Wikipedia talk:Article titles/Archive 23

This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Archive 20

Archive 21

Archive 22

→

Moving the archives

It seems this page was moved without moving the talk subpages (i.e. the archives). Should this be corrected (I assume it should, otherwise they don't show up in the archive search box at the top)? Does anyone have a quick way of doing it (e.g. move the page back and then move it again with the "move subpages as well" option that I believe admins have available)?--Kotniski (talk) 14:59, 6 February 2010 (UTC)

Beyond my skill set. I have left a note with an admin that may be able to take care of this. (here) Blueboar (talk) 17:29, 6 February 2010 (UTC)

Sorry, but I just got around to doing this. The archive pages are all moved. When doing this, I noticed that they were all not that large so I increased the size for future archives. Vegaswikian (talk) 22:07, 6 February 2010 (UTC)

A slightly different question. Should the archives like Wikipedia talk:Naming Conventions (Names and titles)/ Old Archive 2 be moved or left at the old name? Vegaswikian (talk) 22:10, 6 February 2010 (UTC)

This move has left its derivatives at the old name, and has also broken all the shortcuts pointing here. 81.111.114.131 (talk) 23:15, 6 February 2010 (UTC)

Wow, I completely missed this because I've been purposefully avoiding this page for months. Good deal. I'd say just leave all of the old archives where they are, but they need to be linked from here somehow. It'll take a little work, but just get the page names and create an {{archivebox}} with links to them (alternatively, a single navigation page could be created...). Their still there, and perfectly accessible through search, so anything else is really overkill.
— V = I * R (Talk • Contribs) 03:26, 7 February 2010 (UTC)

I added 'Naming Conventions (Names and titles)' archives to the box. I could not find an archive 1. Vegaswikian (talk) 06:19, 7 February 2010 (UTC)

I believe "Naming conventions (names and titles)" (however capitalized) is a past name for WP:Naming conventions (royalty and nobility), not for this page.--Kotniski (talk) 18:26, 7 February 2010 (UTC)

Don't know why, but I just found 3 more in that series that were not included in the subpage listing. I added those to the archive box. If they should be moved to be subpages of another page, we can do that. Maybe some discussion here or simply no objections to your proposal. Vegaswikian (talk) 20:51, 7 February 2010 (UTC)

I agree that the archives for "Naming conventions (names and titles)" belong with the project guideline WP:Naming conventions (royalty and nobility). One thing that changing the title of this page is making clear is just how many project level "naming convention" guidelines are out there. We should check them to make sure that they don't conflict with this policy. Blueboar (talk) 21:08, 7 February 2010 (UTC)

And to gather ideas to improve the policy. Septentrionalis PMAnderson 22:37, 7 February 2010 (UTC)

Yes... perhaps I should have said: ... "We should check them to see if they conflict with this policy... then we should discuss any conflicts and try to resolve them (which might mean changing the guideline, but might also mean changing the policy.)" Blueboar (talk) 15:07, 8 February 2010 (UTC)

Just as a general FYI: If anyone sees any opportunity to consolidate the multitude of Naming convention (my favorite sub-topic) into either this document or any broader document, then you can pretty much count on my implicit support.
— V = I * R (Talk • Contribs) 23:27, 8 February 2010 (UTC)

WP:NPOV and Article titles

This is actually worth a seperate section... and I want to get input from the regulars at WT:NPOV.

I think WP:NPOV does apply to Article titles. Note I did not say that titles must be neutral. I think there is a subtle difference here. I would encourage everyone to go back and re-read WP:NPOV so that you understand what that policy does and does not say. Then think about how it applies how we entitle our articles. Blueboar (talk) 19:26, 9 February 2010 (UTC)

I agree on this point. The relevant section is Wikipedia:NPOV#Article titles. Also of relevance, but not related to article titles directly is WP:AVOIDSPLIT. --Gavin Collins (talk|contribs) 16:22, 10 February 2010 (UTC)

To continue... I want to discuss how WP:UNDUE impacts on article titles. This is a continuation of the discussions we have had above... some of you have expressed the opinion that WP:NPOV should trump WP:COMMONNAME... but to my mind, WP:COMMONNAME is an application of WP:UNDUE. Thus, WP:NPOV can not trump WP:COMMONNAME as they are the same thing.

WP:Undue indicates that being neutral does not mean that we treat all viewpoints equally... instead neutality means we give viewpoints due weight in accordance with reliable sources. If a significant majority of sources express a given viewpoint, then that viewpoint is given more weight in our article. This applies to how we entitle articles. Neutrality in naming articles does not mean we treat all possible titles equally... neutrality means we give each possible title due weight in accordance with the reliable sources. If a significant majority of sources call something by a particular name, or use a particular term in describing it, then that name or term should be given more weight in our article... by using that name or term in the article title. To give alternative names their due weight, we should prominently mention them as being alternatives in the lead. Blueboar (talk) 16:48, 10 February 2010 (UTC)

I don't like the language of WP:ABC trumping WP:XYZ (WP:ANYTHING should try to most lucidly express the current consensus on the matters within its subject area - it doesn't become the consensus), but as to the substance, I agree with Blueboar.--Kotniski (talk) 17:00, 10 February 2010 (UTC)

Thanks for initiating this discussion - it's a critical one. I do think you're focusing on only one aspect of achieving neutrality to the exclusion of some other, equally important considerations. Avoiding undue weight is a central principle, but so is maintaining an impartial tone. The NPOV guidance explicitly says that we have to use an "editorially neutral, point of view", and that "[t]he neutral point of view neither sympathizes with nor disparages its subject, nor does it endorse or oppose specific viewpoints." In the Wikipedia:Neutral_point_of_view#Article_naming section says:

"Sometimes the article title itself may be a source of contention and polarization. This is especially true for descriptive titles that suggest a viewpoint either "for" or "against" any given issue. A neutral article title is very important because it ensures that the article topic is placed in the proper context. Therefore, encyclopedic article titles are expected to exhibit the highest degree of neutrality. The article might cover the same material but with less emotive words, or might cover broader material which helps ensure a neutral view (for example, renaming "Criticisms of drugs" to "Societal views on drugs"). Neutral titles encourage multiple viewpoints and responsible article writing."

That seems to me to be a very clear statement that neutrality is a consideration of its own, separate and apart from which name is most "common." It does say that "proper names for people or events which incorporate non-neutral terms - e.g. Boston massacre, Tea Pot Dome scandal, Edward the Confessor, Jack the Ripper - are legitimate article titles when they are used by a consensus of the sources" - but we're discussing situations where there is no clearly identifiable "common name," and current groups rather than historical events (which, like Biographies of Living People, arguably requires an extra effort for neutrality).

Again, I don't believe that self-identification of a group should be dispositive for Wikipedia article titles, nor that we must always come up with titles that the groups involved would agree with. I am arguing that we should choose article titles that are neutral, do not "disparage" their subjects, that encourage multiple viewpoints and do not "endorse or oppose" a particular point of view. In some cases that may not be the same as the most common usage in the sources we're reviewing. EastTN (talk) 17:39, 10 February 2010 (UTC)

Agreed that article titles are expected to exhibit the highest degree of neutrality, but how do you enforce such a lofty objective? The only way is external validation, and that is by using that title which supports its notability, i.e. the title has been addressed in (or better still defined by) significant coverage from reliable secodnary sources that are indepedent. Another way of saying this is extraordinary names (such as "segmented" article names) must be supported by extraordinary sources. That way, if an article title is highly judgemental (e.g. Moors murders), then the sourcing to support can be externally verified. --Gavin Collins (talk|contribs) 19:19, 10 February 2010 (UTC)

I agree that the first place we need to look is disinterested, neutral reliable sources. For most articles there should be a small set of well-recognized, accepted names in use by such sources that we can look to in selecting an article title. But what makes that work as a basis for achieving neutrality is not the frequency of the usage, but that it has become commonly accepted among sources that are themselves neutral and disinterested. For most articles we should be basing our work on a wide array of neutral and disinterested sources. But if:

A topic is highly controversial, so that most of the sources are not neutral and disinterested, or
No clear consensus regarding terminology has developed in the literature;

then some sort of editorial judgment is required. I believe that in those instances, which should be very unusual (or even "extraordinary", as you put it), making a good editorial call on neutrality will often require more thought than simply basing the article title on the name that is used most frequently. We also need to look at the nature of the sources using each term (are they neutral, for, or against; are they themselves encyclopedic or disinterested), whether the various terms are used purely descriptively or for "advocacy" purposes (is the terminology "spun" - for or against the subject of the article), whether the language is unnecessarily "loaded". In some cases we may be better served to pick up a minority usage because it reflects the most neutral and encyclopedic sources, rather than a more common usage that's "spun" in favor of a particular point of view. In other cases neutrality may be best served by picking a terminology that's not associated with either side of a controversial topic. Again, I really don't believe this should be an issue for something like 99% of Wikipedia's articles. But for the tough cases, I don't think we can necessarily equate "neutrality" with the name or appellation that has the highest word count.EastTN (talk) 21:42, 10 February 2010 (UTC)

