Sklansky's Handicap

[madnak]

This is a response to Sklansky's series of threads in which he basically says that if someone is smarter than you, you should agree with what they say. Okay, that's a slight oversimplification, but you get the idea.

So my question to you, David, is this - if someone who scores well above you on the metrics you're referring to were to come into the debate and disagree with you, that is, if one of the "great minds" you like to mention were to post and tell you that no, smart people are wrong a whole hell of a lot, would you abandon this line of reasoning?

I'm assuming someone who is more knowledgeable than you, who is smarter than you, and who is biased against his own position (as he will generally be, given that your position is that he's some great ubermensch). According to your standards, such a person would be incalculably more likely than you to be right about the subject. Therefore, if he were to dispute your position, you would have to abandon it. Correct?

I believe it wouldn't be outside the realm of possibility for the members of SMP to find such an "anti-Sklansky" with a distinguished record and stratospheric IQ score...

[Utah]

[ QUOTE ]
I'm assuming someone who is more knowledgeable than you, who is smarter than you, and who is biased against his own position (as he will generally be, given that your position is that he's some great ubermensch). According to your standards, such a person would be incalculably more likely than you to be right about the subject. Therefore, if he were to dispute your position, you would have to abandon it. Correct?

[/ QUOTE ]You have introduced a new variable - that the argument has been studied by the handicapper. I believe that you would have to state your question to be:

.....Therefore, if you were likely to get into a dispute in the future about an unknown subject where the previous conditions exist would you agree that it is more likely that the other player will be correct?

[Borodog]

I believe if some "ubermensch" came along proclaiming the theory that, all other things being equal, more intelligent people are not more likely to be correct than less intelligent people, David Sklansky would (rightly) reduce his estimation of that person's intelligence.

[madnak]

The question isn't whether intelligence is a factor, it's whether intelligence is the factor. You're being almost as black-and-white as David. Of course geniuses have a higher probability of being correct; they also have a higher probability of being loony.

I'm a bit surprised you're in agreement with David, actually, because his position very much contradicts the "marketplace of ideas" approach to thought (as well as the idea of meme transmission as a primary innovative vector). I always thought you liked those approaches.

Utah - you're right, it's impossible to prove in reality. That's a pity. Still, we can work in terms of a thought experiment. What if, hypothetically, we were to poll everyone on Earth with an IQ > 180. And what if 80% of them disagreed with David on this issue? That should eliminate the impact of selection bias, right?

[Utah]

[ QUOTE ]
Utah - you're right, it's impossible to prove in reality. That's a pity. Still, we can work in terms of a thought experiment. What if, hypothetically, we were to poll everyone on Earth with an IQ > 180. And what if 80% of them disagreed with David on this issue? That should eliminate the impact of selection bias, right?

[/ QUOTE ]Respectfully, I think you missed the point. This too breaks the original handicap model because it is no longer blind and thus the automatic edge in the original handicap model (barring my awesome insight into the chance that the lower scoring player could actually be right more often in a binary decision model) no longer applies. For example, Skalansky might have been -EV going in but stumbled onto something brilliantly insightful and obviously correct that the other player missed (or that every 99%+ IQer on the planet missed). There is no way he should think he is wrong.

I think he would agree that if such an unlikely person existed that was smarter, more knowledgable, and without bias were to challenge him in a future epic battle of the two most brilliant minds on the planet and if the topic was unknown at this time and the scores were directly relevant to the topic he would lay odds on his opponent. But, only he can answer for sure.

[madnak]

[ QUOTE ]
I believe if some "ubermensch" came along proclaiming the theory that, all other things being equal, more intelligent people are not more likely to be correct than less intelligent people, David Sklansky would (rightly) reduce his estimation of that person's intelligence.

[/ QUOTE ]

By the way, isn't this antithetical to David's position? David is recommending that, when we strongly disagree with the "smart people," we should reevaluate our own opinions, and our own evaluation of our relative intelligence. They've already proved themselves, therefore if they disagree with us, we are the ones more likely to be incorrect. For David to take the opposite approach outlined here, to, upon finding that he's in disagreement with the "smarter people," people who've already proved themselves in various metrics, reevaluate their intelligence level rather than his own... Well, that would show that he's not acting consistently. It would also show that he's using his ego and emotions to determine his positions, rather than his analytical capabilities.

[FortunaMaximus]

[ QUOTE ]
It would also show that he's using his ego and emotions to determine his positions, rather than his analytical capabilities.

[/ QUOTE ]

That's unavoidable for anybody scoring well enough in the metric. But what the counterpoint is for most of those... The blows to the ego and emotional swaying forces the individuals to look at their own analytical capabilities.

If you can incorporate that, the overall sum cannot help but be improved. If you can't, though...

[madnak]

I think it's best for an intelligent person to "over-correct" sometimes. Intelligent people are used to being right, and so it's hard for them to get into the habit of questioning their conclusions. They start to just assume everything they do is right, and as a result someone less intelligent but more perceptive or careful may achieve a greater level of accuracy.

I think it's useful sometimes to step back and think, "what if I'm wrong?" Or even look at your own arguments and try to refute them, try to put yourself in your opponent's place. At worst they give you a better understanding of why you're right, but sometimes they can yield surprising insights that you'd never think to find without consciously suspending your disbelief. These insights aren't necessarily even related to the argument you're having!

But the strange thing is that in spite of the results, it can get hard or even scary. Religion is one subject in particular that really starts to scare me when I consider its truth. I don't think it's all about the fire and brimstone either - it happens with Buddhism, Taoism, all that stuff. It makes me feel like I'm sinking in the ocean. I think it's just the dissonance that stems from questioning hard assumptions that I have.

[Borodog]

madnak,

I think you are stretching David's point too far, that's all. Just because a guy you previously thought was smarter than you says something clearly fallacious (like all else being equal, smarter people are not more likely to be correct), doesn't cause you to throw out logic and adopt nonsensical positions. It should cause you to revise downward your opinion of that person's intelligence.

[aeest400]

Nice post. David is clearly a bright guy, but some of his arguments don't seem to grasp the complexity of various issues about which he posts and the underlying assumptions involved. He analyzes scenarios with examples and thought experiements, when he may be just better off analyzing them (to use a word he seem to apply pejoritively) philosophically. I guess I'm a little irked because I am sometimes aware of philosophers who have discussed issues he brings up in a more rigorous manner. E.g., I think he would be well served by reading some Hilary Putnam or Donald Davidson or Daniel Dennett. That said, I think he is often correct.