Ancestry.com: a new beef

If you have come anywhere close to this blog before, you have heard me grouse about online family trees at Ancestry, Family History Search, and other websites. I have preached ad nauseam that “information” on such sites does not prove anything.[1] It is not even evidence, much less proof. Actual family history evidence — which leads to proof — comes from original sources such as county probate records, deeds, tax lists, state birth and death records, and so forth. Online trees are, at best, clues. For the most part, they aren’t worth the paper it would take to print them. (See, e.g., this post: http://digupdeadrelatives.com/2019/03/01/reprise-what-is-proof-of-family-history/).  

I obviously haven’t bitched and moaned enough. It’s time to kick it up a notch.

A friend with considerable DNA expertise advised Gary and me to take the autosomal DNA test at Ancestry, although we had already tested at FTDNA. He said Ancestry has a larger database and that (at the time) their autosomal results featured something useful called “circles” of people to whom you are genetically related. The “circles” are now gone. What is left is something called “thru lines.”

I haven’t looked at “thru lines.” What I found out right off the bat (according to Ancestry) is that one’s autosomal results aren’t worth a spit unless you have a family tree at Ancestry.

Accordingly, I began to create one. Ancestry purportedly makes it easy by providing “hints.” For example, when I entered the name of a grandparent, a census record in which the grandparent appeared popped up. For the first few generations of a new tree, Ancestry’s suggestions are probably mostly accurate and harmless. There is good information in plenty of readily accessible information in twentieth-century census, marriage, birth and death records. More importantly, most of us know from personal experience the names of our parents, grandparents, and perhaps some great-grandparents. When that is the case, Ancestry’s suggestions, even if erroneous, don’t really matter. No harm, no foul for, say, the twentieth century.

The wicket gets a bit sticky as you make your way into the 19th century. It gets worse the further back in time you go. Let’s assume you have already done a good job researching your family history via conventional paper research in county and other primary records. You will be well-equipped to know whether Ancestry is providing accurate information when it suggests the names of an ancestor’s parents … or whether it is just providing names obtained from other family trees.

I eventually gave up on my autosomal results because of the arrant nonsense Ancestry was suggesting as possible parents for my relatives. Here are examples:

  • Ancestry suggested that the mother of my ancestor “Nancy” Ann Allen Winn Estes (wife of Lyddal Bacon Estes of Tishomingo Co., MS) was Lettice “Letty” Stone. This misinformation gets the “SAY, WHAT?” award. Other than the fact that Letty may also have been from Lunenburg and may have married a Winn — Lunenburg was awash in Winns and Stones in the nineteenth century — that is pure fiction, not fact. There are a million Lunenburg County records proving that “Nancy” Ann Allen Winn’s parents were Benjamin Winn and that his wife’s name was Lucretia (Andrews). Please forgive my hyperbole.
  • Ancestry suggested that Nancy Winn Estes’s husband Lyddal Bacon Estes (“LBE”) married Sally Alston Hunter. We need an emoji here for a big Bronx cheer. Sally Hunter did marry a Dr. Lyddal Bacon Estes (“Dr. LBE”). Dr. LBE and LBE were different men. I wrote about “same name confusion” about these two men here: http://digupdeadrelatives.com/2016/06/04/same-name-confusion-sorting-out-three-men-named-lyddal-bacon-esteslyddal-estes/ The Lunenburg couple — LBE and Nancy Winn — married there in March 1814. Dr. LBE died November 1814 in Maury Co., TN, and his widow was named “Sally” in at least two county records. LBE continued to appear in Lunenburg tax lists after Dr. LBE died. A comment by Shirley McLane’s character Ouizer Boudreaux in Steel Magnolias comes to mind: “these are not difficult questions!”
  • Chesley Estes, son of Benjamin and Frances Bacon Estes of Lunenburg, was not the father of the LBE who married Nancy Winn. Chesley died in Maury Co., TN, having never married and having lived with his parents most of his life. This one at least gets a “close, but no cigar” award: Chesley’s sister Mary Estes was LBE’s mother. Her identity is, I confess, a more difficult question, although Chesley’s lack of children  is not.
  • Benjamin and Frances Bacon Estes were the parents of Dr. LBE who married Sally Alston Hunter and died in Maury Co., TN in 1814. They were not the parents of LBE who married Nancy Winn in Lunenburg in 1814 and eventually settled in Tishomingo Co., MS. LBE died there between December 1844 and March 1845, and Nancy was his administratrix. I’ve written about LBE and Nancy Winn Estes’s family here: http://digupdeadrelatives.com/2017/05/28/identifying-the-children-of-lyddal-bacon-estes-and-nancy-ann-allen-winn-the-follow-the-land-theory-of-genealogy/

When Ancestry tells you it may have identified a parent for one of your ancestors, you can click on a link for the source of the information. You get only one guess for the source. And the winner is: someone else’s family tree. I made the mistake of messaging one of the tree owners about an error (yes, I was kind), but I should have known better. Correcting someone else’s family tree is like trying to teach a pig to sing: it is a waste of your time, and it just irritates the pig. 

