Wikipedia talk:Verifiability/First sentence

Compromise.. with a new approach

edit

Offered in the spirit of compromise... please focus on the concept for now... we can argue about exact wording later if people like the concept. What I suggest is that we keep verifiability, not truth... but not in the lede sentence. I am thinking of something along the lines of the following:

Start the policy off with the following paragraph:

The initial threshold for inclusion in Wikipedia is Verifiability—whether readers can check that material in Wikipedia is directly supported by a published reliable source. Note, however, that while verifiability is a requirement for inclusion, it is not a guarantee of inclusion. Wikipedia has other policies and guidelines that affect inclusion (especially whether specific material is included in a specific article.)

This focuses the lede completely on Verifiability, without the side issue of truth. Truth/untruth issues can then be dealt with in a separate section... something like:

===Assertions of truth and untruth===

An editor's assertion that something is true is not enough for inclusion in Wikipedia. It does not matter how convinced you are that some bit of information is true, if the material is unverifiable - do not add it. In this context, Wikipedia requires "Verifiability, not truth".

An editor's assertion that something is untrue is also not, on its own, enough to exclude information from Wikipedia. In some cases, Wikipedia must include discussion of ideas, theories, and opinions that are considered untrue in order to maintain a Neutral point of view. Discussion of such material, however, should be phrased as opinions, and attributed to those who hold the opinion. They should not be stated as if they were accepted fact. Also note that not all opinions are given the same weight in an article. Additional guidance on when and how to include or exclude verifiable, but potentially untrue material, can be found at WP:Editing policy, WP:Fringe theories and WP:No original research.

Something like this would allow those of us who find the statement "verifiability, not truth" useful in combating POV pushers and Fanboys to continue to use the phrase... but in its proper context, which I hope will be OK to those who are concerned about Wikilawyers. Blueboar (talk) 21:38, 1 September 2011 (UTC)Reply

