talk to." So, I talked to this guy, Tom Troland, and he seemed happy and normal and cheerful, saying, "Oh, this is a great place. We do this, we do that." And I thought, well, that's a great place to go to school. Plus, it's Berkeley, which is a very interesting town. 

Zierler:

Kind of like choosing Haverford all over again, only for graduate school this time. 

Thorstensen:

Yeah, well, basically, I think one way to choose schools is wander in and see what the atmosphere is. That was what it was. Berkeley seemed to have a normal -- of course, the town is crazy, but the astronomy department seemed to be normal and healthy. That was borne out in my time there. At first, I felt a little competitive when I got there, but I sort of got over that pretty quickly. Pretty soon, it was clear that the students are all just sharing information and helping each other. “Oh, I know how to do that.” “Here, take this code.” It was just a lovely place to go to school and we all had a good time. 

Zierler:

Who ended up being your graduate advisor at Berkeley?

Thorstensen:

Stu Bowyer. That's a story in itself. I think Hy Spinrad was hoping that I might work with him, but Bowyer put on this full-court press to get me into his group. Bowyer could turn on a charm offensive like that, and he was able to hire me over the summer or get me to work for him over the summer after my first year at Berkeley. That was the summer they launched Apollo-Soyuz. Apollo-Soyuz was this mission where, what are you going to do with all these Apollo spacecraft after going to the moon? Well, one thing they did was this cool thing where they rendezvoused with a Russian spacecraft in space, and they broke out the champagne which got into the electronics or something. I don't think so, but it was a publicity stunt.

But Bowyer had managed to talk his way onto the Apollo-Soyuz test project with this little extreme-ultraviolet telescope, a kind of crude one. Nobody had ever looked in the extreme-ultraviolet, sort of 500 angstroms or something. Everybody said the interstellar opacity is way too high. The ionization opacity is just way too high. And he said, "I don't know. It looks like the interstellar medium is actually pretty patchy, and there may be some low-density patches you can see through." And he was absolutely right. Bowyer was a remarkable guy. We could talk for hours about Bowyer, and probably will. But he had that thing going, and that was all ready to launch.

So, the summer after my first year, the Apollo-Soyuz went up, it saw some sources, and I did basically some follow up work on that. I also did some ground based observational work, which was totally ill-considered, just completely worthless. But somebody said, "Go out to Leuschner Observatory and take plates of the object we're looking at." And then they never ever used them for anything, which has kind of left a bit of an impression, which is, before you go to the telescope, you've got to have an idea of why you're doing the observations and what you're going to get out of them. Ever since then, I spend far more time planning what I'm going to do and selecting targets than actually observing them. In any case, it was an object lesson, and very interesting.

Meanwhile, I kind of got sucked into the Bowyer group and drawn along on the -- and then shortly after that, Phil Charles showed up. Phil Charles is a -- I don't know if you know him but he was a professor at Oxford and Southampton for a number of years. Phil was this young, kind of enthusiastic pup of a guy, and very British. And we got to working on projects, like optical identifications of X-ray sources, and follow ups of those, and this, that, and the other thing. We just had a wonderful time. We worked really hard on stuff and got lots of good stuff done.

The thing that actually was the incredible stroke of luck was that thanks to, largely, earlier work by Art Davidsen and Roger Malina, and Stu Bowyer was also on the paper because he was on all the papers -- they had started photographing X-ray error boxes, which were bigger than a breadbox at the time, up there in the sky. The telescopes couldn't tell where the X-rays were coming from, except very crudely. They had a plate archive which I inherited of all of these X-ray sources that they looked at. I remember one night I was just -- and I took some plates too on that thing -- one night, sitting in my office in Campbell Hall, I was there looking, what are we going to observe? Again, doing the observing planning. And I looked at a nice picture of Aquila X-1, which is a transient X-ray source. I remember looking at this picture and saying, that one star there must be super red because this is a red plate, and it must be just like super red on this and it's not nearly as bright on blue. You know how it is, sometimes, there's an anomaly and it's sort of ticking in the back of your brain? Well, I think it was like half an hour later, I thought, what about that one star? Star two, very red. What about star two? Could it be variable? And instead of being red, maybe it was up when that picture was taken.

And then I started pulling out the plates of Aquila X-1 and looking at them. And I came to realize, the thing about Aquila X-1 is that it sits there doing nothing, and then about every year, it flares up really bright for a month in X-rays, and then it calms back down. And I was realized that I was seeing star two bright at the same time that the X-rays were going, which was a huge discovery. It was basically finding the optical counterpart of Aquila X-1. This is one of the very first transient X-ray sources identified. It was like, whoa. This is like a career-making discovery because it meant basically, we could write a paper, and everybody said, "Oh, my God. What the hell is -- that's great." And the reason it came about was just because of a lot of sitzfleisch. I don't know if you know the German term, sitzfleisch.

Zierler:

Well, I know, in Yiddish, it literally means you need the ability to sit. You need padding on your tush. 

Thorstensen:

Precisely. It's just sitzfleisch. It was a lucky discovery, but it was chance favoring the prepared. I was down there in the data, kind of just trying to make sense of it, and I found it that way. So, it was very empirical. The reason why that discovery is significant was that Delo Mook -- remember him, back at Dartmouth? He'd been looking for this thing, and he never found it. And when my application came in for the faculty job, he just went, "Oh, my God. I thought that he was a postdoc or something." So, they hired me. I don’t know why, but they did. But that was one thing. Basically, it prepared the ground for my job at Dartmouth. It was a crucial moment to discover Aquila X-1.

Zierler:

John, just to zoom out for a little bit, as you're ensconced in your graduate research, what were some of the bigger questions in the field, and to the extent that you thought about that as a lowly graduate student, how did you see yourself developing and contributing to those bigger questions?

Thorstensen:

That actually -- it's almost embarrassing, but I've almost never really considered bigger questions that much. I'm a person who puts one foot in front of the other. 

Zierler:

Which is just so funny because you're an astronomer. You study the universe. Those are the bigger question kind of places to ask.

Thorstensen:

It is also kind of amusing, because meanwhile, right down the hall, Joe Silk was developing the theory of how perturbations develop in the early universe, which has turned into one of the most stunning things of all time, which is the correlation spectrum of the microwave background anisotropies. That was the fundamental work that ended up leading to that. It was all happening down the hall, and it was all happening in Fourier space. But yeah, basically that's always been a weak point. When I said something about notability in the little note that I sent you, I kind of wonder about notability here. I've never really considered big questions all that much, which is a weakness. 

