The most substantive argument against a pass-fail system is that students would stop working hard if they knew no one else could tell whether they had done A or B, or C level work. This concern would be especially great during the new system's implementation. Believing their late-night study orgies finished forever, students would go positively wild--and probably a little bit lazy, too.

The College could remedy this problem at least partly by providing a very credible threat that slackers would be asked to leave. If the "fail" status is made equivalent to a C or C- and the number of "fails" necessary for expulsion is relatively low, fear would keep students working hard.

Even without this threat, students initial euphoria would die with the realization that the change in grading would not effect post-graduation prospects. Employers and graduate schools would continue to compare applicants against each other, so students still would strive to distingish themselves. But instead of receiving a bunch of letter grades for their efforts, they would have a list of extra-curricular and academic distinctions.

The most significant academic distinction of all would be their degree. For if mother Harvard's educational standards are as outstanding as their reputation, her diploma should embody an unwavering confidence in the graduate, making more specific details unnecessary and even inappropriate.

In reality, of course, such details are both helpful and necessary. Students are not all alike, nor are their performances consistent through college. In theory, letter grades should gage differences between them and help them recognize their growth over time.

But as the system has lost its power, other evaluative mechanisms have filled its place. So why spend the time, effort, and money necessary to compute GPAs? Why not rely on the new mechanisms of differentiation that already work so well?

The answer lies in tradition. The abolition of grades would demand a gargantuan mental shift throughout the college. It is highly unlikely that such a radical solution to grade inflation will be accepted in the near future, if ever.

Still, the problem of grade inflation is so confounding that it may prove useful to keep the most basic answer in mind.