5 8 D O C . 4 4 T H E T H E O R Y O F R E L A T I V I T Y “gravitational field” in its neighborhood (just as a magnet produces a “magnetic field”), and it acts upon the stone and attracts it. Let us imagine a large region of space, so far removed from any stars, etc., that the law of inertia holds within it: Mass points at rest remain at rest, and masses in motion remain in a state of linear, uniform motion. As our frame of reference, we can imagine a large box shaped like a room, in which there is an observer. There is a hook attached to the outside of the ceiling, and a rope hangs from it, on which some sort of being is pulling: this makes the box “fly” upwards with an accelerated motion. What does the observer experi- ence? He feels in his legs the counterforce that is moving toward him from the floor of the box (just as we feel our weight due to the Earth’s gravitational attraction) a body released by the observer “falls” onto the floor of the box (as it would on the Earth), likewise with an acceleration that is independent of its specific weight. This “falling” can thus be interpreted in two ways: The frame of reference (the box) per- forms a uniformly accelerated translational motion, so that the falling bodies “in reality” remain at rest and the floor of the box is moving toward them or else: the frame of reference (the box) is not accelerated, but it is located in a gravitational field that “attracts” the bodies. The same property of the bodies manifests itself de- pending on the circumstances, as “inertia” (the body at rest in an accelerated sys- tem) or as a “weight” (the body falling in a system at rest). We see that the theory of general relativity must lead to a theory of gravitation, for one can “create” a gravitational field by simply changing the frame of reference (“pulling on the box”). Indeed, a consistent treatment of the idea of general relativity led to the laws that are obeyed by the gravitational field.[3] Furthermore, the introduction of a principle of general relativity is supported by the fact that the fundamentals of classical mechanics and of the theory of special relativity are in some respects not satisfactory, as shown by the following consid- eration. As is well known, classical mechanics is based on the law of inertia (I). This law, however, holds as we already know only for frames of reference that are in certain special states of motion, namely, uniform translational motions. It does not hold, for example, for a coordinate system that is rigidly connected to the Earth, since in such a system the fixed stars do not move linearly and uniformly, but along a gigantic circle. It is, however, rather unsatisfactory that the laws of nature are sup- posed to hold only for certain coordinate systems and not for others, without one’s being able to state a reason for this why. This was already recognized by. E. Mach,[4] who therefore called for mechanics to be placed on a new basis. That is what the theory of general relativity aims for its equations hold for any frame of reference, whatever its state of motion. [p. 8]