Zierler:

Besides Bowyer, who was on your thesis committee?

Thorstensen:

Oh, it was cobbled together at the last minute. I think Kinsey Anderson and Forrest Moser. I don't know. Basically, what I did was I wrote a thesis -- I had one chapter where I went after an actual systematic study of something, which was if you look at the variability of the low-mass X-ray binaries, would it be possible to find orbital periods or something, just by looking at the ripple in them? In particular, if you have a normal star being basted by an X-ray source on one side, the one side tends to light up really bright. It happens in Hercules X-1. It happens in another, which was 4U 2129+47, which was a star that Phil and I did a lot of work on.

Basically, what happens then is that as the bright hemisphere rotates in and out of view, you see it getting brighter, you see it getting fainter, you see it getting brighter, you see it getting fainter. So, the thought was maybe you could use this to find orbital periods, but in the meantime, is the light actually even coming from the other star? So, I just took pictures of a whole bunch of these things and found that they were remarkably constant in light. They were not varying up and down, and I did a bunch of statistics to show that most of the light has to be coming from a more or less isotropic source. So, that was a chapter that was systematic, but the rest of it was, oh, here's a paper we already published, and here's another paper we already published. So, it was essentially assembling a thesis, as I'm fond of saying, assembling a thesis with a staple gun. Staple it together and put it together. It was a last-minute thing to just get out. I had a job and wanted to get out of there.

Zierler:

Now, you consciously skipped over the postdoc and went right to a faculty appointment?

Thorstensen:

Oh, it wasn't conscious. I just got a faculty job immediately. 

Zierler:

This is a rare thing. I mean, I'm well-positioned to know how often this happens, and the answer is, not very. 

Thorstensen:

Well, I think it was more common in the past, and it's also more common at institutions which don't -- you won't get that very often at one of the absolute top research universities. Dartmouth is an R1 university now, but it kind of drops in and out of that. 

Zierler:

But it's always enjoyed that Ivy League status. That must have been a factor as well. 

Thorstensen:

Yeah, it is an Ivy League school, but if you compare to, let's say, Princeton, Princeton is a vastly larger graduate school with a much larger research operation. Especially the physical sciences. Princeton's research operation is an order of magnitude bigger. Same thing with Harvard. Harvard-Smithsonian has something like 300 PhDs or something. I don't know what the number is, but it's huge. So, it's Ivy League. As an undergraduate school, I think it's almost second to none. I can't think of a better one, where you'd say, “Oh, yeah, they're better than we are.” But as a large research university, it punches above its weight, but it's not a powerhouse. In any case -- we punch way above our weight, by the way. I don't want to demean our efforts -- but when I say above our weight, I mean compared to the size of the number of people doing the work, our mass-to-light ratio is very low. So, pretty good. 

Zierler:

Besides the obvious benefits of just securing that faculty position, it was tenure track right from the beginning?

Thorstensen:

Absolutely. I mean, why would I not take this job?

Zierler:

I've heard, just to play devil's advocate, that some people see value in the postdoc just to as additional opportunity to develop one's research skills, to broaden your interest a little bit. None of that was compelling to you.

Thorstensen:

Oh, absolutely not. I mean, I'm sure it would have been. When I look back, I'm sure that I could have benefited from a postdoc in all kinds of ways. I was pretty raw when I got here. In some ways, the kinds of things I would have developed as a postdoc I just never did. 

Zierler:

Was the position sort of a unicorn offering? Was the job market particularly strong that year?

Thorstensen:

I think it was actually even more idiosyncratic. Dartmouth actually just had built this telescope. Or not built it, they acquired this telescope a few years before at MDM as a 1.3-meter. The 2.4-meter had not been built yet. And I think they felt a need to take advantage of it, but Mook and Boley were not really in the game. They were doing stuff, but it wasn't producing papers. Mook's interest is optical observations of X-ray sources. His idea of -- one of the things he would do, for example, was take a photometer and just stare at Scorpius X-1, which is the brightest X-ray source in the sky. And Scorpius X-1 jumps around a lot. It flares up, it does this, it does that, and you can see it all happening in optical, and you can correlate it with the X-rays and so on. It is just not -- you get a lot of data, but you don't learn a lot.

He had ideas. Mook was incredibly articulate and could talk about all the great stuff they were going to do and how they were going to figure it all out, but he really just didn't get that far with it somehow. It just didn't work. So, they were in this position where they wanted somebody to kind of continue more or less what Dee was doing, but actually make it go, I think. I remember vividly walking down the hall at Berkeley and looking at the job boards, and there was this flier there from Dartmouth College saying they want somebody to do optical studies of X-ray sources. That's exactly what I do. I had like ten papers in grad school on optical studies of X-ray sources, because I was in the Bowyer group which pushed you to get stuff out. Phil Charles was a powerhouse as well. I'm pretty ambitious and was able to -- really wanted to get it done and wrote a bunch of good papers in graduate school. Aquila X-1, 2129, so on like that.

So, I thought that's exactly what I do, so of course, I immediately applied. And apparently, according to Mike Kurtz -- I don't know if you've ever interviewed Mike Kurtz, but you should. Put him on your list. Michael Kurtz was a Dartmouth graduate student at the time. He had a wildly circuitous career path. I'm off track here, but it's just why you should interview Michael Kurtz. Wildly circuitous career path that led him to a research scientist position at Harvard CFA.

It turns out, he is fascinated by the technology of basically library science, and how you deal with vast amount of information. He was one of the founding driving forces behind the ADS, which I think is the absolute envy of all scholarly fields, that we have the ADS. It is the best thing ever, and he's an incredible guy. Anyway, Kurtz later told me that Dee Mook came to the observatory where the graduate students had their offices, and said, "We have our guy," because I had already beaten him to Aquila X-1. He thought I was a postdoc. So, they thought, if we want somebody to do optical work on X-ray sources, here's this hot grad student who's -- so, it was just weirdly idiosyncratic that I got this job right out of grad school.

So, then what happens is I come here, I give my talk which runs over time and everything like that, but circling back, thanks to my parents who were classroom teacher, I can speak, I can connect, I can work with an audience and such. People said, "That was pretty good," and I was overly confident. Yeah, I can do this. Looking back, some of the things I said I could do, no, I really couldn't have done that. Good thing they never asked me to. So, they ended up hiring me. They hired me over some people who I think would have been at least as good or better. But that's the way it works. So, it was weirdly idiosyncratic. I spent a good bit of time at Dartmouth kind of floundering around, probably better part of a year, not making a lot of progress, which was unfortunate. 

