Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

Let $\mathcal{C}$ be a $2\!-$category and let $(F,G,\eta,\varepsilon)$, $(F',G',\eta',\varepsilon')$ be two paralell adjunctions $A-\!\!\!\rightharpoonup B$ in $\mathcal{C}$. Let $F\xrightarrow{\ \sigma\ }F'$, $G'\xrightarrow{\ \tau\ }G$ be $2\!-$morphisms. If $\mathcal{C}=\mathbf{Cat}$, so that we deal with ordinary adjoint functors, it is proven in Categories for the Working Mathematician, that any of the four equalities

$$\tau=(1_G\bullet\varepsilon')\circ(1_G\bullet\sigma\bullet 1_{G'})\circ(\eta\bullet1_{G'})\,,$$

$$\sigma=(\varepsilon\bullet 1_{F'})\circ(1_F\bullet\sigma\bullet 1_{F'})\circ(1_{F}\bullet\eta')\,,$$

$$\varepsilon\circ(1_F\bullet\tau)=\varepsilon'\circ(\sigma\bullet 1_{G'})\,,$$

$$(1_G\bullet\sigma)\circ\eta=(\tau\bullet 1_{F'})\circ\eta'\,,$$

are equivalent.

Does this result hold in the more general setting of arbitrary $2\!-$categories and adjunctions? If so, how can I prove it?

Unfortunately, Mac Lane uses the definition of adjunctions via isomorphisms of $\operatorname{Hom}$-functors, so that the proof does not work in this more general context.

share|improve this question
Use the 2-categorical Yoneda embedding to reduce to the case of $\mathfrak{Cat}$. –  Zhen Lin Jul 9 at 18:16

1 Answer 1

up vote 1 down vote accepted

Actually, the definition of adjunctions via natural isomorphisms of Hom-functors works in any $2$-category. Specifically, $f\leftrightarrows g$ are adjoint with unit $1\stackrel\eta\Rightarrow fg$ and counit $gf\stackrel\epsilon\Rightarrow 1$ if and only if we have a family of isomorphisms $[fX,Y]\cong[X,gY]$ that is natural in the "$E$-objects" $E\stackrel X\rightarrow C$ of $C$ and $E\stackrel Y\rightarrow D$ of $D$, for every object $E$ of the $2$-category.

One may wonder what kind of Functors $[fX,Y]$ and $[X,gY]$ are, and also between what Categories are they Functors, but one should instead consider them to be just "syntactic sugar" to sweeten one's manipulations of Kan lifts and Kan extensions.

share|improve this answer

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.