User:Saltine/IAR

Just wrote this out as some random thoughts.

"Ignore all rules" is just "Be bold" applied to things other than basic editing of articles.

I interpret it this way: IAR follows from: (1) consensus being Wikipedia's fundamental decision-making process and (2) human nature when it comes to enforcing rules.

Consensus

Most of Wikipedia's policies and guidelines have evolved over a long period from consensus; few policies are sent down from a higher authority. Consensus is Wikipedia's fundamental decision-making process. Policies, with few exceptions, are intended to represent consensus on certain topics. Without policies, a consensus would be required for every single disagreement on the wiki. A policy-less wiki would require a much higher level of participation and discussion from all editors. Speedy deletion is an example of time-saving policy; the criteria for speedy deletion are those that would almost certainly lead to a consensus to delete on AFD. Imagine a consensus-gathering discussion for every single speedy deletion; it would be a huge time sink. The speedy deletion policy was formed to reduce the workload on editors while maintaining consensus.

That said, consensus (i.e. the general opinion of editors on a particular topic) is always changing. Often policy lags behind consensus, or a policy may have too few editors for proper vetting in order to match consensus, or the editors discussing the policy may not be a representative sample of all Wikipedians involved with the topic.

Reiterating, policy is intended as a representation of consensus, consensus being the real basis for deciding what actions are or are not appropriate.

Rules and enforcement

Policy is the closest equivalent to a rule set on Wikipedia. For the sake of discussion, let's assume they're the same thing. Having rules implies that someone will enforce them. On Wikipedia, all editors are tasked with enforcing rules to an extent. Administrators are the only ones who can mete out blocks, but other editors bring rule violations to the attention of administrators. Also, a ban as opposed to a block generally requires consensus. Most people don't care enough or have the time to enforce every rule all the time. People tasked with enforcing rules invoke those they believe to be beneficial more often than rules they view as unnecessary. A real-world example of this is a cop who hesitates to write speeding tickets. This effect is magnified on Wikipedia because editing is not an obligation; an editor can not be admonished as a policeman would for ignoring an obvious rule violation because no one can even know that the editor saw the violation.

This implies that many actions which technically violate the rules occur without any attempt at enforcement, even when the actions are observed by many editors.

Disagreements

Of course disagreements often occur when an editor breaks a rule. The controversy surrounding IAR primarily relates to its involvement in disagreements. Typically, one editor will call out another editor's action as violating some rule. The accused editor will then invoke IAR, claiming that the action was beneficial to the encyclopedia even if it broke a rule. The editor may or may not have expected others to agree when he or she violated the rule.

In the disagreement, each editor is hoping that consensus will decide that his or her side is correct. Even if policy clearly favors one side, a controversial invocation of IAR elicits much discussion. In some cases, consensus will determine that the editor invoking IAR acted in the best interest of the encyclopedia despite the violation of policy. Therefore, an invocation of IAR is an assertion that policy is not in agreement with current consensus. It essentially forces a review of the policy as applied in that particular case. If it is decided to ignore the policy, then that consensus may be cited in support changing the policy.

As a consequence, consensus leading to formation of policies and guidelines is not limited to the discussion page for the particular policy or guideline in question. It is instead "distributed" around different discussion pages.

It follows IAR is not something to fall back on when consensus is not going your way. All IAR does is launch the discussion.

Summary

1. Consensus is true policy; Wikipedia's policies are intended to represent consensus.
2. Policies are necessary to avoid repetitive consensus-gathering.
3. Editors preferentially enforce rules they believe are beneficial or of high importance.
4. As a consequence, rules are "broken" all the time without consequence.
5. IAR is invoked in a disagreement when users believe policy does not represent concensus; it is a "policy review" of sorts.
6. IAR does not trump consensus.