[Previous] How To Ask Questions 2 | Home | [Next] How To Ask Questions 4

How To Ask Questions 3

Caeli: Hey!
Elliot: Hi, Caeli.
Caeli: Last time we talked was fun. I'm glad to be back.
Elliot: That's excellent.
Caeli: So, where were we? Oh yeah. We were talking about persuasion. I don't think you clearly said if persuasion always works, if you are right.
Elliot: It doesn't, but that's OK. Keep in mind that the more important the situation is -- the more critical the error you wish to correct -- the easier persuasion is. So when persuasion fails, we're usually talking about cases where nothing too big is at stake.
Caeli: OK, but even then wouldn't it be better if we got the right answer? If persuasion won't work, maybe we can get it another way.
Elliot: There is no such thing as a way of acting that always does the right answer. It's not possible to have a system that makes mistakes impossible. What we should look for are policies that help spread good ideas to everyone, and help prevent mistakes from spreading, and help eliminate mistakes within ourselves.
Caeli: That sounds wise. OK, how do we do that?
Elliot: Let's compare persuasion and listening to people who feel completely sure that they are right. If someone is sure he's right, and everyone is required to listen to him, good ideas will have an easy time spreading. Someone just has to think of one, and be confident that it's good. Unfortunately, bad ideas will also spread easily. Every time someone mistakes a bad idea for a good one, that will spread just as easily as a good idea. And there aren't any mechanisms for correcting errors built into this system, so once they start to spread, there's nothing to stop them.
Caeli: What about with persuasion?
Elliot: Using persuasion, good ideas will often spread, and they will spread fairly effectively. Bad ideas, on the other hand, will have an uphill battle. Every step of the way, people will challenge them and criticise them. And if someone comes up with a powerful criticism of a bad idea, that itself is a good idea, and a persuasive one that many people will be interested in, so it could spread and cause the elimination of error. Further, if I try to persuade someone of my idea, he may end up persuading me that I was wrong, or I may realise I'm wrong while examining my own idea. So there are multiple levels of error correction.
Caeli: If someone's really sure he has a good idea, isn't that important? If we take that into account, maybe we can find and spread good ideas faster, while still thinking for ourselves some too.
Elliot: It does matter. There are a lot of steps people can take to get their ideas heard. For example, they can write and publish a book. If someone cares enough to do that, more people will be exposed to his ideas. Or he can go on TV, or give lectures. People can put effort into advocating their ideas proportional to how sure they are that it's true and valuable.
Caeli: Oh, that's cool. What if the author of a good idea is really busy, though?
Elliot: If he's too busy to ever tell anyone, then no possible system could spread his idea. But if he does tell some people, they are free to advocate his idea for him, with as much passion as they think is fitting.
Caeli: Are there any of your ideas that you'd like me to advocate for you?
Elliot: That would be nice. But I don't want to say which ones. Just pick whichever you find most interesting or important, or whichever ones come up frequently in your life.
Caeli: OK, I will!
Elliot: I will do the same for you, of course.
Caeli: But I haven't said any ideas.
Elliot: You've said some, but also your questions contain ideas in them. What I meant is that I learn things from you, and I won't hesitate to pass them on when good chances present themselves.
Caeli: Oh, thanks :)
Caeli: Will you tell people that the ideas came from me?
Elliot: Probably not. It's hard to keep track of where my ideas come from, and it's not very important anyway. We should judge ideas based on their merits, not their author.
Caeli: But I want to get credit, so people know I have good ideas.
Elliot: Don't worry about that. Anyone who talks to you will instantly see that you are bright. And if he doesn't, he's silly, so don't think of him.
Caeli: OK, I guess. Maybe I'll come back to this later. What I really want to know about today is you said parents act cruelly.
Elliot: That's right. There are a lot of well known things parents do or say which are cruel. Consider: "You'll understand when you're older", "Do what I say, or else", "Eat your vegetables", "Go to your room", "You can come back when you're ready to apologize", "Because I'm your father, and I said so".
Elliot: And then there's ideas like that children need limits and boundaries. Which are only meaningful and controversial because they mean limits and boundaries that children don't want. And there's the ideas of compromises, discipline, obedience, spoiling children, that "you can't always get what you want", and that a few dollars a week is plenty of money.
Caeli: Wow, that's a lot of stuff. I see why some of them are bad, like "I said so" isn't a good reason. But what about being a father? Don't parents need to be able to make some decisions for the family?
Elliot: As we've discussed earlier, the more critical the case, the easier persuasion is. If a parent can honestly say that something is very important, but for some reason, such as time pressure, he isn't able to explain things to the child now, then won't children voluntarily go along with it?
Caeli: Don't parents try that a lot, and their children don't listen?
Elliot: Yes, but I think you're proving my point. That situation doesn't happen very often. If a parent uses it frivolously, his children may notice and distrust him in the future.
