1
$\begingroup$

Let $R$ be a commutative ring with unit and let $M$ be a $R$-module.

It holds that $\text{Hom}_R(R^n,M)\cong \prod_{i=1}^n\text{Hom}_R(R,M)$, right?

How could we prove this?

Do we have to define a function and show that it is an homomorphism and bijective?

Do we maybe define the following function?

$$\phi : \text{Hom}_R(R^n,M)\rightarrow \prod_{i=1}^n\text{Hom}_R(R,M)=\text{Hom}_R(R,M)\times \text{Hom}_R(R,M)\times \dots \times \text{Hom}_R(R,M)$$

with $f\mapsto (f_1, f_2, \dots , f_n)$, where $f, f_i\in \text{Hom}_R(R,M)$, for $1\leq i \leq n$

If this mapping is correct:

To show that $\phi$ is an homomorphism, we do the following?

$$\phi (f+g)=(f_1+g_1, f_2+g_2, \dots , f_n+g_n) \\ =(f_1, f_2, \dots , f_n)+(g_1, g_2, \dots , g_n)=\phi (f)+\phi (g) \\ \phi (af)=(af_1, af_2, \dots , af_n)=a(f_1, f_2, \dots , f_n)=a\phi (f)$$

Is this correct?

How can we show that the mapping is bijective?

$\endgroup$
  • 1
    $\begingroup$ The question is a special case of the isomorphism $\text{Hom}_R(\bigoplus_{i\in I} M_i, N) \cong \prod_{i\in I} \text{Hom}_R(M_i, N)$ where $M_i, N$ are $R$-modules. $\endgroup$ – user348338 Jun 23 '16 at 17:04
  • $\begingroup$ It's not clear to me how you're defining the homomorphism, but yes, that's one way to prove it. $\endgroup$ – Callus - Reinstate Monica Jun 23 '16 at 17:08
3
$\begingroup$

I will prove that given a family $\{M_j\}_{j \in J}$ of $R$-modules then $Hom(\oplus M_{j},N) \cong \prod Hom(M_{j},N)$ for any given $R$-module. Then your question follows by setting $M_{j}=R$ for all $\thinspace$ $j \in J$.

Assume that $i_{j} : M_{j} \rightarrow \oplus M_{j}$ is the canonical inclusion. Now define the above homomorphism $\phi: Hom(\oplus M_{j},N) \rightarrow \prod Hom(M_{j},N)$ to be given by $\phi(f)= (f \circ i_{j})_{j \in J}$. In fact what we achieve by doing this, is that since $f \in Hom(\oplus M_{j},N) $ the composition $f \circ i_{j}$ is always a map in $Hom(M_{j},N)$, $\forall j \in J$, so the image of $f$ under $\phi$ is in fact always in $\prod Hom(M_{j},N)$. So, $\phi$ defined as above is the isomorphism you are looking for (prove the latter, isn't difficult now).

$\endgroup$
  • $\begingroup$ Shouldn't the codomain be the direct product $\prod_{j\in J}\text{Hom}_R(M_j, N)$ ? $\endgroup$ – user348338 Jun 23 '16 at 17:19
  • $\begingroup$ it is a product.. by $(f \circ i_{j})$ I mean the sequence for all $j \in J$ $\endgroup$ – user321268 Jun 23 '16 at 17:21
  • $\begingroup$ @SpamIAm you are right, as I perceive this isn't our problem in that question. $\endgroup$ – user321268 Jun 23 '16 at 17:23
  • $\begingroup$ @SpamIAm, Finite products and finite coproducts of modules are the same. $\endgroup$ – user348338 Jun 23 '16 at 17:23
  • 1
    $\begingroup$ @SpamIAm You mentioned: ``These happen to be the same in the category of modules, but I think this bears mentioning." This is true for finite products/coproducts, but not the same for arbitrary products/coproducts. $\endgroup$ – user348338 Jun 23 '16 at 17:29

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.