Don’t take ThruLines as gospel either
So a couple of weeks ago The Legal Genealogist focused on the dangers of uncritical acceptance of the automated systems at the DNA testing companies that link us to possible/proposed ancestors by showing how one suggested Theory of Family Relativity at MyHeritage couldn’t be correct.
Based on nothing more than trees that shared a same-name person — Matthew Johnson — MyHeritage had me linked to a family I can’t be linked to, at least not in that way.1
Today let’s share the love.
Here’s one from Ancestry’s ThruLines that’s just as bad.
One of my nemesis ancestors is George Washington Cottrell (b. c1821, probably in Kentucky; d. 1891 in Texas). He was a thorough-going scoundrel whose early appearances in Texas records generally have him running one step ahead of the law.2 Finding even a clue to George’s parents would be terrific.
But the most recent iteration of Ancestry’s ThruLines for George isn’t going to help one bit.
Because, it suggests, his mother was Mary Parsons of West Virginia, who married Thomas H. Cottrell, and had a son George Washington Cottrell, born in 1820-1824.
Opening up the Evaluate button to examine this theory, ThruLines tells me that “None of the trees linked to your DNA matches identify Mary Ann Parsons as the mother of George Washington Cottrell.” So, to begin with, nobody who has tested and who matches my DNA descends from the Cottrell-Parsons family of West Virginia.
ThruLines then tells me that “8 other member trees identify Mary Ann Parsons as the mother of George Washington Cottrell.”
Okay. But there are a few problems with those identifications:
1. Not one of the trees cited as support for George-as-son-of-Mary has so much as a single documentary source for linking any George as a son to Mary, much less my George. The sole citations are to other family trees.
2. All of the trees that do list this George have him well-documented via census records as living in Ohio without a shred linking him to Texas, at a time when my George is equally well-documented in Texas.
3. All of the trees that do list this George have him marrying a different wife than my George, and having different kids from my George, all at the same time as my George.
4. If the trees are right, this George died 16 years before my George.
In other words, this is another same-name-therefore-same-person suggestion.
And if that’s not enough, this particular Cottrell line in West Virginia? Their YDNA says it’s not the same Cottrell family at all. At the basic 12 marker level, we only match at eight of 12 markers. At 25 markers, it’s 15 of 25 markers. The genetic distances only increase as the number of markers tested increase: this isn’t the same family line.3
This isn’t something I’m guessing at. I know this family really well. You see, I happened to research this West Virginia family pretty thoroughly some years ago as part of my kinship determination project for certification by the Board for Certification of Genealogists; my brother-in-law is collateral kin to this family. And no, we’re not the same Cottrell line at all.
Again, I want to stress that these tools are wonderful sources of hints and clues.
But they’re not more than that.
They’re not proof.
And we shouldn’t accept them without critical analysis.
The best tool we have is the grey matter between our ears.
Cite/link to this post: Judy G. Russell, “Equal opportunity DNA criticism,” The Legal Genealogist (https://www.legalgenealogist.com/blog : posted 11 Aug 2019).
SOURCES
- Judy G. Russell, “Just a theory,” The Legal Genealogist, posted 21 July 2019 (https://www.legalgenealogist.com/blog : accessed 11 Aug 2019). ↩
- See Judy G. Russell, “George Washington Cottrell of Texas: One Man or Two?,” National Genealogical Society Quarterly 105 (September 2017): 165-179; pdf online at the Board for Certification of Genealogists Genealogical Work Samples page (https://bcgcertification.org/learning/skills/genealogical-work-samples/ : accessed 11 Aug 2019). ↩
- “Cottrell FamilyTreeDNA Project – Y-DNA Classic Chart,” FamilyTreeDNA.com (https://www.familytreedna.com/ : accessed 11 Aug 2019), comparing the West Virginia Cottrells (Group 2) with the Virginia Cottrells (Group 3) to which my George belongs. ↩
Yep, gotta be really careful. One of my longstanding Ancestry tree nemeses is the supposed father of my Elizabeth Caudle/Caudill/Cordle, who married my gr gr grandfather William B. Mitchener in Sumner Co., Tenn., in 1808. Everywhere I turn, some tree or another tells me she was the daughter of Absolom Caudle. Problem is, my Elizabeth was born about 1789 (1850 census). Absolom submitted a copy of his family Bible as part of his Rev War pension application. HIS Elizabeth was born nearly 20 years after mine, and we know who she married as well. Not my gr gr grandfather.
ThruLines keeps repeating this error. The DNA matches to me through this supposed line/suggested ancestor Absolom may well be valid Caudle-related matches, possibly through one of Absolom’s siblings. But ThruLines just latches onto the widespread but incorrect online trees to explain the DNA match.
Absolom was one of my gateway ancestors on the old “We’re Related” app. I know it’s hard to believe I’m not kin to Brad Pitt, but there you have it.
Sigh…
Thanks for clearly describing a problem that many of us face. There is one more thing I would add that pours salt in the wound and demonstrates once more the tone-deafness of Ancestry. When one clicks on the “Evaluate” button and gets all the reasons why Ancestry thinks it has identified matches, at the bottom of the list is the following statement: “Add [name] to your family tree from ThruLines. Select a record or tree above to add” Nowhere is there an option to report any of the good research you cite demonstrating that the suggestions are wrong and why. So Ancestry’s definition of “Evaluate” seems to be, “Tell us why we are right, but don’t let us know when we’re wrong.”
