by Alex Wagner
The best and last hope for science to redeem my increasingly harebrained idea was my father’s Family Tree DNA’s results:
98 percent European (!)
2 percent “trace” DNA
That broke down to:
50 percent Western and Central Europe
35 percent British
10 percent Scandinavian
3 percent Southeastern European
As far as that “trace” DNA went? Family Tree specified that it was likely from Finland…although it might just be, as the website stated, “background noise.” Before I unpacked that hazy estimation, I stopped first to consider the fact that my father was apparently 10 percent Scandinavian. This was considerably more Scandi blood than either of the other tests had shown—so much more, in fact, that it was clear someone had to be wrong. And then there was the 3 percent Southeastern European blood—might this suggest Jewish blood, after all? Maybe this was the test I’d been waiting for. I held tight to my American Express, convinced against all reason that if I put down just a few hundred dollars more on another, fancier test, fate would deliver me the thing I craved. (I could now understand the mindset of Powerball obsessives who’d drive to middle Jersey just to buy a ticket from the convenience mart where the last winner had bought his. It was an obsession, perhaps. Or maybe I was being thorough?)
Family Tree DNA offered an extra-credit, super-specific test of the male Y chromosome line that would return the ancestral origins of my father’s paternal line, and could be conducted on existing samples—that is, without having to ask him for yet another spit test. I forked over a couple hundred dollars and went for it, pulling the lever on the DNA slot machine like a Vegas newlywed plowed on White Russians.
The results were so scientific that it was nearly impossible to decipher; there was mention of haplogroups and SNPs and charts with “marker values,” none of which I could make heads or tails of. But I did gather that my father had two exact ancestral matches on his Y chromosome: ones from Germany and the United Kingdom.
Once I’d gotten someone on the phone to help me understand these very clinical results, I learned that, apparently, my father was less related—but had similar ancestry—to men in Western Europe, including Belgium, Denmark, France, Ireland, Italy, the Netherlands, Scotland, and Spain. They could offer no specificity on all that Scandinavian blood, which I vowed to take up later (and did). Ultimately, here was more Western European DNA that didn’t really signal anything as far as a narrative-smashing revelation. Despite the percentage of Eastern European DNA in the earlier test, no Jewish DNA matches were returned (something this test could actually scan for with some precision). Unequivocally, it seemed, Carl Wagner had no Jewish DNA on his father’s side. The genetic profile of my paternal line spoke nothing of a complicated or confused narrative. We were what we had always said we were: white, Christian Europeans.
What did the collective lab results have to say about my own saliva and cheek cells? To be fair, I had probably been most interested in what the test results said about my mother’s and father’s genetics—their DNA went back further in time, and could therefore (potentially) unearth more interesting discoveries. And yet the narcissism of the project was a siren song: Maybe, just maybe, something dramatic would be revealed. Some bloodline that had escaped detection in previous generations—and this would be the genetic triumph I’d been searching for.
When it came to me, ole futureface, 23andMe told me I was:
37.5 percent Southeast Asian
22.1 percent French and German
21.2 percent British and Irish
5.7 percent Broadly Northwestern European
5.4 percent Broadly South Asian
4.7 percent Chinese
1.3 percent Broadly East Asian and Native American
.9 percent Broadly Southern European
.4 percent Broadly East Asian
.5 percent Unassigned (!)
.3 percent Broadly European
.1 percent Balkan
and less than .1 percent Central and South African
While AncestryDNA determined that I was:
29 percent East Asian
14 percent Scandinavian
13 percent British
10 percent Central Asian
9 percent South Asian
8 percent Irish
5 percent Polynesian
4 percent Western European
3 percent Italian/Greek
2 percent from the Iberian Peninsula
1 percent Melanesian
Plus a few traces of North African, Eastern European, and Northwest Russian/Finnish
And Family Tree DNA’s Family Finder test estimated that I was:
32 percent Southeast Asian
13 percent British
11 percent South Central Asian
10 percent Scandinavian
8 percent Iberia
8 percent West and Central Europe
7 percent Southeast Europe
4 percent Finland
3 percent Northeast Asia
2 percent Siberia
Plus more “trace” percentages from Oceania and East Europe
Unlike my mother and grandmother, I showed no specific Mongolian DNA. One test showed me to have a hefty dose of Irish (8 percent), while one another found that I had nearly the same amount of DNA from the Iberian peninsula. One test said I was 5 percent Polynesian (“Aloha, hapa!”), but the other two found no Polynesian anything. Beyond the broad breakdown of “Asian” DNA to “European” DNA, it was a toss-up. There were wild variations in these smaller percentages and completely different assessments as to what my genetic makeup was, exactly.
