Eldritch Horror or Cosmic Gardener? The Case of RBH-1

The universe is simply filled with black holes. They come in all sizes, from a few solar masses to upwards of billions of solar masses. Our galaxy, like most large galaxies, has one of these large ones, what we call a "supermassive black hole" (SMBH), at its center. However, in 2023 astronomers reported something very interesting in Hubble Space Telescope observations (van Dokkum et al. 2023): a 62-kiloparsec (roughly 200,000-light-year, about twice as long as our Milky Way's stellar disk is wide) linear feature extending from a galaxy called GX.

Researchers have been testing different possibilities, including an edge-on galaxy. One promising interpretation is that we are seeing evidence of a "runaway" supermassive black hole, designated RBH-1. That is, a truly enormous object, perhaps tens of millions of solar masses, that has been shot out of its host galaxy. In the proposed model (van Dokkum et al. 2026), it is moving at nearly 1,000 km/s (2.2 million mph), producing a spectacular wake containing shocked gas and regions of star formation.

If this interpretation is confirmed in coming years, we are witnessing something profoundly hellish and divine at the same time: a black hole from which nothing can escape is capable of being shot out of a galactic nucleus with such violence that it leaves behind the very conditions for new stars to form.

A galaxy loses its black hole

How does something with the mass of millions of Suns get ejected from its host galaxy? A black hole packs so much mass into such a small space that, once something crosses its event horizon, not even light can escape. And the more massive the black hole, the larger that horizon becomes. But the black hole itself isn't anchored in place. Like any other object, it can move through space, and if it gets enough speed, it can escape the gravitational pull of its galaxy (galaxies have escape velocities just like stars, planets, moons, etc.). The profound question is what could give something so massive that kind of kick.

When two galaxies merge, their central black holes often eventually form an orbiting pair. Interactions with surrounding stars and gas can change their orbits and bring them closer together. When they get close enough, gravitational waves--the literal waving of spacetime--carry away enough orbital energy to cause them to spiral inward and merge.

Gravitational waves carry not just energy, but momentum, much as light does (recall solar sails). When two black holes spiral together and merge, they radiate enormous amounts of energy as gravitational waves. Depending on their relative masses and how they're spinning, those waves can carry more momentum in one direction than another. A merger between black holes of very different masses generally produces a relatively small kick. But two black holes of comparable mass, with spins tilted relative to their orbit, can produce a much larger one. To conserve momentum, the newly merged black hole recoils in the opposite direction. And this is no small kick: numerical simulations (Campanelli, Lousto, Zlochower, and Merritt 2007) show that some mergers can send the resulting black hole hurtling away at several thousand kilometers per second, fast enough to escape even a large galaxy.

There is another possibility: a third black hole encounters an existing binary, and their gravitational interaction transfers enough orbital energy to send one flying out. This three-body slingshot doesn't require the escaping black hole to have just merged. The RBH-1 follow-up paper (van Dokkum et al. 2026) mentioned in the intro above, favors gravitational-wave recoil but retains this alternative.

Tousif Islam, Tejaswi Venumadhav, and Digvijay Wadekar (2026) take the recoil scenario further. If a black hole merger produced RBH-1's enormous velocity, what can we infer about the two black holes that originally collided? Their models favor a mass ratio around 3:1, although with considerable uncertainty, and a rapidly spinning larger black hole. Really rapidly spinning: its inferred spin is around 0.8 on a scale where 1 represents the maximum angular momentum allowed for a black hole of that mass. The models also favor a precessing binary, in which the black holes' spins are misaligned with their orbit, causing the directions of their spins and orbital plane to change as they spiral inward.

These are, of course, simulations based on models of gravitational-wave recoil, not measurements of two black holes that astronomers actually watched collide. The merger would have occurred roughly 70 million years before the stage of RBH-1 we're observing, which is itself billions of years in the past. But as we'll see when we get to the LISA gravitational-wave observatory below, directly observing mergers of supermassive black holes is something we may actually be able to do in the future.

A 200,000-light-year trail of new stars

Once the black hole is shot out of its host galaxy, it doesn't travel through empty space. Galaxies are surrounded by diffuse gas extending far beyond their visible stars, called the circumgalactic medium. In the runaway black hole interpretation, RBH-1 plows through this gas at supersonic speed, producing an enormous bow shock, much like the shock wave ahead of a supersonic aircraft. The shock compresses and heats the gas, which then flows around the shock and into a turbulent wake behind the moving black hole.

