Interview with Jim Wilson

0:00 Okay, so now we actually seem to be working, and it's the 16th of March at a quarter past 10 in the morning, and I'm talking with Jim Lawson. So, well, I guess I wanted to start off. Do we need a quieter place for that, or not? Well, it can help, although this machine is better than I used to have, so it probably will still be a quarter. Oh, you can tell how much noise there is once you watch that. Yeah. So I think we're okay, but we can always go to a quieter place if you prefer. Well, I don't know where a quieter, the only way would be to ask the secretary if they saw an office. I think we're fine. Okay. Well, let me see then. I guess to start off with having explained sort of what I was interested in, I just, I know, of course, about your work recently with Grant Matthews on binary nutrient start coalescence. and your simulations of that. So I wanted to start by asking how you kind of got interested. There is another sort of public room down here I think would be much quieter. Let's go see it. Might as well stop. Well, I'll just put it on again. There we go. So now it's running OK. So I guess I was just asking you about how you got interested in the binary neutron star coalescence problem. Well, I worked for years on various parts of numerical relativity, and I had two students who got their degrees with me working on numerical calculations of gravitational radiation from collapsing actually symmetric systems. They were both from the University of Texas, but they worked with me. They came here and worked with me here. So there is a background in working in gravitational radiation. but that work was very unrelated to any possible measurements,

2:30 especially back then. It's more just what might happen type of calculational work. Sure. And then it was when one of the last students, Charles Evans, to finish his work up, he worked here as a postdoc. He was actually a postdoc at Illinois, but he kept working here, so he finished things up. I got this idea of a way to do binary neutron stars, ignoring gravitational radiation, except at a perturbation level. It's just a general idea that there must be a way to do it without the complexities of gravitational radiation because that is so weak. So this was the idea of treating the kind of equilibrium orbits? Yes, and actually Evans was most unenthusiastic about it, but he helped me get started. He's more academically oriented. He'd rather do things right, right. So I was thinking about it, and one summer, a fellow, I don't remember his name, he's sort of a well-known gravity theorist, I think he's from Florida or someplace. I think I know who you mean, but his name begins with an S, I think. Steve? Detweiler. Detweiler. Detweiler. That sounds like maybe that's it. Is he in Florida? I think so. Okay. Anyway. I think he is.

5:00 So he came here as a Stommer employee, which is a way of earning money, but he worked with us and, well, there really wasn't anything going much at that time, but he sort of showed us how to, a better way to implement conformal flatness or make it, introduce conformal flatness in a more consistent manner than we've been doing or thinking about it there before. Then, as I say, this is twelve years ago, then nothing much happened. Every year I'd work for a week or two on it when I didn't have anything else to do. Then I guess about four years ago, when it looked like the LIGO experiment would really come to being, that's fine as to say, well, let's go to work on this with Grant Matthews. We decided we should really follow through. We've just since been playing around before, just looking at the equation, doing a little numerical work, but not much. Then we began working, putting a really substantial amount of our time in on the problem and eventually a running computer model, I guess, three years ago. And since then we've been making some improvements, I guess we've made quite a number of improvements on the numerical methods and understanding of what is happening. Initially, we didn't expect to see what we saw. We thought more, what I expected more was just this would,

7:30 because these are rather finite objects, that it would just change the relation between the frequency and how the time evolution is, goes, and all that. And one thing we had planned to do was to develop a perturbation theory the gravitational radiation. Now, our calculations of gravitational radiation use this formulation that Thorne published quite a few years ago on all this multiple moment stuff. If you look at the conditions he sets forth, the early part of the thing, paper, we don't Meet those. It's the best way we have at the moment of calculating gravitational radiation, but to the extent one would want to be more quantitative, it's poor. Well, you don't know if it's poor. You don't know whether it's good or it's doubtful that it's good. So we spent a lot of time with a student, Grant Matthews had, Mariotti, trying to develop a way of doing perturbation calculation of gravitational waves off our binary system. And we never figured out how to do it properly. So basically for the late stages where you have your more strong field, fast motion, you haven't been able to do it. So there is some work that's been done by Abrahamson, York, and somebody else on a method of perturbation calculation, which is quite different from what we were looking at. It might be feasible, but in the meantime, we'd come on this increased density of the stars,

10:00 so we lost interest in trying to calculate gravitational radiation. all seem to focus now on the increased density of the stars and particularly what's of interest there besides the fact it changes the gravitational radiation probably some, but the more important we thought was the relation to gamma-ray bursts, because it gives a good source of energy for gamma-ray bursts. So the focus has been more on that the last year. Previous to that, just of interest, had the gamma-ray bursts side of things been much of a motivation, or was it only after you were sort of diverted by finding this increased central density effect? that you decided that maybe that was more interesting than the gravitational waves? Well, yes. Because there are measurements here and now for gamma-ray bursts. Gravitational radiation, that's still four years away. I mean, in a sense, one would like to predict what the gravitational radiation is. There's no reason to predict it four years ahead of time or it's not particularly any value. I think the only value is sort of what I mentioned earlier to keep the experimentalists alerted that they probably don't know exactly what they're really looking for, so they better keep their eyes open. But in a sense, that's already out there as far as we're concerned. But to actually calculate the gravitational radiation is a different issue. And we've been pushed a little by Sam Finn, who would like to work with us, if we ever got our calculations together enough to be able to say something about what we expect to see.

