In his response to some thoughts on academic writing I’d sent him, my friend and former colleague at Touchstone Steven Hutchens added a note about how he writes.
Your suggestion about writing out in one sentence what you want to say is a good exercise, but the method I generally use is different, only because over the years I have found it matches the way I think: What I want to write, long or short, usually comes to me in an instant, connected with an experience or incident or idea, but in a thought that does not yet contain words. It is a Word before words.
I write down the experience, incident, or idea as artfully as I can, and follow through to the conclusion, crafting the words to follow the supervenient idea as I write, and letting them lead me in somewhat the same way fiction writers allow their characters, once animated, to develop on their own. Since the conclusion is contained in the original wordless idea, I will know when I come to it and am finished writing the piece. Then I will go back to fill and polish, smoothing over rough spots, all flaws the result of pushing ahead to finish before the muse departs.
I tend to write this way myself but I can’t claim that what I want to write comes to me in an instant. I often begin with a story I’ve read or an experience or idea of my own, and know I’ve intuited something the story illustrates, but finding out what I’d intuited often takes a lot of work. And the work can frustrate and annoy me, because it’s like digging a trench through clay and rock.
Sometimes after much work I realize that I don’t actually have anything to say about the story I want to tell. Being able to tell the stories and let them carry their own meaning was one of the pleasures of writing the “While We’re At It” section of First Things, but few other places let you do this. Most publications expect you to say something.