I still don’t have any idea what “thru lines” are, or even how to find them. Fortunately, a good genetic genealogist and blogger has explained them here: https://dna-explained.com/2019/03/11/ancestrys-thrulines-dissected-how-to-use-and-not-get-bit-by-the-gators/.

Here’s the bottom line. It has always seemed obvious that many, if not most, family trees on Ancestry and other genealogy sites are constructed by copying other people’s family trees. This is a fast way to spread both bad and good information. Ancestry has now exacerbated and accelerated that process by helping people rapidly construct family trees with information obtained primarily? exclusively? from other peoples’ family trees. Ancestry, bless its heart, is killing credible family history research. That may not be a good long-term business model.

Gary, who likes to predict comments I will receive on my posts, says I’m going to get one saying, “No, Ancestry is just killing antiquated effete intellectual ‘researchers’ who think solving genealogical puzzles by digging through actual records is ‘fun.’”

One final note: if you aren’t familiar with Southern idiom, “bless her/his/its heart” means “what a total idiot.”

See you on down the road.

Robin


[1]Some online trees do provide sources such as census and probate records. Such information is obviously worthwhile provided it is associated with the right person! I can’t tell you how many references to probate records I have seen attached to Mr. X, when the will in question was actually written by Mr. Y, who lived a generation later and lived 6 counties west. Both Mr. X and Mr. Y were named John Smith, but that doesn’t mean they were the same man!

So … Is He My Second Cousin or My First Cousin Once Removed???

I’ve been talking to a friend who is a fifth cousin once removed. “How on earth,” she said, “do you determine our relationship?” I mumbled something incoherent about “rules” concerning degrees of consanguinity before concluding it was impossible to explain without visual aids.

Most of you undoubtedly figured out long ago how to tell a fourth cousin from a fifth cousin, and what “once removed” and “twice removed” mean. It took me a while to get to an “AHA!” moment on those issues. If this stuff is old hat for you, please head for this website’s archives and find an article about genealogical proof, or legal concepts in family history research, or the Scots-Irish.  If you have had any trouble calculating consanguinity, please read on.  There WILL be visual aids.

OK, we all know how to identify a first cousin, right? He, or she, is a child of one of your parent’s siblings, and you share a set of grandparents. The chart below shows a pair of Rankin first cousins, Tom and Robert (the green rectangles), grandsons of John Rankin and Emma Brodnax. Please note: John and Emma, my grandparents, are the only real people on the charts in this article. All others are fictional.

Please notice that there is one generation in between the first cousins, Tom and Robert, and their common ancestors, John and Emma.

OK, moving on, let’s add a generation: Chris and Alex, sons of Tom and Robert, respectively – the green rectangles in the chart below. Chris and Alex are second cousins. Note that there are two generations between the second cousins and their common ancestors, John and Emma.

This demonstrates the general rule: the number of generations between the youngest members of the line and their common ancestors equals the degree of consanguinity. Like so …

  • One generation between the youngest & the common ancestors = first cousins
  • Two generations between the youngest & the common ancestors = second cousins

And so on.

“Removed” simply means that one of the cousins has more generations between himself and the common ancestor than does the other. In the chart below, Tom and Alex are first cousins once removed, since Alex is one more generation “removed” from their common ancestors John and Emma.

Simple, oui? Hahahaha … I still have to sketch a little chart in order to figure out distant relationships. Alternatively, I could just enter the family into my family tree software and let it calculate the relationships. But what fun would that be?

See you on down the road. I’ve got a load of Rankins on my mind …

 

 

What is “proof” of family history?

I’m going to divert briefly from my research slog through various Rankin families of Pennsylvania, Tennessee, Kentucky, and Ohio long enough to address this issue: what constitutes genealogical proof and what does not.

I have a distant cousin (seventh cousins, maybe?) named Roberta Estes. We “met” online via Estes research some twenty years ago.  We finally met in person, spending a week together in Halifax County, VA doing nitty-gritty research among records in the basement of the Halifax courthouse. I knew I had found a kindred spirit when I learned she likes tax and deed records as much as I do.

Roberta writes an excellent blog called “DNA Explained.” A great many of her posts are about DNA “science.” When I have a question about DNA, the first place I go is to her blog and search her Archives.

Roberta’s post today is on a topic that will interest all family history researchers: what is, and what is NOT, genealogical proof. Here is a link to   her post.

