Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. 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

When I studied injective module there is a theorem which say that the two following statement are equivalent:

Let $R$ be a ring, $I$ is a left ideal of $R$, $J$ is a left $R$-module.

  1. For every $f:I\longrightarrow J$ module homomorphism there exist $m\in J$ that $f(r)=rm$ for every $r\in I$

  2. Every module homomorphism $f:I \longrightarrow J$ can be extended to a homomorphism $\bar{f}:R\longrightarrow J$.

Suppose we have 1 and we want to prove 2. Here is my argument:

Let $\pi : R\longrightarrow I$ with $\pi(r)=r$ where $r\in I$ and $\pi(r)=0$ otherwise. Then define $\bar{f}:R\longrightarrow J$ as $\bar{f}=f\circ \pi$.

I felt that my argument is not enough or even wrong, however, I had no idea of constructing such a homomorphism $\bar{f}$.

Please help me. Thanks

share|cite|improve this question
In statement 1., do you mean that for every module homomorphism $f: I \to J$, there exists an $m \in J$ such that $f(r) = rm$ for all $r \in I$? – Martin Wanvik Apr 14 '12 at 15:02
@MartinWanvik : Yes, thank you very much. I have edited it. – Arsenaler Apr 14 '12 at 15:07
up vote 1 down vote accepted

I don't see how your argument shows anything close to what you're trying to do (you appear to be trying to show the converse). To show that statement 2 implies statement 1, you would have to show that given some $f: I \to J$, you can write it as $f(r) = rm$ for some $m \in J$ under the assumption that statement 2 is true. The proof is actually quite easy: given $f: I \to J$, we can extend that to $\bar{f}: R \to J$ using statement 2. Now let $m = \bar{f}(1)$. Then $$f(r) = \bar{f}(r) = \bar{f}(r\cdot 1) = r\bar{f}(1) = r m,$$ which proves the result.

Edit: To show the converse, we can essentially reverse the above proof: given a module homomorphism $f: I \to J$, we can write it as $f(r) = rm$ for some $m \in J$ (by statement 1). We can extend $f$ to $R$ by defining $$ \bar{f}(r) := rm.$$ This coincides with $f$ on $I$, and it is easily seen to be an $R$-module homomorphism, since $$\bar{f}(rr') = rr'm = r(r'm) = r\bar{f}(r'),$$ and $$f(r+r') = (r+r')m = rm + r'm = \bar{f}(r) + \bar{f}(r'),$$ for all $r,r' \in R$.

Second edit: To see why your proof doesn't work, consider $R = \mathbb{Z}$, and $I = 2\mathbb{Z}$. Defining $\pi(z) = z$ if $z \in 2\mathbb{Z}$ and $\pi(z) = 0$ if $z \notin 2\mathbb{Z}$, then we must have $$2 = \pi(2) = \pi(2\cdot 1) = 2 \cdot \pi(1) = 0,$$ so $\pi$ is not a $\mathbb{Z}$-homomorphism. It cannot even be additive, since that would imply that $$2 = \pi(2) = \pi(1+1) = \pi(1) + \pi(1) = 0.$$

share|cite|improve this answer
Sorry. I mean if we have 1 and we want to prove 2. – Arsenaler Apr 14 '12 at 15:17
@msnaber: I've edited my answer. – Martin Wanvik Apr 14 '12 at 15:58
Thank you very much Martin Wanvik. – Arsenaler Apr 14 '12 at 16:11

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.