Sorry - I may have misunderstood your question. How do you "enforce" neutral choices of article titles? The same way we "enforce" neutrality more generally, I guess. I'd like to see the guidelines on article titles stress the importance of neutrality a bit more strongly, and I think they could benefit from some more detailed discussion of the factors editors could/should consider when thinking about cases like this. (And I agree that in cases where we do have significant coverage from neutral, disinterested secondary sources, and there's a consensus among those sources, the guidelines should call for using the consensus terminology - but not all cases will be that neat.) Beyond that, I think we have to trust the process and the community.EastTN (talk) 21:51, 10 February 2010 (UTC)

I am not sure that reliance on "editorial judgement" or a "consensus of the editors" is the right approach to follow because sometimes local consensus may conflict with Wikipedia's content policies. I understand that some article topics may be the subject of "biased" coverage and this may give rise to "loaded" language, but that may be a problem related to the absense of broad based coverage (i.e. a lack of notability) or a lack of reliability of external sources (again, a problem associated with lack of notability).
Having said that, there is no such thing as a neutral or unbiased source. On the contrary, we value reliable secondary sources that are independent for the commentary, criticisms and analysis that provide context for the reader; it is precisely the in-depth opinions of disinterested commentators that is the key hallmark of significant coverage. This is recognised by WP:NPOV which says that "non-neutral terms...are legitimate article titles when they are used by a consensus of the sources". I think it is the purpose of this policy to define what is a consensus of the sources, rather than a consensus of the editors. --Gavin Collins (talk|contribs) 08:54, 11 February 2010 (UTC)

I'm not suggesting that editorial judgment or a local consensus should override Wikipedia's content policies - I simply meant to say that as a practical matter that's how all of these policies (including neutrality) are implemented and enforced (by the community working through the process). As for the consensus of the sources, I understand the point that you're trying to make. But whether we like it or not, editors are making these judgments, because they have to make them. Editors are deciding which sources are treating the subject in a neutral and disinterested fashion, which are partisan, and what that means in determining the "consensus of the sources" - or implicitly treating them all as if they were neutral and disinterested by default. Either way, they're making a decision on it. When there's not a clear consensus among the sources - due to controversy, changing views of the topic over time, or whatever - they're still selecting titles for their articles. The sentence in the WP:NPOV which says that "non-neutral terms...are legitimate article titles when they are used by a consensus of the sources" is important, but it's not the only thing that policy says. It's also relevant that it says "[s]ometimes the article title itself may be a source of contention and polarization. This is especially true for descriptive titles that suggest a viewpoint either "for" or "against" any given issue. ... Therefore, encyclopedic article titles are expected to exhibit the highest degree of neutrality. ... Neutral titles encourage multiple viewpoints and responsible article writing." We need to consider all of the guidance, and in some situations where there's not a clear consensus in the sources it may make sense to base the title on something other than the name that's used by a plurality of the sources. In those case, I believe this is consistent with the content policies. EastTN (talk) 16:53, 11 February 2010 (UTC)

I attempted once to introduce language into WP:Consensus to the effect that a "local consensus" to directly violate major content policies was simply invalid, but the editors at that page rejected it. IMO, the question of consensus is not "Do we agree to say that Mary is (unverifiable disparaging terms here)" but "Do we agree that saying Mary is (unverifiable disparaging terms here) complies with Wikipedia's content policies?" WhatamIdoing (talk) 19:54, 11 February 2010 (UTC)

I concur. Such language was in WP:Consensus when I first saw it, and for some time thereafter. What happened to it? Septentrionalis PMAnderson 21:12, 12 February 2010 (UTC)

Here's my basic problem. It seems to me that we're trying to treat this far too mechanically. Thinking about "Jack the Ripper" may be a good place to start. "Jack the Ripper" is not just a "consensus" name, it's effectively the only name in common use. We pretty much have to use it for the article title, because there aren't any other plausible choices. But if we knew who committed that series of murders, the analysis might be different. For instance, we have an article title Gary Ridgway instead of one titled the "Green River Killer", "Dennis Rader" instead of the "BTK Killer", "David Berkowitz" instead of "Son of Sam", "David Carpenter (serial killer)" instead of the "Trailside Killer", "Raymond Fernandez and Martha Beck" instead of the "Lonely Hearts Killers" and"Vincent Johnson" instead of the "Brooklyn Strangler". (A different decision was made in the case of the "Boston Strangler", where the attribution is disputed, in the case of the "Hillside Strangler", where two individuals were implicated, and in the case of the "Freeway Killer", where there were three independent killers - my point isn't that it always has to go one way or the other, but simply that the analysis becomes more complex.)

The easy ones are easy. But when we get to the hard topics, like "Pro-choice" and the "Pro-life movement", neutrality requires more than simply doing a Google search to determine whether "pro-choice" or "pro-abortion" is used more often, or whether "pro-life" is used more than "anti-choice". I am definitely not saying that we shouldn't fully recognize the importance of considering what the consensus in the relevant literature is. I am saying that we should recognize that more may be required on occasion to achieve neutrality. And yes, I recognize that "simply doing a Google search" is a bit of a caricature of the position stated above, but I'm trying to make the point that in practice, some human editorial judgment is always involved if we're going to do this well - it's not a mechanical measurement of which usage is most common. Let's recognize that editors are having to make judgment calls, and instead to make the process more mechanical, let's give editors guidance on making better judgments. EastTN (talk) 18:22, 11 February 2010 (UTC)

The quotation from WP:NPOV above applies to descriptive article titles, where we are making one up, as the example makes clear. Furthermore, the specific concern with criticisms of drugs, like criticisms of capitalism (or communism), is that it tends to attract ranters on one side of an issue, who quote published rants on their side of the issue, and so make a bad article - not that the title itself is somehow biased.

This is a bad thing; but a limited bad thing. It is not a general principle; it's a warning against using titles which encourage POV forks. We may do well to include exactly that, with a cross-link - but not at the top. Septentrionalis PMAnderson 19:07, 11 February 2010 (UTC)

As far as WP:NPOV has anything to say on the matter, it is that neutrality over contentious issues is important, especially with regard to descriptive titles. With the names of entities and locations, neutrality is not so heavily stressed, which is inline with other policy guidance which has given a significant role to the self-identifying name of an entity. As has been said, neither google hit-counting or "neutrality" are always the golden bullet, because in some instances two names will exist with almost equal usage, one will be favoured by group A, and another by group B. Which is the neutral choice? In such circumstances, it is common sense to consider following the usage of the entity the name actually refers to, or of the group who actually live in the location in question. And this has been the practice of many editors. This also avoids identifying a group by a name they find offensive, or identifying a city or geographical feature by a name that seems to take sides in a territorial dispute. The Dalits v Untouchables example is resonant here. EASTTN's Pro-Choice v Pro-Abortion, Pro-Life v Anti-Abortion example is a similar instance where number-crunching is unlikely to produce as stable or neutral a solution as going with the titles the groups use to identify themselves. Xandar 03:19, 12 February 2010 (UTC)

I disagree that the name the entity calls itself is necessarily the more neutral choice. It might be... but it might not be. Entities often call themselves by names that display a distict POV. I am reminded of the old saw... "The Holy Roman Empire was neither holy, Roman nor an empire".

That said... we do need to accept the fact that sometimes there simply is no "neutral" name for something... situations where someone is going to be offended no matter what name we choose as the title of our article. Blueboar (talk) 03:37, 12 February 2010 (UTC)

I think that is correct. What ever the entity calls itself has to be given due consideration, but not undue consideration, particularly if there are more numerous, more reliable sources supporting a different article title. My view is that contentious article titles can only be arrived at if all the sources are lined up and ranked in terms of reliablity & depth of coverage to reach a point where the it can be established that a clear choice of name has emerged, e.g. Jack the Ripper. This approach is more mechanical in methodology, but the closer we stick to the sources cited in the article, the better the fit for the title, and it will be easier to draft a set of rules along these lines, rather than a more judgemental apporach that is not based on external validation.
Note that this does not apply to nor can it work with topics that are not notable, and which are not the subject of a broad range of sources. Often disagreement over an articles title is associated with lack of notability, but we should set that issue aside for now. --Gavin Collins (talk|contribs) 17:54, 12 February 2010 (UTC)

What the entity calls itself (unless - as usual - it is also what everybody else calls it) is not the most neutral choice. It reflects the point of view of the organization - often, indeed, the point of view of a minority who claim to be the organization.