And then there was the mysterious “Unassigned” or “Trace” category—which appeared on my DNA results, as well as my mother’s and grandmother’s. This was apparently the Ziggy Stardust of DNA: You were an extraterrestrial creature from the genetic version of Mars. This DNA was unlike most of what they’d observed before, akin to a rare cactus or triple-winged bird—or it was suspiciously dubbed background noise. It also appeared to be the opposite of the whole point of this exercise—and I suspected the underlying reason was that the test makers hadn’t yet developed a sufficient answer key.
The discrepancies and super categories and unknowns rang a few alarm bells. It looked like each company might want to put the “unassigned” disclaimer on more than one of its categories. Clearly something was amiss if I was being told I was nearly a tenth Irish and a twentieth Polynesian and maybe a little bit Native American and also…none of these things at all.
Which brings me to the lingering Scandinavian question: AncestryDNA asserted I was nearly 14 percent Scandinavian (and why not? I had always had a particular taste for brown bread and socialized healthcare), while Family Tree DNA determined that I was 10 percent Scandinavian and 4 percent Finnish. But this was also, weirdly, nearly as much Northern European DNA as my father had, if not slightly more—which seemed questionable at best, given the fact that my mother had approximately 0 percent Scandinavian or Finnish blood. And then there was 23andMe, which concluded I had no Scandinavian or Finnish DNA whatsoever. What was going on here?
Presuming the companies stood by their results,* how did they manage the angry or confused customers who didn’t believe—or didn’t want to believe—what the vial of spit or swab of cheek was telling them? Surely I was not the first person in history to receive an ancestry report and find the results less than convincing. Was there some sort of crackerjack counseling team at the ready to soothe the descendant of an Argentinian elder who was being told his blood was really Norwegian Irish? A hotline to call when a report spelled out not Scottish DNA, but southern Italian from Napoli?
As it turns out, there was.
A few calls and emails got me a contact at 23andMe who worked in corporate communications. After an upbeat if mostly inconclusive phone call with Joanna Mountain, the company’s senior director of research—who at one point declared, “It’s all about discovery, and discovery is a plus!”—I was connected to two customer support representatives who dealt specifically with ancestry concerns, the very soldiers on the front lines of any DNA identity crisis for whom I had been looking.
* Even with my limited knowledge of corporate America, I was pretty sure this would be the case.
Because 23andMe is a product of Silicon Valley, everyone who works for the company appears morally uncompromised and in excellent health. I know this because I visited the place for a few precious minutes during the course of my search for the truth. It is a place with free frozen yogurt all day long and snacks that are high in omega-3s; there is a gym where all the employees lift weights together and cheer one another on to lift even more weight together; where the roof deck is outfitted with brightly colored plastic chaise longues in which employees sit and watch the sun set over the California mountains. It is a relentlessly optimistic place, a sharp contrast with some of the fairly serious and complicated questions that arise as a consequence of their services on offer.
When I got on the horn with the two support representatives who’d been delegated to speak with me, Cameron Kruse and Shawna Averbeck, both tamped down the suggestion that the calls they fielded were of any particularly controversial nature. They were well versed with this specific brand of professional reassurance—and explained that “empathy role-playing” was part of their training, though not all of it. “You pick up empathy nuances,” said Kruse, “but being straightforward is key.”
“Usually when people come to us, they’re still in shock and disbelief,” Averbeck explained. “So what we do is just try to answer their questions.”
“We’re there to be part of the process,” Kruse added.
According to them both, there are stages of acceptance that people experience when a DNA test comes back with unexpected results. At first, the science is usually questioned—and, according to Kruse, “we explain that we’re very confident in the data we return. If someone gets Ashkenazi results, we can fully explain how.”
(I didn’t get Ashkenazi results and wanted to know how this could be, but I’d save that for later.)
Once the customer has accepted the validity of his or her ancestry results, user-generated forums (basically genetic chat rooms) are the next step in the process of coming to terms with their new selves. “Talking about options and possibilities with other peers—that’s a really important second step,” said Averbeck.
Both representatives conceded that the testing process can make people feel vulnerable and “opened up to the core” but maintained an upbeat assessment about the ultimate result. “As people accept their results and how valid they are—and as they get genetic perspective—they become proud,” Kruse declared.
Surely, some discoveries were easier to accept as “plusses,” either incidental to a family’s story, or more readily donned as symbols of pride. But what about results that challenged generations of accepted history? Assuming the results were accurate, learning something irrevocably life-altering about one’s people seemed like it would take more than a chat room to come to terms with, and instead would require a certain deftness and decency—as much as empathy. According to the two I spoke with, the handful of employees in this specialized line of customer counseling were selected based on their personalities: “If they’re empathetic or curious, or if they like puzzles, then they would probably be great,” according to Averbeck.
“We do puzzle work behind the scenes,” she said. “It helps target the response.”
Puzzles? I wondered. Were the counselors doing crosswords in between calls?
“We work behind the scenes to help understand [someone’s results],” Kruse clarified, “but we don’t typically tell customers what we determine, because we aren’t positive. It’s tempting to tell them, but there are repercussions to all that. There are consequences.”
