FacebookTwitterYoutubeInstagramGoogle Plus

Gravitational Waves: A New Era of Astronomy Begins

On September 14th, 2015, a ripple in the fabric of space, created by the violent collision of two distant black holes over a billion years ago, washed across the Earth. As it did, two laser-based detectors, 50 years in the making – one in Louisiana and the other in Washington State – momentarily twitched, confirming a century-old prediction by Albert Einstein and marking the opening of a new era in astronomy. Join some of the very scientists responsible for this most anticipated discovery of our age and see how gravitational waves will be used to explore the universe like never before.

The Kavli Prize recognizes scientists for their seminal advances in astrophysics, nanoscience, and neuroscience. The series, “The Big, the Small, and the Complex,” is sponsored by The Kavli Foundation and The Norwegian Academy of Science and Letters.

View Additional Video Information

GRAVITATIONAL WAVES-WORLD SCIENCE FESTIVAL

BRIAN GREENE, PROFESSOR, COLUMBIA: Thank you, thank you. I appreciate that. I’m Brian Greene, co-founder with Tracy. And I also want to welcome you to Day 4 of the 2016 World Science Festival. The program tonight on gravitational waves. It’s a timely program. I trust you all heard the announcement on February 11th or you read about it. The first direct detection of gravitational waves. Amazing, amazing discovery. And we have here tonight a group of the very individuals who made this discovery possible. And I’m going to bring them out in just a little bit. But just a few words at the outset. You know, this discovery just to put it in context, it’s the kind of achievement that happens maybe a few times a century in fundamental science. I mean we were thrilled some years ago for the discovery of the Higgs boson, something that was search for a very long period of time and finally was definitively discovered at CERN. Great discovery. This ranks on par with that kind of experimental observational achievement. And the other side of the story, that will come out in the conversation here tonight, is that whereas for the Higgs boson, just about everybody was convinced that the experiments would succeed.

I mean most people who were following closely the search for the Higgs boson were saying, well if it’s not found that’s the craziness. I mean this has got to work out. The theory says that all of the circumstantial evidence suggests that the particle is out there. And indeed it was found. Now with gravitational waves as we’ll see, there is never really any doubt after, oh I don’t know call it 1950, 1960, that these ripples in the fabric of space were a real feature of Einstein’s General Theory of Relativity. But there definitely was controversy as to whether we would ever have the capacity to catch one of these waves. And we’ll hear about why in just a moment. But what this means is, it takes a certain kind of internal fortitude. It takes a certain kind of vision. It takes a certain kind of stubbornness. It takes a certain kind of independence of spirit to devote your life to this kind of project. And our hats are off to the individuals, they now number roughly a thousand, who made this discovery possible. OK. A little bit of background before I bring the panelists on stage, just to set the stage scientifically. Right, so we’re all more or less familiar by this stage of the game that Albert Einstein gave us a new theory of gravity in 1915. The General Theory of Relativity. The Newtonian picture that we all learn in high school simply spoke about objects pulling on each other through space with some sort of force that was described mathematically but whose mechanism of exerting its influence was not articulated.

Albert Einstein comes along and articulates the way that this force is communicated. And just to give a picture of it, if we can bring up the first little clip here. It goes like this. Objects-bring the lights out a little if you will-like the sun or the planets. They warp the environment of space and time and it’s these warps in curves in space that are able to cause other objects to move along peculiar trajectories if you weren’t aware of this picture. The metaphor that we love to use it’s imperfect to be sure. But if you think about space and time it’s sort of like a trampoline. You put some heavy object in the middle of it. It warps the surface and then if you roll a marble, the marble goes into orbit much like a planet would much the same way the sun warps the environment. And then the planets go into the orbital trajectories that we are familiar with.

[00:04:49] GREENE: OK so Einstein gives this very beautiful new description of gravity in 1915 and then the question of course is, “Is it correct?” And indeed, there were very quickly two pieces of evidence that sealed the deal. The first was an explanation of a puzzle having to do the motion of the planet, Mercury. According to Newtonian gravity, the planet should just trace out the same orbit year after year after year. But observations show that the orbit was actually shifting a little bit. Nobody could give a good explanation of this. Einstein comes along and uses his new approach. This new picture of gravity uses the equations of general relativity and is able to spot on predict this so-called precession of the perihelion of Mercury. Now that was kind of a post diction explaining a puzzle that was already known. Observationally but then he was able to go further and make a prediction for something that had not yet been seen. And indeed, that had to do with the bending of light by the sun. So if you have a distant star, it sends light toward the earth goes in a nice trajectory of that sort. But then later on if the earth is on the other side of the sun and you now consider that star light according to Einstein’s new picture it’s going to go in a curved trajectory as it goes through the curved environment surrounding the sun. And that has the effect of making the position of the star in the sky shift.

Einstein was able to use the equations of general relativity to calculate the angle, and it was just a few years later that teams of astronomers went out and measured the angle between those two positions, taking photographic plates during a solar eclipse so you could see those distant stars compared with photographic plates taken six months earlier. And indeed, the data is a little bit of massaging a little bit of forethought a little bit of actually knowing the answer in this particular case, seemed to show that Einstein’s ideas were correct. Since then, these experiments have been done over and over. There’s no doubt any longer that Einstein’s theory is correct. Now, Einstein then went further and as we’ll discuss here tonight, he wrote a paper in 1916 in which he basically says, he didn’t use quite that language, which basically says mathematically that if space is sort of like a trampoline and it can warp and curve, well if you start to tap the trampoline, disturb the trampling you send ripples going along the trampoline. Those ripples, taking away the metaphor, we see ripples in the fabric of space and that’s what gravitational waves are. Now it’s a very interesting story that we’ll get into a little bit in just a moment with our panelists. Einstein himself, he writes this paper 1960 actually made a little mistake. And maybe a kind of weird concept, Einstein making a mistake in doing calculations in relativity, but relativity is hard. And Einstein was the first one who is actually working on these calculations. So he was making what we would now call mistakes of a beginner, novice mistakes. But he was the guy who was blazing the trail so it makes perfect sense that he might have made those kinds of errors. Anyway, he corrects the error 1918, but then he continues to have a really interesting relationship with this idea. Thinks it’s not quite right. Later papers that suggest that it’s not right.

Bottom line is, others come along develop a very beautiful systematic framework for avoiding the pitfalls that Einstein fell into. And people are absolutely certain that these things are real. What would that mean? It would mean that if you disturb the fabric of space, you should get these gravitational waves, these ripples. And the prototypical example might be something like this: rapidly orbiting neutron stars or black holes sending out this march of gravitational waves. And as we will discuss here tonight, what would that mean for somebody in the wake of these waves? They object would be stretched or squeezed according to the math. And I want to emphasize one thing. This is not to scale. And that is a big deal, because when you actually do these calculations,, as we will discuss the stretching and the squeezing is so small that it is an enormous challenge to detect these gravitational waves. And this is what the LIGO team has done. All right so now let’s get into the actual discussion.

I’m going to bring our panelists on. And let us begin with Rai Weiss, of MIT, is one of the creators of LIGO. He first proposed a large scale interferometer back in 1972 and he has been the scientific and engineering leader ever since. All right. Next up, Barry Barish. Many people say that Barry Barish of CalTech saved LIGO at the time it transitioned into big science and built the two huge detectors in Louisiana and Washington State. He joined LIGO in 1994 and was director from 1997 to 2005. Welcome. Next up, Nergis Mavalvala-I got it, and I’m so proud of myself- of MIT is an astrophysicist who is MacArthur Genius Award winner for her work on LIGO in 2010. She got her Ph.D. from MIT in 1997, postdoc at CalTech from ’97 to 2002, and then joined the faculty of MIT. Welcome. Next up, Frans Pretorius who is a professor of physics at Princeton University. His primary field of research is General Relativity,specializing in numerical solutions to Einstein’s equations, and as we will see, a vital part of this discovery. Finally David Shoemaker, who is director of the MIT LIGO lab and was the team leader of advanced LIGO and spent many years upgrading the technology of the observatories. And lead to the historic detection even before advanced LIGO was officially launched. Welcome. OK. Thank you all for joining us here tonight. Let’s just get into it. So,in the description of this discovery, many have called it, you know, one of one of the great achievements of our age. And because of that, we’re thrilled to note that a filmmaker, who has generously given us some footage, has been following you all around for a little while, Les Guthman, and we are going to begin each section of tonight’s discussion with a little video clip from a documentary that he is making, just to sort of set the scene for each of the discussions. Let’s if you will run the first clip.

(Video Clip)

NARRATOR: Take the entire universe, count every photon that’s coming from every star, and add it all up. That’s how much energy was released by the sun. It’s the most energetic astrophysical object ever observed since the Big Bang. You worry that,in any complicated experiment like this that, you may have made a discovery but you’re not sure that discovery is really true. And I mean that’s many situations in my life where that’s happened. We’ve been through a whole series of exercises to figure out what this was. And many of them, underlying them, are sucked out. We’re like, “Could we have caused this by our own ignorance or anything from accidents to, you know, malpractice. There’s this wonderful thing that goes around, you know, is there some evil genius that has laid this into us? This evil genius gets to be more and more complicated and more and more skilled as you delve into this in a deeper and deeper way. So it’s not just one evil genius, it’s a multiple group of evil geniuses. Every check we’ve done has kind of come back saying, “No, this really came from some astrophysical origin. We believe- I mean nothing in the instrument that we found so far could have done this. And nothing that we have found so far indicates that it could have been some intentional or accidental or malicious injection. And so then you start almost believing it. And that’s pretty cool. I think the remarkable thing is not so much that we saw gravitational wave. I mean that’s remarkable enough, but it isn’t the real centerpiece of this whole discovery. That centerpiece is the fact that we saw a particular source that nature seems to have given us. It’s the best source we could have imagined. It’s the source that says Einstein is right on every goddamned detail, almost. In other words this is a source that doesn’t require any other physics than general relativity. Let’s think of that.

GREENE: So we’re going to we’re going to unpack the discovery in a number of stages. Let me just begin. I recall last fall, rumors started circulating all around that you all had seen a gravitational wave. That was probably- I started to hear about this October, November or something of that sort. First of all, what was your feeling when these rumors started to get out there? You guys had not announced yet, but yet there was this growing anticipation. Was that exciting? Was it irritating? Irritating.

[00:15:07] BARRY BARISH, PHYSICIST: Why? Because if you want to make the case when you do something like this, you want to make the case right. And we weren’t ready to make it right. So it was very irritating and worrying. Because if it did come out, and we had to defend it, we weren’t ready.

NERGIS MAVALVALA, PHYSICIST: I think there was a real process, oh sorry, there’s a real process for us as well. From the time we saw the first signal to believing that it really was of astrophysical origin and not something in the detector. And-

GREENE: I mean, was your initial thought that it that it was not real?

MAVALVALA: Yeah

RAI WEISS, PHYSICIST: Of course. I mean let me say something about that-

GREENE: And you’re not being all modest now. You know it’s really how you felt.

WEISS: No. No, let me tell you how I felt. All of us. I think everybody here and if you disagree with me, raise your hand. However it looked just too good. That was the thing. No no I’m very serious about it. I mean here we’ve gone.

GREENE: You mean if someone was going to fake a signal, this is what they would do.

WEISS: Well, we’ll get to faking it in a second, but let’s just say what it looked like when you first saw it.

WEISS: I mean we’ve gone through a lot of stages expecting signals that maybe just marginal.

GREENE: Yeah.

WEISS: And we had gone and made lovely algorithms to be able to look at the data very carefully. And here is this thing that came on in the morning and you didn’t have to do much filtering. I mean you could look at it with your eye having without a lot of computing. It’s almost nothing more than if you had a radio and you had a base and the treble control and done a little adjusting of it. So you could hear a little bit better or see it a little better. And that’s all we had to do to see this waveform. That was the thing that made us all so happy. It was absolutely beautiful waveform without needing all the paraphernalia that we had developed.

GREENE: Now suspicions of what?

WEISS: That. Yeah what we- I can tell you we- every one of us had some thought. I wish you guys would speak up.

DAVID SHOEMAKER, ASTROPHYSICIST: I think, of course, we did injections. It’s something we regularly do. We inject signals that look just like, you asked, sources that we’re looking for to ensure that pipelines can actually recover the signals.

GREENE: And does everybody know about those injections? Are they a secret?

SHOEMAKER: Most of them, yes, but in our initial LIGO experiments, and we planned also for advanced LIGO to make blind injections, ones which were unknown to all but one or two or three persons who would push the buttons and set the system in motion.

SHOEMAKER: I actually knew a lot about that right at that moment because I was working with the team.

GREENE: But let me just ask you, so if one of those if one of those injections is made, is the point that if someone on the team doesn’t recognize that signal they get fired? Is that basically what it’s about?

SHOEMAKER: It’s certainly if a pipeline that should have detected that signal does not detected, then questions are asked.

GREENE: Yes. So going with what you’re saying about the-

SHOEMAKER: But good, so the- just at that time we were trying to demonstrate that we had the technical capability in this brand new instrument, advanced LIGO, to make injections successfully. And you’re actually having a lot of technical trouble, and I knew that the night before someone was trying out some code that might have been a way to get around the problems that we were having. And so my initial thought was, “Oh good. They finally succeeded in making an injection.” I went to look in the log book and it wasn’t logged there. So that’s that’s when the question really started to come to me that might be a true signal.

GREENE: And so what stage was the apparatus at? It had not actually officially been turned on, right? It was still in a very much of a testing state.

SHOEMAKER: We had just -we had just started an engineering run. We regularly hold these runs which are not to do astrophysical observation, but to run the instrument, leave that alone for a week or a couple weeks, and try and understand if we know how the noise behavior is, if it’s stable, see if it stays locked, see if the operator is able to get into and out of lock successfully. And we had just started the engineering run that was to immediately precede the observing run and we were two days into it when we woke up to this.

GREENE: And now I heard that somebody on the team maybe even on the panel had had suggested to not have the detector actually on for various work that you were doing in order to set it up for the official use of it-

BARISH: Someone was going to do a test that might have knocked it off. And went home and went to sleep instead.

WEISS: A lot of stories like that that we almost missed it because. Yeah. I mean for example, I was in Livingston and I went and we had a problem. I went to fix. And I was told people would take- I told people it would take two weeks to fix it. And I was there the Thursday before the Monday, OK, when the thing happened. So if they’d given me permission to do that we would have missed it.

GREENE: Right.

WEISS: So now there are other stories like the one that Barry just said, you know.

GREENE: Now just want to- when you say missed it just to give it again -we’re going to go into the details of the device in the conversation that goes forward. But, if I understand correctly, so this was an event that happened a long long time ago. Right? Very very far away. More than a billion light years away, right? So this collision of two black holes, which is what this is, sets off this tidal wave that ripples through space and time. And so you’re saying that it’s been traveling toward the earth for like 1.3 billion years and it just so happens that just as it’s rolling by Earth, you’ve got this device that’s just turned on and catches it.

WEISS: I’m sorry, you’re right.

GREENE: So Frans you’re not on the LIGO team even though your work is vital to the success as we will see. So what was your feeling when you began to hear these rumors?

[00:20:33] FRANS PRETORIUS, PHYSICIST: Well yes I’m not on LIGO and actually so I- the way that I first heard these rumors, these people would come up to me and say, “Well have you heard these rumors?” And I felt like I was actually one of the last people to actually hear the rumor. So it was like about two months later, there was someone said LIGO leaked, I won’t break the confidence and say who it was.

But, they sent me this email with this this waveform and the hair went up on the back of my neck, just I got shivers, it was like. Like, “Holy bleep, these things are actually out there.” It was the most astonishing thing to see that. And you know you guys I mean I’ve been studying that course for 20 years. This is the relativity is a fascinating theory. So this is my field. But still you know from a scientific perspective, it’s not just that they’re beautiful remarkable astonishing predictions of a theory. I think as scientists, we want there especially with something which is as astonishing as a black hole, we want from equivalent astonishing level of evidence. And so in some sense I was I guess I was hoping that they existed, but I was sort of agnostic and then seeing that it was you know there was a real revelation on the second thought that went through my mind is that, “No this can’t be right,” looks just like those simulations. Someone, you know, someone did a blind injection and they stupidly just took the most obvious thing that was available.

GREENE: I mean were you concerned about you know sabotage? Was that also part of the thinking?

BARISH: One of the things we spent time on was a trace. Luckily it’s two very independent laboratories. So to make this happen simultaneously limits the options but it’s something that we work very hard

GREENE: And they are in-the locations of the locations and Louisiana and in Washington state.

BARISH: So this had to happen within a few microseconds in both and look alike. That was very difficult but still worried a lot about it.

GREENE: And can I ask another. I don’t know if this is a sensitive question or not, but you know there was an announcement by a different team using a completely different approach looking for a related kind of signal, but not the same one, BICEP2, which found evidence of gravitational waves, indirectly, through its effect on the cosmic microwave background radiation. Some of you may have been here when we had a discussion of this sort, right when that announcement was made, but before it became clear that that was actually then probably not the right interpretation of the data. So that was sort of a premature announcement that, actually got a lot of people irritated, not because the announcement was of something that turned out ultimately not to be the correct description, but people felt that they’d really jumped the gun. These- this team had kind of gone to a press conference as opposed to going through the more traditional channels. Did that influence your thinking at all?

WEISS: I mean look I was- I know a lot about that. Yes. First of all, let me say something positive something negative which is very important. That team that reported that experiment had made a dramatic improvement in the technology of the detection of the polarization which is the, how the electric field points in the cosmic background radiation. The prior experiments had been 20 times worse than any experiment that those guys did.

GREENE: Yes.

WEISS: So that was a dramatic improvement. And what they did, they didn’t have enough channels. In other words, they made an interpretation they did I think

GREENE: channels  meaning frequencies.

WEISS: Well yeah, I’ll get to what it is in a second. But they didn’t have enough channels at the South Pole with the experiment. What they saw is a particular pattern in the sky of the polarization vectors and that’s I want people to ask what that is. But there’s a pattern that particular gravitational waves would make in the plasma of the primeval plasma, and that would then cause these waves that come from the electromagnetic waves, that come from the early Big Bang plasma to have a particular pattern to the polarization through swirly patterns.

GREENE: Yes.

WEISS: And they saw those swirly patterns.

GREENE: Yes.

WEISS: The trouble is that there are other things in nature that make those and that’s what you’re referring to. And they knew that. They knew that-that wasn’t that they were surprised by it. They knew that dust would cause that dust in our own galaxy and also electrons in our own galaxy, that were in magnetic fields. It was a synchrotron. So both ends of what they were looking at but they didn’t have enough detectors of different frequencies to see that. So they made a judgment based on maps that they had. I think they were honest.

GREENE: Yes.

WEISS: OK. And they made a judgment that that the dust couldn’t be as bad as to make what they saw. Sure.

GREENE: And the paper makes that –

WEISS: And the paper makes it very clear. Now where did they make a mistake? They made at least one mistake.

[00:25:24] WEISS: I almost forgive them for that mistake. I don’t feel so bad as some others. It is. I mean there is a dramatic development that in their own technology to be able to do this. And what they did then, is they they assume that the dust wasn’t bad enough to do it. And they published it without going through peer review.

GREENE: Yeah.

WEISS: And the group was too small. That’s the real problem. They all had agreed among each other I had one of my students was on that team. And they had discussed this a lot internally and they all together decided to take this gamble. So what happened is then they needed a bigger group to look at. And that’s what peer review is a very good thing to do.

GREENE: Right

WEISS: And that peer review exposed the fact that there was a good chance that maybe it was dust that they hadn’t gotten rid of

GREENE: Right.

