1

Let $R$ be a commutative ring with $1$ such that every proper ideal is a prime ideal. Prove that $R$ is a field. [Hint: for $a \neq 0$, consider the ideals $(a)$ and $(a^2)$].

Attempt.

Let $a \in R$ with $a \neq 0$. Now, $(a^2)$ is prime. Then $ra^2 \in (a^2)$ for $r \in R$. But I can write $ra^2 = ra \cdot a$. As $(a^2)$ is prime, then either $ra \in (a^2)$ or $a \in a^2$. Either way, $a \in (a^2)$ and so $(a) \subseteq (a^2)$. Trivially, we always have $(a^2) \subseteq (a)$. Thus, $(a)=(a^2)$.

Observe as well that since every ideal is prime, then $\langle 0 \rangle$ is prime. Thus, when $xy \in \langle 0 \rangle$ i.e. when $xy = 0$, then $x \in \langle 0 \rangle$ or $y \in \langle 0 \rangle$, i.e. $x=0$ or $y=0$. So $R$ is an integral domain.

Then as $(a) = (a^2)$, there is $r \in R$ so that $ra^2 = a$. As $R$ is an integral domain, we can cancel $a$ from both sides to get $ra=1$. And so $a$ is a unit, but then $R$ is a field.

rschwieb
  • 160,592
  • Just skip the variable $r.$ $a^2\in(a^2)$ so... – Thomas Andrews Jan 02 '25 at 20:55
  • You can't cancel $a+(a)$ from both sides since $a+(a)=0+(a).$ – Thomas Andrews Jan 02 '25 at 20:57
  • I fact $ra^2+(a)=ra+(a)=a+(a)=0+(a)$ are all the zero element of the quotient ring. – Thomas Andrews Jan 02 '25 at 20:59
  • Your effort is falling short because you are attempting to prove $a$ is a unit when you quotient, which it is not. $a$ is zero in the quotient ring. – Thomas Andrews Jan 02 '25 at 21:00
  • @ThomasAndrews I went to the quotient because I was tempted to say that since $(a)=(a^2)$ then $ra^2 =a$ for some $r \in R$ and then cancel $a$ to get $ra=1$ but I can't cancel since it's not clear that $R$ is an integral domain just yet – Aram Nazaryan Jan 02 '25 at 21:03
  • The condition of the problem means that the zero ideal is prime, so $R$ is an integral domain. – Thomas Andrews Jan 02 '25 at 21:04
  • I understand what you were trying to do, but you can't cancel zero in an integral domain, so that approach is not going to be possible. – Thomas Andrews Jan 02 '25 at 21:04
  • @ThomasAndrews I fixed it now. Thanks for your help – Aram Nazaryan Jan 02 '25 at 21:08
  • 2
    The exercise here is not worded properly. The statement "If $R$ is a commutative ring with unit where every ideal is a prime ideal, then $R$ is a field" is true, but it's only vacuously true: note that the unit ideal is not a prime ideal. The exercise also forgets that the trivial ring is not a field. What the exercise should have asked is: "prove that if $R$ is a nontrivial commutative ring with unit where every proper ideal is a prime ideal, then $R$ is a field". – Joe Jan 02 '25 at 23:05
  • Apart from being a duplicate of a duplicate of a duplicate, notice that this site is not a proof checking machine. Please check out the [tag:solution-verification] tag description: For posts looking for feedback or verification of a proposed solution. "Is this proof correct?" is too broad or missing context. Instead, the question must identify precisely which step in the proof is in doubt, and why so. This should not be the only tag for a question, and should not be used to circumvent site policies regarding duplication. – Martin Brandenburg Jan 02 '25 at 23:16
  • @MartinBrandenburg The step which was in doubt was clarified at the original time of the posting of the question; the question was then edited as comments provided feedback regarding the doubtful step – Aram Nazaryan Jan 03 '25 at 22:44

1 Answers1

1

The question here needs some qualification: $R$ should be required to be non-trivial and the assumption should be that every proper ideal is prime.

You introduced $r$, so that you can talk about a general member of $(a^2)$, but it has led to confusion. If you look at the simplest particular example of a member of $(a^2)$, then the fact that $(a^2)$ is prime should tell us something useful.

