Available here. Attrition appears to be much lower than when I was at Michigan in the late 80s and early 90s. In my class of nine (seven men, two women), the two women did not complete the program, but all seven men did eventually--and that was a good year. The following year, as I recall, only one or two students out of eight or so completed the program, and attrition of half or more of the class seemed to be the norm for the preceding years. A lot has changed in the interim: Michigan now offers better financial aid for all admitted PhD students than it did then; and the Department was a pretty ferocious place at that time (Peter Railton is alluding to this briefly in his Dewey Lecture); I think it is less so now.