Wikidata:Property proposal/historic county

historic county

Originally proposed at Wikidata:Property proposal/Generic

   Done: historic county (P7959) (Talk and documentation)

Description	Historic county or traditional county: the traditional, geographical division of Great Britain and Ireland
Represents	historic county (Q67770276)
Data type	Item
Template parameter	historic county
Domain	historic county of the United Kingdom (Q67376938) county of Ireland (Q179872)
Allowed values	Items which are an instance of (Q21503252) historic county of the United Kingdom (Q67376938) or county of Ireland (Q179872)
Example 1	Dufftown (Q1012028) → Banffshire (Q806432)
Example 2	Luton (Q203889) → Bedfordshire (Q67387552)
Example 3	Pen y Fan (Q1325530) → Brecknockshire (Q547052)
Example 4	Ballymena (Q805451) → County Antrim (Q189592)
Example 5	Carlisle (Q192896) → Cumberland (Q23360)
Example 6	Stanground (Q7598847) → Huntingdonshire (Q67266168)
Example 7	Sunbury-on-Thames (Q2194010) → Middlesex (Q19186)
Source	ONS Index of Place Names in Great Britain’ User Guide page 13 via ONS website ‘Celebrating the historic counties of England’: Ministry of Housing, Communities and Local Government Historic Counties Standard: The Historic Counties Trust
Planned use	Data for towns and villages of the United Kingdom and Ireland
Expected completeness	is complete (Q21874050)
See also	located in the administrative territorial entity (P131): the item is located on the territory of the following administrative entity. Use P276 for specifying locations that are non-administrative places and for items about events. Use P1382 if the item falls only partially into the administrative entity. historical region (P6885): geographic area which at some point in time had a cultural, ethnic, linguistic or political basis, regardless of present-day borders vice-county (P1887): geographical delineation utilized for biological records in the British Isles

Motivation

There are 92 historic counties in the United Kingdom and 26 in the Republic of Ireland, together covering the whole land, so that every geographical feature in Great Britain and Ireland is in a historic county, or on occasion on the border of two of them. Each British historic county has its own data record already.

The historic counties are geographical entities, not administrative areas. Many have administrative areas named after them, but the areas are different and accordingly so is the associated data.

The historic counties are known just as “counties” or “shires” but for distinction from statutorily defined counties they are called “historic counties”, “ancient counties”, “counties proper”, “geographical counties” or “traditional counties”. The Office for National Statistics uses the term “historic county”, as does the Historic Counties Trust. I therefore propose that this be used on Wikidata.

The ONS Index of Place Names in Great Britain’ User Guide clarifies this usage on page 13. This Guide can be accessed on the ONS website

The Historic Counties Trust has published the Historic Counties Standard to clarify any questions about the identification of the historic counties

The historic counties, as noted, are a unit used in the Office for National Statistics database, the Index of Place Names in Great Britain. The Traffic Signs Regulations and General Directions 2016 authorise the erection of boundary signs for historic counties.

More recently, the Ministry for Housing, Communities and Local Government published guidance on community heritage strategy specifically directing local authorities to celebrate the historic counties. The document is named ‘Celebrating the historic counties of England’.

(This Guidance paper is applicable only to England in accordance with the Ministry’s geographical scope.)

The Guidance observes:

“The historic counties are an important element of English traditions which support the identity and cultures of many of our local communities, giving people a sense of belonging, pride and community spirit. They continue to play an important part in the country’s sporting and cultural life as well as providing a reference point for local tourism and heritage. We should all seek to strengthen the role that they can play.” With that in mind, and as authorities will wish to know the appropriate counties for any location subject to the guidance, it will be valuable to be able to include that information as a distinct Statement in articles on towns, villages and other locations. Hogweard (talk) 22:45, 15 February 2020 (UTC)

Discussion

• See historical region (P6885). --Yair rand (talk) 17:38, 17 February 2020 (UTC)
• Interesting, but not the same thing: a 'historical region' is a generalised, past area: the counties are a very specific thing, unique, and of current significance ('historic' not 'historical'). The historic counties (or 'traditional counties') are fixed, defined geographical entities. They are a dataset recognised by the Office for National Statistics and utilised in national policies.Hogweard (talk) 07:06, 18 February 2020 (UTC)
• Why not located in the administrative territorial entity (P131) with end time (P582)? ChristianKl ❪✉❫ 14:45, 18 February 2020 (UTC)

  Presumably because this is intended to be used for current objects that didn't have any temporal overlap with the historical counties. --Yair rand (talk) 17:48, 18 February 2020 (UTC)

