Interview with Thibault Damour

0:00 So, the date is the 11th of October, Tuesday, around 10 a.m. and I'm talking with Professor Crebeau-Demont. So, as I mentioned, what I'm especially interested in is, to begin with at least, the impact of the binary pulsar data. I believe that you worked very specifically to try and study the specific physical system that was involved in the binary neutron star and to bring the abstract radiation reaction problem into agreement with the data that was coming in. And I wanted to know what you felt was the general impact on the field of the discovery of the binary pulsar. Okay. Before maybe, and that would be a way for me to plunge back into the past, maybe I should sketch from my own point of view Binary Pulsar entered the game, because one of the reasons why I caught quickly on the Binary Pulsar phenomenon was that I was interested in the whole stuff before the Binary Pulsar. So let me sketch maybe my own scientific history. That would be good. because I was just thinking back that so I started studies in generativity in France and I got my PhD in France in 1974 June, 5 June 1974 and in fact my work concerned radiation reaction in linearized theories of gravitation, let's say, which is not a very realistic theory of gravitation but the idea was to study renormalization aspect in classical theory

2:30 so it's classical theory for renormalization along the same style that Dirac did for electromagnetism. You remember that Dirac studied the classical electron, classical electromagnetism, and in 1938, if I remember correctly, this problem, and that was the time where they had realized that there was a problem with quantum field theory, I mean in fact quantizing the electromagnetic field, and therefore at this stage they had decided to go back problem of point electric, point-like electric charges, to understand better what was the problem of the self-action of the particle itself, which includes both an infinite part that you renormalize by changing the mass, essentially, and a finite part, which is radiation reaction. And this is the way Dirac got back the famous radiation reaction formula, first obtained Lorenz in non-relativistic form, for Lorenz, in fact, and then Abraham got the relativistic form. I guess this is the history. I must have written about this, that Lorenz got the non-relativistic form, two-thirds of square V dot, something like that, and Abraham got the general relativistic form, and Dirac got back the relativistic form, but by this new way of looking at the thing, which is directly the self-action of a point particle on itself. So anyway, in fact, the topic of my thesis was to look at the same problem, but in theories where you look at spin 2 and spin 0 field interacting with point-like particles, point-like masses, which is, so it was not an attempt to have the full generativity, but just to look at it. So anyway, just in June 1974, so this was a pretty formal work, it was just a training exercise for me, but I was fully aware of the intricacies of the problem when you have point-like particles, what happens, you have to separate the self-action and try to subtract what is the radiation reaction and and at that time I must say

5:00 to continue my history yeah the reason why I had done the thesis on this was that first I should say that in France in 1974 very little activity was going on in the field of relativistic gravity it's in general this was a time where And because of historical reasons, there were still people teaching generativity in Paris, namely in the group of Madame Tonla, and that's where I went because at the time, as a kid, I had been fascinated by generativity, and then I wanted to study that. So when I entered the Ecole Normale Superior in Paris, I went to the only place at the university where there was a good teaching about generativity. But at the time, basically, I knew already more than my teachers. And I had studied, in particular, since my youth, I mean, before university studies in, how do you call it? Secondary school? Yeah, at the end of secondary school, I studied Einstein's papers, and in particular the famous paper with Inferl and Hoffman on the problem of motion, and where they were trying to speak about that generativity because of its nonlinear features and very specific properties about what is the self-action and interaction of bodies and this is the problem of the motion of singularities in the Einstein-Felhoff-Mann and very specific way of dealing with surface integrals and they were looking at the structure of the bodies be them singularities or not They were partly things which were partly unclear in the Einstein-Fell-Offman and partly evidently very nice. And my interest had been especially focused on the Einstein-Fell-Offman type work because as a kid I had discovered a mistake in this work which I had corrected and technically I could integrate better than Einstein could at the time and solve the problem which I realized later that Kerr had done too, independently, but I knew nothing of that.