WEISS: And that’s and that’s where it stands now. It’s still not they haven’t retracted, you know,

GREENE: No, what they saw was what they saw.

BARISH: But back to your original question. So does that kind of thing.

GREENE: Yes.

WEISS: Yes.

BARISH: I think the answer is definitely yes. Yes. We didn’t want to come out with the wrong result. Nobody does. Yes. This particular thing breaks all the rules. If you work in a laboratory, you do something you see something, then you’ll do it better. You do all the checks. This comes by in a quarter of a second. It’s gone. So we only have what we have to look at. We know you talked about, before we came on stage, the certain discovery of the Higgs. They spent a huge amount of money to have two experiments so one can check the other. It’s a way physics does things and gets the right answer. We can’t do that here. So this had to be extreme- and I’ve seen experiments that seem convincing on based on one thing and they’re wrong. So I was very very pessimistic and worried and so forth. This just was so powerful we had to-

GREENE: following up on that, you know I think an interesting question just to get out in the open here is so, Einstein let me take us through the history of this Einstein writes his paper predicting gravitation was back 1916, 1918 interesting, as I described he had a sort of complicated relationship with, the gravitational waves might be say, but why can’t we do this in the laboratory? Why is it that we’re only looking for, you know, colliding black holes or neutron stars? Can’t and why is it possible that we could have done this in a way that was closer to an experiment that we could replicate and do over and over again?

MAVALVALA: Sure, so you’re asking could we build a source of gravitational waves in the laboratory? And the answer is, emphatically, no. It’s really embedded in something that we all learn pretty early in our careers that gravity is a very weak force. And so what are the ingredients that you need for making gravitational waves, is -has so much mass compactified that a laboratory source is not possible. You can imagine taking.

So the example that we love to use is, take a take a dumbbell but make it the most ridiculous dumbbell you ever could think of which is, it’s a it’s a rod that’s a meter long. And you put a one ton mass on each end, and then you spin it at a thousand times per second, and then you go stand if you have a couple of hundred meters away from it, and ask how strong will the radiation be? And it is such a ridiculously small number, it’s like 10 to the minus 42, it’s so small that I’m almost like reluctant to say right. And so. So, no, it’s not possible and that’s why we have to go out to the universe and look for objects that have enormous amount of mass. The black holes we saw have 30 times the mass of our sun. And there, you know, at the time that they’re they’re getting close to colliding, they’re 150 kilometers apart from each other. So if this it’s ridiculous and travelling near the speed of light of half the speed of light.

GREENE: Yeah.

MAVALVALA: And those are the conditions that we need. So, you know, even our ridiculous dumbbell is far from what can be done.

GREENE: Yeah. All right so given that we now know that we’ve got to look out into the cosmos and look at extreme astrophysical bodies ,neutron stars, black holes. Now the question is, how do we get a handle on what we expect to see from Einstein’s General Theory of Relativity? And Frans that’s where I want to move on to the numerical side of the story, which is interesting. I have not read all press accounts by any means, but I haven’t seen that side of the story given as much attention as it might deserve. I think most everyone would agree and you agree with that don’t you? So I just want to spend a little bit of time on numerical relativity is what the name of the subject is and first, we’re going to begin again with a little clip that’ll set up the conversation and then I’d like you to help us take us through why computers and numerical methods are so essential. So we can run this next clip if you will.

[00:30:09] NARRATOR: The first part, the word numerical. It’s because we use computing techniques. So what happens is that the theory of gravity that Albert Einstein came up with others about 100 years ago, last year is extremely complicated set of equations. And they’re very nonlinear and that means that if you have one black hole you could solve analytically. But if I have two black holes they don’t add together. That’s a linear theory. So to solve for two, I have to solve the full Einstein equations with all its nonlinearities and that takes computational techniques. In fact it took us until 2005 to solve it. And it was Frans Pretorius from Princeton. We have to use numerical relativity simulations, which basically simulate the Einstein equations on a computer. And we build what we call templates, that are used in the model search in LIGO. The model search that found the event, used, on the order of 200,000 templates. Also we use wave models, not only to detect the signal, but also to infer the parameters, the properties. Because all the information about the source, had imprinted in the waveform is like a fingerprint. In the waveform.

GREENE: So you’re mentioned in there as the guy right? Who made this all possible. So why do we need numerical methods? Why do we need computers?

PRETORIUS: Well perhaps I can just take a step back and just say a little bit about, you know, the kind of observed that LIGO is.

GREENE: Yes.

PRETORIUS: Right it’s not like a telescope where you know you focus-look at a certain area of the sky you focus, you look, you form an image and you see what’s going on. It’s in some sense a one dimensional seismometer. In some sense, we’re listening to a sound. And so. You know you get a gravitational wave, you know it’s gravitational wave, but it’s a series of wiggles. What does it mean? It’s not giving you an image of the source. So we need solutions to the Einstein equations for various possible sources of black hole collisions that tell us what those wiggles look like. And so we can identify what’s going on. So that’s that’s one thing we need solutions to the fuel equation

GREENE: And the field equations, by the way, are have just magically appeared behind.

PRETORIUS: And look how simple they are.

GREENE: Well that’s actually part of the question that I wanted to explore. I mean, it it looks simple but, we physicists have this way of hiding complexity behind the symbols. And for this part, actually, I just want to take us through, because sometimes in these programs you know it is meant for the general person who just has an interest but we like to sometimes do a little bit of the math and if you don’t like math you can shut this part off. But let’s just do a little tag team here. So this is the form of the Einstein equation. And if you’re doing it a little bit more precisely, it has a few more symbols in it, if you’re not using the natural units that allow us to set certain terms to one. But even that, if we go a little bit further, is hiding this. So this is actually a more complete way of expressing the equations if we’re now going to impact the meaning of that first symbol Gµv. But then you look at that, and you say, “Well wait, what do all those individual symbols mean?” This is called the scalar curvature and that thing G is called the metric. And there’s this little combination of them that gives you the scalar curvature. But wait, what does that really mean? Well that’s hiding this complexity right here, because we have this way of hiding the summation symbols within what we call the Einstein convention, using indices that contract in a particular way. So that’s what that symbol means. And if you go a little bit further and ask yourself what is the first term, Rµv, That’s the Ricci tensor and it’s given by this combination of stuff, but what’s that new guy on the right hand side? That’s the Riemann curvature tensor. What is that equal to? Well that’s equal to this combination, where those gammas are known as the Christoffel symbols. And then you take those and what are they equal to? Well, they’re equal to a particular combination of the metric contract with various derivatives of it. It’s all just to tell you that this is kind of complicated. And the goal of the Einstein field equations, what all of this in some sense is about, is understanding that guy called “G” which is the metric tensor. And just tell us quickly what that is and I have a little visual that will take us through that as well.

PRETORIUS: Right, so so Einstein’s theory’s a theory about space and time, and space time is a geometric structure. And the metric tensor, in some sense, at every point in space time it’s a generalization of the Pythagorean theorem.

GREENE: So actually, you have not seen this before you but you were following exactly what I hope you’d say. So you said the Pythagorean theorem so why don’t we just put that app here for the heck of it right now. So you got two points there and take us through what you’re seeing right here.

PRETORIUS: Right. So you want to know the distance between two points. Now in general relativity, the distance between two points in space is the distance between two, what we call, events in space and time. And so what the metric does, all of Pythagoras’s theorem, we know it’s delta X squared plus delta Y squared plus delta Z squared

GREENE: Slow down, I’m trying to click, I’m trying to click and stay up with you.

[00:35:39] GREENE: There it is. All right.

PRETORIUS: So  that little formula there encodes the geometry of Euclidean space. Now you can also write it in terms of the metric and it’s a relatively simple straightforward metric. And what people in really, long before Einstein’s time when they started looking at general geometries of curved surfaces. They say well how do we describe if we now have a surface that’s got a complicated curvature? We want to generalize what Pythagoras did. This metric is a convenient mathematical way of doing it.

GREENE: And here’s one just example just so people who are following the math can go along. Here’s a curved shape if you will. And the distance between the two points is no longer just delta X squared plus delta Y squared square root, which is what we learned in junior high school, but rather is given by some unusually looking strange combination of delta X squared and delta Y square. And if you curve it a little bit differently, we learned that the distance between those two points might be given by this particular combination of delta X squared and delta Y squared. And finally, just to get to what you’re describing in the general case, we can have combinations that aren’t even just delta X squared and delta Y squared. We can have crossed terms like on the far right delta X, delta Y. And finally to get to that object called “G”, the metric tensor. This is sort of a generalized version of the Pythagorean theorem. Where are you’re going to allow the surface to be curved and that “G” thing there encodes the geometry that Frans is referring to. And Einstein gives us equations, those complicated equations that we saw before, where you can determine the shape of spacetime, if you understand the distribution of mass and energy.

PRETORIUS: Yeah yeah exactly right. And then also.

GREENE: Oh thank you.

PRETORIUS: And perhaps just another example, a historic example, where we might be able to appreciate how complicated those equations are. So the idea is we’ve- at every point. It’s basically a four by four matrix that we need to know over all of space and time. And if we know this, this metric tensor, we can ask any question about the geometry. Is there an event horizon? What do light waves do or the gravitational waves? So the first solution the exact solution was by Karl Schwarzschild in 19- basically 1915 within a few months of Einstein published the equations. And he was able to solve it exactly cause he assumed -well he said let’s do for a spherically symmetric solution. And so the matrix simplifies. He found the solution that turned out to be the first black hole solution. It wasn’t.

GREENE: So he actually beat Einstein to the punch, right? He found the first solution even before Einstein.

PRETORIUS: Einstein couldn’t solve these equations exactly. He solved sort of approximate version to compute the perihelion procession of Mercury. So it’s Karl Schwarzschild that found the first exact solution in spherical symmetry. And that was in 1915. Then if you reduce, if you relax symmetries a little bit so make it something called extremal symmetric to make it look for spinning black holes, it took people almost 40 years. So Roy Curry, 1963, found the solution for rotating black holes. So just not that people weren’t looking for in them and they weren’t trying. It’s just things very quickly get very complicated. And now if you want to understand the metric for two black holes that are colliding, there are no symmetries. Every single term is important and they just- we have absolutely no way of doing it pencil and paper with analytic methods. And thankfully we have computers, which the methods are being developed in the last several decades to solve these kinds of equations using computers.

GREENE: And what what did you do? What was your breakthrough? I mean when I was a graduate student,, you know people were always talking about numerical relativity and then there’s this stumbling block that they couldn’t get through. And now people aren’t so worried about that stumbling block.

[00:39:30] PRETORIUS: Yeah. So it’s a little technical perhaps I can just explained by analogy.

GREENE: Yeah.

PRETORIUS: So, the one thing about those field equations are very complicated, but in some sense they’re even more complicated in the sense that a feature of Einstein’s equations or that- so this is a geometry of spacetime, it’s a coordinate system. But you can use any coordinate system you want. So for example, the Cartesian coordinates system, a spherical polar coordinates system. So one physical space time, you can represent three infinitely many ways. Every single representation is a different metric tensor. Yet they all solve the same Einstein equations. So in some sense, there’s infinitely many possibilities for one spacetime that comes from these solutions. So that’s a feature- it encodes, in some sense, one aspect of the principle of relativity. The problem that if you want to say well now let’s try to solve these equations. So we start with something at an initial time.

Two black holes that are separated. And now we want to understand how do they collide. If you try to put it on a computer it’s what’s called, ill-posed. In some sense, the computer doesn’t know which one of those infinitely many solutions to solve. So the way you get around this and this was known for a long time is well you have to first choose your coordinates. OK. So we do that then but then the real nasty problem comes. That’s what people that try to do in the early 90s when it was realized OK LIGO is being funded. We need to have these templates. Let’s do the numerics and I was OK. Well let’s  do that. We choose the coordinates, we put it on a computer, and things just crashed. You know, it starts with two little black holes, they go for a little bit and then the dreaded NaN shows up. And if you know computers, not a number. The computer’s telling us it’s not getting a number and you can’t go beyond that. And it took actually people a long time to realize that because when something crashes, I mean that could be a bug, you just you didn’t program something right. It’s very difficult to really figure out what’s going on. And it really took probably, a good 10 years I think before people realize that these things called constraints in the field equations which are the problem.

So that’s one way to think of it is, let’s imagine a bowling alley. OK so we want to knock down the 10 bowling ball-that the bowling pins with a bowling ball. And so we throw the ball down the alley and let’s say we knock and spring it into the air so it’s going to go along the bowling track. So one constraint, if you will, is that the bowling ball rolls on the surface, it doesn’t jump off the surface. OK so. It doesn’t seem too difficult. But now, we’re going to get a little bit more careful in that we want to say we want to hit the center pin. We want to get all those 10 pins knocking down so our constraint is, we want that bowling ball to roll right down the center of the alley. Not off to one side or the other. And in some sense, those are the constraints of the Einstein equations where we put them on a computer, there are all these other possibilities which aren’t solutions to the equations that you can get. There’s one solution that is a solution to the Einstein equations.

But what makes Einstein equations so nefarious, and that’s the big struggle, is it’s not just a straight bowling track it’s actually convex. And so this little path that you have to get down is sort of at the top of this cylinder. And people actually didn’t really realize that was happening. So in some sense, we starting our simulations we starting the two black holes, think of it as a rolling pin, and it’s just falling off the side the whole time. I mean people realize that that really was the issue then it was. Well can we somehow- we don’t want to change the solution to the Einstein equations, the straight path to the center. But we can fiddle with the equations to change the the track on either side so we can try to bend it upwards, in a sense. And then some of these huge efforts started in the late 90s to try to find ways of changing the equation such that we can make this a well-posed problem. And so, you know, a very long story that’s quite a lot of the like a lot of technical issues under the rug. That was sort of the breakthrough, was finally coming up with a way to actually make this thing stable.

GREENE: And with that you were then able to calculate numerically with supercomputers. Presumably what shape of the wave you would anticipate receiving if certain kinds of processes happen out there in the universe. I think we’ve been seeing some of those templates. And I hear numbers like you had hundreds of thousands of templates. Why so many templates? What are these things that you need so many of them?

PRETORIUS: What was astonishing about this event, and I think no one expected this, is that you can actually see the waveform directly in the data. So nature has given us a gift which we did we didn’t even expect something as good as this. What was actually anticipated was that you wouldn’t really see such a clear signal. You’d actually just see it in the waves.

GREENE: Could you just tell us what we’re seeing up there?

PRETORIUS: Right right. So the little gray line is-that’s essentially the noise in the detector. The blue lines, that’s the inspiral of the black holes. They start out far apart.

SHOEMAKER: Let me pull back. Well it’s time along the horizontal axis, and it’s  a tenth of a second for every two clicks there. And on the vertical axis, it’s how much space is being distorted by the passing gravitational waves and the noise from the instrument isn’t actually the space time jiggling around in length. It’s electronic noise, mechanical noise, and so forth and so on from the instrument. So I’m sorry. Go ahead.

[00:45:20] PRETORIUS: – we deal with this every day. So we just get way ahead of things. OK. So as time- so these black holes are orbiting each other. They’re emitting gravitational waves. These are these stretching and squeezing of space time, which is what’s being picked up there. These waves carry energy from the binary. And so by conservation of energy, the binary gets tighter. They spiral in closer to each other, they started moving faster. So the frequency increases because they know they’re going faster and the amplitude increases. So it’s a runaway process and eventually they smash together. And that’s when you get to the very peak in the waveform. And then to these two black holes, they smashed together that forms this very distorted, rapidly the oscillating horizon. And one of the remarkable features of black holes in general relativity is something called a No-hair theorem which says that. Any isolated black hole has to be this-for this perfect object stationary doesn’t emit gravitational waves as represented by this curve solution, that I mentioned was found in 1963. And so what we see after the peak is what I call the ring down wave. So the black is essentially losing its hair, it’s shedding all this interesting structure and it’s just settling down to this perfect curved shape.

GREENE: So there’s this beautiful picture here and we’re going to now, in just a moment, go to the details of the actual detection itself. But it seems clear from the outside that the numerics was vital to interpreting and believing and be able to pick out the signal from the data. Now now the breakthrough that allowed this numerical analysis to go forward happened when? In the mid 90s?

PRETORIUS: It was 2003.

GREENE: Now you guys have been building this detector for a long time before that. So were you like prescient? Imagining that people would be able to make this breakthrough? I mean would you have built this device and not have been able to pull out the signal if somebody hadn’t come along and push the numerical methods?

BARISH: Luckily, we were we were too slow. Took us twice as long to do this as we thought it would.

GREENE: But what was your thinking, just so I can get a handle on it.

BARISH: We knew this was a problem. And it was pushed very hard that we find a solution for this particular black hole. Yes.

PRETORIUS: So I mentioned that there are other or other methods of solving the field equations approximately in different regimes. When the binaries are far apart, so they were in very slow early part of the inspiral, there are all these analytic methods called Post Newtonian or perturbant methods, where you can start getting this- the beginnings of this chart.  

GREENE: Which we’ll come to in just a moment but-

WEISS: Let me say something.

GREENE: Yes, go ahead.

WEISS: In defense of our ignorance. We worry no- to us, this business of finding a black hole the first go round was a surprise.

GREENE: Right.

WEISS: We were thinking because- I’ll tell you we could not figure out- we could not tell each other or tell an agency who was supporting us how many of these things we would find. So actually, we’re thinking of another source entirely as what we were designing the apparatus initially

GREENE: And that you had more of handle on.

WEISS: And we knew a lot more about this. OK. And that was neutron stars which are, again things that are of the source of the size of the sun. Sort of the size of Manhattan Island. You have a mass of the sun, the size of Manhattan Island. And that we that we traded on the fact that Hulse-Taylor, which is an experiment that won the Nobel Prize back in 2003. Was it that? Maybe- I’d forgotten within my life, certainly.

GREENE: And mine too.

WEISS: And what they what they saw was something, which was one of the most elegant experiments I think ever done in astronomy, was they saw two stars. One was a pulsar. That pulsar is a neutron star that emits pulses that you can see in the radio, in the radio spectrum. And they noticed that there was a modulation to the pulsar. It was going around at 50 -they had pulses every, what was it every 8? Every eighth of a second or something like- 8 times per second, really, what they saw. And then there was a modulation to that. Sometimes it would speed up, sometimes it would slow down.

GREENE: Yeah.

WEISS: Which indicated that that one pulsar was somehow living with another thing around it. And that turned out to be another pulsar or another neutron star. You just didn’t see the pulses from it. And with that, that was sort of- that business and plotting it for time, they began to realize that these two neutron stars were emitting gravitational waves. It’s the only way they could explain the change of the orbits that were using energy to gravitational waves. And with that, with that, that gave an enormous impetus to the field, our field. And in fact, in the end- I mean let me say a little about the early days because that’s sort of what you’re alluding to. In the very early days, for example when Joel Weber when he began to

GREENE: We’ll come to that in just a moment

WEISS: When he was thinking about it and when people were thinking what might he see when he did see something, they were thinking of supernova. That was the only thing. These stellar implosions that happen once every hundred years in our galaxy. And that was sort of the only thing people could see had enough acceleration. It was and it was-had enough mass that would possibly make a gravitational wave. And that was the beginning of the field. But it turned out to be quite wrong. What the estimates were for that. We still don’t really have good estimates for supernova. They’re getting better. But-and so when the neutron star binary was discovered and people began to see there were binaries of these, that became our goal for detection. And the thing that

GREENE: So let me just just just.

[00:50:49] GREENE: So in that particular case, there was indirect detection of gravitational waves by virtue of the inspiral but you guys wanted to see the direct.

