A topic in the Open Knowledge Graph — a free, open map of 15,290 topics and the order to learn them in.

Fibered Categories

Research Depth 99 in the knowledge graph ☐ I know this ☆ Set as goal
1topic build on this
571prerequisites beneath it
See this on the map →
Comma CategoriesFunctor Categories+2 more→→The Grothendieck Construction
fibered category fibration cartesian morphism Grothendieck fibration descent cleavage

Core Idea

A fibered category (or Grothendieck fibration) over a base category B is a functor p: E → B such that for every morphism f: b → b' in B and object e' in E with p(e') = b', there exists a cartesian morphism (or cartesian lifting) φ: e → e' with p(φ) = f, satisfying a universal property. Intuitively, E is a "family of categories parametrized by B": the fiber E_b = p-1(b) is the category sitting over each object b, and cartesian morphisms provide canonical ways to pull back along morphisms in B. Fibered categories formalize the notion of "varying algebraic/geometric structure over a base" and are central to descent theory, stacks, and the Grothendieck construction.

How It's Best Learned

Consider the codomain fibration cod: Arr(C) → C sending each arrow f: A → B to its codomain B. The fiber over B is the slice category C/B. A cartesian morphism is a pullback square. Verify the universal property of cartesian liftings in this example. Then consider the category of vector bundles over a base space, where the fiber functor sends each bundle to its base space—pullback of bundles provides the cartesian liftings.

Common Misconceptions

Explainer

From your study of comma categories and functor categories, you know how to form slice categories C/b and how to assemble categories of functors into structured wholes. Fibered categories answer a subtler organizational question: how do you talk about a *family* of categories varying over a base, where morphisms in the base tell you how to transport objects between fibers? The motivating example is geometric: over each topological space (or scheme) B, you have a category of vector bundles — small changes in the base space induce pullback operations on bundles, and those pullbacks are the "transport maps" the fibration formalizes.

Formally, a Grothendieck fibration is a functor p: E → B such that for every morphism f: b' → b in B and every object e in E with p(e) = b, there exists a cartesian morphism φ: e' → e in E with p(φ) = f, satisfying a universal lifting property. The universal property says: any morphism ψ: e'' → e in E with p(ψ) factoring through f in B factors uniquely through φ. Concretely, φ is the "best" way to lift f to E above the target e — it is canonical once the target and the base morphism are fixed. The object e' = f*(e) is the pullback of e along f, and this is exactly a pullback in the geometric sense when E is the category of vector bundles.

The fiber E_b over an object b ∈ B is the subcategory of E consisting of objects e with p(e) = b and morphisms φ with p(φ) = id_b. In the vector bundle example, E_b is the category of vector bundles on the single space b. A morphism f: b' → b in B induces a pullback functor f*: E_b → E_{b'}, defined (up to canonical isomorphism) by the cartesian liftings. This is where your comma-category intuition applies: the comma category b'/p in E records all ways to map into a fixed fiber. The codomain fibration cod: Arr(C) → C, sending each arrow g: A → B to its codomain B, has fiber (Arr(C))_b = C/b — precisely the slice category over b. The cartesian morphisms are exactly the pullback squares in C.

A cleavage is a choice, for each (f, e) pair, of a specific cartesian lifting φ_{f,e}. Different cleavages give equivalent total categories, but choosing one converts the fibration into a strict assignment b ↦ E_b and f ↦ f* that almost forms a pseudofunctor Bop → Cat — the composition f* ∘ g* need not equal (g ∘ f)* on the nose, only up to coherent isomorphism. A split fibration is one where the cleavage is strictly functorial: (g ∘ f)* = f* ∘ g* strictly. Most natural fibrations are not split, but every fibration is equivalent to a split one.

The deeper significance is the Grothendieck construction: there is an equivalence (more precisely, a 2-categorical equivalence) between fibrations p: E → B and pseudofunctors F: Bop → Cat. Given a fibration, you extract F by b ↦ E_b and f ↦ f*. Given a pseudofunctor, you build the total category E with objects (b, x) where x ∈ F(b), and morphisms (b', x') → (b, x) given by pairs (f: b' → b, α: x' → F(f)(x)). This is precisely why fibered categories are central to descent theory: a presheaf F: Bop → Set (or Cat) descends along a covering if and only if the corresponding fibration satisfies a gluing condition. Stacks are fibered categories over sites that satisfy the descent conditions, and the entire language of algebraic geometry over varying base schemes is organized through this formalism.

Practice Questions 5 questions

Prerequisite Chain

Understanding Zero → The Number Zero → Counting to Five → Counting to 10 → Counting to 20 → Counting a Set of Objects Up to 20 → Cardinality: The Last Number Counted → Matching Numerals to Quantities → Subitizing Small Quantities → Addition Within 10 → Number Bonds to 10 → Addition Within 20 → Doubles and Near Doubles → Doubles Facts Within 10 → Near Doubles Facts Within 20 → Mental Math Strategies for Addition → Mental Math: Adding and Subtracting Tens → Addition Within 100 → Repeated Addition as Multiplication → Multiplication as Equal Groups → Multiplication: Arrays → Basic Multiplication Facts (0s, 1s, 2s, 5s, 10s) → Multiplication Facts Within 100 → Division as Equal Sharing → Division as Grouping (Measurement Division) → Division: Grouping (Repeated Subtraction) Model → Division: Fair Sharing Model → Division as Equal Sharing → Division as Grouping → Basic Division Facts → Division Facts Within 100 → Multiplication and Division Fact Families → Relationship Between Multiplication and Division → Division Facts as Inverse of Multiplication → Remainders and Quotients in Division → Division Word Problems → Multi-Step Word Problems → Solving Multi-Step Word Problems → Multiplication Word Problems → Division Word Problems → Introduction to Long Division → Factors and Multiples → Prime and Composite Numbers → Equivalent Fractions → Relating Fractions and Decimals → Decimal Place Value → Integers and the Number Line → Comparing and Ordering Integers → Absolute Value → Adding Integers → Subtracting Integers → Multiplying Integers → Introduction to Exponents → Order of Operations → Integer Order of Operations → Variable Expressions → The Distributive Property → Variables and Expressions Review → Introduction to Polynomials → Adding and Subtracting Polynomials → Multiplying Polynomials → Factorial → Permutations → Combinations → Counting Principles: Addition and Multiplication Rules → Introduction to Graph Theory → Propositional Logic Foundations → Logical Equivalences → Set Operations: Union, Intersection, and Complement → Cartesian Products and Relations → Partial Orders → Binary Relations → Equivalence Relations → Injective, Surjective, and Bijective Functions → Categories and Morphisms → Functors → Commutative Diagrams in Category Theory → Commutative Diagrams and Composition → Natural Transformations → 2-Categories and Weak Functors → Natural Isomorphisms Between Functors → Isomorphisms in Categories → Universal Properties → Initial and Terminal Objects → Products and Coproducts → Equalizers and Coequalizers → Limits and Colimits → Pullbacks and Pushouts → Adjoint Functors → Free Objects → Projective Objects and Projective Covers → Homological Dimension in Categories → Exact Sequences in Categories → Ext Functors as Derived Hom → Tor Functors as Derived Tensor Product → Derived Functors → Derived Categories and Derived Equivalences → Localization of Categories → 2-Categories → Fibered Categories

Longest path: 100 steps · 571 total prerequisite topics

Prerequisites (4)

Leads To (1)