Tuesday, September 30, 2014

Against Concentrations

I have been puzzling about how Harvard could get resolve the big problems in undergraduate education. I have a modest proposal.

Of course, the first step is to define what problem, or problems, you are trying to solve. Here are a few.

1) Students are too grade-conscious. This drives them to make anti-educational decisions, taking courses that are graded softly or don't teach them much they don't already know, rather than taking courses they might learn from.

2) Honors, being based largely on grades, reward conservative course choices and punish risk taking. They are also enormously socioeconomically biased, since the people who get good grades on average over four years are the people who were already very well educated before they arrived at Harvard. (I have never seen data on that, but I would love to see a scatterplot of high and highest honors, or GPA, against family income. I'd guess that most of the outliers, the students with low incomes and high GPAs, came from independent schools with aggressive financial aid policies.)

3) Declaring a concentration is seen as a moment of truth, a crisis in identity formation, no matter how much we preach that it's what you know rather than what credential you bear that will determine your future success.

4) The Humanities are terrified that they are not attracting enough enrollments or enough concentrators.

5) The General Education program, a source of pride in theory, is treated with some derision by more talented students, and as game of hopscotch, with some excellent courses and a lot of strange ones, by the bulk of students.

6) We are losing our identity as a liberal arts institution because so many students are concentrating in Engineering and Computer Science.

Pardon a somewhat lengthy digression here. I think #6, though widely bruited, is nonsense, because as far as I am concerned applied science is perfectly consistent with liberal education. The notion that a liberal education is one divorced from utility -- learning for learning's sake, as we like to say -- is a crock. It is just an etymological confusion, where "art" is construed as something other than "science" and "liberal" is taken to mean free from utility.

The liberal arts are neither useless nor artistic. They are called "liberal" because they were the arts taught to the free people of the Roman republic, the citizens as opposed to slaves. They were the arts of citizenship. And of course they included mathematics, since that was among the skills, the useful arts, that Roman citizens were expected to have to take responsibility for their nation. And while I am at it, and while we are still buzzing about the naming of the School of Public Health in honor of the Chans, I have always wondered why the School of Engineering is not already named for Gordon McKay and described properly as a "School of Art." After all, McKay's will is quite explicit:

I am sure there is a good reason; I haven't fought my way through all the codicils. The point is, the campaigning for the Liberal Arts is not just about the Humanities. It is equally about techne.

But back to educational reform. Here is a proposal to deal with all these issues.

I hate "secondaries," what Harvard calls minors. As presently constituted, hey seem to bring out the worst inclinations of Harvard students, their tendency to seek credentials rather than learning. I think students would do less of that if left to their own devices -- but we actually incentivize them to seek secondaries and guts that will raise their GPAs rather than studying the things they would most like to learn. We do this by giving them gold stars for secondaries and giving them honors for GPA, at the same time as we disdain the inevitable consequences of rewarding these things and not others -- students too busy and too competitive to stop and think about who they are and where they are going.

But what if there were ONLY secondaries? Get rid of concentrations. Have departments, and interdepartmental committees, offer "secondaries," and require students to earn at least two, but allow students to earn several. (Of course "secondary" is no longer the right term if there are no concentrations. I'll use it just to convey the idea of a small cluster of courses with some disciplinary coherence and a bit of depth.)

Now that defining moment -- am I going to be an English major or a Computer Science major? -- disappears. You have to be more than one thing, or you don't get a degree.

Well, of course there are damned good reasons why we wouldn't simply do this. We think you are not well educated if you don't know something in depth. We want to produce top scholars too, and that required incentivizing, if not requiring, undergraduates to work at an advanced scholarly level.

Though frankly, we are pretty unclear about the purpose of concentrations. We react with mock horror when students overspecialize, even if they have satisfied all their Gen Ed requirements. We actually reduced the maximum size of concentrations a few years ago; they are not supposed to be mini-PhDs, was the battle cry. But of course the top students are always the ones who do the most advanced disciplinary work. Those are the ones who graduate summa.

So how do we incentivize a deeper education, and the engagement of students in advanced scholarship and research, while not requiring every graduate to have a concentration?