And this is where the star formation comes in. As the shocked gas flows into the wake, some of it can radiate away its heat and cool. If enough gas collects in sufficiently dense regions, gravity can cause it to collapse and eventually form new stars. As RBH-1 continues moving, it leaves behind a long, narrow trail containing both glowing gas and newly formed stars. That's the proposed explanation for the extraordinary 200,000-light-year feature extending from GX.

We can distinguish these processes in the light coming from the wake. Young, massive stars contribute to its blue colors and ultraviolet emission, while excited gas produces distinctive emission lines, particularly from hydrogen and oxygen. The observations indicate both star formation and shocked gas along the structure, although not every bright knot is necessarily a cluster of young stars.

But there's another interesting problem here. The surrounding circumgalactic gas is hot and diffuse, so why doesn't it simply disperse the narrow, cooler wake? And why does the gas in the wake move progressively more slowly, relative to GX, the farther it is behind the proposed black hole?

In a study published in July 2026, Ish Kaul and S. Peng Oh explore this with three-dimensional fluid simulations. As the wake travels through its surroundings, turbulence mixes hot ambient gas with the cooler material already in the tail. That mixture can radiate away energy and cool, adding more relatively cool gas to the wake rather than dispersing it. Because the newly incorporated gas initially moves more slowly relative to GX, it also slows the flow in the wake. This process, called cooling-induced entrainment, helps explain both how the narrow tail persists and why its gas decelerates downstream.

There is an important limit to what these simulations demonstrate. Their "cold" gas is still roughly 10,000–30,000 kelvin, far hotter than the molecular clouds where stars typically form. The simulations also start with cooler gas already present in the tail. They help explain how the wake survives and evolves, but don't establish where its initial cool gas came from or follow that gas all the way into new stars.

How could astronomers possibly know?

The original discovery by Pieter van Dokkum and colleagues combined Hubble images with spectra from the Keck Observatory. The narrow streak pointed toward GX, contained evidence of young stars, and ended in a bright knot of emission from ionized oxygen. A runaway black hole offered an explanation, but a galaxy viewed edge-on can also appear long and thin.

The follow-up, revised in January 2026, added deeper Hubble imaging and spatially resolved spectroscopy from the James Webb Space Telescope. Webb's instrument measures spectra at many positions across the source, rather than combining everything into one spectrum. Shifts in the wavelengths of emission lines then reveal how different regions of gas move toward or away from us.

Near the tip, the gas's line-of-sight velocity changes by approximately 600 km/s across about one kiloparsec. That is a sharp change over a small part of this enormous structure. Together with the shape of the emitting gas, it fits the expected pattern of a tilted bow shock. Modeling the geometry and velocities gives the inferred black-hole speed of about 954 km/s. (That is, astronomers are measuring gas motion and working backward to the proposed object driving it, not directly tracking a black hole across successive photographs.)

What makes the gas actually glow remains a point of contention. In a February 2026 critique (what a title, "JWST spectra are consistent with the edge-on star-forming galaxy scenario for the "runaway supermassive black hole""), Jorge Sánchez Almeida and colleagues argue that the relative strengths of the emission lines match ordinary regions ionized by young stars, consistent with an edge-on star-forming galaxy. That challenges the claim that the line ratios establish evidence of shocked gas. The critique does not, however, provide a competing model of the sharp velocity structure.

The 2026 follow-up by van Dokkum and colleagues also corrected an earlier interpretation. The original 2023 study had tentatively identified a second wake extending from the opposite side of GX, potentially suggesting that more than one black hole had escaped. Deeper Hubble observations found no evidence of this "counter-wake," and the authors withdrew that interpretation.

Van Dokkum and colleagues argue that they've confirmed the first runaway supermassive black hole, and the JWST measurements, especially the gas velocities around the proposed bow shock, make a compelling case. Not everyone is convinced, however, and there's still no secure detection of the black hole itself. Even if RBH-1 really is on the run, we don't yet know whether gravitational-wave recoil or a gravitational slingshot sent it flying. And as we'll see, astronomers have a few other candidates for runaway black holes that complicate the claim of being first.

The other fugitives: 3C 186 and CID-42

RBH-1 is not the only system where astronomers suspect a black hole has moved away from a galactic nucleus. The quasar 3C 186 is about 11 kiloparsecs, or 36,000 light-years, from its host galaxy's center (Morishita et al. 2022). A quasar is an exceptionally luminous "active nucleus", powered by gas accreting onto a supermassive black hole. Unlike RBH-1, it's not hard at all to locate the bright accreting source on the sky here.

