This is a bit of a nerdy quibble, but worth bringing up: our qualitative idea of an "independent" expert and an "independent" test aren't the same thing as probabilistic independence. We think of an independent expert as something like: an expert who examines the evidence without consulting the opinions/notes of other experts before forming their opinion. But that doesn't guarantee probabilistic independence. In fact, most expert judgments will be probabilistically related when looking at similar evidence sets or when operating in the same "universe of discourse," where e.g. "Martian Fish" is never going to be the correct answer.
That doesn't undermine most of your point here, which I think is more right than wrong (especially about the value of counter reliable sources; there are specific people whose social media I check because I knew they tend to be anti-correlated to the truth). But while Condorect gives some conditions under which diversifying your evidential influences can improve things, they're not conditions we often find ourselves in.
Agreed about the limits of Condorcet, and very fair point about the “independent expert” meaning something different in ordinary language. I tried to cheat a little bit by defining both expertise and independence probabilistically. In part, that’s because I didn’t want to commit any crimes against Bayes in the presence of you and Kenny!
I think I have a weird extension of this logic. I sometimes deliberately keep myself ignorant of a topic, in order to leave more space for originality and potential new insights and avoid just adopting others opinions (and it's also more satisfying to figure things out for myself). I suppose I'm choosing to be less informed, in order to (hopefully) be more informative. I guess effectively trying to be the less reliable "independent expert".
It's counterintuitive that the less reliable expert might be more informative, but it's even weirder to think it might be worth choosing to be less reliable/informed yourself.
Although I tend to only do this temporarily and then try to learn what others have found after.
I really think this article is exceptionally timely. We don’t need more polarization and blocking people like Vance isn’t going to cut it. Let’s have more people with the attitude of that democratic representative and less of that writer who think that Iglesias is never right. That writer illustrates why contrarianism is just as bad as being a parrot. Both result from a refusal to do fully independent thinking ! Thanks for this one, Restacking!
Thanks for writing this, hopefully helpful clarifications. In financial circles, we have the useful "alpha" and "beta" metrics that capture types of investment reward with different correlation structures, very helpful in simple investment portfolio design.
On Substack, I am consistently blocking people who are parrots and spammers from my perspective, in order to ensure interesting voices are not crowded out. I'm deeply tired of Adelstein's Pascal's-wager-spam "act now or infinite suffering!" ecosystem, for instance, and I probably would not have been able to see this article through that loud noise if I weren't muting and blocking quite a few of the people there.
> But more recently the Harvard philosopher Selim Berker has come up with arguably the clearest and most authoritative treatment of “opposition” in history—though it’s buried in a paper on the ethical concept of “fittingness.”
"What emerges is a better appreciation of the structural characteristics of fittingness and the other aptic categories, as well as an argument for taking up the nature of oppositeness as a serious philosophical topic that is ripe for further exploration."
Pray tell what did the professor examine and find to be unripe? :)
This is a bit of a nerdy quibble, but worth bringing up: our qualitative idea of an "independent" expert and an "independent" test aren't the same thing as probabilistic independence. We think of an independent expert as something like: an expert who examines the evidence without consulting the opinions/notes of other experts before forming their opinion. But that doesn't guarantee probabilistic independence. In fact, most expert judgments will be probabilistically related when looking at similar evidence sets or when operating in the same "universe of discourse," where e.g. "Martian Fish" is never going to be the correct answer.
That doesn't undermine most of your point here, which I think is more right than wrong (especially about the value of counter reliable sources; there are specific people whose social media I check because I knew they tend to be anti-correlated to the truth). But while Condorect gives some conditions under which diversifying your evidential influences can improve things, they're not conditions we often find ourselves in.
Also, I’m always here for Daniel Rubio’s nerdy quibbles. My kind of content.
Agreed about the limits of Condorcet, and very fair point about the “independent expert” meaning something different in ordinary language. I tried to cheat a little bit by defining both expertise and independence probabilistically. In part, that’s because I didn’t want to commit any crimes against Bayes in the presence of you and Kenny!
I think I have a weird extension of this logic. I sometimes deliberately keep myself ignorant of a topic, in order to leave more space for originality and potential new insights and avoid just adopting others opinions (and it's also more satisfying to figure things out for myself). I suppose I'm choosing to be less informed, in order to (hopefully) be more informative. I guess effectively trying to be the less reliable "independent expert".
It's counterintuitive that the less reliable expert might be more informative, but it's even weirder to think it might be worth choosing to be less reliable/informed yourself.
Although I tend to only do this temporarily and then try to learn what others have found after.
Joey Ramone used to refuse to listen to other guitarists.
I really think this article is exceptionally timely. We don’t need more polarization and blocking people like Vance isn’t going to cut it. Let’s have more people with the attitude of that democratic representative and less of that writer who think that Iglesias is never right. That writer illustrates why contrarianism is just as bad as being a parrot. Both result from a refusal to do fully independent thinking ! Thanks for this one, Restacking!
Cheers, Ken!
Thanks for writing this, hopefully helpful clarifications. In financial circles, we have the useful "alpha" and "beta" metrics that capture types of investment reward with different correlation structures, very helpful in simple investment portfolio design.
On Substack, I am consistently blocking people who are parrots and spammers from my perspective, in order to ensure interesting voices are not crowded out. I'm deeply tired of Adelstein's Pascal's-wager-spam "act now or infinite suffering!" ecosystem, for instance, and I probably would not have been able to see this article through that loud noise if I weren't muting and blocking quite a few of the people there.
That’s a great analogy!
The way you made so many different topics of equal importance fit together is quite remarkable
Blocked (jk)
> But more recently the Harvard philosopher Selim Berker has come up with arguably the clearest and most authoritative treatment of “opposition” in history—though it’s buried in a paper on the ethical concept of “fittingness.”
"What emerges is a better appreciation of the structural characteristics of fittingness and the other aptic categories, as well as an argument for taking up the nature of oppositeness as a serious philosophical topic that is ripe for further exploration."
Pray tell what did the professor examine and find to be unripe? :)