The Fundamental Groupoid of a Space

Core Idea

The fundamental groupoid of a topological space has points as objects and homotopy classes of paths as morphisms, with composition given by path concatenation. Unlike the fundamental group (which depends on a basepoint choice), the fundamental groupoid is base-point-free and captures the full homotopy-theoretic information of the space. It provides a more natural and categorical framework for studying connectivity.

How It's Best Learned

Compute the fundamental groupoid of familiar spaces: the circle, the plane, a figure-eight. Verify that morphisms are invertible and explore how groupoid structure reflects topological properties. Understand the relationship between the fundamental groupoid and fundamental groups at various basepoints.

Common Misconceptions

The fundamental groupoid is not the same as the fundamental group; it encodes information at all points simultaneously. The automorphism group at a point is the fundamental group at that basepoint, but the groupoid structure includes much more.

Explainer

You know that a groupoid is a category in which every morphism is invertible. Objects can be many, not just one, so a groupoid generalizes both groups (one object, all morphisms invertible) and sets (many objects, only identity morphisms). The fundamental groupoid Π₁(X) of a topological space X is the canonical example of a groupoid arising in nature. Its objects are the points of X; its morphisms from point x to point y are homotopy classes of paths from x to y — continuous curves γ: [0,1] → X with γ(0) = x and γ(1) = y, where two paths are identified if one can be continuously deformed into the other while keeping the endpoints fixed.

Composition of morphisms is path concatenation: given a path from x to y and a path from y to z, you travel first along one, then the other, reparametrized to the unit interval. The identity morphism at x is the constant path that stays at x. The inverse of a path γ is the reversed path γ⁻¹(t) = γ(1−t), which traces the same route backwards. Checking the groupoid axioms reduces to standard facts in homotopy theory: concatenation is associative up to homotopy, the constant path is a homotopy identity, and reversing a path gives a homotopy inverse. Every morphism is invertible — that is the groupoid property — because you can always walk backwards.

The fundamental group π₁(X, x₀) based at a chosen point x₀ is the automorphism group Aut_{Π₁(X)}(x₀) in the fundamental groupoid — the collection of all homotopy classes of *loops* at x₀ (paths where γ(0) = γ(1) = x₀). The groupoid sees all basepoints simultaneously. When X is path-connected, all the automorphism groups Aut(x) are isomorphic to each other (conjugate via any path between them), so choosing a basepoint loses no information up to group isomorphism. But when X is disconnected — say, X = {a} ∪ {b}, two separate points — the fundamental groupoid has two objects and only identity morphisms, cleanly encoding the disconnection. No basepoint-based fundamental group can capture this: you'd need to pick a component.

The fundamental groupoid is not merely a notational convenience — it is categorically more natural. Any continuous map f: X → Y induces a functor Π₁(f): Π₁(X) → Π₁(Y), sending points to their images and homotopy classes of paths to their images. This makes Π₁ a functor from topological spaces to groupoids, and the functoriality packages the induced homomorphism on fundamental groups (at any basepoint) into a single, basepoint-free statement. The van Kampen theorem, which computes π₁ of a union of spaces, has a cleaner and more general statement at the groupoid level: Π₁(X ∪ Y) is the pushout of Π₁(X) and Π₁(Y) over Π₁(X ∩ Y) in the category of groupoids.