1
$\begingroup$

If I have a singular matrix $X$ with components $X_{\mu\nu}$:

$t^{\nu}X_{\mu\nu}=0$

By considering now $X_{\mu\nu}$'s as components of a 2-form can I say that:

$X\wedge X=0$ ?

If yes, how?

$\endgroup$

1 Answer 1

5
$\begingroup$

Your condition on $X$ is that it has a kernel, and that by itself does not mean that $X \wedge X$ doesn't have to vanish. For instance in five dimensions, you could have $$X = \begin{pmatrix} 0 & 1 & 0 & 0 & 0 \\\\ -1 & 0 & 0 & 0 & 0 \\\\ 0 & 0 & 0 & 1 & 0 \\\\ 0 & 0 & -1 & 0 & 0 \\\\ 0 & 0 & 0 & 0 & 0 \end{pmatrix}.$$ Then letting $t = (0,0,0,0,1)$, your condition is satisfied, but $X \wedge X$ is not zero.

However, it is true in $2n$ dimensions that a non-zero $t$ exists if and only if $X^{\wedge n} = 0$. That's because $X^{\wedge n}$ is proportional to the Pfaffian of $X$, which is a certain square root of the determinant of $X$. (In odd dimensions, the determinant of an antisymmetric matrix is zero by calculation, while the Pfaffian is set to zero by definition. So $t$ always exists in this case.)

$\endgroup$
3
  • 2
    $\begingroup$ Typo? You meant $X^{\wedge n}$, right? $\endgroup$
    – t3suji
    Dec 23, 2009 at 17:33
  • $\begingroup$ Yes, thanks t3suji! This is the second time that you have improved my answer. $\endgroup$ Dec 23, 2009 at 18:41
  • $\begingroup$ thank you very much, I felt that what I was calculating $X\wedge X$ looks like a Pfaffian but I did not see the relationship. Now I can continue my study of the Plebanski action :-) $\endgroup$
    – Pedro
    Dec 23, 2009 at 21:50

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.