Interview with Misao Sasaki & Martin Rees

0:00 So, it's 2 o'clock in the afternoon of the 8th of October, and I'm speaking with Ms. Alice Sasek, and we seem to be picking up okay. So, well, I thought I'd start by just asking about your own background and how you got started working on issues related to gravitational waves. Well, when I graduated from Kyoto University, I joined this astrophysics theory group in Kyoto University, headed by Chushiro Hayashi, who is a famous astronomer. Perhaps you have heard of this Hayashi track, I mean, on the H.R. diagram. Right. Oh, okay. This is Hayashi. And, well, at that time it was, I think, 76, 75, 1975, I think, yes. And people were quite excited about, I mean, maybe it was a bit sort of a late thing, but, you know, at least in Japan, students were quite excited about the poaching radiation things and all these things. And also, I think that's about the time when the SMAR started working on the numerical relativity things and so on. So, well, with the, you know, Takashi Nakamura, and K.H. Maeda, he's here, too. And another one, Shouke Miyama, and myself, four students at that time, sort of got together and talked about what to do in relativity. I mean, you know, there seems to be a lot of things to be done. And the one thing was, you know, this remember things, I mean, you know, if we could do better simulations of coalescent, I mean, the black hole formation or, you know, collapsing star simulations and detect the gravitational waves from those data, then it might be very, very interesting.

2:30 And also, it could give us some hint of, I'd say, this about whether there was, I'd say, Spenlow's conjecture, I forgot the name right now. Right. Yes, no, but covering up the next similarity is really true or not. So we started to sort of study first. It took us about a few years before we really had the first simulation of axi-symmetric star clubs. But during that time, I was very much interested in the gravitational wave itself because of, well, as you know, the general relativistic, I would say, the equivalent principle, you know, gravitational wave is some kind of a ghost. If you catch it locally, then it goes away, but if you look at it globally, it exists. And so also the problem of the radiation reaction attracted me. So that's the reason why I sort of went into the gravitational wave physics, yes. And well, but the first work I did was with Professor Sato, Kumitaka Sato, and this was exercise problem of interaction between the electromagnetic wave with the gravitational wave. Well, it was fun, but, you know, a very simple calculation. But in any case, that was the first thing. Then, well, when I was in the graduate course, mainly I worked in this formulation of 3 plus 1 splitting of 3 sides in order to be used for the numerical simulations and so on. And then I did some work on how to find the apparent horizon. In fact, that was the first, the only numerical stuff I have done. Well, I mean, calculate some things, like integrate equations and so on, that I use in Google, but you know,

5:00 so finding a parallel horizon is more like simulation, right? So in that sense, that's the only thing I have done by myself. Usually I just sort of write down equations. Anyway, then when I finished the course and then I wrote the paper on radiation reaction thing and that was again the no so what I think that was the real starting point of my my my starting point of this gravitation wave then well meanwhile I you know on that That time, it was about 1980, I think, this very famous scenario, cosmological scenario of inflation came up, and luckily, Katsuhiko Sato, who is another Sato, now in the University Tokyo, who sort of, at almost the same time as Gooth, he sort of thought of this inflationary scenario. But no, unfortunately, because he is Japanese, he cannot think of the good world. He just wrote a paper, like, you know, the title is Exponentially Expanding in the universe or something, you know. Uh-huh, sorry, yes, it did have a catchy name. Yeah, but in any case, well, so, and he found there could be very interesting space-time structures, you know, in this scenario, because, well, in the old inflation scenario, you assume lots of sort of vacuum bubbles nucleated in space-time and they collide and so on. Well, if you consider, for example, island of space with the huge vacuum energy, cosmological constant, while the outside is the vacuum, real, true vacuum, then the inside bubble, it should expand exponentially, while outside is simple, you know, Minkowski space time, so you can't have that kind of expansion. The expansion is limited by the light form.

7:30 And so what we found is that this would give you sort of another universe, both of another universe, island universe. And this was very interesting. And that's another reason why I, well, that's the reason why I was also involved in this cosmological thing. And from that time, I started working in cosmology and gravitational wave things. time. And then, well, when I finished the doctor's course, then I had this poster position in the Yukawa Institute, and Takashi Nakamura was there, and we started to work on the black perturbations, yes. And, well, we sort of wanted to tackle the perturbation of the curved black hole. And we found some interesting results then. Sure. Yeah. And we did some calculations. But then, well, when the equations became been sort of more and more complicated. I sort of gradually lost my interest. And also, at that time, compared to the present day, maybe it's a little bit too early to do those things. And not so many people are interested in this subject yet. So because of those reasons, I sort of stepped aside from the gravitational wave things and went more into the cosmology. So after that, about 10 years, I've been doing mainly in cosmology, yes. And then, I think it was this meeting dedicated for the Rana Israels' 60th birthday took place when I was also about maybe 10 years ago, so I don't quite remember, in Banff,