Well, first of all, having two or three secondaries, say in CS and biochemistry, might be more of an intellectual investment in the future than having a concentration in one or the other. Lots of fields are evolving out of the friction between existing disciplines. A few courses in each of CS and sociology might have been perfect for Zuckerberg.

And secondly, we could incentivize depth by basing honors mostly on senior theses. A department might well decide that the "secondary," which would require only a few courses, was not nearly preparation enough to write a thesis; to be eligible for that you would need several more courses, with some specificity about which courses. When it came time to award honors a department or committee could look at the entire program and give some weight to GPA, but the idea would be to judge people on the depth of their knowledge, as evidenced in part by the quality of the thesis they were able to write. Honors would become more subjective, but also less consequential.

I'll bet we'd get more humanities enrollments if people felt that they did not have to assume the identity of a humanist. Of course, the humanities courses would still have to win in the marketplace of ideas. But there would be less eye-rolling, by parents and peers, when a student announced that one of her secondaries was in English if another were in something more "useful." We could credibly argue, as we cannot today, that the ideal graduate really is broadly educated. (The CS concentrator who does as instructed, and takes one Aesthetic and Interpretive Understanding course, one Culture and Belief Course, and one Ethical Reasoning course, may well not wind up with a credibly liberal education, but if she had a secondary in Philosophy or the Classics, I bet she would.)

And by limiting the size of secondaries, we would make more credible that what makes you liberally educated is not what the scholars in your department think of your academic expertise, it is your facility with thinking and reasoning and analysis and argumentation and presenting well considered solutions to actual problems of the world in all their complexity. Of course those would be rooted in the academic traditions of the disciplines. Educated people are expected to bring the full arsenal of their learning to bear on their problems they confront. Our education would get linked back to our civic purposes, without some childish required civics curriculum.

Now I'd love to know what's wrong with this idea. One thing that is possibly "wrong" with it is that it would end some of the risible cries of "turf warfare." In the politics of higher education, turf warfare is advantageous to those losing the battles, as it is a way to call for the battleground to be leveled artificially, for a shift in resources to make it a fair fight. If we simply acknowledged that the curriculum is largely elective and let students take the subjects in which they were interested -- within the parameters of the regulated marketplace I have argued for -- we would all have a greater incentive to meet students' needs rather than complain about the unfairness of the fight in which we are engaged.

Monday, September 29, 2014

Is this what they mean by "running a university like a business"?

When a university is "run like a business," the implication is that wasteful redundancy and costly inefficiency should be eliminated. But of course determining whether you are being efficient about the way you are running your business, or cutting the soul out of the enterprise, requires defining what business the university is in. My previous post about the marketplace of ideas notwithstanding, I don't want a system that eliminates Egyptology so Computer Science can grow. In fact, Egyptology is holding its own in the marketplace of ideas at Harvard; Peter der Manuelian teaches a cool Gen Ed course to captivated students, including some I have directed to the course. (Interesting fact: der Manuelian took an intro CS course from me while he was an undergraduate here. You can do a lot of things with a little CS, as I keep saying.)

The question of the day is, given that executive compensation in the private sector has grown out of proportion to the wages of the labor force, and (as Paul Krugman details today) the disproportion at the upper end of the executive hierarchy has grown to a degree few of us can imagine, should the same hold for non-profits? After all, the same forces that justify paying big salaries to the op executives of for-profit companies -- competitive pressures, the value returned by top talent, etc. -- justify paying them in charitable non-profits. Don't they?

There is something that doesn't feel right about that, given the commitment these non-profits make to serve the public good -- and to have taxpayers underwrite their operations, if in no other way, by exempting them from taxation.

