Interview with Nils Andersson & Ian Jones

0:00 It's the 23rd of February, 2000, at 20 past one in the afternoon, and I'm speaking with Niels Anderson and Ian Jelman. That's right. Sounds very official, doesn't it? That's right. It's probably a good idea that there was a trouble feeling. Who's that? Is that Niels? So, I don't know, maybe a good place to start is to get an idea of the general research that's being done at Southampton at the moment. You mean in gravitational waves, sources in particular? Yeah, I guess in gravitational waves. For instance, a couple of things I'm aware of are, for example, your work on Arbonne's possibilities and neutral stars and raise stuff on Koshi characteristic matching and that, but beyond that I'm not too. Well, if you add to that Ian's thesis work, which is not strictly, I suppose, originally done here, but... It's being done here, it should have been finished elsewhere. on processing neutral styles if you add that then I think you've more or less covered the themes and your black hole perturbation stuff yes in a way but I think to be a bit unfair to myself perhaps I think I've always done that more for the interest in just general analogies with other problems etc. peculiarities I've never really seen it as example for gravitational waves. But that's just, I suppose, my attitude to it. I mean, you can see the stuff I do with Costas in Cardiff as being relevant, perhaps, in the extension for gravitational waves. But I think the way we're doing it is a little bit like a tangent, where we're looking more at the mechanisms and how they work in detail and thinking about, you know, what does this mean for LIGO? I'm curious actually just touching on that subject as to whether there is a tendency in your mind for topics which have been around NGR for a while to get sucked more towards gravitational waves are more towards

2:30 because of the fact that big detector projects around that doesn't make sense it does, I think it's a good question and I think, in my mind I think there's definitely a trend there I mean there's a huge number of people now that are working on gravitational wave sources in some sense or other, even now I mean it's almost done in astrophysics as well, whereas a few years ago that would just not happen and I think it's I can't remember now I think it was when I was in St. Louis they were writing a paper on gauge conditions and numerical relativity technically it has absolutely nothing to do with gravitational waves I mean it was a mathematical choice of gauge etc and the draft paper started off numerical relativity is important sent a couple of new gauge choices for numerical relativity. You know, they made this tie-in with LIGO that I thought was absolutely fantastic, because it was so far-fetched. I mean, if I had refereed that paper, I would have said, oh, come on! It doesn't mean your stuff is not important. It could be important in its own right. So I think there's a lot of people doing that, trying to make connections with this, because they, maybe, because there is a serious pot of money And I reckon they think more people will read their paper as well, if they can put the buzzword in near the beginning. It could be. Almost all of the wave papers start off by saying, in fact, it's almost the same sentence, isn't it? And it's almost the same paragraph in the next decade. For the first time, we may have a set of gravitation. We have to take this brackets, then LIGO. Not just LIGO, but it's always the same. All of the LIGO, Virgo, Geo, Tama. It's probably in the order of how much money is being spent on the moon. I thought it was a size measurement. It's like measuring up your body parts, saying, oh, mine is longer than yours, and therefore I go first in the last. I don't know. I think that's right. very interesting, I think, would be to compare, as you say, the first paragraph of a large

5:00 number of papers on related topics, but maybe not immediately, sort of, not on gravitational wave data analysis or anything like that, but maybe sources is a good one, to see how many of these key phrases are sort of stolen from some original paper back in the 70s or something, probably by Kip, you know, always to say, in the next decade gravitational waves it's very excited and therefore we're going to have to build this large-scale interferometer kind of thing and then to see how those sentences have essentially not changed at all but just nothing it's like a carbon copy in it a lot of papers every year i think And so for theorists, do you think, you mentioned the sort of pot of money angle, too, that people might be inclined to go towards gravitational waves, so do you think for theorists in Europe that that's a sort of factor, I mean, I've heard of people in the States discussing whether there's more money for gravitational wave work than for types of GR work, and I was wondering if Well, with the limited insight into how these things work, given how you're not just allowed to apply for money for two years, it's hard to say really. But I think, I get the impression that, I mean, there are a lot of branches of relativity, in particular more mathematically minded relativity that are basically on the way out. And gravitational wave and maybe relativistic astrophysics is growing in importance. It used to be that relativistic astrophysics was relativists that liked to think about astronomy as well every now and then. But now it's almost as if astronomers are realising as well that actually this relativity stuff might not be completely useless. So I think that's an important change. And that, of course, means that as a relativist, if you are applying the money to study what an astronomer might think or an astrophysicist might actually think is relevant

7:30 sort of neutral star fluids and stuff, something like that in relativity then it's much easier to convince them that it's worthwhile whereas I think 5, 6, 10 years ago that was much, much harder certainly I mean in Britain now, gravitational waves I think is identified as one of the key areas for PPARC funding so I mean that's a huge change of course it just means that everyone every single relativist is going to say oh look I'm doing important modelling for gravitational waves I don't know what it means in terms of real money but it is And from your point of view, do you think maybe already astrophyses have been sufficiently affected by, you know, detectors like LIGO or other factors that they're... I'm not sure that's true yet. I think they are beginning, you know, I think there are a lot of people that are following the events with interest. But they would still be very skeptical. I spoke to Phil Charles who's a professor from Oxford who's just moved down here last week when we had a joint seminar with them and he basically gave, well my impression of what he said was that he didn't really believe that LIGO etc were going to see anything but he thought it was an exciting possibility so they follow going on, but they probably wouldn't put their money on it. On the other hand, it might be what we would say as well. But you did think, say that you thought maybe that it was a little bit easier to persuade astrophysicists that relativity was relevant nowadays. Is that for other reasons? I think so. Well, no, I think they're perfectly accept now that you need to account for relativity in modeling things. They might be a bit reluctant to do that because they feel that we don't actually have the skills to do it. But I think