10:00 In Canada. And, well, the tip was there and he introduced this, well, Eric Poisson's method of how to calculate, you know, the Hanyola terms and those perturbations and so on. That sounded to me very, that was attractive to me very much was, you know, well, now you do the numerical stuff, but, you know, by analytical calculations, you can get something very interesting and could be very important for those, you know, gravitational wave detectors. Sure. So when I came back, I asked Takashi, you know, so what is going on in the world? Because I was away from the gravitational wave things for 10 years. And he told me many things, and also he told me that he was working with Kitayuki that they know how to calculate very high order terms of the reaction waves from particles. And then, you know, they told me that they found this, you know, logarithmic sort of term, with binding legal, sort of fitting. And when I heard that, you know, there will be this logarithmic term, well, if it is just a simple power series expansion, then since everything can be expanded in power series, So I might not be interested in doing the real analytic calculations that time. But because I heard that there should be logarithmic terms. And if it is a real logarithmic function, then it should be derived analytically. Right? So that's why I was very much interested in doing the, you know, to get these terms. Then, while trying to sort of, while playing around with the relative real equation, I found a way to calculate. Right. And then, you know, Hidayuki is very good at, you know, all this analysis and everything,

12:30 so we get together and calculate to, I forgot the third post-Newtonian, fourth post-Newtonian A lot of work. Once you know how to do it, it became quite easy. It was very fun. At the same time, I don't quite remember, but LIGO was really under construction and also TAMA is approved. Everything became very sort of how do you say the future is bright so you know that's why I thought that I should do much more in the gravitational wave things now because of the detectors and one thing I left it about 10 or 15 years ago was this radiation radiation reaction thing. So I talked to a student of mine, Mino. I mean, he was in research course about this old but very difficult problem of radiation reaction in the gravitational field. He was very, very good to understand the real problems and, you know, to solve the problems and so on. And also, in the end, we found that, in fact, that this, you know, keeps, well, this is phone and heart of paper, you know, match, asymptotic matching, was really the starting point. And by using that technique, you know, we can sort of, without ambiguity, define what the radiation reaction In fact, it's, you know, you shouldn't call it the radiation reaction because even without radiation, I mean, say, for example, in the first or second post neutron order, as you know that we have this order mu over n correction in the equations motion. And this should be also included in this so-called radiation reaction post. It's only the sort of correction to the equation motion to the mu over n order.

15:00 So anyway, so everything sort of fit together very nicely. And now with another split, I'm trying to calculate the explicit radiation reaction form of this of course you know this radiation reaction force is very formal as well but it was important to drive this formal form but then the problem is how to calculate the express form now and And since many other people, including Amosori and their students or his students or some other people in Kotsdam interested in this problem and doing independently many calculations. And recently I found, I think I found another and maybe really a good way of systematically calculating these sort of new abiding correction for us, at least on the Schwarzschild background. So that's what I am right now doing. Interesting. The student you're working at the moment, is that us at this moment? Yes, yes. His name is Nakano. And I was just briefly going to ask, you mentioned, of course, going out to the third or fourth post-Antonian order in your work at Hideyuki. Yeah. Was the order in the expansion that you went to inspired by the needs of the detectors, or was it more sort of internal? I mean, how did you decide, well, this is how far we should work out to? Well, it's both, I mean, one thing was that people were talking that maybe a third post-Newton order is not really enough, or, you know, well, maybe it's not, but you should check it, well, you can check it only when you go one order beyond, right? To see how. Yes, right. How inaccurate the third order is, right? And also, at that time, when we sort of did some kind of order estimate, then, I mean, the order counting, and we found that it's relatively easy to go to the fourth order at once, yes.

17:30 And the next order that was done with the Takahiro Taraka and Hideyuki and myself is the 5.5 post-mintonian order, which took us, well, quite a lot of time. Because then you have to go one order beyond this. And having done that work, what's your opinion of, for instance, how good post-mintonian order tempos are? reasonably okay for, say, something like template-matching or...? As far as you only consider... you don't have to consider these orbits other than the circular orbits. I guess perhaps the third or fourth postmodern order is... I mean, considering the, how to say, the expected, the sensitivity of the, for example, advanced light of the directors, and, you know, I don't think there's no point going farther and farther, well, except from some academic, you know, interest to, interest, and maybe some other And do you think that, and how necessary do you think, say, fully numerical work is likely to be, or will the post-continental work probably work okay, at least with, say, advanced language? Yeah, well, it will be, I guess, necessary, but, well, as far as I see, well, the New Maple thing is very difficult, right, especially when it comes to this sort of matching of data, you know. Sure, there's a long way to go there. Yes, yes, it looks like, yes, very difficult. So what kind of, are you involved at all with data analysis from TAMI? Well, a bit, yes. I mean, in fact that the, I, well, I'm not doing by myself data analysis,

20:00 but I am a member of the data analysis group because they sort of wanted me to be in. I mean, for some budget reasons and all these things Well, this is secret Of course, you know, I do discuss things with Hideyuki and also Takahiro and all these people but I myself do not make any computer calls or anything Are most of the people involved with the data analysis theorists? Well, not really. There are a couple or maybe more students from experiments, but at the moment, yes, I mean, most important work is being done by Hideyuki and Takahiro. Takahiro Tadaka is now in Spain, so he can sort of not do much. So in that sense, now Hideyuki It is, I think, very, very busy, especially, you know, the Tama meetings within a few weeks. It must be really busy. Well, a number of things that you said I found very interesting. One thing I want to go back to briefly, just in case you have, well, just to sort of get some background. I'm interested a little bit in I suppose you could say the history of GR in Japan and you were saying that you got interested in cosmic censorship and other GR type of issues in the mid-70s and so I was wondering when people first got interested in general relativity in Japan and how many people there were already working in the subject? Well, if you just talk about the general relativity, then I think there are essentially two schools. One, which is very old and which is now sort of gone, but I was involved for some time, from Hiroshima University. and they had this very good institute for theoretical physics

