In general, limits and colimits do not commute. It is therefore of interest to list the special conditions under which certain limits do commute with certain colimits. (See also at permuting limits and colimits.)
Let and be (usually small) categories, and a category that has both -colimits and -limits. Then for any functor , there is a canonical morphism
We say that -colimits commute with -limits in if this is an isomorphism for all such . This is equivalent to both of the statements:
If -colimits commute with -limits in , then the same is true in any functor category , since limits and colimits in the latter are both pointwise in .
Also, if -colimits commute with -limits in , and if is a reflective subcategory of with a reflector that preserves -limits, then -colimits also commute with -limits in . This follows because the functor factors as the composite in which all three functors preserve -limits.
In Set, filtered colimits commute with finite limits (e.g. MacLane (1971), §IX.2 Thm. 1 (p. 211)).
In fact, is a filtered category if and only if -colimits commute with finite limits in . More generally, filtered colimits commute with L-finite limits.
By the above remarks, it follows that filtered colimits commute with finite limits in any Grothendieck topos.
Again in Set (and hence also in any Grothendieck topos), sifted colimits commute with finite products. In fact, this is usually taken to be the definition of a sifted category, and then a theorem of Gabriel-Ulmer 71 characterizes sifted categories as those for which the diagonal functor is a final functor.
As a special case, categories with finite products are cosifted.
For more on this see at distributivity of products and colimits.
This means that if is a finite group, is a small cofiltered category and is a functor, the canonical map
is an isomorphism. This fact is mentioned by André Joyal in Foncteurs analytiques et espèces de structures; a proof can be found here.
Let be a set, a connected category, and a functor. Then the canonical morphism
is an isomorphism. This remains true if Set is replaced by any Grothendieck topos.
More generally, if is an (∞,1)-topos, is an n-groupoid, and is a small (∞,1)-category whose classifying space is n-connected, then -limits commute with -colimits in . This follows from the fact that the colimit functor induces an equivalence of (∞,1)-topoi . For example, if is a cofiltered (∞,1)-category or even a cosifted (∞,1)-category, then the classifying space of is weakly contractible and hence -limits commute with -colimits in for any ∞-groupoid .
In general, for any class of limits , one may consider the class of all colimits that commute with -limits and dually. These classes of limits and colimits share many of the properties of the above examples, especially when is a sound doctrine.
Stability of a colimit under pullback looks informally like a “commutativity” condition between colimits and pullbacks, but it is not actually in general an instance of the general notion of commutativity of limits and colimits, though it is an instance of distributivity of limits over colimits. See also pullback-stable colimit for more.
Peter Hilton, Commuting limits, Cahiers de Topologie et Géométrie Différentielle Catégoriques 10.1 (1968): 127-138.
Beno Eckmann, Peter John Hilton, Commuting limits with colimits, Journal of Algebra 11 1 (1969) 116-144.
Armin Frei, and John L. MacDonald, Limits in categories of relations and limit-colimit commutation, Journal of Pure and Applied Algebra 1 2 (1971) 179-197
Saunders MacLane, §IX.2 in: Categories for the Working Mathematician, Graduate texts in mathematics, Springer (1971) [doi:10.1007/978-1-4757-4721-8]
François Foltz, Sur la commutation des limites, Diagrammes 5 (1981) F1-F33
Marie Bjerrum, Peter Johnstone, Tom Leinster, William F. Sawin, Notes on commutation of limits and colimits, Theory and Applications of Categories 30 (2015), 527-532 [arXiv:1409.7860, tac:3015]
Last revised on May 1, 2023 at 06:27:48. See the history of this page for a list of all contributions to it.