Caeli: What about if they're at a restaurant, and the child is disturbing the other customers. I think that's pretty common.
Elliot: Yes, but it's nothing like the kind of emergency I was thinking of. What's the worst that can happen? You're asked to leave the restaurant. That's not very bad. It's nothing worth damaging your relationship with your family over.
Caeli: Wouldn't it be better if the child calmed down long enough for the parent to explain, so they wouldn't get kicked out?
Elliot: Yes, it would. And that can certainly happen. The parent could say, "Please stop. I think you're making a mistake, and I want to tell you why, but first it's very urgent that you lower your voice and stop throwing things." The child will get an explanation right away if he stops, so he doesn't have anything to lose. He doesn't have to take his parent's advice on faith for more than a few minutes. And once he does this, he'll have a better idea whether to do it again in the future. I think children who won't calm down for a little while to talk have almost always tried this many times in the past, and it didn't go well.
Caeli: How would it go badly?
Elliot: Well, the parent might say, "Great, now you're calm. So, you can't act like that in restaurants. You have to be polite to the other people, and it hurts me when you act like an animal."
Caeli: Wow, that's terrible. I wouldn't want any advice at all from someone who talked like that.
Elliot: Yeah, it's unpleasant. It says the child can't have what he wants, but it doesn't explain why in any detail, and it surely doesn't explain why the suggested way of life is nice and enjoyable. And it's manipulative. The parent has chosen to be hurt by behavior he doesn't like as a way to suppress it.
Elliot: By the way, there's a very important fact we haven't yet considered. It is that a parent does not have to take his child to a formal restaurant before explaining what sort of behavior is expected there. It's quite irresponsible to go to one without giving the child any warning. On the other hand, if the child knows what's happening in advance, and has chosen that he does want to go to the restaurant, then the only things that will stop him from acting with great decorum are either if he doesn't know how to, or he changes his mind.
Caeli: So, then what?
Elliot: If he changed his mind about the restaurant visit, perhaps you should leave. Oh well, but really not a very big deal. And if he doesn't understand decorum, despite the lessons he had before coming, that is almost certainly a very small problem. Just remind him, or tell him the parts he doesn't know. If he's truly interested in trying to act appropriately for a formal restaurant, that is, if he does want to be there, then he will be happy to get advice about how to do it better.
Caeli: I see. I guess most of the problems come when the child doesn't really want to be there, or doesn't want to behave.
Elliot: That happens a lot, yes. Another issue is that parents overreact. I've seen parents discipline their kids because they thought the child was bothering me, even though I said he wasn't. The parent refused to believe me, and thought I was just being polite.
Caeli: That's a shame. Why are parents to eager to be kind and helpful to strangers, but not their own children?
Elliot: Our culture values treating strangers with care, and being helpful to them as appropriate. And it's right to do that. It's just that it is also right to be good to our children.
Caeli: If treating strangers well is valuable, then isn't the parent being helpful by making his child do it?
Elliot: I'm sure he's trying to be helpful. But this gets back to using force or persuasion. There's no need, and no justification, to threaten a child if he has made a mistake out of ignorance. Won't he be happy to be told about a very good value that is present in the world, which he can enjoy?
Caeli: Oh, it sounds so much nicer when you put it that way.
Elliot: Indeed. What's going on frequently is that the child doesn't like the parent's advice, because it isn't persuasive, and doesn't seem to be good or nice.
Caeli: If most parents are bad at persuasion, even when they are right, is it understandable if they use force instead sometimes?
Elliot: Well, are they making a large effort to become better at persuasion? Any effort at all? I don't think they can be forgiven if they aren't trying.
Caeli: What if they don't know that persuasion is better? That's just ignorance, so can't we forgive it?
Elliot: Our society values freedom, and voluntary association, and not being forced to do things. Everyone in our culture knows this. If they decide it somehow doesn't apply to children, they are arbitrarily restricting the reach of one of our values to exclude people. It's well known that you shouldn't do this to other groups like blacks or women. But, yes, it's a parochial error and the real issue shouldn't be forgiveness, it should be how to help our society move past this blindness.
Caeli: That's very noble. What do you think would help to remedy this blight?
Elliot: Maybe writing dialogs.
Caeli: Do you like self-reference?
Elliot: Yes :)
Caeli: What are some ways persuasion can go wrong even though you're right?
Elliot: Persuading another person is a matter of communicating your idea and its merits, and discovering rival theories the other person holds, and communicating criticisms of those. Fortunately, the other person will often be helpful and refute some rival theories himself.
Elliot: So, successful persuasion isn't just about being right. It's also about being able to communicate with this person, and finding out about other ideas he has which are relevant, and responding to them. All those steps can go wrong even if your main idea is true.