7:30 So anyway, when I came to Paris and I looked for, at the PhD level, pre-PhD level, where I could study, I wanted to study the generativity, but I had already quite a background in this, and I wanted to understand better the problem of motion in generativity. I remember, in fact, meeting Madame Tonnola and telling her that I knew already about the ancient Feloplan, and I wanted to work along these lines to understand better these mysterious problems of the motion of either straightly sympathetic bodies or singularities. It's not clear in ancient Feloplan, what they are dealing with, but I wanted to clarify that for myself. And that's why later she proposed to me, she proposed that I look as a PhD project, the generalization to gravitational background, gravitational framework of the work of Dirac on radiation reaction and the title of the thesis is in fact Classical Theory of Renormalization, Theory Classique de la Renormalization. And so I had finished this work in June 1974, and I wrote a paper, because technically I found new methods of dealing with self-action of particles, so I have a good technical handle on how to deal with the interaction of punk-like objects and things like that. And this is precisely, so that was in June 1974. And before that, and in fact, I can never remember exactly when I met him, but either in 72, 73 or 74, which is quite big, I met John Wheeler, who gave a lecture here at Institut des Etudes Scientifiques, which was totally disconnected. I did not know about the place here, but John Willow gave a talk here, so I came from Paris by train, like he did this morning, to listen to his talk, which I remember vividly, and then I asked him whether I could go to Princeton, which for me was the mecca for historical reasons, so I wanted to go there.

10:00 And I had, being at the Ecole Normale Superior, I was given there a special fellowship, which was an American fellowship, the so-called Jane Eliza Proctor Fellowship, which was given one every year for people from the Ecole Normale. And that year, in fact, they had to request a second fellowship, because there were two people from the Ecole Normale who were keen on going to Princeton, Bernard Julia and myself. Anyway, we got, both of us finally got, they got an extra fellowship, and so we had this money, so maybe you want to stop that? Maybe that's getting too long for you. No, no, no, very interesting. Okay. So, yeah, so we had... Yes, so anyway, so I asked Wheeler to invite me, I mean, I had a fellowship, so it's not going to cost him anything to be invited by his group, which he accepted, and that's what I did in the fall of 74. I'm saying that because the data are, in fact, important. So conceptually, I've been interested in the point of motion and Einstein and Ferloffman. Technically, I have done a thesis on radiation reaction and classical renormalization theory in pseudo-gravity, generally linearized gravity, et cetera. And then, in the fall of 1974, Bernard Julia and myself went to Princeton early in September to start a postdoc so both postdocs to start a postdoc year as Jane and Eliza Proctor fellows there in the physics department went to the particle physics John Willowswood and this is precisely in the so just, I don't know, a month or two after my arrival I was at Princeton I was technically well prepared and that's when we heard about the discovery which had been done in the meantime, in July, and which was announced in September-October 1974, informally, first, at Princeton, and I guess maybe I heard about it in one of those lunches at the Institute, or directly at the Physics Department there.

12:30 So anyway, so for me, immediately I was, I guess, very fast struck by the interest of this system, and at the time I had started within the group of Wheeler, the group of Wheeler had in fact dissipated away, which means there was Wheeler left, And the only one from the early structure of the group left was Remo Ruffini, but most other people had gone away at the time. There were a few other people left, but I went around to see what people were working on, and nothing was very much of interest to me. but when Raymond Ruffini went back later, when I did in October and so, maybe late in September, then he had problems in black hole physics that I was interested in, in fact on which I had worked because the year before, a friend of mine, Natalie Deruel, who was at the time a personal friend, yeah I should say that, So in Paris, at this place where we had studied under Madame Tonnola and her group, I mean the only teaching place for generativity, Natalie D'Ariel was also there, she was a year after me, but we had met there, and she was from the economy of Jean-Pierre, and she was from the economy of Boris, and so anyway, we had common friends there. So we knew each other, we were friends at the time, and she had been to Princeton a year before me and worked with Remo Ophini on Black Hole's problem, and she had problems she could not solve, so she had asked advice for me during Christmas, Christmas of 73, or 73, 74, so I was aware of what was going on at Princeton with Remo Ophini on Black Hole Physics also at the time. So I was well prepared for working with Remo Uffini on backward physics, and that's when the announcement of the discovery of the binary pulsar came up, as a surprise to all of us. And, in fact, my first published paper was with Remo Uffini on the binary pulsar,

