Wikipedia talk:Genocide

Consider it turned over by me (if needed); that's why I didn't start in my user space. This is basically all of my thinking from the discussions linked, to try and make sense. Hopefully it is helpful. — Very Polite Person (talk/contribs) 17:01, 4 November 2025 (UTC)

I think there are some cases that are archetypical genocides, i.e. Holocaust. But is there some literature that does similar comparison based on control cases? (i know of this ) what comparison are we looking for. User:Bluethricecreamman (Talk·Contribs) 16:58, 4 November 2025 (UTC)

Probably fairly soon someone should do a radical edit on Wikipedia:Genocide and shift most of the content here (or at least add a {{diff}} here for convenient reference), since we don't really want signed discussion on the WP: space page. I'll start a separate section here for a quick thought on how WP:GENOCIDE might be doable. Boud (talk) 18:05, 4 November 2025 (UTC)

Support wholeheartedly. I came to the talk page to say this. The article page should only include proposed policy content. Discussion and thoughts should be on this talk page. Joe vom Titan (talk) 12:31, 29 November 2025 (UTC)

@Joe vom Titan:  Done The brainstorming version is now archived. The draft needs plenty of work - see below for some hints. Boud (talk) 23:23, 28 December 2025 (UTC)

I guess I'm unconvinced about the current 5-question strategy for generating a viable WP:GENOCIDE page. I think that the page would best be descriptive, not prescriptive. And here the problem is the huge range in Wikipedia editing quality and source availability and reliability.

The range of editing attention to genocide-related pages in en.Wikipedia varies hugely depending on media and academic attention and Wikipedia:Systemic bias, leading to a huge variation in the Wikipedia editing quality and the types of sources available.

For the two (main? only?) genocides of the early 2020s, the Gaza genocide had a huge amount of attention, while the Tigray genocide from November 2020 to November 2022, also now as a standalone article at Tigrai Genocide, with an extermination with intent of approximately the same percentage of a population (around 4-10%, depending on the method and source), had very little attention, greatly exacerbated by the tight human and internet blockade of Tigray preventing reports from leaking out and local, federal and international journalists from investigating. I haven't checked whether there's an academic consensus to describe the Tigray genocide as a genocide (I'm using the term here as a non-expert), and I don't see currently active debates among Wikipedians about this in terms of titles or lead text (either for or against the qualification as a genocide). There was this RfC whose result was against the term genocide for the Amhara case, distinct from the Tigray genocide, although also in Ethiopia.

A one-size-fits-all guideline is unlikely to help people who are actually doing the editing. There is generally a time progression of information from online social networks, to human rights organisations, to mainstream news media, to national human rights institutions (such as EHRC), to academic sources, and finally through to independent investigations and possibly prosecutions in courts, although academic non-peer-reviewed sources can be involved at the early stages, benefiting from the social/political scientists' access to knowledge through multiple channels and being able to judge the credibility of the knowledge based on their in-depth background knowledge. Appropriate attribution to sources, starting with subsections of pages before WP:SPLITs to individual pages, and changes from attributed POVs or "allegations" or "accusations" through to Wikivoice generally evolves depending on the degree of reliability that involved editors attribute to the sources, and having sufficient numbers of editors with different personal POVs and life experiences.

I guess the point here is similar to what Buidhe said, that trying to use the Gaza genocide or other well-known consensus genocides as a template for Wikipedia coverage of all other genocides or possible genocides risks being too rigid compared to what genocide scholars view as genocides.

I admit that I really don't have a good idea on how to write a descriptive WP:GENOCIDE draft, so despite my being unconvinced by the 5-question guide, what we have now is at least better than not starting a draft at all. Boud (talk) 18:59, 4 November 2025 (UTC) (clarification Boud (talk) 06:36, 5 November 2025 (UTC))

I agree that the questions are unhelpful. To me, deciding which sources are "correct" is at best a red herring and at worst a means to filter them until we get the end result that aligns with our beliefs. I'm of the opinion that we shouldn't use loaded language in wikivoice at all unless it's a largely uncontested WP:COMMONNAME. For example, Armenian genocide is the default name in the literature, so that's what the article is called and that's how we describe it when we mention it. Similarly, Boston Massacre is only called that on Wikipedia because that's the common name at this point. There's already precedent for this at MOS:LABEL, which is an extension of WP:NPOV#Words to watch. My POV-pusher alarm immediately goes off when someone advocates using a term like massacre or murder on an article where that isn't the term you'd expect to find in most publications on the subject. Thebiguglyalien (talk) 🛸 21:12, 4 November 2025 (UTC)

I completely agree with your proposal of writing a descriptive WP:GENOCIDE, although I disagree with the idea that a non-descriptive (prescriptive?) one would be better than no draft at all, as it risks bringing us into WP:OR territory. Chaotic Enby (talk · contribs) 23:35, 4 November 2025 (UTC)

I put a link to WP:PROPOSAL and a quote from it at the top. Within this section itself, I didn't actually define for people unfamiliar with Wikipedia policy and guideline development what I meant by "descriptive, not prescriptive". WP:PROPOSAL describes this fairly clearly. Boud (talk) 06:59, 5 November 2025 (UTC)

As noted below, Persecution of Uyghurs in China is something that needs to be considered in any analysis since it's gone through a lot throughout its history including once being titled Uyghur genocide with a variety of different wordings e.g. . While most of these have stopped short of calling it a genocide in Wikipedia voice despite the title, it still needs to be considered IMO if this is to be a serious policy especially since a common question is why the article was titled Uyghur genocide for so long but it took ages for us to title the other article Gaza genocide. Note I likely won't be participating in this discussion myself further since I'm unconvinced there's a reason to single out this particular issue compared to the plenty of other cases where we get dispute e.g. some mentioned by Thebiguglyalien but also other stuff like terrorist, etc. Nil Einne (talk) 13:02, 5 November 2025 (UTC)

I would say there are at least a couple of missing questions. One is about the usability of journalistic sources. And another is about how to distinguish where the genocidal aspect of something begins and ends, when it's part of something larger like a war or other violent repression. Even as scholars might agree in recognizing a genocide, they might disagree about its particular shape, and its extent in time and space. Pharos (talk) 04:58, 5 November 2025 (UTC)

The dates are often not agreed upon even when there is a lot of consensus that there is a genocide.

We have reliable sources saying the Armenian genocide ended in 1915, 1916, 1917, 1918, 1923, and even one that says it still continues in some form today. (t · c) buidhe 05:51, 5 November 2025 (UTC)

Also, I think we should talk about when we start labelling or moving articles to "X genocide". Do we do it in real-time, i.e. when the event is happening and there are a lot of emotional charged or driven content out there in reliable sources, or after when there are more historical data or discourse about the event, and undoubtedly more detached. Gaza genocide isn't the first case we have here, we have had Persecution of Uyghurs in China where it was once at Uyghur genocideUyghur genocide when there was much coverage about it in Western media. – robertsky (talk) 12:26, 5 November 2025 (UTC)

The difference between these cases was the volume of scholarly sources available. A blanket rule here would prevent us from complying with NPOV when a large number of scholarly sources do exist. (t · c) buidhe 13:09, 5 November 2025 (UTC)

I am not disputing the volume of scholarly sources for the Gaza genocide. What I am referring to is the probabilities of the reliable sources being emotional charged/driven and hence do we or should we look at such events from a historical lens or not. I am not looking for a blanket rule, but one should be cognizant of such undertones or possibilities for current and future events. – robertsky (talk) 05:12, 6 November 2025 (UTC)

I also felt that most sources are just hyperfocusing on this event due to Israel's involvement.--ZKang123 (talk · contribs) 05:02, 8 November 2025 (UTC)

@Very Polite Person: I have no doubt that you're acting in good faith, but a few notes:

• We already have several gudielines on contentious labels. The prudent thing to do here would've been to start a discussion at one of those pages.
• The questions as you've framed them are not how we usually establish consensus or write PAGs. Please see our guidelines on writing PAGs.
• There's no rush to create a new policy just because Jimbo (or Larry or whoever) said something controversial. In fact, I think that being reactive is an actively bad way to create PAGs.

