24 | Kip Thorne on Gravitational Waves, Time Travel, and Interstellar
What this covers
Kip Thorne and Sean Carroll discuss how the detection of gravitational waves by LIGO culminated a half-century effort that required both experimental ingenuity and theoretical breakthrough. Thorne traces the project's roots to the early 1970s, when the feasibility of detecting such waves remained deeply skeptical among physicists. The conversation centers on why the measurement proved so extraordinarily difficult—achieving sensitivity to displacements roughly one-hundred-trillionths of a meter—and how the collaboration persisted through initial blind tests that made the real 2016 detection nearly unbelievable. The discussion also covers Thorne's dual role bridging theory and experiment, including his pivot toward numerical relativity to generate predictions essential for interpreting observations.
Beyond the detector's achievement, Thorne elaborates on the physics revealed: storms in spacetime created when black holes collide, and the exotic twisting vortices of space that form and shake down through mergers. The conversation then shifts to more speculative territory—wormholes, closed timelike curves, and whether they can actually function as time machines. Thorne presents the mathematical case that quantum physics likely forbids such machines by causing wormholes to collapse at the moment they would become traversable, a form of cosmic protection against paradox. He defends thought experiments as legitimate tools for probing deep questions about physical law. The episode concludes with Thorne's work as executive producer on Interstellar, describing how Christopher Nolan's film balanced scientific accuracy in visualizing black holes and wormholes against narrative necessity.
Thorne argues that LIGO's detection of gravitational waves was fundamentally a half-century team achievement that opened a new observational window onto 'storms' in warped spacetime, and that the same physics-driven curiosity that built LIGO also makes thought experiments about wormholes and time machines genuinely informative about the laws of nature.
- Detecting gravitational waves required measuring mirror displacements ~100x smaller than a proton across 4 km, a feat many doubted was possible.
- The marriage of supercomputer simulations of colliding black holes with observations was essential to interpreting the signals.
- Wormhole/time-machine thought experiments, though speculative, dig into deep questions about quantum physics, information loss, and the consistency of physical law.
LIGO's gravitational wave detection required solving extraordinarily hard measurement and computational problems over decades.
- LIGO's complexity is captured by the fact that it carries roughly 100,000 data channels monitoring environmental and instrumental conditions that might indicate something going wrong, versus just one data channel carrying the actual gravitational wave signal about the pattern of stretch and squeeze.
“the number of data channels carrying information out of the instrument and out of the environment... is not a hundred, not a thousand, not 10,000 but 100,000.”
- LIGO's key measurement trick is that it does not locate any single proton or atom but measures the average position of all the atoms in the mirror, using light that bounces off the thermally jiggling mirror faces while being designed so the signal is only influenced by the motion of the mirror's center of mass and not by the jiggles of individual faces.
“you're measuring the changes in location of huge numbers of them. You're measuring actually the average position of all the atoms in the mirror.”
- Developing the technology and techniques to solve Einstein's equations on computers (numerical relativity) took as long as the experimental effort itself—half a century—and progress was painfully slow despite being pursued by the best people in the world, because the problem was genuinely very hard; Thorne left day-to-day LIGO work in the early 2000s to help drive these simulations because predictions were essential to interpret observations.
“the effort to create the technology and the techniques for these simulations for solving Einstein's equations on computers lasted just as long as our experimental effort, lasted half a century.”
- The motion LIGO detected was about 1/100th the diameter of a proton, roughly a trillion times smaller than the wavelength of the laser light used to measure it and about 10 million times smaller than the individual atoms in the mirror face that are constantly jiggling thermally, which is why the experiment was extraordinarily difficult and cost over a billion dollars.
“for what we finally did detect, it was 1/100th the diameter of a proton or the nucleus of an hydrogen atom.”
- The research that led directly to LIGO began around 1972, when Thorne wrote his first paper on a vision for what gravitational wave detection could achieve and Rainer Weiss independently wrote a technical paper inventing the kind of interferometric detector ultimately built, identifying all major noise sources and ways to deal with each.
“it was 1972 that I wrote with Bill Press, a student of mine at the time, my first paper about a vision for what you can do with gravitational waves if you could detect them.”
