LINK: In favor of niceness, community, and civilisation

A part which seems missing in the discourse -- probably because of politeness or strategy -- is that there are more than two sides, and that people on your side don't necessarily share all your values. When someone tells you: "Harry, look how rational I am; now do the rational thing and follow me in my quest to maximize my utility function!" it may be appropriate to respond: "Professor Quirrell, I have no doubts about your superb rationalist skills, but I'd rather use my own strategy to maximize my utility function." Your partner doesn't have to be literally Voldemort; mere corrupted hardware will do the job.

On the battlefield, some people share the common goal, and some people just enjoy fighting. Attacking the enemy makes both of them happy, but not for the same reasons. The latter will always advocate violence as the best strategy for reaching the goal. (The same thing happens on the other side, too.)

And an imporant part of the civilizing process Scott described is recognizing that both your side and the other side are in a constant risk of being hijacked by people who derive their benefits from fighting itself, and who may actually be more similar to their counterparts than they are to you. And that miraculous behavior which shouldn't happen and seems like a losing strategy, is actually the civilized people from the both sides half-knowingly forging a fragile treaty with each other against their militant allies and leaders.

Which feels like a treason... because it is! It is recognizing that there is some important value other than the official axis of the conflict, and that this value should be preserved, sometimes even at cost of some losses in the battlefield! -- This is what it means to have more than one value in your utility function. If you are not willing to sacrifice even epsilon of one value to a huge amount of the other value, then the other value simply does not exist in your utility function.

So, officially there is a battle between X and Y, and secretly there is a battle between X1 and X2 (and Y1 and Y2 on the other side). And people from X1 and X2 keep rationalizing about why their approach is the best strategy for the true victory of X against Y (and vice versa on the other side).

Civilization is a tacit conspiracy of decent people against psychopaths and otherwise defective or corrupted people. Whenever we try to make it explicit, it's too easy for someone to come and start yelling that X is the side of all decent people, and Y is the side of psychopaths, and this is why we from X have to fight dirty, silence the heretics in our own ranks and then crush the opponents. So we stay quiet amidst the yelling, and then we ignore it and secretly do the right thing; hoping that the part of conspiracy on the other side is still alive and ready to reciprocate. Sometimes it works, sometimes it doesn't; but on average we seem to be winning. And I wouldn't trade it for a "rationalist" pat on shoulder from someone I don't trust.

The problem with Yvain's reply is that he omits the main reason why lying is a bad idea. Yvain compares lying to violence. I don't think this is a good comparison. It's acceptable to respond to violence with violence. It's not a good idea to respond to lies with lies.

Eliezer touched on this issue in his post here where he pointed out that one problem with lying to support a cause is that you'd better be absolutely sure that all your beliefs about the cause and what to do for it are in fact correct. However, the problem is even worse, there is a vicious cycle here since a cause that frequently lies is much more likely to acquire incorrect beliefs.

Think about it this way: suppose you believe that your cause justifies lying, so you lie about it. Your lies attract people to your cause who believe those lies. They in turn make up further lies (that they think are justified based on the lies they believe to be true). And so no until your cause's belief system is full of falsehoods and anti-epistomology. Your cause may ultimately "win" in the sense that it's followers acquire power, but by that point said followers may no longer care about your original goal. Even if they do, they're likely to have so many false beliefs that what they do to accomplish it is likely to be counter-productive and probably have other unpleasant side effects.

Note that the above argument applies to lying but not to violence. Thus in some sense lying for your cause is in fact worse than committing violence for it.

I'm with Scott. It's so natural to think that if your enemies are as ruthless as the Tsars and their goons, you need to be as ruthless as the Bolsheviks to fight them. But we all know how that worked out, and it hardly seems to be an outlier; rather, it seems to be the norm for those willing to sink to their opponents' level. If the goal is victory for our cause, and not just victory for some people who find it convenient to claim to be cheerleaders for our cause, we need to be very careful that our tactics are not training up Stalins within our ranks. Not that I'm advocating total purity at all times and in all respects, but I think before playing dirty you need to make sure you have a much better reason to think it's a good idea than "the other guys are doing it."