Caeli: You've mentioned a few times that communication is hard. That goes against common sense. People hang out to talk all the time, and often use this to relax, and find it easy.
Elliot: How hard it is depends on what you want to say, and your culture. In our culture, some communication is common and easy, because everyone has knowledge to facilitate it. But that's cheating, in a sense, because it doesn't involve much knowledge getting from one person to another, it involves both people already having shared knowledge.
Caeli: OK, so tell me about the case when they don't already share an idea.
Elliot: If someone doesn't understand my idea already, the conventional theory is that I can just tell him, and then he will. But that doesn't make sense. He doesn't know what it is. I can say words that I think can be translated into the idea, but he will only be able to guess at the correct method of translating words to idea, because he doesn't know what he's supposed to end up with.
Elliot: By the way, far and away the best reference on this topic is the book Godel, Escher, Bach by Douglas Hofstadter. See chapter 6 on The Location of Meaning.
Caeli: Can you give a brief summary, now?
Elliot: Sure. OK, imagine sending messages to space aliens. We have to put the message in a language, say English. And we have to do something to indicate that this is a message, and not just random junk, so that they notice and try to understand it. And we can include some hints about how to decode the main idea, that we think might be helpful.
Caeli: Could we just include a dictionary?
Elliot: Yes, and that might be helpful, but unfortunately they won't know how to read it, when they start.
Caeli: Oh, well how will they ever get started?
Elliot: They can look for patterns, and they can make guesses about what things mean, and then try applying the guesses to other parts of the text, and if the guess makes sense in multiple places, then we know the guess has reach and that's a sign it's good and worth trying in more places.
Caeli: This idea of reach seems to have a lot of reach.
Elliot: Yeah. Now consider when we try to say something to another person. We face all the same problems as with a space alien, except for one difference. The difference is that we already have shared knowledge. In fact, we have a lot, including the whole English dictionary. So that makes it a lot easier, because whenever we want to express a new idea, we can say something that's partly new and partly old. Then it's easier for people to get started decoding it. It's like filling in the blanks, instead of guessing the entire thing at once.
Caeli: So this relates to how you were saying communication is easy when you already share knowledge of what you're saying, but hard when you don't already share the knowledge you want to communicate?
Elliot: Yes. Communicating new ideas to a person is exactly the same kind of problem as communicating them to a space alien. It's easier because we have more shared knowledge to start with. But as many people have pointed out, we can expect to have shared knowledge with space aliens too. They will have physicists and mathematicians, and know about logic and morality, just like we do. Communication is hard in both cases because it's hard to guess what idea someone has when you don't already know it.
Elliot: Now, there's a very important fact I haven't mentioned yet. It is that a baby is just like a space alien. What I mean is that he has very little shared knowledge with other people. So communicating to him is very difficult. And that he ever understands anything is amazing. Babies don't have a whole civilisation with foreign language specialists, physicists, mathematicians, and so on, to translate messages. They only have their own brain, and their extreme ignorance.
Caeli: But babies learn language, and lots of stuff. It doesn't seem like a problem.
Elliot: That happens for a few reasons. The first is that babies have fully functional brains. They are very creative. Otherwise learning human language would not be possible. The second is that our culture has evolved traditions about raising children. What that means is that over time ways to raise kids that work less well have been eliminated, and ways that work better have been found. I don't mean better for everything, but for specific issues like raising a child who can talk, we've evolved to be very good at it. (Of course, raising a child to talk intelligently is another matter, and many people would agree with me that there's room for improvement there.)
Caeli: Are there more reasons?
Elliot: Our children are immersed in our culture. There are people talking all the time. They don't have just one message to decode. There are thousands, and if they only decode every hundredth message, that will be fine. It's easier to find patterns in such a big data set.
Caeli: That's cool. So, I read that scientists have shown that young children don't have mature brains yet, not like adults.
Elliot: Those aren't scientists, they are psychologists and "social scientists". And they are not very interesting, so I'd prefer not to talk about them much. Maybe another time. For now, let me just say: if children don't have very functional brains, what is the explanation for how they learn so much?
Caeli: I don't know.
Elliot: Indeed. And neither do those "scientists" that you mentioned.
Caeli: Actually I better go now. But I'll prepare a few questions for next time.
Elliot: OK, see you.

Elliot Temple on October 22, 2006

Messages

Want to discuss this? Join my forum.

(Due to multi-year, sustained harassment from David Deutsch and his fans, commenting here requires an account. Accounts are not publicly available. Discussion info.)