THE NEW DYNAMICS 273
alence principle [S6]. In another modern review, the current situation is sum-
marized as follows:
So far [general relativity] has withstood every confrontation, but new confron-
tations, in new arenas, are on the horizon. Whether general relativity survives
is a matter of speculation for some, pious hope for some, and supreme confi-
dence for others. [W8]
With fervent good wishes and with high hopes for further experiments with rock-
ets, satellites, and planetary probes, I hereby leave the subject of the comparison
between theory and experiment in general relativity.
What did Einstein himself have to say in later years about the three successes?
I described in the previous chapter his high excitement at the time he found the
right value for the precession of the perihelion of Mercury. He still considered
this to be a crucial discovery when he sent Lorentz his New Year's wishes for
1916 ('I wish you and yours a happy year and Europe an honest and definitive
peace'): 'I now enjoy a hard-won clarity and the agreement of the perihelion
motion of Mercury' [E9]. As will be seen in the next chapter, the results of the
solar eclipse expeditions in 1919 also greatly stirred him personally. But, as is
natural, in later times he tended to emphasize the simplicity of the theory rather
than its consequences. In 1930 he wrote, 'I do not consider the main significance
of the general theory of relativity to be the prediction of some tiny observable
effects, but rather the simplicity of its foundations and its consistency' [E10]. More
and more he stressed formal aspects. Again in 1930 he expressed the opinion that
the idea of general relativity 'is a purely formal point of view and not a definite
hypothesis about nature. ... Non-[generally] relativistic theory contains not only
statements about things but [also] statements which refer to things and the coor-
dinate systems which are needed for their description; also from a logical point of
view such a theory is less satisfactory than a relativistic one, the content of which
is independent of the choice of coordinates' [Ell]. In 1932 he went further: 'In
my opinion this theory [general relativity] possesses little inner probability....
The field variables g^ and </>„ [the electromagnetic potentials] do not correspond
to a unified conception of the structure of the continuum' [E12].
Thus we see Einstein move from the joy of successfully confronting experimen-
tal fact to higher abstraction and finally to that discontent with his own achieve-
ments which accompanied his search for a unified field theory. He did not live to
again use tiny effects for the purpose of advancing physical knowledge. Nor have
we to this day recognized any tiny effects which we can be sure pose a threat to
the physical principles with which we, perhaps clumsily, operate.
General relativity does predict new tiny effects of a conventional kind, however.
One of these caught Einstein's attention in 1936 when R. W. Mandl pointed out
to him [ M4] that if an observer is perfectly aligned with a 'near' and a 'far' star,
then he will observe the image of the far star as an annular ring as a result of the
bending of its light by the near star. The idea was, of course, not new. Eddington