  • The idea seems to have potential.—S Marshall T/C 21:49, 1 September 2011 (UTC)Reply
  •   Me likey.--Cerejota (talk) 22:50, 1 September 2011 (UTC)Reply
  • I like. That last paragraph starts wandering into other policies/guidelines and would need work, but that is secondary. North8000 (talk) 23:01, 1 September 2011 (UTC) First sentence of the last paragraph is often per policy true, but as a categorical statement is also sometimes false as a statemetn of policy. I.E. inventing a policy that does not exist. Suggest dropping. North8000 (talk) 02:41, 2 September 2011 (UTC)Reply
  • Works for me. Hans Adler 23:08, 1 September 2011 (UTC)Reply
  • Yes, this approach is a good idea. Thank you, Blueboar. I'd like to see it developed further. --Tryptofish (talk) 00:08, 2 September 2011 (UTC)Reply
  • It's good. It allows editors to keep a WP:Verifiability, not truth link to part of a policy page, and it can be used contradict wikilawyers insisting that (often spurious) verifiability compels inclusion. I know tweaking comes later, but three broad ideas: (1) "Untrue" content: "considered untrue" is too strong for NPOV, it should be something like "not commonly accepted or widely considered untrue". Not everything is as black and white fringe as creationism, and it's gentler to editors who have non-mainstream views on certain subjects. (2) There should also be something for editors wanting to show that a source has made a factual error (for example: birthdates, populations, spellings etc.) and that they should look to WP:RS and WP:RSN when dealing with such contradictions between sources. (3) I suggested this above as an extra sentence for the lede "A reliable source Is one that, according to our policy, is reliably accurate and appropriately authoritative for the content in question." I think something like this would help avoid the problem of people presuming that if a source is RS for one piece of information, it is therefore RS for all information (an abuse of "verifiability"), as well as emphasise to the world how it is we achieve accuracy.VsevolodKrolikov (talk) 01:57, 2 September 2011 (UTC)Reply
  • It needs work but it incorporates improvements that have been discussed, which is good. However, it's trying too hard to keep the nonsense phrase "verifiability not truth" in policy, and it shows. The phrase doesn't add anything to this proposed version of the policy and should not be included. If anyone wants to use the phrase in discussions, they can still use the essay link Verifiability, not truth. --Bob K31416 (talk) 02:00, 2 September 2011 (UTC)Reply
  • I like this proposal also.--Andrew Lancaster (talk) 06:22, 2 September 2011 (UTC)Reply
  • I like this as well, although I am still a fan of having the first sentence end solely in defining verifiability. A full stop where you have your first dash and a reworking of the subordinate clause into it's own sentence. (see proceeding section).Crazynas t 09:20, 2 September 2011 (UTC)Reply
  • I like this too. I do not entirely agree about the characterization of verifiability; what is verifiable is that this is a significant view of the object - reliable sources are our means for verifying this. I don't think this is a meaningless distinction. Be that as it may, Blueboar obviously put a lot of good thought into this and I think it is very helpful. Slrubenstein | Talk 13:23, 2 September 2011 (UTC)Reply
  • I can live with it, it seems a reasonable compromise. Perhaps, now that we are changing things, "the initial threshold" can be changed to "a fundamental requirement" or something similar? There have been a fair number of complaints over the word "threshold" as well, so perhaps that would make those people happy as well :-) Fram (talk) 13:38, 2 September 2011 (UTC)Reply
May I suggest that trying for too many changes at once will make it more difficult to achieve a consensus and compromise. Let's focus on one thing at a time... we can discuss whether to change "threshold" and things like that once we reach a compromise solution as to "not truth". (To be honest... I am surprised no one has objected to my insertion of "initial"... but perhaps that is because I asked us to focus on concept and not wording for now). Blueboar (talk) 15:15, 2 September 2011 (UTC)Reply
I think you're right on all counts. So it does sound a bit awkward......Sincerely, North8000 (talk) 21:39, 2 September 2011 (UTC)Reply
  • Comment - it needs a small change. As it is currently written, it defines "Verifiability" as being able to find a "reliable source". While this is a major part of Verifiability, it is not the entirety of it. Another alternative for having a reliable source is things that are absolutely well know like "most cats have fur", and even more strongly, things that are just absolute logic, like 2+2 = 4, or more complex, if the US tracked its exports in two separate places, maybe East Coast and West Coast, you could (without violating OR) say Total Exports = East + West. These things are Verifiable, but not through Reliable Sources. (please don't make the argument that maybe some reliable sources would cover these particular examples, these are, after all, just examples). -- Avanu (talk) 13:56, 2 September 2011 (UTC)Reply
That's sort of a different topic that is not currently being debated. In that area, this proposal merely repeats the current wording. North8000 (talk) 14:02, 2 September 2011 (UTC)Reply
You are talking about the "common knowledge" exemption... which isn't really an exemption at all. There are lots of reliable sources to support common knowledge statements like "most cats have fur" and "2+2=4". Such statements are extremely verifiable... in fact, they are so verifiable that we don't actually require them to be verified. (Remember, verifiable is not the same as verified)... and while that last point could probably be made in a clearer way, I agree with North that it is a different issue. One thing at a time. Blueboar (talk) 15:05, 2 September 2011 (UTC)Reply
But the two versions do not say or imply the same thing:
Currently,
whether readers can check that material in Wikipedia has already been published by a reliable source
Proposed,
whether readers can check that material in Wikipedia is directly supported by a published reliable source
You change it from merely being published by a reliable source, to being "directly supported", whatever that means. I think the proposed wording still needs a tweak. -- Avanu (talk) 16:16, 2 September 2011 (UTC)Reply
I took "directly supported" from WP:NOR. The idea I was trying to convey is that we don't simply copy what is published verbatim... we re-write what the sources say in our own words... and point to the sources to support what we write. That said, I have no problem with returning it to "already published" if it would be a stumbling block to consensus. Again... so far, I have been going for acceptable concept, and not worrying about perfect language. Blueboar (talk) 17:12, 2 September 2011 (UTC)Reply
  • And I've filled it in, please feel free to edit it boldly. If we can achieve a version that enough editors like, then we can bring the detailed wording back here or to WT:V in search of consensus.—S Marshall T/C 16:52, 2 September 2011 (UTC)Reply
Sounds like a plan to me. I am actually hopeful that this may break the log jam. S Marshal and I are both very "hard to please" (stubborn?) proponents of the two sides in the debate... and if we are starting to agree, that is a hopeful sign! :>) Blueboar (talk) 17:01, 2 September 2011 (UTC)Reply
  • Good improvement. I do think that we need to change "a published reliable source" to "published reliable sources", as verifiability often involves the whole body of literature that you need to be familiar with in order to correctly interpret what some single source says. Count Iblis (talk) 17:36, 2 September 2011 (UTC)Reply
FYI... I have tried out a bold re-phrasing of my second paragraph (the one dealing with untruth). (see this dif at the draft page)... the major change was to admit that assertions of untruth are more complicated (which they are)... and that there isn't one single way to deal with them. (sometimes our policies do need to say "it depends" rather than "always do this".) Blueboar (talk) 18:11, 2 September 2011 (UTC)Reply
  • Looks good with those changes. I'd like to lose the last paragraph. It is EXTREMELY well written and useful, but is an attempt to summarize policies other than wp:ver. But that is minor; this is fine even with that last paragraph in. North8000 (talk) 18:16, 2 September 2011 (UTC)Reply
  • BTW To be a complete proposal it also has to precisely specify the intended change. Like "Replace the first sentence with (the first box) and add (the second box) after the lead, with it's heading as a top level section heading." North8000 (talk) 18:23, 2 September 2011 (UTC)Reply
  • I continue to like the approach proposed here. I'm not sure at what point we start to move from discussing the overall idea to discussing the specific writing, but it looks like some cats in the herd editors are already well down the road of that discussion, and I think we have enough interest expressed that it's a good idea to discuss the details. I would want the "Assertions of truth and untruth" section to be the first section after the lead. I'd also float the idea of making it the end of the lead, instead of a separate section. Start the lead off with a good, succinct statement of what verifiability is, then go on to cover its relationship with truth: that's still a big change from the status quo. --Tryptofish (talk) 20:16, 2 September 2011 (UTC)Reply
  • I could reasonably happily live with this one. The odd tweak to replace threshold with [insert alternative here], f'rinstance, can wait until a bit later. Pesky (talkstalk!) 20:27, 2 September 2011 (UTC)Reply