22:30 founded maybe during the second world war time and headed by I forgot the founder's name anyway he was very much interested in this sort of pure gravitational theories And then maybe you've heard of Nariyai, I mean, who is a quite famous relativist in Japan, who was, I mean, he died probably, but, so he went to this research institute in Hiroshima University, and he formed a very good group of relativistic people. But, well, which is around, say, 1950s, and not so many people are interested in relativity or, you know, it's a very minor group there. I mean, they did some very interesting work. For example, around that time, Nariye found this, well, now famous solution called Nariye solution which is the solution to with some cosmological solution with cosmological constant but which is just different topology from from the space and which has some some kind of cylindrical symmetry or whatever it's very interesting solution and then in the early 60s I think they have worked on, for example, the gravitational theory with the higher order corrections, like the square in Lagrangian and so on, and considered cosmological models and the stability of those solutions and so on. So, you know, they are doing, in that sense, also a bit too early things. I mean, not so many people were interested in it. But also, there was a very interesting story which, well, this I heard from somebody else, so maybe there is some misunderstanding. But there was one professor called Kingra in this research institute. And he was interested in this sort of canonical composition

25:00 And he found what now is called the gravitational anomaly at that time, in the 60s, that you have sort of, by some kind of regularization or something, then you are left with the terms which breaks general covariance. So I hear that this was really the first discovery of the Cartesian anomaly. But, you know, since it was too early, it is not related to strings or anything, so it was sort of forgotten. Anyway, so that was the, you know, one school. And in fact, I was a member for this institute from 1980, maybe 83 or 84 to 88 or 89. It's about four or five years there. And another comes from Kyoto. In fact, this Chumichiro Hayashi was originally a particle physicist, and when he moved into the astrophysics, he also was interested in cosmology. And he did a very nice work concerning this big bang scenario. In the original scenario of Gamow, he assumed that the universe started with pure neutrons. But then, I think it was 52 or so, the Hayashi realized if you take into account the weak interactions, and so on, then that determines the ratio between neutrons and protons in the early days. And so that change is, well, quite a significant country, this light element about those calculations. And because of this, well, he was also interested in cosmology and astrophysics, but as you can see, he's interested from the nuclear physics part of But then this Fumitaka Sato came, joined his group as a student, and first Hayashi told

27:30 him to study those big bang nuclear synthesis and cosmological problems, and so he started to do some calculations, and well, maybe, right, he says, I mean, Sato says, it was in some sense unfortunate, but in another sense it's fortunate that he sort of, he was not very good at numerical calculations, so that, you know, he also did this, you know, nuclear synthesis calculation, but it didn't come out very right. I mean, the calculation, there was some numerical errors or something. And meanwhile, the famous swagger about all these people from the United States have done this, you know, famous nuclear synthesis calculation. So he had to sort of, you know, give up. But because of that, he was looking for something else. And also, you know, he thought that he would never go into numerical things. Then, so he was interested in black holes and so on. And he found this famous solution with his students, the Tomima Sato solution. And then, because of that, you know, many people who are interested in general relativity of black holes sort of gather around Umitaka Sato. Or in some sense, I was one of them. so so that's two schools yeah and then it just so happened that when you and others were getting interested in this sort of field as in addition to these other things like coffee radiation gravitational waves also were or a reasonably big subject at the time, so that attracted your interest in that direction? Was that because of, say, things like the Binary Pulsar result? Well, not really. Well, I'm not sure. Well, in fact, when I first got into the Nubrative course in Kyoto,

30:00 So I was sort of half wanted to do mathematics, but unfortunately, or maybe fortunately, I didn't have any talent in mathematics, I mean not much talent in mathematics, so I decided to do some physics, and then I found that, well, although I am not really, I was not, As a mathematician, I was not very good, but when you come into physics, then I have some mathematical background, which makes me very easy to understand relative theories and all those things, while quantum mechanics sounded too difficult to me. So I thought, well, I just stick to classical theory, and that was my undergraduate age. so I think so maybe motivation is not really a good one sometimes but once I get into the graduate course there are so many subjects to be worked and students and faculty members have their own fields of interest they teach me very interesting problems phenomena in astronomy or you know, astrophysics or other, in physics in general. So I gradually sort of interested in more physical sort of things, you know. You mentioned a number of times the discoveries that were made that, you know, were too early or wasn't sort of real or were significant to them wasn't realized by other people. I suppose that's a really common thing in relativity. Is it a problem that you think has gotten less so over the years, as relativity has gotten to be more of an active field, or is it just something that continues to happen all the time? I know that people... Well, you know, as you know, that if this LIGO project were not approved, then it's still sort of academic, purely academic field. So nobody would be interested in these things, right? So I guess, well, but still

32:30 Well, there are people who are interested in those things, I mean, you know, regardless of whether, you know, it really is connected to the actual experiments or anything, you know? Sure. So, well, in some sense, my starting point was something like that, I mean, I didn't care about, you know, what the real world is or, you know, real experiment could be I was just so interested in what the general relativity implies, or what kind of a picture of world you have from a relativistic point of view, and so on. So, yeah. And is there any sense in which some of these results that people discover at a certain time just don't travel outside the country that they're in, or maybe the working group in that country? Well, another thing is that I think now it's really changed, but in the old days, like 60s or maybe early 70s, Japanese physicists didn't have a lot of chance to go abroad or have discussions with other people in the world or anything. And also because their English, my English, is not so good. Even speaking is not so good. Writing is not so good. So sometimes, well, even if one paper contains really essential sort of result, but people from, say, America or other countries just do not understand the point. because of the bad English. And that really happened several times, I guess. And are the personal contacts important also, presumably, that even if you write a paper, of course, if you haven't met? Yeah, but once you meet, yes, and then even the English is very bad, you can use blackboard, or you can talk and talk

