If you ever wondered about what it takes to get tenure at MIT, this should interest you: Unraveling Tenure, a report at The Tech that "[reveals] one of the most subtle and misunderstood processes at MIT, and [explains how one professor, despite his popular teaching, lost because of it" [Thanks to Incoherent Ponderer for the pointer].
A couple of highlights from the article:
So how are the recommenders chosen? According to [Prof. Patrick H.] Winston, the candidate and his or her mentor make a list of people whom the candidate would like as recommenders, as well as a list of people the candidate would not like. But it is up to the committee appointed to research the candidate to choose who to request a recommendation from, and the committee may choose people from both lists. The candidate never finds out who the committee chooses. [Bold emphasis added]
How to build an international reputation is the tricky part. Winston acknowledges that tenure decisions are based on “short-term reputations,” and he recommends junior professors to tackle “the sorts of things that can end up producing results in a small number of years,” rather than large problems that require “ten years” before a paper can be produced.
“Tenure is never about promise,” he stated. “It’s about accomplishment.” [Bold emphasis added]