Wikipedia talk:Neutral point of view/Fact disputedfact value

Situation
Currently WP:NPOV reads as follows (in part):

A simple formulation
Assert facts, including facts about opinions — but do not assert the opinions themselves. By "fact" we mean "a piece of information about which there is no serious dispute." For example, that a survey produced a certain published result would be a fact. That there is a planet called Mars is a fact. That Plato was a philosopher is a fact. No one seriously disputes any of these things. So we can feel free to assert as many of them as we can.

By value or opinion, on the other hand, we mean "a piece of information about which there is some dispute." There are bound to be borderline cases where we are not sure if we should take a particular dispute seriously; but there are many propositions that very clearly express values or opinions. That stealing is wrong is a value or opinion. That the Beatles were the greatest band in history is a value or opinion. That the United States was wrong to drop the atomic bomb over Hiroshima and Nagasaki is a value or opinion.

Problem
While the implication is that a value-judgment is not a fact, this is not clearly stated. This leads to some confusion on the application of WP:NPOV in that some believe that if a value appears to be predominant, it may be stated in the plain voice of the editor (as facts). This is contrary to the spirit and guidelines of WP:NPOV, WP:V, and WP:NOR. On the contrary, a value ascribed to something is never a factual property, disputed or not.

Proposed solution

 * Insert the following in the "A simple formulation section" in the second paragraph:
 * "Any proposition which is obviously a value-judgment should be expressed as a cited opinion."


 * Insert the following table at the end of the section:

Comments
Please insert any comments you have in this section.


