Friday, May 30, 2014

Darned Boys in Dresses

Records say the darnedest things

We depend upon records to reveal the “truth” about our pasts.

Yet sometimes records have anomalies.
Some are amusing or humorous.
Some are interesting or weird.
Some are peculiar or suspicious.
Some are infuriating, even downright laughable.

Yes, “Records Say the Darnedest Things.”

Records Say the Darnedest Things:
Darned Boys in Dresses

Note this family in the 1870 census.1


Here’s the same family in the 1880 census.2


“Harriet Gray, a female born about 1865 in the 1870 census, has become Harry, a male,” wrote Melinda Daffin Henningfield in an article about the family.3

Other, independent, evidence corroborated the name “Harry.” At that point, I would have been all done. Problem solved.4

But the genealogical proof standard requires another step: resolution of conflicting evidence.5

That step had always been a bit of a mystery to me until Thomas W. Jones’s book, Mastering Genealogical Proof.6 In my mind a matter is not resolved until it is proven. Jones clarified what is needed. “Resolving the conflict requires us to separate the evidence into likely-correct and likely-incorrect answers, discard the incorrect answers, and justify or explain that separation and discarding.”7

In Melinda’s article, I saw this practice in action. She wrote,

The informant for the Gray family in the 1870 and 1880 censuses is unknown. In the 1860s, “[u]ntil about age five, boys were kept in skirts.” (Joan Severa, Dressed for the Photographer: Ordinary Americans & Fashion, 1840-1900 [Kent, Ohio: The Kent State University Press, 1995], 210) The enumerator in 1870, perhaps not being familiar with the family, may have seen Harry in a skirt and heard Harriet.8

Simple, direct, and a reasonable explanation for the conflicting evidence.

Thank you, Melinda for helping teach me about resolution of conflicting evidence.


     1.  1870 U.S. census, Mississippi County, Missouri, population schedule, James Bayou Township, p. 9 (penned), dwelling 71, family 74, Harriet Gray; digital image, FamilySearch (( : accessed 4 May 2014); NARA microfilm M593 roll 792.

     2. 1880 U.S. census, Hickman County, Kentucky, population schedule, Columbus Magisterial District #1, ED 113, p. 10 (penned), dwelling 95, family 104, Harry C.; digital image, FamilySearch ( : accessed 4 May 2014); NARA microfilm T9, roll 420.

     3.  Melinda Daffin Henningfield, “Susannah or Mary: Who Was the Mother of Robert White Gray (1858-1935) of Hickman County, Kentucky,” Crossroads, Winter 2014, 16-25.

     4.  I wrote about growing knowledge and practices—maturity—in a series of articles. “Genealogical Maturity Model,” The Ancestry Insider, blog ( : accessed 4 May 2014). This post contains links to the other articles in the series.

     5.  Board for Certification of Genealogists, Genealogical Standards, Thomas W. Jones, editor, 50th Anniversary Edition (Nashville, Tennessee:, 2014), especially pp. 1-3. Also see “The Genealogical Proof Standard,” Board for the Certification of Genealogists ( : accessed 4 May 2014).

     6.  Thomas W. Jones, Mastering Genealogical Proof (Arlington, Virginia: National Genealogical Society, 2013).

     7.  Ibid., 74. He continues with three situations in which the reasoning can be applied. I’ll not detail them here. Buy the book.

     8.  Henningfield, “Susannah or Mary…”, 20.


  1. There are several other rational explanations to these two records.

    * Hariet age 5 in 1870 was a female who died before 1880, and Harry was a male born after the 1870 census whose age was incorrectly given in the 1880 census.

    * Was Susan age 23 in 1880 the same person as Robert in 1870 age 12? Why isn't Susan in the 1870 census?

    Obviously, the resolution of all of these questions requires much more study - doing a reasonably exhaustive search (e.g., the 1860 census, later census, church, birth, marriage, death, military, newspaper, cemetery, probate, land, court, etc.) for all of the persons, gathering and analyzing all of the evidence, resolving conflicts, and writing a proof argument, etc.

  2. Yay, Randy!

    I think using only two records to prove/disprove is inadequate. There's a 50-50 chance one of the two is incorrect or there is a chance BOTH of the records are incorrect (see Randy's reply). Much further research is necessary.

    I had two children that 'disappeared' from my father's sibling list. It's taken me nearly 20 years to track down their existence and find what happened to them since my father & all of the rest of his siblings are gone. It turns out that even their given names -- repeated in our family lore -- were incorrect. They never appeared on any census forms (born/died in between census years) and were buried in a family cemetery miles from the city they were born in & which is no longer in active use. By using digital records (not yet indexed), I finally found the evidence of their correct names & birth dates, and the local paper (again after an online intensive search) revealed their death dates and burial information. If I'd only gone with the census records, I'd have had to conclude there WERE no children born, let alone that they were twins and died separately in one of the coldest winters the U.S. ever experienced.

    The only way to truly resolve questions is to use as wide a base of information as possible. And continue to keep an open mind, because newly revealed information/records just might blow a very large hole in what you regarded as irrefutable 'proof.'

  3. I followed the link in AI's footnotes to the 1870 Census. Susan appears in the 1870 census several lines below the rest of the family (line 17).

  4. There were two census of Wards Island in 1870. My gggrandfather's sister was enumerated in one as Rike, a male, and not in the other at all. Her name was Christine Friederike Marie. She must have been called Rike. She was 8 years old.

  5. Dear Randy and ponyswimgal,

    Thanks for your thoughtful comments. You are absolutely right. You'll be glad to know that Melinda did perform a reasonably exhaustive search. When I wrote this article I kept being drawn down the path of showing the supportive evidence from Melinda's 10 page article. My article kept ballooning in size and infringing on Melinda's copyright.

    A key piece of evidence was a family Bible that helped define the composition of the family. And, as Karl points out, I neglecting to include Susan in my screenshot for the 1870 census.

    Crossroads is published by the Utah Genealogical Association. The issue containing this article, as well as other back issues, is available to members for download.

    --The Ancestry Insider

  6. My mother said her brother was in a baptism dress in a picture of them, Karl was in a dress and I have many baby pictures of boys in a dress looking outfit.


Note: Only a member of this blog may post a comment.