Core Theory · Article 2 of 60 · The Problems Across Every Field

Cosmology's Cracks

By Vijay Shankar Sharma · 7 min read · Core Theory series

The Singularity and the Hubble Constant That Won't Sit Still

Two specific contradictions in the standard cosmological model motivate everything else examined in this framework: one conceptual, one observational. Neither is a fringe complaint. Both sit in plain view in the peer-reviewed literature, acknowledged by researchers working closest to the problem.

The Singularity's Impossible Location

The Lambda-CDM model proposes that the universe originated from a singularity: a state of infinite density and infinite temperature. Mathematically, a singularity represents a breakdown of the governing equations, not a physical state that actually occurred. The appearance of infinity in a physical model is conventionally understood, going back to Penrose and to Hawking and Ellis, as the model failing at that point, not as evidence that an infinite physical quantity exists in nature. Penrose and Hawking's own singularity theorems, developed in the 1960s and 70s, prove that under very general conditions, General Relativity's equations must break down at a singularity; they do not prove that infinite density is a physical state nature actually reaches. The theorems are, in the mathematical physics community's own language, statements about the limits of the theory's applicability. Treating the singularity as a literal physical starting point, instead of as the signal that a different, more complete description is required at that regime, is an interpretive choice, not a derivation forced on us by the mathematics.

Set that established concern aside and look at a separate, purely logical problem: where was the singularity located? The observable universe today is approximately 94 billion light years in diameter. If the singularity was the point of origin for all matter and space, it must have some definable location relative to the spatial manifold observed today. The standard cosmological answer is that the singularity was everywhere: space itself originated and expanded from that state, so every point in today's universe, including the point you're standing at right now, was once coincident with the singularity.

If the singularity existed at every point, including the current edge of the observable universe roughly 47 billion light years from Earth, then from that edge, space must extend a further 47 billion light years outward, immediately placing us beyond the supposed boundary.

Move to that new edge and apply the identical logic again. The boundary retreats without limit, at every step, no matter how far the distance travelled. "The singularity was everywhere" is therefore logically equivalent to "the universe is infinite," a position directly examined later in this framework, and one that directly contradicts the finite-origin premise the Big Bang model is built on. This is not a measurement problem a better telescope could resolve. It is a structural contradiction sitting inside the model's own explanation of its own starting point.

It is worth being precise about what kind of problem this is, because it is easy to mistake it for a mere quirk of language. A model that cannot state, even in principle, where its own founding event took place, using its own accepted description of what that event was, has not left a minor detail unresolved. It has left the central premise of its own origin story internally unstable, in a way no amount of additional observational precision could ever fix, because the instability lives in the logic of the claim itself, not in the quality of the data supporting it.

A Constant That Has Never Been Constant

A genuine physical constant holds still under remeasurement. The speed of light has not been revised since it was first measured. The charge of the electron has not been revised. The Hubble constant, by contrast, has swung across nearly an entire order of magnitude since it was first proposed:

The implied age of the universe under any given Hubble constant value is approximately 1 divided by that constant. At Hubble's original value of 500 km/s/Mpc, the implied age is roughly 2.0 billion years, younger than the Earth itself, independently dated through geology at 4.5 billion years. At the lowest historically proposed value, roughly 50 km/s/Mpc, the implied age rises to approximately 19.6 billion years. Between the highest and lowest values ever seriously claimed for this supposed constant, the implied age of the universe varies by a factor of nearly ten.

This is now formally known in the literature as the Hubble Tension: a 4-6 sigma discrepancy between the value derived from the Planck-era Cosmic Microwave Background and the value derived from the local distance ladder, described in the peer-reviewed literature as potentially requiring new physics beyond the current standard model. Riess et al. (2019) and the Planck Collaboration (2020) represent the two sides of this tension: Riess's Cepheid-calibrated local supernova ladder consistently returns values near 73 km/s/Mpc, while Planck's CMB-based analysis consistently returns values near 67-68 km/s/Mpc. Both teams have progressively reduced their statistical and systematic uncertainties over successive publications, and the gap between them has not closed. It has, if anything, become harder to explain away as measurement error, which is precisely why the tension is treated as a serious open problem instead of a rounding issue awaiting a routine fix.

A fourth independent method, gravitational lensing time-delay cosmography, exemplified by the H0LiCOW and TDCOSMO collaborations, which measure the time delay between multiple lensed images of a distant quasar produced by an intervening galaxy's gravity, has returned values clustering around 73-74 km/s/Mpc, agreeing with the local supernova ladder and disagreeing with the CMB-derived value, adding a third broad measurement family to a disagreement that now spans time-delay cosmography, the distance ladder, and the CMB, instead of resting on any single technique's potential systematic error.

The age tension has a second, independent check worth naming: the oldest globular star clusters in the Milky Way have inferred ages, from stellar evolution models, clustering close to 12-13 billion years. Any Hubble constant value implying a universe younger than roughly 13 billion years old, which several of the higher historical H0 values do once the standard model's expansion history is folded in, sits in direct tension with these independently dated stellar populations, using a completely separate branch of astrophysics with its own separate systematic uncertainties.

The three independent methodologies referenced above now yield three distinct values, 63, 68, and 73 km/s/Mpc, with local measurements trending systematically downward as measurement precision improves. Notably, the 2026 galaxy-infall papers find that the dynamics of the groups studied are fully explained by the visible baryonic mass of the brightest member galaxies alone, with no dark matter halo required to fit the data.

Why This Matters

Every time observation has contradicted this model's prediction over the last century, the constant has been the thing revised, not the underlying premise that a single number governs a universal expansion rate. A number that has been revised by an order of magnitude, repeatedly, in response to contradicting data, is not behaving like a constant. It is behaving like a symptom of an incorrect premise: that a single universal expansion exists to be measured in the first place.

It is not for lack of trying that the tension remains open. Proposed fixes within the standard framework include early dark energy models, which add a new energy component active only in the early universe to shift the CMB-derived value upward; modified gravity theories, which alter the growth equations to close the gap without new components; and revised local calibration chains, questioning whether Cepheid variable stars are being correctly calibrated as distance markers. None of these proposals has achieved consensus acceptance, and several introduce their own new free parameters, meaning they resolve one tension only by adding another unexplained input to the model, instead of removing an assumption.

This is not a fringe complaint. The tension has been the subject of dedicated conference sessions and review articles since the mid-2010s, with some cosmologists explicitly stating in the peer-reviewed literature that the discrepancy may be signaling the limits of the standard cosmological model itself, instead of a problem solvable within its existing structure.

It is also worth being precise about what would resolve this cleanly and what would not. A single new measurement landing between 68 and 73 would not resolve the tension; it would simply add a fourth data point to a scatter that already spans that range. What would resolve it is a specific, independently motivated physical mechanism, not introduced merely to close the gap, that predicts which of the existing measurements carries the hidden systematic error and why. No such mechanism has yet been proposed and independently confirmed by a method that did not already assume the answer it was built to produce.

Both problems point in the same direction: not that the standard model is careless in its arithmetic, but that its two most central claims, a singular point of origin and a stable, universal expansion rate, do not survive contact with either the model's own internal logic or its own accumulating data.

All DOIs linked below.

← Article 1: Six Layers, One ArchitectureArticle 3: Dark Energy on Trial →
All 60 Core Theory articles → · For full detail, the research papers →