•  Support If this is for the historic counties as required by the MHCLG Guidance of 16 July 2019 as it appears to be then it will be a useful resource to have. They are treated as current geographical / cultural divisions, the guidance observing "The Act did not specifically abolish historic counties, but they no longer exist for the purposes of the administration of local government". The permitted values will need to be limited to those counties if that is possible, or we could get the wrong links. LG02 (talk) 19:05, 18 February 2020 (UTC)
  • The related property vice-county (P1887) is constrained to instances of vice-county (Q7925010), so this property should be similarly constrained to instances of historic county (Q67770276). Owain (talk) 10:31, 19 February 2020 (UTC)
•  Support This seems like a sensible way of handling the case where there is no administrative territorial entity, e.g. Yorkshire (Q163). Taking Leeds (Q39121) as an example, we have located in the administrative territorial entity (P131) of Leeds (Q774015); which is itself located in the administrative territorial entity (P131) of West Yorkshire (Q23083); which is then located in the administrative territorial entity (P131) of Yorkshire and the Humber (Q48063) but nowhere is the actual entity of Yorkshire (Q163) referenced. As a current non-administrative entity, we need a property that describes this relationship. Owain (talk) 20:28, 18 February 2020 (UTC)
  • As suggested, I have added 'permitted values' (I hope this is the right format) and conformed the examples to the preferred style. Hogweard (talk) 13:17, 19 February 2020 (UTC)
    • There is no need to list them all out individually, as the property will have a value type constraint limiting usage to only items which are an instance of historic county (Q67770276). I also deleted the last three you added, which were counties of Jamaica! :) Owain (talk) 14:10, 19 February 2020 (UTC)
•  Support The Office for National Statistics in its Index of Place Names relates each of over 55,000 places in Great Britain to its historic county, as well as to numerous types of administrative area. It says that "While no longer a statistical geography, the historic counties are now included in the IPN for those users who wish to use them for historic, traditional or cultural purposes. They are recommended as a stable, unchanging geography which covers the whole of Great Britain." It is hard, therefore, to argue that they don't form a relevant geography in some contexts, especially, as ONS says, for historic, traditional or cultural purposes. They are also used more widely in reference works. For example, each place description in Britannica is related to its historic county as well as to administrative areas. The Royal Mail permits the inclusion of the historic county name in any UK address. Wikipedia recommends that a reference should be made to the historic county in the lead section of a place name article (in England and Wales) where the historic county has a different name to its current local government area. The ONS has adopted the Historic Counties Trust's Historic Counties Standard. It seems sensible to have a property for this type of entity.Peterjamesb (talk)
•  Comment these seem to be similar to Scottish civil parish (Q5124673) in Scotland. Items where it applies use it in P131 with the qualifier object of statement has role (P3831). I'd do the same here. --- Jura 12:40, 21 February 2020 (UTC)

The trouble is that neither civil parishes in Scotland or historic counties in the United Kingdom at large are administrative entities. Returning to my Leeds example, we would then have an almighty mess of overlapping administrative entities which makes no sense logically. The existing vice-county (P1887) property is the ideal analogue here -- no-one is suggesting that vice counties are administrative entities, but they are independent geographical entities in their own namespace with a unique property to reflect that. Owain (talk) 14:06, 21 February 2020 (UTC)

I think it's fairly clear for civil parishes in Scotland in Scotland that they had been and they way they are include makes it clean. --- Jura 14:16, 21 February 2020 (UTC)

Logically, civil parishes should have their own property too as a non-administrative geography, but that is not the subject of this proposal! We should not be overloading an inappropriate property with lots of qualifying statements as it muddies the logical distinctions that they are supposed to create. There is a way to separate out overlapping non-administrative geographies and we should use it, as we have with vice-county (P1887). Owain (talk) 14:31, 21 February 2020 (UTC)

Actually, vice-county (P1887) is a bit of outlier. Civil parishes are prefectly suitable for P131. --- Jura 14:35, 21 February 2020 (UTC)

The vice-counties are an interesting one - they are used by naturalists and fit no other category. They are based of course on the historic counties.

The historic counties are completely non-administrative, so again none of the suggested properties will fit. Although of no administrative significance, their value as 'cultural capital' cannot be overstated, which is the basis of the MHCLG paper I referenced, and a number preceding it. The Ministry organised the county flag display in Westminster last July with that cultural capital value in mind. Most of those counties are over a thousand years old, predating the formation of England itself, and together they form a comprehensive, coherent geography. There is nothing like them.LG02 (talk) 15:15, 21 February 2020 (UTC)

What are they based on? --- Jura 15:24, 21 February 2020 (UTC)

I think you would have to ask a Saxon monarch to get an answer! The counties of England are over a thousand years old and there is no record of their creation - their names just start appearing in records. Those of Lowland Scotland are barely younger, and based on the English system. Counties in Ireland and Wales appeared as English rule spread there.LG02 (talk) 15:46, 21 February 2020 (UTC)

So they were used for government of the kingdom ? --- Jura 16:00, 21 February 2020 (UTC)

I might be able to answer that, as a scholar of the period. Digging back to the Dark Ages has its limitations though. The word 'county' postdates the Norman Conquest but the counties long predate it, and where known as 'scira' or shires. A few of the counties have their origin as kingdoms themselves or as 'underriciu' (underkingdoms). The system of shires as a whole is best dated from the Kingdom of the West Saxons. As far as we can tell, there were military formations based on the shires, and shire courts were established. That said, there were many boroughs created with their own courts and 'liberties' granted to great magnates and monastic houses which had their own jurisdictions, being geographically within the county but not within the power of its courts. This continued throughout the Anglo-Saxon period and accelerated in the Middle Ages, so the counties have never corresponded with the organisation of judicial functions. They were (and are) geographical divisions primarily, on which functions for the administration of justice and the raising of militias were organised, but did not define those functions. The administration of justice and the militia may have been the motive for defining counties, but it was more that those functions were hung upon an area, not the area defined by the function, as they simply did not correspond.

