A 26.7-Billion-Year-Old Universe Without Dark Matter?
Weighing the CCC+TL Cosmology
By Thomas Lee Abshier, ND ·
This autumn the University of Ottawa announced that an alternative cosmology, one in which the universe is 26.7 billion years old and needs neither dark matter nor dark energy, “has held up against several major cosmology tests.” The model is called CCC+TL, for covarying coupling constants plus tired light. Its author is Rajendra Gupta, an adjunct professor of physics at Ottawa, working with postdoctoral researcher Nikolaos Samaras. The newest paper in the series, a Big Bang nucleosynthesis consistency test, appears in the journal Universe.
I want to take the claim seriously, which means two things: crediting what the work has actually done, and being plain about where the press release says more than the published record supports.
Where I stand
I stand on the side of treating CCC+TL as a live challenger to the standard model of cosmology. Dark matter and dark energy are inferences, not observations, and a model that removes both deserves a hearing rather than a dismissal. But the headline outruns the record. Of the six results the press release cites, two are genuine passes in peer-reviewed print (the light-element test and the galaxy rotation curves), one is disputed in print by another group (the cosmic-chronometer test), one is still under review (supernova time dilation), one is labeled by the authors themselves as a preliminary proof of concept (the Bullet Cluster), and the most important of all, the cosmic microwave background, is unpublished. The model’s real strength is not its list of passes. It is that it makes a sharp prediction about the earliest galaxies that the James Webb Space Telescope can check within the next few years.
The question under the question
Why does every “universe without dark matter” story travel so far? Because people sense, rightly, that something is off. For roughly four decades physicists have searched for the particle that dark matter is supposed to be, and none has been detected. Dark energy is a name for an acceleration of cosmic expansion that nobody can explain from first principles. Together these two unseen ingredients make up about 95 percent of the universe in the standard model, called ΛCDM (Lambda Cold Dark Matter). To many readers this looks like the epicycles of Ptolemaic astronomy: invisible additions bolted on to rescue a framework.
That suspicion is the real question behind the headline, and it deserves a direct answer. The dark sector is inferred, but the inference is not flimsy. It rests on many independent lines of evidence that point the same way: the pattern of peaks in the cosmic microwave background, gravitational lensing by galaxy clusters, the motions of galaxies within clusters, the spacing of baryon acoustic oscillations, and the rotation of galaxies. Any replacement has to account for that whole web, not just one strand of it. Gupta himself states the bar fairly: a new model “must not conflict with the pillars of cosmology and astrophysics.” The right question, then, is not whether ΛCDM has unexplained ingredients (it does), but whether CCC+TL has cleared that bar.
What CCC+TL actually proposes
The model combines two old ideas in a new way.
The first is covarying coupling constants. In standard physics the speed of light, the gravitational constant, Planck’s constant, and Boltzmann’s constant are fixed for all time. Gupta proposes that they change slowly over cosmic history, together, in a way that keeps the locally measured combinations stable. The lineage goes back to Paul Dirac’s large-number hypothesis of 1937, which also proposed that the constants of nature might drift. In CCC+TL, the effects we attribute to dark matter and dark energy come from this slow covariation rather than from unseen substances.
The second is tired light, an idea Fritz Zwicky proposed in 1929: light loses a little energy as it crosses vast distances, so part of the redshift of distant galaxies comes from that loss rather than from the expansion of space. Gupta’s model is a hybrid. The universe still expands, but the observed redshift is split into an expansion part and a tired-light part, and the tired-light part accounts for a modest share of the Hubble constant (roughly 11 of the measured 70 or so kilometers per second per megaparsec).
The consequences are large. The universe comes out about 26.7 billion years old instead of 13.8 billion, and the density of ordinary matter alone is enough to close the books, with no dark matter required.
Because the model changes the Einstein and Friedmann equations themselves, the team could not borrow the standard computer codes cosmologists use. They wrote their own equations and code for each test. That deserves credit; it is a great deal of work. It also calls for caution, because so far every pass has been computed with the authors’ own machinery. Independent implementation by another group is one of the things that would strengthen the case most.
Grading the tests one by one
I grade each result by the kind of evidence behind it, not by the outlet that reported it.
