Three supermassive black holes have been found packed into a galaxy that existed just 1.2 billion years after the Big Bang — an ancient cosmic pile-up that may explain how the early universe produced such enormous monsters before it had much time to do so

Most galaxies get by with a single supermassive black hole. A galaxy called J0148-4214 appears to have collected three.

An international team led by Hannah Übler at the Max Planck Institute for Extraterrestrial Physics has found three actively feeding black holes inside it, the first such trio identified in the distant universe. Their results, published in Astronomy & Astrophysics, place the galaxy at a redshift of 5.0167, or about 1.2 billion years after the Big Bang. Its light spent 12.5 billion years in transit before landing on a telescope switched on in 2022.

Two of the black holes sit near the galactic centre, roughly 620 light-years apart. The third loiters about 5,500 light-years out in the suburbs.

How you spot something that emits no light

Nobody photographed these objects. Black holes are found by what they do to the gas around them, and the evidence here is a broad smear in the hydrogen emission line astronomers call H-alpha. Gas orbiting a heavy compact object moves fast enough that its light Doppler-shifts blue on one side and red on the other, blurring a sharp spectral line into a wide one. Speeds ran from 430 to 2,920 kilometres per second.

The giveaway is that the smear shows up in hydrogen but not in the bright oxygen lines, which come from thinner gas much further out that never moves that fast.

Prising apart the two central objects took spectro-astrometry, which measures tiny shifts in where light appears to come from as you step across a spectral line. Webb cannot resolve them as two separate dots. It can notice that the blue-shifted gas and the red-shifted gas are arriving from slightly different places.

As the Max Planck Society notes in its announcement, without the integral field unit on Webb’s NIRSpec instrument, only one of the three would probably have been noticed.

One of them is eating far too fast

Masses come out at roughly 80 million suns for the primary, 600,000 for its neighbour at the centre, and 2 million for the outlying one. Giovanni Mazzolari, the study’s second author, puts the galaxy’s entire stellar mass at about 1.3 billion suns, which makes the black holes a startling fraction of everything in it.

The runt is the one worth watching.

Despite being the smallest of the three, it is feeding faster than the eighty-million-solar-mass giant beside it, and faster than the Eddington limit permits. That limit marks the point where radiation from infalling matter pushes outward hard enough to throttle its own food supply. Exceeding it means the tidy version of the theory has stopped applying.

Why the early universe keeps embarrassing the models

That giant black holes turn up so early has been an awkward problem for decades. Growing one by steady accretion is slow, capped by that same Eddington ceiling, and starting from a collapsed star leaves nowhere near enough time. Collisions offer a shortcut.

Übler argues the young universe was efficient at shoving massive black holes together, setting up the mergers that future gravitational wave observatories expect to detect. The paper estimates the central pair will spiral into each other within about 700 million years, dragged together by dynamical friction.

The third black hole has two possible backstories. It might be drifting inward to join the central pair, or it might be debris from an even earlier merger, flung to the outskirts by gravitational recoil. Recoil works a bit like pushing off from a canoe: the shove that sends the boat one way sends you the other.

One paper, one galaxy

Restraint is warranted. The authors worked through supernovae, shocks, winds and massive stars as rival explanations for the broad emission before settling on gas discs around black holes as the best fit. Best fit is a long way from confirmed. Masses rest on single-epoch virial relations carrying uncertainties of roughly a factor of three, and the preprint on arXiv flags a possible downward correction of a full order of magnitude.

Three black holes in one galaxy is a first, not a pattern.

What might eventually hear it

The same group has form. In 2024 they used Webb to spot a merging pair of massive black holes in a system called ZS7, which the European Space Agency described as the most distant black hole merger yet found, seen when the universe was 740 million years old.

Collisions like these ring spacetime at frequencies no ground-based detector can hear. That is the job of LISA, the gravitational wave observatory approved by ESA in January 2024 and due to launch in 2035 as three spacecraft flying in a triangle 2.5 million kilometres on a side.

Two of these three black holes would have been invisible without spatially resolved spectroscopy. Every galaxy already on record with just one black hole was counted using instruments too blunt to know whether the count was right.