Race has a place in human genetics research, philosopher argues

Some scientists have argued that bringing race into the field of medical genetics research is problematic or even damaging. But Penn philosopher Quayshawn Spencer has a different view, one based on nuanced concepts of semantics and language, which says there is a racial classification that’s medically useful to reliably sample human genetic diversity. He published his work in the journal Philosophical Studies. 

As an example, Spencer brings up a random sampling of the United States population to look for links between alleles and traits within the human genome. If racial distribution were irrelevant, he argues, the resulting proportion of Pacific Islanders from pulling such a sample—less than 1 percent—wouldn’t matter. But in actuality, Pacific Islanders have extensive genetic diversity so such a sample should really include about 20 percent of this group.

“There have been arguments that suggest it’s a mistake to think about race based on genetics in any way,” Spencer says. “What I show is that you can’t account for the meanings of some of the racial terms we use if you don’t think about some races as ancestry groups in the way that population genetics tells you they are.”  

Spencer approaches the issue from a philosopher’s perspective, and it’s a question his work has been trying to answer for five years. 

He began this research by delving into the five racial classifications on the Census, selected and defined by the Office of Management and Budget (OMB): white, black, American Indian, Asian, and Native Hawaiian or Other Pacific Islander. According to the OMB, these racial classifications were supposed to be comprehensive, based on ancestry, and not be redundant. But applying a philosophical theory called referential semantics, Spencer found that the racial groups didn’t actually fit the government’s intended descriptions. 

To understand this argument, it’s important to understand the philosophical concept called truth conditional semantics, which says broadly that a term’s meaning is what you need to know about that term to discern whether the sentences in which it appears are true. Philosophers fall into two main camps here, descriptivists and referentialists, and for both, context matters a great deal.    

As its name suggests, descriptivism states that a word’s meaning is the description used to fix the reference of the word. Referentialists, on the other hand, argue that a term is defined by the person or item to which the term refers. 

Consider Barack Obama. Who is he? A descriptivist would define “Barack Obama” as the 44th president of the U.S. A referentialist, on the other hand, would point to the specific man being referenced; perhaps that man is the former Commander in Chief, but perhaps he is someone else, such as a neighbor’s new baby.  

This is where context helps. Contemplate the sentence, “It’s possible for Barack Obama to not have become president.” Under the descriptivist meaning given above—in other words, that “Barack Obama” is defined as “44th president of the U.S.”—the sentence is always false. Under referentialist meaning, however, if “Barack Obama” means the actual man himself, then this sentence is true because that man might not have become a U.S. president at all, much less the president following George W. Bush. 

“For every sentence, you’ve got to get the right truth-value,” Spencer says. “When you look at the total array of sentences that we use a term like ‘Barack Obama’ in, what’s common is a reference, in this case, the man Barack Obama.” 

Spencer then applied this thinking to the OMB’s racial classifications. The federal government, for instance, defines white as “a person having origins in any of the original peoples of Europe, the Middle East, or North Africa.” But Spencer found that didn’t align with theories coming out of evolutionary biology, which said that people migrated out of Africa in two waves via Eurasia, one group taking a southern route, the other traveling north. 

“If a white person is anyone who has ancestry from Europe, the Middle East, or North Africa, that’s every single human being except for those with 100 percent ancestry in sub-Saharan Africa,” he explains. In other words, taking these definitions at their face value leads to unintended results and not necessarily to the groups the OMB indicated it was trying to capture.

“But,” Spencer adds, “the oddities melt away when you take the terms’ meanings instead to be references,” and in particular, the five genomic ancestry groups discovered by population geneticists: Africans, Caucasians, East Asians, Native Americans, and Pacific Islanders. 

Spencer published these findings in 2014. His new paper takes them a step further, applying this idea to medical genetics research. Rebutting those who have argued that race should not be part of this field, he states that there is a racial classification that’s medically useful to reliably sample human genetic diversity—although at this point not one that should necessarily be put into practice by clinicians providing medical treatment. 

“Until recently, philosophers weren’t contributing that much to this discussion,” he says. “This new paper takes on the challenge of showing that, using semantic theory, OMB races do have utility in medical genetics. If you think about these races merely as genomic ancestry groups, you can get the ‘right’ results for who is white, who is black, and so on, and that can offer some guidance.” 

Quayshawn Spencer is an assistant professor in the philosophy department in the School of Arts and Sciences. The paper, “A racial classification for medical genetics,” will be published in the May 2018 issue of Philosophical Studies.