Wikipedia talk:Attribution/Role of truth

This Role of truth page provides a forum for Wikipedia editors to explore Wikipedia's operational relationship to truth. Some of these discussions could occur on the TalkPages of the individual Verifiability, Reliable source, and No original research policy pages. But generally on the TalkPages of these policy pages, the discussion would be primarily focussed on improving the policy text. Here, in contrast, we explore various patterns in how those policy pages actually manifest in practice from the problems of the "role of truth" as editors strive to improve Wikipedia pages ranging from Aardvark to Zenith, and more.

This particular branch of this discussion began during the discussions on the development of the Attribution policy page. And initially, this discussion page serves to pull together the discussion on the "role of truth" where the questions arise from the context of examining attribution policy, but where the discussion relates to the "role of truth" and not necessarily to the decisions that must be made about the Attribution policy page.

This thread was initially started as a discussion of the new wording "not whether it is true" without "verifiability" appearing in the same sentence, and discussion of that wording and possible alternative wordings can continue here.

Truth is crucial in articles about living people

So how do we make sure that assertions in articles on living people are true? It seems to me that we have to depend on the published opinions of the reliable sources within any discipline to judge whether any other particular source has verified their facts. That is, for example, in the assertion of facts about a living person, we should make sure that some secondary source has used the assertion of a primary source in a way that implies that the secondary source as a professional within that particular discipline has verified the assertion of the primary source. We could work out this verifiability algebra for any particular Wikipedia page as case-in-point, if you wish. Just attributing the opinion to a reliable source is not enough -- because what is a reliable source that the statement was made may not be a reliable source for whether the statement in the opinion is true. Truth matters, and we must depend on secondary reliable sources to assess that truth for us. And truth matters much more for living persons than it does for the interesting and fascinating rumors on whether or not Shakespeare wrote Julius Caesar. --Rednblu 20:26, 26 March 2007 (UTC) [1]

Oh no it isn't. Attribution, not Truth, is what is important.

Does truth matter? In the extreme case of some secret truth about a famous person. Would we let a whistle-blower publish that secret truth on Wikipedia? Of course we wouldn't. It's a secret so it has not been precviously published and no attribution is possible.

If on the other hand there is a lie which has been widely published, by reliable sources, then that could be included in Wikipedia, along with it's refutation, if that has also been published. Attribution, not truth, is what is important if we want to avoid being sued.Filceolaire 22:06, 31 March 2007 (UTC)

What about the vandalism published on Stephanie Herseth's page in Augusts 2006 claiming she was pregnant by some guy name Ajay Bruno? While all would agree that the vandalism was libelous and not true, and it was later removed. However, the vanadlism itself got quite a bit of play in the South Dakota media and this Bruno guy and is outlandish claims have gotten significant coverage by blogs. The story about the vandalism is "widely published" yet it is not mentioned in the article, because it gives undue weight. I would submit that any false statements about individuals that are proven to be false by reputable sources have no business being included in an article about a living person, no matter how well "sourced" they are, particularly if "the criticism represents the views of a tiny minority" per WP:BLP.Dcmacnut 23:03, 31 March 2007 (UTC)

If that's true, Filceolaire, then Wikipedia is utterly worthless as an encyclopaedia. The problem with your statement that if "there is a lie which has been widely published, by reliable sources, then that could be included in Wikipedia, along with it's refutation, if that has also been published," is that it implies the converse, which is that if there is a lie that has been widely published, but although it is plainly illogical and well-known to be a lie, it cannot be refuted without attribution - no matter how damaging, no matter how plainly false. The criterion for inclusion should be accuracy resting on verifiable facts. Simon Dodd 23:37, 7 April 2007 (UTC)

I agree with Simon Dodd and Dcmacnut, if I understand correctly. First of all, if a statement about a person is false, then it is not very interesting. Wikipedia's purpose is to present interesting or potentially useful facts. So if Wikipedians have established to their satisfaction that a statement is false or probably false, then it probably doesn't belong in the article, (unless maybe there's significant controversy about it), especially not in a biography of a living person -- whether it's published material or some other thing which has convinced the Wikipedians that it's false. If different Wikipedians disagree about whether it's false, that's a different situation -- however, we need to be careful in biographies of living people. If there is good reason to believe that a statement may be false, maybe it needs to be taken out of a biography of a living person pending further investigation. In any case, whether it is true or not is certainly not irrelevant.

There might be no published refutation of the statement, but the living person might privately hold conclusive proof that it's false, which could be brought out during a libel suit against Wikipedia. We don't want Wikipedia going bankrupt having to pay legal fees. Better to delete the information unless we're on solid ground. --Coppertwig 12:58, 8 April 2007 (UTC)

How do you distinguish "false" from "true"?

Your "false" and my "true" may coincide. But of course you've had that pointed out to you enough times now, and simply ignore it. The conflict on WP:V is between those who believe that there is an "objective" truth and those who more realistically believe "truth" is negotiated among people. Wikipedia reflects the latter notion, at least in principle. If you don't like it, you really need to consider whether you should find a fork that believes in a truth you're more comfortable with. Grace Note 03:11, 27 March 2007 (UTC)

Nice point. Do you think we could usefully and simply include the breadth of the Wikipedia truth page in this policy definition? Does "Verifiability, not truth" operationally mean "Verifiability, but represent faithfully the standard of truth used by the Reliable Source's profession"? What do you think? I would appreciate working this out with you comparing the professions of say molecular biology and theology, two different -ologies with different standards of truth. Don't you think Wikipedia does a tremendous job of representing faithfully the corresponding standards of truth on many of the appropriate pages? But there seem to be troubled waters at some of the intersections of the two rivers of, shall we call it, la vérité for want of a better term. What do you think? --Rednblu 10:08, 27 March 2007 (UTC)

In reply to Grace Note: I'll try to do a better job of demonstrating my understanding (such as it is) of others' arguments. I think you mean that I may believe something is false while you believe that the same thing is true. Of course I realize that that kind of thing happens frequently. I'm sorry that I only have partial understanding of what you're trying to say next. You seem to be saying that there is a split between those who believe in objective truth and those who don't believe there is such a thing as objective truth. No, wait, I don't think that's what you mean. I think more likely you mean that everyone actually thinks there's a real world out there with specific characteristics, although we don't necessarily know what they are, but when doing things like writing an encyclopedia there's no use trying to get everyone to agree on what the objective truth is, so a subjective truth is negotiated instead. Sorry, that doesn't sound quite right either. The fact is, I don't quite know what you mean, or how specifically that applies to the wording of the policy. You think I might be happier on some other forum; actually, I like Wikipedia a lot.

Of course I understand that editors will often disagree about what is or is not true, and that the Verifiability or Attribution policy helps save a lot of time because people don't have to keep arguing about what is true, which is sometimes a short, effective discussion but sometimes goes on and on with nobody being persuaded of anything, but they can just see whether the material is in an acceptable source or not. I have no problem with that and am not trying to change that. --Coppertwig 23:31, 27 March 2007 (UTC)

Further reply: In a given situation, if there are no misunderstandings (as there often are), it's possible to establish to what extent each editor believes something to be true, and whether it's stated in an acceptable source. I think getting agreement on "OK, you believe it's false and I believe it's true" is fairly easy, probably easier than finding an acceptable source, and much easier than arguing until both editors agree on it, which could go on for several lifetimes if it's something like the existence of God. When there's disagreement, usually the editors insert prose attributions, and everybody is satisfied that the Wikipedia article is not asserting false statements. None of this logically requires that Wikipedia policy say any more than "whether it is attributable to a reliable source," period. Saying "not whether it is true" is unnecessary. --Coppertwig 23:40, 27 March 2007 (UTC)

Rednblu, that's a very interesting point. I think Wikipedia does quite well in that sense on scientific pages, particularly the technical ones. It presents the SPOV quite faithfully, mostly because those pages are written by practitioners. Few are going to find that exceptionable because, apart from blanket statements, few opposing viewpoints are available on specific subjects (what I mean is, many people do not believe that dinosaurs lived millions of years ago, but they have not taken a view on Diplodocus specifically, so they are not represented in that article). Even if one thought that was problematic, I cannot see any hope of winning the discussion over it. I think that articles of that type have in any case a kind of synthetic truth. They are written in the language of science about science. The language of science can be taken as axiomatic for those articles. If you allow the language of science to be axiomatic, you allow the articles as true so long as they follow from that language. If you do not allow it, you do not allow that the articles are true. I think we should state that explicitly though and not pretend otherwise.

Our pages on "pseudoscience" are presented in the same light though, which is problematic, and I know that you have written about that. The same is true of pages on religion, which is presented from a rather nonneutral position. In neither of these areas is it as clear that we are using a particular language or standard as an axiom. We seem to be writing from a much broader reference point.

The position of most of the policy writers here creates a de facto operational definition of truth, without stating that openly (which is something I take issue with in the discussion on reliable sources at Wikipedia_talk:Reliable_sources_and_undue_weight). Ultimately, we take the position that what the MSM reports, what scientific papers state and academics write is, if not true, correct, and anything else is not. I don't think a truly NPOV encyclopaedia can take any stand on the truth, and our policies on NPOV and sourcing are poor in this regard if our aim is truly to be neutral rather than to reflect the status quo wisdom of the day. For me, this devalues Wikipedia, making it simply a poor man's Britannica. But most editors here seem quite happy to be poor man's encyclopaedians, and run scared from vision or challenge. Grace Note 01:29, 28 March 2007 (UTC)

Coppertwig, I am saying that we cannot address objective truth because no such thing exists. I think you have entirely missed my point, which is that you cannot legislate against including "false" information if there is no way to measure what is true and what is false.

Here is an example. The article on the battle of Englefield, Berkshire said (before I corrected it) that the Saxons were led by King Alfred. They were led by Aethelwulf, a local noble. But were they? There are sources that say so, and I can attribute my edit to them (if I have to; naturally, facts like this are rarely attributed in Wikipedia). But if someone were to reinsert that the forces were led by Alfred, what then?

Have they included "false" information? I cannot know. My sources say Aethelwulf but I don't know that. It might be me who has inserted false information.

I recognise in your reply that you are trying to suggest that the policy should emphasise that I need only be able to attribute my claim. Which I agree with. But you are claiming elsewhere that we should not be explicit that we do not make claims to truth. I think it is essential that we do not.

Think about my example. Think about what it means for it to be true that Aethelwulf led the Saxons. There must be an "objective reality" (we'll allow that for argument's sake). There must have been events in Englefield in 870. But who owns the truth about it? What if Aethelwulf wrote a memoir? Would that be "true"? How can you know? See my note to rednblu above: where our articles are based on subject-specific quasi-axiomatic languages, there is a possibility of referring to the "truth", so long as the "truth" is understood to be wholly dependent on the axioms in question. Otherwise, it is pointless to talk about it at all, except to warn as strongly as we can that we are not dealing in it here.Grace Note 01:29, 28 March 2007 (UTC)

--- Above excerpt extracted from this link.

I would argue from a different perspective. I'd say that in many cases (excluding matters of personal taste for instance) there does exist an objective truth, but our access to the truth is limited, as is our ability to agree on what the truth consists of. Our requirement of attribution, reliable sources, no original research that can't be independently verified, these things are our way to "measure" truth. They serve the goal of approaching the truth to the extent that we are able to do so, and they also serve the purpose of giving skeptical readers the opportunity to look at the evidence to support the various perspectives including the main/central one if such a dominant view exists. They also allow for further research by those looking for a deeper or broader view of the subject. I agree that we don't want to assert that the contents of the encyclopedia are "true," especially not that they are undisputedly true... they are only true to the extent that we can point to our sources and attempt to represent reliable info. But we also don't want to claim that we are entirely unconcerned with the truth. That's not actually the case. If we didn't care at all about truth/accuracy, then we wouldn't bother to require attribution at all. We would permit outright fabrication, deliberate distortion, and fanciful lies if we had no interest whatsoever in truth, or if we really believed that no such thing as truth exists.zadignose 08:56, 3 April 2007 (UTC)

How do I distinguish "false" from "true"? The same way everybody else does: some combination of senses, experience, logic, what I've been told, etc. I think we all agree that from time to time people have beliefs about what is true or false, and that sometimes different peoples' beliefs are in conflict, and that the Wikipedian policy needs to give instructions as to what to do in such cases. One solution is to instruct that personal beliefs should never have any influence on the article writing. That is not the only solution and is not a solution that I support. Another solution is to instruct that everything that is included must be attributed to published sources and that when there is contention, prose attribution should be used. That solution is fine with me and is what the longstanding policies have said. When one believes that a statement is false, seeing it asserted by a Wikipedia article looks wrong; but seeing it presented with a prose attribution looks much more acceptable. --Coppertwig 20:30, 9 April 2007 (UTC)

zadignose - you have hit the nail on the head when you note that "the contents of the encyclopendia ... are only true to the extent that we can point to our sources and attempt to represent reliable info." I don't see where we need to go any further than that. If we can say "this is truthfully the best information available" then we have done our job. I don't see where the real truth matters in terms the contents of this encyclopedia, since

1. That which is attributable is in the vast majority of cases true and accurate, and
2. Over time for those cases where the understanding is wrong it will tend to be corrected.

Over in the physics pages where I edit, people constantly claim that their original research is true. Removing any requirement for truth pulls the rug our from under those people, and in general keeps out very little real truth. Furthermore, in the cases where a real truth is kept out, that means that it was not well known enough to be verifiable, attributable, and notable anyway; and that really is what matters here. --EMS | Talk 22:08, 9 April 2007 (UTC)

Role of truth

This is a continuation of ongoing discussion at Wikipedia talk:Attribution/Archive 12#Role of truth and Wikipedia talk:Attribution/Archive 13#Role of truth, plus discussion moved from Wikipedia talk:Attribution/Community discussion/Archive 1#Role of truth. --Coppertwig 19:44, 23 March 2007 (UTC)

• I object strongly to the new wording "not whether it is true" and the new context for this wording, unbalanced by a word such as "verifiability". Some other editors have the same or similar concerns. See Wikipedia talk:Attribution#Role of truth. --Coppertwig 00:18, 23 March 2007 (UTC)

• What difference in meaning do you see? "Verifiability" always meant that the existence of a reliable source could be verified. Septentrionalis PMAnderson 00:38, 23 March 2007 (UTC)

This has already been discussed in huge depth at Wikipedia talk:Attribution; my entire objection to this page as a separate page is that it is likely to simply turn into a rehash of the same arguments. I beg that this issue simply be recognized as existing (for at least three editors that concur that the present language is problematic) but that this not turn into a repeat of the entire debate. It's simply not a conversation that needs to happen in two places. — SMcCandlish [talk] [contrib] ツ 01:20, 23 March 2007 (UTC)

It has been discussed several times at WT:ATT. Invariably, those discussions have gone quiescent and been archived, because so much else happens on that page. This discussion may last long enough to explain the issue; and if it grows beyond the length this page can take, it can become a permanent subpage. Septentrionalis PMAnderson 15:49, 23 March 2007 (UTC)

That was always there, in WP:V: The threshold for inclusion in Wikipedia is verifiability, not truth.≈ jossi ≈ (talk) 01:33, 23 March 2007 (UTC)

ATT as killed WP:RS see [2] for an example. Betacommand ^{(talk • contribs • Bot)} 02:37, 23 March 2007 (UTC)

Can you elaborate? That is, are you saying ATT has killed RS and this is good? — SMcCandlish [talk] [contrib] ツ 03:01, 23 March 2007 (UTC)

Since some people seem to think the "not truth" wording is problematic, how about the following phrasing?

The threshold for inclusion in Wikipedia is whether material is attributable to a reliable published source, not whether you personally know it to be true.

That's what it has always meant anyway: that information has to be confirmable externally (WP:NOR) and that someone else has to have cared to write it down (WP:NOTE). That Coppertwig and similar arguers keep bringing up this problem shows just how confusing the previous oxymoronic wording (Truth, not truth) was and exactly why we needed to make this change. --tjstrf talk 04:42, 23 March 2007 (UTC)

That is not what it has always meant. Try the first time WP:V was labelled as policy[3]. What it meant then was "your belief that a statement in an article is true is not enough; it has to be possible to check the statement is accurate; providing sources enables this to happen". What it has become in WP:ATT is more like "we don't care whether a statement in an article is true or accurate; all we care about is whether we can find someone who said it". --Henrygb 18:14, 23 March 2007 (UTC)

• The principle "Verifiability, not truth, is the threshold for inclusion" is fundamental to Wikipedia; however, I think that wording it as "Attributability, not truth, is the threshold for inclusion" is less confusing for new users. This is simply because of the English language; in normal usage, "verifiability" refers to something that is certifiably true, while "attributability" refers to something that can be ascribed to a source, whether it is true or not. This better encapsulates the basis of the dictum in question. Walton ^{Vivat Regina!} 09:57, 23 March 2007 (UTC)

• I think that's a very good point.. the phrase "Attributability, not truth, is the threshold for inclusion" seems much better. The word "verifiability" is open to a little more interpretation - for example, i can verify lots of things that shouldn't be included in wikipedia. i understand that in wikipedia jargon "verifiable" implies a reliable source, but in the english language it does not. the word "attributable", on the other hand, is much less ambiguous. it also makes it more clear what is meant by "not truth". Mlm42 15:10, 23 March 2007 (UTC)

• But that is what is wrong. If it can be attributed even if it is obviously not correct then it should not be in Wikipedia except as part of quotations or NPOV balance. If I find someone saying the Ulster Volunteer Force (1912) was established in 1912 (it was in fact in 1913), I am not going to include it except to point out the error. It is not good enough the respond that a source containing errors is not a reliable source, as that simply shifts verifiability a level. Accuracy through verifiability is the key. --Henrygb 18:14, 23 March 2007 (UTC)

I'm moving the following comments of mine here from Wikipedia talk:Attribution#Role of truth. I apologize for my role in creating two ongoing threads on the same topic and hope to merge the two threads here. From the above comments I get the impression that perhaps the users commenting above don't understand the issue I'm raising.

Septentrionalis, you have a good point. The sentence "Not everything which is attributable is worthy of inclusion" is not a very effective solution to the problem. I may disagree with you about whether that sentence would solve the particular examples I mention, but I'll mention below another example where that sentence would do little or no good.

The problem is: The words "attributable ... not whether it is true" do two things:

1. Re-emphasize that material that is true but not attributable is not acceptable; and
2. Suggest that material that is attributable but known to be false is acceptable, even as plain statements (i.e. without prose attribution).

The original wording "verifiability, not truth" did only the first of those two things. When the word "verifiability", which includes "truth" in its meaning, is removed, the second of these two things comes into play. I think the writers of this policy generally intend only the first of the two things, and that the second comes in by accident. I attempted to hold a poll to see whether anyone actually supports the second of the two things, but apparently I didn't explain clearly enough the purpose of the poll. I would like to find wording that does the first thing just as effectively but that doesn't do the second thing.

For reference, the current version of the sentence on the longstanding policy page is

• "The threshold for inclusion in Wikipedia is verifiability, not truth."

and the current version of the sentence on this proposed policy page is

• "The threshold for inclusion in Wikipedia is whether material is attributable to a reliable published source, not whether it is true."

Possible alternative wordings:

• Just delete "not whether it is true". Once "verifiability" is gone, those words are unnecessary.
• Revert to the original "verifiability, not truth".
• As User:DCB4W suggested: "...not solely whether it is true".
• "...not merely whether it is true."
• "Being true is not enough."
• "Some true material does not belong."
• "All material must be attributable."
• "If it's not attributable, it does not belong."
• "...otherwise, it does not belong in Wikipedia."
• "...; only such material is acceptable."
• "...without this, it cannot be included."
• "...; proving that something is true does not get it included."
• "...; however, false or contentious material requires prose attributions (see WP:NPOV)."

Would other users please suggest other alternative wordings that attempt to do the first thing I list above but not the second? Also, please comment on my analysis and whether the second thing is coming in by accident or is actually wanted as Wikipedia policy. Perhaps it would also help the consensus-building process if people would comment on each of the proposed solutions.

Here's the example I promised: Suppose the following were to happen: Someone knowingly inserts false statements into Wikipedia articles (as plain statements, without prose attribution) with the purpose of defrauding someone of money, or of harming someone, or of getting elected, or of starting a war. The person is charged with a crime, and as part of their trial in court, evidence is presented that the person knowingly inserted false statements into Wikipedia. The person uses as their defence: "Oh, but writing something in Wikipedia is like writing something in a fiction story! It isn't fraud, because there is no expectation that it be true! Look at Wikipedia policy -- it says 'not whether it is true'!" On the basis of this defence, the person gets acquitted of their crime. Afterwards, the newspapers and other media, who are to a large extent in competition with Wikipedia for the limited leisure time of potential readers, have lots of fun repeating this example over and over again and repeating that "Wikipedia is not reliable", "Wikipedia is like a fiction story" and "Wikipedia policy itself states that it's not even trying to be true." --Coppertwig 14:36, 23 March 2007 (UTC)

This is a continuation of ongoing discussion at Wikipedia talk:Attribution/Archive 12#Role of truth and Wikipedia talk:Attribution/Archive 13#Role of truth.

(Deleted my comments and moved them to Wikipedia talk:Attribution/Community discussion/Archive 1#Role of truth. --Coppertwig 18:35, 23 March 2007 (UTC))

Moved back. That page is a meta page. This page is for discussion of policy. SlimVirgin ^(talk) 18:42, 23 March 2007 (UTC)

I was asked to clarify what I meant by my comment about the proposed addition of the sentence "Not everything that is attributable is worthy of inclusion" not saying anything. I thought I was pretty clear, it doesn't have a useful meaning, and makes the thing longer. I don't strongly oppose, it doesn't make the thing much longer, and it's not blatantly offensive or anything, so if others agree to add it, I won't fight to the death to remove it. I just don't think it adds anything not obvious to anyone who has ever thought about writing an article. It should be clear that we shouldn't add everything just because we can, we're a summary of secondary sources, not an exhaustive archive. So mark me as a weak oppose. (By the way, it doesn't seem like this section is discussing that any more, but this is where I was asked to clarify.) --AnonEMouse ^(squeak) 15:05, 23 March 2007 (UTC)

i was directed here from Wikipedia talk:Attribution/Community discussion.. it is a somewhat intimidating discussion thread to join, due to the long, philosophical sounding arguments. in any case, personally, i'm in favour of the phrases

• "Attributability, not truth, is the threshold for inclusion in Wikipedia." or
• "Everything in Wikipedia must be attributable to a reliable published source."