Zierler:

Now, as you said before, was the astronomy program still very much an appendage to physics?

Thorstensen:

Very much. It had a little bit of an inferiority complex, I think, because the physics people would say, "Well, that's not physics." In any case, when I got here, my ace in the hole it turned out was teaching. I didn't have a lot of experience as a teacher, but I had soaked in the culture of classroom teacher through my parents and such the whole time. It turned out that I can stand up in front of a Dartmouth liberal arts class of 100 students trying to get through their science requirement, and I can pitch them at the right level, challenge them, but not overwhelm them, and entertain them at the same time. I could do that with great reliability. So, I got great teaching reviews, and Dartmouth prides itself as a teaching institution. Meanwhile, I actually did eventually find my feet and start cranking out work using MDM, which is just what they wanted. 

Zierler:

Now, tenure considerations at Dartmouth equally consider research and teaching?

Thorstensen:

That's a very good question, and I wouldn't want to pronounce on it too authoritatively. 

Zierler:

Or at least your view, in terms of your experience.

Thorstensen:

Yeah, I've actually sat on the tenure committee. Basically, the way that it works is that different people have different skills. I think Dartmouth always wants people to be first rate scholars, at least productive, to the extent that when the letters come in from external examiners, or external experts, they say, "Oh, yeah, this person is really doing nice work in this field," and so on. But they also want people to be good teachers. I think if somebody was a great scholar and a terrible teacher, they probably wouldn't get tenure. If you're a great scholar and somebody who they can look around and say, "Oh, yeah, they've taught this and this pretty well. Some students didn't take to them, but that's okay." But if they're a great teacher and kind of a mediocre scholar, they won't go for it. So, it's the necessary and sufficient type thing. It's pretty much necessary to be a strong scholar, but it's not sufficient to a good teacher. 

Zierler:

Now, as you said, after a short while you found your groove and started writing papers. What was the topic that got you into that groove?

Thorstensen:

What happened was I had an incredibly long observing run my very first semester at MDM. It was like the fall of 1980, and I go out there like 17 nights in a row or something like that, and it's on the 1.3-meter. I had this elaborate preparation done -- I'm not going to do pointless -- where I was going to go and identify X-ray sources from the newly launched Einstein Satellite. I had all the finding charts, and the little candidates marked, and all of that. And I discovered on my first run at this telescope that the 1.3-meter with the spectrograph and equipment that it had at that time was not nearly as powerful an instrument as the 3-meter at Lick which I was used to. Not nearly. And this stuff was almost all too faint to see. So, I was just kind of – “What?” And Joe Patterson -- I hope you've interviewed him. 

Zierler:

No, I haven't, but I know of him. 

Thorstensen:

Oh, he would make the best interview ever. But Joe Patterson was on the top of the mountain doing some work on CVs, and I knew him from conferences and things like that. And Joe says, "Hey, I've got one you might like to look at." And he gives me an object that turns out to be really bright and has features you can see, and there's a point to observing it. I remember just sitting there in the control room looking at this bright star's light spilling around the slit and saying, "I'm finally getting something useful. This is wonderful." And then I started reading up on cataclysmic variables. It turned out there was a huge amount of commonality with the X-ray sources I was working on. They had accretion discs, they had secondaries, and this, that, and the other thing. But they were bright, and they were within reach of my equipment, which meant that I could actually make progress with those.

So, that's where I turned the turret. I turned the turret around to cataclysmics because they were related to problems that I knew about and had a background in already, but they were tractable. I wasn't just spinning my wheels. I didn't have to wait for a giant telescope to come along or compete for a teeny bit of time on Kitt Peak or something like that, the 4-meter. I was able to actually do them from MDM where I had guaranteed time, and I just started cranking them out.

I've been doing that ever since with an interregnum, actually, that Mike Kurtz -- again -- who was at Harvard-Smithsonian, got me involved in a redshift survey with Margaret Geller and John Huchra. Back in the day when redshift surveys could be done by individual investigators -- it wasn't like competing with General Motors or something like that. And I did some work on that. It didn't work out as well as I would have liked, but we did get some papers out of it at least, the century survey strip of the universe and such. This was around the time that Margaret Geller and Valerie de Lapparent and John Huchra found the structures, the voids in the distribution -- the filaments and voids. They really had one of the first wonderful papers about that. So, I got involved in that research to some extent. And I worked on that for a while, but then it felt -- when I came back to working on cataclysmic variables, it felt kind of like coming home. I remember one run where I was doing redshifts where I turned around and worked on a star called PG 0027, which was one of the cataclysmics, and it was like, oh, look at this. I'm getting data that I can see and do stuff with right here, and I don't have to wait until I have 5,000 redshifts to tell anything -- I can see it right here. 

Zierler:

John, were there particular advances in instrumentation that brought you back to cataclysmics?

Thorstensen:

Not really so much. I think the great that that's happened in my generation is of course CCDs. Finally moved to electronic detectors where everything -- when I was at Berkeley, we used Wampler's Lick Image Dissector Scanner a whole lot on the 3-meter, which was an early digital detector. That was revolutionary at the time and fantastic to have, but when CCDs finally got to be up to speed, it was like a whole different thing. I remember in graduate school thinking; “Wouldn't it be great if you could have a whole bunch of tiny photon counters in an array? Then you could actually not worry about these awful properties of photographic emulsions.” They're so nonlinear, and what have you. You could actually get quantitative measurement out of it. The quantum efficiency would be much higher, and so on. And CCDs came along, and once they got the read noise down, they were almost the ideal detector. But that didn't drive my scientific interests that much. It made it possible for me to do the things that I wanted to do anyway much more efficiently and effectively than before. 

Zierler:

Who were some of your key collaborators beyond Dartmouth? Who was making this possible alongside you?

Thorstensen:

Oh, Joe Patterson -- I think he's probably my longest term -- we proceeded to collaborate on a bunch of different papers. I haven't been working directly with him all that much just lately, but we certainly still are in touch. Early on in my career, I was still collaborating with Phil Charles quite a bit. We actually had some runs at Cerro Tololo to work on X-ray sources and had some real good success with that as well. So, I was continuing actually -- part of my effort in my first few years at Dartmouth was continuing some of the stuff I was working on in grad school, except this time with runs at Cerro Tololo. He would send down students to meet me in Chile, and say, "Teach them how to observe." And that's what I did. I've taught a lot of people how to observe. A lot of people.

