Wikipedia talk:Neutral point of view

From Wikipedia, the free encyclopedia

Shortcut: WT:NPOV

The project page associated with this discussion page is an official policy on Wikipedia. It has wide acceptance among editors and is considered a standard that all users should follow. Before you update the page, make sure that changes you make to this policy really do reflect consensus.

Are you in the right place?

For questions or discussions about the application of this policy to any specific article(s), please post your message at either the NPOV Noticeboard (any neutrality-related issue) or the Fringe Theories Noticeboard (undue weight given to a minority view).

The spoken word version of this page is part of WikiProject Spoken Wikipedia, an attempt to produce recordings of Wikipedia articles. To participate, visit the project page.

When starting a new topic, please add it to the bottom of this page, and please sign your comments with four tildes: ~~~~. This will automatically place a date stamp, which will allow us to maintain this page better.

1 FAQ pseudoscience discussion reopened
2 Map issues - India & China
3 Academic criticism
4 On Providing Justification for actions
5 Conflict of interest guidelines
6 Obvious edit
7 RFC at WP:NOR-notice
8 Undue weight in application to articles about pseudoscience
9 How to determine majority viewpoint?
10 Proposal to change wording of UNDUE to eliminate an effective meaning which is likely unintended
- 10.1 Options
11 Bias
12 Giving undue weight to primary sources not mentioned in any secondary sources
- 12.1 The "illustration" analogy
13 Scientific point of view

[edit] FAQ pseudoscience discussion reopened

Moved discussion to Wikipedia talk:Neutral point of view/FAQ#FAQ pseudoscience discussion reopened

[edit] Map issues - India & China

[edit] Academic criticism

I have encountered, of late, a troubling trend of people dismissing peer-reviewed academic criticism as an insignificant and non-notable perspective, and removing accounts of academic criticism from articles. This, combined with the tendency we display by default to have articles based on what comes up on Google bothers me, in that it seems to me to cheapen our coverage. We're an encyclopedia, and part of that means respecting academic perspectives.

Accordingly, I would like to add a quick note somewhere in this policy, parallel to the declaration in WP:RS that "Academic and peer-reviewed publications are usually the most reliable sources when available" that notes that academic and scholarly views are a particularly important viewpoint to make sure is represented when possible.

Thoughts? Phil Sandifer (talk) 20:27, 11 December 2008 (UTC)

I think this is a good thought. It is common good practice for sources of greater reliability to be accorded greater weight. For example, a current university textbook from Oxford University Press detailing ancient Christian history is usually going to be accorded much greater weight than an article in USA Today about the social climate of the early Christian church. Leaving aside questions about the accuracy of news reporting on academic subjects, it is not that USA Today is unreliable, but rather that a work from one of the most reputable academic presses will be considered significantly more reliable and appropriate for the topic than a newspaper article. Vassyana (talk) 01:27, 20 December 2008 (UTC)

I find myself in perfect agreement with the sentiments and concerns expressed above by Phil and Vassyana. Jayen466 01:36, 20 December 2008 (UTC)

I also agree. That said, when we encounter problems such as these, what's the best strategy for dealing with it? Post to the NPOV noticeboard? Something else? --Akhilleus (talk) 01:56, 20 December 2008 (UTC)

Do you have any examples? This seems contrary to my experience. And isn't RS more the place for this? II | (t - c) 02:28, 20 December 2008 (UTC)

The characterization of fringe theories provides many examples. On Jesus myth hypothesis (which asserts that Jesus didn't exist), for instance, there was an editor who insists that quotes from books published by the Oxford University Press shouldn't be used, because the press (allegedly) has a conflict of interest when dealing with religious matters--it publishes bibles, you see? I'm not joking, the full post can be seen here. At the time there was a post to the reliable sources noticeboard, I think, but when this is a systematic problem on an article--that is, editors favor "popular" sources such as newspapers, magazines, mass-market publishers over academic journal articles and monographs, what's the appropriate way to deal with the issue? --Akhilleus (talk) 03:12, 20 December 2008 (UTC)

