Cosmology

Summary

To be written.  The standard model of cosmology graded layer by layer, the rival cosmologies proposed in its place, and the young-earth answers to distant starlight.  25 articles in this section are written;  the rest are planned.  This page will eventually state where they came out, which arguments survived, and which were retired.

The articles

Linked titles are written.  Bold titles are planned.

• The Big Bang — The Big Bang is a stack of six separable claims, not one theory. Expansion, the light elements and the microwave background are very strongly supported. Inflation is barely testable, dark matter is undetected, the cosmological constant is unexplained, and lithium-7 and the matter-antimatter asymmetry are unsolved.
  • The Expanding Universe — Redshift rises with distance, and supernova time dilation shows the shift stretches arrival times as well as wavelengths, which kills tired light. Whether space expands or galaxies move through it is largely a choice of coordinates. The Tolman test cannot separate the two, and a hot start 13.8 billion years ago is a further claim with separate evidence.
  • The Cosmic Microwave Background — The microwave sky is a blackbody to 50 parts per million, scales as T(z) = T0(1+z), carries acoustic peaks whose positions and polarization phases were computed before they were measured, and is gravitationally lensed at 43 sigma. Its temperature was never predicted, its large-angle anomalies are real but small, and no creationist model explains its power spectrum.
  • Big Bang Nucleosynthesis and the Light Elements — Deuterium and helium-4 match what a hot early universe should have made; lithium-7 is over-predicted by a factor of 3.3 and the discrepancy is unsolved. The baryon density fixed by nuclear physics at three minutes agrees with the value fixed by the microwave background 380,000 years later to about a tenth of a percent.
  • Large-Scale Structure and the Acoustic Ruler — Galaxies lie in a web of voids, filaments and clusters. A sound wave frozen at 380,000 years left a preferred galaxy separation of 147 megaparsecs. That length was computed from the microwave background before anyone looked for it, found in 2005, and confirmed to sub-percent precision by DESI. The JWST “impossible galaxies” claim did not survive spectroscopy.
  • The Lyman-Alpha Forest — Quasar spectra carry hundreds of hydrogen absorption lines, each from a different cloud at a different distance. Every line sits at a redshift below the quasar's own, without exception, which kills intrinsic-redshift models. The forest maps the cosmic web, dates the end of reionization near z = 5.3, and weighs neutrinos.
  • Cosmic Inflation — Inflation answers a real puzzle, the horizon problem, and made one prediction that could have failed and did not: the slight tilt of the primordial spectrum, now measured at 8 sigma. But its gravitational waves have never been found, its flatness argument is disputed inside mainstream cosmology, and its own architects disagree in print over whether it can be falsified.
  • Dark Matter — Most of the matter in the universe is measured but not detected. The gravitational case is strong, and in the Bullet Cluster it needs only general relativity and a photograph of the gas. Forty years of laboratory searches have found nothing, the searches are nearly out of room, and several galaxy-scale failures that cold-dark-matter supporters report themselves are unsolved.
  • Dark Energy and the Cosmological Constant — The expansion of the universe is accelerating, and that measurement is solid. What causes it is not known. Quantum field theory overshoots the measured vacuum energy by 60 to 120 orders of magnitude depending on the cutoff, the popular 10^120 is only the Planck-scale end of that range, and DESI's 2024-2026 hint that dark energy evolves is still dataset-dependent.
  • The Hubble Tension — Two good ways of measuring the expansion rate disagree by about nine percent, at 4.5 to 7 sigma. The disagreement is real and unexplained. It is a fight over calibration, not over expansion, and both candidate values still give a universe 12.8 to 13.8 billion years old.
• Rival Cosmologies — The alternatives proposed in place of the standard model, each measured against the same evidence.
  • Tired Light — Fritz Zwicky's 1929 proposal that photons lose energy in transit explains Hubble's law with no expansion and no beginning. It is dead. Any interaction that drains a photon also deflects it, and distant galaxies are sharp. Supernova time dilation, the Tolman test at better than 10 sigma, and the microwave background's blackbody each rule it out independently.
  • MOND and Modified Gravity — Milgrom's modified dynamics predicts individual galaxy rotation curves from the visible matter alone, with one universal constant, and it did so before the data were good enough to test it. It fails on galaxy clusters, the Bullet Cluster and the third acoustic peak of the microwave background. MOND is a successful theory of galaxies and not a cosmology.