WEISS: But I can just say- you asked what was the impetus for us to continue.

GREENE: Yes.

WEISS: And the impetus was really- and all our calculations were then done for a neutron star binary.

GREENE: Gotcha.

WEISS: And in fact, what we still have, when we get to design sensitivity with this detector, we expect to see something like 10, maybe 20, of these a year, of neutron star binaries. So we’re not at design center. We’ll talk about that.

GREENE: Yeah sure, we’ll get to that in just a moment.

GREENE: Anyways, so it wasn’t that the thing that everybody wanted that we should detect black holes. And I know because I think to me that’s the big source to look for. And I remember giving a talk at Cornell Hans-Peter was in the audience at the time and I told him the story that I just told you. And he said he pulled me aside later said, “Don’t you-you’re going after the wrong source.” He said “Go after black holes.” and I said, “Look, tell me how many there are.”

WEISS: He couldn’t tell me that. I mean I know he does. But he’s right. I mean if there were if there are black holes that’s better than mass and stuff like that. So any- in other words, we started the field with ignorance. There’s no doubt. But the idea that there should be gravitational waves coming from some astrophysical sources was the basic idea. We didn’t know exactly where and that was a big controversy in the beginning of the field. You know Kip, who made our, Kip Thorne who was one of the visionaries of this thing has done a lot of calculating of different classes of sources. And, in fact, got us got himself into trouble with some of the real experts in this business because they thought he was a little too optimistic.

WEISS: But anyway, look, all the things that he calculated will probably come about is my guess.

GREENE: At some point.

WEISS: At we some point, yes.

GREENE: So any thoughts on the numerical side of things?

MAVALVALA: Yes I just wanted to add something. Frans said something really important that I wanted to emphasized. So you know when the black holes or neutron stars are far enough apart, we knew what the waveform there would look like. We also knew what the waveform would look like once they crashed and a single object was forming. The part we didn’t know, and that was so important in the breakthrough, was what happens when they’re really close together and about to collide. So in some sense if we hadn’t made that breakthrough, you can think of it as we had all of those bumps and squiggles. We would have had a little gap in the middle of that time line but we wouldn’t know exactly what was going on. And it’s a pretty important region because that’s where the amplitude is the biggest. So. So there was ignorance, but we also had a handle on enough of it that we thought we could go on.

WEISS: Good point.

GREENE: All right so let’s go to the last, oh go ahead please.

SHOEMAKER: One last thing on this point, though, is the real- just an experimenter. I have to say. We built two instruments. If we see the same signal in these two widely separated instruments, but one which we can’t explain by any theory so far. If all of our checks show that it’s a valid signal, then we’ll say this is a gravitational wave, figure out what it is.

GREENE: Even if you don’t know what it is.

SHOEMAKER: And I think that’s a really important feature of the instrument. We built it in such a way that we think we can gain confidence with these two and then with others that will come online.

GREENE: So let’s turn to the instruments.

WEISS: In fact, I’d like to amplify a little on that because that’s really the interest.

GREENE: We will move on at some point.

WEISS: But but that is really- what David is saying is really the heart of the interesting part of this. I mean I mean here we’ve opened a field and we call it gravitational wave astronomy. And if we don’t have something like what David is saying, some new things that we don’t understand at all, it’s not that the field is a failure. But it’s what would defy all the prior history of advances in astronomy. Turns out every time you open some new field, you find something you didn’t know.

GREENE: Right.

WEISS: And what David said, and we’ll go back to that I hope, we have all sorts of internal tests within our own experiment to establish that we, as best as we can, that we’re looking at a gravitational wave and not something else. We should talk about that.

GREENE: Well I would like to turn to now, the actual detection of these of these waves. And we’re going to again begin with a little piece that speaks to the earliest attempts by someone named Joe Weber. Can we roll that video please?

(Video Clip)

[00:54:42] NARRATOR: This is an artifact. It’s one of the bars that Joe Weber used in the 1960s. Joe was the first person to take seriously as an experimentalist, to try to actually make a detection of gravity waves. What we have here, is a very large aluminum bar. The theory was that if a gravity wave comes by and space expands and contracts, if these were free aluminum atoms, they would just move with the local space. But they’re not, because they’re confined by bonds to other aluminum atoms. And so if you think for a moment on those bonds like little springs connecting all the atoms, when the space distorts, the atoms try to move with the space. The springs hold them back, so that puts some tension in the springs. And after the wave went by, Weber’s idea was that strain on those springs, that energy would be released as a soundwave in the bar.

Joe was a controversial character and people don’t give a lot of credence to his claims that he had actually detected gravitational waves. But to me, Joe is a hero because even though he’s a model of what not to do in data analysis, he was the person who had the chutzpah to go and make this first attempt. If it wasn’t for that pioneering work, we might well not be here. And he did think of many of the ideas that he would later cooperate into practice.

GREENE: So Rai, Barry, you were around when these experiments were going on and what’s your view of Joe Weber and this approach and the impact it had on how things developed?

WEISS: Well I mean you really want it for me. I mean

GREENE: I am happy to have it from you

WEISS: Look, in fact it was the beginning of thinking about, at least in my life, of doing it metrically, doing it a different way. I’ll tell you, I mean it’s a little bit of a story, and if you’re asking me you should stop me when I’ve said too much, okay? But the thing is, I was asked, as a young man, to do something which you can do with your eyes closed but I could not even do with my eyes wide open and with every piece of my brain working properly and that it was to teach a course in relativity. OK. I’m an experimenter and I was asked to do that. So that was in 1967, ’68 at the time when Joe not yet published, but he was- there were news all over all over the physics community that he was detecting something. But he hadn’t yet published his result. In ’69 is when he published his result. So here I am, not knowing tensor analysis not that equation, and I will love what you did there, but as you took this thing and unpacked it and it got more and more complicated. Well OK. I had to try to deal with that. OK. And with never having learnt tensor analysis. OK. So anyway, so I can’t tell my department chairman I don’t know anything. I’m not allowed to do it. So what comes up is I teach this course as best as I can and I get to this experiment. I can’t explain it. In fact, the kids in the class ask me to say we’re up to it. And I didn’t. The explanation that Fred Robb just gave I couldn’t have made that, OK, because I didn’t understand. Here I’ve been teaching a course where there is no more on gravitational force, it’s gone. It’s all in geometry. And here’s this thing that the triers are forcing at. So I said nothing. Come on.

So I sat back and I tried to figure out how to explain how you would detect a gravitational wave to people in the most simple minded way I could. I mean with the little bit least amount of that formalism as possible and that the thought occurred to me in this course where I was sort of one day with the students, if that. And I came up with the idea of the most elemental thing you could think of, which was, and this is why I tell you the story. It was like why don’t you throw some masses out there. Put them out in space and put clocks on them. Good clocks I knew about- you could make good clocks. But not as good as I needed. They were good clocks. OK. And so what happens is I say, “Let’s time a light. Light going from one clock and mass to the other one. And then use that as a basis for looking if there’s a gravitational wave or not. Because the gravitational wave as it comes down on that system will cause the time to change because the space between them is shrinking and expanding. And that’s a very straightforward way of thinking about it. Not this molecule molecular thing where you don’t really know what the molecules are up to and everything like that. And I got the kids in the class to do this, a simple problem to do in relatively, just that problem. And that put it aside and now. Now comes the connection to the question you asked me. Well OK in ’69 he writes his paper, which was a stunning paper by the way. And though it’s caused at least 20 groups around the world to do and emulate that experiment because it was such an important result. And by ’72 it was quite clear that most other people did not see anything. And the problem was not.

I can’t I can’t tell you exactly the reason why they didn’t see anything. I mean I know I can tell you that’s exactly why they didn’t see anything, because there was nothing to be seen. But I can tell you what went wrong with Joe’s experiment. I mean I had a theory about that. I don’t want to go- that’s a waste of time right now. My guess is he was not a good enough experimenter, that’s my guess. OK. I mean he probably believes too quickly what he had and he didn’t do enough checks. That’s my guess.

GREENE: Right.

WEISS: And I’m not I don’t know. Let’s leave it at that. You can ask me about that later.

GREENE: Sure.

WEISS: But on the other hand, what happens is that that triggered me to think about well maybe this crazy idea that the Gedankenexperiment that I was thinking about in that course, maybe you could make that into a practical experiment, right? Maybe because I didn’t understand the thing in the first place. Maybe that would detect gravitational waves.

WEISS: And that’s where,, at least in my life others have thought of it too, I’m not the only one by a long shot. But that’s how the thing evolved in my life, that we developed this from a metric thing. I could understand every piece of it, you see?

[01:00:41] GREENE: And we’ll get to a version of that. But Barry, your view of those early days?

BARISH: Yeah. Joe Weber, first, I think I was really an outsider doing particle physics and I thought he was one of the most imaginative, technical, creative guys. And the dilemma for me is how he could be such a terrible scientist at the same time. Which I think he was. This wasn’t the only time he thought he saw gravitational waves. It went through his career, he was bitter, and yet he made a tremendous impact. He had the insight how to build this stuff.

GREENE: And he was never convinced that he didn’t see them is that?

BARISH: Yeah, because better experiments didn’t.

GREENE: But what was his view? I mean he held to this perspective.

BARISH:  He felt he felt he was being mistreated and we didn’t understand, that

WEISS: Well it was even worse than that. Do you mind if I interrupt? Because I’ve lived through it. I mean, you were really doing physics and I was screwing around. But the thing was that the thing that happened was that a very famous physicist, and I blame it on and blame and give credit to, Dick Garwin.

WEISS: Dick Garwin

GREENE: From IBM

WEISS: I know he wasn’t-well

GREENE: I know him from IBM days.

WEISS: That’s exactly right.

WEISS: Dick was- he had free license at IBM to do what he wanted. In fact that’s when IBM still had their labs here in New York. Yeah across the street from your guys.

GREENE: Right.

WEISS: And so he and a guy named Levine, just as a sort of a pleasure for the afternoon they decided to make a little tiny bar, just like what you saw but a well-designed bar, well understood, and well-calculated. In other words a lot of intelligence went into doing that design. He did it and they knew exactly how sensitive they were. OK. And so what happened is- that the acrimony in this whole thing really happened because Weber and Garwin would meet at meetings. Now this is now into the into the late 60s and early 70s. And they would meet and Weber would give us the most recent results. In this case, by the way, he became very serious with the results when he had three detectors. One in his lab in Maryland, another one in the golf course about eight miles away from his lab, and then another one in Argonne laboratory in Chicago. And that’s when he began to really say he’s seen something because he saw co-incident pulses in all three of them about two or three a day. OK. And what happened is that Dick Garwin, who knew his apparatus much better than Joe Weber did, unfortunately for Weber, and he could tell you how much sensitivity- you could you could go to Joe and say, “How sensitive is your bar?” And he couldn’t really tell you. I’ll tell you how much we thought it was sensitive- 10 to the minus 16 was the very best it could have been in these units. And the first detection was 10 to the minus 21. OK. Right. So it’s many orders of magnitude away. But he didn’t know that himself. And now his defense when Dick would get up and say at a meeting, “I built this little thing and I know exactly how much I see. I see nothing at the level of 10 to the minus 16.” Weber would- his response was not a rational response. It wasn’t. “Well I know my sensitivity. I see something at 10 to the minus 17, or something like that, and you’re not sensitive enough.” That’s not the response. The response he made is, “You didn’t do it the way I did it.” And that was the wrong way to approach this whole problem.

GREENE: Very interesting. Wow.

WEISS: You weren’t there.

GREENE: I wasn’t, no. So let’s think about the right way to do it, which you guys have now achieved. So we’ve got some pictures. Tell us what we’re looking at. Barry?

BARISH: That’s a four mile long arms, two of them. Perpendicular to each other. And we make- send light from a laser, split it so it goes down both arms and comes back. And if our arms are equal length, we make them cancel each other.

GREENE: The light cancels against itself. And that’s how do you think about that? Can you give us a way to think about it?

BARISH: The light comes in a waveform and as it goes down one arm and comes back and goes down the other arm and comes back, we reverse one compared to the other so that there’s a black-

GREENE: So like a peak and a trough or kind of crossing each other?

BARISH: Crossing each other and it’s back. And if one of the arms gets a little longer than the other one, they get out of time. And then we see light and that’s the simple simplest picture.

[01:05:07] GREENE: Yeah we’re going to try to do a version of it here if you don’t mind. I’ve tried to do this experiment before, but I’ve always been very nervous about whether it’s going to work or not. I’m not the slightest bit nervous because I’m not doing it. One of you experts is going to do it. Can you bring out this

WEISS: No, no, no, no.

GREENE: Interferometer, if you will, from behind. So here we have  a slightly smaller version of what you just described. So, I’m going to turn it on and then ask one of you guys to explain what’s actually happening here. Wow we actually have some fringes. Can you get those up on up on the screen. I should not be standing here. Nergis would you do this and just take us through what’s happening there?

MAVALVALA: OK. So it’s an interferometer.

MAVALVALA: And what we have here is a laser. It’s actually a laser

GREENE: Can you bring the lights down a little bit over here?

MAVALVALA: – Like a laser pointer and we have a mirror here that’s a special mirror that takes half the light and lets it go through it. And the other half of the light gets reflected from the mirror. Oh you know in the old days when people smoke you would just like light up cigarette and make this all happen. But

GREENE: This is bug spray, by the way.

MAVALVALA: Good, it’ll kill the roaches too. So. So what happens is. The light then reflects off of these two mirrors and comes back to the beam splitter and then, just as Barry explained, if the peaks of the light line up with each with each other, the light in each arm, then you get a bright bar of light on the screen. And in the places where the peaks line up would troughs, the two waves cancel and you get a dark fringe. And so what you can see as you go through looking at those, wherever you see dark stripes those are regions of space where the peaks lined up with the troughs and the light beam cancelled itself. And so our measurement at LIGO was pretty much just that, which is that we operated under normal circumstances where we make the two light beams cancel each other. So it’s dark. And then when the gravitational waves comes by, it gets a tiny bit brighter.

MAVALVALA: And we can measure that tiny bit brighter. One of the thing that’s very interesting to see on this picture here. The darkest- between the dark stripes and the bright stripes is about it’s  a fraction of the wavelength of the light. And so right here on the screen, you’re seeing separations that are of order a fraction of a micron. So about 10 to the minus 6 meters. I’m trying to convert that into units we understand. It’s the spacing is a fraction of a hair of mine. So you can see right here on a screen with a simple apparatus like this and more or less with our eyes in going between bright and dark. We

MAVALVALA: are seeing differences of about, you know, a fraction of my hair between these two arms.

GREENE: Now if you just hold still for one second, can we make this as stable as we can? So everybody please just hold your breath. Don’t move. And we’re getting relatively -my talking is now messing it up. I’m going to talk really softly now. But there you go. And then if a gravitational wave, well obviously this is sensitive to all sorts of disturbances not just gravity waves, but the point is how I mean disturbance by yelling out can you get this to move?

MAVALVALA: Let’s clap.

(Applause)

GREENE: Thank you all very much for coming, thank you. No, but there you go. So- and what you guys need to do, presumably, is to isolate your four kilometer long version so that when people clap, or a tree falls, or someone revs a motorcycle engine, it doesn’t shake like this. So the only way that it will shake is if a gravitational wave rolls by. And presumably that is the challenge. So thank you for this little demo here. So how do you do that? How do you isolate in such a way that the only thing that can make it shake are waves of gravity?

WEISS: Well, a lot of things, but, David should-David, David.

[10:09:22] SHOEMAKER: I think probably if you

WEISS: And I’ll correct him if he’s wrong.

SHOEMAKER: One good thing to do is to think about the two challenges that we have in building this instrument. One is keeping these objects, that are points in space time that should only be influenced by the passing gravitational wave, keeping forces away from them. So they’re not moved by things like seismic noise. And then the other challenge is measuring with ever better precision, with light, the distances between them. But so let’s first talk about this problem of the masses being disturbed by the outside influences. Clearly, seismic waves, and that could be from any source. It could be from the ocean pounding on those shores, which can be seen completely in land. It could be passing trucks and so forth and so on. Wind,, shaking people passing nearby. All of these things caused the ground to move. And we want to keep that motion from getting to the points in space time that we use to watch the gravitational wave.

And the basic principle we use is pendulums. If you have a pendulum, let’s imagine it’s about this long and it has- its swings at about this rate about a half a second per cycle. If I move the top supports slowly compared with that frequency, the pendulum will just follow. But if I move that top and I shake it at high frequencies, frequencies higher than the pendulum frequency this mass is going to want to sit still. It has inertia. It naturally wants to be an isolator for frequencies higher than this pendulation frequency. If you put a bunch of these things in series, which is what we do in advanced LIGO, here you see an image of the bottom of one of these multiple pendulum suspensions. Then each layer acts like another filter against the seismic noise and the system wants to stay still in the inertial space and that’s exactly what we want. We also used more sophisticated systems where for instance we put an electrical sensor next to one of these intermediate pendulums and measure how the outside cage is moving with respect to the pendulum that generates an electrical signal into an amplifier, and then into a motor, where we tell that motion to stop. So it makes a servo controlled stable platform. So through this process of multiple layers of isolation basically using the pendulum idea in different ways, we can suppress the seismic noise by 10 to the 12 orders of magnitude, or something, some incredible amount and get to the point where in fact the mirrors are moving far less due to seismic motion than they would be moving to a due to a passing gravitational wave.

SHOEMAKER: It’s not our only problem. The next thing that happens is that we have to deal with Brownian noise. These mirrors like everything is in equilibrium with the heat bath. Each one of the modes of the optical system’s got energy, thermal energy. You probably heard of Brownian motion which causes, for instance, the needles of a very fine electrometer to shake around. These things move in the same sort of way and to manage that, we have to use materials with very , very low mechanical losses. Gather that energy into very narrow peaks that are outside of the region and frequency, what we’re trying to do the measurement tool. So a lot of it is very fine machine making it to get good instruments with very low mechanical loss and very, very, very slow motion.

GREENE: Now you did this in two steps right? I mean there’s LIGO and then advanced LIGO. So first one was, what, 2002 to 2010? It was operationally roughly. Is that right?  

WEISS: 2008 yeah

GREENE: 2008, and then you went off line for the upgrade. But it wasn’t an unexpected upgrade. This was your plan from the get-go and that’s a kind of forethought that is wonderful. But why did you think of it in a two stage system? Why not just go for the you know the one that would work from the outset?

WEISS: Well that’s almost really part of the whole business of how the project went from table top to being good and large scale. And I’d like Barry to talk about it, but let me give you the philosophy we had. In fact, it succeeded exceedingly well. The philosophy was indeed what you say. Once we got to the point where we want to write a proposal for the large system, which was done in 1989, that’s when the real proposal was written. Prior to that, there had been studies of large systems and how much they cost. That was earlier. But the real proposal for this was written in ’89 with exactly the strategy described. And the reason we did that is because you wanted to at least- and look, this is an expensive project. People were skeptical of the whole concept to begin, with so you had to be very careful. What were they really skeptical of? I’ll

WEISS: be honest with you. The skepticism was the calculations that Kip had done and all of the people had known about gravitational wave sources. Kip Thorne just on about yeah. That was considered flaky. I was- the whole idea of being able to do measurements at 10 to the minus 18 meters, which you know you need to do to get to 10 to the minus 21 strains even over four kilometers. That was considered way flaky. I mean nobody went in- most engineers, when they hear about 10 to the minus 21, they think you’re out of your mind. I mean there’s nothing you do at 10 to the minus 21.