What resonated most with me was her list of things that do NOT constitute “proof.” I have copied part of it below, with my comments and modifications in italics (the numbering has changed from her original list since I deleted a few items):

  1. Proof is not an oral history, no matter how much you want to believe it or who said it. Oral history is a good starting point, not an end point.
  2. Proof is notnot, 1000 times NOT someone else’s tree. A tree should be considered a hint, nothing more.
  3. Proof is not a book without corresponding evidence that can be independently corroborated. Being in print does not make it so. People make mistakes and new information surfaces. Unfortunately, there are also genealogical frauds – see, e.g., Gustave Anjou.
  4. Proof is not a man by the name of Jr. [taken to mean] that he is the son of a man by the same name with the suffix of Sr. “Sr.” often means older and “Jr.” means younger, but not necessarily related. In fact, the suffix can change over time for the SAME MAN: a Robert Rankin who initially showed up in Guilford, NC records as “Robert JUNIOR.” became “Robert SENIOR” after Robert the elder (his father) died.
  5. Proof of a father/son relationship is not just two men with the same name in the same location.
  6. Proof is not just a will or other document … without evidence that a person by the same name as the child named in the will is the RIGHT person.

The lawyer in me, retired though she might be, feels compelled to expand on Roberta’s discussion of “proof.” Namely, I want to draw a distinction between “proof” and “evidence,” and the amount of evidence that is needed to produce a certain standard of proof. 

The definition of “evidence” takes up a full page in Black’s Law Dictionary. Fortunately, the essence of the meaning of “evidence” as it relates to genealogical research is pretty easy to distill. Try this on for size: EVIDENCE is anything that is offered to prove the existence or nonexistence of a fact. In genealogy, evidence might be a deed, a will, tax records, church birth and death records, census records, tombstone inscriptions, and so on.

Notice that the word “prove” appears in that definition. Here is what Black’s has to say about that: PROOF is the effect of evidence.

Boiling both definitions down, evidence is what supports a belief that a fact is proved (or disproved).

If you have ever served on a jury, you already know there are different “standards of proof.” In a Texas criminal trial, the standard of proof requires a defendant’s guilt to be established “beyond a reasonable doubt.” In a Texas civil case, the standard of proof is usually “preponderance of the evidence.”

Another standard of proof lying somewhere between those two is “great weight and preponderance of the evidence.” Law students, who like to boil things down to something understandable, may view it like this:

  • Beyond a reasonable doubt: at least 95% of the facts compel a certain conclusion.
  • Great weight and preponderance: about 65-85% of the evidence supports a conclusion.
  • Preponderance of the evidence: a conclusion is more likely than not – it has the weight of 51% of the evidence.

Naturally, there are parallels in family history research, or I wouldn’t be talking about this.

You frequently see the phrase “conclusively proved” in family history articles. This is roughly equivalent to “beyond a reasonable doubt.” For example, my paternal grandmother’s identity – Emma Brodnax Rankin – is conclusively proved by my birth certificate, my father’s birth certificate, his mother’s will naming him as a son, census records naming him as a son, ad infinitum. Not to mention my recollection of all those awful holiday dinners in her grotesquely overheated house in Gibsland, Bienville Parish, Louisiana. A court would call my testimony about those dinners “direct evidence” based on personal knowledge. If I’m a credible witness, THAT case is closed.

When you see the phrase “conclusively proved,” it means there is really no reasonable argument to the contrary. That is how I use the phrase on this blog.

Use of the word “probably” in family history articles seems to equate with “preponderance of the evidence.” Namely, a conclusion is more likely than not.

Similarly, the phrases “most likely” or “almost certainly” are somewhere in between the other two. There may be a reasonable doubt, but the weight of the credible evidence strongly points one way.

We also need to talk about“circumstantial” evidence, because sometimes there is no other proof of a family relationship. That is particularly true in counties where records have been lost and documentary evidence is limited. “Circumstantial evidence” just means facts that lead to a reasonable inference.

For example, the fact that a 65-year old man named Jedediah Rankin is listed in the 1860 census in a household immediately adjacent to 40-year old Jacob Rankin constitutes circumstantial evidence of a relationship. You can reasonably infer some family connection between the two men because such an inference accords with common sense and experience. If Jacob and Jedediah witness each other’s deeds, that would provide additional circumstantial evidence of a family relationship. If Jacob named his eldest son Jedediah, and Jedediah Sr. was security on Jacob’s marriage bond, those facts would also be circumstantial evidence.

Circumstantial evidence can establish a compelling web of family connections which suggest that only one conclusion seems reasonable: Jacob was Jedediah’s son. It is a powerful tool in serious research.

One last red flag about “proof:” beware the passive voice, a grammatical form that frequently signals lack of evidence. Keep an eye out for these phrases, which appear in many compiled family histories: “it is thought that …” or “it is believed that.”  Hmmmmm…. who believes? And what is his or her evidence? Those phrases rightfully justify a jaundiced eye unless the writer provides evidence supporting the “belief.”

Now, back to Roberta’s list of “not proof,” item #2, someone else’s tree. It may be a fact that “many online trees” show Jedediah as Jacob’s father. That is not even evidence of a relationship between those two men. All it might prove is that many online trees are copies of other online family trees.

Finally, Roberta’s item #1, family oral history, is near and dear to my heart right now. Family legends/family oral traditions nearly always reveal important truths. They usually also contain errors. And we’ll begin my next post with what is probably the most well-known and firmly established oral tradition in Rankin family history.

*   *   *  *   *   *  *   *   *