Another thing, Dartmouth has an off-campus program in South Africa, which hasn't run lately because of COVID, which is too bad, but I've been on all three. It was inaugurated in 2015, and I was on the first of those and got some collaborations going with South African people because we worked at the South African Astronomical Observatory. Those have been just very nice people to work with. I'm trying to think who else recently. During the time I was working on the redshift surveys, of course, Margaret Geller and John Huchra were important collaborators. I wrote three big papers last year during pandemic times because I couldn't go observing. 

Zierler:

Yeah, it's a good time to write. 

Thorstensen:

I had all this time to write. The other thing is I'm almost not teaching now because I'm burning sabbaticals before my retirement. You have to burn your sabbaticals before you retire, so that's what I'm doing. And then next year I go on part-time. So, I'll do a phased retirement over three years. Anyways, so last year I wrote three papers, and two of them were single author. So, basically, I'm not collaborating as much as most people do. I'm not in this giant matrix of people working on a giant project or something. It's just me. It's an unusual way to operate these days. Almost nobody does this. 

Zierler:

Do you have a specific memory of when astronomy at Dartmouth came into its own, and if there was ever a discussion at that point about a breakaway department, which has happened so frequently elsewhere?

Thorstensen:

I don't think there's ever been really serious discussion of a breakaway department. I think that Gary Wegner used to -- he spent several decades here --Gary Wegner used to be a little resentful of being lumped in with the physicists and so on sometimes. But not particularly. But I think that a thing which essentially what we used to do was basically force all of the astronomy grad students through the physics program, as I said. And that just didn't work very well. The students are somewhat different. They have different interests; they have different strengths. Fortunately, I think Brian Chaboyer, who is a fantastic guy, sort of managed to tease away pieces of the astronomy graduate program, and also Ryan Hickox, I believe. I don't remember exactly who did this, but Ryan Hickox, we hired ten years ago or something like that, or maybe a little less. And he's been a tremendous presence and also helped to cement the new regime in place where the astronomers do much more research-oriented track than just a bunch of classrooms and class things and big sit-down formal exams like the physicists do. 

Zierler:

John, a technical question. Do you always use photometric and spectroscopic studies together in trying to understand cataclysmic binaries, and if not, when is the decision to use one or the other?

Thorstensen:

It all depends on what kind of data I get, and whether it's enlightening. A photometric study oftentimes tells you something that the spectroscopic study does not, or vice versa. But sometimes it's just like, well, yeah, it sits there and flickers or something, and the photometry is just not that interesting. So, both of them, they're complementary. It's kind of a question of what's appropriate in what case, and also what data I happen to have. We're very fortunate because we have two telescopes in Arizona. In fact, next week, I'm going to be sitting with the big telescope doing spectroscopic studies, and then the smaller one doing photometric stuff, sometimes in parallel, sometimes not. 

Zierler:

Do you have a sense of when computers became not just nice but really vital to analysis of all of the data? Did that happen slowly, or was there a particular observational run where it just dawned on you there's just too much here for human brains to deal with on our own?

Thorstensen:

I think that the big thing is the digital detectors coming in. Before digital detectors, it would have been possible to do something entirely analog. So, basically, you take photographs of something, you examine them by eye, you write down your impressions, and you have something. And there's never a computer involved. Once a digital detector is involved, there's something reading the data and putting it together in memory and writing it out. So, the computers become part of the chain at that point. And then, from there on, I think it becomes pretty much essential. That happened, actually, before I even got to Berkeley. They had the Wampler scanner, the image-dissector scanner on the 3-meter before I got there. It was almost new when I got there, but that was when it actually happened. 

Zierler:

A nomenclature question. I'm curious the origins of the colorful term superhump, and when it became attached to cataclysmic binaries. 

Thorstensen:

Superhump. The first I heard of it was a paper by Joe Patterson called Superhumps and Super Outbursts, which was just a review of what this was all about. I don't know who first thought of the term superhump, but what it is, is that when certain cataclysmics go into -- certain cataclysmics have outbursts which are especially strong, and others which are normal, which is not quite true, but they have especially long, strong outbursts which are called super outbursts. Some CVs have exclusively super outbursts and never have normal ones.

In any case, what happens is that observationally, the thing gets really bright, and a few days later, light starts oscillating by 10-20% maybe. The light curves are very striking. It goes up, and it wiggles a little bit, and then it starts oscillating as it slowly decays over a week or so or maybe more. And that oscillation, it turns out, you think, oh, that's the orbit. The answer is it isn't quite the orbit. It's a couple percent longer than the orbit, and that's huge. When you're counting a bunch of cycles -- a couple of percent doesn't sound like much, but when you're counting a bunch of cycles, it's an extremely significant difference. It goes way out of phase with the orbit. So, there's something else going on. Apparently, it's a precession -- an accretion disc somehow becomes eccentric, and then undergoes a slow precession. Basically, the timing of when the star catches up with the line of nodes on the disc, which is going to be a little bit later each orbit because of the fact that the disc is slowly precessing, that's what gives you the periodicity. That's the underlying clock. It's a little bit like sidereal and solar time. 

Zierler:

How did you get involved initially with the MDM observatory?

Thorstensen:

I took a job at Dartmouth. 

Zierler:

So, it was right there at the beginning. That institutional connection was built in. 

Thorstensen:

It was already there. I didn't have to do a thing. 

Zierler:

Do you know what the backstory is there?

Thorstensen:

Oh, yes. So, Dee Mook, again, was a grad student at the University of Michigan, and he worked for Al Hiltner. Al Hiltner was the chair of the department, at least for quite a while, a real powerhouse personality. Remarkable guy. Wonderful guy. Dee Mook -- now, way back in the day, I think, in probably the very late '60s or very early '70s, like 1970 or something, the University of Michigan built a 1.3-meter telescope, a 52-inch at the time, in the woods on a park reservation, or something like that, not too far from Ann Arbor in Michigan. They have a building and everything, and it was all elaborately set up. There's an article by Peter Wehinger, I think, in Sky and Telescope from about 1973 or something like that, which shows the new telescope that we have. So, they quickly discovered that it's cloudy in Michigan a lot. And Hiltner got the idea remarkably fast that they really ought to just move the whole thing to a better site. Then, there were a bunch of negotiations.

