People by heritage facts with sources and useful context

Every heritage claim rests on a record of some kind, and every kind of record was created by an institution for a purpose that had nothing to do with describing a people. Knowing what each one was for tells you exactly how far it reaches.

Twenty one points, grouped by source type. The five different variables bundled inside the word heritage are set out on the heritage pillar; this page assumes that distinction and works through the evidence behind each.

What to take away

Every record answers the question its creator asked. A register records a rite; a form records a clerk's categories; a roll records eligibility for something.
Absence in a record series is often a product of the system that produced the records, and it is never in itself a statement about a family's past.
Membership in a community is set by that community. No document, and no test, confers it.

Institutional registers

A parish or congregational register records a rite performed on a date. Baptism, marriage and burial were events in an institution's life, and national archive research guides set out which registers survive for which periods. The register is strong evidence that the rite happened and weak evidence about belief.
The institution had its own membership rules, which may have admitted people from several communities and excluded others. Appearing in one congregation's book does not place a person in one community, and it does not exclude them from another.
Registers were kept unevenly and survive unevenly. Wars, fires, closures and mergers took out whole runs. Denominational archives hold some and not others, and access is at the holder's discretion.
A school, employer or institutional record had its own boxes, with definitions set by that body for its own administration. Those categories are evidence of how the institution classified people, which is a real historical fact about the institution and not a description of the person.

Migration and status records

Arrival and passenger records capture what a clerk wrote, from what was said, in the clerk's spelling, into the fields the form provided that year. The form changed repeatedly; check which version you are reading before assuming what a column means.
The categories printed on migration forms were set by the receiving state, and they changed as its policy changed. They are administrative classes, not descriptions of peoples, and reading them as ethnicity imports a government's filing system into a biography.
A naturalization record names a former sovereign or state. That is a legal status at a moment, and the country facts page works through why it says nothing about a family's origins.
Where enslavement broke the documentary chain, the gap is a product of the system that produced the records. Absence of a name in a series is evidence about how the records were kept, and it is never evidence about a family's past.
Displacement, forced relocation and border changes separated people from the offices holding their records. A missing document frequently means the office is now in another jurisdiction, not that the event never happened.

Testimony

Oral history is a real source with its own standards. Recorded, dated, attributed to a named speaker, with the questions preserved, it is evidence of what that person said and knew.
A family story is testimony about what a family believed. That belief is itself a fact worth recording, and it is a research lead about the past rather than a finding. Record it as an account, with whose account it is.
The person's own description of their background is the primary source for the social parts of the bundle. Ethnicity, religion and culture are constituted by identification and recognition; no external document outranks the subject on them.

Genetic estimates

Populations that married within themselves over generations share more DNA than the documentary relationships imply. Where that history exists, shared-segment estimates read as closer kinship than the paper trail supports, and match lists get long.
A match list is a set of probabilistic relationships, not a pedigree. It suggests where to search. Converting it into a lineage requires documents, and the estimate never becomes one, a limit spelled out in public health guidance on direct-to-consumer genetic testing.

Membership, belief and language

Indigenous nations determine their own citizenship, under their own law. An individual's claim of descent is not membership, and no external body, dataset or test can supply it.
Religious communities define membership by their own law and practice. A religion recorded at a birth is an institutional entry; practice is a separate question; and conversion happens in both directions throughout a life.
Language spoken is not descent. Languages are learned, lost across generations, and shared between unrelated communities. A language on a record tells you what was spoken in a household, which is useful and is not ancestry.
Appearance is not evidence, in any direction, at any strength. Photographs support no conclusion about descent, and reasoning from one is how a great deal of harm gets published.

Families and records

Adoption, fostering, step-relationships and informal kinship produce families whose documentary and biological lines differ. Both are real families. A record series that tracks one is not describing the other, and neither is a correction of the other.
Where sources disagree, record the disagreement. Resolving it by choosing the tidier version discards the most informative thing in the file, and the two accounts often turn out to be answering different questions.
No living person's ancestry, ethnicity, religion, nationality or birthplace is asserted here. Where someone has publicly described their own background and it bears on the subject, we quote it, name the outlet and date it. Otherwise the field stays empty. It is not filled by a name, a face, a place, a relative or a percentage, and no ranked tally of people by origin is published at any level, since such a count measures which archives were indexed rather than the people in them.

What an entry states, and where it stops

The relation, named: descent, ethnicity, nationality, religion, or culture.
The evidence type, and the specific record or statement it comes from.
The date the record was made and the date it was consulted.
The wording of the source, kept rather than paraphrased into an assertion.
Unresolved marked unresolved.

Written this way, an entry can be checked by a stranger and cannot be quietly upgraded by an editor tightening a sentence, which is the failure described on the heritage directory page.

Bottom line

Match the claim to the record that can carry it, and stop where that record stops. A rite, a form, a roll and a test each answer a narrow question, and the questions people most want answered about heritage are answered by the person, or not at all.

Common questions

A documented ancestor is established. What does that support?

That link, and nothing beyond it. A traced line is a fact about the people on it, not a summary of everyone a person descends from, and the arithmetic of ancestral lines is set out on the heritage list page and the pillar.

How should a record's own category wording be handled?

Quote it, attribute it to the body that used it, and date it. Historical classifications carry meanings that have shifted, and translating one into a modern term silently changes what the record said.