15:00 because this is even a funny historical fact that all the papers on the binary pulsar have been published in 1975 including the discovery paper in January 1975 followed by many other theoretical papers, but there is one and only one paper published in December 1974, and this is our paper with Remo Ofini which was written in Princeton but which was sent to Paris at the Contrandu of the Academy of Sciences and at that time the Contrandu of the Academy of Sciences could publish very fast in two weeks so it's just for this reason it was not written before the other ones but it was published faster and this way I remember it was written and this way it was also my first published paper because the paper I wrote on my thesis work it was sent to Niovo Cimento was published on so my interest immediately was brought to this and although our paper does not the standards of the paper I were to write later on the binary pulsar, it contained some allusions to the spin orbit effects and things, the new things that could be obtained, but it was very pessimistic, for instance, on radiation reaction, because in fact in this paper we estimate the energy loss and it seems so small that we conclude that probably it will not be observed ever. So this paper was not concentrated on radiation reaction between Plymouth and the binary pulsar, but that kept me thinking that this was, from my point of view now, I had realized at the time, starting maybe slowly from that time, but still I remember in the fall of 74 being at Stanford and discussing with people like Francis Everett and others, that I had realized at the time that the fact that these objects, neutron stars, are very small, very concentrated, that you have strong self-interaction, high fields, I mean, very strong gravitational fields, pose an absolutely new problem, and that cannot be solved by what already appeared in the literature, which were weak field methods for dealing with the binary pulsar, and at the time, I had already set myself the goal of looking deeply at what I knew about Einstein and Freud of man, radiation reaction, and maybe point-like, which were nearly point-like objects, which was the binary

17:30 clear that I wanted to go deeply in this issue already in 74 which I do not I did not do for a few years because I was very busy with black hole stuff and this is in the intervening years that after some work collaboration with a more fini on how they also quasar extraction of energy from black holes that I came up on the idea that it was very interesting to concentrate on the the surface of the black hole and attribute quantities like resistivity and things like that. That I have discovered in the interview here, so it kept me busy. And up to the year of 1978, yes, which was the year 1978, where I put together my work on this in a paper which was sent to FISRI, which was sent, I guess, in 77, but I put the Second thesis, because in France, traditionally, we are supposed to do two thesis, a PhD, one thing which is supposed to be the equivalent of a PhD, and the second one, which is a test data, which is supposed to represent a big chunk of ten years of work, a sum or something, a synthesis on the topic. And this second thesis, which is not my PhD, which I did in 1974, I finished and defended in January 79, and we can look at, maybe I should look at the date, because data is important there, because there are two coincidences. So this was, yeah, 10 January 1979, and this thesis was on quelque propriété mécanique, électromagnétique, thermodynamique et quantique des trous noirs. So this was the thesis of my work on black hole physics, culminating on the, or what can be done on surface effects in black hole physics ok, so so up to 10th January or in fact before because that I had finished writing the thesis I guess in the fall or maybe December 1978 so that I've been very busy finishing this thing and this is