I don't see the need for this addition as it is covered in WP:SOURCES. It is already stated that the policies should not be read in isolation, and past experience suggests that when 2 or more policies cover the same detail, over time there can be changes to one which brings the them into conflict. So instead of clarity and harmony we end up with confusion and disharmony. --PBS (talk) 11:01, 29 December 2008 (UTC)

Quoting WP:RS#Scholarship sometimes helps. Or a post to WP:RS/N (with fingers crossed). Jayen466 23:09, 6 January 2009 (UTC)

[edit] On Providing Justification for actions

Is there any rule about providing justification for actions occuring in History articles. For example can the words "in response", "because of" be used, with the reasons provided, however obvious and straightforward they may be, may come into question. For example a sentence like " 2 +2 =4 because according to definition of real numbers etc." is different from " Jack shot at Tom because of Tom's threat".

To give a better example how about this "Jack had said that he would shoot Tom but later changed his mind. In a few days Tom threatened Jack, in response Jack shot at Tom". Here with using the words "in response", the reader is left confused about weather Jack shoot Tom because of his threat, or because Jack wanted to shoot Tom anyway. This, I believe, goes beyond NPOV, unless it is rephrased " Jack claimed that he shot Tom because he was threatened". Is there any policy regarding giving reasons to actions and stating them as matter of fact , instead of simply trying to state the facts. The facts should speak for themselves about the reasons if given in a chronological order Kakaka79m (talk) 17:57, 27 December 2008 (UTC)

The key is whether there is a reliable source that discusses the cause of the event. If so, then there is no problem with our including the "because of" in our article. In other words, if a reliable source says someone did something "because of" something else (event to cause), then we can note this fact (citing the source)... but if no reliable source attributes event to cause then we can not do so on our own initiative... to do so would violate WP:NOR. Also, if some other reliable source presents an different cause for the event, we need to mention that view as well. Blueboar (talk) 16:55, 28 December 2008 (UTC)

[edit] Conflict of interest guidelines

I'm trying to get the conflict of interest guidelines rewritten to be clearer in terms of what is allowable from people with conflicts of interest. At the moment they seem to allow most edits, but there are also several parts saying COI editing is strongly discouraged. It is not clear whether COI editing is generally unacceptable or only when it results in bad edits. I have written a proposed revision of the guidelines, but because of my own conflict of interest I really don't want to make any kind of change without consensus. Since obviously a lot of conflicts of interest result in POV problems, people who monitor this should put forward their thoughts at User talk:Helenalex/coirewrite or on the conflict of interest guidelines talkpage. --Helenalex (talk) 05:31, 29 December 2008 (UTC)

[edit] Obvious edit

[1]. I thought this one is a no-brainer, but apparently User:John doesn't think so. Searching for "balanced" sources is not okay in, for example, the evolution article if you are seeking to balance it with creationism. Instead, we are supposed to chose reliable sources. ScienceApologist (talk) 20:43, 31 December 2008 (UTC)

Agree. Additionally, it's problematic given that "balance" isn't defined until later in the article, and the definition is not common usage outside Wikipedia. --Ronz (talk) 20:50, 31 December 2008 (UTC)

disagree. My watchlist popped up User:John's reversion from "A careful selection of reliable sources" back to "A balanced selection of sources". The reversion resonated with me. Too often, I've seen a careful (unbalanced POV-pushing) selection of reliable sources. To address the use of "balanced" before its definition, perhaps the A simple formulation subsection needs to be moved from the Explanation of the neutral point of view section to the Achieving neutrality section so as to follow the subsection on balance which explains "Neutrality weights viewpoints in proportion to their prominence". That seems to me to be be a more logical location in any case, since formulation relates to achieving moreso than to explaining. After making that change in subsection location, I suggest changing the wording to read "A careful and balanced selection of reliable sources".-- Boracay Bill (talk) 00:32, 2 January 2009 (UTC)

Let's choose a word other than "balanced". How about "representative"? The issue is that a balance tends to imply equal amounts which is not inimical to our task. ScienceApologist (talk) 08:40, 2 January 2009 (UTC)

