Huenemanniac

I was in a conversation recently, and somebody tried to explain a small group’s weird behavior in a Nietzschean or Freudian fashion: “They couldn’t exert power over others, so they began punishing themselves” or some such claim. I instantly judged that that explanation couldn’t possibly be right, even though I didn’t have a better one, and that got me wondering what made me so sure.

Here’s what I eventually came to. Consider individuals. If I want to explain or predict an individual’s behavior, many times I need a lot of background information about that individual specifically. Not always: if I see someone approaching a busy street, I can predict they will stop and look both ways before trying to cross, and I don’t need to know much about them to predict this. But if I see them eating breakfast at a cafe, and I want to predict what they are going to do after breakfast, I need more information. Do they have an office job to go to? Are they retired? Do they like to exercise in the morning? Do they have kids to watch? And so on. If I do gather the information and make a prediction, that information won’t help when I shift to another individual and try to make a prediction, since their circumstances are different. The prediction is not transferable.

Now consider large groups. If I gather information about what a lot of people typically do after eating breakfast in a cafe, I might end up with some general information such as that 45% will do errands, 20% will go home and fiddle around, 10% will go to an office, etc. That general information can be used then to make a prediction about any individual I select. Facts about the behavior of large groups can be transferred to individual behavior—not always with success, of course, but with some reliability over the long run.

Finally, consider mid-sized groups. They fall in between, and are the hardest to understand. They are not big enough to provide general and reliable information about their behavior, but they are big enough to make it a headache to try to get background information of all the individuals belonging to them and make predictions.

So when we face mid-sized-group behavior, we often end up making stuff up. We say to ourselves, “There’s a sizable group of people. There must be a single explanation for their behavior, just as reliable as explanations we give for large group behavior.” But probably there isn’t. Probably there are just individuals, each with their own explanation. Some individuals are similar, and they do similar things, and this leads us to think there must be a common explanation for that behavior; and then if we notice some individuals who don’t fit the pattern, we mark them as aberrant exceptions to our made-up explanation (or we add in ad hoc exceptions and provisos).

This is the nature of Freudian and Nietzschean social explanation. They are based on some small-scale pattern (like a handful of upper-class Viennese women, or some perceived relationship that says more about the theoretician’s mind than about anyone else), and they are meant to scale to larger populations, but they just don’t. In fact, when you consider what you know about people, it is hard to imagine these sorts of explanations scaling or transferring. Everybody wants to have sex with their mother? Everybody is trying to tyrannize over others (or themselves)?

Here’s a more common example. “Why are scholars in the humanities writing such ridiculous books and articles these days (you know the type)?” Explanation: they feel complicit in the many forms of oppression that have been practiced by “the West” over the centuries, and out of guilt they are attempting to deconstruct power structures in their disciplines to allow for a greater range of voices to be heard. Well, maybe. But note that only some scholars in the humanities are involved in projects like this; many are not. Some of the ones who are involved do perhaps feel guilty; some are just self-promoting and looking for easy publication opportunities. Some of the ones who are not involved in such projects do feel guilty, but that doesn’t get expressed in the work they do. Some aren’t thinking much of anything about what they are doing, and they simply do what they do because that’s how they’ve been trained. In short, it is a complicated mess.

This mess is a mid-sized group phenomenon, where there sometimes seem to be common patterns among some members, but those patterns are just as commonly violated by other members. There simply isn’t a single phenomenon to explain with a single explanation. There are instead many phenomena, explained in different ways, with mixes of commonalities and differences. It’s turbulent. If we zoom out to a wider angle and talk about trends in higher education, the alleged phenomenon we were trying to explain disappears, as it gets swallowed up by what’s happening in disciplines across the board. There we just see ever-increasing specialization, continued decline (since Reagan) in public funding, changing demographics, and so on. At this level, explanations can exist, since we can measure more stable and reliable trends, and those explanations can with some success be transferred to the behavior of individuals.

So, in short: if you are out to explain a social “thing”, make sure first that there really is a “thing” there to be explained. Behaviors of individuals are things; stable behaviors of large groups of individuals over time are things. Things can be explained. What falls in between these two—just observe, and be skeptical.