In 1964, two radio astronomers named Arno Penzias and Robert Wilson were trying to clean up their radio telescope. The horn antenna they were using at Bell Labs in Holmdel, New Jersey was picking up a faint, persistent hiss. The hiss was there during the day. It was there at night. It was there in every direction they pointed the antenna. It seemed to come from everywhere at once.
They tried everything. They cleaned the antenna. They evicted a pair of nesting pigeons (this is a real story; the cleaning process they politely called "removing white dielectric material"). The hiss did not go away.
What they did not know — and what they would learn from a phone call to a group of physicists at Princeton — was that they had just heard the universe.
The signal that frustrated them is now called the cosmic microwave background, or CMB. It is the oldest light in existence. And the discovery of it remains one of the strongest pieces of evidence we have that the universe began, expanded, and cooled in a way that physicists had predicted on paper a generation earlier.
What the CMB actually is
To understand the CMB, you have to understand what the early universe was like.
In the first few hundred thousand years after the Big Bang, the universe was extraordinarily dense and extraordinarily hot. So hot that hydrogen could not stay together as a stable atom. Every time a proton tried to capture an electron, a high-energy photon would knock the electron loose again. The whole universe was an opaque, glowing plasma — a fog of free electrons, protons, and constantly scattering light. You could not see across it. Light could not travel any meaningful distance before bouncing into another particle.
Then, about 380,000 years after the Big Bang, the universe had expanded enough — and cooled enough, to about 3,000 Kelvin — that protons and electrons could finally bind into neutral hydrogen. This event is called recombination, slightly misleadingly, since they had never been combined before.
The instant the fog cleared, light could travel freely. The photons that had last scattered just before recombination were released into a now-transparent universe. They have been streaming through space ever since.
Those photons are still arriving. Today, after 13.8 billion years of cosmic expansion, they have been stretched by a factor of about 1,100 — so the visible-light glow of a 3,000 K plasma is now redshifted into the microwave part of the spectrum, with an effective temperature of roughly 2.725 K, just under three degrees above absolute zero.
That faint, cold, nearly uniform glow filling all of space is the cosmic microwave background.
How we know what it is
The reason cosmologists are so confident about this story is that the CMB has three properties that fit the Big Bang prediction with almost embarrassing precision.
1. It has a near-perfect blackbody spectrum. A blackbody is an idealized object that emits radiation purely as a function of its temperature. The CMB's spectrum, measured by the COBE satellite's FIRAS instrument in 1990 (Mather et al., Astrophysical Journal, 1990), matches a 2.725 K blackbody better than any other naturally occurring spectrum ever measured. There is no astrophysical "process" that would naturally produce this. There is, however, a hot expanding plasma cooling down, which produces it inevitably.
2. It is almost — but not quite — uniform. The CMB is essentially the same temperature everywhere we point our telescopes, to about one part in 100,000. But the tiny, careful deviations from uniformity (called anisotropies) are exactly the imprints predicted by inflationary models of the early universe. They are the seeds of all later structure: galaxies, clusters, and us.
3. The pattern of anisotropies has structure that fits cosmological theory in detail. Beginning with COBE (1992), continuing with WMAP (2001–2010) and Planck (2009–2013), satellites have mapped these temperature variations with increasing precision. The size and statistics of the patterns let cosmologists infer:
- The age of the universe (about 13.8 billion years)
- Its overall geometry (very close to flat)
- The fraction of ordinary matter, dark matter, and dark energy (roughly 5%, 27%, and 68%)
- The basic conditions of the universe at the moment the CMB was emitted
These numbers come out the same when measured by completely different methods. That kind of independent convergence is what gives cosmologists confidence that the model is something more than a story.
Why the discovery was such a turning point
Before 1964, cosmology was an unusually contested branch of physics. The Big Bang theory had a serious rival in the steady-state model, championed by Fred Hoyle, Hermann Bondi, and Thomas Gold, in which the universe was eternal, with new matter continuously created to keep its density constant.
Both models could explain the expansion of the universe (observed by Hubble and others in the 1920s). The decisive test was whether the early universe should have been hot. Big Bang cosmologists, notably Ralph Alpher, Robert Herman, and George Gamow, had predicted in 1948 that the universe should still be glowing faintly with the cooled relic radiation of that early hot phase. The steady-state theory predicted no such glow.
Penzias and Wilson's measurement of about 3 K of background radiation was the unambiguous detection of that glow. The two received the Nobel Prize in Physics in 1978. The steady-state model never recovered.
What the CMB is still telling us
The CMB is not a settled chapter. Modern cosmology is built on it, and many of the most active questions in the field involve squeezing more information out of it.
- Polarization. The CMB is slightly polarized, and patterns in that polarization can in principle reveal whether the very early universe underwent cosmic inflation — a brief period of exponential expansion. Detection of certain polarization patterns ("B-modes" from gravitational waves) would be one of the most important confirmations imaginable. The search continues.
- Tensions in the data. The Hubble constant inferred from the CMB ("early universe") differs from the value measured from nearby supernovae ("late universe") by a small but stubborn amount. Whether this is a hint of new physics or a measurement issue is one of the active arguments of contemporary cosmology.
- Future missions. Ground-based experiments like the Simons Observatory and proposed successors aim to map the polarization of the CMB to far higher precision over the coming decade.
A small revolution in attention
Sit somewhere quiet at night. The light that fell on your eyes from the moon is about a second old. The light from the sun is about eight minutes old. The light from the brightest stars in the sky is decades to thousands of years old. The light from the farthest galaxies is billions of years old.
And washing across all of it, in every direction, is light older than every galaxy — light released the first moment the universe became transparent enough for it to escape. It is too cold and too low-frequency for our eyes. But it is there, all around us, all the time. A whisper from the universe before there were stars to whisper from.
The cosmic microwave background is the closest thing in physics to a baby picture of the cosmos. The remarkable fact is not just that it exists. It is that, by listening carefully to its temperature, its uniformity, and its tiny imperfections, we have managed to learn the universe's age, its composition, and the conditions of its earliest moments — from a hiss that two engineers spent a year trying to clean off their antenna.
