Changing Our Understanding of the Universe
A Discussion with Kip Thorne and Rainer Weiss
The 2016 Kavli Prize laureates discuss the struggle to detect gravitational waves and how their discovery is changing our understanding of the universe.
By Adam Hadhazy
Albert Einstein made many bold predictions. One that even he thought would never be confirmed was the existence of gravitational waves. His general theory of relativity held that the movement of massive objects, such as black holes—another wild consequence of relativity—would create “ripples” in the fabric of space-time. Should any of these ripples, or gravitational waves, reach Earth, they would be so astonishingly tiny that directly measuring them looked hopeless.
Yet in the same spirit of curiosity that drove Einstein, scientists did not give up the hunt. In the 1970s, breakthroughs by physicists Ronald W.P. Drever, Kip S. Thorne and Rainer Weiss led to the development of an experiment that finally sensed gravitational waves. In 2015, the Laser Interferometer Gravitational-Wave Observatory, or LIGO, registered the slightest of jolts, just one-ten-thousandth the diameter of an atomic nucleus, as passing gravitational waves locally warped space-time. The distinctive signature of the waves showed they had emanated from the collision of a pair of monstrous black holes, 1.3 billion light-years away.
LIGO made this historic detection a century after Einstein unveiled general relativity. A second detection of merging black holes followed in December 2015. Never before had these sorts of events been discernible to science, demonstrating how LIGO has opened a whole new window on the exploration of the universe.
"For the direct detection of gravitational waves," Drever, Thorne and Weiss will receive the 2016 Kavli Prize in Astrophysics.
Science writer Adam Hadhazy hosted a roundtable discussion with laureates:
- KIP S. THORNE– The Feynman Professor of Theoretical Physics at the California Institute of Technology.
- RAINER WEISS– Emeritus Professor of Physics at the Massachusetts Institute of Technology (MIT) and a member of the MIT Kavli Institute for Astrophysics and Space Research.
Ronald W.P. Drever was not available to participate in the conversation, due to illness. The conversation has been amended and edited by the laureates.
Rai, let me start with you. Earlier this year, after the LIGO team announced the first direct detection of gravitational waves, I asked you what it felt like. After all, you had spent four decades and taken hundreds of scientists along with you on the project. You said you felt like a monkey had jumped off your back, but the monkey was still there, walking along on the sidewalk. Now with the second detection made by LIGO, do you feel unburdened? Is the monkey finally out of the picture?
RAI WEISS: The second detection certainly was wonderful. It shows what we reported in February was not a unique event and that maybe we can do gravitational wave astronomy. That’s really what we’re after. The monkey is still yacking in my ear, though, because we’re not at design sensitivity yet for LIGO. We’re about a factor of three off of what should be possible, and for months now, we’ve been stuck not really understanding where a particular noise in the data is coming from. It’s beginning to grate on us.
KIP THORNE: Rai, you’ve advanced through this kind of thing before many times with LIGO.
WEISS: Oh yes, many times. But this one is a little more troublesome, Kip! [Laughter]
THORNE: It’s a standard issue that great experimenters like you and your team face. You will lick it, but it’s a pain in the meantime!
WEISS: It leads to sleepless nights, trying to figure out what you haven’t thought of. But look, Kip’s right. We’ll find the source of this noise. We have a terrific group of young people who are really sharp and now understand the apparatus very well, but they are also mystified.
Is this particular noise necessarily an instrument issue, or could it be the signature of an astrophysical source of gravitational waves we don’t understand?
WEISS: I can’t completely rule out that it’s a foreground noise of gravitational waves. That would be quite surprising because there isn’t anything we know of astrophysically that could be creating this signal.
Kip, as for you, how are you feeling about LIGO now that it has made these two detections?
THORNE: I feel a profound sense of satisfaction and great enthusiasm for the future. Satisfaction because we laid our bets in the right direction on the experimental side, and also on the side of the computer simulations that needed to be done to be ready to extract information about gravitational wave sources when the waves came. I had thought since the early 1980s that the first thing we would see is what we saw, the signature of black holes merging. It all worked out the way that I had expected.
When did you first seriously discuss working together on the gravitational wave detector that would eventually become LIGO? Didn’t it involve a very late night sharing a hotel room in Washington, D.C., in 1975? What happened during that conversation that set you on the path to eventually making LIGO a reality?
WEISS: I can tell the story because it’s vivid in my mind. I had, and have, this enormous regard for Kip. He was the visionary who saw what we might detect. So I was running a committee for NASA, looking at what might be the role of the space program in cosmology and gravitation. I desperately wanted Kip for the committee, but he was busy. He very politely said “no” to me. I did have this brilliant idea, though, that maybe he could come and give testimony to the committee. He agreed he would do that!