So, let $a$ be a non-zero element of $R$. Clearly $a^2 \in (a^2)$ which is a prime ideal by assumption. So by the definition of a prime ideal $a \in (a^2)$. So $a = xa^2$ for some $x \in R$. As the $0$ ideal is prime, $R$ is an integral domain (if $bc = 0$ then at least one of $b$ or $c$ is $0$, so $R$ has no non-trivial zero-divisors). Hence, as we have $1 \cdot a = x \cdot a \cdot a$, we can cancel $a$ on the right, to get $1 = x \cdot a$, giving us that $x$ is an inverse for $a$. So any non-zero element of $R$ has an inverse, i.e. $R$ is a field.

Rob Arthan
  • 51,538
  • 4
  • 53
  • 105
  • I think the OP has worked the answer out based on the comments while I was typing up this answer. I will leave my answer here, because the comments don't give all the details. – Rob Arthan Jan 02 '25 at 21:33
  • The exercise here is not worded properly. The statement "If $R$ is a commutative ring with unit where every ideal is a prime ideal, then $R$ is a field" is true, but it's only vacuously true: note that the unit ideal is not a prime ideal. The exercise also forgets that the trivial ring is not a field. What the exercise should have asked is: "prove that if $R$ is a nontrivial commutative ring with unit where every proper ideal is a prime ideal, then $R$ is a field". – Joe Jan 02 '25 at 23:04
  • @Joe: I've added a proviso to my answer. I think your comment helpfully points out some edge cases that need to be taken care of. The reasons for closing the question seem to me to be quite wrong. – Rob Arthan Jan 02 '25 at 23:12
  • This question seems not to meet the standards for the site, as it is a duplicate of a duplicate of a duplicate, and it is a solution-verification without indication which steps are unclear. Instead of answering it, it would be better to look for a good duplicate target, or help the user by posting comments suggesting improvements. Please also read the meta announcement regarding quality standards. – Martin Brandenburg Jan 02 '25 at 23:13
  • @RobArthan: No problem. I agree that the duplicates originally suggested don't seem close enough, but the one I just suggested in a comment above seems to be an exact match. Perhaps you should add your answer there. – Joe Jan 02 '25 at 23:15
  • 1
    @MartinBrandenburg: the questions cited as duplicates do not mention prime ideals: one is about all ideals and the other is about maximal ideals. – Rob Arthan Jan 02 '25 at 23:15
  • It's an exact duplicate of https://math.stackexchange.com/questions/2845868/let-r-be-a-commutative-ring-with-1-neq-0-prove-that-if-every-proper-ideal-o and many others. Elementary questions like these are duplicates in 99% of the time. Please check out the linked meta announcement which explains in detail what to do instead of answering. – Martin Brandenburg Jan 02 '25 at 23:19
  • 1
    @Joe: thank you. You have found a genuine duplicate. The question should not have been closed without a reference to that. – Rob Arthan Jan 02 '25 at 23:20
  • 1
    @MartinBrandenburg: but that isn't the duplicate cited in the reasons for closure. I am beginning to despair here. MSE questions are being closed without proper visibility of the actual justification for closure. The proposal to reopen was rejected without any useful justification. Yes, we now know that the question is a duplicate, but that's only happened because I complained. – Rob Arthan Jan 02 '25 at 23:24
  • That rschwieb chose the wrong duplicates at first is bad, but it doesn't change that this question is a duplicate. One of the exact duplicates has has been found by Joe afterwards, but anyone can do it (and should to it, instead of answering). When you enter every proper ideal is prime into site search, it is the second search result. There is no justification for your claim "that's only happened because I complained". – Martin Brandenburg Jan 03 '25 at 00:01
  • Your claim "The proposal to reopen was rejected without any useful justification." is also not true, there are currently 2 reopen votes, and more people can vote. But they should in fact be rejected since (a) this question is a duplicate, (b) it is an offtopic solution verification as explained. If you have general issues with the procedure of closing duplicates and offtopic questions, we may continue the discussion on meta. – Martin Brandenburg Jan 03 '25 at 00:07