In their June 2026 revision (Chiaberge, Morishita et al. 2026), Marco Chiaberge and colleagues found something particularly interesting: both the broad emission lines and spectral features associated with the black hole's accretion disk show nearly identical blueshifts. Both indicate motion toward us at about 1,310 km/s relative to the host galaxy. That agreement matters because these signals come from different regions of gas surrounding the black hole, making it much harder to explain the observations as gas moving around an otherwise stationary object. The researchers favor gravitational-wave recoil as the explanation, although they cannot conclusively rule out an offset black-hole binary or ejection through a three-body gravitational slingshot.

CID-42 went the other way. Its two bright optical cores and unusual spectra had made it a promising recoil candidate. But JWST imaging revealed substantial stellar bulges around both components, each containing roughly ten billion solar masses. The proposed runaway instead sits at the center of its own bulge, supporting two merging galaxies rather than an ejected black hole. Only one component is clearly active; it is not a confirmed pair of active nuclei.

Finding the next runaway black holes

RBH-1 was discovered because of its extraordinary wake, but other runaway black holes might not leave such obvious traces. One way to find them is to look for unusual activity far from the centers of galaxies. The NSF–DOE Vera C. Rubin Observatory, with its enormous camera and repeated observations of the southern sky, will be particularly well suited to finding rare objects that suddenly change in brightness.

One possibility is a tidal disruption event, in which a star passes close enough to a black hole to be torn apart by its gravity. Some of the stellar debris can fall toward the black hole, producing a brilliant flare. Finding such a flare well away from the center of a galaxy could be evidence of a displaced black hole.

And here's an interesting wrinkle. In an August 2026 study, Hochart and Portegies Zwart simulate what happens to stars orbiting black holes that receive powerful recoil kicks. Some stars remain gravitationally bound to the escaping black hole, but the kick changes their orbits, potentially sending more of them close enough to be torn apart. The simulations predict a temporary burst of tidal disruption events after ejection. Their models involve smaller black holes than RBH-1 and don't predict what RBH-1 itself is doing today, but they suggest another way to look for these elusive objects.

Of course, a flare far from a galactic center wouldn't by itself prove that a runaway black hole produced it. Astronomers would need follow-up observations and spectroscopy to establish what happened and whether the source is actually associated with the galaxy.

While Rubin could help us find black holes after they've escaped, LISA may help us understand how they get kicked out in the first place. The Laser Interferometer Space Antenna will consist of three spacecraft separated by approximately 2.5 million kilometers, using lasers to measure extraordinarily tiny changes in the distances between freely falling test masses as gravitational waves pass through.

Unlike ground-based gravitational-wave observatories such as LIGO, LISA will be sensitive to the much lower-frequency waves produced when massive black holes spiral together and merge. By measuring those signals, astronomers will be able to constrain the black holes' masses and spins, the very properties that determine how powerful a gravitational-wave recoil kick can be.

LISA won't be able to detect the gravitational waves from RBH-1's proposed merger, which happened billions of years ago and whose signal has already passed us. But it could observe other massive black-hole mergers and help test the physics used to reconstruct what might have happened to RBH-1.

The black hole we still cannot see

In the illustration at the beginning of this post, the black hole is easy to spot: it's the dark circle just behind the enormous bow shock. In the actual observations, things aren't nearly so convenient. Astronomers have yet to securely identify the black hole itself, even as an unresolved point of light.

In the JWST study, van Dokkum and colleagues report something intriguing in the Hubble ultraviolet images: a faint source located roughly where their model predicts the black hole should be, just behind the tip of the shock. But its position alone isn't enough to establish that we're seeing light from the black hole or gas surrounding it. The JWST spectra also failed to reveal the compact source with broad emission lines that the researchers had hoped to find.

Further ultraviolet observations could help establish what that faint source actually is, while better measurements near the bow shock might explain where the cooler gas in the wake comes from. For now, astronomers can trace an extraordinary 200,000-light-year structure, measure the motion of gas around its leading edge, and make a compelling case for a runaway supermassive black hole. They just haven't conclusively found the black hole itself.

A note on the illustrations: I created the illustrations for this post with substantial help from OpenAI's ChatGPT (GPT-6). We started by generating detailed, photorealistic artistic interpretations of RBH-1, 3C 186, the Vera C. Rubin Observatory, and LISA, then transformed them into the blue-line drawing style I use on this site. These are imaginative visualizations, not actual astronomical observations or photographs.