Wikipedia talk:Neutral point of view

WP:NPOV's application under objective qualities

Recently after some scrolling through the policy section of the village pump, and seeing a topic relating to Wikipedia:NPOV, it raised me a question that the main article fails to answer. How does NPOV affect articles and discussion of people, events, and other wiki content about subjects that are, from a neutral and objective point of view, objectively "Bad" in a sense of morality? I think we all can agree that Jeffery Epstein did some horrible things, even from a stance of neutrality and objectiveness. How would NPOV govern such a circumstance? ~2026-11404-95 (talk) 14:32, 24 February 2026 (UTC)

The normal, policy-compliant approach is to describe what bad things the subject did, without writing "BTW all this is obviously morally wrong". When it's obvious, there's no need to state the obvious. When it's not obvious, then some context has to be supplied (e.g., "The organization was fined for doing this on Tuesdays. This was a problem because there was a legal requirement for them to do it on Mondays"). WhatamIdoing (talk) 23:57, 18 March 2026 (UTC)

Additionally, once you have neutrally described the activity, if there's a clear DUE/WEIGHT in sources that opine that the action was bad, then you can add after that attributed discussion to that.

(And a caution on WAID's example, you would need a reliable source to explain why something was legally wrong, likely also with attribution. You can't read the law yourself as an editor and say it was legally wrong.) Masem (t) 00:06, 19 March 2026 (UTC)

Yes, you have to have a reliable source that explains that it was legally wrong. WhatamIdoing (talk) 02:23, 19 March 2026 (UTC)

Pakistani films

Hi all. There is a debate open at User talk:~2026-19304-32#April 2026 regarding verifiability of publications and there usage on Wikipedia on the article for any Pakistani film. Anyone is welcome to participate, thank you! M. Billoo 03:24, 11 April 2026 (UTC)

BALANCE and majority rules

NPOV requires editors "describe, not engage disputes" between viewpoint holders, so long as viewpoints are not fringe or held by an "extremely small minority". I recently came across a discussion where a viewpoint was being put forth in a minority of reputable sources that contradicted a majority. Based on a reading of the above, it would appear that we should not engage the dispute and endorse the views of the majority in wikivoice, but instead describe the dispute and the viewpoints, "fairly, proportionately, and, as far as possible, without editorial bias".

Despite this, WP:BALANCE was cited as supporting "the idea that we should put in wikivoice labels that a majority of sources use". BALANCE indeed requires "describing both points of view" when reliable source "contradict one another and are relatively equal in prominence". How does this requirement that viewpoints should be relatively equal in prominence square with the above reading of not engaging debates in general, not only when they are relatively equal in prominence? Whonting (talk) 06:43, 29 April 2026 (UTC)

Since people will probably ask, this is the discussion: Talk:Missing_scientists_conspiracy_theory#Bias Gråbergs Gråa Sång (talk) 06:48, 29 April 2026 (UTC)

Btw, is your point that Missing scientists conspiracy theory should not use conspiracy theory in title and wikivoice? Gråbergs Gråa Sång (talk) 07:05, 29 April 2026 (UTC)

I don't have an opinion on this particular dispute and don't have the interest to form one. My only point is that if, as described on the talk page, there is a dispute in reputable sources whether something is a conspiracy theory, the simple fact that more support one side doesn't mean their viewpoint should go in wikivoice. Whonting (talk) 07:09, 29 April 2026 (UTC)

, so long as viewpoints are not fringe covers conspiracy theories easily. Simonm223 (talk) 10:36, 29 April 2026 (UTC)

@Simonm223, can I ask you reread the comment you're responding to? Whonting (talk) 10:57, 29 April 2026 (UTC)

Yeah. A lot of it is around the question of whether the FBI is investigating a conspiracy theory or possible chicanery. But considering the current director of the FBI is a conspiracy theory podcaster... Simonm223 (talk) 12:35, 29 April 2026 (UTC)

Like one of the editors on the talk page described this as Three Body Problem-type allegations. Assuming they're accurate about these allegations... Well The Three-Body Problem is about an alien invasion. So I'd say we're firmly in conspiracy theory territory here. Simonm223 (talk) 12:42, 29 April 2026 (UTC)

I'm sure you're right. I don't care for the specifics, which is why I initially asked a general question and why the above comment was qualified with "if" this were a case of there being disagreement in reputable sources, would we defer to a majority or would we avoid engaging a dispute between adherants of a majority and a (non-tiny) minority. Whonting (talk) 12:47, 29 April 2026 (UTC)

Can you present any reliable sources that say there is an alien invasion targeting nuclear scientists? Simonm223 (talk) 12:52, 29 April 2026 (UTC)

If the bulk of high quality reliable sources assert that a certain view is a conspiracy theory or other aspect that falls under fringe theories, we should not be doubting those sources, and thus can express the majority view in wikivoice. These missing scientists seems like a prime example, where the claim some nefarious purpose to their disappearance is tajen but realiable sources as a conspiracy theory. As a counter ecple, the lab leak theory for COVID's origins is generally considered unlikely but not readily dismissed as fringe, so we describe both major origin theories with attribution, though explain most of the scientific body asserting the natural origin as the more likely route. Masem (t) 13:05, 29 April 2026 (UTC)

Thanks Masem. Would you say the lab leak theory is "relatively equal in prominence" with the theory that COVID arose from "natural zoonotic transmission from bats"? Whonting (talk) 13:08, 29 April 2026 (UTC)

wouldn't say it's equal, simply that as a scientific evaluation, most leading organizations and researchers acknowledge it is likely impossible to determine the actual origin so properly call out the difference in theories as more or less probable. As such it's more that WEIGHT still is applied giving more credence to the natural origin, but without dismissing the lab leak theory as FRINGE, so is consider a major viewpoint to be covered. Masem (t) 13:20, 29 April 2026 (UTC)

WP:BALANCE says "However, when reputable sources contradict one another and are relatively equal in prominence, describe both points of view and work for balance." Do you think the qualification of and are relatively equal in prominence is helpful? We should be describing viewpoints and working for balance regardless of whether the viewpoints are relatively equal in prominence (so long as the viewpoint is not fringe or otherwise held by a tiny minority), and making such a qualification can be read as saying we only need describe disputes when viewpoints are relatively equal in prominence. Whonting (talk) 13:26, 29 April 2026 (UTC)

I dont think the viewpoints need to be of equal prominence to make sure all are covered, but how much they are covered will factor in WEIGHT. But keep in mind the "reputable sources" aspect too. If we are only getting one viewpoint from less than high quality sources, while the high quality sources focus on another, the weaker sources likely won't be seen as equal prominence to include. Masem (t) 15:35, 29 April 2026 (UTC)

I wonder whether 'engaging in the debate' is unclear to some editors.

