Like many writing instructors, I teach Hemingway’s “iceberg method” to students. I make sure to remind them that Hemingway was not an “academic” writer but a novelist and writer of short stories but, still, I tell them, he was adamant that writers must know what they’re talking about. If you’re going to write a novel about war, or bullfighting, or love, you better have some knowledge of the subject to start with. Indeed, he would argue, you should have some experience with these things. And, while academic knowledge is not always based on direct, personal experience, there is one section of a typical research paper that can apply Hemingway’s method almost directly. This, it turns out, is the methods section.
Let’s quickly recap what Hemingway meant. In fact, let us let Papa himself explain it.
If a writer of prose knows enough of what he is writing about he may omit things that he knows and the reader, if the writer is writing truly enough, will have a feeling of those things as strongly as though the writer had stated them. The dignity of movement of an ice-berg is due to only one-eighth of it being above water. A writer who omits things because he does not know them only makes hollow places in his writing.
Ernest hemingway, death in the afternoon
How does this apply to writing your methods section? Well, imagine writing the full story, in every detail, of how you gathered your data and analyzed it. This would be more Proust than Hemingway, I guess, and it’s not something you’d ever actually hand in for a grade or submit for publication. But imagine making it so complete that the reader could, not only replicate something like your study, but would in effect be doing the very same study if they did everything you tell them. In fact, imagine giving them instructions for having the very same experience that you did. Imagine putting in physical descriptions of the locations in which you interviewed your subjects or the office where you came up with your survey questions. Describe the bus ride to the field location where you carried out your observations, even what you had for lunch. Put in all your subjective judgments and perceptions, everything you thought and felt. Tell the reader where you are in doubt that you did it right, and which parts you thought at the time were just brilliant. Tell them what your enjoyed and what you suffered through, what gave you pleasure and what gave you pain. Capture every nuance of the process.
Now, think of your reader as someone who has done a similar study. What can you leave out of your account and let this reader fill in with their own experience? What do you have to keep in your account so that the reader can feel that you actually did all the things you leave out, because that’s what the reader would have done to have the experiences you describe. What parts are irrelevant to whether or not your data is of high quality because the reader assumes that you did them in the proper way (and you did them in that proper way, of course). What lengthy descriptions in our imaginary first draft can be summarized in a single bit of jargon (“semi-structured interview”, “control group”, “coding scheme”)? What is the simplest possible statement of your method to a reader who understands your methodology?
We can easily imagine that this version would be (less than) one-eight as long as the “Proustian” draft. That is, you can leave a great deal under the surface and the reader will still know everything they need to know to replicate your study, or at least enough to trust your data. “A writer’s problem does not change,” Hemingway said. “He himself changes, but his problem remains the same. It is always how to write truly and, having found what is true, to project it in such a way that it becomes a part of the experience of the person who reads it.” In research writing, this art of writing “truly enough” begins with an honest statement of your method that your reader understands. Face that problem squarely and you are well on your way.
The iceberg thing reminds me of our discussion of Storytelling as predictive model checking.
Yes, we can definitely talk about “the dignity of movement” of a statistical model. I think of your feather in the pouch of a kangaroo.