0
$\begingroup$

In the book Hamilton's Ricci Flow, we are given the following definition for an inner product on $\Lambda^p\mathrm{T}^* M$. If $\gamma,\eta\in\Lambda^p\mathrm{T}^* M$, then we define $$\langle\gamma,\eta\rangle = p!g^{i_1 j_1}\ldots g^{i_p j_p} \gamma_{i_1\ldots i_p}\eta_{j_1\ldots j_p}.$$ But why is this positive definite? I.e., why is it true that $$\langle\gamma,\gamma\rangle = p! g^{i_1 j_1}\ldots g^{i_p j_p} \gamma_{i_1\ldots i_p}\gamma_{j_1\ldots j_p}>0$$ for $\gamma\neq 0$?

$\endgroup$
1
$\begingroup$

Because the product is coordinate invariant, so you can just take a local normal coordinate system around a point, in which the metric is diagonalized. Then the expression in local coordinates is manifestly the sums of squares of real numbers, and hence non-negative.


Alternatively and equivalently, if you take an orthonormal frame then the inverse metric $(g^{-1}) = \sum_{i = 1}^n (e_i)\otimes (e_i)$. Then $$ \langle \gamma, \gamma\rangle = p! \sum_{i_{1} \ldots i_{p} = 1}^n \left| \gamma_{i_1 i_2 i_3 \cdots i_p}\right|^2 \geq 0$$

$\endgroup$

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.