This is an argument made only by the pedlars of points of view - like the repetitive Xandar. They are not useful contributors to Wikipedia. Septentrionalis PMAnderson 21:09, 12 February 2010 (UTC)

Self-identification

Gavin... Thank you... you just hit the nail on the head as far as the entire "self-identification" debate goes... I think you are correct in saying: "What ever the entity calls itself has to be given due consideration, but not undue consideration". What we have been debating (repeatedly) for the last several months is what constitutes due vs. undue consideration... we have strong disagreement as to where to draw the line between "due consideration" and "undue consideration".

It may be that this is not a line we can draw in a policy... After all, to some extent the line will be different for each article, since the specific circumstances of each article are unique. Blueboar (talk) 19:40, 12 February 2010 (UTC)

It should be given exactly as much consideration as speakers of English generally give it. Often this is a great deal, but if so, it is reflected in usage. Septentrionalis PMAnderson 21:09, 12 February 2010 (UTC)

But that still leaves situations in which there is no clear usage. If there are two or three names that are used essentially equally by the english language sources... one of which is the name used by the entity itself... should we consider that bit of information when determining the title of the article? Is considering this fact "due consideration" or "undue consideration"? I suspect that the answer to that question may be different from article to article. Blueboar (talk) 23:21, 12 February 2010 (UTC)

I don't think you have disagreed with what I said. If there are two or three equally common names, one or two of which are self-identifiers, then English gives some weight to the one or two. Septentrionalis PMAnderson 00:16, 13 February 2010 (UTC)

No I haven't disagreed... I am asking in order to refine what you said. So now we get to the final question... how much weight? Blueboar (talk) 05:18, 13 February 2010 (UTC)

No more than WP:UE would allow, which is often considerable. If there are special cases where self-identification deserves and has received independent weight, I have no idea what they are; I have seen none. Septentrionalis PMAnderson 19:47, 13 February 2010 (UTC)

PMA... WP:UE does not really apply to what I am asking about. If sources (all in english) are relatively evenly split between calling something X or Y (both being english names or terms)... How much weight should the fact that the entity itself self-identifies (also in english) as X be given? Blueboar (talk) 23:20, 13 February 2010 (UTC)

No additional weight whatsoever, in your hypothetical case (in which UE applies equally on both sides). If the sources are evenly split, go to our other considerations: Which is shortest? Which is easiest to link to? Which will readers expect, given what we have called other articles? and so on. If those do not decide, flip a coin. But by that time we are dealing with a case with so many hypothetical conditions that no instances have ever come up or are likely to come up; when policy need not speak, it should be silent. Septentrionalis PMAnderson 19:57, 15 February 2010 (UTC)

I don't think we have to state a weight. We just have to say that it is one of the considerations that editors can/should take into account when coming to a consensus. However I think it should be listed along with other relevant considerations, rather than just leaving a blank. Xandar 02:59, 14 February 2010 (UTC)

PMA's reference to UE makes intuitive sense to me, although I can't explain it.

Perhaps this will help: It usually takes a lot to get a modern English source to not use a self-identifying name. If the sources are actively rejecting a self-identifying name, then on what scholarly basis to we give it special attention? WhatamIdoing (talk) 03:55, 14 February 2010 (UTC)

What is PMA's reference? Can anyone cite this or provide a link to what this means? Apologies, but I have lost the thread's meaning.--Gavin Collins (talk|contribs) 09:35, 15 February 2010 (UTC)

I'm not following the relevance of WP:UE either - it seems to be talking about how to bring in a non-English term when no English common name is available.

As for the rest, I do think we may be asymptotically approaching a consensus. Thinking of the situation where there is no one clear usage, I strongly agree with the formulation that "[w]hat ever the entity calls itself has to be given due consideration, but not undue consideration", and the formulation that "sometimes there simply is no "neutral" name for something." I also agree that the answer to the question will likely vary from article to article, which makes me think that we can't state a particular weight that will apply in all cases). As for self-identification, this suggests to me that editors should consider it, but not to the exclusion of considering other usages and the neutrality of the resulting article title - and certainly not to the extent of giving special privilege to the entity's own particular point of view. When no name is completely neutral, editors should try to choose a title that is as neutral as is reasonably possible (consistent with recognizability, precision, ease of finding the article, etc.). EastTN (talk) 15:41, 15 February 2010 (UTC)

Editors should not, and do not, consider self-identification at all; POV-pushers usually consider nothing else; the proper course of action is to ban them at sight - and thus put an end to the appalling claim that any grouplet has a right to control what other people call them. Septentrionalis PMAnderson 19:57, 15 February 2010 (UTC)

I think that is too extreme. Blueboar (talk) 22:30, 15 February 2010 (UTC)

Then propose a middle view; I stand by the last clause, however. Septentrionalis PMAnderson 23:14, 15 February 2010 (UTC)

I thought this was a middle view. Self identification can be one of many secondary things that might be considered when there isn't a common name identifiable in the sources. Blueboar (talk) 23:23, 15 February 2010 (UTC)

I strongly agree. It is too extreme to say that an editor should never consider self-identification, and in my opinion bordering on a personal attack to imply that only "POV-pushers" would consider what an entity calls itself. (I am on record as saying that I do at least think about what the subject of an article chooses to call itself). It may be a mistake to consider it (I've certainly been mistaken about other things), but it's going too far to suggest that it cannot be anything other than intentional misbehavior. I would argue that the middle ground consists of 1) limiting the discussion to situations where there is no identifiable common name; 2) limiting it to being only one factor among many that may be considered; and 3) making it clear that self-identification cannot be used if it results in a non-neutral article title.EastTN (talk) 23:48, 15 February 2010 (UTC)

There is no middle view between "may" and "must not".

You're looking for an away goals rule: if both names are even on points, give the self-identification name additional weight. Though I don't particularly like it, there is nothing inherently wrong with adopting an away goals rule. But the "give the self-identification name additional weight" clause is a violation of our neutral point of view policy. This is not acceptable. Hesperian 23:43, 15 February 2010 (UTC)

The text Blueboar referred to never suggested that self-identification should be privileged with "additional weight." It simply said that editors could consider whether it made sense, in that particular case where there is no obvious common name for the topic, for a group's self-identification to be used in the title. We're not talking about a "'give the self-identification name additional weight' clause".EastTN (talk) 23:52, 15 February 2010 (UTC)

More importantly, saying one can consider the fact of self-identification in reaching a consensus does not imply a result... it can result in two equal outcomes... 1) deciding to use the self-identified name... or 2) deciding NOT to use it. Either could occur depending on all the other things being discussed. Blueboar (talk) 01:32, 16 February 2010 (UTC)

Either outcome is fine as long as it is for the right reason(s). In both cases, "because it is what the entity calls itself" is the wrong reason. Hesperian 01:43, 16 February 2010 (UTC)

If all you want to say is that the self-identification name has just as much right to consideration as every other candidate name, then that is fine with me. I propose the following wording:

""

Hesperian 00:06, 16 February 2010 (UTC)

All flipness aside, that's exactly what I have have said - that it should be considered on the same basis as every other usage. I have also argued (and still believe) that, in dealing with the case where there is no single recognized "common name" for a subject, less experienced editors could benefit from a bit more guidance than:

""

If we can get past this idea that we're somehow trying to grant organizations the right to pick the title for their own Wikipedia articles, I still think it would be useful to discuss what that additional guidance might look like. EastTN (talk) 00:22, 16 February 2010 (UTC)

Sounds good. However, I'd also like to point out that this policy fails utterly to state that titles that start with the letter H should not be rejected out of hand but should be considered on the same grounds as every other candidate name. I think, in dealing with the case where there is no single recognized "common name" for a subject, less experienced editors could benefit from a bit of guidance here.

The above is only flippant if you can articulate how the two cases differ.

Hesperian 01:14, 16 February 2010 (UTC)

The two cases differ in one very important and significant way... people get into lots of arguments about "but this is what the group calls itself" (self identification) where they don't get into aguments about "but it begins with the letter H". So it is helpful to say something about where and when self-identification can be considered, while there is no need to say something about where the letter H can be considered. Blueboar (talk) 01:40, 16 February 2010 (UTC)

Now you've moved the goalposts again. Let me summarise:

You guys say "Sometimes the self-identification name should be used as the title". I say "Yes, often."
You guys say "Sometimes the fact that a name is a self-identification name should be taken into account when choosing a title." I say "No, never."

