Wikipedia:Administrators' noticeboard

Closer (slakr)

Heya... Just fyi, I invited voorts to provide a better version of the close so that I could improve upon mine (i.e., basically a version of {{sofixit}}) and they refused. Repeatedly. And told me I failed the community. So feel free to take up the mantle here. Thanks a million, and cheers =) --slakr^\ talk / 09:46, 3 May 2026 (UTC)

Now that I've had some time to review some of the input, I'd like to address some of voorts's concerns.

1. More concerningly, slakr's close adopts the RfC framing of "persistant LLM usage", notwithstanding that many editors (including those in support of the proposal) pointed out that our PAGs currently only bar LLM misuse.

I'm now certain that several people did not read the closing text before opening this or commenting on it. The closing text was: "There seems to be consensus for including persistent unsanctioned LLM usage in the blocking policy's list of common rationales for blocks ("WHYBLOCK")." I altered the emphasis here, because it's misquoted a few times, and several people below likely didn't understand the word "unsanctioned" in-context and/or ignored it entirely. At the very least, voorts clearly ignored it when opening this.

2. The close, however, straight up ignores the counterargument to that point that was raised in the RfC: that this proposal cannot and will not change the need for editors to report LLM misuse at ANI/other noticeboards.

I didn't address it because that's not what our policies and guidelines say. Most directly and importantly, the blocking policy does not require noticeboard reports before blocks can happen, however it does strongly recommend warnings (I address this issue in the close, too). Some of the related pages that also don't have a requirement to report to noticeboards? Wikipedia:Writing articles with large language models doesn't have it. Wikipedia:Editing_policy#Artificial_intelligence_additions doesn't have it. Wikipedia:Image_use_policy#AI-generated_images doesn't have it. Wikipedia:Talk_page_guidelines#LLM-generated doesn't have it. Wikipedia:Biographies_of_living_persons#Images doesn't have it. Wikipedia:LLM-assisted translation? Nope. So where is it?

Nothing in the blocking policy pertaining to this requires blockable behavior to first have a noticeboard discussion. We normally do have a discussion if there's some element of uncertainty, of course, but when someone's running roughshod over the wiki and ignoring warnings, it's common to block them after a few warnings and with no noticeboard discussion. So no, I didn't mention it; to have mentioned it would have hallucinated policy out of thin air—like an LLM would do, poetically. Could I have mentioned it? Yes, but as I pointed out when voorts dropped onto my talk page originally: there just isn't enough room or desire for myself to counter-point every single point's counter-point that every single editor wants me to add to a close. We'd basically just be duplicating the entire discussion. Closes are also for brevity.

3. slakr also stated that editors in support of the proposal argued that "It would describe policy as is already being carried out". However, the opposite is true: editors in support argued that admins don't have enough discretion to block editors for LLM misuse. In oppositon to that, the argument was made that admins already have broad discretion to block disruptive editors and that we don't need to codify admin authority in that manner.

Let's digest what actually happened (to the best of my memory):

• While yes, some believed admins didn't already have the ability to block for disruptive/abusive/problematic/disallowed LLM use, they still felt that listing it on the main blocking policy page would accomplish this. That lack of awareness reinforces the argument. If they didn't know, yeah, I can see why they think it needs to be mentioned, since new editors would likely make the same mistake. I think some people might have even made this latter argument, too, but I forget.
• On top of this, some on both sides agreed that admins have the ability to block via the disruptive-editing route; that means they acknowledge that it's blockable. Adding it to the list of common rationales isn't precluded by this assertion.
• Some believed that admins should specifically be able to block not just as "Disruptive Editing" but for disruptive/abusive/problematic/disallowed LLM use specifically. That further refines that they support the ability to block for that.
• Some people believed that admins don't or shouldn't have the ability block disruptive/abusive/problematic/disallowed LLM use. That does not align with existing policy or practice; there was not consensus to overturn existing policy/practice.
• Some people believed that admins should block all LLM use. This does not align with existing policy/practice; there was not consensus to overturn existing policy/practice; this would not align with a "common rationale for blocks."
• Some of those opposing the proposal also believed that admins already have the ability to block disruptive/abusive/problematic/disallowed LLM use. That further acknowledges that it's blockable. Adding it to the list of common rationales isn't precluded by this assertion.
• Combining everything together, I felt there was sufficient consensus that if people believed persistent, unsanctioned LLM use is a common rationale for blocking, then there's not a clear policy-based conflict with listing it as a common rationale for blocking.

Now, did I run numbers to determine whether it truly is "common?" No. I reflected that the participants in the discussion felt that it was. That's the job.

Hope that helps clear things up.

--slakr^\ talk / 18:18, 7 May 2026 (UTC)

RE point 2, you are misunderstanding my objection. It has nothing to do with whether warnings are required. Your close said that editors in support of the change argued "The volunteer burden of mitigating LLM problems is asymmetric and increasing." They did argue that. Specifically, they asserted that this change would somehow reduce the volunteer burden. In response, myself and others argued that they were providing zero evidence that this change would affect "the volunteer burden" and that editors would still need to find LLM misuse and report it to admins.

RE point 3, you're now admitting that your summary of arguments in support is inaccurate and that things weren't as simple as your close makes them seem. voorts (talk/contributions) 18:51, 7 May 2026 (UTC)

"The volunteer burden of mitigating LLM problems is asymmetric and increasing." They did argue that. Specifically, they asserted that this change would somehow reduce the volunteer burden....zero evidence that this change would affect "the volunteer burden".

It's not my job as a closer to predict the future. However, it is important, as a closer, that I mention common arguments and because it wasn't prima facie illogical or otherwise at odds with existing policy—regardless of whether someone had numbers and projections to back it up. Furthermore, I could imagine that elevating it to the main BLOCK page might plausibly help mitigate a problem of lack-of-visibility. As such, I feel that ignoring it entirely would have incorrectly reflected what happened in the discussion. Could the counter-arguments also have been mentioned? Yep. I don't think they need to be in the interests of brevity and not materially affecting the close.

you're now admitting that your summary of arguments in support is inaccurate and that things weren't as simple as your close makes them seem.

1. I don't know how you're arriving at this conclusion, but you're more than welcome to show me the better way to close and summarize this beast. You won't, as we've established, but you're still more than welcome.

2. All closes have to simplify things—especially big ones with extensive discussion. No close is as simple as the underlying discussion, but an ideal close, in my opinion, does simplify things as much as possible so that anyone trying to reference it gets a good-enough overview of what happened.

--slakr^\ talk / 19:21, 7 May 2026 (UTC)

The job of a closer is to evaluate the strength of arguments, not just list them out. I didn't ask you to ignore the argument of point 2. My objection is that you appear not to have even considered the counterarguments when you made the close. Your response here makes it seem like your close was a supervote. voorts (talk/contributions) 19:36, 7 May 2026 (UTC)

Excessively evaluating the strength of the arguments—as I feel you're proposing here—instead of by-default reporting the arguments given is the very definition of a supervote. It's way more important to evaluate the strength of arguments in situations where policies and guidelines are being interpreted. For example, if someone says "Delete - Article should be deleted because it violates BLP," yet you know it clearly doesn't violate BLP as the community typically interprets it, yeah, you squash the argument. When people are proposing changes to a policy itself, you go with what people want so long as it doesn't conflict with (or introduce conflicts with) existing policy. I very clearly stated why I feel this change doesn't conflict with existing policy here, and I very clearly stated in the close why some elements of what people wanted do conflict with existing policy and further steps to consider. --slakr^\ talk / 19:48, 7 May 2026 (UTC)

If 50 editors had said "this change will eliminate misuse of LLMs on wiki", would you have credited that argument? voorts (talk/contributions) 19:54, 7 May 2026 (UTC)

While I can't really speak to hypotheticals that haven't happened, sure, I'll bite. I likely would have reported that there was significant belief (presumably among established editors) that the change will eliminate misuse of LLMs on-wiki—even if I personally felt that would be unlikely. --slakr^\ talk / 20:04, 7 May 2026 (UTC)

Non-participants (LLM usage)

• Endorse Based on a quick count of the bolded !votes, seems like ~80% of participants answered Yes/Support to the RfC question of Should the list of reasons to block be expanded to include "Persistent usage of large language models"?. Sure, consensus isn't decided based on a straight vote count, but an 80% support level for a widely-attended RfC (131 people commented in that discussion apparently) is very difficult to ignore. The closer couldn't have closed the RfC any other way. Some1 (talk) 03:59, 14 May 2026 (UTC)

