The widely-repeated assertion by the Icelandic firm deCODE genetics that James Watson is 16% black and 9% Asian (see, for instance, the new NY Times article "DNA Pioneer's Genome Blur Race Lines") reminds me of one of the least understood contradictions in the conventional wisdom that Race Doesn't Exist:

- The existence of the One-Drop Rule shows that race is an arbitrary social construct.

- Therefore, lots of white Americans must have lots of black ancestors.

But when you stop and think about it, you realize the opposite is true: that the One-Drop Rule is the reason that so few self-identified white Americans have much black ancestry. As I wrote in 2001, when racial admixture testing via DNA was in its infancy:

Among self-identified whites in Shriver's sample, the average black admixture is only 0.7 percent. That's the equivalent of having among your 128 great-great-great-great-great-grandparents (who lived around two centuries ago), 127 whites and one black.

It appears that 70 percent of whites have no African ancestors. Among the 30 percent who do, the black admixture is around 2.3 percent, which would be like having about three black ancestors out of those 128.

In contrast, the lack of One Drop Rule meant that Mexico's black minority has been almost completely absorbed into the general population.

As I've said, racial admixture testing is not always reliable for individuals, but for large sample sizes it works reasonably well. (If anybody has any more recent data than this on American whites, let me know.) As I pointed out in regard to IQ testing, people tend to make a 180 degree wrong assumption about testing in the human sciences: the unexpected reality is that it's much easier to be accurate about a group, whether IQ or racial admixture, than it is to be accurate about an individual.

It was fairly hard to pass visually, but the emotional toll of passing was particularly difficult. A 1953 study by anthropologist C. Stern estimated that 1/4 of people who were 1/4th black and 3/4th white could pass for white. (See Carleton Coon's Living Races of Man, p. 307).

The One-Drop Rule made it wrenchingly hard for even the whitest-looking person with socially-identified black relatives to pass into being socially-identified as white. To pass from black to white socially, an individual typically had to move to a new part of the country and cut himself off from his family because at least some of them would be visibly part-black.

For example, one of the best-known cases of passing is that of the late Anatole Broyard (1920-1990), the distinguished literary critic. His parents were New Orleans "creoles of color," but when he moved to New York to make his career in books, he more or less dropped the black part of his black identity (which, as a native of New Orleans, where the One Drop Rule was an alien Anglo imposition, presumably didn't mean that much to him) and let people assume he was white. His career probably would have been even more successful if he had been publicly black, but he wasn't interested in being pigeonholed as a "black critic."

But this liberation came at a human cost: he cut himself off from his family. His children never met his darker sister until his funeral.

Broyard championed the novelist Philip Roth, and after his death, Roth published a novel, The Human Stain, inspired by Broyard's life.

The 2003 movie version suffers from the casting of Anthony Hopkins as the protagonist, Professor of English Coleman Silk, because Sir Anthony, the laziest of actors, made no effort to appear even subliminally black (he didn't even use an American accent!). And the filmmakers didn't dare put any makeup on Sir Anthony to make him look a little black. But the flashbacks to Silk's life in Newark in the 1940s before passing, featuring the part-black Wentworth Miller of Prison Break as the young Silk, are excellent. (By the way, if anybody wants to make a movie of Broyard's life, Miller looks a lot like him. And, he's got star power.)

Roth's novel makes clear the emotional cost of passing, when the young Silk's clearly part-African mother's responds to his announcement with this moving soliloquy:

"'I’m never going to know my grandchildren,’ she said. ‘You're never going to let them see me,' she said. ‘You're never going to let them know who I am. "Mom," you'll tell me, "Ma, you come to the railroad station in New York, and you sit on the bench in the waiting room, and at eleven twenty-five A.M., I'll walk by with my kids in their Sunday best." That'll be my birthday present five years from now. "Sit there, Mom, say nothing, and I'll just walk them slowly by." And you know very well that I will be there. The railroad station. The zoo. Central Park. Wherever you say, of course, I'll do it. You tell me the only way I can ever touch my grandchildren is for you to hire me to come over as Mrs. Brown to baby-sit and put them to bed. I'll do it… I have no choice.'"