20:00 The moment, so if dates are right, I think this is in December of 1978, that in Munich happened the Texas Symposium, and I attended, I was there in Munich in December of 1978, so you can check the data, but you can check the image, where Joe Taylor presented his results on a preliminary announcement that he was seeing, in fact, a p-dot, I mean, a change of the orbital period, which was compatible with the so-called quadruple formula for the effect of radiation reaction on the binary system. Oui? And I will give you a copy of this material. Thank you. We have more dates and things like that. Thank you very much. Is this on? I should mention in the meantime that I had met Joe before because I had attended, as a student, the Enrico Fermi School in the summer of 75, I guess, that's easy to check in books, which was organized by Raimo Ruffini and where Joe Taylor presented one year of data on the binary pulsar and at that occasion I had, as in summer school in a very nice environment on the Como Lake Lake Como in Italy I met Joe and Marietta and his wife and so I have at least a friendly contact with Joe already at the time which didn't mean and I had already begun in fact writing a paper for this thing speaking of the motion of objects with application to the binary pulsar spin orbit effects in fact in the binary pulsar I had discussed in this thing so I had done a piece of work trying to keep in mind and so on. But anyway, so just to explain that, from my point of view, I had been very busy with

22:30 dark hole physics up to December 78, and this was a lucky, a second lucky coincidence, in fact, for me. The first lucky coincidence was that the Banner-Posar was discovered just when I arrived in Princeton, so I had fast information. If I had been in France at the time, I would have been away from this information, I guess, or not so fast. And I was in a good book at Princeton. And the I was in Europe in December 78 when Joe announced this important thing. And evidently, now, at this same symposium, Mr. Arnold Rosenblum, Dr. Arnold Rosenblum, presented, gave also a talk after the hunt of Joe, saying that he was doing calculations on a radiation reaction by, in fact, the fast flyby. He was not doing the bound system, but the fast flyby of things. And by a very complicated calculation, he was obtaining a result which was in disagreement with the quadrupole formula. Also, I remember that Jürgen Ehlers stood up and said that, in his opinion, all the derivations of the quadrupole formula in the literature were quite unsatisfactory. that John Wheeler was there and stood up to him and said that in his opinion it was not much of a problem, it was clear that the whole thing would be true and the quadruple formula was okay. But anyway, from my point of view, it was very clear that in most of the discussions, people missed the important points that the neutron stars are very condensed compact objects with strong fields and none of the methods in the literature was dealing with that. It was also very clear that the state of the literature was very unsatisfactory and as this was really my initial project that I wanted to work on, really for me it was and I was just free of now my black hole thing, it also meant that this is probably why I never wrote a second FISREF paper on the things which are in my thesis and which are in a conference paper for meeting which is called surface effects in black hole physics which has been published as a proceedings paper but never published as a second piece of paper so and this is why in fact I never published a second paper on that because I stopped working on it and I wanted to put all my energy on

25:00 attacking this problem not in fact thinking much about competition that I I did not care whether other people were going to look at it. I knew from having heard Arnold Rosenblum that this was... I did not believe at all in what he was saying. I knew there was a problem. I mean, it had to be done correctly, but this was not for me the correct way of looking at it. The further papers I heard about, like by Cooperstock and also all this, was to me no interest in this thing. so I wanted anyway for me it was not like a big challenge everybody jumping on it and having to do it fast in fact the idea was not to do it fast the idea was to do it well and not caring about what people were going to do and in fact from the personal point of view this is also the period where for different reasons I had been given a fellowship starting in the beginning of 1979 a fellowship whose purpose was to give me money to travel around the world and just think back about what I wanted to do in life and opened me to the world to the world this was a fellowship which was in a sense the second fellowship of American money I have got because the first one was the one to go to Princeton and this one was money which was coming from Winiretta Singer which was the daughter of Singer, the inventor of the automatic so she had a big fortune and she had married a famous franchise and a French aristocrat Polignac, famous name Polignac, so Mrs Singer Polignac had left money as a foundation and part of this money was given each year to people from the Ecole Rheumat Superior and other schools Ivy League schools in France travel for six months around the world and I had got this thing which means that this opened the parentheses for me because in fact I did that and just at the moment where I said I will jump on this problem in fact I went traveling around the world but I started thinking and I started working effectively on it later in this year 1979 and I started a collaboration with my old friend

