The mathematician William Feller, celebrated for his work on the normal “bell-shaped” distribution, once encountered a professor who proudly demonstrated the distribution of his grades for the semester. The teacher presented a beautiful instance of a bell curve. Feller laughed. The man had just proven that his students’ knowledge of the subject was quite random, just as if the professor had never given any lectures. The effects of teaching should be to unrandomize learning.

A certain amount of knowledge should be the goal of each course, and those who master it should be rewarded with an A. Another student’s learning beyond the syllabus should not denigrate the achievement of those who fulfilled the contract. Nor should the appraisal of any student’s achievement be elevated by the poor performance of classmates. Such competition is fitting for coal shoveling, not education.

Almost every Princeton student has the intellectual capacity to learn almost any course in the catalog. Barring misenrollment and the occasional instance of obstinacy, the only explanation for any grade less than A is that the professor is such a poor explainer that a bright mind cannot decipher sufficient information from the lectures or that the teacher has drained the educational process of motivation through mind-numbing insipidity.

A grade of A in a class averaging C means that the student is talented enough to learn despite the poor quality of her education. The others are being punished for the professor’s hubris in setting standards for his students beyond what he is talented enough to explain.

Coercing professors into this practice is shameful. 

Daniel I.A. Cohen ’67