Talk:Historic counties of the United Kingdom

historic_county field in Infobox:UK place: Thoughts on how we present the Historic County
Following an RfC, a field for ‘historic_county’ has recently been added to the UKPlace Infobox. This comment discusses my views on how to populate this field, with the aim of achieving a consistent approach within Wikipedia. My comment only relates to Great Britain, there already being a ‘County’ field for Northern Ireland which everyone appears happy with.

The RfC is based on the practice of the Office for National Statistics (ONS) and its inclusion of an 'Historic County' field in its Index of Place Names in Great Britain (IPN). The implementation of the RfC essentially requires the new ‘historic_county’ field to match the policy and practice of ONS in the IPN. Wikipedia, of course, is founded on the principles of presenting information which is verifiable and from a reliable source. It is hard to think of a more reliable source in these matters than the ONS or a more verifiable one than the IPN, which was first published in 1831 and has long been regarded as the standard authority on the names and locations of places throughout Great Britain. It should be noted too that Wikidata contains an ‘historic county’ field which uses exactly the same definition as the ONS.

That said, there are several issues to be considered, specifically in relation to detached parts, counties corporate and historic changes in county areas. These issues can only be dealt with in the context of an understanding of what the historic counties are and the ONS’s treatment of them. The ONS’s view of the historic counties is clearly stated in the introductory guide to the IPN.

The ONS is unequivocable on the point that the historic counties are totally separate entities to those local government entities first created by the Local Government Act 1888 and the Local Government (Scotland) Act 1889 and that it was these latter that were abolished by the Local Government Act 1972 and the Local Government (Scotland) Act 1973, and not the historic counties. The ONS states that while no longer a statistical geography, the historic counties are included in the IPN for users who wish to use them for historic, traditional or cultural purposes. The ONS further specifically recommends the historic counties as a stable, unchanging geography which covers the whole of Great Britain.

Whilst the general point that modern local government is distinct from the historic counties is generally understood, a precise, definable approach to their names and boundaries is needed to present them within an information system. Within the IPN, the ONS has adopted Definition A of the Historic Counties Standard published by the Historic Counties Trust (a registered charity which aims to promote the history, geography and culture of the historic counties).

The Historic Counties Standard provides a comprehensive definition of the names, areas and borders of the historic counties of the UK. The Standard is primarily just a restatement of the position consistently taken since 1888 by the General Register Office (GRO) and its successor the ONS. i.e. that the late 19th century legislation created new administrative areas but left the counties themselves unchanged. The Standard also prefers the phrase ‘historic county’ over the GRO’s ‘ancient or geographical county’, ‘historic county’ having now become the standard expression used. The borders of the historic counties were obtained to high accuracy by the Ordnance Survey and presented on the First Edition 1:2500 and 1:10560 maps.

The Historic Counties Standard also addresses the ambiguities of the detached parts and the counties corporate. It follows the precedent of the GRO and does not consider that county corporate status means that a town has literally been removed from its county. With regard to detached parts, it follows the long precedent (e.g. in Lewis' Topographical Dictionaries throughout the 19th centuries) of treating a detached part as being associated with both its parent and host county. The Historic Counties Standard actually provides two definitions. Definition A considers each detached part to be associated with its host county. This is intended to have a general-purpose applicability and has presumably been adopted by ONS for that reason. Definition B considers each detached part to be associated with its parent county. This is intended to be useful to historical studies.

My views then are that (a) the historic_county field in the UKPlace infobox should follow the approach of the ONS in following Definition A of the Historic Counties Standard; (b) where a place lies within a detached part of another county that should be described in the main text of the article; (c) where a place lies within a county corporate that should be described in the main text of the article. Having said that, I have no great issue with the detached part / county corporate information being presented in the infobox but this would need to be done in an appropriate way and without creating 'bloat' of the infobox.

A further issue concerns how to treat situations where an area of land has been transferred from one county to another at a point in history. County boundaries in England have actually been remarkably stable since the 12th century, many for much longer. Those of Wales have been stable since the Laws in Wales Act 1535. There has been more by way of change in Scotland, although even here the number of changes is not large and were generally many centuries ago. My view is that these sorts of changes should be described in the main text of the article rather than in the infobox, partly to avoid 'bloat' but mainly because it would be hard to set them in context in the infobox.

