Modern cosmology rests on a simple idea: look large enough, and matter should smooth out, with no preferred direction across the cosmos. This assumption, known as the cosmological principle, underlies the standard Lambda Cold Dark Matter (ΛCDM) model that assigns roughly 5% ordinary matter, 25% dark matter and 70% dark energy to the universe.
Thanks to a new generation of telescopes on Earth and in space — including the Dark Energy Spectroscopic Instrument (DESI) and the Euclid mission — we can finally test that assumption on unprecedented scales. In our recent paper, we report evidence that the distribution of galaxies does not become directionless on the largest scales we can currently probe. Instead, DESI data reveal coherent directional patterns in galaxy positions stretching across billions of light years.
Why this matters
The ΛCDM model has been remarkably successful at explaining many cosmic observations: the expansion history of the universe, the abundances of light elements from the Big Bang, and the detailed pattern of the cosmic microwave background. Yet several tensions have emerged. Different methods of measuring today’s expansion rate — the Hubble constant — disagree, producing the so-called Hubble tension. Observations from the James Webb Space Telescope have raised questions about how rapidly early galaxies formed. And analyses of very distant quasars and radio galaxies have suggested an unexpectedly large dipole — a pronounced asymmetry between opposite directions in the sky — that conflicts with ΛCDM.
DESI’s three-dimensional galaxy map
DESI is assembling one of the most detailed 3D maps of the galaxy distribution by measuring positions on the sky and redshifts, which are used to infer distance. We asked whether, on the scales accessible to DESI, matter really becomes uniform and directionless — in short, whether the cosmological principle holds.
To test this, we used a statistical approach that examines the directions of galaxy pairs: for a given galaxy, what is the probability of finding another galaxy a certain distance away in a particular direction? If the universe is isotropic and homogeneous on large scales, those pair directions should be uniformly distributed. If matter is arranged into long filaments or walls, pair directions will align along preferred axes.
A persistent cosmic web
Applied to DESI data, the test returned a clear directional signal. Galaxy pairs were not randomly oriented but aligned into coherent filaments and sheets. Crucially, these alignments did not fade away at larger separations. The patterns persisted out to several billion light years in the deepest samples we analyzed. Rather than smoothing into an isotropic fog, the cosmic web appears tangled and organized even on the largest scales we can currently measure.
We compared these observations to suites of simulated universes generated under the standard ΛCDM assumptions. The difference was striking: simulations showed weaker, smaller directional patterns than those present in the DESI data. The strength and spatial extent of the observed alignments are larger than expected if structure grew only under the processes encoded in the simplest ΛCDM models.
Implications and possibilities
If galaxies trace the full matter distribution, including dark matter, then these results indicate that either structures grew more rapidly or coherently than predicted, or that our assumptions about the ingredients and laws governing the universe need revision. Several possibilities include:
– Dark matter with nonstandard properties or interactions that promote larger, longer-range structures than in simple cold, collisionless models.
– A need for a more general cosmological description that allows large-scale inhomogeneities to persist, altering the large-scale behavior assumed by the cosmological principle.
– Other, currently unknown physics affecting structure formation or the interpretation of large-scale surveys.
These findings would not only challenge the cosmological principle but could also bear on other tensions — such as dark energy’s nature and the Hubble tension — by changing how we interpret cosmic observations.
Next steps
This is not a moment for speculation alone; it is a call for more and better measurements. Upcoming and ongoing surveys — more data from DESI, Euclid, and other wide-area mapping projects — will be essential to confirm whether these directional patterns persist and to rule out observational or analysis systematics.
If the signal remains robust, cosmologists will need to develop new models of structure formation and perhaps rethink the large-scale description of the universe itself. Either way, these results show that the largest cosmic scales still hold surprises and that precision mapping of the universe can reveal fundamental gaps in our understanding.
Marco Galoppo is a PhD candidate at the University of Canterbury and Francesco Sylos Labini is research director at the Enrico Fermi Research Institute.
This article is republished from The Conversation under a Creative Commons license.
