User:Hob Gadling/Admit mistakes

Admit mistakes
Admitting and retracting mistakes is one of the basic skills in discussions. As far as I know, there is no Wikipedia rule about it, so, here is at least an essay.

In discussions, this happens again and again.
 * User A makes a mistake.
 * User B says that is a mistake, explains why, and says a few other things.
 * A responds to the other things, ignoring the correction.

The above pattern is very familiar to me because I am a skeptic, and I have a lot of experience with discussing creationists. They never admit mistakes. They never correct mistakes. They never retract false statements. That has a reason: Creationism is based on falsehoods. If a creationist seriously started to correct mistakes, all their arguments would crumble away, and finally, their worldview would break down.

More generally, never admitting mistakes is a crucial precondition for all obviously false worldviews (OFWs). I qualify this with the word "obvious" because, well, obviously, if a worldview is false but not obviously false, there will be very few or no people acting as B in the conversation above.

A proponent of an OFW who corrects mistakes will not stay a proponent for long, so all long-standing proponents of OFWs are people who do not admit mistakes. They could theoretically admit small ones, but I don't think I have ever seen even that.

Someone who cannot admit mistakes still carries with them all the mistakes they ever made in their life. Such people in a position of power are a disaster. I seem to remember that happening a few years ago, somewhere. Should we have a word for someone who cannot admit mistakes and instead, say, attacks the one who corrected them? "Donald" sounds like a good word for it at the moment, for some reason. But on second thought, I don't think so, since making it a noun - "you are a donald" - would make it sound like a personal attack, or, even worse, a comparison with a real person named "Donald". A verb is better, since it denotes a behaviour pattern instead of a character trait.

So, donalding is now defined as dodging and/or attacking when someone caught you making a mistake.

Application to Wikipedia discussions
Admitting mistakes improves discussion culture by demonstrating the existence of a common ground. Routine donalding is a worse cause for a WP:BATTLEFIELD atmosphere than any others. We are looking for truth here. Well, verifiability, actually, but truthfulness is needed for that.

We do not want OFWs propagated as true in Wikipedia, and if donalding is generally frowned upon, the support for OFWs will naturally dry up. Down with donalding!

Different cultures
The goal of science is to find out what is true and what is not. It is based on honesty, and for scientists, honesty is very important. To a scientist, not acknowledging and correcting a mistake in a discussion is the ultimate incivility.

The culture of lawyers is very different. It is a lawyer's job to convince people of a given position regardless of the truth or otherwise of that position. The goal is to win. It immediately follows that honesty is a handicap in that profession. For a lawyer, acknowledging and correcting mistakes in a discussion is probably unfamiliar territory and goes against everything they have learned.

The Wikipedia culture is similar to the scientific one. There are rules against the lawyer mindset: WP:POV, WP:WIKILAWYER, WP:BATTLEGROUND.

So, if you are a lawyer editing Wikipedia, or if you think like one, you should try to overcome that and learn to admit mistakes, otherwise editors who value honesty and the spirit of WP:NPOV will always distrust you. Wikipedia is a cooperative project. If your goal is to win a discussion no matter what - if you behave like a lawyer in that respect - you are not being cooperative, and you are probably better off visiting other websites instead.

Examples
This essay does not apply to cases where there can be different opinions. Here are a few examples.
 * 1) Misquotation
 * "D" is some arbitrary person or website, maybe a reliable or unreliable source outside Wikipedia, maybe a Wikipedia article, maybe an editor on a Talk page
 * D says "X".
 * User A believes "Y", for whatever reason. Y can be true or false, it does not matter.
 * A deduces that from X and Y, "Z" follows.
 * A writes, "D said Z".
 * This is a very clear case of a falsehood. A should instead write, "D said X. If D is right and my belief Y is right, it follows that Z." Keep the facts and your opinion separate. Retract your false statement.
 * 1) Idiosyncratic definition
 * User A says, "X is defined as Y".
 * User B cites reliable sources, such as dictionaries, saying that X is defined as Z, explicitly excluding Y.
 * A cannot find another reliable source saying "X is defined as Y".
 * This is a very clear case of a falsehood. You need to be on the same page as others speaking the same language, not invent your own definitions of words. Retract your false statement.

Counters
There will be opposition to this, and here are a few possible counters.
 * Maybe B did not really refute what A said.
 * Then A should not ignore the refutation completely, but refute the refutation. When A ignores the refutation, the default assumption has to be that B us right.
 * If that is the default assumption, then A does not need to admit it. Everybody knows it anyway.
 * No. If A accepts that their statement was wrong, A should retract it. Otherwise, it can later be revived by linking to it. Also, B will not know whether A has donalded or implicitly accepted the correction.
 * You are saying that when someone does not admit a mistake, they are holding an obviously false worldview!
 * No, I am not. Donalding is a necessary condition for holding an OFW, but not a sufficient one. You can be essentially correct in all your basic assumptions and still donald away minor corrections.
 * How do you know that those allegedly "obviously false worldviews", like creationism, are really false?
 * Because all their reasoning has been refuted. If you disagree, give me one single piece of reasoning that has not. A Gish gallop is invalid, it needs to be one single piece of reasoning, because of Brandolini's law.
 * What if A does not respond at all?
 * If you never hear from A again, then it could be a silent form of donalding, or it could be that A has seen the error of their ways. You simply don't know. But at least they won't bother you again.
 * If I admit a mistake, I will seem weak!
 * Admitting mistakes is not a weakness, it is a strength. It shows that you can afford it, that your worldview will survive corrections without a problem. It is very easy for you to retract your false statement or to amend it to make it true, if it does not hurt your argument. If it does hurt your argument, then your whole argument is based on that very falsehood and does not work. Retract the whole thing.
 * Admitting mistakes takes strength? But not everybody has it. You cannot demand that people overcome their weaknesses.
 * I do not demand it. But if you don't have it in you to stand by your own fallibility, you will be handicapped in discussions because as soon as people notice your weakness, they will mercilessly exploit it by pointing out your earlier mistakes again and again. Onlookers will see that your reasoning is fake, and you will lose the discussions. (Of course, if you admit being wrong, you will lose too, but without the humiliation.)
 * But my worldview is based on many dubious arguments at the moment! Better reasoning will be available soon, when the results of the ongoing research are in! I cannot admit mistakes now, or it may tumble down around me now when there is hope for real evidence!
 * Congratulations, you made the first step. The next step: Try to accept an idea after the evidence for it is in, not before.
 * You are proposing "down with donalding" as a panacea against OFWs. There are no panaceas.
 * No, it is not intended as a panacea. It is just one more tool in the toolbox.
 * One can have different opinions. Maybe what A said is not "false" but just a different take on the situation.
 * In such cases, this essay does not apply. This is about obvious falsehoods only. See the "Examples" section.