Gregory Adams / Getty

In 2014, Ralph Taylor applied to have his insurance company in Washington State certified as a “disadvantaged business enterprise.” The DBE program at the U.S. Department of Transportation was originally designed to help minority- and woman-owned businesses win government contracts. So as proof of his minority status, Taylor submitted the results of a DNA test, estimating his ancestry to be 90 percent European, 6 percent indigenous American, and 4 percent sub-Saharan African.

Government officials reviewing Taylor’s application were not convinced. They saw that he looked white. They noted that he was unable to directly document any nonwhite ancestors. They doubted the underlying validity of the DNA test. And, most relevant to the purpose of the program, they found “little to no persuasive evidence that Mr. Taylor has personally suffered social and economic disadvantage by virtue of being a Black American.” They refused to certify his company. So Taylor decided to sue—out of principle, he says, because other business owners who look white have won DBE certification before. The Seattle Times first reported on the case in detail last week.

Taylor is now challenging how racial groups are defined for this program. “Black Americans,” according to the federal regulations for DBEs,“includes persons having origins in any of the Black racial groups of Africa.” The lawsuit calls this definition “impermissibly vague” and criticizes the lack of “any minimum percentage of DNA, or other objective criterion.” “He considers himself to be Black based upon DNA evidence,” Taylor’s lawyer asserted in a letter included in the lawsuit, which also called DNA “objective” and “unalterable.”

In its four-decade existence, the DBE program has long wrestled with questions of how to determine if someone is a minority. Proof of race and ethnicity “has been a thorn in the side of the DBE program for years,” said a 2001 article in the magazine Government Contractor. But Taylor’s case appears to be the first time, according to Jennifer Sommerville, a lawyer who has written about DBEs, that DNA evidence has come up in a lawsuit over eligibility for the program.

According to several legal experts I spoke with, it might also be the first time a genetic ancestry test is being cited as evidence of race in any type of court case.


Currently, in situations like DBE certification, the legal system generally lets people identify their own race. Title VII employment-discrimination lawsuits are another common scenario where a person’s race is relevant. Richard Levy, an attorney who worked on a $98 million class-action lawsuit alleging racial bias in the New York City Fire Department, says the case relied on self-identification to decide who was eligible for the money. “If they identified as black, they were,” he says.

If DNA evidence somehow seems more tangible and less subjective than self-identification, consider the problems it would also pose.

For one, the accuracy of DNA tests are unproven—and the specific test Taylor took in 2010 is now widely seen as outdated. Today’s leading test companies, such as AncestryDNA and 23andMe, examine around 700,000 DNA markers, comparing them to a database of thousands of people around the world. Still, customers have found that different companies will return different results. And companies also frequently tweak their proprietary algorithms, so results can change from software update to update.

In contrast, the AncestryByDNA test Taylor took looked at just 176 DNA markers, according to government documents, less than one-thousandth of the current industry standard. Despite a similar name, the test has little to do with the more popular one from AncestryDNA. And in fact, a Google search of AncestryByDNA brings up dozens of angry reviews calling it a “waste of money” and warning customers about the confusing name. Taylor has offered before to take another DNA test at the government’s expense, but nothing has come of it.

Even if a perfectly accurate genetic ancestry test did exist, it would not easily settle questions of race. The percentage breakdowns of a test do not map neatly on to racial categories. How many African DNA markers does a person need to have to be considered black? Four percent? Twenty-five percent? Fifty percent? There are no universal cutoffs. Genetic variation is real, but the boundaries of racial categories are socially determined and have constantly shifted over the course of American history. “You cannot rely on DNA evidence alone to decide what is really a socially constructed concept,” says Sheryll Cashin, a law professor at Georgetown University.

A pair of Supreme Court decisions in the 1920s, says Sherally Munshi, also a law professor at Georgetown, is particularly revealing of how racial categories have been drawn. Under U.S. law at the time, only “free white persons” and “persons of African nativity or persons of African descent” could be naturalized as U.S. citizens. A Japanese immigrant named Takao Ozawa brought a case arguing that his white skin qualified him as white. In 1922, the Supreme Court ruled against Ozawa because white persons only referred to Caucasians. Race scientists at the time excluded Japanese from the Caucasian race.

Then, in 1923, a similar case involving an Indian immigrant named Bhagat Singh Thind also reached the Supreme Court.* Since race scientists considered Indians to be Caucasians, he argued that he was eligible for naturalization. The court ruled against him, too. This time, it held that “white person” meant “what is popularly known as the Caucasian race.” The court, in a span of months, appealed to and then discarded the science of its time in constructing an idea of whiteness.

With the advent of genetic ancestry tests, people are once again asking if and how the latest science should be incorporated into an understanding of race. “There are close parallels,” Munshi says. And the answer is far from settled.


Taylor also identified himself as Native American in his DBE application, noting the results of his DNA test. (He says he grew up with stories of Native American ancestors from his father’s family but has no documentation.) As part of the lawsuit, his lawyer filed a public-records request unearthing emails discussing the DBE certification of another business owner with a tribal card showing him to be 1/256 Native American. Taylor’s lawsuit cites this as evidence of the arbitrary criteria of the DBE program.

But Kim TallBear, a professor at the University of Alberta, says this is a misunderstanding of how Native American identity works. Unlike racial categories of black or white, tribal-enrollment criteria is actually quite clear. Each tribe gets to determine who belongs, and membership is often based on tracing direct ancestry to other members of the tribe.

“It might be all these people have Native American ancestry,” TallBear says. “My question is: Who cares? If there’s a particular ancestor that is close enough you can find living family, then you can do that. If there’s nobody for you to find and no tribal community that’s going to claim you, it doesn’t really mean anything.”

Tribal-enrollment offices do use genetic tests to establish parentage. But ancestry tests are irrelevant for enrollment, despite their growing popularity. “Members of the public have been showing up to tribal-enrollment offices and showing them ancestry DNA tests,” says TallBear. “And they’re like, ‘I don’t know what this is. Who are you trying to link?’”

Taylor’s lawsuit is also seeking to void the definition of Native Americans in the federal DBE regulations, which includes “persons who are enrolled members of a federally or State recognized Indian tribe, Alaska Natives, or Native Hawaiians.”


Taylor’s case will be heard in the Ninth U.S. Circuit Court of Appeals in the next few months.

When reached by phone this week, Taylor readily admitted he didn’t think DNA was an objective standard either. “If you’ve got to be 50 percent, what if somebody is 49?” he asked. Ultimately, he just wants to expose the DBE program as unfair. He believes it should be race-blind. The Washington State Office of Minority and Women’s Business Enterprises, which runs the state’s DBE-certification program, did not respond to a request for comment.

It’s one thing to recognize that race cannot be measured in percentages of DNA and that racial categories are not always separated by bright lines. It’s another to say race is entirely irrelevant. The DBE program’s race-conscious policies have been challenged in court before. In a 2005 case, Western States Paving v. U.S., the court ultimately ruled that the DBE program could only apply to groups if they are actually experiencing discrimination. To comply, Washington has had to conduct disparity studies.

The latest disparity study, from 2017, concluded that minority and female business owners continue to face stereotypes, discriminatory attitudes, negative perceptions of competence, and exclusion from industry networks. White men are still more likely to start businesses and make more money.


* This article originally misstated the name the defendant in United States v. Bhagat Singh Thind.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.