Some people eat, sleep and chew gum, I do genealogy and write...

Thursday, November 19, 2020

You Can Read Handwritten Documents! -- Looking at the 1930 US Census Post #5


"United States Census, 1930," database with images, FamilySearch ( : 8 December 2015), Arizona > Navajo > Holbrook > ED 7 > image 13 of 20; citing NARA microfilm publication T626 (Washington D.C.: National Archives and Records Administration, 2002).

Note: This is not really a series although reading all the posts with this topic are useful. If you want to find additional topics, do a Google search for the first part of the post name and add the term "Genealogy's Star, like this "you can read handwritten documents! genealogy's star"]

I am working backward in time looking at the handwriting of each year of the U.S. Federal Census. Handwriting is one of the major challenges in reading old documents. Although the first mechanical writing machines were first used in the 1500s, typewriters did not become common until the 1900s. As genealogists, this means that research back just a few generations will depend almost entirely on handwritten documents. See the following for the history of typewriters.

Weller, Charles Edward. 1918. The early history of the typewriter. La Porte, Ind: Chase & Shepard, printers. Ebook

Vrooman, J. W., Herkimer County Historical Society. (1923). The Story of the typewriter, 1873-1923. Herkimer, N.Y.: Herkimer County Historical Society. Ebook

Handwriting in the 1930s is not much of a change from the 1940s. The major difficulty is usually the handwriting skill of the enumerator. The example above from Joseph City, Arizona is quite readable but if you compare the style to the predominately used Palmer Method of Handwriting, you will see some significant differences. Here is a chart of the Palmer Method characters. 

The Palmer Method of Business Writing. A.N. Palmer Published New York, etc., 1901. Page 29

Here are some of the major departures on the census record.

In this example, the letter variations do not interfere much with the readability. Although some of the other letters, especially letter pairs may cause indexing problems. Here are some of the letter pairs that could cause confusion.

The first example, starting at the top, is "Tanner." In some indexes the name has been indexed as "Tamer." The next example is "Beulah." If the indexer was not familiar with this name, the writing could cause some difficulty. This is the same problem caused by "Reed" and the last example, "Henry." 

The way to avoid problems with these names is to spend the time reading the census and if in doubt about a name, take the time to do some quick research about the person while making the assumption that you can read what has been written. For example, the first name circled above is most likely Clifford Tanner. A search on brought up the name of the individual in the Family Tree. 

You can match the name and his wife's name and the chldren's names. This illustrates an important point, you need to be familiar with the possible names of your ancestors and relatives. If I go a short way down the page, I will find another Tanner, this time, it is my Great-grandfather and his wife. 

The name is written the same way. You may also need to look at more than one census year for the specific location to get alternative spellings of the names. I must admit that years ago when I started looking at the census records, I could not find some of my Tanner relatives in some of the census years because they had been wrongly indexed and it wasn't until I learned to read the entire town's records that I was able to sort out my ancestors from the indexing mistakes


  1. In my small attempts at indexing, I did as you said, using further research to identify the name. My work was repudiated by FamilySearch and I was told to record as written. No more indexing for me.

    1. Since FamilySearch is just now beginning to use OCR software to index records and there is still no adequate handwriting recognition software and FamilySearch is also just barely allowing users to make correction to indexing errors or OCR errors, it seems interesting to me that they will not allow the indexers to do any research to enter correct entries.