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

Given a smooth manifold U, we have a map $\wedge^2\Gamma(U,TU)\to \Gamma(U,TU)$ given by $X\wedge Y\mapsto [X,Y]$, where $TU$ denotes the tangent bundle. Is it possible to describe the map $\Gamma(U,T^*U)\to \Gamma(U,\wedge^2 T^*U)$ corresponding to this map.

share|cite|improve this question
Though Victor's answer sort of tells the moral/ideological story of it, the question is still wrong as stated. The commutator of vector fields is not a well-defined map from the global sections of the exterior square of the tangent bundle, since it is not linear over the functions on the manifold, but only over the constants. – Leonid Positselski Dec 8 '10 at 0:19
Additionally, the spaces $\Gamma(U,TU)$ and $\Gamma(U,T^{*}U)$ aren't dual to each other over the constants. – Victor Protsak Dec 8 '10 at 1:59
Well, when a map between the global sections of two vector bundles is linear over the functions, it does induce a map of the global sections of the dual bundles in the opposite direction. In this case, the exterior differential is not linear over the functions, and the commutator does not even exist as a map of global sections (of the kind stated in the question). – Leonid Positselski Dec 8 '10 at 2:35
up vote 11 down vote accepted

To expand on Leonid's comment, if $\omega$ is a 1-form and $X,Y$ are vector fields, then $$ d\omega(X \wedge Y) = X \omega(Y) - Y \omega(X) - \omega([X,Y]). $$ If the first two terms were not there, then one could say, as in Victor's answer, that the exterior derivative is (minus) the transpose of the Lie bracket of vector fields. The fact that the first two terms are there is symptomatic of Leonid's observation that the Lie bracket is not really a tensorial map.

share|cite|improve this answer
Thanks, I was looking for a formula like the one you have written. – Rex Dec 8 '10 at 18:29
You're welcome. Personally, though, I think that Urs's answer should be the accepted answer to the question which should have been asked. – José Figueroa-O'Farrill Dec 8 '10 at 22:36

The dual of a Lie bracket is the differential in the corresponding Chevalley-Eilenberg algebra.

Background, formulas, details and examples are at

nLab: Chevalley-Eilenberg algebra.

This makes sense for "Lie bracket" understood in the general sense of Lie algebroids and $L_\infty$-algebras and fully generally for $\infty$-Lie algebroids.

In the case at hand, when regarding $T X$ as a Lie algebroid (instead of regarding $\Gamma(T X)$ as just a Lie algebra) the corresponding CE-algebra is the de Rham complex

$$ CE(T X) = (\Omega^\bullet(X), d_{dR}) $$

and the general formula for the dual of a Lie bracket on a Lie algebroid reproduces the familiar formula for the de Rham differential.

share|cite|improve this answer

Yes. The dual of the Lie bracket is the exterior differential that maps 1-forms into 2-forms. See any good textbook on differential geometry.

share|cite|improve this answer
What particular book did you have in mind? – drbobmeister Dec 8 '10 at 2:46
This is morally correct but technically wrong, as pointed out in Jose's answer and the comments to the question (including yours). But, yes, I do agree at a moral level that the bracket and the differential are dual. One way to make this precise, as you probably know but I'll mention it for other readers, is the Koszul duality between Lie algebra/oids and differential graded commutative algebras. When applied to tangent bundle, this Koszul duality returns the de Rham complex. – Theo Johnson-Freyd Dec 8 '10 at 6:41

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.