My Ancestry.com DNA Test: Part 2 (Genetic Ethnicity)

For (sick, nerdy) grins, before my DNA results were in, I made a spreadsheet to try to predict my genetic ethnicity. But first, I did a bit of reading.

In my experience, there are two things that seem to really confuse people about genealogy: cousins and percentages.

I could-should-will do a post someday on how many arguments I've weathered over figuring out cousins. "But he's older, so he can't be my cousin. He must be my uncle." (No.) "If she is my grandmother's first cousin, she must be my third cousin." (No.) "Everyone is probably 4th cousins with each other." (NO.)

But that's another post, one where we could talk about cultural labels ("aunt" and "uncle" as honorifics for all older relatives regardless of relationship) and how English is missing some cool nuances, such as distinguishing between the sister-in-law who married your brother and the sister-in-law who came from the same womb as your spouse. For now, let's talk about percentages.

People say things like, "I'm a quarter Irish, a quarter Polish, and half Italian." It's a nice shorthand way of sharing your cultural influences and identity, especially if you're an American with ancestors who immigrated recently.

But the problem happens when people take their cultural identity as a sort of genetic gospel. I've been reading discussion forums on Ancestry and comments on genealogy blogs about other sites offering DNA tests, and I - a newb myself to these things - just want to bang my head. Over and over I see something like this:

"I had my test done with Company X and it was a RIP-OFF! Don't use them! My great-grandmother was a full-blooded Cherokee and my Mom's side of the family is 100% Swiss, and neither of these showed up!"

A great blog is The Genetic Genealogist, and their "Everyone Has Two Family Trees" post is a must-read. (Please read it if you really care about such things because I'm about to do some simultaneous over-simplifying and rambling convoluting. It's a gift.)

My argument is that many of us actually have three trees.

Tree #1: Cultural Identity

When I was growing up, I asked my mother what we were. Living in the suburbs and, later, exburbs of Detroit, I felt surrounded by a variety of geography-based cultural influences. Polish, Scandinavian, Italian, Russian, French, New York (heh)... (It took a teaching career in an 80 % Hispanic school in Nevada to "learn" that these are all just boring, bland, blank shades of "white" and thus not valid for inclusion in a discussion on multiculturalism. GRRR...) But as far as I could tell, I really was "just white." I couldn't detect a cultural identity in myself based on location. My parents spoke with Southern accents, and none of my friends knew what gumbo was, but otherwise I was a Midwestern girl, a non-participant in my people's rituals except during summer visits to Texas, when things would get really Jesus-y.

My mom said told me I was English, Irish, French, German, and Indian. (We said Indian back then. You know what we meant.) Okie dokie.

It felt right. I didn't ask how she knew. Over the years I became very attracted to different aspects of English and Irish culture, but I don't think it's because I had some notion that they fit into my story. France, too, but less so because of the language barrier. German? Not really. (Is it because I didn't take German in high school? Never saw BOOP! filmstrips BOOP! of its famous landmarks?) Indian? It was the most romantic prospect of them all, but I had some unfocused idea that probably every American had a bit of Indian in them. (If only.)

Weaksauce as it was, that was my first tree. For other people, the links to the old country are a lot stronger. Their paper or DNA genealogies may say something different, but they feel connected to the food, customs, language, and so forth of whatever branches their cultural tree has. It's kind of like religion. You could have three Catholic grandparents, but if they're all dead and you only spend time with your Lutheran grandfather, you're going to feel much more Lutheran.

When I got into genealogy and was pumping my mom for specifics on how she knew we were English, Irish, etc., I was disappointed. It turns out she was mostly guessing based on surnames, possibly carrying on speculation of those before her. Her great-grandmother would put on a brogue at her request, so Mom had assumed she was Irish. (Note: She was born in Arkansas and was at least a 4th-generation American.) Stories about Native American ancestry flitted about, but as years passed some of those stories turned out to be about people who married aunts or cousins, not our ancestors.

In short, nobody really knew. When I started doing genealogy, I was repeatedly warned that there was nothing to be known beyond a few generations, led to believe that my family faded away from Texas, Louisiana, and Arkansas into the Kentucky/Mississippi/Missouri mists, their origins forever lost.

