Skip to main content
Guide

Somebody's relative recognised them and wrote in, decades late

In the summer of 1936 two photographers documented the same crisis. One wrote down names, the other did not. Ninety years on, the Library of Congress credits one identification simply to a family member. That is the whole argument for writing names down while somebody still knows them.

The same summer, the same crisis, two different records

In 1936 the Farm Security Administration sent photographers across the United States to document what the Depression was doing to farming families. Dorothea Lange followed the road west and wrote captions that described a condition: destitute pea pickers in California, a migrant agricultural worker's family, seven hungry children.

In Hale County, Alabama, Walker Evans sat in three houses and wrote down who was in front of him. Allie Mae Burroughs. Floyd Burroughs. Bud Fields. Frank Tengle.

Same programme, same year, same country. One set of families kept their names. The other set became a description of a condition, and stayed that way.

What we found when we checked our own captions

We built a public exhibit from fifteen of these photographs, writing each plaque from the archival record and stopping where the record stops. Then we fact-checked our own plaques against the Library of Congress catalogue, expecting to confirm what we had written.

We were wrong twice, in the direction that matters. Two families we had recorded as anonymous had already been identified, years after the photographs were taken. And the Library credits one of those identifications simply to a family member.

Somebody recognised a face in a federal archive, and wrote in. Nobody paid them, nobody asked them, and the record is better for it forever.

Eight of fifteen have names. Seven do not.

That is the honest count on the wall today, and it is not going to improve on its own. The photographs are ninety years old. Everyone who could have supplied a name from memory is gone, and what is left is the slow chance that a descendant recognises a face.

Your own boxes are the same problem, at an earlier stage. The difference is that the people who can still answer are alive, and reachable, and will answer a photograph in a group chat faster than they will answer a question about the past in the abstract.

  • A name written on a plaque outlives the person who remembered it.
  • A name nobody writes down is lost at the speed of one funeral.
  • The archive itself shows people do write in, given somewhere to write to.

What restoration can and cannot do here

We ran all fifteen through the same restore anyone gets, and judged the results by eye against conditions written down before the run. Two earned a before and after slider. Three were rejected because the model invented detail on a face, including freckles on Allie Mae Burroughs that do not exist in the negative. One came back byte for byte identical to the original.

So the wall shows each photograph as it survives, and the restoration is something you pull across it rather than the thing you are handed. Recovering a face is a task software can attempt. Recovering whose face it is remains something only a person can do, which is why every plaque that has no name asks the question instead of guessing.

Walk the exhibit

Related posts

Naming Anonymous Photos: What the FSA Archive Already Proved · Nostalgia - Living Archive