I N T R O D U C T I O N T O V O L U M E 9 x l i
after he completely revised the theory he held at the time, and, with it, his predic-
tion of the magnitude of light bending. Neither was Einstein averse to modifying
even the field equations of general relativity, once it had been completed, as when
he introduced the cosmological constant. In April 1919, Einstein published a paper
that advanced yet another modification of his gravitational equations. Motivated by
concerns to account for the structure of matter, he introduced modified gravitation-
al field equations that are trace free rather than having vanishing covariant diver-
gence (Vol. 7, Doc. 17). With respect to gravitational light bending, the equations
were equivalent to the ones with cosmological constant which did not appreciably
alter the predicted light bending for light passing near the sun. The modification
amounted to adding a gravitational term to the electromagnetic field energy that
was interpreted as a negative pressure term and was to account for the stability of
a charge-carrying particle, in analogy to a theory by Henri Poincaré, who had in-
troduced a similar pressure term to account for the stability of the Lorentz electron.
Einstein commented on this work in a distinctly ambivalent way: on 22 March he
was “urgently preoccupied with a problem in general relativity which does not give
me peace, day or night” (Doc. 10). A few days later he wrote: “I have found an in-
teresting idea in general relativity. Let’s hope it stands up; in such cases, after a
short while, criticism and dejection generally set in!” (Doc. 16). On the day of the
eclipse itself, he sent a copy of his paper to Theodor Kaluza, commenting that,
“short of something better,” it would adhere to the dualistic conception but that it
may nevertheless be of a certain interest (Doc. 48). A few days later, he expressed
doubts as to whether he had found “the right thing” (Docs. 52 and 59).
In the immediate aftermath of the eclipse announcement, Einstein wrote a short
newspaper article on “Induction and Deduction in Physics” (Vol. 7, Doc. 28). In
what appears to be a direct reflection on the recent confirmation of his gravitational
light-bending prediction, he writes that the actual course of science rests on a
hypothetical-deductive method rather than upon a model of progress by induction.
In this brief piece Einstein puts forth some explicit falsificationist contentions,
namely, that while “the truth of a theory can never be proven,” the theory itself can
be proven wrong by experiment.
Even with two empirical triumphs, and confident in the ultimate success of a
third, Einstein remained on the lookout for further tests. In late 1919 and early 1920
he pursued with Freundlich the possibility that the rotation of very large scale grav-
itational systems, such as globular star clusters, might provide evidence for the ex-
istence of a nonzero cosmological constant (Docs. 197 and
240).[38]
In late 1919, Einstein received a proposal from Wilhelm Hort, an engineer, to
test experimentally the Lense–Thirring effect (the “dragging of inertial frames”)
(Doc. 176). Einstein’s theory was seen here as part of a Machian tradition, since