Below are some quotes from Collins & Pinch, which
I find disturbing. As I recall from another
source, Miller was bitter at the end of his life.
[Miller] rebuilt the apparatus and took readings
again on 4, 5 and 6 September 1924. Miller now
found a persistent positive displacement, and
concluded that 'the effects were shown to be real
and systematic, beyond any further question'.
Miller's experiment was different from the others
in that he pressed ahead with the fourth part of
the protocol and took further readings in spring,
summer and the following autumn. He concluded, in
1925, that he had found an observed motion of the
earth of about 10 kilometres per second -- around
one third of the result that the original
Michelson experiments were expected to find. In
1925, Miller was awarded the 'American
Association for the Advancement of Science' prize for this work.
Thus, although the famous Michelson-Morley
experiment of 1887 is regularly taken as the
first, if inadvertent, proof of relativity, in
1925, a more refined and complete version of the
experiment was widely hailed as, effectively,
disproving relativity. This experiment was not
conducted by a crank or charlatan. It was
conducted by one of Michelson's closest
collaborators, with the encouragement of
Einstein, and it was awarded a major honour in the scientific community.
p. 40
. . .
[various descriptions of improved experiments and
more conclusive results from Miller]
In spite of this, the argument in physics was
over. Other tests of relativity, including the
Eddington observations of 1919 (to be discussed
below), indirectly bolstered the idea that the
theory of relativity was correct and that the
velocity of light must be constant in all
directions. The sheer momentum of the new way in
which physics was done the culture of life in
the physics community meant that Miller's
experimental results were irrelevant.
We have travelled a long way from the notion that
the Michelson-Morley experiment proved the theory
of relativity. We have reached the point where
the theory of relativity had rendered the
Michelson-Morley experiment important as a
sustaining myth, rather than as a set of results.
Results that ran counter to what it was believed
the Michelson-Morley experiment demonstrated were
largely ignored. Think of it this way. The notion
of 'anomaly' is used in science in two ways. It
is used to describe a nuisance - 'We'll ignore
that; it's just an anomaly', and to signify
serious trouble 'There are some troublesome
anomalies in the existing theory.' The
interferometry results started as serious trouble
for the theory of the aether. The null results
passed from anomaly to 'finding' as the theory of
relativity gained adherents. With Miller's
positive claims, interferometry results became,
once more, an anomaly, but this time they were
treated as a nuisance rather than a trouble.
Miller's results were 'just an anomaly that
needed to be explained away'. Miller could not
change the status of his positive readings from
nuisance to troublesome anomaly even though they
were the outcome of the best experiment yet
completed, perhaps the only one which could truly
be said to have tested what it was meant to test.
The meaning of an experimental result does not,
then, depend only upon the care with which it is
designed and carried out, it depends upon what people are ready to believe.
p. 42