35:00 and again, and then if you start to understand each other, then it's very easy, yes, right. Once you've got enough confidence. That's right, that's right. But that's not much of a problem nowadays, right? I don't think so, because, well, there are, I mean, as far as the written English is for some, yes, you can see on, say, those web pages that, I would say, the archives, Because the paper written by Japanese students, it's very hard to read, you know, sometimes. I'm usually very impressed. So anyway, but it's improving a lot, yes. Yeah, I think so. Do most students in Japan, would they have done English since high school or do they specifically try to study English when they get to college or graduate school as well? Well, they learn English from junior high school, so six years in junior high and high school and also in the college for the first two years they study English as well and also we sort of give them English exam when they enter the graduate course yet their English is not so good so I don't know something is wrong about that so probably most of them most of the Japanese government very good English especially for writing do they once they get to graduate school are they then exposed more to English speaking and I think the as I said that has improved a lot from 10 or 20 years ago, while students now have quite a few occasions to meet people like you coming from abroad to Japan. So they have quite a lot of occasions to try to speak in English or try to express themselves

37:30 in English. So that's why, I think, in that sense, English is improving, yes. That's interesting. That's a difficult problem, of course. I think probably the same thing happens, I'm sure it happens between lots of different countries, but also between Russia and Western countries. You seem to have, you know, people there know something already and the other people don't, because they have just unheard about it. I was interested in one suggestion that I think was put forward at the Maldi conference about, or that I heard people talking about at the Maldi conference about TAMA, that there was the possibility that at the level of sensitivity that was achievable, it might to see black holes, so I was curious about the background to that, how likely you think that that scenario is, what sort of the chances are, and also how long would it go about searching for them? Well it's really worth searching for it, because as you know that if those much of experiments, I mean, those detection of gravitational lensing is really significant, I mean, correct. And there is no theory which at least from sort of stand out as a physical point of view, no theory to produce such a huge amount of number of small stars. So this must be something very peculiar, and in some sense, from a general relativistic point of view, only reasonable candidate is black holes, primordial black holes. Otherwise, you can't have such a dark star in those mass range of 0.1 solar mass or something. I mean, if you have, and this is more than 10%, maybe 100% of the halo mass, right?

40:00 It's considered of these machos. And then if these are, say, brown dwarfs or white dwarfs, then all those astrophysical sort of theory of star formation and everything should be sort of reconsidered again. And it's very unlikely. So in that sense, well, you know, you can either bet on some, I'd say, strange mistake in the macho experiments or the existence of black hole machos. but in any case of course this is as I said it's only a candidate so you have to see whether this scenario is really true or not then as for TAMA although well if this black hole match scenario is correct then there is a possibility of detecting it 3 or 4% possibility So it is very important to pursue whether, to see whether, well, even if you can give the upper bound to the number of those machos, it's interesting now. And also, well, very interesting is that once LIGO is done, finished, then LIGO should see those binaries, I mean, since it's much more abundant than the binary neutron stars. Right. Yeah. So in that sense, you can really sort of check or test this scenario. which is, well, I guess it's, you know, it's very important to have some scenarios which can be tested by those detectors. I mean, right? Right? So this is one of the very interesting scenarios which can be really tested by near future interpermitters without any doubts. That's the very important point. Do you see any other scenarios that might be tested by TAMA in the near future?

42:30 By time, it's very difficult, right? I mean, it's, you know, sensitivity is very low. But also, so you just, well, you know this very interesting story about the Kamiokande detector of the neutrinos, you know? From, you know, they observed this, you know, 1987A. Yes, but it was really by chance. I mean, you know, they first, you know, configured their detector to solve the solar neutrino problem. So they found that their detector should be improved. So they shut down their detector, I think, the end of 1986 or something. And in January, they were sort of refilling the water, cleaning up all these detectors and so on. And in the beginning of February, they started to take the data. Right after they started to take the data, this kind of came. I was lucky. So really, really lucky. So in that sense, well, the same thing might happen to Tama. Sure, yeah. That's true. So in Deputy, you say? That's right. As for LIGO, I'm not quite sure, but, well, maybe there are some other scenarios that can be really tested by LIGO. I mean, for example, in the advanced stage, well, it may take a few more years, but I don't remember, can you really test, say, whether the binary neutron star coalescence is related with, say, It's still sort of halfway between, I don't know. For example, you know this. Yeah, I don't know. I'm hoping to go over to talk to Mark Reeves, about how those people view them. But, sure, so probably it'll take some ingenuity to come up with the scenarios that might be testable. Right. Hopefully there are some out there. How did the idea for this, with these match hoes,