So Kip came to Washington, D.C., but he didn’t have a hotel room, and it was in the middle of the summer. I don’t know if you have ever been in Washington in the summer, but all the high schools send their kids there. It’s just impacted. So Kip and I shared my room.
He told me that night he was thinking of building an experimental group at Caltech to look into gravitational radiation, because he already had this wonderful theoretical group. I explained the interferometric technique I’d worked on, that LIGO now uses, and after Kip understood what it was and why it could reach the necessary sensitivity to detect gravitational waves, he really got on board. That was a turning point.
THORNE: The idea for this kind of a gravitational wave detector was conceived in very bare bones by several people in the ’60s, including Rai. But the reason that Rai is regarded as the principle inventor of interferometric gravitational wave detectors is because in 1972, he wrote this paper that was basically a blueprint for the future and the initial LIGO detectors. He identified all of the major noise sources they would likely face and ways to deal with them.
Rai showed that if you made this detector big enough, kilometers in length, it should be able to get into the sensitivity range where we theorists were beginning to understand the waves should be, and where in fact the waves wound up being. This was a tour-de-force article. But Rai published it in an internal MIT report series, rather than the standard literature where everyone would read it, because Rai didn’t think you should publish something like that until you detected gravitational waves! [Laughter]
Nevertheless, that paper reached the community and had a huge impact. That was the thing that really got the field started.
WEISS: I don’t know if I believe all of that, but go ahead! [Laughter]
THORNE: Rai is a modest man and feels uncomfortable, but that is what happened. So yes, that conversation with Rai in the hotel had a huge impact on convincing me that the prospects were good enough. That in part led me to propose to Caltech that they build a group working in this field.
With the two of you united, now in the late 1970s, on pursuing a LIGO-like gravitational wave detector, how did you get the funding?
THORNE: We carried the proposal for my new experimental group to the then-president of Caltech, Marvin Goldberger, who had been a professor at Princeton. He asked a Princeton colleague, Bob Dicke, who had been a mentor to Rai and an inspiration to me when I was a grad student: Is this thing worth doing? Bob gave it a ringing endorsement. That got Caltech to buy-in. They invested roughly $2 million to initiate the construction of a prototype gravitational wave detector.
WEISS: That was a turning point. At the time, we have to admit, the field looked pretty flaky. Joseph Weber, from the University of Maryland, had recently done experiments with these aluminum cylinders and claimed he had detected gravitational waves, but that was widely discredited. Consequently, here was a field whose technology was beyond belief, and we weren’t sure what the gravitational wave sources would even be—that was too much for a lot of institutions, including the one I came from. MIT got on board of course later on, of course, and has been ever since, but it was Caltech first.
THORNE: It was a substantial buy-in by Caltech. That, plus hiring Ron Drever, who had been suggested by Rai and by some other people, got us going.
Speaking of Ron, your fellow 2016 Kavli Prize winner, when he joined up with you guys in 1979, what sort of critical insight did he bring to the table? And how did having him onboard change your creative chemistry?
WEISS: Ron came up with some very clever ideas, and he was able to convince others that they were good ideas. I’ll tell you what they were. We were having a lot of trouble with scattered light in the arms of the interferometer. Ron had an instinct that Fabry-Perot cavities, which are a part of the laser setup and consist of two mirrors facing each other, would help. And Kip did some calculations to show that indeed light scattering problems would be much less serious in a Fabry-Perot.
THORNE: Ron pushed hard that we should use those cavities in the arms of the interferometers for LIGO. That was a principal contribution.
WEISS: You have to remember Ron did not think in equations as much as some of us back then. He is a visual, brilliant guy who imagines things in his head and in pictures. That’s the best way I can describe him. That’s something that was hard for some of us because he couldn’t sometimes explain his pictures so well. And there were two other seminal ideas he had that are very important.
Ron helped come up with the idea of putting a mirror between the laser and the interferometer itself, to make sure that the interferometer got all the light that the laser put out, and that everything that was reflected by the interferometer to the laser would be sent back into the interferometer by this particular mirror. It was effectively the same thing as getting a more powerful laser.
The second idea Ron had, along with other people, was a thing we’re only beginning to use now on LIGO. It’s putting another mirror in the interferometer between the detector and the beam splitter, which helps us get better data. It was added to the second, advanced version of LIGO, which we just started using last year and that made the first gravitational wave detection.
"We really are opening up a whole new way of observing the universe, a way that is going to be central to the human race’s exploration of the universe around us, not just for years or decades, but for centuries into the future."—Kip Thorne
How hard was it to keep the LIGO project on track? Clearly there was creative tension in formulating an unprecedented, multi-million dollar scientific experiment to snag gravitational waves.
THORNE: There were serious conflicts, and I was in the middle trying to mediate them over a period of three years, between ’84 and ’87. We tried to run LIGO with a steering committee at the top. Then we were told in no uncertain terms that we had to get a single director. Robbie Vogt, of Caltech, then took over. Once that happened, it was a new ballgame.