voorts (talk/contributions) 14:00, 5 November 2025 (UTC)

disagree with you voorts. this section here Talk:Gaza_genocide#A_recommendation_of_WP:GENOCIDE was one of the few sections with more light than smoke.
I'd also point to this related appeal from arbcom. Wikipedia:Arbitration/Requests/Case/Palestine-Israel_articles_5#Community_encouraged User:Bluethricecreamman (Talk·Contribs) 14:04, 5 November 2025 (UTC)

I'm not saying nobody should try to write a gudieline here (although I don't think it will be a particularly productive discussion). I'm saying these particular questions that were proposed are odd and that there was no rush to start creating a new page when we need to think about how they interact with existing PAGs. We should be entering these sorts of discussions with an open mind, not a particular set of questions about how we weigh between sources that cabins the framing of the discussion. I've been saying this a lot recently, but rushing to create PAGs without an idea of where the community wants to go with an RFC / where the points of tension are is not a good way to create policy. voorts (talk/contributions) 14:11, 5 November 2025 (UTC)

Ah ok. I see your point.

i posted on npovn about this page to see if others will take a look. I disclose i havent taken a deep look at the rfc questions yet or responses since yesterday, agreed some of the questions may be hard to move into policy space . User:Bluethricecreamman (Talk·Contribs) 14:17, 5 November 2025 (UTC)

If you all slowed down you might actually find points of agreement with people on the "other side" and come to something that more truly resembles consensus. voorts (talk/contributions) 14:18, 5 November 2025 (UTC)

I agree that the current questions posed are not terribly useful. Answers to #1-3 are already codified at WP:SOURCETYPES in the RS policy. #4 and #5 are invitations to engage in OR. _signed, Rosguill ^talk 14:52, 5 November 2025 (UTC)

Enough people seemed interested in my idea (including Wales) that I just wanted to get it down while it was fresh in my mind, in case others wanted to pursue it to whatever ends. It seems like some folks do.

I'm just honestly watching while working on other stuff. This is not my usual lane, and like I said, I'm more than happy to defer to others. I think something like this would be helpful to some end, is all. — Very Polite Person (talk/contribs) 16:24, 5 November 2025 (UTC)

This is not my topic area, but as I read through questions about which sources to consider and how to weigh them, it seems like it would make sense to include a question about quorum among sources, too, especially as it relates to wikivoice.

My impression is that, for scholarly decisions about genocide (as with most applications of a term like that), people most often reach for a pen when they come to a positive conclusion (and that people who are undecided don't rush to write "I am not sure"). So, at what point do we determine scholarly consensus on the issue? In the sciences, we can cite a bunch of reviews and high-profile position statements that summarize many studies, but in the humanities and social sciences, while that happens sometimes, it's not always straightforward (slight variations in definition, extended qualifying language, more need for interpretation, etc.).

There are related guidelines out there, like MOS:LABEL, where there's a high bar for moving from "described as" to "is", but even there I'm not sure we've ever articulated a clear statement about quorum. It just kind of winds up being straw polls asking "are there enough sources yet?" Maybe that's all we can realistically hope for, and I don't have a good alternative -- only suggesting having a question about "quorum among sources for wikivoice" somewhere. — Rhododendrites ^talk \\ 15:49, 5 November 2025 (UTC)

I'm not sure it really moves this past the "are we there yet" straw poll discussion, but one thing to consider is that there's a big difference between a paper arguing for a novel/polemic categorization of something as a genocide (e.g. ) and a paper that presents a genocide as a given (e.g. ). _signed, Rosguill ^talk 15:58, 5 November 2025 (UTC)

Another issue that's generated debate in the past is what if a handful of genocide scholars have examined a particular case and declared "it's a genocide", yet the historians and other Rs discussing the same event don't even mention the possibility. It's hard to figure out what the policy compliant response would be. (t · c) buidhe 15:59, 5 November 2025 (UTC)

There is a guideline for that: Academic consensus. A claim that there is academic consensus over X must be backed by sources that say that explicitly ("there is academic consensus on X" or similar wording), it's not enough enough to round 10, 20, or even 100 sources that individually say "X" and decide ourselves that there is consensus. Cambalachero (talk) 03:12, 10 November 2025 (UTC)

To help, translation of a French book, fr :Bernard_Bruneteau (2019) Génocides, usages et mésusages d'un concept (Genocides: Uses and Misuses of a Concept) with a classification :

[Genocide] can also be grouped based on the interaction between three elements: memory (carried by the victim group and/or its descendants), history (developed through scientific research), and law (established by the international community).

1. The case of canonized ("consacrés") genocides [...] with an alignment of the memory of the victim group (which strongly influences the public sphere through literary accounts, memorial sites, school narratives, etc.), the scientific history of the event (which is widely accepted and occupies a central place in genocide studies), and the law (the event is the very origin of the law or is defined as such by an international criminal tribunal). [...] [= Holocaust, Rwanda, Srebrenica, Khmer Rouge]
2. The case of genocides awaiting full recognition and, as such, sometimes contested or relativized: here, the law is still lagging behind memory and history. In the case of Armenia (1915-1916), Ukraine (1932-1933), and for a long time Cambodia (1975-1979), we have a very active memory disseminated internationally by the diaspora, often long-standing [...] and institutionalized in specific memorials in Yerevan, Kyiv, and Phnom Penh; we also have a massive historical consensus on the genocidal nature of the event: [...] [Nevertheless ] legal delay thus facilitates the challenge to the genocidal classification of the Armenian and Ukrainian events by certain historians beholden to Turkey or pro-Russian (or ex-communist) figures.
3. The case of contested or ignored genocides: here, law and history take a back seat to a social memory that is often activist and seeks to raise global awareness of the reality of past injustices. Three examples of blatant human rights violations during the second half of the 20th century can be cited: Tibet (since the Chinese occupation of 1950), Kurdistan (with Iraqi policy in the 1980s), and East Timor (during the Indonesian takeover in 1975). [...]

His system seems coherent to me, and for example in the case of Gaza, we would be at the second level, a case awaiting full recognition due to the lack of a court verdict. Fabius Lector (talk) 16:09, 9 November 2025 (UTC)

Just to be sure, the 1000-word limit doesn't apply to the brainstorming stage, correct? After all, the point is to send ideas back and forth; it's not yet a formal discussion. --Aquillion (talk) 22:58, 13 November 2025 (UTC)

I think it makes sense not to restrict to the 1000-word limit. After all, we're in the unusual situation where the "page" itself, rather than just the talk page, includes signed comments. Moreover, the discussions seem quite focussed. The fact that we're not going to make any decisions here for any specific atrocity-event articles very likely helps: we're only hoping to summarise what the actual practice is (or has been so far) on these articles, so that readers of the guide can know what to likely expect if they present argument X, Y, or Z in the context of a specific atrocity-event article. Of course, if someone starts walls-of-texting, then someone else may fold those with "show/hide" semi-hidden wrappers. Boud (talk) 15:34, 20 November 2025 (UTC)

There is a requested move discussion at Talk:Southern Kaduna genocide#Requested move 27 November 2025 that may be of interest to editors on this page. Katzrockso (talk) 19:50, 27 November 2025 (UTC)

I've shifted the brainstorming page to Wikipedia:Genocide/Archive 1, shifted the prior debates table to a template at Wikipedia:Genocide/per-event prior debates table, and restarted Wikipedia:Genocide as a rough first draft. I've assumed that we'll only go for a guideline, and that upgrading to a policy could come in the distant future. Please edit directly where controversy is unlikely and start talk page sections here (not on the draft itself) where discussion is needed. Things that I'm fairly sure need work:

• I haven't clarified too well the distinction between "genocide" in Wikivoice versus titles, because these overlap in several cases, but not always; this needs clarification;
• include some perennial links to cases from the prior debates table to illustrate key points based on real evidence – because many people reading the RfC (and later the guideline itself) will be unfamiliar with the full range of cases from poorly documented almost undiscussed ones through to the too-huge-for-almost-anyone-to-thoroughly-read-the-talk-pages ones;
• integrate/add points from Wikipedia:Genocide/Archive 1 that I've overlooked and that are needed;
• put in some guideline-page-acceptable WP:BOLDing and/or italics so that TL;DR readers are more likely to get the key points;
• apart from the {{nutshell}} at the top, I haven't checked too carefully which points should be should consider versus should versus must (I think I haven't put any musts, though consensus might be strong enough for some); individual talk page sections here may be able to converge on good predictions on what RfC consensus is likely to be.