 * I disagree entirely. If the value is one that is only disputed by an insignificant minority, it need not be cited directly.  This agrees with Jimbo's reasoning, as cited in WP:NPOV.  We shouldn't be concerning ourselves with the opinions of insignificant minorities.  Avoiding citing as fact an opinion that only such a minority disputes would make Wikipedia the most indirect encyclopedia in the world, looking like it is avoiding accepting responsibility for any opinions.  Only if a value judgement is significantly disputed do we need to give it a prose attribution.  This is already covered adequately in existing language, so I see no reason to change the policy. JulesH 11:42, 3 June 2007 (UTC)
 * I agree with the basic approach and like the table. Suggest a slight rewording in a couple of places. (a) Suggest calling the first row an "undisputed fact" to distinguish from "disputed fact". (b) Suggest dividing the third row into two, "opinions" and "aesthetics/morality." I fully agree with Fourdee that we need to word things carefully here to avoid creating the impression that Wikipedia requires people to believe in a fact/value distinction or endorses the idea that morality is solely a matter of opinion with no objective or natural basis. I believe careful wording can achieve the same result while avioding endorsing a particular philosphical view.  Finally, I believe that WP:NPOV has the exactly the effect User:JulesH states of making Wikipedia the most indirect encyclopedia in the world. If we don't wish such a result, we should abandon this core policy and take editorial slants and positions like other encyclopedias do. Best, --Shirahadasha 13:59, 3 June 2007 (UTC)
 * I think we need to recognize the is/ought distinction explicitly, and not confuse it with the certainty/uncertainty of facts. Wikipedia shouldn't be taking 'ought' positions ever, and when it comes to facts, we should only be taking the position of our sources. Night Gyr (talk/Oy) 15:27, 3 June 2007 (UTC)
 * User:Fourdee, I'm disappointed that you didn't reference our discussion (bottom of linked section) despite my explicitly asking you to. I continue to strongly believe that your approach, which would ask for sourcing in the face of every subjective statement, is awkward and undesirable.  If there is no significant dispute then even a value judgment/statement of worth should be stated plainly and directly.  For example: "The United States is a powerful country with an important impact of international affairs", "Gengis Khan was a ruthless warlord", "The Middle East is a complex region with strategic military and economic value".  "Important", "powerful", "ruthless", "complex", etc. all express subjective opinions.  While we could preface each statement with "X believe", doing so is undesirable when there is no significant opposing viewpoint, and may even lead to the false impression that there is an opposing viewpoint.  NPOV is about how to deal fairly with disputed viewpoints.  Trying to use it to eliminate all subjective language not directly attributable to third parties, even in the absense of significant disputes, is undesirable.  Dragons flight 16:14, 3 June 2007 (UTC)
 * If you're using subjective words, it might be better to use clearer, more objective and narrow language. 'important' doesn't say much. Better to just say why: show, don't tell. 'influential in international affairs' 'considered a superpower by scholars' 'Genghis Khan's tactics gave him a reputation for ruthlessness' Each one gives more information than simply saying 'important' or 'ruthless' and thereby avoids making troublesome subjective judgements rather than sourceable claims. Night Gyr (talk/Oy) 16:24, 3 June 2007 (UTC)
 * I don't consider it to be clearer or more objective to use phrases like "by scholars" when it ignores everyone else that also agrees, or to cast things in the circumspect form "X was believed to be Y" when "X was Y" is also accurate and undisputed (e.g. "The Earth has a reputation for being round" is certainly not better than "The Earth is round"). Yes, one could add additional clarifiers to explain why something is "important", "powerful", etc., but that doesn't remove the utility of subjective language (for example, your "influential" is still a subjective term, only a more specific one).  Let me put it another way.  We may be able to disagree about when an acorn becomes a tree, but if I point to a mighty oak then we all know it is a tree.  In the same way, we may disagree in specific cases whether X is powerful, ruthless, intelligent, barbaric, old, efficient, etc., but that doesn't mean that there aren't also cases that one can point to where everyone agrees that the subjective language fits.  Dragons flight 17:24, 3 June 2007 (UTC)
 * I'd rather say "20 meters tall" than "big," though, and whatever labels we apply still need to be backed up by our sources. Night Gyr (talk/Oy) 19:01, 3 June 2007 (UTC)
 * Labeling Gengis Khan with the pejorative "ruthless" is definitely not appropriate (already clearly discouraged by the examples in WP:NPOV). I think the policy of the present NPOV is clear, that "some propositions are very obviously values" and that values are not facts.  Whether the earth is round is a fact (disputed or not), whether the earth is good, or powerful, or important, is simply not a fact.  It is not phrasing things neutrally to apply the voice of the article editor to ascribed values that cannot be facts.  The clarification in policy outlined above wouldn't really make any difference to non-disputed articles, since someone has to raise the issue for it to come into play.  However the above table would make shorter work of cases where there is a dispute in wording.  It should be enough to point out that something is not phrased in a neutral voice to have it removed or cited.  Fourdee 21:28, 3 June 2007 (UTC)
 * Or it would promote tortured prose in cases where there is no dispute and no need for clarification. Encyclopedias still ought to strive for compelling prose (one of the FA criteria), and I am against adding constraints on writing and presentation style even in cases when there is no motivation for them.  In my mind the litmus test for the need to apply NPOV and use the voice of published authority is whether there is more than one view that needs to be considered.  If you can't identify a credible alternative view than making undisputed subjective statements in the voice of the article is (and should be) fine.  Dragons flight 22:35, 3 June 2007 (UTC)
 * On reflection, I am inclined to agree that undisputed values can at least sometimes be treated the same as undisputed facts, and attribution would always be necessary only in the presence of a dispute regarding values. As to the issue of using only opinion that is "significant", if it appears in a reliable source, that strikes me as enough, although weight would be another consideration in how much mention to make. Best, --Shirahadasha 23:18, 3 June 2007 (UTC)
 * Is there any value that is not disputed? Isn't just about any published source equally reliable in giving an value-judgment?  I think WP:NPOV already says that values should not be stated uncited.  It also seems to me that it's impossible to phrase things in a neutral voice while ascribing subjective values to them.  WP:NPOV demands a neutral phrasing of facts and mentions that "some things are very obviously values" - which are not facts.  This really only comes into play if there is a dispute; non-controversial articles wouldn't necessarily be affected at all.  I definitely agree with Dragons flight that never using unattributed value-judgments would challenge the elegance of prose because values do read better.  However, most encyclopedias do not have such a clear policy of neutrality, and I don't see that even under the current phrasing of NPOV, values are allowed to be stated as facts.  It's just not "neutral" to my understanding of the term or my reading of the policy.  Fourdee 16:22, 4 June 2007 (UTC)
 * Let's try another example. Suppose I want to write: "Merrill Lynch is a large financial services firm with 15,000 employees, that with a 15% rate of return on their derivatives products has significantly outperformed market averages."  Both "large" and "signficant" are subjective terms and reflect an opinion about what constitues large/significant.  Presumably they add context, since a typical reader would not necessarily know whether 15000 is large or not, or how good 15% is.  Hence removing such terms is to the detriment of the reader's understanding.  Your view of NPOV would say I can't use those words unless I explicitly attribute them, which leads to tortured constructions, e.g. "According to Dow Jones, Merrill Lynch is a large financial services firm with 15,000 employees.  With a 15% rate of return on their derivative products, they have outperformed market averages, which analyst John Doe describes as significant."  Presumably the facts are cited in both cases, but the latter adds many additional words that don't aid in understanding.  In fact, naming an analyst could even be misleading since a novice reader might distrust an unknown name while a knowledgable person would know that everyone in money management would agree that it is "significant".  Complicating our language, in the absense of any differing opinions regarding the accuracy of the description, is not a good way to write an encyclopedia and not consistent with my understanding of NPOV.  Dragons flight 17:27, 4 June 2007 (UTC)
 * I think that's basically how things should be, and believe that is already clearly the policy. "large" and "sigificant" are not facts and would better be replaced with something like "X largest in terms of gross income" etc.  At any rate, any enforcement of NPOV only comes up when someone challenges the article.  I am not proposing any change in policy at all, just a clarification that a value-judgment is an opinion, not a fact.  If someone challenges "significant" or "large" it definitely must be supported under the NPOV policy, and these are basically weasel words and values, shall we say "weasel-values".  Fourdee 17:46, 4 June 2007 (UTC)