I'm not entirely convinced that the relationship between crafting a rational argument and crafting a persuasive argument is nearly as inverse-correlational as implied. On average, lies have a higher manufacturing cost (because you have to tread carefully and be more creative), a greater risk (since getting caught will lower your overall persuasiveness), and a smaller qualitative gain (while lies probably persuade more people, I suspect that they persuade less rationalists than civil debate and are therefore less qualitative overall). There are other means of persuading people without making deliberately irrational arguments. If sound reasoning alone isn't tasteful enough for you, why not season your truth with charm instead of coating it in sophistry? Why not leverage charisma or cordiality? You know - the dark art of sucking up?

While fear is often heralded in psychological communities as the most effective mechanism of persuasion, that doesn't mean it's the mechanism of persuasion with the greatest utility. A well-beaten child might obey best, but obedience isn't the only goal of discipline - nor agreement the only goal of argumentation. Personally, I'd rather treat every worthy cause as an opportunity for non-rationalists to exercise rationality than as an excuse for rationalists to manipulate non-rationalists. This tactic might not win every argument now, but it lays a surer foundation on which to build our arguments in the future.

Whether or not the lawful-goods of the world like Yvain are right, they are common. There are tons of people who want to side with good causes, but who are repulsed by the dark side even when used in favor of those causes. Maybe they aren't playing to win, but you don't play to win by saying you hate them for for following their lawful code.

For many people, the lawful code of "I'm siding with the truth" comes before the good code of "I'm going to press whatever issue." When these people see a movement playing dirty, advocating arguments as soldiers, where you decide whether to argue against it based on whether it's for your side rather than whether it's a good argument, getting mad at people for pointing out bad arguments from their side, they begin to suspect that your side is not the "Side of Truth". So you lose potential recruits. And the real Sith lords, not the ones who are trying to use the dark side for good, will have much less trouble hijacking your movement with the lawful-goods and their annoying code and the social standards they impose gone.

Leaving aside the honor among foes idea, and the "what if you're really the villain" idea, if your cause is really just, then although the lawful-goods are less effective than you, their existence is good for you. Not everything they do is good, but on balance they are a positive influence. You're not going to convince them to attempt to be dark side users for good like you are attempting to be, so stop giving them reasons to dislike you.

Even if you can convince them, the lawful-evils who think they are lawful-goods are listening to your arguments. Most people think they are good. It is hard to tell when you're not good. So the idea that only truly good people are bound by the lawful code is crazy. Lots of lawful evil is an unintentional corruption of lawful good, and this corruption doesn't unilaterally affect your goodness and your lawfulness. They could tell (or at least convince themselves) they weren't really good, if they didn't follow the lawful code, because they think like lawful good people in that respect. The lawful evil people who see you, and know you are opposed to them on the good/evil axis, think they see evil people saying "Forget this honor among enemies thing. We have no honor. Watch me put on this 'I am defectbot' shirt". And that is a much stronger argument to abandon the lawful code of rational argument and become the much more dangerous chaotic evil than what the lawful-goods hear, which is their chaotic good allies telling them to defect.

But in real modern human politics, it's more complicated because although there is one lawful/chaotic axis, there are many good/evil axes. Because there are many separate issues that people can get right or wrong. Arthur Chu thinks that the issue of overriding importance is social justice. So he demands that we drop all cooperation with people who are evil on that axis. He says we aren't playing to win. I can think of 3 issues (2 of them are actually broad categories of issues) that I am confident are more important than social justice, and which are easier to improve than the problems social justice wants to counter. In order of decreasing importance, existential risk, near-term animal suffering including factory farming and wild animals, and mortality/aging.

In real life, you don't demand that your allies be on the same end of every good/evil axis as you. That is not playing to win. A better strategy (and the one Chu is employing) is to pick the most important axis, and try and form a coalition based on that axis. Chu accuses LW of not playing to win, well, I'm just not playing to win along the social justice axis at the cost of everything else. I think different axes are more important.