It seems OK to me for people to still update/add the prior debates table. It should probably be hidden-by-default with an click-to-show button when we start getting near an RfC ready version of the proposal. Boud (talk) 23:16, 28 December 2025 (UTC)

Boud I made an edits to address a minor factual error. FYI, international humanitarian law is just for war crimes. genocide as a crime is governed by national law and international criminal law (for example the rome statute). However, the ICJ is not a criminal court so it doesn't use criminal law.

Also, you may not be aware but there have been cases where the verdict of national courts doesn't match prevailing views in rs. (I recall that there have been more such cases, but I cannot think of them right now). (t · c) buIdhe 23:22, 28 December 2025 (UTC)

Nice correction. I guess it's because genocide or a crime against humanity can, in principle, happen without a war. I've been editing on these topics for quite some time and still made a fundamental error.

The question for the guideline is whether we want to mention the possibility that a decision by a national court under universal jurisdiction might be considered a strong enough argument for Wikivoice genocide or for a title changes. Browsing through the cases listed at universal jurisdiction and comparing to what I'm familiar with at prior debates table, my guess is that universal-jurisdiction convictions were unlikely to have played much of a role in the debates, and were mostly in cases were "genocide" was already widely accepted. Currently that section does not exclude universal-jurisdiction convictions, so the guideline does not obstruct the possibility that future universal-jurisdiction convictions (and the Lithuanian/ECtHR cases you've linked to) might convince Wikipedians. Boud (talk) 23:43, 28 December 2025 (UTC)

The other one I'm remembering off the top of my head is a former ISIS fighter convicted of genocide of Yazidis, but that was well after WP was calling it a genocide, and not a factor in discussions. (t · c) buIdhe 03:40, 29 December 2025 (UTC)

There were over 500 scholarly sources mentioning "Rohingya genocide" before 2021. (t · c) buIdhe 00:13, 31 December 2025 (UTC)

So an overwhelming academic consensus before 2021 (likely to be robust to the problems with Alphabet/Google Scholar). Ironically, the Wikipedia debates on the title or the use of Wikivoice, as far as I can see (see the next section or links in Wikipedia:Genocide/per-event prior debates table), didn't refer to this massive academic consensus at all. Probably because of the ease in finding media and UN/human-rights investigation sources, and the lack of any editors insisting on the need for academic sources? Boud (talk) 00:48, 31 December 2025 (UTC)

It looks like this was just one editor pontificating on the talk page—so maybe don't give that too much weight. (t · c) buIdhe 04:56, 31 December 2025 (UTC)

How much was NPOV debated for actual mass atrocity articles, and how likely would similar debates recur? These are quick notes based mainly on the "Discussions" column, and the "Comments" column. See Wikipedia:Genocide/per-event prior debates table for the links, where I haven't included them here. This analysis may have errors, and is not complete (I mostly dropped the cases lower in the table). Feel free to use the table to make similar analyses for motivating discussions about what should be changed in the draft proposal.

• Masalit - WP:NPOV mentioned only in the 2023 RM, not in the two 2025 RMs
• El Fasher - essentially no Wikivoice/title "genocide" debate
• Darfur 2023–present - created in split, no mention of NPOV
• Gaza
  • RMs, RfCs during Oct 2023 to Feb 2024 (I didn't search, not currently in our table)
  • Feb 2024 RM - irrelevant to NPOV
  • Sep 2025 RfC on first sentence - NPOV is a key theme. It does not override WP:FALSEBALANCE. A key question was achieving informed collective editorial judgment on whether the threshold of academic consensus had been met.
  • Jimbo Nov 2025 - Much, much more heat than light per the closer; several suggestions such as high-quality sources from ... governments and from commentators and Prefer primary official statements seem to be a tiny minority POV, especially in the brainstorming phase for this draft proposal; WP:ATTRIBUTEPOV, i.e. effectively NPOV, was suggested and responses included WP:FRINGE, WP:FALSEBALANCE, WP:WEIGHT, and WP:MANDY. WP:SOURCETYPES and WP:SCHOLARSHIP were mentioned.
• Uyghur
  • AfD
    • Feb 2020 AfD (Ethnocide) - irrelevant
  • RMs
    • Aug 2020 RM passed: Cultural genocide of Uyghurs to Uyghur Genocide - single comment that "Cultural genocide" would have been more NPOV
    • Apr 2021 RM failed: Uyghur genocide to Uyghur cultural genocide - NPOV implicitly discussed, but "alleged" considered WP:WEASELly
    • Jan 2024 RM failed: Uyghur genocide to Uyghur genocide accusation - long scholarly sources + media sources list; mostly WP:COMMONNAME but a few NPOV comments were made.
    • Feb 2024 RM passed: Uyghur genocide to Persecution of Uyghurs in China - closure mentioned WP:NCENPOV; "genocide" was not seen as the wide consensus of the sources.
  • RfCs
    • RfC 30 April–July 2021 Wikivoice-genocide-in-the-lead. Result: Uyghur genocide is the characterization that ... amount to genocide;
      NPOV was mentioned a lot, and together with sourcing was a key argument for characterization.
    • RfC Jan–March 2022 Wikivoice-genocide-in-the-lead. Result: The Chinese government has committed ... in Xinjiang, often characterized as genocide.
      NPOV and the WP:WIKIVOICE section of the WP:NPOV page was included in the closing summary. Useful quote from the closer: The question we're left with is whether there exists a serious debate in reliable sources as to whether these events are a genocide.
• Tigray - single discussion about upgrading to "crimes against humanity" or "genocide"; no major editorial controversy (so far) on Wikivoice or article titles.
  About 7% or so of the Tigrayan population were eliminated from Nov 2020 to Nov 2022. This is an extreme case of lack of media (and Wikipedian) attention.
• Yazidi
  • Nov 2015 - brief objection to Wikivoice genocide as violating NPOV (an editor is harboring NPOV feelings against [Islamic State]) by an IP editor; discussion was dropped
  • unknown date - apparently there was an undiscussed move to Persecution of Yazidis by ISIL
  • Mar 2016 - one comment to change title to "genocide"; apparently done under WP:BRD with no revert
  • unknown date - apparently there was an undiscussed move to Genocide of Yazidis by ISIL
  • Jan 2019 RM failed: Genocide of Yazidis by ISIL to Yazidi Genocide; no objection to "genocide" implied in the discussion; no mention of NPOV
• Amhara
  • Aug/Sep 2023 - strong consensus against "genocide" in Wikivoice due to lack of RS
  • Following the Wikivoice closure, a bold move of the title away from Wikivoice was done and was not reverted
• Rohingya
  • Jan 2021 Talk 1 +
  • Jan 2021 Talk 2 with a sources list
    No mention of NPOV, but instead a broad consensus of media sources, and of a UN investigation and human-rights NGOs, were seen as sufficient WP:WEIGHT to justify Wikivoice.
    Government (Myanmar/China/Japan/Russia) opinions were dismissed, effectively under WP:MANDY (Japan because of trading interests).
  • talk 3
    It looks like no scholarly sources were mentioned in any of the three discussions. Scholarly sources are in a Dec 2025 talk page discussion on casualty estimates.

