To Teach the History of Science? 63 



involved but convincing proof of itJ^ After my lecture, one of the 

 students came to me and told me very gently that I was mistaken and 

 that the infinity of prime pairs had not yet been proved. I bade him to 

 come to my study to discuss the matter. The upshot of our discussion 

 was that the proof by Professor Moore had been shown to be imperfect; 

 arguments used in the theory of numbers are often very subtle and 

 tricky. I had read in Science tbe announcement of Moore's discovery, 

 but the disproof of it had not been registered in Science or I had failed 

 to notice it. The student who gave me that valuable information was a 

 graduate student who had been studying prime pairs for the last two 

 years and knew more about them than anyone else in the university. 



This is the most striking example in my experience of the cooperation 

 which may exist, and should exist, between the teacher and some, at 

 least, of his students. In this case, the student knew very well the topic 

 discussed; in the majority of cases, however, the student does not, but 

 if he be intelligent his queries and his doubts may be very stimulating 

 and oblige the teacher to consider the subject from a new angle. Many 

 of my lectures have been modified because of such queries. Moreover, 

 whenever a student has evoked a point requiring additional explanation 

 or emphasis, I have given the necessary explanation to the whole class,'''"* 

 being careful to name and to thank the student who had prompted me. 



Courses on the history of science have often been intrusted to pro- 

 fessors whose main function was to teach other subjects. Readers who 

 have followed me thus far will realize the utter unwisdom of that prac- 

 tice. The teaching of the history of science is far too important and too 

 difficult to be treated that way. The very fact that it is not yet stand- 

 ardized as is the case for older disciplines ( say, political and diplomatic 

 history, or Greek literature ) increases its difiiculty. The teacher cannot 

 depend, as many of his colleagues do, on excellent textbooks, each of 

 which is the fruit of a long evolution and of continued selection and 

 correction. 



It is generally understood by the administrators of universities that a 

 professor is expected to give about half of his time to teaching and 

 complementary activities, and the other half to research. In this new 

 field, where so much remains to be done and where the work is often 

 slowed up by the absence or the inadequacy of tools, it would be a good 

 policy to allow more than half the time to research. In any case, 

 research would be a very important part of the man's work. It should 

 be realized that the work done by honest historians is difiBcult and slow;"^^ 



""^ The proof was presented at the Wellesley meeting of the American Mathe- 

 matical Association in the summer of 1944. 



''* Except, of course, when the point was not significant enough to be explained 

 publicly or when it was too technical to be explained in the available time. Queries 

 the scope of which is too narrow are generally answered by me in writing. 



''° This statement may seem commonplace to historians; I am making it here for 

 the scientific readers who appreciate well enough scientific difficulties, but not at 

 all historical ones 