GREENE: And 10 to the minus 21 is a fractional stretching of the length-

WEISS: Let’s make it simple. Let’s say over four kilometers. It’s the motion of the mirrors is 10 to the minus 18 meters. And should I-

GREENE: Just to put that into context. How do we think about 10 to the minus 18?

WEISS: Yeah, let me quickly say that that is an easy way. Well I think we should. Let me just. I’ll do what I did once before. You know you start with a meter, and you say OK let’s divide that by a million three times, three times over. OK. So you start with a meter. OK. And then you say let’s divide by a million. We can still sort of imagine that. And what do you add to it? One micron, which is about one hundred- Well 130th the size of your hair. That’s still the

GREENE: Width.

WEISS: People can imagine that. Then you divide by another million. OK. 10 to the minus 12. That’s smaller than an atom. An atom is 10 to the minus 10. So in these units and

GREENE: Including the electron.

WEISS: You’re sort of inside of the atom, but you’re not yet to the nucleus. The nucleus is miles away and now you divide by another 10 to the 6. now you’re 10 to the minus 18 and now you’re in Never, Never Land. I mean most people just can’t imagine that. I mean you’re down to the point where you’re deep inside the nucleus.  

[01:15:20] GREENE: So it’s like a thousand the size of a nucleus right.

WEISS: And that’s the best I can do for you right here.

GREENE: That works. OK. That work for you guys? Very good.

WEISS: And so the people thought you were kind of crazy that you would even be able to do such a thing. So we had multiple things that we had to do. So we had to be very careful. And in that proposal, that’s why we did it this way. We want to take. We had in the program- many people had built prototypes detectors on a table top. You know, and that demonstrated independently, various techniques. For example, the kinds of things David just talked about good vibration, isolation systems then the ability to read with light from an interferometer. You can- you don’t have to build a huge thing to read 10 to the minus 18 meters. You can you can do that on a little instrument. That’s- you don’t- you’re not measuring a strain of 10 to the minus 21. See, for gravitational waves you have to make it bigger. But you could certainly prove in a little instrument smaller table top

GREENE: I see.

WEISS: that you can measure 10 to the minus 18 meters and that took a while but people were getting there. OK. So

GREENE: And what year was that?

WEISS: Oh 10 to the minus 18 is an exaggeration. Ten to the minus 16. Certainly we could do in those days. OK. And that was we’re not talking about a lot of people worked on that. Very good German groups, a group in Munich worked on it. They did much of the initial development of this thing. And then the group in Glasgow and then the group at Caltech and the group at MIT. Those were the principal groups that did a lot of that development. And with that development, you could say to somebody who was a real critic,  “Look we’re not miles away from doing this the individual pieces that you needed. We could do the vibration isolation. You could do the sensing. OK. And now the question is we can’t do the measurement that’s a scientific measurement unless we take that whole thing and put it into something that’s very large. And that’s when the money has to be spent. So you knew- and what we traded on was the fact that you had done these things in the prototypes. That things that people were skeptical of and that the new thing would be that you had to now build it on a large scale. And so we wanted to do- promise no more than what we had done in the laboratories for the first goal and that’s why we made two different things.

GREENE: Two phases.

WEISS: Two phases, because what we had to develop in the laboratory was probably not good enough for some of the more reasonable sources. They were good enough for some of the wild sources that Kip could think of but they weren’t for the more reasonable sources. For example, we could do a neutron star binary with that yeah.

GRENE: Right, OK.

WEISS: And so it was very close and that was unlikely. OK. So consequently, we did right then and there, say, “Look this is a technology development but has a chance, not zero chance, of detecting something.” But we were not very high probability for it. But, we said, “The second- we have a second piece of this. And now here there’s a critical thing. You’ve got to build these facilities big enough so they can encompass both the first version and the second version. And that was the where all the money was going. In other words, all the money was going into making buildings, making vacuum systems that were 4 kilometers big.

GREENE: And how much money are we talking about?

WEISS: We’re talking about down to about 200 million by this time. Barry can tell better, but I think that’s about right.

BARISH: Well up to now 1 billion.

WEISS: Yeah, but not at that time. I mean just for the construction of the facilities and the thing is that -and then we said, “Look, we don’t have to change those facilities to put the better detector into it.” So that was the initial investment was good for both. The things we knew how to do at that time and also what we thought we would do in order- we didn’t have to change the vacuum system. We had-everything would work. The expensive part I think- would have that build-put another something that would be the equivalent of another detector. But you would not have to build the infrastructure right. OK. And that was the proposal.

GREENE: And that went through.

WEISS: And that eventually went through.

GREENE: And you know Barry, you know, I don’t want to embarrass you but everybody that I talked to said that you were the guy who made this happen, right? I mean there was vision there was a plan but you’re the guy that made this happen. What would you say was the big challenge and the big thing that needed to?

BARISH: Well I don’t think we were successful and we’re successful in doing the science. But I must say I never know of anyone who thought this was going to take us 22 years. It’s 1994 when our first detection is 22 years later and I don’t know whether you or anyone thought it was going to take 22 years. But I think probably most of us felt this- these two stages would be half that. So we’re not as good as you-

GREENE: Did you ever lose. Did you lose hope as you were doing this at all?

BARISH: No no no. We’re just not as good as we thought we were. I think.

GREENE: So you just gained humility through the process.

BARISH: So when you point and say “We did it” or “I did it” or something I think.

[01:20:03] WEISS: I want to give him credit. He won’t do it. I want to do it.

GREENE: Please do.

WEISS: Because I’ve lived through this too. You know we were pretty disorganized. OK. I mean that’s why it’s so important to look at the transition from the table top version of this thing to what we call the LIGO full scale thing. OK. It’s true that we had begun, by the middle 80s, to begin our early 80s to look at. In fact I-  MIT did a study of what with industry to see how much it might cost, how you would make a bigger detector than a little tabletop, thing like that. We had and we did a study of it became- in fact, became the basis for two things. It became the basis for a presentation to the NSF jointly by CalTech and MIT together to push the idea that you should build a big one. And we had a lot of help within the NSF, by the way. People in the NSF. But this was exactly what the NSF should be doing. I’m glad they thought that way

GREENE: Rich Isaacson.

WEISS: Rich Isaacson in particular. Yeah. And he then made the investment in the NSF, the intellectual investment and showing the people in the NSF that this was a good thing for them to do. Because it was risky, but it was also, if it worked it had a tremendous payoff. And that’s funny- I think and that’s I think an honest statement of what was going on. And so what happened then was that we were able to get this collaboration going. But it turns out the principles were not-that includes myself and Ron Drever, who is unfortunate not here right now, and Kip Thorne, we were the ones who- well we’re not- we were not necessarily the most strategic thinkers, I’ll be honest. And in fact, we squabbled with each other- not several of us did. And what happened is that in fact, in 1986, when NSF had started looking at this whole thing, again Richard Garwin played a very important role in it at that time. He wrote a letter to the NSF says, “Look he’s heard about this thing, you know.” Now why he was so important is because of the experience he had with Webber. But he was honest in his thing, by the way, he wasn’t totally critical. He was trying to help. And he was critical but also he was trying to help he says, “Look if you’re going to do this,” he said to the head of the physics branch of NSF, “if you’re going to do this, you ought to have a study about it. A study with people who are not in the field. Let them judge this.” And that was put together and we got some very good people doing it. People who are not in the field. Nobel Prize winners who are experimenters, many of them a lot of people now in the field. But who were not in the field then. And they gave an enormous endorsement to them. Endorsement was, don’t mess around, build two big facilities right away. Build them right away- why don’t make that- don’t make any more prototypes. We- if you want to know you’ve got to get to the science and you’ve got to find out what are the problems when you make this bigger. And they have recognized that and they also said, “Don’t build just one, because they want to see the science.”

GREENE: Sure.

WEISS: If you want to have two. They saw that right away. That was part of the strategy. And we haven’t yet written the proposal. You see, that- all of this gets incorporated later in 1989 in the proposal. And- but the thing that they were very skeptical of was look the- here was this three of us who were not adept at running things. We couldn’t quite agree on things, among other things. And they say, “Get yourself a single director. ” And that was bingo. Oh no, that wasn’t him right away.

GREENE: When are we going to get to Barry? I’m hanging by the edge of my seat here.

WEISS: We’re going to get to Barry. First, it was a fellow named Robbie Vogt, who helped us write that proposal. And then it became Barry. And Barry really pushed the queue.

BARISH: Actually the key to success is one that people don’t really point to. And this was something that was very hard to sell to the NSF, but it wasn’t just building this big thing. The NSF actually supported us as we built this unfinished building. We were able to concentrate on all the stuff that David talks about. That rather than the usual thing whether you go and you build a big thing, is they take away all the money or the work that you do in your labs to develop all the things we don’t know how to do and come back after you fail or succeed. And they kept us funded to do all the work toward the second generation. In 1998, 1999 when we finished this.

WEISS: But you insisted on it.

BARISH: I insisted on it. I thought we would never get to the second step without that if they bought off on it. Now that to me was the key to success, not any little individual.

GREENE: Right. We’re starting to run a little short in time. I want to get to our final section, which is just looking toward the future, right? We’re going to also begin this with a little final clip. If you could run that please?

(Video Clip)

[01:24:47] NARRATOR: In the era when we were trying to convince people that LIGO should be built, for me, this was the most exciting thing is the likelihood that we will see waves from phenomena that we never dreamed of. After all, every time a new window has been opened up in the universe, unexpected things have been seen, and this will surely not be an exception. Gravitational waves are the only form of radiation that is so penetrating that it would travel unscathed through the very hot and dense material. In the earliest fraction of a second in the life of the universe. So if we are ever going to actually see the Big Bang, the only way that we will actually see it is through gravitational waves. I, therefore, regard that the ultimate Holy Grail of gravitational wave astronomy is probing the Big Bang. Some of the strings from which everything is made. Fundamental strings we call them, which are submicroscopic, far smaller than the size of an atom or an atomic nucleus that some of these strings are expanding the cosmic size by inflation. And there are calculations using tentative versions of string theory that show this may be the case. If that happens the result are then what we call cosmic strings. And these cosmic strings are simple enough that we can predict with high confidence the shapes of the gravitational waves that are produced when two such cosmic strings collide pass through each other and with high probability, recombine. So you wind up with two strings that have kinks of themselves and those kinks that travel down the strings at the speed of light, emitting gravitational waves. The resulting wave forms are very well understood and are quite simple and the estimates of the strings of those waves are such that LIGO has a real shot at seeing them.

GREENE: So I love the fact that he speaks of string theory as if it’s absolutely right, which is sort of a beautiful thing for somebody like me. But he really means that if string theory is correct that there is a chance we could see them. The evidence of strings through these gravitational waves the undulating strings would create. But you know we don’t have a lot of time left but starting with you David, when you think to the future, you know, both in terms of the devices and the things that we might find, what gets you most excited?

SHOEMAKER: Well I think there are two things, two directions we can go to go well beyond the advanced LIGO and well beyond probably makes it- from initial LIGO to advanced LIGO with a factor of 10 in sensitivity because we look at the amplitude of the waves being able to reach out a factor of 10 times further brings a thousand times as many possible sources into our reach. Probably the next step would be another factor of 10 on ground. And there are a couple of different ideas of how to do that. I mentioned that one of the limits to what we can see is the thermal noise. And one thing you could do is to refrigerate the masses down to a fraction of a Kelvin, or maybe something not quite as cold to reduce the thermal noise. Another way you can do it-this is a strain in space. If you have a meter stick, it will change by 10 to the minus 21 meters when the wave goes by. If you have a 10 meter stick, it will change by 10 to the minus 20, ten times more. So we have four kilometer long arms to take advantage of that. Now, if we had, say 40 kilometer long arms, that would make the signal ten times bigger without any of our noise sources growing that- by that same amount. So there are a couple of ideas that we have for how we can grow ground based interferometers into systems that could reach even further back and say get to the point where we could see the very first black holes see all the black holes in the universe. I think though the other thing there was this image up on the screen a little while ago of a constellation of satellites in space.

It’s obvious that that’s an experiment that must be done. It would probe not the same frequency range of gravitational waves so it’s a little bit like the radio astronomy an optical astronomy, but it would be also a place where you can instead of having a mere 40 kilometers long arm, you could have a million kilometer long arms. And have a signal that is that much larger and be able to probe, with incredible precision, the waveform of the gravitational waves use it as a- small black holes orbiting around big black holes as a way of mapping out the curvature of spacetime in more detail. It’s an experiment which must be done and we think we understand how to do it. I think we’re getting closer to the point where we can know you have a good date for sending something like a LISA, a gravitational wave antenna, and the Europeans are pushing this so much more than we are. Well you know there’s a bit of an unhappy history there, for a while, there was a really good joint mission between the U.S. and Europe and the technology was working along well. The James Webb Space Telescope got much much more expensive than anyone thought. And the way the walls work for money and NASA is that if there’s one space experiment that’s in trouble in science, it eats all of the other ones up. And that caused us to break a number of relationships we had with the Europeans to build space missions, and this was one of them. So the Europeans have gone ahead. They actually have. They have a date of 2034 for a launch of a space based mission. They’d actually like to bring that in a bit. And their calculations for technical readiness say 2029 2030. The U.S. has got a small effort that we’re trying to catch up with them in the sense that we’re at least going to play a minority partner in it and to participate in this really wonderful experiment. Maybe we can even grow our own participation based on the successes of LIGO and also a recent experiment- a test experiment was put up by the Europeans called LISA Pathfinder. And while the data are not yet publicly released, I think we’ll see you know performed really remarkably well. It’s it’s on it’s way it’s still a ways away.

[01:30:44] GREENE: So Frans, what about the future of the numerical methods? Do you see that playing a vital role going forward? New templates? New ideas in that domain?

PRETORIUS:  Yeah. And in particular, if we think of adding neutron stars to the mix. So we are going to hopefully learn a lot about black holes in the universe with upcoming observations other other events. But in terms of the template, we pretty much have that control. But the one thing which we don’t have under control is neutron stars. And you have the. Einstein’s equations the the geometric side the G on the left hand side and this Tµ which is the matter on the right hand side with black holes at zero. We just have to hit zero. We don’t worry about matter. It’s purely vacuum. Neutron stars now, of course, they are they made of matter. And the thing that’s both interesting and challenging about it is matter in those extreme conditions set the density of neutron sources into- the gravitational force is so strong that packs all of the neutrons together in basically one big atom with a mass of the sun. So example- if you can compress the Earth down to those densities, it would be about the size of a football field. It’s incredibly dense matter. We don’t understand those properties. And so when we tried to do- to try to solve what the gravitational waves would look like, especially when, say two neutron stars collide, they form a hyper-massive neutron star. They might collapse to a black hole. But a black hole rips a neutron star apart. We need to model that right hand side. Isn’t it’s interesting with the Einstein equations, that we use words “modeling” and “simulation,” but with the left hand side, the geometry, we’re actually not modeling anything. We’re actually solving Einstein’s equations.

There’s some numerical error, but it’s actually solving those exact equations. They’re the corresponding equations for neutron star matter. That’s the fundamental laws of particle physics, quantum electrodynamics, quantum chromodynamics, electromagnetism. There’s no way in hell we could ever solve those equations in the same way that we do the Einstein equations. So we have to do now, in the real sense of the word “modeling,” come up with an approximation for those things. And because we don’t understand math at those densities, the models are incredibly crude. Basically blobs of fluid. So. The two things that are interesting about neutron stars one is just when they collide, they’re are going to emit photons. And so we will hopefully see both gravitational waves, perhaps a gamma ray burst. Perhaps these interesting things called Killer Nova, which might be what’s responsible for all the gold and platinum and all that all the bling in the universe. As opposed to the stars which produced the stuff that make that make us exist. So there’ll be a lot of interesting information that will come from the electromagnetic signatures but also, they will tell us really what matters doing at those extreme densities. So they might be clues, not just to what happens with the populations of neutron stars and black holes, but it might be telling us something about matter at densities that we can probe in the laboratories on Earth. And so that’s a very exciting theme going forward in the simulations.

GREENE: And Nergis, is gravitational waves going to be your focus going forward?

MAVALVALA: Yeah, I think so-one thing that we know- there’s an “L” in LIGO and there’s an “L” in LISA, and that’s the laser. The laser light is our meter stick. And the laser light is is- light is quantum, it’s made up of photons and the quantization of the light is a big problem. It’s a limitation to how well we can build future detectors. So for me, that’s an area that I’m really interested in. How do you engineer quantum states of light that can help you do better? And we’re doing that in our labs now and so yes, there’s plenty to do. And for me personally, the field is, it’s just begun. We’re filled with ideas and now we just have to go build better instruments

GREENE: Very good. How about you two guys, young gentlemen? What are you thinking about for the?

BARISH: For me. I think we’ve opened a whole new area of observation. And I can’t imagine that we know all the things we’re looking for. And so we’re going to be surprised and I’m looking for the big surprises. I’ve lived through that in particle physics when in 1975, we found a totally new set of particles and a new way to see things. I think the same thing will repeat itself here.

[01:35:14] GREENE: Rai? You want to take us out on a final thought?

WEISS: Well I’ll give you the same thought that Kip had. I’ll just amplify what Kip-Kip said something very important. He said that this field is fundamentally different in astronomy than what we now know, except for neutrinos astronomy, that’s closer to us. But an electromagnetic astronomy, you’re dealing with things that scatter. The thing that you see has come out of a source. It isn’t what actually went on deep inside the source. You’re seeing something that has been bounced around by all the processes that are going on in the source. For example, when you look at the sun, you can’t see the inside of the sun. Why? Because the light has been scattered so often.

GREENE: And the thing is, it takes a heck of a long time for it to come-

WEISS: Diffuse out through all of that stuff. So it turns out that the idea that you are looking now at the dynamics of things deep inside of the astrophysical systems we see only the outside of, says to me that we’re going to learn an incredible amount of things. It’s going to make a- I think it’s going to make a change, a vast change just in the way people look at astronomy. I’m hoping also that this is not a faint. Because I’ve worked in other areas that I- eventually I’d like to see this really have an impact on cosmology. And I think if we build the bigger the bigger systems that we’re talking about, we will do some kinds of cosmology. We might be able to do the cosmic metric better, for example, to measure the same things that see with light you see now you might have a different might have a different set of systematic problems that might be able to help out where we are now and not knowing things well enough. I wish we could see the Big Bang. I don’t think this instrument I know this instrument won’t do it. There are very fancy ideas on paper that allow you maybe to do this. I hope the people the CMB, the people who did the BICEP, will get there first because then we’ll have really known something in my lifetime.

GREENE: Well when you guys get there, we will invite you all back for another conversation. So this has been a fascinating discussion. I- actually everybody join me in congratulating the team with fantastic achievements. Thank you.

Leave a Reply

Your email address will not be published. Required fields are marked *

View More Comments
Load More

Gravitational Waves: A New Era of Astronomy Begins

On September 14th, 2015, a ripple in the fabric of space, created by the violent collision of two distant black holes over a billion years ago, washed across the Earth. As it did, two laser-based detectors, 50 years in the making – one in Louisiana and the other in Washington State – momentarily twitched, confirming a century-old prediction by Albert Einstein and marking the opening of a new era in astronomy. Join some of the very scientists responsible for this most anticipated discovery of our age and see how gravitational waves will be used to explore the universe like never before.

The Kavli Prize recognizes scientists for their seminal advances in astrophysics, nanoscience, and neuroscience. The series, “The Big, the Small, and the Complex,” is sponsored by The Kavli Foundation and The Norwegian Academy of Science and Letters.