12:30 The difficulty there is the way we're doing our numerical calculations, we can only look at very close orbits near the end. The gravitational wave detectors become sensitive at 10 hertz and peak at around 100 or a little less. So they aren't interested in or their problems aren't related to where we're looking or our present calculations. So to get something of interest for the gravitational wave detectors, people, we need to be able to calculate the stars when they're far apart. Is the problem there simply the amount of time that you have to run the code for, for the larger orbits? Yeah, that's how one treats the gravitational field and getting enough accuracy when they're far apart. A reasonable number of zones. Because you need a bigger sort of grid? Yeah, you can't grid up everything. That would be way too expensive. Anyway, and it may not make much difference so far as the 10 hertz, I can't believe that the effects we're talking about would have any effect. I think at 100 hertz they're beginning to be appreciable, but that would probably be when you first start seeing deviations. This is my guess from the back of the envelope calculation. At 100 hertz, you start seeing deviations. So as far as experimentalists being able to detect that there is a gravitational wave from 10 to 100 hertz prior to the present system is quite adequate, so it doesn't make any difference with the results of our work.

15:00 So for instance, the increase in central density, does that start to have an effect at 100 hertz or would it be more like 500 hertz or a kilo hertz? A little effect at 100 hertz and a large effect at 500 hertz. So it just means the whole late signal will be far different. But I think that would be very exciting for those observationalists. Maybe they can detect gravity waves, and then they might learn some physics afterwards. from how the final signal looks. So if the two bodies, the two neutron stars, are spiraling into each other, and then at some point, like, say, 500 hertz or thereabouts, they were actually to undergo, say, collapse into black holes, then you would expect that to alter the subsequent signal dramatically. So would it be very... No, not the subsequent. At that point. As they start collapsing, they are giving off mass. I mean, they're losing mass by radiating neutrinos, which means also they're losing angular momentum. So the whole signal will be quite different towards the end. Towards the end, the energy loss in neutrinos can be larger than the loss by gravity waves so that the signal will be dominated by neutrino losses, but it's not gravity waves. You know, how the phasing and all that goes. So the main way that you might... So I'm sort of interested just in how we'll go about picking out the point at which, say, one or other of the collapses occurred with your gravitational wave detector. Presumably the signal afterwards is altered by the fact that the mass of the system has changed.

17:30 At the time, I guess there's going to be a memory signal, but presumably you won't necessarily expect to see that. Well, we really haven't looked at this much. As I say, we've kind of diverted from gravity waves, but say there's this chirp mass. Right. Now, the chirp mass they can find early on, I think. I presume. So if they find the chirp mass, then they know they've got a good idea of the mass of the stars. And if they're sort of equal, then you can make... and you're in good shape to interpret the late signal to the degree the late signal doesn't look like the simple point-particle signal, then there's a lot of data there to work with. Well, of course, if it does agree with the point-particle calculations, then we're wrong. But there is something to be learned there. I remember going to one meeting, one of my first ones, to a gravity wave, I think it was at Penn State, two or three years ago this summer. Yes, okay. I heard several people remark, neutron star binaries are not interesting. The only interest for them is that they are the first place we'll find gravity waves. They were considered there's no physics involved, nothing to learn. And, I mean, that was expressed several times. So, I mean, that's ingrained in the theoretical thinking, which has something to do with all this argument, I think. That's interesting. This was before. I think I gave a talk about our results there, but it was sort of a wishy-washy talk because we'd just gotten the results and I didn't really understand them. So I gave a sort of blah-y talk on purpose. Right. So then subsequently when you did start to talk about these interesting aspects of the results, like the increase in central density and so on, do you feel that the, well, you said that the reaction was sort of, you think,

20:00 influenced by this bias that the theorists in the community had that there wasn't going to be anything interesting there. Right, yeah, that they'd solved it all. That was loud and clear. Well, so before going on to that, which is an aspect that I'm interested in, the kind of interchange that followed, which I followed to some extent, the interchange that followed your results. As a graduate student, I was somewhat interested in the problem from the point of view of the signal analysis side. Yeah, yeah, yeah. So I myself was waiting for the results because I was interested to see what they were going to be. Sure. We were aware, as a matter of fact, just having been interested in the idea of signal analysis, extracting some sort of physics from the neutron star thing, we were aware that we were actually going to need simulations like this if we were going to be able to say anything. Well, I'll say this interesting thing there. Twice I talked to Thorne about this. Yeah. Well, it was about two years ago and three years ago. And I started to try and say what we were seeing. And he, both times, he sort of got off on a side issue. And I think it was influenced by what his, he says, Oh, you're doing such wonderful work. you'll be able to study the coalescence and get the equation of state of neutron star matter because we will have established the mass of these stars and if you can calculate the final coalescence, that will be wonderful. That's the physics he was interested in. See, our code, I don't think, could calculate the final coalescence. It's a rather queer situation. I think our code to physics isn't adequate. People are doing it with Newtonian and post-Newtonian, which I think is absolutely ridiculous. I don't think the results have any meaning. Yes, we were kind of interested, I think it's fair to say, in the coalescence side. So you don't think that at present there are codes that can really address... Well, there's this grand challenge to make a real 3D code with all components of the metric in it, not enforcing conformal flatness.

22:30 I think that'll be around by the time, or probably before. But that's the only hope, I think, to get physics out of that final coalescence. It's very qualitative ideas. So your code was really aimed at addressing sort of an area between the kind of early in spiral up to some neighborhood like 10 or 100 hertz, which you feel is probably reasonably well approximated by the point mass calculations. Or up to the point of the last stable orbit. That's why I think we're still pretty good at the last stable orbit. That is to say that the analytic calculations are still pretty good? No, that our calculations... Your calculations, yeah, from there, from where the analytic thing leads off up until about the last stable orbit. Yeah, right, yeah. And beyond the last stable orbit, I think we become doubtful. Right. Presumably because you're not going to have any more orbits that you can maintain or some sort of stable equilibrium. Well, in a way, and maybe we might do it, is to, say, take a binary at the last stable orbit and watch it go all the way in. We could do that. Now, I think our calculation would be far superior to anything heretofore. So it would have some value, but I don't think it would be very at all quantitative. There's a difference between being quantitative and getting a better qualitative feeling of what goes on. So I think we could do that, and we may. One of the things that we had been interested in that I and a couple other people that I was working with as a graduate student had been interested in was this question of where are the last stable orbit onsets in the case of, say, more or less equal mass neutron star binaries. And, I mean, there had been various estimates of where that point might lie up to that point,