My suggested approach would see Wikipedia present a clear and accurate representation of the historic counties, dealing consistently with the ambiguities around county corporate status and detached parts, but also relating the historic counties to those subsequent administrative areas which were initially based upon them. Importantly, it will be presenting a definable entity as ‘historic_county’, one based on a reliable and verifiable source. Peterjamesb (talk) 12:01, 1 July 2021 (UTC)
 * This totally aligns with my views on what a historic county is and how it is best presented in the Infobox:UK place. It also seems sensible to me that, although they are also important aspects of a place's history, detached parts, inter-county transfers and the counties corporate are best dealt with in the appropriate section of the main text of a place's article. I do think, though, that further consideration of where a modern place lies in one or more historic counties is worth having, especially because modern urban sprawl has meant that what was, for instance, in the 19th century two or three separate places is now treated as just one place and this can be tricky to determine and then reflect in the Infobox. Perhaps this is one instance in which the modern administrative local government boundaries of a place can help us to inform how we determine the historic county/counties for that place? An example is that modern Redditch lies wholly in the jurisdiction of Worcestershire County Council, but the Warwickshire/Worcestershire border runs through the centre of the modern town. Districts of modern Redditch such as Ipsley and Matchborough that lie in Warwickshire would have (probably) been considered as seperate settlements in the 19th century and before then. —Songofachilles (talk) 16:46, 1 July 2021 (UTC)
 * If we are talking about towns which have expanded across county borders in relatively recent times then I would refer the individual urban areas, such as Ipsley and Matchborough, to the historic county in which they lie. As for the entry for the main town, I would just list the county in which the town centre is located and with which the town has traditionally been associated. There are also towns which have long lain in more than one county (e.g. Lamberhurst). For these, the infobox entry should definitely include both counties (or all three in the case of Mossley!) Peterjamesb (talk) 21:58, 1 July 2021 (UTC)
 * That is certainly the approach I have taken -- suburbs listed in their own county, town centre defines the county for the main article. This is consistent with the way that sprawl into different local government districts are treated too. Owain (talk) 07:42, 2 July 2021 (UTC)
 * I have been adding the historic_county field to articles in line with the ONS Index of Place Names, which is the verifiable source. I agree with you about the counties corporate -- they are not in the IPN (e.g. Bristol is split between Gloucestershire and Somerset) so we should not ascribe places to them. Despite this, I have had a user revert one of my additions to a Bristol article, claiming "not since 1373". I didn't want to get into an edit war, but it is good to clarify that the source data backs up my position. I think in a few cases it may be desirable to list the detached parts too -- e.g. in the case of Dudley, we could have "Historic county: Worcestershire (detached in Staffordshire)" or words to that effect. As you mentioned, this is entirely compatible with the Historic Counties Standard on which the IPN is based. I have also had an editor request that they would like the Ridings of Yorkshire added too. Although the IPN does not subdivide Yorkshire, I think it is entirely possible to have "Historic county: Yorkshire (West Riding)" or the like, which (as we are not actually changing the county) would still be acceptable. Thoughts? Owain (talk) 19:49, 1 July 2021 (UTC)
 * Many Bristolians certainly take the county corporate status to mean Bristol is a county on a par with Gloucestershire and Somerset. I don't think that's the case with others of the counties corporate. No-one thinks Haverfordwest is not in Pembrokeshire. Don't see it worth getting in an edit war over this issue. The Gloucestershire/Somerset split could be dealt with in the main article text. I'd be in favour of only referring to the largest of the detached parts in the infobox. Most should be discussed in the article text. Can't see a good reason not to include the riding in Yorkshire entries. Although the ridings are not counties, it is common in works of reference to list the riding name in brackets after 'Yorkshire'. Peterjamesb (talk) 21:52, 1 July 2021 (UTC)
 * It is a fallacy to presume the historic county stands for just county cricket - I have read this as an argument against listing the historic county on Wikipedia. The historic county works in a number of areas such as family history and genealogy. In grass roots sport the historic county is the county primarily used for its local governance which is based on historic counties. From Wildlife Trusts to campanology are rooted in their historic county, this is certainly the case in Middlesex where all types of organisations are alive and kicking. So much of everyday life continues to be rooted in historic counties and not the administrative or ceremonial counties. Wikipedia is supplying the information and knowledge required by millions of people who need the historic county as a source to pursue an interest, hobby, local history and so much more. The Office for National Statistics deem it essential to use the historic county so who are we to argue with that. Not to publish the historic county leaves a big gap in the information of the enquirer who rely on Wikipedia to supply what they need to know. MiddlesexHeritage (talk) 10:47, 2 July 2021 (UTC)