As always, no matter what one-line summary of the policy we give, there will always be exceptions.. for example, the first phrase seems to imply "Everything attributable is worthy of inclusion", which may not be true. Nevertheless, i think the first phrase gets to the point in a fairly unambiguous way, that is understandable to newcomers. The second phrase is simply less emphatic, without the not truth phrase.

Also, regarding a couple examples i saw, such as "The leader shook hands with God". Reliable sources may say this, but, there are also reliable sources that contradict this (such as books published regarding the non-existence of God). to disregard the contradiction would be POV.. it would have to be qualified to "These newspapers claimed the leader shook hands with God." Similarly for the Santa Claus argument.. if there is reason to doubt the claims, such as published sources regarding the mythology of Santa Claus, then you can't simply state "A flying object pulled by flying reindeer was seen in the sky," since there are other reliable sources that contradict this, and would imply that newspapers may intentionally lie to convince children.. so it would be POV to not say so. Mlm42 15:55, 23 March 2007 (UTC)

Thank you, user User:Mlm42 for your suggestions. I like the second one. The first one suffers from exactly the same problem as the "not whether it is true" wording -- i.e. saying "not truth", without balancing it by any word such as "verifiability".

Would people participating in this discussion please make it clear whether you support having the policy say something that encourages or explicitly allows knowing insertion of false statements as plain statements without prose attribution? Or whether you think it's unimportant whether it does or not? To me, it is of fundamental importance to the very meaning, purpose and credibility of Wikipedia. It may not be important to you, but would you please state where you stand on this: do you disagree with me that the words do that? Or do you see that they do that, and think it's not of much importance? Or do you actually want to encourage inclusion of such known false statements?

With regard to user Mlm42's second suggestion: any comments on this? If you oppose changing to that wording, would you please explain why? How about giving examples of real or hypothetical situations where the "not whether it is true" wording would work better than user Mlm42's second suggestion? --Coppertwig 22:06, 23 March 2007 (UTC)

---

Truth has no role on Wikipedia, other than by coincidence. ^High_InBC^{(Need help? Ask me)} 16:01, 23 March 2007 (UTC)

You are overstretching. BTW, please provide reliable sources to prove your statement. Or say in a more modest way, like, "IMO truth has no role on Wikipedia". `'mikka 16:29, 23 March 2007 (UTC)

---

<< Truth has no role on Wikipedia, other than by coincidence. >> ("attributed" to User:HighInBC}

I really admire User:Coppertwig for insisting repeatedly for so long that we must deal with the real problem here, which may be Wikipedia's relationship to "truth." I use "truth" for want of a better English term. Surely, when we, as users, click on that browser link to Wikipedia to look for some answer for ourselves, we come to Wikipedia looking for the "truth." Isn't that the "truth"? And surely as world-travelers, by now we know that the "truth" is: There are many truths. And surely as users of Wikipedia looking for answers, we want some reliable source for each of those many truths to have verified each claim in the Wikipedia page. If that is so, then our job as Wikipedia editor is to make sure that each of the -- perhaps contradictory -- truths on the Wikipedia page has been verified by some reliable source. Yes? So how do we do that? We do that by clear and self-consistent Wikipedia policy text. Attribution is only one small part of our job. Too many well-meaning Wikipedia editors read the current Attribution policy text and follow it faithfully to produce incredibly well-attributed Original Research that does not match with anything that any published reliable source has shown to be verified. I say Bravo! by the way to everyone who has assembled, nurtured, and protected the Wikipedia page on truth. It really manifests Wikipedia's strong relationship to "truth" by ensuring that each claim on that page has been verified by some reliable source, including the implied ranking of which "truth" is more true. --Rednblu 17:59, 23 March 2007 (UTC)

if editors faithfully follow what is said on ATT, then they will not write well-attributed Original Resarch, as that policy clearly states that we should not do this. Blueboar 18:09, 23 March 2007 (UTC)

I understand that you think that is so. And I hypothesize that each of the over 50% of Wikipedia editors who oppose the "merger" can give you a counter-example page to your assertion in which the well-meaning editor followed the current Attribution policy text faithfully to produce incredibly well-attributed Original Research that does not match with anything that any published reliable source has shown to be verified. Anyone? --Rednblu 18:34, 23 March 2007 (UTC)

Too many well-meaning Wikipedia editors read the current Attribution policy text and follow it faithfully to produce incredibly well-attributed Original Research that does not match with anything that any published reliable source has shown to be verified.??? I think that you may have missed something. See the nutshell: Everything in Wikipedia must be attributable to a reliable published source. How can you engage in "well-attributed Original Research" given this principle of ATT? ≈ jossi ≈ (talk) 20:01, 23 March 2007 (UTC)

In reply to user User:Mlm42: If a source says that God doesn't exist or that reindeer don't fly but doesn't mention the particular incident, it can't be used as a Wikipedia citation that the particular incident didn't happen, because of WP:NOR. Suppose there are no citations saying the incident didn't happen. The Wikipedian editor may be convinced that it didn't happen or that it probably didn't happen or that it's controversial. Are they allowed or encouraged or practically forced to put false information into the article or to allow another editor to keep in the false information (as plain statements without prose attribution)? I oppose the "not whether it is true" wording (unbalanced by any word such as "verifiability") for this reason. --Coppertwig 00:40, 24 March 2007 (UTC)

I disagree with you here; if a source says "many newspapers knowingly put false claims of santa claus sightings around Christmas time", and then a newspaper claims to have a santa claus sighting, surely the source can be used to refute this. to require a mention of every single incident would be silly. Mlm42 11:17, 24 March 2007 (UTC)

Role of truth 2

(Comments moved from Community discussion page)

In reply to tjstrf: No, that wording has essentially the same problem. The problem is that "not truth" implies two things: (1) true but unattributable material is unacceptable and (2) false but attributable material is acceptable. The second of these is undesirable. As far as the second item is concerned, it would be best to say nothing about truth, and let people use common sense and the status quo in trying to decide whether a Wikipedian writing a Wikipedia article means what that one says.

The word "verifiability" contains the idea of "truth" in its definition, so it negates the second item. If "verifiability" is taken away, something else needs to replace it; unless there is a consensus that when someone knowingly writes false statements into a Wikipedia article, as plain statements without prose attribution, they are actually helping to write the encyclopedia; is anyone willing to admit that they support that? And if so, what do you think the ultimate purpose of Wikipedia is? I mean, if you think the purpose of Wikipedia is to write a summary of what's in the literature regardless of whether it conforms to reality, then what is the purpose of writing such a summary? To complete homework assignments with the purpose of getting good marks regardless of whether one learns anything useful? To leave a record, after our civilization falls, of the falsehoods we believed so that some later civilization can learn from our mistakes? Or what?

I think the purpose of Wikipedia is to provide information that conforms to reality a large proportion of the time, and that requiring attribution is merely a method of trying to achieve that goal. --Coppertwig 18:51, 23 March 2007 (UTC)

I deny that Wikipedia's sense of verifiability ever implied "truth"; it implies that the sourcing of our assertions was verifiable. WP:V did, and does, say so at some length. Our only hope to conform to reality is the faith that our sources conform to reality; if they err, we must. (To some extent, this can be avoided by choosing better sources; by acknowledging, exempli gratia, that a newspaper, however reliable on other days of the year, is not a reliable source when it prints flying reindeer stories; but this is a minor point.) Septentrionalis PMAnderson 21:18, 23 March 2007 (UTC)

Ever? Try [4] when WP:V became policy. It said "accuracy" and I can live with that, but verifiability was all about being able to check facts. --Henrygb 22:32, 23 March 2007 (UTC)

---

<< See the nutshell: Everything in Wikipedia must be attributable to a reliable published source. How can you engage in "well-attributed Original Research" given this principle of ATT? >>

Here is how it is done by well-meaning editors following religiously the murky and self-contradictory Wikipedia policy text. Following faithfully the current incoherent text of the Attribution policy page including its wonderful Nutshell summary, one prolific Original Research writer protested it "was not original research since it was a simple reporting of facts and no conclusions were drawn." And each fact reported was surely "attributable" because every fact in the statement was "attributed" to a "reliable source." What was missing from the murky text of current Wikipedia policy was the requirement that every claim on a Wikipedia page must have been verified by a published reliable source. That is, following the insights of User:Coppertwig, the fundamental Wikipedia value here is truth. And at Wikipedia we require that each assertion on a Wikipedia page must have been verified as truth by a published reliable source. Mere attributability of each "everything in Wikipedia" to a published reliable source is not enough -- because much of what reliable sources publish -- even the exact words -- will not show that the well-meaning Wikipedia editor's assertion has been verified as truth. --Rednblu 01:24, 24 March 2007 (UTC)

We never had in Wikipedia a rule to verify anything as truth. Not in WP:V and not anywhere else. The policy of WP:V was designed to disallow material that cannot be attributed to a reliable published source, that is the same to say that can be verified that it was published in a reliable source. Again: we never had a policy that forced editors to provide assertions that something has been verified as truth. ≈ jossi ≈ (talk) 01:40, 24 March 2007 (UTC)

This from WP:V's lead: ≈ jossi ≈ (talk) 01:47, 24 March 2007 (UTC)

The threshold for inclusion in Wikipedia is verifiability, not truth. "Verifiable" in this context means that any reader should be able to check that material added to Wikipedia has already been published by a reliable source.

Agreed; WP:V as it stood was not about truth. In many cases, especially with divisive topics, we can never hope to get all editors to agree that everything in the article is true. The NPOV policy mandates that, from time to time, we must have things in articles that not all sides would accept as true. CMummert · talk 01:51, 24 March 2007 (UTC)

Yes, very good. Though we can never hope to get all editors to agree that everything in the article is true, we often get all editors to agree on what the published reliable sources have verified as true, do we not? Look at the truth page itself. In my opinion, that is exactly what the editors did there. They ensured that all the conflicting and jarring truths about truth were verified as true by some published reliable source. So it seems that User:Coppertwig and others have been right all this time in boring us out of our mind with the accusation: "The threshold for inclusion in Wikipedia is verifiability, not truth." -- that's nonsense because verifiability requires truth. Perhaps the correct phrase is something like "The threshold for inclusion in Wikipedia is verifiability. 'Verifiable' in this context means that any reader should be able to check that material added to Wikipedia has been asserted as factual by a published reliable source." --Rednblu 03:20, 24 March 2007 (UTC)

we often get all editors to agree on what the published reliable sources have verified as true, do we not? No, we don't. You have got it wrong, Rednblue. Verifiability is NOT about truth. Read the policy. It is about "that any reader should be able to check that material added to Wikipedia has already been published by a reliable source." Period. ≈ jossi ≈ (talk) 04:23, 24 March 2007 (UTC)

The implication I'm getting from that is that including information you know to be false is a good thing as long as it is published in a reliable source. Which is stupid; nobody should be able to deliberately add information they know is false and think they're following policy. -Amarkov moo! 04:27, 24 March 2007 (UTC)

I am sorry to inform you, my friends. Verifiability is very much about truth. We cannot reinvent the English language. Use your dictionary -- any dictionary. You will find no reputable English dictionary on earth that says that "Verifiability is NOT about truth." Every English dictionary still in print and selling on Amazon.com will define "verifiability" as very much about truth. Let me alert you to the particular situations where it is deadly to make the mistake that you are making in asserting that "Verifiability is NOT about truth." First, if you put together articles about living people merely from the facts in "attributions" to published reliable sources, you will get pages full of assertions that are not true. And you will get us all into trouble, my friend, if you do that. At least with respect to articles about living people, "Verifiability is very much about truth." Would you grant me that? Would you grant -- us -- that? I mean, excuse me while I yield the floor to User:Coppertwig, who is my teacher in this matter. --Rednblu 05:15, 24 March 2007 (UTC)

I think the issue here is not whether or not verifiability should be about truth, but whether it can be about enforcing truth. On paper it's a great idea, but on paper Soviet Communism works great. (Oh yes, I went there) In my opinion we could (and should) have a paragraph or section explaining this little dilemma. Just off the top of my head, perhaps it could go something like this:

"Though it may seem confusing, the point of Wikipedia's policy on verifiability is to ensure all content can be looked up or otherwise verified by the casual reader, regardless of the inherant "truth" of the information. Ideally all material on Wikipedia should be pure, incontrovertable truth, but in practice this is impossible. Truth can be at many times a subjective designation; there can be multiple, widely held, and well-reasoned opinions on what constitutes "truth" for a given topic. In absence of an objective measure of truth Wikipedia must instead rely on an appropriate approximation: the combined weight of the sources backing up the information.

— Me, just now

I think having this paragraph (or a paragraph like it) featured prominently in the policy would help to clarify the matter. -- Y|yukichigai (_ramble argue ^check) 09:06, 24 March 2007 (UTC)

I think it is important to simplify matters wherever possible. For example, perhaps it can be summarize with a statement like:

• "The threshold for inclusion of material in Wikipedia is attributability, not its perceived truth."

A statement like that, which questions the objectivity of the word "truth", and gives a link to the article on truth, would probably get the point across. Mlm42 11:27, 24 March 2007 (UTC)

Sorry, should have been more clear. I was both intending for that paragraph to clarify the issue of why Wikipedia's threshold for inclusion is attributability, and to head off statements that "Wikipedia doesn't care about truth." In that regard I think it works fine, as does your simple statement regarding the threshold of inclusion. -- Y|yukichigai (_ramble argue ^check) 11:55, 24 March 2007 (UTC)

Rednblu: "Verifiability is very much about truth. We cannot reinvent the English language." Sorry to inform you this, but we can and we did. It's called a term of art, and many people all over the world redefine their languages all the time using them. JulesH 11:34, 24 March 2007 (UTC)

Rednblue, Coppertwig, and perhaps (sorry, not entirely sure) JulesH are seriously muddying the waters here. "the fundamental Wikipedia value here is truth. And at Wikipedia we require that each assertion on a Wikipedia page must have been verified as truth by a published reliable source" - this is simply not true. From its inception Wikipedia has been about verifiability, not truth. Verifiability does not mean that we can verify that someone believes it to be true, it means that we can verify that someone believes it. You may think that for someone to believe something means that they believe it to be true, but the whole point is that it simply doesn't matter. In this context we are not twisting the meaning of the English word verifiability. Larry Sanger was using it to make a philosophical argument- one I do not believe he claimed was original to himself - that there is no objective basis for truth-claims but there can be an objective basis for verifiable claims - and I repeat, "verifiable " does not mean we verify that they are true, it means we veify that someone (who is not a Wikipedia editor) has actually made the claim. In this sense, verifiability is central to our NPOV policy. The reason people can do original research while citing verifiable/attributable/reliable sources has nothing to do with "verifiability not truth," it has to do with the way people can take a variety of claims found in diverse sources and string them together to make a new argument or synthetic claim. Simple example: I quote one newspaper article that says "all men are mortal" and another newspaper that reports that "Socrates was a man" and then I write in a Wikipedia article "therefore Socrates was mortal." I cited two sources but have still done OR thus violating this policy. I cannot add to Wikipedi "therefore Socrates was mortal" unless I can find another verifiable/attributable source, e.g. "Aristotle wrote (citation) that if both a and b are true, then Socrates was a mortal." What is important is not that this is true (violates NPOV and V), and certainly not that it logically follows from a and b (for which we have sources) (violates NOR and V) — what is important is that we can verify that Aristotle made the claim. Slrubenstein | Talk 13:49, 24 March 2007 (UTC)

Completely agreeing with Slrubenstein on this. Verifiability means that we can verify what someone believes, not to verify what is true in terms of Objectivity. -- Vision Thing -- 14:00, 24 March 2007 (UTC)

---

<< and I repeat, "verifiable " does not mean we verify that they are true >>

"Well said!" I say. We would not want Wikipedia policy text to encourage the editors following that text to verify under their own processing power that the assertions in a Wikipedia page are true. Rather, we would want the Wikipedia policy text to require the editor following that text to ensure that some published reliable source has verified that each assertion in a Wikipedia page is true. And we must correct the murky, incoherent, and self-contradictory Wikipedia policy text to make this clear -- that the Wikipedia editor is to ensure that published reliable sources have verified that the assertions in a Wikipedia page are true. This is actually the operant Wikipedia standard being enforced today. We have to make Wikipedia policy context conform to reality. Let us consider a concrete example, the global warming page. Al Gore, bless his heart, tells us that we must face this Inconvenient Truth. And so we must. The proper Wikipedia standard on the global warming page is truth -- not the editor's truth, not the crank's truth -- but rather what the published reliable sources have verified as truth. And of course, there are often conflicting and jarring truths on any topic. The Wikipedia page on truth lays this out for us very clearly. Accordingly, we want the Wikipedia policy text to clearly state that it is the duty of any editor changing the text on global warming to ensure that some published reliable source has verified as true whatever is on the global warming page. The statement of standard Everything in Wikipedia must be attributable to a reliable published source is no longer sufficient. Rather the statement of standard must be compatible with the best practices on Wikipedia which is something like Every assertion in Wikipedia must be verified as true by a published reliable source. --Rednblu 15:32, 24 March 2007 (UTC)

Rednblu, your request is to change core policy in a direction that contravenes the principles upon which this project is based on. Not a happening thing, IMO. Please re-read Slrubenstein's response. ≈ jossi ≈ (talk) 15:40, 24 March 2007 (UTC)

I disagree with you as to what the current consensus is on this issue. So do others. So let's have it in Wikipedia:Attribution/Poll and get broader opinions. --Audiovideo 03:13, 25 March 2007 (UTC)

The wonderful thing about the truth, is that we can each have our own. If we refer to theTruth page, we see that there is a Constructivist theory of truth, where "that truth is constructed by social processes" and "all of our knowledge as "constructed," because it does not reflect any external "transcendent" realities" plus there is a Consensus theory of truth, where "truth is whatever is agreed upon" and many others. Do we seriously expect to get editors that in many cases can not agree on what are reliable sources, to agree on what is the truth? If we place more emphasis on truth we might solve nothing, but create even more edit wars by people that insist that the article reflect the(ir) truth. --Xagent86 (Talk | contribs)

Thank you, friends, everyone, for engaging with this here. I hypothesize that Wikipedia's relationship to truth is the real trouble that we must deal with underneath all of this confusion and what looks like hysteria at this time. But I have great hope. I was moved to tears on finding what excellent quality the editors had produced on the truth page. I have no idea how to proceed next on this. I am just guessing like we all are. Thank you, Xagent for applying the analysis of the truth page to our situation here as excellent and useful questions. I found Slrubenstein's, Jossi's, and VisionThing's analysis above to be very useful in implying, to me at least, that Larry Sanger created the "Verifiability, not truth" construction as a means of getting Wikipedia pages to record the personal truths of 1) reliable sources instead of the personal truths of 2) edit warring Wikipedia editors. For our Wikipedia situation, I am sure all of you will see some better and fuller application of the wonderful and very useful categories of truth described on that truth page. In any case, I agree with Audiovideo above -- in that it seems to me that the editors of Wikipedia have a very strong dedication to truth as a many-pathed approach to describing the facts of reality -- as seems to me is sketched very well and quite usefully on that Wikipedia truth page. Why do the defenders of the current Attribution policy page so strongly assert that Wikipedia has no interest in truth? That is a puzzle that we must untangle and resolve, it seems to me. --Rednblu 06:49, 25 March 2007 (UTC)

With all due respect, Rednblu, could you please try to have a perceivable thesis in your posts? For as long as I've seen you, you've had a consistent habit of making extensive, lengthy posts to policy talk pages that I can never pull any substance from beyond "we need to talk about this more" and that you generally but vaguely disagree with the implementation of our policies in certain controversial subject areas. --tjstrf talk 06:59, 25 March 2007 (UTC)

The current Attribution policy page is a humongous change from prior Verifiability and NoOriginalResearch policy. The current Attribution policy page encourages Wikipedia editors to do Original Research that makes assertions that are not true about living persons. What do you think we should do about that? --Rednblu 07:42, 25 March 2007 (UTC)

What should I do about it? Why say "It does not", of course!