25:00 but we were interested in more reliable figures based on that, and that's something that your code clearly has addressed. So I was interested. I know that the increase in the central density of the objects was regarded as a surprise. Probably, it's fair to say, a big surprise by a lot of the viewers. has there been a similar reaction to your results on the last stable circuit orbit the last stable orbit, that's a queer one because with point particle calculations you don't know what M over R really means it's just an expansion parameter when you get in close so it's saying the last stable orbit is in such a radius I don't think it means anything Yes, there is such. There have been some post-Newtonian calculations, I think, looking at it. I think it's a better way. I think it's too fuzzy a concept to be worrying a great deal about what it means and whether our calculations are very good. We came, I thought, close enough to the point particle calculations that I didn't think there was any significant difference. I wouldn't attribute anything to the difference. There's a relatively modest correction. Yeah. And in fact, you know, the question of what's the distance? How do you measure the distance between two stars? Well, we quoted measuring the proper distance, which is something we can calculate, but it's sort of an odd thing, too. something that might be better to talk about is say the frequency at the last stable orbit which of course would be something more measurable and it's also a better concept so far as being able to extract it from calculations Sure.

27:30 Sure. Yeah, and as you say, it's more of an interest from the point of view of a detector anyway, and if something interesting shows up in the signal, for instance, at that point, then you'd know about it in terms of its frequencies. Well, to go back then for a minute to the increase in the central density effect. Not having ever worked in numerical relativity myself, or not having done much numerical work, I was sort of interested in what you mentioned earlier about the fact that the result was not, in that regard, was not what you expected. And I was curious about, during the whole process of finalizing the code and presumably debugging the code, But is there a sense in which you have kind of a struggle between results that you don't expect to know whether they're some sort of artifact of the numerical side as opposed to something physical that's happening, or is it somehow different from that? Well, when one gets a result that's unexpected, One goes back and reviews how one is doing the calculations, of course, and tries to understand it. I think one thing that we sort of have held fast to through all this discussion is the most arguments seem to be of our critics are based on tidal effects. And our calculations clearly include tidal effects. I have no ifs, ands, or doubts about that. I think that is irrelevant, that the tidal effects, which are stabilizing, is far weaker. As we're talking about a much stronger effect, it has to do with rotation.

30:00 Now, but it is a very subtle effect because it's not so simple as tidal forces which almost Let's do it back of the envelope in a crude way to see, yes, it's stabilizing. We have pointed out several terms in the equations that will lead to enhancing the compression, but the way they're involved in the final equations, which are so extremely non-linear, it's hard to really understand them. By any simple calculation, I think. We could be wrong, but it has nothing to do with our critics' criticisms. are irrelevant. That's the main key, is those critics are all off on other simple-minded calculations of tidal forces for the most part. So they're not addressing anyway at all the reality of our calculations. And I think that's what gets us upset. And it has the bad effect that we've talked about gamma-ray bursts being relevant to our model and we're always dismissed by the statement. Caltech has shown your calculations are incorrect. Forget. We don't believe your gamma-ray burst model. Well, see, it's the average astrophysicist, he has to, he can't read our papers. He has to depend on it being accepted by the community of others. Then he, so I don't blame these people for saying they don't believe us. Right. Because they have to, majority has to rule in this case.

32:30 as far as their opinions go, so it is a bit upsetting in a way. So what means do you see of sort of escaping from the dilemma that, as you say, the community that you'd like to address, for instance, gamma-ray people really can't sort of judge your work on their own independent assessment because, as you say, they lack the expertise to do so. Yeah, they can't. Yeah, so they have to go by the reaction of people. In a way, we feel that very few of our critics have read our paper. The way they write their papers, we don't think they've read them. We think they've read the abstract, the introduction, and the conclusions. And haven't tried to understand what we've done. In fact, one of them admitted he'd never read our paper except the abstract. I mean, at least he was honest. So it's sort of a strange business. Of course, I guess that's why you're looking into it. You realize it's a little strange. Sure, it's interesting. Yeah. Sure, it's interesting from my point of view for that reason. Well, for instance, I suppose one possible way out of this memo would be sort of replication of the results. Do you see that as a possible... Oh, yeah, that'll come along, but I think that we need a full general relativistic calculation that doesn't assume no gravitational radiation. Well, we can put gravitational radiation at the quadrupole level, I mean, do a real full GR, and there is the Grand Challenge, which is dedicated to doing that, but I think it'll be several years. I mean, it's all going to come out in the wash. But in the meantime, as you say, you have the difficulty that it's difficult to get a hearing for the physics side. Yeah. Yeah. Well, I brought a thing on our gamma ray burst model that you might find interesting. This is something we just finished last week, last Friday.

35:00 This is our rebuttal to our critics. Oh, I'd be very interested. So you would like that. Thank you very much. The only thing wrong is this is an awfully thick paper. I don't know if anybody would read it. Sure, well, I guess that's a problem, obviously. Yeah, that's a problem. Once things are sort of settled in people's minds, you can't, there's so much in the literature, you can't go around reading everything. Certainly no astrophysicist who might be interested in gamma rays are going to read this. Right. And that isn't something that I'm interested in, about getting an audience, especially, as you say, in a circumstance where maybe people have already made up their minds. When I talked to Grant Matthews at Notre Dame... That was quite a while ago. Last summer, was it? It was October. Oh, not so long ago. About six months or so. And at the time, he mentioned, certainly, that one of the difficulties was that referees for your paper were liable to keep harking back to the papers of your critics, such as from Caltech and others, and say, well, we want you to address such and such an issue so that you were sort of having to keep going backwards, too. Well, see, we do have a paper that is probably what inspired it, which we submitted to FISRAV letters I think probably early last fall or May and September and one referee accepted it, the other said we had to give a section which addressed the points of our critics so that the work would be believable well we were sort of bothered by that first and we realized fellow's correct. You know, when there's so much doubt out there, what's the meaning of the next paper? So that's, so then we started work on this. But this is a very difficult paper to write. It went round and round and it's gone many changes. Well, it did go out

