Property talk:P27

From Wikidata
Jump to navigation Jump to search

Documentation

country of citizenship
the object is a country that recognizes the subject as its citizen
Representscitizenship (Q42138)
Data typeItem
Template parameteren:Template:Infobox person (nationality) ; see Template:Infobox person (Q6249834).
Domain
According to this template: person, term
According to statements in the property:
human (Q5), character that may be fictional (Q21070598), fictional character (Q95074), fictional human (Q15632617), group of humans (Q16334295) and person (Q215627)
When possible, data should only be stored as statements
Allowed valuesstate (Q7275), mainly sovereign state (Q3624078), no value for statelessness (note: this should be moved to the property statements)
ExampleNelson Mandela (Q8023)South Africa (Q258)
Napoleon (Q517)First French Empire (Q71084)
Mahatma Gandhi (Q1001)British Raj (Q129286)
Sergio Mattarella (Q3956186)Italy (Q38)
Format and edit filter validationAbuse filter #92
Robot and gadget jobsDeltaBot does the following jobs: See Maintenance requests below
Tracking: sameno label (Q42533253)
Tracking: usageCategory:Pages using Wikidata property P27 (Q23908972)
Tracking: local yes, WD noCategory:Articles with Template:Bio and country of citizenship (Italy) not in Wikidata, but available on Wikipedia (Q24047817), Category:Articles with Template:Bio and country of citizenship (United States) not in Wikidata, but available on Wikipedia (Q24047805), Category:Articles with Template:Bio and country of citizenship not in Wikidata, but available on Wikipedia (Q20115759), Category:Country of citizenship not in Wikidata, but available on Wikipedia (Q19414348)
See alsocountry of origin (P495), country (P17), country for sport (P1532), permanent resident of (P5389)
Lists
Proposal discussionProperty proposal/Archive/1#P27
Current uses2,864,052
[create] Create a translatable help page (preferably in English) for this property to be included here
Value type “state (Q7275), fictional country (Q1145276), nation (Q6266), dependent territory (Q161243): This property should use items as value that contain property “instance of (P31)”. On these, the value for instance of (P31) should be an item that uses subclass of (P279) with value state (Q7275), fictional country (Q1145276), nation (Q6266), dependent territory (Q161243) (or a subclass thereof). (Help)
Exceptions are possible as rare values may exist.
List of this constraint violations: Database reports/Constraint violations/P27#Value type Q7275, Q1145276, Q6266, Q161243, SPARQL, SPARQL (new)
Type “human (Q5), character that may be fictional (Q21070598), fictional character (Q95074), fictional human (Q15632617), group of humans (Q16334295), person (Q215627): element must contain property “instance of (P31)” with classes “human (Q5), character that may be fictional (Q21070598), fictional character (Q95074), fictional human (Q15632617), group of humans (Q16334295), person (Q215627)” or their subclasses (defined using subclass of (P279)). (Help)
Exceptions are possible as rare values may exist.
List of this constraint violations: Database reports/Constraint violations/P27#Type Q5, Q21070598, Q95074, Q15632617, Q16334295, Q215627, SPARQL, SPARQL (new)
Qualifiers “start time (P580), end time (P582), statement is subject of (P805), sourcing circumstances (P1480), end cause (P1534), has cause (P828), nature of statement (P5102): this property should be used only with the listed qualifiers. (Help)
Exceptions are possible as rare values may exist.
List of this constraint violations: Database reports/Constraint violations/P27#Allowed qualifiers, SPARQL, SPARQL (new)
Conflicts with “sex or gender (P21): male organism (Q44148), female organism (Q43445): this property must not be used with the listed properties and values. (Help)
Exceptions are possible as rare values may exist.
List of this constraint violations: Database reports/Constraint violations/P27#Conflicts with P21, SPARQL, SPARQL (new)
Contemporaries:
if [item A] has this property (country of citizenship (P27)) linked to [item B],
then [item A] and [item B] have to coincide or coexist at some point of history. (Help)
Exceptions are possible as rare values may exist.
List of this constraint violations: Database reports/Constraint violations/P27#Contemporary, SPARQL (new)
Which items to use?
Country Earliest Latest Note
Kingdom of the Netherlands (Q29999) 1815 present
Netherlands (Q55) - - don't use

Constraint One of moved at the bottom of the page - too many items...

