edit in Google

Census Roll of the Three Affiliated Tribes at Fort Berthold Indian Reservation

This digitized dataset was found on the National Archives Catalog website. The data is of a census roll for the Three Affiliated Tribes located on the Fort Berthold Indian Reservation in central North Dakota between 1877 and 1902. The three tribes include the Mandan, Hidatsa, and Arikara Nation. This specific census roll is a part of a larger data collection series of Indian reservation censuses from the Record of the Bureau of Indian Affairs, maintained from 1877 to 1999, and organized by the Department of the Interior, Office of Indian Affairs at the Fort Berthold Agency; The specific individual who collected and recorded this record group, which is number 75, is unknown.

Between 1819 and 1864, the Upper Missouri Agency was responsible for keeping records of the total Indians living in the entire Northern Plains of the U.S. Although, there were complications with this broad data collection system; The agency did not have an organized central location, it was a very large span of land to keep track of, and the jurisdiction boundaries were not accurately defined.[1] Over its lifespan, the agency was subdivided to represent more concentrated and defined territories. By 1864, the agency was completely divided, by 1870 the Fort Berthold Reservation was created, and in 1871 there was an agent who was assigned to manage the Fort Berthold Agency located in central North Dakota.[2] 

The Fort Berthold Agency census record for the Three Affiliated Tribes between 1877 and 1902 includes 394 pages, with about 10 to 20 individuals per every other page. The structure of the data follows a two-page format; The left-hand page includes the names (Indian name, English name, or both), birth years, sexes, and Indian roll number of the individual recorded. The race of the individuals is also identified with the first initial of their tribe name. On the right page are the corresponding descriptions of the individual's relationship with the head of their household/family, usually a father or husband. An example of the description might be “son of number 84” or “wife to number 38 and daughter to number 72.”  Frequently, although not consistently, there will be miscellaneous descriptions of an individual’s personal information, such as “reads and speaks English” or “lives with…”  

The first major theme of the dataset to be noted is that towards the end of the dataset and beginning on page 348, descriptions of the individual’s relationship to the head of household, which would have been written on the right-hand side of the page, are no longer recorded. Likewise, roll number and race (tribe initial) are not present, either. Instead, what is recorded is the name and sex of the individual, along with an uncategorized one or two-word description on the same page. Some of these include “looking for medicine… many dances… spotted tail… yellow… laughing at battle…weasel woman...” (see page 366 of the dataset). Although, many pages after 348 do not include these types of short, miscellaneous notes. Perhaps this shift is due to a lack of available information, and taking down quick, two-word notes was the best that the agency could do. Likewise, the agency left the right page - which is where the relationship to the head of household descriptions would have been - blank, perhaps to indicate that if the information was ever found, it could be entered there at some point. [incert image of page 350 from digitized dataset]

A second theme is the impact of time on the context and visualization of the dataset.  The beginning pages of the census record are visually more faded and smudged than the ending pages, which have more sharp, dark handwriting. It can be assumed that over the decades, the ink faded from the natural wear and tear of the book. Likewise, it is possible that the quality of ink used to record this data improved over the years, which is why the writing from the ending pages, still from over one hundred years ago, is of quality and legible. On the contrary, in terms of context, the first half of the dataset contains information that is much more dense and concise than the second half of the database. Here, nearly all the categories (sexes, races, birth years, etc.) are filled out. The ending pages typically do not indicate birth years nor descriptions on the right side of the pages for the individual’s relationship to the head of the family. Some ending pages have this information present, but many do not. It can be assumed that since the right side of the pages and the column of where the birth date and race would have been left blank, that the information was not available, but the agency was keeping space if it were ever found. [Incert Census Roll Image A and Census Roll Image B]