Every time I agree with 1., you proceed as though I have agreed with 2. Every time I disagree with 2., you proceed as though I have disagreed with 1. It is really starting to piss me off. Can you guys please figure out which of these two assertions you are proposing to insert in the policy, and then stay on topic. Hesperian 01:50, 16 February 2010 (UTC)

To be clear: If the independent sources are truly equally divided, with exactly half calling the organization X, and half calling the organization Y, you would never consider self-identification as even a tie-breaker, correct?

Then what would you use to make a decision? We cannot, as a purely practical matter, place the article under both titles, and arbitrary solutions, such as flipping a coin, have never enjoyed support on Wikipedia. WhatamIdoing (talk) 02:20, 16 February 2010 (UTC)

I can do no better than quote PMA above:[1]

No additional weight whatsoever.... If the sources are evenly split, go to our other considerations: Which is shortest? Which is easiest to link to? Which will readers expect, given what we have called other articles? and so on. If those do not decide, flip a coin. But by that time we are dealing with a case with so many hypothetical conditions that no instances have ever come up or are likely to come up; when policy need not speak, it should be silent. Septentrionalis PMAnderson 19:57, 15 February 2010 (UTC)

Hesperian 02:27, 16 February 2010 (UTC)

Arbitrary solutions (flipping a coin, number of letters in the name) are routinely and firmly rejected by the community. "Easiest to link to" is supposedly (as far as anyone can make out, which perhaps isn't very far) what has been intended by the long-standing, if opaque, "favors readers over editors" clause, and thus that criteria has been rejected by the community. That leaves you with "consistent with what we called other articles" as the only possible tiebreaker.

If this point is unhelpful (as it will be in a significant proportion of disputes), then is there anything else that (1) you would consider and (2) hasn't been previously rejcted by the community? WhatamIdoing (talk) 03:36, 16 February 2010 (UTC)

You've ignored PMA's "Which is shortest", a concise paraphasing of our "conciseness" criterion. And you need to read Wikipedia:Article titles#Deciding an article title again: see if you can find the phrase "and to which editors will most naturally link from other articles". I believe coin-flipping was a tongue-in-cheek suggestion by PMA, intended to suggest what is stated explicitly in the next sentence: "by that time we are dealing with a case with so many hypothetical conditions that no instances have ever come up or are likely to come up".

Frankly, I think the notion that we need self-identification because we won't be able to make decisions without it, is the weakest argument that has been put forward in this discussion. There are an abundance of criteria on the policy page; if no consensus emerges from them, adding yet another criterion is going to make it harder to find consensus, not easier. And even if it did make it easier, I'd still be opposed, because I'm opposed to a policy that makes it really, really easy to arrive at the wrong decision. Hesperian 04:05, 16 February 2010 (UTC)

WhatamI hasn't ignored conciseness; he disagrees with it, as he disagrees with much of this policy. His claims of community consensus require citation; discussions on this page are not the voice of the community, and there has been no such consensus, even here. Indeed, as Hesperian points out, many of these "rejected" points are part of this policy page - and, more importantly, of daily discussions on move requests. Septentrionalis PMAnderson 04:27, 16 February 2010 (UTC)

Aye, she certainly does disagree with "The choice of article titles should put the interests of readers before those of editors," and precisely because she favors reasonable concision (but not wholesale abbreviations) and appropriate precision, even if this might occasionally and slightly favor "editors over readers" -- or, far more likely, give them equal consideration -- assuming that this is what this opaque phrase is supposed to address, which nobody seems to know. Despite my repeated questions, nobody has ever produced an explanation for this verbiage that wasn't promptly contradicted by someone else, but several people seem determined to keep this meaningless and/or confusing text in the policy, perhaps because it sounds so high-minded and poetical, rather than providing any practical or necessary advice to editors. WhatamIdoing (talk) 04:47, 16 February 2010 (UTC)

And so we reach yet another impass, this time with three view points instead of two...

self-identification should be a major criteria, always to be considered even when there is another name that is more common.
self-identificication should never, ever be considered under any circumstances.
self-identification should be considered a legitimate secondary tie breaker... along side many other tie breaking considerations.

Have I stated the view points correctly? Is it likely that anyone will change their view? If not, I don't see any way to resolve this impass without broader community involvement. Perhaps it is time to toss these three options to the wider community and do another RfC. Any objections? Blueboar (talk) 03:18, 16 February 2010 (UTC)

Does anyone actually support #1 (self-identification is more important than widespread common use by English sources)? WhatamIdoing (talk) 03:37, 16 February 2010 (UTC)

Well, that was Xander's original position... although, to be fair, he seems to have backed off from that. It may still be Storm Riders position (he/she hasn't commented recently so I am not sure if his/her position has changed)Blueboar (talk) 03:39, 16 February 2010 (UTC)

Since there seems to be constant confusion between "self-identification should never be considered when choosing a name" and "the self-identification name should never be chosen", care need to be taken in phrasing an RfC to avoid conflating the former with the latter. Hesperian 03:51, 16 February 2010 (UTC)

Yes, I got that... your position is the former. You object to the consideration, not the use.Blueboar (talk) 03:59, 16 February 2010 (UTC)

So is mine. Very often, for example, there is only one name for an organization (of any frequency), which must therefore be both common usage and self-identification. Septentrionalis PMAnderson 04:27, 16 February 2010 (UTC)

Using PMA's example... everyone would agree on the name... those who support #1 would choose the name because it was the self-ID, while both position #2 and position #3 would choose it because it was the common name. Both 2 and 3 agree with what happens when there is only one name of any frequency ... but they differ over what we can and should consider when there are two or more names that have equal frequency in the sources (one of which is the self ID name). Am I correct that you do not think we should consider self-identification even in that situation... as a tie breaker? Blueboar (talk) 04:47, 16 February 2010 (UTC)

Exactly. Going back to my earlier post about "What ever the entity calls itself has to be given due consideration, but not undue consideration", I think that this needs a major qualification. This is what goes on in the real world, but in the context of Wikipedia, editors are not free to create article titles that can't beverfied by reliable, third party sources:

self-identification should not be considered, whether or not there is another name that is more common. Although self-identification is the source of information that most secondary and tertiary sources get their information from, from a Wikipedia standpoint the entity is a primary source, and as such does is not an independent source of information;
self-identificication should never be considered in isolation, because it is a primary source , and should never be considered in the absence of signficant coverage from reliable secondary sources. If there are no sources that address the entity directly and in detail, chances are that it is not notable at this time;
self-identification based on the primary source and should not be considered a legitimate tiebreaker. In the real world, identities can change quickly, but because of splits and schisms, there is not always a one-one relationship between the old and the new identity. Reliable secondary sources are required to verify that the correct title is being used to identify the correct entity.

In short, secondary sources must be used to confirm identity in every case, or else we failing to make proper identity checks using external validation. --Gavin Collins (talk|contribs) 09:59, 16 February 2010 (UTC)

Gavin, I agree with what you just said... but as I see it, in the "tie breaker" senario all your concerns are dealt with... when we consider self-identification in that senario we are not considering the name in isolation. The self-identification is being considered due to signficant coverage from reliable secondary sources. The situation assumes that we have looked at the reliable sources, and found that there is equal significant coverage between two or more names. We have reliable secondary sources to verify that the correct title is being used to identify the correct entity. What we don't have is a clear majority of secondary sources agreeing on any one name. In that senario, I think it is reasonable to include self-identification as one of many "things to consider" that will help us choose between the names that are used in reliable secondary sources. Blueboar (talk) 15:26, 16 February 2010 (UTC)

On that I disagree, and perhaps an example would be useful to illustrate the points I have made. In many secondary sources the terms "Persian" and "Farsi" are often used to describe the Persian language, but Persian is not the term used by the speakers of the language itself. Reliable secondary sources show term "Persian" and "Farsi" both have the same root (Greek Pérsis), which is why the English nomenclature is "Persian". This might be a special case, but I would say this is an example of where reliable secondary sources, even if they are in the minority, have sufficient weight to over rule self-identification. This example shows to me that, in theory at least, secondary sources should be used to confirm an article's title in every case. I am open to counter arguements, but I feel this is a clear and unambigious starting point. --Gavin Collins (talk|contribs) 15:55, 16 February 2010 (UTC)

I think you misunderstand my position... To adapt your example, let us invent a language... and assume that out of a total of 100,000 reliable secondary sources, 49,000 use "term X" and 48,000 use "term Y" and 1,000 use "term Z". We can rule out "term Z" since it is used by a distinct minority of sources. WP:COMMONNAME tells us to use either "term X" or "term Y", but since they are used essentially equally, it does not tell us which to use.