Contemporaries:
if [item A] has this property (P27) linked to [item B],
then [item A] and [item B] have to coincide or coexist at some point of history.
Exceptions are possible as rare values may exist.
List of this constraint violations: Database reports/Constraint violations/P27#Contemporary, SPARQL
This property is being used by:

Please notify projects that use this property before big changes (renaming, deletion, merge with another property, etc.)

Discussion[edit]

Maintenance requests[edit]

List here some automated requests that you may find useful for maintaining this property:

Renaming[edit]

In the spirit of having the most unambiguous property label, I'm changing "Nationality" to "country of citizenship". "Nationality" has several possible meaings, and, perhaps more importantly, several possible phrasings - if I were to have an item, would it say my Nationality were "America", "American", or "Americans"? Imagine how much more complicated it would be if I lived elsewhere, or had changed citizenship... this way, if there's a need, we can also have a "country of residence", "country of birth" (though we already have a "place of birth"), "previous countries of residence" - whatever. But "Nationality" is far too vague. — PinkAmpers&(Je vous invite à me parler) 21:20, 4 February 2013 (UTC)

Not sure in look of the description and comment of PinkAmpersand for german language: sich ergebenden Rechte und Pflichten einer natürlichen Person in einem Staat. Is this btw the right position to discuss such heavy weight decisions? Conny (talk) 14:06, 5 February 2013 (UTC).

The value for this property should be "United States of America" or "American"? --β16 - (talk) 22:05, 8 February 2013 (UTC)

United States of America. --09:29, 6 March 2013 (UTC)

"Sovereign State"[edit]

As was pointed out at Wikidata:Property_proposal#Unrecognized_country_of_citizenship, the description specifies that the object should be a "sovereign state", but that may be inaccurate. Hong Kong is a Chinese dependency, yet an Hongkongese passport is really not the same thing as a maintstream Chinese passport, so that it makes sense to have "Hong Kong" as a country of citizenship. Should I change "sovereign state" to the more vague "country" ? --Zolo (talk) 09:29, 6 March 2013 (UTC)

old times[edit]

"the object is a sovereign state that recognizes the subject as its citizen". That may seem fine for today, but it is not so suitable for other periods. For instance en:Dante's infobox states that Dante is an Italian. There was no sovereign Italy at the time, and actually, I am not sure that "recognizing the subject as its citizen" is something that makes much sense in the 13th century. At the same time, it sounds sort of obvious to most people that Dante is Italian. How should we handle such cases ? --Zolo (talk) 19:09, 11 March 2013 (UTC)

I don't know, but this is a problem even in fairly recent times. For instance, I just removed the statement that William Dawson (1852-1923) was a New Zealand citizen, because New Zealand citizenship did not exist before 1949. Presumably he was a British subject, although I'm not sure if that's the same thing as having British citizenship. Many people from elsewhere in the British Empire would have a similar status. Even Canadian citizenship (the earliest in the British Commonwealth, besides Irish) was only created in 1946. --Avenue (talk) 15:16, 25 March 2013 (UTC)
Are there any news for old time cases ([1])? In dewp it is solved via category Historic Person (Italy). Thank you, Conny (talk) 13:23, 4 March 2015 (UTC).
I'd also like this resolved. For example, there are many Britons (the Celtic group of tribes described by Celtic Britons (Q849967), not the modern group described by British people (Q842438)) on Wikidata who existed before the concept of nation-state existed on the island of Great Britain (that came about in the time of the Anglo-Saxons). Many currently point to the United Kingdom as "country of citizenship" because Wikipedia says they're "British", but the UK did not exist yet. Is it okay to use ethnic group (P172) in these cases, and point to their tribe or tribal group, and leave country of citizenship (P27) empty? I think that there should be a property which defines a "place a person is from" (how we use demonyms in English) without needing to bother with concepts like citizenship, which didn't exist for many people. People can be "from a place" without being recognised as a citizen. People can also obtain citizenship of a place without being a 'person from' (even in their own opinion) of that place. Maybe residence (P551) is a good way to do that? --BurritoBazooka (talk) 01:03, 16 September 2015 (UTC)

