3

$R$ is a ring and $H$ is the intersections of all non-zero left ideals of $R$.Then prove if $H$ not equal to zero then $H$ is a two sided ideal in $R$ and that we have $H^2=0$ or else $R$ is a division ring

Since $H$ is the set of all non zero left ideals then we can find a $x$ which belongs to $H$ and hence $H$ is itself a left ideal . Since a left ideal not necessary equals right ideal.Then how to prove it please someone explain.

Krish
  • 7,262
  • @Harry49 That's not really necessary with this question. It is entirely legible without MathJax. If you really feel that it should be done, it is probably going to be done faster if you do it yourself. – Arthur Sep 05 '17 at 13:26

1 Answers1

3

If $H$ is nonzero, then $H$ is obviously a minimal left ideal. For any $r\in R$, $Hr$ is another left ideal, in fact, a homomorphic image of $H$. Because $H$ is simple, $Hr$ is either $\{0\}$ or simple. If $Hr$ is nonzero, then because we know $H\subseteq Hr$ and $Hr$ is simple, we have $H=Hr$. So in either case, $Hr\subseteq H$, and $H$ is a right ideal.$^\ast$ It's obviously a left ideal (being an intersection of left ideals.) So it is a two-sided ideal.

If $H=R$, then we obviously have a division ring. Otherwise, $R$ has nontrivial maximal left ideals. In that case, $H$ is contained in the intersection of all maximal left ideals, a.k.a. the Jacobson radical.

But the Jacobson radical annihilates simple left modules including $H$. Therefore $\{0\}\subseteq H^2\subseteq J(R)H=\{0\}$ implies $H^2=\{0\}$.


$^\ast$ Alternatively: $H$ is then contained in the left socle $Soc(_RR)$, which is always a two-sided ideal. (Help if you need it.) But if $Soc(_RR)$ was a direct sum of more than one minimal left ideal, $H$ could not be contained in the other minimal ideals. So $Soc(_RR)=H$ is a two-sided ideal.

rschwieb
  • 160,592
  • Sir I am not familiar with the concept of socle and Jacobson radical annihiltes .Can you explain it more simple. – user476538 Sep 05 '17 at 16:44
  • @user476538 Well, I changed the part about the socle to be a little more elementary, but I don't see any reason to change the part about the Jacobson radical. That is a fundamental thing for rings, and if you don't know about it yet, you should be learning it ASAP. – rschwieb Sep 05 '17 at 18:01
  • 1
    Thank you Sir! For such an simpler explanation and I learnt about Jacobson radical. – user476538 Sep 06 '17 at 00:26
  • @user476538 glad you made progress ! – rschwieb Sep 06 '17 at 02:36
  • Is there a more elementary solution that contains only concept from undergraduated class? (I am aware to the fact that those concepts are fundamental, but it's still interesting). – I am not Paul Erdos Sep 06 '17 at 11:43
  • @IamnotPaulErdos Sure, you could just pull the guts of the proof that the intersection of maximal left ideals annihilates simple left modules out and make a completely elementary proof. I don't think I will do that here... Besides, don't people learn about the Jacobson radical in an undergrad course? I'm pretty sure that was in my first course on ring theory... – rschwieb Sep 06 '17 at 12:57
  • Okay, I think the view with Jacobson redical is more natural. Sometimes there are two undergraduated course in algebra: a mixed of group theory and ring theory, and a second course on field theory (undergraduaded studends can take more advanced courses, off course).Thanks – I am not Paul Erdos Sep 06 '17 at 13:13
  • @rschwieb Why does $R$ have nontrivial maximal left ideals when $H\neq R$? – karparvar Sep 28 '21 at 14:25
  • @karparvar All rings with identity (I see no reason to believe we are in a situation other than that here) have nontrivial maximal right ideals if they are not division rings. – rschwieb Sep 28 '21 at 14:29
  • @rschwieb I mean just the very point. Here in this problem, we are not assumed that $R$ has necessarily an identity. – karparvar Sep 28 '21 at 14:41
  • @karparvar In all public contexts I'm aware of (stackexchange, wikipedia, etc) the common practice is to assume identity is available, unless explicitly noted otherwise. So you may be nearly alone in perceiving what I've done as "a problem." I'm going to do some checking though... I think this argument probably works even without identity. – rschwieb Sep 28 '21 at 15:33