27:30 known for many years, and other people from the institute Henri Poincaré in Paris, so the work lasted in fact something like three years, I mean it was started in 79, 80, 81, And up to 1982, where we had obtained, in fact, I had obtained, because I prepared my mind alone, the last one, the final equations of motion for two bodies, taking into account the fact that they are condensed, so you need a special method to match the strong field of these objects to the weak field outside. and with, I mean, the technical difficulties that you have to push the calculation to very high order, cubic order in the non-linearity in the weak field thing, fifth order in the other C, and we had solved the full problem. And nobody else, in fact, had succeeded in getting near to this problem at the time, in getting the full equations of motion. So this work was finished in June 1982, I guess, and I sent my compte rendu of the Academy of Sciences paper in June, or July, when was this, in June 1982 at the Lesouches School, which was organized by Nathalie D'Arrel and Svipia. So from my point of view, so maybe I should stop here for a moment. Do you want to go back to the more general framework now? Sure, I mean, obviously... Maybe you should ask questions now. Sure. Well, I'm obviously interested in your personal interaction with the work, since that's what you can tell me about most precisely. And I'm very interested to hear the framework of how you came to work on it. One probably brief question. You were saying that you felt quite confident that the previous work was unsatisfactory to various degrees. At the same time, how did you feel about the answer that is the validity of the quadruple form itself? Did you feel that the previous work was likely to have derived the correct answer by proper methods, or were you open as to what they were? Yeah, in fact, I was open because, okay, the controversy was multi-prone.

30:00 I mean, people said this point, which is unsatisfactory. One of the arguments was that the so-called quadruple formula in Lifshitz is what I call in some of my favorite far-field quadruple formula. This is talking about energy loss at infinity, and this is not directly what you observe. you need a heuristic argument to connect that. I was not worried about this heuristic argument. I mean, we tried to have the direct proof of what you observe is what you get. You want to look at the motion, and that's what you want to solve directly so that you have no heuristic weak links in your thing. But now, as a physicist, I was convinced that the heuristic thing is okay, which means that when you have energy loss, you should pay for it somewhere. except that I was aware of subtleties like for instance there is more energy going away in electromagnetic radiation than gravitational radiation so if you write directly in fact energy loss you can get nothing because or you should get twice the answer or something like that because the total budget of energy loss at infinity is the sum of electromagnetic energy loss and gravitational energy loss so how do you attribute one to the other one so this was a subtlety but I was not very much worried And then I deal with one for me to separate the two. What I was really worried about was that when you look at the Landau-Liefschitz now derivation of this energy loss at infinity, there are some things which are not very clear from some integrals. But this was not very worrying. I knew that technically one could probably clean this up. This was not the main point. But the main point that they were dealing with weakly self-gravitating objects. and they appeared in their derivation something like mass density rho, a concept which is valid only for weakly self-gravitating object but which becomes uncertain by 20% when you speak about the neutron star when you define in the law really the main point for me was the following if there is a quadruple formula proven by Landau-Lipschitz, independently of these heuristic things, it is a formula which says that some observable quantity will be connected to a quadrupole computed in which you must inject the mass of the objects and the masses that you have to put in the quadrupole formulas is the integral of a Newtonian density rho on the volume of the bodies.