Meanwhile, I'm not sure of the exact timing, but Dee Mook took a job at Dartmouth around that time. And somehow, through a whole bunch of phone calls and letters and whatever, a collaboration was formed of Michigan and Dartmouth and MIT to move the Michigan telescope to Kitt Peak and operate it there with a consortium. So, it all happened right about them. Right in, I'd say, the mid '70s. I think the telescope took first light on Kitt Peak in '76 or something. So somehow before my time, all of these crazy negotiations and stuff happened. I remember people saying kind of in a wiseacre kind of way that the telescope was Mook's dowry, because he had actually married Al Hiltner's daughter in the meantime. By the way, if the name Mook sounds familiar, do you remember Robby Mook? 

Zierler:

Yeah, sure. 

Thorstensen:

That's Dee's son. 

Zierler:

Oh, my goodness. 

Thorstensen:

Hilary Clinton's campaign manager in 2016. 

Zierler:

Right, right. 

Thorstensen:

I kind of cry when I think of 2016 anytime. But in any case, that's when they moved the telescope to Arizona in '76. By the time I was hired at Dartmouth in '80, it was kind of clear that Dartmouth wasn't getting as much out of it as they thought they might. That's one reason why I was such an attractive candidate, because I was kind of a telescope jockey. 

Zierler:

And even though it was rather forgettable to you when I asked in the beginning about your affiliation as president, when did that happen, or at least when were you on the trajectory of leadership?

Thorstensen:

That's a very funny question actually, because what happened was that the consortium was reorganized a couple of times. In early days, there wasn't even anything written down really, and I think -- I want to say about 1990 or something like that, what was it? MIT became very restive, and Paul Schechter had been incredibly helpful in getting things together. So, maybe it was a little later in the '90s, but Paul Schechter had been very committed to the observatory and done this and that. He was actually the person who pushed the hardest for just to get the optics of the 2.4-meter refigured, which was an enormously wise thing to do, especially in retrospect. He helped with some of the instrumentation, and stuff like that. In any case, basically, what happened was the MIT came to realize that this was not enough observatory for their ambitions. And they wanted to get into something much bigger, but apparently the way the politics worked at MIT, they couldn't do this without sacrificing something.

So, they would sacrifice their involvement with MDM. They announced that they wanted to pull out of MDM, and then there was this astonishingly contentious business of who was going to replace them. MIT, weirdly enough, somehow felt like they really, really wanted Ohio State to replace them. We were thinking maybe Harvard would replace them, because Harvard was interested as well, and Harvard would have been a terrific partner. We were also working with Harvard to some extent at the time. And then, it's really odd. Why do they have standing to decide who's going to replace them? So, what we decided after, there was kind of a mess, and it all ended up working out okay. Ohio State bought a part of it. Columbia bought a part of it, and so on.

But also, as part of the whole trauma of the reorganization, we decided we had to have an actual charter drawn up, a legal document describing just what this is. So, we organized as a nonprofit corporation under the laws of the state of New Hampshire. And it's owned by Dartmouth, and Columbia, and Michigan, and Ohio State, and Ohio University in shares. Now it's all spelled out. Well, if somebody wants to leave, or sell down part of their stock, or get more, or part of their share, here's how you go about it. You offer first right of refusal to the other partners. They have this long to respond. Then you go out on a -- so, it's all spelled out. So, this kind of foolishness about people getting angry at each other over these things is gone.

I remember one meeting when John Tonry was trying to convince us to get this all wrapped up and do what he wanted. He said, “Maybe we should just do this as astronomers, and not get the lawyers involved.” And I remember turning around to him and saying, because of the fact that I have all these lawyers in my family – there’s teachers but there’s also a bunch of lawyers, and they’re great people. I said, “Sometimes lawyers are your friends.” If there’s anything I hate it’s people who just blanket dismiss lawyers and politicians as being all bad, because they’re not. But that kind of took the air out of that foolish argument. In any case, how did this go then?

So, then what happened was I think the first president was Doug Richstone at Michigan. He basically seldom if ever observed. He was the chair at Michigan, and things kind of went along okay. We had some reorganizations from that. And then, Jules Halpern did it for a while at Columbia. He's excellent with detail. Really, really good with detail. And then he didn't want to do it anymore, in part because the observatory kind of seemed like it was drifting along. It wasn't clear what was going to happen with it or how it's going to work. So, then basically he said, "I don't want to do it anymore," and people sort of turned to me, I guess. Here I am, a relatively senior professor at one of the major shareholders. I observe all the time, so I have a good sense of what it's like on the ground. So, I said, "Sure, I'll do it."

So, basically, it fell to me by default. It wasn't like I ascended to a position of leadership or something like that. It fell to me by default. There had been all kinds of things, personnel stuff at the observatory, which took up a long time -- one thing which was truly bad, I had an employee who essentially went off the rails. We had another great employee who had some drawbacks, but eventually he retired, and now I think we've just managed to put together a tremendous staff. They basically run the place. They consult with me about stuff, but they basically run the place. The other thing is that the observatory is administered through Dartmouth, and we have a person in the administration, Kate Soule, who just loves to do this kind of stuff as a service, who's also an English professor's kid, who likes to be of service to things. So, she actually has been great at providing professional backdrop for, like, “Okay what about the budget?” She budgets the whole damn arts and sciences. She does MDM on the side. So, basically, the place runs itself pretty much, except it needs some direction from time to time.

What I have done, I think one of my contributions, or the contrition to the observatory that's most significant, is that as director, I observe all the time. That's very seldom the case in astronomy that you have a -- the director is usually somebody who sits in a remote office and tries to figure out what's going on, but I'm there all the time. We have a tiny, tiny staff, and if I see something that's -- there've been a couple of times where I've said, "This instrumentation is just really not up to snuff. We ought to figure out if we can do something to rig something to make this more effective." And I've done that a number of times, and actually multiplied and helped the power in place quite a bit. The other thing I've done is I've written documentation, and I've written software. I wrote a code, and I have a whole other side which is kind of what my dear colleague in the department, Jim LaBelle, calls recreational computer programming. And I'm also observing all the time.

So, many, many years ago, actually weirdly enough -- this thread actually starts when I was an undergraduate and I wrote a code to write a map of the sky as seen from anyplace anytime. It's 1972 or '73. But I've always kind of liked time in the sky. It just has this aesthetic appeal to me. Where are the things in the sky? The geometry of it is beautiful, and the other thing is that it's so wonderfully predictable. It's great for computer stuff. So, I've written the time in the sky code over and over and over again. Many times. 

Zierler:

Why does it need to be written over again?