45:00 is it that you have two of the primordial black holes colliding? Well, you have sort of an abundance of primordial black holes. Then, assuming that they produce sort of randomly in space, then you can calculate the probability of forming binaries, yes. So from that estimate, you get this kind of number. Right. And how did the idea for that scenario come about originally? Was it just inspired by this gravitational lensing work? Pardon? I was wondering how the idea, when the idea for the scenario came about. Well, the scenario for black holes for machos was not new. I mean, people discussed a lot about this because that's sort of only, primary black hole could be the only possibility of forming such a lot of matches in low mass range. So, but they, well, if you have those black holes, then, you know, you may have some binaries. Once you think of binaries, then, you know, it automatically connects to this binary coercion and then can give you some reasonable, yes, collapsing rate. Sure. I was just wondering if it was recent that people actually did the calculation for the rate of binaries in the thing. Was that only done recently? Or were some calculations of the possible event rate done? Because I know, as you said, that the idea of matches being from real black holes has been around a long time. But I guess I'd only just personally heard of, I was just wondering if the, you know, looking at possible event rates for the gravitational waves, if that was directly inspired by the current round of detectors or if that too had been around for a long time. Well, yes and no, because another reason is that after this discovery of gravitational

47:30 lensing, you know, those macho groups searched for other, how to say, phenomena, I mean, similar to the gravitational lens, right? And they announced that they might have found some binary event. Yes, and that was another reason, I mean, you know, so, well, machos could be binary. Right. So, since you mentioned, since of course you work both in cosmology and gravitational ways, I'm interested to ask if, what do you think are the possibilities of detecting, say, well, gravitational and gravitational ways of cosmological origin with the neuronal detectors? Of course, at Cardiff, we have Nina Grishok, who's also very interested in that problem. So I was wondering if you want, well, if, for instance, you see any testable scenarios arising in the air cosmology from detectors like LIGOR or PANEL. Do you see any possible cosmological scenarios that could be tested? Well, maybe not by LIGO or TARMAS. Well, you know, usually all this, well, maybe, of course, it depends on the scenario, but usually you have really long wavelengths, gravitational wave perturbations. Well, for example, those could be detected by the cosmic microwave background and I said it would be some horizon scale where one oscillation takes a cosmic time, a cosmic age. So, yeah, well, for short wavelengths, well, maybe, for example, some people discussed about those bubble collisions and, you know, gravitation wave produced by those collisions or, you know, some gravitation wave produced from the cosmic strings and so on. And I'm not following much of these scenarios,

50:00 but maybe in these scenarios, we can have relatively high frequency. I mean, still it's quite low compared to, say, Tama or LIGO, So you have to wait for Lisa, I think, but maybe enough for Lisa could detect some, you know, well, in the cosmological sense, high frequency wave, I guess, at, say, 10 to minus 4, 3 or 4 halves. So Lisa is probably, might well be interesting for the body of the cosmology, and actually see some of the ways. Are there any null results that would be interesting here that could eliminate certain scenarios? Right. I mean, for example, those constant timing data also already excludes, you know, the amount of gravitation wave to a very low value. And, of course, it depends on the spectrum of the gravitation wave. I mean, you know, so the spectrum frequency, I mean, I guess, well, those timing data excludes the frequencies shorter than at least distance between the pulsar and the error, I guess. But for very high frequencies, usually, I'm not quite sure, so maybe in the visa range, Maybe there is quite a sort of severe constraint from those pulsar timing data. I have to look up some papers. So the pulsar timing might already have put a serious constraint from the point of view of theorists producing the top-long test. I've been meaning to look into that. Excuse me. Just a minute. Just a minute. What time is it? What time is it? What time is it?

52:30 What time is it? I'm ready. I'm ready. I'm ready. Should I let you go? No, no, well, he will cook dinner this evening for me. I mean, not for us, actually. Yes, that's right. Well, that's an important subject. Right. Well, in any case, we've covered most of the ground already. So I guess I was just going to return then to the more recent work that you mentioned that you've been working on. I gather that the plan at the moment is that there will be a successor to Tama, which will be... We hope. We hope so, yes. But nobody knows. I mean, it only depends on how successful Tama will be. At the moment, it's very successful. But, well, we should go a bit further down to, yes, in the insensitivity. Sure. And do you, is your own work on radiation and radiation reaction problem motivated by, you know, the possibility of future detectors and that, or is it still, mostly as it was when when you started out it's still you know you were interested in what you can find out within the theory uh both yes yes i mean uh of course it's uh you know let's say uh since you know because of this sort of success of the black hole partition you know calculations that the further to include those you know reaction force in the calculations then it will give you you know sort of a very good sort of testing ground of the theories or models or you know well and also in the uh this uh you know laser detector case then the main target is black holes right? Because of very low frequency. So then you really need, you know, what kind of gravitational

55:00 wave you can detect, I mean, waveforms or, you know, what kind of trajectories you plot from those gravitational wave data or, you know, whether you can really determine, say, for example, curve parameter, all these things are related with this calculation, right? So, in both ways, I mean, you know, if, for example, this calculation turns out to be very useful, for those, for example, Lisa or even for LIGO, then, of course, it would be very interesting. But even if it turns out to be not really useful, from an academic point of view, it's an interesting question. Right. So... Sure. Well, so in that sense, both. So you mentioned that you, if I remember correctly, that you're working at the moment on actually calculations of the radiation reaction force at inch four or two. So the other, I know that it's been, when I started at Caltech, in fact, one of the things that Kip told me to look at at first was the Gauss cell formalism for the radiation reaction force, and we decided after all it wasn't very good, we went and did other things. But since then, of course, with the work of Munoz, Mark and others, there have been lots of progress on the conceptual side of working on the formal radiation reaction. Are there still kind of formal conceptual hurdles to overcome? I don't think so. It's really just more calculating. That's right. And do you see the calculations that you might make that you're working on now as being producing, basically, as enabling one to produce templates that would be usable by Lisa or some sort of space-based detector?

