The problem with family trees

My least favorite story-that's-occupying-too-much-media-attention this week is the Al Sharpton-Strom Thurmond family-ties affair. As a former newspaper editor, I know all too well why the story of the uncovering of the connection between a prominent African-American civil-rights leader and famous white racist senator is irresistible. But should it really be more than a mildly amusing footnote, rather than a front-page feature? And now we hear that Sharpton wants a DNA test? Give it a rest.

First of all, we're not talking about blood relations. The story goes that Sharpton's great-grandfather, Coleman Sharpton, was owned as a slave, by Strom Thurmond's great-great-grandfather. They aren't even distant cousins, for pete's sake. One of Thurmond's distant cousins owned one of Sharpton's ancestors.

Mike Ward, one of the genealogists with Ancestry.com, which found the link, reportedly called it "amazing." Which is itself amazing, because an alleged expert in genealogy should be anything but surprised when this sort of thing pops up.

Do the math. Go back five generations and you'll find a lot fewer than six degrees separates most of the country -- probably more like one or two at the most. I remember reading a piece in Harper's a while back in which the author was most distressed to learn that statistically, just about everyone of European ancestry can squeeze out a genealogical connection to Charlemagne. (Here's a brief post pointing out that with 67,108,864 direct grandparents, how could one not be related?)

The real reason this story has so many legs is that so many people are obsessed with genealogy. Next to porn, it's the leading raison d'etre for DSL connections. Given that so much of who we are is determined by our genes (roughly 60-40 in favor of nature over nuture, according to the prevailing consensus it would seem), this should not be a surprise. But it turns out that merely tracing your family tree is anything but a reliable method for deducing where and whom you come from.

For one thing, records in much of the world are incomplete and inaccurate. That's one reason why I have never really taken a big interest in tracking the history of my father's side of the family. The name Hrynyshyn isn't all that common, but record-keeping didn't seem to be a big priority in rural Ukraine back in the day. I can't get much past the great-grandfather stage. (On my mother's side, we think we can go back to the early 17th century -- something to do with a crony of Oliver Cromwell and a castle in Northern Ireland, but it's all very murky.)

Second, there is the inconvenient truth that a certain percentage of children are the result of undisclosed extra-marital affairs. As Carolyn Abraham of the Globe and Mail wrote four years ago in her article "Mommy's Little Secret"

It's now widely accepted among those who work in genetics that roughly 10 per cent of us are not fathered by the man we believe to be dad. Geneticists have stumbled upon this phenomenon in the course of conducting large population studies and hunting for genes that cause diseases such as cystic fibrosis. They find full siblings to be half-siblings, fathers who are genetic strangers to more than one of their children and uncles who are much closer to their nieces and nephews than anyone might guess. Lumped under the heading of "pedigree errors," these so-called mis-paternities, false paternities and non-paternities are all science jargon for the unwitting number of us who are chips off someone else's block.

Ten percent may not seem like a big number, but compounded errors, like interest, can produce some startling results given enough time. After four or five generations, that's a lot of stray DNA floating around in the family gene pool.

Finally, as Steven Pinker pointed out in his wonderful book, The Blank Slate, even when it comes to the nurture factor, the primary determinants of one's character aren't even parental. Peers, it turns out, have a stronger influence on how we turn out than dear or mom and dad. On average, of course.

Gene-tracking does have its uses. You might want to know whether you're susceptible to certain inherited diseases, for example. That sort of thing is going to pose great ethical challenges in the coming years as DNA sequencing gets cheaper, and insurance companies get more interested in evaluating genetic risks.

But for anyone out there interested in uncovering your roots, it probably makes more sense to flip through your high school year book than go digging through government records.

The Reverend can settle down now. Even if it does turn out that one of the late segregationist senator's ancestor's sired one of Sharpton's, so what? We already know Thurmond's philandering ways produced a daughter with his African-American housekeeper. Do we really another DNA test to remind us that we're all the same under the skin?