And then there was always the fear among individuals who were passing that they'd have a child who was clearly part-black. (Your child can inherit from you genes that aren't evident in your looks.)

But, if you can successfully pass, your descendants will tend to be increasingly white by ancestry, while the descendants of your siblings' who didn't pass will tend to get blacker because they will be in socially different gene pools for choosing spouses. (For example, there are, I believe, two lineages descended from Sally Hemings's 1/8th black sons: Madison's is socially identified as black and Eston's as white.)

So, with a reasonable picture in our heads of just what was required of an individual to pass, let's see how credible the claim that James Dewey Watson Jr. is 25% nonwhite now sounds. I'm going to spend some time going over this because it might help people understand how to evaluate genetic claims (by seeing, for example, if they make sense in human terms of who marries whom), and because it explains a little about what America was like.

The first chapter of Watson's autobiography Avoid Boring People is available as a five megabyte PDF file, complete with pictures of his parents and maternal grandmother.

On his mother's side, his grandfather was a Scottish immigrant, while his mother's mother (see picture to the right) was the daughter of Irish immigrants who initially took up farming in the Midwest. So, the search for blacks and Asians should concentrate on his father's side of the family, who were of English stock.

But that would mean his father would be 50% nonwhite, and one of his paternal grandparents would likely be 100% nonwhite.

One place to start is by looking at the photo (not online) on p. 265 of Watson's new autobiography, Avoid Boring People. It shows Watson at the 1967 wedding of his cousin Alice. Standing alongside him are his sister, his father, and his paternal grandfather. In other words, the Watsons were not split up like the Broyards were.

Further, just from looking at the picture, I'd say these Watsons are just about the five whitest people in the whole world.

The reality is that the Watson family was way too socially fashionable for too long to be significantly black in a profoundly anti-black America. For example, Watson's paternal uncle William Weldon Watson IV was appointed chairman of the Yale Physics Department in 1940. If somebody who was one-third black was a Yale department chairman in 1940, it would be big news.

Watson's father (see picture to the left) started work at the Harris Trust Company in Chicago before WWI. Watson's paternal grandfather was a stockbroker and his paternal grandmother an heiress. The scientist's paternal great-grandfather was a hotelkeeper in ritzy Lake Geneva, WI and married a banker's daughter. His paternal great-great-grandfather was a friend and prominent supporter of Abraham Lincoln.

You could hypothesize that Watson was the product of an illicit affair between his mother and a man who was half nonwhite, or between his paternal grandmother and a man who was completely nonwhite, but the circumstantial evidence makes this unlikely. Watson was the first-born child, born three years after his parents wedding. His parents had his sister a couple of years later and stayed together for the rest of their lives. So, it doesn't sound like Mrs. Watson stuck Mr. Watson with a cuckoo's egg.

Similarly, Watson's father was the first-born of four sons, a couple of years after his parents' wedding. And he was born in northern Minnesota!

Or you could hypothesize that James Watson had several different ancestors who were all part non-white, but that's just pushing the passing problem back farther in time, and multiplying the improbability of it all.

Now, it's quite possible that Watson's distant frontier-era ancestors include blacks and American Indians (I'm dubious about the 9% Asian figure). When people were moving around and communications were slow, it was easier to pass. But, their descendants would tend to get whiter because they had passed into the white marriage pool.

So, what likely happened is that Watson had a few nonwhite ancestors fairly well back in the past, and their versions of the genes used as genetic markers in deCODE's analysis kept, via the luck of the draw in the sexual reproduction shuffle, kept coming up, greatly exaggerating his overall nonwhite ancestry.

Enough detail. The point is that when you think about genes, you need to think about genealogy.