Can DNA prove your hypothesis?

Can DNA prove your hypothesis?
I love using DNA to help my genealogy! But let’s talk about how (autosomal) DNA works when I’m trying to find/prove the parents of a brick wall ancestor.

For example, my theory is that my 4x-great-grandfather Walter Griffith is the brother of John and Charles Griffith, two of the confirmed-by-probate-records sons of Hezekiah Griffith. See image 1.
Image 1.
Griffith DNAsleuth
If I have a matching DNA segment with John Belford Griffith’s descendant (let’s call him Cousin Bob), does that prove my theory that my ancestor Walter was John’s brother and Hezekiah’s son?
No. It is one piece of evidence that’s worth adding to my collection of evidence. Just like I consider the fact that my Walter named his oldest son Hezekiah one piece of evidence supporting my hypothesis. But that DNA match isn’t *proof*. Why not?
Several reasons.
  1. How complete are our trees? Hezekiah Griffith is just one of Cousin Bob’s 128 5x-great-grandparents.  Likewise, that unnamed fill-in-the-blank space in my pedigree chart is one of 128 slots in that generation. Bob doesn’t have all the other names filled in on his family tree back that far, and neither do I. So that DNA segment we match on might have come down to us from some other shared ancestor we don’t know yet; maybe not a Griffith at all. For a very insightful discussion of this, see Blaine Bettinger’s blog post How Much of Your Family Tree Do You Know and Why Does That Matter? 
  2. How big is the match? Segments under 10cM are what Blaine Bettinger calls the “danger zone”, too likely to be a “false positive” to rely on, especially without other confirmed matches on the same segment. See his blog post  The Danger of Distant Matches. What if the matching segment is 10 cM? Or 20? The ISOGG Wiki explains that “fewer than 40% of 10 mB segments are within the last 10 generations. Larger segments can still date back quite some time and it was found that around 40% of 20 mB segments date back beyond 10 generations.” (Check out that ISOGG Wiki article here, citing Speed and Balding’s study  Relatedness in the post-genomic era, published in Nature Reviews Genetics, 16:33-34.) So a single segment match 20cM or less might come from a Most Recent Common Ancestor farther back than we hope.
  3. How accurate is my tree, and how accurate is my match’s? I know my tree has had mistakes in the past. A death certificate gave me the names of parents, so I put them in my pedigree chart, only to discover later that the death certificate was wrong. I may discover more glitches in my tree in the future. I may even learn that someone in my tree is wrong because of an unknown adoption. As for Cousin Bob? Is his line of descent back to Hezekiah Griffith solidly documented? Or did he copy from other people’s trees or unsourced family histories? Maybe I have to do more research on his line for him.
SO… if matching someone doesn’t prove who our common ancestor is, why bother with DNA? (And believe me, I definitely DO bother with DNA.)
old San Francisco brick wall
Most of the time, I’m not counting on DNA to *prove* kinship when I don’t have strong documentary evidence. I’m hoping that DNA will connect me to new cousins who either have documents or artifacts I don’t, or their family trees will give me new leads to look for evidence.
I’ve matched someone with a family Bible that gave me the birthplace of my great-great-grandmother, which led to my identifying her parents and adding a new generation to my pedigree chart. Yay! On another line, a cousin had a marriage record I needed. I’ve matched kin whose ancestors came from a specific area in Ireland, helping me pinpoint where our common line originated. I’ve met cousins who were able to identify people in my old family photos. I’ve connected to distant relatives who shared pictures of my ancestors, photos I’d never seen.
DNA has revealed unknown family stories too. A young widow in the 1830s gave her little boy to her brother-in-law’s family to raise. The boy’s surname didn’t change, but DNA from his biological mother, passed down to today’s generation, suggests what really happened.
Adding DNA to my genealogy toolkit has definitely added value to my quest to discover more about my family history. NotprovenBut a single match doesn’t prove who the common ancestor is.

All this is not to say that we can’t use DNA to *prove* a question of kinship. It’s just not as simple as finding we match someone who is descended from the person we think we’re descended from. Experienced genetic genealogists have constructed convincing proof arguments with thoroughly exhaustive documentary research supported by DNA results with *multiple* match pairs and analysis that may include concepts like triangulation. Check out recent issues of the National Genealogical Society Quarterly for examples. It can be done.
But when we first discover a DNA match to someone whose family tree has the person we *think* we might be descended from? Don’t pop that champagne cork just yet. It is worth celebrating as a new piece of evidence. The way we’d rejoice if we just discovered in city directories or census records or tax files that our known ancestor was a neighbor of the man we think was his father. Similar to a find like that, a DNA match can make our kinship case a little stronger.
But it’s not case closed.
Ann Raymont © 2017

7 thoughts on “Can DNA prove your hypothesis?

  1. Jane Chapman

    Thanks … this is a good reminder of the importance of seeking out, and using, a range of evidence from multiple sources when seeking to prove/disprove our hypotheses … a really good message for us all to heed … in genetic genealogy, genealogy generally and/or any other area of research, assessment or evaluation we undertake.


  2. dan

    Perhaps we should just drop the whole idea of “proof” and “proving”? These are leftovers from an age when genealogy was less rigorous than today. Contemporary historiography attempts to not be the hagiography so common in the genealogy practices of previous centuries.

    And, I don’t know if I would exalt a piece of paper over DNA. There are many, many instances of records that are either intentionally false, or unintentionally due to lack of knowledge by the originator.

    Liked by 1 person

  3. Andreas West`

    I first of all disagree with the “danger zone” of less than 10cM matches. These are random matches (meaning it’s calculated on all of someone’s GEDmatch matches). Taking a match that is:

    a) triangulated
    b) has an identified common ancestor between both family trees
    c) maybe even is further validated by phasing

    won’t surely give the same percentage of “false positives”. What Blaine is actually measuring is the quality of the IBD algorithm used, I think it was GEDmatch one.

    In order to triangulate you need 3 legs anyway, hence you have 3 independent family trees that will hopefully overlap. Usually, you get more DNA cousins in a TG, again with each one you have more chances to put them all into the same tree.

    So it’s a lot less black & white. Can DNA prove your hypothesis? Yes and no, it depends on what exactly you do and how you present your case.

    But we should avoid to painting such a negative picture that is only leading to people categorically denying even the best prove, because the article from Mr/Mrs SoundAndSo said so.


    1. Christopher Schuetz

      Andreas, I think you are saying very much the same thing. Ann is using the language of family history and you are largely using the language of data handling. Outside DNA – just using documentary sources – people will often pick one John Andrews and say that has to be their John Andrews. They ignore the fact that John Andrews might have moved, or that the record source they are using has gaps and this is one of them; and so on. I think Ann is just saying that the same think applies to DNA genealogy. Isn’t finding another match to triangulate just finding more evidence?
      Your points are valid and add to Ann’s case, and show readers more ways in which they can strengthen a case for proof.
      My ultimate standard of proof comes from my profession – would I be prepared to defend my conclusions in a court of law, where the penalties could be severe. One notch down from this is to imagine myself defending my position among my relatives and DNA peers, where a sloppy job of proof could bring ridicule and humiliation. If i am tired and maybe likely to miss something, those thoughts tighten my approach.

      Liked by 1 person

  4. Karen Sipe

    Blaine if you had a DNA match from a descendant of 3 or 4 of the sons of Hezekiah would you feel confertable saying DNA broke your brick wall?


  5. Josiah Austen

    I was studying a Jim Rohn ebook considered one of the nice private improvement authors when it all started to click for me and I
    was lastly in a position to really look and efficiently find
    the suitable woman . The consequence was
    what is named a normal Distribution curve.



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s