Ancestry gave me another bad NAD today, or a New Ancestor Discovery, who is absolutely, positively, unquestionably, not my ancestor. But this time, they did me the huge favor of assigning someone that was immediately familiar to me, and I can share with you the “logic” of how this erroneous connection happened. You can then use this same process to work on unraveling your own New Ancestor Discoveries – now that you know what to look for.
Let me first say that genetic genealogy based on inferences has the ability to give you hints you would not otherwise have, like with DNA Circles and NADs, but these inferences that Ancestry arrives at by a process they call “network theory” can also lead you badly astray – like the logic that says your ancestor’s sister’s husband is your ancestor. Of course, I am assuming here that you are not double descended – and I know positively that I’m not. I went through the proof process with the first bad NAD that Ancestry gave me, although I never figured out the logic of how I was assigned that original Bad NAD couple, who is now gone.
Blaine Bettinger recently explained Ancestry’s network theory quite well in his blog, “Creating DNA Circles – Exploring the Use of Genetic Networks in Genetic Genealogy”.
Ancestry has consistently refused to provide us with the triangulation tools we need, via a chromosome browser, and we are left to do the best we can with genetic networks and other inference methods. Triangulation confirms descent from a common ancestor, while network theory connects people who are related to each other, suggesting common ancestors – like my new bad NAD.
My new bad NAD is Robert Shiflet, the husband of Sarah Clarkson/Claxton, the sister of my ancestor Samuel Claxton. Both Samuel and Sarah share parents, Fairwick Claxton and Agnes Muncy. However, Robert Shiflet is not related to me by blood, but of course, his children are – through his wife.
This chart, below, shows how all of the people we’ll be discussing in the bad NAD group descend from common ancestors, Fairwick Claxton/Clarkson and Agnes Muncy. You can see that three groups descend from Sarah Clarkson and Robert Shiflet through son Fairwick Shiflet and daughters Elizabeth and Rhoda. I descend through Samuel Clarkson, brother to Sarah.
Here’s Robert Shiflet, my newly arrived bad NAD, at Ancestry.
By clicking on the New Ancestor, you can see how I connect to the people that Ancestry has used to determine that Robert Shiflet may be my ancestor.
The NAD circle is made up of three family groups, where several closely related individuals have tested, so they are counted as “one” and not as separate matches.
There are two individuals in each of the three family groups.
All of these people descend from Sarah Clarkson/Claxton and Robert Shiflet. Ironically, Sarah, who is not listed as a NAD, is the daughter of my ancestor.
In fact, as irony would have it, two of these same groups ARE in the Fairwick Clarkson/Claxton DNA Circle and along with me, these are the only two other members of that circle that I match.
So, if you’re judging from the number of connections only, the NAD circle, with 3 groups totaling six people looks stronger than my Fairwick Clarkson Circle with only 2 groups totaling 4 people. I checked each tree of each individual within the Shiflet Circle and have summarized the results below.
|Participant||Family Group||Sarah Listed?||Fairwick Listed||Fairwick Circle|
|CT||Martha Patsy||Yes, as Sarah “Sallie” Clarkson||Yes, Fairwick Claxton||Group is in circle|
|Charlene||Martha Patsy||Yes, as Sarha Clarkson Shiflet||No||Group is in circle|
|DL||DL||Yes, as Sarah A. Claxton||No||No|
|JL||DL||Yes, as Sarah A. Claxton||No||No|
|DB||Barbara||Yes, as Sarah “Sallie” Clarkson||Yes, as Fairwick Claxton||Group is in circle|
|DJ||Barbara||Yes, Sarah H. Clarkson||Yes, as Fairwick Clarkson||Group is in circle|
Please note that in this case, the spelling of Sarah’s name was quite different. It was spelled Clarkson, Claxton and in one tree, she was listed as Sarah Clarkson Shiflet, with Shiflet as her surname. Her first name was misspelled in one tree. This could be why Sarah was not listed as a NAD along with Robert, whose name was consistently spelled the same way.
Still, because two of these family groups are members of the Fairwick Claxton/Clarkson Circle, one would think that it would be immediately evident that since we DO share an upstream ancestor, when utilizing our trees, that the husband of my ancestor’s sister is not my ancestor – but I am related to his descendants by virtue of his wife’s parents – so of course I match the DNA of his descendants. That does NOT mean I descend from him.
The linchpin that may have triggered Ancestry to create a NAD may have been that I match one set (family group) of Robert Shiflet’s descendants that aren’t in the Fairwick group. The reason the DL group is not in Fairwick’s circle, if you look at the trees, is because the DL group does not list any parent for Sarah – so they can’t be in Fairwick’s circle because Fairwick isn’t listed in their tree. It would make a lot more sense for Ancestry to give the DL group Fairwick as a NAD than to give me Robert Shiflet as a NAD.
So, take all NADs with an extremely large grain of salt – in fact – the whole shaker would be appropriate here or maybe something the size of rock salt.
So far, the NAD score, out of 5 that have been assigned to me, 3 are proven to be incorrect. Two, the Larimers, the jury is still out – well, sortof.
The jury isn’t entirely out on the Larimer’s actually, because when I look at the group of people in the Larimer NAD circle, I discovered all 5 people who I match on my Andrew McKee line. Hmmm….
These people ALSO connect to John and Jane Larimer – on a completely separate line from Andrew McKee. In another group, I find another ancestral surname where I connect with the entire group. So, I’m guessing that it’s circumstantial that all of these people descend from John and Jane Larimer – and that John and Jane have nothing to do with me just because I match their descendants through two of my other known lines. I don’t actually match anyone else in that group – although a lot of them match each other. As it turns out, all of this “network theory” matching is a red herring this time – because of intermixed multiple family lines.
Can I prove positively that I don’t share any ancestor upstream with John and Jean Larimer? Nope, I can’t, but given the trend that I do see, it looks like the NAD was based on other family connections that circumstantially are connected to the Larimers as well. And I can tell you, from what I do know about my genealogy, that I don’t descend from Jean and John Larimer. There is no vacancy in my tree that fits their ages, so they are not my ancestors.
So, I guess that really makes the score:
- Ancestry – 0
- Bad NADs – 5
The sad part is that it also makes my score 0 – and leaves me begging for the chromosome browser that we so desperately need and would eliminate all of this tail-chasing. A chromosome browser wouldn’t leave us guessing about whether the Larimer segments were the same segments as the McKee segments. We would know positively whether they were or not – no guessing, tail chasing or network theory needed.