In that situation we need to look at other considerations, we need to ask further questions... for example: "Is one term sholarly usage while the other is more colloquial?" "Is usage determined by the age of the source (with more modern sources using one term and older sources using the other)?" These are just a few of the many questions that need to be asked... and what questions are asked will be slightly different from article to article ... but one of these many questions should be: "Is one of these terms what speakers of the language use"? None of these questions inherantly carries more weight than any other. It is the over all weight, taking all the answers together, that determines which term should be used for the article title. Blueboar (talk) 17:52, 16 February 2010 (UTC)

And in the "Persian"/"Farsi" example, we're not talking about two English names for the same subject - we're talking about an English term and a transliterated Persian term. While the term is used in some linguistic literature, it isn't used frequently enough by English speakers to have become a common English name. The real problem occurs when there are multiple terms in the English language for a single topic, none of which is clearly the "common name" for it.EastTN (talk) 20:05, 16 February 2010 (UTC)

And I think the tie-breaker scenario is a fantasy; I have never seen an example of an absolute tie between other considerations, much less one where self-identification was thrown into the scale and tipped it. This page should not invent procedures which the community does not use. Septentrionalis PMAnderson 17:05, 16 February 2010 (UTC)
- I've seen a few situations in which the balance was so close that it might as well have been an absolute tie, although I can't remember one offhand for which self-identification (an issue that rarely affects my main subjects) would have been relevant. For example, is it "lymphoblastic leukemia" or "lymphocytic leukemia"? It's the same thing, and just totting up the number of hits on your favorite web search engine won't tell you the subtle difference (the choice of terms indicates the age of the patients the source is most accustomed to dealing with). WhatamIdoing (talk) 19:05, 16 February 2010 (UTC)

- Even if it is a fantasy, it might be useful guidance to rule out the use of self-ID in non-fantasy cases by saying something like "Self-ID should be used only as a last resort, i.e. in cases where two or more names are judged to be equally valid when all other factors are taken into consideration." —Codrdan (talk) 03:23, 17 February 2010 (UTC)

Going back to Blueboar's three choices . . .

self-identification should be a major criteria, always to be considered even when there is another name that is more common.
self-identification should never, ever be considered under any circumstances.
self-identification should be considered a legitimate secondary tie breaker... along side many other tie breaking considerations

. . . I would argue that neither #1 nor #2 is ultimately consistent with the current practice of most editors or with a neutral point of view. Editors do not use self-identification to trump well-established common names (nor do I believe they should). Where there isn't a well-established common name, editors do discuss what an entity calls itself as they choose article titles - but even in that case, they do not always end up basing an article title on it (again, even in that case I don't think that self-identification should always be the deciding factor). Beyond inconsistency with how editors actually behave, the first two options share a common fundamental flaw - they both single out a particular viewpoint for special treatment (privileging it in the one case, and excluding it in the other). If there is no common name, then editors have to base article titles on other, secondary factors. In practice, they look to a number of things. Which usages are most common, even if none of them rises to the level of a "common name"? What title would most clearly describe the subject? What does the entity call itself? What term or terms do the highest-quality, most encyclopedic sources use? Which terminology is most neutral? Which search terms are users most likely to use? All of these are legitimate considerations, and the correct answer will vary from situation from situation (that's why we have human editors, instead of some sort of "article titling bot"). I believe that the guidance would be more useful and more consistent with actual practice if we used two or three sentences to describe option 3 and lay out some of the more common/important secondary considerations editors may need to consider when there is no clear common name for the subject of an article. EastTN (talk) 20:05, 16 February 2010 (UTC)

I oppose any mention of option 3, as contrary to WP:NPOV. It is not legitimate under any circumstances, and those who insist on mentioning it are proposing language applicable only once in a blue moon, even under its own terms, wrong then, and predictably and profoundly abusable. Septentrionalis PMAnderson 20:17, 16 February 2010 (UTC)

I agree with Pmanderson in this regard. In answer to EastTN, clearly "Farsi" is the most commonly used, and most commonly cited. Yet it is the quality of the sources that provide knockout evidence that topic should be refered to as Persian language. I don't think you are contradicting each other, but clearly we have not reached the point where we can agree on what is consitutes knockout evidence of what decides an article nameyet. My view is that it is the reliablity and quality (in terms of significant coverage) of a source that matters particularly where the source provides a defintion to support the article title. Usually such a definition is obtained from the entity itself (i.e. self-definition), but where a reliable secondary source defines a topic better than self-indntiication does, that is the best source of all because it is independent of the entity.--Gavin Collins (talk|contribs) 21:59, 16 February 2010 (UTC)

Absolutely. I have no problems with the idea that "quality of sources" is an important part of a consensus discussion. Blueboar (talk) 00:38, 17 February 2010 (UTC)

Indeed. In a perfect world (which this isn't), we wouldn't bother with conciseness, precision, consistency, etc. We would simply follow usage in reliable sources. If multiple names are in usage in reliable sources, we would examine usage more deeply. There is a right way and a wrong way to do that.

The wrong way is to associate the various names with points of view, and then choose a point of view to endorse—that would be a violation of our commitment to neutrality, and that is precisely what the self-identification proposal would have us do.

The right way is to associate the various names with shared editorial values. What name is used by sources that, like us, value neutrality? What name is used by sources that, like us, are there to inform and educate, not merely to titillate and entertain? What name is used by sources that are written for a broad audience rather than the specialists in a tiny subfield? Hesperian 01:03, 17 February 2010 (UTC)

I completely agree... where I think we disagree is whether any of this conflicts with asking "what do the natives call it?" as one of many questions we should ask when reaching a consensus on what to title an article. I don't think there is a conflict. Blueboar (talk) 03:09, 17 February 2010 (UTC)

Taking into account "what the natives call it" equals "associating the various names with points of view, and then choosing a point of view to endorse". Hesperian 03:39, 17 February 2010 (UTC)

I agree we have to steer well clear of this approach. For me the rule for deciding what an article title should be is closely aligned with WP:GNG:

Recognizable –If an article title has been the subject of significant coverage from reliable secondary sources that are independent of the entity, it is presumed to have been recognized by the world at large as being the article title for that particular topic.

Wording along these lines needs to to into the section Deciding an article title, as the wording provides little useful guidance at the moment. --Gavin Collins (talk|contribs) 09:00, 17 February 2010 (UTC)

I think closer to what we mean is "If an entity has received significant coverage from reliable secondary sources [that are independent of the entity - why??] under a particular name, it is presumed that that name will be a recognizable title for the article." (It's not what the world at large has recognized - but what our readers will recognize. And there's no reason to say the title - it is likely not to be unique in this respect.)--Kotniski (talk) 09:35, 17 February 2010 (UTC)

Why should we discount reliable sources that fail to be independent of the subject? --Kotniski (talk) 13:08, 17 February 2010 (UTC)

Independent sourcing is an important indicator that a topic is not a neologism. For example, in Wikipedia:Articles for deletion/Socionomics (2nd nomination) the decision to redirect the article was made because the topic had no independent sources as all the published journals came from commentators who were affiliated with the topic of Socionomics. In this case, the notability of topic was disputed, in which case so too was the article title. --Gavin Collins (talk|contribs) 13:41, 17 February 2010 (UTC)

That's for notability; but once something is notable, aren't all sources equal as far as recognizability of the name is concerned? I'm thinking of a hypothetical situation where an organization is much discussed in non-English sources, but the only English translations of its name are on its own website and in one other source - does that mean the website is discounted? A kind of anti-self-identification rule? --Kotniski (talk) 15:03, 17 February 2010 (UTC)

Notability's relationship with article titles is similar to the Chicken or the egg causality dilemma; if an article topic is notable, then its likely that its title is recognised by reliable secondary sources, and vice-versa. If a topic is notable, but its title is not recognised, then you have a situation which could give rise to a content fork because reliable secondary sources are better than self-indentification from the primary source. --Gavin Collins (talk|contribs) 15:46, 17 February 2010 (UTC)