updated proposal

edit

Taking into account many of the comments made so far... my current proposal is as follows:

Change the lede paragraph to:

The initial threshold for inclusion in Wikipedia is verifiability—whether readers can check that material in Wikipedia has already been published by a reliable source. Note, however, that while verifiability is a requirement for inclusion, it is not a guarantee of inclusion. Wikipedia has other policies and guidelines that affect inclusion (especially whether specific material is included in a specific article.)

Add a new section (right after the lede) to deal with the issue of truth/untruth... as follows:

==Assertions of truth and untruth==

An editor's assertion that something is true is not enough for inclusion in Wikipedia. It does not matter how convinced you are that some bit of information is true; if the material is unverifiable, do not add it. In this context, Wikipedia requires "verifiability, not truth"'.

Assertions of untruth (ie an editor's assertion that some bit of information is untrue) are a more complicated issue. If the dubious information is not supported by a source, it should be challenged; but the question of whether to tag the information as needing a citation or to remove it immediately it depends on the nature of the information (see: WP:Burden, below). If the dubious information is supported by a reliable source, the problem should be discussed on the article talk page, with reference to policy concepts such as maintaining a neutral point of view (and especially the sub-concept of due weight). In many situations, a simple rewording to present the information as an opinion rather than as an accepted fact can resolve issues of verifiable but potentially untrue information.

Again, we can tweak the exact wording (and argue about things like "threshold vs. criteria" etc.) later, but I now formally propose something close to this (and I will add it to the proposals list on the main WP:Verifiability/First sentence page). Blueboar (talk) 13:22, 3 September 2011 (UTC)Reply

