nLab absolute colimit




An absolute colimit is a colimit which is preserved by any functor whatsoever. In general this happens because the colimit is a colimit for purely “diagrammatic” reasons. The notion is most important in enriched category theory.

Of course, there is a dual notion of absolute limit, but it is used less frequently.


The term “absolute colimit” is actually used for two closely related, but distinct, notions.

Particular absolute colimits


A particular colimit diagram in a particular category CC is an absolute colimit if it is preserved by every functor with domain CC.

This definition makes sense also in enriched category theory: for any VV, a weighted colimit in a particular VV-enriched category CC is an absolute colimit if it is preserved by every VV-functor with domain CC.

Note, however, that a conical colimit in a VV-category CC may be preserved by all VV-functors without being preserved by all unenriched functors on the underlying ordinary category C oC_o.

Generalizing in a different direction, absolute colimits in SetSet-enriched categories can be regarded as the particular case of postulated colimit?s in sites where the site has the trivial topology.

Weights for absolute colimits


For a given VV, a weight Φ:D opV\Phi\colon D^{op} \to V for colimits is an absolute weight, or a weight for absolute colimits, if Φ\Phi-weighted colimits in all VV-categories are preserved by all VV-functors.

Absolute colimits of this sort are also called Cauchy colimits. A VV-category which admits all absolute colimits — that is, all VV-weighted colimits whose weights are absolute – is called Cauchy complete. By the characterization below, it is equivalent to admit limits weighted by all weights for absolute limits.


Both types of absolute colimits admit pleasant characterizations.

Particular absolute colimits

For a particular cocone μ:FΔA\mu \colon F \to \Delta A under a functor F:ICF\colon I\to C (all in the Set-enriched world), the following are equivalent:

  • μ\mu is an absolute colimiting cocone.

  • μ\mu is a colimiting cocone which is is preserved by the Yoneda embedding C[C op,Set]C \hookrightarrow [C^{op},Set].

  • μ\mu is a colimiting cocone which is preserved by the representable functors C(F(i),):CSetC(F(i),-)\colon C\to Set (for all iIi\in I) and C(A,):CSetC(A,-)\colon C\to Set.

  • There exists i 0Ii_0\in I and d 0:AF(i 0)d_0\colon A\to F(i_0) such that

    1. For every iIi\in I, d 0μ id_0 \circ \mu_i and 1 F(i)1_{F(i)} are in the same connected component of the comma category (F(i)/F)(F(i) / F).
    2. μ i 0d 0=1 A\mu_{i_0} \circ d_0 = 1_{A}.

The equivalence of the first two is basically because the Yoneda embedding is the free cocompletion of CC. The third clearly follows from the second. The fourth follows from the third by inspecting exactly what preservation by those representables means in terms of colimits in Set (as is explained in more detail in the special case of absolute coequalizers). Finally, it is straightforward to check that the fourth implies that μ\mu is colimiting, and it is clearly a property preserved by any functor.

Note that in particular, the fourth condition implies that AA is a retract of F(i 0)F(i_0). Also, the first half of the fourth condition by itself characterizes absolute weak colimits.

It is also possible to prove directly that the third condition implies the first two, without extracting the fourth condition along the way. Namely, Let BB be the full subcategory of CC consisting of the objects F(i)F(i) and AA. Then FF defines a functor IBI\to B; call it FF'. Note that AA is also the colimit of FF' in BB. Moreover, by the equivalence of the first two conditions, AA is an absolute colimit of FF', since by hypothesis it is preserved by all representable functors out of BB. Therefore, this colimit is in particular preserved by the inclusion BCB\hookrightarrow C, along with its composite with any functor out of CC; so AA is an absolute colimit of FF.

Weights for absolute colimits

Let VV be a Bénabou cosmos and J:KAJ\colon K ⇸ A a VV-profunctor. Then the following are equivalent:

  • JJ is a weight for absolute colimits (i.e. JJ-weighted colimits in any VV-category are preserved by all VV-functors)

  • JJ has a right adjoint J *J^* in the bicategory VV-Prof.

  • There is a weight J *:AKJ^*\colon A ⇸ K such that JJ-weighted colimits coincide naturally with J *J^*-weighted limits.


Particular absolute colimits

Of course, every colimit weighted by a weight for absolute colimits is itself a particular absolute colimit. But it may also happen that a particular colimit may be absolute without all colimits of that shape being absolute. For example (in ordinary category theory, with V=SetV=Set):

We can also say something about non-examples.

  • Initial objects (in SetSet-enriched categories) are never absolute. If 00 is an initial object, then it is never preserved by the representable functor C(0,):CSetC(0,-)\colon C \to Set.

  • Similarly, coproducts in SetSet-enriched categories are never absolute.

Weights for absolute colimits

In ordinary SetSet-enriched category theory there are not very many weights for absolute colimits, but we have

In fact, this example is “universal,” in that an ordinary category is Cauchy complete iff it has split idempotents, although not every absolute colimit “is” the splitting of an idempotent. More precisely, the class of absolute SetSet-limits is the saturation of idempotent-splittings.

In enriched category theory there can be more types of absolute colimits. For instance:

New kinds of absolute (co)limits also arise in higher category theory.

  • for (∞,1)-categories (enriched over \infty-groupoids), splitting of idempotents is a universal absolute colimit.

  • in stable (∞,1)-categories (which are enriched, in the (,1)(\infty,1)-categorical sense, over the (,1)(\infty,1)-category of spectra), initial objects and pushouts are absolute, and therefore so are all finite colimits.


If K:ACK : A \to C is a functor, then a colimit in AA is KK-absolute if it is preserved by the nerve N K:C[A op,𝒱]N_K : C \to [A^{\mathrm{op}}, \mathscr{V}] of KK. This generalises absoluteness, taking KK to be the identity functor.


Last revised on June 7, 2023 at 17:37:04. See the history of this page for a list of all contributions to it.