Cosmology is in big trouble, and I believe the primary cause of this trouble is metaphysical, not scientific. It is a third manifestation of the same underlying cause of the Hard Problem of Consciousness and the Measurement Problem in QM: physicalism’s lack of conceptual space for an observer.
The Hubble Tension
In recent years, a large and persistent discrepancy has emerged between independent measurements of the Hubble constant (H0) – the parameter that describes the rate of cosmic expansion. Resolving this conflict, known as the Hubble tension, is one of the most pressing challenges in contemporary cosmology. It has prompted serious reflection on the assumptions underpinning ΛCDM.
There are two primary and independent methods used to determine the value of the constant, and they yield results that differ well beyond the range of mutual error bar. The first method infers H0by analysing temperature fluctuations in the CMB. When interpreted within the ΛCDM model, this method yields a value of 67.4±0.5 km/s/Mpc. This approach is model-dependent. It depends on assumptions made within ΛCDM (especially inflation, see below) which do not necessarily apply to other cosmological models.
The second method derives the constant from observations of astronomical objects in the local universe, using the so-called cosmic distance ladder. This process involves calibrating the intrinsic brightness of Cepheid variables. A Cepheid variable is a type of massive star that pulsates in a regular cycle, changing in brightness with a well-defined period. The crucial characteristic of Cepheids is the direct relationship between their pulsation period and their intrinsic brightness (luminosity), a relationship known as the period-luminosity law, discovered by Henrietta Swan Leavitt. This law makes them powerful “standard candles” for measuring vast cosmic distances: by observing a Cepheid’s pulsation period, astronomers can determine its true luminosity and then calculate its distance by comparing it to its observed apparent brightness] and Type Ia supernovae1. The SH0ES (Supernovae, H0for the Equation of State) collaboration, among others, has consistently obtained higher values of 73.0±1.0 km/s/Mpc. This method is relatively model-independent.
The discrepancy between these two values now exceeds5 standard deviations, which makes it highly unlikely to be attributable to statistical error. While it has been suggested that unrecognised systematic errors may be responsible, extensive reanalyses and cross-checks using different methods and observatories have failed to eliminate the discrepancy. Very recently, the James Webb Space Telescopehas essentially eliminated the possibility that the Hubble Tension is just a measurement error in the distance ladder. JWST’s high-resolution infrared data has confirmed the Cepheid distances to an unprecedented degree. The tension is now a “Crisis of Physics,” not a “Crisis of Data.”
The Hubble Tension suggests there is a deep flaw in our understanding of the universe’s early conditions, the nature of Dark Energy, or the validity of the ΛCDM model itself. Possibilities under investigation include modifications to the physics of the early universe (such as early dark energy or extra relativistic species), revised models of Dark Matter, and even exotic proposals involving varying fundamental constants or departures from GR.
Dark Matter
Dark Matter has never been directly detected, but regardless of that it is now thought to comprise approximately 85% of the matter content of the universe and about 27% of its total energy density. The hypothesis of Dark Matter was not introduced for a single reason, but rather emerged as a unifying explanation for multiple independent observational anomalies across different astrophysical and cosmological scales. In each case, visible (baryonic) matter alone proved insufficient to account for the observed gravitational effects.
1. Galaxy Rotation Curves
The original and most famous evidence for Dark Matter came from the study of spiral galaxy rotation curves. According to Newtonian dynamics, the rotational velocity v(r) of stars orbiting at a distance r from the galactic centre should decrease with distance once outside the bulk of the visible mass, roughly following: v(r) ∝ 1/sqrt(r). However, beginning with the work of Vera Rubin and others in the 1970s, it was found that rotation curves tend to flatten at large radii: stars and gas far from the galactic centre orbit at roughly constant velocities, rather than slowing down. This observation suggests the presence of an extended, invisible halo of mass surrounding each galaxy, whose gravitational influence maintains the high orbital speeds. The discrepancy between the mass inferred from starlight and the mass required to explain the rotation curves is substantial – typically an order of magnitude or more.
2. Galaxy Cluster Dynamics
Earlier still, in the 1930s, Fritz Zwicky observed that galaxies in the Coma Cluster were moving too rapidly to be gravitationally bound if the cluster contained only the mass visible in stars. Applying the virial theorem to estimate the total mass required to keep the cluster from dispersing, he found that the luminous matter fell short by a factor of up to 100. This mass discrepancy in galaxy clusters was later confirmed through X-ray observations of hot intracluster gas (which itself requires deep gravitational wells to remain bound) and gravitational lensing studies showing that much more mass is present than can be accounted for by visible matter.
