Graduate education in the humanities may have its problems, but don't try to tar science with the same brush. In a NY Times Op-Ed, by Dr. Mark Taylor, the chairman of Columbia's religion department, we're told that graduate education in general is in need of a major overhaul.

Graduate programs train students for jobs that most of them won't get:
Most graduate programs in American universities produce a product for which there is no market (candidates for teaching positions that do not exist) and develop skills for which there is diminishing demand (research in subfields within subfields and publication in journals read by no one other than a few like-minded colleagues), all at a rapidly rising cost (sometimes well over $100,000 in student loans).
Graduate programs are a cover for slave labor:
The dirty secret of higher education is that without underpaid graduate students to help in laboratories and with teaching, universities couldn’t conduct research or even instruct their growing undergraduate populations. That’s one of the main reasons we still encourage people to enroll in doctoral programs. It is simply cheaper to provide graduate students with modest stipends and adjuncts with as little as $5,000 a course — with no benefits — than it is to hire full-time professors.
As a cure, we need more interdisciplinary training involving all disciplines:
Abolish permanent departments, even for undergraduate education, and create problem-focused programs. These constantly evolving programs would have sunset clauses, and every seven years each one should be evaluated and either abolished, continued or significantly changed. It is possible to imagine a broad range of topics around which such zones of inquiry could be organized: Mind, Body, Law, Information, Networks, Language, Space, Time, Media, Money, Life and Water.
That last suggestion scares me: I can think of nothing worse than to take a science graduate program, or even a scientific problem, and lard it up with input from the humanities. Don't get me wrong - I love the humanities and arts (my B.A. is in music), and I think they are worthy academic disciplines. But the last thing a physics PhD student needs to do is take a course on "Space" or "Networks" taught by an English professor. The humanities aren't sciences, they don't solve problems like sciences, and they shouldn't try to be sciences. Throwing some engineers, biochemists, historians, and philosophers in a room with the assignment of solving the problem of water or alternative energy is a recipe for a large waste of time.

Collaborations need to be nurtured but not forced - they have to grow naturally. A political scientist may very well want to consult a historian or philosopher, and an economist may want to get together with a psychologist, but that doesn't mean that every discipline has something relevant to say about every issue. I don't need to hear from a religion professor about networks, and the English department doesn't need to hear my opinions on narrative theory or the influence of Wallace Stevens.

When it comes to grad programs, the humanities may want to take a page out of the playbook of the sciences, because many of the problems faced by humanities graduate students are non-issues in the sciences.

1. Graduate students in the sciences aren't "being trained for teaching positions that do not exist." Science grad students actually do very little teaching; they do research, acquiring skills which they can use in a variety of careers. It's true, there are more science grad students interested in university faculty positions than there are positions available; however more and more students are aspiring to and choosing other career paths for which a PhD in the sciences is appropriate.

You can't get around the fact that someone with a PhD in Medieval Studies is going to have more limited job opportunities than someone with a PhD in biochemistry, and thus Medieval Studies programs should limit the number of students they take on. But providing more flexibility for students, as Dr. Taylor suggests, is a good idea - one that is already being implemented in science graduate programs.

Taylor suggests that suggests "the division-of-labor model of separate departments is obsolete and must be replaced with a curriculum structured like a web or complex adaptive network." I don't know what a "complex adaptive network" means when it comes to curriculum design (ironic, given the name of this blog), but again, the sciences are ahead of the curve. Most science graduate programs have now been, for years, extremely flexible about which departments students work in. Student are brought in under an 'umbrella program', and they then get to choose a laboratory home in a variety of departments.

Even undergraduate science programs are getting into the interdisciplinary act.

One could envision something similar in the humanities - new graduate students choose a theme based on their interests, and then, as their ideas for a dissertation shape up,  go work in the English or History or Art departments. Students (who should be helped to come up with realistic career goals) should tailor their programs to match their career aspirations.

2. Overspecialized research is not the problem in the sciences that it is in the humanities. A thesis project "on how the medieval theologian Duns Scotus used citations" may be of interest to very few people. In the sciences, interdisciplinary work (as opposed to specialization) has been the name of the game for some time now.  By the time most grad students finish, they have acquired a broad scientific backrgound, even though, by necessity, their thesis projects are fairly specialized (which is the only way to make progress on a scientific problem - think big, but then get to work on the details).

Here, Dr. Taylor has a worthwhile suggestion, one that, again, is already implemented in the sciences:
Transform the traditional dissertation. In the arts and humanities, where looming cutbacks will be most devastating, there is no longer a market for books modeled on the medieval dissertation, with more footnotes than text. As financial pressures on university presses continue to mount, publication of dissertations, and with it scholarly certification, is almost impossible.
Grad students in the sciences don't write books - they write papers that may or may not be thematically related. All graduate students shouldn't be forced to carry out a book-length treatment of some specialized topic; by writing papers, they have the freedom to shift gears, tackle new problems, and gain breadth as their research interests develop.

The second part of Taylor's suggestion is less plausible:
For many years, I have taught undergraduate courses in which students do not write traditional papers but develop analytic treatments in formats from hypertext and Web sites to films and video games. Graduate students should likewise be encouraged to produce “theses” in alternative formats.
Writing clearly and persuasively is a key skill that should be honed in grad school, so why throw that away by producing these in alternative formats? If the concern is that students graduate without job skills, the response shouldn't be to get rid of one major element of a grad program that develops the most broadly applicable job skills.

3. Science grad students aren't exploited quite so badly as their humanities colleagues. The grad student-slave labor problem is real, but there is an important distinction when it comes to the sciences. Humanities students who have to teach classes in order to get any sort of living stipend are being drawn away from their ultimate goal - a dissertation. Every hour spent teaching or preparing for a class is one hour away from the research needed to graduate.

In science grad programs, students don't get paid to teach - they get paid to work in the lab. The key difference is that the lab work, which grad students are getting paid to do, is in fact the dissertation research necessary to graduate. So while humanities students have to spend much of their time away from their dissertation research in order to earn subsistence wages, science grad students get paid subsistence wages while working on their dissertation research.

In the sciences, that hardly counts as a "dirty secret" - you get paid to work in a lab on your PhD thesis, and you're fortunate to have a faculty advisor who did some heavy lifting to get the lab funded.

Unfortunately, the humanities don't have any hope of getting the kinds of funding that scientists get, so the problem of slave labor in humanities graduate programs is more intractable. Every grad student should be guaranteed at least some time free of teaching to make progress on the dissertation. To make sure there is money for such teaching-free time, departments should make an effort to cut down on the slave labor: it's better to spend the limited money providing a healthy research environment for a smaller pool of students with real career prospects in the field, than to spread the money thin on a large group of graduate students without realistic career prospects, but who can teach for next to nothing.