A Bar Born Where None Should Be
18 May 2026, Yanjiang
A stellar bar, over 7 kiloparsecs long, is revealed in the gas-rich galaxy GN20 just 1.5 billion years after the Big Bang, challenging theories of bar formation.
Imagine a dance floor flooded with several inches of water. Dancers would slip; any attempt at a rigid, straight‑line formation would dissolve into sloshing chaos. For decades, galaxy theorists imagined the gas‑rich disks of the early universe like that flooded floor—utterly hostile to the formation of stellar bars, those majestic, elongated assemblages of stars that act as cosmic thoroughfares, funneling cold gas into a galaxy’s core and feeding a central black hole. The standard script said that bars could only arise billions of years later, once a disk had exhausted most of its gas and settled into a cool, orderly stellar ballroom.
A preprint (arXiv:2605.15273) led by Leindert Boogaard of Leiden Observatory and Axel Weiss at the Max Planck Institute for Radio Astronomy has just thrown a bucket of cold water—metaphorically—onto that tidy picture. In the galaxy GN20, sitting at a redshift of 4.055, a mere 1.5 billion years after the Big Bang, the team has identified a stellar bar as clean and crisp as any seen in the local universe. And GN20 is not a gas‑starved, mature system. It is drenched in cold molecular gas—so much so that gas makes up about three‑quarters of the galaxy’s baryonic mass. A bar, it seems, can take root in exactly the kind of turbulent nursery that was supposed to preclude it.
When the Floor Was Supposed to Be Dry
For decades, the picture of bar formation was built on the principle of secular evolution: a disk must first become dynamically cold and dominated by stars, not gas. In a gas‑rich disk, the dissipative nature of the gas was thought to smear out the delicate orbital resonances that a bar requires. And at cosmic dawn, when galaxies were veritable gas‑choked furnaces, bars were considered all but impossible. The few barred galaxies glimpsed at high redshift were either lensed, low‑resolution, or already past their gas‑rich youth. Yet GN20 refuses to cooperate.
Using the JWST’s mid‑infrared MIRI camera, the researchers peered into the rest‑frame near‑infrared light at 1.1 microns—a wavelength that traces old stellar populations and cuts through the obscuring dust of vigorous star formation. There, unmistakably, sat a stellar bar approximately 7 kiloparsecs from end to end. The bar’s isophotes—curves of equal surface brightness—displayed all four morphological signatures that astronomers use to identify a genuine bar: the ellipticity of the isophotes jumps sharply above 0.3, the position angle stays locked over the bar’s entire length, a distinct drop in ellipticity marks the bar’s tip, and the position angle twists by at least 10° between bar and disk. A Fourier analysis independently confirmed a dominant m=2 mode—the mathematical fingerprint of a bar. There was no mistaking it: GN20 had a stellar backbone.
A massive stellar bar lurks within a gas-rich galaxy from the early universe. This hidden structure funnels material to the galaxy’s center, driving extreme star formation in the distant past. (Source: arXiv:2605.15273)
But the deeper shock was not the bar’s presence; it was the setting. Simultaneous observations of dust and molecular gas with the NOEMA interferometer showed that GN20 is baryon‑dominated—the combined mass of stars and gas outweighs dark matter in the inner parts. Crucially, those baryons are overwhelmingly not stars. The team estimates that gas accounts for roughly 75% of the baryonic budget, and the gas fraction—the share of baryons in the form of gas—hovers around 60%. This is a galaxy whose gravitational landscape is sculpted by a sea of cold, star‑forming fuel. And yet, a bar exists—apparently old, with a mass‑weighted age of about 300 million years, and stable.
A Disconcerting Disguise
Behind the crisp image, however, lurks a thorny question. Earlier studies of bars at cosmic noon—roughly redshift 1.5—by Costantin, Kalita, and collaborators have underscored how subtle bar identification can become when only a single filter is used. Dust, which glows fiercely in the mid‑infrared, can sculpt elongated features that masquerade as a stellar bar. The GN20 bar is detected primarily in one MIRI band (the F560W filter), though its signature also appears at longer and shorter wavelengths. Could a dusty veil be impersonating a stellar structure?