- The LIGO collaboration regularly tests its system with blind injections—where a few members secretly wiggle the mirrors electronically to mimic a gravitational wave and send a fake signal through the entire detection pipeline up to writing a paper—so when the real first signal arrived, Thorne and the team were sure for several months it was a blind injection rather than a genuine detection.
“That's the best way to test the system, is an end-to-end test where a group of about something like three or four members of the collaboration are assigned the test, go in electronically wiggle in mirrors in this instrument”
- LIGO consists of interferometric detectors with four 40-kilogram mirrors hung from overhead supports by quartz fibers, arranged two per arm in an L shape with arms 4 kilometers long; a passing gravitational wave pushes the mirrors on one arm together while pushing the other arm's mirrors apart, and laser interferometry monitors this pattern of stretch and squeeze.
“we have four mirrors. Each of them weighs about 40 kilograms, 100 pounds... they hang from overhead supports by Quartz fibers.”
- LIGO cannot detect mergers of supermassive black holes (millions to hundreds of millions of solar masses) because bigger black holes produce slower oscillations and longer-wavelength waves with periods of minutes to hours, and at those time scales noise on Earth from weather and human activity is horrendous; detecting them requires space-based detectors like LISA with arms roughly 10 light-minutes apart, far larger than the Earth's ~0.1 light-second size.
“the bigger the black holes when they merge, the slower the oscillations of the waves as they pass, which means the longer the wavelength of the waves... Noise on the Earth is horrendous.”
The Nobel Prize should recognize large collaborations, not just individuals, because major scientific breakthroughs require teamwork.
- The Nobel Prize for gravitational waves should have gone to the entire LIGO team rather than just three individuals, because some major scientific breakthroughs can only be achieved through large collaborations, and the Nobel committee has an obligation to educate the public about the importance and process of collaboration.
“This prize should have gone to the LIGO team who pulled this off, and not to just three people. In this case, this could never have been done by the three of us. It really was a team effort.”
- Thorne believes the Nobel Prize is, on balance, good for science because it is more effective than anything else at reaching broadly throughout the world to non-scientists and giving them a sense of the importance, power, and beauty of science.
“It is more effective at reaching broadly throughout the world to non-scientists and giving them some sense of the importance, the power, and the beauty of science. More effective than anything else that we have.”
- The Nobel committee maintains that its principal goal is to educate the public about the value of science, and that three individuals are better icons for science than a big team, which is why they refuse to award prizes to collaborations despite having the legal power to change.
“our principal goal with the Nobel prize is to educate the public about the importance of science, the value of science, what has been done, and three individuals are better icons for science than a big team”
- The Breakthrough Prize handled the gravitational wave recognition better than the Nobel by giving two-thirds of the prize to the LIGO collaboration and one-third to three individuals, demonstrating a workable model for recognizing both teams and key contributors.
“with LIGO, they gave two thirds of the prize to the collaboration, and one third to three individuals. And I thought that worked well.”
- Thorne jokes that Nobel Prize winners are usually 'has beens.'
“Nobel Prize winners are usually "has beens."”
Merging black holes are rare but detectable across the universe because they emit strong waves observable over vast distances.
- Thorne was not worried that there would be nothing detectable in the sky; it seemed clear to him at roughly the 95% confidence level that black holes and neutron stars orbiting and merging existed and would be detectable, and these became the primary design-driving sources by about 1980.
“it seemed clear to me, not with 100% confidence. But up in the 95% confidence level at least, that we would have black holes that orbit around each other and merge”
- When two black holes collide they merge objects made not of matter but only of warped space and warped time, creating a veritable storm in the shape of space and the rate of flow of time, analogous to the surface of a stormy ocean; this phenomenon was not understood even theoretically until supercomputer simulations of Einstein's equations a few years before detection revealed it, and observation has now verified it.
“you're colliding two objects that aren't made from matter. They have no solid surface... They're made only from warped space and warped time.”
- Mergers of two 30-solar-mass black holes are quite rare on a per-galaxy basis—roughly one per million years in our galaxy—but because there are so many galaxies in the observable universe, such events become detectable.
“One in a millions years.”
- It was the marriage of supercomputer simulations with the observations—each making huge progress on exactly the time scale required—that enabled deep understanding of what was being seen, because the simulations provided predictions that could be compared with observations to deduce the details of the black hole storms.
