My other comments in this thread seem to have misread Tom’s intention, so let me try again. I’ll summarize what I think the point is, so if I’m off, it’ll be obvious.
The point is that, while rationality has been important and effective on the level of Western civilization, it isn’t really a super power when it comes down to individual behaviour. Thus, claims about rationality being all-important are dramatically overstated; there are much more important criteria for success.
My criticism of this, and the reason that I think it’s generating debate the way it is, is that the first half is simply false, and the second half is so obvious that people refuse to interpret it correctly.
Western civilization isn’t particularly powered by rationality. There’s a pretty complex cultural and institutional framework that powers the West and its prior success, some of which is thoroughly irrational (Mercantilism, anyone?). Rationality may have played some role in the West’s rise to dominance—it certainly did in a technological context—but you don’t need to be rational to point guns at people, even if you do need to be rational to invent guns.
Conversely, at the individual, rationality is easily outweighed by other factors. If you want to be a baseball player, hand-eye coordination, speed, and strength top rationality. For a politician, pick charisma. For a designer, pick social skills and taste. For a scientist, if you have basic competence in your field, pick skills conducive to publicity and obtaining grants. The only special thing about rationality is that, insofar as all of these professions must make decisions, additional rationality will help all of them, whereas running faster will not help most politicians.
There’s also an issue of survivorship bias. There are a whole lot of very irrational people. Therefore, saying, “Being irrational isn’t so bad; look at so-and-so, he did pretty well,” is actually not a useful argument. I guarantee you that, if your goal is being a lottery winner, being less rational will serve you extremely well. If a million irrational people all set out to be movie stars, ten succeed, and the rest are miserable, this does not make irrationality a good strategy.
I agree, but, ceteris paribus, I expect you’d see higher returns to these abilities. That is, if you had to choose between being better at math and being better at getting grants, you’d want to pick the grants, particularly over a certain threshold of math skills. This, however, assumes a lot about how you define “success” as a scientist. I may be overly cynical; I am not in the sciences.
Actually, I’ve looked at the data here, and I’d say to pick math skills. Professors at top-25 departments in mathematical sciences who took SATs at age 12 scored, at that time, 3.75 SD above the mean, which is FAR above the threshold for entry to any scientific field even at the PhD level (and there’s some regression from SAT scores at age 12 to math ability making the correlation even more impressive).
This sounds correct at the margin of an average scientist or average LW reader, but your original post could’ve been read as implying an average human baseline, where math probably is more important.
My other comments in this thread seem to have misread Tom’s intention, so let me try again. I’ll summarize what I think the point is, so if I’m off, it’ll be obvious.
The point is that, while rationality has been important and effective on the level of Western civilization, it isn’t really a super power when it comes down to individual behaviour. Thus, claims about rationality being all-important are dramatically overstated; there are much more important criteria for success.
My criticism of this, and the reason that I think it’s generating debate the way it is, is that the first half is simply false, and the second half is so obvious that people refuse to interpret it correctly.
Western civilization isn’t particularly powered by rationality. There’s a pretty complex cultural and institutional framework that powers the West and its prior success, some of which is thoroughly irrational (Mercantilism, anyone?). Rationality may have played some role in the West’s rise to dominance—it certainly did in a technological context—but you don’t need to be rational to point guns at people, even if you do need to be rational to invent guns.
Conversely, at the individual, rationality is easily outweighed by other factors. If you want to be a baseball player, hand-eye coordination, speed, and strength top rationality. For a politician, pick charisma. For a designer, pick social skills and taste. For a scientist, if you have basic competence in your field, pick skills conducive to publicity and obtaining grants. The only special thing about rationality is that, insofar as all of these professions must make decisions, additional rationality will help all of them, whereas running faster will not help most politicians.
There’s also an issue of survivorship bias. There are a whole lot of very irrational people. Therefore, saying, “Being irrational isn’t so bad; look at so-and-so, he did pretty well,” is actually not a useful argument. I guarantee you that, if your goal is being a lottery winner, being less rational will serve you extremely well. If a million irrational people all set out to be movie stars, ten succeed, and the rest are miserable, this does not make irrationality a good strategy.
Wait, really? I was given to understand that there was, like, math involved. And what we call “epistemic rationality.”
I agree, but, ceteris paribus, I expect you’d see higher returns to these abilities. That is, if you had to choose between being better at math and being better at getting grants, you’d want to pick the grants, particularly over a certain threshold of math skills. This, however, assumes a lot about how you define “success” as a scientist. I may be overly cynical; I am not in the sciences.
Actually, I’ve looked at the data here, and I’d say to pick math skills. Professors at top-25 departments in mathematical sciences who took SATs at age 12 scored, at that time, 3.75 SD above the mean, which is FAR above the threshold for entry to any scientific field even at the PhD level (and there’s some regression from SAT scores at age 12 to math ability making the correlation even more impressive).
In my experience actually being in the sciences would make you more cynical.
This sounds correct at the margin of an average scientist or average LW reader, but your original post could’ve been read as implying an average human baseline, where math probably is more important.
That’s why you’ve got postdocs ;)
Postdocs need shmoozing and grant-writing skills too—ask me how I know.
How do you know, oh friend Cyan?
I’m a postdoc sorely feeling the lack of schmoozing and grant-writing skills. :-p
See http://lesswrong.com/lw/1ko/on_the_power_of_intelligence_and_rationality/1d5t