View Additional Video Information

Moderator

Brian GreenePhysicist, Author

Brian Greene is a professor of physics and mathematics at Columbia University, and is recognized for a number of groundbreaking discoveries in his field of superstring theory. His books, The Elegant Universe, The Fabric of the Cosmos, and The Hidden Reality, have collectively spent 65 weeks on The New York Times bestseller list.

Read More

Participants

Barry BarishPhysicist

Barry Barish is an experimental physicist and a Linde Professor of Physics, emeritus at Caltech. He became the Principal Investigator of LIGO in 1994 and was LIGO Director from 1997-2005. Barish led the effort through the approval of funding by the NSF National Science Board in 1994, and the construction and commissioning of the LIGO interferometers in Livingston, LA and Hanford, WA in 1997.

Read More
Frans PretoriusPhysicist

Frans Pretorius is a Professor of Physics at Princeton University. His primary field of research is general relativity, specializing in numerical solution of the Einstein equations.

Read More
David ShoemakerAstrophysicist

David Shoemaker is the Director of the MIT LIGO Lab and the Leader of the Advanced LIGO Project to build the detectors used in the discovery of gravitational waves.

Read More
Nergis MavalvalaPhysicist

Nergis Mavalvala is a physicist whose research links the world of quantum mechanics, usually apparent only at the atomic scale, with some of the most powerful, yet elusive, forces in the cosmos.

Read More
Rai WeissPhysicist

Rai Weiss is known for his pioneering measurements of the spectrum of the cosmic microwave background radiation and his seminal leadership in the conception, design and operation of the laser interferometer gravitational wave detector; remarkable scientific achievements recognized by his roles as a co-founder and an intellectual leader of both the COBE Project and LIGO.

Read More

Transcription

GRAVITATIONAL WAVES-WORLD SCIENCE FESTIVAL

BRIAN GREENE, PROFESSOR, COLUMBIA: Thank you, thank you. I appreciate that. I’m Brian Greene, co-founder with Tracy. And I also want to welcome you to Day 4 of the 2016 World Science Festival. The program tonight on gravitational waves. It’s a timely program. I trust you all heard the announcement on February 11th or you read about it. The first direct detection of gravitational waves. Amazing, amazing discovery. And we have here tonight a group of the very individuals who made this discovery possible. And I’m going to bring them out in just a little bit. But just a few words at the outset. You know, this discovery just to put it in context, it’s the kind of achievement that happens maybe a few times a century in fundamental science. I mean we were thrilled some years ago for the discovery of the Higgs boson, something that was search for a very long period of time and finally was definitively discovered at CERN. Great discovery. This ranks on par with that kind of experimental observational achievement. And the other side of the story, that will come out in the conversation here tonight, is that whereas for the Higgs boson, just about everybody was convinced that the experiments would succeed.

I mean most people who were following closely the search for the Higgs boson were saying, well if it’s not found that’s the craziness. I mean this has got to work out. The theory says that all of the circumstantial evidence suggests that the particle is out there. And indeed it was found. Now with gravitational waves as we’ll see, there is never really any doubt after, oh I don’t know call it 1950, 1960, that these ripples in the fabric of space were a real feature of Einstein’s General Theory of Relativity. But there definitely was controversy as to whether we would ever have the capacity to catch one of these waves. And we’ll hear about why in just a moment. But what this means is, it takes a certain kind of internal fortitude. It takes a certain kind of vision. It takes a certain kind of stubbornness. It takes a certain kind of independence of spirit to devote your life to this kind of project. And our hats are off to the individuals, they now number roughly a thousand, who made this discovery possible. OK. A little bit of background before I bring the panelists on stage, just to set the stage scientifically. Right, so we’re all more or less familiar by this stage of the game that Albert Einstein gave us a new theory of gravity in 1915. The General Theory of Relativity. The Newtonian picture that we all learn in high school simply spoke about objects pulling on each other through space with some sort of force that was described mathematically but whose mechanism of exerting its influence was not articulated.

Albert Einstein comes along and articulates the way that this force is communicated. And just to give a picture of it, if we can bring up the first little clip here. It goes like this. Objects-bring the lights out a little if you will-like the sun or the planets. They warp the environment of space and time and it’s these warps in curves in space that are able to cause other objects to move along peculiar trajectories if you weren’t aware of this picture. The metaphor that we love to use it’s imperfect to be sure. But if you think about space and time it’s sort of like a trampoline. You put some heavy object in the middle of it. It warps the surface and then if you roll a marble, the marble goes into orbit much like a planet would much the same way the sun warps the environment. And then the planets go into the orbital trajectories that we are familiar with.

[00:04:49] GREENE: OK so Einstein gives this very beautiful new description of gravity in 1915 and then the question of course is, “Is it correct?” And indeed, there were very quickly two pieces of evidence that sealed the deal. The first was an explanation of a puzzle having to do the motion of the planet, Mercury. According to Newtonian gravity, the planet should just trace out the same orbit year after year after year. But observations show that the orbit was actually shifting a little bit. Nobody could give a good explanation of this. Einstein comes along and uses his new approach. This new picture of gravity uses the equations of general relativity and is able to spot on predict this so-called precession of the perihelion of Mercury. Now that was kind of a post diction explaining a puzzle that was already known. Observationally but then he was able to go further and make a prediction for something that had not yet been seen. And indeed, that had to do with the bending of light by the sun. So if you have a distant star, it sends light toward the earth goes in a nice trajectory of that sort. But then later on if the earth is on the other side of the sun and you now consider that star light according to Einstein’s new picture it’s going to go in a curved trajectory as it goes through the curved environment surrounding the sun. And that has the effect of making the position of the star in the sky shift.

Einstein was able to use the equations of general relativity to calculate the angle, and it was just a few years later that teams of astronomers went out and measured the angle between those two positions, taking photographic plates during a solar eclipse so you could see those distant stars compared with photographic plates taken six months earlier. And indeed, the data is a little bit of massaging a little bit of forethought a little bit of actually knowing the answer in this particular case, seemed to show that Einstein’s ideas were correct. Since then, these experiments have been done over and over. There’s no doubt any longer that Einstein’s theory is correct. Now, Einstein then went further and as we’ll discuss here tonight, he wrote a paper in 1916 in which he basically says, he didn’t use quite that language, which basically says mathematically that if space is sort of like a trampoline and it can warp and curve, well if you start to tap the trampoline, disturb the trampling you send ripples going along the trampoline. Those ripples, taking away the metaphor, we see ripples in the fabric of space and that’s what gravitational waves are. Now it’s a very interesting story that we’ll get into a little bit in just a moment with our panelists. Einstein himself, he writes this paper 1960 actually made a little mistake. And maybe a kind of weird concept, Einstein making a mistake in doing calculations in relativity, but relativity is hard. And Einstein was the first one who is actually working on these calculations. So he was making what we would now call mistakes of a beginner, novice mistakes. But he was the guy who was blazing the trail so it makes perfect sense that he might have made those kinds of errors. Anyway, he corrects the error 1918, but then he continues to have a really interesting relationship with this idea. Thinks it’s not quite right. Later papers that suggest that it’s not right.

Bottom line is, others come along develop a very beautiful systematic framework for avoiding the pitfalls that Einstein fell into. And people are absolutely certain that these things are real. What would that mean? It would mean that if you disturb the fabric of space, you should get these gravitational waves, these ripples. And the prototypical example might be something like this: rapidly orbiting neutron stars or black holes sending out this march of gravitational waves. And as we will discuss here tonight, what would that mean for somebody in the wake of these waves? They object would be stretched or squeezed according to the math. And I want to emphasize one thing. This is not to scale. And that is a big deal, because when you actually do these calculations,, as we will discuss the stretching and the squeezing is so small that it is an enormous challenge to detect these gravitational waves. And this is what the LIGO team has done. All right so now let’s get into the actual discussion.

I’m going to bring our panelists on. And let us begin with Rai Weiss, of MIT, is one of the creators of LIGO. He first proposed a large scale interferometer back in 1972 and he has been the scientific and engineering leader ever since. All right. Next up, Barry Barish. Many people say that Barry Barish of CalTech saved LIGO at the time it transitioned into big science and built the two huge detectors in Louisiana and Washington State. He joined LIGO in 1994 and was director from 1997 to 2005. Welcome. Next up, Nergis Mavalvala-I got it, and I’m so proud of myself- of MIT is an astrophysicist who is MacArthur Genius Award winner for her work on LIGO in 2010. She got her Ph.D. from MIT in 1997, postdoc at CalTech from ’97 to 2002, and then joined the faculty of MIT. Welcome. Next up, Frans Pretorius who is a professor of physics at Princeton University. His primary field of research is General Relativity,specializing in numerical solutions to Einstein’s equations, and as we will see, a vital part of this discovery. Finally David Shoemaker, who is director of the MIT LIGO lab and was the team leader of advanced LIGO and spent many years upgrading the technology of the observatories. And lead to the historic detection even before advanced LIGO was officially launched. Welcome. OK. Thank you all for joining us here tonight. Let’s just get into it. So,in the description of this discovery, many have called it, you know, one of one of the great achievements of our age. And because of that, we’re thrilled to note that a filmmaker, who has generously given us some footage, has been following you all around for a little while, Les Guthman, and we are going to begin each section of tonight’s discussion with a little video clip from a documentary that he is making, just to sort of set the scene for each of the discussions. Let’s if you will run the first clip.

(Video Clip)

NARRATOR: Take the entire universe, count every photon that’s coming from every star, and add it all up. That’s how much energy was released by the sun. It’s the most energetic astrophysical object ever observed since the Big Bang. You worry that,in any complicated experiment like this that, you may have made a discovery but you’re not sure that discovery is really true. And I mean that’s many situations in my life where that’s happened. We’ve been through a whole series of exercises to figure out what this was. And many of them, underlying them, are sucked out. We’re like, “Could we have caused this by our own ignorance or anything from accidents to, you know, malpractice. There’s this wonderful thing that goes around, you know, is there some evil genius that has laid this into us? This evil genius gets to be more and more complicated and more and more skilled as you delve into this in a deeper and deeper way. So it’s not just one evil genius, it’s a multiple group of evil geniuses. Every check we’ve done has kind of come back saying, “No, this really came from some astrophysical origin. We believe- I mean nothing in the instrument that we found so far could have done this. And nothing that we have found so far indicates that it could have been some intentional or accidental or malicious injection. And so then you start almost believing it. And that’s pretty cool. I think the remarkable thing is not so much that we saw gravitational wave. I mean that’s remarkable enough, but it isn’t the real centerpiece of this whole discovery. That centerpiece is the fact that we saw a particular source that nature seems to have given us. It’s the best source we could have imagined. It’s the source that says Einstein is right on every goddamned detail, almost. In other words this is a source that doesn’t require any other physics than general relativity. Let’s think of that.

GREENE: So we’re going to we’re going to unpack the discovery in a number of stages. Let me just begin. I recall last fall, rumors started circulating all around that you all had seen a gravitational wave. That was probably- I started to hear about this October, November or something of that sort. First of all, what was your feeling when these rumors started to get out there? You guys had not announced yet, but yet there was this growing anticipation. Was that exciting? Was it irritating? Irritating.

[00:15:07] BARRY BARISH, PHYSICIST: Why? Because if you want to make the case when you do something like this, you want to make the case right. And we weren’t ready to make it right. So it was very irritating and worrying. Because if it did come out, and we had to defend it, we weren’t ready.

NERGIS MAVALVALA, PHYSICIST: I think there was a real process, oh sorry, there’s a real process for us as well. From the time we saw the first signal to believing that it really was of astrophysical origin and not something in the detector. And-

GREENE: I mean, was your initial thought that it that it was not real?

MAVALVALA: Yeah

RAI WEISS, PHYSICIST: Of course. I mean let me say something about that-

GREENE: And you’re not being all modest now. You know it’s really how you felt.

WEISS: No. No, let me tell you how I felt. All of us. I think everybody here and if you disagree with me, raise your hand. However it looked just too good. That was the thing. No no I’m very serious about it. I mean here we’ve gone.

GREENE: You mean if someone was going to fake a signal, this is what they would do.

WEISS: Well, we’ll get to faking it in a second, but let’s just say what it looked like when you first saw it.

WEISS: I mean we’ve gone through a lot of stages expecting signals that maybe just marginal.

GREENE: Yeah.

WEISS: And we had gone and made lovely algorithms to be able to look at the data very carefully. And here is this thing that came on in the morning and you didn’t have to do much filtering. I mean you could look at it with your eye having without a lot of computing. It’s almost nothing more than if you had a radio and you had a base and the treble control and done a little adjusting of it. So you could hear a little bit better or see it a little better. And that’s all we had to do to see this waveform. That was the thing that made us all so happy. It was absolutely beautiful waveform without needing all the paraphernalia that we had developed.

GREENE: Now suspicions of what?

WEISS: That. Yeah what we- I can tell you we- every one of us had some thought. I wish you guys would speak up.

DAVID SHOEMAKER, ASTROPHYSICIST: I think, of course, we did injections. It’s something we regularly do. We inject signals that look just like, you asked, sources that we’re looking for to ensure that pipelines can actually recover the signals.

GREENE: And does everybody know about those injections? Are they a secret?

SHOEMAKER: Most of them, yes, but in our initial LIGO experiments, and we planned also for advanced LIGO to make blind injections, ones which were unknown to all but one or two or three persons who would push the buttons and set the system in motion.

SHOEMAKER: I actually knew a lot about that right at that moment because I was working with the team.

GREENE: But let me just ask you, so if one of those if one of those injections is made, is the point that if someone on the team doesn’t recognize that signal they get fired? Is that basically what it’s about?

SHOEMAKER: It’s certainly if a pipeline that should have detected that signal does not detected, then questions are asked.

GREENE: Yes. So going with what you’re saying about the-

SHOEMAKER: But good, so the- just at that time we were trying to demonstrate that we had the technical capability in this brand new instrument, advanced LIGO, to make injections successfully. And you’re actually having a lot of technical trouble, and I knew that the night before someone was trying out some code that might have been a way to get around the problems that we were having. And so my initial thought was, “Oh good. They finally succeeded in making an injection.” I went to look in the log book and it wasn’t logged there. So that’s that’s when the question really started to come to me that might be a true signal.

GREENE: And so what stage was the apparatus at? It had not actually officially been turned on, right? It was still in a very much of a testing state.

SHOEMAKER: We had just -we had just started an engineering run. We regularly hold these runs which are not to do astrophysical observation, but to run the instrument, leave that alone for a week or a couple weeks, and try and understand if we know how the noise behavior is, if it’s stable, see if it stays locked, see if the operator is able to get into and out of lock successfully. And we had just started the engineering run that was to immediately precede the observing run and we were two days into it when we woke up to this.

GREENE: And now I heard that somebody on the team maybe even on the panel had had suggested to not have the detector actually on for various work that you were doing in order to set it up for the official use of it-

BARISH: Someone was going to do a test that might have knocked it off. And went home and went to sleep instead.

WEISS: A lot of stories like that that we almost missed it because. Yeah. I mean for example, I was in Livingston and I went and we had a problem. I went to fix. And I was told people would take- I told people it would take two weeks to fix it. And I was there the Thursday before the Monday, OK, when the thing happened. So if they’d given me permission to do that we would have missed it.

GREENE: Right.

WEISS: So now there are other stories like the one that Barry just said, you know.

GREENE: Now just want to- when you say missed it just to give it again -we’re going to go into the details of the device in the conversation that goes forward. But, if I understand correctly, so this was an event that happened a long long time ago. Right? Very very far away. More than a billion light years away, right? So this collision of two black holes, which is what this is, sets off this tidal wave that ripples through space and time. And so you’re saying that it’s been traveling toward the earth for like 1.3 billion years and it just so happens that just as it’s rolling by Earth, you’ve got this device that’s just turned on and catches it.

WEISS: I’m sorry, you’re right.

GREENE: So Frans you’re not on the LIGO team even though your work is vital to the success as we will see. So what was your feeling when you began to hear these rumors?

[00:20:33] FRANS PRETORIUS, PHYSICIST: Well yes I’m not on LIGO and actually so I- the way that I first heard these rumors, these people would come up to me and say, “Well have you heard these rumors?” And I felt like I was actually one of the last people to actually hear the rumor. So it was like about two months later, there was someone said LIGO leaked, I won’t break the confidence and say who it was.

But, they sent me this email with this this waveform and the hair went up on the back of my neck, just I got shivers, it was like. Like, “Holy bleep, these things are actually out there.” It was the most astonishing thing to see that. And you know you guys I mean I’ve been studying that course for 20 years. This is the relativity is a fascinating theory. So this is my field. But still you know from a scientific perspective, it’s not just that they’re beautiful remarkable astonishing predictions of a theory. I think as scientists, we want there especially with something which is as astonishing as a black hole, we want from equivalent astonishing level of evidence. And so in some sense I was I guess I was hoping that they existed, but I was sort of agnostic and then seeing that it was you know there was a real revelation on the second thought that went through my mind is that, “No this can’t be right,” looks just like those simulations. Someone, you know, someone did a blind injection and they stupidly just took the most obvious thing that was available.

GREENE: I mean were you concerned about you know sabotage? Was that also part of the thinking?

BARISH: One of the things we spent time on was a trace. Luckily it’s two very independent laboratories. So to make this happen simultaneously limits the options but it’s something that we work very hard

GREENE: And they are in-the locations of the locations and Louisiana and in Washington state.

BARISH: So this had to happen within a few microseconds in both and look alike. That was very difficult but still worried a lot about it.

GREENE: And can I ask another. I don’t know if this is a sensitive question or not, but you know there was an announcement by a different team using a completely different approach looking for a related kind of signal, but not the same one, BICEP2, which found evidence of gravitational waves, indirectly, through its effect on the cosmic microwave background radiation. Some of you may have been here when we had a discussion of this sort, right when that announcement was made, but before it became clear that that was actually then probably not the right interpretation of the data. So that was sort of a premature announcement that, actually got a lot of people irritated, not because the announcement was of something that turned out ultimately not to be the correct description, but people felt that they’d really jumped the gun. These- this team had kind of gone to a press conference as opposed to going through the more traditional channels. Did that influence your thinking at all?

WEISS: I mean look I was- I know a lot about that. Yes. First of all, let me say something positive something negative which is very important. That team that reported that experiment had made a dramatic improvement in the technology of the detection of the polarization which is the, how the electric field points in the cosmic background radiation. The prior experiments had been 20 times worse than any experiment that those guys did.

GREENE: Yes.

WEISS: So that was a dramatic improvement. And what they did, they didn’t have enough channels. In other words, they made an interpretation they did I think

GREENE: channels  meaning frequencies.

WEISS: Well yeah, I’ll get to what it is in a second. But they didn’t have enough channels at the South Pole with the experiment. What they saw is a particular pattern in the sky of the polarization vectors and that’s I want people to ask what that is. But there’s a pattern that particular gravitational waves would make in the plasma of the primeval plasma, and that would then cause these waves that come from the electromagnetic waves, that come from the early Big Bang plasma to have a particular pattern to the polarization through swirly patterns.

GREENE: Yes.

WEISS: And they saw those swirly patterns.

GREENE: Yes.

WEISS: The trouble is that there are other things in nature that make those and that’s what you’re referring to. And they knew that. They knew that-that wasn’t that they were surprised by it. They knew that dust would cause that dust in our own galaxy and also electrons in our own galaxy, that were in magnetic fields. It was a synchrotron. So both ends of what they were looking at but they didn’t have enough detectors of different frequencies to see that. So they made a judgment based on maps that they had. I think they were honest.

GREENE: Yes.