Thorstensen:

Because the old systems die. The first one I wrote is an APL, which is a language called A Programming Language. It was Ken Iverson's great symbolic thing. It is possible to find APL these days, but nobody ever uses it in astronomy. And then I wrote one in PL/I, which was, weirdly, the language at Dartmouth when I arrived. Then I wrote it in C; then I wrote it in Java; then I wrote it in Python. So, it's like crazy stuff. One after the other. Systems died, and basically old, old software doesn't work after a while because you can't find something to run it on. But the Java version called JSkyCalc, and is probably my greatest contribution to astronomy, actually, which basically shows you a map of the sky, and it's got these controls and stuff like that. It's sort of a little like a desktop planetarium program, except it's designed to tell you the actual number that a professional astronomer wants, like on the spot.

And people love it because it's so easy to use and intuitive. You just fill in some numbers, and you click on this, and you see what's happening. And I wrote it for use on the telescope. So, basically, it's like 2:30 in the morning. I've got these targets, I've got this, I've got the moon over here, I've got this. What can I do? And you can very quickly figure out all the observability constraints. Okay, well, then how long do I have before sunrise? It's all right there. Anyway, I took that and adapted it to MDM and interfaced it with telescope control systems so that now you can use that to figure out what's going on, and then tell the telescope – I like this target here, I'll click on it, it'll tell the telescope to go to it, and there it goes.

So, that kind of stuff has been really useful for the observatory. People liked it enough that somebody at Palomar, Jennifer -- what's her name? Milburn or somebody at Palomar, one of the programmers at Caltech, said, "Hey, can I steal this?" And she proceeded to take it and did a wonderful job of refactoring it, because it's a terrible piece of code in computer science terms. But she put it all back together, and now, at least as of a couple years ago, it ran the 200-inch at Palomar. So, that's a nice thing to have done. The other thing is that people know me through this. I'll be in elevators at conferences, and somebody will say, "Oh, you wrote that. Well, thank you." So, it's this tremendous publicity, or at least name recognition generator. Who's going to read these papers about these obscure stars I study? But they'll use that. 

Zierler:

John, to bring our conversation closer to the present, what have been some of your interests in recent years in white dwarf binaries?

Thorstensen:

Oh, I haven't actually done hardly anything with white dwarf binaries. I do have an occasional collaborator, who's been a great collaborator, named Stephane Vennes, who is now in the Czech Republic. He was at University of Montreal for a while. Then he was at Berkeley under Bowyer when the EUV Satellite was flying after I left. But now he's in the Czech Republic. But he is a great white dwarf guy. He and his wife and collaborator, Adele Kawka, have touted me onto some nice objects.

One of them was a fabulous, very low-mass white dwarf in orbit around another white dwarf, which being higher mass is therefore smaller. White dwarfs are weird because they have an inverse relationship between mass and radius. The two of them are orbiting, and you can only see the low-mass one, but you could see the line shifting as it orbits, and apparently in the photometry, you can see these subtle effects due to the fact that one star is lensing the other one gravitationally. It's just great.

But the big discovery, which was an actual Nature paper, was one I had not much do with, but I helped with. That was one just three or four years ago. That was one in which he says there's this weird spectrum of the star. It had this very, very high velocity and these weird lines. And he just said, we need to know if it's moving. So, since I had a run at MDM, I just hammered it every night. It wasn't that faint. So, I could get a nice spectrum every night. The lines are just sitting there and they're -- I don't know -- 900 km/sec, and they're all from some bizarre heavy elements. That's this one which is this super high velocity star ripping across the sky. Now, with Gaia, they've got an actual distance. They've got a proper motion. They have a transverse velocity, because Gaia is so good. And we have the radial velocity and so on. It's easily escaping from the galaxy, and the only thing that makes any sense is that it is a failed supernova, which is a phenomenal discovery. I had very little to do with it. I know Stephane; Stephane knows me. He said, "Can you get some spectra of this and see if the velocity is constant?" I did, and it was. So, I can't claim any credit for being the intellectual author of this at all, but I helped. 

Zierler:

John, between your teaching interests and your perch as a senior person in the field who's looking at becoming emeritus at some point, I'm curious what your sense is in terms of what graduate students in these research areas are interested in, and how that might suggest where the field is headed from here?

Thorstensen:

Well, I think basically astronomy is becoming dominated by giant surveys. So, graduate students in binary stars are interested in doing things like -- largely interested in taking enormous datasets from enormous projects, and just mining them for all they're worth. We've come to the era of largely keyboard astronomy, as people say, where you just sit and interact with an enormous dataset through your keyboard and mouse. There is tremendous interest in machine learning, because again, these datasets are so huge that you can't look at it all.

Curiously, Michael Kurtz, who I mentioned earlier, foresaw this in 1981. He wrote one of the first theses on automated spectra classification. He saw that as a problem which at that point was very quickly going to come into prominence because there was no way that a human was going to look at all of those spectra anymore, because there were just going to be too many of them. He's a brilliant guy, Michael. Very foresightful. And then he went and figured out the ADS. He's one of those people who you've got to talk to at some point. But where it's going, I think it's going to places like, what can we learn that's from these enormous datasets? They're there. That's where the frontier is at the moment. We're getting comprehensive view of these things, and real numbers for how many there are and things like that.

In the meantime, in my own part, as these enormous datasets develop, they also throw up huge numbers of things that need follow up. I hadn't mentioned ASASSN yet. I don't know if you've heard of that, but it stands for something, Automated Sky something Supernovae. What it is, it's a beautiful little project based at Ohio State. It's the severalth generation of projects that started in Poland, actually, with the Copernicus Institute. Bohdan Paczy?ski, who was one of the great souls of astronomy -- Paczy?ski and his proteges realized, why don't you just photograph the whole sky all the time and see what kind of variable stars you find? It's the electronic descendant of the Harvard plate stacks. It's gone through various generations.

Now, the current one is this one that actually has these -- they're little telescopes that are about this long -- like, 6-inch or 5-inch telescopes. Just telephoto lens I think they buy from B&H Photo. Those are feeding these fairly large, but also off the shelf CCD cameras. The mounts, they had like four of them on one little mount, and they're all automated, and they're all over the sky. There's like four or five stations. There's one in Chile, and one in South America, and one in Hawaii, and various other places. And they basically get the whole sky every night with these. Essentially, the Zwicky Transient Facility is doing this similarly, but without quite as much comprehensive coverage. The thing about ASASSN is they're kind of crappy little telescopes, and they can only find stuff that's fairly bright.

So, everything they find is fairly bright. But things that just pop up into visibility with a big telescope, they're going to fade down to the point where you would need like a 10-meter for a week to get anything. 99% of what LSST is going to get is going to be stuff like that, plus of course, a haul of amazing supernovae, which is one of the main science drivers for LSST. Some vast amount of the stuff that LSST gets is just going to be impossibly faint to follow up. But ASASSN, most everything is something that people could get. So, that's been a huge thing.

