Which came first, the black hole or its galaxy's stars?
A Cambridge team writing in Nature has weighed a black hole in a little red dot seen when the universe was 700 million years old. At 50 million solar masses, it outweighs its entire host galaxy, suggesting some black holes grew before their stars did.
For a decade, the James Webb Space Telescope has been turning up objects it was not really designed to explain. Among the oddest are the little red dots: compact, red, faintly glowing smudges seen at a time when the universe was still a toddler. Astronomers have argued over what they are, with one camp holding them to be supermassive black holes wrapped in dust, and the models that back that view tending to inflate how heavy those black holes must be. A paper published on 27 May 2026 in Nature, by Ignas Juodžbalis and colleagues, reports the first direct mass measurement of one of them, and the result unsettles the usual story of how galaxies begin.
The object is Abell2744-QSO1, a little red dot first detected by Webb through a gravitational lens in the Pandora’s Cluster field. Rather than inferring the black hole’s weight from its brightness, which is where earlier estimates may have run away, the team measured the speed of gas whirling around it at different distances. Rotation speed at a radius is a direct readout of the gravity pulling on the gas, and from that the black hole mass follows without much dependence on the dust or the accretion rate.
The number is 50 million solar masses. That is large in absolute terms, but what is arresting is the comparison with the host galaxy. All the stars in that galaxy put together weigh less than half what the black hole weighs. In the nearby universe, by contrast, the central black hole sits at roughly one part in a thousand of its host’s stellar mass, a ratio that has long been read as evidence that black holes and galaxies grew up together, each feeding the other. Here the ratio is inverted. The black hole is not a small resident inside a grown-up galaxy; the galaxy is a thin sprinkling of stars around a monster that was already nearly full size.
The implication is that, at least in some early systems, the black hole assembled before the stellar population caught up. The favoured explanation for how such a heavy hole could form so early is that a large cloud of gas collapsed straight into a seed black hole of tens of thousands of solar masses, rather than starting from the death of the first stars and climbing slowly. That direct-collapse route was proposed precisely because star-sized seeds seemed too small to reach these weights within the first few hundred million years. This object, at a redshift corresponding to a universe about 700 million years old, is the cleanest evidence yet that such fast-tracked growth actually happened.
The authors are careful to note that one object is not a census, and that their model still carries assumptions about the gas dynamics. The measurement depends on tracing the rotation out to the radius where gravity is dominated by the hole, and on the lens geometry being understood. Yet the direction of the result is robust: a direct dynamical mass, at this epoch, of an object that the field had suspected was a black hole all along.
What it upends is the tidy co-growth picture. If some black holes reach 50 million solar masses while their galaxies are still gathering their first stars, then the early universe was not a place where stars lit up galaxies and later spawned holes at their centres. It was a place where dense gas could collapse, ignite and grow very quickly, and where the starlight arrived afterwards. The next step is to repeat the measurement across a population of little red dots, to find out whether this one is an outlier or the first of a kind. Either answer tells a different story about the first billion years than the one textbooks currently carry.