  What percentage of those !votes actually answered the RfC question as opposed to something else entirely? voorts (talk/contributions) 12:30, 14 May 2026 (UTC)

Participants (LLM usage)

• I too was surprised by this close, for pretty much the reasons given in the initial post here - it seems to be closer to a a vote count than to a proper analysis of the actual substance. I stress I'm not saying it was just a vote count, just that the number of bolded supports seems to have been given more weight than they should given that many of those supports seems to have misunderstood the actual issues under discussion (which is very sadly all too common when it comes to AI/LLMs on Wikipedia - the whole thing is a lot more complicated and nuanced than the simple "AI bad" many seem to desire it to be). Thryduulf (talk) 22:39, 2 May 2026 (UTC)

  @Slakr Voorts was absolutely correct to refuse to do your job as closer for you, just as I would have been, and just as every other involved editor would have been. If you don't understand that then you really shouldn't be closing RFCs. Thryduulf (talk) 11:50, 3 May 2026 (UTC)

  Seconding Thryduulf here, if there are issues with the closure, it shouldn't be the responsibility of other editors (and especially not of involved editors!) to write a new one. Chaotic Enby (talk · contribs) 11:58, 3 May 2026 (UTC)

  Perhaps it's my inexperience with the close-review process, but isn't it inevitably the responsibility of another editor to write a new close if people decide the old one has issues? Like, say it's overturned; someone else is going to eventually have to close it with a new close. --slakr^\ talk / 14:42, 3 May 2026 (UTC)

  If the close is overturned, a new editor will evaluate the discussion with a fresh set of eyes and draft a new close. The close review process does not require providing a new draft close up front. It also shouldn't require that because the purpose of a close review is to determine whether the closer reasonably evaluated consensus, not how the discussion should be closed. voorts (talk/contributions) 15:46, 3 May 2026 (UTC)

  I see. I guess, then, the main question is whether or not I closed it unreasonably. I'd love to get your version of what reasonable is, but clearly that's not going to happen, so I dunno what else I can do to help. *shrug* Best wishes. --slakr^\ talk / 21:21, 3 May 2026 (UTC)

• The requester's argument relies on the fact that slakr's close adopts the RfC framing of "persistant LLM usage", while the exact words that slakr used were actually persistent unsanctioned LLM usage, which indeed corresponds much better to LLM misuse. Chaotic Enby (talk · contribs) 01:47, 3 May 2026 (UTC)

  That's fair. voorts (talk/contributions) 01:49, 3 May 2026 (UTC)

• "The closer didn't take my argument into consideration" isn't a valid objection to an RFC closure (it's a common misunderstanding of WP:CLOSE and WP:NOTAVOTE.) While closers do weigh the strength of arguments, they only do so by assessing what the community believes and discarding arguments that are not grounded in policy. When both sides present reasonable policy-based arguments, it is not the closer's role to select which argument is stronger - that is up to the community. Per CLOSE, If the discussion shows that some people think one policy is controlling, and some another, the closer is expected to close by judging which view has the larger number of responsible Wikipedians supporting it, not personally select which is the better policy. The rationale for this review request amounts to saying "but I think our arguments were right and theirs were wrong", which is an unworkable standard to challenge an RFC (every experienced editor who weighs in on an RFC is obviously going to think their arguments are correct and that everything the other side says is wrong or is easily rebutted.) No matter how strong you may think your arguments were and no matter how much you personally treasure your rebuttals, they failed to convince the community; nor were the arguments on the other side groundless to the extent that a closer could reasonably disregard them. As a result, there was only one way this RFC could have been closed. --Aquillion (talk) 02:23, 3 May 2026 (UTC)

  I would like to see the closure overturned and reopened for a new close. Obviously I would prefer if the close then comes out the other way, but I am not advocating that that occur through a close review. This is an consequential change to policy and the close should document how the closer weighed the relevant arguments, not list a few arguments that were made on reach side and then write a paragraph describing next steps. voorts (talk/contributions) 02:39, 3 May 2026 (UTC)

Again, this misunderstands what closes are supposed to be. A closer's job is not to weigh every argument and decide which one the closer feels is strongest; it's to evaluate what argument the community thinks is strongest. This means that an argument that the community clearly disregarded can be likewise disregarded by the closer - it makes no difference how strong you feel it is if you failed to convince people. Closers has some room to disregard arguments that are not grounded in policy, or to disregard opinions with no rationale that don't really seem to have actually considered the argument they are endorsing, but in situations where there is a clear numerical majority favoring a policy-compliant position from editors using rationales that are reasonable under policy, the closer does not have the option to close it in any other way. What you are requesting (ie. a "new close" because you feel the closer didn't address an argument you personally found compelling, where you hope the closer closes it the other way based on their personal analysis of the arguments) is against policy. Even if you somehow got it re-closed, and even if the new closer disregarded WP:CLOSE and inappropriately closed it the other way, it would immediately be brought back here and overturned again, because there is simply no viable reading of consensus that would support the outcome you want. RFCs aren't simple numerical votes, but they do represent the actual consensus of the community, rather than just the opinion of the closer; if you want to overturn an RFC with, from a quick eyeball, a roughly 2:1 majority in one direction, you need a better argument for how the closer misread consensus than "you didn't address my counterargument!" It's not the closer's job to address every individual point raised in the RFC; it's their job to determine whether those points convinced the community. If you want to get it overturned you have to actually argue that the community sided with you, and that your argument actually convinced them; these things are just clearly not true. --Aquillion (talk) 02:17, 5 May 2026 (UTC)

This issue is that many of the supporting comments are not grounded in policy and/or grounded in a misunderstanding of policy. For example LLM use is not prohibited by policy, but rather LLM misuse. Misuse is rather overbroadly defined, but not to the extent that it covers all use, indeed some uses are explicitly permitted. Thryduulf (talk) 02:38, 5 May 2026 (UTC)

@Thryduulf: quick question: which part of the closing text do you feel claims/implies that consensus prohibits all use of LLMs? --slakr^\ talk / 23:20, 5 May 2026 (UTC)

It doesn't do so explicitly, but I can't see how to arrive at the consensus it does without implicitly endorsing the comments that do as being in accordance with prior community consensus. Thryduulf (talk) 00:06, 6 May 2026 (UTC)

It's probably worth noting that although the close said There seems to be consensus for including persistent unsanctioned LLM usage in the blocking policy's list of common rationales for blocks, WP:DBLOCK was actually updated with simply persistent usage of large language models. So I guess the memo was not heeded to the tee, though I suppose anyone can correct this. Wh1pla5h99 (talk) 00:10, 6 May 2026 (UTC)

Yikes; I intentionally crafted that closing text to preclude the implication that it was any LLM usage, as that would, like others have mentioned (and I mention in my close), create a confusing conflict with existing guidelines. --slakr^\ talk / 00:21, 6 May 2026 (UTC)

• What an egregious decision. I will never block anyone for productive LLM usage, and I will contest any such block that I encounter. The idea of accepting a proposal that "any administrator would be able to block on sight for LLM usage" means that it's now prohibited to use an LLM to modify an existing userspace page. Consider User:Nyttend/sandbox/without, which has a lot of links, including some with non-ASCII characters; I could have used an LLM to mark the lines that have these characters, but no longer. Perhaps I should first get some minor sanction for it (e.g. a level-1 warning), so I won't be blockable for doing this without sanctions. Nyttend (talk) 07:47, 3 May 2026 (UTC)

I guess I should go ahead and formally oppose this close. Now, the blocking policy is at odds with other policies and guidelines. Consider Wikipedia:User pages, which permits the use of user subpages for experimentation and plans for future Wikipedia work — an LLM is a potentially useful tool that can be used in my user sandbox, but now if I do it persistently, I can get blocked. See also WP:OWN, which acknowledges that Wikipedia offers wide latitude to users to manage their user space as they see fit — just dumping in this prohibition creates a conflict with this policy, too. And it's just bizarre that the new addition says that I could be blocked for the uses I mentioned if I'm unsanctioned, but it implies that I can't be blocked under this addition if I've been sanctioned for it. Nyttend (talk) 07:47, 3 May 2026 (UTC)

