While reading Michael Berube's most recent post about transforming MLA interviews into Skype interviews, I had another idea (which I do hope I'm not unwittingly echoing--I haven't seen it suggested on any of the various threads I've been following):
Make doctoral-granting institutions commit to paying the full cost of each student's job search. For as long as the job search takes. Heck, the MLA could even require that any such institution that wants to advertise through the MLA job list, let alone keep up MLA interviews themselves, must have such a system in place. (The university is paying to send the faculty there? Then they can afford to send the students there.)
"But wait!" I can hear you wail. "That's expensive! What about all of our students who are on the market for four, five, six years--"
Er, well. If your students are on the market for that long and cannot find TT work, then perhaps your desire to train graduate students has no relationship to their employment opportunities?* And then, perhaps, you might consider training many fewer graduate students? (Yes, that might require you to spend more time teaching--gulp--undergraduates. Your life will not end.) Or, alternately: you might consider committing seriously to other forms of training, so that students are better prepared to take their skills to other professions?**
Now, there are potentially serious objections to this plan--the most serious being that cash-strapped regional campuses, possibly be the only avenue for higher ed available to many graduates, will have difficulties. I would imagine that in practice, as opposed to in theory, you would need to scale contributions in proportion to the university's finances. But make everyone contribute something. And, just as pragmatically, enforcement. The MLA can't reach into any given program's pockets, but it can certainly regulate which departments are allowed to post or interview where. Then there are the other big professional associatons with whom one might network (e.g., the AHA), the AAUP, and the accrediting bodies, which, as a group, would be able to exert more pressure.
*--I should note here that some programs are very successful in placing students at campuses that normally don't hire at the MLA, such as community colleges. I'm also well aware of Marc Bousquet's argument that we have underemployment instead of overproduction, but until we have a mass groundswell of governmental support for university funding, in practice we pretty much do have the latter instead of the former.
**--I'm on the record as saying that you usually don't need a Ph.D. for non-academic work, which makes me suspicious of the "let's just include more professional tracks so that we can keep all our graduate students!" approach, but if we must do this, then let's do it well.