10:00 when you talk, you can certainly talk to a lot of astrophysicists about relativistic effects and they will say, yeah, that's what we should do. I don't think I've used to feel really. But maybe that's not a huge change. Maybe it's just me growing up. But it seems to me that the R-mode is because of these neutron stars is an interesting example of gravitational waves making itself, you know, making itself relevant in astrophysics. Would you agree with that? Oh, I think, maybe yes, I think it's interesting how that sort of change in, a dramatic change if you like, by fluke, in how relativists viewed an effect that had basically been known for 20 years or more. how that suddenly made it into the minds of astrophysicists as well and some people started, you know, astrophysicists started thinking about this with their style, which is often sort of more hand-waving and, you know, say, oh, if the number is 46, then maybe... Oh, somebody here. Hello. It's not an epidemic here. No, it's not an epidemic at all. I think so yes I do think that did make a difference, it sort of brought something that several relativists or many relativists were perfectly aware of into the minds of people that hadn't really taken it seriously before. It's a good thing too because a lot of the difficulty with noise and generalizing lots of the problems that people are stuck on, just on the classical physics, I mean the non-raltimistic read are about the classical theory of modes. So, there's so much work to be done, but it sort of can be contributed to by people that don't know a lot of relativity. I mean, there's of course lots of relativity things as well to do with other ones. In fact, I would argue that a lot of the work that we would need to understand to make serious progress on those issues is work that has already been done by astrophysicists. In

12:30 stellar pulsation and things like that in Newtonian theory so I mean or as well in the sort of geophysics for the earth there's a lot of stuff that we could learn where we could benefit from having you know, picking these people's brains so in that sense I think it is interesting because it's across border topic where you need you need both sides desperately and I think for a long time these instability gravitational wave instabilities in neutron stars was just it was something that was driven by a few people you know the enthusiasts if you like and astrophysicists thought well you know maybe but it's not going to be that important so that I think is suddenly the numbers came out completely the other way, and it looked as if it was going to be very strong. And then they started thinking about it. So, actually, it's a topic I'm interested enough in to ask a little bit about the history. So you were saying that this is something that had been known for 20 years. You mean, actually, that the gravitational wave instability in the R modes had been known? No, not in the R modes, but in other modes. So the only, when the only, the main difference between the, if you go back to Chandra's paper on instability, which is like 1970, the only difference really between, and his work and then, you know, the formal foundation for how one should distinguish these instabilities, etc. by Friedman and Schutz. if you go from those which are like early 70s up until the R-mode thing the only difference between the early studies and the R-modes is that in the early papers they focused on modes that have to become unstable as you spin the star up and then it just conspires in such a way that this doesn't happen until the star actually spins very fast so the instability is never present until the star spins very fast whereas if you take the R-mode case goes unstable immediately even if the star spins very very slowly you know in an ideal perfect

15:00 fluid situation so that's the sort of race if you like where one guy first has to become unstable and then become important as well or strong enough to be important whereas in the other case it can just start growing towards importance immediately as you spin the star up and i think the interesting thing i mean there are very interesting things in in the the early days around late 70s where ian and i have reflected on before that there's these two two main guys that did the first armada papers of papa loiso and pringle who did a nice little piece of work on that and the same year published a paper on gravitational wave instabilities looking at how gravitational waves could, the idea that gravitational waves from an unstable mode could balance accretion in a system so that it sort of would hover at a point that they did gravitationally. And, I mean, they must have been, won't get on your mini disc, but they must have been that close because they had the R modes there, they knew about the instabilities they knew the character of the modes and they could have seen immediately that these modes would fall into this category so it's tantalizingly close and if they had done that of course I mean if the field would have probably been completely different so I think I know a lot of people kicked themselves when we did discover it I mean because our modes are in the Newtonian stellar pulsation textbooks and stuff like that. So it's not really, it's not a secret as such. And the characterization of the instability is very simple. So in that sense, it's quite remarkable. So it's interesting because part of what you were saying, as I gathered, was that in this case, it's an example of people coming from different fields and they have different bodies of knowledge. modes and then you have say relativity people maybe who know about gravitational latency and so on and work going back to chatter

17:30 since you know they don't know about each other's body at work it's difficult for them to put things together but also interestingly you could actually have a situation in which you have the same people know both sides still don't put it together yes that's right any kind of a reason that you could put your finger on why it was discovered at a certain point? Was there just a chance? The R-modes and such, that was just pure I think it was just a fluke maybe I mean so saying there was sheer luck I suppose I'm putting myself in a bad light but I think so I think there is one element one reason why why I looked at that kind of problem I mean the calculation that led to sort of saying that arm roads are unstable it wasn't a sort of circuitous route if you like because it was a messed up calculation trying to do something else that then sort of showed this and then I had to start asking well what is going on I think it had been traditionally in relativity. If you go back in stellar pulsation to the earlier papers in relativity, which is, I mean, Kip's group in the late 60s, they decided already at that point that if you want to get gravitational waves out, you have to have large fluid motions, like large density variations. because that's what the quadruple formula tells you. And so they focused on kind of spheroidal, quasi-radial oscillations and the modes that follow from that. And that kind of set a trend, I think, in relativistic stellar pulsation all the way up to the early 90s. So, for example, in the Friedman-Schutz papers about instabilities, there is a clear statement saying that, you know, we should never expect modes other than these to be important because of, you know, radiation reaction and the quadruple formula, etc.

