Are you Black, Hispanic, Asian, Caucasian? What does that question mean, exactly? Does the answer matter? You may have relatively little hesitation in answering the first question, even though the second may stump you, thus throwing some doubt on your first answer. As for the third question, it seems obvious that “race” matter, but it is much less obvious as to why, or whether it should.

I’ve been writing about race from the point of view of a biologist and a philosopher of science for a while now (see this, and this, and this), and it still puzzles me just how much confusion, not to mention acrimony, the subject engenders. On the one hand we have strong “realists” about race, who maintain that it is simply ludicrous to claim that races don’t exist. On the other hand there is an equally strong “antirealist” camp, which claims not only that races don’t exist, but that to insist that they do in itself betrays a dangerously racist attitude.

In some sense, the race debate reflects the broader divide between “naturists” and “nurturists” about other human traits, especially behavioral and cognitive ones. I have written about that debate as well, and have been accused by both camps of falling into the other one. (The answer is neither: I’m an “interactionist,” believing that any complex human trait is the result of inextricable, non-linear interactions among genes, developmental processes, and environments, to the point that even asking questions like “what percentage of X is due to genes/environment?” is nonsensical.)

Back to race. It’s a good bet that when debates keep going in circles like that for years and then decades, the answer doesn’t lie in more empirical evidence, but rather in a mix of conceptual clarification and parsing of ideological posturing. Since I’m a philosopher, not a sociologist or political scientist, I’ll leave the latter to the pertinent experts and focus on the former instead.

It is hard to find sound, clear, and intelligent treatments of the issue of race in either the popular or the primary literature. But I know I can rely on my good friend (and former PhD mentor) Jonathan Kaplan (Oregon State, Corvallis) for those qualities, regardless of the occasional disagreements the two of us may have on specific issues.

Sure enough, Jonathan has published a must-read paper on race in Philosophy of Science, co-authored with Rasmus Grønfeldt Winther, entitled “Realism, Antirealism, and Conventionalism about Race.” Their argument is as clear and compelling as it is simple: “races” are real, not real, and conventional — depending on the context. Let me explain, beginning by introducing you to (or reminding you of) the pertinent terminology.

“Realism” in philosophy of science refers to a position that attributes positive ontological status to a given object or concept. So, for instance, to say that one is a realist about electrons within the context of the Standard Model of physics means that one thinks that electrons are real, actual particles, with specific physical attributes.

“Antirealism,” by contrast, is the judgment that a hypothetical construct plays a theoretical role, without necessarily having a positive ontology. To continue with the above example, an antirealist about electrons would say that these are hypothetical entities that play an instrumental role win the Standard Model (e.g., they are necessary for calculations aimed at predicting the result of experiments), but that one doesn’t need to take the further step to believe that they are physically real. If this sounds strange, realize that the history of science is full of entities that were postulated for theoretical reasons but turned out not to be physically real, from Ptolemaic epicycles to caloric fluid to cosmic aether.

Finally, to be a Conventionalist about something is to say that one may adopt a particular vantage point for specific purposes, but change it if the purpose changes. For example, we can think of light as a physical phenomenon that behaves like a particle under certain circumstances and as a wave under different circumstances. But it is not very helpful to say that light really is alternatively a particle and a wave. In an entirely different realm, we could say that the boundaries of New York City include certain geological areas and exclude others, and such boundaries matter very much for economic, political and other purposes. But it’s not like New York City really ends at those boundaries in a deeper metaphysical sense.

Now that we are clear on the differences among realism, antirealism and conventionalism, let us go back to races. Here too Kaplan and Winther distinguish three possibilities: races as bio-genomic clusters, as biological entities, and as social entities.

Let me discuss them in reverse order, which I think will help clarify things. To think of races as social entities is to take the standard “nurturist” position: human races are not biologically grounded, but they are the result of social practices. Conversely, to say that races are biological is to adopt the classic “naturist” stand: races are identified by some deep genetic dissimilarities among certain human populations, of which external markers such as skin color are the most obvious outward manifestation.

Lastly, a bio-genomic cluster is a technical term from population genetics, which indicates the fact that one can study genetic similarities and dissimilarities among individuals belonging to different populations and objectively “cluster” them (using a number of well known statistical techniques) into distinct groups.

We need to understand exactly what this means, because the idea is crucial to the very satisfying argument that Kaplan and Winther end up making. Statistical clustering can be effective, and objective, even when it is based on a small subset of variable markers. For instance, it is a well known fact that an overwhelming portion of the genetic variation among human beings (about 85%) is the result of within-population differences, and only a small fraction is attributable to between-population differences. Despite this, one can focus on the smaller fraction and identify a number of groups (clusters) that are indeed objectively distinct from each other based on that very limited number of markers. Bio-genomic clustering, therefore, is usually carried out for specific purposes (e.g., to focus on populations where certain genetic conditions have unusually high incidence, as in the case of Ashkenazi Jews and Tay-Sachs Disease), but it is useful only for those purposes. Moreover, different purposes may require a focus on different markers, thus yielding different — and yet still objective — clusterings.

With all of the above in mind, then, Kaplan and Winther conclude (and, I think, are obviously correct) that the most sensible positions concerning race are: conventionalism about bio-genomic clusters, antirealism about biological races, and realism about social races.

This means that we understand that bio-genomic clusters are “real,” in the sense of being objectively reproducible outcomes of certain data analyses, but that their realism is conditional on specific purposes and ceases to hold when the purpose of the analysis changes.

We are antirealists about biological races because they do not exist in any deep biological sense. Biologists use the term “race” to indicate subspecies of animals or plants that are genetically dramatically differentiated from each other, and that are likely heading toward a status as independent species (if the evolutionary phenomena that created the differentiation in the first place keep holding). The empirical evidence here is overwhelmingly clear that there are no human biological races in that sense.

And we are realists about social races because social identification as “Black,” “Hispanic,” “Asian,” “Caucasian,” or whatever — regardless of whether it is generated by the individual or imposed from the outside — has very real socio-economic as well as psychological consequences.

Kaplan and Winther’s parsing is both beautifully simple and in accordance with all we know from biology and the social sciences. It ought, then, to spell the end of the race debate. Something tells me, however, that that would be far too optimistic a prediction to make, unfortunately.

MASSIMO PIGLIUCCI IS THE K.D. IRANI PROFESSOR OF PHILOSOPHY AT THE CITY COLLEGE OF NEW YORK. HIS BACKGROUND IS IN EVOLUTIONARY BIOLOGY AND THE PHILOSOPHY OF SCIENCE, THOUGH HE HAS RECENTLY DEVELOPED A KEEN INTEREST IN STOICISM. HIS MOST RECENT BOOK (CO-EDITED WITH MAARTEN BOUDRY) IS “PHILOSOPHY OF PSEUDOSCIENCE: RECONSIDERING THE DEMARCATION PROBLEM” (UNIVERSITY OF CHICAGO PRESS, 2013). HIS OTHER WRITINGS CAN BE FOUND AT PLATOFOOTNOTE.ORG.