37:30 to a referee and came back with a lot of criticism, but most of it was good criticism. He wasn't tossing it out. So this is a revised version that we just finished, which I think is far better than our first one. It's good to have a critic criticize it. Anyway, I suspect you caught Grant at a rather sensitive moment. Yes, I just happened to coincide with that. So, yeah, I sort of bring it up because one of the things that I was interested in was one might naively tend to think that maybe it doesn't matter so much anyway nowadays about the whole refereeing process since you can get your paper out, you know, maybe through e-print archives like the one at Las Almas. But as you say, perhaps the much more important problem is that people just don't have time to read all of it anyway. so just having a channel doesn't necessarily mean that yeah being published means nothing especially to me because it means something to Grant because being in his position they count the number of publications it's far more serious for Grant and his career depends on it but for me fact that I would like to have some interest shown by astrophysicists who, I say, depend on having some words from the gravitational community that may be of interest, not a present situation where it's just been thrown out. So just to finish off that subject about, you know, how the channel by which maybe, for instance, you can reach the astrophysicist

40:00 another possible channel, I suppose, would be sort of conferences and so on where... Well, yes and no, see. Two invited talks at APS meetings have been given on why we're wrong. See, it's the old boys' club business. Nobody will invite us to give a talk. See, we can give the little 10-minute talks, in fact, Grant has. But those usually may or may not be well attended. Of course, they're very brief. They're too brief for an astrophysicist to listen and understand whether he wants to believe us. Of course, I haven't given any of these talks, so I don't know if Grant has. So, in fact, conferences are actually more of an avenue through which the sort of official verdict from the gravitational community can reach the astrophysics community. Yeah. So I want to, to the extent it's, okay, this is the real world, in fact, I have a son-in-law who's in the academic world. He just laughed at me. He says, you've got to learn the real world. See, I've been working at the lab most of my life and only got interested. More on retirement and doing this kind of stuff. True. So he had no sympathy for me. He said this happens all the time. Yeah. That's the way the world is. the academic world runs. In a sense, it all comes out in the wash eventually, but... Sure. But it's a bit of a weight, especially... Yeah. That's the scientific method, I guess, is foolproof in the end. The... Well, so to get back then from that subject to something that you mentioned earlier

42:30 about how you felt that the criticisms that have been made haven't really been addressing your result at all, really. And I was curious about that, because I, just having talked to a few of the people who are involved one way or another, did get a sort of impression that people were talking past each other to a certain extent. Right. And I was curious if you had any ideas as to the reasons for this. I think it's personal. This last time, I think, Grant went down to Caltech to talk to that group there. And he'd been sort of asked to, and Grant called me up and asked that I'd come down with him. I said, nothing's going to happen there. They're just going to talk about what their view is. They will just talk among themselves. He says, oh, no. He says, I'll tell them what our position is, blah, blah, blah, blah. Well, the day after he got back, he says, you were right. They didn't listen to one word I said. They just really talked among themselves. They addressed me, but they were really talking among themselves of why they were right. You see, that's, when you get down to these personal things, science has nothing to do with it. listen. A lot of people I don't listen to. So it's this. As you say, it's going this way. I wouldn't care what the gravity folks think except that I have more interest in this astrophysics and it was a different issue altogether. But you mentioned while the gravitational theorists have largely been sort of negative in their reaction, the experimentalists have been a bit more positive. But perhaps... But for a very different reason. They're more pragmatic. Right. I think they weren't making a judgment on whether I was right or wrong.

45:00 They were making a judgment that they don't trust theorists in general when the theorists talk as if they know what they're doing about something complicated. Right, so for them it was a salutary reminder that things may not be all sorted out yet. Right, yes. But at the same time, presumably, because they're still a few years away from any actual detections, it's not that critical of an issue for them. Right. In fact, what I said, we aren't going to follow really what they're suggesting. We have to follow to a certain degree what they're suggesting, but we know better than to follow it all down the line. I was curious as to whether one of the reasons why some of the critics, for instance at Contact especially, have been focusing on questions like title attractions or post-Newtonian parameters is connected with the fact that they're trying to translate the results from your work which is based in numerical relativity into the kind of language that that they're familiar with from a more analytic approach and I was wondering if there is a sort of a certain level of incomprehension between the type of concepts and approaches that you have on the analytic side and that on the numerical side I mean would you be inclined to think that that's part of the barrier there or is it just I mean is I would say one thing that sort of got us in trouble was we wrote an FJ article at the end. We showed how velocity terms would come into the potential and strengthen the potential. And then we said that one could do, this effect might show up if you did the hydrodynamics properly. Now, to do the hydrodynamic, there's no reason to use post-Newtonian hydrodynamics in a post-Newtonian theory. That is one thing I've really found awful out in the community. The concept that you do post-Newtonianism,

47:30 you expand the hydrodynamic equations as well as the gravity. I mean, they're totally unrelated. See, particle physicists would never think of connecting gravity and special relativity. You know, they do their special relativity properly. So the hydrodynamic equations are really just doing special relativity in a curved coordinate. Right. So you would keep that exact? Yeah. In fact, we talk sometimes of could we make our code post-Newtonian by field calculations and just do the same hydrodynamics and see if we get our same results. I think we would, but that's just, I'm not sure. It may or may not, but I think you can do the hydrodynamics correctly in the relativistic sense without doing the field calculations correctly. They're really disconnected. and my reading and what's published on post-Newtonianism I think they're stuck in a rut so but you think that in principle it would be quite possible to modify the code so that you only head up to a certain post-Newtonian level of approximation on the field equation post-Newtonian people should think They're thinking on post-Newtonian field calculation that's completely separate from post-Newtonian hydrodynamics. Actually, the post-Newtonian hydrodynamic equations are more complex than the full relativistic ones. I don't know if you ever looked at them. They're awful. Yeah, well, I see. Yeah, they get pretty grungy. Yeah. So there is a certain difference in, or there seems to you to be a difference between this approach amongst the gravitational people, that they're inclined to just take everything and do it to a certain post-itone level, and what somebody, for instance, from a particle

