Every pair $F \dashv G$ of adjoint functors $F: \mathcal C \to \mathcal D$, $G: \mathcal D \to \mathcal C$ induces a monad $\mathbb T = (T,\eta,\mu)$ on $\mathcal C$. Given a monad $\mathbb T = (T,\eta,\mu)$ on $\mathcal C$, we define $\operatorname{Adj}(\mathbb T)$ to be the category of adjunctions inducing $\mathbb T$. Its objects are adjoint pairs of functors $F \dashv G$ between $\mathcal C$ and some category $\mathcal D$ such that the monad induced by the adjunction is $\mathbb T$, i.e. $GF = T$, the unit is $\eta$, and $G\varepsilon F = \mu$ where $\varepsilon$ is the counit. The morphisms between two such adjunctions $F \dashv G$, $F' \dashv G'$ are functors $H: \mathcal D \to \mathcal D'$ such that $HF=F'$ and $G' H = G$.

I am now worried whether this is really a category. Consider the example where $\mathcal C$ is a one-object category with only the identity morphism and $\mathbb T = (T,\eta,\mu)$ is the trivial monad. For every category $\mathcal D$ which has an initial object $I$, we get an adjunction $F \dashv G$, where $F: \mathcal C \to \mathcal D$ sends the one object of $\mathcal C$ to $I$, and $G: \mathcal D \to \mathcal C$ sends everything to the one object (it's easy to check that this is really an adjunction). Its induced monad on $\mathcal C$ is necessarily $\mathbb T$. So $\operatorname{Adj}(\mathbb T)$ contains at least one object for each category having an initial object. However, $\operatorname{Adj}(\mathbb T)$ itself has an initial object, namely the Kleisli adjunction, which in this case is just $1_{\mathcal C} \dashv 1_{\mathcal C}$. If $\operatorname{Adj}(\mathbb T)$ is a category, it seems that in a way it contains itself, in the form of the adjunction $\mathcal C \rightleftarrows \operatorname{Adj}(\mathbb T)$ defined by the Kleisli adjunction. (How) Is that possible?

up vote 7 down vote accepted

Well, either one works in a framework where self-containing categories are permitted (but I do not know of any), or else one works more carefully. So I would say something like, fix a universe $\mathbf{U}$ such that $\mathcal{C}$ is a $\mathbf{U}$-small category, and let $\mathbf{Adj}(\mathcal{C})$ be the category of $\mathbf{U}$-small categories equipped with an adjunction etc.; then $\mathbf{Adj}(\mathcal{C})$ will not be $\mathbf{U}$-small, and so will not member of itself.

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

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