Describing differing views is not 'engaging in the debate'. 'Engaging in the debate' means trying to persuade the reader that one side is correct and the other side is wrong. A neutral article dispassionately informs the reader of the facts without engaging in persuasive rhetorical techniques. If the article summarizes the subject by presenting statements of facts (including plain statements of facts about opinions, e.g., "Alice Expert's favorite color is green") in plain English, and it does not contain appeals to reason, emotion, values, etc., then the article is not 'engaging in the debate'. Such an article is not engaging in the debate even if a reader might be more inclined to agree with one side or the other after reading it. WhatamIdoing (talk) 23:08, 29 April 2026 (UTC)

Such emotive writing would be engaging the debate, but if the COVID article wrote "COVID arose from natural zoonotic transmission from bats", even with later comment that "some scientists hold that it may have arisen from a lab leak, that would also be engaging the debate.

Do you see there being value in including the phrase "and are relatively equal in prominence"? Whonting (talk) 23:45, 29 April 2026 (UTC)

• I disagree that your COVID example is "engaging in the debate". "Engaging in the debate" sounds more like presenting a list of reasons why the reader ought to believe that COVID either arose from natural zoonotic transmission or as a consequence of scientific research. Merely stating that the main position as a fact, and then noting the existence of disagreement, is not engaging in the debate.
• The phrase "and are relatively equal in prominence" is not just valuable, but actually the critical point. We have multiple different scenarios to deal with here. They roughly work like this (simplified to just two "sides"):
  • One widely accepted, one minority view: "Universal literacy is the goal of most educational systems, and teachers around the world work to teach reading to all children, including children with dyslexia, deafness, blindness, intellectual disability, or other difficulties. However, some scholars have argued that, for a small number of children, educational resources should be directed away from reading and towards other forms of training or care. As Prof. I.M. Portant puts it, 'Yak herders have survived for thousands of years without first learning how to read, and there's every likelihood that they'll continue surviving even if they don't all spend eight years sitting in a classroom'."
  • Two equally prominent views: "Windows and Macs are both personal computing systems. Each has its champions, with Windows tending to be favored in the bookkeepers, and Macs tending to be favored by graphic designers."
  • One widely accepted view, with one tiny minority view: "Donald Trump lost the 2020 election for US president", except in articles about the tiny minority view, such as Attempts to overturn the 2020 United States presidential election and Big lie#2020 stolen election claims. In those articles, the overall presentation should sound more like "Donald Trump lost the 2020 election for US president, and a few people didn't believe the facts".
• The phrase "and are relatively equal in prominence" is important here because it's telling editors that the approach I've described under "Two equally prominent views" is the one that they should take in this case (and not the approaches described in the other two).

WhatamIdoing (talk) 02:12, 30 April 2026 (UTC)

I don't think we need to interpret what engaging the debate "sounds like" -- we have WP:SERIOUSLYCONTESTED making clear that if something is genuinely in dispute it should not be stated as fact but opinion, and we are also told "not to promote one particular point of view over another" - stating one view as fact unambiguously does this.

To your examples:

• One widely accepted, one minority view: This doesn't state any view as fact and thereby does not engage a dispute.
• Two equally prominent views: This doesn't describe any view as fact and thereby does not engage a dispute.
• One widely accepted view, with one tiny minority view: I said above that "engaging the debate" does not cover "tiny minorities". Moreover, I am not sure there is even a tiny minority of reputable sources making such claims.

Based on the examples you give, it sounds like we are in agreement as to how views of different proportions should be presented, but we disagree on whether "and are relatively equal in prominence" supports such constructions. For the first, why do you not think we are "describing both points of view and working for balance"? Whonting (talk) 02:30, 30 April 2026 (UTC)

When did I say that Wikipedia is not "describing both points of view and working for balance"? WhatamIdoing (talk) 03:00, 30 April 2026 (UTC)

This comes from WP:BALANCE: "However, when reputable sources contradict one another and are relatively equal in prominence, describe both points of view and work for balance." In the example you gave, the views were not relatively equal in prominence. Whonting (talk) 03:27, 30 April 2026 (UTC)

I gave three examples. Two of them are not supposed to be "relatively equal in prominence". The one that does (Macs vs Windows) is an example of two views being held with relatively equal prominence. (In case it's not clear, "equal market share" is not the same as "equal prominence". The general POV in reliable sources is not that one is the best, but that which one is the best depends on what you want to do with it.) WhatamIdoing (talk) 07:49, 30 April 2026 (UTC)

Do you read "work for balance" to mean "work to represent views with equal emphasis" or "proportionately, so as not to give undue weight"? Whonting (talk) 08:28, 30 April 2026 (UTC)

When the two views are basically equal (←the context for that sentence), then there's no difference between "work to represent views with equal emphasis" and "proportionately, so as not to give undue weight". WhatamIdoing (talk) 17:35, 30 April 2026 (UTC)

I understand that when views are basically equal there is no difference between the two. Here is another way to read the text: WP:BALANCE describes an approach to editing: "describe both points of view and work for balance", a clarification of what that means in practice ("This involves describing the opposing views clearly, drawing on secondary or tertiary sources that describe the disagreement from a disinterested viewpoint.") and conditions that require such behaviour ("when reputable sources contradict one another and are relatively equal in prominence"). If such conditions are not met, "describing the opposing views clearly, drawing on secondary or tertiary sources that describe the disagreement from a disinterested viewpoint" is not required.

If this sounds academic or nitpicky, I only raise it because WP:BALANCE was cited as supporting "the idea that we should put in wikivoice labels that a majority of sources use", and because I see the notion that we should be deferring to a majority of reliable sources in various discussions. Whonting (talk) 04:41, 1 May 2026 (UTC)

A label is not exactly the same thing as a viewpoint.

We should usually but not always adopt the label/name used in a majority of English-language reliable sources. There are many complexities and nuances in making these choices. Here are two of them, by way of illustrating times when we should WP:Use our own words instead of mindlessly following a majority of reliable sources:

• If that label or name is now wrong, then we should use a name or label that is correct. For example, when a corporation changes its name, editors should not say "Oh, dear, we have to use the old name because a majority of reliable sources are from the years before the name change".
• If that label or name represents a viewpoint (e.g., freedom fighter vs terrorist or rebel), we should be careful about adopting that label, and we should generally make a note about the choice (e.g., "Critics say that Israel's response to Hamas killing and kidnapping well over 1000 people now amounts to genocide, but supporters say that destroying Hamas isn't the same thing as destroying the Palestinian people, so it's not actually genocide"). The main categories (and some common examples) for this are in Wikipedia:Manual of Style/Words to watch.