The final theme of this data set is that it appears as though the outline or structure of the book in which this data was recorded was actually meant for something else. The pages are titled “Record of Employees” and the majority of the designated categories are not filled out by the record-taker according to the descriptions. For example, there is a column titled “were employed” and “date of nomination.” Neither of those category descriptions is used for their intended purposes. This might indicate that the Department of the Interior was using this record booklet because it was available and the majority of the categories - such as race, sex, birth year, etc. - were relevant to the census roll data that was collected. Likewise, the Elis Island Records (another student’s digitized dataset) appears to have this same issue, although not as severe. The heading of the Elise Island Records is titled “List or Manifest of Alien Immigrants for the Commissioner of Immigration.” Yet, the “Alien Immigrants” part is crossed out, likely indicating that the agency no longer wished to have that included in the description, but wanted to continue to use the document outline/format, as the categories were still applicable to the data that was collected. [Incert Census Roll Image C and Census Roll Image D]. [Incert Elis Island Record image]

This series of the Indian Census Roll of the Three Affiliated Tribes was digitized into microfilmed copies, preceively 198 digital images - a total of two physical pages per image, with around 10 to 20 entries per image[3].  The census is considered a record of the Department of the Interior, Bureau of Indian Affairs, but copies of the records can be found at the Family History Library in Salt Lake City, their Regional Archives, the National Archives, and Ancestry.com[4]. It likely took the National Arcives to digitize microfilm copies of this census roll around 3.5 hours, assuming that it would take around one minute to digitize one image, which includes two physical pages. The Indian Census Rolls were taken between 1877 and 1939 when congress mandated that the Commissioner’s Office of the Bureau of Indian Affairs start collecting census data on Indian populations in the United States.[5] This was because around the turn of the century, Indian territory was becoming a central focus for the U.S. government, and outtisde pressure was put on Congress and the Federal Government to address this land concern, as well. The Bureau of Indian Affairs was responsible for the the jurisdiction over Indian lands, and evenautlaly ended the responsibility of managing these lands but allowing them to be transferred to the U.S. government. In general, nearly all historical records regarding Indian populations were created by the U.S. government - the Bureau of Indian Affairs. In addition to census records, the Bureau was responsible for maintaining health records, school records, correspondence, annuity rolls, and allotment records. Additionally, records of these three tribes - called the Three Affiliated Tribes - were taken together because the devastation of the smallpox epidemic brought the three tribes together for economic social survival.[6] 

Likewise, Kinnahan’s Charting Progress: Francis Amasa Walker’s Statistical Atlas of the United States and Narratives of Western Expansion illustrates the lack or insufficient recording of Native Americans. In turn, this left the map for westward expansion blank or “unfinished” in the territories which Native Americans inhabited, therefore establishing the understanding that such land is available for white settlers and U.S. economic purposes. It is important to note that Walker’s Atlas only included a population that was broken down into the following categories: "free white males under sixteen, free white males over sixteen, white females, free blacks, and slaves." Likewise, by this understanding, Native Americans were not considered to be an important group. Interestingly, and around the same time period, the Bureau of Indian Affairs was collecting the census data on Three Affiliated Tribes inhabiting the central North Dakota territory, Fort Berthold. This was an established Native American reservation, which might have excluded the territory from being incorporated into other U.S. population and territory reports.

[1] “American Indian Census Rolls.” FamilySearch Wiki, https://www.familysearch.org/wiki/en/American_Indian_Census_Rolls.

[2] “Records of the Bureau of Indian Affairs, Reservation Census Series 1877-1938.” National Archives and Records Administration, Department of the Interior. Office of Indian Affairs. Fort Berthold Agency, https://catalog.archives.gov/id/7570275.

[3] “Records of the Bureau of Indian Affairs, Reservation Census Series 1877-1938.” National Archives and Records Administration, Department of the Interior. Office of Indian Affairs. Fort Berthold Agency, https://catalog.archives.gov/id/7570275.  

[4] “American Indian Census Rolls.” FamilySearch Wiki, https://www.familysearch.org/wiki/en/American_Indian_Census_Rolls.  

[5] “MHA Nation History.” MHA Nation, https://www.mhanation.com/history.  

[6] “Ibid