Definitions

Popularly a castle

What an ordinary person would call a castle and travel to see. No scholarly standing whatsoever, and by far the most useful band for anyone actually planning a visit.

Criterion

A general audience, shown the structure without context, would call it a castle — evidenced by how the structure is named and described by its own operator, its national tourism body, and general reference works. This is a claim about usage, and it is measured, not adjudicated.

Includes

• Anything marketed, signposted or commonly referred to as a castle
• Structures failing every scholarly band but universally called castles — Neuschwanstein is the type case

Excludes

• Structures called castles only by this dataset

Closure: open

Not closable, ever, and this is the band's point. Popular usage has no register and no boundary, so any single global integer for it — 2,435, for instance — is a selection presented as a census. This band exists to hold that number and label it correctly.

This band will never close. No register bounds it and none could, so it publishes a distribution or a range with its assumptions visible — never a fabricated total. A coverage cell that tried to give it a world population figure would fail the build.

Who defines it this way

• canonical-castles — Usage band; verdicts cite the naming sources rather than a definitional authority (source) — this project's own band; no external authority

Bounded by

• Wikidata — Wikimedia Foundation · 36716 matching entries
• OpenStreetMap — OpenStreetMap Foundation · 28544 matching entries
• The Castle Map — thecastlemap.com · 2435 matching entries

In this dataset

Derived from the records, not typed. See counts and coverage.

Note

This is the band thecastlemap.com actually built while presenting it as the general one. Recording it as a band rather than dismissing it is the point: popular usage is real, widely held, and the reason most people arrive at a castle dataset in the first place. It is simply not a census.

Assessed records

Structures somebody applied this criterion to, with the reasoning.

• Moritzburg Castle — yes: Marketed and signposted as Schloss Moritzburg and routinely rendered 'Moritzburg Castle' in English-language tourism copy.
• Neuschwanstein Castle — yes: Very probably the most photographed 'castle' on Earth, universally signposted, marketed and described as one by its own operator and by German national tourism.
• Palace of Versailles — contested: English usage overwhelmingly calls it a palace, French usage calls it a chateau, and castle-aggregator datasets file it as a castle - thecastlemap ranks it #1 by fame. Usage genuinely differs by language, which is the whole difficulty.
• Sham Castle — contested: Named, signposted and visited as a castle, but its own name concedes it is not one. Whether a structure that announces its own falsity counts as popularly a castle is genuinely unclear.
• Elmina Castle — yes: Universally called Elmina Castle, a World Heritage site and one of the most visited historic sites in Ghana.
• Dún Aonghasa — contested: A major signposted heritage attraction that many visitors experience as a castle-equivalent, but Irish sources generally call it a fort, and a commenter in the originating thread named it specifically as something he would NOT consider a castle. Usage genuinely splits.
• Trim Castle — yes: A major visitor attraction operated and signposted as a castle by the Office of Public Works.
• Chittorgarh Fort — contested: Called a fort in Indian usage and in UNESCO's inscription, but routinely rendered 'castle' in general-audience English travel writing. Usage differs by language and audience.
• Himeji Castle — yes: Japan's most visited castle, a National Treasure and World Heritage site, universally called a castle in both Japanese and English.
• Crac des Chevaliers — yes: Universally described as a castle in scholarship, tourism and general reference - frequently as the best-preserved crusader castle in the world.

Definition file · dataset licensed CC BY 4.0