# What is Negative Evidence?

Negative evidence is a conclusion drawn from a record's silence. It is not the same as failing to find something, and knowing the difference is what turns an empty search into a usable result.


Most family trees record what was found. The searches that came back empty leave no trace at all, which is why the same registers get read twice and why a question nobody has investigated looks identical to one that has been settled.

Negative evidence is the methodology for using those empty searches rather than discarding them.

## The silence has to be meaningful

The word doing the work in the definition is *should*.

A man appears in a county tax list every year from 1840 to 1849, then disappears from it while the list continues to be kept and continues to name his neighbours. He is not in the local burial register either. The list should contain him and does not. That absence is an argument: he probably left around 1850, and you now have both a date range and a direction to search.

Contrast that with searching a parish register whose volumes for the years you need were destroyed in a fire. You also found nothing. But the silence is a fact about the archive, not about your ancestor, and no conclusion follows from it.

Both searches produced nothing. Only one produced evidence.

## Negative evidence versus negative findings

These two are constantly used interchangeably, and the distinction is worth keeping.

| | What it is | What it gives you |
| --- | --- | --- |
| Negative finding | A search that returned no result | A research fact to log |
| Negative evidence | An inference drawn from a meaningful silence | An argument to test |

A negative finding is always worth writing down. It becomes negative evidence only when you can say why the record should have been there.

## What it is used for

**Establishing a date range.** Someone present in one source and absent from the next gives you a window for whatever happened between them.

**Eliminating a place.** If a complete death register for a town shows no entry across the years someone could have died there, they died somewhere else. Eliminating places is how people get found.

**Resolving a conflict.** When two records disagree, the absence of a third can tip the argument.

**Establishing a non-event.** Sometimes the conclusion is simply that the thing never happened: a woman who never married, a couple who never divorced. That is a fact about a life, not a limit on a search, and it is recorded differently.

## How to record it

The unit is not "no marriage found". It is a short note containing what you looked for, the specific source, the span you covered, how you searched it (indexed or page by page, which spelling variants), and anything that limited the search.

"Searched for a marriage, nothing found" is a feeling. "Read the Kalisz parish marriage register page by page, 1889 to 1900, checking four spellings of the surname; the 1893 volume is water damaged and roughly forty entries are unreadable" is a finding, and it tells you exactly what to do next.

Historically this lived in a research log kept apart from the tree, which is why it so often went missing. GEDCOM 7 now carries it in the file itself, so the searches you recorded travel with your data instead of dying in whichever program you happened to use.

In KleioBase, negative evidence is recorded on the person it concerns rather than in a separate log. Dated entries appear on that person's timeline in the chronological place that explains the gap, struck through so they cannot be mistaken for something that happened, and they stop the app suggesting you look for a record you have already searched for.

Canonical: https://kleiobase.com/what-is/negative-evidence