I was thinking "proportional" before I say the suggestion for "representative." Either is better than "balance" unless someone wants to try all the reorganizing that Boracay Bill suggests. --Ronz (talk) 21:12, 5 January 2009 (UTC)

[edit] RFC at WP:NOR-notice

A concern was raised that the clause, "a primary source may be used only to make descriptive claims, the accuracy of which is verifiable by any reasonable, educated person without specialist knowledge" conflicts with WP:NPOV by placing a higher duty of care with primary sourced claims than secondary or tertiary sourced claims. An RFC has been initiated to stimulate wider input on the issue. Professor marginalia (talk) 06:30, 3 January 2009 (UTC)

[edit] Undue weight in application to articles about pseudoscience

In application to an article about a theory which is generally considered pseudoscience, does WP:Undue weight mean that the majority of the article should be specifically about criticisms? At Talk:Bates method, an editor has argued for that, citing UNDUE. PSWG1920 (talk) 21:17, 3 January 2009 (UTC)

WP:ASF. The majority of the article should provide context to the reader as to why the particular pseudoscientific idea is notable and what the facts are surrounding it. In general, it's best to avoid a "criticisms" section outright and instead incorporate factual context into articles. By the way, you probably want to ask this question at WP:NPOVN to get more visibility. ScienceApologist (talk) 21:49, 3 January 2009 (UTC)

I don't think that's right. To "incorporate factual context" as suggested above is really to synthesize a debunking, to editorialize, to frame the fringe idea in the scientific point of view, as opposed to presenting it on its own terms, in a neutral point of view. I agree with WP:ASF, but to choose facts to put a fringe idea into "context" is to distort the idea itself, and contravenes WP:NOR. What's better is to find reliable sources that assert facts and opinions about the topic idea itself, and report those. Don't go looking for contrary facts, just report what's been published about the topic idea itself. If it hasn't elicited much critical reaction, you won't have to say much. Dicklyon (talk) 22:02, 3 January 2009 (UTC)

I'm looking for insight on how WP:UNDUE applies to a fairly general type of article. Ideally the policy itself should better explain that. PSWG1920 (talk) 22:14, 3 January 2009 (UTC)

The policy is intentionally vague in order to accommodate a wide range of subjects. The basic idea is that Wikipedia shouldn't try to artificially "balance" ideas to make them appear more respectable or more considered than they actually are. ScienceApologist (talk) 22:16, 3 January 2009 (UTC)

I'm unclear on what you're asking. Do you mean for the type "generally considered pseudoscience"? Or something more general? It seems pretty clear to me where the page starts out:

An article should not give a minority viewpoint on its subject as much of or as detailed a description as a more popular view, and will generally not even mention a viewpoint which is in the tiny-minority relative to the overall subject. For example, the article on the Earth does not mention modern support for the Flat Earth concept, a view of a distinct minority.

I agree with this interpretation and example. But some editors fail to note that it says "on its subject" and "relative to the overall subject". The "subject" here is the article topic, if I read it correctly. In other words, an article on Flat Earth would represent all the viewpoints on the flat earth idea, not be dominated by also including all the stuff you can find on why a round Earth is the mainstream view; that material can be in the Earth aricle, can be linked, can be mentioned, but is pretty much off-topic except where it's part of the published commentary on the flat earth idea. In fact, Flat Earth looks like a good example of how fringe ideas should be presented. Dicklyon (talk) 22:24, 3 January 2009 (UTC)

Actually, "on its subject" and "relative to the overall subject" were added by me earlier today in an attempt to make more clear what seemed to be the intended meaning per the example given. PSWG1920 (talk) 22:30, 3 January 2009 (UTC)

Oh dear me! I should have checked. Obviously I agree with you here, but we'll have to wait and see if that's the consensus. Dicklyon (talk) 22:33, 3 January 2009 (UTC)

In regards to what I'm asking, see the talk page discussion linked to at the top of this thread, and scroll down to the comments I referred to. Basically I want to know if that is a valid interpretation of UNDUE. PSWG1920 (talk) 22:36, 3 January 2009 (UTC)

