3 0 2 D O C U M E N T 1 9 0 N O V E M B E R 1 9 2 0
190. From Willem de Sitter
Arosa,[1]
4 November 1920
Dear Einstein,
It was with very great enjoyment that I read your Leyden inaugural
speech.[2]
What pleases me particularly in it is that you emphasized with such finality the un-
tenability of a purely mechanical explanation of nature. Even as a student (so,
around the time of 1894), I always bristled when someone explained matter by the
ether or by electricity, only to turn around again and seek material explanations for
the ether! That always seemed preposterous to me. Now you have decided to call
the -field the “ether,” and you show convincingly that this ether is just as good
as “matter,” if not better, as a physical primal substance [Ur-ding]. In my opinion,
there is consequently no reason left to look for a material carrier of inertia. Mach’s
requirement also seems to me simply to be a residue of the quest for a mechanical
explanation of nature (on the basis of action at a
distance).[3]
The ether is the carrier
of inertia. The material points are just discontinuities in the ether, i.e., in the -
field; the field itself is what is real.
From this point of view, though, it also seems to me that the ether of your system
(ds2
) does not have any more advantages over mine
( );
My system has the advantage that it avoids the inconvenience of closed space in
that it makes the “journey around the space”
impossible.[4]
In your theory, there are
ghosts of the Sun that are visible there (but not materially) where the Sun once was,
for inst., 500,000,000, 1,000,000,000, 1,500,000,000, etc., years ago. Astronomi-
cally (and geologically) these are short periods. A substantial portion of the objects
appearing to us as stars would thus just be ghosts. From that it would follow that
there would have to be many more young (apparent) stars than old ones. In fact,
there are many more old stars than young ones, however. [This can be interpreted
as indicating either that the creation of stars is essentially already coming to an end,
hence that only a few new ones are being added, or that the young (i.e., “giant”)
stages are passed through much faster. The second interpretation seems to be the
true
one.][5]
It would be possible to get rid of these ghosts by assuming absorption.
But I do not believe that. The loss of light, according to Rayleigh’s law, along a path
that is traveled in 500,000,000 years with the appropriate density for “world
matter” would, according to your theory, be only about
1/100.—There
is, of course,
g
g
2 d –
c2dt2
+=
ds2= 2
d –
2 c2dt2
cos +
d spatial line element = =
2
d
r2
d
R2 2
d
2 2
d
2
sin + sin +=
r
R
---
.
=