Suppose $\forall x_1,\ldots,x_n,\in R^n$, denote $x_1=(x_1^1,x_1^2,\ldots,x_1^n),x_2=(x_2^1,\ldots,x^n_2)$. Define $\psi:V\times V\times \dots\times V\to R$ as follows: $$\psi (x_1,\ldots,x_n)=\det \begin {pmatrix} x^1_1 & \dots &x^1_n \\ \vdots & \ddots &\vdots \\ x^n_1 &\dots &x^n_n\end{pmatrix}.$$ Then is $\psi$ multilinear?

  • 1
    $\begingroup$ What have you tried? What do you think? Have you explored the simpler case of $n = 2$? $\endgroup$ – Tom Oct 27 '14 at 1:25
  • 1
    $\begingroup$ how do you define the determinant? $\endgroup$ – James S. Cook Oct 27 '14 at 1:29
  • $\begingroup$ Sum over the permutation for entries along the diagonal multiplied by signature, $\endgroup$ – pxc3110 Oct 27 '14 at 2:11
  • $\begingroup$ I think Tom had good advice, write out what you know for $n=2$. Or, think about $\psi(x_1+cy, \dots , x_n)$ and see if you can simplify it. Then, exploit the symmetry of the determinant in term of column exchange... $\endgroup$ – James S. Cook Oct 27 '14 at 2:42
  • $\begingroup$ Well, I think exchanging the column yields a minus sign isn't it? $\endgroup$ – pxc3110 Oct 27 '14 at 22:05

You probably mean $\psi \colon V^n \to R$.

Yes, $\psi$ is multilinear. This fact is one of the fundamental properties of determinants.

  • $\begingroup$ sure, but, if he has the misfortune of thinking the determinant is defined by the cofactor expansion then there is something for him to prove... $\endgroup$ – James S. Cook Oct 27 '14 at 1:29
  • $\begingroup$ Yes, the approach taken to prove it will depend on the definition of determinants used. $\endgroup$ – user187373 Oct 27 '14 at 1:32
  • $\begingroup$ exactly, if we used my preference for determinant the multilinearity is manifest in the definition. $\endgroup$ – James S. Cook Oct 27 '14 at 2:43

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.