One way I’d put it: it seems like you’re anchoring the bar for good philosophy to a level where it’s very rare to find examples of good philosophy. Setting a high bar is an understandable reaction to noticing everyone else missing this big obvious thing (because clearly everyone else is setting a bar that’s too low), but I don’t think it’ll get you where you want.
To come back to this, there may be a persistent misunderstanding where you think my bar for good philosophy is so high that I rarely find examples of it. (I just came across an old thread where you asked me for examples of what I considered good philosophy, which I gave but you didn’t reply after that.) I think there are plenty of examples of good philosophy (additional examples: I appreciate academic meta-ethics for delineating various possible answers, UDT is based on previous decision theory and Eliezer and Nesov’s ideas/hints on LW, Bostrom and Ord’s Moral Parliament). When I say that humans are bad at philosophy I instead mean a combination of:
the collective competence is too low relative to the size of the challenges (e.g., solving philosophical problems in AI safety and other tech advances in time)
most humans care very little about philosophy (making it unsafe to have AIs obey/defer to humans)
among those who care, only a minority follow a reasonable philosophical tradition like analytical philosophy
individually most philosophers (even analytical philosophers) seem overconfident in their own ideas/positions
As I just articulated in a different comment, my interest in engaging with your ideas is bottlenecked on whether it seems like you’re building up a set of crisper, clearer concepts, versus invoking a very broad skepticism. Obviously there’s a spectrum between these, and I don’t claim you’re right at the latter end, but you seem too close to it for me to get much value out of many of your comments. In particular, both of the examples you gave in the last thread (the naturalistic fallacy and moral/normative uncertainty) are much closer to “finding new ways of being skeptical” than “developing useful conceptual frameworks”.
To be clear, I like the examples in this post more (especially UDT/decision theory and the moral parliament stuff). I don’t know if you have enough other examples to allow you to catalogue habits of thinking and intellectual patterns that seem unusually productive, but if you do that seems like an interesting direction (and if not, you could probably do a similar thing by looking at patterns in which philosophy is very bad vs less bad).
To cite one field that I’m especially familiar with, consider probability and decision theory, where we went from having no concept of probability, to studies involving gambles and expected value, to subjective probability, Bayesian updating, expected utility maximization, and the Turing-machine-based universal prior, to the recent realizations that EU maximization with Bayesian updating and the universal prior are both likely to be wrong or incomplete.
Insofar as this is a central example of the kind of thing you want to understand and make progress on, I think your mistake might be simply calling it “philosophy”, when in fact the thing people were doing back then bears very little resemblance to modern academic philosophy. The term “natural philosophy” seems more accurate, but probably even better is “pre-paradigmatic science”.
To come back to this, there may be a persistent misunderstanding where you think my bar for good philosophy is so high that I rarely find examples of it. (I just came across an old thread where you asked me for examples of what I considered good philosophy, which I gave but you didn’t reply after that.) I think there are plenty of examples of good philosophy (additional examples: I appreciate academic meta-ethics for delineating various possible answers, UDT is based on previous decision theory and Eliezer and Nesov’s ideas/hints on LW, Bostrom and Ord’s Moral Parliament). When I say that humans are bad at philosophy I instead mean a combination of:
the collective competence is too low relative to the size of the challenges (e.g., solving philosophical problems in AI safety and other tech advances in time)
most humans care very little about philosophy (making it unsafe to have AIs obey/defer to humans)
among those who care, only a minority follow a reasonable philosophical tradition like analytical philosophy
individually most philosophers (even analytical philosophers) seem overconfident in their own ideas/positions
As I just articulated in a different comment, my interest in engaging with your ideas is bottlenecked on whether it seems like you’re building up a set of crisper, clearer concepts, versus invoking a very broad skepticism. Obviously there’s a spectrum between these, and I don’t claim you’re right at the latter end, but you seem too close to it for me to get much value out of many of your comments. In particular, both of the examples you gave in the last thread (the naturalistic fallacy and moral/normative uncertainty) are much closer to “finding new ways of being skeptical” than “developing useful conceptual frameworks”.
To be clear, I like the examples in this post more (especially UDT/decision theory and the moral parliament stuff). I don’t know if you have enough other examples to allow you to catalogue habits of thinking and intellectual patterns that seem unusually productive, but if you do that seems like an interesting direction (and if not, you could probably do a similar thing by looking at patterns in which philosophy is very bad vs less bad).
EDIT: I just saw this other list from this old post, which I also like:
Insofar as this is a central example of the kind of thing you want to understand and make progress on, I think your mistake might be simply calling it “philosophy”, when in fact the thing people were doing back then bears very little resemblance to modern academic philosophy. The term “natural philosophy” seems more accurate, but probably even better is “pre-paradigmatic science”.