Which brings me to the article I received today that got me thinking about all this -- about not Harvard, but the University of Chicago. It's written by David Mihalyfy of the Harvard class of 2002, now a graduate student at Chicago. It documents the amazing run up in compensation for several top administrators at that university.
New analysis of tax data from publicly available IRS 990 forms shows that eight high-level UChicago administrators have received more than $7.6 million in compensation increases since 2007-2008, even as the school moved toward and suffered a credit downgrade.
Over five years, administrators enjoyed pay increases of between 40 percent and 135 percent, and as a result each received $450,000 to $3.3 million from cumulative increases by the end of 2012-2013, the most recent year for which tax data is available.
UChicago thus ended up paying $2.5 million more annually for the combined services of these eight people — an increase from $3.4 to almost $6 million per year.
And we wonder why the public is skeptical that we have done everything we can to hold down the cost of our product.

Bonus links:

A fascinating Bloomberg piece on the history of fundraising for endowment, an art first developed into a science by none other than Harvard's greatest president, Charles William Eliot.

A lovely Bryan Marquard obituary for one of the all-time great marketers of ideas, Irven DeVore. And a terrific David Warsh column giving a nod to the marketing skills of David Malan, and pitching an important idea about computer science education.

I will have to return to say more about Homi Bhabha's citation of Deresiewicz's epistemological dogma ("a subjective perspective," -- "subjectivism should not always be confused with solipsism"):
“We ask of a scientific proposition, ‘Is it true?,’ but of a proposition in the humanities we ask, ‘Is it true for me?’ ”
(Deresiewicz actually attributes this formulation to Lionel Trilling.) It does not take the "willful ignorance" that Bhabha attributes to me to wonder what this means or where it will take us.

Here, by the way, is the full video of Deresiewicz's appearance at Harvard, complete with Bhabha's introduction.

Thursday, September 25, 2014

You're kidding, right? Carlyle tracked its LPs at their annual meeting?

The huge global asset management firm The Carlyle Group, at its annual meeting, issued ID badges to its limited partners--its investors in other words, the people who make the business happen--so it could track their comings and goings at the meeting. These are not employees who imaginably signed away all their privacy rights when they agreed to work for the firm. As the story explains,
Carlyle Group has more than 1,650 investors from 78 countries, according to its website. And hundreds of those investors show up at Carlyle’s annual meeting, which according to one LP is something like the Burning Man festival of the private equity industry.
The LPs who learned about it -- there was apparently no advance notice given -- were furious. One guy thought about giving his badge to a woman LP for the night so whoever was sleuthing their movements would have some fun analyzing the data.

After the infamous email privacy scandal at Harvard a couple of years ago I should not be surprised by anything. How could anyone at Carlyle have thought this was a bright idea? I get it at one level -- some wizard realizes something is possible, nobody ever told him anything about not doing stuff like this, and he might learn something, so why not?

But it's bizarre. It demonstrates a shocking myopia about the big picture. What information did they hope to get that would be more important than the outrage and mistrust a disclosure would precipitate? If your LPs don't trust you in this business, you are cooked. Why would anyone trust Carlyle with anything now?

Wednesday, September 24, 2014

The (Regulated) Marketplace of Ideas

CS50, Harvard's introductory computer science course, is now the largest undergraduate course at Harvard. That is, it's the largest course in undergraduate enrollments, not even counting its enrollment from the professional schools, nor its Extension School and HarvardX cousins. That news was picked up by Fortune and by Business Insider. When I comment, I try to fight back against the lazy angle that Harvard students are just seeing CS as a way to make a quick buck--because for most students, I don't think that's the rationale at all. It's a very well-taught, fun, and empowering course. And as I said, "[Harvard students] have figured out that in pretty much every area of study, computational methods and computational thinking are going to be important to the future." As I told Business Insider,  the "course enrolls students from other disciplines who realize that computational thinking and skills are valuable in their own discipline, whether that’s economics or biochemistry or music or even the Classics." (I went on to plug the release of the Loeb Classical Library in an online edition, but somehow Business Insider didn't print that part.)

As I told Business Insider for a different story, "most of the people who are majors are converts from other fields, people who are switching over from all disciplines. This course is really kind of a conversion experience for a lot of people." I used the same metaphor in the long piece the Crimson published about CS50: “We are evangelical about our subject. … We want to compete for students. We want to take all the students who thought they couldn’t do computer science. We want them to understand that it’s going to going to be hard and fun. .... We’ve been doing the shenanigans for years.” (See this earlier blog post for the Confi Guide description of an ur-CS50 course I taught in the 1970s, if you think that David Malan invented clowning in CS at Harvard. And I didn't invent it either, I just picked up where Bill Bossert and Chuck Prenner left off.)

