Big Bang Cosmology
The expansion, the microwave background, and light element abundances give an age of 13.8 billion years.
7 min readUpdated
The Big Bang model describes the universe as having expanded from a hot, dense early state about 13.8 billion years ago. It is not a theory of how the universe began, but of how it developed from an early condition onward.
The name was coined derisively by Fred Hoyle, who preferred a steady-state universe. The model won on evidence, and the steady-state alternative was abandoned when its predictions failed.
The main evidence
| Observation | What it shows | Source |
|---|---|---|
| Galaxy redshifts | Galaxies recede at speeds proportional to distance | Hubble, 1929 |
| Cosmic microwave background | Uniform 2.7 K radiation from all directions, the relic of the hot early universe | Penzias and Wilson, 1965 |
| CMB blackbody spectrum | Matches a perfect blackbody to within measurement error — the most precise blackbody known | Mather et al., 1994 |
| CMB anisotropies | Temperature fluctuations of about 1 part in 100,000, matching predicted density variations | Planck Collaboration, 2018 |
| Light element abundances | ~75% hydrogen, ~25% helium-4, with predicted traces of deuterium and lithium-7 | Cyburt, Fields, Olive & Yeh, 2016 |
| Galaxy evolution | Distant galaxies, seen earlier in time, are smaller and more irregular than nearby ones | Williams et al., 1996 |
| Large-scale structure | Distribution of galaxies matches simulations grown from the CMB fluctuations | Springel et al., 2005 |
The microwave background
The strongest single line of evidence, and it was predicted before it was found.
Ralph Alpher and Robert Herman calculated in 1948 that a hot early universe should leave residual radiation, now cooled by expansion to a few degrees above absolute zero.
Arno Penzias and Robert Wilson detected it accidentally in 1964, while trying to eliminate noise from a radio antenna at Bell Labs. They found a persistent signal from every direction that they could not remove, and initially suspected pigeon droppings in the horn. A companion paper by Dicke, Peebles, Roll, and Wilkinson supplied the cosmological interpretation.
The COBE satellite measured its spectrum in 1990 and found it matched a blackbody curve so precisely that the error bars were smaller than the width of the plotted line. The steady-state model has no mechanism to produce a blackbody spectrum filling the sky.
Subsequent missions — WMAP and Planck — mapped the tiny temperature fluctuations, whose angular size distribution constrains the universe's geometry, composition, and age. The 13.8-billion-year figure comes primarily from these measurements.
Nucleosynthesis
In the first few minutes, the universe was hot enough for nuclear fusion. The model predicts specific proportions of light elements, determined by the density of ordinary matter and the expansion rate.
| Element | Predicted | Observed | Source |
|---|---|---|---|
| Hydrogen | ~75% by mass | ~75% | Cyburt, Fields, Olive & Yeh, 2016 |
| Helium-4 | ~25% by mass | ~24–25% | Aver, Olive & Skillman, 2015 |
| Deuterium | ~2.5 × 10⁻⁵ relative to H | Matches | Cooke, Pettini, Jorgenson, Murphy & Steidel, 2014 |
| Helium-3 | ~10⁻⁵ | Matches | Bania, Rood & Balser, 2002 |
| Lithium-7 | ~10⁻¹⁰ | Observed ~3× lower — an unresolved discrepancy | Sbordone et al., 2010 |
Helium abundance is the key test. Stars produce helium, but not nearly enough to account for 25% of ordinary matter, and the oldest, most metal-poor stars still show roughly this proportion. The lithium-7 discrepancy is a genuine open problem and is stated as such in the literature.
What the model does not claim
Several common descriptions are inaccurate.
- It is not an explosion in space. It describes space itself expanding, with galaxies carried apart rather than flying through a pre-existing void.
- There is no centre. Every observer sees the same recession pattern.
- It does not describe the origin of the universe. The model applies from a fraction of a second onward; conditions at t=0 are outside current physics.
- It does not require the universe to have come from nothing. What preceded the hot dense state is unknown.
Common objections
"The Big Bang means the universe came from nothing"
Frequently asserted by both proponents and critics of theism, and it is not what the model says.
The equations describe expansion from a hot dense state. Extrapolating backward, general relativity breaks down at the Planck scale, where quantum gravity would be required and no accepted theory exists.
What preceded that state, or whether "preceded" is meaningful, is not addressed by the model. Proposals include a bounce, eternal inflation, and a quantum origin, none of which is established.
The philosophical argument from the universe's beginning is separate and is treated under the cosmological arguments in Philosophy.
"Redshift could be caused by something other than expansion"
Tired light, proposed by Fritz Zwicky in 1929, holds that photons lose energy travelling through space, producing redshift without expansion.
The proposal was taken seriously and was tested rather than dismissed.
It fails on time dilation. Distant Type Ia supernovae take longer to rise and fade than nearby ones, by exactly the factor (1+z) that expansion predicts — a supernova at z=1 evolves at half speed. Tired light predicts no such stretching.
It also cannot produce a blackbody spectrum, since scattering that removes energy from photons would blur images of distant galaxies, and distant galaxies are not blurred.
"Dark matter and dark energy are invented to save the model"
A fair concern in form: two components making up about 95% of the energy content are inferred rather than directly detected.
They are inferred from independent lines of evidence rather than added once to patch a single problem. Dark matter is required by galaxy rotation curves, galaxy cluster velocities, gravitational lensing, the CMB power spectrum, and structure formation — five distinct observations.
The Bullet Cluster is the most direct case: two colliding clusters where the gravitational lensing mass, mapped independently, is separated from the visible hot gas. That separation is difficult for modified-gravity alternatives, which predict lensing to follow the visible matter.
Dark energy was not predicted and its discovery in 1998 surprised the field, which is not the pattern of a component invented for convenience.
"Distant galaxies look too mature, which contradicts the model"
Observations from JWST have found galaxies earlier and more massive than many models anticipated, and this is a genuine result reported in the literature.
It is a problem for galaxy formation models rather than for the expansion framework. The age of the universe comes from the CMB and the expansion rate, not from how quickly galaxies assembled.
The general trend also still holds: distant galaxies are on average smaller, more irregular, and more actively star-forming than nearby ones, which is what evolution over time predicts.
Unexpected results prompting model revision is ordinary, and the revisions concern star formation efficiency rather than the age.
"The Hubble tension shows cosmology is broken"
Real and unresolved. Measurements of the expansion rate from the CMB give about 67 km/s/Mpc, while measurements from Cepheids and supernovae give about 73. The discrepancy exceeds the stated uncertainties.
This is a serious open problem and is treated as such by cosmologists, with proposals ranging from unrecognised systematic errors to new physics.
The disagreement is about 10%, which does not bear on whether the universe is billions or thousands of years old. Both values imply an age near 13 to 14 billion years.
What the evidence shows
The model predicted the microwave background sixteen years before it was detected, and predicted its blackbody spectrum, which was confirmed to extraordinary precision in 1990.
It predicts light element abundances from first principles, and the hydrogen, helium, and deuterium values match, with lithium-7 remaining discrepant.
The competing steady-state model was abandoned because it could not account for these observations, not because of institutional preference — its leading proponent held out against the change.
Open problems exist, including the Hubble tension and the lithium abundance. Neither concerns the age scale, which is fixed at 13.8 billion years by CMB measurements and independently supported by the ages of the oldest stars and by distant starlight.