When DNA won’t work

Go for the paper trail

Here’s a news flash from The Legal Genealogist, a/k/a the geek who never met a DNA test I wouldn’t take:

dna.qDNA isn’t always the answer.

Sometimes it just can’t answer the question you have.

Case in point: which of three sisters born in Germany in the late 1700s is my third great grandmother?

That question came up this week because a dear friend of mine is in Germany right now and, since he was going to the archives in Sachsen-Anhalt where my father’s paternal side ancestors were from, I asked him to see what he could find out for me.

And we came up against this problem: my second great grandmother Friedrike’s parents are listed in her baptismal records as Friederik Geissler and Johanna Sophia Schumann.

But the records from Ossig, where Friedrike gave birth to my great grandfather Hermann Geissler, show her mother as Rosina (Schumann) Geissler. a native of Kayna.

And the Schumann family of Kayna had three daughters:

• Maria Rosina Schumann, born 3 February 1789;
• Maria Sophia Schumann, born 17 August 1791; and
• Johanna Christiana Schumann, born 13 December 1793.

So, as my friend puts it,

Johanna Christiana Schuman could be Friedrike’s mother, or
• Maria Rosina Schumann could be Friedrike’s mother, or
• Maria Sophia Schumann could be Friedrike’s mother.

“There is no doubt Friedrike’s mother is one of the three,” my friend the researcher reports. “There is only Schumann family in Kayna at the right time, that is the Elias Schumann family. But will the real sister please stand up?”

Do you see the problem with trying to answer this with DNA?

First, we can’t use YDNA at all. That’s the kind of DNA passed from father to son to son — it can’t be used when there’s a female in the line at all.

Second, we can’t use mitochondrial DNA either. Yes, that’s the kind that’s passed down the female line, but there are two reasons why it won’t work here:

• It’s not an unbroken female line. Yes, my great grandfather Hermann has his mother’s mtDNA, but my grandfather Hugo Ernst got his mtDNA from his mother, a different line altogether, and my father got his mtDNA from his mother — another different line — and I got mine from my mother — yet another different line.

• Even if it had been an unbroken female line, or we could find surrogate candidates to test, all three Schumann sisters would have exactly the same mtDNA, inherited identically from their mother Christine Seÿffarth Schumann. There’s nothing to distinguish the mtDNA among the sisters.

Third, autosomal DNA won’t give us any definitive answer here either. Assuming we could find living descendants of all three sisters to test, the relationship is distant enough that the sheer random process by which today’s living descendants would have inherited autosomal DNA from our common ancestors would make it unlikely in the extreme that we could distinguish descendants of one sister from descendants of another.

This is exactly the kind of problem that forces all of us who love DNA as a tool to say, over and over, it’s not something we can use instead of a paper trail. It works, when it works at all, with the paper trail. And sometimes it doesn’t help at all.

Here, as in so many genealogical problems we wish DNA could solve for us, we have to go for the paper trail.

Wish us luck!!

Print Friendly
This entry was posted in DNA. Bookmark the permalink.

13 Responses to When DNA won’t work

  1. Is there a marriage record for Friederik and any of these ladies?

    • Judy G. Russell says:

      Not that we’ve been able to find yet. Not married in the town where she was born, or where he was born, or where their oldest children were born, so we’re still hunting. Sigh…

      • Richard Slaughter says:

        That is what I was thinking is that they have such great records in Germany. Think what would be there had it not been for such destructive wars. I have a 7.2 cousin match on Gedmatch. Our grandparent set that immigrated were 2nd cousins from Germany. Henckel and Dentzer. My cousin in Germany is a Dentzer. I only solved this one because I belong to a Yahoo group that has his cousin in it and she had him tested. She knew in about 2 sec. how we were related. She sent me some incredibly awesome records.
        I was very lucky. You never know what matches you will get tomorrow.
        I would bet on Rosina as the records are later and more correct. They put down Johannna Sophia on the baptismal record which is two names combined and the record with Rosina is hers alone.