Bottask proposal[edit]

Should my bot check links to this items on a regular basis and fix wrong entries if this could be done automatically? Look at the list Wikidata:Database_reports/Constraint_violations/P27#Values statistics there are some common false entries like Q29520 which should be: Q148. Or e.g. Q1110 which should be deleted if I am correct. --Sk!d (talk) 14:58, 9 May 2013 (UTC)

I asked User_talk:Kizar#P107:Q4167410 to fix Q29520. I don't think he minds if you do it. There is also Wikidata:Bot_requests#Q21_on_the_P17_report. --  Docu  at 15:04, 18 May 2013 (UTC)
Note: not all China (Q29520) should be changed to People's Republic of China (Q148), some should use Taiwan (Q865) instead. --Stevenliuyi (talk) 16:37, 26 May 2013 (UTC)
What about older periods ? Should we relabel Q9008156 from "Imperial China" to "Empire of China" ? --Zolo (talk) 16:49, 26 May 2013 (UTC)
I'd leave it as Imperial China, only because that's the main alias used on Imperial China. But, seems either is acceptable. Danrok (talk) 17:02, 26 May 2013 (UTC)
IMHO Imperial China (Q9008156) is not suitable because: (1) There are many periods in Chinese history (e.g. Five Dynasties and Ten Kingdoms period (Q242115), Sixteen Kingdoms (Q683551)) are not considered as "empire"; (2) There are some tricky cases such as Yuan dynasty (Q7313). Although Yuan is considered as an imperial Chinese dynasty, it was part of Mongol Empire (Q12557). I think we need to find a way to specify the dynasty and/or kingdom a person lived, but I'm not sure what's the best way to do it. --Stevenliuyi (talk) 17:32, 26 May 2013 (UTC)
Another problem is about the definition of "citizenship". It's a well-defined concept for all modern countries, but many historical countries don't have this concept. For instance, can we say that Johann Adam Schall von Bell (Q76732) and Ferdinand Verbiest (Q378277) are citizens of China? --Stevenliuyi (talk) 07:31, 2 June 2013 (UTC)

I will now start to remove United Federation of Planets (Q1110). I will not change China (Q29520) as my bot does not know if People's Republic of China (Q148) or Taiwan (Q865) is correct. Also should i change England (Q21) i don't think this is correct as it is only a part of United Kingdom (Q145) which seems to be correct and has many more entries. --Sk!d (talk) 23:09, 15 June 2013 (UTC)

How to do when a person had several nationalities, successively or at the same time ?[edit]

The most prominent example I could think of, right now, is Albert Einstein ? should he be considered, German, Swiss, apatrid, American ? it seems there is a "unique value" constraint on that property…

For people who became american (a lot of them), but came from another country ? which nationality should be prefered ? the "original", the "present", the "nationality at time of death" ?

Also what about people who have more than one passport (just because laws vary around the world and in some countries, gaining a nationality, through marriage for example, does not make you loose your previous nationality, or for Jewish people, all around the world, that have simultaneously a US/French/GB (or any other country) passport, and an Israeli one ?

In my opinion, this value should not be limited to one, or there would need to be some "dates" values linked… and even in that case, that does not solve the "double-nationality" value ?

What is the consensus, now ? and is there one ? --Hsarrazin (talk) 01:40, 28 May 2013 (UTC)

There is no unique constraint on this property. Have a look at Albert Einstein (Q937), there are already several values for P27. Soon it will be possible to add a qualifier with the period of time the citizenship refers to; the Date-Datetype is supposed to go live by this week. Then it'll be possible to distinguish between "double nationalities" and former nationalities. Does this answer your question?--Kompakt (talk) 06:08, 28 May 2013 (UTC)

medieval and early modern persons[edit]

Recently this property has been added by bots to items on medieval or early modern persons, probably using Wikipedia-categories as sources. This applying of a rather modern concept is a rather bad idea IMO. The concept of nation (Q597897) only begins to emerge in this time, the terms are used quite loosely (cf. e. g. http://bonaelitterae.wordpress.com/2011/02/19/britons-bretons-and-germans/). There are enough problems already concerning the 19th or 20th century (cf. e. g. Wikidata:Forum#Deutschland = Deutschland?). And the relevant Wikipedia-categories are usually ill-defined, heavily contested and often misapplied. --HHill (talk) 17:19, 14 May 2014 (UTC)