  • The Doppler Model of Galaxy Redshifts — Lisle's Doppler model uses a wrong surface-brightness equation and fails supernova and CMB tests. JWST galaxy sizes remain anomalous, but the model neither explains them nor yields a young universe.
  • Halton Arp and Discordant Redshifts — Halton Arp argued that quasars are ejected from nearby galaxies and carry non-cosmological redshifts. His alignment cases came from an eye-selected sample, so they could not measure a probability. The redshift periodicity he predicted failed direct tests in 2dF and Sloan. Lensing, resolved host galaxies and Lyman-alpha absorption each independently place quasars at their redshift distance.
  • Steady State and Plasma Cosmology — The steady state was a good theory and its own field killed it with radio source counts, quasar redshifts and the microwave background. Plasma cosmology never produced a working cosmology. Neither gives a young universe, and citing them to show the Big Bang is contested proves less than creationists think.
• Young-Universe Cosmologies — Eight young-earth answers to distant starlight, checked against their own literature. Created light implies a record of events that never happened, the decaying speed of light was refuted by a creationist's reanalysis, and Humphreys and Hartnett both walked away from their own models. Lisle's synchrony convention survives, but by his own account it changes no observation.
  • The Cosmic Distance Ladder — Astronomical distances are measured, not assumed. Parallax is pure trigonometry. Three geometric methods bypass the ladder entirely and agree with it: eclipsing binaries, water masers, and the SN 1987A light echo. Every rung carries assumptions, and this article names them. None of the errors is large enough to shrink the universe.
  • SN 1987A — One supernova, about 167,000 light years away, measures its own distance twice by geometry, times its own light echo with a stopwatch, and shows titanium-44 decaying at laboratory rates. It is the single best worked case against a 6,000-year light travel time, and it constrains every young-earth answer differently.
  • Light Created in Transit — The oldest young-earth answer to distant starlight holds that God created the light beams already on their way. It cannot be refuted, and that is the problem: SN 1987A's neutrino burst, light curve, ring echo and titanium-44 lines would then record an explosion that never happened. Most young-earth astronomers abandoned it for that reason.
  • The Decaying Speed of Light — Setterfield and Norman argued that 163 historical measurements of c since 1675 show a real decline, solving distant starlight. A weighted refit by an ICR department head found no trend, the apparent decay traced to one mishandled seventeenth-century point, and a dimensional constant cannot change on its own. Creationists retired the argument.
  • Humphreys' White Hole Cosmology — D. Russell Humphreys proposed in 1994 that gravitational time dilation near the centre of a bounded universe let billions of years pass in the distant cosmos during one earth day. Two physicists showed a boundary changes no clock rate inside it. Humphreys revised the model in 2008 and later set it aside.
  • Hartnett's Carmelian Cosmology — John Hartnett built a young-earth cosmology on Moshe Carmeli's five-dimensional relativity, claiming galaxy rotation without dark matter, expansion without dark energy, and starlight in transit during creation week. He retracted it in print five times between 2015 and 2024. His own bounded solution turned out identical to the unbounded one.
  • The Anisotropic Synchrony Convention — Lisle's synchrony convention rests on real physics: relativity fixes only the round-trip speed of light, and the conventionality debate is genuinely open. But a coordinate choice changes no observation, and it leaves proper time untouched, so ASC converts the distant starlight problem into the appearance-of-age problem rather than solving it.
  • Coordinate Cosmologies: Dennis and Tenev — Phillip Dennis and the Tenev-Baumgardner-Horstemeyer team answer distant starlight by relabelling time on standard spacetime rather than changing physics. Both are mathematically sound and observationally empty. Dennis calls his own model an open subset of FLRW, so it predicts exactly what the big bang predicts, and proper time still makes distant stars old.
  • Faulkner's Miracle — Danny Faulkner's 2013 proposal answers distant starlight with a Day Four miracle rather than a mechanism. He says so himself, and says it may have nothing testable in it. The proposal is coherent, its Hebrew argument is weaker than it looks, and it leaves the ages of nearby objects untouched.

Verdict

To be written.