WhatamIdoing (talk) 18:09, 1 May 2026 (UTC)

Those examples are good. I'll ask two questions and if the answers to either are no I'll leave this here:

1. Do you sometimes see editors express the sentiment that we should go with the majority view of sources when there is a dispute?
2. Do you believe it is believable that some editors would read WP:BALANCE in the way I describe above?

Whonting (talk) 04:48, 2 May 2026 (UTC)

1. Yes, of course. And sometimes they are correct (it's 'the measles vaccine', not the 'baby killing deathwish industrial waste disposal system'), and often the choice between options is not very important anyway (e.g., choosing between a Spanish spelling and a Dutch spelling, when no true English version exists; deciding whether to refer to Hillary Clinton as "Hillary" or as "Rodham Clinton" when you need to differentiate between her and her husband in a paragraph).
2. I find that a very large number of misconceptions about Wikipedia's policies are believable. Nobody reads the directions in the first place, and if they do, they tend to misunderstand it, misquote it, and twist it to support whatever it is that they wanted to do in the first place. We could literally have a rule that says "Don't let the pigeon drive the bus under any circumstance whatsoever this means you I really mean it this time I will block you from here to infinity!", and someone would turn up soon afterwards saying "but that means I can let the pigeon drive the bus over here, right?"

WhatamIdoing (talk) 23:37, 2 May 2026 (UTC)

• This is too contextual to answer in the abstract (beyond the very broad statements that you've already gotten.) Things that matter, and which even very experienced editors often disagree over, include:

• What qualifies as a reliable source in a topic area, and what the best reliable sources are.
• Whether some aspect is fringe.
• Whether the sources are relatively equal in prominence.

One other thing that I haven't seen mentioned above is that per WP:BESTSOURCES, "prominence", in Wikipedia terms, is not just about numbers but quality and expertise; if we have a lot of really high-quality sources on a topic, then we simply ignore clearly lower-quality sources, we don't weigh them by number or word-count or something. An overwhelming consensus among academic experts could trump a more divided or even utterly opposed reception among non-expert news media, opinion commentators, and so on. This is the case on topics such as the creation-evolution controversy, the efficacy of faith healing, the effectiveness of vaccines, global warming, the Shakespearean authorship controversy, and more; it is easy to find lower-quality news sources that both-sides these issues, but since the academic consensus is clear-cut we can state that consensus as fact in the article voice. If you looked at eg. more credulous news coverage by non-experts, you might say "but there's a bunch of people speculating that Shakespeare's plays were written by someone else! How can we state in the article voice that he was the author?" But the best sources - reputable historians specializing in that topic - basically unanimously agree that he wrote his plays. --Aquillion (talk) 16:08, 30 April 2026 (UTC)

These are all good considerations, particularly the point on BESTSOURCES. I have others, though I haven't raised them so as not to detract from the key point: the attitude that floats around Wikipedia sometimes, that we should be deferring to a majority of reliable sources is only true when that claim is not the subject of genuine dispute among reputable sources. Whonting (talk) 16:17, 30 April 2026 (UTC)

When there is a genuine dispute among reputable sources, then there are generally also reliable sources directly saying that there is a genuine dispute. When a large number of reliable sources say there is a genuine dispute, Wikipedia editors should defer to those reliable sources.

For example: How some popular drugs (e.g., GLP-1s and SSRIs) work is unknown. There is a genuine debate among reputable sources about whether this mechanism or that mechanism or some other mechanism is what's actually going on. A majority of reliable sources (including basically all secondary sources) simply state this outright. Editors should definitely be deferring to those reliable sources. WhatamIdoing (talk) 17:43, 30 April 2026 (UTC)

related to academic sources, we should also avoid trying to prematurely assign what are the predominantly viewpoints to some event, and wait for longer term sources, like academics, to describe those from a higher level overview. Near term sources like newspapers and websites will often skew the perception of an event, so trying to judge what are majority or minority views in the short term can be problematic. That's where as an encyclopedia, if we are going to include those short term assessments, it is far better to write describing the differing viewpoints and maybe not necessary give weight to one or the other outside of FRINGE cases. Masem (t) 17:17, 30 April 2026 (UTC)

Right. So for example the question of what crashed to kick off the Roswell incident in 1947 is absolutely settled in serious academic sources (spolier: a balloon), even though it's a hot debate in trashier, even "prominent" sources (UFO books, newspapers, TV shows). In general one high-quality source trumps an infinity of lesser ones for knowledge. Same principle applies for whether Ivermectin treats COViD-19, various "lab leak" narratives, whether tylenol/vaccines cause autism, etc. etc. A frequent WP:PROFRINGE stance is to insist something is seriously contested when in the WP:BESTSOURCES it simply isn't. Bon courage (talk) 03:46, 1 May 2026 (UTC)

If I can take this off topic a bit, just as distinctions can be made between BESTSOURCES and non-BESTSOURCES, distinctions can be made between types of disputes; those over facts (Water is H2O), analysis (this happened because of Y), and normative disputes/disputes over opinion (knowledge should or shouldn't be free/beetroot tastes good).

One article may cover multiple disputes: one on an election between two parties has the dispute between two parties (party A advocated for more Y, saying it would bring prosperity, party B advocated for Z, saying it would quell the flow of bad thing) and the dispute in reputable sources over how to characterize the election (the election was characterized by misinformation / party B utilized crises effectively). Whonting (talk) 04:55, 1 May 2026 (UTC)

So long as Wikipedia maintains a focus on duly summarising the accepted knowledge published on a topic these distinctions, while they may exist, are moot for our purposes. Bon courage (talk) 05:06, 1 May 2026 (UTC)

I don't think so, such a focus tells us nothing about how articles should be framed or what should go in wikivoice vs be attributed. Here are two ways the lead of Elon Musk salute controversy could be written based on the body, both summarising accepted knowledge but in very different ways. Do you believe either one better adheres to commitments such as "describe the dispute" and WP:OPPONENT? Whonting (talk) 04:52, 2 May 2026 (UTC)

I would say the first one, as it appears to be the one neutrally based on reliable sources. -- LCU ActivelyDisinterested «@» °∆t° 11:37, 2 May 2026 (UTC)

Both are based on reliable sources, why do you say the first is neutrally done so? It is framed around the interpretation that it was indeed a Nazi or fascist salute. The second states what the gesture was before describing how it was interpreted. It also discusses the substances of the positions. Whonting (talk) 11:43, 2 May 2026 (UTC)

You may need to show your workings, the current version sources can be found in the article. If you have sources for your version it would be helpful to show them.
I never said it's neutral, I said it neutrally based on reliable sources. The first would be my opinion, the second an observation on whether the content reflects the sources without editors inserting their own bias. -- LCU ActivelyDisinterested «@» °∆t° 11:54, 2 May 2026 (UTC)

I drew from the sources in the article, if there's a particular claim you would like cited inline lmk. To the rest, hopefully if you reread my comment you will see I was clumsily also saying neutral based on the sources. Whonting (talk) 12:38, 2 May 2026 (UTC)

It may be neutral to the sources, but that's where the tone also has to be reviewed. Rushing to call it being interpreted as a Nazi salute rather than explaining the gesture and then moving it to discuss how it was interpreted is how we should describe a controversy, since we're not predisposed to taking a side in the second version. This is a case where the number of voices that complained it seemed to be Nazi salute is the louder voice, both views (one that it was a Nazi salute, the other that it was just a gesture) are of equal prominence that we should not rush to take one side here.