You seem to be challenging the very existence of this property. A similar proposal was archived. --Nemo 16:22, 15 May 2014 (UTC)
Thanks for the link! No, challenging the very existence of this property is not my intention, I just want to see it applied correctly. Biographies of persons living in the 19th or 20th century create the bulk of items on humans in Wikidata, to most of these the concept of country of citizenship (P27) can easily be applied (and yes I'd prefer the approach proposed on Project chat). --HHill (talk) 17:16, 15 May 2014 (UTC)


Property only useful for persons from this and last century/propery values too complicated?[edit]

First let me say that this is probably not the first time something like this has been discussed, but I think the problem has not been resolved yet (we talked this about on the German project chat page a few days ago).

One problem with this property is outlined above at #medieval and early modern persons: Some states changed quite often, it's really a mess to get this right for persons from those states. E.g. the concept of a citizenship of the state Germany starts to exist in the year 1913 (maybe also 1871, not sure about that), everything before was either another state or at least did not know the concept "citizenship" from a legal point of view. What could be an alternative is a property "nationality of citizenship", how it was proposed originally (see https://www.wikidata.org/wiki/Wikidata:Property_proposal/Archive/1#P27). Then this property could also be correct for no-longer existing states (or state-like concepts) if we define e.g. nationality=German as nationality of any state that was a German state (now what's a German state? :). To summarize this a bit: At the moment I find this property a bit too generic and useless, maybe as first step the description should really emphasize that this should only be used for persons who really have a citizenship of this modern state (and not any previous state which has been "merged" into that state). And also emphasize to leave this property if one is not sure about that. Another example how complicated this topic can be: On the German Wikipedia the talk page for the category "Germans" is 180 KB in size. Lots of discussion happened on which persons should be in this category and of what categories that one should be a sub-category (e.g. of "persons by citizenship"). The same discussion/problem that was discussed there applies in large parts to Wikidata as well. --Bthfan (talk) 17:10, 24 July 2014 (UTC)

BTW: If we decide to create/use a property "nationality" (German/French/...) we would still need a text to explain who is actually German (or of another nationality), but for that we could reuse discussion results from the various Wikipedias. Right now this property is just too difficult for many states IMHO (due to wars, merger of countries, seperations, etc.), it would be some kind-of nightmare to fix this for all persons. --Bthfan (talk) 17:39, 24 July 2014 (UTC)

Well, not all properties are useful for any Wikipedia or any biographic article.
In the Russian language Wikipedia, infoboxes in biography articles generally illustrate the changes of country of citizenship over the time. Example: here.
To store idiosyncratic definitions of one or the other Wikipedia, categories there can be used or specific new properties could be created. --- Jura 15:07, 1 August 2014 (UTC)

Leave this constraint at the bottom of the page[edit]

This constraint comprises so many items using Template:Q that it exceeds the maximum allowed - causing script errors and not listing labels as expected.

This constraint was still active (see constraint report), but invisible on this page because of the comment tags and hard to find when editing the page because it was buried amongst the discussion. I've moved this back next to the other constraints and replaced Template:Q with plain links to solve the problem of Template:Q being called too many times. If we really don't want it, we should just remove it. - Nikki (talk) 12:29, 5 November 2016 (UTC)

Use of this property in the scope of history[edit]

Pandora's Box

I hope I can approach this topic without opening a Pandora's box (Q937164) (oder auf Deutsch: ohne ein Fass aufzumachen), but I keep thinking about the use of this property in the scope of history. I'm sure that many Wikidatans are aware of the problems connected with the use of this property for historic persons. In short, many persons are said to have been citizens of states that didn't exist at the time. Example: Giovanni Boccaccio (Q1402) lived in the 14th century, five centuries before the state of Italy (Q38) existed. I'm no expert on Italian history, so I don't know how to fix this. all I know is that citizenship is a different thing than ethnicity (for that, we have ethnic group (P172)).