 * This discussion will need to be widely advertised. I got here by accident, for example.
 * My view is that relying on the ONS as the only source to describe historic counties is a major failure. It's clearly not correct in some cases and difficult to use in others. If you're going to add it, add it properly - including exclaves, enclaves, cross-border sprawl, county corporates and everything else. Otherwise what you're doing is deliberately choosing to mislead.
 * And, again in my view, if it's complex then it must be referred to properly in the article body (not just the lead) with multiple sourcing before we even think about sticking it in the infobox. And we don't need it in every infobox - some places used specifically slimmed down infoboxes deliberately. Blue Square Thing (talk) 12:49, 2 July 2021 (UTC)
 * The Index of Place Names is based on Definition A of the Historic Counties Standard. As I have previously suggested I would be more than happy to use definition B data to add the "X detached in Y" that you suggest. However, the county corporates are not defined by any dataset, so would be impossible to use. Yes, they should be mentioned in the main body of an article, but data cannot be verifiably sourced, and they are deliberately excluded from the ONS data for this reason. Incidentally, in which cases is it not correct? Feel free to download the entire dataset from https://geoportal.statistics.gov.uk/datasets/index-of-place-names-in-great-britain-july-2016 and see for yourself. Owain (talk) 12:58, 2 July 2021 (UTC)
 * "The county corporates are not defined by any dataset" so what? They are defunct historical units properly described by reliable sources. We don't need a single dataset of today's geography to know Haverfordwest had county corporate status. WP can describe that just fine and should do so in the relevant articles as they are important to the history of those areas. The same is true with any other changes in county borders, we don't need a current database saying "X is in Y". We need a reliable source saying "before 1752 X was in Y". It doesn't make much difference if we want to explain a location was in Britannia Prima or if it was in Avon. Same rules apply, if the fact is pertinent it should be fully described in the body as important to the place's history.
 * The difference with the historic counties is the desire to present them as a present-day unit, which requires consistent modern boundaries. The historic county standard describes a methodology to produce consistent modern boundaries from the historical mapping, and to do that it is necessary to make decisions about anomalies (eg river changes).
 * The Standard also says the following "There were few reliably documented changes to the borders of the historic counties during the centuries before [end of 19th C], any such changes were not such as to undermine the continuity of the counties as geographical and cultural entities". This acknowledges changes took place, but explains these changes do not affect the character of the counties.
 * So this means that the standard ignores minor, but documented, historical changes, and includes minor, but invented, modern changes by the Historic County Trust itself. That is enough to question if it is a reliable source, or if a degree of caution is needed.
 * If you want an example of an anomaly consider: Dalwood and Bridgerule. Both were transferred to Devon in 1844, but IPN shows Dalwood as Devon and Bridgerule as Cornwall. Both places were affected by the same act in the same way, so why shouldn't they be treated identically?--51.7.92.61 (talk) 19:22, 2 July 2021 (UTC)
 * The Index of Place Names is based on Definition A of the Historic Counties Standard. - yep - and that's one reason why there's a problem. It's one definition out of several possible ones. And one that isn't something I'd consider scientifically robust enough. Blue Square Thing (talk) 01:35, 3 July 2021 (UTC)
 * Agree totally that any complexity in the historic county of a place must be dealt with fully in the main body of the article. Not convinced one should try to reflect that in the infobox or the lead. Maybe it would be better to just not fill the infobox in such articles. We should bear in mind though that complexities around detached parts, counties corporate etc. do not affect the vast majority of places. My main interest is in Wales in which it is perfectly clear and unambiguous which historic county each place is in. The historic counties are understood by most people and still have huge cultural resonance. For me listing Glamorgan as the historic county of Maesteg is a simple, unambiguous statement of geographical fact, a fact important to understanding what makes Maesteg what it is. Obviously, I don't agree with your view on ONS. If we can't treat the UK Government's statistics agency as a reliable and verifiable source then Wikipedia's whole philosophy and approach is fundamentally flawed. Peterjamesb (talk) 19:39, 2 July 2021 (UTC)
