MathOverflow is a question and answer site for professional mathematicians. It's 100% free, no registration required.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

A real $(p,p)$-form $\Omega$ on complex manifold $X$ is called strictly positive $(p,p)$-form if $$(-\sqrt{-1})^{p}\Omega(v_1,\bar{v}_1,...,v_p,\bar{v}_p)>0$$ for any nonzero $v_1,v_2,...,v_p\in T^{1,0}(X)$.If $\omega$ its Hermitian form.

Does the $\omega^p$ is strictly positive $(p,p)$-form?

share|cite|improve this question

Actually, you don't have definition quite right. You have to assume that $v_1,v_2,\ldots,v_p$ are linearly independent over $\mathbb{C}$, not just nonzero, otherwise the quantity you have displayed will certainly be zero.

With this corrected definition, though, one clearly does have $\omega^p$ is positive for any real, positive $(1,1)$-form $\omega$ on a complex manifold. This follows immediately, for example, from the fact that the symmetry group of a positive Hermitian inner product on a complex vector space $V$ acts transitively on the set of complex $p$-planes in $V$.

share|cite|improve this answer

I will try to clarify a little bit your question and at the same time to give you an answer. All that I say, you can find on Demailly's book "Complex Analytic and Differential Geometry".


Let $V$ be a complex vector space of dimension $n$ and $(z_1,\dots,z_n)$ coordinates on $V$. We denote by $(\partial/\partial z_1,\dots,\partial/\partial z_n)$ the corresponding basis of $V$, by $(dz_1,\dots,dz_n)$ its dual basis in $V^*$ and consider the exterior algebra $$ \Lambda V^*_\mathbb C=\bigoplus\Lambda^{p,q}V^*,\quad \Lambda^{p,q}V^*=\Lambda^p V^*\otimes\Lambda^q\overline{V^*}. $$ Let us first observe that V has a canonical orientation, given by the $(n, n)$-form $$ \tau(z)=idz\wedge d\bar z_1\wedge\cdots\wedge idz_n\wedge d\bar z_n=2^n dx_1\wedge dy_1\wedge\cdots\wedge dx_n\wedge dy_n, $$ where $z_j=x_j+iy_j$. In fact, if $(w_1,\dots,w_n)$ are other coordinates, we find $$ dw_1\wedge\cdots\wedge d w_n=\det(\partial w_j/\partial z_k)dz_1\wedge\cdots\wedge dz_n, $$ $$ \tau(w)=|\det(\partial w_j/\partial z_k)|^2\tau(z). $$ Definition. A $(p,p)$-form $u\in\Lambda^{p,p}V^*$ is said to be positive if for all $\alpha_j\in V^*$, $1\le j\le q=n-p$, then $$ u\wedge i\alpha_1\wedge\bar\alpha_1\wedge\cdots\wedge i\alpha_q\wedge\bar\alpha_q $$ is a positive $(n,n)$-form. A $(q,q)$-form $v\in\Lambda^{q,q}V^*$ is said to be strongly positive if $v$ is a convex combination $$ v=\sum\gamma_si\alpha_{s,1}\wedge\bar\alpha_{s,1}\wedge\cdots\wedge i\alpha_{s,q}\wedge\bar\alpha_{s,q} $$ where $\alpha_{j,s}\in V^*$ and $\gamma_s\ge 0$.

It is straightforward to see that strongly positive implies positive and, moreover, the concepts of positive and strongly positive coincide in bidegree $(0,0)$, $(1,1)$, $(n-1,n-1)$ and $(n,n)$.

Now, you have that all positive form $u$ are real, that is satisfy $u=\bar u$. In particular, in terms of coordinates, if $$ u=i^{p^2}\sum_{|I|=|J|=p}u_{I,J}dz_I\wedge d\bar z_J, $$ then the coefficients satisfy the hermitian symmetry relation $\overline u_{I,J}=u_{J,I}$.

A form $u=i\sum_{j,k}u_{jk}dz_j\wedge d\bar z_k$ of bidegrees $(1,1)$ is positive if and only if $\xi\mapsto\sum u_{jk}\xi_j\bar\xi_k$ is a semi-positive hermitian form on $\mathbb C^n$.

Proposition. If $u_1,\dots,u_s$ are positive forms, all of them strongly positive (resp. all except perhaps one), then $u_1\wedge\cdots\wedge u_s$ is strongly positive (resp. positive).

Proof. The proof is immediate from the very definition.

What you are asking follows straightforwardly from what I have written (in particular the last proposition), taking all $u_j=\omega$.


If $\omega$ comes from a (positive definite) hermitian metric, then you can always choose coordinates such that your form $\omega$ is given by $\omega=i\sum_{j=1}^n dz_j\wedge d\bar z_j$. Then $$ \omega^p= i^{p^2}p!\sum_{|I|=p}dz_I\wedge d\bar z_I. $$ I leave you the pleasure to compute $\omega^p(v_1,\dots,v_p,\bar v_1,\dots,\bar v_p)$ and to discover how this quantity is related with the determinants of the order $p$ minors of the matrix whose columns are given by $v_1,\dots,v_p$.

share|cite|improve this answer
Thanks dear diverietti,Robert Bryant – Gran Murra Dec 26 '11 at 0:54

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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