About & methods
Historical investigation works best when the reader can see how the answer was reached, and where it stops.
I’m Jorge Aguirre, writer and editor of The Uncaptioned Past. This blog follows unidentified photographs and other archival mysteries: who or what is in the picture, where it was made, and how anyone might work that out.
The first two cases offer different starting points. The Historical Society of Cheshire County has identified a mill in Swanzey, New Hampshire. An ornate hall photographed by Ripple Bros. still has no secure name or location. Both have details worth following, and both leave work for the next person who looks.
You can reach me at clues@theuncaptionedpast.com with a lead or correction. The address forwards to my inbox. This is an independent blog; contributor and repository credits describe their contributions, without implying affiliation or endorsement.
What supports a finding?
Each case separates four kinds of evidence:
- Direct observation: what can be seen in the photograph or mount. If a small scan or incomplete view makes a feature ambiguous, the article says so.
- Documentary evidence: an identifiable record, such as a dated plan, directory, newspaper, accession entry or labelled photograph, linked where possible.
- Repository identification: a finding supplied by someone at a collection. The article names the contributor and explains whether the reported comparison material has been inspected.
- Hypothesis: an explanation to test. A resemblance or suggestive address can start a search; it needs support before becoming a finding.
Corroboration means independent support. Three websites repeating the same caption are one chain of information, however often the caption appears.
Why a case is identified, open or a dead end
An identified case explains the basis for accepting a name. That might be a repository identification, a documented visual match or independent sources that converge. Identifying a building does not necessarily establish its date or original use.
An open case names the missing evidence or unresolved conflict. It gives readers specific questions they might be able to answer.
A dead end records a test that reached its limit: what was checked, what came of it and what could make the lead useful again. It can save the next researcher a wasted search without declaring the mystery unsolvable.
Public research notes hold the fuller evidence trail, access limits and negative results. Research uses free public sources.
Credit, correspondence and images
An identification belongs to the person who supplied it. Archivists, local historians, readers and repositories receive credit for their contributions. Before publishing a reader’s name or attachment, I will agree on credit and permission.
Correspondence is summarized where it helps explain the evidence. Full private exchanges, contact details, email headers and internal identifiers stay out of the site.
Photographs link back to their source records and carry creator, repository and rights information where known. A repository’s “No known restrictions on publication” statement is quoted as its assessment. Source images appear without generated reconstruction; any future crop, annotation or adjustment will be labelled. A comparison image will be shown only when it has actually been inspected and can be reproduced.
Corrections and updates
Cases carry a review date and an update history. A substantive correction explains what changed and why, leaving any superseded conclusion understandable. Small spelling and formatting fixes need not clutter the record.
Send a correction with the claim and its source. A small detail can matter: a date on a plan, a caption on another print, or an alteration that makes a proposed match impossible.
Returning findings to repositories
Useful findings should reach the institution caring for the original. An identification submitted to a repository should include supporting references and the limits of the evidence.
The site distinguishes a finding prepared, submitted, acknowledged and added to a catalogue. The mill identification has been submitted to the Library of Congress for consideration. A catalogue amendment has not been confirmed.