Loading field map
Ordinary category theory works with hom-sets: for any two objects, the collection of morphisms between them is a set. But many mathematical structures come with richer collections of maps—topological spaces with continuous maps form a topological space themselves, chain complexes with chain maps form a chain complex, and so on. The central insight of enriched category theory is to replace hom-sets with hom-objects taken from a suitable monoidal category V. This shift, initiated in the 1960s, transforms category theory into a flexible framework that can internalize the structure of the very categories it studies.
A V-category C consists of a collection of objects together with, for each pair of objects a,b, an object C(a,b) in a fixed monoidal category (V,⊗,I). Composition is not a function but a morphism in V: C(b,c)⊗C(a,b)→C(a,c). Identity is a morphism I→C(a,a). When V is the category of sets with Cartesian product, a V-category is exactly an ordinary locally small category. Enrichment thus generalizes ordinary category theory by allowing the hom-objects to carry additional structure—topological, algebraic, or homotopical—while preserving the categorical architecture.
The systematic development of enriched category theory began in the 1960s with the work of Samuel Eilenberg and Max Kelly, and later with Brian Day and others. The key challenge was to define not only V-categories but also the appropriate notion of functor and natural transformation that respects the enrichment. A V-functor F:C→D is an assignment on objects together with a family of morphisms in V: F{ab}:C(a,b)→D(Fa,Fb) that commutes with composition and identities. A V-natural transformation α:F⇒G is a family of morphisms αa:I→D(Fa,Ga) satisfying a V-enriched version of the usual naturality condition. These definitions, codified in Kelly's 1982 monograph Basic Concepts of Enriched Category Theory, provide a complete categorical language that works entirely within V.
Two technical innovations distinguish enriched category theory from a mere translation of ordinary concepts. The first is the theory of weighted limits and colimits. In ordinary category theory, limits are defined by cones; in the enriched setting, the shape of a diagram is not enough—one must also specify a weight, a V-functor W:D→V, that determines how the limit object is built from the hom-objects. Weighted limits generalize conical limits (which correspond to constant weight) and are essential for constructing enriched Yoneda embeddings, ends, and coends. They are the correct notion of limit for V-categories because they are preserved by V-functors and satisfy the enriched Yoneda lemma.
The second innovation is change of base. Given a monoidal functor F:V→W, one can transport enrichment from V to W: every V-category gives rise to a W-category by applying F to hom-objects. This allows mathematicians to move between different enriching contexts—for example, from simplicial sets to topological spaces—and to compare results across different enrichments. Change of base is a methodological tool that reveals the unity behind seemingly disparate enriched structures.
Enriched category theory provides one of the main models for higher category theory. A simplicially enriched category (a category enriched over simplicial sets) is a strict model for (∞,1)-categories, where the hom-simplicial sets encode higher-dimensional morphisms. Similarly, dg-categories (enriched over chain complexes) model linear (∞,1)-categories. These strict models coexist with weak models such as quasicategories (which are simplicial sets satisfying the horn-filling condition). The relationship is complementary: enrichment gives a strict, algebraic framework that is easier to work with computationally, while quasicategories offer a more flexible, homotopy-coherent approach. The tension between strict and weak models is a central theme in contemporary higher category theory, and enriched category theory remains a vital tool for constructing and analyzing strict higher structures.
Today, enriched category theory continues to expand. Enriched ∞-categories, which combine enrichment with the language of ∞-categories, are an active area of research: one enriches over an ∞-category with a monoidal structure, yielding a framework that unifies classical enrichment with homotopy theory. Enriched monads and enriched algebraic theories extend the reach of categorical logic, allowing the study of theories in enriched contexts (e.g., linear logic via enrichment over the category of vector spaces). There is ongoing debate about the 'correct' model for enriched higher categories: should one use complete Segal spaces, quasicategories, or simplicial categories? Each has advantages, and the field is pluralistic. What the leading approaches agree on is that enrichment over a monoidal (∞,1)-category is the right starting point; they disagree on how to handle coherence and composition in the weak setting. Enriched category theory, born from the simple idea of replacing hom-sets with structured hom-objects, now underpins much of modern categorical mathematics, from homotopy theory to quantum algebra.