Indeed, that's what everyone fights about. I lean more toward your interpretation than to Ronz's. Dicklyon (talk) 22:49, 3 January 2009 (UTC)

I don't really see how the majority of any article even can be specifically about criticisms unless it's a "Criticisms of..." type article. Otherwise the reader would be left wondering what exactly is this thing which is being criticized. PSWG1920 (talk) 16:14, 4 January 2009 (UTC)

I think I see this more clearly now. The opening paragraph of UNDUE states that "An article should not give a minority viewpoint on its subject as much of or as detailed a description as a more popular view". Taken literally (which some editors do), this would necessitate that an article about a minority viewpoint devote most of its space specifically to the contrasting majority view, i.e. criticisms. (Previously it said "articles that compare views", but an article about a minority viewpoint would still fall under that description as it necessarily compares it to the majority view.) The question is, is that the intent? The example given leads me to believe that this clause is in fact intended for general subject articles. If so it needs to be amended. PSWG1920 (talk) 19:22, 4 January 2009 (UTC)

What about adding "which is not specifically about a viewpoint"? So (what is currently) the first paragraph would say "An article which is not specifically about a viewpoint should not give a minority viewpoint on its subject as much of or as detailed a description as a more popular view". The example given, Earth, is clearly not an article about a viewpoint, so I think that would fit with the intent. PSWG1920 (talk) 19:39, 4 January 2009 (UTC)

Perhaps it'd be best if everybody editing pseudoscience articles just calmed down for a while and waiting for the upcoming ArbCom decision on the topic. But, both as previous history goes and comments of some arbitrators in ongoing problem, NPOV policy means not attacking a fringe idea, but absolutely including and sourcing why the scientific community considers it pseudoscience throughout the article as appropriate. Not having the criticism there gives major undue weight to a minority view.DreamGuy (talk) 22:44, 3 January 2009 (UTC)

[edit] How to determine majority viewpoint?

Pardon if this issue is covered elsewhere, but I cannot find it: Is there a policy or discussion somewhere on how to determine "(i)f a viewpoint is in the majority" or "in the minority"? I did not see it in the FAQ. The project page only states to use "commonly accepted reference texts...." But what if both sides of an issue can give references? This occurs with the historical accuracy of accounts found in ancient texts accepted as scripture, such as Noah's Ark. Is there a recommended process that should be followed? Is it necessary to start listing scholarly references? How can WP editors sort through such issues? Thanks, SteveMc (talk) 02:25, 4 January 2009 (UTC)

If there's not an overwhelming unbalance of viewpoints in sources, treat them as balanced. Dicklyon (talk) 02:40, 4 January 2009 (UTC)

Sure, but how do we prove sources? Quantifying them? Qualifying them? SteveMc (talk) 17:33, 4 January 2009 (UTC)

WP:RS gives guidance about evaluating the quality of sources, and at a more basic level WP:V requires a basis in reliable third party sources, with selfpublished and questionable sources being used carefully, subject to restrictions. If there are good academic sources giving different expert views, we show both views and can aim to evaluate the degree to which each has wide acceptance. Evidence can be discussed on the article talk page if there's a dispute. . . dave souza, talk 17:53, 4 January 2009 (UTC)

Thanks all, I found enough to help me at the locations mentioned above, leading me to the pageWP:SYNTH on WP:NOR. SteveMc (talk) 22:26, 4 January 2009 (UTC)

[edit] Proposal to change wording of UNDUE to eliminate an effective meaning which is likely unintended

From WP:UNDUE:

An article which compares views should not give a minority viewpoint on its subject as much of or as detailed a description as a more popular view, and will generally not even mention a viewpoint which is in the tiny-minority relative to the overall subject. For example, the article on the Earth does not mention modern support for the Flat Earth concept, a view of a distinct minority.

I suggest changing "which compares views" to "which is not specifically about a viewpoint".

An article which is not specifically about a viewpoint should not give a minority viewpoint on its subject as much of or as detailed a description as a more popular view, and will generally not even mention a viewpoint which is in the tiny-minority relative to the overall subject. For example, the article on the Earth does not mention modern support for the Flat Earth concept, a view of a distinct minority.