The other thing I wanted to mention is I've become even more appreciative of the enormous contribution of amateurs in this one field. Amateurs don't make that big a contribution across large swatches of astronomy, but in cataclysmic variables, the things come up to be pretty bright, and a lot of them, there's a guy in Belgium, Tonny Vanmunster, who follows the superhumps -- he gets a dozen superhump periods a year or more out of that. Another thing is that there are these people largely in Eastern Europe who somehow find the time and the energy, despite it not even being their job, to look through these enormous publicly available databases of variable stars and pick out interesting ones. There's a guy in Poland, Gabriel Murawski, who I've just come to realize -- he's characterized 3,000 of them, and he's a dentist. It's amazing. 

Zierler:

I guess, you take whatever help you can get. 

Thorstensen:

He's a dentist, and meanwhile he's come out with 3,000 things. He says, "Oh, this one's one of those, and that --" And they're all new. It's unbelievable. So, yeah, these huge surveys, it's this wild ecosystem of new stuff going on which is really fun. In my field, the particular niche and particular skills I have, the recreational computer programming for sifting through the 3 million stars in the VSX database and trying to figure out which ones are this and that. Those things are all still going along. So, it's very nice. 

Zierler:

Well, John, for the last part of our talk, I'd like to ask a few broadly retrospective questions, and then we'll end looking to the future. So, the first one is funding sources. Who have been some of the key institutions that have funded your work and that of your closest collaborators?

Thorstensen:

Well, basically, for me, the major funding source -- the thing is that I'm in a funny position in a way, in that I'm one of the unusual people for whom money doesn't matter that much. Fortunately, as a personal thing, I've always lived deeply within my means and stuff like that. I don't have expensive tastes. So, I'm not out for personal aggrandizement hardly at all, except for name recognition. I love name recognition. But for money, it doesn't -- and for much of my career, I've been unfunded, which is wild. However, I did get some nice NSF grants, especially mid and a little bit later career, to hire students, keep the enterprise going, pay the page charges, get me out to Arizona and back, and so on. And those have been, of course, deeply appreciated.

Also had a little bit of funding from NASA when I've proposed for satellite observations of some kind, but I've done very, very little of that over time. Oh, yeah, the other thing was that the redshift survey work that I did had some funding, partly because the hard parts of those proposals were worked on by the folks at Harvard who were the best in the world. So, that helped a lot. But I've given up at this point -- I was applying for funding until a few years ago at NSF, and I actually was successful until 6 or 8 years ago. I had a succession of modest NSF grants that kept things going. But then, I think a few years back I just decided that it would actually be almost more ethical to leave the money on the table for people who need it more. And also, because all I need to do is get myself to Arizona and back a few times a year, and I can do that easily. And also pay page charges when I publish papers, which I would like to do more. Some results can get published by other people who pay the page charges. I also have a modest pot of money built up, which should be enough for some years of that. So, basically, I'm such a small and individual operator, I just didn't need that much money. 

Zierler:

Given your background, your family business in teaching, and how important it is to you personally, what's been most satisfying to you? Ranging from those big classes where you're teaching non-science majors to perhaps upper-level classes where you're really formative and encouraging students to pursue this in graduate school. 

Thorstensen:

I think probably for me, you get more immediate gratification from big classes. Actually, it turns out also, the other thing is that I haven't taught astronomy in quite a while. Hardly any. Actually, I've taught some observational classes, but that's about it. I mostly taught physics, and I mostly taught intro physics to scientists and engineers. So, that's where the department needed my talents. That's very rewarding. People actually do go on to make use of this stuff. I think the big astronomy classes were very rewarding to teach partly because you get such an incredible cross section of the student body, and you see so many different people with so many different motivations.

Speaking of funding, it was kind of amusing. I had a student in one of my classes years ago -- this didn't end up being consequential at all, but it was just funny. This one guy whose name was Van Citters, and his father was Wayne Van Citters, who was the head of the astronomy division at NSF. So, Wayne Van Citters held the purse strings for my whole field, and I had his kid in my class. And Wayne -- it turned out that this Van Citters kid turned out to be an excellent student who just loved the class. So, I liked that. He's now actually the chair of the engineering department at Dartmouth. So, he's gone on to other things. But basically, it's been very rewarding to see students who you teach and help form going on to being great. It's just incredibly rewarding to see that, and I've seen many students go on to be just great. It doesn't always matter what the field is. There's a woman named Gwen Rudie who's at Carnegie who's phenomenal, and I taught her first astronomy class. I taught Parker Fagrelius her first astronomy class. She's running DESI. And so on. So, those have been things that I've found -- I think just seeing students blossom, take stuff in and then blossom and go on to become great is what's really the most rewarding about that. 

Zierler:

John, last question, to tie it all together, past, present, and future, for you, cataclysmic binaries is the gift that has always given. So, I'm curious, in all of your work, in all of your research, what surprised you the most, and what surprises might be out there that you haven't uncovered yourself?

Thorstensen:

Well, that's a good question. I think the most -- the moment that sticks in my head as being the single most surprising thing was when I was working on a star which has some phone number, but it's known as EI Piscium. This star already was weird because it has 65-minute photometric period, a superhump or something, and it's 65 minutes. That's already short. That's below the -- most of them, they cut off at about 70 minutes, unless they're helium secondaries, in which case they go down to very, very low periods, even 5 minutes. But 65 minutes is already an anomalously weird period. So, I was looking at this, I analyzed the data, and I took it and stacked it up into a grayscale display which looks as if you're following the spectrum through the orbit. So, each line of the image was a different orbital phase, and you could see the bright lines and the faint lines and stuff like that. It's almost like an old-fashioned strip spectrogram, but with time axis on it. I brought this up on the screen, and my jaw dropped because there was a K star, an obvious K star wiggling back and forth with a 65-minute period. That was -- for an object of this kind, the secondary has to be an extremely low-mass star. And a K star is not extremely low-mass.

So, this was the most dramatic example ever of a cataclysmic in which the secondary star had begun to evolve significantly before all of the rest of this orbit rearrangement happened, so that what you were looking at was the stripped remnant of a normal star. And the reason why it was a K star, and I figured this out pretty quickly, was because essentially it has much more helium in it than it should because of the burning products, which changes the whole stellar structure in a way that it's much hotter than it should be for its mass. I found a similar one -- that moment was like so instantaneous that I just reached for the phone and called Joe Patterson, because I was working on it with him. I remember just looking at this and literally my hand just starting to reach for the phone. 