Light elements from the Big Bang (the new Universe paper). The model’s predicted amounts of hydrogen, helium, and the other light elements match those of the standard model. That is a real consistency check, and a model that failed it would be finished. But matching ΛCDM means matching what ΛCDM already gets right, namely deuterium and helium. The press summary does not say whether CCC+TL does anything about the lithium-7 discrepancy, the one place where the standard model’s nucleosynthesis prediction has disagreed with observation for decades. My grade: a pass, not yet a win.
Cosmic chronometers. These measure how fast the universe has expanded at different epochs by comparing the ages of old, passively evolving galaxies. The press release says the model “agreed” with them. That is the claim I find most overstated. In February 2026, a group led by Lei Lei at the Purple Mountain Observatory published a test in Monthly Notices of the Royal Astronomical Society using 32 chronometer measurements. They found the data strongly favored ΛCDM over CCC+TL, and more seriously, they found an internal tension: the value of the speed-of-light variation index that fits the supernova data is rejected by the chronometer data at a likelihood ratio of about 1.7 × 10⁻¹⁴.
Gupta replied in June 2026. His argument is that chronometer ages are not truly model-independent. They come from stellar population models built on standard physics and local calibrations, and if the constants of nature vary, those stellar clocks have to be recomputed before they can be used to test the model. Applying a correction factor, he finds the CCC+TL and ΛCDM fits statistically indistinguishable. He also concedes, to his credit, that this correction is phenomenological rather than derived from first principles, and that a definitive test would require recalculating the stellar models themselves.
I think both sides have a legitimate point. Gupta’s physical objection is sound: if the constants drift, every clock calibrated on fixed constants is suspect. But a correction factor chosen to remove a discrepancy is not the same as a prediction that avoids it. Until the stellar models are rebuilt under CCC+TL, the honest word is “disputed,” not “agreed.”
Galaxy rotation curves. In Monthly Notices, the team applied the model to all 175 galaxies in the SPARC database and reproduced their rotation curves without dark matter. That is a substantial, peer-reviewed result. I would add one piece of context: rotation curves are the strand of dark-matter evidence that alternatives have most often matched. Modified Newtonian Dynamics (MOND) has fit them well for years. The harder evidence lies elsewhere, in cluster lensing and in the microwave background.
Supernova time dilation. This is where tired light has historically failed. If distant galaxies are truly receding, a supernova’s brightening and fading should appear stretched in time by a factor of (1 + z), where z is the redshift. Pure tired light predicts no stretch at all. Since 2001 the measured stretch has matched the expanding-universe prediction closely. CCC+TL keeps expansion, so it gets part of the stretch automatically, but the remainder has to come from its own mechanism. The team’s paper on this is still under review. I cannot grade it until it is published, and I regard it as one of the two decisive tests for the model.
The Bullet Cluster. This pair of colliding clusters is the iconic piece of evidence for dark matter, because the mass seen by gravitational lensing follows the galaxies rather than the hot gas that holds most of the ordinary matter. The team reports that in CCC+TL the extra gravitational effect also tracks the galaxies. They themselves call this an early proof of concept that needs more study, and I take them at their word. It is a lead, not a result.
The cosmic microwave background. This is the other decisive test. The pattern of acoustic peaks in the microwave background is where the standard model’s dark matter is most tightly constrained, and tired-light models have long struggled to explain why that radiation has a nearly perfect blackbody spectrum. Gupta says the model meets “the yet-to-be-published cosmic microwave background challenge.” Until that work is published and others can check it, it counts neither for the model nor against it.
The early-galaxy argument: the model’s strongest card
Since 2022 the James Webb Space Telescope has found galaxies that look surprisingly large and mature at very early times. In the standard model the universe was only about 300 million years old during “cosmic dawn.” In CCC+TL, the same observed redshift corresponds to an age of about 4.6 billion years, which gives galaxies far more time to grow. The team’s calculations, reported in the journal Galaxies, predict early galaxies about twelve times larger in size and mass than the standard model allows. As Gupta puts it, “Impossible early galaxies become predictable galaxies.”
In fairness, ΛCDM researchers have proposed ways to ease the tension (bursty star formation, different stellar mass distributions, and overestimated masses in some early objects), and refined measurements have softened some of the most extreme early claims. But the CCC+TL prediction has the virtue of being falsifiable in a clean way. The model says galaxies exist well beyond the distances JWST has reached so far. JWST’s GLIMPSE program plans to search for galaxies at redshifts that may exceed 50, where the standard model leaves little or no time for massive galaxies to form. If GLIMPSE finds mature galaxies there, ΛCDM is in serious trouble. If it finds nothing, CCC+TL loses its main motivation. Either outcome is a real result, and that is how science should work.