“Consequences” sounded ominous. This was the first time the cheery emissaries from 23andMe had offered a suggestion of the perils inherent in their work—unspecified “consequences” traumatic enough that the counselors and puzzle solvers at the company opted not to reveal them to unsuspecting customers.
“You have to know your role,” concluded Kruse. “And 23andMe prepares us really well for that.”
How confident were these counselors in their results? Kruse said that in his line of work, ancestry concerns “ran the gamut,” but he said calls about Native American genetics are common and sometimes necessitate certain sleuthing.
“Someone might be confident that they have Native American ancestry—and then the genetic test doesn’t reflect that,” he explained. “Maybe there were tales in their family history about Native American ancestors, and then [those tales] turned out not to be true.”
Kruse was confident that whatever data 23andMe was using behind the scenes to determine this kind of ancestry was the right stuff: “We have a good Native American reference population,” he assured me. “If there’s a Native American in there, it should be reflected.” (As long, a company representative later stipulated, as the Native American was not “very far back” in the generations.)
But what, exactly, was a “Native American reference population”? I kept hearing about these so-called “populations,” and while I didn’t expect that each of these companies had actual Native Americans in the lab helping to analyze spit samples, I also had no idea what this meant practically. In fact, I had absolutely no idea how any of this worked.
What I did gather from this optimistic and efficient phone call—before I hung up the phone, marveling at my ignorance—was that whomever these Native Americans were…23andMe believed they were definitely getting the correct data from them. I suppose that it wasn’t surprising that an emissary with a bellyful of frozen yogurt and omega-3 fatty acids, a man with access to a winning California sunset every night, would be so confident—as if determining whether there was “a Native American in there” was akin to making sure there was a tomato in your BLT: Pick up the bun, lift the lettuce, look for the tomato, and it’s either in there or not. Simple as that, dude.
Obviously, it wasn’t that simple.
* * *
—
With the BLT metaphor stuck firmly in my head, I realized I had absolutely no idea what the bacon, the lettuce, or the tomato represented in this particular field of study, or where these sizable and presumably very accurate populations of Navajo, Iroquois, and Sioux tribespeople were being housed. Though my saliva and cheek cells had been priority airmailed to California and Utah and Texas, I had not the faintest clue what these companies were looking for or how they interpreted their results. Basically, I knew nothing. I’d blindly assumed science would work it all out and give me the answer, and that the answer would remain uncontested because, hey, what the fuck, it was science. And I believed in climate change.
I began trying to unravel what exactly was going on by starting with the type of tests themselves.
There were multiple references to “Y chromosome” and “mitochondrial DNA” testing on all the ancestry websites. I had made my father take one of those Y chromosome tests. I remembered from biology class that men had X and Y chromosomes, and women had two X chromosomes.*1 So men and women both carry X chromosomes (even though men, unlike women, don’t pass the X chromosomes on to their children), so it turns out that male ancestry is mostly tested via the Y chromosome—which passes down genetic traits from father to son, ad infinitum.
Women, including genetic detectives like myself, are tested using mitochondrial DNA—which reveals genetic traits passed down from mother to daughter in the same way that Y chromosomes reveal genetic clues inherited along th
e paternal line. (For the record, mothers also pass their mitochondrial DNA to their sons, but those sons can’t, in turn, pass it to their children. It travels with moms only!)1
In other words, the primary—and most reliable—DNA tests (Y chromosome and mitochondrial DNA) reveal details about either your father’s father’s line (if you’re a man) or your mother’s mother’s line (if you’re a woman). By all accounts, these kinds of tests provide conclusive evidence if you need an answer on something such as paternity or maternal health, or, in my case, your father’s father’s father’s potentially Jewish DNA.
But if you want something more comprehensive, like, say, ancestry through the generations, there are significant blind spots if you use only these two tests. A mitochondrial DNA test will give me reliable information on my mother’s mother and her mother and her mother before that. But the further back in generations you go, the bigger an issue this becomes. Not only will this test leave out all the fathers and grandfathers from back in the day, it will also leave out any information about the mothers of all those fathers and grandfathers, as well as the fathers and grandfathers of all those mothers.
I had to draw out a chart to confirm just how many ancestors this would exempt, and as it turns out, that number doubles with every generation. By the time you got to my great-great-grandparents (the parents of Henry Wagner and U Myint Kaung and Daw Thet Kywe) a mitochondrial DNA test would return information on only one out of sixteen of them.
Like so:
So this was a problem.
Scientists recognize this flaw and, in recent years, have developed what’s known as autosomal testing, and here’s where the story gets real interesting. (This is where the BLT becomes a lot more like a triple-layer hoagie.) Autosomal testing essentially scans the genetic material found on chromosomes one through twenty-two. (The X-Y stuff is what happens on the twenty-third chromosome.) There’s a ton of DNA material inherited from both parents on these chromosomes, so the process effectively searches for what are known as genetic “markers” of certain ethnic groups.