The Poincare Conjecture: The Problem That Made a Mathematician Refuse a Million Dollars

In 2010, the Clay Mathematics Institute offered Grigori Perelman $1 million for solving one of the seven Millennium Prize Problems. He said no. He’d already declined the Fields Medal in 2006. “I’m not interested in money or fame,” he reportedly told a journalist through his apartment door in St. Petersburg.

The problem he solved had been open for 99 years. It’s called the Poincare Conjecture, and understanding it requires rethinking what you know about shapes, surfaces, and the meaning of “same.”

What the conjecture says (in plain language)

Imagine you have a rubber band stretched around the surface of an apple. You can slide it, shrink it, move it around, and eventually contract it down to a single point without ever lifting it off the surface. The apple’s surface is “simply connected.” Every loop can be shrunk to a point.

Now try the same thing with a doughnut (a torus). A rubber band that goes through the hole can’t be shrunk to a point without tearing it or leaving the surface. The doughnut is NOT simply connected. It has a hole that loops can get stuck around.

In two dimensions, this completely characterizes the sphere: the only closed, simply connected 2D surface is the sphere. No other shape has this property.

In 1904, Henri Poincare asked: does the same thing work one dimension higher? If a three-dimensional shape is closed (finite, no boundary), and every loop on it can be shrunk to a point, must it be a 3-sphere?

Formally: every simply connected, closed 3-manifold is homeomorphic to the 3-sphere \(S^3\).

What the technical terms mean

Simply connected. Every closed loop can be continuously shrunk to a single point without tearing or leaving the space. The apple surface is simply connected. The doughnut is not.

Closed 3-manifold. A three-dimensional space that is finite (compact), has no boundary (no edges), and locally looks like ordinary 3D Euclidean space at every point. Think of it as the 3D analog of a surface: if you zoom in close enough, it looks flat, but the global shape might be curved.

3-sphere \(S^3\). This is NOT a ball. A ball is a solid 3D object. The 3-sphere is a surface that lives in 4D space, just as the ordinary sphere (2-sphere) is a surface that lives in 3D space. You can’t visualize it directly because your visual system is 3D, but the mathematics describes it precisely.

Homeomorphic. Two shapes are homeomorphic if one can be continuously deformed into the other without tearing or gluing. A coffee mug is homeomorphic to a doughnut (both have one hole). A sphere is NOT homeomorphic to a doughnut. Topology cares about connectivity, not geometry.

Why dimension 3 was the hardest

This is the counterintuitive part. The Poincare Conjecture in higher dimensions was proved first.

Stephen Smale proved it for dimensions 5 and above in 1961 (Fields Medal, 1966). Michael Freedman proved it for dimension 4 in 1982 (Fields Medal, 1986). The two-dimensional case is classical, known since the 19th century.

Dimension 3 resisted all attacks for nearly a century. The techniques that worked in higher dimensions (where there’s “more room to maneuver”) failed in 3D. And the 4D techniques relied on properties specific to that dimension. Three dimensions sat in a gap: too constrained for high-dimensional methods, too complex for low-dimensional ones.

The Clay Mathematics Institute named it one of seven Millennium Prize Problems in 2000, each carrying a $1 million reward. As of 2026, it remains the only Millennium Problem that has been solved.

Perelman’s proof: Ricci flow with surgery

The key idea came from Richard Hamilton, who introduced Ricci flow in 1982. Ricci flow is a way of deforming the geometry of a manifold, smoothing out its curvature over time. It’s analogous to the heat equation: just as heat diffuses from hot spots to cold spots until the temperature is uniform, Ricci flow smooths out curvature differences until the geometry becomes uniform.

Hamilton’s program was: start with any 3-manifold, run Ricci flow on it, and watch the geometry simplify until you can identify the underlying topology. If it simplifies to a round sphere, the manifold was a 3-sphere. If it simplifies to something else, it wasn’t.

The problem: Ricci flow develops singularities. The curvature can blow up at certain points, creating pinch points where the manifold tries to tear itself apart. Hamilton couldn’t handle these singularities systematically.

Perelman solved this in 2002-2003 by introducing “Ricci flow with surgery.” When a singularity forms, you cut the manifold at the pinch point, cap off the resulting pieces with spherical caps, and continue the flow on each piece separately. Perelman proved that this procedure terminates after finitely many surgeries and that the final result classifies the manifold.

He posted three papers on arXiv (the open-access preprint server), never submitted them to a journal, and gave a series of lectures at MIT and Stony Brook in 2003. The mathematical community spent several years verifying the proof. By 2006, three independent teams confirmed it was correct.

Perelman actually proved something bigger

The Poincare Conjecture was a special case. Perelman’s proof actually established the Thurston Geometrization Conjecture, proposed by William Thurston in 1982. This conjecture classifies ALL compact 3-manifolds by decomposing them into pieces, each of which carries one of eight possible geometric structures.

The Poincare Conjecture falls out as a corollary: if a 3-manifold is simply connected and closed, the only possible geometric structure (under Thurston’s classification) is the round 3-sphere geometry. So it must be a 3-sphere.

Proving the full Geometrization Conjecture is like proving that every building can be decomposed into rooms of eight specific shapes. The Poincare case is just: “If a building has no internal walls, it must be a single room of shape 1.” True, but far less than the full result.

Why Perelman walked away

Perelman declined the Fields Medal in 2006. He declined the Millennium Prize in 2010. He resigned from the Steklov Institute of Mathematics. He effectively withdrew from mathematics entirely.

He never fully explained his reasons. In limited public statements, he expressed dissatisfaction with the mathematical community’s ethical standards, particularly around credit and collaboration. He felt that the verification process had been used by others to claim partial credit for his work.

“I’m not interested in money or fame. I don’t want to be on display like an animal in a zoo.” This was reported by a journalist who visited his apartment in St. Petersburg, where he lives with his mother.

Whatever his reasons, the result stands. The Poincare Conjecture is proved. The classification of 3-manifolds is complete. And the only person who has ever solved a Millennium Prize Problem chose not to collect the prize. In a field obsessed with recognition, Perelman’s refusal may be the most mathematically interesting thing about the entire story.

Frequently Asked Questions