  • Support as updated But, Blueboar, it's going to get lost in the crowd on that hidden poll page. Plus no change is ever going to occur just based on that page. Should we noodle on it here / build consensus that it is THE compromise proposed change, and then discuss the next step? North8000 (talk) 13:31, 3 September 2011 (UTC)Reply
OK, you did it already. But suggest we also noodle on it here / build consensus that it is THE compromise proposed change, and then discuss the next big step. North8000 (talk) 14:01, 3 September 2011 (UTC)Reply
  • Well, I think Blueboar's proposals show a thoughtful effort to address diverse concerns. My understanding is that the slogan has always been shorthand for "inclusion or exclusion of material in Wikipedia will not be bnased on whether they are true or false (because people cannot reach a consensus as to how to prove this) but instead will be based on whether or not we can verify that a significant group of people hold the view in querstion (because people can reach a consensus as to how to prove this)." I think the questions have always been (1) how best to explain this and (2) where is the appropriate place to explain this - or appropriate places. It used to be explained at NPOV. I peresonally think some of it should be explained at V and some should be explained at NOT. But I think most people are settled that it should be explained at V. That leaves the first question, and Blueboar is making valuable contributions. Slrubenstein | Talk 11:19, 4 September 2011 (UTC)Reply
(Thanks for the kind words, SLR... I do try. But let's not make this about the editors... let's focus on the proposals. I have no problem with someone pointing out that I am being inconsistent (I often am... especially on issue where I have sympathy with both sides of the debate), or saying they think my proposal goes too far, if that is what they think. Criticism is good... It gives me something to work with as we continue to work towards compromise.) Blueboar (talk) 12:46, 4 September 2011 (UTC)Reply
Of course you are right (but ... I did begin my comment by referring to the proposal, rather than you). Based on North 8000's comment in the thread currently at the bottom of the page (to which I just replied) I do think that the policy (if not the first line) perhaps needs clarification but I would like to know what you think - you can comment on the thread at the bottom (responses to the "truthiness" proposal) unless you want me to move it up here ... I am not making a specific proposal (except perhaps my expanded version of the slogan) but the point I think is: if a view is verifiably significant, as evidenced by reliable sources, then it cannot be excluded from an article just because one or more editors consider the view to be false (if needed, I try to explain this more fully in the thread below. I honestly thought it was obvious, but if you think I am wrong about that ... well, feel free to comment on my perception!) Slrubenstein | Talk 17:48, 4 September 2011 (UTC)Reply

A variation on Blueboar's proposal

edit

Since we're just discussing things here, I'd like to float a variation on the idea. I'm thinking, specifically, of some of the objections in the poll, that: (1) it's seen as too long, and (2) there are disagreements about the last paragraph (which strike me as the kinds of things where proposals can get sidetracked by the details). I'm also anticipating that there will be editors in the larger community who will still insist on seeing the words "not truth" somewhere in the lead.

So I'm suggesting not proposing a section about "Assertions of truth and untruth", although eventually a section expanding on the idea could still be a good thing. Instead, I propose this, all in the lead:

The initial threshold for inclusion in Wikipedia is verifiability—whether readers can check that material in Wikipedia has already been published by a reliable source. While verifiability is a requirement for inclusion, it is not a guarantee of inclusion. Wikipedia has other policies and guidelines that affect inclusion (especially whether specific material is included in a specific article).

An editor's assertion that something is true is not enough for inclusion in Wikipedia. It does not matter how convinced you are that some bit of information is true; if the material is unverifiable, do not add it. In this context, Wikipedia requires "verifiability, not truth".

--Tryptofish (talk) 18:36, 4 September 2011 (UTC)Reply

I could certainly live with it... but I am concerned that (like the current language) it ignores the issue of what to do about verifiable, but potentially untrue material (and that, I think, is the elephant in the room and the issue that underlies this endless debate.) Blueboar (talk) 19:12, 4 September 2011 (UTC)Reply
WP:V talks about that. Something might simply be untrue because of the way it is presented or phrased. Saying "All Cats Have Fur" is probably untrue. But saying, "According to Bob Roberts, PhD in Cats, 'all cats have fur'" makes this a true statement. What Bob said is still open to being false, but Wikipedia's presentation of it is true. -- Avanu (talk) 19:23, 4 September 2011 (UTC)Reply
Thanks. No argument here. But I'm just thinking about what has the best chance of consensus, and perhaps it's better to leave that issue for the next round of discussion. (Maybe this is Bob K's issue that I didn't understand before, but I'm starting to agree.) --Tryptofish (talk) 19:19, 4 September 2011 (UTC)Reply

Hmmm... I am not sure if we could get a consensus if we continue to ignore the elephant. That elephant is essentially the reason why people are calling for "not truth" to be cut. But perhaps it could be addressed with just one sentence as follows:

The initial threshold for inclusion in Wikipedia is verifiability—whether readers can check that material in Wikipedia has already been published by a reliable source. While verifiability is a requirement for inclusion, it is not a guarantee of inclusion. Wikipedia has other policies and guidelines that affect inclusion (especially whether specific material is included in a specific article).

