Brian Sarnacki has pulled together some of the recent discussion of grad school reform that has been taking place.
I don’t have much to contribute, but I wanted to address one of Mills Kelly’s suggested reforms: decreasing the number of years students commit to finishing a Ph.D. As Kelly notes, according to the AHA, the average Ph.D. takes eight years to complete, and small programs tend to do better at producing Ph.D.s more quickly.
Kelly’s reform, in short, looks this:
Year 1-2
12 credits of course work
6 credits of advanced reading
Qualifying examsYears 3-5
Dissertation research and writing
I think the shorter time frame is a crucial adjustment that graduate programs in history need to consider. One of the keys, as I have argued previously, is for doctoral students to know (generally) what dissertation topic they want to pursue. The M.A. is the time for intellectual exploration; if you are going to study for a Ph.D., then you should enter the program with a clear research agenda. You’ll certainly tweak it as you go through your coursework, but by the time you take your exams, you should be well on your way to researching the dissertation.
Program size, advisor availability, teaching expectations, dissertation topic, family expectations, life in general–all make it difficult to standardize reforms that will satisfy everyone. Nevertheless, I agree with Kelly: graduate programs need to encourage their students to move more quickly toward graduation.
Jacksonian America: Society, Personality, and Politics 
As a PhD candidate, I write as an adjunct instructor, not a professor out in the field. I posted on Brian’s website as well, but, in a nutshell, I think that PhD coursework is essential, and generally seminar-based anyway, as Terry suggests. Professors-in-training need to learn traditional methodology (to get a sense of the context of the field), but also methods training for the digital age. We need to understand the politics of history (how do we avoid another “Enola Gay” debacle, for example). We need pedagogical training.
Perhaps the PhD should no longer be a one-size-fits-all degree. Because PhD graduates enter diverse careers, perhaps the PhD itself can split into 2-3 types of programs, geared for different career paths. If the PhD splits into various types, however, the intention should focus on collaborative authority, not hierarchical.
In the interest of full disclosure, I got an MA in the 1970s but didn’t complete the Ph.D.
What about eliminating the 12 hours of course work, which I assume are upper-level undergrad courses. Instead, have grad students take reading and writing graduate-level seminars. The reading seminars could emphasize historiography and problems, and be geared to the qualifying exams. The writing seminars could advance research on the subject area of the dissertation, and require frequent short writing assignments. The first seminar would be a combination reading-writing seminar, stressing reading good historical literature and the techniques of writing historical prose.
Terry,
The coursework depends on the program. I took one or two undergrad/grad combo courses at the doctoral level, but most of my coursework was graduate only.