Overall, the relevance of NPOV has ranged from highly relevant in several debates for at least one high-participation case (Uyghur), to not needing explicit mentioning at all in some of the low-participation cases. Several cases are intermediary.

I've made some changes, mostly additions, to the draft proposal, based on these notes. Boud (talk) 00:36, 31 December 2025 (UTC)

It seems to me that there could be an alternative approach, that titles could be constructed in such a way that we don't have to spend so much time and effort going over whether to include "allegations" in a title or not. If would just take a slightly different default for how we format the title to allow flexibility in how the lede is expressed in Wikivoice.

For example, rather than Xinjiang genocide we could have had Genocide in Xinjiang from the very beginning, and it could still be named that today, despite an apparent shift in scholarly consensus. The changes would be gradual editing to the lede, not a whole involved RM. At different points in time, it might be appropriate to use different wording as the sources evolve, e.g.: A genocide in Xinjiang is recognized by a consensus of genocide scholars... vs. A genocide in Xinjiang has been alleged by several countries and major human rights groups....

In short, I think we may be putting too much authority into the title itself, rather than in the article's actual content. Pharos (talk) 04:41, 31 December 2025 (UTC)

While this is an interesting suggestion, I don't think this is how our content is perceived by viewers. When we renamed the Gaza genocide article there were a lot of news papers claiming that Wikipedia recognized the genocide, even though it was several months later that it got put into wikivoice. And debates on the wording are likely to be even more contentious than debates on the title. (t · c) buIdhe 04:51, 31 December 2025 (UTC)

Nice idea, but it would be difficult to justify in the guideline at the moment, since it doesn't seem to have been proposed and accepted in any actual cases so far. Guidelines are supposed to be primarily descriptive, not prescriptive, even if they often de facto become prescriptive. We might be able to include it as a possible (innovative) alternative to Allegations of ..., although (more or less) as buidhe says, the nuance might not be noticed by the media, and might not shorten RM debates. Boud (talk) 12:33, 31 December 2025 (UTC)

I think the obstacle we stumble over, again and again, is trying to compress a somewhat nuanced statement into the literally 2-5 words of a title. It's not really possible to convey any gradations at all in so confined a space, and we just shouldn't attempt it. Instead of finer gradations, we're forced to make the leap from "Allegations of X are" to "The X is" at a certain arbitrary point. Rather that jamming Wikivoice into the title, Wikivoice on the scholarly consensus would be better employed a couple of pixels down, once we can write a meaningful English sentence on the subject. A title is just a variation on a headline, and let's not make the common media literacy mistake of relying too much on the headline rather than the actual story. If we are to avoid attention-grabbing headlines about seemingly sudden shifts in Wikipedia's editorial policy, we should be wary of overreliance on headlines ourselves. It is true this is not a looking-backward proposal, but I do think that sometimes new ideas are called for. Pharos (talk) 04:42, 8 January 2026 (UTC)

I empathise with the motivation, but we would need an idea that would have a fair chance of wide support. I agree that the switch from Allegations of X to X is qualitative. In terms of pure numbers of research papers stating X, I would say it's clear from debates both on mass atrocities and other Wikipedia debates that 51% in favour is insufficient for "broad consensus", while 99% in favour is sufficient. Having an intermediary state that allows a more gradual transition might actually be a good idea. An en dash is one possibility.

Allegations of a genocide in X which may evolve to X – genocide and maybe then to X genocide (depending on the sourcing and RMs, of course)?

An argument against this is that X genocide generally means genocide of the Xians, not genocide by the Xians (TODO: this was debated a few times; and should be added to the guideline as an FAQ), while X – genocide suggests both directions are within the scope.

What we really want is something like Mass atrocity against the Xians that many genocide researchers consider a genocide but not (currently as of 20YY-MM) the overwhelming majority, which is more than a 2-5 word limit. I think X (genocide) would have the same problem as with an en dash. Possible genocide of the Xians is no better than Allegations of. Evolving from Allegations of a genocide of the Xians to X mass atrocity would be seen as downgrading the significance of scholarly research at a time when the scholarly view is increasingly in favour of "genocide"; moreover neither Wikipedia mass atrocity nor Wiktionary wikt:mass atrocity consider mass atrocity to be a notable term. X atrocity crime or X mass atrocity crime would tend to be seen as implying that a court conviction has been made, as in WP:KILLINGS#Flowchart. However, in the spirit of WP:COMMONNAME, a title does not necessarily aim to use formal terminology. The lead of atrocity crime focuses on mass atrocity crime(s), suggesting a distinction between a "mass atrocity" (scholarly POV in ordinary English) vs a "mass atrocity crime" (legal English).

So how about adding X mass atrocity and/or Mass atrocity against the Xians to our draft guideline proposal as innovative alternative titles to be considered in RMs?

My feeling is that by the time there is sufficient Wikipedia attention to a mass atrocity for there to be editorial disputes, the evidence of there having been or there being an ongoing mass atrocity is already strong enough in WP:RS. So this might be acceptable in RM debates as an alternative to replace Allegations of ..., and as an alternative to X genocide if/when the consensus is only "big" but not overwhelming. Boud (talk) 13:05, 8 January 2026 (UTC)

"Mass atrocity" arguably implies a legal conclusion even more so than genocide given that war crimes and crimes against humanity never took off as a sociological concept in the same way as genocide. And it's rarely going to be a common name. In practice we end up using "Persecution of X by Y" or similar, although this name is often beset by OR issues. (t · c) buIdhe 13:20, 8 January 2026 (UTC)

I agree, my concept is that we would treat such articles as being about the issue of genocide in a particular context, on the analogy of Torture in Venezuela or Racism in Japan. Rather than giving something a common name when it perhaps doesn't really have one. Pharos (talk) 01:14, 12 January 2026 (UTC)

the difference is there is no reasonable dispute that people have been tortured in Venezuela and that racism is a thing in Japan (t · c) buIdhe 03:22, 12 January 2026 (UTC)

I have added an alternative draft at Wikipedia:Genocide/Draft 2, with stricter rules over what gets to be called a genocide and what not. It would be a good idea to have both types of options on the table when proposing to make a policy over this. Cambalachero (talk) 05:35, 3 January 2026 (UTC)

It's worth pointing out that this proposal runs into problems that lead to changing the list of genocides inclusion criteria in the first place. Namely that going by international courts means excluding most events including the Holocaust—blindlynx 19:19, 4 January 2026 (UTC)

This proposal doesn't match the evidence of the actual decisions that have been made in en.Wikipedia. Per WP:PGE, The major content policies, in particular, arose out of the community's actual practices, and thus are correctly considered descriptive pages, even when they describe the community's long-established and widely supported practices in unflinching terms. Any page may use—and many should use—clear, firm, and direct language when describing a firmly established practice. We do not have a firmly established practice; we have a huge diversity of actual decisions, ranging from huge debates to essentially nothing.

As for the emphasis on stating Genocide allegations, however, are not a scientific discussion, that contradicts the evidence of the debates, where "scientific" is used in the wider (beyond-English) sense of "scholarly study based on evidence, definitions, and scholarly debate" rather than the (common-in-English) hard-sciences-only sense, that's unlikely to reach consensus in an RfC, given what we've seen in the various debates. Boud (talk) 15:03, 6 January 2026 (UTC)

There is a big number of users (including even Jimbo Wales) who do not agree with the current use of wikivoice. It would be good for them to have an option B, rather than just "I oppose the proposal".

