1

Is it true that $\langle x^2+y^2-1\rangle$ is prime in $\mathbb{C}[x,y]$ ?

user26857
  • 52,094
random_guy
  • 433
  • 3
    Yes, it is: there is no polynomial $f(y)\in \mathbb C[y]$ such that $f(y)^2+y^2-1=0$ and it is monic. (If don't like this use Eisenstein.) – user26857 Jan 22 '17 at 21:02

2 Answers2

2

One way to see that this ideal is prime is to show that the quotient ring is an integral domain. By change of variables $u=x+iy, v=x-iy$ you get that $$\mathbb{C}[x,y]/\left<x^2+y^2-1\right>\cong \mathbb{C}[u,v]/\left<uv-1\right>\cong \mathbb{C}[t^{\pm 1}].$$

user26857
  • 52,094
Ofir
  • 8,015
  • I wonder how would you solve the problem over $\mathbb R$ instead of $\mathbb C$, and more generally over any field of characteristic $\ne 2$? – user26857 Jan 22 '17 at 21:14
  • @user26857 Over $\mathbb{R}$ you can use the embedding $\mathbb{R}[x,y]/<x^2+y^2-1> \hookrightarrow \mathbb{C}[x,y]/<x^2+y^2-1>$ in order to get that the quotient is an integral domain. Also, the same argument used for $\mathbb{C}$ should work for every field which contains the square root of -1, and if it does not contain, you should add $\sqrt{-1}$ as in the argument for $\mathbb{R}$ – Ofir Jan 22 '17 at 21:19
  • @Orfir Is it obvious that $\mathbb{R}[x,y]/\left<x^2+y^2-1\right>\hookrightarrow\mathbb{C}[x,y]/\left<x^2+y^2-1\right>$? (Not mentioning the general case you suggested.) – user26857 Jan 22 '17 at 21:48
  • @user26857 It is not trivial, but not that hard to prove. What you need to show is that if $f\in \mathbb{R}[x,y]$ is divisible by $x^2+y^2-1$ over $\mathbb{C}$, then this is also true over $\mathbb{R}$. This is true since if $f=h(x,y)(x^2+y^2+1)$, then $Re(f)=Re(h)(x^2+y^2-1)$. – Ofir Jan 22 '17 at 22:56
  • Right. Btw, this property holds in general: if $f,g\in K[X_1,\dots,X_n]$ and $K\subset L$ is a field extension such that $f\mid g$ in $L[X_1,\dots,X_n]$ then $f\mid g$ in $K[X_1,\dots,X_n]$. – user26857 Jan 22 '17 at 23:02
-4

Yes. Since the conic $x^2+y^2=1$ is an irreducible curve in $\mathbb{C}^2$.

More than irreducible, its projective closure is smooth. One can simply check the vanishing of the partial derivatives.

math
  • 1,664
  • 2
    And how exactly do you prove that the curve is irreducible? – Mariano Suárez-Álvarez Jan 22 '17 at 21:05
  • How to prove the irreducibility ? – random_guy Jan 22 '17 at 21:05
  • 3
    Are you serioussly suggesting that someone who cannot prove that polynomial irreducible use the fact that smooth curves in projective space are irreducible to do it? – Mariano Suárez-Álvarez Jan 22 '17 at 21:09
  • @ Mariano Suárez-Álvarez. The answer meant to show an alternative way of looking at the problem.
    Many of the OP’s questions have the (alg. geometry)/(comm. algebra) tags. So I would not assume that he is not capable of understanding this possible approach. Even if that were the case, the answer can very well work as a motivation/invitation to explore related notions. I really don't see how this can be harmful.     – math Jan 22 '17 at 22:40