View Message

Subject: Perils of taking census records as the gospel truth
Author: Cleveland Kent Evans   (Authenticated as clevelandkentevans)
Date: March 20, 2008 at 12:36:11 PM
I've been having fun the last couple of days checking out the names in a book called Bad Baby Names published by two guys who work for the website The book is a compilation of interesting names they have found in the records on that site. The great majority of those records are US Census Records.

I should put in a plug for as I have realized even more than before what a gold mine of data they have. You can look at the original census records for the entire United States, and they have the indexes of the records completely computerized so you can look up people by either last or first name, or any combination of the same, and find out how many times such a name occurs in the indexes to the censuses.

However, what a bit of research shows is that a lot of the supposed "real names" they have in the book are simply wrong. The most obvious example is "Hades Fryher", the supposed name of a girl living on Long Island in 1920. If you look at the record, at first glance the first name does seem to be Hades. But if you then pay attention to how that particular census taker wrote his capital G's, you can figure out that this girl was named Gladys, not Hades. His weird "G" and the "l" get together to look like a very fancy sort of H, but that's not what he meant to write. And that is completely confirmed by finding the family in the 1930 census, when the census taker that year had much better penmanship and clearly wrote "Gladys" as the name.

Before recent times census takers were not hired because they had good penmanship or good spelling. And they wrote down what they heard a name to be. People at different times use different forms of their names, and census takers freqeuntly misheard, misinterpreted, and misspelled what they were told. And they were more likely to do this if the name was uncommon and so one they were unfamiliar with.

I checked this out by looking up the records for one of my own great-aunts in the US censuses on My great-grandparents named her after a family friend called Alexander Tennant, and her official first and middle names were Alexandria Tennant. As with many people born in the late 19th century, she was usually called by her middle name. So the family knew her as Tennant or Tennie most of the time.

In the 1900 census, she is listed as "Alexander T.", as if her parents had given her the normal male form of the name. So someone looking for "girls given boys' names" in 1900 would think she was a case of that, when she wasn't.

In the 1910 census, she is listed as "Tennet". And whoever prepared the index, not being familiar with that, read it as "Jennet". So if you look for her according to her real name in the 1910 index, you won't find her.

In the 1920 census, she is listed as "Alexandria T.". That is her real name, but perhaps a bit misleading, since she wasn't called Alexandria in everyday life.

In the 1930 census, she is listed as "Tennant". However, once again the indexer misread the census taker's bad handwriting and came up with a name he or she was more familiar with, and so in the index to the 1930 census she is listed as "Toussaint"! And people will wonder why a 35 year old woman in rural West Virginia had a male French first name, if they simply count up the Toussaints found in the index.

So one really has to be careful in interpreting these records. Here my great-aunt is listed in four different ways -- six different ways if you count the indexes -- and someone who didn't already know what her real name was would be very confused.

This message was edited by the author on March 20, 2008 at 12:39:26 PM

Messages in this thread: