Tuesday, December 1, 2015
A Survey of Surnames
Ever wonder how far and wide your ancestral surnames have been circulating?
As I trawl through my nine-hundred-plus matches on Family Tree DNA, I try to find surnames held in common with those supposed distant relatives. Of course, I'm hoping to light upon the very name which, shared, will be our most recent common ancestor. That, however, turns out to be the elusive mystery person, the one who refuses, seemingly, to be uncovered.
In the meantime, I see lots of other commonalities among the surnames of my DNA matches—names like Smith and Jones and even Taylor. Those, as it turns out, are hints not worth pursuing.
So I started wondering just which of all those "common" surnames might be worth pursuing. I went looking for a scorecard of surnames—at least those in the United States, if nothing else.
I did find one on Wikipedia, giving the ranking for surnames currently found in the North American continent, itemized by countries: Canada, Mexico, Cuba among them. For the United States, the top one hundred surnames gleaned from the 2000 census were listed by rank, along with their comparative rank in the previous enumeration.
I found out that the top seven surnames in America happen to be among those in my ancestry, too. Perhaps that explains why I'm having such a hard time matching up with my Davis ancestral line—ranked seventh for the 2000 census, down from sixth place for the previous decade. Or why it's even fruitless to pursue any matches with descendants of the Taylor surname—now slipped down to thirteenth place from its top-ten category in 1990.
Surely, I think, there are some of my surnames snagging a more reasonable ranking—at least for cousin matching—and take a look farther down the list. Remembering my discovery of relationship to the famed Meriwether Lewis of Lewis and Clark renown, I'm amused to spot the relative ranking of the surnames Lewis and Clark—with Clark at rank twenty five besting Lewis by one spot. Carter and Phillips, two additional family names in my maternal tree, also snag contiguous spots, sitting at forty six and forty seven, respectively.
But who am I kidding? All of the top one hundred surnames in America represent seventeen percent of my fellow Americans. That's almost one fifth of the people in this country.
On the other side of what surely must be shaping up to be a bell curve of surname distributions, about fourteen percent of Americans have a surname which occurs fewer than one hundred times in the entire population.
So, what might those other surnames be, I wondered. I headed to the United States Census Bureau to see if they could provide any further enlightenment. There, they provide a downloadable readout of rankings—again based on the 2000 census, but this time extending the list to the top one thousand surnames.
You might think that a thousand surnames would give a researcher plenty of wiggle room. But that isn't what happened. I noticed quite a few of my family's surnames didn't make it on even this list—mostly, names from my father's side of the family, like the elusive Puchalski, but surprisingly also Laskowski. And Taliaferro, that surname that gained me entrance into the Daughters of the American Revolution? Not even in the Top Thousand.
That Top Thousand list managed to steal my attention for not a few minutes. That handy "Find" function got a workout, fetching—or coming up empty, as the case may be—surname after surname from my genealogy database, checking out how high or low each one's ranking turned out to be. I'm a sucker for numbers that way.
However, if I wanted details on the years prior to the 1990 census, my former go-to website—a collection no longer found online at its former www.hamrick.com URL—needs to be replaced by another resource. For now, I'm settling for the surname resource at Ancestry.com, but I'd sure like to have the long view of surname histories I now only have in my notes from years ago. I'm curious to see how some of these auxiliary lines stack up over time—like my relatively rare Taliaferro.
Of course, just because a shared surname is much too common doesn't mean that my chances of finding a nexus with a DNA match won't happen. It's just that it's less likely. Perhaps the lower down the ranking list a surname occurs, the more likely it might be that someone else will match, but I'm not sure, right now, that I can claim that. Those are the kinds of speculations best left to the statistical wizards among us.