Thanks for emphasizing the need to verify, rather than blindly accept these hints! ThruLines is only as accurate as the Ancestry.com trees it relies on, or even less so, as it can make huge leaps based on nothing more than surname similarities. It really concerns me to see all the ThruLines “success stories” posted elsewhere, where people say they have used it to add many people to their trees. Are they blindly clicking? I’m also really tired of seeing it called “TrueLines”!
Thanks for another great article, Judy!
I LOVE ThruLines and Theories of Family Relativity! They always deliver value!
After careful research of what is presented, I discover one of the following valuable things:
1. I confirm something I already knew. I love that! Yeah me!
2. I learn something new about my family. I love that! Voyage of discovery!
3. I find out that the suggestion is ab-so-lute-ly reee-diculous. And that gives me hilarious stories to share online and over adult beverages with other genealogists…and we all get a good laugh! Humor!!
Anything that triggers careful research that gives me value – I am all for it!
Do I believe these Theories and ThruLines at face value? I’m with you, Judy – not on yer life!
I feel like platforms like Ancestry and MyHeritage are improving their DNA offerings, but there is room for more improvement. For example, I’d like to see not only my ThruLines, but I’d like to see a cluster of DNA matches around different family lines that may not already be documented links on my tree. Right now, I have over 80,000 DNA matches. That’s great. Having better tools to sort and cluster these matches is what’s needed. And tools that do the work for the consumer. I think it’s too much to ask the average user to sort it all out. Perhaps this may be a wish list, but these platforms are making a lot of money off consumers and need to invest some of it into improving their platforms.
I *am* related to that Cottrell line in WV, and, regardless of the DNA, there is some similarity to your George, tho’ yours sounds much more industrious, at least in running ahead of the law. A cousin in my Cottrell line was described as, “…not afraid of hard work. He could sleep right next to it.”
Definitely a predisposition to mischief! One of my favorite finds is the Virginia court case where Jackson Cottrell (son of Thomas and Mary), his uncle Joseph Parsons and Joseph’s brother-in-law Daniel McCune all get sent to prison for murder in 1843… 🙂
Thrulines gave me a hint. My new DNA match is… Me!
Lots of mistakes on the trees on Ancestry so everything still needs to be researched if possible. People had the same names and lived in the same places at the same time even way back in time. Per DNA we are somehow related but ThruLines is not absolute. There were 2 John Pates in early Nacogdoches but one died 1837 and the other 1874. However, the records are still mixed up on the Ancestry trees. There were two Daniel Ramey/Reameys in Virginia in the same timeframe. One was in Henry County, VA. and died between 1803-1805 in Henry Co. The other was a Rev. War soldier from Augusta County, VA. who moved to Kentucky and died there. Folks still mix the two men up and both are listed as a “Colonel” in the Rev. War on records. Even some books have it mixed up. Bottomline is to not trust everything without doing research on the actual records if you can find them-wills, probate, death, birth, and church records-oh and a family bible which is rate but wonderful.
Broke down and purchased actual membership in Ancestry when they offered an incredible price. Looked at my husband’s family. My sister-in-law who a few years ago bubbled about doing the family tree has the WRONG man as our husband’s father. And the tree goes from bad to worse. Our husband’s mother was wife #2 and there was a wife #3. All three wives gave him children. I remember when he died, hubby’s brother called us asking if we wanted to contest the will. [Hubby didn’t.] Father in tree died much later than above info and only had 1 wife. ARG!!
Problem, I know better, but real father doesn’t show up in any normal sources, and this wannabe father has lots of sources. Oh the joys!
Is there a way to get rid of the match once you have evaluated and disproved it?
It’d sure be nice but… no. At least not as of now.
On DNA Matches, using “Tools,” I can hide the match but not delete it. Haven’t tried to fix things on Thru Lines. Ancestry “thinks” it is coming up with great new features but alas, they aren’t.
I don’t want to hide matches–only the incorrect ThruLines.
I haven’t spent much time with Thrulines but I will say, every time I hop over to FamilySearch I am appalled at the misinformation attached to my family tree. I don’t bother correcting the info there; all of my concentrated efforts are in Ancestry, but I wonder about others who are relying on FamilySearch for their family tree research.
I use FamilySearch extensively for researching their excellent extensive collection of records. I do not use trees as research, period, and I ignore “suggestions”. They are dead-end rabbit holes. When I have done considerable research on a line using various actual sources, and feel that I am on the right track, I will check FS trees to see if someone else is doing research on the same line and is using reliable research protocols, including valid citations (NOT indexes or trees!). If things look ok, I will contact the person(s) responsible, and ask if they’d be willing to share and compare info. This has resulted in new leads several times, which I then research to verify and make sure citations are valid. I contribute to some collatorial lines when I have verified info so that the folks working on that line can make use of it. It’s also an opportunity to gently explain what is actual evidence and what is not. Sometimes I offer guidance to someone who is clearly a newby, pointing them toward good training resources. Beyond that, I do not go. I do not use Ancestry, recently pulled my data from MyHeritage out of disgust. I grew tired of the ubiquitous errors thrown at me- my fav is the one that has my grandmother as the daughter of my grandfather. How in the world did they do that? And I was getting it over and over, against my will, as people copied and pasted. It was a waste of time and money. I can use both to treat myself to a research trip and have some fun while I’m at it.