20:00 The strongest modes are going to be the ones that have the largest density variations. Which, of course, if you take the R mode into account, is completely wrong. So then it was mainly Chandra and Valeria Ferrari who decided to reformulate after Chandra's work on black holes they decided to reformulate stellar perturbation as a scattering problem for gravitational waves just try to look at it as a gravitational wave propagating on the background and then if you do that it doesn't make any difference is whether you're in that quasi-radial kind of perturbation or in a toroidal kind of shearing motion. Because for gravitational waves, you know, if you have gravitational waves, say, an H plus H cross combination that hits a star, they don't, gravitational waves don't really care. And essentially, gravitational waves of both polarizations exist, so they can drive both kinds of motions. And so if you view the problem from that end, then you can look at say gravitational wave modes that are trapped by the space time of a star and they can have both flavours as well and we had done a little bit of work trying to understand how the curvature of a star affects gravitational waves it sort of traps gravitational waves so I think for me it was kind of natural to think about both kinds of perturbations, both sort of spheroidal and toroidal motions. And so it just happens that in relativity it's easier to look at the toroidal perturbations, I mean the odd parity or whatever you want to call them, because the equations are easier. So I started off, when I started off doing rotating systems, I started off doing that, because no one had done it, and I thought, well, you know, it's a good starting point, because it's simpler, so why did I start off doing that, down and then and so that's where that result came when it came but so in a sense it's a fluke but because I didn't really think about what I would see but at the same time there was a sort of reason for doing it, it wasn't just

22:30 just coincidence And what was it you were trying to look at when you said you were sort of looking for something else? Well, the thing that we were interested in at that time, there were two things, essentially. One was the potential detectability of the various modes that belong essentially to the gravitational wave sector respond to a lot of fluid motion but our gravitational waves are trapped in a star for a very short time and then leak out. So that was one so you need and then when you have a rotating system these the two different kinds of modes coupled together and I was interested in seeing and how that coupling worked out. If that could be beneficial to detection or if it would lead to a signature as you spun the star up, that would be distinguishable. So questions like that subsequently dropped and I haven't really gone back to it, but I suppose one day they're still interesting. And were you at St. Louis at that point? Yeah, I was in St. Louis at that time. Which is another interesting reflection given that I hear rumours that that group are now working on R-Mods, which is quite ironic given that they had no interest in them at all while I was there. so in your thesis work it's also a neutron star that's right it's on something that more than one sense is opposite to our modes because I'm looking at pre-procession so that's something that's possible because there's a solid phase I've sometimes speculated that say 1% of a star by moment of inertia is solid and 99% is fluid, then that's about the ratio of the number of papers you get on free processions to papers on fluid loads.

25:00 It was taken up at first in the Sillerman papers in 1978 or something, and I suppose since then there may have been just really a handful of papers, gravitational waves from And so it's, in terms of people working on it, it's a real minority source. It's the sort of source that's always mentioned in some review of gravitational wave sources. But when you actually look at how many people have worked on it, it's very small, very small. So I'm doing a PhD that will be from Cardiff under Bernard Schertz, in which I'm revisiting free procession. I'm trying to do it in a more realistic way than people have generally looked at it taking into account for instance that the star isn't perfectly rigid it's an elastic shell with a lot of fluid inside it and try to get a better picture of how of what the important physical factors are try to identify astrophysical mechanisms that might keep free procession alive because if you just take a neutron star and you make it There'll be gravitational radiation reaction and there'll be internal damping as well. So you've got to try and keep these things alive in some way. So that's what I'm looking at. So another way which is different from, say, the arm melting at least initially, is the gravitational wave going to dampen? Oh yeah, it's not gravitational wave unstable. I thought for a while it was actually. another story in fact in my when i came to look at how the radiation reaction worked i followed the method of an old paper by patotti and anil and in fact that paper was published in 73 the year i was born so i looked at that paper and it was a very very difficult paper to follow but i could roughly see what the first line was so i went through it and i repeated it and I got this peculiar result out that under some circumstances the thing was gravitational wave unstable so I incorporated that into my thesis and then Kurt Cutler sort of looked at it and he didn't believe it and went away and we had an argument we ended up finding an error in this paper and this paper was claiming a remarkable result

27:30 it's hardly been referenced since it was written, hardly anyone seems to be aware of it, even though it had a seemingly unusual and unexpected results in it and it turns out that these things aren't gravitationally wave unstable and if you put some extra terms in that we're missing that gravitational radiation reaction does doubt the wobble but presumably for 27 years there's been a paper sitting around of what is more or less a gravitational wave instability or claiming such an instability but it never claims it in words and it's very difficult to follow the notation it's a completely un-understandable paper really and it was only when I saw that it was saying something interesting and got excited and other people I didn't believe that and started following it through and we found that there was an error in that so you were the first one to notice what it was actually saying that, yeah, I mean, as I said, I could only really understand the first line of their calculation, and so I repeated it. And even when I got the result, it took me months to realise that there was a way in which, there was a circumstance in which it would... Hadn't you, actually, when you first came to Cardiff, didn't you actually look into the mechanism, mechanisms of these kinds of instabilities a little bit? Isn't that one of the first things you did. So what I'm trying to get at is you were mentally prepared for what those kinds of instabilities were about, whereas I think a lot of people wouldn't be. And if you're not, then it's easy to actually come up with something. I certainly, I mean, Bernard, at the very beginning, had explained the CFS, and this isn't a CFS instability, although there may be... I mean, I've not tried to use the CFS that at some deep level you might be able to do that but I did know that there were such things as instabilities but because it wasn't the language that people normally use for free procession it took me a long time in my own mind to realise that there could be such a thing as an instability tomorrow 24 hours too early but apart from that it's alright