This is common problem on WP nowadays that it may seem right to rush to call out the gesture as bad, as on average, WP editors and sources lean left, and see that as a problem, but we have to put aside our personal outrage to still keep to a neutral tone and approach on these articles, which can lead to misreading the proportion of sources and treating minority views as outright FRINGE-type arguments. Masem (t) 12:20, 2 May 2026 (UTC)

Thanks Masem. When you say "of equal prominence", I would think that more authorities/reputable sources espousing one would mean they are not of equal prominence? I did try in text to express the prominence of each view to that effect. Whonting (talk) 12:45, 2 May 2026 (UTC)

If sources don't handle it that way then that would be the editor introducing their own bias into the description. If sources are split on how it's described then both should used, rather than pushing on to prominence over the other. -- LCU ActivelyDisinterested «@» °∆t° 15:44, 2 May 2026 (UTC)

Sources themselves are subject to bias which we have to make sure we work around to keep a neutral approach to our articles. I remember when this happened that several not-quite-top-tier reliable sources said straight out it was a nazi gesture. It should be obvious that we have no idea if it was an initial reference to a nazi gesture, and that's how our tone should reflect it, though still make sure the majority viewpoint, that most sources saw it as such rather than the explanation given by Musk and others. Masem (t) 16:30, 2 May 2026 (UTC)

Clarification on NPOV/due weight for allegations and controversies

Moved from Wikipedia:Village pump (policy)

If I may suggest, I would like to raise a possible clarification on NPOV and due weight for allegations and controversies.

NPOV asks us: "carefully and critically analyze a variety of of reliable sources" and present views: "fairly and proportionately" so content reflects the relative prominence of views in those sources. Policies like NPOV, NOTSOAP, PROMO, NOTSCANDAL all point in same direction. Meaning, it's not just "is this verifiable?" but also "how much weight does this carry"? in practice, something can pass the "it's in a reliable source" test, but not the weight test. For example, a claim in a single source can make its way into a wikipedia page even when there hasn't been much follow-up or wider coverage. This is especially relevant to serious allegations, where limited or one-off reporting without broader or continued coverage can give more weight than the sources support. It can look bigger on the page than in the wider world of sources.

We can see that we have clear guidance for promotional or advocacy content like WP:PROMO, which helps keep things from getting too heavy on the positive side. The flip side, how to handle lightly covered negative or controversial content relies more on general due weight rules. So, the question is: when something is verifiable but minimally covered with possibly no follow-up or wider lasting coverage or updates, how should we decide if it has enough prominence to include without giving undue weight?

One way might be to clarify this with something like this: "Allegations or controversial content is generally appropriate when supported by significant coverage in independent, reliable sources. Content based solely on isolated or limited reporting or single sourced one-off reporting, even if verifiable, may be excluded if its inclusion would give undue weight. The presence of a denial does not by itself establish due weight". This would help illustrate how due weight applies in a common edge case. If is notable enough for inclusion, it would have atleast a follow-up or an investigation or multiple sources with some in-depth reporting and this clarification will not impact inclusion of such content or any discussion around such content.

Curious what others think. Would something like this be useful, perhaps as an example within WP:UNDUE or WP:PROPORTION? NicoR8 (talk) 17:55, 28 April 2026 (UTC)

We have stronger protections for BLPs, which covers the vast majority of problematic, poorly-verified allegations. I think you're correct that for other topics, like companies, the rules are just due weight. I can imagine situations where this would be less than ideal. Generally editors use their common sense to reach reasonable outcomes, though, so I'd have to see evidence that this is a real and persistent problem before considering modifying policy. Toadspike [Talk] 19:54, 28 April 2026 (UTC)

@Toadspike Yea that makes sense and I suppose commonsense plays an important role and I saw BLP covers many of the more sensitive cases and that is what placed me in dilemma about this edge case in non-BLP situations like as you said companies or deceased people. For example, someone's claim might be attributed, but without much follow-up or wider coverage (Talk:Paul Crouch#Question about BLP Policy) it's veriifiable, yes, but how much weight it should carry isn't always obvious.

That's when things can get a little wobbly. Editors can look at similar content and land in different places (Wikipedia:Neutral_point_of_view/Noticeboard#Ron_Wyatt per one editor " it seems to be in reasonable shape" and Special:Diff/1351575584 - per another editor "a bunch of unacceptable sources. Is there any secondary coverage of this" ), same ingredients, different pot. Another example Wikipedia:Neutral_point_of_view/Noticeboard#Millsaps College article – section weight and neutrality is waiting for help since April 10 ( might not be quite related ), and would have helped if some NPOV/UNDUE clarification was available.

A small clarification or example(s) based on what policy already says, not a new rule, might help steady things in these edge cases. Just making the guidance a bit clearer so we don't end up going round in circles. Of course, like any policy, it would still rely on commonsense, but with a bit more consistency and some example-based thinking to guide it, so editors are a bit more aligned in how to apply the same due weight rules consistently. For the kind of cases you thinking, an example like that could help make the guidance clearer in practice. NicoR8 (talk) 11:58, 29 April 2026 (UTC)

I think that two types of examples could be useful to editors:

• Avoid "criticism" and "reaction" sections. The goal is for someone reading about the subject to get both pro- and anti- sentiments in the same place. Otherwise, some readers will see only the main section and not see the criticism, and other readers will skip to the criticism section and never see the supportive information. For example:
  • When writing about a creative work, business, or product, try to put all related things into the same section (e.g., ==Products== or ==Corporate history==): "In 2026, Big Business, Inc. produced a cheaper version of their popular widget, which was criticized for having lower quality." Don't put "In 2026, Big Business, Inc. produced a cheaper version of their popular widget" in the ==Product develompent== section and then have a separate ==Criticism== section that says "In 2026, a cheaper version of their main product was criticized for having lower quality".
  • When writing about an event or politics, try to put al related things in one section (e.g., ==Recommendations== or ==Campaign themes==): "This report recommended changing the initial requirements, which was criticized for excluding people and not following scientific standards." Don't put "This report recommended changing the initial requirements" in one section and then put "The report was criticized for excluding people and not following scientific standards" in a separate ==Reactions== statement. (If the reaction is perfunctory or generic (e.g., a politician offering Wikipedia:Thoughts and prayers after a disaster), then it should normally be omitted.)
• Only include criticism when it is significant or unexpected. For example:
  • Omit Captain Obvious, like "Demicans disagreed with Republocrats".
  • Try to write for your opponent, and especially to search for sources that support other viewpoints.
  • Think carefully about the difference between a viewpoint being popular, especially within a filter bubble, and a viewpoint being presented in high-quality sources.
  • etc.