As for "science", select a meaning and stick to it. If by science we mean hard science, genocide is outside its field of study, as it is strictly a human society phenomenon. If we mean soft science, then there's zero reason to consider that some opinions have weight and others can be dismissed. Cambalachero (talk) 20:41, 7 January 2026 (UTC)

The opinions we care about are well informed and well reasoned. As a proxy, we generally use "reputability". Not sure if that contradicts what you are saying. Rollinginhisgrave (talk | edits) 22:16, 7 January 2026 (UTC)

If Wikipedia lists so many more "genocides" than real-world recognized genocides, that's a big problem of original research that needs to be fixed. Cambalachero (talk) 20:23, 7 January 2026 (UTC)

This is a false premise. Most genocides that are alleged by somebody are not recognized by Wikipedia, and no genocides are recognized by literally everyone. Like all other content we decide based on reliable sources not an official authority, which I doubt is considered original research by anyone but you. (t · c) buIdhe 21:53, 7 January 2026 (UTC)

Your proposal to go off of court rulings would mean four events would be called genocides in wikivoice Srebrenica (ICJ), Rwanda (Rwanda Tribunal), Cambodia (Cambodia Tribunal) and Darfur (ICC) which is way further away from anyone's understanding of the term—blindlynx 23:05, 12 January 2026 (UTC)

with the Rohingya, Ukraine, Masalit and Gaza genocides pending—blindlynx 23:31, 12 January 2026 (UTC)

Very well put: Wikipedia is currently using the term in a way that is completely divorced from the real world. Cambalachero (talk) 01:41, 13 January 2026 (UTC)

Just because something isn't widely known does not mean it doesn't exist or didn't happen. That's why we should go off of scholarly opinions. Your proposal would exclude events that are decidedly widely knows genocides though – namely the Holocaust – so it's a non-starter—blindlynx 21:20, 13 January 2026 (UTC)

Not commenting on the proposal as a whole, but I find The inclusion of a minority opinion that an atrocity event is not a genocide may constitute WP:FALSEBALANCE (see also: WP:FRINGE), and in some cases might in effect be genocide denial. very troubling. Minority views ≠ FRINGE views, and the two should be kept distinct. We in fact have an obligation to report all significant views in proportion to their appearance in reputable sources, of which minority views are still significant, and deferring to a majority will often be engaging a dispute. Rollinginhisgrave (talk | edits) 08:55, 4 January 2026 (UTC)

I believe both you and boud are saying that fringe views are a subset of minority views, so I don't see the dispute here. (t · c) buIdhe 16:22, 4 January 2026 (UTC)

I suppose I am taking issue with "see also" rather than "see". Are we okay with rewording to "The inclusion of a fringe opinion that an atrocity event is not a genocide will constitute WP:FALSEBALANCE, and in some cases might in effect be genocide denial?" Rollinginhisgrave (talk | edits) 16:36, 4 January 2026 (UTC)

From my perspective the coverage of claims "no genocide occurs" is not a dichotomy but a wide spectrum, and coverage of a "no" view could easily be false balance even if it's not a fringe view. For example, devoting equal weight to Yes and No views when the former is say two-thirds of reliable sources would be false balance. (t · c) buIdhe 16:45, 4 January 2026 (UTC)

Yes, I don't think there is any call for dedicating equal weight to Yes or No views if that does not reflect the balance in the literature. Rollinginhisgrave (talk | edits) 16:49, 4 January 2026 (UTC)

This statement is probably wrong because denial is a common aspect of genocides and will likely be due a mention even when it's very much fringe in the scholarly literature (see Armenian genocide for example) (t · c) buIdhe 17:01, 4 January 2026 (UTC)

Yes, you're right. I think we can resolve that by adjusting the wording, either swapping "will" for "may" (vague), or by clarifying that it would be FALSEBALANCE if those views were included as a significant, reputable perspective on the subject (more wordy). Do you have any suggestions? Rollinginhisgrave (talk | edits) 21:44, 4 January 2026 (UTC)

The text stands on the assumption that the event is indeed a genocide, and that's not always the case. Not every case is as clear cut as the Holocaust. What if there is non-trivial controversy? What if the allegation of genocide is itself the fringe one? (For example, the White genocide conspiracy theory) Cambalachero (talk) 23:41, 4 January 2026 (UTC)

The sentence already makes it clear that it's the majority view. The main issue I have with Boud's proposal is that it is vague on which circumstances it "may" apply, but it may not be possible to be more specific yet remain grounded in previous discussions. (t · c) buIdhe 00:01, 5 January 2026 (UTC)

We are not looking for "majority views" however. Rollinginhisgrave (talk | edits) 00:01, 5 January 2026 (UTC)

I agree that the sentence is vague. I also agree that based on my browsing of the (incomplete but still long list of) previous debates, I don't see an easy way of "be[ing] more specific yet remain[ing] grounded in previous discussions". Quite likely someone else can read these more thoroughly than I have and write a less vague but still accurate summary to improve the sentence. There are a lot of debates there. At Wikipedia talk:Genocide#Role of NPOV in prior debates and decisions I noted WP:FALSEBALANCE for the Sep 2025 RfC for Gaza, and WP:FALSEBALANCE, WP:FRINGE, WP:WEIGHT, WP:MANDY, all coming up in the folded Jimbo-launched-debate. Boud (talk) 15:15, 6 January 2026 (UTC)

I'm not coming from the perspective of previous discussions but from the perspective of someone who views the quoted text to contradict the NPOV policy. I understand you wrote the green highlighted text? When you said "The inclusion of a minority opinion that an atrocity event is not a genocide may constitute WP:FALSEBALANCE", were you talking about instances where the opinion is FRINGE / held by a tiny minority? I think that clarification would get us some way towards disambiguating. Rollinginhisgrave (talk | edits) 15:25, 6 January 2026 (UTC)

The whole point of trying to write a guideline that has a chance of achieving Wikipedia consensus is to base it on actual practice from the community of people who edit mass atrocity articles or are uninvolved people who closed RMs, RfCs, and other editorial debates on these topics, not theoretical expectations. I'm very likely the author of the green highlighted text. Yes, I'm thinking of cases such as Talk:Gaza genocide/Archive 12#RfC on first sentence. My impression is that debates varied between people saying that NPOV dominates versus others saying that WP:FALSEBALANCE and/or WP:FRINGE dominate, for the decision on Wikivoice in this particular case. I think that the closer Beland of that RfC might want to look at the Wikipedia:Genocide#Threshold of consensus section, and this talk page section, and maybe see if the third sentence, or the paragraph as a whole, can be improved, with the aim of getting towards a draft that has a chance of consensus in an RfC. Of course, the Gaza case is not the only one. Boud (talk) 16:04, 6 January 2026 (UTC)

I'd consider Fringe and Falsebalance fundamentally a part of NPOV, not opposed—avoiding wiki voice for fringe views itself violates NPOV. Imagine starting an article about vaccines and autism with "Many scientists believe that vaccines don't cause autism" (t · c) buIdhe 16:08, 6 January 2026 (UTC)

Agree with buidhe here. Not having been a part of these discussions, it's quite worrying if FALSEBALANCE is being set against NPOV; I find it hard to imagine what that would even look like given FALSEBALANCE is part of NPOV. Rollinginhisgrave (talk | edits) 16:15, 6 January 2026 (UTC)

I agree that my wording was imprecise. By NPOV versus WP:FALSEBALANCE and/or WP:FRINGE, I meant the in a nutshell basic part of NPOV versus its nuances. WP:FALSEBALANCE is part of the WP:NPOV policy page and WP:FRINGE is summarised and pointed to on the WP:NPOV policy page. Nuances are absent from in-a-nutshell TL;DR, by definition. Boud (talk) 16:35, 6 January 2026 (UTC)

Oh I see. Describe disputes, but don't engage in them" which accounts for the bulk of the nutshell is generally linked to WP:WIKIVOICE or WP:YESPOV, sometimes WP:IMPARTIAL depending on emphasis. Rollinginhisgrave (talk | edits) 16:41, 6 January 2026 (UTC)

I have no opinion on this text; I leave it to interested editors. -- Beland (talk) 07:33, 7 January 2026 (UTC)

@Rollinginhisgrave, I also don't understand the purpose of this clause. Alaexis_¿question? 18:57, 29 March 2026 (UTC)

Please do, fix or make proposals here and then do or fix them after getting rough consensus. Feel free to add other subsections for missing tasks. Feel free to add comments such as "Seems to be done" or "Still needs a lot of work". Boud (talk) 13:25, 8 January 2026 (UTC)

