nLab weak adjoint

Weak adjoints

Weak adjoints


A weak adjoint is like an adjoint functor but without the uniqueness of factorizations.

Importantly, this means that unlike adjoint functors, weak adjoints are not unique, and so a weak adjoint of a given functor is a structure whereas a functor having an adjoint is a property.


Let G:DCG:D\to C be a functor. We say GG has a left weak adjoint if for each xCx\in C there exists a morphism η x:xGz\eta_x:x\to G z such that for any morphism f:xGyf:x\to G y there exists a (not necessarily unique) morphism g:yzg:y\to z such that f=Ggη xf = G g \circ\eta_x.


  • A weak limit is a weak right adjoint to a constant-diagram functor.


Of course, if the factorizations gg are always unique, this is precisely an ordinary left adjoint. A weak adjoint can also be regarded as a stronger form of the solution set condition in which the solution sets are required to be singletons. The “dual” property, in which the solution sets need not be singletons but the factorizations are unique, is called a multi-adjoint.


Weak adjoint functors along with weak colimits were defined in:

Last revised on November 1, 2022 at 12:48:52. See the history of this page for a list of all contributions to it.