How does a does a data set like the US Census deal with people who defy your expectations? How can a data set encompass those that its designers never imagined? These questions cannot be answered with the published numbers, the final facts. They can only be answered by looking for deep stories within the data, stories like this one:
Margaret Scattergood had been one of the few women in the room where the census questions had been debated. A year and two months later, Scattergood faced those questions herself. He had to fit into the framework his colleagues had built. Scattergood had been an outlier in the Commerce Department auditorium; it would also be an outlier in the census sheets.
The 1940 manuscript census programs record the results of the conversations that took place at millions of doors as more than 120,000 census takers crisscrossed the nation and began asking questions. We are told, for example, that an enumerator named Richard Gray visited the Scattergood home in Fairfax County, Virginia, on May 25, 1940. He arrived late and arrived at this cottage almost a month after he was supposed to rural areas had to be listed; he had probably tried to list the household before but was unable to find the residents at home. On that visit, he appraised the home’s value at an impressive $50,000. It was (and is), by all accounts, beautiful, bucolic and grand.
Gray then listed three residents: Florence Thorne, 57, white, unmarried, with four years of college education, the “assistant editor” of a “labor union”; Margaret Scattergood, 45, white, unmarried, college-educated and “researcher” for a “labor union”; and May Stotts Allen, 50, divorced and (apparently) totally unschooled; she was listed as “W” for white, which was later crossed out with a darker “Ng” for “Black”. Gray was supposed to mark which of the people in the household she was talking to directly, but she didn’t in this case, so it’s impossible to say whether Scattergood came up with the questions herself.
We know how Gray made sense (in census terms) of these three middle-aged women living together. He made the eldest, Florence Thorne, the “head” of the household, writing her name first. Allen listed the last, as the “Maiden,” related by her servile status. Scattergood, in the middle, became a “Partner.”
“Partner” is a curious label, a term that can have a mixture of meanings. Partners can run companies or law firms. Some of us have partners in crime. These days, partner mostly means lover or partner. Used by gays and straights alike, married or not, couple now often indicates a long-term intimate connection. This usage is not even new; in his 1667 masterpiece paradise lostJohn Milton made the fathers of mankind partners, putting into the mouth of the lover of ur, the mouth of Adam, speaking of Eve, this lamentation: “I stand before my judge . . . for accuse my other self, the companion of my life.”
Is this what the census taker had in mind when he tagged Margaret Scattergood as a partner? There were just over 200,000 other people in the continental United States labeled as a couple in 1940. Were they all romantically involved?
For the majority in the first century of the US census, it was not possible to be labeled as a partner, as Scattergood was. The “Relationship” column, which asked all individuals in a household to explain their position in relation to the “head” of the family, did not appear until the 1880s.