MathOverflow is a question and answer site for professional mathematicians. 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

Hi, I apologise if this is the wrong place for this question but i need to ask it somewhere. The question is whether the right, (perhaps left?), adjoint to $Hom(-,A)$ exists in $\bf Set$ and how to deduse it.

I would very much like it to be the disjoint union $-\oplus A $ but im not quite sure how to deduce it or if its simply whishfull thinking.

Alternativly, is there an other adjoint (left or right) to the disjoint union.

I am writing an undergrad paper and I am aware of the coproduct. I'm asking since i hope the deduction of a dual adjoint will tell me a little of the general workings of category theory and help me understand how to make "computations" in concrete categories.

Also, if you have the time and will, is there a right adjoint to the covariant Hom functor?

Thanks in advance or sorry for wasting your time

share|cite|improve this question
up vote 5 down vote accepted

To speak of right (or left) adjoints for a contravariant functor $F: C\to D$, one needs to decide whether to view it as a functor from $C^{op}$ to $D$ or as a functor from $C$ to $D^{op}$. What the one viewpoint calls a left adjoint, the other will call a right adjoint. One therefore often speaks instead of two contravariant functors being "adjoint on the right" (or on the left). In this language, $Hom(-,A)$ is adjoint on the right to itself, which means that morphisms from $X$ to $Hom(Y,A)$ are in natural bijective correspondence with morphisms from $Y$ to $Hom(X,A)$; here "natural" means (as usual for adjointness) with respect to both $X$ and $Y$.

share|cite|improve this answer
Thx, I appreaciate all answers are good. what really had me confused was the contravariant nature and my own hopes. – Tobias Feb 19 '12 at 23:08

Left adjoints preserve colimits, but $\mathrm{Hom}(-,A)$ does not because $\mathrm{Hom}(\varnothing,A)\neq\varnothing$.

Right adjoints preserve limits, but if $A$ is not terminal, $\mathrm{Hom}(-,A)$ does not because $\mathrm{Hom}(\{*\},A)\neq\{*\}$. (if $A$ is terminal, the functor is constant equal to $\{*\}$ and has a left adjoint which is the constant functor equal to $\varnothing$).

You can do the same reasoning to deduce that $-\oplus{}A$ does not have adjoints, unless $A=\varnothing$.

share|cite|improve this answer
Nice explanation. – Samuel Vidal Feb 19 '12 at 20:22
You forgot the contravariance. Whichever hand it is, we do have $\hom(\emptyset,A) = \lbrace\ast\rbrace$, so we're preserving that limit, at least. – Theo Johnson-Freyd Feb 20 '12 at 2:44

Covariant $Hom$ takes $x$-element sets to $x^n$-element sets. If $f$ is a right adjoint, we have $y^{x^n}=f(y)^x$, and there is no function with this property.

But product is left adjoint to covariant $Hom$, as follows:

$Hom(X,Hom(A,Y))=Hom(X \times A,Y)$

share|cite|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.