And there's also the fact that for some causes, "lawful" people (people who play by the rules of rational discourse) are much better to have as allies. If we use bad statistics and dark arts to convince the masses to fund FAI research, they may as well fund Novamente as MIRI. Not all causes can benefit from irrational masses. Something like MIRI can't afford to even take one step down the path to the dark side. When you want to convince academics and experts of your cause, they will smell the dark arts on you and conclude you are a cult. And with the people you will attract by using dark arts, your organization will soon become one. The kind of people who you absolutely need to do object-level work for you are the kind of people who will never join you if you use the dark arts.

If you take a pluralistic "which axes are important" approach instead of the one that Chu takes, then there is a lot to be said for lawfulness, because it tends to promote goodness*, a little. And when get a bunch of lawful-goods and lawful-evils together and you nudge them all a little toward good through rational discussion (on different axes), that is pretty valuable. Because almost everyone is evil on at least one axis. And such a community needs a policy like "we ask that you be lawful,[follow standards of rational discourse] not that you are good [have gotten object-level questions of policy right]," because it is the only defensible Schelling point.

*If you haven't caught on to how I'm using "law vs chaos" and "good vs evil" axes here by now, this may sound like moral realism, but when I mean by "law" is upholding Yvain-style standards of discourse. What I mean by "good" is not just being moral, but being moral and right given that morality about questions of ethics.

Chu's position -- at least, as presented at Yvain's blog -- seems to dip into the realm of being a guardian of truth. To me, that position is always scary... even if it comes from the "good guys".

I'd be interested in people's opinions of how dire the real world consequences have to be before it's worthwhile debating dishonestly

I, for one, have the impression me that the more dire the consequences, the more important honesty in arguments becomes. So, I don't really get your dilemma.

Scott's examples have a fair amount of selection bias. If you take Chile, Russia, NK, or Zimbabwe, those who play dirty prevail. However, I agree that building a walled garden and making it attractive to join is a far better strategy whenever feasible.

It is rather amusing to treat this exchange as a debate on consequentialism.

Nice. This is another confirmation of something that's becoming increasing apparent to me, and raises the same issue I've been thinking about.

I'm of the rationalist libertarian persuasion. We value truth, honesty, and a lack of coercion in human interaction. When you argue, you argue honestly. You don't lie, you admit when the other side scores points, etc. Politically, you respect the freedom and equal rights of others, and don't use force to violate those rights. But we live in a world of people who do not share those values. By our lights, these people are engaged in war against us. Yet we don't respond in kind. We're in a war, we're being shot at, but we never shoot back.

For all the talk of guns and self defense, rationalist libertarians are basically social pacifists. You could add the nerd persuasion to the list of pacifisms - in this case pacifism in the near politics of negative sum social status wars.

All this pacifism, from people who generally think it's moronic when actual bullets are involved. It's rather peculiar.

Whatever we might think of the best way for humans to interact, surely that is contextual, and the fact that another person is shooting at you should count as a relevant part of that context. If we never make such people pay a price in violating what we consider the ceasefire of the war of all against all, why would we ever expect them to stop?

Libertarians in the US are fond of the quote (variously attributed) that "The price of liberty is eternal vigilance". Vigilance is just the start. The true price is making violators of that liberty pay.

It always makes me happy when my ideological opponents come out and say eloquently and openly what I’ve always secretly suspected them of believing.

It's nice to get those occasional windows into an alien mind. Probably worthwhile to go to various ideological walled gardens on the internet disguised as a native.

Scott just responded here, with a number of points relevant to the topic of rationalist communities.

I would assume there was supposed to be a link there?

Obviously, at some point being polite in our arguments is silly.

I think you seldom convince someone to change his opinion by name calling.

I once went to a talk about the implications of neurology on economics. Unfortunately for the professor who gave the talk he had a badly dressed conspiracy theorist in his audience who was upset about the professor providing a new way to justify the economic status quo. That talk would have benefited from throwing out the conspiracy theorist instead of being nice to him. The reason isn't that the conspiracy theorist is dangerous but because wasted valuable time of a talk about neurology on economics.

