What to know about people by heritage directory, and why it matters

Pick any heritage claim attached to a well known name and try to find where it came from. Not a page that repeats it: the first page that said it. The search usually ends in one of two places, and neither is a document.

Either the trail runs cold at a short unsourced line in an editable reference, or it runs in a circle: site A cites site B, site B cites site C, and site C's own footnote points back to site A. This page is about how that loop forms, how to break it, and what an entry looks like when somebody has.

What to take away

• Repetition is not corroboration. Fifty sources carrying identical wording are one source and forty nine copies.
• Heritage claims are unusually prone to circular reporting because they are short, interesting, rarely sourced, and almost impossible to disprove.
• Each copy strips a qualifier. A careful attributed sentence becomes a flat assertion in about three hops, and the assertion is what gets indexed.

How the loop closes

The sequence is consistent enough to be described step by step.

1. A line appears somewhere with low editorial cost: an open reference entry, a forum post, a fan page, a hobbyist database. It has no source.
2. An aggregator or a content site copies it, because it fills a field that was empty and nobody else has anything better.
3. The copy is now a published page on a site that looks like a source.
4. Someone editing the original entry adds a citation, and the nearest available citation is the copy.
5. The claim now has a footnote. The footnote points at a page whose only basis was the unfootnoted original.

Nothing dishonest has to happen at any step. Each participant is behaving reasonably given what is in front of them, and the result is a claim with a source that is itself.

Why this subject in particular

Heritage claims have the exact properties that make a line travel.

They are one clause long, so they fit anywhere. They answer a question readers actively wonder about, so they get clicks. They are rarely supported in the original, so there is no document to contradict, and the documents that would settle the question, such as a parish or civil register, are not what anyone consults. They cannot be checked against anything public, because the underlying facts are either private, legal, or a matter for the person to state. And they feel like the kind of thing that must be recorded somewhere, which stops people from asking whether it is.

Machine-readable knowledge bases accelerate all of it. A value imported once is served to many products, and the products display it without the import history. How a cataloging convenience becomes an apparent biographical fact is set out on the country directory page.

Signals that you are looking at a loop

• Identical phrasing across sites that have no other similarity. Copying preserves word order; independent reporting does not.
• A distinctive detail with no anchor, for example a specific region or a specific relative, that never appears with a document, a date or a name of a source.
• Every citation is lateral. Each source cites another site rather than a record, an interview or a published work.
• No earliest version has a source. Sort the candidates by earliest datable publication and look at the oldest. If it has no basis, nothing later added one.
• The subject has never said it. For a claim about a living person, the absence of any statement by them is the loudest signal there is.

Breaking the loop is the same work in reverse: order the sources by date, find the earliest, and see what it rests on. If the answer is nothing, the claim is unsupported no matter how many pages carry it. That is the same discipline described for historical records, applied to modern web sources instead of archives.

A register entry like this one records a rite on a date. It is the kind of document a heritage claim would need and almost never has, which is why a claim of that shape circulates instead of being checked. The habit of tracing a statement back to a document is the same one described in the surname literature, where a name is a lead rather than a finding.

Qualifier decay

Watch what happens to one sentence as it is copied:

Stage	What the sentence says
Original	In an interview published by a named outlet on a date, she described her family as coming from a place
First copy	She has said her family comes from that place
Second copy	Her family comes from that place
Third copy	She is of that origin

The first version is durable. It remains true regardless of what is later established, because it reports a statement rather than a fact about ancestry, and it names who said it and when. By the fourth version, the attribution, the date and the person's own voice have all been removed, and what remains is a claim the publisher cannot support and the subject may reject.

This decay is not caused by bad faith. It is caused by editors tightening prose. Which is why the qualifiers have to be structural, sitting in fields, rather than stylistic and available to be trimmed.

What a responsible entry holds

• The statement in its original form, quoted, not paraphrased into an assertion.
• Who said it, in which outlet or publication, on which date.
• The relation being described: descent, ethnicity, nationality, religion, or culture, named explicitly rather than bundled as heritage.
• The date you retrieved it, since online sources change and disappear.
• The disagreement, where sources differ, kept rather than resolved into the tidier version.
• An empty field where there is nothing, marked as unresolved rather than filled.

An entry built like this is shorter and less satisfying than the one it replaces. It is also the only kind that survives the subject correcting the record.

Corrections do not travel

The last part of the loop is the part nobody plans for. When an error is fixed at the origin, the copies do not update. They keep serving the old claim, they keep ranking, and they keep being cited, including by the corrected original if a future editor goes looking for a footnote.

So the fix has to be upstream of publication. Do not publish an unsourced heritage claim in the first place, because you will not be able to retrieve it. Where a claim has to be corrected, keep the earlier value and the correction date visible rather than silently overwriting, so that anyone comparing your page to a stale copy can see which is which.

Who carries the cost

Not the publisher. A false statement about a person's ancestry, religion or origins attaches to them, follows them through search results, and is used against them by people looking for a reason. Claims of this kind are also made deliberately, by others, for political purposes, and repeating one without a source launders it.

That is why this site asserts no living person's ancestry, ethnicity, religion, nationality or birthplace, quotes and dates self-description where it is relevant, and publishes no ranked tally of people by origin. The full reasoning is on the heritage pillar, and the category machinery those tallies depend on is on the heritage list page.

Bottom line

Ask one question of any heritage claim: what did the earliest version rest on? If the answer is another page, you have found a loop rather than a fact. Publish the statement with its speaker and its date, or publish nothing, because the correction will never catch up with the copies.

Common questions

A claim appears in a printed book. Does that settle it?

Print raises the cost of publication; it does not add evidence. Check whether the book cites a source and whether that source predates the online versions. Books also copy from the web, and dated editions make the direction of copying easier to establish than on the web.

Is it acceptable to write that a claim is widely reported?

Only when the article's subject is the reporting itself, and only if it says what the reports rest on, which is usually each other. Used as a hedge, the phrase repeats the claim while pretending not to.