 * I'm not saying ONS isn't a reliable source. But the Historic County Standard was created by a 3rd party organisation with a strong POV, and the ONS has adopted that data. That data therefore needs consideration before using, just as you'd consider if a piece of climate change "science" produced by Trump's EPA is appropriate. The historic boundaries produced by OS differ to those from ONS, due to a different methodology used (OS derived its own borders based on the time of the 1889 acts).--51.7.92.61 (talk) 19:55, 2 July 2021 (UTC)
 * I have little to add here except to say it is an excellent proposal following commendable research by Peterjamesb. There will be the usual complainers with over analysed unsubstantiated opinion but that will be our cross to bear. Roger 8 Roger (talk) 20:13, 2 July 2021 (UTC)
 * Actually, there is one important thing I want to say, not directly related to the proposals above. IMO, the main cause of these HC endless discussions is around whether they currently exist or not. This has a polarising effect that creates contradictions everywhere and feeds countless edit wars. I think there needs to be a proper debate to make it unambiguously clear that HCs are or are not extant entities. I was going to start that debate in the UK geography/counties project talk page, which i think is the best location to get the current guideline amended (the current position is that they do not exist). However, I am very aware of the need to keep all HC discussions in one place so I will not start that new discussion right now. I hope somebody with more technical skill in WP and with more time can arrange this - a debate leading to a consensus that HCs do exist. Once that is done I think a lot of the probablems we face with circular discussions will disappear. Bleating opposers will continue to be heard from the sidelines but that is where they will stay. Roger 8 Roger (talk) 20:34, 2 July 2021 (UTC)
 * I agree with the core of this, the issue is the argument on the current status of historic counties. Previous consensus established taking the view they do not exist, but that consensus may have changed. If they don't, then everything should be reset to before the start of the template's RFC. If they do, details about how to deal with edge cases is needed. The problem with your statement is the last bit, you aren't open to the possibility that your position will fail to get consensus nor to any potential compromise solution. It also veers close to personal attacks on those who disagree with you.
 * The template's RFC closure ought to be reviewed at the AN as "no consensus" would be accurate summary. I believe the correct solution is a properly conducted RFC on the core issue, not this side matter. In order to hold that discussion, the current situation needs to calm down.--51.7.92.61 (talk) 20:53, 2 July 2021 (UTC)
 * Hmm, I am happy to find myself (almost) agreeing with you. My last comment mentioned by you was indeed intended to ruffle a few feathers, as it has done, but I was careful to avoid any personal attacks because I have no personal problem with anyone. I accept though that deep breaths and calming down the tensions is needed by everyone, and that includes being extra careful in how we express our views. I do take slight exception though to the accusation of my supposed closed mind. In fact, from what I have seen over the last few years it is the established 'they don't exist' group that has had a closed mind. Only now that the genie is out of the bottle is that old school group of editors apparently willing finally to sit down around the table and talk. You seem to be saying that a discussion about HCs current position as dead or alive is warrented but that it should take place somewhere specific to that debate and with people having had the opportunity to establish calmed nerves. If you are saying that, then I am with you. BTW, I accept consensus, as I have done for many years. When I have been accused of edits that contradict current HC consensus, what is nearly always overlooked is that my edits in general comply with WP policy, patently obvious commonsence or clearly cited sources, and that all trumps consensus. The problem has not been with my edits but with the consensus. Roger 8 Roger (talk) 23:32, 2 July 2021 (UTC)
 * Both "they exist" and "they don't exist" are simplified points of view - neither is overwhelmingly factually correct. And it's absolutely unnecessary to try to reach a consensus on this imo - as you've said, you disagree with the current consensus. If it changed I'd disagree with the new consensus. Nothing's going to change either of our minds. But it's not unreasonable to acknowledge that historically places were administered by different places than they are today. That's totally reasonable - and adding well sourced history describing that, in all its complexity, is cool. But just relying on one source set at one point in time is poor history and poor sourcing. Blue Square Thing (talk) 01:42, 3 July 2021 (UTC)
