The Soundness Standard: When Family History Research on an Ancestor Is Finished

Family history research is finished on an ancestor when: the facts that establish the person are held at the confidence the evidence supports, the search covered the record groups that exist for that place and time, the reasoning is written down, and the conclusions are honestly defensible. That is the Soundness Standard, and this page is its permanent home.

The field already has a rigorous standard for when a single conclusion is trustworthy — the Genealogical Proof Standard, developed and refined over decades. It answers its question well. Is this specific claim proved?

It was never built to answer a different one. When is my work on this person finished, sound enough to write up, hand down, or give to a community archive?

Why nobody ever told you to stop

Family history research gets treated like a science where new data shatters theories, and "always subject to revision" gets used as permanent permission never to close anything out.

That reading forgets the other half of how science actually works. Scientists publish. They release the current state of knowledge, with the evidence and the reasoning exposed, and let it stand until something better comes along. Publishing is not a betrayal of rigor. It is the point of rigor.

The Soundness Standard puts the publication back. It honors the proof standard completely, and then does the thing that standard was never built to do. It tells you when to stop, certify, and release.

You do not lower the bar. You meet it, and then you clear it.

What "sound" means

Research on an ancestor is sound when all four of these are true.

1. The linkage facts are held at honest confidence. The ancestor's birth, marriage and death, and each child's, because the children's facts are what establish the relationships. Each stated at the level of confidence the evidence actually supports.
2. The search covered the record groups that exist for this ancestor's place and time. Which groups those are is a finite list, not an open-ended one, and the list names the authority it was built from. The results were logged, including the searches that found nothing.
3. The reasoning, the searches, and the negative findings are recorded in a Soundness Record.
4. The conclusions are honestly defensible. Stated confidence matches the evidence, gaps are named rather than hidden, and another researcher could follow the reasoning.

That is the whole standard. Soundness is not certainty, and it is not "I found everything." It is this: here is what the records support, here is what remains open, and I have written it down in a form someone else can check.

New records may surface later. When they do, the record gets revised, exactly the way a book issues a new edition. But the work was completed, recorded, and put into living hands. That is the thing the unfinished folder never achieves.

The stopping rule nobody drew

There are thousands of hours of good instruction on how to research well. Search more than one platform. Contact the local society. Pursue every record group. Get the offline file that was never digitized. All of it sound advice, and all of it framed the same way: as more things to keep doing.

Nobody converted the legwork into a place you can stand and say done.

The people who never finish are not all avoiding the work. Many of them simply never had a way to know they had enough. You cannot fault someone for failing to cross a line that was never drawn.

The Soundness Standard draws it. Same legwork the field already teaches, made terminal. A thing you complete, record, and sign.

The Soundness Record

The standard's artifact is the Soundness Record. One per ancestor, filed at the front, so that anyone who opens that folder in fifty years understands everything behind it.

It opens with the facts to be proved: the required linkage facts, kept deliberately separate from your own optional questions. Whether he was at Brandywine, what she did during the strike year, why the family left the county. Those questions enrich the life. They do not establish it, and mixing the two lists is how people end up researching forever.

It confirms the research-completion checklist for this person's place and era. Its body is the research log: what was searched, what was found, what was not found and what that absence means, and the reasoning behind every call. It closes with the verdict:

• What is proved
• What is probable or possible
• What is conflicted — where sources disagree and the disagreement stands
• What is unsupported — where you searched and the record was not there
• What is open — where nobody has looked yet.

Searched-and-empty and not-yet-searched are different findings and the Record keeps them apart, because collapsing them is how a body of work comes to look more finished than it is.

It is not narrative and it is not short. A suspiciously brief Record usually means the reasoning got amputated again — and producing one used to mean weeks of typing, which is most of why so few exist.

The Record is also where your skill finally gets to show, and that turns out to matter for the writing. When the reasoning has a home of its own, it stops leaking into the story. The research stays in the Record and the chronicle stays about the ancestor.

"Honestly defensible" has teeth

That phrase would be a loophole without a hard definition, so here it is. All three must be true.

Stated confidence matches the evidence. Gaps are named, not hidden. The reasoning is auditable, meaning another researcher could take your citations and follow your logic to the same place.

"I declared it done because I was tired" fails that test. "I declared it sound because the reasoning holds and the gaps are honest" passes it.

A chat transcript is not a Soundness Record. A pile of notes is not a Soundness Record. The Record is the reasoning made legible and owned by you. If AI helped, its role belongs in how you describe your method, never in place of your reasoning. You cite the record, not the tool.

Where the standard fits

The Soundness Standard and the Research Quality Check answer different questions, and they work as a pair.

The Research Quality Check is the check you run on one ancestor's research. The Soundness Standard is the definition of finished that makes a clean check mean something. If you want the wider research side, the AI for genealogy guide covers it.

After an ancestor's research is sound, you may elect to write the chronicle: the story a relative will actually read. The STORI Method is how that gets done, and if you want the wider picture of using AI in this work first, start with Is AI Safe to Use for Genealogy Research?

The narrative never substitutes for the record. It sits on it.

The standard is taught and practiced inside the Chronicle Makers community, but you do not need us to apply it. Pick the ancestor you have researched longest. Write down the four conditions, and check your work against them. You will know within an hour whether you have been finished for years.

The Soundness Standard™ — a Chronicle Makers framework · Denyse Allen · chroniclemakers.com/soundness-standard/ · First published August 2026. It builds on the Genealogical Proof Standard and replaces none of it.