Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

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

I'm reading in a paper: "Let $\theta$ be a $(1,0)$-form with respect to $I$ ($I \theta = -i \theta$)". Here $I$ is the almost complex structure.

Any idea why it says $I \theta = -i \theta$ and not $I \theta = i \theta$?

NB: It is not a typo, the author repeats this several times.

share|cite|improve this question
What paper? Some more context would probably be helpful. – Micah Aug 12 '14 at 19:44
I don't think it will help, but here it is: see top of p.16 in – Seub Aug 12 '14 at 19:51
up vote 1 down vote accepted

It is a matter of convention, and some authors do not follow it. If $Z$ is a $(1,0)$ vector field, then $IZ = iZ$. If $\alpha$ is a $(1,0)$-form, then one has $I^* \alpha = i \alpha$, where $I^*(\alpha)(-) = \alpha(I-)$ is the (pointwise) pullback of $\alpha$ by $I$. Ok, so one can, say, define $I$ as $I^*$. But then you have a rather annoying sign showing up. Namely, denote by $g$ the map $T\otimes \mathbb{C} \to T^* \otimes \mathbb{C}$ obtained by complexifying the map $T \to T^*$ induced by the metric (also denoted by $g$). Then you have $g \circ I = -I^* \circ g$. In order to avoid the appearance of a minus sign, some authors instead define $I = -I^*$ instead of $+I^*$. This issue is briefly discussed in Chern's complex manifolds without potential theory (along similar lines).

Another advantage of the $-I^*$ convention in hyperkähler geometry is the following. Consider $K = IJ$. This implies that $K^* = J^* I^*$ (since the $*$ operation is contravariant). This can be annoying, if one uses the $+I^*$,$+J^*$,..., convention. However, using the $-I^*$,..., convention, one recovers the familiar $K = IJ$, also for $1$-forms.

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.