50:00 physics background would do. Do you have any sense as to why the gravitational field is different in this respect? I mean, do the gravitational people are different? I think what it is is probably historical. When people first started thinking of post-Newtonian expansions for relativity, they were just thinking of it as sort of an academic exercise. So it would just expand everything. And I think that sort of got stuck in their thinking. So it might be an artifact in the fact that historically there wasn't much experimental input. Right, so I think it's more historical. And I may be doing some of them a disservice to. I don't know what they're really thinking. Now, I know from reading some of the papers and from talking to Grant that there was a certain amount of misunderstanding about a paper that Alan Wiseman wrote on trying to identify if this effect showed up at the first post-Newtonian level. And I remember reading Alan's paper that he said that in one of your papers, I don't know if it was the Apjay paper that you spoke of, that you had said that, oh, that this should show up at the first post-Newtonian level. And then he went ahead and sort of calculated to that. I said, no, I don't see it. And I know talking to Grant that he said that he was saying that Alan had sort of gotten the wrong in the stick as to, I think Grant was saying that maybe it would show up at the second person, but not at the first. But I forget now. Well, Grant went through Weissman's paper and saw he'd done incorrectly what we think is an important part of the problem,

52:30 how the velocity terms are, how the hydrodynamics is treated. He wrote to Weissman and told him he'd left some terms out that should be put in, but Weissman didn't change it. Now, there's something a little strange going on there. It would have been easy to change. Why did Weissman not want to add more higher order terms at that point? Now you get down to psychological problems, not physics, I think. I sort of suspect he got a first result by leaving out some terms, and it was the result he wanted, so he went ahead. And part of this may be my fault. See, I went to this conference about a year ago on gravity wave detectors. At first, Don Lye had written a paper saying we were wrong, and then Lye and Weissman had Well, at a coffee break, thinking as I probably should have, I started making remarks about Caltech, how they didn't understand what was going on, blah, blah, blah. It turned out the people who were from Caltech Theory Group, and Weissman was there later, but the ones I was talking to were Brady and somebody else who's written a paper. Oh, Scott Hughes, yeah. So I suspect I'm a villain, so far as Caltech is concerned, or an inspiration to their work.

55:00 well it's interesting the psychological aspects you mentioned and you alluded to the possibility that of course when you get to a point where you get the result that you expect there's always a temptation to stop at this point and I know that when I was in the Caltech group which is only really going back as far as when your results first were reported I mean, we were obviously surprised, and I've sort of tried to analyze in my own head what the psychological response, because one of the peculiar aspects was that I felt that we were not only kind of surprised, but disappointed. You know, we were thinking, oh, this might make it much harder, for instance, to get out interesting physics, funny enough. And after a while, I realized, well, it's hard to say why one would think that. I mean, personally, as you say, for instance, there might be quite a lot of interesting physics that might come out as a result of this or so. But the mere fact that the result is unexpected and therefore you're not inclined to believe it somehow seemed to play into the fact that you didn't think that it would be. Well, I saw a queer thing that goes on, too. I've talked to three different gravity theorists who are very independent of all this. And they all three gave me sort of a queer rundown. They said, Thorn has never looked at strong gravity. He has a prejudice against anything in strong gravity. All of his work has been in perturbation theory, gravitational waves, so on and so forth. And they say he's probably afraid of it because he doesn't have... He's a great expertise on gravity waves and things around close to black holes and so on. You know, he is, in some sense, Mr. Gravity.

57:30 But that's sort of a dangerous place when there's parts you don't understand. Right. Which I thought was a rather interesting comment, see. Again, they're giving the psychological part. Yeah. That this is a place that Thorne doesn't understand what's going on. So he'd like to sweep it all away. Right, yeah, I see. But as you say, there does seem to be a strong psychological component to the reaction. I like to tell people we may be wrong, but it has nothing to do with our critics. or what our critics are doing has nothing to no relevance to our right it's all sort of a psychological game in a way so you think anyway that the increasing central density effect might show up in a post-entonian calculation if one went to a sufficient distance like for instance Well, I think it might turn up at first, if I had to bet $10, but not more, I would bet the first, if you did the field calculations at first order post-Newtonian and did the hydrodynamics full up relativistically, I think it'd show up. Now, getting back to the question of the validity of the criticisms that have been made, when I talked to Grant, he said that he drew a sort of distinction between what the Caltech group had been saying and what the Illinois group and Cornell and Illinois group had been saying. Well, see, there the disagreement is much more subtle. Well, there the problem is they're doing a different problem. Unfortunately, a lot of people said, well, Cornell people have also proved you wrong.

1:00:00 See, but they don't understand that they are not doing hydrodynamics at all. they're putting in a fixed motion and seeing what comes out, which I think is good work. It's just misinterpreted. I have talked to Shapiro, and he says, yeah, and so on. But even there, he doesn't deny that they're doing a much different problem. but then he who I think knows gravity pretty well sort of has some doubts if we're correct because of all the other people so even there it has its effect I don't it would be interesting to talk to Tukowski because I think he's a pretty sharp character he knows what's going on I don't know what his opinion would be. So if I could get a picture then of where the effect is sort of arising, Shapiro, Tchaikovsky so on there at work has none of the rotations so they have a kind of a locked, tidally locked system going around. And as you say, the tidal effects actually tend to stabilize the central density. But then when you have a non-tidally locked system coming in and when you include at least first post-Newtonian gravitational field terms, then some sort of rotations are set up within each body. Right. Well, see, what we see is that the non-rotating or almost non-rotating stars the most stable configuration, because that's a key point.

