3
$\begingroup$

In case this question appeared here I apologize for the preposterous usage of computer space and attention of participants. I searched over there, for a similar or the same question for an answer but resigned. :)

I am reading:

S. B. Niefield, Cartesianness, topological spaces, uniform spaces, and affine schemes, J. Pure Appl. Alg. 23, 1982, 147–167, where the author extends the Day-Kelley characterization of exponentiable objects in $Top$ to slice categories $Top/X$.

Let $A$ be a category with finite limits and $T$ is an object of $A$. It is clear that a product $X\times T$ becomes an object over $T$ via the projection $\pi_2 : X\times T\rightarrow T$. This induces a functor $T^*:A\rightarrow A/T$ which is clearly right adjoint of the forgetful functor $\Sigma_T :A/T\rightarrow A$.

Now the Proposition 1.1 states: A functor $F:B\to A/T$ has a right adjoint if and only if $\Sigma_T \circ F$ has a right adjoint.

To prove that if $\Sigma_T \circ F$ has a right adjoint then $F$ has a right adjoint the author argues: Suppose $\Sigma_T \circ F\dashv G'$, For $X$ object of $B$, let $\sigma_X:X\to G'T$ be the right adjunct of $FX$ considered as morphism $\Sigma_T(FX)\to T$. Then, if $f:X\to X'$ is a morphism of $B$, the diagram $\require{AMScd}$ \begin{CD} X @>f>> X' \\ @V \sigma_X VV= @VV \sigma_{X'} V\\ G'T @>>id_{G'T}> G'T \end{CD} commutes.

I stuck on this: WHY THE DIAGRAM COMMUTES?

Then using naturality (since the commutation of the above diagram means that $\sigma $ is a natural transformation) the author just builds the right adjoint $G$ for $F$ as equalizer of

$G'p_Y :G'Y\to G'T$

and

$\sigma _{G'Y}:G'Y\to G'T$

where $p_Y: Y\to T$ is an object of $A/T$.

My Question is why the diagram commutes. Thanks in advance.

$\endgroup$
5
$\begingroup$

If $f:X \to X'$ is a morphism, then we have an induced morphism $F(f) : F(X) \to F(X')$. By definition of $A/T$, this is a commutative diagram

$\require{AMScd}$ \begin{CD} \Sigma_T(F(X)) @>\Sigma_T(F(f))>> \Sigma_T(F(X')) \\ @V VV @VV V\\ T @>\mathrm{id}>> T \end{CD}

This gives the desired diagram

\begin{CD} X @>f>> X' \\ @V \sigma_X VV @VV \sigma_{X'} V\\ G'T @>>\mathrm{id}> G'T \end{CD}

because the bijection in the adjunction is natural.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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