Thursday, September 29, 2011

About the Danish West Indies Census

One of the reasons I decided to start this blog was to share information about the records, resources, and techniques of Virgin Islands genealogy. Periodically, I will post information on the types of records I find to help other researchers (who may not care about my particular family) understand what's available, how to get it, and how to use it. It seems natural to begin with the most commonly sought genealogical records: the censuses.
Detail from 1841 Census showing my great-great grandmother, Sophia Andersen and her mother Eliza Scott
Population Censuses were taken at irregular intervals starting in the mid 1800s. These census records are much like US state censuses in that they were collected locally and maintained locally. For St. Croix, that means that the papers were kept in an environment best described as an archivist's nightmare. Tropical heat and humidity has destroyed many records. Luckily, many of the records survive and are available in microfilm or digitized form.

  The earliest censuses I have found date from 1835. Censuses were also taken in 1841, 1846, 1851, 1855, 1857, 1870, 1880, 1890, 1901, and 1911. The US bought the islands in 1917 and immediately conducted a full census using the US 1910 census forms. Since the enumeration was conducted in 1918, this was used as the 1920 US census and appears in that collection. Starting in 1930, the Virgin Islands participated in the regular US censuses.
The original Danish census records are kept at the Danish Archives, Statens Arkiver, in Copenhagen, but a few pages are in the NARA Record Group 55 in College Park MD. The FHL has images of the Copenhagen documents and the Virgin Islands Social History Association (VISHA) has indexed the St. Croix portion. They are working on the St. Thomas records (including St. John), but they aren't available yet. The St. Croix index is available at the VI Family Roots site, and also at, which has images of the pages.

When I first saw the document images, I noticed something unexpected. Although the official government language in the islands was Danish, the pre-printed census form, and attached instructions, were in English. I later discovered that English was the default language of the island, and most people didn't speak Danish at all. Enumerators filled them out in whatever language they personally spoke, either English or Danish. This is most commonly seen in street names. As an example, Fisher St. is sometimes listed as Fiskergade. Hill St. is listed as Bjergade. Most are in English though. (If this seems odd to you, as it did to me, take a look at the 1930 US census for Puerto Rico. The whole form is in Spanish.)

Signature of Sophia Andersen on 1841 census
The Danish West Indies census forms are far simpler than the US census, recording only a small bit of information about each person. Each city address was recorded on its own sheet, sometimes with multiple households indicated. The plantations (estates) usually needed multiple pages. Children were usually listed in descending order of age. Other than that, there isn't a lot of consistency in how things were recorded. Religion is usually indicated, but dates of baptism are sporadic. Sometimes the Head of Household signed the sheet (nice when you can find it) but usually only the Enumerators signed them. They also usually indicate the property owner, which is extremely helpful in identifying family connections. Fields include:
  • Name
  • Sex
  • Address
  • Age
  • Religion
  • Date Baptized
  • Birthplace
  • YearArrived
  • MaritalStatus
  • Occupation
  • Relationship to Head
  • PropertyOwner
Prior to the abolition of slavery in 1848, the census listed slaves as well as free people, so they are particularly valuable for researching the "unfree" population. Unfortunately, slave or free status and racial divisions were not recorded.

Often the responses recorded are amusing as well as informative.  In one census for my family there was an unrelated 9-year old girl who was listed as a "stranger".  Sometimes there is extra editorial content. For Occupation, one respondent listed,"shoemaker, but at present can scarcely make a living by it in consequence of the great importation of foreign shoes" I'll post more of these responses from time-to-time.

As helpful as censuses are to genealogy, they are more so for St. Croix. Most people were born, lived, and died on the same island, an island with at population of about 20,000. This makes it possible to find people by "process of elimination". I was able to locate my great-grandmother's maiden name this way. I'll share the details of that investigation in a future post. They are also a great source of statistical data to better understand the composition of the population and changes over time.


  1. We need for the Methodist, Moravian and Catholic Churches to put their records on line. The churches in the Virgin Islands have a wealth of information about the people who have lived there. The Virgin Islands Government also need to cooperate in sharing their information on births, marriages, and deaths. The Social Security Death Index used to be a great source of information about our relatives who have died. Now they have stopped being of assistance in documenting the lives of our ancestors.

  2. That is completely true. I am fortunate that most of my family was Lutheran for over a century and I have been able to glean so much from those records. Sadly, they converted to Moravian in the mid 1800s and I have a hard time assembling from there. Some of the records are in NARA RG55 for the late 1800s. They may be of help.

  3. Great blog! Have you heard any recent updates on the indexing of the st thomas census records?

  4. Slavery was abolished in 1848, the date listed above says 1948. Sometimes autocorrect can incorrectly fix things in your document or your fingers can fly across the keyboard so fast, it's easy to hit the wrong key and not notice. I know I am guilty of this. I noticed this error and just wanted to alert you. The blog and project is really nice but I wish the records were more elaborate and chronologically defined. It is hard to find a complete file but something is always better than nothing. Thank you!

    1. Thanks for the catch! You are so right, the records are very hard to find. The two large repositories, NARA and RA have very minimal descriptions of their collections. Within each record set there is little organization at all. I just spent some time on a project at NARA and found boxes of completely unorganized documents, tied together with string, that appear to have been untouched since the 1950s.