PS, note that the close also conflicts with Wikipedia:Writing articles with large language models/RfC, which acknowledges that More specialized, constructive use cases do exist, from using them as research assistants to citation-formatting tools, and this guideline does not aim to restrict any of these use cases. If I do those things persistently, now I can be blocked for that, since nobody's going to sanction me for doing those things. Nyttend (talk) 07:50, 3 May 2026 (UTC)

• Hi! "Unsanctioned" is a synonym of "disallowed", it doesn't mean that you can only be blocked if you haven't been sanctioned yet. Chaotic Enby (talk · contribs) 07:59, 3 May 2026 (UTC)

  Sanction is one of the classic autoantonyms. ~2026-86111-3 (talk) 11:01, 3 May 2026 (UTC)

Discussion (LLM usage)

• Is LLM usage ever a reason to block per se, regardless of whether any other PAGs have been violated? WP:LLM (which presumably we are being pointed towards to find out what constitutes unsanctioned LLM usage) suggests yes, but paradoxically says that this is because Text generated by large language models (LLMs) often violates several of Wikipedia's core content policies. I think this is the crux of the issue; we shouldn't ban people from using tools because they are faulty in the hands of the incompetent, we should instead ban the incompetent users per WP:CIR. Slop is slop. Whether it comes from a human or an LLM. Just bin the slop, and the users producing it. That is already quite clear in policies and I don't think admins are confused about it. Both WP:LLM and this close RfC result seem to be technophobic overreaches. All that matters is whether the published material is acceptable or not, according to the usual policies. If "aversion to writing style" is the motivation, which it seems to be for some, then that should be made explicit, and discussed. As it stands we don't ban people for writing style AFAIK. Wh1pla5h99 (talk) 23:44, 2 May 2026 (UTC)

  Is LLM usage ever a reason to block per se, regardless of whether any other PAGs have been violated? No. WP:LLM is only about article content. Specifically, it says: "the use of LLMs to generate or rewrite article content is prohibited, except for basic copyedits and translation of material from other language Wikipedias as outlined below." voorts (talk/contributions) 23:50, 2 May 2026 (UTC)

  WP:LLMDISRUPT and WP:LLMTALK state that repeatedly using unreviewed LLM-generated content is disruptive. voorts (talk/contributions) 23:50, 2 May 2026 (UTC)

  Right but even if the close had said Persistent usage of large language models to generate or rewrite article content I would make the same argument. We shouldn't police what tools people are using, just the outcome of their tool use. Wh1pla5h99 (talk) 23:54, 2 May 2026 (UTC)

  This discussion is about whether consensus was correctly evaluated at the RfC, not to relitigate the merits of the RfC. voorts (talk/contributions) 00:17, 3 May 2026 (UTC)

  I posted this before I saw @Kowal2701's comment below. voorts (talk/contributions) 00:18, 3 May 2026 (UTC)

  You said the close adopts the RfC framing of "persistent LLM usage" but it actually says persistent unsanctioned LLM usage, which seems to mean LLM misuse. Are you hoping for a different outcome in the close, or just proper acknowledgement of the different arguments given? Wh1pla5h99 (talk) 00:32, 3 May 2026 (UTC)

  I'm hoping for a closer to actually evaluate all of the arguments in the discussion and properly determine consensus. voorts (talk/contributions) 01:05, 3 May 2026 (UTC)

  Both WP:LLM and this close seem to be technophobic overreaches The reasoning behind WP:LLM is that more often than not, editors contributing to articles using LLM output are not properly reviewing such content. Thus, the community has decided that editors can only use LLMs in articles for basic copyediting of their own writing and translation. voorts (talk/contributions) 23:53, 2 May 2026 (UTC)

  Solving a problem that doesn't exist. Admins have full license to block for disruptive editing of any kind. Wh1pla5h99 (talk) 23:56, 2 May 2026 (UTC)

  Close reviews are not for relitigating the question (or discussing something vaguely related), they're for assessing whether the close was correct Kowal2701 (talk, contribs) 00:09, 3 May 2026 (UTC)

Closer (Beland)

The decision about word choice here was made by following a clear pattern of reliable sources not using the phrase "persecution" when they are speaking in an objective voice, reinforced by numerical superiority of participants.

The last paragraph in this close does not explain how the decision was made, but rather is just a personal aside to people who are about to write a strongly worded complaint about the outcome on spurious grounds because it was emotionally unsatisfying. (This is not to say there can't be legitimate complaints.) It should not be read as setting a precedent or binding on anyone, and people are free to disagree and have a different philosophy of neutrality.

I agree with the people writing complaints that neutrality includes respecting the due weight of reliable sources, and confronting readers with the facts, even if it makes them uncomfortable. As far as I can tell, we are doing that in this case. When deciding whether to be mad about the outcome, I ask well-meaning advocates to consider the actual consequences - using neutral language may mean more readers with what we would consider incorrect and harmful views are actually exposed to the facts. And I personally think that is and should be part of the purpose of the neutrality policy - generating trust in Wikipedia by not using unnecessarily biased language. There are two reasons we don't say "state-sanctioned murder" instead of "capital punishment" or "death penalty", even though some would argue that the first term is accurate. 1.) It's not what the vast majority of reliable sources say in objective voice, and 2.) among the terms that are equally accurate, we use editorial discretion to try to pick those that readers on any given side will not perceive as biased. It's fair to disagree over what that means in practice for any given case, as did the participants in this RFC. -- Beland (talk) 18:30, 4 May 2026 (UTC)

Non-participants (Trump)

While I don't disagree with the consensus found by the closer, I strongly disagree with the method, and am worried about the precedent that it might set regarding our interpretation of NPOV. The exact sentences in the close I am having issues with are:

But as some participants pointed out, a significant portion of our readers think that these policies are helping trans people get over some sort of recoverable delusion or social contagion. Much as I think that point of view is hurtful and ignorant and dangerous, I do want those people to keep reading Wikipedia and trust its content and learn more facts about Donald Trump and trans people.

This is not a way to write an encyclopedia. NPOV is about fairly assessing and representing reliable sources, and a bias in favor of what will please the most readers is still a bias, and will not make Wikipedia more trustful in any way. Chaotic Enby (talk · contribs) 12:58, 4 May 2026 (UTC)

Agree with Chaotic Enby - that phrasing makes me uncomfortable and I've been trying to figure out why & how to put it into words. It's a little like POV cosplaying as NPOV; there is an element of conspiracy theory when it comes to the Trump administration, so saying that we've based our decision on "tricking" them into reading Wikipedia (I know it's not tricking them, but it could easily be seen that way by the distrustful) isn't something I feel we should do.

Almost every large website bases their content on what will get the biggest audience, I stay here because that's not what Wikipedia is about. Our decisions should be based on real vs. not real, true vs. not true, provable vs. unprovable - follow where the evidence (reliable sources combined with our policies) lead us. Clear, open and transparent decisions for all to see, with no preference for any belief or background. Blue-Sonnet 13:34, 4 May 2026 (UTC)

Also agree with the above. The consensus was found correctly but it takes some reading to see what's actually being factored in. I think most of us agree that something being believed by lots of Trumpists, or any other partisan group, is not a valid metric in an RfC even if the belief were true, and I don't believe that Beland was considering this as an actual factor in the close. In his own words, these were a few thoughts before I get a slew of complaints that this outcome is a crime against the dignity and human rights of trans people. My suggestion would be that in contentious RfCs, closers make it clearer what portion of the close is a summary of the votes and what portion is commentary about the close's context, implementation, and implications. Thebiguglyalien (talk) 14:02, 4 May 2026 (UTC)

Time for the usual comment, but it applies I stress that I am not accusing Beland of transphobia here (they made their stance quite clear) but the comment But as some participants pointed out, a significant portion of our readers think that these policies are helping trans people get over some sort of recoverable delusion or social contagion replace the word "trans" with "gay" or "black". Now how does it look? Here at enwiki we are really bad at making excuses for transphobes. They need to be in the same bucket as racists, misogynists and homophobes and have no place here. Black Kite (talk) 14:30, 4 May 2026 (UTC)

Oh, and this should have been closed as no consensus because a very large number of the "No" !votes are completely evidence-free. Black Kite (talk) 14:30, 4 May 2026 (UTC)