3. Gravitational Lensing
GR predicts that massive objects curve spacetime and thus bend the paths of light – a phenomenon known as gravitational lensing. When distant galaxies or quasars are viewed through massive intervening structures like galaxy clusters, the degree of lensing observed allows cosmologists to infer the total mass along the line of sight. In many such cases, especially with strong and weak lensing maps, the lensing mass significantly exceeds the luminous mass, reinforcing the existence of large quantities of invisible mass. Importantly, gravitational lensing provides a direct measure of total mass, independent of dynamical assumptions.
4. The Bullet Cluster and Analogous Collisions
One of the most striking pieces of evidence comes from observations of colliding galaxy clusters, such as the Bullet Cluster (1E 0657-56). In these systems, the visible baryonic matter slows and interacts during the collision, while the gravitational mass, inferred from lensing, appears to pass through relatively undisturbed. The spatial offset between the baryonic mass and the total gravitational mass strongly suggests the presence of non-collisional mass, consistent with Dark Matter that interacts gravitationally but not electromagnetically. Similar signatures have been found in other merging clusters. This is the strongest evidence against Modified Newtonian Dynamics (MoND). Dark Matter is a necessary placeholder for a real gravitational effect that MoND cannot explain.
5. Large-Scale Structure Formation
Another key motivation for Dark Matter arises from the need to explain the formation of cosmic structure: the growth of density fluctuations into galaxies, clusters, and filaments in the early universe. The standard model of cosmology assumes that the tiny fluctuations observed in the CMB grew over billions of years into the structures we observe today. However, calculations show that baryonic matter alone, coupled to radiation before recombination, cannot grow fast enough to account for the observed structure, especially on small scales. Dark Matter (being non-baryonic and non-interacting with radiation) can begin clumping earlier, seeding gravitational wells into which baryons later fall. Simulations of structure formation match observations only when Dark Matter is included.
6. Cosmic Microwave Background Anisotropies
Precision measurements of the CMB have revealed tiny fluctuations in temperature across the sky, corresponding to density variations in the early universe. The detailed angular power spectrum of these anisotropies depends sensitively on the composition of the universe. The best-fit models to CMB data require a significant component of cold, non-baryonic Dark Matter to reproduce the relative heights and positions of the acoustic peaks. This result is independent of galaxy dynamics and provides a cosmological-scale confirmation of Dark Matter.
In summary
Despite its success in explaining these phenomena within the ΛCDM framework, the true nature of Dark Matter remains unknown. Candidates range from weakly interacting massive particles (WIMPs) to axions, sterile neutrinos, and more exotic possibilities. Decades of experiments have yet to yield definitive evidence for its identity.
What this has got to do with philosophy
The connection with philosophy comes via another cosmological problem known as the Fine Tuning Problem. Not just the physical constants but all sorts of other features of the cosmos, especially the early cosmos, are ridiculously fine tuned for life (even though we can’t locate any life beyond Earth). Scientists typically treat this as a problem to be solved – they look for dynamic, law-governed mechanism, operating forwards in time, to explain anything that looks like fine tuning. However, this cannot possibly work in all cases, so fine tuning is left as a brute fact in need of explanation. Various explanations have been proposed, especially variations of multiverse theories (all possible universes exist, we just happen to be in one that supports life) and theological explanations (God did it). But regardless of which is the correct explanation, the brute fact remains: we live in a cosmos/timeline which appears to be fine tuned for us to be here. NOW…if we are forced to accept fine tuning then it makes no difference how many different instances there are. 20 examples of fine-tuning are no more difficult to explain than 2, since all of them can have the same explanation. This changes everything. Why? Because of inflation.
Cosmic Inflation
Inflation has long been regarded as one of the most successful theoretical advances in modern cosmology. Introduced in the early 1980s, it purports to explain why the observable universe appears so flat, homogeneous, and isotropic, despite the apparent lack of causal connection between distant regions in the early universe.
Inflation was introduced to address several deep puzzles that arise when the universe is assumed to have evolved according to classical relativistic physics from the very beginning: the Horizon Problem, the Flatness Problem and the Monopole Problem. To solve these problems, inflation posits that the universe underwent a brief period of exponential expansion immediately after the Big Bang. This expansion would stretch a tiny, causally connected region to encompass the entire observable universe (solving the horizon problem), drive the geometry of the universe toward flatness (solving the flatness problem) and dilute any relic particles with empty space (avoiding the monopole catastrophe). However, inflation itself requires finely tuned initial conditions. It demands the existence of a hypothetical inflationary field (the “inflaton”) with a specific potential, appropriate dynamics, and a graceful exit mechanism to end inflation without reheating the universe too violently. Inflation trades one set of mysteries for another, and does so on the assumption that the early universe actually existed as a classical, physical state, evolving forwards in time in a manner determined entirely by the laws of physics.