It does not. WP:A states "Any unsourced material may be removed, and in biographies of living persons, unsourced or poorly sourced contentious material must be removed immediately." Additionally, it states that information which "introduces an argument without citing a reliable source who has made that argument in relation to the topic of the article" (emphasis mine). The page is quite clear on the matter, the inclusion of original research or synthesis in BLP articles is forbidden, so your assertion that it encourages this is patently false. --tjstrf talk 08:05, 25 March 2007 (UTC)

Your answer to Rednblu's suggestion that our policy encourages OR that makes false assertions about living people, which is quite accurate, is to suggest that we should fix it by removing unsourced material. Way to miss the point, which is that if your source lies, you lie. Personally, I feel that the "include it if it's sourced" point of view is a serious divergence from what a responsible encyclopaedist should do, because it places the emphasis wrongly. Although everything we include should be sourceable, so as not to represent our own opinions, being sourceable should not be a defence for injudicious edits. All too often an editor will say "I am including this utter bullshit because I have a source for it". And it remains true that one person's "reliable" source is the next person's pack of lies. Grace Note 22:53, 31 March 2007 (UTC)

I think it is evident now that Rednblu is merely trolling. I and others have explained several times why s/he is wrong, and his/her response is always something like "Yes, excellent point, I agree: ..." and then goes on to insist on the opposite of what we are saying. for example, above in response to my comment Rednblu writes that I am right that "we would want the Wikipedia policy text to require the editor following that text to ensure that some published reliable source has verified that each assertion in a Wikipedia page is true." That is not what I said and it is wrong to put words in my mouth andmisrepresent my views. What I said was the opposite, and Wikipedia policy is the oppositeof what Rednblu wrote. I am not going to play Rednblu's games by pretending or putting words in his/her mouth. Rednblu wrote the above quoted remark. Rednblu, you are wrong, dead wrong. And you cannot change Wikipedia policy on this matter. Slrubenstein | Talk 12:28, 25 March 2007 (UTC)

I think this is simply a case of the word "not" getting accidentally left out of a sentence. You quote Rednblu as saying that you said something; but when I look at Rednblue's actual comment, it says the opposite: the word "not" is in there, very close to the beginning of what you quote. I think when you notice the word "not" you may find that there is no problem. --Coppertwig 23:51, 25 March 2007 (UTC)

Rednblu wrote: "Rather, we would want the Wikipedia policy text to require the editor following that text to ensure that some published reliable source has verified that each assertion in a Wikipedia page is true." This is wrong: it is not what I said, and it is not our policy, and it indeed makes a mess of and undermines the whole idea of verifiability, not truth. Slrubenstein | Talk 10:19, 26 March 2007 (UTC)

Sorry. I guess Rednblu said two very similar sentences and I was looking at the wrong one. Still, please assume good faith. I think it's very easy to have misunderstandings when talking about this stuff (even if you do know whether the word "not" is in the sentence or not). There are a lot of fine distinctions that can be made, many of which are seen by some people but missed by others. It's not necessary to assume that someone is deliberately misrepresenting your views. --Coppertwig 01:10, 28 March 2007 (UTC)

Believe it or not, Steven, it is not "trolling" to disagree with one of your pronouncements. I read Rednblu as agreeing with your premise and suggesting that he takes that as his basis for his desired policy shift. He is asking that we do not simply verify that someone said it. He asks that we verify that someone asserted it as the truth. The difference is important to him, although I'm yet to figure out why. Grace Note 22:57, 31 March 2007 (UTC)

I never said it is trolling to disagree with me, people disagree with me all the time. I don't even think it is trolling to misrepresent me once, as you have Grace Note - people will good intentions misunderstand one another frequently. However, I do believe it is trolling to misrepresent someone consistently, which is what rednblu has done. As for the nature of the disagreement, I have made a claim about what our policy is. We can argue over whether or not this is what our policy should be, and there too we can disagree. But it is dangerous when someone consistently disregards and misrepresentes our policy, and disingenous too. Slrubenstein | Talk 11:48, 1 April 2007 (UTC)

Role of the truth 3 : "Sweeping Responsibility under the Rug" or "Axiomatic Introduction into Wikitruth"

Please allow me to start from a set of statements which when taken together IMHO may shed some light on truth vs. wikipedia.

1. "Wikipedia is Encyclopedia", i.e., a compendium of a certain category of human knowledge (some knowledge is unwikipedic, but may be accumuated elsewhere: wiktionary, wikitravel, wikicookbook (or whatsitname), etc.)
2. Wikipedians are harvesters of this knowledge (and since "anyone can edit", the whole humankind may be w'ians)
3. Without going very deep into philosophy about the meaning of the terms "truth", "knowledge", "information", "universe", etc., the knowledge has its own very simple but very deep hierarchy, which is grasped in two simple items 1 & 2 below:
   1. Statements about something in the "whole world around us"
   2. Opinions stated by some people about certain "statements about something"
   3. Corollary: the "opinions stated by some people" are also 'something in the "whole world around us"', which leads to the infinite chain of "opinions of some people about opinions of some people about opinions...
4. Regardless the definitions of "truth" (whose dicdef is remarkably tautological in wikipedia, btw :-), their common denominatior is that "truth guides us", i.e., the utility of truth, so that we can predict and plan, and our actions are based more than on stimulus-response chains. It is so even in most fatalistic philosophies without "free will".
5. Any kind of "truth" maintained by a certain category of people deserves to be reported in Wikipedia, because this type of knowledge has the utility beyond this category: e.g., knowing "their truth" (i.e., "opinions of some people about...", see item 3.2), we can predict their behavior, because "their truth" guides their behavior. Which "certain category of people" to be reported is judged by WP:NOTABILITY. It may be as small one as a single person: we do report on certain notable kooks, because they influence the world, if only "making a load buzz". Notice also that Wikipedia doesn't care whether the kook truly believes in his "truth" or simply pulls our leg, because the "kook's truth" (item 3.2) may be confronted in wikipedia with item 3.3, with all consequences of WP:NPOV.
6. Note: While the previous statement smacks strongly of a particular theory of truth known as "Consensus theory of truth", this similarity is superficial. Once again, I am to discuss only what some people recognize as truth, not whether it is truth indeed.
7. A Wikipedia's foundation principle (i.e., a principle beyond any dispute) is the "wiki process" as the decision mechanism on content, which is usually read as consensus decision-making (unless overriden by the ultimate authority). A corollary is an egalitarian absence of a special role of experts as judges of any "truths". Wikipedia:Notability is the major workhop of decision-making: despite its lowly label as "guideline", WP:NOTE is where the ultimate fate of any content lies, and this fate, "to be or not to be", is decided by the whole community (even if implicitely, by the absense of any "nays").
8. In absence of any mechanism for the incorporation of "revelation" in the process of writing wikipedia, the whole actual content of any article is a collection of two categories of statements:
   1. "Undisputable truths", i.e., those perceived to present no trouble for consensus decision-making (but even these may be freely questioned). Among other things, this itemn includes direct citations from external sources.
   2. Opinions contributed by wikipedians. (It may seem paradoxical, but consider this: even if a wikipedian refers to some very reliable source, it is nothing but their implicit opinion that the referred source says this or that).
9. "Experts" in "real world" are recognized as such by a "certain category of people" and these people delegate the experts the right to utter and judge "truth".
10. "Wikipedians" within Wikipedia and "experts" within "real world" share the same important trait: both of them are subject to peer review by their peers. In wikipedia the circle of peers is the whole wikipedia community. "Real life" experts have their own rules of the game; sometimes the circle of peers consist of a single person (and in this respect a kook and the Pope are similar, the difference being the number of people that delegate them the rights to "truth").
11. If "experts" are entitled to utter and judge "truth" recognizable by a "certain category of people", then if this category is deemed notable by wikipedia community, their opinions as reported by wikipedians are allowed in articles. What is more, in many cases opinons of experts are safely declared as "facts", e.g., entries from demographic tables.
12. In some cases even "experts" may recognize their limitations and formulate their statements as "opinions" or "hypotheses". In other cases experts may disagree, especially experts speaking for different categories of people. This is where WP:NPOV kicks in.
13. It is technically possible that a certain wikipedian is confirmed to be a "real-life expert", and barring any other provisions a viable scenario could be that they write something in wikipedia from their own name, while his peers perform the same scrutiny they do at conferences, during journal peer reviews, etc. After all, what's big deal: he is an expert, i.e., in charge of "truth", his "peer review" is one mouse click away in wiki environment, why an intermediate step? It must be a decision of the community sorry, wikipedia is simply not a publisher of original thought; just as wikipedia is not a wiktionary, we simply say: there are other media for doing this. All the more, "original research" is a no-no for "ordinary" wikipedians, about whom we don't know whose truth they are representing.

The above train of thought was intended to demonstrate that WP:NOR is independent of any interpretations of the meaning of the term "truth" and in fact is a consensus-defined part, although a nontrivial and groundlaying one, of WP:NOT. While its importance goes a beyond simple WP:NOT item, it is still an atomic decision: shall we allow experts to publish their new research in wikpedia or not? With all due respect, we are saying "not", as simple as this. As a byproduct, we cancel out various kooks and trolls. Mukadderat 07:39, 27 March 2007 (UTC)

Before I proceed with a similar exposition to demonstrate:

• WP:V is equally intependent of "truth"
• WP:V and WP:NOR are independent of each other
• WP:ATTR is a tool to enforce both WP:V and WP:NOR,

I would like to see opinions as to whether all the above is understandable and reasonable.

And a final piece for today: while in current formulations it can be shown (i.e., I can try to convince you) that WP:V may be ensured by means other than WP:ATTR, it may also be shown that WP:ATTR may be extended to cover these extra means as well. Therefore for all as of today technically WP:ATTR may be equivalent to WP:V + WP:NOR. And therefore in restructuring the policies related to "truth" the main decision is the choice which policy is to be considered as primary or "core":

1. procedure-oriented rule WP:ATTR, i.e., attribution as a way to ensure that statements in articles reflect truth for some notable category of people
2. goal-oriented rule WP:V, i.e., contributions to wikipedia must reflect truth for some notable category of people and we do not preclude other possible ways to enforce truth (an example of a possible way is to allow recognized experts to write from their name; cool down, it is just an example, I will be the first one to vote against such a possibility :-)

My heart is strongly against the procedure-oriented approach, because it gives a priority to a bureaucratic way of decision-making, with all related corollaries of the Parkinson's Law.

I hope that my reasoning brings the moment of decision closer, rather than further by wasting your time for reading my rant. -- Mukadderat 07:39, 27 March 2007 (UTC)

Our previous three main policies (NOR, V and NPOV) were all linked, and ATT and NPOV are linked. Also, I am not sure what you mean by V is independent of truth. The basis of Wikipedia's NPOV policy - which also informed NOR - is "verifiability, not truth." Clearly, verifiability is presented as a principle distinct from "truth." But it is also presented as our alternative to "truth" as a criteria for inclusion. I think it is impossible to understand V if one does not understand how it is a radical alternative to "truth" as a criteria for inclusion/exclusion for the encyclopedia. Slrubenstein | Talk 09:50, 27 March 2007 (UTC)

Colleague, thatnk you for your interest in my text. I confess I wrote it primarily to set my own thoughts clear, rather than to influence politics: I am not a political person. Now, I would like to underline my intendent accent once more: I am demonstrating that the important relations of WP:NPOW (and WP:V, later) are not that they are "distinct" or "alternative" to "truth". I am saying that they are independent of (the definition of the) "truth". IMHO this point of view has at least three very important traits

1. We don't have to struggle to figure out our definition of truth
2. We don't put wikipedia in opposition to "truth" (which may be read from words "distinct" or "alternative")
3. And as a consequence, we preempt ourselves from further criticism of the validity of wikipedia's information. (I am trembling in horror waiting some nasty blogger detects this discussion and tells to the world that wikipedians do not care for truth! and seek for its replacement!) Mukadderat 01:15, 28 March 2007 (UTC)

Mukadderat, I agree with most of the (too) long essay above, but I have two objection.

1. This policy actually emphasizes that not every thing needs to be attributed, what is required is attributability, which can be precisely defined as "reflect truth for some notable category of people". Thus, the policy is striving for the very same ideal that you are discussing!
2. Why then the procedural focus on attribution? Perhaps I lack imagination, so can you give one real example of an appropriate way to "reflect truth for some notable category of people" other than attribution? One option is of course original research, but that's not allowed. True, that's an independent decision, but that doesn't make it unrelated: our goal is V, and we have decided we can't do OR, so all that remains is ATTRIBUTE!

There are some qualification, for example, we might allow purely analytic truths that we deduce by logic, and WP:ATT does make that exception. I any case, your argument was interesting, I hope you found this and Slrubenstein's response helpful. --Merzul 00:07, 28 March 2007 (UTC)

Yes, I agree that the essay is somewhat long, but the problems with short essays for important topics is that they are either too simplistic or require certain leaps of logic or faith, or express someone's opinion, which may be easily refuted by "OK, it is your opinion. let's vote" my goal was to invoke logic based by interconnecting already accepted postulates of wikipedia.

About Objection 1: I don't intend to replace our policies: I am showing their logical place within some less disputable (or even indisputable) rules of wikipedia. Re: "Attritabutability": sorry I don't see the ground of your disagreement; so far this very long text only was intended for WP:NOR. I anoly promised to continue. And in my future text it will be logically see that "immediate attribution of everything" is an overkill.

About Objection 2: Re: "procedural focus": Because the policy simply dictates "how" to decide on inclusion of text, i.e., the procedure of decision, ingnoring and even "substituting" the notion of "truth". It is a well-known stumbling log in histories of many organizations that when procedures take priorities over strategic goals, then it leads to big troubles. Re: "real example other than attribution" Again, I wanted to cover this in the continuation, but since you are impatient, here is one: things of real world may manifest themselves. Imagine that instead of nice photos wikipedia will have a posibility to incorporate webcam feeds into articles. Now let us make this more encyclopedic: wikipedia has a direct feed from a device which measures the distance between continents (we know they are drifting) and the very current distance value is displayed right here: <<13,207.3 km>>. A quite encyclopedic piece of info and very verifiable. And some bureaucratically minded people may say that this data are attributable (attributable to the device in question). But some other people may readily seen that this way of verefiability has its own very specific issues to resolve and rules to follow and hence better to be handled by a quite separate policy. If you think it is a fantasy, let's talk in 5 years. If you say that we shall scratch our head when it starts itching, I will say such an attitude often leads to disasters in many projects, and if you can easily make forward-looking provisions to eliminate future restructuring, then why not. Mukadderat 01:15, 28 March 2007 (UTC)

Sorry for being very brief, but I just want to clarify my first objection as I think that was the crucial one. You are essentially arguing that ATT is taking a procedure-oriented focus, and I was saying that it's not entirely so. ATT defines a goal-oriented ideal, that of attributability, and I think that captures the idea that we want material on Wikipedia to "reflect truth for some notable category of people", basically we want material to be attributable to experts in the relevant fields. I think you are saying that there are truths held by most people, which nevertheless can't be immediately attributable to them, for example, there are certain manifest facts that would be considered true by most people and yet they don't write it down in a published source that they think so. Well, manifest sources are available today, but this is a highly problematic issue! For example, citation indexes do serve to verify that a certain person has not published an article in the relevant field! It is not original research to report that a citation index returns zero published articles, and it is immediately verifiable. But who is it attributable to? It's a data stream, you can't attribute statements to such a source, and that might be a good thing! As it does capture what Mr Wales would like V and OR to be, I'm sure he wouldn't want us to include information that some living person hasn't published a single article on subject X and yet is selling products in this field and so on. Once we get used to attribution, we will no longer cause Jimbo such headaches, it's a pity he believes this policy weakens NOR when in fact it strengthens it. Sorry, I completely diverted from the philosophical discussion at hand, I'm going to sleep now, it's safe to ignore me :) --Merzul 03:27, 28 March 2007 (UTC)

One thing we need to be clear about (and this si semantics, not epistemology) is unattributed = unsourced, but unattributable = original research - in practice, they are not the same thing. Slrubenstein | Talk 10:40, 28 March 2007 (UTC)

I thought of another compromise that I would accept, though I wouldn't find it completely satisfactory: changing the period to a colon at the end of the sentence, so it would read "The threshold for inclusion in Wikipedia is whether material is attributable to a reliable published source, not whether it is true: Wikipedia is not the place to publish your opinions, experiences, or arguments." I forget whether I or anyone had suggested this earlier.

I have no problem with the original words "verifiability, not truth". My issue is with having words such as "not whether it is true" or "not truth" without qualification by the presence of a word such as "verifiability" or "solely" or any of many other suggestions. One suggestion is to include the sentence "Not everything that is attributable is worthy of inclusion" somewhere in the first section. I see that one too as a compromise. If you search for "possible alternative wordings" above you can find quite a few alternatives I've suggested. I'm sure others can be thought up.

I'd like to find at least one user who supports the new wording who is willing to discuss back-and-forth with me long enough to arrive at an understanding like "OK, I understand that you see this meaning in these words, while I see that meaning. And I understand that you want the words to mean this, for this reason, while I want them to mean that, for that reason," and to consider at least three or four possible alternative wordings. --Coppertwig 13:44, 29 March 2007 (UTC)

I don't mind any of these proposals, I only object to the Conservopedian "not solely if it is true". I have posted your proposals below, feel free to edit them. --Merzul 18:24, 29 March 2007 (UTC)

I accept the current wording as a compromise:

• Wikipedia is a compendium of well-established knowledge. The threshold for inclusion in Wikipedia is whether material is attributable to a reliable published source, not whether we think it is true: any reader should be able to verify that material added to Wikipedia has been published by a reliable source.

I particularly like having the link to "Doesn't Wikipedia care about truth?" as a clarification. Thank you very much, everyone. --Coppertwig 12:39, 18 April 2007 (UTC)

Clarification: I accepted it merely as a compromise, not as a version I fully accepted or liked. If, for example, the link to "Doesn't Wikipedia care about truth?" is removed, I no longer accept it even as a compromise. I strongly object to new policy saying "not truth" or "not whether it is true" without satisfactory (to me and others) qualification. --Coppertwig 17:06, 14 July 2007 (UTC)

Coppertwig's compromise proposals

Current form	Compromise 1 (colon)	Compromise 2 (not worthy of inclusion)
The threshold for inclusion in Wikipedia is whether material is attributable to a reliable published source, not whether it is true. Wikipedia is not the place to publish your opinions, experiences, or arguments.	The threshold for inclusion in Wikipedia is whether material is attributable to a reliable published source, not whether it is true: Wikipedia is not the place to publish your opinions, experiences, or arguments.	The threshold for inclusion in Wikipedia is whether material is attributable to a reliable published source, not whether it is true. Wikipedia is not the place to publish your opinions, experiences, or arguments. Not everything that is attributable is worthy of inclusion, see undue weight.

I just noticed the irony in the second proposal: not worthy of inclusion. I wasn't intentionally implying that the sentence about things not worthy of inclusion is not worthy of inclusion. But that was actually the objection against it, if I recall correctly, that it's obvious and so it isn't worthy of inclusion. --Merzul 22:10, 29 March 2007 (UTC)

Here is a central problem with our policy structure (and the reason it's so beloved by the likes of Steve Rubenstein). We take the philosophical basis that no claim is more valuable than another: that one does not need to engage in the discernment of the "truth" of a claim. However, having taken that as square one, we proceed to introduce means to value claims. We define certain sources as "reliable" (which surely means something like "more likely to be truthful") and we note that some things are "worthy of inclusion". But what are the criteria for worth here? Presumably, "more likely to be true" is one of them. "We agree they are more important" is another. But judging that one thing is important and another not is surely contrary to NPOV (unless our reason for judging it so is "more people say it in the Western press", which, preposterously, is our basis for considering something important). So we can pretend to be value neutral (which is what the "neutral" in NPOV more or less represents) while introducing our values by the back door. I don't know what our articles would look like if we were truly neutral. I don't know whether it's even possible with the rather linear approach we have taken. (Naturally, when I think about this, I think about the clitoris article, which I tried hard to make inclusive. It represents one point of view: that pictures of genitals are acceptable, while dismissing another, that they are not, totally. The general argument for this is that an encyclopaedia should include this, that or the other. However, by defining an encyclopaedia in a particular way, we destroy neutrality. Instead of NPOV, we leave ourselves with an "encyclopaedia point of view". A solution that encompassed both POVs: dual articles, one with image, one without, was entirely dismissed by liberal-value zealots. But this is the kind of creative solution, possible in this medium, that an encyclopaedia truly interested in "neutral" presentation of views, should include. Instead, we have a push to create "stable" versions of articles. I do not know how we can distinguish between saying "this article is stable" and "this article is true".) Whether it matters that we take this approach to constructing the encyclopaedia depends on what your vision of it is, and how much you mind that it's simply a compendium of liberal, orthodox statements of how the world is. (This is not to mention even those areas in which the orthodox media simply have not had much to say, or are not informed enough to have much to say; or those in which the focus of orthodox media leads to representations of subjects that are not comprehensive: for instance, biographies of people who have made the news for chiefly negative reasons.) Personally, I prefer a broader interpretation. Grace Note 23:21, 31 March 2007 (UTC)

I think the third proposal (including WP:UNDUE) is the best. To put it quickly, clearly two broad interpretations of truth oppose themselves here (and I will include a third one):

• the first one believes in absolute truth (from a Revelation or otherwise; it does not accept rival truths). This is what Wikipedia explicitly opposes, in order to be able to struggle against fanatism.
• The second one believes in some relativity of truth (cultural relativism & others), and is the one supported by Jimbo Wales, when he said somewhere that we should abstain ourselves from judging what is true, or not, and simply attributes opinion to individuals (or institutions — it's quoted here, see Wikipedia talk:Attribution/Role of truth#Do not change the core policy).

Both have found their limitations here, and in fact Wikipedia has worked on another model, which more or less inspired itself by works of the scientific community: if one publishes something, one should publishes the material which goes with it, in order to let the reader verify the statement. Without sources, a Wikipedia article is useless, even if true, because the reader can not verify what is been said. Hence the importance of attribution.

However, this seems to bypass the importance of WP:UNDUE which helps to restrict statements which clearly are false (even if correctly attributed) and marginal, fringe views. This is very important, because else, many articles would give way too much space to fringe, clearly false, views. Dumping in the meanwhile any relation between Wikipedia and knowledge.

In other words, if an encyclopedia has something to do with knowledge, if Wikipedia is an encyclopedia, and if knowledge is related, in some way or another, to truth, it is an illusion to think that Wikipedia can work without reference to truth.

Thus the importance of not dismissing truth so quickly, and of maintaining UNDUE weight. Last comment: we have noted here that reliable sources are not always the same, depending on the context. This points to another problem: if you take the case of a court judgement, one can hardly assert that a witness is "reliable" in an absolute, essential way (that is, granting him the equivalent of Papal infallibility). One can only try to see if this or that statement is reliable.

Thus, when a globally reliable source says something which is obviously false, this must be included in the article (as it is not every day that such things happen), but with caution and alternate views (I seem to oppose my support of UNDUE, but I don't think I'm contradicting myself, the nature of the problem makes for complexity). This is particularly important in politics, which are made of lies and of public denials: a public denial of a government, when loads of others reliable sources have pointed out the reverse, must be included, but according to UNDUE: that is, it must not be presented as the absolute truth.

