2-natural transformation?
Two morphisms $C \stackrel{L}{\to} D$ and $D \stackrel{R}{\to} C$ in a 2-category $\mathcal{C}$ form an adjunction if they are dual to each other.
There are two archetypical examples:
If $A$ is a monoidal category and $\mathcal{C} = \mathbf{B}A$ is the one-object 2-category incarnation of $A$ (the delooping of $A$), so that the morphisms in $\mathcal{C}$ correspond to the objects of $A$, then the notion of adjoint morphisms in $\mathcal{C}$ coincides precisely with the notion of dual objects in a $A$.
If $\mathcal{C}$ is the $2$-category Cat, so that the morphisms in $\mathcal{C}$ are functors, then the notion of adjoint morphisms in $\mathcal{C}$ coincides precisely with the notion of adjoint functors.
The notion of adjunction may usefully be thought of as a weakened version of the notion of equivalence in a 2-category: a morphism in an adjunction need not be invertible, but it has in some sense a left inverse from below and a right inverse from above. If the morphism in an adjunction does happen to be a genuine equivalence, then we speak of the adjunction being an adjoint equivalence.
Essentially everything that makes category theory nontrivial and interesting beyond groupoid theory can be derived from the concept of adjoint functors. In particular universal constructions such as limits and colimits are examples of certain adjunctions. Adjunctions are already interesting (but simpler) in 2-posets, such as the $2$-poset Pos of posets.
At the cost of some repetition (compare adjoint functor), we outline how one gets from the hom-functor formulation of adjunction in Cat to the elementary definition in terms of units and counits. This will motivate the definition in the section that follows, which is elementary (definable in the first-order theory of categories) and portable to any 2-category.
We start from a familiar example. Let $U: Grp \to Set$ from groups to sets be the usual forgetful functor. When we say β$F(X)$ is the free group generated by a set $X$β, we mean there is a function $\eta_X: X \to U(F(X))$ which is universal among functions from $X$ to the underlying set of a group, which means in turn that given a function $f: X \to U(G)$, there is a unique group homomorphism $g: F(X) \to G$ such that
Here $\eta_X$ is a component of what we call the unit of the adjunction $F \dashv U$, and the equation above is a recipe for the relationship between the map $g: F(X) \to G$ and the map $f: X \to U(G)$ in terms of the unit.
Now we work more generally. Suppose given functors $L: C \to D$, $R: D \to C$ and the structure of an adjunction in the form of a natural isomorphism
Now the idea is that, a la the Yoneda lemma, $\Psi$ should be completely describable in terms of what it does to identity maps. With that in mind, define the unit $\eta : 1_C \to R L$ by the formula $\eta_c = \Psi_{c, L(c)}(1_{L(c)})$. Dually, define the counit $\varepsilon : L R \to 1_D$ by the formula $\varepsilon_d = \Psi^{-1}_{R(d), d}(1_{R(d)})$. Then given $g: L(c) \to d$, the claim is that
This may be left as an exercise in the yoga of the Yoneda lemma, applied to $\hom_D(L(c), -) \to \hom_C(c, R(-))$. By duality, given $f: c \to R(d)$,
(In fact, we spell out the Yoneda-lemma proof of this dual form below.)
Finally, these operations should obviously be mutually inverse, but that can again be entirely encapsulated Yoneda-wise in terms of the effect on identity maps. Thus, if $\eta_c \coloneqq \Psi_{c, L(c)}(1_{L(c)})$, the recipe just given for $\Psi^{-1}$ yields back
and this is one of the famous triangular equations: $1_L = (L \stackrel{L \eta}{\to} L R L \stackrel{\varepsilon L}{\to} L)$. Note that juxtaposition in the diagram above is neither functor application, nor vertical composition, nor horizontal composition, but is actually whiskering. By duality, we have the other triangular equation $1_R = (R \stackrel{\eta R}{\to} R L R \stackrel{R \varepsilon}{\to} R)$. These two triangular equations are enough to guarantee that the recipes for $\Psi$ and $\Psi^{-1}$ are indeed mutually inverse.
Thus, it is perfectly sufficient to define an adjoint pair of functors in $Cat$ as given by unit and counit transformations $\eta: 1_C \to R L$, $\varepsilon: L R \to 1_D$, satisfying triangular equations as above.
One thing often heard is that the definition of adjunctions via units and counits is an βelementaryβ definition (so that by implication, the formulation in terms of hom-functors is not elementary). This means that whereas the hom-functor formulation relies on a background category of sets, the formulation in terms of units and counits is purely in the first-order language of categories and makes no reference to a background model of set theory. It is therefore a perfectly serviceable definition of adjunction without assumptions of local smallness.
We claim that $\Psi^{-1}_{c, d}: \hom_C(c, R(d)) \to \hom_D(L(c), d)$ can be defined by the formula
where $\varepsilon_d \coloneqq \Psi^{-1}_{R(d), d}(1_{R(d)})$. This is by appeal to the proof of the Yoneda lemma applied to the transformation
For the naturality of $\Psi^{-1}$ in the argument $(-)$ would imply that given $f: c \to R(d)$, we have a commutative square
Chasing the element $1_{R(d)}$ down and then across, we get $f: c \to R(d)$ and then $\Psi^{-1}_{c, d}(f)$. Chasing across and then down, we get $\varepsilon_d$ and then $\varepsilon_d \circ L(f)$. This completes the verification of the claim.
An adjunction in a 2-category is a pair of objects $C,D$ together with morphisms $L: C \to D$, $R : D \to C$ and 2-morphisms $\eta: 1_C \to R \circ L$, $\epsilon: L \circ R \to 1_D$ satisfying the equations
and
variously called the triangle identities or the zig-zag identities. We call $L$ the left adjoint (of $R$) and $R$ the right adjoint (of $L$). We call $\eta$ the unit of the adjunction and $\epsilon$ the counit of the adjunction.
When interpreted in the prototypical 2-category Cat, $C$ and $D$ are categories, $L$ and $R$ are functors, and $\eta$ and $\epsilon$ are natural transformations. In this case (which was of course the first to be defined) there are a number of equivalent definitions of an adjunction, which can be found on the page adjoint functor. Conversely, the definition in any 2-category can be obtained by internalization from the definition in $\Cat$.
The definition of an adjunction may be nicely expressed using string diagrams. The data $L: C \to D$, $R : D \to C$ and 2-cells $\eta: 1_C \to R \circ L$, $\epsilon: L \circ R \to 1_D$ are depicted as
Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β Β
(where 1-cells read from right to left and 2-cells from bottom to top), and the zigzag identities are expressed as βpulling zigzags straightβ (hence the name):
Β Β Β Β Β Β Β Β
Often, arrows on strings are used to distinguish $L$ and $R$, and most or all other labels are left implicit; so the zigzag identities, for instance, become:
Β Β Β Β Β Β Β Β
See at monad β Relation to adjunctions.
An adjunction in Cat is a pair of adjoint functors.
For $A$ a monoidal category and $\mathbf{B}A$ its incarnation as a one-object 2-category (its delooping), an adjunction in $\mathbf{B}A$ is a pair of dual objects.
For $\mathcal{C}$ a 2-groupoid, an adjunction in $\mathcal{C}$ is an adjoint equivalence.
Catsters, Adjunctions (YouTube)
Wikipedia, Adjoint Functors