Galaxies & Cosmology

Cosmology has become a precision science with two enormous holes in it. The universe is 13.8 billion years old, it is expanding, and that expansion is accelerating; the geometry is flat and the composition is roughly 5% ordinary matter, 27% dark matter and 68% dark energy. Every part of that is measured to a few percent, and the last two terms are names for things nobody has identified.

The confidence comes from the cosmic microwave background, which is the best-measured thing in astronomy: light released when the universe first cooled enough to be transparent, 380,000 years in, carrying temperature ripples of one part in a hundred thousand. The size of those ripples is a ruler of known length, and how large it appears now measures the geometry and the contents at once. Dark matter is not inferred from one measurement but from four that have nothing to do with each other — galaxy rotation curves, the ripple spectrum, the way clusters bend light behind them, and the abundance of light elements — and they agree on how much there is. That is why it survives despite forty years of increasingly sensitive detectors on Earth failing to catch a single particle of it.

Two active problems dominate. The Hubble tension: the expansion rate measured locally with supernovae comes out around 73, and the value derived from the microwave background around 67, in the same units — and thirty years of shrinking error bars has widened rather than closed the gap, which is the opposite of what a measurement error usually does. And the early universe: Webb keeps finding galaxies larger, brighter and more mature than the models say should exist that soon after the Big Bang. Neither is a crisis yet. Both are the kind of disagreement that has historically preceded something interesting, and both are now precise enough that one of the two numbers has to be wrong in an interesting way rather than a boring one.