In two words: Wikipedia can not abstain itself from a reference to truth (least it becomes a battlefield for rival, and fringe, opinions, held by fanatics from all environments), and can not either abstain itself from UNDUE. Tazmaniacs 18:46, 2 April 2007 (UTC)

Role of truth 4: Conservapedia

Conservapedia Commandment	Wikipedia content policy
Everything you post must be true and verifiable. Do not copy from Wikipedia or other non-public domain sources.	The threshold for inclusion in Wikipedia is whether material is attributable to a reliable published source, not whether it is true.

I really find this contrast interesting. I'm proud to be a Wikipedian precisely because we are not defining the truth. I love NPOV. --Merzul 18:44, 29 March 2007 (UTC)

Count me in... ≈ jossi ≈ (talk) 19:44, 29 March 2007 (UTC)

What is interesting is the use of "or" in Do not copy from Wikipedia or other non-public domain sources. LOL... ≈ jossi ≈ (talk) 19:45, 29 March 2007 (UTC)

Conservapedia doesn't use the GFDL, so if they were to copy from Wikipedia it would be a copyright violation. What they are saying is totally appropriate.--agr 21:03, 29 March 2007 (UTC)

If the only standard is that there's an attribution somewhere, Wikipedia will degenerate into a gossip column. If the only standard is that a statement is true, Wikipedia will become an online collection of math identities (although there may be problems there with the NOR policy!) The problem is that the attributable false trumps unattributable truth. I would expect that most entries would be both true and attributable, and that those of questioned attribution or questioned truth should be so labeled. We should be grown-up enough to admit that sometimes "the truth" isn't known, and that sometimes what we know can't be attributed. htom 21:08, 29 March 2007 (UTC)

The only standard is attribution? That's clearly false attribution! WP:ATT is saying no such thing. There are two concepts:

1. What sources to use, and how much to rely on each source? (covered by NPOV, especially Undue weight, and notions of reliability.)
2. What is the appropriate use of our sources? (covered by ATT.)

We need to develop a culture of fair and honest attribution. Perhaps it is verifiable that the policy says "not whether it is true", but the reasoning above is taking it out of context, and is ignoring the very important qualifier about source reliability. Saying that the standard is that "there's an attribution somewhere" is therefore not attributable to WP:ATT. (See how useful the ATT terminology is, I don't know how I would argue this using V and NOR. Well actually I do, but still.) :) --Merzul 22:04, 29 March 2007 (UTC)

Very reliable sources have been known to be wrong. htom 22:15, 29 March 2007 (UTC)

Yes, but the way to counter that on Wikipedia, is to dispute the reliability of the source in question, either by

1. finding other sources who dispute the statement, or
2. showing some other evidence of why an otherwise reliable source isn't reliable, e.g. has copied from Wikipedia :)

Also, attribution doesn't always have to be literal and verbatim. It is far more important that it honestly captures the idea. Let's say you are citing a mathematics textbook and there is a small flaw in a formula. If a few of our esteemed mathematicians agree that an index or something is wrong. Why shouldn't they fix it? The idea is still attributable to the source! Unless somebody challenges this, for a valid reason, and not just "Oh no, it's not the same, I can't verify that it says so!" Finally, for the killer argument: so what if reliable sources are wrong? Here are two possible cases:

1. I cite Nature about a fact, but it turns out to be wrong.
2. You write something you are certain is true, but it turns out to be wrong.

When experts in the field will evaluate Wikipedia, which scenario would you prefer? --Merzul 22:46, 29 March 2007 (UTC)

Neither. I'd prefer that they find the citation to Nature, and a bold editor who noted that the information was being challenged. Or that someone found an error in something I wrote, and corrected it with a reference to the correctness. (And then I could correct their RS's problem and explain that I was correct, of course, or make /blush/ noises in the talk page.) htom 23:26, 29 March 2007 (UTC)

The process that you are indicating here is in quite stark violation of WP:NOR, we are not encouraged to refute a source like Nature. Sure, if you find a fault, and point it out on the talk page, that's good. If I agree with you, I will remove the material, but if I think Nature is right, then I'm sorry, you can scream all you want, but pseudo-science has no place on wikipedia. (I'm quite sure you are not promoting pseudo-science, but what makes your bold fault-finding any better than that of the pseudo-science advocate?) --Merzul 16:22, 31 March 2007 (UTC)

Define Truth

This may be a rehash of arguments already made, but I've never seen it put as distinctly as it should. The statement "the threshold for inclusion in Wikipedia is attributable to a reliable published source, not whether it is true" has different meanings. To the casual reader, "not whether it is true" is a loaded statement. Most everyone not familiar with Wikipedia would read that and assume that patently false or libelous statements are perfectly acceptable, as long as you can prove that someone said them. Example (ludicrous for emphasis): What if some political pundit says "Politician A is an alien from Mars that eats puppies"? You are able to attribute that statement to someone. The fact that the statement was uttered is true, but it would be obvious to anyone that it is false. No one would object that a statement such as this should not be included in Wikipedia. It is libelous, factually wrong, and inappropriate. Wikipedia is not a gossip rag or the National Enquirer.

We need to more clearly define what we mean by truth. Truth as in accuracy is different than truth as in "I believe it to be true." Examples of this would be "Global Warming Is an Immediate Crisis" and "global warming is the greatest hoax ever perpetrated on the American people." Both statements are true, i.e. the statements exist and were said by the people attribted. However, reasonable people can disagree on whether they are factually true, i.e. is global warming a real.

The Wiki-definition of truth is in this second category, one that gets at underlying truth that is based on personal belief (Does God exist?) versus factual truth (Moon revolves around the Earth). We need to ensure it is clear that the policy is designed to guarantee NPOV by including both sides of controversial arguments in articles. However, it is not always that simple. In some articles, like the Flat Earth Society you could include arguments that the "Earth is Flat," because it is appropriate in the context of the article. But I do not think you necessarly need to give the "Flat Earth" argument equal time in the article about the Planet Earth. In short just because something is printed and can be attributed doesn't mean it should be included. The goal is to keep personal bias out of articles, not to exclude facts or to include false information in the name of completeness. Editors should be use their collective judgement in crafting articles. Dcmacnut 21:42, 29 March 2007 (UTC)

• Please see my essay #Role of the truth 3 : "Sweeping Responsibility under the Rug" or "Axiomatic Introduction into Wikitruth", which says that we don't have to define "truth" at all, and the policies will still work. Mukadderat 21:50, 29 March 2007 (UTC)

I had seen that, and went back to read it in more detail after your post. I see your points, but the purpose of my post was, like yours, to put my own thoughts in some cohesive form. What I think you are describing is a world where "truth" is simply "truth" in all its forms. The majority of editors would agree with you, I think. We as Wikipedians know what we mean when we say "truth." However our policies are there to explain to newcomers just what Wikipedia means. Those newcomers might not understand our definition of truth. That's the point I was making. When we say "verifiability, not truth" or some form thereof, we need to be clear in what we are saying. I think a definition of truth is necessary. We need to differentiate between "truth," as in verifiable facts, versus "truth," as in "personal opinion." My global warming example one. Both Al Gore and Senator Inhofe have their own opinions about global warming, and both view their own view as fact, but they are still opinions. My fear is that people are focusing too much on the assumtion that by saying "no truth" that Wikipedia is endorsing false statements in its articles, even if one can prove those false statements were uttered by someone else. Bottom line, Wikipedia is not a place for editors to debate their own opinions or the opninios of others by proxy. Somoneone once said that "everyone is entitled to their own opinions, not their own facts." Wikipedia needs to be factual and unbiased to the greatest extent possible, and we need to be clear to newcomers just how Wikipedia's policies are meant to ensure that.Dcmacnut 23:41, 29 March 2007 (UTC)

So how do these various competing standards play out where truth really matters, such as in the theory of gravity? Isn't "verification as fact" much more important than "attribution" to either Al Gore or The Other? --Rednblu 00:20, 30 March 2007 (UTC)

WP:NPOV is the rule to "play out" between standards. WP:CITE should specify rules which say that the opinion of Al Gore about gravity are irrelevant, while his opinions about US politics are, even if many think they suck. Please undertand the major point expressed in different ways by different people: wikipedians cannot judge "truth", even if it "really maters". However important "the real truth" is, we cannot judge it, period. All we can do is to report other's judgement, in a balanced way. Mukadderat 00:50, 30 March 2007 (UTC)

Gravity is a really good example. While we can agree gravity exists, the how and why is still a matter of scientific contention. If we want to publish the "truth" about gravity, we should publish in a scientific journal, or if our theory is offbeat, maybe some other publication. At Wikipedia all we can do is refer to the most reliable sources we can, and present them. --Michael Johnson 01:11, 30 March 2007 (UTC)

Ok. Then I would like to examine the "role of truth" specifically for gravity. What is it that gives the gravity page the high quality that it has? I would say that truth is a major component of the high quality in the gravity page. That is, by my standards, the gravity page summarizes quite well what the ReliableSources have Verified about gravity, and there is No original research by the editors who have written the page. Furthermore, the gravity page gives me Attributions to where I can find the original writing of those ReliableSources that will lay out for me the standard of truth by which those ReliableSources have accomplished their Verification of the assertions on the gravity page. And, as User:Mukadderat says, neither the 1) editors nor the 2) ReliableSources have established "the real truth." Furthermore, in my estimation, both the 1) editors and 2) ReliableSources have reported others' judgment, in a balanced way -- given the methods for Verification that the ReliableSources have specified. And, as User:Michael Johnson says, there are elements of even gravity for which truth has not been determined in a way that ReliableSources can Verify among each others' works as being truth. And I find that the gravity page gives me good indications where these unknown areas are. What can I extract from this? It seems to me that Attribution plays only a small part of constructing a quality page such as gravity. What matters is Verifiability; I want only the Attributions that point me to the ReliableSources that have 1) defined explicitly their method of Verification of truth and 2) actually have verified each other's conclusions. Attribution is merely a means of pointing me to the ReliableSources that have Verified as truth within their explicitly defined methods of Verification the assertions in the Wikipedia gravity page. That is, the fundamental element of quality in the Wikipedia page is Verification. By analogy, Verification is the gold; Attribution is merely the map to get to the gold. Both are important. But for the gravity page, the mere map to anything without Verification is worthless, in my opinion. --Rednblu 04:12, 30 March 2007 (UTC)

I'm starting to see what you are trying to get at, I agree "accurate" or call it even "true" information is the goal, I also agree that the way we reach this is that we attribute to sources that within the established fields of expertise have verified as "true" (in their own operative definition of the term) whatever assertion we use. However, I hope you do realize how incredibly confusing your argument is, and it is also greatly in contrast with the accepted definition in WP:V.

• You are saying verifiable on Wikipedia means we are only including information that has been verified by experts to be the truth as they define it in their field.
• WP:V says verification means that "any reader should be able to check that material added to Wikipedia has already been published by a reliable source." Nowhere does WP:V talk about things being verified as true!

I don't mind your definition of verifiability, it's actually quite good and sophisticated. But that is not what WP:V is about at all. If WP:V would use your definition of verifiability explaining it carefully, I wouldn't mind; but having a policy called verifiability that then immediately defines verifiability to be attribution is quite ridiculous. And I still don't understand why WP:ATT is being blamed for this distinction, when it is clearly in the second sentence of WP:V. --Merzul 16:58, 31 March 2007 (UTC)

Well said! Your thoughts give me many further insights, and there will be lots of time for those. Does someone else have a comment? There seems to be a wide difference of opinion about what WP:V means, particularly in the "role of truth." Is that true? --Rednblu 17:19, 31 March 2007 (UTC)

Rednblu, in scientific articles you are probably right, and a formulation at least somewhat similar to yours would be useful. However, while science is a/ willing and able to describe its basis for considering things true and b/ generally built on axioms that are implicit in its descriptions, so that contesting its truths amounts to contesting the axioms it is built on (and consequently not doing science), other areas are not as clear cut. Still, I do think that what you're suggesting more clearly describes what we actually do here, or what many aim to do. Grace Note 05:28, 2 April 2007 (UTC)

Would you agree that the standard in the scientific method is "Verifiability, not truth"? --Rednblu 05:42, 2 April 2007 (UTC)

Hmmm. I think scientists would more readily agree that "verifiability is truth". Grace Note 06:10, 2 April 2007 (UTC)

Can you think of a situation where 1) "Verifiability, not truth" would lead to a result different from 2) "Verifiability is truth"? --Rednblu 06:47, 2 April 2007 (UTC)

Do not change the core policy

This entire discussion is irrelevant and any claim that privileges the idea of "truth" is in conflict with our core policies. According to the original core policy, NPOV, what is "objective" is not gravity, but that some people hold certain beliefs about gravity. And it is these beliefs, completely regardless of truth or truthfulness, that are represented in Wikipedia. Here are some quotes from the first version of the policy, from 2002:

• Perhaps the easiest way to make your writing more encyclopedic, is to write about what people believe, rather than what is so. If this strikes you as somehow subjectivist or collectivist or imperialist, then ask me about it, because I think that you are just mistaken. What people believe is a matter of objective fact, and we can present that quite easily from the neutral point of view. --Jimbo Wales
• But even on such pages, though the content of a view is spelled out possibly in great detail, we still make sure that the view is not represented as the truth.

These ideas provided the basis for another core policy that was developed several years ago, namely, or verifiability policy the essence of which is "verifiability, not truth." Slrubenstein | Talk 14:23, 31 March 2007 (UTC)

I disagree. The article on gravity states "Gravitation is a phenomenon through which all objects attract each other." That is a statement of objective fact. Gravity exists. There is no viable theorgy that "gravity does not exist" and as such you don't need NPOV on the question of whether gravity exists. Where individual beliefs or opinions come in, and where Wikipedia NPOV and verifiability is essential, is in opinions of how gravity works. That is where you state objective "beliefs" about the theory of gravity to explain the various viewpoints on the subject in a neutral manner. There can be alternative theories, but the central fact that gravity "is a phenomenon through which all objects attract each other" is not open for debate. Atlernative theories get are listed in the article on gravity, but they do are not used in such a way to claim that gravity does not exist. The article on the Moon landings is another good example. The article starts from the the basis that the Apollo moon missions in fact took place. It is verifiable. There is a small section describing accusations that the moon landings were faked, but that section is written appropriately in NPOV by stating "so and so believes they were faked" and "so and so thinks the moon landing conspiracy theorists are flat wrong."

The point I was trying to make in some of my comments is that there are certain objective facts that exist as true. I guess my biggest problem is changing the "verifiability, not truth" to "whether material is attributable to a reliable published source, not whether it is true." The verifiability leaves open for some analysis if whether the statement is libelous or false. Limiting the standard to just attributability leaves open to any false or libelous statements as long as you can point to someone that said them. Jimbo has also said that just because you can attribute something to someone doesn't mean it should be included in Wikipedia, particularly contentious statements about living persons. I know I may be oversimplifying matters, but the purpose of policy is to explain our standards of conduct to outsiders. Those that already edit Wikipedia "get" what we mean by "truth." Someone new probably wouldn't.

In its simplest terms you can declare "2+2=4" as fact, and you'd be hard pressed to find any reliable, verifiable source that claims instead that "2+2=5" (George Orwell's 1984 notwithstanding).Dcmacnut 17:17, 31 March 2007 (UTC)

But if somebody believes the moon landing was a hoax, clearly they think that is the truth. Now, do you want to engage in a long argument with them about the truth? Or do you prefer saying "Dear sir, what we personally think is true is not relevant to our discussion here, we present what the majority of reliable sources have published on the matter". If you imagine yourself being in a content dispute with someone, who thinks the moon landing didn't happen, who thinks all the NASA-presented evidence is bogus, do you really want to discuss truth with them? I accept the sophisticated view of truth given by Rednblue, but would you want to explain such notions of truth with various conspiracy theorists? Isn't it easier to just tell them "I'm sorry, truth doesn't matter, we represent the mainstream scholarly view", and go on with it.

And regarding the libellous facts, the way I understood the matter, what was inserted wasn't actually inaccurate, was it? It was one-sided and paid unnecessary attention to an obscure law-case, but I think it was accurate. It was giving undue weight to something that had not been covered by secondary sources, but I think what they had inserted was in fact "the truth". I might be wrong, but at least on Talk:Gillian McKeith, there is an immense drive to include an infinite amount of criticism about her. I think all the criticism is true, she is most probably a complete quack who knows very little about real nutrition science, but what I said of her is quite disrespectful and it would violate WP:BLP to give so much weight to so much criticism. In these cases, I don't see that truth is helping us. --Merzul 18:08, 31 March 2007 (UTC)

Would Verifiability help? It is clear that "truth" would not help. --Rednblu 18:41, 31 March 2007 (UTC)

Yes, of course, but the only question is how practical and helpful is the terminology. I personally find the attribution terminology easier to work with. If somebody synthesizes material, let's take my favourite example (a real one):

"People suffering from mental illness often hallucinate.^[1] Religious figures are known to have been inspired by visions.^[2]"

Both are cited, and no conclusions are drawn in the text. I would ask the editor to not use these sources next to each other because this usage isn't exactly attributable to the second source, which never implied any connections with mental illness. I would ask for a source that explicitly makes this connection, because it is a controversial one. For me, attribution has the connotation of being fair and scholarly, respecting the context in which one uses sources. I often ask myself, would the sources in question, if they could edit wikipedia, agree to being used in this way. I find this way of thinking a lot easier than the WP:SYNT policy. --Merzul 19:17, 31 March 2007 (UTC)

I agree with you more than you think. My point all along has been that we need to strive for truth in wikipedia articles, otherwise they will forever be viewed as inaccurate. What you seem to be saying is that we don't need to include every "crackpot" version of the "truth" in the article. I would agree that I would not want to give undue credence to moon hoax conspiracy theorists, because those views are readily disputed by mainstream knowledge and reliable sources. But the way I read the "not truth" portion (where attribution is the only constant) means that we couldn't even include a discussion of the moon landings because it is one person's view of the "truth," if you look at it from the standpoint of a moon hoax adherent. I believe your standard of ensuring wikipedia "represents the mainstream scholary view." Simply relying on something being "attributable" to someone is not by itself a firm enough standard in my opinion.Dcmacnut 18:16, 1 April 2007 (UTC)

Irrelevant discussion

This discussion is not relevant for Wikipedia. Wikipedia was never about "truth" but about reporting what published sources say about a subject. If some editors do not like this concept, they can go and edit other wikis such as Conservapedia. ≈ jossi ≈ (talk) 01:10, 1 April 2007 (UTC)

A yet another example of blatant disrespect of some wikipoliticians to fellow wikipedians. `'mikka 19:19, 6 April 2007 (UTC)

No one is forcing you to take part in any discussion. Wikipedia is about whatever its members think it's about. If you don't like that concept, I suggest you fork off and make your own wiki. Grace Note 09:17, 1 April 2007 (UTC)

Grace Note, no one is forcing you to participate in Wikipedia if you reject our core policies. Not, instead of disrespecting Jossi why don't you take his point seriously that core policies - in this case, one that has guided Wikipedia since its inception and which has the support of the overwhelming majority of Wikipedians - do matter? Slrubenstein | Talk 11:40, 1 April 2007 (UTC)

I stand by my statement. You re welcome to discuss nuances within the core policies, but I would continue to argue that this discussion and its peculiar timing detracts from this project. ≈ jossi ≈ (talk) 18:53, 1 April 2007 (UTC)

Steven, you have an unfortunate habit of using "we" to exclude others when in fact "we" includes them. Wikipedia does not have laws handed down on tablets. Its laws are made by the people it consists of. If they are flawed, it's perfectly reasonable that we should talk about fixing them. If you don't believe they are flawed, it's perfectly reasonable for you to defend them. I don't see your doing so though. I see you trumpeting their rightness without bothering to substantiate your position. Jossi simply wishes to close down debate. I will continue to disrespect that just as I do in every instance that a KoolAider insists on trying to do it. Grace Note 05:06, 2 April 2007 (UTC)

Grace Note, the character of my reaction has to do with the fact that certain people, including Rednblu and others involve din this discussion, have not been saying "here is a policy that we have had for many years but it is fundamentally flawed, can we have an open discussion;" rather, they have been making claims that Wikipedia is in fact based on privileging truth. I think you misunderstand my point. I am not arguing that the policy is right and opponents are wrong. I am arguing that the policy states that verifiability is not truth, and that verifiability is the standard for inclusion, not truth. In short, I am making an argument (yes, very rigidly) about what the policy states. If you and others want to change the policy I would happily join in with my reasons for taking whatever position I end up taking. But at least have the honesty to say, "We are proposing a change in the policy" rather than - and I am not sure what I am about to say applies to you, but I am sure it applies to Rednblu and several others - disingenuously assert that you (they) ar enot changing the policy but rather anyone who claims "verifiability not truth" is somehow weakening the policy. Slrubenstein | Talk 09:28, 3 April 2007 (UTC)

Something must be true to be verifiable. Verifiability standards are a measuring rod for truth/accuracy.

The comment from Henrygb above is quite relevant, and I copy it here:

That is not what it has always meant. Try the first time WP:V was labelled as policy[3]. What it meant then was "your belief that a statement in an article is true is not enough; it has to be possible to check the statement is accurate; providing sources enables this to happen". What it has become in WP:ATT is more like "we don't care whether a statement in an article is true or accurate; all we care about is whether we can find someone who said it". --Henrygb 18:14, 23 March 2007 (UTC)

An important distinction exists in how we can interpret the sentence "The threshold for inclusion in Wikipedia is verifiability, not truth."

One possible interpretation is "if a statement is true, but not verifiable, that doesn't meet the threshold for inclusion in Wikipedia." This interpretation has the policy demanding that facts be verifiable. It's a reasonably tough standard, and details are provided to help define what makes a fact verifiable. It also means that, if you can meet the standards of verifiability, you can say in an article "China is a country in Asia," rather than "so and so claims that China is a country in Asia, but this is disputed by so and so's uncle."