The only argument made for why these sources should be avoided is that assertion that they are biased in favor of Western governments, which is not, by itself, a criterion for unreliability. I would argue that these sources are actually much more reliable than ideologically captured "scholars" in certain disciplines that this proposal holds up as reliable.^[a] Partofthemachine (talk) 06:22, 16 January 2026 (UTC)

the wider problem is that news sources aren't particularly consistent in calling things 'genocides' and they don't interrogate why they refer to something as such. Really that section should just be a less general version of WP:BESTSOURCES. Which when it comes to genocide stuff tends to be scholarly ones—blindlynx 17:30, 16 January 2026 (UTC)

Journalism is a good source when it's stuff in the journalist's area of expertise. Like most scholarly fields, legal issues and social science/history doesn't tend to be one of them. Newspapers are already listed as poor sources or not cited in many areas of the encyclopedia (medicine, hard science, mathematics, etc.) (t · c) buIdhe 18:07, 16 January 2026 (UTC)

Medicine and mathematics are different because they generally don't significantly intersect with politics/current events. There is also recent precedent for using news coverage for politics-related crimes:Assassination of Charlie Kirk was moved to its current title mostly because the term assassination was commonly used in news reports. Partofthemachine (talk) 01:54, 18 January 2026 (UTC)

It's common to have media covering scientific studies, legal disputes, historical events, etc. and it's also pretty common to see the media misunderstanding / misrepresenting the details of same. Why should they be expected to be any better when it comes to genocide?

That's why we usually cite the generally accepted scholarly viewpoints, not news articles. (t · c) buIdhe 19:58, 19 January 2026 (UTC)

And what if we focus on the assertion that media is biased towards Western governments? That's a classic tenet of conspiracy theorists. It is built on the premise that "the media" is a homogeneous group, that all media are the same, and that all of them work in unison. It is also built on the premise that the government controls the media and dictates to them what to say; which may be the case in quasi-dictatorhips or complete dictatorships, but not everywhere; certainly not in the United States. And the conclusion? That otherwise reliable sources can not be trusted because they are "biased", and so we should stick to academics... but academics are also biased. Case in point: the assertion links to the "Propaganda model" by Herman and Chomsky, and we have a full article about the political positions of Noam Chomsky. Cambalachero (talk) 07:16, 18 January 2026 (UTC)

That's already covered by wp:fringe and various other policies regarding wp:sourcing including wp:great wrongs. We're not here to re litigate existing policies but to fill in a gap between them—blindlynx 16:05, 19 January 2026 (UTC)

Have you actually read the discussion? I'm saying that this proposal is based on ideas that belong to conspiracy theories. Of course that I'm aware of the policy against fringe theories, in fact my answer is built on the premise that fringe theories are not allowed here. As for sourcing, the idea that otherwise reliable sources are suddenly not reliable for a given topic needs to be explained; and with a better explanation than "all Western sources are biased against the East". As for great wrongs... if we follow that idea, we would wait for the courts to decide, as in every other case of someone being accused of a crime. Cambalachero (talk) 19:01, 19 January 2026 (UTC)

Perhaps i should be clearer in what i'm saying. I'm agreeing with you. The media section should focus on why media isn't as good as scholarly sources. Though on reflection there should also be a section about western bias being a bad argument—blindlynx 00:54, 20 January 2026 (UTC)

You have made your opinion well known across multiple articles, that if you personally disagree with scholars publishing on whether something is genocide they are hyper-woke garbage and should be ignored. -- Cdjp1 (talk) 19:51, 28 March 2026 (UTC)

We shouldn't have a statement about the problems with media baked in in our policy. It's not "empirical" at all and it is linked to the Propaganda model article which is about a theory and has a "citations needed" tag.

I agree that media should be ranked below scholarly sources but there is no need to add statements that would immediately get a {{cn}} tag if they appeared in a regular article. Alaexis_¿question? 19:18, 29 March 2026 (UTC)

Proposed RfC text:

Should WP:GENOCIDE be a Wikipedia guideline?

@Blindlynx, Bluethricecreamman, Buidhe, Butterscotch, Cambalachero, Cdjp1, Katzrockso, KetchupSalt, LDW5432, M.Bitton, Partofthemachine, Pharos, Robertsky, Rollinginhisgrave, Rosguill, Very, and Voorts: you are the top 10 contributors apart from me to the WP:GENOCIDE draft guideline, and to its talk page, sorted alphabetically (17 people), according to the page statistics tools. Is the current draft ready for an RfC? If not, then please edit or describe the work that still has to be done, or edit and start talk page sections to see if we can converge on things that risk leading to a confused RfC discussion or to rejection of the draft guideline. I suggest a time scale of a few weeks several days to see if we can converge. Boud (talk) 18:23, 28 March 2026 (UTC) (edit likely timescale, based on initial feedback Boud (talk) 14:08, 29 March 2026 (UTC))

I just want to say thanks to you all turning my rough idea into something helpful and needed. — Very Polite Person (talk/contribs) 18:34, 28 March 2026 (UTC)

Thanks for launching the process. Boud (talk) 20:43, 28 March 2026 (UTC)

I can't comment on if it is ready for a RfC as such an assessment is beyond my ability. Looking over what has come about from individuals efforts, I think the text as stands is good, and has avoided what were my worries about it turning into a template for OR assessment. -- Cdjp1 (talk) 19:33, 28 March 2026 (UTC)

I have no experience with launching RfCs so can't really comment on that but i do think the text is good and an accurate refection the general consensus around when we call atrocities 'genocides' (particularity over at the list)—blindlynx 00:00, 29 March 2026 (UTC)

One can never be ready for a RfC. The feedback here may not be a representative of the wider community. Chances are that there may be staunch opposition no matter what given that it would cut into one's POV of what's considered as genocide. While crystal ball-ing as well, the RfC might end up being ECP-ed as it would weigh in on various CTOPs (Ukraine war and Gaza war).

I suggest just go ahead with a RfC. This proposal has undergone quite extensive feedback, comments, and changes. There's nothing much to change or adjust (if needed) without a wider discussion. – robertsky (talk) 00:30, 29 March 2026 (UTC)

Overall it looks very good, better than I expected it to be. My one criticism is that the two segments on assessing genocidal intent seem likely to cause confusion. My understanding of what’s written is that relevant scholars identifying genocidal intent is a very important piece of evidence towards establishing something as genocide if-and-only-if scholars further link this intent by individuals/groups to genocidal violence occurring, otherwise it’s not relevant to this question. If that understanding is correct, it may be better to present this class of sources as its own section, rather than having it split across both a “Do” and “Don’t” section in a way that could lead to selective misreading. _signed, Rosguill ^talk 00:57, 29 March 2026 (UTC)

I think this is a good idea —blindlynx 12:12, 29 March 2026 (UTC)

I'll start a separate section on this point, so that the overall "Are we ready?" discussion doesn't get too messy. So far looks like we may converge in more like several days than a few weeks ... Boud (talk) 14:03, 29 March 2026 (UTC)

Would be good to stress test before an RFC a bit by asking folks from WP:NPOVN to see if there is anything disqualifying, but from my quick lookover looks good enough id vote yes User:Bluethricecreamman (Talk·Contribs) 15:45, 29 March 2026 (UTC)

WP:NPOVN notified. Boud (talk) 17:59, 29 March 2026 (UTC)

imo as a guideline this is WP:CREEP, it’d be good as an information page or explanatory essay on the topic though Kowal2701 (talk, contribs) 22:37, 29 March 2026 (UTC)

I was going to comment the same thing. It's fairly well written, thought out and responsive to a lot of the concerns brought up in previous discussion, but I don't think guideline is the right tag for it. Katzrockso (talk) 23:35, 29 March 2026 (UTC)

Could you explain why not a bit more?

