The excellent Fabio Rojas offers bluntly excellent advice for grad students in the latest Inside Higher Ed. He begins by distinguishing "short-clock" from "long-clock" disciplines:
Some disciplines kick people out in four to six years. I call these
"short clock." Engineering, economics, and biomedical sciences fit this
mold. Others are more extreme. In many cases, you can stay in graduate
school for 10 (!) years and still be considered "fresh." I call these
For short-clock disciplines, Fab's advice is short and sweet:
Short clock disciplines do not expect much from grad students. You
don't need a long list of publications or even a terribly
well-developed paper - because you've only been working on it a year or
so. These disciplines tend to rely heavily, almost exclusively, on
adviser recommendations and Ph.D. program reputation because there is
not much else to go on. The bottom line is that most students who make
it to candidacy will soon be kicked out, whether they like it or not.
So get smart: get an adviser with a good track record and make sure your job market paper is great.
For long-clock disciplines, he's got more to say:
Unlike short clock disciplines, you will not be kicked or nudged out
after X years. You will be allowed to drift indefinitely. If you don't
finish your dissertation, no one will remind you. If you dedicate all
your time to teaching, no one will care. Even if you do finish your
dissertation, people will sit on it for semesters and nothing will
happen. To blunt, the graduate school system is not designed to help
you graduate in a reasonable amount of time. It's designed to waste
The ten million dollar question:
So how on earth do people graduate in departments where no one lifts a finger to help you? A few paths:
Once you get published in real journal, then many faculty will let you
graduate. Why? Publication is often a prerequisite for a job. If you
are published, no one feels bad about letting you go on the job market.
It also shows that you are serious about your career. The higher ranked
the journal, the better.
Demand it: Sometimes you simply have to be pushy. I've seen cases
where a person has published, written their dissertation, and still
nothing happens. You just have to say (politely), "What else can I do
to complete my degree?" If that fails, see the graduate chair or dean.
Be a jerk. If people aren't letting you graduate, they are costing you
money and wasting your time.
Get a job: In some programs, they don't let you graduate until you
get a job. If that's the case, graduation is actually simple. Publish
first (or write a good job market paper in short clock fields). Then go
on the job market. When you get the job offer, you'll see that the
dissertation hearing gets scheduled fairly quickly.
My advice for long-clock disciplines is a little more radical: Stop before you start. Get an econ Ph.D. instead - and call your interests "economics." Odds are, you'll have tenure in econ before you would have found your first job in history.