One might try to construe the policy another way, such as "if it's verifiable but NOT true, it can be included." But this catches on a little hitch. The word "verifiable" includes "truth!" Something untrue can not be verifiable by definition.

However, if we substitute the word "attributed," or "attributable," then the possible interpretation is loosened. It can easily be construed as "if it's attributed but NOT true, it can be included." Then an arguement can be made to open the Apollo 11 article with the sentence "The Apollo 11 moon mission was a complete fraud."[5]

I'm sure no one here wants to support such an interpretation. And whatever standards are in place, it's impossible to absolutely avoid distortion, untruth, and contested facts in all articles. But I think we should recognize that our standards of verifiability exist as a measure of truth, that they are in place to attempt to set some standard for what we regard as true, and that this is quite different from claiming that Wikipedia is not concerned with truth, only attribution.zadignose 00:43, 2 April 2007 (UTC)

Agreed.Dcmacnut 01:19, 2 April 2007 (UTC)

But an argument can be made for opening the Apollo 11 article with a statement that it was a fraud. (I'm not saying I'm making, or intend to make, that argument.) You cannot verify that it was true that it succeeded. You can only verify that someone says it was. It's a very good example of something that has been sufficiently difficult to verify for some that they are sceptical of its having happened. (Whereas events that are much less well attested are not contested. That's a curiosity of historiology, I suppose. See my note above about Englefield.) The question is not whether we intend to present the truth, but how we decide which versions of the "truth" we should promote and which we should demote. Zadignose seems to be labouring under the entirely unsubstantiable notion that there is a "truth" that we could be posting if only it weren't for all the editors trying to distort it. The best approach you could make is to suggest that we should post whatever we can agree is "true". Which is more or less what we do. We have votes, more often brawls, in which whoever can muster more people who think what they want included is "true" wins and prints their truth or at worst, can elevate their truth above other truths.

It is a problem, although possibly not our problem, that people have the same modernist view of an encyclopaedia as Zadignose, and believe that it should be in some sense "true". But we are aiming for "what people think is true", not "what is true". We post what historian X thinks is true; what newspaper Y claims is true; what writer Z once said was true.

As I noted, in some instances, it's easy to decide, although it's a problem that we're not explicit about our basis for the decision. In others though, our policies are doomed to be incoherent. You cannot on the one hand claim to be neutral -- in other words, not have a prejudice to one version of truth -- and on the other more heavily weight particular interpretations. This is a fundamental problem with our core policies and a source of a great deal of the conflict on this website: partly because people will game whatever policy you write to try to pin it down.

One of the positives about shifting the policy title to WP:Attribution is that it more accurately notes what we require: not that edits can be verified, but that they are checkable in the sense that you can go look that someone actually did claim what we say they did. It would have been doubly positive if it had been written with more of an eye to explaining the basis for it and other policies here, so that we could settle what we're aiming for and why. Grace Note 05:24, 2 April 2007 (UTC)

Mukadderat: In response to your axiomatic analysis: I agree with you that WP:NOR seems (to me too) to be independent of any particular definition of "truth". I think we don't need to spend much time arguing about different definitions of "truth". All we need to do is figure out what Wikipedians are supposed to do and what the wording of the policies should be; I don't think deep philosophical discussions about "truth" are needed for that.

IMO the idea of Mukadderat was to supplement the "IlikeItThisWay" arguments by some solid logic and his whole point was that yes, it is none our business to define "truth" but not because "we'all say so" but because it is simply unnecessary for the purposes and goals of wikipedia. `'mikka 19:23, 6 April 2007 (UTC)

Merzul: As I understand you, you like that Wikipedia is not claiming that what it says is true. I think I like that too. I think lots of Wikipedians like that. It's normal to state something, as is done in many Wikipedia articles. To state something and then add "... and it's true!" is redundant at best and probably hubris. I also wouldn't like a policy that said that Wikipedians have to prove or guarantee that what they add to articles is true. It just wouldn't be very workable. I think the way we do things now is just fine. My point is that saying nothing -- i.e. not saying that we guarantee truth or anything like that -- is just fine, but that saying "not whether it is true" (in the way the new WP:ATT wording says it) is going too far in my opinion. Do you think we really need those words "not whether it is true", and if so, why? What do you think of my compromise proposals and other proposals above?

Merzul, thank you for displaying my two compromise proposals in a box. Users should note that I also gave a list of other alternative wordings I would be satisfied with; search for "Possible alternative wordings" above.

Merzul, you said, "I don't mind any of these proposals, I only object to the Conservopedian "not solely if it is true"." Just to clarify: Which proposals are you indicating acceptance of? E.g. the list above "Possible alternative wordings", including "not merely whether it is true"? Or some more limited set of proposals on this page? The way this page has been refactored it's hard to tell, sorry. Anyway, your input may be very valuable in working towards consensus, because your position appears to me to be a sort of middle ground. I hope you'll continue participating in this discussion.

Filceolaire said, "If on the other hand there is a lie which has been widely published, by reliable sources, then that could be included in Wikipedia, along with it's refutation, if that has also been published. Attribution, not truth, is what is important if we want to avoid being sued." In the case of biographies of living persons, I disagree with you here. Suppose there's a widely published defamatory statement about a living person, and suppose Wikipedians have an unpublished but convincing source (e.g. a highly respected Wikipedian reports having received a letter from someone involved) which convinces Wikipedians that the statement is false. Wouldn't we be safer from being sued if we removed the statement from the article? And I think it would be better to remove it not only on grounds of being sued. It would just be more likely to be an accurate article if we remove things like that. Do you think leaving it in would make us safer from being sued?

Grace Note: you said "Coppertwig, I am saying that we cannot address objective truth because no such thing exists. I think you have entirely missed my point, which is that you cannot legislate against including "false" information if there is no way to measure what is true and what is false." OK, I think maybe I understand what you mean about objective truth not existing. For things in remote history, we really can't know what actually happened. We can only work with the records we have, imperfect as they are. I think I pretty much agree with you: I think we can't legislate effectively or enforceably against including false information if there is no way to measure what is true and what is false. I'm not proposing that we have a policy that legislates against including false information. I think maybe we already have a policy similar to that about biographies of living persons. I'm not suggesting any changes to those policies at this time. I'm suggesting deleting "not whether it is true" (or any of a number of other proposed changes in wording mentioned above, etc.) None of those proposed changes in wording seem to me to be legislating against including false information. Do you think they are? I'd like to just go back to saying nothing about whether the articles are true or not.

Grace Note, you also said, "I recognise in your reply that you are trying to suggest that the policy should emphasise that I need only be able to attribute my claim. Which I agree with. But you are claiming elsewhere that we should not be explicit that we do not make claims to truth. I think it is essential that we do not." If I understand this right, you think it's highly important that we not only avoid making claims to truth, but that we also say something in the policies to state clearly that we are not making claims to truth. How about something like this? "Wikipedia cannot guarantee that everything in its articles is true." or some similar wording. I would be OK with something like that. Or to Wikipedians, "If you've checked that something is stated in a reliable source, that's normally good enough; you don't have to go further than that and verify that it's actually true." There is a subtle distinction here with the "not whether it is true" wording, which raises its head when there is actual evidence of falsehood present rather than just lack of evidence of anything other than the reliable source. --Coppertwig 23:51, 2 April 2007 (UTC)

Forgetting phylosophical depths, the title of this section is an agreeable statement, as long as wikieditors don't forget that the statement "Some say 'Moon is made of green cheezz'" is also true albeit incomplete and unbalanced. It is also verifiable, but raises the issues of undue weight if included in some articles. `'mikka 19:27, 6 April 2007 (UTC)

In reply to mikka's comment further above: I agree that we don't need to define truth in order to go about the business of writing the encyclopedia. Individual Wikipedians may be motivated by a desire to communicate the truth, but this need not be discussed in policy. Your statements about the moon and green cheese are also quite acceptable and I think accepted by everybody in this discussion. --Coppertwig 01:54, 7 April 2007 (UTC)

Back To Basics: A Fundamental Question

Here is a basic question, the answer to which might help explain a certain point of view.

1. Why do we want information to be attributable/verifiable at all? Why not explicitly allow all statements, whether true or false, attributable or unattributable, original research or otherwise unsourced?

I suggested above how I would approach this question. But if we in fact are not interested in truth, then what other reason would we have to require attribution?zadignose 09:11, 3 April 2007 (UTC)

Colleague, there is no section #measuring_truth in this page. Mukadderat 21:54, 6 April 2007 (UTC)

The answer to all your question lies in your wrong presupposition "if we are in fact not interested in truth". We are interested in "truth", but the point is (see #Role of the truth 3 : "Sweeping Responsibility under the Rug" or "Axiomatic Introduction into Wikitruth") that wikipedia is encyclopedia, i.e., is is supposed to report notable knowledge, not create nor judge it. Old encyclopedias heavily relied on expertise of article authors. Wikipedia took a radical approach of not assigning a special weight to experts. WP:NPOV teaches us that various opinionns about the world must be reported, provided "proper balance" and "undue weight" issues are considered. In this form the issue of checking the "truth" is irrelevant to creation of wikipedia, however strange it may sound. The latter statement is not my opinion, it is a logical conclusion from certain fundamental, indisputable rules of wikimedia and wikipedia. If you disagree with it, I will be happy if you point out an error in my reasoning in the referred section. Mukadderat 21:54, 6 April 2007 (UTC)

Oh. You concluded above that 'WP:NOR is independent of any interpretations of the meaning of the term "truth"', which I can accept; but it does not follow from this that 'the issue of checking the "truth" is irrelevant to creation of wikipedia.' I think it's fine for the policies to say nothing about truth. However, a search for truth (accuracy in the sense of conformity to the real world) and a desire to present that truth to huge numbers of other people as a gift is highly relevant to the process of writing Wikipedia: it's the motivation, at least for most Wikipedians I think, for pouring hours and hours of volunteer time into the effort. The policies do not require checking the truth but do require checking for attribution; however, when Wikipedians go through the motions of checking for attribution, they are, in most cases at least, intentionally and actually checking for accuracy in the sense of conformity to the real world, at least at a good level of confidence. The policy does not explicitly require conformity to the real world, but it happens nevertheless. At least the policy doesn't disallow it!!!  :-) --Coppertwig 01:36, 7 April 2007 (UTC)

Good statement, in my opinion. Does this help solve our policy problem? Anybody? --Rednblu 04:49, 7 April 2007 (UTC)

When editors check for attribution, they are checking to make sure that the claim that some people (or group x of people) hold the view y is accurate. They are not checking to make sure that "the view y" itself is accurate. there is a big difference here, and it is this difference that is essential to our policy. Slrubenstein | Talk 13:29, 7 April 2007 (UTC)

Ok. But, for a high quality page like gravitation or truth, how do you select the "people (or group x of people)" whose "view y" is reported on the Wikipedia page? --Rednblu 18:19, 7 April 2007 (UTC)

I think we would start with people who have research gravitation. If people have researched it from diverse points of view (e.g. philosophers and physicists) we would make that distinction and present the distinct views. If there is a majority view and a minority view, we would represent those. With minority views we do need some criteria to keep out crank theories so we could limit it to people published in peer-reviewed journals. Slrubenstein | Talk 13:07, 9 April 2007 (UTC)

When I look at the HistoryFile of the GravitationPage, it seems to me that there are many, many fully Attributable published views of ReliableSecondarySources on "gravitation" that have been excised repeatedly from the gravitation page, and in my opinion appropriately. In particular, it seems to me that there is a lot of fully Attributable material on "gravitation" even published in the peer-reviewed journals of highest repute that does not belong on the gravitation page. So the standard for inclusion on the gravitation page is appropriately much, much higher than Attributability. --Rednblu 05:22, 10 April 2007 (UTC)

What are examples of the courses deleted? It is possible that the issue is multiple criteria for inclusions under ATT, rather than "higher" or "lower." That said, we do acknolwedge fringe theories which should either be excluded or given minimal space. Slrubenstein | Talk 11:18, 11 April 2007 (UTC)

Thank you for your sage comments. Your comments have given me many hints and therefrom many useful insights. But I would like to make this a placeholder, a platform for others at this time. I picked the gravity and truth pages for the cases-in-point here because, as far as I know, I have approached those pages only as a user, and not as an editor. So the editors of those pages would have the best data to answer our questions here. What do others say? --Rednblu 16:49, 11 April 2007 (UTC)

I think that you are doing a wonderful job of throwing mud at an excellent policy change. The reason why a lot of potentially relevant material is excluded from gravitation is that it would only clutter up the article if it was present. Gravitation is an general overview of the topic. You want it to provide a good, comprehensive overview of the history and major areas of study related to the topic. However, things like black holes are details of the major topic covered by that article. In fact, the old gravity article got split because it had gotten way too big, and "gravitation" was chosen as the title of the overview article on the topic.

This rejection of anciliary material from articles like gravitation has nothing to do with attribution, and it involves issues that this policy is not intended to cover. This policy cannot and will not cover all of the considerations as to what is or is not acceptable in any given article. Much of that issue is decided by the consensus of the editors rather than policy anyway. Indeed, what I am realizing is that this policy is irrelevant to mature articles on broad topics like gravitation and trurh because they have long ago surpassed the standards embodied here. I for one am happy to leave it at that.

It seems to me that your argument is analogous to claiming that a stop sign should not be placed at an interesection because stopping there is necessary but not sufficent to establish someone as a good driver. The goal of the stop sign is not to make people good drivers but instead to keep the not-so-good ones from hurting others. Similary, the purpose of this policy is not to make all articles good, but instead to keep the not-so-good ones from being a detriment to this encyclopedia. --EMS | Talk 18:36, 11 April 2007 (UTC)

That is a nice concise metaphor and model, congratulations for "stop sign," thank you. Probably Wikipedia policy is useful only for establishing some understandings that would minimize conflicts among editors. So it would be important to define and describe a "stop sign" so that everyone would recognize it as a "stop sign" just to minimize conflicts over what "stop sign" is. Just as one example that WP:ATT does not define and describe adequately what the "stop sign" is, WP:ATT in operation, by its poor redesign of previous policy, gives editors good arguments for reinserting baseless, false, and untrue statements about living persons--all constructed from fully attributable statements to ReliableSecondarySources. Hence, WP:ATT fails to define and describe an adequate "stop sign." So the conflicts between editors can be settled only by wonderful edit war heroics, gang warfare, or the edict of some just dictator. An adequate "stop sign" might require that that some secondary ReliableSource has actually published a Verification that the assertions in the statement are worthy of being considered fact--by the appropriate Verification standards of the relevant profession. --Rednblu 19:30, 11 April 2007 (UTC)

All that I see here is more mud. If the material really is inapproproate for an article on the person in question, then its insertion fails WP:NPOV, amd WP:ATT is irrelevant. However, there are times when a baseless allegation can be quite notable, especially if it had a significant impact on a person's life and/or carreer.

BTW - I would like you to name the article where these well documented but untrue alleagations are a problem. --EMS | Talk 20:26, 11 April 2007 (UTC)

Fair question, I would say. The problem here is not WP:NPOV. The problem is "Verifiability, not truth", which from the Wikipedia:Attribution/Poll is a very malleable phrase with at least two big splits in the Wikipedia community over what it means about truth. The problem manifests in its most obvious form on pages about living people, who read the page about themselves and are appalled at how the truth has been distorted. The general pattern is they first fight with the editors over their page. For example, Richard Dawkins was able to correct the fully Attributable untruths in his page. Others have not been able to correct fully Attributable untruths in the Wikipedia page on their life. Many of them complain to the just dictator, so the UserContributionsPage of the just dictator is a good source of links to pages with this problem where he had to intervene to remove the fully Attributable untruths that violated "Verifiability, not truth." I select a random problem page targeted in this edit by the just dictator where the editors accurately protested that everything removed was fully Attributable because "it was a simple reporting of facts and no conclusions were drawn." What was missing from that editors' process and imposed by the just dictator was the requirement that some secondary ReliableSource has actually published a Verification that the overall assertion was worthy of being considered fact--by the appropriate Verification standards of the relevant profession. You may have a different explanation for the problem. If you do, I would like to hear it, thank you. --Rednblu 22:15, 11 April 2007 (UTC)

I really don't see the issue here. Something that was somehow misrepresented got caught and fixed. The people who made the mistake were using what they thought was a reliable source. When the issue of its veracity was raised, people dug deeper and the mistake was resolved. Without WP:ATT I only see a lot more stuff like this, not less. You seem to be saying "Oh my God! This stop sign won't stop a drunk driver on a icy road. It is therefore ineffective." What matters is that which is stopped or tempered. This policy cannot stop honest mistakes from being made! It can and IMO does make it harder for those mistakes to last. --EMS | Talk 02:39, 12 April 2007 (UTC)

So, we are back to the question of the ReliableSource and the "issue of its veracity" are we not? Would you agree that a necessary condition for being a ReliableSource on the gravity page is that the ReliableSource must use the standard of "Verifiability, not truth" in assessing what is being asserted? --Rednblu 16:50, 12 April 2007 (UTC)

Not at all. That would require the editor to do a potentially infinite regression of verifications, and that is both impractical and improper. You establish a paper trail at the start, and if it gets challenged then you go deeper. So the editor's burden with a source is ensuring that it is reliable and relevant, not that it is verified in turn. --EMS | Talk 17:23, 12 April 2007 (UTC)

Hmm. Well, I must admit that "Verifiability, not truth" is a very malleable standard.  :)) But there seem to me to be very practical interpretations of "Verifiability, not truth" as well as the "infinite regression" possibility you highlight above. Verification after all has to be approached as a falsifiable process--not as an "infinite regression". So, for example, if Brian Greene's "The Elegant Universe" is written in a way that the author 1) presents his assertion as verified by his profession and there is a 2) Publisher's Weekly review that makes a statement like "Brian Greene, professor of physics and mathematics at Cornell and Columbia universities, makes the terribly complex theory of strings accessible to all," thus implying that Brian Greene is considered by his profession as having verified according to the standards of his profession what he wrote in "The Elegant Universe," then the Wikipedia editor has established a falsifiable claim that Brian Greene's "The Elegant Universe" is a ReliableSource for the assertion in the gravity page that "Many believe the complete theory [for gravitation] to be string theory." Where such a two prong standard for ReliableSource would be useful is in calming the edit war on global warming where the very weak WP:ATT standard encourages the insertion of all kinds of ridiculous "attributions to ReliableSources" that blatantly violate "Verifiability, not truth." And the only way the good guys can win under the very weak WP:ATT standard is to endure the My Goodness! extravagant and recurrent edit war there. May the good guys never give up their edit war until we can come up with some clear and self-consistent policy text to explain which ReliableSources should be used to establish what is on the global warming page! --Rednblu 20:33, 12 April 2007 (UTC)

(outdent) I would be careful about holding up global warming as an example here. There are plenty of attributable statements from reliable sources on both sides there, and there are some amazingly strong opinions on that issue. In a case like that WP:NPOV is going to be the first recourse in deciding overall content, as WP:ATT is only going to winnow out the most extreme and unlikely statements on both sides. Any POV warring is a different issue that this totally beyond the scope of this policy. --EMS | Talk 21:03, 12 April 2007 (UTC)

Well said, thanks for working me through this! I would make you co-author of my publication from this, but I would want your permission.  :)) And, following my line on your thoughts, as a user of the global warming page, I would like to know all the assertions of those researchers who have used the standard "Verifiability, not truth" in doing their own work. Thus, for the global warming page as for the gravitation page, I would say delete it if it is "fully attributable to a ReliableSource" who violates "Verifiability, not truth" in her own writing. --Rednblu 21:44, 12 April 2007 (UTC)

I really think that you are totally on the wrong track with extending the verifiability to the sources. If a new study is relevant and is being taken seriously in the media and the press, then that study deserves mention in the article as a matter of notability. The results of said study may be yet to be verified, and a totally new study will not be built on a paper trail of published works. If the study is notable and relevant, and we have reliable primary and secondary sources that document it, then it belongs and that is that. Instead, you would seek to use WP:ATT to suppress it, and this is an abuse of this policy. --EMS | Talk 03:45, 13 April 2007 (UTC)

I don't know. Here is the falsifiable hypothesis: The evolution page exhibits no edit war as long as the editors excise from the evolution page whatever assertion that is "fully attributable only to a ReliableSource" who violates "Verifiability, not truth." Who would want the evolution page to report any ridiculous new study being taken seriously in the media if the study is "fully attributable only to a ReliableSource" who violates "Verifiability, not truth"? Of course, the evolution page has the "Social and religious controversies" section, but everything reported there is Verified by ReliableSources who themselves use the professional standard of "Verifiability, not truth" in their own work. Would you agree? --Rednblu 05:33, 13 April 2007 (UTC)

I disagree for many reasons, most significantly that, despite your use of the language of "verifiability, not truth", your use of that term treats it as if it said "verifiability, IF truth". In other words, your standard requres that we be arbiters OF truth as part of the verifiability process. This is a comprehensive misundertanding of what verifiability means. In the words of Inigo Montoya, I don't think that word means what you think it means.

