In lieu of an abstract, here is a brief excerpt of the content:

214 21 Fisher: The Fable of the Wasp It will be seen then that the difference between Prof. Fisher and myself is not a matter of mathematics—heaven forbid—but of opinion. student 1938, 367 By how much may we expect the yield of variety B to exceed that of variety A if they were sown alternatively on the same soil in the same season? student 1926, 126 If Gosset was the Bee, his difficult friend Fisher was the Wasp. Gosset patiently tried for a quarter century to teach Fisher about human relations, such as the importance of being kind and telling the truth and practicing humility and giving credit to other scientists and being accurate about history . He tried to teach the Wasp about the Wasp’s own ␤-self, hoping Fisher would get around then to analyzing the how much of his ␤-coefficients. But this Wasp was not an apt student in matters of scientific ethics. He was not inclined to give anyone beyond himself scientific credit. Fisher died without acknowledging, for example, Edgeworth’s 1908 original insights on maximum likelihood, misleading historians of statistics for half a century.1 The Bee nudged him along in those 150 letters (Gosset [posthumous ] 1962). But the Wasp, obsessed as he was with his own ␣-self, wouldn’t listen, even after he made significant errors in his sometimes brilliant but uneven and difficult youth.2 The Wasp, in the end, never did make time for his ␤-self. The Bee did his best. He tried, for example to explain to the Wasp how to design and evaluate experiments economically, tasks at which the Bee Page 1 of Student 1908a. Karl Pearson encouraged Gosset to publish “The Probable Error of a Mean,” though Pearson did not see much value in it. The article went unnoticed until 1912, when Fisher wrote to Gosset to tell him about “degrees of freedom .” Fisher saw in Gosset’s articles of 1908 (1908a, 1908b) a revolution in the life and social sciences and for Fisher himself a major reputation in statistics, and produced them both. (Reproduction courtesy of Oxford University Press and Biometrika.) Fisher: The Fable of the Wasp ⱐ 215 had tried, failed, practiced, and improved upon for decades before the Wasp took his first post at Rothamsted. Gosset actually ran experiments on things that mattered to a bottom line, your uncle’s favorite beer, for example , and at the time the world’s largest brewer (Ziliak 2008a). Still the Wasp wouldn’t listen. In July 1923 Fisher, who had recently published results on one of his first experiments, wrote sharply to Gosset from Rothamsted—a post Gosset helped him to secure—asking Gosset how he would have designed the experiment.3 “How would I have designed the exp[eriment]?” Gosset replied to his adopted student. “Well at the risk of giving you too many ‘glimpses of the obvious,’” Gosset, an experimentalist with by then two decades of experience, wrote, “I will expand on the subject: you have brought it on yourself! The principles of large scale experiments are four,” he explained to Fisher in a reply dated July 30. “There must be essential similarity to ordinary practice. . . . Experiments must be so arranged as to obtain the maximum possible correlation [not the maximum possible statistical significance] between figures which are to be compared [like Leamer and other oomph-ful scientists, Gosset thought in terms of upper and lower bound estimates, best and worst case scenarios]. . . . Repetitions should be so arranged as to have the minimum possible correlation between repetitions (or the highest possible negative correlation). . . . There should be economy of effort [net pecuniary advantage in the 1905 sense].”4 Fisher shrugged. The economic approach to the design of experiments was too difficult. He never did try Gosset’s way.5 Fisher Was “Insensitive to Fellow Humans” Gosset’s methods sweetened the Guinness bottom line. But the Wasp, a Guinness drinker, didn’t care. Though an undoubted genius, a major figure of science in the twentieth century, Fisher was considered by friends and associates to be a blusterer. And yet he coveted the insights of the Bee, the anonymous Student, who was in his own way a genius. And the Bee’s lessons were packaged in a sweet scolding, perhaps irresistible. “I have come across the July J. A. S. [Journal of Agricultural Science] and read your paper,” the Bee wrote to the Wasp. “I fear that some people may be misled into thinking that because you have found no...

Share