WEISS:I think part of this struggle was the transition from table-top physics, like in a university lab, to big-time, big-budget physics. We almost killed ourselves in the process, but Robbie helped us write a first-class proposal, understandable not only by experts, that got money from the National Science Foundation in 1988. Then Robbie organized the project to improve the collaboration between the disparate groups at Caltech and at MIT.
THORNE: Then in 1994, Barry Barish became laboratory director. He transformed the entire LIGO project from a small, troubled R&D effort into what it is today. He recognized it had to be expanded from what was roughly 40 people at Caltech and MIT into a collaboration of many institutions and hundreds of people in order to pull it off successfully. It was so complex and so difficult that it required this much larger collaboration. Barry conceived how to do that.
Kip, you got out of day-to-day involvement with LIGO in the early 2000s to further develop numerical relativity, which uses supercomputers to simulate the behavior of cosmic objects governed by general relativity, such as black holes. How did this work help in understanding the black hole collisions LIGO is witnessing?
THORNE: It was very much on our minds that those computer simulations needed to be in hand by the time we might begin to see gravitational waves with LIGO. But in the 1990s, there were huge problems in the field. These great computational scientists could collide two black holes head-on, but when they tried to have the black holes go in orbit around each other, as should happen in Nature, they couldn’t even get them to go around once before the computers crashed. By 2001, I got alarmed because I was expecting that Advanced LIGO would be operational in the early 2010s, roughly a decade into the future. It was not at all clear that the simulations would be in hand by then.
So that was when I pulled out of day-to-day involvement with LIGO and initiated a numerical relativity group at Caltech, in collaboration with Cornell, to try to help move this effort along. One of our post-docs, Frans Pretorious, had a huge breakthrough in 2004 that cracked the logjam. So by last year, several research groups could simulate the black hole collisions that produced the waves LIGO saw. Fortunately! Because the waves’ signal power was almost all from moments near the holes’ collision, when numerical relativity is the only way we have to compute the predicted waveforms, for comparison with observation. That was crucial for understanding what LIGO saw.
Let me add that just as I was not involved in the experimental work developing LIGO, I also was not involved in the hands-on work on numerical relativity computer simulations. I simply provided vision as to what calculations were important to do and what information was important to extract.
WEISS: Now Kip is being too modest. [Laughter] The fact is that I would go to Kip whenever I got stuck. He has this wonderful way of explaining things, and he solved many major problems for us.
Finally, after decades of work, you got to share LIGO’s success with the world earlier this year. Why do you think the detection of gravitational waves captured the public’s imagination so strongly?
WEISS: I think the answer has two ingredients: Einstein and black holes. Black holes have some sort of enormous consequence to people who have never heard anything about general relativity. It’s amazing. I walk around and people tell me they’re scared and interested by black holes because they are so big and eat everything. It’s a visceral thing that people somehow want to know about.
THORNE: I think we succeeded in getting across the fact that we really are opening up a whole new way of observing the universe, a way that is going to be central to the human race’s exploration of the universe around us, not just for years or decades, but for centuries into the future.
Several other gravitational wave detector experiments are currently underway, such as VIRGO and GEO600, both in Europe. Another, KAGRA, is scheduled to open in Japan in 2018. A third LIGO instrument is under development for operation in India, circa 2023. How will having all these detectors boost the field?
WEISS: What we cannot do with LIGO alone is tell you where a gravitational wave source is in the sky. That is enormously important for science. We want to connect gravitational wave detections with the rest of astronomy and do observations using electromagnetic observations—in gamma-rays, x-rays, visible light, infrared and radio—to learn more about where the waves are coming from. With a network of detectors, we can determine where that source is. That is one fundamental reason for having a network.
"People tell me they’re scared and interested by black holes because they are so big and eat everything. It’s a visceral thing that people somehow want to know about."—Rai Weiss
Looking farther into the future, scientists and engineers are hoping to put a gravitational wave detector in space, sometime in the mid 2030s. Called eLISA, it will consist of three spacecraft a million kilometers apart in a triangle formation. The spacecraft will beam lasers to each other in order to detect passing gravitational waves using interferometry, just like LIGO. What excites you about taking gravitational wave science into space, and what will we detect with eLISA that we cannot detect while earthbound?
THORNE: Space-based gravitational-wave observatories will go after waves with wavelengths 10,000 times longer than LIGO detects. That’s the same factor as radio waves to light. So eLISA is like radio astronomy, if LIGO is like optical astronomy. And just as radio waves show us very different things about the universe than light, so eLISA will show us very different things than LIGO.
The set of phenomena that can be studied with something like eLISA is really exciting. You’ll be able to look at supermassive black holes, and at things falling into supermassive black holes. You can map out the full space-time geometry of a black hole with exquisite precision.
