You've probably heard that the black hole in Interstellar was a simulation of unprecedented scientific accuracy. You may also have heard that its creation led to an "amazing scientific discovery" having to do with the shape of its accretion disk, which loops over and under its dark, central shadow. Neither of these claims is true.
Above: What Gargantua's accretion disk "would truly look like to an observer near the black hole." | All images via James et al., licensed under CC BY-NC-ND 3.0
We spoke with Caltech physicist Kip Thorne, who served as both science advisor and executive producer on the film, to learn the real story behind Interstellar's black hole, Gargantua. We also spoke with astrophysicist Jean-Pierre Luminet, whose groundbreaking work on black holes in the 1970s was essential to the depiction of Gargantua in the film.
A lot was made of the scientific accuracy of Interstellar in the lead-up to the film's release. Director Christopher Nolan had worked closely with Thorne to see that as many plot points as possible were grounded in Real Science (or, more commonly, Speculative-Albeit-Imaginable Science).
Thorne even wrote a companion book to the film called The Science of Interstellar. The book elaborates on the scientific bases for the film's various geological, astrophysical, and cosmological scenarios, and does an admirable job distinguishing scientific fact from conceivability and guesswork. Anyone who took issue with the film's handling of science would do well to check it out; the film may play fast and loose with science here and there, but, as Thorne explains in his book, the decision to do so was almost always arrived at after much deliberation. Nolan, Thorne insists, did his homework.
Thorne did his fair share of intellectual legwork, as well. Through his collaboration with Nolan and Double Negative, the film's visual effects studio, the renowned theoretical physicist used Albert Einstein's equations on general relativity to help build what's been hailed as the most accurate depiction of a black hole ever committed to film. And as if that weren't sexy a story enough, several outlets reported that creating the film's black hole had led to "an amazing scientific discovery" (a discovery about what, exactly, we'll get to shortly).
But Gargantua, the massive black hole that appears in the film, was not actually as scientifically accurate as it could have been. And that "amazing" scientific discovery? Thorne tells io9 the discovery wasn't as amazing as most outlets made it out to be. In fact, he says, many people actually missed what the discovery was, in the first place.
In a new paper, published Friday in the journal Classical and Quantum Gravity, Thorne, along with co-authors Oliver James, Eugénie von Tunzelmann, and Paul Franklin (all three of Double Negative), describes the methods used to create the black hole featured in Interstellar, and why a more scientifically accurate simulation was excluded in favor of a flashier, "less-confusing" one.
A series of images from the paper shows, in order of increasing visual accuracy, what Gargantua would have looked like had science won out over story and spectacle.
The first image shows what Thorne and his co-authors describe as a "moderately realistic" accretion disk, the gyre of matter that orbits some black holes, and appears here to wrap over and under a spherical hole in spacetime.
The second and third images show what the accretion disk would look like, given the increasingly intense color-changing effects of Doppler and gravitational frequency shifts. (I'm simplifying, but these shifts characterize how light moving quickly toward and away from an observer affect the perceived color and intensity of that light.) The third and final image (which I've enlarged, below), write Thorne and his colleagues "is what the disk would truly look like to an observer near the black hole."
Notice how the right side of the accretion disk, from our vantage point, appears to change color and waste away? While it is arguably the most realistic-looking black hole of the lot, Nolan feared its appearance would be too confusing for mass audience. In fact, the black hole could have looked even stranger, still. The simulation above shows what the black hole looked like after reducing its spin from 0.999-times its maximal value (a plausible but improbably fast spin, but one necessary to produce the huge time dilations experienced by those characters in the film who visit Miller's planet) to 0.6-times maximal value. Were the disk spinning at full-speed, the left side of the black-hole's shadow would appear to collapse into a flat, vertically-oriented boundary, and multiple images of the accretion disk would appear to emanate from this edge.
"It would have looked a lot more puzzling" Thorne tells io9. "The black hole plays a big role in the movie," he says, and it does so without a detailed explanation of what it is you're seeing (the fact, for example, that the gas appears to wrap up and around the top and bottom of the black hole due to an effect known as gravitational lensing).
"It's quite spectacular," Thorne says, "but if you add the Doppler shift and the gravitational frequency shift, the right side of the disk becomes so dark you can hardly see it, and the left side becomes so bright that it dominates in a really puzzling way." Bring the black hole's rotation up to .999-times its maximal value, and any remaining symmetry basically vanishes. The result is a very accurate model of a very specific genus of spinning black hole, says Thorne, but at the potential expense of clear, compelling storytelling.
Gargantua plus lens flare minus frequency shifts. It's also spinning slower than expected for a black hole with its time-dilating properties.
The cinematic version of Gargantua benefited from some additional movie magic.
