Yes. Dark matter has gravity, and that gravity is the main reason we know dark matter exists. Every rotation curve, gravitational lens, and cluster collision that revealed the missing mass depends on dark matter attracting ordinary matter and other dark matter gravitationally. The open question is not whether dark matter gravitates, but whether it does anything else.

That answer sounds circular at first, so it helps to separate two claims. The first is that dark matter responds to gravity exactly as general relativity predicts for any mass-energy. The second is that dark matter exerts no significant force besides gravity. The evidence strongly supports the first claim and places tight but not airtight limits on the second.

Mechanism: How Gravity Reveals Invisible Mass

Gravity is the only fundamental force that cannot be screened or neutralized. Electromagnetic forces cancel in bulk matter because positive and negative charges pair up. Gravity has no negative charge analog, so every kilogram of dark matter adds to the total gravitational field of a galaxy, cluster, or cosmic web filament.

Fritz Zwicky first applied this logic to the Coma Cluster in 1933. Galaxies near the cluster's edge moved as if the cluster held roughly 400 times more mass than its visible stars provided (Zwicky, Helvetica Physica Acta, 1933). Vera Rubin and Kent Ford confirmed the pattern in individual spiral galaxies in the 1970s. Stars far from a galaxy's center should slow down under Newtonian dynamics, but their orbital speeds stayed nearly flat at 200 to 250 kilometers per second.

The deeper mechanism is gravitational lensing. Mass curves spacetime, and curved spacetime bends light. When a massive cluster sits between Earth and a background galaxy, the cluster's total mass, visible and dark, produces arcs and multiple images. The lensing signal maps mass directly and does not care whether the mass emits light. Lensing surveys consistently find that dark matter outweighs ordinary matter by roughly five to six to one.

There is also a structural reason dark matter could not simply be dim ordinary matter. Big Bang nucleosynthesis, the fusion window in the universe's first minutes, fixed the abundance of hydrogen, helium, and lithium. Those atomic ratios match observations only if ordinary matter makes up about 5 percent of the universe's energy budget. The remaining matter must be something that does not participate in nuclear reactions.

Quantified Context: The Cosmic Mass Budget

Current Planck satellite measurements put the universe at roughly 68 percent dark energy, 27 percent dark matter, and 5 percent ordinary matter. In absolute terms, dark matter amounts to about 2.4 × 10^53 kilograms spread across the observable universe, several times the mass of all the gas and stars combined.

At galaxy scales, the numbers repeat with striking consistency. The Milky Way's rotation curve stays flat at about 220 kilometers per second out to at least 80,000 light-years, far beyond the visible disk. Fitting that curve requires a dark matter halo of roughly 1 to 2 × 10^12 solar masses. The visible stars and gas supply only about 60 billion solar masses, a fraction of the total.

Evidence type What it measures Typical dark-to-visible mass ratio
Spiral galaxy rotation curves Orbital speed versus radius 5 to 10
Galaxy cluster lensing Total mass via light bending 5 to 6
Cluster galaxy velocities Velocity dispersion of member galaxies 5 to 10
Cosmic microwave background Global energy budget About 5.4
Big Bang nucleosynthesis Ordinary matter ceiling Requires non-baryonic mass
Bullet Cluster collision Mass-gas separation Mass leads gas

Every row assumes dark matter couples to gravity. No alternative has matched this table as economically.

Edge Cases: Collisions, Dwarfs, and the Bullet Cluster

The Bullet Cluster, formally 1E 0657-56, is the sharpest test. Two galaxy clusters collided, and the hot gas between their galaxies, which dominates the ordinary matter, slowed and stalled through ram pressure. Gravitational lensing shows the bulk mass had already passed through and separated from the gas. Dark matter behaved like a collisionless fluid while ordinary gas behaved like a fluid with friction. The mass-to-light offset is about 150 kiloparsecs, and it is hard to reproduce without dark matter that gravitates but barely interacts otherwise.

