Historians’ ethics, women’s history, and the infinite archive
Lately, I’ve been thinking a lot about historical sources and the implications of digitizing them. I work a lot with a particular collection of government papers that’s full of ordinary women’s letters about their experiences as mothers during the early 20th century. It’s hundreds of linear feet, and only a small part of it has been microfilmed. It’s easy enough to get to; you don’t have to be a professional historian to read the collection; and it’s open to the public. But it’s still physically located in one place, which limits the number of people who can read it. (The curious can read a published selection of the letters in Raising a Baby the Government Way.)
The collection’s also a fabulous target for digitization. It’s part of a federal government collection held by NARA; it’s civilian, unclassified records, and as far as anyone can tell me that means that it’s in the public domain. It’s not encumbered by intellectual property issues or digital-photography restrictions. In fact, I’ve spent weeks looking through it, digital-camera-photographing materials that are relevant to my research— the ability to do that has literally made my current project possible. There’s nothing stopping me from putting that material online, save my own limited time and budget— and I’ve been thinking hard about doing just that. We need more women’s history sources in the infinite archive.
But here’s the problem: the letters in it are full of intimate stories, and this poses a problem that I haven’t seen any digital-methods historians addressing.1 Right now, because they’re physical papers stored in boxes, these documents enact a kind of privacy through obscurity, and any intelligent computer user knows that security through obscurity is a horrible way to keep a secret.
When historians quote from this collection, they usually obscure identifying details, out of respect for the privacy of people who may still be living, but they cite it in a way that it can be found by other historians in the physical archive. So I’ll do the same here and quote an entire letter in full, omitting identifying personal information.
This is a letter written to Eleanor Roosevelt by an African-American mother who faced a complicated family situation and was looking for some help. (I have an image of it which I’m not posting here, because the content of the letter would require me to redact it heavily.) It’s written on hotel stationery from Chattanooga, Tennessee, in pencil; the handwriting gets looser and less composed over several pages, as its author got more upset about the problem she was describing. (There aren’t any paragraph breaks in the original, either.)
Dear Mrs. Rosevelt,
Will you please help me I am a negro girl I am married. I am going to tell you every thing so you will under stand. I married my husband Nov 23 1938. Diden any body know until June 1939. I was going to have a baby. My husband sisters did not like it be cause he married so they did every thing to break us up and succeeded We got a baby born Nov 5 1939 name [redacted] in Atlanta Ga. The baby was born 7 months come be fore time I want a birth certificate I cant get a regular job without it I have wrote to the Bureau of Vital Stitictis for a cercitific in Atlanta 2 months a go hasent heard from them will you please help me I cant have my baby be cause I cant make my husband keep me without it. His sisters wont let him help me but I can get help if I get the cificitate I cant bring the baby home until I get some help but they wont help me in Atlanta the babys name is [redacted]. Fortha [Father's] name [redacted] Mother Susie Mae [redacted] Please help me I wont my baby my brother has got her untill I can take care of her
Susie Mae [last name redacted, with an address.] 2
I could describe a number of interpretive points about this document, but for now, trust me: this is an interesting letter—both for the gritty detail it gives about one woman’s hardships, and for what it says about how Americans learned to use birth certificates. But for me as a historian, this letter writer’s personally identifying information is Really Important. Since I know her name, her age (and thus her approximate birth year), and her address, I can use a bunch of online tools to find out about her.
Thanks to a digitized, searchable 1930 federal manuscript census, I’ve looked her up. I can tell you all about her husband and his family. I could probably use the Digital Sanborn Maps to find out more about the house she was writing from, and I might even be able to find it on the Google Maps Street View. Since I know her child’s name, and her child was born in the late 1930s, it’s entirely possible that I could find the “baby” described here, who would be about 70 years old today. (That’s a step many historians wouldn’t take unless they really needed some information. For my project, it’s not important.) But most of what I’ve just described is absolutely standard historical research practice: when you want to find out about someone, there are ways to learn if you’ve got the patience to work through them. It’s just that modern digital research tools make the legwork faster.
If I were putting this collection online with good semantic web metadata, it’s entirely possible that I might add automatically-generated links to such services: “Search for this person in the US manuscript census,” “Investigate this address,” and such. (There’s a workshop next weekend in Toronto that’s going to be thinking about the technical issues involved in building APIs for digital-humanities web services. What I’m describing is entirely feasible, from a technical perspective—if not today, then very soon.)
In short, I could make the lives of Susie Mae M____’s descendants very… interesting. And, given the content of this letter, probably not in a good way.
Not all of the letters in the Children’s Bureau papers are nearly so scandalous. Many of them are questions about childrearing in one form or another, and many are from women in tiny rural towns. Any really useful project would require that the images of the documents have metadata added: the author’s name, city, state, and such; lots of work, possibly automatable or crowdsourceable, but the results would be amazing. (Wouldn’t it be cool to google your great-grandmother’s name and find a letter she wrote in the 1920s, when she was trying to learn about the best diet for your then-2-year-old grandfather?) And putting the collection online would make these thousands of letters more accessible than they are now, where they’re hardly even indexed.
What are the ethics of putting materials like this online, in bulk, for use by digital-methods historians? I’m not sure I know. On one hand, if I revealed the last name of Susie Mae M____ by putting the images of that letter online, I might find myself answering some very testy correspondence by her descendants, who don’t want their family’s historical “dirty laundry” aired on the Internet.
On the other hand, historians who work on the history of women and sexuality have spent a lot of time talking about how defining and separating “private” from “public” does political work. I’m not willing to say that just because Susie Mae M____’s descendants might object to that letter’s publication on the internet, we should never digitize “sensitive” material. If we keep “private” or “scandalous” materials out of digital primary source collections, then (to borrow Estelle Freedman‘s phrasing) the burning of letters continues online, even if the paper copies still survive.3
What do you think the best approach is? What haven’t I taken into consideration?
- Then again, I’m a finite person. Do comment and let me know what I should be reading. ↩
- Susie Mae M___, Chattanooga, Tennessee, to “Mrs. Rosevelt,” May 11, 1940, Folder 4-2-1-2 Birth Registration, Box 730, Records of the Children’s Bureau, RG 102, National Archives, College Park, Maryland. All spelling and punctuation is consistent with the original. ↩
- We could put it in a walled garden where only “serious scholars” can search for these materials, but that approach is fraught for entirely different reasons. (First: define “serious scholar.”) ↩