I love the research of Robert Boice, author of the 1990 book Professors as Writers. Although his research has been incorporated and stated somewhat more accessibly by Paul Silvia in How to Write a Lot, which I also love, Boice is The Man when it comes to research, data, hard numbers, or whatever else you want to call it. They make a great pair. Booklab doesn't focus as much on how we feel about writing, and neither does Boice. Instead, we care what works, and that makes Boice our go-to guru.
Today's tidbit is from Boice Chapter Five, Generative Writing, and his fondness for what he calls the "results-first approach." Boice thinks you have to write something before you can edit or perfect it, but he notices that most academic authors do the opposite, trying to be perfect on the first bounce. He isn't recommending spontaneous writing (he thinks that's a big Fail for different reasons), nor is he suggesting binge writing. Generative writing is more substantive, but it is still produced without judgment. It creates the stuff that can then be edited by a completely different part of the brain.
Over the course of the semester I'll post tidbits from Boice. A great way to read him is to buy both books, read Silvia first, and then read Boice second for backup material.