57:30 Yeah, so that's what I hope, yes. Sure. That would be interesting. I was at the last meeting of the binary black hole ground challenge in the US and Kip was there and he was getting them all to bet on whether they would be finished before LIGO was detecting binary black holes. So I was tempted to say that he should try and make the same bet with people working on the radiation reaction problem in the perturbation case because at least they had a much better chance because at least it was going to take a longer time. But I don't think he'd take that bet. I guess just speaking of the Grand Challenge Alliance, I was struck that in the U.S., at least on the numerical side, you had this tendency towards these big collaborations developing for people to work on problems related to gravitational waves. Is there any similar movement in Japan, or is it... Well, in Japan, the group is not really large. I mean, essentially, the leader is Takashi, Takashi Nakamura, who are sort of in charge of that, you know, numerical part of the relativity. Right. And right now the Ohara and also Shibata. Shibata is now in the, I think, Illinois. I'm at it. Okay. Well, as far as their calculations are concerned, they are making a lot of progress. Well, as I said, it's very difficult to relate those calculations with actual templates or anything. But, well, even, you know, if it is not sort of so much useful for those, you know, template construction, well, as physics or astrophysics, it's very important and interesting to see what goes on in those, you know, binary coalescence, right? Right. Yeah. So in that sense, I think they are doing very good work now.

1:00:00 Yeah. But unfortunately, the group is not so large. But most of the people who are working sort of collaborate at some level? Well, for those real simulation things, I think, as I said, Nakamura, Ohara, and Shibata are the only three. It's the normal size of a group. Thank you. Well, we can just take a break and have a coffee with you. Sure, yeah, that would be great. We can do that. Refresh. Yeah. Okay. I was going to ask about sort of the data analysis issues and how that is dealt with in terms of who has access to the data and so on, which is something that, as far as the European detectors, that they're only just beginning to work out how to handle this. But in the case of TAM, is somebody like yourself able to get your hands on data if you wanted to, or is it just that people who are directly working on it like hidey if you just... I don't think that part is not determined yet.

1:02:30 I mean, well, I guess be open to public or sooner or later. But at the moment, you know, they don't have time to consider those subjects. Well, that's after you've done everything, all the books and... So perhaps you're going to have a meeting, right? Yeah. So that, I mean, you just asked. I was just curious if it's something else. It's not something that people are really worried about. I mean, it's not. They are worried about the future of the projects. So, you know, I mean, if, for example, I mean, it's sort of a pity that if people from TAMA or near the TAMA group did not analyze the data while somebody else did something, data analysis of Sama and found something very interesting, then it's kind of a shame. But still, I mean, even if it's a shame that it's very nice to have some interesting results, which gives you more future for this project in Japan, right? So in that sense, everybody's welcome, I guess. True. Yeah, but the rest also depends on sort of our experimenter's pride, I mean, what they think. Sure, there's a tension that I would say, you know, do you want to keep it to your style? Yeah, in some sense, I mean, as a theorist, I mean, you know, for me, I mean, very personally, sort of selfishly, it doesn't have to be tower. I mean, anything else is okay, right? which sort of test my theory or scenarios right now, which can be, you know, which gives me sort of, you know, very interesting data as well. Right? So that's, yeah. Yeah. Sure. So, yeah, I suppose in the sense that here's position is different. Do you find that you have much contact with the experimentalists, for instance, on TAMA, or is it only occasionally at a meeting or that?

1:05:00 Only occasionally, yes. Especially recently, I don't meet them very often, except those several people who were involved in data analysis. So there are sort of meetings of the whole data analysis group? Right. But you said that Himyuki is certainly doing that, and so we haven't got a chance to talk to him. Yeah, and recently we bought, I think it's nine decades of our machines, which you will see in Osaka, you know, sort of, you know, sitting in the computer room, nine machines. And this machine is called Tamako, means children, child of Tama or children of Tama. Okay. And that's the analysis. Yes. And Hideki is working on it very hard. That's a big job. So that's all the questions that I can think of. Thank you very much. You're welcome. Very interesting. Cellar mass objects falls into a mass black hole is a detectable signal at quite a big range. And the question is what that rate is, rather hard to estimate. But that's another quasi-periodic type of signal for LISA. So would your impression be that the prospects for LISA are essentially better than, say, Earth-based detectors because, for instance, they can see these kind of low-frequency periodic waves? Well, I think it's less likely to get a null result in the sense that the occasional supermassive black hole coalescences with very strong signals, there's a debate about the rate of those, there may be only one per decade, and so the next best signal would be merges of black holes as, say, 10th to 5th solar masses, and the issue there is whether low mass black holes form in small galaxies at high redshift.

1:07:30 And my former student, Martin Heneldt, did a paper in 1994 estimating this, and there hasn't been much progress since then, about what the formation would have to be in small galaxies. Because the point is that if every galaxy is a result of a merger of a lot of small components, then you get many events in building up a big galaxy, and that pushes up the event rate to more than one per year. but then the other class of events is stellar mass objects falling into 10 to 6 stellar mass black holes that's rather hard to estimate but that would be a quasi-peotic fairly weak signal which would be something which I guess would be on the level with the sort of 10 minute binaries which is the other type of source that we see well you mentioned the possibility of non-results And that was a topic I'm sort of interested in. Are there reasonable scenarios that are of interest to astrophysics that either LISA or LIGO type detectors might be able to set limits on with null results based on their projected sensitivities? Well, I think that... Take about LISA first. I think if it doesn't detect strong signals at the rate of, say, one per year that will tell us that there aren't a lot of million solar mass black holes forming in small galaxies at high redshift because if there were, then the mergers of those would go higher vent rate and it would also obviously constrain these other things but that would be less surprising In the case of LIGO No, most predictions of LIGO phase one predict nothing. It could be that the rate of binary coalescence is higher because some binaries fall fairly close already and then coalesce quickly. The only other grounds of optimism might be if a significant fraction of supernovae are sufficiently non-spherical that they give gravitational radiation. It's normally assumed that supernovae will not be the dominant signal

