D O C . 4 8 R E P L Y T O R E I C H E N B Ä C H E R 2 0 3

49. Response to Ernst Reichenbächer,

“To What Extent Can Modern Gravitational Theory

Be Established without Relativity?”

[Einstein 1920k]

DATED 20 November 1920

Published 17 December 1920

In: Die Naturwissenschaften 8 (1920): 1010–1011.

The question if the theory of gravitation can also be established and justified with-

out the principle of relativity must, in principle, undoubtedly be answered with

“yes.” Then, why the principle of relativity? First, I answer with a comparison. The

theory of heat certainly can be developed without using its second theorem; then,

why use the second theorem?

The answer is obvious. When there are two theories that in one field do justice

to the totality of ascertained experience, one prefers the one that needs fewer mu-

tually independent assumptions. From this point of view, the principle of relativity

is, for electrodynamics and for the theory of gravitation, as valuable as the second

theorem is for the theory of heat, because it would take many mutually independent

hypotheses to reach the conclusions of the theory of relativity without using the

principle of relativity. Until now, all attempts to avoid the postulate of relativity

have shown this.

Aside from this, the introduction of the general principle of relativity is also jus-

tified from an epistemological point of view. For the coordinate system is only a

means of description and in itself has nothing to do with the objects to be described.

Only a law of nature in a generally covariant form can do complete justice in this

situation, because in any other way of describing, statements about the means of

description are jumbled with statements about the object to be described. I mention

Galileo’s law of inertia as an example. In detailed formulation, it necessarily

sounds like this: material points sufficiently distant from each other move uniform-

ly in straight lines—provided the movement is referred to a suitably moving coor-

dinate system and time is suitably defined. Who does not sense the embarrassment

in this formulation? But deleting the second clause would be dishonest.

[1]

[p. 1010]

[2]

[3]

[4]