Pages

Wednesday, September 02, 2020

Finding my family's African roots

I've been looking into my family's genealogy for a couple of decades now so it was a bit of a surprise when my brother took a DNA test and found our ancestry 99% from Northwest Europe and 1% Western Africa.  The Europe part was expected.  I've been able to research most of my ancestors back to the year 1800, or about 5 generations.

Was it on Dad's side of the family?

The African part was unexpected but not totally unexpected.  Growing up my brother, sister and I always assumed we had African roots from my paternal grandmother who had very curly hair and darker skin than my other grandparents.  A DNA test from my Dad disproved our childhood theory entirely.  His DNA results came showed him more British than the Queen of England, proof that looks can be deceiving.

Was it on Mom's side of the family?

That left only one possibility but to prove it, I had my Mom take the test.  Like my brother, it also came back 99% European and 1% West African.

I expected it to be 2% but there are lots of reasons for this result.  These are very small numbers and so her percentage could be 1.4% and his 0.7%.  Rounding would make both 1%.  In fact basic math would prove this out.

If a person had one European parent and one African parent we can assume their child will have 50% European DNA and 50% African DNA.  If you assume their offspring had nothing but European ancestry after that the next generation would be 1/4, the 1/8, then 1/16, then 1/32, and finally 1/64.

If my Mother is 1/64, she'd have 1.56% African DNA.  The thing is DNA isn't always handed down 50/50.  Sometimes it is handed down 55/45 and once you've lost that portion of that DNA it's gone.  It is easily possible my Mom's African DNA is lower than 1.5% which is why she and my brother both showed at 1%.

Getting this information gave me a starting point and with my family tree in hand I was pretty sure I knew the answer.  We'd been told from a young age that we had Cherokee heritage on my maternal grandfather's side of the family.  Once I started looking into genealogy, I found this story is common in many families, especially in the South.  Most times when people look into the details they find this story was a cover to hide African roots.

Isabel Roberts

Unlike most families I had been able to find the person who it is claimed was native American.  Her name is Isabel Roberts, born in 1832 close by the Shawnee Reservation near Indian Lake in Logan County, Ohio.  As of now, I haven't been able to find out much other than her name and the fact she is supposedly the child of Arthur Roberts.  We know nothing definitive of her mother which certainly added to my suspicions.

As i looked into slavery, I found out many uncomfortable truths of how it was handled as a legal matter of course.  The laws were sometimes different by jurisdiction but the chains of bondage passed through the maternal line.  If a white man had a child with a black woman, their offspring was a slave to the black woman's owner.  If a black man had a child with a black woman, their offspring was a slave to the black woman's owner.  If a black man had a child with a white woman, their offspring was free.

The assumption was white women would never do this willing and many unexpected pregnancies of consensual intercourse ended with a lynching.  I haven't researched what happened to most of these children but I'd guess it depended on the 'mercy' of the girl's father.

Isabel claimed to be white in every census and having no other obvious subjects, I made an assumption Isabel was our most likely ancestor of African descent on my mother's side.  I later figured out this assumption was wrong.

The Leeds Method

I don't have a lot of experience working with the DNA side of genealogy but it has helped me clear a few roadblocks.  For instance, it helped me match back 4 generations from my Dad's DNA to his 2xgreat-grandfather and find the father of his grandmother.  It really is a great help.

One of the ways of arranging DNA information that helps adoptees is called the Leeds method.

When you get your results back from the DNA service you get a list of other people who have taken the test and match to you.  The services take this one step further and will let you know if that person is related to someone else on your list.

It's a little bit confusing so here's a simple example:

  • Paternal Grandfather A
  • Paternal Grandmother B
  • Maternal Grandfather C
  • Maternal Grandmother D

You are related to all 4 of the above.  Your father's sister and her children would be related to the A family and the B family.  They would not be related to the C and D.  A second cousin would only be related to one of the above.

The Leeds method has you separate all your matches so you put them into 4 catagories, depending on the relation to your grandparents.  A first cousin would appear in 2 categories.  My brother would appear in all four for my Mom.

I did this for my Mom and then took it a step further to match to her great grandparents.  I wanted to see if I could find a pattern.  The next thing I was see how many of my Mom's matches had any African ancestry.  Unsurprisingly, many did.  Surprisingly, none of them were descendants of Isabel Roberts.

Here is the breakdown of West African DNA results matched to my Mom's 8 great-grandparents:

  • Bennett - 14/33
  • Bable - 1/29
  • Boutwell - 1/37
  • Frankenberry - 0/6    (I want my cereal royalties!!!)
  • Keezer - 1/14            (this person also has Bennett DNA)
  • Peters - 0/17
  • Reed - 3/51
  • Trout -  0/25
The fact that I could do this shows how little families of different races mixed during the last 200 years.  I'm sure this is especially true of my family who were mostly rural white farmers living in Ohio.  All it would take is one descendant of one ancestor to marry outside their race over this entire span and it would easily add 1%.  You see the results above and because of them, I think it is safe to say the Bennett tree is the place to look.

(If you are wondering why the Bennett line isn't 33/33, the numbers are so small that every generation makes it less likely to find enough DNA to call it a match.  When results go under 0.5%, these services probably wouldn't show anything at all.  I suspect if my brother's kids took the test, they wouldn't get a West African result.)

Mom's family tree

To break this down a little further we need to see a simple family tree from my Mom's dad.

This tree goes back five generations and if you follow the Bennett-Boutwell-Boutwell-Roberts line you'll find Isabel.

I broke down my Mom's DNA comparison by those who connected to the Bennett-Keezer line and the Boutwll-Bable line.

All her matches who had African ancestry were in the Bennett-Keezer line.

This means the ancestor could be a Bennett or a Keezer if not for an unfortunate yet I'm sure scandalous happy event.  