The reason I am suggesting this is that I believe the current wording has an effective meaning which is likely unintended, though some will disagree that it is unintended. An article about a minority viewpoint meets the definition of an article which compares views, since it will necessarily be contrasted with the majority view. The current wording therefore has the effect of saying that an article about a theory which is generally considered pseudoscience should devote most of its space specifically to the majority view, i.e. criticisms. But in my opinion, this would normally be unfeasible, lest the reader be left wondering what exactly it is which is being criticized. I would further point out that the example given, Earth, is not really an article which compares views, which is why I doubt that this effective meaning is intended.

In interest of full disclosure, it was this discussion which got me started here. PSWG1920 (talk) 21:49, 4 January 2009 (UTC)

You seem to be trying to fix something that isn't broken. What you are describing would essentially make articles on fringe topics be POV-forks written from the views of believers in those topics. The point here is that fringe topics be explained, but in no way give the idea that thy are anything but fringe beliefs. I don't think the current wording of the policy is unclear. DreamGuy (talk) 22:35, 4 January 2009 (UTC)

I would say that the current wording if taken literally makes an article on a fringe topic into a "Criticisms of..." type article. Of course the criticisms should be included and presented as the majority viewpoint (as is explained later in the section), but that does not mean that said criticisms should necessarily encompass the majority of the article's space. PSWG1920 (talk) 22:44, 4 January 2009 (UTC)

From my experience, there seem to be more problems with regards to removing or not including criticism in fringe articles than in the other way around. We could add something like "Articles on a fringe topic may go into more detail about the beliefs and views of the fringe topic's proponents, but care should still be taken to explain the mainstream view of the fringe topic, notable criticism, and not to make the fringe topic seem more prominent or accepted than it is." - but the suggestion cuts articles on fringe subjects free of undue weight completely. Alternatively, we could borrow a quick summary of WP:FRINGE, the relevant expansion of NPOV into handling of fringe subjects. Shoemaker's Holiday (talk) 01:23, 5 January 2009 (UTC)

The later statement that "Minority views can receive attention on pages specifically devoted to them—Wikipedia is not a paper encyclopedia. But on such pages, though a view may be described, the article should make appropriate reference to the majority viewpoint wherever relevant, and must not reflect an attempt to rewrite majority-view content strictly from the perspective of the minority view." shows how undue weight applies to articles about fringe topics. My suggestion did not involve changing the meaning there at all. PSWG1920 (talk) 01:31, 5 January 2009 (UTC)

I have no doubt it wasn't meant to, but it'd have probably been used that way. We'd probably be better off saying more here than simply giving an exception to a rule that should still (at least partially) hold. For instance, the reader of Holocaust denial should make it very clear what the widely-accepted facts being denied are, and perhaps some of the evidence for them, as a basic requirement of putting the minority view in context. Shoemaker's Holiday (talk) 02:53, 5 January 2009 (UTC)

Can you look at the discussion I linked to above and see if you think that is a valid interpretation of UNDUE? PSWG1920 (talk) 03:03, 5 January 2009 (UTC)

What do you think of this (I'll get to the discussion in a bit)

An article should not generally give a minority viewpoint as much of or as detailed a description as a more popular or mainstream view on its subject, and will generally not even mention a viewpoint which is in the tiny-minority relative to the overall subject. For example, the article on the Earth does not mention modern support for the Flat Earth concept, a view of a distinct minority. However, for articles specifically related to the minority views, a bit more latitude is available to explain what the minority views are, but it should be clear that the minority view is a minority view, and the article should explain the mainstream view. In some cases, the evidence backing the mainstream view may be necessary to put the minority view in the overall context of mainstream thought on the issue. For instance, Holocaust denial explains what well-accepted facts on the Holocaust are generally denied by the holders of that fringe viewpoint, and briefly explains the types of evidence and the sheer amount of said evidence that backs the mainstream interpretation. Shoemaker's Holiday (talk) 02:53, 5 January 2009 (UTC)