 * "Assert facts, including facts about opinions — but do not assert the opinions themselves" bears repeating. I don't think there is any question that value-judgments are intrinsically opinions (especially given the examples and statements in the policy).  Fourdee 17:46, 4 June 2007 (UTC)
 * Don't ignore the following lines. "Assert facts, including facts about opinions — but do not assert the opinions themselves. By 'fact' we mean 'a piece of information about which there is no serious dispute.' ... By value or opinion, on the other hand, we mean 'a piece of information about which there is some dispute.'"  As before, an undisputed opinion is to be treated as fact.  And as before, I disagree, in my reading you are proposing a significant (and detrimental) change in policy.  Dragons flight 17:56, 4 June 2007 (UTC)
 * Doesn't any value fall under "some propositions are very obviously values or opinions"? It follows: if some things are obviously Ys, and X is obviously a Y, X is one of those "some things" that are obviously a Y.  It seems to me that the policy tried to make 2 categories out of 3, and lumped "value or opinion" together with the implicit assertion that a value be treated as an opinion.  Value is only mentioned in association with opinion in those paragraphs, and the examples given of things that are "very obviously" opinions (not facts) are all values.  Again, put very simply, the policy directly equates value with opinion with the phrase "value or opinion" - it already says that a value is an opinion...  Shrug.  Obviously have a difference of opinion here. :-)  Fourdee 18:25, 4 June 2007 (UTC)
 * Also I think you're forgeting the effect of WP:V - "Editors should provide a reliable source for quotations and for any material that is challenged or is likely to be challenged, or it may [(is allowed to)] be removed." - You are already required to provide citations for "any material that is challenged". Nothing would change in that regard.  If someone questions "significant" it must be cited, under either version of the policy.  If nobody questions it it's not affected, under either version of the policy.  Fourdee 18:48, 4 June 2007 (UTC)
 * Something can be cited while still being conveyed in the voice of the article. Obviously, we have many citations for undisputed facts without prefacing each fact by who said it, e.g. "NASA says the Earth is round.".  Including citations and identifying sources in text are not the same issue.  Dragons flight 18:56, 4 June 2007 (UTC)
 * That's right - values should always be attributed (whether cited due to being potentially controversial or already having been contested, or not). Values cannot be stated as facts. Fourdee 15:45, 5 June 2007 (UTC)