There's an example of this problem at Dunning–Kruger effect: The phenomenon that Dunning and Kruger identified has received a lot of attention, but they didn't name their discovery after themselves -- in fact, they didn't really name it anything during the initial burst of media attention, which is not an uncommon thing in that field. An editor perpetually insists that the now-common eponym is a neologism and original research, because Wikipedia used the (trivially predictable) eponym before it was adopted widely. I don't, however, think that there's any risk of a content fork arising as a result. WhatamIdoing (talk) 18:20, 17 February 2010 (UTC)

I am not sure this is entirely correct and I have posted my concens about this issue on the article's talk page[2]. There is a problem with such editorial disuputes: it is impossible to prove that a topic is neologism, in the same way that it is impossible to prove that a topic is not notable.
My view is that the article Dunning–Kruger effect is thinly sourced, and the fact that is being discussed in isolation from almost identical topics such as Groupthink or Illusory superiority is not helping. Perhaps if it goes to mediation, these issues will be resolved to everyone satisfaction.
I am not saying that my esteemed colleague WhatamIdoing is wrong in his assessment, but I think this is a case where there are insufficient sources to demonstrate that the term "Dunning–Kruger effect" is a title used by the world at large for the phenomena it describes. In a case like this, it is important not to compartmentalise a topic if there are reliable secondary sources that suggest that alternative, more widely cited terms are in use and are more generally accepted. This is another reason why self-indentification should be given due consideration, but not undue consideration. --Gavin Collins (talk|contribs) 09:51, 18 February 2010 (UTC)

Examples

I think that the elephant in the room is the debate over the naming of the Roman Catholic Church from which this recent debate has its seeds. In the case of the Catholic Church how does one assess which secondary sources are independant? Another two article titles, the Republic of Ireland and the city of Derry, over which usage is split and both article would probably end up at their alternative names if self-identification was given more weight. BTW the Republic of Ireland is under an Arbcom decision and a poll on the name to be used. -- PBS (talk) 11:00, 17 February 2010 (UTC)

With respect to Derry, the long-standing solution there is to call the county Londonderry and the city Derry. That solution keeps both peace and balance and has a grounding in the history of the place names in Ireland (that I won't go into here), common names, "self-identification" and manitaning NPOV across multiple interrelated related topics.

The result of the Ireland/Republic of Ireland poll was to maintain the status quo (island at Ireland, state at Republic of Ireland) for another two years. The questions around that also demonstrate that there are many factors to be considered besides simply a legal name for an entity when considering disambiguation, primary topics, and the appropriate teamtment of topics. -- RA (talk) 11:57, 17 February 2010 (UTC)

Its a solution, but it needs to backed up by some form of external validation that can withstand a formal challenge. Although I participated in the Ireland/Republic of Ireland poll, I think polls are a wholly inappropriate method of determining names, because there will be a lingering suspicion that voting will have been conducted along partisan lines, rather than in accordance with Wikipedia policy.--Gavin Collins (talk|contribs) 12:32, 17 February 2010 (UTC)

A poll is not the best way to resolve things - neither is trowing tables and chair at each other, which is all that was happening. Both of the choices (Ireland/Republic of Ireland or Derry/Londonderry) are externally valid and supportable by policy and guidelines. It is not only the passions involved, but more cricually the mutliplicity of considerations when it comes to topics such as these, that make naming these articles less than clear-cut.

The answer is illusive because the question is difficult. That is the nature of these topics. The solution to that of course is not to simply ammend policy and guidelines to restrict the question to only one answer.

With particular reference to Derry/Londonderry, that solution has survived challenge and consensus quite amicably for six years now. It would be unfortunate for someone on this talk page to "legistlate" poorly (whilst of course believing that they were doing right) only to cause havoc elsewhere. Questions like these need to be resolved appropriately with respect to the subject matter being discussed. This page cannot make those calls because it does not know the subject matter. -- RA (talk) 13:10, 17 February 2010 (UTC)

This leads me back to something I said earlier... each article is unique... so the solution to one title dispute will not work if applied to another title dispute (and may even make the dispute worse).

Perhaps the solution to this endless debate is to simply say "If an examination of the sources does not indicate a clear common name, editors should title the article through consensus" and leave it at that. In other words, perhaps we need to intentionally remain silent on what things should or should not be considered when attempting to reach a consensus. As far as this relates to "self-identification", don't explicitly say that "self-identification" should be considered, but don't explicitly say it shouldn't be considerd either. Let those who are trying to reach a consensus determine what should and should not be discussed. Blueboar (talk) 14:43, 17 February 2010 (UTC)

That's what we say at the moment. I'm quite happy with it, although I think even that is too much - commonness of name is only one of the factors we consider (as the first section of the policy makes clear), so it's wrong to say "first try to find the commonest name, and if that fails consider other things".--Kotniski (talk) 15:10, 17 February 2010 (UTC)

If you phrase it that way, I agree with you... that phrasing implies that a simple majority wins over other important criteria, and that is not what this policy intends. If, on the other hand you phrase it as: "first see if there is an obvious common name that is used by reliable english language secondary sources, if so use that, and if not consider other things" it becomes much clearer, and in line with what we intend. I do think that WP:COMMONNAME carries more weight than the other criteria (it's why we list it first)... but the useage must be fairly obvious. Blueboar (talk) 15:28, 17 February 2010 (UTC)

I disagree that a simple majority wins over other important criteria, and the lead example in this context is article on Persian language being used in preference to the more commonly used term Farsi on the strength of coverage from reliable secondary sources. If any article title dispute would go mediation, it would not be settled on the basis of a majority, but on the strength of external sourcing. How else could title disputes ever be settled if editors knew that external validation could be always be overuled by a majority? Once we abondon sourcing in favour of arbitary majority, we will be creating a set of rules based on subjective importance, rather than on verifiable evidence. --Gavin Collins (talk|contribs) 16:00, 17 February 2010 (UTC)

Apart from the fact that the Farsi example is flawed, since it involves English v foreign language usage, very few of the naming disputes under discussion can be solved in the way Gavin Collins suggests. What he seems to be stating is that if there isn't a clear common name, find a clear common-name in reliable sources. In most cases this is going round in circles. We are talking about areas where there is no clear referenced name. There also seems to be misunderstanding regarding NPOV. Addressing a person by its chosen name is not POV. This has been made clear in WP policy for at least the past five years. I can direct people to the references if they wish. It is also clear that Wikipedia editors do use self-identification as a criteria in naming. Those who now wish to dictate a doctrinaire ban on this are the ones trying to impose new policy from above. In the Derry - Londonderry dispute "Derry" was chosen for the town partly because that was the name the town had chosen to self-identify by, while "Londonderry" is used for the county because that is the name the County has decided itself to use. We do the same with places like Canton, China, in the policy on the Church of Jesus Christ of Latter Day Saints, and as a factor in the resolution of the Macedonia dispute. So the simple thing policy should continue to do is follow what actually happens on WP, and state that self-identification is a factor to be considered now matter how certain individuals dislike it. Xandar 23:30, 17 February 2010 (UTC)

Yes, please do provide references to back your assertion that "Addressing a person by its chosen name is not POV. This has been made clear in WP policy for at least the past five years." Not that it is particularly relevant, since consensus can change. Hesperian 00:18, 18 February 2010 (UTC)

"Addressing a person by its chosen name is not POV". Perhaps... but we are not addressing a person... We are determining a title for an article. And titling an article because a person or group prefers that name can be considered POV. Blueboar (talk) 00:41, 18 February 2010 (UTC)

I agree with you on the POV issue, but the title of an article refers to the article's subject. Maybe that's what Xandar meant, in which case the real question is whether or not his "WP policy for at least the past five years" references back up his claim. —Codrdan (talk) 01:02, 18 February 2010 (UTC)

The main naming convention dealing with the issue of self-identification was the Wikipedia:Naming Conflict page, stable from 2005 to 2009. It was "merged" into this page last year, which is how this argument has arisen. The page was deferred to from this page and others for rationale and specifics of the common name policy, and clearly stresses the NPOV nature of self-identifying names. Wikipedia Manual of Style also advocates the use of self-identifying names. This specific naming convention is based on self-identification. All of this is long standing, and unchallenged until recently. Xandar 02:04, 18 February 2010 (UTC)

You source for the assertion that "Addressing a person by its chosen name is not POV. This has been made clear in WP policy for at least the past five years" is a guideline that didn't exist five years ago, and doesn't state that addressing a person by its chosen name is not POV? <shakes head> You know, we actually make progress here when you are not around. Hesperian 03:11, 18 February 2010 (UTC)

Xandar, what would you have done with The Artist Formerly Known As Prince ten years ago? The sources contrived not to use a name, in a sort of "he who must not be named" way, and his self-identification was useless (images can't be put in a URL). WhatamIdoing (talk) 02:06, 18 February 2010 (UTC)

