In general relativity the christoffel symbols 'pop' out of the metric $g^{\mu \nu} dx_\mu dx_\nu$. When $g^{\mu \nu} dx_\mu dx_\nu$ is integrated it is a measure of distance between the $2$ points. Using this as a starting point we have:

$$ \frac{8 \pi G}{c^4}T^{\mu \nu} = G^{\mu \nu}$$

where $T^{\mu \nu}$ is the stress energy tensor and $G^{\mu \nu}$ is the Einstien tensor. In the weak field limit one can derive Poisson's equation:

$$ \nabla^2 \phi = 4 \pi G \rho $$

However, when Newtonian gravity approximated and integrated from the Poisson equation, we get:

$$ U_g = - G \frac{m_1 m_2}{r}$$

where $r$ is the displacement. More precisely if I look at the action in general relativity:

$$ S_{GR} = \int (g^{\mu \nu} dx_\mu dx_\nu)^{1/2} $$

This is a measure of distance. However the action in Newtonian gravity is:

$$ S_{NG} = \int (\frac{1}{2}m_1 v^2 + G \frac{m_1 m_2}{r}) dt $$

where $r$ is displacement.

I feel I missed something subtle on how distance became displacement. Does anyone mind sharing a proof with this point in mind?

  • 1
    $\begingroup$ What is the difference between the distance between two points and the displacement between two points? $\endgroup$ May 7, 2020 at 3:30
  • $\begingroup$ also why do you need distance as starting point to Einstein field equation? And did you consider that density in case of two point masses is given by $ \rho=m_1\delta(r-r_1)+m_2\delta(r-r_2) $? Might be where your problem lies? $\endgroup$
    – Umaxo
    May 7, 2020 at 4:38
  • $\begingroup$ @probably_someone I would makea difference when you integrate something like $\vec F.d \vec r$ $\endgroup$ May 7, 2020 at 5:20
  • $\begingroup$ @Umaxo usually one does begin with distance though right? $\endgroup$ May 7, 2020 at 5:23
  • $\begingroup$ @MoreAnonymous Can you explain what the difference is? I'm trying to understand precisely how you're defining these two terms. $\endgroup$ May 7, 2020 at 5:42

1 Answer 1


The displacement is hidden in the metric.

Notation $(X^0,X^1,X^2,X^3)=(t,x_1,x_2,x_3)$. Newton's equations are given by $$m\ddot{x}^{\alpha}(t)=mf^{\alpha}(x(t)),$$ where $$f^\alpha(x(t))=-\left[\frac{G M}{r(t)^2}\right]_\alpha$$ i.e. $\alpha$ component of the field. Taking Newtonian gravity as a curvature in spacetime, the required equations of motion are

\begin{equation} \ddot{X}^0=0 \\ \ddot{X}^\alpha-f^{\alpha}(X(t)) \dot{X}^0 \dot{X}^0=0. \end{equation}

We know that the autoparallel (geodesic) equation in spacetime obtained by minimizing the action $$\int d\lambda \sqrt{g_{i j}(\gamma(\lambda))\dot{\gamma}^i(\lambda) \dot{\gamma}^j(\lambda)},$$ (where $\gamma$ is the curve parameterized by $\lambda$) is given by $\ddot{X}^a+\Gamma^{a}_{b c}\dot{X}^b \dot{X}^c=0.$

Comparing with the above equations of motion gives us Christoffel symbols as

$\Gamma^\alpha_{00}=-f^\alpha$ which is a function of displacement. All other Christoffel symbols are zeros.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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