I feel like where this is now is a good summary of the general concensus about when things are called 'genocides'. In my understanding a guideline is something that captures that so that we don't have to explain it at length during every discussion. Which is presently what's happening in most discussions about genocides. —blindlynx 23:56, 29 March 2026 (UTC)

WP:NOTCREEP (part of the same essay): Additional instruction can be helpful when it succinctly states community consensus regarding a significant point, but it is harmful when the point is trivial, redundant, or unclear. I think it's harder to be more succinct (brief) while still covering key points. The current text is clearly not trivial, given the huge and repeated discussions in e.g. the Uyghur and Gaza cases; I don't think it's redundant, because it's for a quite specific class of Wikipedia articles that are sometimes highly contested and get media attention to Wikipedia (and other times achieve consensus with almost no discussion) rather than a wide class of articles; and the current emerging consensus here seems to be that it's clear. Boud (talk) 11:19, 30 March 2026 (UTC)

Just my opinion, but this appears already covered by existing PAGs, and we don’t have any content guidelines about a specific/niche topic like this. Imo it’d be better as an explanatory essay on NPOV, but it appears very much about the issues of the day rather than something that’s useful in the long term. Imo something more useful would be expanding WP:BESTSOURCES, there’s very little in our PAGs that actually says "prioritise scholarship", there’s only really WP:ACADEMICBIAS (an essay). It’s a pain trying to convince someone that scholarship is the best respected and most authoritative reliable sources and media is comparatively rubbish Kowal2701 (talk, contribs) 13:18, 30 March 2026 (UTC)

I think that these comments will be most useful once the RfC is launched. Here we can propose the text that we see as having the best chance of consensus as a guideline, and the community will accept/reject/modify.

But a few thoughts: we do actually have some niche topic guidelines: WP:ASTROART, WP:CHART, Wikipedia:Indic transliteration; genocides are not just issues of the day: we have several in the 2020s (so far) and no guarantee that the 2030s will be genocide-free; prioritising scholarship: there are many topics (e.g. minor films/film stars/singers/lists of head of state visits) that are WP-notable but unlikely to be covered by academic studies. Boud (talk) 17:58, 30 March 2026 (UTC)

I agree that this is far too specific to be a guideline. We rarely have guidelines focused on particular topic areas, largely for the manual of style. Katzrockso (talk) 19:26, 30 March 2026 (UTC)

Looks rfc-ready to me. Gråbergs Gråa Sång (talk) 08:14, 30 March 2026 (UTC)

WP:ASTROART and WP:CHART are on very specific topics, both in some sense provide specific ways in how to interpret WP:RS in those particular contexts, and neither is MOS-related. Boud (talk) 11:09, 1 April 2026 (UTC)

Please include Wikipedia:Genocide/Draft 2, so that those who do not agree with the current text have an alternative proposal. Cambalachero (talk) 15:16, 30 March 2026 (UTC)

Could others please state support/oppose for rewording the RfC question to:

Should WP:GENOCIDE (Draft 1) or Wikipedia:Genocide/Draft 2 be a Wikipedia guideline?

instead of Should WP:GENOCIDE be a Wikipedia guideline? Boud (talk) 17:58, 30 March 2026 (UTC)

I am not sure that this would be an appropriate RfC question. Several people have expressed support for this draft, but no one except the proposer has supported the other one, so an rfc on making the second draft a guideline would be premature. It would be more appropriate to keep the previous RfC question and allow alternatives to be brought up in comments or a separate RfC (if and when it can pass wp:RFCBEFORE). (t · c) buIdhe 13:39, 31 March 2026 (UTC)

It's clearly not ready for an RFC because it contradicts itself by claiming that otherwise reliable mainstream news publications are unreliable because of "an empirical pattern of pro-government bias in the mainstream Western media, likely similar pro-government biases in the media of non-Western democracies", but also (correctly) pointing out that bias is not a criterion for unreliability. Partofthemachine (talk) 23:03, 4 April 2026 (UTC)

You're comparing two places in the text.

The second place states: Stating that a source asserting or denying genocide is biased, without presenting evidence that that source is unreliable in this context, is a weak argument, where I've added italics for emphasis. The first place refers to, as you correctly quote, an empirical pattern ..., i.e. it presents evidence that these sources (as a class) are unreliable in this context. So there's no contradiction.

Nevertheless, you're right that although bias and unreliability are related, they're not equivalent, and the wording needed clarification. See this correction. Boud (talk) 10:57, 5 April 2026 (UTC)

I propose that someone other than me close this section by posting the RfC. We don't have literally 100% consensus, but it seems to me like we're close enough to consensus and that the current version of the proposal "has a fair chance of success with the broader community", where I've replaced "solid" by "fair", since I think we've done as well as we can to maximise the probability of community support, but predicting the reaction is too speculative to justify "solid". Since I'm the dominant contributor by byte and edit counts, I suggest that someone else than me close this discussion by judging when we've achieved sufficient consensus and starting the RfC per WP:RFCOPEN and Wikipedia:Policies and guidelines#Creating a request for comment. Boud (talk) 09:29, 2 April 2026 (UTC) (see below Boud (talk) 16:54, 4 April 2026 (UTC))

Is it too late for a quick review of the Media section? Currently it says that mainstream and reliable-source media are poor sources. The discounting of media sources literally for being "mainstream" is crazy, and I'm surprised that clause has made it this far Placeholderer (talk) 18:49, 2 April 2026 (UTC)

It is not literally because they are mainstream; they are poor sources in this context as stated per the evidence, but tabloids and so on in Western democracries are even worse (and media in autocracies worse still). The wording of the version of 11:02, 1 April 2026 looks correct to me, but I agree that it could be modified to make it clear that we're not saying that non-mainstream sources are necessarily better than mainstream ones, or that unreliable sources are better than reliable ones. Does this fix of 22:59, 2 April 2026 make the sense sufficiently clear?

Pedantically, this fix doesn't cover all "categories" of sources (e.g. logically, non-mainstream reliable sources, and mainstream unreliable sources, except that unreliable sources are typically not thought of as "mainstream"). The category of non-mainstream reliable sources could include sources like The Conversation, but that's a crossover between "media" and "scholarship", and even if articles from The Conversation have been used in genocide debates, that sounds like too much detail. Boud (talk) 23:11, 2 April 2026 (UTC) (clarify Boud (talk) 23:12, 2 April 2026 (UTC) + Boud (talk) 23:15, 2 April 2026 (UTC))

Could 1. Mainstream media that are reliable sources are usually considered as poor sources... (status quo) be swapped with 2. News media that are reliable sources are usually considered as poor sources...? (Or, maybe better, 3. News media are usually considered as poor sources...) I had (mis)interpreted 1 as saying that once a source meets both conditions of being reliable and mainstream was when the point applied. 2 could be misinterpreted as a nudge towards sources that aren't reliable Placeholderer (talk) 02:51, 3 April 2026 (UTC)

How about this edit of 09:38, 3 April 2026? I left in "mainstream" rather than "news", because WP:NEWSORG does actually say News reporting from less-established outlets is generally considered less reliable for statements of fact, and it would be false to say that "news media" usually qualify as reliable sources (since "news media" will tend to be interpreted to include sources that look newsy and pretend to be giving serious news). I removed the parenthesized sentence, because there's no longer any likely misinterpretation of encouraging the use of unreliable sources. I did the edit directly, on the assumption that if someone other than me closes this and posts the RfC, s/he will actually browse this conversation properly first. Boud (talk) 09:45, 3 April 2026 (UTC)

Actually, your option 3 is better than my 09:38 edit. I've implemented that at 10:41, 3 April 2026, but keeping the intrawiki link. It's clearer to focus on what we're saying rather than start off by warning against what we're not saying. Is this OK? Boud (talk) 10:46, 3 April 2026 (UTC)

Looks good to me! Placeholderer (talk) 22:12, 3 April 2026 (UTC)

Per this meta-discussion about the RfC, I'll start the process without waiting further. Scroll down to #Draft RfC on WP:GENOCIDE and please comment there. Boud (talk) 16:54, 4 April 2026 (UTC)

See the point by Rosguill in the above section: My one criticism is that the two segments on assessing genocidal intent seem likely to cause confusion. ... I've done a quick first attempt at unifying these. Please edit directly or discuss here.