30:00 So it doesn't have to build up the anticipation kind of deal. So even when I repeated their calculation and got the same result, it still took months for me to realize that if a particular parameter was negative rather than positive, the thing was unstable, because I wasn't expecting it either. But then I did realize it. do you think i mean i know now that that result sort of is gone a bit but how much do you think from that that point when you when you drew that conclusion how much do you think that was coincidence and how much do you i mean luck in just seeing oh what about what about this because that's that's exactly the way that i feel about the almost that i knew that there weren't something like R-modes. In fact, I have an old transparency from a talk I gave once, where I listed the various possibilities, you know, if you add more physics to a neutron star model there will be a new class of modes, basically, with every piece of physics you add. And I had R-modes down as one. Which I was amused to find later on. But, I mean, it's even my point. I wasn't expecting to find that. So when I did find it, and it was at a late stage when I thought I'd found all the interesting things, I did the calculation one night, and I was convinced that it was wrong. And I'd had a glass of wine or something, so I thought, well, I'll try and redo it. Maybe I've got tips in Mr. Sign or something. And then I looked at it the next day, and it was just the same. So then I thought, well, maybe there is something going on. It was very peculiar. Yeah. And so then you were saying that what happened subsequently was that having found what seemed to be an instability based on what the production had done, that Kurt Cutler... my external because he was immediately suspicious of the result so we started having phone conversations about by the Lord it was right and he gave arguments and I it's my best to refute them and I sailed for a several month period

32:30 but eventually after a number of conversations converged on a point of view where he basically spotted a class of terms that had been missing There is a radiation reaction problem, because when you do the radiation reaction problem, one way of doing it is to take a radiation reaction potential and use that to get a radiation reaction force. And you can use that to get a radiation reaction torque. So as this body's wobbling around, processing around, you can let this torque act on it and see what happens. And that's fine if the body's rigid. and if the body isn't rigid, what the force also does it doesn't only exert a torque but it changes its shape slightly and if you put the extra change in shape in you get a cancellation between the largest part of the torque and a term to do with the change in shape and it leaves a much smaller answer that's stable always, never unstable I think that's quite remarkable in a way if you call, you know just call on the question saying well does gravitational radiation reaction produce a change in shape of a rotating a wobbling object a rigid object I mean would you a solid object in such a way as to cancel other terms it just sounds incredible doesn't it it's a leading order the two effects like the torque and the changing shape cancel Oh, I, can we, uh, maybe you two can continue. Yeah, we'll do that. All right, we can go to the biophys. Okay. I can talk to you more later on. Yeah. That's all right. Time again. The, uh, so who was the author, along with Bertotti, did the original author, or authors ever get involved in the... No, we're going to post our paper soon on the pre-print server, so I don't know if I'll comment then, but making sure that we're writing it sufficiently politely so we start to upset them. So it's Batotti and Anile, so that's A-N-I-L-E. So that's from the year I was born, that paper. It's an interesting story of course you say it being such an old paper and then you being the first

35:00 Kirk claims to have used some sort of electronic database and found that it's been referenced four times I must admit I've checked on ADS and I can't find I haven't it certainly hasn't been referenced more than that, it's a very small number of references So how did you come upon the paper? I would say the reason that it had gone unnoticed and unreferenced is that it was not easy to follow and it wasn't easy to make contact with the notation that was used with the sorts of quantities that people usually think about But in geophysics or in relativity, it was very difficult to know what the equations meant. And there was never a sort of a verbal statement either. That's probably a bit harsh, but it was a difficult paper, though. Yeah, sure. I made a paper. so I'm kind of curious actually because I guess we've been discussing this whole process of discovering effects which is something I'm kind of interested in I don't know if it's sociology maybe it's psychology but I was wondering about the kind of psychological process of that kind of discovery I mean was it exciting when you it was yeah It was exciting. First of all, it was exciting because I had these formulae for months, and I thought I knew about them. But I'd never considered the case where basically the body was prolate as opposed to oplate, because I thought things would just come out in the same way. but because there were the terms missing the leading order term did change sign in such a way that what was unstable what had been a stable radiation reaction effect what I've been thinking of as being stable could in a prolate case be unstable and that was very exciting because it immediately made these things more interesting sources and I started thinking in my head how can I put this together to give a nice picture in a paper and I started thinking about the low mass x-ray binaries that loads of people

37:30 and gravitational resources are writing papers about these things and thinking can I say something interesting about them and I thought I could. So I wrapped together a lovely final chapter in which I'd taken this exciting effect and I'd try to put it in a context and it all seemed to be going very nicely. It was very exciting when I first became aware of it. yeah so um and so was there a corresponding sense of disappointment absolutely yeah this one was a lot worse than the yeah seems to have fought further than you climbed in the first place and of course it's a it's funny i guess funny as it says of course at the same time you're actually still learning a lot but it's sort of disappointing when something that you were hoping to find I mean, it's still, there's still interesting things there, but not the original things that I'd hoped for. So, from the point of view of gravitational wave detection, what do you think is the upshot? I think the upshot is that at least for neutron stars that carry oblate, that means fat, oblate deformations, I don't think they're going to be very good sources, but I think they're going to be rather poor sources. And it may even be that with an advanced LIGO detector, I know that they keep changing the names of what the final, but even with some of the best noise curves that I've seen in the literature, we may not see 3D processing stars unless they're much closer than anyone thinks I don't think they're going to be very good sources at least in any of the standard scenarios as I said they've been mentioned so often in reviews without anyone ever having really sucked their teeth into them and saying just how strong are they going to be and so speculation that was sort of long overdue. I mean, it could have been done. All the ideas in my thesis will be pretty old ones, really. It could have been done decades. Sure, but as you say, no one has really... No one's...