The second point could probably be expanded into half a dozen useful essays. WhatamIdoing (talk) 20:39, 29 April 2026 (UTC)

Hey @WhatamIdoing useful points. Your two points, fold critique into context and skip the obvious are about how to serve up crticism that already passes the Inclusion test. Good seasoning, all good. Dilemma is one pot earlier: when does a thinly-sourced allegation earn its place at the table at all? Picture a non-BLP page like a company or organization or deceased person, basically any non-BLP. For a thinly-sourced allegation, nobody else picks it up, no follow-up, no investigation, the story just sits there like a single grain of rice on the plate. WP:V says yes, WP:DUE shrugs. Commonsense used by editors looking at the same text walk away with different answers and round and round potentially inconsistently across pages. So a small example within WP:UNDUE, nothing fancy, not a new rule, just a little lighthouse for editors. Something like: A claim's appearance in one (or thinly-sourced) reliable source establishes verifiability but does not by itself establish due wieght. Where coverage is limited to isolated reporting without follow-up or engagement by other reliable sources, inclusion may give the claim undue weight". NicoR8 (talk) 11:46, 6 May 2026 (UTC)

I think that if you can't fold the critique into context, that should be a warning that you might be headed into UNDUE territory.

For example, consider Vector Marketing. Pretty much everything published about the company is criticism of their main business model: inexperienced young people buy (or put a security deposit on) a sales kit, get a small amount of training, and are unleashed to sell the product (which is kitchen knives) to their friends, family, neighbors, etc. In response to criticism and a few lawsuits, they stopped charging for the sales kit and maybe started paying for the training program (at least sometimes). Consequently, that should be in ==History== instead of ==Criticism==. The fact that you could put it in ==History== shows that this criticism is DUE (though in the case of that company, the only question is how much criticism, not whether any should be there).

But there's another lawsuit that's rather unlike the others: One of the new sellers was an 18-year-old woman who met a 25-year-old man and offered to do an in-home sales demo for him. Shortly after she arrived, he drugged her and raped her. He went to prison, and she sued the company, claiming that they should have prevented this through the provision of safety training. There are few sources about the lawsuit being filed, I see no follow-up sources about the lawsuit, and there is no indication that the company took any action or made any changes in response. Therefore I would lean towards not including this information. WhatamIdoing (talk) 22:02, 6 May 2026 (UTC)

Ahh, now I see what you mean @WhatamIdoing. Vector Marketing makes the point clear where the general criticism has an arc, so it folds into ==History== naturally. Maybe a small note in WP:UNDUE that ties them together: "Where an allegation or controversy can only stand as an isolated item - without follow-up, coverage, response, or broader narrative to fold it into, its inclusion may give the claim undue weight." captures both the structure test you gave and the sourcing angle. Feel free to change, your phrasing will probably be cleaner. Maybe worth circling back to Toadspike for their earlier points? NicoR8 (talk) 16:26, 7 May 2026 (UTC)

We should probably move this discussion to Wikipedia talk:Neutral point of view. WhatamIdoing (talk) 18:06, 8 May 2026 (UTC)

Hey @WhatamIdoing I followed Template:Moved talk and completed the move. NicoR8 (talk) 19:19, 8 May 2026 (UTC)

Regarding the "reasonable shape" comment - as I was the editor who made it - I had glanced at the article and seen what looked much like a pretty normal article about a pseudoarchaeologist of marginal notability. At the time I watchlisted the article and put a pin in it for later as I didn't see it as top-priority and even mentioned that I was a bit rusty on pseudoarchaeology articles which hadn't been a focus of my editing for quite some time. I actually agree with the later editor (@ScottishFinnishRadish) with regard to the overall quality of the sources in the article. I felt little need to go back to the noticeboard and revise my comment as it was already in safe hands. Simonm223 (talk) 11:49, 12 May 2026 (UTC)

Thanks.

I think this idea is closer to WP:BALASP than to WP:DUE. In particular, I think it fits in with this paragraph:

An article should not give undue weight to minor aspects of its subject but should strive to treat each aspect with a weight proportional to its treatment in the body of reliable, published material on the subject. For example, a description of isolated events, quotes, criticisms, or news reports related to one subject may be verifiable and impartial, but still disproportionate to their overall significance to the article topic. This is a concern especially for recent events that may be in the news.

What do you (all) think? WhatamIdoing (talk) 01:12, 9 May 2026 (UTC)

That too works @WhatamIdoing. In this place, something like this with few changes can fit well right after this first para you gave. "Also, for example, where an allegation or controversy can only stand as an isolated item - without follow-up, wider coverage, or a broader narrative to fold it into - its inclusion may be disproportionate. A denial by itself does not make inclusion proportional." NicoR8 (talk) 11:41, 12 May 2026 (UTC)

Does anybody object to adding a sentence like this? WhatamIdoing (talk) 19:47, 12 May 2026 (UTC)

I second everything in this discussion and support inclusion. Stronger barriers to prevent criticism-collecting are desperately needed (especially since there are editors trying to erode away WP:ONUS, one of the few tools we have besides WP:BALASP). Also note there are over 5,000 BLPs with section headings that include the words "criticism" or "controversy". Thebiguglyalien (talk) 00:26, 17 May 2026 (UTC)

• I suspect this is too contextual to come up with one hard-and-fast rule (it's really a huge chunk of NPOV and DUE, and to an extent RS.) Some considerations:

• First, due weight is relative. Weigh the coverage something has received (in terms of breadth, duration, quality, etc) against what's already in the article. "We can't weigh this more / less than XYZ" is one of the strongest and most reliable WP:DUE arguments - with the obvious caveats that editors can still disagree about relative weight and that, even when it's true, it can be solved either by adjusting the weight given to the proposed addition (possibly to zero) or by adjusting the weight to the other points of comparison. Often uncontroversial but insignificant things accrue in articles because nobody bothers to challenge them, and going "wait we have all of this minor stuff but people are objecting to a bit more text for this big scandal?" might mean we just need to clean the minor stuff up.
• Second, there's a bunch of things that can cause something to be due or undue, whose meaning shifts depending on the subject and its overall coverage. I think that sustained coverage, broad coverage, coverage in high-quality sources, coverage that relates it to the subject's main source of notability, and coverage that itself treats the aspect as important are all significant, but I'd be reluctant to turn any one of them into a red line; various combinations of them can lead to something passing the bare minimum threshold for inclusion.
• "Broad coverage" perhaps ought to be elaborated on because I'm not sure we have guidelines discussing it - in general, especially when talking about WP:BIASED sources or the like, if coverage is all from the same "informational bubble", so to speak, or from sources that are all biased in the same direction, I would say that this gives it less weight; a story that has been covered by every news article that reports the position of Party XYZ but has failed to make the jump outside that is due less weight. Personally I feel that we could stand to elaborate a bit more on the relationship between BIASED and due weight in particular.

