Module 6: The first three minutes and after · Lesson 6.2 · about 15 min
Big Bang or Steady State? The great argument
From 1948 to 1965 two pictures of the expanding universe fought it out: a hot beginning, or a universe with no beginning that makes new matter to stay the same forever. How radio source counts, quasars, helium and the cosmic microwave background settled it.
Module 6 · about 15 minutes. Read first: 6.1 Gamow, Alpher and Herman and 4.8 The cosmological principle. An expanding universe does not have to have a beginning, if it can make new matter as it grows. For seventeen years that idea was a serious rival to the Big Bang. By the end you will be able to explain how each picture made its own predictions and which observations decided between them.
1. A universe with no beginning
In 1948, in Cambridge, three young scientists, Hermann Bondi, Thomas Gold and Fred Hoyle, disliked the idea of a beginning. A creation event cannot be studied, and in the 1940s the Big Bang also had an embarrassing problem: Hubble’s expansion rate, about 500 km/s per megaparsec, made the universe only about 2 billion years old, younger than the Earth’s rocks.
They proposed the perfect cosmological principle: the universe is the same not only everywhere (Lesson 4.8) but also at every time. It expands, so galaxies drift apart; and new matter appears, very slowly, so that new galaxies form in the gaps. A fountain in a park looks the same all day though its water keeps changing: the Steady State universe is a fountain, not a firework.
2. How much new matter?
To keep the density fixed while every cubic metre of space grows, new matter must appear at a steady rate. A cube of space of side L grows in volume as L³, and L grows at the rate H, so the volume grows at the rate 3H:
- n
- atoms per cubic metre, which in the Steady State never changes (Hoyle took the critical density, about 5 hydrogen atoms per m³)
- 3H0
- how fast each cubic metre of space grows in volume: three times the expansion rate, because space grows in three directions
- Physical meaning
- About one new hydrogen atom per cubic metre every billion years: one atom in a large room every few tens of millions of years. Far too slow to notice, so the theory could not be tested by looking for creation directly.
Creating matter from nothing breaks the law of conservation of energy, and critics said so. Hoyle answered that a Big Bang creates the whole universe at once, which is no better. The argument could only be settled by looking.
3. Four tests
The two pictures differ most in what they say about the past. In a Big Bang universe, looking far away (Lesson 2.1) shows a younger, denser, different universe. In a Steady State universe, the far past must look like today.
4. After the argument
Most astronomers accepted the hot Big Bang after 1965. Hoyle never did. With Geoffrey Burbidge and the Indian astrophysicist Jayant Narlikar, a former student of Hoyle’s who later founded the Inter-University Centre for Astronomy and Astrophysics (IUCAA) in Pune, he proposed a quasi-steady state model in 1993, with matter created in bursts. It never matched the precise measurements of the microwave background, and few cosmologists follow it today.
The argument was good for science. Because the Steady State made firm predictions, it pushed astronomers to count radio sources, hunt for quasars and measure helium, and in the end it was decided by observation, not by taste.
A rivalry with a famous name. On 28 March 1949, explaining the two theories on BBC radio, Hoyle called the rival idea a “big bang”, and the name stuck (Lesson 4.4). Hoyle and Ryle, both in Cambridge, argued in public for years, and Ryle’s first radio counts (1955) turned out to be partly wrong, which made Hoyle’s camp distrust the later, better ones. Science settled it anyway.
- 1948 Bondi and Gold; Hoyle: the Steady State, with continuous creation.
- 1949 Hoyle says "big bang" on the radio.
- 1952 Baade doubles the distance scale; the age problem eases.
- 1961 Ryle and Clarke: too many faint radio sources for a Steady State.
- 1963 Schmidt: quasars are far away; they belong to the past.
- 1965 The cosmic microwave background is found.
- 1993 Hoyle, Burbidge and Narlikar's quasi-steady state model.
“The Steady State was a silly idea that only stubborn people believed.” It was a serious, elegant theory held by excellent scientists, and it was more testable than the early Big Bang: it allowed no change at all, so any sign of change anywhere could disprove it. That is exactly what happened.
- Steady State (Bondi, Gold and Hoyle, 1948): the universe expands but looks the same at all times; new matter appears at 3H0n, about one atom per cubic metre per billion years.
- It solved the 1940s age problem, which went away when distances were corrected in the 1950s.
- Radio source counts (more faint sources than allowed) and quasars (common only far away, long ago) showed the universe changes with time.
- 25% helium everywhere and the cosmic microwave background (1965) fit a hot beginning and settled the argument.
- Hoyle, Burbidge and Narlikar’s quasi-steady state (1993) kept the idea alive but does not fit the CMB.
Check yourself
Sources
- H. Bondi and T. Gold (1948), The steady-state theory of the expanding universe, MNRAS 108, 252 https://doi.org/10.1093/mnras/108.3.252
- F. Hoyle (1948), A new model for the expanding universe, MNRAS 108, 372 https://doi.org/10.1093/mnras/108.5.372
- M. Ryle and R. W. Clarke (1961), An examination of the steady-state model in the light of some recent observations of radio sources, MNRAS 122, 349 https://doi.org/10.1093/mnras/122.4.349
- M. Schmidt (1963), 3C 273: a star-like object with large red-shift, Nature 197, 1040 https://doi.org/10.1038/1971040a0
- F. Hoyle, G. Burbidge and J. V. Narlikar (1993), A quasi-steady state cosmological model with creation of matter, ApJ 410, 437 https://doi.org/10.1086/172761
- H. Kragh (1996), Cosmology and Controversy: The Historical Development of Two Theories of the Universe, Princeton University Press
Your ticks stay in this browser. Save them as a file on the course home.
Updated 2026-11-22. Numbers are rounded for teaching; measured values come from the sources listed in each lesson. Simulations run in your browser; nothing you type is sent anywhere.