Why the concern remains plausible
Even if CCC+TL ultimately fails, the instinct behind it is sound, and I do not think readers who feel that cosmology has lost its footing are being unreasonable.
- The searches keep coming up empty. Decades of direct-detection experiments and collider searches have not found a dark matter particle. Each null result is reinterpreted as narrowing the parameter space rather than as evidence against the hypothesis. That may be correct, but a hypothesis that survives every null result is one we should keep examining.
- Dark energy is a label, not an explanation. The value of the cosmological constant predicted by quantum field theory misses the measured value by roughly 120 orders of magnitude, the worst mismatch in physics.
- The standard model has open wounds. The Hubble tension, the disagreement between the expansion rate measured locally and the rate inferred from the early universe, remains unresolved, and recent survey results have hinted that dark energy may not be constant after all.
- The constants have never been proven constant over cosmic time. Their stability is an assumption, tested to high precision only over limited ranges. CCC+TL’s covariation is built to evade those tests, which is both its cleverness and its vulnerability.
- Institutional momentum is real. Careers, grants, detectors, and simulation codes are built on ΛCDM. That is not a conspiracy; it is how scientific investment works. But it means an alternative must be developed by a handful of people writing their own code from scratch, as Gupta and Samaras did, while the standard model enjoys thousands of researchers smoothing its rough edges. Alternatives deserve hard scrutiny. They also deserve to be scrutinized with the same tools.
What would settle it
I would consider CCC+TL to have earned its headline when these things happen:
- The cosmic microwave background analysis is published and independently reproduced.
- The supernova time-dilation paper is published and survives scrutiny.
- Stellar population models are recomputed under the model’s varying constants, so the cosmic-chronometer test can be run fairly.
- An independent group implements the model’s modified equations in its own code and reproduces the passes.
- JWST’s deep surveys, GLIMPSE above all, report what is or is not there at extreme redshift.
Conclusion
CCC+TL is a serious, hard-working attempt to explain the universe with fewer invisible ingredients, and it has passed real tests. It has not yet passed the hardest ones, and one test the press release counts as a pass is disputed in the literature. I do not think the standard model has earned the right to treat dark matter and dark energy as settled facts, and I do not think CCC+TL has yet earned the right to replace them. The next few years of JWST data, and the publication of the microwave-background and time-dilation work, will tell us much more. I will be watching.
Sources and their limits
- University of Ottawa, “Theory predicting older universe without dark matter, dark energy passes key cosmological tests” (press release, 2026; republished on Phys.org): https://www.uottawa.ca/en/news-all/theory-predicting-older-universe-without-dark-matter-dark-energy-passes-key-cosmological
- R. P. Gupta et al., “A Big Bang Nucleosynthesis Consistency Test of the CCC+TL Cosmology,” Universe (2026). DOI: https://doi.org/10.3390/universe12070193
- R. P. Gupta, “On Dark Matter and Dark Energy in CCC+TL Cosmology,” Universe 10, 266 (2024): https://doi.org/10.3390/universe10060266
- L. Lei et al., “Stringent constraint on the CCC+TL cosmology with H(z) measurements,” Monthly Notices of the Royal Astronomical Society (2026): https://arxiv.org/abs/2508.04277
- R. P. Gupta, “Testing the CCC+TL cosmology with cosmic-chronometer measurements of the Hubble parameter,” arXiv:2606.27391 (2026): https://arxiv.org/abs/2606.27391
- E. L. Wright, “Errors in Tired Light Cosmology” (UCLA), on supernova time dilation and the microwave background spectrum: https://www.astro.ucla.edu/~wright/tiredlit.htm
This essay leans on the published record: the press release, the abstracts and summaries of the papers, and the published critique and reply. I have not seen the team’s unpublished microwave-background and time-dilation work, so I treat those claims as leads, not results. The published record also has a known tilt. Critiques of alternative cosmologies are written by a large community with mature tools, while the alternatives are defended by a few people with code they wrote themselves, so the mainstream side of any dispute will usually look more polished than its relative merits alone would justify. The press release, for its part, is the interested party’s own account, and it left out the published challenge to its chronometer claim. Readers should weigh both accordingly.
Recent Comments