>harald says... > >>Acceleration effects are not identified as gravitational fields in >>Newtonian physics (which, as you now know, you didn't know); and >>neither is that the case in SRT. In those theories acceleration is >>"absolute", and no gravitational fields are caused by acceleration.
The idea that "gravitational fields are caused by acceleration" is, as I have pointed out many times, a very misleading way to describe the equivalence principle. It's not that acceleration effects are a kind of gravitational field, it's that gravitational fields are a kind of acceleration effect (they are an indication that you are accelerating relative to the local inertial frame).
You don't need a theory of gravity in order to know what things "look like" in an accelerated coordinate system. You just take the description in inertial coordinates, and apply a coordinate transformation. It's just calculus.
But if you do this for SR, you find some interesting things:
1. All objects accelerate downward (unless acted upon by a force) at a rate that is independent of the composition of the object. A thrown object will follow, not a straight line, but a parabola (roughly speaking). 2. Clocks that are "higher up" run faster than clocks that are "lower down".
These are effects of acceleration that are derivable without any mention of any theory of gravity. The key observation behind the equivalence principle is that fact (1) is true (or approximately true) for objects in a uniform gravitational field. So perhaps it is possible to interpret *gravitational* fields as a kind of acceleration field due to the use of noninertial coordinates. If so, this identification makes a definite prediction, namely (2) that clocks that are higher up in the gravitational field should run faster.
So the real insight in General Relativity is not that acceleration effects are gravitational fields, it's that gravitational fields are a kind of acceleration effect. To make this work, it is necessary to have curved spacetime, so that the notion of being unaccelerated can vary from place to place (to allow for nonuniform gravitational fields). The same insight could have been done with Newtonian physics, but historically it wasn't. Newtonian gravity can likewise be described in terms of curved spacetime, and under this description, the "force of gravity" becomes just an acceleration effect.