Departmental cultures run very deep, I have learned over the years. The idea that professors are supposed to be evangelical about their subjects is not universal. In some departments the attitude seems to be "We will teach, however reluctantly, whoever shows up, and complain about it if they are not intellectual enough"; in others it is, "I want to teach the students I want to teach; why should I teach students who don't want to be in my classes?" I wish I knew how many of the course "lotteries" (in scare quotes because they are often not based on random selection) are secondary admissions processes driven by faculty desires to select among the select. It is as though the limited concentrations, which were officially banned pursuant to Paul Martin's report on concentrations around the time the Core Curriculum came into being, are being recreated piecemeal as a nexus of limited-enrollment courses. Enrollment limits may be inevitable in some cases, but they are far too prevalent now, with no rules of which I am aware about when they are appropriate. Surely the most bizarre lottery story I have heard is that for Humanities 10, a brand new gateway course designed to draw students to Humanities concentrations and reverse the gradual decline in Humanities enrollments. It sounds like an absolutely wonderful course, vaguely resonant with the first term of the old Hum 5 course I took in the mid-1960s. Humanities 10 drew several hundred students--but was capped at 75. I wonder how many of those turned away took Ec 10 or CS50 instead, and will never return to the humanities except to satisfy their unattractively named "Aesthetic and Interpretive Understanding" requirement?

I think departmental cultures are actually not Harvard cultures so much as cultures of the various academic guilds. That would make sense--economists draw their norms from economists elsewhere, for example, not from the norms in the English department. One of the consequences of the unwinding of the faculty community at Harvard--I am old enough to remember when lots of us went to the Faculty Club to drink before heading home, something unthinkable today--is that we learn less from each other about how to behave. We learn from our departmental colleagues, and that's it. Given the self-perpetuating nature of departments, and the way junior faculty are groomed within the department before the department awards them tenure, these norms become very ingrown.

The variation in norms really hit home when I attended the Mahindra Humanities Center's event surrounding the publication of William Deresiewicz's book Excellent Sheep. I had published a dialog with Deresiewicz in the Chronicle of Higher Education in August. I was startled when Homi Bhabha, head of the Center, opened his introduction of Deresiewicz by quoting from that dialog. He first claimed that the unnamed author had drawn a cordon sanitaire around his own discipline and cast the social sciences to the wind (or words to that effect). He then began to quote me, identifying me only as an "amiable colleague":
It seems to me that the process of promotion and tenure has had a particularly noxious effect on the humanities. We used to count on the humanities faculty to open students’ eyes to what it means to be human. Now that is not why humanities professors are hired, incentivized, or promoted. Their social conscience, when they feel called to exercise it, is manifested mostly in normative political posturing that is divisive and chilling to discourse on campus, and of no great civic, educational, or maturational value to students. Isn’t the so-called humanities crisis, the declining numbers of students choosing to study the humanities even at the top institutions, really part of the picture you paint in your book—of institutions that provide lots of freedom, and lots of busyness, but little support for self-understanding? 
"With friends like these," Bhabha sonorously rumbled, "who needs enemies?"

Except that Bhabha quoted only the italicized sentence, not the surround, the argument that the tenure process, desiccating everywhere, has bled the soul out of the part of the institution where we should most hope to find it.

I made a similar argument in Excellence Without a Soul, but it seems to have stung more here, perhaps because the humanities faculty feel so much more beleaguered now. In that book I wrote "We have forgotten that we teach the humanities to help students understand what it means to be human," and I quoted a humanities editor as saying "The demands of productivity are leading to the production of much more nonsense." The reference to politics as a proxy for the more traditionally humanistic values was written at a moment when a leading question among humanities scholars was whether their conferences should follow the lead of the American Studies Association in boycotting Israel. Is that really an instructive issue for those ruing the decline in interest in the humanities among undergraduates? 