1:10:00 because the predicted emission from a supernova is so far below what you get from a coalescing binary, despite supernovae being thousands of times more common to not dominate. But it could be that some significant fraction of supernovae would give a rather more powerful signal than is normally assumed in supernova models. so that they are significant to the event rate. And I suppose there's slight ground to optimism in the evidence recently for non-spherical effects and possible jets and gamma ray emission from some supernovae. That might push up the event rate a bit. But whether those will be detected, I don't know, because first of all, the power emitted is not as high as you get in a coalescing binary. and also if you don't have a clean periodic signal or one whose template you know as the case of coalition binary then obviously it's harder to pull the signal out of the noise I've heard it suggested but it's not a theory that I know much about there's a theory by Bethart and Brown which might predict a large... more black holes Right. And that, therefore, for instance, even LIGO 1, according to the suggestion of her, or maybe a more advanced LIGO, might be able, for instance, to put a null result that would falsify, that's not quite the word, or put a severe constraint on the predictions of that theory. Is this because they expect relative to raise when they fall? I think the idea... Or when they merge. I think the idea is that when they merge, that if they didn't see... I see. That there wouldn't be that many binaries around and therefore there wouldn't be these huge numbers of binaries. I see, so they're predicting more black hole binaries than in the... I guess so, yeah. Yeah, that's a bit of not too certain. So I was interested in if there are any examples of that, astrophysic models that you're interested in that you think might be testable? Well, I mean, obviously there are speculative sources that could be detectable, like cosmic strings and

1:12:30 bubble collisions in the early universe and things like that. The primordial gravitation waves, particularly by most theories, would not be detectable, I think, either by laser or by LIGO. So as a general rule, there might be, as you say, of our predictions, which you could test, but probably nothing that's really expected by astrophysicists. Right. I was interested also in gamma-ray burst models. Is that something that an astrophysicist would be interested in? Presumably there wouldn't be anything that would be tested if LIGO doesn't see something, at some point? Well, some of the gamma-ray burst scenarios involve compact binaries, but the event rate for gamma-ray bursts no higher than the event rate that's been assumed already. So the gamma-ray bursts, if there wasn't a beaming factor, are one every tenth of the seventh years per galaxy. And you've got to mark that still well below the rates that have been assumed for bio-coalescence, the advanced and herbal rates of about 10 to minus 5 per year or something like that. So I don't think anything in gamma-ray burst observations provides any grounds for further optimism for LIGO, except, as I said, possibly the evidence that a lower-level burst from some nearby supernovae, because that may support the view that some subset of supernovae significantly asymmetrical. But that's the only sort of new angle which in my view gamma-ray burst studies provide relevance to LIGO. So it might, as you said, provide more hope that supernomers might... Yes, but still I think if you put in numbers it probably is not going to be as promising as the binaries. If LIGO does see signals from coalescing compact binaries, do you think that would be important from the point of view of gamma-ray bursts? Is that still considered a very likely source of gamma-ray bursts? Well, I think it would be interesting because it's certainly one of the candidates for gamma-ray bursts. But, of course, the gamma-ray bursts we observe are very far away.

1:15:00 So the chance of seeing a gamma-ray burst from such a local source to be detected in correlation waves is very small indeed. The, when people in the sort of gravitational wave community that I used to work in, I still have a bit of contact with, talk about event rates from say neutron star binaries, they always had this figure which originated I think in a paper master of Finney in 1991 or so, of three per year up to 200 mega parsecs. and I noticed that figure sort of stayed the same when people in my field talk about it over the years and I was curious if still in some sense valid to talk about that, I mean I know that some of the numbers that go into the calculation have changed over the years with new estimates to the actual Yes, I would say it's still very uncertain I mean the best experts are people like Ed Van den Hovel, he's the leading expert But there's always the possibility that there's a class of biomes that form very close and emerge very quickly, and so they don't appear in our present day samples of biomes, but none of this could dominate the rate, but I think that's as good a guess as any. The statistics we are our sponsor. And I guess just the last question, I've been given the impression by some people involved in LISA that astrophysicists are more interested in LISA than in LIGO, on the basis that it might see more astrophysically interesting things. Is that something that you'd agree with? Well, that's true in my case. Yes, in that I think LISA will see something, whereas I'm not convinced that LIGO will see anything. If LIGO sees something, that's a bonus. Right, okay. So in that sense, you think it's a more viable kind of detector, as far as we can tell now? Well, I mean, of course, we don't. It leads us a long way downstream, so whether it can actually achieve sensitivity as advertised, we can't be sure about, I suppose. but if it can, then the excitement would be that it could detect supermassive black hole coalescences with a very high signal to noise, which would actually allow you to study the waveforms in a way that you can't do for any of the events that LIGO expects to detect.