40:00 Do you think that's because people sort of sense that it mightn't be a great source, or just because it was a challenging problem to deal with it all at once? That's a good question. I suppose it's probably because they sensed deep down it may not be a great source but even so you'd have thought that with so many papers being written on binary and spiral with so many other sources being looked at in such detail there may have just been a few more papers on procession and it could have sort of cleared up some important issues in our area I suppose certain areas become fashionable and certain areas aren't and this wasn't it's interesting you should say that it was completely left away there were probably I don't know certainly less than a paper a year much less than that in fact and in terms of estimating the field strengths I think there are probably only five or six papers in total And that's since 1978, when they were first talked on. So you're talking on paper every four or five years on average, something like that. Yeah, that's not very good. But it is one of those things, of course, that a lot of the time, it does happen that a potential source of gravitational waves doesn't really pan out once it gets looked at. My wife, who's an astronomer, used to joke when I was a graduate student and looking through things like that, that we ought to keep track of all of the things that we found that wouldn't be sourced for a long time. Where I think it works is that you're as optimistic as your ignorance allows you to be, and as you learn more, you realize that you're being over-optimistic, and you sort of put a rein your opinions in a bit. I don't know if anyone's done this and if anyone had maybe it would be someone like you but just to plot a graph the sort of the consensus as to how strong a source is going to be as a function of year starting in about 1970 and just see as it gets closer and closer the good detectors being built or better detectors people on the basis theoretical ideas becoming less and less optimistic because I'm sure that people are sensing that they need to be played a bit safer, especially now that they're potentially just a few years away from these things working.

42:30 Yeah, I haven't actually drawn such a graph, although it might be nice to do, but I have been thinking along the similar lines, and it does seem as if that's one sociological factor that operates, as you say, that people get more cautious, actually, as the time comes closer, when they may be proven to be wrong. so here at Southampton will you continue working on similar topics as soon as this thesis is wrapped up I'll spend most of my time on R modes there are so many issues in common between free procession and fluid modes in fact there may even be ways in which one sort of motion could drive the other but I hope I'll obtain an interest in both but it'll be mainly R modes That would be what, the North was the discoverer of the instability, so if we don't study on my chair, then it would be silly. Still most of you are working with the others. That's right. I sometimes talk problems still with other members of the group, but he's the only one that I'm really working with. The others are probably in some more mathematical relativity. And does that type of approach to relativity interest you at all, or do you want more inclined to stick with it? I'm inclined to stick with it. I think that theory has run a very long way ahead of experiment. And I'm anxious to do work that will hopefully bridge that gap, not extend it. So, from your point of view, LIGO is Geo, Virgo, Gravitational Wave Detection, broadly speaking, is an exciting area to work with. Absolutely, yeah. I mean, I applied for my PhD, didn't know much about it, and I even applied some places to do particle physics, but it was fairly random that I ended up doing gravitational waves. but I feel as if I've fallen on my feet slightly, it's clearly an interesting time to be in this field.

45:00 Well, I hope that someone wasn't saying it was the clinical thing in 1970, but I'm sure it is a good place to be. Yeah, I think that seems, it does seem to be a very vibrant field in the last few years. So speaking of source strength and so on, I guess when dealing in the case of neutron stars, you have to have that there are lots of them known, and therefore there are pretty good astrophysical estimates of how many are out there that can be expected. What is that? There are reasonable estimates. I mean, presumably better, for instance, than estimates of the number of neutron star-binders. Oh, yes. Much better, yes. But is there still a lot of room for... There's room for manoeuvre. I would say one of the biggest uncertainties is their initial spin period, rather than the total number of stars, how fast they're born spinning, and of course how large their departure is for rexysymmetry when they're born as they age. I suppose the actual number should be reasonably well constrained, because if they are formed in supernovae explosions, we know pretty well how often they happen. So we should be able to track the total population. But the actual number of fast-spinning ones close to us, that isn't quite so well known, but even that's much better known, I think, than the one in the spiral. And are the estimates of the number of that kind of object years, to get back to this idea of how things vary with time, constant estimate or does that could change around a lot of do you know I don't really know I don't know about that there is a I talked about people being optimistic

47:30 but you know the book 300 Years of Gravitation there's a nice little piece in Kip's art chapter, an idea I think it's due to Blanford, what would happen this is the most optimistic scenario if there's a population of neutron stars that don't emit anything other than gravitational waves and just spin down and you can do some statistics then and work out how close the nearest fast spinning star is so, I mean I can't remember quite what answers they get out but that's the most optimistic thing you can do and that hasn't been completely refuted you know, that sort of idea because by definition you can't really until you've got the graphics when people are still interested in objects like the crab, you know, the all well-known well-studied objects people talk about those people like Bernard will produce talks where they'll plot the crab as a point on a signal frequency curve, as an upper, they'll put the upper band on its signal strength from saying that all of its spin-down energy is coming out as gravitational waves so people are still looking at the old familiar objects in a very optimistic way and what do you think? that their optimism is probably not well-grounded at this point, or is it? I don't think. I think they're right to plot the curves, because you never quite know. But I suspect that the gravitational wave amplitude may be orders of magnitude less. And given that these points are just above the sensitivity curves of the detectors, it could be that even these objects are undetectable the first generation, or even advanced. I can't think of any source for the ground-based instruments, but I could say, you know, we really will see that. It's still possible in my mind that they may see nothing, maybe. Have you published your thesis? No, I'm currently reworking yet, so that will happen in a few months. I was kind of curious because I think talking to at least one guy, I thought of a vague impression that maybe astrophysicists who were interested in neutron stars but who weren't relative to this might have a tendency to do at least back of the envelope calculations about whether strong gravitational waves would be expected coming from this kind of source.

