Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Suppose I have a comonad $G$ on a category $\mathbb{C}$.

If $C$ is a preorder then I can define a comonad (i.e. interior operator) $\square$ on the set $\mathcal{P}\uparrow(C)$ of upwards closed subsets of $C$ by taking preimages. This preserves all limits and all colimits and has a right adjoint monad (i.e. a closure operator).

I would like to know if (and how) this can be generalized to the presheaf category $\text{Set}^{C^{\text{op}}}$, since upwards closed sets are generalized by presheafs.

Given a functor $F : C \to C$ one can define a geometric morphism $F : \widehat{C} \to \widehat{C}$ (where $\widehat{C} = \text{Set}^{C^\text{op}}$) where the inverse image functor $F^*$ is given by precomposition by $F^{\text{op}} : C^{\text{op}} \to C^{\text{op}}$ and the direct image functor $F_*$ is given on objects as $F_*(Q)(C) = \text{Hom}\left(\text{Hom}(F(-), C), Q\right)$. Now obviously one gets a comonad on $\widehat{C}$ since $F^* \dashv F_*$, but this does not use the structure of $F$ in any way and doesn't seem to me to be the generalization of the construction on preorders.

My question is, given a comonad $G$ on $C$ is $G^*$ in general a comonad on $\widehat{C}$ and is perhaps $G_*$ a monad right adjoint to it?

share|cite|improve this question
up vote 3 down vote accepted

The operation $\mathbb{C} \mapsto [\mathbb{C}^\mathrm{op}, \mathbf{Set}]$ is a 2-functor $\mathfrak{Cat}^\mathrm{co op} \to \mathfrak{CAT}$, so it must in particular send comonads to monads. For instance, given $\epsilon : G \Rightarrow \mathrm{id}_{\mathbb{C}}$, we get an induced natural transformation $\epsilon^* : \mathrm{id}_{[\mathbb{C}^\mathrm{op}, \mathbf{Set}]} \Rightarrow G^*$ etc.

That said, there is also a 2-functor $\mathfrak{Cat} \to \mathfrak{CAT}$, which sends a functor $G$ to its left Kan extension $G_!$. This doesn't reverse natural transformations, so it sends comonads (resp. monads) on $\mathbb{C}$ to comonads (resp. monads) on $[\mathbb{C}^\mathrm{op}, \mathbf{Set}]$. We can say the same thing for the 2-functor $\mathfrak{Cat} \to \mathfrak{CAT}$ that sends a functor $G$ to its right Kan extension $G_*$.

share|cite|improve this answer
This is enough. Upwards closed sets are in fact a subset of covariant functors on a preordered category. So given a comonad on $\mathbb{C}$ we get a monad on $\mathbb{C}^{op}$ which is then mapped to the comonad on $[\mathbb{C}, \mathbf{Set}]$. Thank you. – Aleš Bizjak Oct 22 '13 at 17:43

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.