Making the Grade
Nearly half of all grades at Harvard University last year were A's and A-minuses, we learned last week. America's most venerable institution of higher education is also the national capital of grade inflation. If you attend small humanities classes at Harvard, your chances of coming away with an A or A-minus are nearly two in three. Ninety-one per cent of Harvard students graduated with honors (summa, magna or cum laude) this June. At Yale it was only 51 per cent, and at Princeton only 44. All this, according to a report issued last week by the university itself. (Herself? I've never been clear about the gender of the noun "university" — though I am clear that if I say "gender" instead of "sex" when speaking of human beings, I shall be carpet-bombed with emails from angry conservatives accusing me of selling out to the language police.)
Whenever this topic comes to my attention, I find myself thinking of wooden mushrooms. By way of explaining this, permit me to take you on a wee trip down Derb Memory Lane.
I attended University College, London, the oldest-established, and in several major disciplines the most prestigious, of London University's fifty-odd colleges. U.C.L. was founded in 1826, at a time when the Anglican church was felt to have too much of a stranglehold on higher education in England — which is to say, on Oxford and Cambridge, the only English universities then existing. The anti-Anglican spirit of the founding was captured in a phrase of the time: U.C.L., it was said, was an establishment for "Jews and Welshmen." Its most notable founder was in fact the great eccentric and Utilitarian philosopher Jeremy Bentham, whose preserved head was, and still is, in accordance with the terms of his will, kept in a box over one of the interior doors.
I went to U.C.L. to study mathematics, with which I have been having a sort of unrequited love affair on and off since childhood. I mean, I love math, but it doesn't love me — I am not actually much good at it. U.C.L. had a stiff math program. There was no nonsense about majors or minors: we did three years of undiluted math. "Elective" meant that in the third year you were permitted to choose whether you wanted to take extra courses in Functional Analysis, Celestial Mechanics, Mathematical Logic or Algebraic Topology. The only other thing we were permitted, in fact required, to study was German, the second language of math. (Though Germans will tell you it's the first. At enrolment a few of us smugly announced that we had already learned German at school, and had exam passes to prove it. Unimpressed, the department said it would be unfair if our classmates had to study for a language requirement but we didn't, and shipped us off to a Friday-afternoon Russian-for-dummies class, Russian being the third language of math. The class was held at the nearby School of Slavonic and East European Studies, where I briefly dated the entire third-year Hungarian department. Nice girl.)
So there we were, forty-odd students, thinning out to thirty-odd by course end, being flogged through higher mathematics by some quite distinguished personalities — and, of course, some much less distinguished research assistants on starvation pay. The grading system was mathematically elegant in its simplicity. At the end of the second year you took an exam. At the end of the third year you took another exam. Based on these two exams, you were awarded a degree. The classes of degree awarded were as follows: first, upper second, lower second, third and "general." (The "general" meant that you had survived the three years without dropping out, shown up at the examination hall, and written your name on the exam paper.) In my graduating class of thirty-odd, there were only three firsts, every one an outstanding mathematician. One of them was well-known for never taking notes. I used to watch him in lectures. Most of the time he seemed to be looking out the window. At other times, I thought he was sleeping. One of the others was close to being mad. He used to eat raw onions — just bite into one, as if it were an apple. He had some bizarre theory about the nutritive powers of onions.
I myself got a third. Partly this was just not being very good at math, but I can't pretend that was the whole story. I know, looking back, that if I had truly busted my hump, I could have got a second for sure, perhaps an upper second. The main thing that got in the way was those wooden mushrooms.
See, the student union lounge at U.C.L. had games tables. One of them was a pool-type game in which, instead of having pockets in the corners of the table, there were ball-sized holes actually in the surface of the table itself. Each of the high-scoring holes was guarded by a wooden mushroom that stood in front of it. The only way to get a ball into one of these holes was to play it off the back and side cushions. If you did this, you got the big points. If you knocked down a mushroom you got no points, and lost your break score. If you knocked down the red mushroom — which, of course, guarded the highest-value hole — you lost your entire game score.
For some reason this stupid game took a grip on me in my third year. With a classmate, another ne'er-do-well character like myself, I played the mushroom game all day and every day. I had done quite well in my second-year exam, the equivalent of a borderline upper-second, but my third year was wiped out by that damn game. When, that June, I sat down in the exam hall and opened the final paper, I was dismayed to find that it contained no questions at all about wooden mushrooms, only a lot of incomprehensible stuff about Banach spaces, homology functors and stress-energy tensors. (Huckleberry Finn, my playing companion, did even worse than me, and ended up with a "general" degree. Shrugging it off with fine aplomb, he became a folk singer.)