“it was the marriage of the simulations with the observations that really led to our being able to understand in-depth what was being seen.”
- Thorne expected black hole mergers to be the first detection because heavier objects emit stronger waves: black holes 10 times heavier than neutron stars can be seen 10 times farther, which means a thousand times greater volume of the universe (the cube of distance), so even though black holes are perhaps 100 times rarer than neutron stars, the vastly larger observable volume made them the most likely first source.
“With the black holes, 10 times heavier, you can see them 10 times farther. That means the volume of the universe you can see them through is 10 cubed... A thousand times greater volume.”
- Thorne's favored explanation for the unexpectedly heavy 30-solar-mass black holes is that smaller black holes form inside globular clusters, sink to the bottom of the cluster through gravitational interactions with less massive stars, find each other and merge, and then those merged holes merge again, building up fairly quickly to pairs of 30-solar-mass black holes.
“you begin with smaller black holes that form inside, what we call globular clusters... They sink to the bottom, they find each other, they collide and merge, and then that merged hole, merges with another merged hole, and you build up fairly quickly to two 30 solar mass black holes.”
Wormholes can theoretically become time machines but probably cannot exist naturally and would likely self-destruct.
- A wormhole can be turned into a time machine by taking one mouth and accelerating it through the universe at high speed and back: time slows for the traveling mouth as seen externally, so the two mouths' clocks differ when reunited, yet through the wormhole interior the clocks always run at the same rate, creating an inconsistency that constitutes a time machine.
“You just take one mouth and I put my wife in one mouth. She carries that mouth out through the Universe in her spaceship and comes back at high speed and time for her slows down, as seen in the external universe.”
- Even though relativity permits turning a wormhole into a time machine, an advanced engineer must also obey quantum physics, and because quantum physics guarantees at least a little fluctuating matter is always present, the analysis indicates with high probability that the wormhole will self-destruct at the very moment it is turned into a time machine—suggesting the universe is reluctant to permit time machines.
“quantum physics says, with high probability, that the wormhole is going to self-destruct the moment you turn it into this time machine.”
- Wormholes made purely from warped spacetime without matter will self-destruct (implode), so to hold one open you must put some kind of matter inside it, and that is where the problem becomes very difficult because it involves the messy, poorly understood matter side of Einstein's equation rather than the clean curvature side.
“if you have them made from warp space time without matter, then they self-destruct. So you have to put some kind of matter in them to hold them open, and that's where it becomes tough.”
- Wormholes almost certainly cannot exist naturally; the chance an advanced civilization could make one is larger but still small, and even if made they probably implode before anyone could travel through, and the chance a civilization could stabilize one for travel is small but not zero—the research points strongly toward the conclusion that they probably cannot exist, and if they can, very probably not naturally.
“I think the chances that they exist naturally are exceedingly small. The chances that they can be made by a very advanced civilization, are bigger, but still small.”
- You cannot move faster than light locally, but the speed limit only forbids two objects close enough that there is no significant warping of spacetime between them from exceeding light speed relative to each other; since the universe expands and distant regions recede faster than light, a wormhole with large spacetime warping between its mouths lets you effectively travel across the galaxy faster than light without violating relativity.
“Sure you can go faster than the speed of light, we all know, but not locally.”
Quantum mechanics in the presence of time machines produces information loss and multiple classical solutions, not parallel timelines.
- Once quantum mechanics is properly formulated in the presence of a time machine (closed timelike curves), one discovers that information gets lost—a conclusion most physicists dislike but which Thorne finds wonderful, leading into the information loss paradox where he holds an iconoclastic minority view.
“once you've done that, you discover that information gets lost.”
- Quantum mechanics can be formulated in the presence of a time machine using the Feynman sum-over-histories approach generalized by Gell-Mann and Hartle, and doing so probably yields a single unique consistent solution from the multiple classical possibilities; this is not about branching into parallel timelines but about finding the one most probable trajectory that actually happens.
“the quantum physics probably has a solution to get around this, and there's probably one unique solution.”
- In Polchinski's Paradox—a billiard ball sent into a wormhole time machine emerges in time to knock itself off course and prevent its own entry—nature avoids the contradiction not by sacrificing logical consistency but by sacrificing predictability: the ball receives a gentle self-collision that slightly alters its trajectory, and there turn out to be multiple classical solutions where previously it seemed there were none.