I think WhatamIdoing example perfectly illustrates the trap that self-identicaton can lead to - sometimes the primary source gets it wrong. Earlier I made the point that where self-identification has changed, this policy needs to say that editors should examine the secondary sources on the subject to ensure the correct term is being used, and to confirm that the name change has been accepted by the world at large. In short, multiple secondary sources are always better than one primary source.
In answer to Xandar's comments about the Persian language controversy, I think dismising it as a foreign language issue is missing the point: Persian and Farsi share the same root, so they are essentially the same term. There may be a shift in usage, by which I mean Farsi may well become the accepted term in the future, but there needs to be evidence to show that this shift has occured.
In order to stop disputes going around in circles, we must evaluate the evidence from secondary sources in order to make a reasoned judgement, as is done in mediation cases. We can write rules based on evidence (as oposed to personal opinion), and I think Xandar has shown us the way forward when he writes "if there isn't a clear common name, find a clear common-name in reliable sources". --Gavin Collins (talk|contribs) 10:18, 18 February 2010 (UTC)

Isn't that redundant? After all, the only way to know that a name is common in the first place is through use in reliable sources... so it sounds like you are essentially saying "if there isn't a clear common name (as shown through use in reliable sources), find a clear common-name in reliable sources." Or are you making a distinction that I am not seeing? Blueboar (talk) 15:00, 18 February 2010 (UTC)

I meant to say that if there isn't a clear common name (as shown through use by the entity itself and in reliable sources), find a clear common-name in reliable sources. Give the primary source due consideration, but not undue consideration. --Gavin Collins (talk|contribs) 15:25, 18 February 2010 (UTC)

Ah... interesting... I had not considered the primary/secondary nature of the sources before. I will have to think about that. Blueboar (talk) 15:43, 18 February 2010 (UTC)

Thinking on this a bit more... the primary/secondary issue is moot... as we are assuming that there are reliable secondary sources that support the self-identification as well as reliable secondary sources that support any other name we are considering. So it really comes down to a choice between secondary sources. Blueboar (talk) 19:22, 18 February 2010 (UTC)

That's right; we have to examine our sources more closely, and possibly accept some and reject others. Since we couldn't get a clear-cut answer to the question "What name do reliable sources use?", we dig deeper and ask "What name is favoured by the most reputable, the most reliably neutral, the most accurate sources?" Often when there is a naming dispute, the more reputable sources will favour the self-identification name, in which case the self-identification name should be used here. But in cases where the more reputable sources do not use the self-identification name, nor should we. Hesperian 23:39, 18 February 2010 (UTC)

The trouble with that is a) identifying the "more reliable sources" and then b) estimating the usage in same sources once these had been selected. These processes are liable to produce more argument not less, and may still not resolve into a clear advantage for any name. This is not what editors actually do anyhow. The use of self-identification as a factor in name choices is quite widespread, and there is no reason for a dramatic change of policy as would be required to rule this out. It should therefore continue to be documented in this policy page. A major change of policy, such as forbidding use of self-identification in naming would require a very wide-spread consultation across Wikipedia, since the vast majority of editors most concerned, the ones writing the relevant articles are not here. Xandar 01:46, 19 February 2010 (UTC)

"These processes are liable to produce more argument not less" So be it. As I said above, I'm opposed to a policy that makes it really easy to come to the wrong decision.

I dispute Xandar's claim that this would be a major policy change, and also that claim that editors use self-identification; but I'm happy to see an RfC on this, so long as it isn't written by Xandar, whose previous RfCs were hideously biased. Hesperian 01:59, 19 February 2010 (UTC)

A few things may be helpful for the RfC:

The language Xandar quotes is a few sentences of a much longer guideline, most of which had nothing to do with self-identification.
It was written by a single editor, ChrisO, who happens to have strong feelings about the name of the Republic of Macedonia.
It was accepted as a guideline after very little discussion, none of which addressed self-identification, either way. Most of the guideline was uncontroversial guidance on other subjects, and most of its content is still, in substance, on this page and the naming conventions.
The key phrase of the self-identification section of the guideline is quoted only in these references in Wikipedia. Some of them are supporting some article title now rejected (Kyiv, Myanmar, Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel, Macedonia - for the Republic), objecting to some phrasing now accepted (mentioning Malvinas in the first sentence of Falkland Islands, using Bollywood - I still don't understand what alternative was being pushed), or otherwise against consensus. The others are all Xandar's personal noise about the phrase being consensus, and what everybody does. Septentrionalis PMAnderson 06:13, 19 February 2010 (UTC)

In answer to Xandar, your argument that "self-identification as a factor in name choices is quite widespread" has some merit in it, because usually secondary sources take their lead from the entity itself (i.e. the primary source), and so secondary sources usually confirm that self-identification is correct. In this case, primary and secondary sources are congruent, but in such a case, there is little likelyhood of there being an editorial dispute.
However, this is not support giving undue weight to self-indentification where there is a dispute. Focusing on the primary source as a means of selecting an article's title is not supported by any of Wikipedia's content policies, all of which are primarily based on what relaible secondary sources say about a particular topic. In short, over-reliance on self-identification is just not supported by any content policy. --Gavin Collins (talk|contribs) 12:05, 19 February 2010 (UTC)

I'm not sure what "over-reliance" on self-identification means. What is being argued for here is that it be listed as a factor for consideration. As I have said WP:NamingConflict, the Manual of Style and other policy guidance have been strongly in support of self-identification as a factor in naming. This goes beyond what has been suggested as a compromise solution here. Articles like Republic of China also go beyond the suggested wording since self-identification is clearly the primary factor used by editors OVER and above, common name. Attempting to ban the use of self-identification even as a factor in naming decisions would be a very new and major change of policy, and as far as I can see not one driven by either actual WP practice or even by any pressing problems arising out of the use of self-identification in article naming. Xandar 01:41, 21 February 2010 (UTC)

"Over-reliance" on self-identification in this context means that the primary source (the entity itself) is being used as the source of information, rather than secondary sources. Since WP:V says that "articles should be based upon reliable, third-party published sources with a reputation for fact-checking and accuracy", it seems to be that giving priority to primary sources goes against this prinicple.
A good example of this is the Hong Kong Special Administrative Region of the People's Republic of China which is the offical name, but there is no dispute about its article name which is widely used in secondary sources, i.e. Hong Kong. Unlike the Hong Kong, the names recognised by reliable secondary source for those of Republic of China and the Republic of Ireland may be in transition (and hence are disputed). At some point, a consensus may emerge, but it will be based on secondary, rather than primary sources.
The main problem with relying on self-identification as I see it is that the primary source is not indepedent. In this context, independence means that a source is free from pressures associated with a strong connection to the subject matter (such as, but not limited to family relationships, close political affiliation, business dealings or other benefical interest) that may compromise, or can reasonably be expected to compromise, the sources reputation for reliablity.
In short, article titles should be should be based upon reliable, third-party published sources with a reputation for fact-checking, accuracy and independence. --Gavin Collins (talk|contribs) 10:15, 21 February 2010 (UTC)

That is probably the most coherant argument that I have read so far. Thank you Gavin. Blueboar (talk) 16:17, 21 February 2010 (UTC)

A summary of the debate with respect to Ireland/Republic of Ireland:

With repsect to Republic of Ireland, the crucial aspect to that debate was the existance of another entity of the same name, Ireland. If that entity did not exist or if the two coincided then there the article on the state would simply be at Ireland. Since another entity called Ireland does exist and the two "Irelands" do no co-incide, some form of dabbing is be required. Hence the problem.

In deciding how to dab, it was considered by some that Ireland formed a primary topic. If so, so Ireland would not be a straight dab page and the article on the state would have to be dabbed somehow. The question then was how to dab the article on the state: Ireland (state) (not Ireland (country) for specific reasons) or Republic of Ireland. These were the two most viable options and reflected guidelines on how to dab: either using a distinguishing word or using an alternative common name. It was considered by some that dabbing by an alternative common name (Republic of Ireland) would be superior to dabbing by a partentesided word. That choice sarcrificed the "self-identification" name in favour of not using a parenthesised word.

(The situation is analogous with China/People's Republic of China/Taiwan/Republic of China. In some ways it is also analogous to Korea/South Korea/North Korea.)

