18 October 2013

When the Male Line Fails-Where Did the Chain Break?

If a male line DNA test indicates two researchers are not related, does it mean that their supposed direct line male ancestors are not related? It seems to me that is not necessarily the case. I'd be interested in thoughts on this little scenario.

The Oldmans

Kevin Oldman has researched his ancestry back seven generations to Henry Oldman, who appeared in Kentucky in 1820 from nowhere.

Bubba Oldmann has researched his ancestry back seven generations to Thomas Oldman, who also appeared in Kentucky from nowhere in 1820.

Thomas and Henry Oldman lived in the same Kentucky county in 1820 and 1830, but their listing on those two population census schedules are the only documents on which they appear together. Their farms are in the same township, but not adjacent. After 1830, Henry leaves Kentucky for Indiana. Thomas dies in Kentucky. No record ties them together other than the census. There's no hint they were related other than their enumerations in the same census district in 1820 and 1830.

Kevin and Bubba learn of each other through the internet and discover that they are potentially related through a common Oldman ancestor. They speculate that Henry and Thomas are most likely brothers or first cousins.

Kevin and Bubba decide to have a y-DNA test performed to determine if they share a common male ancestor. The test indicates that Kevin and Bubba do not have a common strict paternal line ancestor. Kevin and Bubba conclude that Henry and Thomas are not related based upon the test results. They continue to compare research notes, but are left convinced they are researching different Oldman lines. After all, if Henry and Thomas were related through a paternal ancestor, it would have showed up on the y-DNA test Kevin and Bubba took wouldn't it?


There's something Kevin does not know.

Kevin's ancestor, Clueless Oldman (son of Henry), thinks he was the father of all the children his wife had. Clueless is not aware of the fact that during his marriage his wife had an ongoing relationship with an unmarried neighbor. During her lifetime, the wife was relatively certain that the neighbor was actually the father of two of her children and not Clueless. One of those children Clueless' wife had with the neighbor is Kevin's Oldman ancestor.

So "technically" Kevin's not really and Oldman at all--that's what the paper trail says, but it's not the biological reality. Kevin is not actually a descendant of Henry Oldman.

And it turns out that Henry and Thomas Oldman were indeed brothers.

Bubba Oldman and Kevin Oldman are not related--because Clueless Oldman didn't father all his supposed children. And that has nothing to do with whether or not Henry Oldman and Thomas Oldman were brothers.

The Inverse

Readers with a mathematical or logical bent, may notice that there's a logical fallacy at work here. The original statement is:

  • If Kevin and Bubba share a common ancestor, then Thomas and Henry are related. 
This is a valid statement based upon DNA theory. If Kevin and Bubba share a common male ancestor, then Thomas and Henry share a common male ancestor. But the statement:
  • If Kevin and Bubba share a common ancestor, then Thomas and Henry are not related.
is not true and cannot be derived from the original statement. It is the inverse of the original statement and the believing the that the inverse is true is referred to as the "fallacy of the inverse."


Geolover said...

This is an excellent example of how YDNA can point to common ancestry but not specify who the common ancestor is.

The situation could have been complicated if actual father of sons Clueless thought was his, was actually a paternal uncle of Clueless or even Clueless' paternal grandfather. The YDNA results for Kevin and Bubba should show a common paternal-line ancestor -- again not identifying who exactly it was. And not identifying who the male parent of Thomas and Henry was.

Kevin and Bubba should also show some extremely close autosomal DNA matching.

Your post shows the fallacy in the "triangulation" arguments so commonly used in interpretation of YDNA test results.

Lynn David said...

If two men of the same surname are tracing roots back to some areas of Germans they may have a disconnect as well, as shown by their DNA. However, it need not be one involving an extra-marital affair. So there are situations which arise.

One occurance would be when a woman, who first married the original inheritor of the farm and bore him male children, then remarries after the death of her first husband.

In this case males born to the first marriage carry the male DNA heritage and surname of the farm while their half-brothers born in the second marriage carry the farm surname but a 'foreign' DNA.

Another would occur when the inheritor of the farm is a female in the farm family and her children take the surname of the farm irrespective of their father's surname.

In this case the male children again have the farm surname but a foreign DNA. This male children may, however, have 2nd cousins (or even 1st cousins?) who do carry DNA endemic to the surname and farm. Here two researchers would have to delve into earlier generations to discover their relationship.

The first case occurs in my own paternal line. My immigrant ancestor was the half-brother to five elder males of the same surname (but different fathers). I used to preface the surname when referring to those 5 brothers and their descendants with "the real" - however then I found out that in the prior generation they fell prey to the second case in their line when an elder woman inherited the farm. Going further back on my own paternal line I discovered that my immigrant ancestor's father did not even share the surname of his father, but took the surname of the farm on which he was born.

So all in all, in areas where this naming practice occurs, one's DNA heritage (Y-chromosome) can scarsely be tied to a paternal surname except for short-term runs.

Unknown said...

My great great grandfather Peter Hopfer was born in Northumberland Co. PA in 1828 or 1829, at age 7 he moved with the Glidewell family to Sullivan Co. So we are assuming he was an orphan. WE can't seem to find who his parents are. There was one Liberty Hopfer who died in that county in the same twp, but we have no connection the Peter as of yet. She was married to Jacob Hopfer a War of 1812 Veteran who dies in 1865 in Northumberland Boro where Peter was born. WE hoped they were his parents, but why would his father give him up like that? Then we found on Peter's Civil War record that he was "borne a Hoffer" Well there are plenty of Hoffers in the area and surrounding counties just across the river who decended from Matthias Hoffer of Lancaster, PA. BUt I can't find any connection to them, even though Matthias had a son named Peter, he's not our man as a possible father. Although his other sons could be.I know I need to she some Overseers of the Poor Orphan's Court books if there is any. It's just disheartening looking for the past 39 years.