My great-granduncle (Pat, mentioned in the last post), was considered the best hope. "My mother was very English, very proper," he said, idolizing the woman who'd died when he was a child. "My father was Irish. He ran away and sailed to Ireland and learned all about our family, the Irish ancestor who took his wife's name to escape the law."

For five years or so, I heard through the grapevine that this was a true family story. Then I actually met Pat, and we became besties, and he admitted that maybe he'd embellished or misremembered some details over the years. My research showed that if either his mother or father were very English or Irish, it was several generations back. (But I did finally find an Irish link before he died. His father's mother's mother's mother was described in a census as being born in Ireland, but her name remains unknown to me. AND she married a Smith. Arrrrrrgh!)

In short (way too late), Tree #1 is your emotional attachment to geography. I understand why people are reluctant to let it go, even when nothing in a DNA test can invalidate your cultural identity.

 

Tree #2: The Map of Paper Trails

The second tree is the one that says Grandma was born in this year, in this place, and she married this guy, and they lived here, and here are their children, and these were their parents - repeat as far as possible in every direction, hoping to fill in the beautiful mural of the context of you as time goes on.

Since my family couldn't pass along non-USA geography-centered traditions (not to disparage the gumbo or the Jesus), I couldn't wait to see what the historical records would say.

Well, I've said it before and I'll say it again. Bring on the chorus of dancing Springsteens because I (and the next five generations) were all boooorrrrrn in the USA, boooorrrrn in the USA. (And that greatx4 grandfather born in England was a bare blip on a boooorrrrn in the USA radar.)

As I mentioned at the top of this post, I tried to predict what the ethnicity results might be for my DNA test, and to do this I looked at my paper trail. By the eight generation, when I only had a few known non-USA ancestors, I turned into my mom and started making guesses based on surnames, or on those ancestors I knew of further up the tree (fractional percent of the tree though they may represent).

The best I could do was this:

Original predicted ethnicity

(I tried to use the same groupings as Ancestry.com, thus "British Isles" instead of English, Irish, Scottish, Welsh.)

It's funny, though, because in my research, which nationalities have showed up loud and proud first? English. Irish. French. German. My mother always did have good instincts.

But when I start looking at the bigger picture, so much is still unknown. That's why, once again, I have to shake my head at those people who curse their DNA tests for being "clearly" inaccurate. I've looked at 600+ trees over the past week, and - not to be boastful about something that often comes down to luck - my ancestral tree goes back further than most. (We're talking documented ancestors. Don't get me started.)If I have that much "unknown" in my tree in the first eight generations when those first eight generations are mostly complete, less-complete trees are full of at least as much possibility. It's not because I have all those USA people, either. Migration and intermarriage are not a USA exclusive.

(Of course, those people's DNA tests may very well be inaccurate; I'm just arguing against assumptions based on what percent ethnicity people perceive to be in their trees. It's easy to forget how fast just a few blank spots can multiply.)

Despite all of my reading on the fallability of DNA tests (or rather, the interpretation of DNA tests), and despite those blanks in my tree, and despite having a firm idea of how culture and historical records and DNA were three separate things, I still "felt" like I knew what my results would probably be. Probably a lot of British Isles and Central European with a touch of Scandinavian, right?

Wrong.


Tree #3: You and Only You

Unless a person is an identical twin, siblings don't look just alike. They inherit different traits (or don't) in differing amounts (or not at all) from each parent. So why do people expect to have inherited all of their ancestors' DNA? Probably for the same crazy reason that people act like they know their entire tree and overlook the influence of empty branches. The emotional attachments developed in Tree #1 stay with us.

The aforementioned Genetic Genealogist has another great post, this one discussing how you may take autosomal DNA tests with three different companies and seem to get three very different results, but it's not that the tests are different, just the interpretations.

