Would a C-minus by any other name still smell as rank?

Last week, Kenan Professor of Government Harvey C. Mansfield Jr. '53 announced that students enrolled in his class, Government 1061: "Modern Political Thought," will receive two separate grades. The first will reflect the grade that Mansfield thinks the student deserved, the second is the one that will actually appear on the student's transcript. Ostensibly, this allows Mansfield to maintain his principled stance against what he describes as "Harvard's system of inflated grades" without penalizing students who elect to take Mansfield's class instead of one with a more generous grading curve.

There's something unsettling about this two-grade system. Getting that A-that's-really-a-B is kind of like setting a world record, but having an asterisk next to your name because of some favorable wind condition. In the end, you're never quite sure what the accomplishment really means.

But the real problem with Mansfield's theory is the deeper assumption that Harvard grades--inflated, corrected or otherwise calibrated--are actually accurate measures of a student's academic achievement. They are not, at least not in the way that most of us think they are.

Each class here at Harvard has a grading system. Most will have midterms and finals. Larger lecture classes may also evaluate section participation, problem sets or papers. Students enrolled in seminars will complete research papers and those taking science classes will perform lab work. In a few rare cases, some students (the truly unfortunate) will have to make oral presentations. The conventional view, shared among course instructors, is that such systems will ultimately yield a single grade that reflects a student's performance.

But it isn't quite clear what kind of performance is being measured. One possible response is that grades measure "learning." But learning, in the pure sense, is the act of acquiring knowledge. A humanities student taking an introductory chemistry class might "learn" more than his pre-med peers, but that doesn't necessarily translate into a higher grade. Measuring learning requires taking into account a student's knowledge of the subject material at the end of the term and how much it differs from that student's knowledge at the beginning of the term.

In this light, it might be safe to say that most classes don't seem to care how much students learn as much as what it is that students know. But what kind of knowledge do grades measure? A multiple-choice test in macroeconomics, for example, might measure how well a student can apply a particular set of theoretical concepts to solve problems. But a research paper in that same class might focus on how well a student can make logical, reasoned arguments about economic policy. And section participation is purportedly a test of critical reading and thinking skills.

Doing well in any of these requirements depends heavily on knowledge of the course material. But there is also a considerable degree to which each of these requirements depends heavily on factors outside the course material. One example is writing ability--a weak writer with a strong command of the course material may often receive a lower grade on an essay assignment, compared to a stronger writer with a weaker command of the material. Identification (ID) questions (particularly when the ID's themselves are distributed ahead of time) test sheer memorization more than they evaluate conceptual understanding.

Harvard's grading system is replete with these types of inconsistencies and imperfections. A final course grade, particularly in classes with a single final exam or paper, is by no means a perfect depiction of knowledge or learning. On the other hand, grades aren't a completely inaccurate picture, either.

The point of all this is not to disparage students with high academic records, or even to suggest that traditional course grading schemes should be abandoned. Rather, it is a plea to be honest with ourselves about what grades actually mean. If Mansfield wants to give his students two grades, so be it. But there is nothing about Mansfield's non-inflated grades that make them any more accurate as gauges of learning or knowledge. Saying otherwise is simply misleading.

Faculty and students alike should be more willing to acknowledge the extent to which grades are tangential to course material. This would do much to dispel Harvard's grade obsessive culture and shift emphasis toward the pursuit of learning for its own sake.

Richard S. Lee '01 is a social studies concentrator in Pforzheimer House. His column appears on alternate Thursdays.