If you replaced it with "gay" or "black" it would not be true that a "significant portion" of readers believe it (or, in the latter case, it wouldn't make any sense). Here's an interesting question - say it was the 50s and a good chunk of people still believed homosexuality was a choice or something that people could "get over", would it be ok to disregard their opinion completely when figuring out how to express something "neutrally"? I take it that you would say yes, but can you see how this might raise eyebrows in a project that brands itself as neutral. Wh1pla5h99 (talk) 14:47, 4 May 2026 (UTC)

You'd be surprised what percentage of people don't believe that sexual orientation is not a choice. But my point was a semantic one - by writing things like this we are making excuses for transphobes. ArbCom managed to detoxify the area a little by banning some of the worst offenders, but Wikipedia still treats it as somehow "different" from the other behaviours that we deal with above, and we need to make it clear that it isn't. Black Kite (talk) 14:52, 4 May 2026 (UTC)

Meh. Just my view but let people have different opinions. No need for an inquisition any time someone makes a claim that could be construed as making excuses for transphobes. Wh1pla5h99 (talk) 15:01, 4 May 2026 (UTC)

Thank you for proving my point. Black Kite (talk) 15:03, 4 May 2026 (UTC)

This is a deflection. Alexandraaaacs1989 (talk) 15:15, 4 May 2026 (UTC)

Trans people are not an opinion. I think your above statement is extremely hurtful. It essentially reduces people's humanity to a debatable subject. In the same way it is disgusting to spread the "opinion" that, idk, the shoah was good (or the "belief" it did not happenfor that matter) transphobia is, in my view, not just a "view" or "opinion". Beland did not do this, but I just want to have said that I totally reject "let people have different opinions" when it comes to dehumanising "opinions" or ones that, indeed, contribute to persecution. Slomo666 (talk) 18:54, 5 May 2026 (UTC)

I'm not arguing that people should be welcome to spew hate on here, but that if people are (thoughtfully, in good faith, and with a willingness to listen to other people) expressing their opinion, then we should not simply kick them out for fear of any offense such opinions may cause. Again, this is just my view. Censoring anything that has the potential to offend any person or group would make this quite an unproductive endeavor. For example, I'm sure ample offense is caused to religious people when they are met with skepticism about their beliefs, yet such skepticism is indispensable to this project. In any case, this is a tangent that I don't think we need to hash out right here and now, and as such I will preemptively agree to disagree and move on. Wh1pla5h99 (talk) 19:19, 5 May 2026 (UTC)

I don't agree with your characterisation, but if it had been as simple as you frame it, then I would gladly agree with you. My issue is how, in my view, you make light of this. I will gladly acccept your invitation to Agree to disagree. Thanks for your response anyways, happy editing, Slomo666 (talk) 19:25, 5 May 2026 (UTC)

I am gay and have some conservative religious in-laws who did start out believing homosexuality was a choice, with horrific you're-banned-from-the-house, I-don't-want-to-know-your-name homophobia, not in the 1950s but the 1990s. Through repeated and sometimes unwanted exposure to LGBT people like me, they have gained a lot of empathy and are now warmly welcoming and try to use respectful language. I don't know how differently they speak when I'm not around, but I definitely change my language when I'm around them to avoid unnecessarily upsetting them, including not swearing.

I do try to give some grace for the sake of continuing a conversation. One thing people really hate is to be constantly corrected in the language they use even when they mean no disrespect, and that can derail a conversation on the grounds that political correctness or wokeness has gone overboard and is obnoxious, and even allies make that complaint sometimes. So I might hear "women" used to mean "cisgender women" and not stop to correct it but I would reply with the latter as a reminder of and exposure to the idea that trans women are considered women. I don't make excuses for their homophobia and transphobia, and when they referred to sexual orientation as a "lifestyle choice" or someone's gender identity as "pretending they are a woman" I try to correct their perceptions by relaying my personal experience or those of LGBT+ friends and others. I think this approach is working. I had a good conversation with my mother-in-law about whether she ever "decided" she was straight, and learned some interesting things about how sexual attraction works differently for different people. And another about how gender seems to be an indelible feature of some brains, exemplified by the large percentage of intersex people who do not identify with the gender they were raised as, even if they appear physically gender-conforming because of genital surgery as infants. This is not the sort of information she gets at church or from her conservative friends and family. We still have to make the case about racial language; their religion only started allowing black people to become full members in 1978. ("I'm pretty sure you can't refer to African-American areas as 'the hood'." "But that's what my students called it!" "You are white and they are black." "That's racist.") You're right that racial acceptance is a lot farther along than trans acceptance, but not meeting people where they are isn't necessarily going to advance either cause.

I can't think of any LGBT-related terminology that I change, but I do change how I refer to the President and the Trump Administration. At home I might say "the regime" or simply "Voldemort", but that would cause immediate eye-rolling and derail any political conversation I was having with my conservative relatives (which do happen...sometimes in a civil fashion; I remember a good one about how their family would benefit from universal health insurance). To my shock, despite a lot of almost involuntary Obama- and Biden-bashing, one of them started referring to Trump as "the crazy man" in his second term.

I know there are folks who would prefer to be more confrontational and use language only from the "right" side and that's fine and not always unhealthy for our society, but it seems like an approach better suited to a street protest than the calm and academic pages of an avowedly neutral encyclopedia. -- Beland (talk) 19:38, 4 May 2026 (UTC)

Thanks for that, and of course I agree with all of it apart from the last line. I don't think we actually need to be "confrontational" to maintain neutrality, but my point was that by even giving any wiggle room to people who want to introduce their biases into an encyclopedia because "a lot of people think like this" we are actually reducing our neutrality. I know a lot of people (mostly the same people that I've just mentioned) roll their eyes at the phrase "safe spaces" but we do need to consider how trans people view editing Wikipedia if we are not strict on calling out transphobia, or even language that normalises transphobia. I don't believe you did the latter, incidentally, but it doesn't take much for people to seize upon it and say "well, yes I said that, but..." Black Kite (talk) 19:51, 4 May 2026 (UTC)

I know there are folks who would prefer to be more confrontational and use language only from the "right" side and that's fine and not always unhealthy for our society, but it seems like an approach better suited to a street protest than the calm and academic pages of an avowedly neutral encyclopedia. I'm sorry, but you're arguing against a strawman here. The criticism against the close doesn't come from people wanting to use the language from the "right" side (in fact, relatively few here disagree with the result of the close), but that neutrality doesn't come from finding the lowest common denominator between readers. Neutrality isn't writing what will please the most people, it is checking what the sources actually say. A populist bias is still a bias. Chaotic Enby (talk · contribs) 06:42, 5 May 2026 (UTC)

Following what the sources actually say in objective voice is in fact what determined the outcome of this RFC. The last paragraph in my close message was just pointing out that a byproduct of doing that is transphobic people and Trump supporters are now more likely to read more about this topic, and hoping that thought helps being on the losing side feel less bad. Is it that this thought instead makes some people angry?

I will say that in other cases, though, checking the reaction from partisans is quite helpful. For example, a few years ago a dispute arose over whether Puerto Rico and other territories are part of the United States. Is it an integral part? A colony? A possession? A nation? Occupied territory? Reliable geographical and technical and legal sources are kind of all over the place, as are political opinions. Which aspect of belonging should we base our language on, potentially elevating it in importance? For the purposes of the lede to the article United States, editors negotiated the phrase "asserts sovereignty over", which is acceptable for people from all of these points of view. I would generally think that objectionable wording is worth changing, as long as it doesn't create another problem, like inaccuracy compared to sources or objections from a different point of view or the gratuitous use of unfamiliar words. -- Beland (talk) 18:02, 5 May 2026 (UTC)

What Whonting wrote below inspires a bit of clarity: one aspect of neutrality is accurately describing facts or opinions following due weight in reliable sources. If people stop reading because we do that well, that's their problem. Another aspect is when we are describing facts in wikivoice, choosing language that sounds objective to people from a diversity of viewpoints, to avoid the perception we are speaking from a particular POV. If people stop reading because we do that badly, it's our fault. -- Beland (talk) 18:13, 5 May 2026 (UTC)

Another aspect is when we are describing facts in wikivoice, choosing language that sounds objective to people from a diversity of viewpoints The problem is that these are two ultimately incompatible goals: either we stick to wikivoice and reflect the words use by the sources, or we choose to be more cautious about using wikivoice and tone down our wording to sound objective to more people. To take a (slightly) more extreme example, what about young Earth creationists, who represent 40% of the US population? Should we say that T. rex is "claimed by evolutionary scientists" to have lived 68 to 66 million years ago, to avoid angering them? There is a difference of degree, but, I believe, not really of principle. Chaotic Enby (talk · contribs) 18:59, 5 May 2026 (UTC)

I agree we should not say that. The first aspect always takes precedence. It's certainly worthwhile to tell our readers how those dates have been determined and be honest about the uncertainty. Our article Tyrannosaurus already does the latter; I just added a link to Geochronology to do the former. I've had sidewalk arguments with young-Earth creationists, and one of the things they'll tell you is that the dinosaur skeletons seen in museums are fake. For the sake of intellectual honesty with all our readers but also credibility with this demographic, it behooves Wikipedia to note which skeletons were really found as-is, which are pieced together from multiple finds, and which are partial or complete copies or recreations. I just tweaked some captions on the T. Rex article to do that; the article Sue (dinosaur) does a good job of explaining which parts are a reconstruction and why.

