On Carroll's spacetime and geometry book, page 67, the book gives the component form of vector field commutator $$[X,Y]^\mu=X^\lambda\partial_\lambda Y^\mu- Y^\lambda\partial_\lambda x^\mu \tag{2.23}$$ following sentence

Yet another fascinating exercise is to perform explicitly a coordinate transformation on the expression (2.23), to verify that all potentially nontensorial pieces cancel and the result transforms like a vector field.

I am trying to understand this sentence.I think

(1) $X^\lambda\partial_\lambda=X^{\lambda^\prime}\partial_{\lambda^\prime}$ is invariant under coordinate transformation;

(2) $Y^{\mu^\prime}=\frac{\partial x^{\mu^{\prime}}}{\partial x^{\mu}}Y^{\mu}$ is just the transformation of a vector field.

So where is the nontensorial pieces?

  • 4
    $\begingroup$ Well, $\partial_{\lambda'}$ will also hit the change of basis matrix. This is what gives rise to the non-tensorial part, but in the commutator it gets canceled. $\endgroup$
    – Gold
    Jan 16, 2023 at 14:27
  • 1
    $\begingroup$ @Gold, thank you very much! I think one critical point is that $\frac{\partial}{\partial x^{\lambda^{\prime}}}\frac{\partial x^{\mu^{\prime}}}{\partial x^{\mu}}=\frac{\partial^2 x^{\mu^{\prime}}}{\partial x^{\lambda^{\prime}} \partial x^{\mu}}=\frac{\partial^2 x^{\mu^{\prime}}}{ \partial x^{\mu} \partial x^{\lambda^{\prime}}}=\frac{\partial}{\partial x^{\mu}}\frac{\partial x^{\mu^{\prime}}}{\partial x^{\lambda^{\prime}}} $. So the two non-tensorial terms can cancel with each other. Would this be right? Thanks! $\endgroup$
    – Daren
    Jan 17, 2023 at 14:34
  • $\begingroup$ You are mixing derivatives in two different coordinate systems there. The argument is much simpler though, I have posted an answer with an outline. $\endgroup$
    – Gold
    Jan 17, 2023 at 15:33

1 Answer 1


This is actually simple. Since as you point out $X = X^\lambda \partial_\lambda$ and $Y=Y^\lambda\partial_\lambda$ as vector fields (not their components) are themselves invariant, we can evaluate the transformation of the commutator as follows

$$[X,Y]^{\mu'}= X(Y^{\mu'})-Y(X^{\mu'})=X^\lambda \partial_\lambda Y^{\mu'}-Y^\lambda \partial_\lambda X^{\mu'}.$$

Then we just write $X^{\mu'}$ and $Y^{\mu'}$ in terms of $X^\mu$ and $Y^\mu$ in the original unprimed coordinates

$$[X,Y]^{\mu'}=X^\lambda \partial_\lambda \left(\dfrac{\partial x'^\mu}{\partial x^\nu}Y^\nu\right)-Y^\lambda \partial_\lambda \left(\dfrac{\partial x'^\mu}{\partial x^\nu}X^\nu\right).$$

Now use the product rule to get

\begin{eqnarray} [X,Y]^{\mu'}=\dfrac{\partial x'^\mu}{\partial x^\nu}[X,Y]^\nu+(X^\lambda Y^\nu-X^\nu Y^\lambda) \dfrac{\partial^2 x'^\mu}{\partial x^\nu \partial x^\lambda} .\end{eqnarray}

The second term is then noted to be the non-tensorial part. Observe, though, that it vanishes because $\frac{\partial^2 x'^\mu}{\partial x^\nu \partial x^\lambda}$ is symmetric while $X^{\lambda}Y^\nu - X^\nu Y^\lambda$ is anti-symmetric under the exchange $\lambda \leftrightarrow \nu$. As a result

\begin{eqnarray} [X,Y]^{\mu'}=\dfrac{\partial x'^\mu}{\partial x^\nu}[X,Y]^{\nu}.\end{eqnarray}

One can also prove that $[X,Y]$ defines a vector field without resorting to coordinates and this is quite instructive. The key thing about being a vector field is that $[X,Y]$ should obey the Liebnitz rule when acting on functions, i.e., we need


This is a more abstract definition of a vector field: a vector field is a derivation acting on smooth functions. Using this definition it is straightforward to prove $[X,Y]$ is a vector field:

\begin{eqnarray}[X,Y](fg)&=&X(Y(fg))-Y(X(fg))\\&=&X(Y(f)g+fY(g))-Y(X(f)g+fX(g))\\ &=& [X(Y(f))g+Y(f)X(g)+X(f)Y(g)+fX(Y(g))]\\ &&-[Y(X(f))g+X(f)Y(g)+Y(f)X(g)+fY(X(g))]\\ &=& X(Y(f))g+ fX(Y(g))- Y(X(f))g-f Y(X(g))\\ &=& g[X,Y]f+f[X,Y]g.\end{eqnarray}


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.