Also, in most articles, the bare minimum needed for one sentence devoted to something is usually not that high; total exclusion is often a high bar. So sometimes it's more useful to start with that minimal coverage. Due weight is also about placement and focus, after all. And how we frame it also matters; if the sources don't treat something as important, then we shouldn't word it in a breathless manner. That said, another consideration is "are we giving this overall perspective too much weight" - a common problem is editors who add every single opinion piece or biased source that argues a particular position, giving each of their arguments massive amounts of depth. Even if the position itself is WP:DUE, the massive amount of focus that this results in may be undue, so it might make more sense to just combine them all into one sentence. --Aquillion (talk) 01:29, 17 May 2026 (UTC)

@Aquillion, I like your phrase of "We can't weigh this more / less than XYZ". It sounds very practical. Would you please find a way to put that in this policy?

I agree with you about "broad coverage". Maybe an essay defining Wikipedia:Broad coverage would be the place to start? WhatamIdoing (talk) 19:31, 17 May 2026 (UTC)

Trying to fold in Aquillion's considerations, but yeah, the broader "we can't weigh this more / less than XYZ" point likely needs separate treatment. This can be stirred further: "Also, for example, where an allegation or controversy is covered only in isolated reporting, editors should consider whether it is treated as significant in the body of reliable sources on the subject. If not, including it may give undue weight to a minor aspect of the subject. A denial by itself does not make inclusion proportional." Keeping it close to BALASP itself using "undue weight" and "minor aspect" in first para, so this adds clarification to that. NicoR8 (talk) 00:05, 18 May 2026 (UTC)

@NicoR8 Quite the discussion, I've proposed another version below. Anchoring this to retained inclusion after the initial breaking-news directly onto the WP:Notnews and WP:Recentism paragraph already in WP:Balasp, and it picks up @Aquillion's point that "sustained coverage" is one of the strongest signals of due weight. It also speaks to @WhatamIdoing's Vector Marketing example, where the absence of follow-up sourcing on the isolated lawsuit was precisely what tipped it toward exclusion. Also I tightened the denial clause in your original wording. A subject's denial is a downstream consequence of the allegation being published, not independent evidence that the matter carried weight in reliable sources, and without that line, editors can highlight the denial as if it were a balancing second "side."

"Also, for example, where an allegation or controversy reported as breaking news was previously included in an article, editors should consider its retained inclusion on the merits of its sustained coverage as evidenced by what has been published in reliable sources following the initial report. Otherwise, inclusion of an allegation or controversy in the absence of sustained coverage risks undue weight. A denial of allegations by an involved party itself does not merit inclusion and also risks undue weight."

LeónGonsalvesofGoa (talk) 13:34, 23 May 2026 (UTC)

A denial of allegations by an involved party itself does not merit inclusion could be read to mean that a denial of allegations should never be included, which is obviously not true. I presume you are intending to say that allegations being denied does not guarantee inclusion? Thryduulf (talk) 14:45, 23 May 2026 (UTC)

If the allegations are about a crime or similarly serious claim about a BLP, editors consistently include the BLP's denial. We should respect the community's ordinary practices in such cases. WhatamIdoing (talk) 18:39, 23 May 2026 (UTC)

I'm not sure what a Wikipedia version of this would look like, but many sources have a formal right of reply policy. - Palpable (talk) 20:13, 23 May 2026 (UTC)

What we're mostly looking at is something like this:

• Monday: A hundred websites report that a celebrity was arrested on suspicion of drunk driving.
• Tuesday: A hundred websites report that the celebrity denies all guilt.

This isn't proof that any of this should be in a Wikipedia article. If something belongs in Wikipedia, it should look more like this:

• January: Reliable sources report that a celebrity was arrested on suspicion of drunk driving.
• February: The celebrity's publicist reports that the celebrity is in a detox facility.
• March: Reliable sources report that the trial has been postponed.
• April: Reliable sources report that a legislator has introduced the Celebrity Rehab Act to encourage film studios to screen actors for high-risk drug and alcohol abuse.
• May: Reliable sources report that the celebrity plead guilty to lesser charges.
• June: The celebrity's publicist announced that they're in negotiations with an anti-drunk-driving public service announcement.
• and so on.

Or at least like this:

• January 2021: Reliable sources report that a celebrity as arrested for drunk driving.
• June 2021: The celebrity plead guilty to lesser charges.
• January 2022: The news story appeared in the end-of-year reviews in some newspapers.
• Next couple of years: The arrest is mentioned occasionally in unrelated news articles (e.g., "Chris Celebrity, who hasn't been seen drinking in public since their arrest in January 2021, sipped ginseng tea while answering my questions").

WhatamIdoing (talk) 23:47, 23 May 2026 (UTC)

Yea, thanks, I think the Monday/Tuesday example catches the point well. An allegation reported one day and denied the next may still just be one short news wave. Thryduulf's point taken too. The point about denial would be narrower, that a denial by itself does not make the allegation proportional for inclusion. Some tweaks:

"Also, for example, where an allegation or controversy is covered only in isolated or short-lived reporting, editors should consider whether it is treated as significant in the body of reliable sources on the subject, such as through sustained coverage, follow-up reporting, or later discussion in reliable sources. If not, including it may give undue weight to a minor aspect of the subject. A denial by itself does not make inclusion proportional." NicoR8 (talk) 20:04, 26 May 2026 (UTC)

I think that "short-lived" is a good way to describe it. WhatamIdoing (talk) 20:34, 26 May 2026 (UTC)

I agree with your points here but I don't think the wording is great as it has negatives and positives in one long sentence. Perhaps something like:

Editors should consider the significance of the allegation or controversy in the context of the totality of reliable sources on the subject. For example, something that is covered only in isolated or short-lived reporting is less likely to be significant than something which has sustained coverage or follow-up reporting. Something for which discussion exists in reliable sources that significantly post-date the event happening or allegation being made is more likely to be due for inclusion than something which lacks this. A denial does not impact whether an allegation should or should not be included, but where allegations are included it will normally be appropriate to also include verifiable responses to them by the accused.

