Hmm, I think the problem is slightly different. The job of a tenured professor in 1980 was pretty sweet. Upper middle class salary, great job security, a soap

Hmm, I think the problem is slightly different. The job of a tenured professor in 1980 was pretty sweet. Upper middle class salary, great job security, a soap box from which to influence impressionable minds, freedom to pursue intellectual interests, etc. Naturally, as with any career track that looks pretty sweet, people rush to enter the field. Thus the great numbers of grad students. There are basically three ways a guild profession can react to an influx of new demand: a) lower the salary and benefits of the master guildsmen to make the profession less desirable and to allow more overall hiring, until supply and demand balance out. b) make the entrance requirements more and more onerous in order to restrict entry c) use a lottery system. Option a) is out of the question because it hurts the powerful senior professors who run the existing system. Option b) is the path the universities took. Earning tenture requires more years in grad school, more years as a junior professor, and more risk of never reaching the goal. And the actual research requires mastering extremely esoteric readings and/or mathematical techniques that have virtually no use for actually doing productive research. But requiring mastery of these techniques simply became the default filter. When you have too many grad students, the default is to make the tests harder until you have sufficiently reduced the number who can pass. They then become the new professors, and thus have an inordinate love and respect for the esoteric test material that allowed them to get their current job. The result is the death spiral you describe - burnt out professors who are out of touch with reality. Option c) a lottery, is the route academia probably should have taken. Set some sort of bar for admittance into grad school - great GRE's, great grades, and great recommendations, and then simply lottery off the spots. Keep the spots few enough so that there is a clear path to a tenured professorship. But the real underlying problem is that academia is not accountable. Academia is suffering from the problems that plague any self-selected priesthood - over time its practices grow more esoteric and out of touch with reality. There are a few good departments out there, but most departments neither do good teaching nor good research. Unfortunately, the body that has the lawful right to regulate academy (Congress and State legislatures) are too dumb and too fickle to do the job well. What "reform" happens takes the form of temper tantrums that do little to fix the system. In the end, I think fixing academia is thus a "regime change complete" problem - ie. not possible in the American political system as it currently exists.