O ano da fotografia (The year on the photograph)

The photograph arrived at the town museum in a plain envelope. Six people stood outside a shop whose window carried the words Bell Street Bakery. On the back, someone had written “1934?” in pencil; the question mark was almost as faint as the figures' faces. I was asked to prepare a caption for a small exhibition about work in the neighbourhood. By the end of the afternoon, I had written, “Staff of the Bell Street Bakery, 1934.” It sounded satisfyingly complete.

The trouble was that our evidence was less complete. Trade directories showed that a bakery traded at that address between 1931 and 1947. It did not tell us when the photograph was taken or whether the people in it worked there. They might have been customers, relatives or visitors. The pencilled date was a clue, but we could not identify the writer or discover what it was based on. My caption had quietly turned two reasonable possibilities into two apparent facts.

At the planning meeting, one trustee argued that visitors would lose interest if every label announced what we did not know. I could see the point. A wall of cautious language can make even a well-supported fact sound doubtful. The curator therefore suggested beginning with what the directory established, then naming the limits in a separate sentence. Her proposal was neither a confident story nor an apology for having no story at all.

We changed the caption to: “Outside a bakery on Bell Street. A business at this address appears in directories from 1931 to 1947. The date of the photograph and the identities of the people remain unconfirmed.” The question mark on the envelope went into the catalogue record, with a note about its uncertain author. That detail was not put on the wall: it was useful for future researchers, but it would have made the public caption harder to read without making it more accurate.

During the exhibition, a visitor said that one of the figures might be her aunt. She remembered a family story about the bakery, although she could not say which person in the photograph it described. We recorded her account under her name and invited her to bring any letters or photographs that might support it. We did not add the aunt's name to the label. A new lead can change a research plan before it changes a conclusion.

The revised caption did not answer the question that first drew me to the image: who were those people? Yet it gave visitors a firmer place from which to ask it. One person noticed a mark on the shop sign that we had overlooked; another asked why the envelope carried a question mark. Their questions might lead nowhere. They were still better than the false certainty of my first sentence, which would have made further questions seem unnecessary.