WEISS: OK. And they made a judgment that that the dust couldn’t be as bad as to make what they saw. Sure.

GREENE: And the paper makes that –

WEISS: And the paper makes it very clear. Now where did they make a mistake? They made at least one mistake.

[00:25:24] WEISS: I almost forgive them for that mistake. I don’t feel so bad as some others. It is. I mean there is a dramatic development that in their own technology to be able to do this. And what they did then, is they they assume that the dust wasn’t bad enough to do it. And they published it without going through peer review.

GREENE: Yeah.

WEISS: And the group was too small. That’s the real problem. They all had agreed among each other I had one of my students was on that team. And they had discussed this a lot internally and they all together decided to take this gamble. So what happened is then they needed a bigger group to look at. And that’s what peer review is a very good thing to do.

GREENE: Right

WEISS: And that peer review exposed the fact that there was a good chance that maybe it was dust that they hadn’t gotten rid of

GREENE: Right.

WEISS: And that’s and that’s where it stands now. It’s still not they haven’t retracted, you know,

GREENE: No, what they saw was what they saw.

BARISH: But back to your original question. So does that kind of thing.

GREENE: Yes.

WEISS: Yes.

BARISH: I think the answer is definitely yes. Yes. We didn’t want to come out with the wrong result. Nobody does. Yes. This particular thing breaks all the rules. If you work in a laboratory, you do something you see something, then you’ll do it better. You do all the checks. This comes by in a quarter of a second. It’s gone. So we only have what we have to look at. We know you talked about, before we came on stage, the certain discovery of the Higgs. They spent a huge amount of money to have two experiments so one can check the other. It’s a way physics does things and gets the right answer. We can’t do that here. So this had to be extreme- and I’ve seen experiments that seem convincing on based on one thing and they’re wrong. So I was very very pessimistic and worried and so forth. This just was so powerful we had to-

GREENE: following up on that, you know I think an interesting question just to get out in the open here is so, Einstein let me take us through the history of this Einstein writes his paper predicting gravitation was back 1916, 1918 interesting, as I described he had a sort of complicated relationship with, the gravitational waves might be say, but why can’t we do this in the laboratory? Why is it that we’re only looking for, you know, colliding black holes or neutron stars? Can’t and why is it possible that we could have done this in a way that was closer to an experiment that we could replicate and do over and over again?

MAVALVALA: Sure, so you’re asking could we build a source of gravitational waves in the laboratory? And the answer is, emphatically, no. It’s really embedded in something that we all learn pretty early in our careers that gravity is a very weak force. And so what are the ingredients that you need for making gravitational waves, is -has so much mass compactified that a laboratory source is not possible. You can imagine taking.

So the example that we love to use is, take a take a dumbbell but make it the most ridiculous dumbbell you ever could think of which is, it’s a it’s a rod that’s a meter long. And you put a one ton mass on each end, and then you spin it at a thousand times per second, and then you go stand if you have a couple of hundred meters away from it, and ask how strong will the radiation be? And it is such a ridiculously small number, it’s like 10 to the minus 42, it’s so small that I’m almost like reluctant to say right. And so. So, no, it’s not possible and that’s why we have to go out to the universe and look for objects that have enormous amount of mass. The black holes we saw have 30 times the mass of our sun. And there, you know, at the time that they’re they’re getting close to colliding, they’re 150 kilometers apart from each other. So if this it’s ridiculous and travelling near the speed of light of half the speed of light.

GREENE: Yeah.

MAVALVALA: And those are the conditions that we need. So, you know, even our ridiculous dumbbell is far from what can be done.

GREENE: Yeah. All right so given that we now know that we’ve got to look out into the cosmos and look at extreme astrophysical bodies ,neutron stars, black holes. Now the question is, how do we get a handle on what we expect to see from Einstein’s General Theory of Relativity? And Frans that’s where I want to move on to the numerical side of the story, which is interesting. I have not read all press accounts by any means, but I haven’t seen that side of the story given as much attention as it might deserve. I think most everyone would agree and you agree with that don’t you? So I just want to spend a little bit of time on numerical relativity is what the name of the subject is and first, we’re going to begin again with a little clip that’ll set up the conversation and then I’d like you to help us take us through why computers and numerical methods are so essential. So we can run this next clip if you will.

[00:30:09] NARRATOR: The first part, the word numerical. It’s because we use computing techniques. So what happens is that the theory of gravity that Albert Einstein came up with others about 100 years ago, last year is extremely complicated set of equations. And they’re very nonlinear and that means that if you have one black hole you could solve analytically. But if I have two black holes they don’t add together. That’s a linear theory. So to solve for two, I have to solve the full Einstein equations with all its nonlinearities and that takes computational techniques. In fact it took us until 2005 to solve it. And it was Frans Pretorius from Princeton. We have to use numerical relativity simulations, which basically simulate the Einstein equations on a computer. And we build what we call templates, that are used in the model search in LIGO. The model search that found the event, used, on the order of 200,000 templates. Also we use wave models, not only to detect the signal, but also to infer the parameters, the properties. Because all the information about the source, had imprinted in the waveform is like a fingerprint. In the waveform.

GREENE: So you’re mentioned in there as the guy right? Who made this all possible. So why do we need numerical methods? Why do we need computers?

PRETORIUS: Well perhaps I can just take a step back and just say a little bit about, you know, the kind of observed that LIGO is.

GREENE: Yes.

PRETORIUS: Right it’s not like a telescope where you know you focus-look at a certain area of the sky you focus, you look, you form an image and you see what’s going on. It’s in some sense a one dimensional seismometer. In some sense, we’re listening to a sound. And so. You know you get a gravitational wave, you know it’s gravitational wave, but it’s a series of wiggles. What does it mean? It’s not giving you an image of the source. So we need solutions to the Einstein equations for various possible sources of black hole collisions that tell us what those wiggles look like. And so we can identify what’s going on. So that’s that’s one thing we need solutions to the fuel equation

GREENE: And the field equations, by the way, are have just magically appeared behind.

PRETORIUS: And look how simple they are.

GREENE: Well that’s actually part of the question that I wanted to explore. I mean, it it looks simple but, we physicists have this way of hiding complexity behind the symbols. And for this part, actually, I just want to take us through, because sometimes in these programs you know it is meant for the general person who just has an interest but we like to sometimes do a little bit of the math and if you don’t like math you can shut this part off. But let’s just do a little tag team here. So this is the form of the Einstein equation. And if you’re doing it a little bit more precisely, it has a few more symbols in it, if you’re not using the natural units that allow us to set certain terms to one. But even that, if we go a little bit further, is hiding this. So this is actually a more complete way of expressing the equations if we’re now going to impact the meaning of that first symbol Gµv. But then you look at that, and you say, “Well wait, what do all those individual symbols mean?” This is called the scalar curvature and that thing G is called the metric. And there’s this little combination of them that gives you the scalar curvature. But wait, what does that really mean? Well that’s hiding this complexity right here, because we have this way of hiding the summation symbols within what we call the Einstein convention, using indices that contract in a particular way. So that’s what that symbol means. And if you go a little bit further and ask yourself what is the first term, Rµv, That’s the Ricci tensor and it’s given by this combination of stuff, but what’s that new guy on the right hand side? That’s the Riemann curvature tensor. What is that equal to? Well that’s equal to this combination, where those gammas are known as the Christoffel symbols. And then you take those and what are they equal to? Well, they’re equal to a particular combination of the metric contract with various derivatives of it. It’s all just to tell you that this is kind of complicated. And the goal of the Einstein field equations, what all of this in some sense is about, is understanding that guy called “G” which is the metric tensor. And just tell us quickly what that is and I have a little visual that will take us through that as well.

PRETORIUS: Right, so so Einstein’s theory’s a theory about space and time, and space time is a geometric structure. And the metric tensor, in some sense, at every point in space time it’s a generalization of the Pythagorean theorem.

GREENE: So actually, you have not seen this before you but you were following exactly what I hope you’d say. So you said the Pythagorean theorem so why don’t we just put that app here for the heck of it right now. So you got two points there and take us through what you’re seeing right here.

PRETORIUS: Right. So you want to know the distance between two points. Now in general relativity, the distance between two points in space is the distance between two, what we call, events in space and time. And so what the metric does, all of Pythagoras’s theorem, we know it’s delta X squared plus delta Y squared plus delta Z squared

GREENE: Slow down, I’m trying to click, I’m trying to click and stay up with you.

[00:35:39] GREENE: There it is. All right.

PRETORIUS: So  that little formula there encodes the geometry of Euclidean space. Now you can also write it in terms of the metric and it’s a relatively simple straightforward metric. And what people in really, long before Einstein’s time when they started looking at general geometries of curved surfaces. They say well how do we describe if we now have a surface that’s got a complicated curvature? We want to generalize what Pythagoras did. This metric is a convenient mathematical way of doing it.

GREENE: And here’s one just example just so people who are following the math can go along. Here’s a curved shape if you will. And the distance between the two points is no longer just delta X squared plus delta Y squared square root, which is what we learned in junior high school, but rather is given by some unusually looking strange combination of delta X squared and delta Y square. And if you curve it a little bit differently, we learned that the distance between those two points might be given by this particular combination of delta X squared and delta Y squared. And finally, just to get to what you’re describing in the general case, we can have combinations that aren’t even just delta X squared and delta Y squared. We can have crossed terms like on the far right delta X, delta Y. And finally to get to that object called “G”, the metric tensor. This is sort of a generalized version of the Pythagorean theorem. Where are you’re going to allow the surface to be curved and that “G” thing there encodes the geometry that Frans is referring to. And Einstein gives us equations, those complicated equations that we saw before, where you can determine the shape of spacetime, if you understand the distribution of mass and energy.

PRETORIUS: Yeah yeah exactly right. And then also.

GREENE: Oh thank you.

PRETORIUS: And perhaps just another example, a historic example, where we might be able to appreciate how complicated those equations are. So the idea is we’ve- at every point. It’s basically a four by four matrix that we need to know over all of space and time. And if we know this, this metric tensor, we can ask any question about the geometry. Is there an event horizon? What do light waves do or the gravitational waves? So the first solution the exact solution was by Karl Schwarzschild in 19- basically 1915 within a few months of Einstein published the equations. And he was able to solve it exactly cause he assumed -well he said let’s do for a spherically symmetric solution. And so the matrix simplifies. He found the solution that turned out to be the first black hole solution. It wasn’t.

GREENE: So he actually beat Einstein to the punch, right? He found the first solution even before Einstein.

PRETORIUS: Einstein couldn’t solve these equations exactly. He solved sort of approximate version to compute the perihelion procession of Mercury. So it’s Karl Schwarzschild that found the first exact solution in spherical symmetry. And that was in 1915. Then if you reduce, if you relax symmetries a little bit so make it something called extremal symmetric to make it look for spinning black holes, it took people almost 40 years. So Roy Curry, 1963, found the solution for rotating black holes. So just not that people weren’t looking for in them and they weren’t trying. It’s just things very quickly get very complicated. And now if you want to understand the metric for two black holes that are colliding, there are no symmetries. Every single term is important and they just- we have absolutely no way of doing it pencil and paper with analytic methods. And thankfully we have computers, which the methods are being developed in the last several decades to solve these kinds of equations using computers.

GREENE: And what what did you do? What was your breakthrough? I mean when I was a graduate student,, you know people were always talking about numerical relativity and then there’s this stumbling block that they couldn’t get through. And now people aren’t so worried about that stumbling block.

[00:39:30] PRETORIUS: Yeah. So it’s a little technical perhaps I can just explained by analogy.

GREENE: Yeah.

PRETORIUS: So, the one thing about those field equations are very complicated, but in some sense they’re even more complicated in the sense that a feature of Einstein’s equations or that- so this is a geometry of spacetime, it’s a coordinate system. But you can use any coordinate system you want. So for example, the Cartesian coordinates system, a spherical polar coordinates system. So one physical space time, you can represent three infinitely many ways. Every single representation is a different metric tensor. Yet they all solve the same Einstein equations. So in some sense, there’s infinitely many possibilities for one spacetime that comes from these solutions. So that’s a feature- it encodes, in some sense, one aspect of the principle of relativity. The problem that if you want to say well now let’s try to solve these equations. So we start with something at an initial time.

Two black holes that are separated. And now we want to understand how do they collide. If you try to put it on a computer it’s what’s called, ill-posed. In some sense, the computer doesn’t know which one of those infinitely many solutions to solve. So the way you get around this and this was known for a long time is well you have to first choose your coordinates. OK. So we do that then but then the real nasty problem comes. That’s what people that try to do in the early 90s when it was realized OK LIGO is being funded. We need to have these templates. Let’s do the numerics and I was OK. Well let’s  do that. We choose the coordinates, we put it on a computer, and things just crashed. You know, it starts with two little black holes, they go for a little bit and then the dreaded NaN shows up. And if you know computers, not a number. The computer’s telling us it’s not getting a number and you can’t go beyond that. And it took actually people a long time to realize that because when something crashes, I mean that could be a bug, you just you didn’t program something right. It’s very difficult to really figure out what’s going on. And it really took probably, a good 10 years I think before people realize that these things called constraints in the field equations which are the problem.

So that’s one way to think of it is, let’s imagine a bowling alley. OK so we want to knock down the 10 bowling ball-that the bowling pins with a bowling ball. And so we throw the ball down the alley and let’s say we knock and spring it into the air so it’s going to go along the bowling track. So one constraint, if you will, is that the bowling ball rolls on the surface, it doesn’t jump off the surface. OK so. It doesn’t seem too difficult. But now, we’re going to get a little bit more careful in that we want to say we want to hit the center pin. We want to get all those 10 pins knocking down so our constraint is, we want that bowling ball to roll right down the center of the alley. Not off to one side or the other. And in some sense, those are the constraints of the Einstein equations where we put them on a computer, there are all these other possibilities which aren’t solutions to the equations that you can get. There’s one solution that is a solution to the Einstein equations.

But what makes Einstein equations so nefarious, and that’s the big struggle, is it’s not just a straight bowling track it’s actually convex. And so this little path that you have to get down is sort of at the top of this cylinder. And people actually didn’t really realize that was happening. So in some sense, we starting our simulations we starting the two black holes, think of it as a rolling pin, and it’s just falling off the side the whole time. I mean people realize that that really was the issue then it was. Well can we somehow- we don’t want to change the solution to the Einstein equations, the straight path to the center. But we can fiddle with the equations to change the the track on either side so we can try to bend it upwards, in a sense. And then some of these huge efforts started in the late 90s to try to find ways of changing the equation such that we can make this a well-posed problem. And so, you know, a very long story that’s quite a lot of the like a lot of technical issues under the rug. That was sort of the breakthrough, was finally coming up with a way to actually make this thing stable.

GREENE: And with that you were then able to calculate numerically with supercomputers. Presumably what shape of the wave you would anticipate receiving if certain kinds of processes happen out there in the universe. I think we’ve been seeing some of those templates. And I hear numbers like you had hundreds of thousands of templates. Why so many templates? What are these things that you need so many of them?

PRETORIUS: What was astonishing about this event, and I think no one expected this, is that you can actually see the waveform directly in the data. So nature has given us a gift which we did we didn’t even expect something as good as this. What was actually anticipated was that you wouldn’t really see such a clear signal. You’d actually just see it in the waves.

GREENE: Could you just tell us what we’re seeing up there?

PRETORIUS: Right right. So the little gray line is-that’s essentially the noise in the detector. The blue lines, that’s the inspiral of the black holes. They start out far apart.

SHOEMAKER: Let me pull back. Well it’s time along the horizontal axis, and it’s  a tenth of a second for every two clicks there. And on the vertical axis, it’s how much space is being distorted by the passing gravitational waves and the noise from the instrument isn’t actually the space time jiggling around in length. It’s electronic noise, mechanical noise, and so forth and so on from the instrument. So I’m sorry. Go ahead.

[00:45:20] PRETORIUS: – we deal with this every day. So we just get way ahead of things. OK. So as time- so these black holes are orbiting each other. They’re emitting gravitational waves. These are these stretching and squeezing of space time, which is what’s being picked up there. These waves carry energy from the binary. And so by conservation of energy, the binary gets tighter. They spiral in closer to each other, they started moving faster. So the frequency increases because they know they’re going faster and the amplitude increases. So it’s a runaway process and eventually they smash together. And that’s when you get to the very peak in the waveform. And then to these two black holes, they smashed together that forms this very distorted, rapidly the oscillating horizon. And one of the remarkable features of black holes in general relativity is something called a No-hair theorem which says that. Any isolated black hole has to be this-for this perfect object stationary doesn’t emit gravitational waves as represented by this curve solution, that I mentioned was found in 1963. And so what we see after the peak is what I call the ring down wave. So the black is essentially losing its hair, it’s shedding all this interesting structure and it’s just settling down to this perfect curved shape.

GREENE: So there’s this beautiful picture here and we’re going to now, in just a moment, go to the details of the actual detection itself. But it seems clear from the outside that the numerics was vital to interpreting and believing and be able to pick out the signal from the data. Now now the breakthrough that allowed this numerical analysis to go forward happened when? In the mid 90s?

PRETORIUS: It was 2003.

GREENE: Now you guys have been building this detector for a long time before that. So were you like prescient? Imagining that people would be able to make this breakthrough? I mean would you have built this device and not have been able to pull out the signal if somebody hadn’t come along and push the numerical methods?

BARISH: Luckily, we were we were too slow. Took us twice as long to do this as we thought it would.

GREENE: But what was your thinking, just so I can get a handle on it.

BARISH: We knew this was a problem. And it was pushed very hard that we find a solution for this particular black hole. Yes.

PRETORIUS: So I mentioned that there are other or other methods of solving the field equations approximately in different regimes. When the binaries are far apart, so they were in very slow early part of the inspiral, there are all these analytic methods called Post Newtonian or perturbant methods, where you can start getting this- the beginnings of this chart.  

GREENE: Which we’ll come to in just a moment but-

WEISS: Let me say something.

GREENE: Yes, go ahead.

WEISS: In defense of our ignorance. We worry no- to us, this business of finding a black hole the first go round was a surprise.

GREENE: Right.

WEISS: We were thinking because- I’ll tell you we could not figure out- we could not tell each other or tell an agency who was supporting us how many of these things we would find. So actually, we’re thinking of another source entirely as what we were designing the apparatus initially

GREENE: And that you had more of handle on.

WEISS: And we knew a lot more about this. OK. And that was neutron stars which are, again things that are of the source of the size of the sun. Sort of the size of Manhattan Island. You have a mass of the sun, the size of Manhattan Island. And that we that we traded on the fact that Hulse-Taylor, which is an experiment that won the Nobel Prize back in 2003. Was it that? Maybe- I’d forgotten within my life, certainly.

GREENE: And mine too.

WEISS: And what they what they saw was something, which was one of the most elegant experiments I think ever done in astronomy, was they saw two stars. One was a pulsar. That pulsar is a neutron star that emits pulses that you can see in the radio, in the radio spectrum. And they noticed that there was a modulation to the pulsar. It was going around at 50 -they had pulses every, what was it every 8? Every eighth of a second or something like- 8 times per second, really, what they saw. And then there was a modulation to that. Sometimes it would speed up, sometimes it would slow down.

GREENE: Yeah.

