Topology

Topology begins with a deceptively simple provocation: what is left of a shape once you strip away all notion of distance, angle, and rigid measurement? The answer, it turns out, is a great deal — a rich landscape of structure that has reshaped nearly every corner of modern mathematics and theoretical physics. Where geometry asks how long, how wide, how curved, topology asks only whether: whether two points can be connected, whether a surface has a boundary, whether a loop can be shrunk to a point without tearing the space it lives in. These qualitative questions, once posed carefully, demand powerful new machinery to answer, and that machinery has proved unexpectedly universal.

The subject traces one of its earliest landmarks to Leonhard Euler, who in 1736 solved the problem of the Königsberg bridges by observing that the answer depended only on how the landmasses were connected — not on any particular distances or shapes. His solution inaugurated a combinatorial way of thinking about space that would eventually grow into the full edifice of topology. But the modern subject was born in the late nineteenth and early twentieth centuries, when Bernhard Riemann’s work on surfaces, Henri Poincaré’s invention of what he called analysis situs, and Georg Cantor’s set-theoretic revolution converged into something genuinely new. Poincaré in particular glimpsed the depth of the enterprise: his conjecture about three-dimensional spheres, posed in 1904, would not be resolved until Grigori Perelman’s proof in the early 2000s, a hundred years later.

Felix Hausdorff gave topology its axiomatic foundations in 1914, introducing the abstract notion of a topological space through open sets rather than through distances or coordinates. This abstraction was radical: it meant that topology could speak about spaces that had no obvious geometric realization, that continuity could be defined without ever measuring how close two points were. Around the same time, L.E.J. Brouwer was proving results that would become cornerstones of the field — the fixed-point theorem that bears his name (any continuous map from a closed disk to itself must fix some point), and the invariance of domain, which showed that continuous bijections between Euclidean spaces of the same dimension behave in the way intuition demands. These early decades established that topology was not merely a weakened form of geometry but a discipline with its own profound problems and techniques.

The branch is organized here into two sub-topics, each representing a distinct but deeply related mode of investigation. General Topology lays the foundational language: what a topological space is, how continuity is captured by open sets, and which properties — compactness, connectedness, separation — give spaces their essential character. It develops the theory of manifolds, the spaces that locally resemble Euclidean space and provide the natural setting for differential geometry and physics, and it introduces knot theory, where the question becomes whether one loop embedded in three-dimensional space can be continuously deformed into another without passing through itself. General topology also explores covering spaces and the fundamental group, the first bridge between the continuous and the algebraic, where loops based at a point encode information about holes in a space through the language of group theory.

Algebraic Topology takes that bridge and builds a highway. The central insight, developed by Poincaré and vastly extended through the twentieth century by figures such as Emmy Noether, Heinz Hopf, Samuel Eilenberg, and Saunders Mac Lane, is that to every topological space one can associate algebraic invariants — groups, rings, modules — that remain unchanged under continuous deformation. If two spaces have different invariants, they cannot be homeomorphic, no matter how clever the attempted deformation. Singular homology assigns a sequence of abelian groups to a space, capturing the presence of holes in each dimension: zero-dimensional connected components, one-dimensional loops, two-dimensional voids, and so on. Cohomology refines and dualizes this picture, introducing a ring structure that encodes how cycles intersect. The Mayer-Vietoris theorem provides a powerful computational engine: the homology of a space assembled from pieces can be computed from the homology of those pieces and their overlap, in exact analogy with inclusion-exclusion. Higher homotopy groups, fiber bundles, spectral sequences, characteristic classes, K-theory, cobordism, and index theory all build on these foundations, each extending the reach of algebraic methods into the topological world.

What unifies these two sub-topics is a shared conviction that the right way to understand a space is to ask what survives deformation. This conviction has had extraordinary consequences. Topological data analysis uses persistent homology to detect shape and structure in high-dimensional data sets, finding loops and voids that statistical methods miss. Theoretical physicists use fiber bundles and characteristic classes to formulate gauge theories and string theory. Knot invariants computed by algebraic topology have connections to quantum field theory that are still being actively explored. The classification of surfaces, manifolds, and higher-dimensional spaces remains one of the driving programs of geometry and topology alike, with the tools of both sub-topics working in tandem. Topology has earned its place not as a curiosity about coffee cups and donuts, but as the language in which questions about the shape of the universe are most naturally asked.