WEISS: There are also projects that want to use pulsars to detect gravitational waves. Pulsars are neutron star remnants of massive stars that send out repeating beams of radiation. The waves pulsars can detect could have wavelengths of about three light years to maybe half a light year, and would come from black holes even larger than eLISA’s.
THORNE: That would reveal the biggest black holes in the universe. And with another technique, studies of the polarization of cosmic microwaves, we will also have a shot at observing ultra-long wavelength gravitational waves from the earliest moments of the universe. So there are these four different wavelength bands, including LIGO’s, that will all be opened within 20 years and probably three of those four within the next 10 years.
A key thing I want to point out is that eLISA is a stripped-down version of the original LISA, which was a joint project of the European Space Agency and NASA. But NASA stopped funding it in 2011. I think eLISA is dangerously underpowered for the physics it is intended to do.
WEISS: It’s absolutely critical that the United States get back in, and I think that may yet happen. They should not build eLISA. They should build LISA, and I think all the people in Europe believe that, too. I think that’s all going to get straightened out in the next year or two.
LISA, by the way, has taken a big step forward. The European Space Agency launched a LISA Pathfinder mission in December and their tests show that they can really do the full LISA experiment.
Getting back to LIGO again, so far it has detected gravitational waves from the mergers of two black holes. What other gravitational waves sources are out there that LIGO can detect?
THORNE: As Advanced LIGO’s sensitivity improves by that factor of three that Rai was referring to earlier, we will see three times farther into the universe. That actually works out to being able to observe 3 cubed or about 30 times more of the universe, because there are three dimensions in space. So it’s a huge improvement. At that point, we can expect to see probably a few black hole mergers a week. We also expect to see neutron stars spiral together and merging, and black holes tear neutron stars apart.
If we’re really lucky, we’ll at some point see the gravitational waves of a supernova explosion. That’s exciting because we don’t understand what goes on in the core of a massive star during a supernova, when the core implodes and then forms a neutron star. A huge amount of energy is released and somehow that energy is used to blow off the star’s outer layers. But by seeing both the gravitational waves and particles called neutrinos from that event, this problem is very likely to be solved.
WEISS: What Kip was just talking about isn’t the end of the line. There will be further improvements to LIGO that improve its sensitivity by yet another factor of three. With that, we’ll get more sensitive tests of general relativity because we’ll have better signal-to-noise ratios. People are also thinking about systems that are even more sensitive.
THORNE: Instead of interferometers with four-kilometer-long arms, they could have 40-kilometer-long arms.
WEISS: That will actually take us into the realm of cosmology, back in time to when the first black holes formed. But for that you need to change the facilities. That’s a big expensive step and it will clearly be an international project.
LIGO’s discoveries have been roundly received as confirmation of Einstein’s general theory of relativity. Could later detections of gravitational waves actually open up cracks in Einstein’s theory?
THORNE: Gravitational waves will test general relativity to high precision in this domain that was never tested before, where space-time is highly warped and dynamical. Gravity is just very, very strong in these situations. There could be a crack that opens up there. I’m very skeptical, but there are many people who are quite hopeful that that may happen.
By the middle and second half of this century, I expect the central piece of cosmological physics that will be explored using gravitational-wave detectors is the first second of the life of the universe. There, we are likely to begin to see some cracks in general relativity.
To wrap up, it’s a poetic coincidence that LIGO made its monumental detection a century after Einstein predicted the existence of gravitational waves. If you were magically to have a coffee with Einstein now, how do you think he would react to the news?
WEISS: That’s something I’ve pondered from the very beginning of all this. I am an Einstein devotee—all of us are—but I’ve studied a lot about Uncle Albert. I think he would be absolutely delighted, but the first thing he would ask about is probably the technology. I have a piece of evidence for that. At MIT, a friend of mine, Henry Stroke, ran into Einstein in the street. Henry wanted to explain to Einstein what he was working on, which was nuclear physics. Einstein rolled his eyes and wanted to get away from him. Then my friend said, “Oh, by the way, before you walk away Dr. Albert Einstein, we also work on atomic clocks.” And that stopped Einstein dead in his tracks.
What Einstein wanted to know was how you made a tick-tock, timekeeping method out of a thing, in this case an atom, that had ten billion oscillations per second. That was the thing Einstein asked first. That’s absolutely fascinating to me. Of course Einstein would be interested in the rest of it, but mainly, “How did you do it?”
Also, as I’ve said before, I would desperately love to see Einstein’s face as we told him about the discovery.
THORNE: I also would like to watch Einstein’s reaction, to watch him think through the discovery that was made and his thought processes. I’m sure he would be enormously interested in the phenomena we’re seeing with LIGO, and how that can be used to test his general relativity theory. Einstein was an enormously curious person.