The key there is brief. But sometimes a brief statement of evidence against a fringe idea can come across as ridiculous, and actually make the discussion of the fringe idea look credible by comparison. In the example cited, Holocaust denial, there's a small paragraph in the middle that looks absurdly out of place: "According to researchers Michael Shermer and Alex Grobman, there is a "convergence of evidence" that proves that the Holocaust happened. This evidence includes...:." The article would actually be improved by removing this attribution of the mainstream view to two unknowns, I think; there's plenty of other information in the article that clarifies that the dominant view of history is not in agreement with the fringe idea; we don't need to review how historians work for that idea to be clear. Dicklyon (talk) 03:34, 5 January 2009 (UTC)

[edit] Options

As I see it, there are basically two choices, which I have illustrated. The first would make explicit a logical implication of a literal reading of what is already stated in WP:UNDUE. The second would remove that implication. Once one or the other is settled on, it could then be worked on further if necessary. PSWG1920 (talk) 19:56, 5 January 2009 (UTC)

I've tried what may be a better compromise: I moved the paragraph on articles on minority viewpoints upwards, so that it clarifies and expands on the point just raised. Shoemaker's Holiday (talk) 21:13, 5 January 2009 (UTC)

Thanks, I think that was a good move, but the basic problem still exists. Does the requirement that the majority viewpoint get more space apply to an article about a minority view or not? If it does then that should be made explicit; if it doesn't then the earlier statement should be qualified. PSWG1920 (talk) 21:19, 5 January 2009 (UTC)

These additions are fine but still do not solve the problem. The NPOV faq does not address this particular issue as far as I can tell, and whatever WP:FRINGE says, it's only a guideline, and therefore won't override the literal meaning of a policy. PSWG1920 (talk) 22:00, 5 January 2009 (UTC)

I'd say not necessarily, so long as care is taken not to present disputed views of the minority as undisputed fact. But we need to be a bit careful how we clarify that. Shoemaker's Holiday (talk) 22:03, 5 January 2009 (UTC)

Of my two alternatives, which one would you favor? PSWG1920 (talk) 22:16, 5 January 2009 (UTC)

I find both problematic: Alternative 1 says criticism must dominate. While writing on some subjects is dominated by criticism, this is not always true. Our article on Flat Earth, for instance, quite rightly concentrates on the history and anthropology of the idea. Our article on HeadOn (an American homeopathic product), concentrates on a commercial which forms the product's only real claim to notability (evidently, anyway - I neither own a TV or live in America, but there is strong evidence for it having become a meme in America). Other subjects, such as Holocaust denial, have almost all the reliable sources devoted to criticising and explaining problems with that view. In general, I'd probably say that there should be sufficient criticism to make sure the reader is not misled, whatever that means in the particular article. It may be that a couple sentences are enough; it may be necessary to make it the majority of the article in order to have it appropriately balanced. To some extent, our best-quality sources can lead us, but this does mean judging quality of sources appropriately, and being suitably wary of "crank" journals (ones set up specifically to promote a viewpoint) when judging sources. Shoemaker's Holiday (talk) 02:57, 6 January 2009 (UTC)

I like your recent changes, but the first paragraph of UNDUE if taken literally still implies that any article on a fringe topic should devote most of its space to criticisms. PSWG1920 (talk) 04:00, 6 January 2009 (UTC)

Eh, the second paragraph clarifies, I think that too much clarification and we lose the underlying message. Shoemaker's Holiday (talk) 04:10, 6 January 2009 (UTC)

In my experience, as long as the first paragraph is written the way it is, some editors will attempt to apply its literal meaning to any fringe article they find, or at least tag it for a neutrality dispute on that basis. PSWG1920 (talk) 05:00, 6 January 2009 (UTC)

I still like the wording "an article which is not particularly about a viewpoint" for the first paragraph. I think that is likely the intended distinction from the situation described in the next paragraph. PSWG1920 (talk) 06:43, 6 January 2009 (UTC)

[edit] Bias

wikipedia will never be fully neutral. humans have a tendency to think of themselves as superior to other animals. i found many examples of bias based on species difference and scientific ignorence.--00:42, 6 January 2009 (UTC)Guppy22 (talk)

[edit] Giving undue weight to primary sources not mentioned in any secondary sources

I would like us to think about whether it would make sense to add a sentence or two about undue use of primary sources in the WP:Undue section.