1:02:30 Now, whether that would show, how that would show up in post-Newtonian calculations with the hydrodynamics done correctly I don't know I suspect say Weissman would have set up a co-rotating neutron star system that's the simplest one to set up so he could have put in a velocity term and showed it and seen it didn't make any difference And then we move. I don't know, you know, what he went through or how he pursued the whole business. So essentially then, both the Shapiro group and the Caltech group are missing this key point of the rotation that are set out for perhaps different reasons. See, the theme we've presented in this paper, all the calculations our people have, critics have done, have been corrected by their assumptions, but they've left out too much. whether that goes over or not I don't know my criticism is rather low key one thing that I wanted to bring up it may not be something that you've given any thought to but you mentioned that at the moment you're interested more directed towards the gamma ray burst mod and that in some sense the gravitational wave situation doesn't seem so critical after all the detectors won't be ready for some time it may not be relevant to the detectors anyway except after they find one to analyze the late time maybe so yeah so that would be the thing

1:05:00 I was going to bring up I mean, because, as you say, these effects would show up relatively high in the frequency, it wouldn't serve to hinder attempts of actual detection because that would be in the 100 hertz range or so. So you don't think, so I suppose that was going to be the basic question. I mean, you don't see any way in which the fact that the theoretical community rejected the result for the minute, it would seem. And the fact that the theoretical community in this instance, perhaps unusually, is sort of involved with the analysis side, at least proposing methods of analysis, you don't really see any problem that the detection effort might actually be hindered because important effects wouldn't be included. As you say, that would only come in once you were trying to analyze the same. Well, we do want to work on the gravitational radiation business, but as I say, I think to make any sense, we have to figure out how to do good numerical calculations of stars that are far apart, which doesn't fit very well the scheme used so far. Because we should connect into the—it'd be nice if we could connect into the post-Newtonian say at 100 hertz, where we think our corrections, so to speak, are small. This is sort of a non-trivial effort, so I think we need to do it. working around not doing it but there are other things that we dealt with first

1:07:30 essentially it was I'm just curious about that aspect, because it seems to me that one of the interesting things about this field is that the theorists are so heavily involved. So even if, as you say, the experimentalists apparently have a healthy skepticism with regard to what the theorists are saying, but it does seem as if it won't entirely be the case anyway that once the detectors are online, the theorists will become irrelevant. I mean, they do seem to have some sort of continuing role on the data analysis side. I mean, especially groups like the tech group and stuff. Yeah, it might be a problem with the theorists. To the extent they really think about it, they realize maybe calculations really aren't very important or don't have quite the importance. See, they've gone up to the seventh half, the seventh power of V over C. Right. I remember a couple of years ago talking to Bob Wagner over at Stanford, he and a couple of others were the originators of how to do the post-Newtonian point particle business. How did he get upset? He says all that stuff is nonsense, he says. once they go beyond maybe the two and a half post-Newtonian he says that's just garbage there's no reason in the world to believe it converges with such non-linear equations he went on and on he was upset that these people were pushing those as absolute reality the, apart from any structure of the stars or anything like that. So he was a non-believer. Oh, and I got sort of the same thing from Jim Anderson, too. Yes, I've sort of heard that from him.

1:10:00 He was one of the early post-Newtonian, you know, when they were really trying to figure Yes, I know Kip has been very anxious to persuade Jim Anderson, for instance, to go further in the present time of the end, but he's been very reluctant because he doesn't see that it's necessarily going to be meaningful. Right. Well, as I say, these two critics think they've gone far too far already. Well, Wagner said the only reason to go to two-and-a-half is that gives you the radiation. So far as really what the orbit is doing, he says, I wouldn't believe the two very much. Well, it's how you use it, too. I think what he's talking about is using it for a close binary system, not, you know, probably the very high orders post-Newtonian makes sense for the neutron star binaries In the sort of 10 to 100 hertz, right? You mean? No, I mean. Oh, I see. A few days. For the binary pulse. Yeah. A few days, yeah. I was wondering about. Yeah. Sure, for instance, the work such as that done by Thiebaud. Yeah. Yeah, I see. I'll be interested to read this paper on the gamma reverse model because this is something that I don't know that much about one thing I've heard people say is that in the coalescence of the two of say two neutron stars in the original model, orthodox model to let the two neutron stars come together and merge and then presumably then form maybe a black hole. I heard somebody comment recently that they thought that for the idea of gamma-ray bursts that there might be some problem with enough of the energy escaping from the black hole to account for the great luminosity of gamma-ray bursts.

1:12:30 I was wondering if... Well, I've heard some of those talks, and I think they're done yet. I mean, neutrinos don't escape from black holes. There's some calculations were done, Newtonian, post-Newtonian, on the final coalescence, and I think they got an important qualitative number out of it. I mean, that the coalescence is such a fast process, even Newtonian, that you can't extract enough energy for a gamma-ray burst, as if you put some limits on sigma T4 and times the time and you know the radius, you can't get anything out or you only get very weak. As you're down to millisecond, timescales are much smaller. Are those calculations you feel not likely to be applicable in the more As I said, those calculations, doing Newtonian in such extreme, strong field situations, I think it's sort of meaningless, except to order of magnitude timescale, which is, of course, you can just look at how fast things are going to spiral in and come up with the time state. So it has to be that what one would have to find, and some people talked about this, is that this was a model that as the stars came in, they'd pull each other apart, and they'd

