Working note · from our own desk
When a parish register is lost, most people stop. They shouldn't. Tax censuses of households — revision lists — survive in state archives when church books have burned, and they are built in a way that hands you two things a parish register never will: a death date for a man whose burial entry is gone, and, if you read the column headings carefully, the name of a father who was already dead before the record was written. This is how we read them.
The structure
A revision list is not a snapshot. Each household entry carries, for every male, two ages side by side: what he was recorded as at the previous revision, and what he is now. Between those two numbers the clerk writes what happened — died in such a year, taken as a recruit, moved to another village, born since the last count.
That structure is why the layer is so strong. The list is self-linking: the "previous revision" column of one book must match the "present" column of the book before it. When it does, you are not guessing that two men in two documents are the same person. You are reading the same entry, carried forward by a clerk who had the earlier book open in front of him.
From a household numbered 71 in an 1811 list, under the heading "the children of Ilya son of Ivan":
Three lines, and they contain: two brothers, a grandson, a death date, and the fact that Tymofiy was born after the previous count. Then the next list, fifteen years later, opens its "previous revision" column with 45 for Semen and 9 for Tymofiy — exactly these numbers. The chain is closed without a single leap of faith.
Method
A matching surname proves nothing. Spelling drifts from book to book — we work with one name that appears in thirty-five documented forms, including one broken in half across two lines by a clerk who ran out of room. What proves kinship is a chain of things that cannot all agree by accident:
Where these disagree, we write down that they disagree. A family history that quietly smooths over a bad match is worth less than one that says plainly: this link is probable, not proven.
The trick worth knowing
Look again at that heading: "the children of Ilya son of Ivan".
Ilya himself is dead — he isn't in the list, only his children are. But the clerk named him the way people were named: given name plus father's name. Which means the document silently records Ivan, a man who died before the register was written, who has no birth entry anywhere, and whose existence is stated by nothing but his son's patronymic.
That is not a guess. It is how naming worked, and it is accepted evidence of a father's given name. It is also the honest limit of what it proves: you have his name and roughly his generation, and nothing else. On a family chart we mark such a person as named through a patronymic rather than found in a record, because a grandchild checking the work deserves to know which is which.
This layer is where most of the depth in a Central and Eastern European family line actually comes from. If your search stopped when the parish registers ran out, it probably stopped one document too early.
Send us the surname, the village as your family says it, and one exact date. We will tell you whether a revision series covers that place — and what it is likely to hold.
Write to us