WHO would want the evolution page to report on a study that was being taken seriously in the media? Why, people who write encyclopedias would. Whether you or anyone who feels they have the "truth" of the matter thinks the page is "ridiculous" is not the issue. As was said at the beginning of this thread, "wikipedia is encyclopedia, i.e., is is supposed to report notable knowledge, not create nor judge it." Breadth of reporting IS the mark of notability for knowledge. Your assertion that we should only use sources which use OUR standard for inclusion in OUR pages is non-sensible. By that definition, we could not use news stories unless we were a news organization. And by your rules, taken only a slight bit more ad absurdium than you've already taken them, we would have very few articles indeed, if any. Jfarber 13:12, 13 April 2007 (UTC)

You have posed some wonderful lines of inquiry in my thinking. 1) What are the various hermeneutical usages in Wikipedia for each of the individual words "Verifiability," "not," and "truth"? Wow! Thus, you have brought to my attention that "not" in that phrase can mean $A\subseteq B$, $A\cap B$ generally, $A\cap B=\varnothing$, or something else as Wikipedia editors interpret and use the phrase "Verifiability, not truth." 2) On what pages should Wikipedia editors report the findings of the professional journalist's level of "Verifiability, not truth" and on what pages should they report the findings of the evolutionary biologist's level of "Verifiability, not truth"? 3) From what you wrote, I immediately jumped to see if I could find an example of the editors on the evolution page excising material that, in my judgment, would immediately be embraced by evolutionary biologists with whatever outcome if that material met the evolutionary biologists' standard of "Verifiability, not truth." I found one good example. But I'm not sure that is what you wanted for the direction from the, at least, three possibilities you made me think of for this discussion. I will be tied up here for a bit, but I will check in occasionally. --Rednblu 17:42, 13 April 2007 (UTC)

I'm not sure I understand your other two comments -- or I understand them, and think they have so twisted my meaning as to be utterly meaningless. As near as I can tell, you have decided that "not" can mean several things, which is just silly -- as truth CANNOT BE DEFINED by an encyclopedia, it is wholly moot to worry about HOW verifiability and truth can or cannot fit together; the issue should be entirely what do we mean by verifyability, and saying that it is not truth seems to be clear enough for all definitions of "not" without requiring that we worry about whether, if, and which definition of "not" we might want to worry about. But as for #2...your question makes NO sense. The same rules should apply for all pages. This is not an issue of which kind of attribution/information/value GOES where. And there is nothing at all new OR different from "the way wikipedia has always been" to say so. As for #3, I don't even understand your criteria for your "if" statement. Jfarber 18:33, 13 April 2007 (UTC)

WP is incapable of determining truth

Hi.

Well, it isn't. And some things are of disputed truth -- for example, the validity of certain religious claims. If Wikipedia were to be able to deermine their truth, then it would be famous around the world not as an encyclopedia, but as this incredible entity that managed to magically end centuries of debate! That's not the point of Wikipedia. And if we are to exclude things that are not true, then we could also not have articles on falsified scientific/pseudoscientific theories such as N-rays, polywater, phlogiston, geocentric model, etc. But in a comprehensive encyclopedia, I do not see the problem of including those things. mike4ty4 23:36, 6 April 2007 (UTC)

How about if we just don't make any claims about truth? How about if Wikipedia presents itself the way most history books, science books etc. present themselves -- they just say things? Most science textbooks don't say "When deciding what to include in this book, we didn't consider whether the things are true or not as a criterion." If they say nothing like that, nobody accuses them of claiming that "everything in this book is true" (which is probably false).

I think you mean that the policy must say "not whether it is true", otherwise we can't have articles about pseudoscience. I disagree -- I think the policy can just say nothing about truth, and we can have those articles; I don't see why not. Or even if the policy stated that articles must be true (and I'm not proposing that the policy should say this) then we could still have those articles, because the articles could say "J. Smith claims that ..." and it would be true that J. Smith made the claim, even if the claim is probably false. Anyway, for controversial topics, prose attributions are supposed to be given -- I think that's in WP:NPOV. --Coppertwig 01:46, 7 April 2007 (UTC)

"Verifiability, not truth" would trim to what is in the phlogiston page, it seems to me -- because what is in the phlogiston page is Verifiable by ReliableSources, including the judgment that "Eventually, quantitative experiments revealed problems, including the fact that some metals, such as magnesium, gained weight when they burned, even though they were supposed to have lost phlogiston." "Verifiability, not truth" seems to work quite well for the phlogiston page, would you agree? --Rednblu 04:44, 7 April 2007 (UTC)

I've been thinking about why people would passionately support including the words "not whether it is true" in the policy.

One reason could be to have an easy escape from having to participate in unending discussions of the truth of pseudoscientific claims. They can just point to the policy and say "I'm not interested in discussing whether that's true or not."

But there are other possible wordings that would give a similar escape, so I suspect there's a more fundamental reason for supporting the words "not whether it is true". Here's one possibility: Some of the supporters of this wording may be people who believe that the universe doesn't really exist. Such people tend to be misunderstood and ridiculed. People even think they're joking when they state their beliefs -- people can't even imagine someone actually believing that. So when these people see "not whether it is true" written into the policy of such an important institution as Wikipedia, they feel great relief as if they're finally understood or vindicated.

But then I thought maybe there was an even more fundamental reason for supporting that wording, because people don't really need to see their private beliefs about the universe expressed in Wikipedia policy, which is supposed to be more about how Wikipedians are to behave than about the philosophical basis of the universe. I'm guessing it's this: that in order to have mutual respect and consensus decision-making, it works well to show respect for others' beliefs. At the extreme, one can say things like "I believe that your opinion is equally as valid as my opinion," or "I believe that both our opinions are merely opinions and have nothing to do with objective truth." These behaviours can be associated with pleasant feelings of generosity and of working for the good of a group rather than for oneself. They can be freeing: they allow consensus-building to proceed. It's more than that. It's tied in with the belief that huge numbers of people can get along using consensus methods -- or that huge numbers of people must be able to get along using consensus methods or else the planet is in big trouble. I can imagine these sorts of thoughts leading to a passionate support for including the wording "not whether it is true". --Coppertwig 14:07, 8 April 2007 (UTC)

Well, I'm a strong believer in "not whether it is true" simply because I can't imagine any content dispute where Truth would be a helpful criteria. This is the first reason you suggest. The other reasons don't seem to apply to me. I believe there is an external physical reality and I subscribe to some naive version of a correspondence theory of truth, and when I'm in a content dispute I always believe my position is the ultimate truth and the opponents position is absolutely and ultimately false, except perhaps now :) Well anyway, even in real content disputes, I try to respect the other person's view, but I most certainly don't consider it "equally valid as my opinion".

For example, when I was in a content dispute over whether the ontological argument was rejected by Aquinas, I was absolutely, ultimately, and indisputably right, and the other editor was absolutely, ultimately and indisputably wrong. This is not a matter of interpretation, there is no scholarly debate on this issue, so I was trying to insert a factually true statement, but I was denied this because another editor had interpreted Aquinas himself, and thought Bertrand Russell and Graham Oppy (and many more) have no idea what they are talking about. I was ridiculed for wanting to include "cited untruths". Now, since what he called "cited untruth" was in fact The Ultimate Truth, I conclude that including such "cited untruths" can only be a Good Thing for Wikipedia. And that's basically all there is to my passion.

On the other hand, I have seen so much source-based POV-pushing recently that I'm no longer so fanatical about this. Also, having more fully understood the "truth"-supporter position, especially the discussion with User:Rednblu above and here, I've come to realize we all want the same thing, but as you can see below, I'm not the one who needs to be convinced. (And I'm supposed to be on a Wikibreak.) --Merzul 00:50, 9 April 2007 (UTC)

Thank you very much for this reply. I hope we can continue this discussion.

There's no use having the policy say anthing about absolute, ultimate truth, because in general people won't agree on what it is. The policy needs to deal with situations where all editors agree on what is true, and with situations where some editors don't know or don't care what is true, and particularly with situations where editors disagree about what is true. What the policy can work with (or can ignore) is editors' opinions about what is true, not what is actually true.

You can't think of a content dispute where truth would come into the question. I can think of some. (Truth in the sense of what the editors believe is true, not ultimate truth.) Here's one: Suppose one editor wants an article on God to say "God exists", and backs that up by a citation of Aquinas. Another editor can argue "Not everyone believes that God exists; for example, I don't. So let's insert a prose attribution." In this case, an alternative would be to find Reliable Sources stating that God does not exists, but there could be other examples where there is no such Reliable Source. For example, suppose there is only one source giving the number of soldiers at a battle and it says that an army had 7000 soldiers. One editor might want the article to say "The army had 7000 soldiers." But another editor could argue, "I don't think the army actually had exactly 7000 soldiers. It may have had approximately that number. Or the number may have been exaggerated by the source, which is likely to be biased. So, let's insert a prose attribution." I would side with the one wanting to insert the prose attribution.

Here's another content dispute: Suppose all the editors agree that some information is false. One editor may want to include the information, with prose attribution. Another editor may argue that the information is not important or interesting, because it is false. (Sometimes false information is important and interesting. Often it is not.) In this case, perhaps the information should be included or perhaps not, but I don't think it's a valid argument to say that whether it is true or not is irrelevant in determining whether it's interesting enough to include. Or, it may be in a biography of a living person; there, if the editors agree that the information is false, it should certainly or almost certainly be deleted, I suppose. WP:ATT should not contradict the policy on biographies of living people.

I think the most common compromise is: if one editor believes something is true and another believes it is false, it gets included but with a prose attribution. This is not the only possible outcome. The material could be excluded on a basis of undue weight, for example; or it could be included without prose attribution on the grounds that it's widely supported in the reliable sources and that there is no significant controversy about it in the reliable sources.

It's not only content disputes where truth comes in. Suppose there is only one editor and no dispute, but the editor has selfish reasons for inserting certain (attributable) information even though the editor believes it is false. I don't want the policy to support that sort of behaviour. Note that this is quite different from the "cited untruths" you were talking about, which are cases where some editors actually believe that the material is true.

Some editors believe that the opinions of editors as to what is true or not does sometimes properly influence the writing of articles, while others believe that such opinions should never influence the writing. A good compromise, then, is for the policy not to state that opinions of editors are relevant, and not to state that they are irrelevant, but to say nothing on that topic. Then we can have sentences with consensus support. Previously, (as you describe in your characterization of the opposition at Wikipedia talk:Attribution/Poll#Attempt to summarize-More serious opposition criticism-3.Truth), we had consensus for "verifiability, not truth" because the ambiguity in it allowed users with different opinions to support it.

To explore your position further: suppose there are no content disputes where truth is relevant. Why does it follow that the policy should say "not whether it is true"? What does it mean to you, and what other wordings would convey a similar message?

Note that my position is different from a position of requiring that editors verify the truth of what they write. --Coppertwig 13:24, 9 April 2007 (UTC)

Another example of a content dispute where truth (or what editors believe to be true) is relevant in my opinion is given in the 3rd paragraph of my initial post at Wikipedia talk:Attribution/Archive 12#Role of truth which started this thread. --Coppertwig 13:48, 9 April 2007 (UTC)

If there is a consensus that some item is no appropriate for inclusion in Wikiepdia (and this can be because it obviously is not true), then it can be omitted. Other times, an editor must defer to the common or known opinion. Item: I am an independent researcher. I do not believe that black hole exist. Should I be allowed to edit the black hole article to state that black holes definitely do not exist? I really think not. It is not a matter of whether I am right or not. It instead is a matter that it is not commonly known that I am right. Overall, I have found that this standard keeps a lot of trash out of this encyclopedia. In fact, I do not consider it to be fair to the reader for he or she to read that "black holes do not exist" if every scientist he or she talks to afterwards will say "of course the exist". In that case, I will have done the reader a major disservice, and if he or she should not use Wikipedia again I will have also done Wikipedia a disservice.

If you want an alternate wording, it is that the information in Wikipedia should be reliable: It should demonstratably reflect existing human knowledge. In fact, yuo have set a standard above wrt to "truth" that reflects this. You say that '... the articles could say "J. Smith claims that ..." and it would be true that J. Smith made the claim, even if the claim is probably false.' If you want to bring "truth" into this, that is the way to do it, but that kind of truth arises from the contents being attributable to a reliable sorce, hence the name of this policy. --EMS | Talk 15:35, 9 April 2007 (UTC)

That's openminded of you to realize that it wouldn't be good to change what Wikipedia says about black holes. That's the sort of openmindedness that's needed in order to make Wikipedia follow its NPOV policy. I figure that if a certain percentage of Wikipedians are able to set aside their personal beliefs and biases and aim for NPOV articles then, along with compromises between the editors with different extreme views, they can succeed in making the Wikipedia articles sufficiently NPOV. I think that's what usually happens and that's why Wikipedia works.

You said "If there is a consensus that some item is no appropriate for inclusion in Wikiepdia (and this can be because it obviously is not true), then it can be omitted. " Yes, exactly. I think that the editors on a page can discuss and argue about what to include and what to leave out and that they can use whether something is true, in their opinion, within those arguments, especially if they all agree about whether it's true or not. It seems to me that apparent truth is probably more relevant than any other factor in deciding what to include and what to leave out; it is of course not the only factor. The various factors are largely not spelled out in policy but are left up to the judgement of editors. Some users believe that whether something is true or not (or whether editors believe it is true or not) should never have any influence on decisions about what the article says. However, I don't believe there is broad consensus about this, and I think if people look at how the articles are actually written they will see truth and falsehood (or apparent truth or falsehood) being used quite often. The whole meaning of what is a "reliable" source for a given statement is based on perception of likely truth.

If I understand correctly, you're suggesting the possible alternate wording of changing the words " The threshold for inclusion in Wikipedia is whether material is attributable to a reliable published source, not whether it is true." in the policy to "The information in Wikipedia should be reliable: It should demonstratably reflect existing human knowledge. " This solves my objection to "not whether it is true," but I don't think it adequately expresses the need for published sources as opposed to just an editor's conviction that something is true. --Coppertwig 20:20, 9 April 2007 (UTC)

However, for example, with your God analogy, many editors are going to have differing opinions. A lot may say that God does exist, a lot may say that He does not. If we are to require "truth" as part of the bar for inclusion, then we would create endless debates that could never be settled. How do you expect to get a general agreement on, say, the God issue? You can't! Not with WP, that's for sure. They'll bicker endlessly about whether or not God exists. The God's existence debate has been going on for, most likely, millennia, and to expect Wikipedia to be able to solve it would be silly. If we require "truth" on these issues, then the God article would never be stable at all. It would constantly be flipping to and from "God exists" to "God does not exist" so fast it would make a humming sound! And because the opinions differ, it would suggest that someone else's opinion is not valid (if it says "God exists" then we could state in the article about atheism that atheism is a false belief. Similarly, if it says "God does not exist", we could state in all the aritces about the different concepts of God that they are "myths", etc.), and since we cannot determine ultimate truth, it is still going to be a perspective. That would raise questions of neutral point of view, another core Wikipedia principle. If we are to use our opinion that something is true to guide the editing, then we are creating a bias. Some other group may hold a differing opinion. The best option is to simply attribute each opinion, like in the God article, attribute the views like God exists/God does not exist to various prominent adherents, get statistics about their prevalence as well, again with attribution, etc. Truth as a criterion in terms of editor consensus is a way to create a bias. Truth as a criterion in terms of ultimate truth is useless since Wikipedia cannot determine that and everyone would be trying to get their view in as "truth". mike4ty4 00:30, 15 April 2007 (UTC)

Please comment on possible wordings of policy

We need to try to move towards consensus. Would people please comment on possible alternative wordings (or if you think only one wording is acceptable, explain in detail why)? If you search for "possible alternative wordings" above you can find a list of suggestions I made. There's also a box above with some compromise wordings. You can also suggest other wordings. Which wordings are or are not acceptable, and why? Thanks. --Coppertwig 02:01, 7 April 2007 (UTC)

I have lots of ideas here, but I am more interested in hearing ideas from other people. So I am listening. --Rednblu 04:34, 7 April 2007 (UTC)

You want a comment? Sure. As exemplified by this discussion alone, the concept of truth is inherently fuzzy; it is impossible to reach real consensus on what is "true" for all proposed items in all cases which might be included in any article. On the other hand, the concept of verifiability can be defined, enumerated, and demarked -- that is, the extent to which we can agree that something is or is not verifyiable, especially if we stick to including language which acknowledges what people said about a topic without trying to reach full consensus on whether that statement was "true" or not, is pretty clear. TRUTH is not an attainable standard for a 'pedia, especially one which uses consensus as its guidance system. Any linguistic compromise which tries to allow for TRUTH will end up suffering from the same problems, though more insidiously, as a policy which uses the word TRUTH. I see no consensus to explore change. Jfarber 14:43, 7 April 2007 (UTC)

"I see no consensus to explore change." Neither do I. The sentence I like is "The threshold for inclusion in Wikipedia is whether material is attributable to a reliable published source, not whether it is true."

A lot of wasted K here. I particularly appreciate that we have a sub-header called "Defining truth." When you get the answer, I have some old professors that would be curious. Marskell 15:30, 7 April 2007 (UTC)

Coppertwig, I do not think that this line is worth pursuing. The concept of "verifiability not truth" is one of the foundations of Wikipedia. ≈ jossi ≈ (talk) 15:44, 7 April 2007 (UTC)

Jossi, it is not. It is an ugly buzzword we are discussing to get rid off. Your attemts to shut down any discussions of your favorite policies without looking into merits of arguments are called "policing" and will be marked as such. `'mikka 17:28, 9 April 2007 (UTC)

Well, we could give up and just not have policies. Or we could give up on trying to establish consensus on policies and do it by voting or something. But if we're going to continue trying to have consensus policies, then I see no other option than to try to find a wording we can build consensus around. If you see another option, please describe it.

As I've said before, I'm satisfied with the longstanding wording, "verifiability, not truth". Here's a suggestion: the new wording can be "The threshold for inclusion in Wikipedia is verifability, not truth. Everything in Wikipedia must be attributable to a reliable published source." What do people think of this wording?

I'm still looking for anyone who supports the proposed new wording "attributable ... not whether it is true" who is willing to discuss back-and-forth with me until we can get to the stage where we can agree on a statement like "OK, you see the words as meaning this, while I see them as meaning that. And you want them to mean this, for this reason, while I want them to mean that, for that reason." There's been a lot of discussion here about the philosophy of the meaning of truth, but still a lack of discussion about the wording of the policy. I appreciate Grace Note taking the time to discuss related ideas with me, but I don't actually know which wordings Grace Note supports or opposes. --Coppertwig 13:19, 8 April 2007 (UTC)

In reply to Jfarber: When you say you see no consensus to explore change, do you mean we should keep the original words "verifiability, not truth"? Or do you mean that you think that "attributable ... not whether it is true" has already been established as policy? --Coppertwig 16:46, 8 April 2007 (UTC)

I think the second phrase you mention is not yet established as policy, and note that only a very few of the votes to keep ATT allowed that the language should be set in stone -- most folks I read in support either didn't mention it, or stated clearly that they expected a proposal to keep was also a proposal to give it the attention to language that it would need to be most effective, IF we decided we wanted such a policy. If anything, there were enough people on both sides of the pollvote interested in language issues to suggest (to me) that there is neither consensus nor majority desire to keep the CURRENT LANGUAGE of ATT, whether or not folks support of the basic premise of the ATT page.

I also think it is wordy and unnecessary to say "not whether it is true". If we must mention the word truth at all -- and it seems advisable, as a pre-emptive measure -- the clearest way to connect the word not and the word truth/true, in my opinion, seems to be the former. Mnemonics would better be served by slogans here, anyway -- since the goal is, in part, to make sure we can quote the shortform of policy as easily as possible, when it's needed. The phrase works; it's accurate and tight; IMHO, just because we're exploring the organizational structure of policy is no mandate to monkey with the effective language we use to "meme out" our policies throughout the 'pedia, and certainly not a mandate to monkey with the basic underpinnings of what counts, and what should count, as usable information for our purposes. Jfarber 18:22, 8 April 2007 (UTC)

Marskell: You've said that you like one particular sentence. Unfortunately, a number of editors have expressed objections to that sentence. In order to try to achieve consensus, we need to (a) discuss why you like it, and (b) consider other wordings. Would you be willing to discuss the wording you like with me long enough to have a good chance of arriving at an agreement like "OK, you see the words as meaning this, and I see them as meaning that. And you want them to mean this, for this reason, while I want them to mean that, for that reason."  ? What do you think of the other suggested wordings above? --Coppertwig 12:38, 9 April 2007 (UTC)

Orignial research

From User_talk:ScienceApologist#General_time_dilation:

Why do you think that [ General time dilation ] shouldn't be in Wikipedia if this may actually be the reason why we see the Hubble redshift as confirmed by Supernova Project? ... [I]f GR is so successful, why not let the readers of Wikipedia to know that some of Wikipedia pages tell a true story even if it is not the exact purpose of encyclopedia in your opinion? Why do you think that suppressing truth is even better? [bolding mine]

This type of thing is the reason why the poilicy calls for verifiability instead of truth. Simply put all kinds of people are happy to come out of the woodwork claiming "truth". I will admit that there is a "truth" out there in terms of the real world as it really is. But how does one determine what this is?

In this case, the author of the above quote has his own view about general relativity which is very much at odds with accepted scientific opinion and therefore also the human knowledge that Wikipedia has chosen to document. Suppose for a moment that it is "true". Given that it is not a part of the knowledge base that Wikipedia is to document, so what? If its being true is not a part of human knowledge, then it doesn't belong here. How then do we know what does belong here? Though verifiability and more importantly good attribution.

Suppose that we did want to express the truth. Then how would we determine that this person's view is true instead of merely mistaken? The answer would be through a research program of some kind. However, that is an expensive, time consuming process which neither the Foundation nor the volunteers here can affort to do. Also, if it came out against this person's opinion, he would not be satisfied.