Zierler:

And there's no question who you're going to call.

Thorstensen:

No, I was going to call Joe. And Joe was flabbergasted as well. So, we wrote a beautiful paper on that. I thought it was a really nice one. It involved Isabelle Baraffe from Europe, the stellar structure person. I think probably one of the most significant things I found also, which was slower developing -- oh, yeah, actually the discovery moment in this was also fairly sudden. There's a little bit of a story, but Howard Bond, who you may know -- he's at Penn State. He's also somebody you should interview. He's amazing, and he has an enormously interesting career. He discovered the first optical counterpart of a gamma ray burster. He's got chops. He's really great.

In any case, Bond had written a paper about the first radio selected cataclysmic, and it was discovered by the FIRST survey at the VLA. That's its acronym. So, it was already kind of a joke -- the 'first' radio-selected cataclysmic. So, they had this star, and it was a radio source you could see. It flickered a little bit, and it had some emission lines, and it looked like a cataclysmic variable. I actually refereed that paper, and I said, "This is fine. This is good." So, then, since I just look at every goddam thing I can find, I was at the telescope, and I said, "I wonder about that first star." So, I just took a look at it, and I went over and set on it and took a spectrum of it. It was like the last night of the run. I took a spectrum of it, and it was a G star. It had a solar-type spectrum. These are the vermin of the skies. They're all over the place.

So, it was like, that's not a cataclysmic. I wonder what happened? So, it was the next year by the time I got around to looking at it again, because it had gone into the sun and stuff. And this time, I said, "Okay, I'll bet I got the wrong star." I had assumed I had the wrong star. I very seldom get the wrong star because I'm very careful, but I might have gotten the wrong star. I might have been in a hurry. I don't know. So, I go back and look at it, and this time I'm very careful with the chart, and make certain that I have everything perfect. And it's a G star. And I went back, and I think I looked at it again the same night, and it's a G star. So, I said, "Okay, now I'm going to be really, really certain it's a G star." So, I went back -- so, I had these spectra of a G star, and the next day -- I reduce my data right away and analyze it almost immediately, and the next day I have the reduced data, and I remember going, "What the hell? This thing is a G star, but it's the same object as before."

So, it occurred to me to measure the radial velocities of the two spectra. They were different by 400 km/sec. I said, "Oh. These two spectra taken a couple hours apart are different by 400 km/sec. This thing is a binary." So, I continued to hammer on it for the rest of the night, and the rest of the run actually. I keep going back, still a G star, still a G star. It's like Bart Simpson with his science experiment -- “still just a potatoe”. It's still a G star, but it's oscillating with a 4.8-hour period. So, I think, well, okay, this is very interesting because here you have this cataclysmic thing that turned off, and now you're just seeing the secondary. But a G star at a 4.8-hour orbit is already too hot. It's sort of like the EI Piscium thing, except a little less extreme. So, that was already weird.

Meanwhile, Brian Warner and Patrick Woudt in South Africa, who are now dear friends of mine, published a paper showing photometry of a whole bunch of things, and they show this thing, and it has this oscillation at 4.8 hours, very regular. It kind of comes up, and is flat, and then goes down to a big broad dip, and is flat. And Brian describes it as a heating effect light curve. I thought, well, that's interesting. So, I went and got Joe Patterson's protege Eve Armstrong to take some data with the 1.3-meter, some photometric data, and I got that and started putting it all together to try to figure out what's going on. It looks as if -- and sure enough, it showed the same light curve that they did. Sure enough, this G star is being heated on one side by an invisible source of radiation. What's going on?

So, I had a long delay while I wrote code to figure out the illumination of this star, and how bright it should be, and synthesize a light curve to see if I could constrain it. One of the parameters is the inclination of the orbit. Are you looking at it face on, or are you looking at it edge on? I could tell it couldn't be too edge on, because otherwise the light curve would have more amplitude. So, when I constrained the inclination, it turned out that the object could not be a white dwarf. It was too heavy. It was an exotic neutron star binary of some kind, or possibly a black hole, but it was probably a neutron star.

So, I published a paper that said, look at this thing. It's wild. You have a G star, which is also only .2 solar masses, which can't possibly be right, but it is. And look at this. And then this paper just sank without a trace. I think one person followed up. There's one follow up paper where some people took some X-ray data or something. It sank without a trace. Glug, glug, glug. Until -- I think it was two years later -- I get a phone call from Scott Ransom at NRAO. He says, "John," -- I hardly know this guy. He says, "We looked at your system. It has a 500 Hz millisecond pulsar in it." Just all of a sudden, it was a combination of, “Oh my God, I was right,” and “Why didn't I push it just a little farther?” So, this thing turned out to be what's known as the first redback. It's a transitional object between low-mass X-ray binaries and basically detached systems. So, what happens is sometimes, when Bond saw it, and there was later another outburst seen, it flares up when the G star fills its Roche lobe somehow, spills some material over, creates this accretion source which is like a low-mass X-ray binary, and looks a little like a cataclysmic, and then it turns off. And that's when I studied it.

So, the thing is because of the work I had done before this, Scott Ransom, when they found this thing, they were randomly pointing the greenback telescope around because their steering system was working right, and this thing just blew them off the console. They just happened to point the radio detector in the right direction. Then it's, what the hell is that? They go into SIMBAD and start looking around to see what it is. Oh, that's it. And sure enough, very quickly, they had the orbit because the Doppler effect of the pulsar, it's the same object. It's clearly the same object. So, that became cottage industry. People did a lot of work on that. But it was just because I was sniffing around at all the different cataclysmics that I happened to have this all ready for them. I was really glad I had the perspicacity to say this is something really unusual and I should push this paper to fruition. 

Zierler:

Intuition might not be a scientific concept, but it can certainly move the science forward.

Thorstensen:

Oh, yeah. Well, sometimes you look at something, and when something just isn't right, that's often really the time you're going to make a discovery. It's often just you screwed up, and it just isn't right. You got the wrong star or something. That was my first instinct. But, no, I didn't. I got the right star. 

Zierler:

John, it's been an absolute pleasure spending this time with you. I'm so glad we connected through that email from Nick Suntzeff, and I'm just so happy that we were able to do this. So, thank you so much. 

Thorstensen:

Oh, yeah. This is not exactly a typical path through science, or trajectory through science, my own career, but it somehow all worked out. 

Zierler:

It somehow all worked out.