1:17:30 So the signal to noise would just be high enough to make it really interesting. I just got invited to the inauguration of LIGO in Louisiana. Oh, okay. I suppose it would be interesting to see that I'm fascinated by the idea of making such a giant interferometer actually work. Well, the main cost is in the vacuum pipe, isn't it? It's certainly big. Well, thank you very much. Okay. So you're doing a PhD, are you? Well, I did my PhD in the States at Caltech, in physics and in history of science. Oh, I see. It's really a physics PhD, but I ended up doing as much history science as I did. So what was the history? The history was of the orbit of the dynamic of binaries by gravitational waves. In particular, the controversy or debate in the 50s and 60s over whether it actually happened at all. Oh, yeah, right. For a considerable time, people like Bondi suggested that maybe it just didn't happen in periodontics. the news function and all that so it was really looking at that and then a little bit later the quadruple form of controversy to some extent as well so it was sort of that yes that was in the 60s I suppose so it was interesting and I went around and talked to some of those people I really enjoyed it so I enjoyed it so much that I decided to continue on and there's a chap in Cardiff I think you met him Harry Collins at the British Association but he was he wrote something on Webber and so in recent times he decided to actually look at and these newer detectors follow the story behind that so you're working with him sort of in parallel I'm looking at the theorists and how they relate to the problem of detecting gravitational waves especially people who do theorists who are interested in data analysis for LIGO the numerical people who are trying to solve the binary back hole the grand challenge problem mainly in Potsdam now that's right, it's a big group there and I was looking at people like

1:20:00 who did this work who had this neutron star binary simulation which no one really believed yes, that's right The gamma ray burst came up there, too, because when they were having so little success in getting the relativists to believe the results, they decided that they should try coming up with a gamma ray burst model based on the idea that the two stars would be coming in and then they'd collapse separately. And then the collapse would release burst of neutrinos and some other gamma ray burst and so on. But I think what they found is that they'd already lost sufficient credibility in their own community that they sort of weren't able to interest another community, saying astrophysics community, in taking off their own work. Yes, well, I didn't look at it, but anyway, yeah, I mean, I think it's one of these results which probably can't be exactly repeated, but may, but it will probably not be corroborated. You have to decide in these things whether you're going to make an effort to follow up and study in detail something or whether you'll place your efforts in some other direction. Right, yeah. Sure, this is the problem. It takes such a big effort to, as you said, to try to corroborate or find out exactly where those are came from. Yes, yes. But I mean, you know, I can think of lots of other observations of that, claims of that kind. Well, not just theoretical computer results, which, you know, probably were wrong, but no one knows why. But lots of observations of that kind. There are quite a few cases where very exciting observations have been claimed. And, you know, one tries to make sense of them and you count the difficulty. but then people sort of lose interest because they suspect that it was just some mistake and that it wasn't real. And I can think of several cases like that. So it's a common phenomenon. Sure. If someone sees a sort of rare event and they can't repeat it, or if people repeat it and don't find it, then it's very hard to pick out what the first guy actually did wrong, if they did something wrong. things just get sort of forgotten about that

1:22:30 yeah it's interesting I suppose you have to do that for a practical point you may sometimes be wrong in your judgement but I think we all have to all through my work I have to decide if there's some fascinating results should I take it seriously or should I be suspicious and suspect it'll run away next year do you think there are markers that you can use to to pick out what's like? Well, I mean, obviously, one should, based on the judgment of people closer to the experiment and one's assessment of the more incredible the result is, the higher the standard proof you quite rightly expect, as it were. So the more, yeah, the more unexpected. Yeah, yeah. actually to Francis Everish, the Gravity Probe B guy, and he was talking about, I think the results of Gravity Probe A, and he was saying, well, you know, I believe these results, and I thought for a long time why I believe them, because it's such a difficult experiment, and I know that's such a lot could go wrong, and I came to the conclusion it was only because I admired the people doing experiments so much, you know, I thought they were honest and thorough, and so I figured they must be right, and they said, the only thing that worries me is, you know, I think I'm pretty honest and pretty thorough, and I know I make lots of mistakes, so I guess they could do it. Yes, yes, yes. But obviously some people have more credibility than others, and once you see the micro-background experience, all these things, you know. But Gravity Pro B, I mean, I think does have a problem, because I heard Francis Everett talk about that when I was a student here in the 60s, and at that time, it was a quick, cheap experiment, and we only had the post-Newtonian GR terms to 10%, whereas now, of course, we have the post-Newtonian terms to 1,000 plus the binary parser, etc., so the level of competence in GR is much higher, so much higher that I suspect if Gratty Probe B gets the expected results, no one will be surprise, if it gets the unexpected results, no one will believe it. So the most exciting result of Gravity Probe B would be a request for another half billion dollars to do it again. And so that's why I think the Americans are crazy to be doing it.

1:25:00 Right, it's sort of lost perhaps its validity as a theory testing experiment. Because there have been so many other tests of the same parameters. Well, it's true. It is testing something a bit different. And no one doubts it's a very, obviously, extremely ingenious experiment, but it would be nice to know the answer. But the point is that the burden of proof on any experiment which gets a discrepant result with GR is now stronger than the burden you would impose in the 60s when the end of the GR was not very strong. And so I think if it gets a discrepant result, my prediction is that most people will not believe that discrepant result. And they will say, well let's do it again. If it was done again by this technique, that would be extremely exciting. But I think the first time round people will believe it. And I suppose the problem is it's so expensive to do it again. That's right, that's right. So if I had been an American I would have cancelled it ten years ago. So, okay, well thank you very much. Good to see you anyway.