View Full Version : False interpretation of "race" in Virginia Marriage Records, 1936-2012"

08-20-2016, 05:28 AM
This doesn't really concern Ancestry as a DNA testing company, but I ran into what appears to be a rather "interesting" error in interpreting some records they've put online.

Specifically, the issue is with "Virginia Marriage Records, 1936-2012".

I was doing some research for a grandniece and grandnephew of mine. Their mother is my wife's niece. She is white, but her husband is African American. At the time, I was looking specifically at ancestors of her husband's.

I was pleased to find a marriage record one of his collateral relatives. I think it might have been an uncle. [Edit: Initially I think I said the record was about my niece's husband's grandparents, but that was incorrect. The point remains the same, though.] Anyway, I was startled to see that he was listed as "white". The parents, who are included in the record, are consistently listed in the U.S. censuses as "Negro" or "Black".

The thing is, though, when I viewed the scan of the actual document, I saw that it does not say that the groom's race is white. Instead, in the box marked "race", I could see the letter "C". But that raised a question in my mind -- was it perhaps, "C" for "Colored"? To find out, I started looking at other records in the set.

What I found was that in most instances, the box for race usually had the word spelled out in full: either "White" or "Colored". Sometimes, "Col'd" was used, as well. There were also occasional single-letter abbreviations, either "W" or "C".

Clearly, then, "C" did not mean "Caucasian", and the writer of the summaries seemed to have believed. Yet, in every instance in which I found the single letter "C" for race, the summary interpreted this as "White". This seems to be a fairly big "oops", in my opinion.

What I suspect is that whoever made the summaries may be young enough to believe that surely an official state document wouldn't use a word like "Colored" for race. But, yeah, they really did that.

08-20-2016, 05:29 PM
I don't mind seeing the term "mulatto", which is prevalent throughout the 19th century, but get annoyed when in the the 20th century all of a sudden the same people are listed as "Black".
I have heard of at least one case of a person whose ancestor's were classified as African-American having their DNA tested and coming up 100% European.
dp :-)

08-20-2016, 07:07 PM
Yeah, it can get pretty strange.

I guess I'm straying from the topic of Ancestry as a company (and especially as a DNA-testing company), but you might be interested in the case of a woman named Susie Guillory Phipps. In 1977, she needed a copy of her Louisiana birth certificate. But when it arrived, she was stunned to find that it identified her as "colored"; all her life she'd believed she was white, and so did everyone around her. Besides, that she'd twice married white men -- including her current husband. Fortunately, because of the U.S. Supreme Court's decision in the Loving case in 1967, she at least wasn't in any legal jeopardy.

Nevertheless, she wanted to get her birth certificate changed. However, under Louisiana law anyone with at least 1/32 African ancestry was considered black or colored, tor whatever term was used at the time. According to the genealogist working for the state, Phipps was actually 3/32. That's less than an eighth.

So Phipps pursued the matter in court and lost. It went all the way to the level of the Louisiana Supreme Court, which ruled against her; and even the U.S. Supreme Court, which declined to hear the case. That, of course, allowed it to stand.

Eventually, Louisiana repealed the law, but they didn't make it retroactive. So it doesn't apply to Phipps. In some ways it might be considered to no longer matter, since the anti-miscegenation and other discriminatory laws no longer apply. I still think the U.S. Supreme Court should have heard the case, though, and overturned the state courts.

But ... I digress. My main reason for posting earlier had just been as a caution when looking at the transcripts or summaries of various records. Clearly, the people making them don't always really understand what they're looking at. I presume that the same person must have seen some of the documents that wrote race out in full as either white or colored, so I really can't understand how they could have interpreted "C" as "caucasian". But that's the only thing I can figure to explain why the records marked with "C" for the race are showing "white" in the summary.

EDIT: Forgot to include my references.


2nd EDIT:

One further point, which I think was made by one of the "expert witnesses" for the plaintiff, is that it really isn't possible to say how much "African" ancestry Phipps' slave ancestor(s) had to begin with. [I say "slave" ancestors, but I can't recall right now whether any source I've seen identifies her nearest black/African ancestors as having been slaves, or Free Persons of Color. It isn't really relevant to my point, though.]

Sally Hemings, for example, was 3/4 European, even though she was a slave. Today, she has descendants who identify as white; and descendants who identify as black or African American. (The descendants in both groups are widely believed to also be descendants of Thomas Jefferson.)

Obviously, in none of them can start with Sally Hemings as "100% African" if they're trying to determine their own degree of African admixture.

08-20-2016, 07:31 PM
I might also note that members of the family that got me started on this topic are actually listed as "Mulatto" in either the 1870 or 1880 census (I don't remember which right now). But this is another case that also involves poor transcription. They're living in a household in which the "head of house" is white, indicated by "W" in the race column. So are the members of his family. The family I'm interested in consists of a family servant (not a slave, since it was after the Civil War) and his family.

In the original document, it's possible to tell that they do not have a "W", but an "M". They do look a little similar, except the "tail" of the W always goes up; the "tail" of the M is always down. (If I'd do a screen capture and post, you'd immediately know what I mean.)

Yet I can't find any subsequent census that says "Mulatto". All say "black" or "negro" or some such thing.


It was the 1870 census, which is significant in that relationships to the head-of-house only came to be actually listed -- rather than perhaps implied -- in 1880.

08-20-2016, 07:45 PM
Here's the difference between "W" and "M" in the 1870 census document I referred to earlier. I really can't understand why the transcriber wasn't able to distinguish the two. It definitely wouldn't have been a trivial distinction in 1870!