An editor's assertion that something is true is not enough for inclusion in Wikipedia. It does not matter how convinced you are that some bit of information is true; if the material is unverifiable, do not add it. In this context, Wikipedia requires "verifiability, not truth". For guidance on the inclusion or exclusion of verifiable but potentially untrue material, see WP:NPOV (and especially WP:UNDUE).

Just a thought. Blueboar (talk) 21:11, 4 September 2011 (UTC)Reply

That wording seems fine to me. -- Avanu (talk) 21:13, 4 September 2011 (UTC)Reply
Yes, that's fine with me too. --Tryptofish (talk) 20:28, 5 September 2011 (UTC)Reply
  • Leaving "not truth" in the lede is not in any sense a compromise. The proposed version is the status quo, in only slightly different language, and it fixes little. The compromise that was gaining consensus was to leave it in the policy but put it in a separate section where its meaning and context could be explained. And Blueboar is quite right about the elephant in the room.—S Marshall T/C 23:04, 4 September 2011 (UTC)Reply
I hope that editors in this talk will think carefully about the significance of S Marshall's comment. If anything ever comes out of this page, and is put in front of the entire community, there will be various impediments to acceptance. These impediments include not only those editors who are opposed to any change to the policy, and there are many of those, and not only the tendency of Wikipedian culture to reject any change, but also the fact that a few of the strongest advocates for change are going to oppose anything that doesn't kill "not truth" by driving a stake through its heart (and perhaps will oppose anything that might achieve consensus, because then there would no longer be something to discuss endlessly). I have given up on trying to convince the latter group, not least because there are so few of them (however vocal they may be here), but I hope that other editors will realize that catering to that group will mean nothing will get community consensus. --Tryptofish (talk) 20:38, 5 September 2011 (UTC)Reply
Irrespective of whether I'm a bizarre character who enjoys keeping this talkpage locked in eternal discussion for my own perverse amusement, I don't see a consensus forming in favour of this idea anyway. We should, perhaps, return to Blueboar's version.—S Marshall T/C 19:24, 6 September 2011 (UTC)Reply
There's no hurry, and I think you are the only one who is really objecting. Blueboar's proposal was made in the spirit that it is subject to revision, and the most recent iteration was actually proposed by Blueboar, with a sentence to address that "elephant in the room". --Tryptofish (talk) 19:32, 6 September 2011 (UTC)Reply
S Marshal... Do you actually object to tryptofish's revision (with my added sentence)... or do you simply prefer my original version more than his version? If the latter, then I actually agree with you... I don't object to trypto's version, but I prefer mine. (of course my actual preference is staus quo... but I accept that others object to it) Blueboar (talk) 14:19, 7 September 2011 (UTC)Reply
(And I, in turn, do not object to Blueboar's earlier version, although it's actually farther from the status quo. Either way, there's a bit of word-tweaking that still needs to be done.) --Tryptofish (talk) 15:11, 7 September 2011 (UTC)Reply
I see it as virtually no advance over the status quo. From my point of view, it's a token gesture rather than a change.—S Marshall T/C 15:54, 7 September 2011 (UTC)Reply
But the exact same wording if moved out of the lede and into to a section immediately below the lede would be acceptable? Blueboar (talk) 20:38, 7 September 2011 (UTC)Reply
I very much want to hear an answer to Blueboar's question, too. --Tryptofish (talk) 22:57, 10 September 2011 (UTC)Reply

Wait, is this yet another proposal to remove the phrase "verifiability, not truth" from the opening sentence? The same proposal that fails month after month to gain consensus? Jayjg (talk) 01:45, 11 September 2011 (UTC)Reply

Yup, and it's a lot milder than most of the rest that has been discussed. Editors who live on this talk page had better be prepared to hear a LOT more editors who agree with Jayjg if and when anything gets placed in front of the larger community! --Tryptofish (talk) 21:46, 11 September 2011 (UTC)Reply
And I'm still interested in an answer to Blueboar's question. --Tryptofish (talk) 18:49, 12 September 2011 (UTC)Reply
I think we're all aware that there's a significant gap between "not in any sense a compromise" and "not in any sense a compromise that is acceptable to S Marshall". WhatamIdoing (talk) 21:14, 12 September 2011 (UTC)Reply
Then we shouldn't get sidetracked by that. --Tryptofish (talk) 16:37, 13 September 2011 (UTC)Reply

