This winter I’m taking a course on urban education. Our first topic: segregation and desegregation in schools.

Firstly, what do we mean by segregation? As a working definition, I’ll offer that segregation is the spatial pattern of people across some attribute. So we could talk about segregation by race, by income, or by favorite ice cream flavor. Once we pick something to measure against, we find that every city is segregated according to this definition. What matters is in what way the segregation manifests and the consequences on the populace the pattern has. Segregation patterns can be uniform, with all groups distributed more or less evenly within a region, or clustered. Likewise, we could also calculate the extent to which subpopulations are isolated from each other—which also gives a rough estimation of how often members of one group is likely to run into someone outside of their group. I think when we talk about ‘segregated’ groups, we typically mean highly clustered populations that are isolated from the other groups in the city.

I don’t think that clustered, isolated groups are necessarily bad on their own. I love visiting the North End and Chinatown. Because they’re both T-accessible, it’s easy for me to get there. (Though, both neighborhoods have had rough pasts.) And Harvard Square is the nicest place I’ve ever lived. Score one for segregation!

Moral judgments aside, self-selection can have a big influence on patterns of segregation, at least it can in models. The positive feedback loops reinforce small, individual choice to generate large-scale patterning. Schelling’s model of segregation is a classic, good first example of what I mean. In this model individuals exhibit only a slight preference to have neighbors that are similar to them. The individuals in this model are not racist. (Or maybe they are. I don’t have a good functional definition of racism yet.) When individuals find themselves in a neighborhood that is too unlike themselves, they move somewhere else at random, possibly to a neighborhood more dissimilar from themselves than the last. Even with this mild, partially blind behavior, a totally segregated structure emerges.

In more relaxed models that completely ignore race, even more realistic patterns of segregation form. In this class of model, individuals simply choose to live in the nicest area they can afford. As if by magic, isolated poor and rich neighborhoods form. Depending on the details of the model, wealthy suburbs appear spontaneously. If we use socioeconomic status as a proxy for race, it’s the same old story. Except this time, we have a systems-level mechanism that generates isolated, poor communities that lack the power to advocate for equitable resources and very rich communities with disproportionately high share of public goods insulated by a buffer of middle class individuals. Race was not the cause; money was.

When was ask whether it’s morally justified for a white family to send their kid to a predominantly white school, I think it’s important to know what about the school is so attractive. Do all parents value differentiated cultural and social understanding across many kinds of experience? Are they likely to value it more than a pretty campus or reputation of success by its graduates? Sure, in some cases the choice may be motivated largely by racism. But I’d expect that in many cases, it’s mostly a matter of ensuring access to the most and best resources possible for their child. It just so happens that low-resource groups aggregate, even in the absence of race.

I believe that diversity (of background, experience, perspective, and the like) is important in schools because, as has been mentioned a few times by others, students learn how to navigate social situations outside of school from the people they meet in school. But when we talk about diversity, do we really mean racial diversity? As an example, imagine that an elite, wealthy, mostly white college in the Northeast has recently been chastised for admitting a student body that is not sufficient diverse. Consequently, the school begins recruiting wealthy black students from Africa, some of whom attended the same boarding schools as students already enrolled in the college. In time, the student body comes to be half white, half black with an even mix in all classes and housing situations. In what sense, if any, has the college increased diversity on campus? Do you think the college has produced the diversity they were previously lacking?

While I think that racial segregation is a problem, I don’t think race is necessarily the capital-C cause. In a world without racism, economic segregation will still exist. But I’m willing to bet that in a world with no financial disparity, a lot of the troubles we associate with racism would evaporate. And so, I think race will play a secondary part in the solution to segregation. In fact, I think that race may even obscure the issue of access to equitable education for all. (I’m not sure if that’s what we’re really trying to achieve, but I think it’s a good start.) Instead, I believe that the struggle of the American education system is one of power and status. As such, I think we should talk about resource allocation (including strategies that move students to resources as well as bringing more resources to students), causes and effects of socioeconomic segregation, and cultural and pedagogical practices that systematically discourage/motivate students to learn the skills required to become an informed and capable citizens.