Jenny Keezer died at age 52 and James Bennett being a spry man of 49 years, decided to remarry a 1st cousin, 19 year old Rennie Bennett.  They had four children and I found quite of few of their ancestors matched to Mom.  All of them had twice as much African ancestry as her.

This proves the African ancestry came from this Bennett line and not the Keezer line.  If it came from the Keezer line, they'd show no West African DNA.  As it is, the Bennett-Bennett descendants have twice as much West African DNA as the Bennett-Keezer kids because of the cousin marriage.

My theory is further proved by the fact that none of my Mom's Keezer-only relations have any African ancestry (there were 3 of them who've take a DNA test at the time of writing).

Bennett-Linnabary

I am fairly confident my ancestor with African heritage comes from the Bennett-Linnabary line.  Can we go further?

Yes we can but it's at this point the evidence become a little more spotty.  I found two Linnabarry-only ancestors and neither of them showed any African ancestry but that's not enough evidence to use as proof.  When you are talking about 1% amounts of DNA, it isn't uncommon for the tests to show no result at all.  

Relying on the Bennett-only names is problematic because they could be from just about any generation.  Still, we do see many of them with African ancestry adding to a suspicion it came further back on that line.

To prove it definitive with DNA you'd need to find someone on the maternal line - daughter to daughter and have them take a specialize DNA test.  As we don't even know the person's name we'd need to test, that's a problem (and it is difficult to find living ancestors due to privacy laws).

Luckily there's another option - investigating the stories of the Bennett-Linnabary ancestors.

But first, a little bit of math

At the start of this story, I did some math to show how my Brother and Mom could both have 1% African DNA.  To do this I showed my Mom most likely is 1/64th.  If you work that backwards, her father would have been, 1/32.  Her grandfather Jacob Budd 1/16.  Her great-grandfather James Bennett (mentioned above) would have 1/8.  I am 99.999% confident this is true.

That brings us to Jacob Bennett and Lucretia Linnabary, one of whom would have had 1/4 African ancestry.  If my math is correct, one of their parents has 1/2 African DNA.

We're really close.  We need to look at their parents.

The Linnabary's

Andrew Linnabary was born in 1813 in Delaware, Ohio, the child of a family that came to Ohio from Luzerne, Pennsylvania going back at least three generations.  Before that the family likely came from New Jersey and possibly Germany as some researchers thing the name morphed from Linnaburger.  With this information I'd say it is unlikely he'd have any African heritage.  

During the Civil War, Andrew joined the 38th Ohio Infantry as a Corporal of Company F.  He served from August 1861 to April 1862 and died at home from sickness acquired in the war.  He is buried in Little Auglaize Cemetery in Paulding, Ohio.

Sarah Young was born in 1816 in Pennsylvania and her family moved to Delaware as she married Andrew there in 1836.  The name on the marriage certificate says Sarah Young though oddly, her name is listed as Sarah Jones on all her children's death certificates (remarried?).  

Her father is said to be a man named George Young of Kentucky who moved to Pennsylvania.  This information is dubious but listed for completeness.  Sarah died in 1901 and like her husband is buried in Little Auglaize Cemetery.

The Bennett's

William Bennett was born in 1812 and the 1850, 1860, 1870, and 1880 census all agree he was born in Ohio but nothing specific.  We do know he moved around a lot as three census where in different Ohio counties - Hardin, Putnam, and Paulding.  

The Bennett family moved to Paulding County sometime before 1870 and stayed there for the next 3 generations, some still live there today.

William died in 1881 and is buried in Potter & Klein Cemetery in Paulding County.

According to the above census records William's wife, Nancy Clark, was born in Virginia in 1809.  Her children's death certificates all list her maiden name as Clark so we can be pretty confident that is correct.  She died in 1887 and is buried in Potter & Klein Cemetery in Paulding County, Ohio and her gravestone is what gives us her birthdate and death date.  Beyond that we have little concrete evidence.

So who is the source of our African roots?

Given the information we have right now there is no way to know for sure.

I am pretty confident the Linnabary's aren't the source as neither of the Linnabary-only matches on my Mom's DNA showed any West African DNA.  Also, the information the details we have on their lives makes it unlikely.

The Bennett's on the other hand are much more likely.  There are rumors about both of them though not enough proof for me to mention at this time.

What we do know if the family moved around a lot, probably as 'for-hire' farm work.  In addition to the census data we see above, the birth certificates for their kids show them living in Union County in 1832, Hardin County in 1841, back to Union in 1843, to Hardin in 1845, Union in 1848, and eventually to Putnam and Paulding.  To be fair, all of these counties are near one another so it wasn't like they were moving far.

The census records for both recorded them as white.

The question is what is most likely and as William listed his birthplace as Ohio and Nancy as Virginia, Nancy is the obvious candidate.  Of all my Mom's ancestors, Nancy is the only one who was born in a southern state as far as I've been able to determine. 

If this is true it raises all sorts of questions.  Was she the product of a master-slave?  If that is true and it happened in Virginia, she too would have been a slave.  How did she get to Ohio?  Was she the daughter of a freedman?  I hesitate to go further on this topic as it isn't one beneficial for too much speculation.

What is true?

Based on all the data I've seen, I'm 99% sure a Linnabary ancestor or a Bennett ancestor was a slave and I suspect that person was Nancy Clark's mother or father.

I did not do this exercise to prove anything.  I suspect 1/128th of me has West African DNA.  It's not much.  This doesn't suddenly give me black cred or anything beyond a little more knowledge of my ancestry.

However, as a son of rural Ohio, it does give me a sort of obstinate pride.  If Nancy is indeed the source of this DNA, I am quite aware that many of her progeny would be horrified to hear of the truth of their past.

I wish I could tell each and every one.

No comments: