The Method
Four steps, and one standard of proof
Work of this kind has traditionally required a trained specialist proceeding record by record. This is an attempt to write it down as a protocol anyone can follow.
Normalize the name
Treat spelling drift across languages and scripts as expected noise rather than contradiction. Helman, Hellmann, Hilman, and in Cyrillic Gelman. Lando, Landa, Landau. This is the point where an ordinary name search fails, and where most people stop.
Reconstruct the logistics
Locate each person in time and place through the administrative trail: arrest, ghetto, transport, camp registration, transfer, liberation. Someone who cannot be found by name can still be found by where the paperwork had to put them.
Correlate with AI
Link one individual across archives, languages and scripts at a scale that is not feasible by hand. This is what makes the search possible from a desk, and it is the only step here that is new.
Verify and weigh
Check every linkage against the primary document, and weigh evidence by source type and independence. AI accelerates the search. Human judgment decides.
The standard of proof
An identification is admitted only where independent record types corroborate one another without contradiction: camp forms, civil registers, census returns, victim databases, testimony. Every linkage is verified against the primary document. Where the evidence supports a probable conclusion rather than a proven one, it is stated as probable.
Why women are the hardest to find
In this region, marriages that were religious only were not recognized by the state, so children were entered in the civil register under their mother’s maiden name, with the father acknowledging paternity in the remarks. My grandmother’s own birth record carries no surname for her at all. Her mother is the only parent in the column. Her father appears once, in the margin, swearing the child is his, and signing his name.
Most people searching for a grandmother are searching the one name they were told, and she is filed under another. It is the most common wall, and it is the reason step one comes first.
Sources cross-referenced
Arolsen Archives · Yad Vashem · USHMM · USC Shoah Foundation Visual History Archive · JewishGen Subcarpathia and Máramaros vital records · the 1921 Czechoslovak census · Buchenwald and Terezín rosters and card files · displaced-persons and Red Cross tracing records · cemetery transcriptions · community memorial books.