As far as word choice specifically, I think a good example is our article Empirical evidence for the spherical shape of Earth. It's an interesting collection of facts for lots of different audiences, but one demographic that relies on it is people who have flat earth beliefs or uncertainty, and others having discussions with them. Using the word "empirical" instead of "scientific" is a deliberate choice. For this demographic, it can avoid a whole debate on what science is, what are legitimate scientific methods, and a whole bunch of conspiracy theories about how scientists are a cabal protecting the secret of the true shape of the planet. Those are all beside the point of the facts the article is trying to convey - that when people go out and observe these things, these are the results they get. Some flat earthers value that type of information, and a few actually go out and take observations of their own. -- Beland (talk) 22:24, 5 May 2026 (UTC)

Chaotic Enby, if you have a read of WP:OUTRAGE and to a degree WP:OPPONENT which are found at Wikipedia:Neutral point of view#Balancing different views, does this complicate your view? Whonting (talk) 14:30, 4 May 2026 (UTC)

WP:OUTRAGE clearly says that we have no responsibility for how neutral content we write is generally perceived, which is exactly my point: we shouldn't bend what sources say to be more palatable to the average reader. Chaotic Enby (talk · contribs) 14:33, 4 May 2026 (UTC)

But if content is perceived as neutral, then whatever bending has to take place has already happened. -- Beland (talk) 18:04, 5 May 2026 (UTC)

I meant this more as an expression of "one of our guiding principles is we care about not alienating readers, so when possible try to meet them where they are". In that section, it is argued that "writing neutrally consistently" will best accomplish that, but doing so is not at odds with other manifestations of writing neutrally (e.g., as a rule not putting normative claims in wikivoice). Whonting (talk) 00:51, 6 May 2026 (UTC)

• One more in support of the close conclusion except for those sentences. Kudos to Beland otherwise, this was a tough, charged RFC, and we shouldn't hold those few sentences against them, much. --GRuban (talk) 15:37, 4 May 2026 (UTC)
• I think there is a core of a good point in the controversial part of Beland's close, even if I don't entirely agree with how Beland expressed it. There is some value to the appearance of neutrality, which is a distinct concept from NPOV (which, really, shouldn't have the word "neutral" in it at all, since what we actually mean is consensus POV, which is often non-neutral). The appearance of neutrality is never a strong enough consideration to justify asserting falsehood as truth or truth as disputed; we can't say "the Earth is probably round but might be flat" just to make the flat-Earthers happy. But when it comes to how we characterize politically loaded topics, I do think there is something to be said for considering whether the reader is going to say "Huh, I sure don't feel like I'm reading a neutral Wikipedia article right now. I feel like someone's trying to convince me of something, when I come here to get the facts and reach my own conclusions."
  That's one of a few reasons that these RfCs over word choice often bring out the worst in Wikipedians, because they conflate the substantive question with the stylistic one. Beland's close errs inasmuch as it implies (maybe unintentionally) that there is any reasonable debate on the substantive question, which is whether Donald Trump persecutes trans people; even a diehard Trump supporter would acknowledge all the necessary facts to conclude "yes" and might only disagree with the word choice. But the word choice, the stylistic question of whether "persecute" is the right word for Wikipedia to use, is not insignificant. Adolf Hitler's article does not use the word "persecute" or "persecution" in the encyclopedia's voice at any point. Joseph Stalin's does so just once, in passive voice, regarding persecution of clerics during the Soviet cultural revolution. Even our article on Nero, the historical figure I'd most associate with the word "persecution" specifically (as opposed to genocide, ethnic cleansing, etc.), only uses the word in quotes or paraphrases. As such, I think it's reasonable to say that, on a stylistic level, describing a politician's actions as persecution generally exceeds the level of emotion readers expect in Wikipedia articles, and that this is, again, not on its own enough to motivate a particular close, but not an invalid concern either, and one worth mentioning in an RfC clearly heavily effected by the emotive weight of that word.
  I'd really like to see a communal push toward less emotive word choice in political articles—one coming from a place of caring about good article-writing, and not just the objections of the losing side in various ideological disputes, which for good reason the community will always take with a shaker of salt. I think a few high-profile "wins" for one side or another in such articles have emboldened people. But the thing that gets us the trust of our readers is their sense that we give the facts for them to make their own opinions. For me, as someone who's been more a reader than an editor the past few months, it jumps out to me the moment that it feels like a Wikipedia article is trying to lead me to a conclusion. Above all else it insults my intelligence: If the conclusion is true (which, again, it is here), then I should be able to reach that on my own based on the facts of the matter. And if I actually were too stupid or too biased to reach the correct conclusion, there's no reason to think that telling me the correct opinion will sway me. I think that's often how people with strong political views think other people think, but in my experience the exact opposite is often true.
  Anyways, endorse. -- Tamzin^{[cetacean needed]} (they|xe|🤷) 19:56, 5 May 2026 (UTC)

  Did people agree somewhere that rather than actually being neutral, Wikipedia was merely supposed to have the appearance of neutrality? This is news to me as someone who has not been involved for very long. Wh1pla5h99 (talk) 20:08, 5 May 2026 (UTC)

  I'm very confused where you get "merely" from this when I've gone out of my way to stress that the appearance of neutrality is a separate concern from (the thing we somewhat misleadingly call) neutrality. Wikipedia articles should reflect the consensus point of view and, when it does not conflict with that objective, should be written in a way that readers will perceive as neutral. Style is always secondary to substance. There absolutely are cases where an emotive word is necessary, for instance in calling a genocide a genocide if that's the consensus term. But I haven't seen any good argument that this is such a case. -- Tamzin^{[cetacean needed]} (they|xe|🤷) 20:20, 5 May 2026 (UTC)

  I think you are wrong that Wikipedia articles should reflect the consensus point of view. That is not one of the purposes of striving for consensus among editors. We should try to reach a consensus about how to express something neutrally, not how to express something in a way that we personally or collectively like, or is closest to our collective perspective. Forgive me if I'm wrong about what you mean by "consensus POV", but I don't see what else it could mean, if it is distinct from NPOV, which you reject. Wh1pla5h99 (talk) 20:28, 5 May 2026 (UTC)

  Consensus POV as in the POV that is the consensus among reliable sources. We call it "neutral point of view" for kind of arcane historical reasons; I don't reject the NPOV policy, I'm just pointing out the obvious fact that that's not what "neutrality" means when most people say it. It's good that Wikipedia is not neutral in the way most people mean. A truly neutral Wikipedia wouldn't be able to say that the Earth is round or that the Holocaust happened. -- Tamzin^{[cetacean needed]} (they|xe|🤷) 20:43, 5 May 2026 (UTC)

  I think that when there is consensus about facts in reliable sources then we present that consensus-fact as a fact rather than a POV. I don't think this extends to opinions (per WP:OPINION), which should always be attributed. In other words, there are some things that are not matter-of-fact based (they are unprovable, immaterial, emotional etc.), and we should not present those things in the same way we present historic or scientific facts. Wh1pla5h99 (talk) 20:53, 5 May 2026 (UTC)

  As far as I am aware, we do not, but we also don't attribute literally everything that is subjective, and I think we'd spin out of control if we had to. Slomo666 (talk) 21:11, 5 May 2026 (UTC)