Although I'm sure this could be improved upon. Thryduulf (talk) 20:42, 26 May 2026 (UTC)

That is a fair point. Trying to fold in Thryduulf's point, I agree the wording was trying to carry too much in one bag and doing too much heavy lifting, mixing the negative and positive points. Splitting them makes things clearer.

"Also, for example, editors should consider the significance of an allegation or controversy in the context of the body of reliable sources on the subject. Isolated or short-lived reporting does not by itself establish that the matter is significant to the subject. Sustained coverage, follow-up reporting, or later discussion in reliable sources can indicate that inclusion is due. When sustained or later coverage is lacking, including it may give undue weight to a minor aspect of the subject. A denial by itself does not make inclusion proportional, but if the matter is due for inclusion, verifiable responses should normally be included as well."

This keeps the two sides in separate buckets, while still tracking the existing BALASP point. NicoR8 (talk) 00:06, 30 May 2026 (UTC)

I'll read this again when I'm more awake but an immediate thought is that we don't want verifiable responses should normally be included as well. to be an unintended license for extensive reactions sections featuring everyone and their dog's opinions on twitter. Adding "by the accused" (or similar) after "responses" should sort that. Thryduulf (talk) 00:30, 30 May 2026 (UTC)

Hmm, I suppose someone could wikilawyer that to exclude statements by the accused actual lawyer or agent, but do we really need to state "by the accused or their representatives"? Thryduulf (talk) 00:45, 30 May 2026 (UTC)

Ah, makes sense. Folding that in, maybe the last sentence could be "A denial by itself does not make inclusion proportional, but if the matter is due for inclusion and the subject has denied the allegation, the denial should normally be included as well." (NB: borrowed a few words from BLP's denial wording). What you proposed "by the accused or their representatives" also works. NicoR8 (talk) 01:19, 30 May 2026 (UTC)

Sorry for the late reply on this. I prefer "accused" rather than "subject" as the latter could be misunderstood where the subject of the article is the party making the accusations, so putting it all together:

"Also, for example, editors should consider the significance of an allegation or controversy in the context of the body of reliable sources on the subject. Isolated or short-lived reporting does not by itself establish that the matter is significant to the subject. Sustained coverage, follow-up reporting, or later discussion in reliable sources can indicate that inclusion is due. When sustained or later coverage is lacking, including it may give undue weight to a minor aspect of the subject. A denial by itself does not make inclusion proportional, but if the matter is due for inclusion and the accused has denied the allegation, the denial should normally be included as well."

I think "or their representatives" is sufficiently implied in that wording that we don't need to make it explicit, but I absolutely will not oppose a footnote after "denied" along the lines of "either directly or through lawyers or other representatives"). Thryduulf (talk) 19:07, 4 June 2026 (UTC)

I also think that "or their representatives" is sufficiently implied. If we find that we're wrong, we can add the footnote later. WhatamIdoing (talk) 19:35, 4 June 2026 (UTC)

I like the wording, but can I suggest a slightly shorter version of the last sentence:
"A denial by itself does not make inclusion proportional, but if the accusation is due for inclusion then the denial should normally be included as well"
The subclause "and the accused has denied the allegation" is redundant, as the subject must have denied it for there to be a denial. -- LCU ActivelyDisinterested «@» °∆t° 20:47, 4 June 2026 (UTC)

"A denial by itself does not make inclusion proportional, but if the accusation is included, then any denial should normally be included as well"? WhatamIdoing (talk) 22:43, 4 June 2026 (UTC)

"any denial" is better than "the denial" as there isn't necessarily going to be a denial (e.g. there could be no response (yet), an admission or a "yes, but"). However, I worry that "any" will be misconstrued to allow denials by people other than the accused or their representatives. I can't bring the details to mind but I recall a case not too long back where a party leader or CEO or someone like that stridently said words to the effect of "Jee Bloggs strenuously denies these allegations" shortly after they were made with an "actually, I don't deny them", and then there are semi-regular occasions when Trump contradicts his cronies. The first denials might be worthy of inclusion but should not be regarded as automatically so. This means I still favour the apparently redundant formulation. Thryduulf (talk) 23:05, 4 June 2026 (UTC)

OK I see your point. I didn't see that the sentence is trying to define both the denier and the denial, which means the subclause isn't redundant. -- LCU ActivelyDisinterested «@» °∆t° 23:12, 4 June 2026 (UTC)

Either works for me.

On the other change I suggested, I think that "if the accusation is included" is better than "if the accusation is due for inclusion", because it doesn't matter why the accusation is included, and having an UNDUE accusation in the article without the denial is worse than having an UNDUE accusation with a denial. I could see a POV pusher saying "You said this accusation of harassment is UND, and that line in NPOV says denials should only be included if the accusation is DUE". WhatamIdoing (talk) 23:13, 4 June 2026 (UTC)

That's a good point and I agree. Thryduulf (talk) 23:43, 4 June 2026 (UTC)

So where are we at on this? I came on this talk page to raise this issue, having completely forgotten about this discussion. Are we ready to include some statement about events of no lasting consequence? Thebiguglyalien (talk) 17:59, 4 June 2026 (UTC)

@Thebiguglyalien very nearly I think, see my most recent comment. Thryduulf (talk) 19:09, 4 June 2026 (UTC)

Building off NicoR8's proposal and suggestions from Thryduulf, WhatamIdoing's modification from "due for inclusion" to "is included" (which forecloses the WP:Game interpretation where a POV pusher uses an undue finding to remove denial), and retaining "and the accused has denied the allegation" part, here's another version.

Also, for example, editors should consider the significance of an allegation or controversy in the context of the body of reliable sources on the subject. Isolated or short-lived reporting does not by itself establish that the matter is significant to the subject. Sustained coverage, follow-up reporting, or later discussion in reliable sources can indicate that inclusion is due. When sustained or later coverage is lacking, including it may give undue weight to a minor aspect of the subject. A denial by itself does not make inclusion proportional, but if the accusation is included and the accused has denied the allegation, the denial should normally be included as well. LeónGonsalvesofGoa (talk) 04:51, 9 June 2026 (UTC)

That looks good to me. Thryduulf (talk) 10:04, 9 June 2026 (UTC)

Do we want this to link to the WP:DENIALS policy? WhatamIdoing (talk) 17:24, 9 June 2026 (UTC)

That sounds reasonable to me. See revision below.

