A couple years ago, I mentioned that one thing that has stuck with me very strongly is my 10th grade English teacher writing "Sally is a tireless reviser of her work."
This week, at the beginning of my meeting with my advisor, he pulled out a copy of The Language Instinct, which he's been reading recently, and read me this quote: "A banal but universally acknowledged key to good writing is to revise extensively. Good writers go through anywhere from two to twenty drafts before releasing a paper. Anyone who does not appreciate this necessity is going to be a bad writer."
Pinker also writes, "Expository writing requires language to express far more complex trains of thought than it was biologically designed to do. Inconsistencies caused by limitations of short-term memory and planning, unnoticed in conversation, are not as tolerable when preserved on a page that is to be perused more leisurely. Also, unlike a conversational partner, a reader will rarely share enough background assumptions to interpolate all the missing premises that make language comprehensive. Overcoming one's natural egocentrism and trying to anticipate the knowledge state of a generic reader at every stage of the exposition is one of the most important tasks in writing well. All this makes writing a difficult craft that must be mastered through practice, instruction, feedback, and--probably most important--intensive exposure to good examples."
I am really trying to keep this in mind this week as I work on version 5 of Experiment 1 and version 3 of Experiment 2. Fortunately, this evening as I was basically re-writing the results of Experiment 2 from scratch (and with the stats from what I think is Experiment 2 analysis version 4 or 5), I realized that I was making improvements that I want to incorporate in Experiment 1 as well, even though the Experiment 1 results section is looking pretty good at this point already. So yes, I'm seeing evidence that supports this whole revision thing in my own work right now.
I also realized that I have been thinking of myself as working on the same 15 pages for weeks now, but that actually my two experiments together are up to 29 pages...and counting. This helps satisfy the part of my mind that is occupied with the calculation of the ratio of pages written to total number of pages needed (about 60).
And I will just say that I find writing empirical papers much harder than review papers. Clearly and thoroughly presenting the results of a somewhat large number of 2 x 2 x 2 factorial design experiments and tying the results together in a way that makes sense is hard. (We are not "biologically designed" to understand the results/implications of our own three-way interactions, let alone someone else's.)
I should enjoy the fact that I am currently writing a paper that is supposed to be long because one of the hardest things of all is to be clear and thorough and insightful and concise.
Subscribe to:
Post Comments (Atom)
1 comment:
One of the things that worked out so well about the last paper I wrote (that Laguerre planes one) was that I wrote it so far in advance that I was able to write a completely different version afterwards, greatly improving it. When I was younger and we didn't do everything on computers, I found that writing a draft over from scratch would improve the writing a lot, but I'm usually too lazy to do that when I can just edit on the screen.
"Write early, write often" is obviously the key here. It seems like for any serious paper, you can't even really figure out how it basically should go until you've written an entire draft.
Post a Comment