Home

Colimits

Colimits are a central concept in category theory that generalizes how a diagram of objects and morphisms can be glued together into a single object. Given a small category J and a functor F: J → C, a colimit of F consists of an object colim F in C and a cocone consisting of morphisms ι_j: F(j) → colim F for each object j in J, such that the following universal property holds: for any object N in C and any cocone η_j: F(j) → N, there exists a unique morphism u: colim F → N with η_j = u ∘ ι_j for all j in J. Equivalently, the colimit represents the functor sending N to the set of cocones from F to ΔN, where ΔN denotes the constant diagram at N.

Colimits generalize several familiar constructions. If J is a discrete category with as many objects as a

Existence and preservation: Not every category has all small colimits, but many common categories do (for example

In practice, colimits provide a unifying framework for constructions such as quotients, unions, and gluing along

family
{A_i},
the
colimit
is
the
coproduct
(a
disjoint
union
in
Set).
If
J
has
two
objects
and
two
parallel
arrows,
the
colimit
is
a
coequalizer.
A
pushout
arises
when
J
forms
a
span
A
→
B
←
C.
More
generally,
colimits
can
be
taken
over
any
shape
J
and,
in
many
categories,
they
can
be
built
by
composing
simpler
colimits.
Set,
Ab,
Cat,
Top).
Colimits
are
stable
under
change
of
diagram
shape
in
the
sense
that
they
intertwine
with
functors.
A
fundamental
fact
is
that
left
adjoint
functors
preserve
colimits,
while
right
adjoints
preserve
limits.
common
subobjects
across
various
mathematical
contexts.