I put "may be relevant to the content" because, e.g., a scholarly source A may say that Minister X of Y-land, which is massacring Z-ians, stated "Zs are [dehumanising description]" and that that statement constitutes hate speech, but A might not go all the way of asserting that that's evidence of genocidal intent. So it would likely be relevant to the overall event (in a Background or Reactions section), but not help much for a genocidal-intent-in-Wikivoice debate. Boud (talk) 14:03, 29 March 2026 (UTC)

"The balance between WP:POVNAME versus WP:NDESC may fail to achieve consensus, in which case the strength of sourcing may be the decisive factor." After reading several times I still don't understand what is meant. (t · c) buIdhe 04:30, 31 March 2026 (UTC)

In that case, it needs to be rewritten more clearly. It's meant to be a succinct summary of the key arguments of the 3 May 2024 debate and closure. It's fairly clear that Joe Roe meant The main arguments ... options 1 and 2 ... instead of The main arguments ... options 2 and 3 ..., since the former is the only interpretation that makes sense (@Joe Roe: correcting this with {{s}} and a brief note on the edit would be useful to reduce confusion). Let's see if a better wording for WP:GENOCIDE can be chosen to summarise Joe Roe's summary of that major debate:

• The main arguments ... descriptive effectively says that people arguing for options 1 or 2 (in favour of qualifying the word "genocide" with "question" or "accusation") were saying that option 3 was unacceptable because it would violate WP:NDESC;
• people arguing for option 3 argued that
  • the source analysis supported 'genocide' as a neutral descriptor, i.e. option 3 satisfies WP:NDESC and options 1+2 violate WP:NDESC;
  • and/or
  • that the presence of a statement in an article title does not imply that the statement is factual effectively says that WP:POVNAME applies (as in antigravity does not imply antigravity's existence and Alexander the Great does not imply Alexander's greatness);
• the arguments in favour of this title [Option 3] generally had a stronger grounding in reliable sources, which I reworded to "the strength of sourcing may be the decisive factor".

In other words (simplifying the groups of people for the sake of the summary, and neglecting that the 'and/or' is a non-exclusive 'or' for simplicity):

• Option 1+2 people disagree with a subgroup of Option 3 people about which option is preferred under WP:NDESC; 1+2 say their option is preferred, 3 say that theirs is preferred;
• Option 1+2 people disagree with the other subgroup of Option 3 people; this subgroup of Option 3 people say that WP:POVNAME overrides WP:NDESC;
• Overall, no matter whether Option 3 is justified under WP:NDESC or WP:POVNAME, it's better sourced, and thus is a fair summary of the consensus.

How about this?

A dispute in which WP:POVNAME is opposed to WP:NDESC may fail to achieve consensus, in which case the strength of sourcing may be the decisive factor.

Getting the typo fixed will help reduce confusion for people who click through to the RM. Since Joe Roe states "I no longer actively edit the English Wikipedia" and seems to edit about once a month, we might have to wait a lot before the typo is fixed, unless there's an acceptable way to edit that faster (maybe a CTOP arbitrator?). Boud (talk) 13:53, 31 March 2026 (UTC) (already fixed; see below Boud (talk) 20:57, 8 April 2026 (UTC))

I still think that "A dispute in which WP:POVNAME is opposed to WP:NDESC may fail to achieve consensus" is hard to understand what consensus is failing to be achieved. Maybe "if the majority of sources use the term genocide, consensus may favor its use even if editors disagree whether it is a descriptive title or a POVNAME" (t · c) buIdhe 13:57, 31 March 2026 (UTC)

It may also be a good idea to avoid all acronym links, and rewrite them in normal speech. Cambalachero (talk) 14:04, 31 March 2026 (UTC)

How about this?

If the majority of the sources use the term "genocide", then consensus may favour that, even if editors disagree whether it is a neutral descriptive title or a common but non-neutral name.

The acronym links are hidden, but available for clarity. While the Gaza debate is "hot", I think it's worth referring to in order to show that this is not just a bunch of people inventing a guideline from thin air. Boud (talk) 14:21, 31 March 2026 (UTC)

Precisely because it's hot (meaning, a heavily contested "consensus" that will not last for long, like the one to keep X as "Twitter"), it may not be a good idea to use it as an example. Cambalachero (talk) 16:22, 31 March 2026 (UTC)

I support Boud's version. While a "hot" example may not be ideal, it does show some support in prior discussions. (t · c) buIdhe 01:07, 1 April 2026 (UTC)

 Done Boud (talk) 11:11, 1 April 2026 (UTC)

The 3 May 2024 RM close was already corrected (I didn't check who did the fix - it wasn't me). One example of a permalink to the corrected version is: this oldid link with an internal anchor. This has The main argument in favour of options 1 and 2 were ..., which makes sense. Boud (talk) 20:57, 8 April 2026 (UTC)

I fixed that on Talk:Gaza genocide. So many people were enthusiastic about participating in the debates, but after they were over, and a few people tidied, too few people double-checked the tidying. Feel free to double-check my fix. Boud (talk) 21:11, 8 April 2026 (UTC)

"NPOV may override a reliable-source majority if the minority opinion is considered strong enough" This is already in policy, and is really difficult and vague; what does "strong enough" mean? What counts as "genocide denial"? I think one way to make this clearer is that "a claim is considered fringe, not seriously contested, or genocide denial if the vast majority of the sources making that claim are non-reliable and there are few scholarly sources making it". That's a better litmus test as it is more concrete. VidanaliK (talk to me) (contributions) 16:14, 31 March 2026 (UTC)

The expression "is considered strong enough, to be judged on a case-by-case basis" is within the Threshold of consensus section, whose whole focus is that trying to formalise a threshold is unlikely to gain consensus - some cases will be obvious and others will require a difficult judgment by the closer. That section adds nuance, not a litmus test. So strong enough is defined empirically by the evidence (wiki researchers could use the table to as a hint to inspire a rigorous study). See genocide denial for more on genocide denial.

This proposed guideline is intended to be descriptive of how debates have happened and concluded, not prescriptive of how they should happen and conclude. The aim is to reduce time spent in arguments that are likely to be ignored or distracting. I don't see (though I might have missed them) cases where the closer of an RM or RfC found consensus that the alternative to genocide in the title or Wikivoice is a WP:FRINGE claim or is genocide denial, even though some participants of the debates have asserted these. Writing a sentence with a definite if (in the logical sense) would be contrary to the theme of the section, which is to say that the threshold to determine consensus will vary from case to case (and be likely difficult to quantify). Boud (talk) 17:54, 31 March 2026 (UTC)

Just for convenience, in case anyone is concerned about the WP:RFCBEFORE pre-RfC process:

• at WP:GENOCIDE, the post-brainstorming version was edited from 28 Dec 2025 to 8 Jan 2026, growing from 7,179 bytes to 10,351 bytes;
• at WP:GENOCIDE, we then had an unofficial pause from 8 Jan to 28 March 2026;
• here at Wikipedia talk:Genocide, we had a corresponding unofficial pause from 20 Jan to 28 March 2026;
• at WP:GENOCIDE, starting from the 28 March first edit through to 10:41, 3 April 2026, there was quite a bit of tidying, especially clarifications, with the length reduced slightly from 10,340 bytes to 9,969 bytes.

Boud (talk) 11:30, 3 April 2026 (UTC) (fix incorrect link Boud (talk) 20:01, 5 April 2026 (UTC))

{{draft RfC}} {{rfc|policy|hist|pol}} Should WP:GENOCIDE be a Wikipedia guideline? 16:54, 4 April 2026 (UTC)

This is no longer a draft, and is now a proposed RfC. See below at #RfC on WP:GENOCIDE. Boud (talk) 12:46, 5 April 2026 (UTC)