32:30 In fact, you compute the quadrupole, which is the integral of rho times xi, xj, then you take the symmetric trace-free part of it, but in which you have this newtonianly defined mass density, rho. And I was very, I mean, focused on the fact that in neutron stars, there is an ambiguity in what you call rho, and the mass differs from the integral of rho defined as the barium density, mass density by 20 or 40 percent, something like that. And now, it was not very urgent at the time, because there was in fact a discrepancy or there was one sigma error bar of the order of sigma, 20% in the announcement of Joe. But this allowed the possibility that indeed strong field effects in generativity would introduce not basic mistakes, basic errors in the quadruple formula, but significant cell gravity differences in the quadruple formula of the order of 20%. And this was this aspect that I most wanted to clarify, and also clarify all the fine points that we did not want to have heuristic reasoning, that we wanted to look directly at the problem of motion. So for me, this was the important thing, where I had darts, and indeed I had an open mind, and this was a bit not really disappointing, because it was also what allowed really to finish the thing and prove something, not so interesting, let's say, from a physical point of view, to conclude that whether you are dealing with black holes or neutron stars or ordinary stars finally there is one way of writing the fallible formula where all the information about the structure of the object is put in one object the mass but this was directly connected I should say with my thesis work of 1974 now the original thesis work which was precisely the problem of renormalization of mass and whether you can put all the self-gravity in one mass I was deeply interested in from the start. But because of Einstein-Fell-Lofman also from the beginning in a sense I knew one could work with outside the object surface quantities and characterize the mass of the object from the outside and not by integrating the density, which I never wanted to do. I never wanted to work with density or in the neutron stars and integrate something to get a credible formula

35:00 but work a la Einstein-Fell-Lofman outside the bodies and things like that. but still with this openness that maybe the result could be genuinely different from the naive quadruple formula. And really I felt there was something to be proven. Although people did not usually appreciate that there was something very important to be proven, because the final result looked the same as heuristics, saying, OK, there is a mass of the object, it's the same mass which enters in Kepler's law that you use to write that the objects move on elliptic orbits, and the same mass that you write in the radiation reaction. at all evident that it should be the same mass. And this was a hard point to be proven. And I must say that in the quadruple controversy when you look at all the things, most of the papers don't even talk about that. They don't even realize that this was the main point. They argue on wrong things and periferic things, but not on this main thing. to get back to the something you touched on with the actual data itself, you felt that there was room in the results from the binary pulsar that there was some room for corrections to the protocol formula due to self-gravity. Yes, independently of the definition, yes, definitely so. And observationally, there was room in the sense that I would figure somewhere here that this is probably the thing. I mean, the error bounds were of the order of 20% or something like that, so there was room for something going on for sure. And in fact, it was not even well centered at the one signal level. So, later on things concentrated. Well, going then just briefly after the period that we discussed up to now, some people, I believe, have continued to express some doubts about various aspects of the radiation reaction problem, quadrupole formula controversy. Not surprisingly, maybe in view of what you said, that there were many different objections to various aspects of the theory. Do you, at this point, feel that there are any of these continuing reservations, if you're aware of any, are legitimate? Or do you feel that, on the whole, that most people are satisfied and that that's the reasonable stance? I don't know what you are alluding to. I have not seen what you are alluding to.

37:30 Well, here I think that probably the controversy seems to have ended publicly pretty much at the period that we're talking. But I believe that certain people continue to hold various what we could call unorthodox positions I know for instance Cooper Stonker you mentioned earlier has even produced a recent paper in which he returns to an earlier epoch of controversy concerning gravitational waves and suggests that perhaps they cannot carry energy at all referring to an old argument of Rosen for instance in the 50s regarding the energy momentum pseudotensor in the waves yes to me this does not add clarity to the discussion in any case Because after that, the field, I mean, entered, for instance, in view of, okay, because for many years, the binary pulsar has been this jewel, and then gravitational wave research has been partly motivated more by the prospect of having detectives of gravitational waves. And this is why, for instance, a lot of work now is going on, going beyond the quadrupole formula. And this was, in fact, the paper I was starting, finishing this morning, because recently, and this is work now, which is going on several places in the world, but I must say most of the techniques which can go beyond that have been done in France in collaboration with Luc Blanchet. we can improve on the quadrupole formula by going two orders in power of the V over C beyond that, and the second post-Newtonian thing. So, to me, there is no doubt. I mean, the first order of the quadrupole formula, then you can go beyond that. This is for weakly self-gravitating systems, and one still needs to match things when there are neutron stars in the business, and now I don't want to spend too much time that when you have neutron stars, you can justify all that, because I have no doubt that this is the case. So we concentrate more technically, so it's more technical work, but...