As GG points out, different countries have different databases. Those databases are constantly being refined. (Ancestry points this out in the test results, meaning that as time passes, my ethnicity results WILL change.) Some entries in the database are imperfect, too. I can't find the article right now, but some of those entries came from people here in the States being gathered, tested, and asked to provide a four-generation genealogy. Well, who's to say that this person's genetic input didn't mostly come from a few empty branches in the fifth generation? They're reporting "England, England, England," unaware of, say, an Egyptian great-great-grandmother. The more people who test, the more such anomalies are rustled up and properly re-categorized.

Understanding as much about migration as possible is important, too. Expecting French but your results say Scandinavian? Blame the Vikings. If your results say Italian, blame the Romans. And so on. And those are just the big sweeps we know about. Small migrations happened all the time (look at how the USA was formed - we didn't invent the idea of moving elsewhere for a better life), right down to the far-off ancestor whose journey was never recorded for the annals.

(And then there's the DNA problem of the far-off ancestor's wife, who may have been a little too sweet on the local goatherd while her husband was at war. But none of us like to think about that because that's a brick wall built for the ages. So, shhh.)

The GG post mentions another post that says at the 10th generation, only around 10-12% of our ancestors in Tree #2 may show up in Tree #3.

I just created a 10-generation ancestor report for myself. That's 1024 theoretical ancestors, but my greatx4 grandparents were cousins, and who knows how many times other relatives married? So I don't know how many distinct ancestors are in the first 10 generations, but in my database I only have 357 names so far.

In other words, I only know ~35% of my ancestors through the first 10 generations. (Which is actually a good-size amount.) That means 65% of my family tree between me and my great-great-great-great-great-great-great-great-grandfather is a mystery. And that means that, even though the world was once a smaller, less connected place where people didn't move around as much, you can't be surprised by what shows up in your Tree #3. Sixty-five percent is a wide enough door for all kinds of unexpected people to slip through, and for all I know, most of my DNA is coming from that big 65% hole.

Delving back into Tree #2 kind of talk, by the 10th generation, these are the known countries in my tree:

  • Belgium
  • Canada
  • England
  • Finland (one 10th-generation ancestor)
  • France
  • Germany
  • Ireland
  • Scotland
  • Spain (one 10th-generation ancestor)
  • Sweden
  • Switzerland (one 10th-generation ancestor)
  • USA

In fact, these are the only countries anywhere in my tree, unless you count some early medieval locations.

So, imagine my surprise:

Ancestry pie chart

I don't recall which happened first: boggling over what it said or boggling over what it didn't.

The hefty presence of the British Isles is no surprise. (I hope someday they can categorize it further. If only Ancestry allowed you to download your raw data. Sigh.) But I definitely wasn't expecting the total lack of Central European (France/Germany) genetic markers. (I assume they're somewhere in the 3% Uncertain.)

But the big news is that there's a whole new texture of white bread on the table. Eastern European? Does this mean I can finally live up to my (Hungarian) name?

And an entire 7% Finnish/Volga-Ural, aka Russian? Is my one known Finnish ancestor behind this? (The spotlight swings to Hendrick Andersson, who arrived in New Sweden - aka Delaware - in 1654, a blacksmith who lived in the Finnish community there until his death 40 years later.)

Or does it come from my great x 32 grandmother, Anna of Kiev?

Back to the big chunk of Eastern European. Mike's theory: "You're part gypsy." Me: "We prefer the term 'travelers,' thanks."

Not to indulge in silly stereotypes, but sure, why not, could be. Could be lots of things. I wonder how close the connection is. I wonder which country. (Did you know that 99% of Estonia's population is blue-eyed? And all blue-eyed people are said to be related to a common ancestor from 6000 years ago?) Are we talking deep-East (maybe tying in to the Finnish/Volga-Ural), or just on the border?

Will I ever know?

05 August 2012 |

Previously: My Ancestry.com DNA Test
Next: Vanessa


Hamsters

 WE BUILT A HOUSE 

 RABBITS TOLERATE US 

 RECENTLY PLAYED 
 BOARD GAMES: 



 CRUISE REPORTS: 

Carnival Elation (2009)
Carnival Splendor (2009)
Carnival Spirit (2010)
Carnival Spirit (2011)
Carnival Splendor (2011)
Norwegian Pearl to Alaska (2012)