One complicated example I do know a bit about is Germany. In the 19th century, people living in the German Reich were citizens of their respective territories, e.g. the Kingdoms of Saxony, Bavarias, Württemberg, Prussia, Hannover, not to speak of all the small duchies. When we go back further, it gets more particular and way more complicated. For simplicity's sake, let's look forward. On July 22, 1913, citizens of states within the German Reich were additionally declared citizens of the Reich itself. This meant that if you were a citizen of Prussia at the time, you automatically became 'Reichsangehöriger'. This 'double citizenship' (though a legal scholar might disagree with the term) remained until Februar 5, 1934, when Nazi legislation dissolved former citizenships of particular states. Also, after the annexation of Austria in 1938, on Juli 3 all Austrian citizens were declared German citizens (deutsche Staatsbürger).

After World War II, it gets more complicated. After the foundation of the Federal Republic of Germany and the German Democratic Republic, the former saw all citizens living in the GDR as German citizens, while on February 20, 1967 the GDR declared all of its citizens 'citizens of the GDR'. So when I was born, I was considered a citizen of the GDR by the GDR and a German citizen by the FRG. The confusion ended on October 3, 1990, when the FRG incorporated the GDR into itself.

Other legal concerns aside, the citizenship of persons on Wikidata should be reviewed according to the time they lived. When can, I think, argue, that every person considered a 'Reichsangehöriger' since 1913 can be called a "German citizen" and rendered as such. For people living in the GDR, we could try looking for the places they lived (or died) in the time between 1967 and 1990 to determine if they were likely to have been GDR citizens.

What we should do with the citizens of particular German states before 1934 (especially before 1913), is another matter.

And then there are innumerable other cases in other parts of the world, but I'll leave that to people who are familiar with them. Jonathan Groß (talk) 09:47, 18 July 2015 (UTC)

I added an illustration to your post ;)
I had been wondering if we should try to develop constraints that check dates of birth/death/floruit against values used in P27. --- Jura 13:27, 19 July 2015 (UTC)
Another point is that some of labels in use for some of the former countries might not be suitable in conjunction with this property. "citizen of Weimar Republic" sounds like a city-state. Maybe it would need a dedicated item with a label different from the one for the period in history. --- Jura 12:15, 25 August 2015 (UTC)


Naturalization[edit]

@Jonathan Groß, Jura1: : good points ; somebody raise a similar question today on the french chat (Wikidata:Bistro) about a painter from Netherlands and who got naturalised in France in 1547 (Corneille de Lyon (Q720941)). I suggested to put the claim like this :
< Corneille de Lyon (Q720941) View with Reasonator View with SQID > country of citizenship (P27) View with SQID < Kingdom of France (Q70972) View with Reasonator View with SQID >
statement is subject of (P805) View with SQID < Letters patent of naturalisation (Q3236974) View with Reasonator View with SQID >
start time (P580) View with SQID < "1547" >
Does it seem good to you? How can it be improved? (maybe with more precise or other qualifiers).
Cdlt, VIGNERON (talk) 15:22, 13 September 2015 (UTC)
"start time" seems fine. "Subject of" would generally lead to an item about the nationality of this person. Maybe there is a better one than "Determination method"? --- Jura 15:35, 13 September 2015 (UTC)

subproperty of country?[edit]

The country (P17) property means "located in country", not "is from country". In fact country (P17) is a subproperty of location (P276). Therefore, this property should not be a subproperty of country. (If someone is a citizen of a country, it doesn't mean he/she is located in that country.) I propose to remove subproperty of country (or change to another). Jefft0 (talk) 17:32, 13 November 2015 (UTC)

@Jefft0: this property was add by Filceolaire (Special:Diff/223221491). I don't think that country (P17) really means « located in country ». The english description of P17 is « sovereign state of this item » wich seems rather consistent with description of P27 : « the object is a country that recognizes the subject as its citizen ». I think we should keep it. Cdlt, VIGNERON (talk) 14:53, 19 February 2016 (UTC)

statelessness (Q223050)[edit]

It seems we have around 1000 items with a country of citizenship (P27) statelessness (Q223050) (query). That's not a valid country and the intended meaning seems to be the same as country of citizenship (P27) novalue. Is there any reason why they shouldn't be changed to country of citizenship (P27) novalue? - Nikki (talk) 14:55, 5 November 2016 (UTC)

