 People get on my case for not having a hard and fast rule for when I should wear a sweater and when I shouldn't. I mean, it's a matter of degrees. One of the persistent themes of thunk, besides an unhealthy obsession with Star Trek the next generation, has been to point out some of the complexities of ideas that are usually understood as irreducible or monolithic. Philosophy is one of the topics that obviously fits this bill, but we've also examined some of the nuances of rational thought, design, engineering, all sorts of stuff. We get a lot of interesting and enriching insight out of analyzing the machinery that's ticking away under and around these concepts. So when I bring up a paper with a title like this, I'm sure that at least a few of you might react somewhere on the order of, what? Okay, relax, we're not giving up on any of that stuff, we're going to qualify it. It's pretty clear that Kieran Healy intended to turn some heads with this paper, and given that his ultimate goal is a recalibration of his field of sociology, it's not a bad strategy to be a little inflammatory. We're going to cover some of his arguments against what he sees as a scourge of vacuous clamoring for additional nuance in that field. But to be honest, I think his points can be expanded without issue to some other fields, including some that we've covered here in an excruciatingly nuanced fashion. And they might serve as a helpful rubric for how to approach the process of dissecting ideas without getting lost in an endless hall of mirrors of exceptions and provisos. I'm going to try and tease a more general principle out of his extremely entertaining writing, please just read the paper. Healy's target isn't really nuance itself, he's very clear that it's just a feature of the landscape of sociology, sometimes useful, sometimes not. The problem that he's addressing is a rise in demands for additional nuance in any theory, every theory, for its own sake, untethered from any justifiable connection to the actual contents of the theory in question. It's trivial for anyone to listen to any carefully thought out model of human societies and say, well, that's well and good, but aren't things a little more complicated than that? Or aren't you leaving some important aspect of the world out without the slightest amount of thought invested in how those questions actually bear relevance to the subject material, whether it's helpful to know the answers, or whether it's remotely possible for even the most diligent researcher to find the answers. The very nature of abstraction necessary for creating sociological theories requires leaving a bunch of stuff out, which invites a trivial critique, why don't we add them back in? You can use a term like honor cultures to mark out some interesting similarities between societies in Japan and the Middle East, but, I mean, in the process of creating this abstract category that focuses on a particular aspect of how people live in those cultures, you're leaving out so much stuff, differences in politics, economics, class, gender roles, all kinds of things. Those things are definitely important, there's volumes upon volumes written on them, but the temptation to add them back into the analysis of honor cultures is misguided. Add them all back in, and we've deluded the theory past the point of being useful or predictive or explanatory or anything of this sort. All we're giving is a one-to-one description of the societies in question. Japan is like this, the Middle East is like that. There is a reason we were talking about both of them, but hell if I can remember why. Japan plays into Healy's first critique of unbounded demands for nuance, on the grounds that the process of abstraction is necessary to have any theories at all, and reflexively asking for nuance is essentially asking to dissolve the theory in favor of a brute empirical description. He also notes that while it's very easy to say something like, well, how does class play into this, the reality of what sort of data gathering and analysis you'd need to respond in any meaningful way is a task for several well-funded research departments over a couple of decades, unless you can make a meaningful connection between that concept and the theory's subjects yourself, just saying X is important and your theory should address it is an impossible request to increase scope without any demonstrable benefit besides saying now it addresses race too. Adding these additional vectors of analysis willy-nilly also runs the risk of overfitting the theory to the particular societies it's built around, as we've discussed. Now obviously the accuracy and explanatory power of a theory are most important for evaluating its merit, and the issues that arbitrary demands for nuance raise for that endeavor are most important to diffuse, but there are some other fluffier aspects that are also degraded by those demands. We don't usually talk about the aesthetics of models, but there's something valuable about finding a compelling way of looking at the world, something that sticks in people's minds and rewards their attention. Beauty. Elegance. Healy suggests something like, interestingness as a good measure. It's true that there's some aesthetic value to a very nuanced theory, but acting like that's the only thing that has any value? Well, have you ever met anyone who can't enjoy a basic-ass mac and cheese or a fun action romp because they aren't nuanced enough? Basically that. Again, this is a criterion that is considerably less important than a theory's accuracy, but if adding nuance doesn't make it any more predictive, and in fact makes it impossibly boring, there are also strategic considerations. This is about how to create models that will have the greatest impact. It's well and good to formulate a neat and correct theory, but if the result is that it gets published and subsequently forgotten, you probably shouldn't have bothered. Nuanced texts can be very impactful, but asking to plaster some nuance onto a theory for its own sake probably isn't going to make the theory more complete and influential as a result. Society cites the incredibly far-reaching success of economic theories, which are often cartoonish caricatures of human decision making, but still manage to influence totally unrelated fields, like child-rearing and philosophy, simply because they're very powerful ideas. Looking at all these arguments, it's easy to get a sense why Healy might scoff at the ever-expanding call for additional nuance in sociological theories, comparing them to an invasive species that has totally taken over an ecosystem. His critiques also land uncomfortably close to home in a lot of the subjects that we've been dissecting here on THUNK. To be fair, the targets of our inquiries are usually well-established theories and ideas, things that tons of people know about and hold some sort of opinion on already. Obviously, we don't have to worry about exposing enough nuance that they will somehow lose influence or become less interesting. Okay, maybe that last one, if I'm being totally honest, but it's probably useful to bear in mind that by saying things aren't as simple as that, by calling to expand these ideas to incorporate or account for some important part of the world, we are, in some sense, being anti-theoretic. We're corroding the substrate of abstractions necessary to build something meaningful. Now, sometimes that's 100% justified. We don't have to feel bad about breaking down abstractions that are misleading or routinely compel people to make irrational decisions. A racist might have a neat little conceptual framework that falls apart when it comes in contact with genetics, and that's a good thing. But maybe we should be cautious about trying to cram the whole complexity of the world into every model that we encounter just because, just for the sake of completeness. The perfect is very often the enemy of the good. Maybe instead, we should look at models of the world and, rather than asking if they're nuanced enough, we should ask, are they useful? Do they lead us to make accurate predictions? Are they interesting? Do they reward investment of thought? And are they powerful? Do they change the landscape of thought in some important way? It's not a complete list, but do you find these compelling criteria? Please leave a detailed comment below and let me know what you think. Thank you very much for watching. Don't forget to bubble up, subscribe, blah, share, and don't stop thunking.