Fix one governed identity per entity in a curated registry — preferred label, variant set, disambiguating notes, and a persistent identifier — so every catalog entry cites it rather than re-spelling the name, and one edit propagates everywhere.
An authority record is a curated registry entry that establishes the preferred form for naming a single entity — person, body, place, work, or subject — together with its variants, disambiguating context, and cross-references, so every other catalog entry resolves to one canonical identity by citing its identifier. Its distinctive discipline is deferral: descriptive records cite the identifier rather than copy the name, so a rename or merge is effected once and propagates automatically.
How would you explain it like I'm…
The Official Name Card
In a big library, one special card says the one official way to write a person's or place's name, and lists their nicknames too. All the other cards just point to that special card, so if the name ever changes, you fix it in one place.
The Library's Master Name Entry
An Authority Record is a library's official entry for one person, group, place, or subject. It picks one preferred way to write the name and lists the other ways people write it, like nicknames, pen names, or misspellings. It also adds facts like birth and death years, so two people with the same name don't get mixed up. Every book record in the catalog points to this one entry by its number instead of spelling the name again. If the name ever needs to change, the library fixes it once in the Authority Record, and every book that points to it is fixed too.
Canonical Name Registry Entry
An authority record is a curated entry in a library or archive system that fixes the preferred form of a name for a single entity and ties all its variant names to one identity. It includes a persistent identifier, the preferred heading (like 'Twain, Mark, 1835–1910'), a list of variants (pseudonyms, former names, transliterations, misspellings), notes that distinguish it from namesakes, links to related records, and a history of corrections, merges, and splits. The key discipline is deferral: other catalog records don't restate the name, they cite the authority record's identifier. That makes it the single source of truth for naming, so one change there spreads everywhere. Examples include the Library of Congress Name Authority File, the Getty vocabularies, ORCID, and ISNI; VIAF is different in that it clusters many national libraries' records under one shared identifier.
An authority record is a governed registry entry that establishes the preferred form for naming one entity — a person, corporate body, place, work, or controlled subject — along with its variants, disambiguating context, and relations, so that all other catalog records can resolve to a single canonical identity by citing it. Its structure comprises a persistent unique identifier, a preferred label (the authorized heading), a variant-label set (aliases, transliterations, former names, pseudonyms, misspellings), scope notes such as dates, field, and place of activity to separate namesakes, relation fields linking broader, narrower, or associated records, and a change history recording corrections, merges of duplicates, and splits. Its defining feature within bibliographic control is deferral: descriptive records cite the identifier rather than re-specifying the name, so a rename is made once and propagates to everything that cites it. Major implementations include the Library of Congress Name Authority File, the Getty vocabularies (ULAN, AAT, TGN), ORCID, and ISNI, each enforcing one preferred form per entity in its scope; VIAF instead clusters authorized forms contributed by many national libraries under one shared identifier despite their local variation. Content follows standards such as RDA and ISAAR(CPF), and exchange uses MARC authority format or BIBFRAME. Adjacent systems like database keys, canonical URLs, DNS, or CAS numbers share some functions but lack the curated controlled-vocabulary layer.
The authority record lives within bibliographic control and the wider library/archive/museum/scholarly-publishing tradition, where the full role-set assembles in one maintained registry artifact.
Library name and subject cataloging — the Library of Congress Name Authority File governs personal names, bodies, and works.
Federated authority clustering — the Virtual International Authority File clusters national-library records under a shared identifier.
Museum and gallery cataloging — the Getty vocabularies (ULAN, AAT, TGN) govern how a collection refers to its entities.
Researcher and public-identity registration — ORCID for researchers, ISNI for public identities.
Bibliographic metadata exchange — cross-institution sharing via MARC authority and BIBFRAME under RDA and ISAAR(CPF).
Naming the authority record makes legible the difference between a name string and a governed identity, and between holding a name and citing it. Namesake collisions, scattered aliases, renames, and merges stop being unrelated nuisances and become one thing: success or failure of deferral to a single point of truth. The question shifts from "what name do I write?" to "which authorized identity does this resolve to?"
A catalog accumulates combinatorial naming trouble — every entity surfaces under many strings, every string can collide, every mention is a site where trouble recurs. The record collapses that field of decisions to one governed object per entity: disambiguation, variants, preferred form, and change history resolved once. Every descriptive record reduces to a citation; a global rename becomes a single edit.
Holding "one governed identity per entity, cited not copied" as the object licenses diagnostic reads (a conflated heading is due for a split; non-converging variant searches signal a missing record; stale forms after a rename mean entries copied the string), interventionist prediction (one edit reaches exactly the citation set), boundary-drawing (scope-note versus split), and an order-of-resolution constraint (identity settled at the registry before description).
Within bibliographic control the concept transfers as mechanism, its role-set and diagnostics intact across LCNAF, VIAF, the Getty vocabularies, ORCID, and ISNI — co-instances of one governed-identification discipline. Beyond it, the generalizing insight — one canonical identifier per entity with variants resolving to it — is carried by the parent primes identification and representation. DNS canonical names, database surrogate keys, and CAS registry numbers instance those parents but drop the curated-vocabulary-plus-standards layer that makes an authority record its own thing.
An Authority Record is a Representation whose target is one entity and whose governed record medium preserves identity-relevant labels, notes, and relations.
The sanctioned heading string is a strict internal constituent of the full Authority Record, which additionally stores variants, notes, identifiers, and history.
A Persistent Identifier is the record's strict machine-stable constituent, cited by descriptive records so name edits propagate without rewriting citations.
Versioning is an internal constituent of the maintained record: corrections, preferred-form changes, merges, splits, and retirements are logged over time.
Strip bibliographic standards from an Authority Record and one maintained identity with many variant routes remains—the Alias-to-Authority Mapping core.
Hierarchy paths (42) — routes to 12 parentless roots
Authority Record sits in a sparse region of the domain-specific corpus (64th percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.