50:00 And, for instance, this guy was saying, well, you know, I think that the order is going too, too low and so on. So I was wondering if, if you had any, if you, you know, A, talk to people with just an astrophysics background much who weren't into relativity, whether they were sort of interested in what the relativists had to say on this kind of score, and also whether you think it might be true that probably they, you know, may have done calculations at least simple calculations, but sort of not publish them because in their field it's not that interesting. I must admit I've not met a huge number of people in this category but those that I have it's always slightly disappointing at how entrenched everyone is in their own field. People are often at best just aware that there are some new detect gravitational wave detectors being built but the only people I've met who've made back of the end estimates are people who are self-confessed relativists They're either in the field or it seems like they're out. I've not met many people, many astrophysicists who've had their own little hand at having a go. I mean, no, I don't think I have. Right. They don't even seem to be too aware of it. I think people are very much involved in their own little seed, and I'm sure it's true of us as well. I suppose it's a bit like the same phenomenon, Niels was referring to in the case with RMOs, where you can have two different communities that know different things, but since they don't know what each other knows, I think people judge their own success by how much they sort of impress their immediate peers and their immediate colleagues. and that so you don't quite care as much if you were to get a paper out in something that none of your own friends would read you know there's always that pressure that's doing that's immediately of interest yeah to your own circle and as a matter of fact I suspect that there's something that I'm familiar with being active in two fields where the practitioners really don't talk to each other there's even I think probably the job effect there in that, you know, if you're writing papers in more than one field, you know, supposing we'll say for example that you're split 50-50 between the two fields, well that means that, you know, one bunch will know half of what you've done and the other bunch will also know half of what you've done, but there'll be nobody who knows actually everything you've done.

52:30 So I guess probably there are real practical considerations as regards whether it's advisable or even if it were practical to get yourself noticed by two different groups. So how long will you be here? Another two and a half years. I had a three year contract. This is my first year. And when did you start it at Cardinal? 95? No. Yeah, I started October 95. So I spent three years there. I've been here one and a half years. But I've been working on my PhD for most of the first year here. I came here at short notice because they had some money to employ someone for one year. and most people wouldn't, you know, the postdocs from overseas and so on were slightly reluctant to come for a year, but it was so easy for me and I needed else already, so I came, but now I've got this three year contract here. So, just out of curiosity, you were actually based at Cardiff to write your PhD, but I were there Bernie was most of the time. Well, that's absolutely right. I didn't see an awful part of him. So did you mostly work away on your own? On the whole, yes. I would hopefully see him and he came back and we would have a meeting for an hour or two. So when I did meeting, we would often have quite substantial sessions. And once or twice I visited Potsdam as well for a week or so. I spent much of the time working on my own. There were people I had to talk to. I think in terms of guiding their research, it was a combination of my own views and bernards. I wasn't really supervised by anyone else.

55:00 Well, interesting. It's impressive to be able to walk away. Well, I'm afraid it's taken a long time, so I don't know. It didn't work perfectly. Well, that's probably to be expected. It takes a little longer, I guess. When you have to push it sometimes just by yourself. Because it takes a while, I think. I think it's especially in the early stages that you really benefit from having an advisor around a lot more of it. where you're still flying in your feet. Anyway, thanks very much. It's a pleasure. Very interesting. So, I'll just say quickly that it's 23rd of February at 2.30 in the afternoon, speaking with Professor Chris Clark at Southampton University. and so I'm just sort of interested in general in the background to work at St. Hampton on Koshi characteristic matching and stuff that relates to general relativity and how I understand right now that the group is, Talisier is busy involved in preparing for a large European network tackling. Psygala? Yeah. And of course that's somewhat of interest from the sort of sociology, sociology and science point of view and how you're getting on these large collaborations. But I'm sort of also interested in the background of how the group got interested in this kind of research. I think I would have kept notes on this if I'd known this. But anyway, you'll be able to trace back specific dates through the publications. Yeah. I suppose it was a gradual thing because we were interested in algebraic computation as a result of Radian Lerno's work. We've been in that area for quite a long time, certainly for longer than I've been here, so over the last 20 years or so probably. So having that algebraic background, I think a strategic decision was made to go into numerical relativity because it looked to be the way in which the subject would be developing and it would seem natural as a way of building on our expertise in algebraic computation.