  Well said, Tazmin. I think combining that with Wh1pla5h99's earlier formulation which I also find clarifying, I'd say Wikipedia aims to present external-consensus facts in language generally perceived as neutral, to take a neutral position when there is no external consensus on the facts, and to describe opinions but for the sake of neutrality not assert them in wikivoice, and our idea of a neutral balance opinions is to convey due weight rather than create a false balance. Is that a better nutshell for WP:NPOV than the current one: "Articles must not take sides, but should explain the sides, fairly and without editorial bias. This applies to both what you say and how you say it." ? Maybe it's too complicated, or maybe the concept is just inherently complex, or maybe I'm too easily distracted. -- Beland (talk) 22:50, 5 May 2026 (UTC)

  That would be a great start for an explanatory essay, I believe! Glad to see that we could all come to an agreement! Chaotic Enby (talk · contribs) 00:09, 6 May 2026 (UTC)

  I absolutely agree with the core goal of avoiding loaded language (which we, in a way, already do through Wikipedia:Words to watch). Having it be a separate concept altogether from NPOV is essential, as this is a separate question from evaluating source consensus, and we don't want one to bleed into the other. I really like the way you worded this as appearance of neutrality, and would be happy with anyone making it an essay, although I can already foresee the "Wikipedia openly admits they want to hide their non-neutrality!!!" criticism from some people. Chaotic Enby (talk · contribs) 20:09, 5 May 2026 (UTC)

  I don't know if I would call WP: LABEL which is a sub section of words that may introduce bias a separate concept from NPOV. It seems it is a guideline to make sure we are in line with NPOV. --Kyohyi (talk) 20:25, 5 May 2026 (UTC)

  @Tamzin, do you think as a general rule Wikipedia should not be putting normative claims in wikivoice? Persecute for instance has connotations of unfairness: is it appropriate for wikivoice to say things are fair or unfair? Whonting (talk) 00:50, 6 May 2026 (UTC)

• No comment about the rest of the close, but those final sentences are not anything I could sign on with. I find the "cosplaying as NPOV" criticism to be appropriate. Furthermore, it seems to me rather naive to expect that anyone who is truly upset with Wikipedia, or animated by a personal transphobic conviction, to be mollified by our tweaking word choices. People who want to be upset will find things to be upset about. People who cash in by making others upset will find things for their audience to be upset about. You can try to avoid a verbal minefield, but people acting in bad faith will just lie about what you say. It was true for creationists on Usenet 30 years ago, it was true for antivaxers on the blogosphere 20 years ago, and it's true for transphobes now. Stepwise Continuous Dysfunction (talk) 20:02, 5 May 2026 (UTC)

  Well said. Slomo666 (talk) 21:13, 5 May 2026 (UTC)

• Endorse the close. It is clear what Beland's opinion is, and it is quite separate from the summary of the discussion and the conclusion. The conclusion is based on the summary, which is based on the discussion. The review requested just suggest a disagreement with the consensus outcome. No need to change anything about the close: on Talk:Donald_Trump#RfC:_Transgender_persecution?. Graeme Bartlett (talk) 00:00, 6 May 2026 (UTC)
• Endorse per Tamzin and others above me. Jessintime (talk) 00:12, 6 May 2026 (UTC)
• Endorse largely per Tamzin and the concurring comments. I too might have worded things a bit differently, but IMO Beland substantively got it right and deserve credit for their handling of a high voltage discussion. -Ad Orientem (talk) 14:28, 7 May 2026 (UTC)
• Endorse close, IMO wording used in the last paragraph wasn't ideal (or "bad" per se) but it was closed correctly & I don't have any other concerns. Blue-Sonnet 01:35, 8 May 2026 (UTC)

• Endorse, with caveats. Now, had I participated in this !vote, I doubt I could have lodged a straight support or oppose. Because, personally, I think it should be possible to thread the needle between using more muted/conservative variations on language summarizing mainstream views of the character and motivation of the social suppression at work here, while at the same time heavily leveraging the more express language about persecution through attributed statements, representing the WP:PROPORTION honestly enough so as to emphasize that this is a broadly accepted and well received interpretation of the behaviour of the subject, his administration and its political allies. This would allow us to preserve editorial credibility in wikivoice, without risking problematic sanitizing of the underlying topic.
  Now, is this approach consistent with what consensus adopted in the discussion, or how Beland summarized that consensus? Well, certainly there were some !votes which echo some of the tent-poles of my position, and you could argue that the close is not directly inconsistent with a "less contentious label, except where attributed" approach. But I do think that, for all its length, Beland's precise wording in that close did fail to address some of the nuance I would have liked to have seen. However, fidelity with what I think is the obvious approach to the problem under policy, after the discussion has run, is not the relevant test for whether or not Beland's close was a reasonable distillation of consensus. The overall feedback was to apply the precautionary principle to to summary verbiage in the forward-facing content. That was the reality with which Beland had to contend. Did they also stamp a bit of their own view on the close? Yes, unfortunately, I think they did--to a degree where I think they need to be more careful about that the next time they decide to close such a heavily disputed discussion. But ultimately, the close was also consistent with the more heavily-represented views.
  And personally, my view is that when we are talking about an interpretation of existing policy/broader community consensus upon which reasonable minds may disagree, then the closer inherits more responsibility to adhere more strictly towards the more numerically advantaged position. That is to say, this is pretty far from a case where the "no" !votes were so obviously and completely out of step with longterm/community level consensus that Beland could have eschewed the popular view, even if they wanted. The view they summarized was both the slightly more well-represented one and also a perfectly reasonable interpretation of what the relevant PAGs require in this situation, though hardly the only one.
  To those who feel there is a threat of whitewashing here: I am sympathetic, and rest assured, none of this goes one whit towards changing my own view of the reprehensibility of the behaviour and policies in question, the animus behind it all, or the profound human and societal costs. But 1) sometimes the words chosen by the good humanitarian and the good encyclopedian are not going to be the same, and 2) to the extent anyone might want to argue they should be in this case, that argument did not prevail in the consensus, and the close only reflects that reality. SnowRise ^{let's rap} 08:05, 12 May 2026 (UTC)
• Endorse The close was fine. Sure, it could have been worded differently but most things could be expressed with more appropriate language. Johnuniq (talk) 08:56, 12 May 2026 (UTC)
• Endorse The closing statement would've been better and less controversial without the last paragraph, but if editors are allowed to give personal opinions like that as part of their RfC closure, then I guess it's fine. Some1 (talk) 04:07, 14 May 2026 (UTC)
• Endorse — The close is fine (e.g., on issues of interpreting wikivoice/NPOV policy), but the last paragraph wasn't a choice I would have made, because parts of it feel clearly contradictory to WP:CENSOR. Regardless of the personal beliefs of the readers, the critical point of NPOV is that we cover things, well, neutrally/objectively as based on sources. People trust us—I trust us—to be an unfeeling tome of knowledge independent of preconceived notions or sensibilities. --slakr^\ talk / 20:18, 15 May 2026 (UTC)

Participants (Trump)

• As a non-admin, I agree that Beland (whom I respect and in no way intend to attack by making critiques here) prioritized perception of neutrality over actual neutrality (by under-weighing expert opinion in sources in the discussion) based on the aforementioned quote and from subsequent WP:NPOV/N discussion comments. For example, he said WP:NPOV does not contemplate the political consequences of our content, only the perception of neutrality by readers. My impression is that a significant consideration in closing note's decision to use the wording "restrict" instead of "persecute" was how "restrict" is less likely to be perceived as biased (even if the term more accurately summarizes expert consensus than "restrict"). While I'm okay with RfCs not going my way, the reasoning behind the close didn't sit well with me either and I think a re-evaluation of the closing decision is due given the reasoning behind the original close. Alexandraaaacs1989 (talk) 12:32, 4 May 2026 (UTC)
• Not alienating readers was given in the close as a reason for having a neutrality policy, not the reason for the RfC result. Beland also wrote the sources presented do not support that claim. This word is used in opinion pieces or attributed to third parties, but rarely or never used in objective news voice. If you are taking issue with the idea that we should accommodate readers with different points of view on Wikipedia (hardly a scorching hot take) then a close review is probably not the place for that. Wh1pla5h99 (talk) 12:52, 4 May 2026 (UTC)
• The issue is that the non-alienation paragraph specifically advocates for us to downplay the language surrounding pseudoscience over concerns of perceived neutrality. This, as a precedent, would be disastrous not just for our coverage for the attempts of the Trump administration to exterminate trans people but also for things like our coverage of COVID-19, autism and vaccine safety, climate change and several other topics where there is a political argument that is out of touch with the science. Simonm223 (talk) 13:08, 4 May 2026 (UTC)