So we come back to verifiability and attribution. Ironically, I find that to be the best way of getting at the truth of the situation. I won't claim that it can deliver the truth in terms of what is. However, it can deliver the truth about what is known about a topic and what disputes there are relating to the topic. Indeed, the truth about the state of human knowledge is what matters here. In general, that will also be the truth about that which is, but it need not be. After all, whenever someone doing original research makes a seminal discovery, our understanding about what the underlying truth most likely is changes. --EMS | Talk 00:19, 8 April 2007 (UTC)

Nice analysis, in my opinion. I particularly found useful your summary that "the truth about the state of human knowledge is what matters here." Would you agree that "Verifiability, not truth" would work for us as a standard to get at "the truth about the state of human knowledge" and, at the same time, avoid Original Research? --Rednblu 01:11, 8 April 2007 (UTC)

Yes, I would. I actually see it as a better standard than WP:NOR itself, since to be verifiable there needs to be some writings on the subject that are independent of the creator of the ideas. WP:NOR itself only says that you need to achieve publication in a peer reviewed journal, but in reality there are highly regarded journals that do not use peer review to aid in the selection of articles, and other journals that not held in high esteem but which do use peer review. The latter have provided occasional headaches on the relativity pages. However, most topics written up in a respected peer reviewed journal are unlikely to merit immediate inclusion in Wikipedia. Those that do are accompanied by articles in other journals that act as secondary sources on the issue. --EMS | Talk 03:16, 8 April 2007 (UTC)

Would you say that WP:ATT weakens the standard expressed in the statement "to be verifiable there need to be some writings on the subject that are independent of the creator of the ideas"? --Rednblu 06:17, 8 April 2007 (UTC)

From Wikipedia:Attribution#Primary_and_secondary_sources:

Edits that rely on primary sources should only make descriptive claims that can be checked by anyone without specialist knowledge. and

Wikipedia articles should rely on reliable, published secondary sources wherever possible.

That really is the correct way of expressing this IMO. You cannot have hard and fast rules about this stuff, but you can set the basic standards up in a way that is more rigorous. These statements are reasonably rigorous. They still allow an awful lot in, but the stuff that cannot pass these tests does not belong here. Overall, this is a more rigorous test than WP:NOR, and that is why I approve of it. --EMS | Talk 16:55, 8 April 2007 (UTC)

So would you say that the statement "should rely on reliable, published secondary sources wherever possible" adequately captures the standard that the editors have applied in maintaining the high quality of the gravitation and truth pages? --Rednblu 18:28, 8 April 2007 (UTC)

I can't speak about the truth article since I don't edit it, have not read it, and am not interesting in evaluating it (and that is the truth :-) ). As for gravitation, I will say that to the extent that this policy's mandates have been applied there that it is a good article, and to the extent where it has been less than faithfully applied it can be improved. (For example the GR section lacks citations but is verifiable through secondary sources. Even so, it would be better to have the citations.) Let's just say that being a high quality article takes more then adherence to the Attribution standards. Even so, a good article cannot be created without adhering to this policy. --EMS | Talk 14:46, 9 April 2007 (UTC)

Would you agree that Attribution standards are necessary but not sufficient for a high quality gravitation page? --Rednblu 15:16, 9 April 2007 (UTC)

Yes. --EMS | Talk 18:27, 9 April 2007 (UTC)

Such a thing is easier and more feasible for WP to check than the objective or scientific truth of a claim itself. mike4ty4 23:24, 16 April 2007 (UTC)

That's how I see the attribution standards: necessary but not sufficient. In other words, material must be attributable to be included, but there are also many other factors that can influence editors in whether to include material or not. Some of these factors are spelled out in other policy or guideline pages and some are up to the good sense of the editors. --Coppertwig 20:23, 9 April 2007 (UTC)]

Rednblu, I know I have often been curt with you but please read what follows carefully because I am really trying to address your concerns as seriously as I can. No one stated that NOR or ATT is sufficient to guarantee high quality articles. I would argue and I think most would agree with me that no one policy is sufficient to guarantee quality articles, and that not even all policies together would be sufficient to guarantee quality articles. I think that the original intention when jimbo and Larry Sanger created Wikipedia was to consider the wiki nature of the project as the principal element that would produce quality articles: an ever-increasing number of people contributing to an encyclopedia, where the good judgement and knowledge of different people would over time eliminate crap. A pure wiki would have no other regulation than the self-regulating dynamics of an anonynmous group of people, limitless in number, who can add, change, and delete anything. I think that Jimbo and Larry recognized that this anarchic dynamic actually needed some - but limited - regulation, which led first to NPOV (essential to turn a heterogeneous and anonymous group of people into something like a community), where V was implicit in NPOV, and then an explicit V and NOR policies, as well as civility policies. But historically, these policies have served to help people with different views get along with one another, and to help limit abuses - not to guarantee the production of excellent articles. It is the wiki nature of the project itself that bears that burden, harnessed to the explicit purpose of this wiki community, which is to produce an "encyclopedis" (see the first of the five pillars). Rednblu, I would like to take you at good faith and not question your motives to ensure Wikipedia is not only a good but a great encyclopedia. I respect your motive and share it. However, whereas some encyclopedias aspire to achieve this goal by hiring PhD.s to constitute an editorial board, who then hire PhD.s and graduate students to write articles (i.e. rely on a certain notion of expertise), Wikipedia aspires to achieve this goal through a faith in large numbers of people acting as freely as possible and in good faith. It took a loooong time to create ArbCom i.e. to acknowledge not only that some people do not act in good faith but that the Wiki dynamic (a huge number of people just deleting the crap) alone was insufficient to counter people acting in very very very bad faith and we thus needed something more. It also took a looong time to develop additional policies like NOR meant to guide and support editors acting in good faith who like you want to help create a great encyclopedia i.e. to acknowledge that the wikidynamic alone is not sufficient. I am being sincere, Rednblu, historically the question was not whether any policy is sifficent to produce a high-quality encyclopedia, but rather, can a wiki-community (wiki meaning practically anarchic) be sufficent to produce a high-quality encyclopedia. Over time we have discovered it is not entirely sufficient. Nevertheless, it is the wiki nature of the project that distinguishes it and that we all count on to produce a great encyclopedia. Anything else we add - a committee, a policy, a process - we do so very very cautiously and only when we are convinced that the wiki-dynamic is insufficient. Since I contributed to some of our policies I am obviously among many people who think that the wiki nature of the project is insufficient - yet, I nevertheless consider the wikinature of the project to be its core and distinguishing principle and I think we really have to be as exceedingly cautious in adding to it (e.g. policies) as possible. rednblu, with all due respect I think the only, the one, missing element you are searching for is simply a statement that "We are writing an encyclopedia" ... "We are a wiki community, but we are a wikicommunity specifically dedicated to writing an encyclopedia" ... "We are a wiki community, and we are an encyclopedia." It is the idea of an encylopedia that drives us towards quality. I think if you look at the five pillars you will see how all of them follow naturally from conjoining a wiki-community to an encyclopedia. And I say this with all due respect: if you want higher standards, then you ought to get a job for Encyclopedia britaninica or Nupdia. i am not being sarcastic, I have a lot of respect for EB and am glad it exists and don't think it should cease to exist or that people should cease to work on it. But Wikipedia is motivated by a very different ethos, one that is based not on expertise or policies but on Wiki. It is a grand experiment and like any it may fail. If it fails it is not a major loss to the world of knowledge as EB and other encyclopedias will endure. But if it succeeds, it will succeed as a wiki where quality is ensured not by policies but by the lack of them, by freedom, by millions of people working bit by bit over many years, producing a living thing that at any given time has different strengths and weaknesses. Slrubenstein | Talk 11:51, 11 April 2007 (UTC)

Historical question

Some people above are tracing the "verifiability, not truth" phrase to Sanger. Is this correct? I thought it only went back a couple of years. --Abu-Fool Danyal ibn Amir al-Makhiri 20:54, 10 April 2007 (UTC)

Someone may have more to add. I would consider most important how the concept of "Verifiability, not truth" actually has been used in the HistoryFiles of the Verifiability, gravity, and truth TalkPages. Of course, you have to search on "verif" because it has many forms, does it not? Here are the Early gravity TalkPage and Early truth TalkPage in which you might search on both "Sanger" and "verif", Here is Larry's essay that would become the first draft of the "truth" page in the 2002 "automated conversion" to modern day Wikipedia. The last three sentences of Larry's first draft of the "truth" page read as follows.

"But as usual you have been introduced to the terrain: among different theories of truth there are the correspondence theory, the redundancy theory, the coherence theory, and pragmatism. With such a variety to choose from at the very least you should be convinced that you don’t have to rest content with any sort of relativism that says that truth is just the same as belief. You can do a heck of a lot better than that." (Larry Sanger, sometime before March 2001 in writing the essay that comprised the first draft of the truth page.)

"Verifiability, not truth" is a careful assertion of the professional variety of truth that matters in high quality Wikipedia pages such as gravity and truth, I would say. --Rednblu 01:14, 11 April 2007 (UTC)

The word "verifiability" was not in the original policy, but the idea was implicit: According to the original core policy, NPOV, what is "objective" is not gravity, but that some people hold certain beliefs about gravity. And it is these beliefs, completely regardless of truth or truthfulness, that are represented in Wikipedia. Here are some quotes from the first version of the policy, from 2002:

• Perhaps the easiest way to make your writing more encyclopedic, is to write about what people believe, rather than what is so. If this strikes you as somehow subjectivist or collectivist or imperialist, then ask me about it, because I think that you are just mistaken. What people believe is a matter of objective fact, and we can present that quite easily from the neutral point of view. --Jimbo Wales
• But even on such pages, though the content of a view is spelled out possibly in great detail, we still make sure that the view is not represented as the truth.

These ideas provided the basis for another core policy that was developed several years ago, namely, or verifiability policy the essence of which is "verifiability, not truth." However, the most detailed discussion of why wikipedia is not about truth or objectivity (except as discussed above, by Jimbo) will be found in older versions of the NPOV policy (look at almost any from 2003 or 2004) - note, these older versions of the NPOV policy say that we should "attribte" views to others rather than present views as facts and rather than presenting our own views. In 2003, Jimbo made a comment on the listserve that became our first NOR policy: From a mailing list post by Jimbo Wales:

If your viewpoint is in the majority, then it should be easy to substantiate it with reference to commonly accepted reference texts.

If your viewpoint is held by a significant scientific minority, then it should be easy to name prominent adherents, and the article should certainly address the controversy without taking sides.

If your viewpoint is held by an extremely small minority, then whether it's true or not, whether you can prove it or not, it doesn't belong in Wikipedia, except perhaps in some ancilliary article. Wikipedia is not the place for original research.

The NOR policy was created December 2003 and was developed into its more or less full form by the middle of 2005. "Verifiability not truth" was added to the Verifiability policy by User:Uncle G on August 25, 2005. Slrubenstein | Talk 14:23, 31 March 2007 (UTC)

Thanks for the info, especially the hint about finding "attribute". More than the general principles, I was also wondering about the specific phrases. After all, it seems the subtleties of wording are partially at issue. As for "verifiability, not truth", I can trace it further back into NOR, which goes back to a draft rewrite at the start of 2005 (around the time we started putting the ticky-boxes on things). I lose the thread at this point. Is that as far back as it goes? --Abu-Fool Danyal ibn Amir al-Makhiri 16:39, 11 April 2007 (UTC)

I think you are right. My recollection is that the NPOV policy always had the claim that we are not dealing with truth, and the claim that we need to show that the views represented are actually held by others (not editors). The explicit link between these two points was I think first made when we developed the Verifiability and NOR policies, in part to show how they fit in with (and in no way contradicted) NPOV. Slrubenstein | Talk 10:42, 12 April 2007 (UTC)

Scientific controversies

-- Copied from the Talk:Evolution page [6]

I Noticed that there was not a section for scientific controversies. Even though it is taught in most schools that there is no opposition to this theory in the scientific community I found some articles even here on wikipedia about controversies surrounding evolution. I created this section to cover those topics and tried to develop a small paragraph addressing theses arguments from a neutral perspective.B89smith 18:31, 14 April 2007 (UTC)

The fact that Evolution -- like any science -- is a field where hypotheses get tested and supported (or rejected) and concepts and ideas sometimes change, does not imply that there is scientific "controversy" as to whether Evolution is true (as we define anything to be "true"). There is no controversy among scientists that all life evolved from shared ancestry, via mechanisms of selection and population dynamics as outlined in part by Darwin, Wallace, Fisher, and the Modern Synthesis. Like any science, controversy sometimes exists on smaller points, such as the actual structure of a phylogenetic tree, or whether or not sympatric speciation can occur in sexually reproducing organisms. But those controversies do not mean that phylogenetics is under scrutiny, or speciation does not occur. Normal scientific debate about specific topics does not mean that the core structure and tenets of the science are under any debate whatsoever. TxMCJ 18:56, 14 April 2007 (UTC)

This issue is addressed in the Talk:evolution/FAQ. You may be interested in Objections to evolution. Cheers, Gnixon 18:34, 14 April 2007 (UTC)

I still think it should be included on this wiki since social controversies is a section. It would be inconsistent to not include scientific controversies. As soon as we remove this section from the wiki we remove the objectivity of this article and replace it with a subjective opinion on evolution. This greatly lowers wikipedia's credablity down from a encyclopdia to a chat forum discussing what wikipedia members think about evolution.B89smith 18:45, 14 April 2007 (UTC)

As everyone here will attest, I am very interested in the problem you are having inserting this material that is fully attributed to many Reliable Sources.  :)) Would you agree to the principle that the evolution page should contain only material that satisfies the standard of "Verifiability, not truth"? --Rednblu 19:16, 14 April 2007 (UTC)

You may want to propose revising the FAQ. Gnixon 19:19, 14 April 2007 (UTC)

Rednblu - This is getting silly. You are acting like someone who is shaprenning their sword for an edit war. IMO, you can ask that the material used in an article like global warming be of the highest relevance and reliability. You can even ask that a study not be mentioned unless it is referenced by multiple reliable sources, which themselves must be listed as references in the article. However, the study itself is legitmate primary source material, and primary materials will NEVER meet your standard. Please leave off of this. --EMS | Talk 06:37, 15 April 2007 (UTC)

Really? You may be right, but let us examine the empirical data to see if what you say is "Verifiable, not true." So let us examine a random "primary source material." Let the next string of numbers be drawn from the lottery basket, and we have Annalen der Physik, 1905. 17 (10): p. 891-921. And let's just take a random assertion from the first page of that particular "primary source material"--and we have the following.

• "Beispiele ähnlicher Art, sowie die mißlungenen Versuche, eine Bewegung der Erde relativ zum 'Lichtmedium' zu konstatieren, führen zu der Vermutung, daß dem Begriffe der absoluten Ruhe nicht nur in der Mechanik, sondern auch in der Elektrodynamik keine Eigenschaften der Erscheinungen entsprechen, sondern daß vielmehr für alle Koordinatensysteme, für welche die mechanischen Gleichungen gelten, auch die gleichen elektrodynamischen und optischen Gesetze gelten, wie dies für die Größen erster Ordnung bereits erwiesen ist."

Now obviously this rather original assertion in this "primary source material" challenges any competing ideas about Special relativity, and probably force, mass, and acceleration, so the question you pose is whether this "primary source material" meets the Wikipedia standard of "Verifiability, not truth" in regard to that assertion. I would say that operationally it does--because the assertion in that ReliableSource is "Verifiable, not true" by the following two prong test. First, that assertion is "Verified, not true" in Halliday, Resnick & Walker, Fundamentals of Physics, Seventh Edition, p. 1023 as follows.

• The laws of motion are the same for observers in all inertial frames. No one frame is preferred over any other.

Second, the Halliday & Resnick "Fundamentals of Physics" textbook abides by the "Verifiable, not true" standard of physics because it is in its seventh edition and is listed on the Wikipedia list of physics textbooks that are considered by physicists to abide by the "Verifiable, not true" standard for the assertions made in them. Thus in general, it would seem that the assertion A in any "primary source material" would be considered as "Verifiable, not true" if that assertion A has been "Verified, not true" by the writings of most of the scholars in the profession. Notice that No original research by the Wikipedia editor is involved because the Wikipedia editor merely represents faithfully what the ReliableSources who use the "Verifiable, not true" standard of physics have published about the assertion in the "primary source material." --Rednblu 02:30, 16 April 2007 (UTC)

IMO, you have sought to refute me and ended up sayng the same thing: The secondary sources back up and set the notability of the primary sources. Ideally under WP:ATT, all attribution should be like that. Use the existing policy and seek that material that most strongly meets those standards, and you will lead the articles you edit to a good place. However, if you try to go beyond where WP:ATT is intended to take you, then you will end up with endless useless battles on your hands, amd a lot of editors who are unhappy with you. --EMS | Talk 06:09, 16 April 2007 (UTC)

Good warning! We shall see. --Rednblu 06:13, 16 April 2007 (UTC)

Great Job!

It appears you've all worked very hard to define Wikipedia's policy of "verifiability". Bravo! I was just wondering what a "reliable source" was when I happened across this:

• "Reliable sources are credible published materials with a reliable publication process; their authors are generally regarded as trustworthy or authoritative in relation to the subject at hand. How reliable a source is depends on context. In general, the most reliable sources are books and journals published by universities; mainstream newspapers; and university level textbooks, magazines and journals that are published by known publishing houses."

Then I just had to ask what's a "reliable publication process"? Who decides whether an author is "generally regarded as trustworthy or authoritative in relation to the subject at hand"? Who decides the reliability of a source in relation to its "context"?

I was also wondering about the quote I keep seeing on this page about attributability:

• "reflect truth for some notable category of people"

I was just wondering what this actually means? What's a "notable category of people", and who decides whether a category of people is notable or not? And how do you decide whether or not something reflects "truth" for that category of people? Is that done by a scientific survey or something?

I didn't find any links on these items. Maybe someone here can enlighten me? Thanks. SqlPac 17:29, 9 June 2007 (UTC)

Truth requires non-contradiction

If a source can be found which contradicts another source, it is not a matter of:

• What is said in most sources.
• Which source is closer to the subject.
• Whether one source is more emotional than another.
• Which source was closer to the time when the subject existed.

What matters is that the article doesn't contradict itself, or that does not contradict what has been established in other articles. Wikipedia should be coherent. Also, the ability to attribute something to a source does not make it true, and just because something is true does not make it attributable. Articles should be able to examine the facts according to an understanding. But one understanding should not contradict another understanding in other articles. Truth implies non-contradiction, but non-contradiction does not imply truth, but understanding. Truth is not for truth's sake, but it is to achieve understanding of the material under what has already be lit under the light of others.◙◙◙ I M Kmarinas86 U O 2¢ ◙◙◙ 23:49, 12 June 2007 (UTC)

Hmmm. So if 100 articles on Wikipedia stated that "1 + 1 = 3", and one article stated that "1 + 1 = 2", then "1 + 1 = 3" is the truth as defined by Wikipedia because truth is created by consensus. Does that accurately summarize this? Thanks. SqlPac 01:45, 13 June 2007 (UTC)

Not at all. What is said in most articles (consensus) has nothing to do with it. In fact, what you showed me is a contradiction. It would be fine to have these phrases in a different context, like if these 100 articles said something like "so-and-so said '1 + 1 = 3'" with sources to show that it is not original research. Being openly truthful about a subject basically means being honest about the material which is about the subject, and only its material. This is in line with WP:V and WP:A.◙◙◙ I M Kmarinas86 U O 2¢ ◙◙◙ 02:18, 13 June 2007 (UTC)

I don't see anything "contradictory" about my example. Consider if 100 articles said "1 + 1 = 3" and sourced it, but the Addition article says "1 + 1 = 2" without a source (as it stands right now). Why would 100 articles need to qualify their statements with "so-and-so said"? Sounds to me like the Addition article would need to provide a source and qualify their statement with "so-and-so said" in light of the overwhelming 100 articles with sources contradicting it. So you're saying that truth has nothing to do with the factual accuracy of an article, but rather with qualifying statements with "so-and-so said..."? If that is the case, then there are literally thousands of articles that need hundreds of thousands of their sentences qualified with "so-and-so said..." E.g., "so-and-so said 1 + 1 = 2. so-and-so said 2 + 2 = 4. etc." (from the Addition article). SqlPac 16:03, 13 June 2007 (UTC)

While I'm thinking about it, let me give you a "real-world" example. In the article SQL, there has been a debate for a while about whether "SQL" is currently used as an acronym for "Stuctured Query Language" or if it no longer is considered an acronym, but rather a proper name. The former ANSI and ISO SQL standards stated that "SQL" was an acronym. I can pull up 100's of college text books, research papers, database product documentation, technical books, and technical journals that expand "SQL" to the words "Structured Query Language" explicitly. I can also pull up a lot of information by experts and academicians (sp?) who are highly regarded in the field, and who state that "SQL" no longer is an acronym. The ANSI and ISO standards themselves no longer expand the acronym anywhere in their text, although they do not specifically state anywhere that SQL is *not* an acronym. So what's the truth in this situation? So we need to state that "so-and-so says that SQL is an acronym, but other people like so-and-so say it is not?" Is that the non-contradictory definition of truth you're aiming for here? Thanks. SqlPac 16:31, 13 June 2007 (UTC)

"So what's the truth in this situation?" You just said it to me.◙◙◙ I M Kmarinas86 U O 2¢ ◙◙◙ 04:05, 14 June 2007 (UTC)

So you're saying the truth is found by keeping the facts unnecessarily cryptic. Or maybe you're saying that the truth is found only by qualifying every statement of fact in every single article with "so-and-so said?" As for my example, which is the truth? Based on what I told you (which are all easily verifiable facts), is SQL: (a) an acronym or (b) a former acronym that is now a proper name "with no official expansion"? Or (c) is the truth here only to be found by weaseling around the issue in the first sentence of the article? No need for a cryptic answer. A simple (a), (b), or (c) will do just fine. SqlPac 04:19, 14 June 2007 (UTC)

(b)◙◙◙ I M Kmarinas86 U O 2¢ ◙◙◙ 04:45, 14 June 2007 (UTC)

That's a surprising answer, considering there are hundreds of respected textbooks, research papers, vendor documents, and other sources including direct quotes from notable subject matter experts (some members of the committees that write the standards in question) in the field that disagree with (b). So the proper answer is to ignore hundreds (if not thousands) of reliable sources completely and go with the hundreds of other reliable sources who say the opposite. And this is the essence of truth? SqlPac 04:58, 14 June 2007 (UTC)

I am find this to be a very odd thread. Wikipedia is not a provider of "truth", and so non-contradiction is not required. OTOH, I agree with teh statement that "Wikipedia should be coherent". I for one see nothing wrong with presenting contradictory viewpoints on a topic. In fact, Wikipedia by its own rules is expected to do just that when multiple notable viewpoints exist. The coherence comes from saying that "group A call for X, while group B calls for Y", and explaining just what X and Y are and even why they are incompatible.

