I want to prove Sean Carroll's "spacetime and geometry"'s eq.(2.78): $$ \mathrm{d}(\omega \wedge \eta)=(\mathrm{d} \omega) \wedge \eta+(-1)^p \omega \wedge(\mathrm{d} \eta) \tag{2.78} $$ where $\omega$ is a $p-$form, $\eta$ is a $q-$form. There are many versions of proof online, but most of them use directly the wedge product $\wedge$, which is more mathematics. But I want to use the component version to prove it. Start with $$(A \wedge B)_{\mu_1 \cdots \mu_{p+q}}=\frac{(p+q) !}{p ! q !} A_{\left[\mu_1 \cdots \mu_p\right.} B_{\left.\mu_{p+1} \cdots \mu_{p+q}\right]} \tag{2.73}$$ and $$(\mathrm{d} A)_{\mu_1 \cdots \mu_{p+1}}=(p+1) \partial_{\left[\mu_1\right.} A_{\left.\mu_2 \cdots \mu_{p+1}\right]} \tag{2.76} $$ The L.H.S of (2.78) is $$\begin{aligned} \mathrm{d}(\omega \wedge \eta)&=\mathrm{d}(\frac{(p+q) !}{p ! q !} \omega_{\left[\mu_2 \cdots \mu_{p+1}\right.} \eta_{\left.\mu_{p+2} \cdots \mu_{p+q+1}\right]}) \\ &=(p+q+1)\frac{(p+q) !}{p ! q !}\partial_{\left[\mu_1 \right.}\omega_{ \left[\mu_2 \cdots \mu_{p+1}\right.} \eta_{\left.\left.\mu_{p+2} \cdots \mu_{p+q+1}\right]\right] } \\ &\stackrel{?}{=}\frac{(p+q+1) !}{p ! q !}\partial_{\left[\mu_1 \right.}\frac{1}{(p+q)!}(\omega_{ \mu_2 \cdots \mu_{p+1}} \eta_{\mu_{p+2} \cdots \mu_{p+q+1} }+ \text{alternating sign sums })_{\left. \right]}\\ &=\frac{(p+q+1) !}{p ! q !}\partial_{\left[\mu_1 \right.}(\omega_{ \mu_2 \cdots \mu_{p+1}} \eta_{\mu_{p+2} \cdots \mu_{p+q+1} })_{\left. \right]} \\ &=\frac{(p+q+1) !}{p ! q !}(\partial_{\mu_1 }\omega_{ \mu_2 \cdots \mu_{p+1}} \eta_{\mu_{p+2} \cdots \mu_{p+q+1} }) \end{aligned}$$ where in the $\stackrel{?}{=}$ step, I use the fact that all the $(p+q)!-1$ terms can equal to the first term under suitable change of signs.

Now, for the R.H.S of (2.78) $$\begin{aligned}(\mathrm{d} \omega) \wedge \eta&=((p+1)\partial_{\left[\mu_1 \right.}\omega_{\left. \mu_2 \cdots \mu_{p+1}\right]})\wedge \eta \\ &=(p+1)\frac{(p+q+1)!}{(p+1)!q!}(\partial_{\left[\left[\mu_1 \right. \right.}\omega_{\left. \mu_2 \cdots \mu_{p+1}\right]} \eta_{\left. \mu_{p+2} \cdots \mu_{p+q+1} \right]} ) \\ &=\frac{(p+q+1)!}{(p)!q!}(\partial_{\mu_1 }\omega_{ \mu_2 \cdots \mu_{p+1}} \eta_{\mu_{p+2} \cdots \mu_{p+q+1} } )\end{aligned} $$ where in the last step I use a similar method with $\stackrel{?}{=}$. But, this already equal to the L.H.S of (2.78). So what's wrong with my derivation?

  • $\begingroup$ I think you weren't careful enough about the derivatives. On the LHS, you are taking derivative of the product of $\omega$ times $\eta$, whereas the RHS term you computed has only a derivative of $\omega$. If you expand the derivative of the product on the LHS, you get a second term which you can only capture with the $\omega \wedge (d \eta)$ term $\endgroup$ Feb 19, 2023 at 15:50
  • $\begingroup$ Would Mathematics be a better home for this question? $\endgroup$
    – Qmechanic
    Feb 19, 2023 at 17:56

1 Answer 1


I think your derivation is correct, except that $$ d(\omega \wedge \eta)_{\mu_1 ... \mu_{p+q+1}} = \frac{(p+q+1)!}{p! q!} \partial_{\mu_1}(\omega_{\mu_2 ... \mu_{p+1}} \eta_{\mu_{p+2} ... \mu_{p+q+1}}) \neq \frac{(p+q+1)!}{p! q!} (\partial_{\mu_1}\omega_{\mu_2 ... \mu_{p+1}}) \eta_{\mu_{p+2} ... \mu_{p+q+1}} $$

Instead $$ d(\omega \wedge \eta)_{\mu_1 ... \mu_{p+q+1}} = \frac{(p+q+1)!}{p! q!} (\partial_{\mu_1}\omega_{\mu_2 ... \mu_{p+1}}) \eta_{\mu_{p+2} ... \mu_{p+q+1}} + \frac{(p+q+1)!}{p! q!} \omega_{\mu_2 ... \mu_{p+1}} (\partial_{\mu_1} \eta_{\mu_{p+2} ... \mu_{p+q+1}}) $$ The first term, as you've already shown, is equal to $d\omega \wedge \eta$ and for the second term: $$ (\omega \wedge d \eta)_{\mu_1 ... \mu_{p+q+1}} = \frac{(p+q+1)!}{p! q!} \omega_{\mu_1 ... \mu_p} (\partial_{\mu_{p+1}} \eta_{\mu_{p+2} ... \mu_{p+q+1}}) $$ which differs from the previous second term by a cyclic permutation of $\mu_1 ... \mu_{p+1} $ with sign $(-1)^p$

  • $\begingroup$ Great! Yeah my mistake for derivative! Thank you very much! $\endgroup$
    – Daren
    Feb 20, 2023 at 4:06

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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