The 95% Problem: Setting the Stage

The 95% Problem: Setting the Stage

For the entirety of human history, the stars we see, the planets we walk on, and the air we breathe have been the only matter known to exist. Philosophers from Aristotle to Descartes assumed that the cosmos was, at bottom, made of the stuff we could touch. Even after the revolutions of the twentieth century — quantum mechanics, general relativity, the discovery of galaxies — the basic intuition held: the universe is mostly ordinary matter, and we just need to find more of it.

That intuition is now dead. Cosmological observations, sharpened over the last three decades by space telescopes and galaxy surveys of extraordinary precision, have arrived at a conclusion so startling that some physicists initially resisted it. The atoms that make up stars, planets, and people — every proton, neutron, and electron that high-school chemistry catalogues — account for only about 5% of the total energy content of the universe. The remaining 95% is, in the literal language of the field, "dark." We cannot see it directly. We do not know what it is. And yet, by every independent measurement we know how to make, it is there.

This is not a small mystery, a loose end, a missing decimal point. It is a problem of a kind that has no real precedent in the history of physics. The previous "missing" things — Neptune's gravity before its discovery, the neutrino before its detection, the positron predicted by Dirac — were all *predicted* before they were *found*. In each case, theory ran ahead of observation and told observers exactly where to look. Here the situation is reversed: observation has accumulated, with no theoretical road map, two enormous components of the cosmos that have refused, for over twenty years, to yield to direct detection. This book is about those two components — dark matter and dark energy — and about the third, equally consequential component: the human effort to figure out what they actually are.

Before we can talk about candidates, detectors, or future missions, we need to fix in the reader's mind just how *much* of the universe is dark, how *precisely* we know that number, and what we *do* and *do not* mean when we use the word "dark." That is the work of this opening chapter.

A Quantitative Census of the Cosmos

The single most important table in modern cosmology is short enough to fit on a postcard, and it summarises roughly thirteen billion years of cosmic history. It comes from combining the measurements of multiple observatories, but the dominant contributions are the Cosmic Microwave Background (CMB) anisotropies measured by the *Planck* satellite (final results released in 2018) and the Baryon Acoustic Oscillation (BAO) measurements from the Dark Energy Spectroscopic Instrument (DESI), whose first three years of data were published in 2024 and updated in early 2025.

The parameters that matter for the cosmic census are dimensionless ratios called density parameters, written with the Greek letter omega: $\Omega_i = \rho_i / \rho_c$, where $\rho_i$ is the energy density of the *i*-th component and $\rho_c$ is the critical density — a special reference value that the universe would need to have if it were exactly spatially flat (a notion we will define carefully in a moment). Three density parameters dominate the conversation:

| Component | Symbol | Planck 2018 value | DESI 2024 (combined) | |---|---|---|---| | Baryonic (ordinary) matter | $\Omega_b$ | $0.0486 \pm 0.0010$ | $0.0467 \pm 0.0015$ | | Cold dark matter | $\Omega_c$ | $0.260 \pm 0.003$ | $0.266 \pm 0.003$ | | Dark energy (cosmological constant) | $\Omega_\Lambda$ | $0.691 \pm 0.006$ | $0.685 \pm 0.007$ | | Total | $\Omega_{\rm tot}$ | $1.000 \pm 0.002$ | $1.000 \pm 0.002$ |

A reader unfamiliar with cosmology might wonder how on earth anyone can quote a percentage of the universe with three decimal places of precision. The short answer is that these numbers are not measured directly; they are inferred from the *statistics* of the CMB temperature map and from the *spatial pattern* of galaxies at large scales. Both datasets are enormous — *Planck* measured the temperature of the sky at millions of pixels, and DESI has catalogued tens of millions of galaxy positions — and both rely on physics that is, in principle, calculable from first principles. The numerical precision is the result of a kind of triangulation: different physical effects at different epochs leave different statistical fingerprints, and a single underlying cosmological model has to explain them all simultaneously. The tiny error bars are the price of that consistency.

Notice two crucial features of the table. First, the *total* density parameter is, within measurement error, exactly equal to 1. The universe, on the largest scales, is spatially flat — a statement we will unpack in the next section. Second, the baryonic component is barely 5% of the total. If you were to add up every atom in every star, planet, galaxy, and gas cloud in the observable universe, you would have less than a twentieth of the cosmic budget. The other nineteen-twentieths is *dark*, in two flavours: cold dark matter (~27%) and dark energy (~68%).

```mermaid flowchart LR A["Cosmic Mass-Energy Budget"] --> B["Ordinary Matter: ~4.9%"] A --> C["Cold Dark Matter: ~26.5%"] A --> D["Dark Energy: ~68.5%"] B --> B1["Stars"] B --> B2["Planets, gas, dust"] B --> B3["Neutrinos (~0.1%)"] C --> C1["Unknown particle(s)"] C --> C2["Holds galaxies together"] C --> C3["Forms cosmic web"] D --> D1["Drives acceleration"] D --> D2["May be vacuum energy"] D --> D3["Discovered 1998"] ```

This pie-chart representation — the iconic image of modern cosmology — is one of the most reproduced figures in any scientific field, and yet the *meaning* of each slice is often glossed over. Let us give each one a careful definition.

Ordinary (baryonic) matter. The term "baryon" is a holdover from particle physics; it refers to any particle made of three quarks, of which the proton and neutron are the only stable examples in everyday matter. Strictly speaking, electrons are *leptons*, not baryons, but cosmologists use "baryonic matter" loosely to mean all matter made of standard-model particles that interacts via the electromagnetic force. $\Omega_b \approx 0.049$ includes the mass of every star, planet, galaxy, gas cloud, dust grain, and free-floating atom in the observable universe. It also includes a non-trivial contribution from *neutrinos*, the nearly-massless particles that stream through space at nearly the speed of light. The neutrino contribution is small — roughly $\Omega_\nu \approx 0.001$ — but it is, in a literal sense, part of the "ordinary" matter budget. In popular accounts, the 5% is often described as "everything we can see," but this is slightly misleading: roughly half of the baryons in the universe are not in stars. They are in the warm-hot intergalactic medium, in diffuse gas filaments, and in the haloes of galaxies — they emit so little light that they are essentially invisible to optical telescopes, but they are still made of atoms and so they still count as baryonic.

Cold dark matter (CDM). The "cold" in CDM is a technical term meaning that the particles were moving slowly compared to the speed of light when they fell out of thermal equilibrium in the early universe. This matters for structure formation: hot dark matter (like standard neutrinos) would zip through gravitational wells too quickly to clump, while cold dark matter can settle into small, dense haloes that grow into galaxies. $\Omega_c \approx 0.265$ refers to matter that is *gravitationally* abundant — it pulls on galaxies, bends light, slows galaxy clusters — but emits essentially no electromagnetic radiation of any kind. It cannot be seen. Its presence is inferred entirely from its gravitational influence on things we *can* see. We will spend most of Chapters 2–6 e