1
$\begingroup$

My professor said that a hyperplane in projective space can be assumed to be $V(x_0)$ by a linear change in coordinates. Im not quite sure what he means. A hyperplane in Hartshorne is defined to be the zero set of a linear homogeneous polynomial. So say $H = V(x_0 + x_1).$ Why can we assume $H = V(x_0)?$ Are the two isomorphic under some isomorphism of $\mathbb{P}^n?$ The only isomorphism I can think of is sending $x_0, x_1$ to $x_0$ is the pullback ring homomorphism of the respective coordinate rings but this is not an isomorphism.

$\endgroup$
  • $\begingroup$ Perhaps you could send $x_0$ to $(1:0:\dots:0)$, $x_1$ to $(0:1:0: \dots: 0)$, etc? A little linear algebra should prove that this is a linear automorphism of $\mathbb{P}^n$. $\endgroup$ – bounceback Oct 25 '18 at 22:30
0
$\begingroup$

The map \begin{align*} \mathbb{P}^n &\to \mathbb{P}^n\\ [x_0 : x_1 : \cdots : x_n] &\mapsto [x_0 + x_1 : x_1 : \cdots : x_n] \end{align*} is an isomorphism of $\mathbb{P}^n$ mapping $V(x_0+x_1)$ to $V(x_0)$.

This is really just an incarnation of changing bases. More generally, given vector space $V$ of dimension $n+1$ over a field $k$, a choice of basis $e_0, \ldots, e_n$ induces an isomorphism $\mathbb{P}(V) \to \mathbb{P}^n$ such that the following diagram commutes. The homogeneous coordinates on $\mathbb{P}(V)$ are basically just the linear maps $e_0^*, \ldots, e_n^*$ in the dual basis.

$\hspace{3.75cm}$

So, given a hyperplane $H = V(\ell)$ in $\mathbb{P}^n$ where $\ell = a_0 x_0 + \cdots + a_n x_n$ is some linear form, we can complete $v_0 = (a_0, \ldots, a_n)$ to a basis $\{v_0, v_1, \ldots, v_n\}$ for $k^{n+1}$. Letting $e_0, \ldots, e_n$ be the standard basis for $k^{n+1}$, then the change of basis map sending $e_i \to v_i$ induces a map $\mathbb{P}^n \to \mathbb{P}^n$ sending $V(\ell)$ to $V(x_0)$.

$\endgroup$

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.