Threshold and Verifiability

edit

"The initial threshold for inclusion in Wikipedia is Verifiability" Since this page is primarily about Verification, rather than the threshold for inclusion. I'd like to see the sentence order changed to reflect this. Something like:

  • "Verifiability allows material in Wikipedia to be checked against reliable sources. It is one of the five main criteria (see Five Pillars, right) that allows us to assess whether material may be suitable for inclusion in Wikipedia. --Iantresman (talk) 11:56, 7 October 2011 (UTC)Reply

FYI, There is a discussion at Wikipedia talk:Verifiability/First sentence/Procedural#RfC management, RfC bot tagsUnscintillating (talk) 02:39, 4 November 2011 (UTC)Reply

There is a discussion at Wikipedia talk:Verifiability/First sentence/Procedural#anonymous ad hoc change to the RfC textUnscintillating (talk) 16:22, 5 November 2011 (UTC)Reply

Alternatives? Or basis?

edit

All this discussion rather intimidates me, so perhaps someone could just tell me. The first sentence suggests that verifiability is an alternative to truth. Would it not be more correct to say that verifiability is the basis of any knowledge of truth? That is, it is not sufficient that a statement be true, it must be shown to be true, or at least why it might be thought to be true. ~ J. Johnson (JJ) 22:45, 9 November 2011 (UTC)

No... the point is to say that verifiability is a fundamental requirement for inclusion, while truth is not necessarily a requirement for inclusion. Blueboar (talk) 22:51, 9 November 2011 (UTC)Reply
I guess the point is (trying to find words most can agree with or follow):-
  • Verifiability is a fundamental requirement for inclusion. In practice this means that for anything debateable, someone needs to have published that something is true. We are collecting information from outside sources.
  • Collecting truth is a background aim for us as people making an excyclopedia, but un-verifiable truth is not useful or relevant when deciding what is good enough to include on Wikipedia. I would suggest that this is true by definition in any encyclopedic ("tertiary") collection of information about what has been published, simply because we are not aiming to "produce" new findings.
  • Just for completeness sake I will mention that it seems to be a widely accepted practice to allow discussion of un-verified information when deciding NOT to include something. The "rules" are less strict about when NOT to include something, but WP:NEUTRAL does require consideration in many such cases.
Does that wording match the understanding of others?--Andrew Lancaster (talk) 11:08, 10 November 2011 (UTC)Reply
Both of your statements seem to suggest that verifiability and truth are separate, and that we require one but not the other. Yet the concept of truth keeps hanging around, as in "background aim" and "un-verifiable truth".
Another way of looking what I am asking about is: what is the connection between verifiability and truth? Are we possibly doing the one for the sake of the other? ~ J. Johnson (JJ) 18:33, 10 November 2011 (UTC)
"Truth" is a red herring word because in common usage it is used to both matters of opinion and also accurate statements. Of the two meanings, I think that verifiability helps achieve accuracy, (including accurate coverage of errors and opinions as such) which is the common meaning of "information" as in the foundation's objective statements. North8000 (talk) 22:17, 10 November 2011 (UTC)Reply
I certainly agree that "truth" is a slippery concept. And as it is so often bandied about without due consideration that I have become reticent about using it. But I think there is some validity to the concept, slippery though it be. Therefore my question: have we given up on "truth" as such, choosing this other concept of "verifiability" as an acceptable alternative? Or: are we using verifiability as a way of determining possible truth? ~ J. Johnson (JJ) 19:41, 12 November 2011 (UTC)

The short answer to all three of your questions IMO is "no".  Elements of the encyclopedia, on the basis of expediency, dismiss truth as a consideration, and will argue that verifiability (i.e., the existence of authoritative published material documenting the claim) displaces truthfulness.  As an unrelated process, many of the content policies avoid discussion about truth in deference to the usage on WP:V.  However, WP:Editing policy still states that "...on Wikipedia a lack of information is better than misleading or false information", and Jimbo continues to back this clause.  As a practical matter, verifiable material may or may not be accurate, see WP:Inaccuracy#Verifiable material may or may not be accurate.  Truthfulness and verifiability are independent variables with a statistical correlation.  Because of these last two factors, the usage of the words "truth" and "true" on the WP:Verifiability policy page are confounding, and your questions reflect this confounding of ideas.  What we are left with IMO is populist sentiment to confound this policy page with the words "truth" and "true".  Unscintillating (talk) 17:51, 13 November 2011 (UTC)Reply