40:00 So the controversy of the quadrupole formula is old stuff, I mean, this is the zeroth order of all these things, and now one wants to refine to have these quadruple binaries, and you work on that too, no? so I mean this is not a lively controversy anymore from that point of view yeah that's more or less what I was wondering what your opinion would be on that so regarding then going on beyond the quadruple formula but by the way you always use this as something I strongly objected to you use this phrase quadruple formula which I never use myself only within quotes because I think, maybe, I don't know if it is clear to you, to me, it was clear from the beginning, a lot of the confusion of the quadruple formula controversy was that people did not understand what was meant by quadruple formula. They had vague ideas, they thought, they knew what they thought, and they were not talking about the same thing. And part of clarifying the confusion was to define clearly what you were talking about, which physical process you were talking about, what quantity you were observing, and things like that, which most people, I mean, but this is all stuff. One of the relevant problems of the past is that many people have been fascinated by this theory, but they've plunged into it and not having a clear idea that this is a physical theory, and in a physical theory you have to know at each moment what you're talking about, what you observe, and discuss only observables. Like in quantum mechanics, there is only one role in generativity. If you discuss observables, you never make mistakes. and we can make technical mistakes, but never make conceptual mistakes. And a lot, because in the meantime, for instance, when doing this work, I also went back to study the literature on radiation reaction in generativity and all that, which is an enormous literature of the years, and most of the people who had tried to do it, they had some technical obstacles, but also they had conceptual obstacles. They were not having a clear idea of what they were talking about. they were discussing only coordinate effects and things like that, and there are still people saying, but what you're talking about is not coordinate effects. No, I mean I should say that maybe at this point to clarify fully this point that in the binary pulsar you are not talking about coordinate effects and things like that

42:30 and this is at the point where most people have said okay, the protocol formula is okay, let's stop thinking about it. We continued working on it because we realized that what Joe Taylor is observing is not the motion of two neutral stars in a certain coordinate system, but what he is observing are atomic times, I mean, recordings of atomic times measured of pulse, electromagnetic pulses received from the binary pulsar on the Earth, measured by an atomic clock on the Earth, which is a sequence of proper times. And then we realized that, really, to clean up the thing, we have to make the theory also of the observations, really, of the pure observables in the binary point star. And this is something that very few people worked at. In fact, after the initial work of Blanford and Tchaikovsky, nobody else cared about that. And this is this line of work that we did alone with Nathalie de Aguero and which was finished recently with Joe Taylor, which, in fact, occupied me for a while. which turned out to be surprisingly rich, because in discussing the way the observations are made in the binary pulsar, then one can find many new things, many new tests of the strong field limit of generativity and alternative theories, and new tests that have been recently applied in the newly discovered binary pulsar PSR 1534-PERS-12 concept. So, in fact, really going back to observables was a rich thing for the binary pulsar. So from my point of view, the binary pulsar has not been only important in coming at a good moment to suggest that one should really clarify the problem of motion and radiation reaction, but going beyond that, getting new tests of strong field aspects of general relativity, which was always, for me, the important point of the binary pulsar, this strong self-gravity of the neutral star, and really the binary pulsar continues to be an inspiration for that, at least for me, although most people in the States did not work on that, without competition, in fact. People were not interested in the States and the barbershop. But you found it, as you say, a very useful test bin for strong field events, which did not previously exist to us. Yes, exactly. And this was why the... I'm saying that because you insist on the