 * I maintain there is no distinction between "facts", "disputed facts" or "opinions". Just consensus.  When the consensus is overwhelming, it doesn't need to be continually cited as justification for the wikipedia entry.  When the consensus is not obvious, it must be cited by third party reputable sources.  And isn't it fantastic, that's just what wikipedia requires.  Try subsitituing Pluto for Mars in your Fact statement about Planets and you will see what I mean. --Work permit 01:33, 5 June 2007 (UTC)
 * I'll just repeat that value-judgments are never facts, they are subjective values ascribed to something. Color, size, position, number, weight, that someone published something, etc. are facts (whether disputed or not, they are like facts).  Goodness, importantance, relevance, worth, etc. are not ever facts, whether they are particularly controversial or not.  Fourdee 15:45, 5 June 2007 (UTC)
 * Comment: The above is an opinion, not a fact. Others may disagree. Best, --Shirahadasha 01:38, 6 June 2007 (UTC)
 * I agree that while there can be mamy opinons on the issue of whether all propositions neatly divide into the two classes "fact" and "opinion", For most purposes, the relevant distinction is simply whether a proposition is disputed or not. The only reason the fact/opinion distinction matters is because the encyclopedia may be subject to greater legal liability for publishing what certain legal systems call a "fact" than for publishing what these laws call an "opinion", because only legal "facts" can be defamatory. We shouldn't be debating what's true or not here. We should be discussing what strictures we have to live within. Like it or not, laws are one of them. Suggest adding the phrase "for legal purposes" or similar to make clear that Wikipedia isn't necessarily endorsing the idea that matters of morality etc. are not facts; it's simply abiding by it for legal purposes. Best, --Shirahadasha 01:48, 6 June 2007 (UTC)
 * I agree with Shirahadasha's assertation that the distinction between "facts", "disputed facts", and "opinions" is just one opinion. I will go futher and assert it is WP:OR, with a particular WP:POV on epistemology.  Fourdee asserts mars is a planet is a fact.  I maintain that it is a widely held consensus opinion on the definition of a planet and how it applies to mars.  As a simple example of my point, I refer to theplanetary controversy of Pluto. Fourdee's statement that tennis is played with rackets is just false, it can be played with a paddle.  His statement that water is liquid is a trivial and circular "fact", since water (in typical usage) refers to the liquid form of the compound.  I conclude from these three examples that the only fact Fourdee feels can be asserted without assertation is one given by it's own definition.  I show that even with his examples, one must use third party reputable sources if the definition and it's application is not self-evident.--Work permit 05:06, 7 June 2007 (UTC)
 * As another example, the statistician and quality guru W. Edwards Deming, whoi followed a version of epistemological pragmatism based on work by Clarence Irving Lewis, argued that there is no such thing as a fact. Note that the Pragmatism article identifies "a denial of the fact-value distinction" as one of the characteristics of this philosophical approach. Best. --Shirahadasha 03:52, 8 June 2007 (UTC)

All human organizations, and rules governing them, make innumerable epistemological (and other) assumptions. I'm saying that the policy should go ahead and explicitly make the assumption that the moral/aesthetic worth of something is not a fact. This debate could devolve endlessly into what assumptions are made by current policies, what philosophers support them, etc. Thats (in my opinion) needlessly reductionist and not to the point - the current NPOV policy already tries to separate facts from "values or opinions" it just doesn't do it clearly enough to match its intent, in my humble opinion. :-) Fourdee 14:48, 14 June 2007 (UTC)