Ullica Segerstrale examines the Millikan oil-drop experiment
on a number of levels. The details of discrepancies between his published
results and his laboratory notebooks are described. Specifically,
he took data on more oil-drops than he actually reported, even though his
paper stated that all oil-drops were included in his data analysis.
Also discussed is the treatment of the Millikan story by various authors,
and a concern is raised about the dangers of basing an ethical analysis
on brief "canned" versions of a complex situation. Lastly, Segerstrale
addresses the ethical implications of Millikan's data analysis, the extent
to which such procedures are accepted in the scientific community, and
their impact upon the scientific community. She asks, "Is the ethical
accountability now increasingly required from science in principle attainable
in a system which is so constructed that one gets rewarded for being a
quick-and-dirty first rather than a conscientious second?"
DISCUSSION
Another interesting sidelight
to the Millikan oil-drop experiment involves the viscosity of air, a quantity
necessary for his calculations but, as it was later discovered, the source
of a significant portion of his deviation from what is now the accepted
value for the electronic charge. John Thomsen has pointed out (American
Journal of Physics 36:368 (1968))that in Millikan's 1911 paper he mentioned
that he had two other methods for finding e without relying on the viscosity
of air. He rejected these approaches as being unnecessary.
The question was posed in that 1968 note, "Are such modifications possible?"
To date, no one has come forth with an answer.
It has been pointed out by Sharon
Traweek in Beamtimes and Lifetimes (Cambridge, MA: Harvard University Press,
1988) that science is often presented in textbooks as a "history of saints".
Well-known scientists are put on a pedestal. Perhaps the reaction
of present day physicists to the issues raised by Millikan's oil-drop experiment
are better understood in light of this observation. They were not
so much defending Millikan's methods as they were his reputation as a "saint".
Why, if Millikan and so many
others fudged their data, did their results hold up? How likely is
it for wrong methods to give good results? One factor in the Millikan
case is that Millikan was not after the "answer" but rather he was after
reducing the error (and doubt about the theory's correctness). Would
there have been any ethical problems if Millikan had in fact reported all
of his data and then proceeded with the same data analysis?
Finally, some technical points were
raised. First, many physicists would whole-heartedly agree with Millikan's
dropping data associated with drops containing a net charge greater than
20-30 e (although arguably he should have reported having done so).
Extracting the value of the elementary charge from such drops is problematic
at best, given the equipment of his times. Secondly, it is misleading
to imply that the fractional charge theory which Millikan was fighting
has some credibility in light of the present theory of quarks. No
generally accepted theory involving quarks also involves one being able
to isolate a quark and thus being able to measure a fractional charge.