"Accuracy" can sometimes be hard to define, but substitution of the word "truth" for accuracy creates a straw man that deliberately adds needless complication. That is because "truth" has two common meanings: #1. Accurate information. #2 People's opinions. For an example of the latter, if someone said: "here's the truth about Obama:" you know that what comes next is going to be their opinion. That is the second meaning of the word "truth" North8000 (talk) 20:37, 13 November 2011 (UTC)Reply

Unscintillating kind of confuses me. Perhaps you were first stating the position of certain "elements", then giving your own view? At any rate, you say there is a correlation, but is the correlation purely incidental, and verifiability is an ultimate good in itself independently of "truth"? Or is the goodness (desireability) of verifiability derived in part because it may correlate with (lead to) "truth"?
On the other hand, North8000 seems to deny that there is objective truth, and supplants it with "accuracy". Well, I think there is much more to "truth", and its discovery, than mere accuracy; I would say accuracy is just the very first stage of verification. However, my query isn't really meant to get into the nature of either truth or verification, but concerns the relationship between them. As to popular misusage, I certainly agree that confuses matters, but I don't think such misusage is relevant to this discussion. ~ J. Johnson (JJ) 21:26, 13 November 2011 (UTC)
On the contrary, I most certainly think that there is objective truth. North8000 (talk) 21:32, 13 November 2011 (UTC)Reply
The point initially confuses many editors that verifiable material may or may not be true.  Under WP:V policy, it is verifiable, and WP:V is fully satisfied on this point, that Dewey defeated Truman in 1948.  We know this because we have an authoritative source, as documented at Dewey Defeats Truman, to back us up in saying so.  The reasons that we would not use this source to make this claim in the encyclopedia, are not to be found in WP:V.  Unscintillating (talk) 23:19, 13 November 2011 (UTC)Reply
I'd say that verifiability is useful independently of truthfulness.  There are four logic states involved: verifiable+true; verifiable+false; unverifiable+true; unverifiable+false.  The populist sentiment focuses on the case of "unverifiable+true" as being crackpot theories, satiring truth in the essay WP:The Truth, and trademarking "truth" written out as TruthTM.  This viewpoint conversely assumes that all cases of "unverifiable+true" are actually cases of "unverifiable+false", and this in turn has left the policy without clear guidance for the real cases of "unverifiable+true".  WP:Inaccuracy focuses on the case of "verifiable+false".  As far as "verifiable+true", I think we assume that anything that is verifiable is also true until we have evidence to the contrary.  More to the point, we assume that anything that is verifiable is publishable as truthful until we have evidence to the contrary.  Unscintillating (talk) 23:19, 13 November 2011 (UTC)Reply
For whatever it may be worth, I am inclined to think that ultimate "truth" is unknowable (at least to absolute certainity"), and in any given instance we're really chasing after its shadow. (Like Plato said.) And verifiability is practically nothing, but for being the first step in getting any approximation of truth. Which could suggest "Verifiabiliy, not towards truth." But this a deeper hole than I want to get into for now; I'll leave it to the rest of you deep thinkers. :-) ~ J. Johnson (JJ) 22:02, 14 November 2011 (UTC)

To start with, I'm setting aside the second common meaning of "truth" (opinions, dogma) and deal with just the meaning which basically means "accurate, and not misleading". IMHO the answer is They are two separate words which have no inherent relationship until you define the framework for the "relationship" question. If you define the framework as cause/effect in Wikipedia, I would say that verifiability helps achieve truth/accuracy, in those cases where objective truth/accuracy exists. If you are talking about the relation of the two "sets" in Wikipedia (material which meets wp:verifiability requirements, and material which is accurate/true (for those cases where object truth/accuracy exists)) I would say that they are separate sets which partially overlap. North8000 (talk) 22:43, 14 November 2011 (UTC)Reply