A defining consideration throughout the debate was the many participants so vehimently resented the state being at any title other than it's official name. Many editors considered that a slight against the state (and consequently the nation). The counter argument was that there was no such slight and that Republic of Ireland is commonly used to distinguish one entity from the other in secondary sources. (That particular aspect of the debate was coloured by some "anti-ROI" editors arguing that the term was a slight against the nation from historically weak positions e.g. that the term Republic of Ireland was a "British impostion" whereas in fact it was introduced by the Irish government in declaring the state to be a republic in 1949, something that was resented by the British.)

The result of the vote of 234 ballots was 43% in favour of the state being at Ireland (state) and 54% in favour of the state being at Republic of Ireland. 7 ballots were "wasted" (a concept of the voting system used). -- RA (talk) 10:12, 22 February 2010 (UTC)

I'm interested in the history of the Chinese article titles - it seems not to be an entirely analogous situation, since the People's Republic is obviously the primary topic for the term "China", and that's the kind of name we would ordinarily be using for a country (we don't normally include "Republic of" etc. unless there's a real ambiguity involved). Was there some political/nationalist pressure against doing the obvious thing here? Is it worth revisiting?--Kotniski (talk) 10:48, 22 February 2010 (UTC)

"Was there some political/nationalist pressure against doing the obvious thing here?" Not substantially. The "political/nationalist pressure" more typically came from "anti-ROI" participants as noted above.

Bear in mind that like Scotland and England, Ireland is a traditional country of the region and is still a hugely culturally and practically significant. History of Ireland, Culture of Ireland, Geography of Ireland, People of Ireland, Sport in Ireland, Christianity in Ireland, basically everything except Government of Ireland, points to the island being the primary topic. Pick up any general book on "Ireland" and it will almost invariably be about the island not the state. Instances where "Ireland" refers to the state only are a very limited subset (e.g. Economy of Ireland).

"Is it worth revisiting?" The matter is the subject of an ArbCom ruling the decision to be binding for two years. This is being clarified. -- RA (talk) 11:09, 22 February 2010 (UTC)

Sorry, I wasn't clear - I was asking about China, not about Ireland.--Kotniski (talk) 11:17, 22 February 2010 (UTC)

Oh ... no idea about China, but it looks the same. -- RA (talk) 11:49, 22 February 2010 (UTC)

Both Chinese governments agree that "China" is a larger entity which includes the de facto territory of both states (they both claim to be its legitimate government). Furthermore, the history of China begins several centuries BC; the history of both states spans less than a century. Therefore distinguishing China from People's Republic of China and from Republic of China represnets a real difference, as well as minimizing the max of unhappiness. Septentrionalis PMAnderson 05:46, 24 February 2010 (UTC)

The thing about history applies to many countries I would have thought (they have a history which covers entities often very different from the modern state), but we still name the articles with the common short name of the country - as far as I can see the only reason we treat China differently is the claims of the respective governments, which ought to be ignored for naming purposes. Still, if it makes people happy... --Kotniski (talk) 08:55, 24 February 2010 (UTC)

But China is almost unique in that two governments both claim to be the country. We should not agree with only one of them, and we cannot agree with both. What do we do about Cyprus, I wonder? Septentrionalis PMAnderson 03:40, 25 February 2010 (UTC)

We should follow usage, right? No-one says "China" meaning the Republic of China. What governments claim should be irrelevant to us.--Kotniski (talk) 08:44, 25 February 2010 (UTC)

No, but many people will, when the question comes up, say that Taiwan is in China, that Taipei is a Chinese city. It would be a Beijing POV to say that they're in the People's Republic. Septentrionalis PMAnderson 12:46, 25 February 2010 (UTC)

No-one says "China" meaning the Republic of China. Actually, there are a lot of sources that do so. It may not be the majority, but it is a significant minority. And older sources will do so almost exclusively (heck, back in the 50s and 60s "China" was represented in the UN by the ROC... so any UN documents of the time are referring to Tiwan when they say "China".) That said... I think PMA gets it right in the China case... Disambiguation is needed... Save China for geographic and historical articles... use the full governmental names when referring to the political entities. Blueboar (talk) 22:19, 25 February 2010 (UTC)

> two governments both claim to be the country

Do they really say that? And are you sure they mean it literally? Most governments just claim to be the sole legitimate rulers or representatives of a country. The name "China" is usually associated with the geographical region, the people, and thousands of years of history that include a whole series of governments. Obviously people use it as a shorthand for the ROC in modern political contexts, but the scope of WP is wider than that. —Codrdan (talk) 23:11, 25 February 2010 (UTC)

The fact remains that the reason we use Republic of China rather than the name in the sources - Taiwan - is that ROC is the self-identifying name of the entity. The same applies to Daluts rather than Untouchables. Once again I would draw people's attention to the fact that THIS discussion is about when there is no single clear name in the sources. GavinCollins seems to believe that no use of self-identification should be considered by editors. That, as I have said, would be a major policy change. Xandar 00:54, 26 February 2010 (UTC)

No, the reason we have articles on Taiwan and the Republic of China is that they are different concepts; even geographically, they do not comprise the same territory. Exclude the de jure claims of the RoC, which extend further into India than the PRC does or has; still the Republic rules Quemoy, Matsu, and the Pescadores, not rationally part of an article on the island of Taiwan. Septentrionalis PMAnderson 01:09, 26 February 2010 (UTC)

Not at all. The state we refer to as Republic of China is known in the sources and throughout the world chiefly as Taiwan, and secondarily as Chinese Taipei or Formosa. Xandar 01:38, 4 March 2010 (UTC)

There is already an article at Taiwan. Clearly "Taiwan" is a legitimate title for both an article about an island and an article about a state. Given that it is not possible to give two articles the same title, at least one must be disambiguated, either with a disambiguation term or by use of a more precise alternative title. Your assertion that "We use RoC because we value self-identification" is logically equivalent to the assertion that "if we didn't value self-identification then we wouldn't have used RoC." So tell me then, what titles would we have used if we didn't value self-identification? Personally, I'm inclined to think we would have used precisely the titles we are currently using. Hesperian 01:53, 4 March 2010 (UTC)

General Principles

Here is a suggested wording of the approach which might be useful to take forward:

Article titles should be should be based upon reliable, third-party published sources with a reputation for fact-checking, accuracy and independence.

The reasons for doing so originate from Wikipedia's content policies, which I have adopted to support this. NB: italics have been used where I have amended the original policy wording. I am putting these ideas out now so we can kick these ideas about.

WP:OR: To demonstrate that you are not presenting original research, you must be able to cite reliable sources that support the article title directly, and that directly support the article title as it is presented. An article title generally should not rely on unclear or inconsistent passages nor on passing comments.
WP:SYNTH: Do not combine material from multiple sources to create an article title not explicitly supported by any of the sources. If one reliable source says A, and another reliable source says B, do not join A and B together to create an article title C that is not mentioned by either of the sources. This would be a synthesis of published material to create a new article title, which is original research.
WP:NPOV: A careful selection of reliable sources is also critical for producing article titles with a neutral point of view. An article title should clearly describe, represent, and characterize all the disputes about a topic, but should not endorse any particular point of view.
WP:V: An article title that is challenged or likely to be challenged must be attributed to a reliable, published source supported by an inline citation in the body of the article itself. The source should be cited clearly and precisely, with page numbers where appropriate, and must clearly support the article title as presented. If no reliable third-party sources can be found in support of an article title , Wikipedia should not have an article on it.

Forgive the choice of wording, but these ideas approximate the approach that I think would have to be adopted in a mediation case. Can anyone suggest improvements or a better approach? --Gavin Collins (talk|contribs) 09:56, 4 March 2010 (UTC)

Yep, this is the right idea. I would say "should be based upon usage in reliable, third-party sources", to make it clear that we are not expecting to find a source that says "its most common name is X". I also think the second half is unnecessarily wordy; it might suffice to simply write "should be based upon usage in reliable sources". At the very least, I would eliminate "published", and replace "independence" with "neutrality". I'm not sure whether "accuracy" should be included here: verifiability not truth, and all that guff. Hesperian 10:23, 4 March 2010 (UTC)

This suggestion conflicts with current practice, this policy and other policy considerations. See for example 'Dismissal of U.S. attorneys controversy.' in the section about descriptive titles. Dmcq (talk) 15:11, 4 March 2010 (UTC)

I think a lot of the problem here is that "current practice" conflicts with itself... there is no single "current practice". Different articles have used different practices. That's what makes working this policy so contentious. To resolve this, we need to focus on what best practice is... to guide us in explaining what future practice should be. We may not agree as to what is best practice, but at least thinking in those terms will put us in the right direction. Blueboar (talk) 19:01, 12 March 2010 (UTC)