Also, for example, editors should consider the significance of an allegation or controversy in the context of the body of reliable sources on the subject. Isolated or short-lived reporting does not by itself establish that the matter is significant to the subject. Sustained coverage, follow-up reporting, or later discussion in reliable sources can indicate that inclusion is due. When sustained or later coverage is lacking, including it may give undue weight to a minor aspect of the subject. A denial by itself does not make inclusion proportional, but if the accusation is included and the accused has denied the allegation, the denial should normally be included as well (Wp:Denials). LeónGonsalvesofGoa (talk) 02:37, 11 June 2026 (UTC)

WP:DENIALS should be capitalised but other than that it looks good to me. Thryduulf (talk) 09:42, 11 June 2026 (UTC)

Religious affiliation and the limits of "Attribute statements of opinion"

When we attribute statements of opinion in contentious topic spaces, I see debates where editors insist on labelling every scholar or author by their background. For articles of religion, do we need to label each individual as 'a Muslim scholar,' 'a Catholic historian,' or 'an atheist critic,' to maintain neutrality?

If a scholar is writing from a position of academic peer-review, does such constant labelling actually violate NPOV by implying a inherent bias where none may exist? It can make articles incredibly cumbersome to read. At what point does the policy's requirement to attribute opinion stop being about clarity and now inadvertently creating bias? LeónGonsalvesofGoa (talk) 01:51, 12 May 2026 (UTC)

There are two primary types of "religious scholars"; those who are acting on behalf of a religion (regardless of whether they have any recognized authority within said religion) and those who engage in religious studies or theology from an academic perspective. I believe that, in a vast majority of circumstances, the former should be labeled while the latter is usually irrelevant unless their religion (or lack thereof) is deemed noteworthy by reliable sources. - ZLEA ^TǀC 02:08, 12 May 2026 (UTC)

For cases of the latter, we should also be careful to not confuse the scholar's field of study (if they specialize in a specific religion) from their practiced religion, as they may not always be the same. - ZLEA ^TǀC 02:12, 12 May 2026 (UTC)

I think it is intended to be a way of downplaying the source (Well, he would say that, wouldn't he?), but I think that it can be done appropriately, especially when the point is to highlight differences between two groups ("The Catholic theologian said...The Protestant professor said..."). WhatamIdoing (talk) 02:48, 12 May 2026 (UTC)

• Knowing when to attribute is sometimes complex, but we certainly wouldn't attribute everyone by faith automatically (ie. not simply because they are of that faith.) We don't attribute every person by nationality in discussions over nationalistic disputes, after all. My thoughts:

• If we're talking peer-reviewed papers or high-quality academic stuff, or generally people being cited for their expertise, you'd usually only attribute using their area of expertise, if anything. There are exceptions when a scholar has a particularly heterodox position and is mostly known and being cited for that position (which you'd probably establish with high-quality secondary sourcing), or for when high-quality secondary sourcing makes it clear that they're considered to be arguing from eg. religious perspective XYZ.
• If you're citing them via a high-quality independent secondary source, it often makes sense to just go with how the secondary source frames them.
• When citing an opinion piece by someone who lacks expertise... well, honestly, I wouldn't do that in the first place. But if we are, and they're talking about something that touches on their faith, then I feel it's reasonable to attribute them using that faith; when you're citing a statement of opinion by a non-expert, you need to contextualize their opinion.
• If a source is making WP:FRINGE claims it should always be attributed as clearly as possible (and probably not used at all outside of articles about that theory, per WP:ONEWAY.) Like, if someone is arguing that the earth is 6000 years old, the only context where we should even cite them is in articles about Young Earth Creationism, and in those contexts we would obviously have to identify them as such. This can get slightly dicey with some of the anti-evolution types (who sometimes try to obscure their position or claim they reached it independently from their faith, partially because doing so is necessary for them to get around the establishment clause in the US when they want to legislate in favor of their beliefs), but this can usually be cut through by relying on high-quality secondary sources; if they don't exist, we probably shouldn't be citing them at all anyway.
• The way they frame things in their own text also matters - eg. on The Bible there was just a bit of discussion over a source that was being cited for something eyebrow-raising, and when I dug it up the source was good but it was from a section where the author said basically "now, everything above is my expert opinion. But speaking as a Christian, and as far as my personal faith goes..." If we were citing that particular section we would attribute it with their faith because that's how they frame it. Reputable scholars writing scholarly texts that are also intended to be read by a general audience will often write things this way, and in that case how and whether to attribute might depend on what part we're citing. With lower-quality sources it sometimes gets dicier, but I think it's reasonable for editors to sometimes make use their own judgment and say "here, in this part, the author is clearly talking about their own faith and their own personal beliefs related to that faith, and we should attribute it as such; whereas in this other part they're talking from their expertise."
• Some sources are overtly and intentionally written as apologetics. Those should always be attributed as such.

I suspect that most disputes over whether to attribute come down to disagreements over what category to put someone in, which is usually best settled by looking at their expertise, their reputation, and how secondary sources treat them. --Aquillion (talk) 04:24, 17 May 2026 (UTC)

@Aquillion, I have been reflecting on your last response. You are right that blanket labelling is not what anyone is proposing. Overt apologetics, WP:Fringe claims, opinion pieces by non-experts touching on their faith, and passages where the author themself shifts from scholarly to confessional register are all places where attribution clearly earns its keep. @ZLEA's earlier two-category framework (acting on behalf of a religion versus engaging with it academically) maps onto your taxonomy, but the difficulty arises when editors treat open identification with a tradition as itself sufficient to push a scholar into the "acting on behalf of" bucket, even where they publish through independent academic presses, are subject to peer review, and are routinely cited by unaffiliated scholars in the field. Under WP:Rs it plainly is not, but talk-page practice often drifts that way. Suppose Professor AB openly identifies as a Christian and publishes a peer-reviewed work through an independent academic publisher, and Professor CD is sceptical of all religion and criticises AB's views. What I tend to see in practice is something like "Christian scholar AB states that Christians believe in XYZ. CD states that Christians do not believe in XYZ." AB gets labelled while CD's position is presented unmarked, even though CD's hostility to religion is just as relevant to how a reader should weigh the claim. The asymmetry is the problem. It is easier to label someone a follower and quietly discount their views than to label someone a critic and appear to undermine theirs, leading to an one-directional attribution that does not actually serve neutrality. LeónGonsalvesofGoa (talk) 12:00, 23 May 2026 (UTC)

In that example (which I've seen), I think we're treating the anti-religious POV as the default. It would be reasonable in such cases to add a description to either both or neither.

Another possible problem is that we're overfocused on WP:INTEXT attribution. If these are two generally held positions, then why are individual names being given at all? Alternatively, some statements about what Christians do/don't believe might need a completely different framing, e.g., "It is the official position of the Roman Catholic Church that abortion is evil, but most of their members in the US support abortion rights." WhatamIdoing (talk) 19:15, 23 May 2026 (UTC)