The idea that Wikipedia should present "truth" means that there is only one valid POV on a topic. Very often that is the case, but for Wikipedia's purposes the "truth" to be documented is that said POV is the overwhelming supported one and not that the POV is true. For cases where an ongoing controversy exists, once again we want to write about the topic and the controversy surrounding it, presenting as accurately as possible the views of the competing groups, and without outright endorsing either POV. In such cases, there will be contradictory statements in the articles, but that is a reflection of the controversy and not an indictment of Wikipedia. So our goal is to write "truthfully" about the human understanding of a topic, and not to find/declare the truth about the topic. --EMS | Talk 14:52, 14 June 2007 (UTC)

Something like whether SQL is an acronym is a bad example because whether it is an acronym depends on usage by people. It's quite possible for some people to use it one way and some to use it another; there's no definitive truth about what it means outside of the people who use it.

Consider the covered bridge example, though. The reliable sources say the bridge is closed to traffic. A guy visited the bridge and it has traffic, but he hasn't published his observation anywhere. It would be stupid to write that the bridge is closed to traffic just because all the reliable sources say it is, because that's false. Ken Arromdee 04:37, 15 June 2007 (UTC)

That's a bad example because it is Original Research, plain and simple.

Whether SQL *is* an acronym is not the central issue here. It might be, it might not be. Your response about some people using it one way and some using it another is a straw man as far as Wikipedia policy is concerned. Consider: taking a poll of people to find out how they use "SQL" constitutes Original Research. AFAIK this is prohibited by Wikipedia policy. The vast majority of people who use SQL are not notable or attributable per Wikipedia policy. So what does some people possibly using SQL one way, and other people using it another way have to do with Wikipedia policy and Wikipedian truth? Whether it's "true" or not doesn't really matter on Wikipedia, does it? It's all about what you can source from the "best" sources, whether their contributions to the body of human knowledge are "true" or not. SqlPac 04:50, 15 June 2007 (UTC)

It doesn't violate any Wikipedia rules to do original research to determine whether to include something in an article, as long as the original research itself is not in the article. (We do it all the time when, for instance, doing a Google test).

Let me make sure I understand your logic here, using your bridge example:

1. Several attributable sources publish that the bridge is not open to traffic.
2. Intrepid Writer #1 visits the bridge himself, and decides that the bridge is open to traffic.
3. Because original research is not allowed, our writer cannot write in the article that the bridge is open to traffic.
4. Because he has performed his original research, however, our writer will now specifically not say anything about bridge traffic in the article.
5. Enter Writer #2, who lives 2,000 miles away from the bridge and cannot visit it personally. He updates the article adding that the bridge is not open to traffic, citing the attributable sources that Writer #1 chose to ignore.
6. Writer #1 apparently can remove the sourced statements based on his own original research.
7. Writer #2 comes back and demands his properly sourced material be put back in.

Now you are asked to mediate this discussion. You don't live near the bridge and cannot personally confirm Writer #1's assertions, and you have several attributable sources that tell you the opposite of what he is claiming. What's your decision? Include the statement based on attributable sources or exclude it based on this one guy's Original Research? SqlPac 15:41, 15 June 2007 (UTC)

In this case, we can't take a poll and write in the article "87% of people polled use SQL as an acronym". But what we *could* do is remove all statements which say "SQL isn't an acronym". Putting in a conclusion produced by original research is prohibited. *Excluding* something on the basis of original research is not prohibited.

So you say to include some items in an article that can be sourced properly, and arbitrarily exclude other contradictory statements that can also be sourced properly? *Excluding* something on the basis of original research is a weasel-move, and will only work up to the point that someone else who hasn't performed the exact same original research decides to add the excluded item back in, properly sourced. IMHO, the only difference between *including* something based on original research and *excluding* something on the basis or original research is that *excluding* might take people a little longer to discover. SqlPac 15:41, 15 June 2007 (UTC)

Likewise, for the covered bridge example (and ignoring WP:IAR, for which it's a classic case), we couldn't write in the article "this bridge has traffic", since that is original research. However, we could delete statements that say the bridge has no traffic.

See above. This works until someone else who locates the attributable sources and adds the properly sourced statements that "the bridge has no traffic" back in. Then everyone gets to decide whether to keep something in that is properly sourced, or delete it based on one guy's original research. SqlPac 15:41, 15 June 2007 (UTC)

And that's basically what the role of truth question is. Do we want to include statements that are in reliable sources and attributable but are false? Ken Arromdee 15:05, 15 June 2007 (UTC)

Even by deleting a single statement that contradicts the results of one guy's original research, that one guy is adding his viewpoint to the article. A viewpoint which "is held by an extremely small minority". And, that viewpoint, "whether it's true or not, whether you can prove it or not, doesn't belong in Wikipedia, except perhaps in some ancillary article." After all, Wikipedia "is not the palce for Original Research." SqlPac 15:41, 15 June 2007 (UTC)

By that reasoning you could never say "I did a Google search, this isn't notable enough to include". After all, the Google search is only the opinion of the one guy who did the Google search. In fact, you couldn't even take original research out of an article, since your decision that a passage is original research is *itself* original research. You're "adding the viewpoint" that a passage is original research when you take it out.

What Google search are you talking about exactly? Quoting an attributable source is not the same as expressing the opinion of the guy who added the properly sourced quote, is it? SqlPac 03:44, 16 June 2007 (UTC)

People often edit articles by saying "I did a Google search, this is notable enough to include" or something like that. A Google search is original research on the part of the person doing the search. If original research was not allowed as a way of deciding what goes into an article, this very common procedure would be prohibited. Ken Arromdee 17:11, 16 June 2007 (UTC)

Claiming that not including something is really a form of including something twists policies beyond all sensibility. Removing something is not "adding a viewpoint". It's removing.

Claiming that not including something is the same thing as removing something that is properly sourced and that is already part of the article, based solely on your original research, is expressing a viewpoint, is it not? Or is it OK to go through randomly deleting properly sourced statements from articles because my personal research does not agree with the so-called experts? Basically you are saying that it's OK to remove properly sourced statements from articles if you personally think they are wrong? This does not bode well for articles on religion, abortion, etc., does it? SqlPac 03:44, 16 June 2007 (UTC)

Sure, claiming that "not including" is the same as "removing" is a viewpoint. I have no reason to disagree with this viewpoint. It's permitted in either version.

But even if I disagreed with it, I wouldn't say it's a bad idea *because it's a viewpoint*. I'd say it's a bad idea for some other reason.

As for the articles about religion, abortion, etc., the same applies. I would not say "taking that out is bad because doing so is original research"; however, this wouldn't stop me from saying it's bad for some other reason, either a different reason or a more specific reason. Ken Arromdee 17:11, 16 June 2007 (UTC)

We're not discussing other reasons. Someone might be, based on their original research, of the opinion that one side of the abortion debate should not be expressed. Based on what you've said, it would seem that this is good enough reason for a person to delete the information they view as "false" from the article. Is this an accurate assessment of your position? If not, what reason justifies editing the article to put back the recently deleted information? SqlPac 23:45, 16 June 2007 (UTC)

If someone has an opinion that one side of the abortion debate should not be expressed, we could not say "that deletion is bad because all deletions based on original research are bad." We could, however, be more specific and say (for instance) that "should not be expressed" is not a factual claim and that we don't want to delete material based purely on the kind of opinion that does not involve factual claims. That would, in fact, prohibit certain deletions based on original research--but not every last one.

If, say, the article said "abortions are not prosecuted in Zenda", an editor who was prosecuted in Zenda for having an abortion should be able to take the line out of the article. Ken Arromdee 17:03, 17 June 2007 (UTC)

The editor has a responsibility under WP:V to source his information. Fortunately for the editor in question, prosecutions tend to leave a nice long paper trail of official government/court documents that can be easily sourced by the accused. Unless you're referring to the Zenda Top-Secret Supreme Abortion Court that operates in the shadows and is not "officially" a government institution anyway. In that case the credibility of the editor's claims will be questioned for other reasons. For all we know, the editor here might simply have been told, by the voices in his head, that he was prosecuted. Changing the content without properly sourcing it in your example is irresponsible on the part of the editor, and accepting this guy's changes as fact is irresponsible on the part of everyone else.

Your example is too easy. Let's try a harder one: If your editor is "pro-life" based on his original research, according to you it is not unreasonable for him to remove all "pro-choice" information from the abortion article, correct? (After all, the "pro-choice" information contradicts his own original research.) How about addressing some of my questions at the end of this thread? SqlPac 19:45, 17 June 2007 (UTC)

(re-indenting)

One of the frustrating things about this discussion is that people say things that are just not what actually happens. There's some reluctance about accepting official court documents as sources, since they are not necessarily published in a way that anyone can just go and get. Especially a prosecution in a foreign country. But you're forgetting that this is an *example*. Even if you disproved the example, it doesn't mean that such things don't happen. There have been cases where it is not possible to prove a source false using Wikipedia rules (other than WP:IAR), but where by any sensible non-Wikipedia standards, the information can be shown to be false, or at least unreliable enough to leave out. Ken Arromdee 22:49, 17 June 2007 (UTC)

I'm not sure what you mean by "things that are just not what actually happens." Any, and every, thing happens. In your example, it appears that we have an editor who deletes content from an article and uses a non-verifiable court action against him in a foreign land as his justification. In addition to being apparently unreported by any verifiable sources, there is apparently no record anywhere that the courts of the foreign land took any action against the person. To me that sounds less likely (though not impossible) than the alternative, in which a court action (no matter where) would leave some sort of evidence that there actually was a court action.

In the United States, and in many other countries, court documents are published as a matter of public record. Many professional writers, reporters, and researchers, who get paid to write use court documents as a primary source all the time. A lot of the work put out by these same people is used as a source by Wikipedians, all the time. I would consider reluctance to use official court documents as a source fairly unreasonable. Especially since you'll often source the writings of others who use official court documents as their source.

BTW, if you look at controversial articles, like abortion, you will see that they are often edit-locked. I chose the abortion article as an example because of its very contentious edit history that exemplifies this conversation. As you scroll through the edit history for it, you can see people deleting content that does not match their point of view. This is a thing that not only "can actually happen", it "does actually happen". Another example of such an item is the Tupac Shakur article that is changed every week when someone adds the claim that his birth name is "Lesane Parish Crooks." There's no real evidence of this anywhere; even his official birth certificate states otherwise. These things happen all the time... SqlPac 01:36, 18 June 2007 (UTC)

I'm not claiming that any edit to an article is automatically okay. There can be specific kinds of original research on which it's a bad idea to base a deletion. It just doesn't generalize to all original research. Ken Arromdee 17:11, 16 June 2007 (UTC)

Perhaps all the specific situations you allude to should be thoroughly documented, so that people will know exactly when it is OK to base editing decisions on original research and when it is not OK? SqlPac 23:45, 16 June 2007 (UTC)

And the guy who took out the attributable-but-false information would be smart to put something in the talk page or in the edit summary stating why he took it out, so someone else won't just put it back. And the correct response to "it was properly sourced" is "not every properly sourced statement belongs on Wikipedia".Ken Arromdee 19:10, 15 June 2007 (UTC)

And the guy who took out the attributable-but-false-based-on-his-personal-opinion information should add a note stating that "I removed this, because I personally believe it not to be true; no matter what you say"? I can see this resulting in a significantly smaller amount of information on Wikipedia.

In practice, this has not happened, as long as people are reasonable. (Unreasonable people, of course, are a problem with any rule.) Ken Arromdee 17:11, 16 June 2007 (UTC)

I don't know that this hasn't happened in practice. I know of at least a couple of situations in which people got into arguments and edit wars, possibly because of the perceived ambiguity in policy that you point out. I don't think it's a matter of being reasonable, as the guy who changed article content based on his original research was being perfectly reasonable according to some people. Also the guy who changed it back based on attributable sources was being perfectly reasonable according to others. In the case of such an edit war (which has happened), who is right and what is reasonable according to you? SqlPac 23:45, 16 June 2007 (UTC)

I have no idea. But I have yet to be convinced that deleting verifiable-but-false information is subject to worse edit wars than lots of other things for which edit wars happen. It just isn't a legitimate criticism of a policy to say "that results in edit wars"; everything we do results in edit wars. Ken Arromdee 17:03, 17 June 2007 (UTC)

So the two should keep editing the article back and forth, for all eternity? What is reasonable? I gave you an example of two reasonable people with differing opinions reverting each other's work ad infinitum. AFAIK, the policy states quite simply that the guy with the sources wins in this case, and the guy with the voices in his head telling him to delete the sourced statements loses. But then again, you seem to have a different reading of policy than I do. SqlPac 19:53, 17 June 2007 (UTC)

You do exactly the same thing you do when you have an edit war over anything else. Those don't go back and forth for all eternity; this one is no more likely to do so. Ken Arromdee 22:49, 17 June 2007 (UTC)

And what do you do? That's the question. From what I've seen, the guy with the sources wins the argument. Obviously, "for all eternity" is a bit of an exaggeration, but how about since 2001, with no end in sight? SqlPac 01:36, 18 June 2007 (UTC)

(re-indent)

"What do you do if this rule has the same problem that every other rule has?" is not, to me, convincing. Ken Arromdee 18:18, 20 June 2007 (UTC)

That might be one reason I didn't say that. The question was, and still it, "what do you do in the situation provided?" E.g., where two reasonable people—one has verifiable, attributable sources to back up his edits, while the other edits based solely on the voices in his head—get into a debate? If it's perfectly alright to edit content based on the voices in your head and nothing else, then I ask you again, why are there so many policies and talk pages on Wikipedia? What exactly is the point? SqlPac 19:34, 20 June 2007 (UTC)

Basically people should be allowed to eliminate content their personal experience disagrees with, despite proper sourcing. OTOH they are not allowed to add content that their personal experience agrees with if they cannot properly source it. So the only content allowed is content that can be properly sourced, and that *everyone's* experience agrees with. I guess the only question left is, if everyone agrees on all content *and* all content is properly sourced, why are there so many policies and talk pages like this? They would seem completely unnecessary. SqlPac 03:44, 16 June 2007 (UTC)

It seems to me that No Original Research, Neutral Point of View, Attribution, Verifiability, etc., are designed to prevent one person's personal experience or original research from overtly influencing content. This makes sense if you think that the original research of one person may be a statistical fluke or anomaly when compared to the established research of 100's of experts and professionals. To continue your example, maybe the vehicles your intrepid writer saw on the road were actually construction vehicles and the road is closed to normal traffic after all. Of course I'm sure it's possible to squeeze the blood out of any policy (this is not limited to Wikipedia) with enough bureaucratic and lawyerly word-wrangling. SqlPac 04:06, 16 June 2007 (UTC)

I'm also seeing wikilawyering here. I'm not sure how to work things out here, when as best I can tell many of the participants here are hell-bent on disrupting the policy process instead of facilitating it. IMO, attributability is what is important, and sources cannot be considered reliable of they are regularly wrong. For the matter at hand, if all available reliable sources say the road is closed, then note that according to several sources the road is closed, and include the citations. As I indicated above, it is what is generally known that is the issue here, and not that which is the absolute truth. So for example if you can show that the highway department web site now says that the road was reopenned on such-and-such a date, then that can also be cited as evidence of the road being open. I'm not going to say that one's personal experience is totally irrelevant here. Often that is how people know what to look for. Even so, at the end of the day, it is the ability to attribute the statements made here that matters and not one's personal experience in the matter. If the best information available is that the road is closed, then Wikipedia should reflect that and properly attribute it. --EMS | Talk 02:14, 17 June 2007 (UTC)

That's basically the idea I got from the policies mentioned, but we seem to have turned a corner here where some think it's OK to alter content based on personal, non-attributable, original research—so long as it's not "adding unsourced content". That seems like it completely circumvents the spirit (if not the exact wording) of WP:OR, WP:ATT, WP:V, etc. If deleting properly sourced content based on non-attributable, non-verifiable, original research is actually OK, then there's a big gaping hole in Wikipedia policy that should probably be fixed. A more cynical person might suggest that people are unreasonable—they will follow the exact wording of policy while ignoring the spirit and purpose of that policy. SqlPac 04:36, 17 June 2007 (UTC)

IMO, the reliance on sources comes from the fact that individuals are often not trustworthy, but the word from respected sources very often (but not always) is reliable. There will indeed be times that a normally reliable source is mistaken or outdated. However, the best way to have content in Wikipedia that is reliable is to use the available sourcing and not to rely on personal observations. An individual saying "but that road is open" can be wrong for any number of reasons, but their saying "the highway dept web site says that the road is now open" is something that can be verified and once verfied will cause people to correct the article. (The check of the highway department web site could have come from the editor finding the road open, BTW. IMO, there is nothing wrong with that, as they are backing up their discvoery with the word of a reliable source.) --EMS | Talk 05:04, 17 June 2007 (UTC)

I'm probably a bit cynical, but I think the reliance on published sources also goes to CYA, since it's hard for to claim misrepresentation by Wikipedia if the material isn't made up by Wikipedians. I personally agree with the need to source material properly, especially controversial material. I'm more concerned with getting clarification on some of the areas of policy that are not so well-defined. SqlPac 07:31, 17 June 2007 (UTC)

You reference to CYA is true and astute, but please realize that is also cover's Wikipedia's rear end too. If there is something specific in the policy that bugs you, then now is a good time to name it (or name it again), as this thread is so large that it hard to see what you are refering to. I will say in general that there does need to be some "wiggle room" in these policies, and that as a practical matter people's experience and judgement do come into play here, and they should. If an item is important, then it should be listed and attributed. If it cannot be attributed, then the decision is one of whether to leave it out or seek assistance in finding a source for it. Sometimes, a group consensus that a given statement is OK will exist, and the need for attribution will await a time when the statement is contested. --EMS | Talk 23:10, 17 June 2007 (UTC)

I actually posted previously (in the thread above this one, in fact) that I wanted clarification on what constitutes a "good source", etc. The definitions in the policies seem to rely on other items that are not defined. SqlPac 01:36, 18 June 2007 (UTC)

(<<<outdent) Don't forget: There are always prose attributions as a way out. The easy way to end the edit war is to include the disputed, sourced material, but to include it in quotation marks or as an indirect quotation such as "Agency X states that ...". Then the person who vociferously objects to the material on the grounds that it's false can accept the Wikipedia article as being acceptable and apparently true.

I agree that Wikipedia articles should not contradict themselves or other Wikipedia articles. Where there is controversy, use of prose attributions is a good way of avoiding contradictions. It's not a contradiction to say that A said X and that B said Y, even if X and Y are contradictory; all you're saying is that those things were said.

Most of the time, in most articles, such prose attributions are not needed. For highly controversial topics, they may be needed extensively within an article. --Coppertwig 17:00, 14 July 2007 (UTC)

Coastlines

An example which illustrates the distinction between truth and knowledge is coastline lengths. Most encyclopedias give the length of coastline for those countries or geographical entities which have a coastline, and Wikipedia is no different. However, these figures are essentially arbitrary. The length of a coastline depends almost entirely on how it is measured. A coastline is approximately fractal, and its length grows without bound as it is measured on smaller and smaller scales.

So there is no truth or verifiability in the length of coastlines. The best we can do is attribute the figures. If the consensus is that Wikipedia should be based on truth, then I hereby offer to go through all of Wikipedia's geographical articles and remove the coastline length. I guess I should probably do the same with river length as well, since this has a similar problem. I sincerely hope that this is not the consensus, as doing this will require a lot of work! Geometry guy 20:46, 28 June 2007 (UTC)

I see your point. However, the length of coastlines need not be totally arbitrary nor increase without bound. There could be a standard unit of length used in measuring the coastlines; and/or irregular shapes which are normally only sometimes on the coast (as tides or ordinary waves go in or out) could be ignored.

If the lengths of coastlines are more-or-less meaningless, then I would support either removing them or putting prose attributions, as in "Souce X gives the length of the coastline of this country as Y." Another alternative might be inserting a unit or explanation of how they were defined or measured. Are they of any use to anyone? Maybe they are actually somewhat standardized in how they're measured, and useful for calculating things like how many cottages can reasonably be built on the coast. --Coppertwig 16:41, 14 July 2007 (UTC)

What is being proposed?

Hello again,

What exactly is being proposed to have to be considered "true" in order to be included in Wikipedia? That some opinion is attributable, or that the opinion itself is true? If it is the latter, this runs into all sorts of problems. How would we then deal with statements for which the truth value is not known with certainty, such as the existence of God, UFOs, etc.? Thes things seem reasonable to include in an encyclopedia. We can't push one opinion as truth otherwise that would violate the neutrality principle, which is just as fundamental a Wikipedia policy and principle as verifiability and attribution. mike4ty4 09:04, 4 July 2007 (UTC)

I don't think anyone is proposing that material has to pass an objective truth test in order to be included: or if anyone is suggesting that, please give a link to where that suggestion is made.

The way to deal with things where the truth value is not known, such as existence of God, is how we do it now, for example using prose attribution, such as "Acquinas says that God exists". Everyone can agree that the quoted person said the quoted text.

Note discussion at Wikipedia talk:Attribution#The second sentence. --Coppertwig 16:31, 14 July 2007 (UTC)

Factual accuracy

Factually accuracy is required for Good Article status and Featured Article status. It is more important is that the article is coherent with itself and its sources it is attributed to. Attribution is not enough. The approach away from factual accuracy is original research. So avoiding original research should be enough for factual accuracy.

The facts are determined by the consensus among a group of intellectuals. Facts are created by man. Facts are justfied by rationality.

The truth is not determined by the consensus among a group of intellectuals. The truth is not created by man. Truth is not justified by rationality.

The former is more important for Wikipedia.◙◙◙ I M Kmarinas86 U O 2¢ ◙◙◙ 18:08, 14 July 2007 (UTC)

This page is an archive. Please do not edit the contents of this page. Direct any additional comments to the current talk page.