The principle I am trying to get at is this: If there is a substantial body of secondary sources available on a topic, then Wikipedians should feel free to use and cite any primary sources (e.g. affidavits, self-published websites, quotations from an author's literary works, etc.) that are cited in these secondary sources. However, we shouldn't really bypass the body of secondary sources to select and cite primary sources ourselves, unless we can demonstrate that this primary source (i.e. affidavit, self-published website, quotation from an author's literary works, etc.) has been used in this way in a secondary source (news source, scholarly publication, etc.).

In my experience, misuse of primary sources, in a way that is unsupported by the existing secondary literature, can cause Wikipedia articles to deviate significantly from the picture presented in the most reliable sources. That is a NPOV and due weight problem. (Such primary-source use has arguably been a major factor in causing the fourth Scientology-related arbcom case in as many years: [2] [3]).

There has been a recent discussion at NOR talk concerning this topic – http://en.wikipedia.org/wiki/Wikipedia_talk:No_original_research#Primary_sources:_The_novel_example – but it has been stated there that this might be more appropriately addressed here. Thoughts? Jayen466 17:43, 6 January 2009 (UTC)

I have a feeling that what you are really discussing are situations when an article quotes statements from a primary source out of context (which is a form of Original Research). Correct me if I misunderstand. Blueboar (talk) 18:14, 6 January 2009 (UTC)

Yes, sometimes this concerns quotes out of context. This is certainly the case in one current discussion I'm having (third opinions welcome). Another, slightly different case I recall from German Wikipedia was an editor quoting 300 words or so of an obscure speech that made Rudolf Steiner (who was anything but) sound racist. (German WP subsequently introduced a policy against Wikipedians using primary sources directly whenever there is a substantial body of secondary literature available. It's been like that over there for the past 2 or 3 years, and it's been very successful and uncontentious.)

The other example I gave over at WP:NOR was the Bible – I could quote 10 bible passages damning homosexuals and fornicators to hell, or stoning, and make it sound like an entirely bloodthirsty document. But those quotes are not the main passages discussed in secondary sources, and if we allow this sort of thing what we arrive at is a caricature. With controversial groups and figures that have attracted online communities dedicated to disparaging them, as in the example of Scientology, the risk of our ending up with a caricature in Wikipedia is of course much greater, as Wikipedians working in such areas typically seem to be more familiar with the online material than with the scholarly literature. Jayen466 19:05, 6 January 2009 (UTC)

(outdent) Is the concern here about using material from primary sources out of context (as determined by fair reading of the whole source and perhaps any secondary sources relevant to it), or about using a primary source which itself has not been synthesized from, or referenced in, (or refuted by,) one or more secondary sources? Or both? The first is a case of accurate (or not) representation of the source and is a editorial issue, but one which can evolve into a disruptive editing issue. The second is an issue about reliability of sources, in particular notability, keeping in mind that peer review can act in some sense as a proxy for notability in some cases. Baccyak4H (Yak!) 19:47, 6 January 2009 (UTC)

It is a number of those things. I'll try and give some examples.

An affidavit filed with a court and available on a private website alleges that notable living person X, who's been the subject of hundreds of news articles, physically abuses his employees, hitting them in the face, kicking them in the groin, etc. No newspaper or other secondary source has reported on this affidavit and the claims made in it. My conclusion: The affidavit is not a good source for the person's BLP.
Attack sites on the Internet feature isolated quote X which sounds suggestive of Y. There is no evidence of this quote being used by reliable sources (news media or scholarly publications) to ascribe Y to the author of X, even though there is a plethora of reliable published sources. My conclusion: If no RS can be found, don't use the quote. Focus on what RS focus on.
A number of court documents are available on private websites. The matters described in these court documents have not been covered by any published source, even though there are published sources discussing the entities concerned. My conclusion: If no RS can be found that deals with these matters, don't use the court documents as sources in WP.
Publication A by notable author B contains passage C that a Wikipedian finds weird, outrageous, immoral, whatever. Author B has been discussed in hundreds of newspaper articles and dozens of scholarly publications. None of them mention passage C as representative or otherwise remarkable. My conclusion: Don't reproduce passage C in the WP article on author B if no secondary source can be found that mentions it. Focus on what reliable secondary sources focus on.

