Select Page

The dangers of uncritical acceptance

There are some really cool tools to be used these days to help analyze our DNA results up against our genealogical tree research.

But there’s a reason why they have names like Theory of Family Relativity (at MyHeritage) or ThruLines (at Ancestry) and not Proof of How We’re Related.

And we all need to exercise extreme caution when we review the hints and tips and clues we get from these tools, and not simply accept what they tell us as if it were proven fact.

The simple fact is, every one of these tools uses largely undocumented online family trees to suggest a possible relationship between two people who share some DNA.

The less documentation there is in these trees, and the more distant the relationship, the more likely it is that the theory will end up being disproved rather than proved.

Case in point.

In the new iteration of MyHeritage’s Theory of Family Relativity, I have a DNA match to an individual who shares less than 10 cM of autosomal DNA. That’s a pretty distant match, hard to trace under the best of circumstances. MyHeritage suggests, in its Theory of Family Relativity, that this match and I could be fourth cousins once removed through my Johnson line.



Almost as bad as Jones or Smith or Williams.

It begins by suggesting that there’s a 100% link between my tree and the tree of another person that includes my mother. I’ll buy that one, especially since tree #2 lists her father, my grandfather, with known information.

Theory of Family Relativity

And it climbs that tree to his mother, Martha H. (Johnson) Cottrell, who was born in 1857, and from Mattie to her father Mathew Johnson, my nemesis second great grandfather. Nemesis because I only have a few records — census, marriage, tax — and a possible place of birth, Virginia. No parents, no specific location in Virginia. And a surname of Johnson.


So I’m intrigued to see a link with an assigned 75% confidence level to another Matthew Johnson who’s going to be the linchpin in this theory of how I’m related to this DNA match, because this Matthew has a father in the tree where he appears.

But there’s just one little problem with this Matthew.

He died before the birth of the daughter from whom I’m descended.

No chance that Mattie is his posthumous daughter. This Matthew died 41 years before Mattie was born.

He wasn’t born in the same place as my Matthew. He didn’t die in the same place as my Matthew. So how we get to 75% confidence that he’s the same guy is beyond me.

Which means that even these very-high-confidence-level suggestions are often going to turn out to be just a theory, with nothing at all except a name in common to produce even a suggested link.

Now don’t get me wrong here. I’m going to look at every shred of evidence in every one of these trees and hints and theories on MyHeritage and in every ThruLine at Ancestry. I’m looking at enough of a brick wall with Matthew that I’ll consider any hint three times before throwing it out. Maybe my tree is wrong. Maybe my Matthew is the son of another Matthew and I’m missing a generation.

But what I can’t do — ever — is just accept that the suggested John Johnson born in 1750 is my Matthew’s father and drop him into my tree.

Because there’s a reason why these tools have names like Theory of Family Relativity or ThruLines and not Proof of How We’re Related.

These are hints and tips and clues — not evidence.

Cite/link to this post: Judy G. Russell, “Just a theory,” The Legal Genealogist ( : posted 21 July 2019).

Print Friendly, PDF & Email