        • Judy G. Russell says:

          I’m personally betting on Maria Rosina myself, but we’ll have to see what the rest of the paper trail tells us.

  2. Richard Slaughter says:

    Certainly not DNA solvable unless you get lucky with the right German descendent that knows his genealogy. Or, you find a whole lot of folks with a Schuman GG grandmother to test.
    I have a possible NPE situation with a GG grandfather from Scotland. I have a paternal 1st cousin who matches someone from Australia who has Scot ancestors in that exact same area at the same time. I do not match her which is odd and probably means she is a little further out. The Autralian cousin also matches my maternal aunt and not to me.
    Then I have another distant cousin that matches a maternal 1st cousin1R on my mother’s maternal side and she also matches the same maternal aunt above on my mother’s paternal side. Now really folks!

    • Judy G. Russell says:

      Not even with a whole lot of Schumann descendants, Richard: the chances are just too great that recombination would have scrambled things in the interim.

  3. Carolyn Lea says:

    I have found it difficult following lines in German records because not only do other families (related probably?) use the same names, the siblings within a family share the same name in a different order. Johann, Hermann, Friedrich, Heinrich, August and their feminine variants were so common.

    In one family the daughters were:
    Justine Marie Dorothee
    Marie Dorothee Elizabeth (same as mom)
    Johanna Catherine Marie Dorothee

    Then there is the fact that there can be as many as 4 (or more?) forenames (given names) and while one is usually used in daily use (Rufname or appellation name – the name the person goes by from childhood on and often not the first in the list) – I have found any one of those might appear on different records.

    Really hard to follow – especially when topped off with old German script!

    Good luck in your search.

    • Judy G. Russell says:

      That script sure doesn’t help things any, does it? Thanks for the good wishes. I’m hoping we end up finding marriage records for at least some of the sisters to help narrow this down.

  4. Israel P. says:

    I’m thinking you should be looking for the death records of the women. That happened in Jewish families all the time – the young wife dies and leaves young children, so the family arranges a second wife – a younger sister or a niece, for instance. Keeps the greater family intact, as it were.

    One of the records may have been created by someone who thought the step-mother was the actual mother.

    • Judy G. Russell says:

      Already have my 3rd great grandmother’s death record (she died in Ossig in 1864). It simply identifies her as Rosina. So the question is which woman had the call name of Rosina.

  5. Judy G. Russell says:

    A comment emailed from Ida Skarson McCormick points out that there’s one kind of DNA test that might work in this case:

    For xDNA testing one would need the right kind of descendants for Johanna, Rosina, and Sophia to test. The line likely to be most productive is the one that starts with a woman as #1 on a pedigree chart and with a mix of males and females goes to her #21 great-great-grandmother, from whom #1 gets an average of 25% of her xDNA (#21 = #1’s paternal grandmother’s paternal grandmother). That’s twice as much X as great-great-grandparents #22, 23, 26, 27, 29 provide #1 on average and 4 times as much X as great-great grandparents #30 and 31 provide #1 on average.

    Since there is less recombining of the X due to the males in the lineage, perhaps the xDNA will be more durable than autosomal. The testing great-great-granddaughters of the 3 sisters may have received separately identifiable X, due to recombining in each female in the line.

  6. Rondina says:

    My paternal family’s surname project has over 500 members. I have three 67 Y-DNA matches. They actively recruit participants in the UK to test and pay for these, but it is a common Welsh surname that spread over the UK before the the 18th century. I think that I have finally convinced them that they need to complete research here before trying to determine where the family was that migrated from the UK. They haven’t come close to straightening out the lines here and DNA isn’t going to solve that problem. We can all trace back to the same family that spread out in the states. There is one person who has “exhausted research” in the US with online resources and CDs who pushes the agenda. Ahhhhh! I finally had to say something. On the other hand a new 67-marker match led me to the correct SC county where my second great-grandfather came from.

    • Judy G. Russell says:

      I only wish I could research this family here before worrying about the migration. The problem is, I’m first generation American on my father’s side. He was born in Germany and emigrated when he was not yet four.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>