  • You can't add statements and sitelinks on "novalue". Are there cases where the current solution is problematic?
    --- Jura 15:03, 5 November 2016 (UTC)
    I'm not asking whether Q233050 is correct, because it isn't (nobody is a citizen of "statelessness", even if humans can figure out the intended meaning). It is not the "current solution" either, there are plenty of items correctly using novalue. I only wanted to know whether novalue could mean anything other than statelessness.
    Q233050 will continue to exist either way, so you can continue to add statements and sitelinks to it either way. If the link between novalue and Q233050 is so important, that does not mean we should abuse the data model and use Q233050 instead of novalue, it means we should find a proper generic way to express what novalue means which can be used for any property. For example, a new property "novalue defined as" to be used on properties would allow us to create a statement like P27 "novalue defined as" Q233050.
    - Nikki (talk) 18:19, 6 November 2016 (UTC)

Former country[edit]

When a person was a citizen of a former country, it is still valid.. only current countries are an invalid restriction. Thanks, GerardM (talk) 15:12, 29 April 2017 (UTC)

Agree - this doesn't seem to be a very helpful restriction at all. Andrew Gray (talk) 18:16, 17 May 2017 (UTC)
Symbol support vote.svg Support Jefferson Davis (Q162269) certainly had country of citizenship (P27) Confederate States of America (Q81931). Let's get rid of the one-of constraint. Runner1928 (talk) 22:07, 22 June 2017 (UTC)
currently the one-off constraint seems to need to explicitly enumerate every country or former country. Is there a way we can adjust the constraint to be an instance of any subclass of country (Q6256) (which would then also include historical country (Q3024240)) --Oravrattas (talk) 07:38, 5 September 2017 (UTC)

Wikidata:Property proposal/Nationality[edit]

There is the ongoing Wikidata:Property proposal/Nationality perhaps tackling some of the problems above. --Marsupium (talk) 19:11, 29 August 2017 (UTC)

Permit "proposed country"[edit]

The constraint with this property currently does not permit instances of proposed country (Q28864179). It should because individuals should be able to identify with the citizenship of their choice even in the absence of external recognition. Blue Rasberry (talk) 20:55, 11 May 2018 (UTC)

Multiple UK values - reality check pls[edit]

I've done an exercise to amend P27 values for people from the UK, along the lines of ensuring that all of the flavours of UK the person was a citizen of, are represented in their record. Some of those flavours are:

  • Q145 United Kingdom of Great Britain and Northern Ireland 12 April 1927 -
  • Q174193 United Kingdom of Great Britain and Ireland 1 January 1801 - 12 April 1927
  • Q161885 Kingdom of Great Britain 1 May 1707 –1801
  • Q179876 Kingdom of England

Thus for a person born in 1900 and expiring in 1950, I'm giving them two values: Q145 and Q174193 ... and I'm working on adding PQs such as end date and end cause - e.g. https://www.wikidata.org/w/index.php?title=Q12285904&type=revision&diff=712129483&oldid=712088876

User:Pierrette13 has announced that they do not see the sense in this, and has reverted a couple of my edits. See talk. As far as I understand, Pierrette13 objects that the 1927 change from United Kingdom of Great Britain and Ireland to United Kingdom of Great Britain and Northern Ireland is a matter of legal form only (Gtranslate: "In my opinion, it can not be said for the same person that she comes from the United Kingdom and that she comes from the United Kingdom of Great Britain and Ireland, that makes no sense, especially since the question nationality, and not the legal form of the country.")

My view is that our 1900-1950 person was a citizen of United Kingdom of Great Britain and Ireland, when a state of that name existed. They were also a citizen of United Kingdom of Great Britain and Northern Ireland when a state of that name existed. My use case is a wish to query wikidata for people who were a citizen of United Kingdom of Great Britain and Ireland, and find our person; and to query wikidata for people who were a citizen of United Kingdom of Great Britain and Northern Ireland, and find our person. If the person's record does not contain both values, then that query can only be done inferentially by looking at date of birth.

