212
10 Statistics and Other Concerns
about the gullibility with which the scientific world had received those stories and
had Cook’s story assumed to be true. Kapteyn rose up against this with force; he
considered it natural that a priori a scientific man should be believed at his word
and not even to contemplate the possibility of misconduct. Tampering aroused
his deep indignation. Even after years his wrath could flare up over a physicist
who had come up with a theory, but when Kapteyn approached him to consult
certain books to test the theory, he replied: ‘But you cannot expect me to gather
facts against my own theory!’ And speaking of a well-known English biologist,
in whose writings he had found evidence of tampering, he said: ‘That’s the one
man I hate.’ Twice in his life he personally encountered men who were not fair
in science. He was horrified about this, it was impossible for him to put himself
in that mental attitude. He was finished with him forever, they had failed in the
elementary requirements of the code d’honneur and were, according to him, no
longer worthy to carry the torch.
The English biologist in this passage was most likely Pearson.
Kapteyn returned to these matters in 1916. He said that he had decided to
do this only after his retirement, because it took him too much time. But he
was offered help by the mathematician Marie Johan van Uven (1878–1959),
who had been a teacher at a gymnasium in Utrecht for a long time and had
been appointed professor in Wageningen in 1913. The Agricultural University
of Wageningen was not formally a university and actually had no professors,
but they made an exception for him, because the Technical University of Delft
(which had professors) had already offered him a professorship. Kapteyn and
van Uven wrote a long article, which actually consisted of three articles (two
by Kapteyn and Van Uven separately, on theory, and a third together on applications), [92]. In addition to mathematical details, they paid a lot of attention
to Kapteyn’s point of view that the form of a correlation should say something
about its cause, i.e. why particular correlation should give rise to a certain
form of the correlation function. This is no longer believed to be the case. An
extensive study of the Kapteyn versus Pearson case has been published by Ida
Stamhuis and Eugene Seneta [96], to which I refer for further details.
In the meantime Kapteyn had tackled another statistical subject, which
he published in the leading British astronomical journal Monthly Notices of the
Royal Astronomical Society. This occurred in 1912 and concerned the definition
of the correlation coefficient. This is a number that indicates the extent to
which two quantities, for example the mass and the luminosity of a star, are
correlated. This coefficient is a number between zero and one, such that when
there is no relationship between those two quantities it is equal to zero and
unity if they correlate perfectly. Kapteyn wondered why one should use that
number as such and not the square of it or maybe even a higher power or a root.
10 Statistics and Other Concerns
about the gullibility with which the scientific world had received those stories and
had Cook’s story assumed to be true. Kapteyn rose up against this with force; he
considered it natural that a priori a scientific man should be believed at his word
and not even to contemplate the possibility of misconduct. Tampering aroused
his deep indignation. Even after years his wrath could flare up over a physicist
who had come up with a theory, but when Kapteyn approached him to consult
certain books to test the theory, he replied: ‘But you cannot expect me to gather
facts against my own theory!’ And speaking of a well-known English biologist,
in whose writings he had found evidence of tampering, he said: ‘That’s the one
man I hate.’ Twice in his life he personally encountered men who were not fair
in science. He was horrified about this, it was impossible for him to put himself
in that mental attitude. He was finished with him forever, they had failed in the
elementary requirements of the code d’honneur and were, according to him, no
longer worthy to carry the torch.
The English biologist in this passage was most likely Pearson.
Kapteyn returned to these matters in 1916. He said that he had decided to
do this only after his retirement, because it took him too much time. But he
was offered help by the mathematician Marie Johan van Uven (1878–1959),
who had been a teacher at a gymnasium in Utrecht for a long time and had
been appointed professor in Wageningen in 1913. The Agricultural University
of Wageningen was not formally a university and actually had no professors,
but they made an exception for him, because the Technical University of Delft
(which had professors) had already offered him a professorship. Kapteyn and
van Uven wrote a long article, which actually consisted of three articles (two
by Kapteyn and Van Uven separately, on theory, and a third together on applications), [92]. In addition to mathematical details, they paid a lot of attention
to Kapteyn’s point of view that the form of a correlation should say something
about its cause, i.e. why particular correlation should give rise to a certain
form of the correlation function. This is no longer believed to be the case. An
extensive study of the Kapteyn versus Pearson case has been published by Ida
Stamhuis and Eugene Seneta [96], to which I refer for further details.
In the meantime Kapteyn had tackled another statistical subject, which
he published in the leading British astronomical journal Monthly Notices of the
Royal Astronomical Society. This occurred in 1912 and concerned the definition
of the correlation coefficient. This is a number that indicates the extent to
which two quantities, for example the mass and the luminosity of a star, are
correlated. This coefficient is a number between zero and one, such that when
there is no relationship between those two quantities it is equal to zero and
unity if they correlate perfectly. Kapteyn wondered why one should use that
number as such and not the square of it or maybe even a higher power or a root.