Cosmology model that ditches dark matter clears a Big Bang nucleosynthesis test

A rival to the standard model of cosmology, one that drops dark matter and dark energy entirely, has cleared a test against the universe's oldest chemistry, according to research publicized this week by the University of Ottawa. The result does not overturn the mainstream picture of the cosmos, and the model remains a minority view, but its authors say it has cleared a hurdle that has sunk other alternatives before it.

The model, known as CCC+TL, has been built over several years by Ottawa physicist Rajendra Gupta. It combines two ideas: that fundamental constants of nature, including the strength of gravity, have slowly weakened since the Big Bang, and that light loses energy as it crosses vast distances, an old hypothesis called tired light. Together, Gupta argues, the two effects can reproduce galaxy rotation and cosmic acceleration without invoking dark matter or dark energy, which the standard model holds make up roughly 95 percent of the universe. Earlier CCC+TL papers put the universe's age at about 26.7 billion years, nearly double the standard estimate of 13.8 billion.

A test the model could have failed

The new paper, by Gupta and physicist Nikolaos Samaras, was posted ahead of print and later published in the journal Universe. It checks CCC+TL against Big Bang nucleosynthesis, the production of hydrogen, helium and trace lithium in the first few minutes of cosmic history, and one of cosmology's most tightly constrained benchmarks. Models that get it wrong are usually abandoned quickly. Running a standard nucleosynthesis code with CCC+TL's physics substituted in, the authors found predicted light-element abundances matched the conventional model's own predictions to within roughly one part in a thousand to ten thousand, once a key scaling factor was set to three rather than one.

The fit was not uniformly clean. Using a lower late-time baryon density, drawn from a supernova-brightness survey, eased a long-standing mismatch between predicted and observed lithium-7 abundance, a known problem for the standard model too, but worsened the match for deuterium. The authors concluded that nucleosynthesis data alone cannot determine which baryon-density estimate is correct, for either model.

Still an outlier

Gupta's group points to other work it says strengthens its case: an analysis of rotation curves for 175 galaxies that it says CCC+TL reproduces without dark matter, and a claim that the model explains why the James Webb Space Telescope keeps finding galaxies that look too massive and too mature for their age under standard cosmology. According to Phys.org, the model predicts galaxies will turn up at redshifts beyond where Webb's planned GLIMPSE survey has yet looked. A separate look at the Bullet Cluster, usually cited as strong evidence for dark matter, is described by the researchers themselves as only an early proof of concept.

Significant pieces are still missing. A fit to the cosmic microwave background, cosmology's most demanding test, has not been published, and a companion paper on supernova time dilation remains under peer review. No independent cosmologist is quoted backing the new nucleosynthesis result, and so far the paper's two authors are its only validators.

That caution has trailed Gupta's work before. When an earlier version of the model made headlines in 2024 for eliminating dark matter, University of Queensland cosmologist Tamara Davis told Cosmos magazine:

Extraordinary claims require extraordinary evidence. You have to go and test it, and test it in different ways.

Passing a nucleosynthesis check is one such test, and CCC+TL cleared it. Whether it can survive a full comparison with the cosmic microwave background, the test that has retired most dark-matter-free models before they got this far, remains to be seen.