  The RfC had nothing to do with pseudoscience, and that's not at all the point he was making. He was offering thoughts on why being neutral is a good idea, and certainly not "disastrous" as you claim. Wh1pla5h99 (talk) 13:16, 4 May 2026 (UTC)

  The paragraph at issue explicitly references "social contagion" as a viewpoint to avoid alienating. That's pseudoscience. It was in the closing statement. This is the issue before us. Pseudoscience is the central reason for this closure challenge. Like maybe you missed the several times myself and other editors have said that the reason to review the close is more about that para than the no !vote. There's a very good reason why. Simonm223 (talk) 13:20, 4 May 2026 (UTC)

  People believe in unproven hypotheses all the time. I don't see that as a reason to censor those views on talk pages, pretend no one expressed them in RfC discussion etc. Wh1pla5h99 (talk) 13:35, 4 May 2026 (UTC)

  Are you advocating for treating pseudoscience as a legitimate viewpoint on Wikipedia? Alexandraaaacs1989 (talk) 13:44, 4 May 2026 (UTC)

  I'm pretty sure it applies to talk pages. I have no issue with people being "corrected". Wh1pla5h99 (talk) 13:55, 4 May 2026 (UTC)

  Woah your comment just metamorphosed. I'm in favour of people disagreeing without trying to have opposing views banished, if that's what you mean. Wh1pla5h99 (talk) 13:58, 4 May 2026 (UTC)

  I'm not trying to be mean, but I'm more than a little surprised by this. Pseudoscience is... well, fake by definition, so accepting an inherently contradictory position for the sake of extending inclusivity to those with opposing views seems extremely misguided if our goal is to write an accurate Wikipedia. And I'm not trying to censor you here but I don't consider this point to be debatable. Alexandraaaacs1989 (talk) 14:08, 4 May 2026 (UTC)

  I disagree with extending inclusivity to those with opposing views being extremely misguided, on NPOV grounds among many other reasons. For example if someone said "puberty blockers are definitely safe, fully reversible, and improve outcomes" I would certainly no try to have them silcenced on Wikipedia (or anywhere), though I may believe their ideas to be scientifically unfounded. And if people aren't trying to pass their views of as science there is really no reason to reach for the "pseudoscientific" hand wave. Its a disagreement. We don't have all the facts on everything. The world is a complicated place. Wh1pla5h99 (talk) 14:16, 4 May 2026 (UTC)

  Especially for medical issues (like your example claim about puberty blockers), we absolutely do want to give more weight to scientific results. We can discuss the political aspects separately, but political opinions should be separated from evidence-based findings. If someone wants to add a medical claim to an article based on a political op-ed, they shouldn't, and requiring quality sources is one of Wikipedia's key policies, not censorship. Chaotic Enby (talk · contribs) 14:30, 4 May 2026 (UTC)

  Of course. I was referring to what people can and can't express on talk pages; what ends up in an article is a separate issue. Wh1pla5h99 (talk) 14:38, 4 May 2026 (UTC)

  No one is really advocating for banning content on talk pages in this discussion. It's about the close, which was flawed because it gave weight to pseudoscientific arguments. Simple as that. Alexandraaaacs1989 (talk) 15:11, 4 May 2026 (UTC)

  Not really, it explained that the word was not neutral, and then made a case for neutrality being worthwhile. If you want to argue that neutrality is not worthwhile because many millions of people hold views that ought to be disregarded then fine, but at least be accurate about what the issue is. Wh1pla5h99 (talk) 15:26, 4 May 2026 (UTC)

  If you want to argue that neutrality is not worthwhile because many millions of people hold views that ought to be disregarded then fine This is the bandwagon logical fallacy. We do not write articles based on opinion polling - we write them based on RS, and RS do not promote pseudoscience. Wikipedia only uses RS, so it does not use pseudoscience. Please listen to what others here are saying: pseudoscience is never to be conferred credibility on Wikipedia. Alexandraaaacs1989 (talk) 15:49, 4 May 2026 (UTC)

  I'm not really seeing how calling a view hurtful and ignorant and dangerous is conferring credibility to it. By just acknowledging that it exists? Also this was said in a talk page not an article. Wh1pla5h99 (talk) 15:55, 4 May 2026 (UTC)

  Simonm223 spelled it out for you at the top of this thread. Please re-read carefully and try to fully understand the implications of the first sentence here. Alexandraaaacs1989 (talk) 16:05, 4 May 2026 (UTC)

  Yes, I understand that his and your view is that we should alienate people who hold such a viewpoint. I think we are going round in circles now. Wh1pla5h99 (talk) 16:13, 4 May 2026 (UTC)

  I'm not really seeing how calling a view hurtful and ignorant and dangerous is conferring credibility to it. Closing an RfC on the basis of seeking to avoid alienating pseudoscience confers credibility to pseudoscience in the process. That is why a group of editors here is arguing Beland's closing note was problematic. Please read what we have told you and think about it carefully. Alexandraaaacs1989 (talk) 16:49, 4 May 2026 (UTC)

  The point is that we shouldn't be thinking about that at all. Wikipedia isn't meant to be popular, it's meant to be accurate. (By "thinking", I mean openly stating it was a factor in the decision, which also risks legitimising it) Blue-Sonnet 16:56, 4 May 2026 (UTC)

  If you are arguing that the purpose of neutrality is not to have people of different beliefs read/trust the content of Wikipedia (which was Beland's framing), then fair enough. I would say it is one reason among many for a neutrality policy. Other reasons are that people would be excluded/discouraged from collaborating based on their views, that presenting opinions as facts is unencyclopedic and dimishes credibility, etc. The original NPOV policy statement gives reasons like avoidance of edit warring, and also that when it is clear to readers that we do not expect them to adopt any particular opinion, this is conducive to our readers' feeling free to make up their own minds for themselves, and thus to encourage in them intellectual independence. Wh1pla5h99 (talk) 17:23, 4 May 2026 (UTC)

  I guess my position is more that, by openly stating it in the decision, we're drawing attention to it and inadvertently putting us on one side or the other, depending on the reader and their perception/view of the subject - current US politics is such a hot-button topic that IMO it's best to just not go there and stick to objective & clear facts that everyone can see on the page in front of them.

  The more you add, the more can be misconstrued & twisted - especially if you say you want to aim the content at a specific audience. They'll ask "why are you doing that?

  What's in it for you? Why don't you want the other side? Why you you singling us/them out for special treatment? What's your agenda?"

  If we stick to policy, facts and neutrality - regardless of reader or subject - we have a much better chance of true objectivity. That way, the truth can speak for itself. Blue-Sonnet 18:20, 4 May 2026 (UTC)

• I basically agree with everyone in the non-participants section that the result of the close was fine (I'd rather the consensus ran the other way, but it clearly didn't), but the language of the close was not, especially the final paragraph. Loki (talk) 16:26, 4 May 2026 (UTC)

  Even up at the top I said I was more concerned about the last para of the review then the actual !vote that it was closed to so I'd say, as the opener of the challenge, I also agree with the non-participants. Like if it turned out that the close was overturned as a result of this I'd not say no to it but what I am really looking for, and why I opened the review, was that paragraph of the close note. Simonm223 (talk) 16:53, 4 May 2026 (UTC)

  I know I'm late to the discussion but I appreciate your efforts here.

