ncatlab.org

coequalizer in nLab

Contents

Context

Limits and colimits

limits and colimits

1-Categorical

2-Categorical

(∞,1)-Categorical

Model-categorical

Contents

Idea

The concept of coequalizer in a general category is the generalization of the construction where for two functions f,gf,g between sets XX and YY

X⟶g⟶fYX \stackrel{\overset{f}{\longrightarrow}}{\underset{g}{\longrightarrow}} Y

one forms the set Y/ ∼Y/_\sim of equivalence classes induced by the equivalence relation generated by the relation

f(x)∼g(x)f(x)\sim g(x)

for all x∈Xx \in X. This means that the quotient function p:Y⟶Y/ ∼p \colon Y \longrightarrow Y/_\sim satisfies

p∘f=p∘gp \circ f = p \circ g

(a map pp satisfying this equation is said to “co-equalize” ff and gg) and moreover pp is universal with this property.

In this form this may be phrased generally in any category.

Definition

Definition

In some category 𝒞\mathcal{C}, the coequalizer coeq(f,g)coeq(f,g) of two parallel morphisms ff and gg between two objects XX and YY is (if it exists), the colimit under the diagram formed by these two morphisms

X ⟶g⟶f Y ↘ ↙ p coeq(f,g).\array{ X && \stackrel{\overset{f}{\longrightarrow}}{\underset{g}{\longrightarrow}} && Y \\ & \searrow && \swarrow_{\mathrlap{p}} \\ && coeq(f,g) }.

Equivalently:

Definition

In a category 𝒞\mathcal{C} a diagram

X⇉gfY→pZX \underoverset{g}{f}{\rightrightarrows} Y \overset{p}{\rightarrow} Z

is called a coequalizer diagram if

  1. p∘f=p∘gp \circ f = p \circ g;

  2. pp is universal for this property: i.e. if q:Y→Wq \colon Y \to W is a morphism of 𝒞\mathcal{C} such that q∘f=q∘gq \circ f = q \circ g, then there is a unique morphism q′:Z→Wq' \colon Z \to W such that q′∘p=qq' \circ p = q

    X ⟶g⟶f Y ⟶p Z q↓ ↙ ∃!q′ W \array{ X &\stackrel{\overset{f}{\longrightarrow}}{\underset{g}{\longrightarrow}}& Y &\overset{p}{\longrightarrow}& Z \\ && {}^{\mathllap{q}}\downarrow & \swarrow_{\mathrlap{\exists ! \, q'}} \\ && W }

Properties

Relation to kernel pairs

Proposition

In any category:

(e.g. Borceux 1994, Prop. 2.5.7, 2.5.8, Taylor 1999, Lemma 5.6.6

Relation to pushouts

Coequalizers are closely related to pushouts:

Proposition

A diagram

X⇉fgY⟶pZ X \underoverset {f} {g} {\rightrightarrows} Y \overset{p}{\longrightarrow} Z

is a coequalizer diagram, def. , precisely if

X⊔X ⟶(f,g) Y ↓ ↓p X ⟶ Z\array{ X \sqcup X &\overset{(f,g)}{\longrightarrow}& Y \\ \big\downarrow && \big\downarrow{^\mathrlap{p}} \\ X &\underset{}{\longrightarrow}& Z }

is a pushout diagram.

Conversely:

Proposition

A ⟶f 1 B f 2↓ ↓ p 1 C ⟶p 2 D \array{ A &\overset{f_1}{\longrightarrow}& B \\ {\mathllap{{}^{f_2}}}\Big\downarrow && \Big\downarrow{{}^\mathrlap{p_1}} \\ C &\underset{p_2}{\longrightarrow}& D }

is a pushout square, precisely if

A⇉q 2∘f 2q 1∘f 1B⊔C⟶(p 1,p 2)D A \underoverset { q_2 \circ f_2 } { q_1 \circ f_1 } {\rightrightarrows} B \sqcup C \overset{(p_1,p_2)}{\longrightarrow} D

is a coequalizer diagram.

(Here B→q 1B⊔C←q 2CB \overset{q_1}{\to} B \sqcup C \overset{q_2}{\leftarrow} C denotes the two coprojections into the coproduct.)

Examples

References

Coequalizers were defined in the paper

for any finite collection of parallel morphisms. The paper refers to them as right equalizers, whereas equalizers are referred to as left equalizers.

Textbook accounts:

Last revised on May 1, 2023 at 08:39:10. See the history of this page for a list of all contributions to it.