WEISS: Which indicated that that one pulsar was somehow living with another thing around it. And that turned out to be another pulsar or another neutron star. You just didn’t see the pulses from it. And with that, that was sort of- that business and plotting it for time, they began to realize that these two neutron stars were emitting gravitational waves. It’s the only way they could explain the change of the orbits that were using energy to gravitational waves. And with that, with that, that gave an enormous impetus to the field, our field. And in fact, in the end- I mean let me say a little about the early days because that’s sort of what you’re alluding to. In the very early days, for example when Joel Weber when he began to

GREENE: We’ll come to that in just a moment

WEISS: When he was thinking about it and when people were thinking what might he see when he did see something, they were thinking of supernova. That was the only thing. These stellar implosions that happen once every hundred years in our galaxy. And that was sort of the only thing people could see had enough acceleration. It was and it was-had enough mass that would possibly make a gravitational wave. And that was the beginning of the field. But it turned out to be quite wrong. What the estimates were for that. We still don’t really have good estimates for supernova. They’re getting better. But-and so when the neutron star binary was discovered and people began to see there were binaries of these, that became our goal for detection. And the thing that

GREENE: So let me just just just.

[00:50:49] GREENE: So in that particular case, there was indirect detection of gravitational waves by virtue of the inspiral but you guys wanted to see the direct.

WEISS: But I can just say- you asked what was the impetus for us to continue.

GREENE: Yes.

WEISS: And the impetus was really- and all our calculations were then done for a neutron star binary.

GREENE: Gotcha.

WEISS: And in fact, what we still have, when we get to design sensitivity with this detector, we expect to see something like 10, maybe 20, of these a year, of neutron star binaries. So we’re not at design center. We’ll talk about that.

GREENE: Yeah sure, we’ll get to that in just a moment.

GREENE: Anyways, so it wasn’t that the thing that everybody wanted that we should detect black holes. And I know because I think to me that’s the big source to look for. And I remember giving a talk at Cornell Hans-Peter was in the audience at the time and I told him the story that I just told you. And he said he pulled me aside later said, “Don’t you-you’re going after the wrong source.” He said “Go after black holes.” and I said, “Look, tell me how many there are.”

WEISS: He couldn’t tell me that. I mean I know he does. But he’s right. I mean if there were if there are black holes that’s better than mass and stuff like that. So any- in other words, we started the field with ignorance. There’s no doubt. But the idea that there should be gravitational waves coming from some astrophysical sources was the basic idea. We didn’t know exactly where and that was a big controversy in the beginning of the field. You know Kip, who made our, Kip Thorne who was one of the visionaries of this thing has done a lot of calculating of different classes of sources. And, in fact, got us got himself into trouble with some of the real experts in this business because they thought he was a little too optimistic.

WEISS: But anyway, look, all the things that he calculated will probably come about is my guess.

GREENE: At some point.

WEISS: At we some point, yes.

GREENE: So any thoughts on the numerical side of things?

MAVALVALA: Yes I just wanted to add something. Frans said something really important that I wanted to emphasized. So you know when the black holes or neutron stars are far enough apart, we knew what the waveform there would look like. We also knew what the waveform would look like once they crashed and a single object was forming. The part we didn’t know, and that was so important in the breakthrough, was what happens when they’re really close together and about to collide. So in some sense if we hadn’t made that breakthrough, you can think of it as we had all of those bumps and squiggles. We would have had a little gap in the middle of that time line but we wouldn’t know exactly what was going on. And it’s a pretty important region because that’s where the amplitude is the biggest. So. So there was ignorance, but we also had a handle on enough of it that we thought we could go on.

WEISS: Good point.

GREENE: All right so let’s go to the last, oh go ahead please.

SHOEMAKER: One last thing on this point, though, is the real- just an experimenter. I have to say. We built two instruments. If we see the same signal in these two widely separated instruments, but one which we can’t explain by any theory so far. If all of our checks show that it’s a valid signal, then we’ll say this is a gravitational wave, figure out what it is.

GREENE: Even if you don’t know what it is.

SHOEMAKER: And I think that’s a really important feature of the instrument. We built it in such a way that we think we can gain confidence with these two and then with others that will come online.

GREENE: So let’s turn to the instruments.

WEISS: In fact, I’d like to amplify a little on that because that’s really the interest.

GREENE: We will move on at some point.

WEISS: But but that is really- what David is saying is really the heart of the interesting part of this. I mean I mean here we’ve opened a field and we call it gravitational wave astronomy. And if we don’t have something like what David is saying, some new things that we don’t understand at all, it’s not that the field is a failure. But it’s what would defy all the prior history of advances in astronomy. Turns out every time you open some new field, you find something you didn’t know.

GREENE: Right.

WEISS: And what David said, and we’ll go back to that I hope, we have all sorts of internal tests within our own experiment to establish that we, as best as we can, that we’re looking at a gravitational wave and not something else. We should talk about that.

GREENE: Well I would like to turn to now, the actual detection of these of these waves. And we’re going to again begin with a little piece that speaks to the earliest attempts by someone named Joe Weber. Can we roll that video please?

(Video Clip)

[00:54:42] NARRATOR: This is an artifact. It’s one of the bars that Joe Weber used in the 1960s. Joe was the first person to take seriously as an experimentalist, to try to actually make a detection of gravity waves. What we have here, is a very large aluminum bar. The theory was that if a gravity wave comes by and space expands and contracts, if these were free aluminum atoms, they would just move with the local space. But they’re not, because they’re confined by bonds to other aluminum atoms. And so if you think for a moment on those bonds like little springs connecting all the atoms, when the space distorts, the atoms try to move with the space. The springs hold them back, so that puts some tension in the springs. And after the wave went by, Weber’s idea was that strain on those springs, that energy would be released as a soundwave in the bar.

Joe was a controversial character and people don’t give a lot of credence to his claims that he had actually detected gravitational waves. But to me, Joe is a hero because even though he’s a model of what not to do in data analysis, he was the person who had the chutzpah to go and make this first attempt. If it wasn’t for that pioneering work, we might well not be here. And he did think of many of the ideas that he would later cooperate into practice.

GREENE: So Rai, Barry, you were around when these experiments were going on and what’s your view of Joe Weber and this approach and the impact it had on how things developed?

WEISS: Well I mean you really want it for me. I mean

GREENE: I am happy to have it from you

WEISS: Look, in fact it was the beginning of thinking about, at least in my life, of doing it metrically, doing it a different way. I’ll tell you, I mean it’s a little bit of a story, and if you’re asking me you should stop me when I’ve said too much, okay? But the thing is, I was asked, as a young man, to do something which you can do with your eyes closed but I could not even do with my eyes wide open and with every piece of my brain working properly and that it was to teach a course in relativity. OK. I’m an experimenter and I was asked to do that. So that was in 1967, ’68 at the time when Joe not yet published, but he was- there were news all over all over the physics community that he was detecting something. But he hadn’t yet published his result. In ’69 is when he published his result. So here I am, not knowing tensor analysis not that equation, and I will love what you did there, but as you took this thing and unpacked it and it got more and more complicated. Well OK. I had to try to deal with that. OK. And with never having learnt tensor analysis. OK. So anyway, so I can’t tell my department chairman I don’t know anything. I’m not allowed to do it. So what comes up is I teach this course as best as I can and I get to this experiment. I can’t explain it. In fact, the kids in the class ask me to say we’re up to it. And I didn’t. The explanation that Fred Robb just gave I couldn’t have made that, OK, because I didn’t understand. Here I’ve been teaching a course where there is no more on gravitational force, it’s gone. It’s all in geometry. And here’s this thing that the triers are forcing at. So I said nothing. Come on.

So I sat back and I tried to figure out how to explain how you would detect a gravitational wave to people in the most simple minded way I could. I mean with the little bit least amount of that formalism as possible and that the thought occurred to me in this course where I was sort of one day with the students, if that. And I came up with the idea of the most elemental thing you could think of, which was, and this is why I tell you the story. It was like why don’t you throw some masses out there. Put them out in space and put clocks on them. Good clocks I knew about- you could make good clocks. But not as good as I needed. They were good clocks. OK. And so what happens is I say, “Let’s time a light. Light going from one clock and mass to the other one. And then use that as a basis for looking if there’s a gravitational wave or not. Because the gravitational wave as it comes down on that system will cause the time to change because the space between them is shrinking and expanding. And that’s a very straightforward way of thinking about it. Not this molecule molecular thing where you don’t really know what the molecules are up to and everything like that. And I got the kids in the class to do this, a simple problem to do in relatively, just that problem. And that put it aside and now. Now comes the connection to the question you asked me. Well OK in ’69 he writes his paper, which was a stunning paper by the way. And though it’s caused at least 20 groups around the world to do and emulate that experiment because it was such an important result. And by ’72 it was quite clear that most other people did not see anything. And the problem was not.

I can’t I can’t tell you exactly the reason why they didn’t see anything. I mean I know I can tell you that’s exactly why they didn’t see anything, because there was nothing to be seen. But I can tell you what went wrong with Joe’s experiment. I mean I had a theory about that. I don’t want to go- that’s a waste of time right now. My guess is he was not a good enough experimenter, that’s my guess. OK. I mean he probably believes too quickly what he had and he didn’t do enough checks. That’s my guess.

GREENE: Right.

WEISS: And I’m not I don’t know. Let’s leave it at that. You can ask me about that later.

GREENE: Sure.

WEISS: But on the other hand, what happens is that that triggered me to think about well maybe this crazy idea that the Gedankenexperiment that I was thinking about in that course, maybe you could make that into a practical experiment, right? Maybe because I didn’t understand the thing in the first place. Maybe that would detect gravitational waves.

WEISS: And that’s where,, at least in my life others have thought of it too, I’m not the only one by a long shot. But that’s how the thing evolved in my life, that we developed this from a metric thing. I could understand every piece of it, you see?

[01:00:41] GREENE: And we’ll get to a version of that. But Barry, your view of those early days?

BARISH: Yeah. Joe Weber, first, I think I was really an outsider doing particle physics and I thought he was one of the most imaginative, technical, creative guys. And the dilemma for me is how he could be such a terrible scientist at the same time. Which I think he was. This wasn’t the only time he thought he saw gravitational waves. It went through his career, he was bitter, and yet he made a tremendous impact. He had the insight how to build this stuff.

GREENE: And he was never convinced that he didn’t see them is that?

BARISH: Yeah, because better experiments didn’t.

GREENE: But what was his view? I mean he held to this perspective.

BARISH:  He felt he felt he was being mistreated and we didn’t understand, that

WEISS: Well it was even worse than that. Do you mind if I interrupt? Because I’ve lived through it. I mean, you were really doing physics and I was screwing around. But the thing was that the thing that happened was that a very famous physicist, and I blame it on and blame and give credit to, Dick Garwin.

WEISS: Dick Garwin

GREENE: From IBM

WEISS: I know he wasn’t-well

GREENE: I know him from IBM days.

WEISS: That’s exactly right.

WEISS: Dick was- he had free license at IBM to do what he wanted. In fact that’s when IBM still had their labs here in New York. Yeah across the street from your guys.

GREENE: Right.

WEISS: And so he and a guy named Levine, just as a sort of a pleasure for the afternoon they decided to make a little tiny bar, just like what you saw but a well-designed bar, well understood, and well-calculated. In other words a lot of intelligence went into doing that design. He did it and they knew exactly how sensitive they were. OK. And so what happened is- that the acrimony in this whole thing really happened because Weber and Garwin would meet at meetings. Now this is now into the into the late 60s and early 70s. And they would meet and Weber would give us the most recent results. In this case, by the way, he became very serious with the results when he had three detectors. One in his lab in Maryland, another one in the golf course about eight miles away from his lab, and then another one in Argonne laboratory in Chicago. And that’s when he began to really say he’s seen something because he saw co-incident pulses in all three of them about two or three a day. OK. And what happened is that Dick Garwin, who knew his apparatus much better than Joe Weber did, unfortunately for Weber, and he could tell you how much sensitivity- you could you could go to Joe and say, “How sensitive is your bar?” And he couldn’t really tell you. I’ll tell you how much we thought it was sensitive- 10 to the minus 16 was the very best it could have been in these units. And the first detection was 10 to the minus 21. OK. Right. So it’s many orders of magnitude away. But he didn’t know that himself. And now his defense when Dick would get up and say at a meeting, “I built this little thing and I know exactly how much I see. I see nothing at the level of 10 to the minus 16.” Weber would- his response was not a rational response. It wasn’t. “Well I know my sensitivity. I see something at 10 to the minus 17, or something like that, and you’re not sensitive enough.” That’s not the response. The response he made is, “You didn’t do it the way I did it.” And that was the wrong way to approach this whole problem.

GREENE: Very interesting. Wow.

WEISS: You weren’t there.

GREENE: I wasn’t, no. So let’s think about the right way to do it, which you guys have now achieved. So we’ve got some pictures. Tell us what we’re looking at. Barry?

BARISH: That’s a four mile long arms, two of them. Perpendicular to each other. And we make- send light from a laser, split it so it goes down both arms and comes back. And if our arms are equal length, we make them cancel each other.

GREENE: The light cancels against itself. And that’s how do you think about that? Can you give us a way to think about it?

BARISH: The light comes in a waveform and as it goes down one arm and comes back and goes down the other arm and comes back, we reverse one compared to the other so that there’s a black-

GREENE: So like a peak and a trough or kind of crossing each other?

BARISH: Crossing each other and it’s back. And if one of the arms gets a little longer than the other one, they get out of time. And then we see light and that’s the simple simplest picture.

[01:05:07] GREENE: Yeah we’re going to try to do a version of it here if you don’t mind. I’ve tried to do this experiment before, but I’ve always been very nervous about whether it’s going to work or not. I’m not the slightest bit nervous because I’m not doing it. One of you experts is going to do it. Can you bring out this

WEISS: No, no, no, no.

GREENE: Interferometer, if you will, from behind. So here we have  a slightly smaller version of what you just described. So, I’m going to turn it on and then ask one of you guys to explain what’s actually happening here. Wow we actually have some fringes. Can you get those up on up on the screen. I should not be standing here. Nergis would you do this and just take us through what’s happening there?

MAVALVALA: OK. So it’s an interferometer.

MAVALVALA: And what we have here is a laser. It’s actually a laser

GREENE: Can you bring the lights down a little bit over here?

MAVALVALA: – Like a laser pointer and we have a mirror here that’s a special mirror that takes half the light and lets it go through it. And the other half of the light gets reflected from the mirror. Oh you know in the old days when people smoke you would just like light up cigarette and make this all happen. But

GREENE: This is bug spray, by the way.

MAVALVALA: Good, it’ll kill the roaches too. So. So what happens is. The light then reflects off of these two mirrors and comes back to the beam splitter and then, just as Barry explained, if the peaks of the light line up with each with each other, the light in each arm, then you get a bright bar of light on the screen. And in the places where the peaks line up would troughs, the two waves cancel and you get a dark fringe. And so what you can see as you go through looking at those, wherever you see dark stripes those are regions of space where the peaks lined up with the troughs and the light beam cancelled itself. And so our measurement at LIGO was pretty much just that, which is that we operated under normal circumstances where we make the two light beams cancel each other. So it’s dark. And then when the gravitational waves comes by, it gets a tiny bit brighter.

MAVALVALA: And we can measure that tiny bit brighter. One of the thing that’s very interesting to see on this picture here. The darkest- between the dark stripes and the bright stripes is about it’s  a fraction of the wavelength of the light. And so right here on the screen, you’re seeing separations that are of order a fraction of a micron. So about 10 to the minus 6 meters. I’m trying to convert that into units we understand. It’s the spacing is a fraction of a hair of mine. So you can see right here on a screen with a simple apparatus like this and more or less with our eyes in going between bright and dark. We

MAVALVALA: are seeing differences of about, you know, a fraction of my hair between these two arms.

GREENE: Now if you just hold still for one second, can we make this as stable as we can? So everybody please just hold your breath. Don’t move. And we’re getting relatively -my talking is now messing it up. I’m going to talk really softly now. But there you go. And then if a gravitational wave, well obviously this is sensitive to all sorts of disturbances not just gravity waves, but the point is how I mean disturbance by yelling out can you get this to move?

MAVALVALA: Let’s clap.

(Applause)

GREENE: Thank you all very much for coming, thank you. No, but there you go. So- and what you guys need to do, presumably, is to isolate your four kilometer long version so that when people clap, or a tree falls, or someone revs a motorcycle engine, it doesn’t shake like this. So the only way that it will shake is if a gravitational wave rolls by. And presumably that is the challenge. So thank you for this little demo here. So how do you do that? How do you isolate in such a way that the only thing that can make it shake are waves of gravity?

WEISS: Well, a lot of things, but, David should-David, David.

[10:09:22] SHOEMAKER: I think probably if you

WEISS: And I’ll correct him if he’s wrong.

SHOEMAKER: One good thing to do is to think about the two challenges that we have in building this instrument. One is keeping these objects, that are points in space time that should only be influenced by the passing gravitational wave, keeping forces away from them. So they’re not moved by things like seismic noise. And then the other challenge is measuring with ever better precision, with light, the distances between them. But so let’s first talk about this problem of the masses being disturbed by the outside influences. Clearly, seismic waves, and that could be from any source. It could be from the ocean pounding on those shores, which can be seen completely in land. It could be passing trucks and so forth and so on. Wind,, shaking people passing nearby. All of these things caused the ground to move. And we want to keep that motion from getting to the points in space time that we use to watch the gravitational wave.

And the basic principle we use is pendulums. If you have a pendulum, let’s imagine it’s about this long and it has- its swings at about this rate about a half a second per cycle. If I move the top supports slowly compared with that frequency, the pendulum will just follow. But if I move that top and I shake it at high frequencies, frequencies higher than the pendulum frequency this mass is going to want to sit still. It has inertia. It naturally wants to be an isolator for frequencies higher than this pendulation frequency. If you put a bunch of these things in series, which is what we do in advanced LIGO, here you see an image of the bottom of one of these multiple pendulum suspensions. Then each layer acts like another filter against the seismic noise and the system wants to stay still in the inertial space and that’s exactly what we want. We also used more sophisticated systems where for instance we put an electrical sensor next to one of these intermediate pendulums and measure how the outside cage is moving with respect to the pendulum that generates an electrical signal into an amplifier, and then into a motor, where we tell that motion to stop. So it makes a servo controlled stable platform. So through this process of multiple layers of isolation basically using the pendulum idea in different ways, we can suppress the seismic noise by 10 to the 12 orders of magnitude, or something, some incredible amount and get to the point where in fact the mirrors are moving far less due to seismic motion than they would be moving to a due to a passing gravitational wave.

SHOEMAKER: It’s not our only problem. The next thing that happens is that we have to deal with Brownian noise. These mirrors like everything is in equilibrium with the heat bath. Each one of the modes of the optical system’s got energy, thermal energy. You probably heard of Brownian motion which causes, for instance, the needles of a very fine electrometer to shake around. These things move in the same sort of way and to manage that, we have to use materials with very , very low mechanical losses. Gather that energy into very narrow peaks that are outside of the region and frequency, what we’re trying to do the measurement tool. So a lot of it is very fine machine making it to get good instruments with very low mechanical loss and very, very, very slow motion.

GREENE: Now you did this in two steps right? I mean there’s LIGO and then advanced LIGO. So first one was, what, 2002 to 2010? It was operationally roughly. Is that right?  

WEISS: 2008 yeah

GREENE: 2008, and then you went off line for the upgrade. But it wasn’t an unexpected upgrade. This was your plan from the get-go and that’s a kind of forethought that is wonderful. But why did you think of it in a two stage system? Why not just go for the you know the one that would work from the outset?

