User:RegentsPark/Verifiability RfC

This page summarizes my conclusions from reading the various opinions on the RfC.

Preliminaries

Proposed changes

1. The removal of the word ‘truth’ from the lede and addition of a sentence indicating that other policies may apply
2. Adds a new paragraph that explains what “verifiable, not truth” means. The para also indicates which other policies may apply when deciding when to include or exclude material.

Stated rationale for changes

1. clarifies that verifiability is not a sufficient reason for inclusion
2. clarifies that editorial judgement about accuracy is allowable, particularly in the case of ‘untruths’.
3. removes truth from the lede (which could be distracting and confusing for new editors)

Threshold stats

A total of 444 !voters commented on the proposal. The number of supporters was 276 while 149 editors explicitly opposed the proposals. 19 'neutral/abstain' comments were without supporting or opposing the proposal. This translates to:

• A 65% support percentage from amongst support/oppose !voters
• A 62.5% support percentage from amongst all !voters

Main oppose arguments

1. Fundamental disagreement: These arguments centered around the word 'truth', mainly that 'not truth' is a central tenet of wikipedia, that wikipedia does not care about the 'truth' and therefore an 'assertions of untruth' section is unnecessary. The underlying assumption, sometimes explicitly stated, in these oppose !votes is that the change in wording results in a change in policy.
2. Opens up the possibility that truth/non-truth is debatable without sourcing and/or provides a blank check to POV pushers to keep coming back with arguments. Unnecessary disputes about truth would arise. This is a stronger version of 1 above and there is a large overlap between the oppose !voters in these two groups. Some oppose !voters worded this differently, that people adding OR is a significant problem on wikipedia but that the new wording downplays this. (By my count, 41 !voters explicitly made either point 1 or point 2.)
3. Current statement - “Verifiability, not truth” is a clear statement (more powerful statement). Many oppose !voters felt that this (new) wording only adds confusion and/or the proposed change is worse than the current statement. Current version is more powerful. The premise that the current statement implies all verifiable material is includable is incorrect. Many editors also felt that 'verifiability, not truth' is a recognizable and “rallying cry” for wikipedia. Somewhere between 55 and 60 editors made this argument explicitly.
4. The distinction between truth and verifiability is better made up front in the lede because it is the fundamental tenet, and because few people read beyond the first couple of sentences anyway. This is a variation of the 3rd argument and a little less than 50 editors made this one. (The union of 3 and 4 is about 85 distinct editors.)

Main additional points in support

(The main support arguments are listed above in the Preliminaries section. I'm assuming that most !voters in this section supported on that basis and am therefore only listing the main additional thoughts of support !voters here.)

1. The proposed wording is the best compromise available
2. The wording may not be perfect but can be tweaked once the initial change is in place
3. The proposed wording does a great/adequate job clarifying the meaning of ‘verifiability, not truth’
4. Because of the additional clarity provided by the proposed wording, it will be harder for (tendentious) editors to misrepresent policy
5. The proposed wording codifies the need for editorial discretion
6. Verifiable and ‘not truth’ are opposite in meaning (based on the root of verify)
7. We seek truth on wikipedia. The new wording clarifies that.
8. There is no truth. The new wording clarifies that. (A very small minority of !voters used this rationale but it is interesting in juxtaposition with point 8!)
9. Verifiability, not truth was added without discussion and removing it from the lede will fix that (one !voter)

Decision

No consensus

My decision is that the RfC should be closed as "no consensus". The primary reason is that WP:V is an important policy document on wikipedia and changes to it should only be made with strong community support. A 65% (or 62.5%) level of support does not provide that necessary degree of comfort. Additionally, it appears that many editors, on both sides of the aisle, believe that the changes are a policy change rather than a mere clarification. Since that was not the intent of the changes, and since policy is, in the final analysis, solely contained in wording, I suggest that going back to the drawing board is probably a good idea regardless of the level of support.

Takeaways

However, I believe that the RfC itself has been very useful. While it is hard to gauge consensus for specifics, there are several useful takeaways that should help guide the next steps in this process.

1. The support for the clarification paragraph (proposed change number 2 above) was higher than the support for changes to the lede. At least 9 oppose !voters expressed willingness to see that change and, considering the number of !voters whose opposition mentioned only the lede, it is possible that there will be the necessary consensus for making that change. However, since the changes were presented as a package, it is not possible to be sure about this without a separate RfC. But, it should be noted that there is some concern about the wording (or length of wording) in that para and several support !voters indicated there unhappiness with the wording (while supporting the clarification itself).
2. There doesn't appear to be the necessary consensus for removing the "verifiability, not truth" phrase from the lede.
3. However, the RfC has clearly shown that the phrase "verifiability, not truth" is the source of much confusion on wikipedia and, at the minimum, needs clarification. For example, there is a significant minority of editors who believe that the purpose of the encyclopedia is to embody truth, not just verifiable 'truth'.

Disclosure

It would be disingenuous to say that, after reading everything twice over, I have no opinions of my own. So here they are.

Surprisingly (to me anyway), I'm unable to decide whether I would have supported or opposed the changes. I believe that the purpose of the changes is clarification, not a policy change, and much of the clarification provided agrees with my interpretation of the policy so I'm not uncomfortable with the changes. After reading the comments on the RfC, it is more than clear to me that the policy needs clarification. The fact that the same words can be interpreted to mean that we seek only truth and that we don't seek the truth at all is a clear reflection on the confusion in the minds of the editors. That a significant number of editors believe that the encyclopedia embodies, or should embody, 'truth' was also interesting. That tells me that the definition of truth is fungible and we, as a group, either need to be clear on what we mean by it or not use it at all. On the other hand, I also agree many of the oppose !voters that the new wording is unclear (I guess it reflects the fact, or should I say 'truth', that it is a compromise). I'm also troubled by the fact that many editors - on both sides - seem to think that the policy itself is changing (though I don't see that at all). Would I have !voted on the support or oppose side at the beginning of the RfC? I don't know. But, it is quite clear to me that a change of this magnitude, real or imagined, doesn't (as yet) have the necessary consensus. If this RfC finally closes as 'no consensus', I hope that the initiators of this RfC are willing to go back to the drawing board and revisit the wording. I suggest working to change the wording incrementally. Given the unwieldy number of editors in the project, it might be easier to get consensus on small changes, but that's just a personal thought.