cody-bryce comments on Rationality Quotes August 2013 - Less Wrong
You are viewing a comment permalink. View the original post to see all comments and the full post content.
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (733)
-Robert Downey Jr.
I think it's good to be well-calibrated.
It is usually best to be socially confident while making well-calibrated predictions of success. The two are only slightly related and Downey is definitely talking about the social kind of confidence.
Good point. I'm still not sure I like his framing of social interactions as getting people on "your" team (which I may be partly biased in by the source of the quote), but the objection in my initial post isn't a good one.
I think it's best to be well-calibrated, use that to choose your team as one that's going to succeed, and then to be confident.
Maybe I'm misunderstanding the quote, but this seems to wither if you have something to protect. If I'm having surgery, I don't really want the team of expert surgeons listening to my suggestions. I shouldn't be on my team because I'm not qualified. Highly qualified people should be so that my team will win (and I get to live).
Well, I think the thrust of the quote had more to do with being confident in your own projects. But I'll try to do an answer to your point because I think it's important to recognise the limitations of domain specialists - some of whom just aren't very good at their jobs.
If you're not on your team of expert surgeons, you're gonna be screwed if they're not actually as expert as you might think they were. There's a bit in What Do You Care What Other People Think? Where Feynman is talking about his first wife's hospitalisation - and how he had done some reading around the area and come up with the idea that it might be TB - and didn't push for the idea because he thought that the doctors knew what they were doing.
[Feynman moves onto less likely possibilities]
[Gets convinced to lie to her that it's Hodgkins - lie falls through]
=====================
Point being, disinvolving yourself from decisions is not a no-risk choice, and specialists aren't necessarily wise just because they've sat through the classes and crammed some sort of knowledge into their heads to get a degree. Assigning trust is a difficult subject.
There's a book called The Speed of Trust - and that's pretty much what you give up in being involved in complex decisions where you're not a specialist and where the specialists are actually really good at their jobs - a bit of speed.
Expert surgeons tend to think that more problems should be solved via surgery than doctors who aren't surgeons. Before getting surgery you should always talk with a doctor who knows something about the kind of illness you are having who isn't a surgeon.
After the operation is done doctors will ask you if everything is alright with you. If you try to understand what the operation involved you will give your doctor answers that are likely to be more informative than if you just try to place all responsibility onto another person.
Especially if you feel something that's not normal for the type of operation that you get, it important to be confident that you perceive something that's worth bringing to the attention of your doctor.
Having had big operations (one with 8 weeks of hospitalisation and one with 3 weeks) myself I think not taking enough for myself in those context was one of the worst decisions I made in my life. But then I was young and stupid about how the world works at the time.
Only if you're not the one with the responsibility to do something to protect it. I don't know the context of the quote, other than apparently being from an interview (with the actor, not any character he has played), but I read it as being about your own efforts to accomplish something. In such matters, you are the first person on your team, and you won't get any others on board by telling them you're not sure this is a good idea. Once you've made the decision that you are going to go for it, you have to then go for it, not sit around wondering if it's the right decision. If you're not acting on a decision, you didn't make it.
That may be a better wording of what I was trying to say here.
This works as a rationalization growing from the conclusion that others should be "on your team". If on well-calibrated assessment you yourself are not "on your team", others probably shouldn't be either, in which case projecting confidence amounts to deceit.
(Unless I don't understand what you are saying) I reject whatever definition 'deceit' is given such that the above claim is true. Behaving in a socially confident manner is different in nature to lying.
I was using "confidence" in a more specific sense, as in "overconfidence", that is implying that you know what you are doing, in the case where you actually don't. "Socially confident manner" might in contrast (for example, among many other things) involve willingness to state your state of uncertainty, as opposed to hiding it (including behind overconfidence).
This seems reasonable. Misleading about probabilities is deceptive. To be fair on Robert Downey, it doesn't seem likely that that is the the usage he was making in the quote.
Jehovah's Witnesses (or insert your cult of choice) who secretly don't believe in what they're selling, army recruiters who have secretly come to know and reject the horrors of war, insurance salesmen who sell useless policies:
All these (and many others) can be deceitful even without telling you their respective lies explicitly, just by using their social capital / community standing / aura of authority to signal their allegiance to their tribe, lending it credence in a deceitful (dishonest because not in tune with their well-calibrated assessment) manner. The similarity to lying comes from social cues (such as exuding confidence in one's role) and 'explicit' lies being forms of communication both.
It is possible to deceive others while using social confidence signals. Such signals are instrumentally useful for even vital for this and many other purposes. But this is not the same thing as the confidence being deceitful.
A somewhat similar sentiment: http://lesswrong.com/lw/2o3/rationality_quotes_september_2010/2kol
Why shouldn't they be? The idea that if you don't rate yourself highly no one should is just an excuse for shitty instincts.
Obviously it's a useful piece of nonsense to tell yourself. People are more likely to come to your side if you are confident. But the explicit reasoning is reprehensible. (not that any explicit reasoning probably went in, it's such a common idea that it is repeated without thought. It's almost a universal applause light.)
This is more of an irrationality quote. A bit of of paper thin justification for a shitty but common sentiment which it's useful to adopt rather than notice.