57:30 Not quite sure, you know, precisely when that is, but as I say, you can trace that from the publication dates. And I suspect that at that stage, certainly Raiden Verneux was in contact with a number of groups in the USA, but I think in particular Nigel Bishop came here and visited, so it was really a collaboration between Nigel Bishop, Raiden Verneux and myself which really focused work on that particular method. I'm particularly interested in that method rather than other methods. I think there was a sort of a number of people trying different areas. I've been collaborating with people in the Max Planck Institute for Astrophysique in Garkey for quite a long time and they're interested in various sorts of approaches including sort of characteristic bits, hyperbolic approaches and similar things. So I think there was a sort of general feeling in the air that this was one of the various options being looked at. So I think it was a sort of part, it was largely a strategic decision based on where we thought the subject would be going, where our interests were at the time, what connections we had, who was actually there at the time, Well, just as a curiosity, because, as I mentioned, one of the things I'm interested in is gravitational waves. And my impression here is that the subject of gravitational waves has become of increasing prominence in recent times, probably because of the fact that there are these big content of boundaries. Yeah, it's what you put at the introduction to all your graph proposals. That's right. We were just talking about that. The sort of standard paragraph that everyone copies from one to the other, those time dramatic observations will be coming from. We've been producing this paragraph for the last eight years, of course. So, was that a factor, well, say, at least for the funding? Yeah. Or any other reasons as well? Yes, I think so. I mean, it's not

1:00:00 I suppose that's been a factor for gravitational waves. My perception of the sort of overall change in the subject over the last 20 years, I suppose, has been that for a long time it was an extremely ivory tower sort of subject and made very little contact with conventional observations or the sort of procedures which are regarded as conventional science. And so it was, to some extent, marginalised. And the feeling that I had and still have, despite some evidence of the contrary, was that this situation is changing through actually two things. One is experimental relativity, including gravitational wave detection, satellite observations and so on, and the other we suspected then and I suspect now is using numerical relativity as a sort of equivalent of observations so that you can do a simulation rather than an observation and if one got decent numerical relativity test beds set up then that would have a similar role in gravitation well in general relativity to observational work and experimental work in other sciences I mean, the whole thing has proved a lot more difficult and slower than we had dreamed it would have done ten years ago. But I still think, in principle, that's a legitimate way of looking at things. So I think we were backing numerical relativity, not so much because of the link with gravitational radiation, but because we hoped that numerical relativity would be playing a major part in getting general relativity more back into the mainstream of science. Right. And you still think that that's likely the case, but at the same time, of course... At the same time, we've got the observational work in gravitational radiation. yeah which um so in fact there's a sort of quite a nice um symbiosis between the numerical work and the experimental work in that you know the two need each other for corroboration

1:02:30 so yeah i think um it still looks that wasn't to me as though we were right to go down that path We knew it would be a difficult path to go down because it's an area where by and large the cache only goes to the big groups, so we realised we were in heavy competition at the time. Right. Do you see a similar split to follow the theoretical experimental analogy between the people doing relativity and then the numerical people who are playing the role of the experimenters? Is it that there are two different groups of people, by and large, who offer the work in different ways? Yeah, there are a number of splits, actually. I think there is that split, certainly, because general relativity has been on the pure end of theoretical physics as well. So in some split that you get in high energy physics between the theoretical work and the experimental work. But in the case of relativity, the theoretical end looks much more like pure mathematics than is the case with high energy physics. And the split is sort of confused because of differences between European and United States traditions as well. You have a European traditional which is very linked with pure mathematics and so that pretty much polarizes the way things look in Europe whereas in the United States you don't have that sort of rather specific or mathematical tie-up so it looks a bit more like the high energy physics split in the states right and so that the relativity theories in the states are or something more like a general body of yeah they're they're theoreticians but um they're sort of pragmatic theoreticians whereas i got this book last week um marcus creeler's space-time foundations of general relativity and differential geometry, and you open it, and you get lemma 2.1.1. Let NB assess and Curley use subscript to alpha and so on, which is, it says foundations of general relativity,

1:05:00 but it's essentially a pure maths textbook. Right. Sure, yes, very different from that in our school. And in the case of the European tradition, Do you think that that arises because the relativists at some point were actually coming out of the mathematics tradition as opposed to the physics tradition? Yeah, yeah, which is, you know, why this group is in a building level of mathematics, and same for all the groups, no, that's not true, actually, same for most of the groups in the UK, anyway. And do you see a cultural clash then between the relativists of, well, let's say, for a start, I was wondering if there is kind of a cultural clash then between the relativists with a more mathematical background and a less mathematical background? I think there can be, yeah. I mean, the person I just fragmentally quoted is someone who takes a very rigorous approach to the pursuit of theoretical general relativity, and he's certainly been involved in quite sharp professional controversies with relativists who take a more pragmatic and laid-back approach and the sort of thing which is more common in high energy physics. So I think there's a genuine tension there as to what's the sort of fruitful way of progressing a theoretical relativity. I guess I've gotten some flavor of that tension, as you say, looking at the history of gravitational waves, where you, I think, did have some kind of tension along those lines, for instance, in the quadruple form of the countries. Yes, yes. Yes, there's quite interesting personalities involved like that, people who happen to have state their personal reputation on particular ways of looking at it and defending it very passionately. Sure, yeah, obviously the personal factor comes into it. The numerical relativists then, just to continue on that sort of scheme, do you see them as coming from the more mathematical tradition or the more the last mathematical tradition or

1:07:30 I think sort of a different I think by and large from the less mathematical tradition that seems to be the way it's worked out yeah I'll become certainly certainly that's yeah I think that's by and large true though for a number of different reasons I I think in the UK and parts of Europe, or at least scientists who have been brought up in the European tradition, that's happened because of the sort of traditional animosity between pure mathematics and computation. Computation is associated with doing things approximately and pure mathematics doing it accurately and precisely. I mean, this is a total travesty. And so you tend to have a situation where the people with a pure mathematical inclination or training go towards the theoretical side of relativity, and those with more of a physics background, or indeed a computer science background, go towards numerical relativity. There are some, I think it doesn't hold quite so much in the case of the states. You get people there like Beverly Berger, who is pretty sharp mathematically and very committed to numeric relativity. So I think it's a reflection of the culture from which people come. I think the States is in many ways a more unified One of the impressions that I had was that part of the split involved a division of labor to some extent between the mathematical relativists who would, as it were, be the ones who would provide algorithms which numerical people would then sort of turn into actual code. I'm wondering if that is true in your experience. Yeah, yes, that's certainly the case. Yes, certainly, not uniformly true.