WEISS: Well that’s almost really part of the whole business of how the project went from table top to being good and large scale. And I’d like Barry to talk about it, but let me give you the philosophy we had. In fact, it succeeded exceedingly well. The philosophy was indeed what you say. Once we got to the point where we want to write a proposal for the large system, which was done in 1989, that’s when the real proposal was written. Prior to that, there had been studies of large systems and how much they cost. That was earlier. But the real proposal for this was written in ’89 with exactly the strategy described. And the reason we did that is because you wanted to at least- and look, this is an expensive project. People were skeptical of the whole concept to begin, with so you had to be very careful. What were they really skeptical of? I’ll

WEISS: be honest with you. The skepticism was the calculations that Kip had done and all of the people had known about gravitational wave sources. Kip Thorne just on about yeah. That was considered flaky. I was- the whole idea of being able to do measurements at 10 to the minus 18 meters, which you know you need to do to get to 10 to the minus 21 strains even over four kilometers. That was considered way flaky. I mean nobody went in- most engineers, when they hear about 10 to the minus 21, they think you’re out of your mind. I mean there’s nothing you do at 10 to the minus 21.

GREENE: And 10 to the minus 21 is a fractional stretching of the length-

WEISS: Let’s make it simple. Let’s say over four kilometers. It’s the motion of the mirrors is 10 to the minus 18 meters. And should I-

GREENE: Just to put that into context. How do we think about 10 to the minus 18?

WEISS: Yeah, let me quickly say that that is an easy way. Well I think we should. Let me just. I’ll do what I did once before. You know you start with a meter, and you say OK let’s divide that by a million three times, three times over. OK. So you start with a meter. OK. And then you say let’s divide by a million. We can still sort of imagine that. And what do you add to it? One micron, which is about one hundred- Well 130th the size of your hair. That’s still the

GREENE: Width.

WEISS: People can imagine that. Then you divide by another million. OK. 10 to the minus 12. That’s smaller than an atom. An atom is 10 to the minus 10. So in these units and

GREENE: Including the electron.

WEISS: You’re sort of inside of the atom, but you’re not yet to the nucleus. The nucleus is miles away and now you divide by another 10 to the 6. now you’re 10 to the minus 18 and now you’re in Never, Never Land. I mean most people just can’t imagine that. I mean you’re down to the point where you’re deep inside the nucleus.  

[01:15:20] GREENE: So it’s like a thousand the size of a nucleus right.

WEISS: And that’s the best I can do for you right here.

GREENE: That works. OK. That work for you guys? Very good.

WEISS: And so the people thought you were kind of crazy that you would even be able to do such a thing. So we had multiple things that we had to do. So we had to be very careful. And in that proposal, that’s why we did it this way. We want to take. We had in the program- many people had built prototypes detectors on a table top. You know, and that demonstrated independently, various techniques. For example, the kinds of things David just talked about good vibration, isolation systems then the ability to read with light from an interferometer. You can- you don’t have to build a huge thing to read 10 to the minus 18 meters. You can you can do that on a little instrument. That’s- you don’t- you’re not measuring a strain of 10 to the minus 21. See, for gravitational waves you have to make it bigger. But you could certainly prove in a little instrument smaller table top

GREENE: I see.

WEISS: that you can measure 10 to the minus 18 meters and that took a while but people were getting there. OK. So

GREENE: And what year was that?

WEISS: Oh 10 to the minus 18 is an exaggeration. Ten to the minus 16. Certainly we could do in those days. OK. And that was we’re not talking about a lot of people worked on that. Very good German groups, a group in Munich worked on it. They did much of the initial development of this thing. And then the group in Glasgow and then the group at Caltech and the group at MIT. Those were the principal groups that did a lot of that development. And with that development, you could say to somebody who was a real critic,  “Look we’re not miles away from doing this the individual pieces that you needed. We could do the vibration isolation. You could do the sensing. OK. And now the question is we can’t do the measurement that’s a scientific measurement unless we take that whole thing and put it into something that’s very large. And that’s when the money has to be spent. So you knew- and what we traded on was the fact that you had done these things in the prototypes. That things that people were skeptical of and that the new thing would be that you had to now build it on a large scale. And so we wanted to do- promise no more than what we had done in the laboratories for the first goal and that’s why we made two different things.

GREENE: Two phases.

WEISS: Two phases, because what we had to develop in the laboratory was probably not good enough for some of the more reasonable sources. They were good enough for some of the wild sources that Kip could think of but they weren’t for the more reasonable sources. For example, we could do a neutron star binary with that yeah.

GRENE: Right, OK.

WEISS: And so it was very close and that was unlikely. OK. So consequently, we did right then and there, say, “Look this is a technology development but has a chance, not zero chance, of detecting something.” But we were not very high probability for it. But, we said, “The second- we have a second piece of this. And now here there’s a critical thing. You’ve got to build these facilities big enough so they can encompass both the first version and the second version. And that was the where all the money was going. In other words, all the money was going into making buildings, making vacuum systems that were 4 kilometers big.

GREENE: And how much money are we talking about?

WEISS: We’re talking about down to about 200 million by this time. Barry can tell better, but I think that’s about right.

BARISH: Well up to now 1 billion.

WEISS: Yeah, but not at that time. I mean just for the construction of the facilities and the thing is that -and then we said, “Look, we don’t have to change those facilities to put the better detector into it.” So that was the initial investment was good for both. The things we knew how to do at that time and also what we thought we would do in order- we didn’t have to change the vacuum system. We had-everything would work. The expensive part I think- would have that build-put another something that would be the equivalent of another detector. But you would not have to build the infrastructure right. OK. And that was the proposal.

GREENE: And that went through.

WEISS: And that eventually went through.

GREENE: And you know Barry, you know, I don’t want to embarrass you but everybody that I talked to said that you were the guy who made this happen, right? I mean there was vision there was a plan but you’re the guy that made this happen. What would you say was the big challenge and the big thing that needed to?

BARISH: Well I don’t think we were successful and we’re successful in doing the science. But I must say I never know of anyone who thought this was going to take us 22 years. It’s 1994 when our first detection is 22 years later and I don’t know whether you or anyone thought it was going to take 22 years. But I think probably most of us felt this- these two stages would be half that. So we’re not as good as you-

GREENE: Did you ever lose. Did you lose hope as you were doing this at all?

BARISH: No no no. We’re just not as good as we thought we were. I think.

GREENE: So you just gained humility through the process.

BARISH: So when you point and say “We did it” or “I did it” or something I think.

[01:20:03] WEISS: I want to give him credit. He won’t do it. I want to do it.

GREENE: Please do.

WEISS: Because I’ve lived through this too. You know we were pretty disorganized. OK. I mean that’s why it’s so important to look at the transition from the table top version of this thing to what we call the LIGO full scale thing. OK. It’s true that we had begun, by the middle 80s, to begin our early 80s to look at. In fact I-  MIT did a study of what with industry to see how much it might cost, how you would make a bigger detector than a little tabletop, thing like that. We had and we did a study of it became- in fact, became the basis for two things. It became the basis for a presentation to the NSF jointly by CalTech and MIT together to push the idea that you should build a big one. And we had a lot of help within the NSF, by the way. People in the NSF. But this was exactly what the NSF should be doing. I’m glad they thought that way

GREENE: Rich Isaacson.

WEISS: Rich Isaacson in particular. Yeah. And he then made the investment in the NSF, the intellectual investment and showing the people in the NSF that this was a good thing for them to do. Because it was risky, but it was also, if it worked it had a tremendous payoff. And that’s funny- I think and that’s I think an honest statement of what was going on. And so what happened then was that we were able to get this collaboration going. But it turns out the principles were not-that includes myself and Ron Drever, who is unfortunate not here right now, and Kip Thorne, we were the ones who- well we’re not- we were not necessarily the most strategic thinkers, I’ll be honest. And in fact, we squabbled with each other- not several of us did. And what happened is that in fact, in 1986, when NSF had started looking at this whole thing, again Richard Garwin played a very important role in it at that time. He wrote a letter to the NSF says, “Look he’s heard about this thing, you know.” Now why he was so important is because of the experience he had with Webber. But he was honest in his thing, by the way, he wasn’t totally critical. He was trying to help. And he was critical but also he was trying to help he says, “Look if you’re going to do this,” he said to the head of the physics branch of NSF, “if you’re going to do this, you ought to have a study about it. A study with people who are not in the field. Let them judge this.” And that was put together and we got some very good people doing it. People who are not in the field. Nobel Prize winners who are experimenters, many of them a lot of people now in the field. But who were not in the field then. And they gave an enormous endorsement to them. Endorsement was, don’t mess around, build two big facilities right away. Build them right away- why don’t make that- don’t make any more prototypes. We- if you want to know you’ve got to get to the science and you’ve got to find out what are the problems when you make this bigger. And they have recognized that and they also said, “Don’t build just one, because they want to see the science.”

GREENE: Sure.

WEISS: If you want to have two. They saw that right away. That was part of the strategy. And we haven’t yet written the proposal. You see, that- all of this gets incorporated later in 1989 in the proposal. And- but the thing that they were very skeptical of was look the- here was this three of us who were not adept at running things. We couldn’t quite agree on things, among other things. And they say, “Get yourself a single director. ” And that was bingo. Oh no, that wasn’t him right away.

GREENE: When are we going to get to Barry? I’m hanging by the edge of my seat here.

WEISS: We’re going to get to Barry. First, it was a fellow named Robbie Vogt, who helped us write that proposal. And then it became Barry. And Barry really pushed the queue.

BARISH: Actually the key to success is one that people don’t really point to. And this was something that was very hard to sell to the NSF, but it wasn’t just building this big thing. The NSF actually supported us as we built this unfinished building. We were able to concentrate on all the stuff that David talks about. That rather than the usual thing whether you go and you build a big thing, is they take away all the money or the work that you do in your labs to develop all the things we don’t know how to do and come back after you fail or succeed. And they kept us funded to do all the work toward the second generation. In 1998, 1999 when we finished this.

WEISS: But you insisted on it.

BARISH: I insisted on it. I thought we would never get to the second step without that if they bought off on it. Now that to me was the key to success, not any little individual.

GREENE: Right. We’re starting to run a little short in time. I want to get to our final section, which is just looking toward the future, right? We’re going to also begin this with a little final clip. If you could run that please?

(Video Clip)

[01:24:47] NARRATOR: In the era when we were trying to convince people that LIGO should be built, for me, this was the most exciting thing is the likelihood that we will see waves from phenomena that we never dreamed of. After all, every time a new window has been opened up in the universe, unexpected things have been seen, and this will surely not be an exception. Gravitational waves are the only form of radiation that is so penetrating that it would travel unscathed through the very hot and dense material. In the earliest fraction of a second in the life of the universe. So if we are ever going to actually see the Big Bang, the only way that we will actually see it is through gravitational waves. I, therefore, regard that the ultimate Holy Grail of gravitational wave astronomy is probing the Big Bang. Some of the strings from which everything is made. Fundamental strings we call them, which are submicroscopic, far smaller than the size of an atom or an atomic nucleus that some of these strings are expanding the cosmic size by inflation. And there are calculations using tentative versions of string theory that show this may be the case. If that happens the result are then what we call cosmic strings. And these cosmic strings are simple enough that we can predict with high confidence the shapes of the gravitational waves that are produced when two such cosmic strings collide pass through each other and with high probability, recombine. So you wind up with two strings that have kinks of themselves and those kinks that travel down the strings at the speed of light, emitting gravitational waves. The resulting wave forms are very well understood and are quite simple and the estimates of the strings of those waves are such that LIGO has a real shot at seeing them.

GREENE: So I love the fact that he speaks of string theory as if it’s absolutely right, which is sort of a beautiful thing for somebody like me. But he really means that if string theory is correct that there is a chance we could see them. The evidence of strings through these gravitational waves the undulating strings would create. But you know we don’t have a lot of time left but starting with you David, when you think to the future, you know, both in terms of the devices and the things that we might find, what gets you most excited?

SHOEMAKER: Well I think there are two things, two directions we can go to go well beyond the advanced LIGO and well beyond probably makes it- from initial LIGO to advanced LIGO with a factor of 10 in sensitivity because we look at the amplitude of the waves being able to reach out a factor of 10 times further brings a thousand times as many possible sources into our reach. Probably the next step would be another factor of 10 on ground. And there are a couple of different ideas of how to do that. I mentioned that one of the limits to what we can see is the thermal noise. And one thing you could do is to refrigerate the masses down to a fraction of a Kelvin, or maybe something not quite as cold to reduce the thermal noise. Another way you can do it-this is a strain in space. If you have a meter stick, it will change by 10 to the minus 21 meters when the wave goes by. If you have a 10 meter stick, it will change by 10 to the minus 20, ten times more. So we have four kilometer long arms to take advantage of that. Now, if we had, say 40 kilometer long arms, that would make the signal ten times bigger without any of our noise sources growing that- by that same amount. So there are a couple of ideas that we have for how we can grow ground based interferometers into systems that could reach even further back and say get to the point where we could see the very first black holes see all the black holes in the universe. I think though the other thing there was this image up on the screen a little while ago of a constellation of satellites in space.

It’s obvious that that’s an experiment that must be done. It would probe not the same frequency range of gravitational waves so it’s a little bit like the radio astronomy an optical astronomy, but it would be also a place where you can instead of having a mere 40 kilometers long arm, you could have a million kilometer long arms. And have a signal that is that much larger and be able to probe, with incredible precision, the waveform of the gravitational waves use it as a- small black holes orbiting around big black holes as a way of mapping out the curvature of spacetime in more detail. It’s an experiment which must be done and we think we understand how to do it. I think we’re getting closer to the point where we can know you have a good date for sending something like a LISA, a gravitational wave antenna, and the Europeans are pushing this so much more than we are. Well you know there’s a bit of an unhappy history there, for a while, there was a really good joint mission between the U.S. and Europe and the technology was working along well. The James Webb Space Telescope got much much more expensive than anyone thought. And the way the walls work for money and NASA is that if there’s one space experiment that’s in trouble in science, it eats all of the other ones up. And that caused us to break a number of relationships we had with the Europeans to build space missions, and this was one of them. So the Europeans have gone ahead. They actually have. They have a date of 2034 for a launch of a space based mission. They’d actually like to bring that in a bit. And their calculations for technical readiness say 2029 2030. The U.S. has got a small effort that we’re trying to catch up with them in the sense that we’re at least going to play a minority partner in it and to participate in this really wonderful experiment. Maybe we can even grow our own participation based on the successes of LIGO and also a recent experiment- a test experiment was put up by the Europeans called LISA Pathfinder. And while the data are not yet publicly released, I think we’ll see you know performed really remarkably well. It’s it’s on it’s way it’s still a ways away.

[01:30:44] GREENE: So Frans, what about the future of the numerical methods? Do you see that playing a vital role going forward? New templates? New ideas in that domain?

PRETORIUS:  Yeah. And in particular, if we think of adding neutron stars to the mix. So we are going to hopefully learn a lot about black holes in the universe with upcoming observations other other events. But in terms of the template, we pretty much have that control. But the one thing which we don’t have under control is neutron stars. And you have the. Einstein’s equations the the geometric side the G on the left hand side and this Tµ which is the matter on the right hand side with black holes at zero. We just have to hit zero. We don’t worry about matter. It’s purely vacuum. Neutron stars now, of course, they are they made of matter. And the thing that’s both interesting and challenging about it is matter in those extreme conditions set the density of neutron sources into- the gravitational force is so strong that packs all of the neutrons together in basically one big atom with a mass of the sun. So example- if you can compress the Earth down to those densities, it would be about the size of a football field. It’s incredibly dense matter. We don’t understand those properties. And so when we tried to do- to try to solve what the gravitational waves would look like, especially when, say two neutron stars collide, they form a hyper-massive neutron star. They might collapse to a black hole. But a black hole rips a neutron star apart. We need to model that right hand side. Isn’t it’s interesting with the Einstein equations, that we use words “modeling” and “simulation,” but with the left hand side, the geometry, we’re actually not modeling anything. We’re actually solving Einstein’s equations.

There’s some numerical error, but it’s actually solving those exact equations. They’re the corresponding equations for neutron star matter. That’s the fundamental laws of particle physics, quantum electrodynamics, quantum chromodynamics, electromagnetism. There’s no way in hell we could ever solve those equations in the same way that we do the Einstein equations. So we have to do now, in the real sense of the word “modeling,” come up with an approximation for those things. And because we don’t understand math at those densities, the models are incredibly crude. Basically blobs of fluid. So. The two things that are interesting about neutron stars one is just when they collide, they’re are going to emit photons. And so we will hopefully see both gravitational waves, perhaps a gamma ray burst. Perhaps these interesting things called Killer Nova, which might be what’s responsible for all the gold and platinum and all that all the bling in the universe. As opposed to the stars which produced the stuff that make that make us exist. So there’ll be a lot of interesting information that will come from the electromagnetic signatures but also, they will tell us really what matters doing at those extreme densities. So they might be clues, not just to what happens with the populations of neutron stars and black holes, but it might be telling us something about matter at densities that we can probe in the laboratories on Earth. And so that’s a very exciting theme going forward in the simulations.

GREENE: And Nergis, is gravitational waves going to be your focus going forward?

MAVALVALA: Yeah, I think so-one thing that we know- there’s an “L” in LIGO and there’s an “L” in LISA, and that’s the laser. The laser light is our meter stick. And the laser light is is- light is quantum, it’s made up of photons and the quantization of the light is a big problem. It’s a limitation to how well we can build future detectors. So for me, that’s an area that I’m really interested in. How do you engineer quantum states of light that can help you do better? And we’re doing that in our labs now and so yes, there’s plenty to do. And for me personally, the field is, it’s just begun. We’re filled with ideas and now we just have to go build better instruments

GREENE: Very good. How about you two guys, young gentlemen? What are you thinking about for the?

BARISH: For me. I think we’ve opened a whole new area of observation. And I can’t imagine that we know all the things we’re looking for. And so we’re going to be surprised and I’m looking for the big surprises. I’ve lived through that in particle physics when in 1975, we found a totally new set of particles and a new way to see things. I think the same thing will repeat itself here.

[01:35:14] GREENE: Rai? You want to take us out on a final thought?

WEISS: Well I’ll give you the same thought that Kip had. I’ll just amplify what Kip-Kip said something very important. He said that this field is fundamentally different in astronomy than what we now know, except for neutrinos astronomy, that’s closer to us. But an electromagnetic astronomy, you’re dealing with things that scatter. The thing that you see has come out of a source. It isn’t what actually went on deep inside the source. You’re seeing something that has been bounced around by all the processes that are going on in the source. For example, when you look at the sun, you can’t see the inside of the sun. Why? Because the light has been scattered so often.

GREENE: And the thing is, it takes a heck of a long time for it to come-

WEISS: Diffuse out through all of that stuff. So it turns out that the idea that you are looking now at the dynamics of things deep inside of the astrophysical systems we see only the outside of, says to me that we’re going to learn an incredible amount of things. It’s going to make a- I think it’s going to make a change, a vast change just in the way people look at astronomy. I’m hoping also that this is not a faint. Because I’ve worked in other areas that I- eventually I’d like to see this really have an impact on cosmology. And I think if we build the bigger the bigger systems that we’re talking about, we will do some kinds of cosmology. We might be able to do the cosmic metric better, for example, to measure the same things that see with light you see now you might have a different might have a different set of systematic problems that might be able to help out where we are now and not knowing things well enough. I wish we could see the Big Bang. I don’t think this instrument I know this instrument won’t do it. There are very fancy ideas on paper that allow you maybe to do this. I hope the people the CMB, the people who did the BICEP, will get there first because then we’ll have really known something in my lifetime.

GREENE: Well when you guys get there, we will invite you all back for another conversation. So this has been a fascinating discussion. I- actually everybody join me in congratulating the team with fantastic achievements. Thank you.