Mathematics
Elliptic Operators and the Index Theorem
Quick fact
The Atiyah–Singer index theorem, proven in 1963, shows that the analytic index of an elliptic operator—the difference between the dimensions of its kernel and cokernel—equals a purely topological quantity computable from the manifold's geometry. This links the seemingly analytic notion of solvability to global topology, with applications from particle physics to pure mathematics.
Why this is interesting
You know that solving a differential equation often depends on the shape of the space. But what if the number of solutions minus the number of obstructions tells you something about the shape itself?
Read the full explanation
Understanding Elliptic Operators and the Index Theorem
Imagine a differential equation on a curved surface, like the Laplace equation for heat distribution. Its solutions form a vector space, and sometimes there are obstructions that prevent a solution from existing. The difference between the number of independent solutions and the number of obstructions is the 'index'. For a special class of differential operators called elliptic operators, this index is finite and, surprisingly, does not change when you perturb the operator slightly. This stability hints that the index is not just an analytic accident but a reflection of the underlying space's topology, its global shape. The index theorem makes this precise: it tells you that this analytic number can be computed using only the topological data of the manifold and the operator's symbol.
A deeper explanation
An elliptic operator is a differential operator whose principal symbol is invertible for every nonzero cotangent vector. This condition ensures that the operator is Fredholm on appropriate Sobolev spaces: its kernel is finite-dimensional, its range is closed, and its cokernel is finite-dimensional. The analytic index is defined as the integer dim(ker) − dim(coker). The Atiyah–Singer index theorem states that this analytic index equals the topological index, which is obtained by integrating a cohomology class—constructed from the operator's symbol and the manifold's characteristic classes—over the manifold. The proof involves showing both indices are invariant under certain deformations and then verifying equality on a set of 'generating' cases. This theorem unifies many classical results, such as the Gauss–Bonnet theorem and the Hirzebruch–Riemann–Roch theorem, and has profound consequences in geometry, topology, and mathematical physics.