By contrast, when my colleague Fawwaz Habbal, tried to explain that engineers deal with questions of value all the time--he gave the example of a course in which students have to "solve" the problem of Fukushima, including deciding on a triage protocol when only a limited number of people can be evacuated--Bhabha seemed to draw his own cordon sanitaire, suggesting that the engineers would build the tools but only the humanists dealt with questions of value.

What was amusing about the projectile Bhabha launched in my direction (I am sure he did not realize I was there) was that a few minutes later Deresiewicz accurately quoted me as arguing that not all subjects were equally important to college education, adding, correctly, that I thought that the humanities were more important than Computer Science. The humanities, as he and I understand them, not necessarily the weird humanities courses constructed for the Gen Ed curriculum.

Though I appreciated Deresiewicz quoting me accurately there, I found the whole event pretty disappointing, as a showcase for the humanities or even for the kind of "critical thinking" the humanists keep telling us their teaching trains students to do. When challenged about his habit of sweeping overgeneralization, Deresiewicz had three kinds of responses.

  • Yes, there are exceptions, the students are not all excellent sheep. But it's true generally.
  • Other people have said the same thing, so it must be true.
  • After I said it, a lot of people wrote me to say I was right, or gave me other examples to support my argument.
There is a name for the fallacy behind the last one: Confirmation bias. You say something, a lot of people agree with you, some people disagree with you, you hear more of what the first group has to say, and you conclude that you were right.

Overall, I just don't think Deresiewicz is a nuanced thinker. He seems not to recognize that he has serious Oedipal issues--his book's way-too-much-information confessional about breaking away from his father was embarrassing to read. His Freudian conflicts have now turned patricidal. He is striking a dagger into the heart of the hated academy that gave him breath. The news one learns from the book that his father was a professor just makes the the transference more obvious.

To loop back to where we started. The Harvard curriculum is largely elective. It is not quite elective enough, in my opinion, because the lines between Gen Ed categories were drawn too finely, as a compromise in a complex turf war among disciplines. So some good courses that are "general education" by any normal person's standard get rejected as Gen Ed courses, and some Gen Ed courses are on topics of dubious general value even though they fit the Gen Ed guidelines. Still, students generally have a lot of choice of what courses to take, and they are not eager to make those choices. So they wind up looking for metrics--easiness, grading softness, time of day--that make choices possible with little thought about the underlying educational values.

One of the marvels of CS50 is that nobody would choose it by easiness. Most students choose it because it is fun and it is empowering and it is useful (not as a way to get a job, but to make a better Loeb Library, etc.). But what information do we generally offer students about exercising their freedom of choice? The Q (student evaluation) guide, which is all but worthless as a course selection tool.

So I ask myself: Why shouldn't we have an actual free market competition for enrollments, with some rewards to departments for building their enrollments? Well, one good reason is that faculty would be incentivized to use dishonorable means to build enrollments -- grade inflation, light workload, etc. Well, guess what -- that happens already. A certain ethical reasoning course has gotten extraordinarily popular lately, with everyone from  overstressed seniors writing theses to members of our athletic teams. 

So I don't believe in a totally free market -- the controls on how we teach and how we grade should actually be stronger than they are. But a regulated, competitive market, with rewards for gaining market share, would improve teaching across the board. If the workload and grading were regulated, why shouldn't there be market incentives to teach courses that inspire students, and to lift the enrollment caps on courses like Humanities 10? It would be a way to break through the logjam that my amiable colleague Professor Bhabha blasted me for pointing out, that the nature of humanistic scholarship these days makes teaching such courses very difficult and unrewarding, even when they the faculty would love to carry their enlightenment to a broad swath of undergraduates. The folks who do it, and they certainly exist, are heroes. Helen Vendler. Greg Nagy. Michael Sandel. There are others, too. There is a reason why students adore them, and it's not because their courses are guts. It's because they speak to students' souls.

Thursday, September 4, 2014

Upcoming Talks

I'll be speaking on Privacy at the University of Rhode Island on Tuesday, September 9, and about anonymity at Cornell on Wednesday, September 17.  Hope to see some friends at both places -- looks like great lineups for both conferences.