Most situations however aren't like that. While I would have wanted the person thrown out of that talk I would happily talk with someone who holds the same position face to face or at a forum like LW.

I personally don't attempt to manipulate people against their will but if I would want to do so, dishonesty isn't the most straightforward way. If you interact with a person in that way you bring up resistance.

If you create rapport you can ask them whether they are happy with how their life is going and hit much deeper. That means both being able to do real psychical damage that goes deeper than a cheap insult and having a lever that's strong enough to change opinions.

I think that a conception of heroic morality (basically, whether or not to use TDT, or choosing between act and rule utilitarianism) may be at the heart of many of the choices to be cooperative/nice or not. Many people seem to assume that they should always play the hero, and those more virtuous ones who don't seem to think that you would never be able to play the hero.

As an example, consider assassinating Hitler. It's not clear how Hitler could reprise this -- he is already killing people who disagree with him, and he is a single tyrant while you are an invisible individual. This does not apply, however, if you are in equal factions, say Fascists and Communists.

I don't understand all the consequentialist arguments against playing dirty. If your only objections are practical, then you're open to subtle dirty maneuvers that have very high payoffs.

A really simple example of this would be to ignore articulate opponents and spend most of your energy publicly destroying the opposition's overzealous lowest-common-denominators. This is actually how most of politics works...

... and also how this conversation seems to be working, since the Scott Alexander side seems more intent on arguing through hyperbole than addressing the actual spirit of what is being suggested. A simple example could be to deliberately misinterpret what the other side is saying, and force them to clarify themselves ad nauseam until they run out of energy, "conceding" the issue by default.

AND JUST LIKE THAT WE'RE BACK TO THE IRONY

Just think about how much more persuasive fighting dirty sounds if the whole fate of the human race hangs in the balance. As is, there is an underlying assumption that we have infinite time to grind down our opposition with passive logical superiority.

"How dire [do] the real world consequences have to be before it's worthwhile debating dishonestly"?

M̶y̶ ̶b̶r̶i̶e̶f̶ ̶a̶n̶s̶w̶e̶r̶ ̶i̶s̶:̶

One lower bound is:

If the amount that rationality affects humanity and the universe is decreasing over the long term. (Note that if humanity is destroyed, the amount that rationality affects the universe probably decreases).

T̶h̶i̶s̶ ̶i̶s̶ ̶a̶l̶s̶o̶ ̶m̶y̶ ̶a̶n̶s̶w̶e̶r̶ ̶t̶o̶ ̶t̶h̶e̶ ̶q̶u̶e̶s̶t̶i̶o̶n̶ ̶"̶w̶h̶a̶t̶ ̶i̶s̶ ̶w̶i̶n̶n̶i̶n̶g̶ ̶f̶o̶r̶ ̶t̶h̶e̶ ̶r̶a̶t̶i̶o̶n̶a̶l̶i̶s̶t̶ ̶c̶o̶m̶m̶u̶n̶i̶t̶y̶"̶?̶

R̶a̶t̶i̶o̶n̶a̶l̶i̶t̶y̶ ̶i̶s̶ ̶w̶i̶n̶n̶i̶n̶g̶ ̶i̶f̶,̶ ̶o̶v̶e̶r̶ ̶t̶h̶e̶ ̶l̶o̶n̶g̶ ̶t̶e̶r̶m̶,̶ ̶r̶a̶t̶i̶o̶n̶a̶l̶i̶t̶y̶ ̶i̶n̶c̶r̶e̶a̶s̶i̶n̶g̶l̶y̶ ̶a̶f̶f̶e̶c̶t̶s̶ ̶h̶u̶m̶a̶n̶i̶t̶y̶ ̶a̶n̶d̶ ̶t̶h̶e̶ ̶u̶n̶i̶v̶e̶r̶s̶e̶.̶