Dwarf galaxies sketch the opposite edge case. Some ultra-faint dwarfs contain almost no visible matter yet show velocity dispersions implying substantial mass. Segue 1, one of the faintest known galaxies, has a mass-to-light ratio that could exceed 1,000. These objects are natural laboratories for testing whether gravity behaves as expected at extremely low accelerations, the regime where modified gravity theories diverge from dark matter.

The counterintuitive detail: dark matter's gravity is not perfectly smooth. Computer simulations that include only gravity produce halos with dense central cusps, but many observed dwarf galaxies show flatter cores. This core-cusp problem is a live research frontier and a genuine caveat, not a solved issue. It suggests either that dark matter has mild self-interactions or that feedback from star formation reshapes halos.

Misconceptions That Refuse to Die

The most common error is treating dark matter as a kind of cosmic dust. Dust is ordinary matter, absorbs light, and would show up in absorption spectra. Dark matter is transparent to electromagnetic radiation at every wavelength tested. It does not block starlight and does not glow.

A second error confuses dark matter with dark energy. Dark matter clumps and pulls inward, seeding galaxy formation. Dark energy is a uniform pressure that accelerates cosmic expansion. They are unrelated components with opposite effects on structure growth.

A third error assumes dark matter must be a particle already discovered. The leading candidates are weakly interacting massive particles, axions, and sterile neutrinos, but no confirmed detection exists as of 2025. Null results from detectors like LZ and XENONnT have pushed WIMP cross-section limits below 10^-47 square centimeters for certain masses, ruling out broad swaths of parameter space without ruling out the framework.

Red-flag warnings for reading popular coverage:

Claims that a lab has "found dark matter" almost always describe a detector anomaly later shown to be background noise. Modified gravity proposals such as MOND fit galaxy rotation curves well but struggle with cluster lensing and the cosmic microwave background. Any source that says dark matter "has no gravity" or "is just a math error" is contradicting five independent lines of evidence. Simulation images labeled as photographs are visualizations; treat them as models, not observations.

Practical Actions: How to Evaluate a Dark Matter Claim

You do not need a telescope to judge whether a claim holds up, but you do need a checklist. The steps below follow the same logic professional reviewers apply.

Ask whether the claim addresses at least two independent evidence lines. Rotation curves, lensing, nucleosynthesis, and the microwave background must all be explained. A theory that fits one and ignores three is not competitive. Check the scale. Results from particle detectors apply to specific mass and interaction ranges. A null result at one mass does not eliminate all dark matter candidates. Distinguish detection from inference. All current evidence for dark matter is gravitational and indirect. Anything labeled a direct detection requires replication by a second experiment. Look for the word "model." Simulations and halo fits are models constrained by data. Treat quoted mass values as ranges, not exact measurements. Demand a stated threshold. Legitimate particle papers report confidence intervals and exposure times. "Significant" without numbers is a marketing word, not a scientific one.

For students, the most useful practical step is to reproduce a rotation curve fit yourself. Public datasets from the SPARC catalog provide observed velocities and radii for over 100 galaxies. Fitting a dark matter halo profile against those points teaches more about the evidence than any review article.

Limitations: What Gravity Cannot Tell Us

Gravity tells us that dark matter has mass. It does not tell us what dark matter is. A particle with gravitational interactions only is indistinguishable, on cosmological scales, from a particle with extremely weak extra interactions. That ambiguity is the field's central frustration.

Laboratory limits define the boundary. Direct detection experiments now constrain spin-independent WIMP-nucleon cross sections to around 10^-48 square centimeters at masses near 40 GeV, and axion haloscopes have covered broad microwave frequency bands without a confirmed signal. These are genuine limits on specific models, not verdicts on the whole concept.

The honest position is asymmetry. Dark matter's gravitational behavior is measured to a few percent precision on cluster and cosmic scales. Its microscopic nature is unknown. Gravity is the door we have opened; everything behind the door remains in shadow. The question in this article's title has a firm answer, and the answer explains why the search continues.