“The billiard ball goes in, it gets hit a very gentle blow by itself on its way in. So it goes in, it comes out on a slightly different trajectory than it was supposed to”
Gravitational waves are real physical phenomena that carry energy and are produced by colliding black holes stretching spacetime.
- Feynman's sticky-bead thought experiment demonstrated that gravitational waves carry real physical energy: beads on a stick with slight friction get moved back and forth by a passing wave while the stiff stick resists stretching, so the beads rub against the stick, heat it up, and could even start a fire with strong enough waves.
“let's take a stick and we'll put some beads on the stick, and there's a little bit of friction between the beads and the stick. And when the gravitational waves go by, they move the beads back and forth because they can slide.”
- A gravitational wave is a ripple in the shape of space and time produced by sources like two colliding black holes; unlike a water ripple that moves up and down, it stretches space in one direction perpendicular to its propagation while squeezing it in the other perpendicular direction, so freely floating particles get pushed apart and together as the wave passes.
“A gravitational wave can be described heuristically as a ripple in the fabric or the shape of space or of space and time, that is produced, in our case, by two colliding black holes”
- The reality of gravitational waves remained controversial among some relativity theorists all the way into the 1980s because the difficulty lay in truly understanding the physical consequences of the mathematics, and older-generation physicists who struggled with the math were not convinced until detector planning forced clarity.
“the issue of really understanding what the mathematics were saying was even harder in the case of black holes than gravitational waves. So that skepticism also... lasted I think up until the early 1970s when it was finally laid to rest.”
Thought experiments about colliding objects through wormholes can reveal deep truths about the laws of nature.
- In areas of physics beyond where experiments can currently be done, simple thought experiments—such as what happens to billiard balls colliding through a wormhole-based time machine—can sometimes dig pretty deeply into the laws of nature, which is why Thorne values such speculative reasoning.
“thought experiments like these, simple thought experiments like what happens to billiard balls if they collide through a worm hole that has a time machine built into it. Simple thought experiments can sometimes dig pretty deeply into the laws of nature.”
- Although the term 'wormhole' was coined by John Wheeler and the concept is often credited to Einstein and Rosen in 1936, the underlying idea actually goes back to Hermann Weyl around 1922; Wheeler pushed hardest to understand them, intuiting that at the Planck length scale—where space and time become probabilistic and fluctuate due to quantum physics—one would find a froth of fluctuating wormholes.
“the phrase was coined by John Wheeler. The concept actually goes back to Herman Vile, about 1922.”
- Thorne holds that anything which shows the laws of nature are different from what you expect is wonderful because you can learn wonderful things from it, so he would welcome rather than fear being able to build a time machine despite the logical paradoxes involved.
“Anything that says that the laws of nature are different than you expect is wonderful. You can learn wonderful things.”
Interstellar used scientifically accurate visualizations of black holes and wormholes based on solving Einstein's equations.
- Thorne obtained his executive producer credit on Interstellar off the bat through negotiation by a well-connected attorney (Ken Safran), based on the brainstorming he and Lynda Obst had done and the expectation he would remain lead science advisor throughout, illustrating that Hollywood credits depend heavily on legal representation rather than scientific contribution alone.
“Oh, that happens by having a very good attorney.”
- For Interstellar, Christopher Nolan insisted that black holes, wormholes, and astrophysical objects be visualized as accurately as possible by solving the equations for light propagation from source through the environment and the IMAX camera optics to film; Thorne provided the equations to Double Negative's chief scientist Oliver James who programmed them, producing scientifically grounded imagery.
“he wanted everything that involve visualizations of black holes and wormholes and astrophysical objects to be done as accurately as possible through computer simulations and solving the equational propagation of light”
- Almost everything in Interstellar is compatible with the known laws of physics under the agreement that science should not get in the way of making a great movie; the one place of substantial departure was the visualization of the trip through the wormhole, which used creative license because the scientifically accurate versions were all dull, with only minor glitches elsewhere documented in Thorne's book.
“the agreement between Christopher Nolan and me, from the outset, was I told him I wanted everything to be compatible with the laws of nature and he said, 'I'm enthusiastic to do that. As long as it doesn't get in the way of making a great movie.'”