  The "yes" side argued NPOV does not apply or favors "persecution" because lots of reliable sources use the word "persecution" in an equivalent way to wikivoice, considering this dispute settled or at least giving weight to use of that term. But looking at the context of how the word is used, the sources presented do not support that claim. This word is used in opinion pieces or attributed to third parties, but rarely or never used in objective news voice

  I don't recall a significant portion of editors stating NPOV "doesn't apply", at least not in the sense that this seems to imply, but that may only be my perception of it. With articles such as Persecution of homosexuals in Nazi Germany for example, we have the benefit of many years of hindsight, yet still so few experts, and the fact that the Nazi's lost the war didn't hurt either. It was not until the 1970s that gay Holocaust victims would start speaking out and receiving public acknowledgement. I'm also reminded of the Gay agenda and how the Anti-gender movement was criticized for encouraging discrimination, undermining human rights protections, and promoting misinformation and rhetoric against LGBTQ people. The notion of a pervasive and institutionalized "gender ideology" is all about creating the effect of a moral panic and conspiracy theory. We all know what makes this article so exceptional, and at least giving the appearance of best intentions, towards "making X great". I do not begrudge anyone here in particular, perhaps I'm even a bit jealous of those that do not bear such weight. As the Buddhists say the only permanence is impermanence. Cheers. DN (talk) 07:08, 14 May 2026 (UTC)

  I think that's important for those editors who have used this RfC to seek the erasure of the Trump regime's persecution of trans people from other articles to remember. This RfC represents a local consensus on a single page and I hope the closer will at least confirm that they did not expect it to be binding on the entire project. Simonm223 (talk) 10:07, 15 May 2026 (UTC)

  I think their answer to my question at least implies they do not consider their close to bind any other discussions. Slomo666 (talk) 12:41, 15 May 2026 (UTC)

  Yes, the close itself points out the inconsistency with the title of Persecution of transgender people under the second Trump administration and leaves "any resolution of the inconsistency to future discussion." Editors may consider the findings of this RFC to be persuasive with regard to other articles, or not. It's out of scope of the closed discussion to conclude that other discussions came to an incorrect conclusion; there may be additional evidence and reliable sources raised in other discussions which support using the term "persecution" in general or in specific contexts (such as discussing a controversy over the term). -- Beland (talk) 18:04, 15 May 2026 (UTC)

• The result of the close was fine, IMO, based on familiarity with how reliable sources address this subject. Anythingyouwant (talk) 19:53, 4 May 2026 (UTC)
• I've no problems with Beland's closure & therefore I endorse the closure. GoodDay (talk) 20:15, 4 May 2026 (UTC)

• The last paragraph honestly was such an extreme case of WP:FALSEBALANCE, it was horrifying. By the logic displayed there, neutrality means we have to find the middleground between whether covid vaccines and sick masks are good or not, or whether climate change is real; even when the exact anti-trans beliefs Beland sought to appeal to in stating But as some participants pointed out, a significant portion of our readers think that these policies are helping trans people get over some sort of recoverable delusion or social contagion. Much as I think that point of view is hurtful and ignorant and dangerous, I do want those people to keep reading Wikipedia, have already been declared FRINGE at FTN.Snokalok (talk) 05:31, 5 May 2026 (UTC)

  That discussion declared a specific organization fringe, which also subscribes to anti-vax ideas, etc. -- Beland (talk) 17:09, 5 May 2026 (UTC)

  Neutrality is about avoiding stating opinions in Wikivoice entirely, and sticking to facts. It is not about finding a conciliatory "middle ground" between opinions. Wh1pla5h99 (talk) 18:04, 5 May 2026 (UTC)

  Well put. It also means when stating facts, to choose our language carefully so as not to appear to be speaking from a specific opinionated point of view. -- Beland (talk) 18:07, 5 May 2026 (UTC)

• (Not sure if I should be in (non-)participants or here) I agree with many of the criticisms posted here. To me, the last paragraph struck a nerve too. I don't think that is what neutrality on wikipedia is meant to mean. The second paragraph, imo, implied a bit too much (IMO) that volume of !votes is a deciding factor, which means Voting rather than !voting. I want to point out I actually appreciate the penultimate paragraph a lot, however. That said, the reason I am here is actually very different:
  I would like to ask Beland why he did not close as "No Consensus" for (among others) the reasons I stated during the RfC. This entire RfC was the result of a link to the Persecution of transgender people under the second trump administration article in the lede (which some have stated should never have been there to begin with). Why not simply close the RfC to allow the discussion there to be finished first?
  The current outcome leads to an inconsistency between the articles that cannot be explained. If the term is banned here, it certainly should not be an article title there. (and vice versa: if the article title is sustained there, this RfC's outcome makes no sense, or reference should only be removed from the lede (I think that would be better anyways)) The outcome of whichever RfC was to close first would prejudice the other RfC. Now we've made the RfC there so much messier. I reiterate myself: I think the discussion should have been centralised. This was a bad RfC.
  Happy editing everyone, Slomo666 (talk) 19:18, 5 May 2026 (UTC)

  I did not close as "no consensus" because the consensus was clear. As far as I know, there wasn't a discussion about moving Persecution of transgender people under the second Trump administration active at the time I closed this RFC. The previous one had already concluded and failed to move. A new one has been opened after this RFC closed. As I stated in the close, there are different NPOV rules for titles vs. article prose, and I did not think it was in the scope of my job as closer to rule on whether the two must be consistent. If two different RFCs on two different pages come to conflicting decisions, I think it's reasonable to have a third discussion where participants from both are invited, if the closer of the second one can't see an obvious reason why the first one was incorrect.

  I'm not sure why it matters whether numerical superiority or alignment of sources with arguments is more important if they both point in the same direction. "Reliable sources say" are not magic words; I do feel comfortable overriding a numerical majority if the cited sources straightforwardly do not support what they claim they support. That was not necessary in this case. -- Beland (talk) 22:37, 5 May 2026 (UTC)

  My mistake. I misremembered. I still think it would have been better to 'declare ourselves incompetent', (judicial term where I'm from for declining jurisdiction) but I understand why you did not see a need to do so. That said, as far as I can tell, it would be perfectly in the scope of a closer to recognise the downstream effects of an RfC on other issues, and to decide on the venue for the discussion if necessary.

  As for the last paragraph of your answer: it might not matter in this case, because they did point in the same direction, but pointing this out as you did suggests there is any relevance in numerical volume, which as far as I was aware, there is not, and I would hate to see language like this more often as it may lead to people believing that Wikipedia is a democracy. (It is a minor complaint, not a major issue, but I do hope this kind of language won't become standard)

  Your answer here, again, places that "numerical superiority" in front. A mistake, in my view: you don't need to "override a numerical majority", since such a "majority" does not matter unless it absolutely needs to. A priori, the arguments matter and nothing else. Thanks for your answer(s). Happy editing and good luck with the rest of the review. Slomo666 (talk) 23:11, 5 May 2026 (UTC)

  Numerical voting is indeed taken into consideration when determining consensus. Though the guideline Wikipedia:Closing discussions § Consensus says "Consensus is not determined by counting heads or counting votes", it also says "nor is it determined by the closer's own views about what action or outcome is most appropriate" (supervoting). The specific guidance on counting is: "If the discussion shows that some people think one policy is controlling, and some another, the closer is expected to close by judging which view has the larger number of responsible Wikipedians supporting it, not personally select which is the better policy."

  Some discussions are easy to close on policy grounds, but many (at least where I work, on the controversial discussions that come to WP:CR) involve conflicting rules as described. Many also involve disputes over applying a policy based on sources or claims made, how to read sources to discern controversial facts, how to apply a policy to choice of phrasing, or questions that mostly involve editorial discretion like arbitrary style choices or how to organize an article or set of articles. In these cases, I do discard arguments that are nonsensical, clearly wrong, or clearly misinterpret or contradict policy. But after that, I feel I generally have to respect a supermajority because that is the only way I can make a decision without imposing my own judgement. -- Beland (talk) 02:13, 6 May 2026 (UTC)

Discussion (Trump)

• I find some of the reasoning in the !votes to be far more concerning than the close. Seeking out sources to support a predisposed belief is not neutral editing. Using emotionally charged terms when non-charged terms exist is textbook POV pushing. Trying to define a contentious term yourself and then deciding whether that definition applies is pretty much exclusively the domain of WP:CPUSH or WP:CIR (usually a bit of both). And I'm almost tempted to say that citing WP:OBVIOUS or WP:SKYBLUE for political labeling should qualify for a tban on the spot. Thebiguglyalien (talk) 14:02, 4 May 2026 (UTC)
• I agree with Beland to the extent that editors really need to stop engaging in politico-culture wars over the use of a single word as a descriptor, and that we should show, not tell. I get why people don't like Beland's last paragraph, but closers have some leeway to make a statement beyond the actual arguments and I don't view that aspect of their close as supplanting their evaluation of consensus. That said, I haven't read the discussion and don't plan to so I'm not going to !vote here. voorts (talk/contributions) 16:56, 4 May 2026 (UTC)