1:10:00 I mean, I suppose Jeff Winneker is a case in point. I don't know how much actual coding Jeff does, actually. I might as well say, not too much, maybe. But he's certainly heavily involved in the details of the numerical relativity, even if he doesn't actually spend all that much time sitting down writing code. but in his case he's certainly been at the leading edge of developing the algorithms and the concepts of asymptotic analysis and also involved in doing the numerical analysis of it and there's a sort of it looks as though there are many cases where the two activities have fed off each other in the sense that he's been developing algorithms stimulated by the development of code. It's been a quite tightly intermeshed operation as far as he's concerned. And I think that's also been the case with the way things have worked here. I mean, James has firmly set his face against writing any code, but the rest of us have been involved in sort of both sides of the operation. I was going to ask that. So the model that Southampton has decided to follow was that of being involved both in the algorithmic and the coding. Yeah, yeah, yeah. And in practice, it's noticeable in the case of appointing research students and research assistants that in some cases we've argued that there might be a niche for someone to come in just doing numerical analysis, generating code, simply picking up other people's algorithms and implementing those. um that's i don't know that i know of any case where that has been done successfully you know

1:12:30 in a pure way i think you know for any for it to work successfully there needs to be some in some active involvement of the person who is doing the numerical analysis in the development of the algorithms um otherwise there's just not sufficient communication there for it to work and I think in the past we may have made mistakes by thinking that you could separate them considering people even if not actually engaging them who were purely numerical analysts who just had an interest in coding up other people's algorithms so it's too difficult for that, basically. You hit points where you're saying, well, this algorithm is just not working, there's no, it just doesn't code, you can't do it that way, and so you've got to go, there's got to be some backwards and forwards. Right. So it doesn't, as a model, the idea of splitting the algorithm we developed or the code, doesn't seem to work or not. I don't think it works, yeah. Yeah. And because it's just not clear how to split the work off? Because the algorithm needs to be designed in order to be efficiently codable. So for instance, you can cast the equations in many different forms which will imply different procedures numerically, these are going to have varying degrees of efficiency. You're trying to get the maximum possible efficiency in actually generating the code, so you want the algorithm to cast the equations in a form which you can code up efficiently. It could, I mean, further down the line, you can imagine that the case, it's probably more or less the case now in some centres, but you can imagine a situation where there was so much expertise in terms of ranges of tools and packages which were available that any old equation which

1:15:00 somebody threw up could immediately rather automatically be coded up and run without any trouble at the moment that's not the case because one's just sort of struggling to get the thing to work and so there's got to be this sort of symbiosis between the two I suppose some people would argue it's it's always it's always been like that and there's there's a bit of controversy over these big fluid dynamics packages that which are set up as general purpose packages for general problems and there's a and they're used by design engineers with limited understanding of the structure of the equations underlying them there's a lot of controversy over that and is appropriate and should one in fact be looking in more detail about the nature of the equations and tailor-making the packages to fit the equations otherwise you get garbage coming out of it so in other words you could make the argument that once you have had a much you know much greater experience solving problems in a given field that in principle then you are then you could do the that in practice again some people argue that you know I think yeah but it's still problematic I think once a field is well staked out so that you know what sort of problems are doable and what sort of problems are doable once it sort of turns into a routine engineering as it were then you can have a complete separation of two areas but I mean in the case say of fluid dynamics if you're looking at flow inside corners or something like that or flow being injected through slots and things like this where you get regions of high velocity gradient, then it's well known that that's a difficult problem where if you just throw a standard numerical algorithm package at it, it's liable to run into difficulties and so you really need to look very carefully the sort of algorithms that are going on and just how you code up those algorithms in these well. We have lots of debates on this between numerical relativity and the fluid dynamics colleagues. They would claim that their area is pretty much more highly developed

1:17:30 than general relativity in this respect and that if we used their ways of looking at it much more in more ready-made packages and just mop up the existing algorithms with ready-made packages. To which we reply, well, no, it's not like that, because we have things like coordinate freedom, which you don't have. Right. My impression was that there had been some efforts to try to do that, to take advantage of the experience gained in other fields and the practice of other fields by recasting the outside equations in laparoscopic forms, yeah, which was entirely reasonable. I get the impression that it's not quite so wonderful as it was first thought to be, but it looked So, in fact, in practice, it again may well have turned out that it's not so even trivial a manner to import the experience of other things. But it's interesting, because I hadn't gotten that before, you were saying that the fluid dynamics people were actually sort of looking at numerical relativity and saying, well, you really ought to try to take... I mean, this is just a coffee room conversation, but I think they take place in lots of different Well, the coffee room conversations are often the most interesting things, maybe, from the historical point. So, just as opposed to sort of conclude, at what kind of stage do you think the numerical relativity field is at the moment? In some sense, how far away is it from reaching that stage where, at least in some areas, you do have that? I really don't know. I've essentially been out of it for 18 months, so I really don't. No, I don't know where it is right now. Well, thanks very much. Okay, alright. That's very interesting.