45:00 quadrupole formula. And for me, this was always very periferic. This catch-wear of quadrupole formula was not the main point. It was already a striking proof that people had a confused mind just to call it quadrupole formula, for instance. And just looking more deeply, it gave more information about gravity. So, you spoke just now of the impact of the binary phosphor on the problem of radiation an opportunity moment for that. So you have the sense on the whole that it was, if not quite necessary for that to come along at that point, that at least it was of great help in order to allow the problem to be cleared up at that point. Okay, so even more than that, it's true that, for instance, I mean in physics, sometimes Sometimes you really need to be dragged by experiments or observations, otherwise an old problem drags along without a clear solution because it's not the best people who are attracted by this problem and then the confusion injures on and there is not the strong motivation. I mean, in this thing something is observed and you want to get numbers which are correct which is a big challenge you are not just discussing a few vague ideas and things like that, or concepts you want new concepts finally you also want to get new numbers and precise numbers and really, even from my point of view with all the interest I have conceptually in the thing, it's true that the object had not existed have done a tenth of the work I have done on it because really it was urgent. Not urgent in the sense of month because that was not the question of really getting it fast but really necessary to do the work. And one should add that in fact this has been beyond the problem of radiation reaction the binary pulsar has been a tremendous luck to generativity because the number of problems where generativity is an essential tool in the real world has been for a very long time extremely small that's why the field withered away tried out like that because there was

47:30 no incentive for an experiment and even the big discovery starting in the 60s and 70s of quasars, pulsars and all that than binary X-ray sources, did not bring that many conceptually interesting problems. They were not very clean systems. In quasars, you still don't know whether they are black holes or not. So the big renewal of generativity studies was the concept of black hole, which by itself, independently of observations, was such a tremendous conceptual thing that really it brought life to the topic for a very long time. in the end it was a bit disappointing and still now in the sense that I remember arriving in the States in 1974 people were saying already at the time Signus X1 is a confirmed candidate as a black hole it's still the best candidate but it's not more confirmed, I guess it is more confirmed but it is not proven to be really a black hole there is no direct proof NASA has been recently over optimistic about announcing that they have discovered the proof When you look at things, the data are so far from the inner part that really there is no other thing. And a lot of the studies there do not have very genuinely generativistic effects entering in many of these things. And maybe these systems are so dirty that it's difficult to point a place where really generativity is essential. So, although now gravitational radiation observations in the coming years probably will now already providing this very important incentive to do work along directions which are connected with reality, for a big block of time between 1974 and now, I mean, big 20 years, in my opinion, the binary pulsar has provided absolutely unique and very timely. and what we have been lucky this I've been lucky in the sense that I was 23 in 1974 so this was just the right moment for I think interesting problem to work on yes it's really tiny so by the impulse I is that they are present for we for whom the value process has been the most important thing to have interesting problems and really incentive otherwise otherwise I would not have done this a lot of this way.

50:00 Yeah, I see. So you touched on what I was trying to get at earlier, and so you spoke of the continuing fruitfulness of the binary pulsar data for comparing theoretical studies with the experimental data. So specifically in the problem of the radiation-reaction problem of motion arena, and you spoke of your work to move on to higher orders. Does the binary pulsar data at that point provide any further tests going beyond what you've done previously, or do you have to... Not for radiation reaction. Sure. It studies other aspects, not for gravitational radiation, but it contains still other surprises, Because I still remember the day in, I mean, I'm happy I can't remember it, but it was just a few years ago, I guess in 1990 or something like that, when I received an email of Joe Taylor saying that for the first time after 10 years of agreement between theory and experiment, there started to be two or three sigma discrepancy. of disorder, I mean, something which one was not used to, because before that, the agreement between now the finalized theory and experiment was always one sigma, a beautiful agreement, and then it started to have a two or three sigma discrepancy between the two, and Joe, with whom I had already worked at the time, contacted me because I wanted to explore with me from the theory, I mean, putting together theory and experiment if one could explain away this thing before announcing it, if there was a real discrepancy or whether it could be. So it is at any moment it can become a lively problem again. In fact, we found for this thing that there was a Newtonian effect of the acceleration of the binoculars and the galaxy, which is of the solar system and the binoculars within the galaxy very small effect, which people had thought about from the beginning, but they had estimated to be very small. But this was a mistake. In fact, this thing is not small. They were putting assumptions, concluding that it was small, but it was not right.

52:30 Thank you.