Inflation Fine-tuning Problems
Inflation was brought into ΛCDM to solve the fine-tuning problems mentioned above, but it does so at the expense of introducing the fine-tuning problems described below.
The Reheating Precision Problem
Inflation ends when the potential energy driving exponential expansion decays into ordinary matter and radiation – a process known as reheating. For the universe to resemble what we observe today, this reheating must occur with extraordinary precision in both timing and efficiency. If reheating happens too early, the universe may not inflate long enough to solve the horizon and flatness problems. If it happens too late or too inefficiently, the universe could be left too cold, too empty, or dominated by relics incompatible with structure formation. The temperature of the universe after reheating must fall within a narrow window to allow nucleosynthesis, matter-radiation equality, and galaxy formation to proceed correctly. This the Reheating Precision Problem, and it reveals that solving fine-tuning problems via inflation creates as many problems as it solves.
The Reheating Mechanism Problem
In addition to the need for precision, there is also a fundamental lack of clarity about the microphysical mechanism of reheating. In most inflationary models, the process by which the inflaton field decays into the standard model particles is only sketched in, relying on speculative couplings, parametric resonance, or perturbative decay schemes. No experimentally verified mechanism or standard field-theoretic interaction has been confirmed to realise this transition. The detailed dynamics of how the vacuum-like energy of inflation converts into a hot, thermalised plasma (the birth of the observable universe as we know it) remain deeply uncertain. This is the Reheating Mechanism Problem: the mechanism must not only exist but execute precisely under extreme conditions without observational guidance, further compounding the implausibility of accidental success.
The Inflaton Field Problem and the Origin of Cosmic Inflation
Inflation requires the existence of a scalar field with a very specific potential energy landscape – flat enough to drive rapid expansion, then steep enough to decay into standard particles. Yet no known field in the Standard Model of particle physics behaves this way. The inflaton could never be observed, and its origin, nature, and physical justification remain completely unknown. It is a hypothetical entity postulated purely to make the inflationary model work. Moreover (surprise, surprise!) the inflaton field must possess extremely finely tuned properties:
The shape of its potential must produce the right amount of inflation.
Its quantum fluctuations must generate the correct amplitude and spectrum of primordial density perturbations.
Its decay (reheating) must convert its energy into matter and radiation without destroying structure or producing unwanted relics.
These requirements amount to an elaborate layer of theoretical scaffolding with no empirical foundation. Despite decades of searching, we have found no B-mode polarisation in the CMB that would definitively prove the “simplest” inflation models. In most models, the inflaton is simply inserted by hand, without derivation from deeper theory. Furthermore, even if we accept inflation as a real event, the questions keep on coming. Why did inflation start at all? What determined the inflaton field’s initial conditions, or when and how it ends? Why did the universe begin in a state conducive to inflation in the first place? Inflation is the epitome of ΛCDM epicycles: it’s fine-tuning all the way down.
If we accept fine tuning, we do not need inflation
If we accept fine tuning as a brute fact, we do not need inflation as a mechanism to try to explain it away. In a fine tuned cosmos, the Horizon Problem and the Flatness Problem cease to be problems at all, because they are just even more examples of extreme fine tuning! Does that mean we can just get rid of inflation? Not quite, because we still haven’t accounted for the missing monopoles. If inflation didn’t dilute them away, why didn’t they collapse the early cosmos with their gravity, and why can’t we find them now? The answer is obvious. If the cosmos is fine tuned then so can the monopoles. Fine-tuned monopoles then become a feature rather than a bug – they can be exactly the correct sort of monopoles which bind together to form inert “monopolium” (+ve/-ve pairs), which can then become the dark matter which is needed as gravitational scaffolding so large scale structures can form, which is necessary for life to evolve. And the reason why haven’t found them is that cosmologists aren’t even looking for the right kind of monopoles. A paper last year describes the right ones: The physics of monopolium | Two-Phase Cosmology
Why this matters for the Hubble Tension
If inflation didn’t happen then our models of the early cosmos are completely and utterly wrong. That means the early universe figure for H0 (67km/mps) is nonsense – it is a model-dependent figure extrapolated from the CMB, but the model is broken. The late universe figure is a local measurement, which is presumably correct. A completely new model of the early cosmos is required, but the Hubble Tension is no more. And exactly the same reasoning applies to the S8 tension.