- Thorne, Stephen Hawking, and Lynda Obst wrote the treatment for a follow-on science fiction movie several years ago before Hawking passed away, and the project has an option sold to a studio with a hired screenwriter, following closely the original treatment story.
“Stephen Hawking and Lynda Obst and I wrote the treatment for a follow-on movie several years ago before Stephen passed away.”
- Christopher Nolan works in the opposite manner to Steven Spielberg, who carries more than a dozen movies in development simultaneously; Nolan does one thing at a time and will not even decide on his next project until his previous film has been out for several months and he has gone through all its publicity.
“He does one thing at a time. He doesn't even make a decision about the next thing he's gonna do until the previous movie has been out for a few months.”
Feynman's curiosity-driven private calculations and bold skeptics like Braginsky drove progress in understanding relativity and detection.
- Feynman's startling conference insights came not from preparing for publication but from years of private curiosity-driven calculations done simply to understand how things work, after which he felt satisfied without needing to write papers, building an enormous storehouse of understanding he could draw on.
“I did some long calculations back several years ago just trying to understand how things work. And once I understood, I was satisfied. I didn't need to write a paper describing this, I just understood and I was satisfied.”
- There was even more skepticism about black holes than about gravitational waves because understanding what the mathematics was saying was harder in the black hole case, and that skepticism was not finally laid to rest until the early 1970s.
“the issue of really understanding what the mathematics were saying was even harder in the case of black holes than gravitational waves. So that skepticism also... lasted I think up until the early 1970s when it was finally laid to rest.”
- Because LIGO's $1.1 billion funding came from the National Science Foundation, NSF program director Richard Isaacson advised Thorne to find funding elsewhere for his time-machine research so that a congressman attacking him for working on time machines could not jeopardize LIGO, leading Thorne to acknowledge that work as funded by a Caltech endowment rather than the NSF.
“You don't want to screw things up for LIGO if some congressman comes in and starts hacking at you for working on time machines. And maybe you can find funding elsewhere for that work, just to keep yourself safe”
- Vladimir Braginsky, one of the best experimenters Thorne worked with and the second person after Weber to make a mark in the field, was so skeptical for about 15 years of the interferometric approach because of the many things that could go wrong that he pursued a simpler method—until he visited Caltech and MIT in the late 1980s, saw the progress and plans, went home, shut down his whole operation, and joined LIGO.
“he was just so very skeptical for about 15 years because of the complexity that these instruments would have to have.”
- Most physicists are either theorists, who build mathematical models and deduce consequences using physical intuition, or experimenters, who design experiments and build apparatus to test the laws of nature; Thorne deliberately positioned himself at the interface, even spending a year doing nuclear physics experiments at Princeton to prepare to interact with experimenters.
“most physicists work essentially entirely on theory, developing models for the universe... An experimenter designs experiments, builds apparatus to perform them in order to investigate the laws of nature”
Merging black holes create vortices of warped space that shake down to fewer vortices as the black holes merge.
- Sticking out of each spinning black hole are twisting vortices of space, like tornados—counterclockwise at the north pole and clockwise at the south pole; when two black holes merge you can get up to six vortices (four from the holes plus two from orbital angular momentum), but black holes cannot have more than two, so the vortices fight and shake down to two, a behavior of empty space discovered only recently through simulations.
“sticking out of each spinning black hole is a twisting vortex of space. It's very much like the twist of a tornado.”
The United States is an unreliable long-term partner for big science missions due to congressional politics and budget swings.
- While LISA is expected to be on a fast track with NASA likely rejoining as a junior partner, the Chinese are pushing very hard to build a simpler analogue to LISA faster, potentially making them the first to detect gravitational waves from space.
“the Chinese are pushing very hard to do a simpler analogue to LISA faster.”
- The United States is an unreliable partner for big long-term science missions because NASA pulled out of a signed agreement with the European Space Agency on LISA after a cost overrun on the James Webb Space Telescope; this is not something NASA can control but results from the congressional election system and policy swings with new administrations, forcing Europe to scale the mission down dangerously.
“NASA had a cost, a big cost overrun on James Webb Space Telescope, and NASA pulled out of a signed agreement with the European Space Agency to do this mission.”