 * BSqT, I hear where you are coming from and I think it has merit. If I were to start a RfC to change the current position it would be 'to remove the phrase in the guidelines that says:'we do not take the view that the HCs still exist ...' (I quote from memory that might be slightly flawed). I would not request a new phrase stating that they do exist. Effectively I would sit on the fence and watch as everyone tried to work out how best to incorporate HC into wiki articles once that 'they don't exist' phrase was removed. Having said that, I do believe HCs exist, but not necessarily only in the way many people perceive the meaning of the word 'exist'. That means I might well be dragged into debates and take the side of the 'they do exist' editors. I think the debates that are happening now are doing exactly what I have just said - they are trying to work out how best to incorporate HCs into WP now that the new IBX field has been created. That is not exactly the same as a consensus that the 'they don't exist' guideline is removed but it is close to it. And that then is why I think we should take the further step of now getting those guidelines changed as well because the guidelines are the root cause of the problem. If we do not change the guidelines there will still be a niggling unresolved problem with how we handle HCs. At this point I am slightly wary of your middle ground view because I cannot immediately see how it can be enacted in a way that will please everyone. An example of what I think is a middle ground approach often found in articles is "X is a town in Y county and was/is historically in Z county". IMO this is just meaningless nonsense. The tense of the verb 'be' is the cause of edit wars, so the problem is not solved, and the phrase is ambiguous anyway so it doesn't really solve anything. Roger 8 Roger (talk) 02:59, 3 July 2021 (UTC)
 * I'd prefer "historically it was part of..." - there's nothing contentious about that tense. It was - but in the context of the sentence it's clear and the POV that it still is is still supportable. To say "it is" is, I think (but I may be wrong), much more contentious and is certainly confusing for readers. How can it be in this and that? For readers not versed in the history of counties (i.e. 99.9% of readers) it's confusing. Blue Square Thing (talk) 11:32, 3 July 2021 (UTC)
 * The point about Wales is absolutely solid - and in cases such as this it's easy to do. Sorted. The problem is that the ONS stuff is ambiguous at times at best and misleading at worst - St Olaves, depending on which line of the spreadsheet you look at, was either in Norfolk or Suffolk. Hmmm - that's somewhere I know very, very well indeed and one of those, by whatever definition you want to use, is flat out wrong. Blue Square Thing (talk) 01:35, 3 July 2021 (UTC)
 * Re. St Olaves, I think it's because the River Waveney is the county border. So while the centre of St Olaves is in Suffolk (see the Wikishire map showing the boundary, there might be some parts of the village, such as St Olaves Marina (I've never been to St Olaves and I'm just using Google maps to see what it there, obviously you know it really well) that are west of that boundary hence both counties are listed. We only have Suffolk in the Infobox, though. As you know, the whole area comes under Norfolk County Council for administrative purposes since 1974. I don't see any issue with finessing the ONS data so the Infobox reads something like: "Suffolk (village centre); Norfolk (west of River Waveney)" —Songofachilles (talk) 09:56, 3 July 2021 (UTC)
 * So we can't just use the ONS data - as is being done throughout the UK - we have to interpret it. Based on? I'm only able to come to a conclusion because I drive through it every day; you because you've done some map interpretation. Do either of us actually have any reliable sources we can bring in? What about Haddiscoe Station? Is that part of St Olaves? The problem is the data source is difficult to use and, in my view, that requires too much interpretation to make it reliable. It's too simplistic an approach to take - we should be looking for better sourcing to check (and, btw, the answer is Suffolk). Blue Square Thing (talk) 11:32, 3 July 2021 (UTC)
 * The issue in St Olave's is the built up area. ONS has multiple listings because the built up area is in two districts (and in this case historic counties). Only sensible thing to do with that sort of case is ignore extent of the built-up area in the infobox. Otherwise you introduce a ton of complications to multiple entries, which could include the parish, the district, all sorts of counties, constituencies etc. That's nothing to do with historic counties specifically.--51.7.92.61 (talk) 11:05, 3 July 2021 (UTC)
 * In this case, sure (but see my reply above). In other cases, that might be a terrible thing to do. Blue Square Thing (talk) 11:32, 3 July 2021 (UTC)
 * Forget about historic counties. Built-up areas straddle civil parish, district and unitary authority boundaries all the time, including all the big cities. Near to St Olave, the built up areas of Great Yarmouth, Belton, Thurlton, Lowestoft, Loddon all cross parish boundaries. Given higher weighting to the built-up areas for the infobox would be a much more radical change than adding historic county.--51.7.92.61 (talk) 12:08, 3 July 2021 (UTC)