(c) by CLARENCE BROWN

A few weeks ago a student in one of my courses at Princeton got an F as his final grade, and, owing to this failure, did not march with his classmates to receive the diploma. This makes me feel bad, even though I no more "gave" the grade than a doctor who watches the wavy line on the oscilloscope go flat kills the patient. Still, I hate failure and insist on my right to participate in the sorrow that failure brings.

I took great comfort, however, in a fact that I learned incidentally in connection with this case. The student was merely one of some thirty-five students who did not march at Commencement because of academic failure.

Why is that comforting? It is a great solace to know that an institution's backbone is on the mend. It is a comfort to know that the official transcript, the historical record of four years of academic performance, is edging toward historical truth. It is by no means there yet, but it is on the move.

And recent news stories suggest it is on the move elsewhere as well. Stanford has restored the F, abolished a score of years ago.

Thanks to what is badly misnamed "grade inflation," the transcript produced by most major universities in this country has been as genuine as a field-grade commission in the Army of the Confederacy, which you can buy at roadside novelty stands.

The economic metaphor represented by "inflation" suggests that grades are a kind of currency. They are not. Grades are, or should be, a tiny language, a symbolic system or code. The elements are very few--usually only A, B, C, D, and F--and they refer to another system, that of values, that is cloudy and relative. A+ means "the very best" and F means "the very worst," and the others mean stages in between. Vague this other system may be, but we all understand it and live by it.

Grade inflation, however, has robbed the grading system of all right to the word "system." When the grades have been reduced to A and A-, one has not a system but a binary code.

Though that, too, is wrong. A binary code (like the one on which this computer is running) is structured on the model of Yes/No or On/Off.

The reason why A/A- is not a binary code is that no one knows what A- means. No means no and off means off, but A- means the whole range from "pretty good" to "total failure."

The grading charade will be put right eventually, but I have a plan to speed up the process. My plan calls on the end user of the transcript, the employer, to rebel against the fraud that has been practiced against him for these last couple of decades. The employer, if he produces a product for sale, is constrained by the "truth in labelling" laws. Why has he meekly submitted all this time to accepting as the truth an official record that is a web of outrageous falsehoods?

My plan calls on the CEOs and the boards of GE, Seagrams, the Morgan Bank, Disney, and others to announce to the universities that they will get no more grants until they somehow find the courage to admit that not every graduate is supremely gifted. A few are very good, a few are very bad, and the great majority fall at various points in between.

In their own research, scholars spend their lives searching for the truth and discriminating small differences. Until they begin to do so in reporting on their own students, they will have to go elsewhere for their donations. Companies ought to be fed up with the A+ graduate who thinks that 7 is the sum of 3 and 2 and that spelling is an opportunity for personal expression.