Is there / what is the consensus on P27 carrying multiple UK values? @Pierrette13: --Tagishsimon (talk) 22:23, 17 July 2018 (UTC)

The article en:United Kingdom is sitelinked to United Kingdom (Q145), but as far as I can see the history of the article doesn't start at 1927, and the article doesn't even mention any change in 1927. I suggest changing the inception on United Kingdom (Q145) to recognize the longer period of the state's existence, even if it has changed over time. Then no multiple citizenships are necessary just because the form of the state changed a bit. Ghouston (talk) 00:26, 18 July 2018 (UTC)
It's probably worth saying that wikidata items are defined by their properties, not their sitelinks. And you're placing quite a lot of dependence on an article that seems to know nothing of the w:en:Royal and Parliamentary Titles Act 1927, which established the naming of the state described by Q145. I'm kinda understanding your response to say "even though a succession of states and their start and end dates can be defined - e.g. in Q145, Q174193, Q161885, Q179876 - we should instead have an overarching Generic UK item. And instead of making people citizens of the state they were a citizen of, we make them a citizen of the (made up) Generic UK? Is that really what you mean? And if so, does that hold for only the first two states - United Kingdom (Q145) & United Kingdom of Great Britain and Ireland (Q174193) or, as the en.wiki article goes further back, are you suggesting wrapping Kingdom of GB, and Kingdom of England citizenships into the generic UK? --Tagishsimon (talk) 04:49, 18 July 2018 (UTC)
Yeah, I'd say there needs to be a generic UK item to link with the Wikipedia articles, and since they are already linked with United Kingdom (Q145), that's probably the right item. How far back to go, I don't know. In most cases I'd expect there's enough continuity at each change that a citizen (subject) retains their status, so it doesn't really make sense to talk about a new citizenship. Ghouston (talk) 05:49, 18 July 2018 (UTC)
country of citizenship (P27) has been kicked backwards and forwards a lot, as to whether it should be read narrowly as connecting somebody to a particular state that existed in their lifetime, or should be read more widely to allow a painter to be designated eg French or Italian or Flemish at a time when those states may not have existed in their current form, or even (Flemish) not necessarily existed as independent states granting 'citizenship' at all. I have always found the whole area a minefield, that I wish somebody would definitively clean up, and have never known what value to give a P27 statement for somebody born before 1900, so confess that as a result I have left it blank for somebody else to add, leaving the decision to them. Pinging @Multichill, Andrew Gray: who have both at times had thoughts or suggestions in this area. Jheald (talk) 09:04, 18 July 2018 (UTC)
I also stopped adding country of citizenship (P27). We currently don't have a proper way to model the connection of a historical person with a contemporary country. Probably easiest to show with examples:
Of course you run into the same problem for England and the UK. This problem needs to be solved before doing any large scale editing. Multichill (talk) 19:33, 18 July 2018 (UTC)
It's not easy. Australian citizenship didn't exist before 1949, yet Australia as a federation has existed since 1901 and the term "Australian" was in use in the 19th century. To say that the United Kingdom was created in 1927 or the Netherlands perhaps in 1815 or 1945 or China in 1949 is to ignore much of the history of these countries as states. Ghouston (talk) 01:38, 19 July 2018 (UTC)
A German state has existed for quite a while. It's boundaries, name and form of government have changed over time, but does it make sense to speak of separate citizenships for Weimar Republic (Q41304), Nazi Germany (Q7318), West Germany (Q713750) and Germany (Q183)? Ghouston (talk) 08:03, 19 July 2018 (UTC)
jurisdiction (Q471855) marks a boundary. --Succu (talk) 21:09, 31 July 2018 (UTC)
Yes, we only have one item United States of America (Q30), not one for each time its boundaries changed. Even though the state today is far larger than the one founded in 1776, the continuity is favoured over the discontinuity. Ghouston (talk) 22:59, 31 July 2018 (UTC)
As Jheald says, I made some notes on it earlier in the year (building on the Wales/Australia discussions), with the hope that we could have a broad discussion about the best way forward - but it's a really complex issue and it's been really hard to pull it all together clearly. I'll try and find some more time to work on it since the problem certainly won't go away... Andrew Gray (talk) 20:54, 21 July 2018 (UTC)