For instance, the visualization team modeled the black hole using bundles of light rays instead of individual ones, which evidently smoothed the appearance of the accretion disk. Lens flare was also added, to simulate the scattering and diffraction of light expected to occur in the lens of an IMAX camera trained on an accretion disk like the one in the film. Thorne tells io9 that "Chris [Nolan] loves lens flare," and that, as a physicist, he actually found it "rather annoying" (the artifactual glare actually hides some scientifically interesting images of the gravitational star field behind the black hole, says Thorne – more on that later); but he assures us that the decision was not made gratuitously.
"The addition of lens flare is absolutely true to the real world characteristics of IMAX lenses," he says. The simulation above is a variation of the accretion disk seen in the final film, and eschews the effects of fast spin, Doppler shift, and gravitational shift in favor of lens flare.
Thorne says he was disappointed to see the most accurate black-hole simulations excluded from the film, but that he understood Nolan's decision. "I completely agree with Chris [Nolan]," he says. "I would not want" a lopsided, color-shifted black hole "used in a fast paced movie using science as a venue for an exciting story." The goal, he says, was to make the film as scientifically accurate as possible while producing images that could be understood by a mass audience.
But disappointed though he may have been to see the most accurate black hole simulations go unused, Thorne says he was more upset by articles that misrepresented his scientific contributions to the film. The original headline of a feature at Wired (it has since been changed) captures the tone of a spate of articles, published around the time of the film's release, that Thorne found particularly troubling:
Thorne took issue with this article, and ones like it, for two reasons. The first has to do with the use of the "D" word, and the use of modifiers like "amazing."
"There were no profound discoveries made in the simulation of these black holes," he tells io9. Thorne was brought on board to provide the visual effects team at Double Negative with instructions for how to map light from an accretion disk to the "local sky" of a virtual IMAX camera near a spinning black hole. Double Negative then turned Thorne's prescription into "Double Negative Gravitational Renderer" (aka DNGR), a fast, high-resolution package of image-generating code that differs from the techniques commonly used by physicists (you'll recall that Double Negative could incorporate things like lens flare into its renderings, which can actually be counterproductive for an astrophysicist). Thorne and the Double Negative team then used their unconventional technique to observe black holes in – pardon the phrase – a new light.
"What we stumbled upon" were not so much discoveries but "cute little mysteries," Thorne tells io9. His two newly published papers, he says "are about trying to understand these mysteries, and the weirdness they give rise to in the simulated images."
One of the cute little mysteries Thorne is referring to has to do with the way light from an accretion disk bounces around the lens of a virtual IMAX camera. Another has to do with the way stars behind a spinning black hole, lensed by the gravitational deformation of spacetime, appear from a vantage point close to said black hole. The lensing, it turns out, gives rise to a complex, fingerprint-like structure along the left side of the black hole's shadow. You can see it in simulations that do not include an accretion disk. One of the central ironies of this observation, in particular, is that the star field lensed by Gargantua is effectively invisible in Interstellar, washed out by the far-brighter lens flare of the black hole's accretion disk.
Sure, Thorne admits, he and Double Negative got two publications out of their observations, but these are technical papers, he says. They're as much for visual effects artists as they are for scientists. "It's nothing profound," he reiterates, "it's just fun."
The second issue Thorne had with articles trumpeting scientific discovery is that they tended to confuse what new insights had been gained in the first place.
From the Wired piece and others, Thorne said, "it sounded like what we found dealt with the apparent shape of the accretion disk, when it actually has to do with gravitational lensing patterns of stars behind the black hole."
"I was not happy with that," says Thorne.
In truth, astrophysicists have been simulating accretion disks in this way for decades, going back all the way to the groundbreaking work of French astrophysicist Jean-Pierre Luminet, in the late 1970s. What Thorne told me over the phone echoes what appears in his recent publication:
There are no new astrophysical insights in this accretion-disk section of the paper, but disk novices may find it pedagogically interesting, and movie buffs may find its discussions of Interstellar interesting.
I ran all of this by Luminet himself, who confirmed Thorne's account of things and shared in his disappointment:
"I agree with his assessment," Luminet responded via e-mail, "and I deeply respect Kip Thorne's honesty... he was rather embarrassed by the use of the phrase 'This particular black hole is a simulation of unprecedented accuracy' used in the Wired article... [in such a way] that people might assume it refers to the disk." In fact, Luminet confirms, the unprecedented accuracy refers neither to the accretion disk, nor the black hole as it appears in the film.
I asked Luminet whether he agreed with Thorne that the simplified appearance of Gargantua was a necessary compromise, made for the sake of clarity.
"I don't necessarily share this point of view," Luminet replied.
No matter how you slice it, he says, your typical moviegoer knows practically nothing about the properties of black holes. Why, then, would an audience necessarily be more confused by a more complex, realistic image?
"I think that the censorship came from the Hollywood producers," says Luminet, "who arbitrarily decide what is good or not for the public."
Read the paper published by Thorne and his co-authors at Double Negative in the latest issue of Classical and Quantum Gravity, or arXiv.org. Double Negative has also created a website for all film clips discussed in the paper, which you can visit here.