Wikipedia:Navigating conflict

Encountering conflict when editing Wikipedia is just as inevitable as encountering storms when sailing. Sail for long enough, and it will happen. It seems appropriate to describe a few practical skills for navigating these storms.

De-escalation
When you're in a conflict with another person, there's an adversarial relationship established. Anything that reinforces that relationship tends to prolong a conflict. Thus, it's great to weaken that dynamic. There are various good ways to do that. You can probably invent many examples; here are two:
 * "I can see that you want the article to be well-sourced and reliable. Let's see how we can make that happen." - Recognize their good-faith, and highlight your common purpose. (If they're adding rather than deleting content, "well-sourced and reliable" might change into "accurate and informative".) You can follow this immediately with a serious editing question - make it concrete and direct - and if you take a professional tone, they're likely to rise to it, after you've just affirmed and complimented their motives. Now you're both back on track.
 * "Thank you for digging up those links. I'll need a few minutes (hours, days) to read them now." - Posts that say nothing more than "Thanks" or "I'm sorry" or "That was a good edit" can make a real difference, especially cumulatively, over time.

Each of us writes in our own voice. Thus, you might be someone who would not phrase those comments as above, and that's fine. There is still some way of getting the same messages across in your voice. Don't say it if you can't say it with sincerity.

You can dignify another editor as a human without compromising your commitment to encyclopedic principles. In fact, doing so is a good way to make it clear to any reader that you recognize the distinction between the editor and the edits.

Note: Some editors don't respond well to this technique: if you're too nice, they don't trust you. You've got to play it by ear.

Write for the other guy
See: Writing for the enemy. Put "his" facts in the article, neutrally. Represent his perspective better than he does. It's difficult, but if you can do it to the point of ending a conflict, you win on multiple levels.

Revert piecewise
Boring but useful: If someone reverts several changes you've made, or makes several changes that you want to revert, don't go back to your last preferred version. Break it into pieces. Make one small edit - a single source, or one sentence. Wait a day. Repeat. This can pare a conflict down to the heart of the matter, and it lowers the heat generated by rapid back-and-forth edits. Slow but steady "wins the race".

Remember that edits are not protected by default. If someone questions content in an article, they may remove it, and it should not be re-added until any questions about it are resolved. The questions may prove to be groundless, or not well-based in policy or in good sense, but this should be clearly established on the talk page for all to see.

Widen scope of discussion
Don't argue with one person on one page. Once you've gone three rounds, you can tell whether you're getting anywhere. If not, post a note somewhere relevant and ask what others think. The more people in the room, the lower the heat, on average. When you ask the question, phrase it as neutrally as you can. RFC is one way to do this, but you can do it informally as well.

Many talk pages have banners at the top linking to one or more WikiProjects; these are good places to start. You may also want to try talk pages of relevant policies and guidelines. Word to the wise: some project space talk pages are dominated by one of several conflicting Wikipedia philosophies. It is therefore sometimes desirable to ask in more than one place. If you do this, it is considered good form to indicate in each venue that you are cross-posting to multiple venues.

Just the facts
Stay super-focused on edits; don't mention motives. When someone is editing tendentiously, it's very easy to want to "call them out" on it. That's a bad idea, because then you end up arguing about that, and you're one step further removed from improving an article. Also, the less you are willing to descend to any kind of ad hominem, the better you look to outsiders. (That may sound cynical and Macchiavellian, but it's true.)

Eventually, it doesn't matter if the person at the other end of the dispute is Hitler or Gandhi - the edits have to stand or fall on their own merits. We might as well refuse to talk about anything else. It doesn't matter whether the other guy is editing with an intentional or unintentional bias, and we'll never be able to tell the difference. So, don't say he's biased, don't say he's lying, don't say he's acting like a jerk. These facts will speak for themselves. Just talk about the edits, and the sources, and any relevant policies and guidelines. You can float above a lot of friction that way.

Behavior policies (and guidelines)
Don't cite WP:CIVIL (Civility policy), WP:AGF (Assume good faith), or WP:NPA (No personal attacks) at someone, unless you're an utterly uninvolved third party. (Even then... gently.) Telling someone that they've just "violated" a behavior policy tends to exacerbate conflict, and to generate heat. The point is not to expose wickedness or to see justice served; it's to get back to editing.

Instead, say something like, "Hey, what you just said offended me." Many editors, confronted with this, will say, "I'm sorry, I didn't mean to. What was offensive about it?" If someone says you have offended them, then "How so? I didn't mean to," is an excellent (and hopefully honest) reply.

Telling someone that you feel offended gets to the point, whereas citing policy tends to provoke a defensive reaction: "I did not attack you! How dare you accuse me?" If you tell someone that you were upset by what they said, and they don't react well, then you want to go back to "Widen scope", while remembering, "Just the facts". The behavior will speak for itself, and people accuse themselves far more effectively than any opponent can accuse them.

The numbers in the list below indicate how many steps removed a topic is from the actual encyclopedia. The idea is to keep the number small:


 * 0. Editing an article
 * 1) Talking about another person's edit
 * 2) Talking about another person
 * 3) Talking about whether you violated WP:NPA by talking about the other person
 * 4) Talking about whether the other person violated WP:DTTR (don't template the regulars) by warning you of violating WP:NPA
 * 5) Talking about the fact that WP:NPA is a policy, while WP:DTTR is at best a guideline.

Content policies (and guidelines)
Citation of content-related policy is somewhat different. It's certainly more encyclopedia-focused to say that you think an edit violates WP:NPOV (neutral point of view) than it is to say that a post violates WP:NPA (no personal attacks). However, remember that even the content related policies and guidelines are not laws, and are only as strong as the reasons behind them.

Asking whether a particular policy or guideline should apply in a given situation is appropriate, if it seems to be questionable. It is sometimes suggested that if someone doesn't like a guideline, then they should get the guideline changed first, and then come back to the specific article. This is backwards. Consensus is developed at the article level, and later abstracted to the guideline.