This is a screen shot of one page in David Tong's QFT notes.

This is a screen shot of one page in David Tong's QFT notes.

To clarify some of the notations, $p$ means the 4-vector momentum, while $\vec{p}$ denotes the 3-momentum, and similarly for $x$ and $\vec{x}$.

I have two questions:

  • In my understanding, causality forbids the propagation of a particle from one point to another that is spacelike separated, that is to say, the two point function has to vanish for spacelike intervals. Is this requirement equivalent to the equation (2.86) in the image that operators commute?

  • How is the integral in equation (2.89) zero? I couldn't get to understand David's arguments.

  • 2
    $\begingroup$ Tyler answered your second question, and I think this will answer your first one. $\endgroup$
    – gravitaunt
    Sep 21, 2016 at 2:34
  • $\begingroup$ I just found that the following chapter in David's notes expands a little on the mathematical equivalence of transition amplitude being zero and the operators commute for spacelike intervals. But the link you attached provids many physical insights into this! $\endgroup$
    – JamieBondi
    Sep 21, 2016 at 15:31

1 Answer 1


Let's just look at the second term in the integral (ignoring some pesky constants, and seeing that $E_p=E_{-p}$) \begin{align} \int_{-\infty}^{\infty} d^3p \frac{1}{2E_p}e^{-i\mathbf{p}\cdot\mathbf{x}}=-\int_{\infty}^{-\infty} d^3p\frac{1}{2E_p}e^{i\mathbf{p}\cdot\mathbf{x}}=\int_{-\infty}^{\infty} d^3p \frac{1}{2E_p}e^{i\mathbf{p}\cdot\mathbf{x}} \end{align} In the second equality I sent $\mathbf{p}\to-\mathbf{p}$. So, the whole integral will vanish...

  • 1
    $\begingroup$ This does not seem right: We have $\exp(\pm i\mathbf{p}\cdot\mathbf{x})= \cos(\mathbf{p}\cdot\mathbf{x})\pm i\sin(\mathbf{p}\cdot\mathbf{x})$. So the contribution of the cosine should remain nonzero. Or do I miss something? $\endgroup$ Sep 21, 2016 at 4:53
  • $\begingroup$ I don't see why that would affect anything....it will still cancel the first integral. $\endgroup$ Sep 21, 2016 at 5:47

Your Answer

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

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