1:15:00 go in to form a black hole, but they'd leave a ring of substantial amount of material rotating around, which would then form a disk to, and then that disk as it came in would be a source of energy. So, in a sense, that could make sense, but then to do post-Newtonian or Newtonian calculations So look at that. I guess that's where I get sort of there. The calculations aren't even worthwhile for a qualitative answer, I don't think. So in the case where... I mean, the scenario could be correct, but I wouldn't believe the calculations have any relevance to whether the scenario is correct or not. Now, in the case where the bodies might undergo a collapse individually before some sort of merger, what would be the mechanism generating the gamma ray burst in that instance? Well, you won't have disks. It's a picture in here. I'll show you. See, this is the time, so the compression starts several seconds earlier, so this is the luminosity neutrino, so they were emitting neutrinos at quite a rate for many seconds beforehand, an appreciable wrap for 10 or 20 seconds, so that, and here are curves of, these are results from binary calculations, this is the energy of compression that's available for neutrinos for a single star, but then this is U squared, which is the three

1:17:30 velocity, magnitude squared of the three, of the spatial part of the four velocity. but these terminate these terminate by forming black holes so that's really what sets the energy is how much compression occurs before a black hole now see a stiff star or a massive star forms a black hole rather early so you don't get much energy out So the lightest star gives the most energy, which is the reverse of what might first think. But in a sense, it takes a long time to go from here to there, just because of the gravitational This is one of our important results, the photon number spectrum that gets 200 kilovolts, which is sort of a very typical peak energy for the spectrum and also they see a lot more high energy photons than you expect from a thermal distribution. This is just a thermal distribution, so it cuts off very quickly. And this is the energy as a function of time, and you can find we had one that somebody sent us which just plopped, an environmental one just plopped right down. Of course, there's one thing about gamma-ray bursts, there's an enormous variety. Everyone's different. A typical one is a little problem unless you have some way that the model can have a lot

1:20:00 of variation. Explaining a small number is nothing if you can't explain, if there isn't something in the model that will produce variations, so we've got to reproduce what you call the average gamma reverse very well but it's the variance that's the real problem the gamma is it a producer just associated with the neutrino production? yeah they do So the neutrinos making electron-positron pairs, and then the pairs annihilating and eventually the whole system expanding and releasing the photons. I was just curious because, again, this is a topic that I know nothing about. I think Grant had mentioned to me that one of the things he was interested in with regard to neutron star binaries, binary coalescences, was the possibility of nucleosynthesis of heavy elements in them. Is that something that would be radically changed by the neutron stars collapsing? Let me say, I'm not very interested in that, but it's more Grant's background. He started out as a nuclear physicist, and he spent a lot of time... In fact, he works about half his research on nuclear astrophysics still. idea of explaining R process stuff and P processes of it. Is there some way of ejecting material out of this binary that would produce different elements or something? I'm not very interested because I can't think of any reason any material would be ejected to speak of. As these stars are cooling, they're going to ablate some material off the star. That's one of the key points of this,

1:22:30 is how much material comes off. Actually, that's what I'm spending most of the time working on now, is trying to calculate how much material will be ablated off the stars. But I'm interested more in how it affects the gamma-ray burst rather than… But it would come off under very hot circumstances, come off with a very high neutron component. But if you just throw a bunch of neutrons even out into the interstellar medium, what are they going to do? They'll find hydrants. Well, most of them will, of course, decay to protons anyways for a short time. I can't see how they... They have to run into something, some already heavy elements or something in a short time. Another thing I was curious about asking you was, we talked a little bit about, you know, how one goes about relating the numerical results with the kind of expectations that analytic people have, like, you know, where does this show up in a post-intentional expansion and that kind of thing. I was curious, since we mentioned also the issue of sort of replication of the result by future code that may come along, for instance, from the Grand Challenge people. If you have any experience or any sort of sense of what happens in the instance where two codes disagree on the results, where they're actually treating the same issue as opposed to not quite the same issue, how difficult or maybe easy a problem is it comparing across between the two codes? Very, very difficult. See, I worked for years in the supernova problem, and different groups are always getting different answers. Efforts are made, as they should be, to try and make comparisons, but codes are written

1:25:00 in such different manners it's never pinned down you could say I did such and such and you'll say I did such and such and so on but how doing that different really affects the final answer you don't know I mean you know you've done you haven't solved the same you haven't solved the problem the same way or you haven't solved the same problem in a certain sense. But what the meaning of that is, when the calculations get very complicated, is, well, no. As if you understood each little effect so well, you'd know what to do. In these complex systems, you don't really understand how important details are leaving this out or putting it in is crucial one thing that somebody mentioned to me who I talked to was that in some areas one of the problems that people had was with certain hydrodynamic codes I forget now this person was saying one of the problems was that people were quite secretive about code and that therefore this presented a problem because if they got different results they were a little bit anxious of revealing details of the code because it was sort of something that you'd invested so much time in. Is that something that has risen in your experience? I don't think there's much secrecy. And it's a problem of being forthright about all the test problems one has done, all the checks and so on, and how well they do. but I'd be willing to send a listing of our code to anybody who wanted it,

1:27:30 but that would be a challenge for them to find out anything from that. It would always be easier to start to make their own. I really want it, sure, yeah. Actually, one of the difficult problems of working on complicated codes like this is work on them for a while and go off and do something else and then come back. It's even hard for me to find my way through if I haven't looked at it for six months or a year. Right. Yeah, I can imagine that must be the case, even on the run of the very simple codes that I worked with. Yeah. You've got to remember all the interconnections of this, that, and the other thing. Does some of the issues that you spoke of there regarding the difficulties of cross-comparing between different codes with presumably different approaches also has some relevance to the problem of comparing between people taking more analytic approach and people and the results of a numerical work. I mean, you were saying it's difficult to know within a code which of the various details is actually the strong one so that it may be difficult also to pin down. I mean, the impressions I've got with the Caltech group is that they want somebody to say to them, well, look, in your terms, here's where this thing shows up. Well, it would be nice, say, if one could make a definite statement about here's where it shows up, and we seriously consider just writing a post-Newtonian code, first order post-Newtonian, with, in my hypothesis, if we do the hydrodynamics, so to speak, I think it'll show up but I'm not sure

