One would like to say: This is what took place here; laugh, if you can.

Ludwig Wittgenstein

I don’t do empirical work myself, but I often talk to authors who use interviews and observations in their ethnographic research. You don’t have to be an ethnographer, however, to be able to distinguish between experience and understanding — between what happens to us and what we make of it — and I want to use this distinction to say something that I hope will be useful to people who write analyses of social life. As always, I will imagine that the problem is one of writing down what you know for the purpose of discussing it with other knowledgeable people, i.e., that you are writing “for academic purposes”. I will also assume that you are going to go about this in a deliberate, disciplined fashion, composing one paragraph at a time that supports, elaborates or defends your claims. In this case, I’m going to assume that you will mainly be interested in supporting your claims; that is, I will imagine that you want to get your reader to believe that what you are saying is true.

To keep things simple, I will assume that your data consists of interview transcripts and field notes. (Data can be messier than this. The interviews may be represented only by notes jotted down after the fact; the field may be little more than a memory to you at the time of writing. But I will leave this to you and your methodological conscience to work out between you.) What this means is that the materials that are “given” to you for analysis are records of things people have said and done. You can report what people told you in answer to your questions, what they said to each other in conversation, where they went and what they did, how they sang and how they danced. And you can describe the material conditions under which they did this, the settings in which the action and dialogue took place. “Social establishments—institutions in the everyday sense of that term—are places such as rooms, suites of rooms, buildings or plants in which activity of a particular kind regularly goes on,” said Erving Goffman (Asylums, p. 15). As a social scientist, your empirical object, i.e., the thing you try to experience in a particularly scientific way, is simply this “activity of a particular kind”. You are “particular”, if you will, both in your choice of methods and in your choice of objects.

Now, while what people have said and what they have done is “given” to you (as data, on the assumption that we trust your methods), what these words and actions mean is open to interpretation. Your analysis will attribute significance to the discourse and behavior of your research subjects (which, perhaps ironically, together constitute your object of research). You will not just tell your reader what words were spoken or what deeds were done, you will tell us what the people you studied thought, felt, imagined, believed, desired, hoped, and feared — what they loved, even, and what they hated. You will not just describe their movements but their intentions, not just their performances, but the projects those performances represent. While a novelist like Hemingway may be content to present an experience such that it becomes part of the reader’s experience, an ethnographer must provide the reader with an understanding of it too. And it must be the ethnographer’s understanding that the writing conveys. The reader must become aware, not just of what happened or what was said, but of what the writer took it to mean. Only in this way are the ethnographer’s ideas exposed to the criticism of peers. You must tell us, not just what you think took place, but why you think it matters.

There is a straightforward way to deal with this in your writing practice. When setting up your writing moment the day before, make sure your key sentence states, not just a matter of fact, but an interpretation of one or more facts. In your key sentence, don’t just say that someone did or said something, say that they were unhappy or successful or competent or deceptive or struggling or celebrated or surprised or critical. That is, make a statement that is not immediately true of your data, but one that can only be known by carefully analyzing your data, teasing out the details, uncovering their hidden meaning. Then resolve to support that interpretation by citing your interview transcript or your field notes. Select from your data the items that, taken together, make your statement about the mental states or social relations of the people you have studied easier to believe. What would someone have to say in an interview to make a reader believe, on that basis alone, that she was unhappy? What did you observe someone do that could make your reader believe he was struggling? Write the paragraph with that very specific aim in mind: to support your interpretation of a piece of human behavior.

“In the room the women come and go,” wrote T. S. Eliot, “talking of Michelangelo.” But why did they come and where did they go and what did they think of Michelangelo? As a poet, Eliot didn’t owe us more than the image, which he presumed would leave us with a feeling, and that feeling was all he wanted us to take with us to the next strophe. But as a social scientist, you can’t just leave it at that. Or, if you do, you must be very certain that your reader feels exactly what you want them to feel — ideally, that they think what you think of it all — in short, that they get your meaning. We can imagine writing every paragraph in your analysis as I have suggested — the key sentence always an interpretation of what the data shows — and then, satisfied that each of your attributions of meaning to your subjects has been adequately supported, removing every key sentence so that the reader will have to get your point without your making it explicit. Perhaps this is what Wittgenstein meant in his remarks on Frazer’s Golden Bough, where he said “we must do away with all explanation; and description alone must take its place.” Of course, he also believed that philosophy ought to be composed like poetry. Far be it for me to discourage a little poetry in your writing, but let’s not make social science more difficult than it needs to be. Surely, we want to say to Wittgenstein, one can say, “This is what took place here. And this is what it meant.”

See also: “Reading, Talking, Seeing.”