A stellar bar spanning 23,000 light-years is hidden in a distant, gas-rich galaxy seen just 1.5 billion years after the. (Source: arXiv:2605.15273)
The team’s multi‑wavelength suite offers strong—though not airtight—reassurance. NIRCam and NOEMA data show that old stars, ongoing star formation, and dust all share the same elongated orientation, a classic hallmark of a true bar rather than a chance alignment of dusty clumps. The alignment is not perfect everywhere, but it is persuasive enough to make the case strong. Still, the ghost of dust contamination cannot be entirely exorcised. The authors themselves call for follow‑up spectroscopy to seal the identification. For now, the evidence leans hard toward a real bar, but the judgment rests on a delicate stack of interlocking measurements.
From Enemy to Ally
A second, more foundational tension emerges from recent simulations of radial migration in gas‑rich disks. Zhang and colleagues have suggested that in gas‑dominated galaxies, stars can ride waves of radial mixing that may scramble the orderly orbits a bar demands. If that picture held uniformly, the bar in GN20 should have been torn apart. Yet it survives, and the team’s own comparison with new hydrodynamic simulations—developed by Bland‑Hawthorn, Tepper‑Garcia, and others—reveals a telling inversion. In baryon‑dominated disks with high gas fractions, the gas does not just disrupt; it cools the disk dynamically, lowering the velocity dispersion of the stars and actually accelerating the onset of bar instability. The simulations produce galaxies with a striking resemblance to GN20, complete with a crisp bar and asymmetric spiral arms, and they predict that a bar can emerge within a few hundred million years, not the billions required in gas‑poor scenarios.
What had been cast as a bar‑killing swamp thus becomes, on closer inspection, a bar‑nurturing floodplain. The gas that was supposed to prevent order can, under the right conditions, coax it into existence. This is not a conscious agency, but the physical result of a disk that, when dominated by baryons, can shed its thermal energy quickly and settle into a more ordered state—exactly the kind of state that is ripe for a bar to crystallize.
Rebuilding the Cosmic Nursery
GN20 therefore forces us to re‑examine a foundational chapter of galaxy assembly. If bars can spring to life in gas‑rich disks barely a billion years after the Big Bang, then they may be among the earliest coherent structures to shape a galaxy’s destiny. By channelling gas inward, early bars could feed the rapid growth of supermassive black holes and ignite starbursts that, in turn, quench further star formation—processes that define the adolescent years of the cosmos.
The prospect upends the comfortable narrative that bars are latecomers, signs of a galaxy entering its middle age. Instead, bars may have been present almost from the cradle, re‑engineering their host galaxies while the universe was still a wild, gas‑soaked frontier. The cosmic timeline of galaxy evolution may need a substantial rewrite.
The true power of GN20’s bar lies not in what it says about one object, but in the doors it creaks open. Future JWST observations—especially deep spectroscopy with NIRSpec—will need to test whether other distant, gas‑rich disks also harbour hidden bars, and whether the bar in GN20 represents an exceptional case or a common phenomenon. For the moment, a stellar bar born where none should be forces us to confront a more inventive early universe than we ever imagined, one in which the very ingredients that were supposed to inhibit structure were instead the catalyst that built it.
— Yanjiang
Yanjiang is an online editor of LoomSci.com.
References
- L. A. Boogaard et al., A stellar bar hidden in an extreme gas‑rich disk galaxy at z=4.055, arXiv:2605.15273
- Kalita et al., Galactic bars are already mature at Cosmic Noon: bar strength and flatness at z ~ 1.5, arXiv:2512.04163
- Zhang et al., Enhanced rates of stellar radial migration in gas‑rich discs at high redshift, arXiv:2512.09030