Those are some real-life examples. Jayen466 20:01, 6 January 2009 (UTC)

Thanks. These are all interesting cases and they have different issues. There are some important things to flesh out here, I would think. Baccyak4H (Yak!) 20:58, 6 January 2009 (UTC)

I'd suggest that while they are all different, what they have in common is that they seek to use a primary source without the back-up of a body of secondary sources according that primary source the same significance. That could be one obvious way of addressing it. Can you think of others? Jayen466 23:20, 6 January 2009 (UTC)

[edit] The "illustration" analogy

I tend to see (direct use) of primary sources as a kind of illustration. This fits well with images (or other media sources) that are used as illustrations and over at WP:NOR have their separate section to explain how these sources are to be approached NOR-wise (WP:NOR#Original images), a description that without much ado could be fitted for other primary sources like direct quotes.

I adopted and developed this idea in essays and proposals like Wikipedia:Use of primary sources in Wikipedia, Wikipedia:Wikipedia is a tertiary source and Wikipedia:Sources - SWOT analysis.

My idea is that ideally an article would have 30 to 50 percent of its surface devoted to "illustrations". For some topics illustrations would be mostly images (e.g. The Gates); for others mainly quotes (e.g. Tacitus). Going over 50% of the surface of an article being illustration would call for a move of most of it to commons (images) or wikiquote (quotes), or even wikisource for some. Less than approx. 20% would not make it to "Good Article" for example, and for a "Featured Article" approx. 30% would be a minimum. Well, this would be my answer to Jayen's question/suggestion above. In fact this idea includes Jayen's suggestion derived from German Wikipedia, without therefore being anti-primary source (...or anti-illustration). --Francis Schonken (talk) 20:29, 6 January 2009 (UTC)

[edit] Scientific point of view

Why is the issue of scientific point of view not discussed in this article? --Wet dog fur (talk) 15:51, 7 January 2009 (UTC)

The notion that there is such a thing as "Scientific point of view", which is distinct from neutral point of view and must be balanced (presumably by other points of view that reject science) does not enjoy consensus on Wikipedia. The question is under arbitration at the moment, in the case Fringe science.

A proposed principle in that case reads:

Science is not a point of view

3) While scientists have points of view, scientific inquiry itself is a methodology, and cannot hold one. That coverage of a topic is primarily scientific does not prevent it from being (nor obviates the need to be) neutral.

Hope that helps. --TS 16:57, 7 January 2009 (UTC)

And since NPOV depends on things like verification with reliable sources, not giving undue weight to fringe theories, the SPOV (if there was one, but like TS says, it really doesn't exist) equals the NPOV. OrangeMarlin ^{Talk• Contributions} 17:07, 7 January 2009 (UTC)

It seems that Scientific consensus is a point of view; also see Wikipedia:Scientific consensus. It is what is usually taught in science classes at universities. This issue should be discussed in the NPOV article. --Wet dog fur (talk) 17:28, 7 January 2009 (UTC)

Wikipedia:Scientific consensus is an essay started by Ed Poor. As I remarked above, this matter is under active arbitration. --TS 18:51, 7 January 2009 (UTC)

And Ed's POV is well-known on this project. Scientific consensus misrepresents how science develops theories and research. We don't sit around and take a vote. BTW, where is under active arbitration? OrangeMarlin ^{Talk• Contributions} 19:15, 7 January 2009 (UTC)

For the ongoing arbitration case, see Wikipedia:Requests for arbitration/Fringe science. --TS 19:21, 7 January 2009 (UTC)

After the arbitration case, it may be beneficial to either discuss the issue of scientific consensus in the NPOV article or to at least mention why it is not being discussed, that is so people do not wonder why it is not included. --Wet dog fur (talk) 19:32, 7 January 2009 (UTC)