Apart from the three guys who got firsts and a couple who were awarded the despised "general," the class was pretty evenly divided between seconds and thirds, with half a dozen upper seconds. There were no surprises. We'd all been going to class together for three years and knew each other's abilities pretty well. The guys who got firsts deserved them. I deserved my third, and my pal deserved his "general." I didn't hear anyone complaining.
I realize, of course, that this experience can't be translated to Harvard. Too many things are different. Our system — forty of us all together in nearly all our classes for three years — is not followed at American colleges (nor at U.C.L. either, nowadays, I'm told). My education was state-funded, while the parents of Harvard students are paying truckloads of money for their kids to attend the place, and will be angry if there is no visible return on their investment. Mathematics is a subject in which it is easy to discern who is, and who isn't, much good. You set a problem; Freddy First solves it by an elegant and brilliant method even you yourself hadn't thought of; Suzie Second, after a couple of false starts, solves it just as you intended it to be solved; Theodore Third gets half-way to a solution after five pages of floundering, then gives up. Excellence is much harder to judge in less crunchy disciplines, I understand that.
And there are all those other pressures, of course. Note that I have been referring to our firsts as "guys." We did have a sprinkling of girls, about five as I recall, but none got a first. There have been some fine women mathematicians — we had one on the faculty — but they are awfully rare. To say this, or even just to declare it implicitly by the way you give grades, is of course rank heresy in the politically correct world of today's academy, and is a sure path to a major lawsuit and a world of hurt. Better just to give the whole top half of the class an A grade. A fortiori with race: our class had two Chinese, an Indian and a Burmese, but no blacks at all. If a situation like that occurred at Harvard in 2001, it would force the resignation — if not the ritual seppuku — of the school's entire administration, a clamorous national scandal, and cases before the Supreme Court.*
There is another factor, though. Last week's report from Harvard notes that the higher grades may also be deserved, as students work harder and are better prepared. When I read that I laughed — talk about excuses! On reflection, though, I think I see their point. My occasional contacts with people out of good American schools the past few years suggest to me that they do indeed work very hard, much harder than I and my classmates were expected to. Even allowing for the distortions of "affirmative action," the ethos of American higher education is now firmly, in fact intensively, meritocratic. The old idea of a university was that it should be, as well as a center of scholarship, an agreeable place for well-heeled young men to fritter away three or four years under modest supervision, emerging with the famous "gentleman's C." This notion survived into the 1970s, with enough potency to infect even working-class kids like myself and my pal, who should have had more sense. It seems to me that notion is now perfectly dead. I can well imagine that older faculty members, impressed with the diligence of their students by comparison with what they remember of their own time at college (supposing my impressions are correct), might be inclined to award up.
You could argue that, even if this is true, it doesn't justify giving A grades to half your students — that grading should give students some idea of how they rank among their peers, not how they compare with their fathers' generation. That sounds right to me. Attempts to measure educational attainment, or any other kind of mental ability, across generations turn up some very knotty conundrums, like the famous Flynn Effect. On a scholarly email list I belong to there is a discussion in full flow right now about whether people learn more at school today than they did in the past. Accredited experts — people who are paid a salary to make intensive studies of these things — disagree quite bitterly about the answer.
The Derbyshire system for college grading, which I believe would deliver as much as can reasonably be expected of a grading system, would be:
- Rank a student among his classmates, not according to some abstract or historic standard.
- If it can be fairly done (in higher education, I don't think it very often can), give a supplementary rank showing how the student places among students of the same subject nationwide.
- Identify the five per cent or so of truly brilliant students.
- Identify the five per cent or so of slackers and no-hopers.
- Find some fair and sensible way to put the other ninety per cent into three or four categories by ability and effort: upper second, lower second, third; or A minus, B plus, B, B minus.
Such a straightforward system would, of course, be revolutionary, and very dangerous, in the modern academy.
* The American Mathematical Society reported in February that "U.S. new doctoral recipients 1999-2000" in the mathematical sciences break down as: 14 black males, 803 other males, 6 black females, 296 other females. Just under half — 537 out of 1,119 — of these math Ph.D.s were U.S. citizens. (Footnote to the footnote: I originally wrote that sentence as "Just over half…" and only noticed my mistake a nanosecond before hitting the "send" button to dispatch this piece to the noble webmaster. No wonder I got a third.)