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

Let $M$ be a $2d$-dimensional manifold. We say that $\omega \in H^2(M)$ has the Hard Lefschetz Property (HLP) if multiplication with $\omega^j$ is an isomorphism $H^{d-j} \to H^{d+j}$. This holds for the first Chern class of an ample line bundle on a smooth projective variety, this is the classical Hard Lefschetz Theorem.

My question is whether there is a generalization to classes in $H^4, H^6,$ ... etc.

My proposed definition is that $\eta \in H^k(M)$ has the k-HLP if multiplication by $\eta^j$ from $H^i$ to $H^{i+jk}$ is invective (surjective) if $|i-d|$ is greater (smaller) than $|i+jk-d|$. Then 2-HLP is HLP and if $\omega$ has the HLP then $\omega^j$ has the $2j$-HLP.

Are there other examples of classes with the $2j$-HLP? E.g. Chern classes of vector bundles satisfying an appropriate ampleness condition?

share|cite|improve this question
I like this question! One little fact which is relevant but doesn't answer your question: if one chooses any $d$ ample classes and uses them to define maps $H^0 \to H^2 \to H^4 \dots \to H^{2d}$ then hard Lefschetz and the Hodge-Riemann relations are still satisfied. Hence any product of $j$ ample classes will give you a $2j$-HLP. Cattani is the expert, e.g. – Geordie Williamson Oct 5 '13 at 22:51
Hard Lefschetz can be seen as "coming from $H^*(\mathbb{P}^n)$". Next on the scale of "easy" cohomology rings are Grassmannians. I wonder if the Chern classes of the tautological bundle give $2j$-HLP's. There are explicit formulas (Pieri rule) so it shouldn't be difficult to decide... – Geordie Williamson Oct 5 '13 at 22:59

As Geordie pointed out in the comments, products of Kähler classes have the $2j$-HLP.

In the paper "The mixed Hodge-Riemann bilinear relations for compact Kähler manifolds" by T-C. Dinh and V-A. Nguyên, the authors introduce the notion of Hodge-Riemann form.

A real form $\Omega$ of type $(k,k)$ is called a Hodge-Riemann form if at each point there is a continuous path $\Omega_t$ with $\Omega_0 = \Omega $ and $\Omega_1 = \omega^k$, where $\omega$ is a fixed Kähler form, in such a way that for each $t$ the map $\alpha \mapsto \Omega_t \wedge \omega^{2r} \wedge \alpha$ is an isomorphism for $0 \leq r \leq \min\{p,q\}$.

They proove that the Hodge-Riemann forms satisfy the Hard Lefschetz Theorem. Using a result by V. Timorin this result apllies for $\Omega = \omega_1 \wedge \cdots \wedge \omega_k$, where $\omega_j$ are Kähler forms.

share|cite|improve this answer

I guess I knew that I should've looked in part 2 of Lazarsfeld's book before posting this question, but it was checked out of the library...

Anyway, the speculation in the last paragraph is correct: if $X$ is a smooth projective variety and $E$ is an ample vector bundle on $X$ of rank $e$, then $c_e(E)$ has the $2e$-HLP. This is exactly the statement of Theorem 7.1.10 in Positivity in Algebraic Geometry II. (Lazarsfeld only considers multiplication by a single top Chern class, not powers of it, but this is equivalent by replacing $E$ by $E^{\oplus r}$).

share|cite|improve this answer

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.