We are in any evet after a thousand years a bit beyond that period of why they might have been devised - they have been a geographical reality for countless generations. Hogweard (talk) 18:23, 21 February 2020 (UTC)

• I think depending on how one interprets that, the "historic counties" could go into located in the administrative territorial entity (P131) or historical region (P6885). --- Jura 08:16, 22 February 2020 (UTC)

I don't think those fit the position. Any 'administrative entity' property is wrong: they are not administrative at all, only geographical. Historical region is problematic, because 'historical' means 'in the past'. The counties are 'historic' in the sense of 'rooted in the past' in the same way that a building may be called a 'historic building', but it is not 'historical'. (We could have 'traditional county' as a name instead - I was just going by the terminology in official statistics.)

The purpose of having the data is to allow applications and outside projects to interrogate the data. The traditional counties are unique and exact: an application looking the find the traditional county for any given town or village will want to get that single answer data, not a collection of unrelated regions and past regions. Hogweard (talk) 14:01, 22 February 2020 (UTC)

I'm sure the ancient history was interesting for some but I am interested in the current use for the historic / traditional counties, and that reality is that they are unique, geographical areas, with no administrative significance but of importance for the cultural capital inherent in them. I appreciate the observation that the point is to be able to get the data. Any town, unless it is on a county boundary, will have one historic county only as they are a defined, coherent topographical system, and so it should spit that single answer out. LG02 (talk) 14:25, 22 February 2020 (UTC)

• The use of located in the administrative territorial entity (P131) or historical region (P6885) allows to do that. Contrary to Wikipedia, Wikidata doesn't limit values to "current" administrative layers. Obviously, users would like to have a separate property for every layer and country or state, but it wouldn't really be practical to do that. --- Jura 14:36, 22 February 2020 (UTC)
• If the traditional counties were administrative or were historical, you might have a solution, but they are neither. Hogweard (talk) 16:04, 22 February 2020 (UTC)
•  Oppose I think we should have a generic approach for this usecase. If it isn't located in the administrative territorial entity (P131) or historical region (P6885), one needs to be available. --- Jura 12:17, 23 February 2020 (UTC)

Virtually the exact same discussion was had about the the vice-county property at Wikidata:Property_proposal/vice-county. Everything said about that property applies here. To wit: "vice-counties are not part of any hierarchy of administrative areas", "they don't represent any organisational unit at all", "they exist as entirely independent concepts that have exactly no relation to each other", "the boundaries of ceremonial and administrative counties can and do change, but vice-counties are static by design". Vice-counties are, for the most part, subdivisions of the historic counties, so to repeat myself, everything that was said about the vice-county property equally applies here. Owain (talk) 12:37, 23 February 2020 (UTC)

We do have some pretty niche Properties already, such as GENUKI ID (P7352), and, as Owain points out, vice-county (P1887). Those are useful properties from a data-management perspective but still of narrow use. To have those and not to be able to have a label for the most fundamental and popular geographical demarcation in the British Isles would seem bizarre. Hogweard (talk) 17:31, 23 February 2020 (UTC)

• external-id properties are different. Eventually we should probably review P1887 as well and try to insert it into a parent property. --- Jura 09:04, 24 February 2020 (UTC)

•  Support exactly per Owain and all the many explanations made about how these are not administrative areas, exactly analagous to vice-county (P1887). I really don't understand Jura's opposition or attempts to shoehorn these defined non-administrative geographical areas into something they are not. I've got no objection to creating a new superproperty of which this and vice-county (P1887) can be instances of, but I'm not aware that we have anything existing - it's certainly nothing in the hierarchy of administrative areas or anything that would require use of an end date - and it could not replace vice-county (P1887) as that's an entirely parallel system. For example Bedminster (Q2660843) is located in the administrative (and ceremonial) county of City of Bristol (Q21693433) (1974 to 1996 Avon (Q929902)), the vice county of North Somerset (Q17581811) and the historic county of Somerset. There are multiple instances of administrative, ceremonial, vice and historic counties that all have the same name but cover different geographical areas. British human geography is complicated and confusing, but that is no excuse for not getting it right. Thryduulf (talk) 10:51, 24 February 2020 (UTC)
• @Hogweard, Yair rand, ChristianKl, LG02, Jura1, Owain:  Done --Fralambert (talk) 01:22, 11 March 2020 (UTC)