1:30:00 if that was done that might help a lot so many people have done post-Newtonian and sort of feel happy with it and to the extent it would show up they probably feel they understand it better it's not a black box it's not as much a black box So in principle, at least, it's possible to, with a considerable amount of extra work, but it's possible in principle, at any rate, to go back and sort of restructure the code so that you're putting things in terms of analytic constructs, like the post-intention. Yeah, well, it's not quite an analytic concept. Yeah, I don't mean that you'd be dealing with it analytically. It's just that my sense of it is... Yeah, it'd just be a matter of doing things in a manner that people feel more familiar with. as it is the problem that people's thinking is primarily Newtonian. Then they think of all relativity as adding a little bit here and a little bit there. That's the real thinking. And it's sort of odd because black holes are such extreme, but they're so extreme that there's a black hole here and there's sort of everything else there. You work out black holes, which are the most extreme, but in some sense, they're sort of simple. What's in between is where people don't understand it, but there isn't much in between. These neutron star binders are probably as close to something in between as there is.

1:32:30 You know, enormous work has been done on perturbations of black holes. So you do a lot of that, and then you do a lot of perturbations of Newtonian physics out here. You know, that's post-Newtonian or perturbation theory or something. Right. Sure. Yeah, so as the ANAT, I think that's what I was sort of trying to get at that. Now, I was curious if there have been, I mean, the sort of criticisms that I'm familiar with are those from the Caltech group and from Shapiro and Tarkovsky group. Has that been pretty much it as regards critical response? Well, there's been several others, which are some lesser ones, they're listed here. There's four through thirteen or something way down here, so there's all those. These Japanese ones are sort of mixed, they're so complicated to understand what's going on there and whether they're seeing this or not, so that's the lesson. Flanagan, his is a kind of expansion which is related to what Thorne did, but he does it in far more detail, but answering, criticizing his work was tough. but I think I think some of the assumptions are bad he really ends up he's really just looking at tidal effects

1:35:00 I have looked at all of these grand tests. So I suppose that's sort of an inherent problem, that sort of unevenness of the relation that is more or less necessary from your point of view to actually carefully analyze the critical response and to rebut them, but that it's not so much necessary for other people to, they don't have to feel like they have to... Yeah, nobody's rebutting them, they're all just following the bandwagon. See, Flanagan's work is so hard to understand what he's really doing. is our real problem. But I think he's ignored some things that are important. I think he's In the end, he's getting just principally the title effects. Actually the way the referee's report was written, we sort of think that maybe Flanagan was the referee. So, we won't be surprised if it comes back with a criticism of our criticism. Well, we'll see. So, I guess I'm curious as to what you sort of see happening next to it. Is it likely that you'll be sort of obliged to put this coat on the back burner for a while

1:37:30 until the circumstances change so that you have more of an audience for it? Or do you feel that you'll push forward looking at the gamma ray burst side of things and hope that the astrophysicists will be more interested in it? I'll probably go just where I want, but one would like to have somebody read one's papers I'm more interested in the astrophysics involved, and I think nobody's going to read the papers, which is a little disconcerting, but I think just keep plugging along is about all one can do. It is a little disturbing, see, I had an NSF grant for travel and some expenses. It's a meniscal one, by usual standards, because I retired, but it was rejected this last fall, a continuation of it. And I was told by the program director, people didn't like my work. So it hurts in a way, but for granted, see, it's more serious. Right. You know, his career depends on him. I feel more upset about his future. but he spends half his time on nuclear astrophysics where he's a good authority and has no problem but I mean there are already sort of knock on effects he would like to work more on gamma ray bursts too but you know nobody's going to read a gamma ray burst paper

1:40:00 that put any real time in on it. So, I mean, there are different practical ways in which we're sort of going against the current orthodoxy. Yeah. I've had run up against this before. I did a lot of work in nuclear physics, which was... ill-received by the community, except by two people, Hans Bethe and Jerry Brown. I don't know if you know them, but, well, Bethe's a big man on nuclear physics. Eli, but the nuclear physics people, twice I got referees who said I should just mind my own business and stay out of nuclear physics, period. I mean, that's a referee's report. I mean, these gravity people are nice. Rather than be civilized. Yeah, but actually, I'm talking to a grad about this once. Yeah, he says, that's why I got out of nuclear physics properly. He says, when I realized I could use, he got his degree in nuclear physics. I think he worked with Fowler at Caltech for a while, too, so as soon as I saw that it could be applied to astrophysics, he says, I didn't want to do any more theoretical It's a peculiar community that, I think, talks to themselves even more than gravity folks. Right. Interesting. So there's something for you to investigate someday. Right. The gravity people are not necessarily the most insular. I think they're less insular. Good. That's good to know. Which community would you sort of view yourself as coming from, sort of straight after? Which community would you sort of view yourself as coming from? Well, see, I've, since I worked on nuclear weapons engineering for 20 years or more,

1:42:30 in a way, I am an outsider. Right. See, I do what I please. and I think my style maybe I think bothers people too so because as you mentioned earlier you haven't had to be part of this sort of academic when I did get a NSF grant three years ago two of the referees made statements to the fact that I shouldn't be working on such things. It's sort of queer because they said my work was good, but then just of a sentence at the end, they didn't think I should be supported. That's right. People are very provincial to speak, I think, of what it really amounts to. As these two referees for the grant were just putting some of their afterthoughts of where they thought money should go. And it shouldn't go for a person like me. I think they should sort of keep it within the field. But I know in the rules for refereeing grants, you're not supposed to make any comments of that kind. They're considered a no-no, so the reviewer would ignore them, so. Unfortunately. Yeah. well thank you very much that's very interesting I enjoyed it very much