We actually see this across a lot of fields besides philosophy, and it’s not LW-specific. For example, simply adding up a few simple scores does better than experts at predicting job performance.
It’s been shown that expertise is only valuable in fields where there is a short enough and frequent enough feedback loop for a person to actually develop expertise—and there is something coherent to develop the expertise in. Outside of such fields, experts are just blowhards with status.
Given the nature of the field, the prior expectation for philosophers having any genuine expertise at anything except impressing people, should be set quite low. (Much like we should expect expert short-term stock pickers to not be expert at anything besides being lucky.)
Of course, one could argue that LW regulars get even less rapid feedback on these issues than the professional philosophers do. The philosophers at least are frequently forced to debate their ideas with people who disagree, while LW posters mostly discuss these things with each other—that is, with a group that is self-selected for thinking in a similar way. We don’t have the kind of diversity of opinion that is exemplified by these survey results.
However see my comment above for evidence suggesting that the views of the specialists are those they brought with them to the field (or shifting away from the plurality view), i.e. that the skew of views among specialists is NOT due to such feedback.
It’s been shown that expertise is only valuable in fields where there is a short enough and frequent enough feedback loop for a person to actually develop expertise—and there is something coherent to develop the expertise in
What do you think philosophy is lacking? An (analytical) philosopher who makes a logic error is hauled up very quickly by their peers. That’s your feedback loop. So is “something coherent” lacking? Phil. certainly doesn’t have a set of established results like engineering, or the more settled areas of science. It does have a lot of necessary skill in formulating, expressing and criticising ideas and arguments. Musicians aren’t non-experts just because there is barely
such a thing as a musical fact. Philosophy isn’t broken science.
OK, so philosophers manage to avoid logical errors. Good for them. However, they make more complicated errors (see A Human’s Guide To Words for some examples), as well as sometimes errors of probability. The thing that philosophers develop expertise in is writing interesting arguments and counterarguments. But these arguments are castles built on air; there is no underlying truth to most of the questions they ask (or, if there is an underlying truth, there is no penalty for being wrong about it). And even some of the “settled” positions are only settled because of path-dependence—that is, once they became popular, anyone with conflicting intuitions would simply never become a philosopher (see Buckwalter and Stich for more on this).
Scientists (at least in theory) have all of the same skills that philosophers should have—formulating theories and arguments, catching logical errors, etc. It’s just that in science, the arguments are (when done correctly) constrained to be about the real world.
Why is that a problem? Science deals with empirical reality, philosophy of science deals with meta-level issues. Each to their own.
Because if there is no fact of the matter on the “meta-level issues”, then you’re not actually dealing with “meta-level issues”. You are dealing with words, and your success in dealing with words is what’s being measured. Your argument is that expertise develops by feedback, but the feedback that philosophers get isn’t the right kind of feedback.
I don’t know what you mean by “fact of the matter”. It’s not a problem that meta-level isn’t object level, any more than it’s a problem that cats aren’t dogs. I also don’t think that there is any problem in identifying the meta level.
Philosophers “don’t deal with words” in the sense that linguists. They use words to do things, as do many other specialities. You seem to be making the complaint that success isn’t well defined in philosophy, but that would require treating object level science as much more algorithmic than it actually is. What makes a scientific theory
a good theory? Most scientists agree on it?
I don’t know what you mean by “fact of the matter”.
An actual truth about the world.
I don’t know what you mean by that. Is Gresham’s law such a truth?
What makes a scientific theory a good theory?
Have you read A Technical Explanation of Technical Explanation?
My question was rhetorical. Science does not deal entirely in directly observable empirical facts—which might be what you meant by “actual truths about the world”. Those who fly under the Bayesian flag by and large don’t either: most of the material on this site is just as indirect/meta-levle/higher-level as philosophy.
I just don’t see anything that justifies the “Boo!” rhetoric.
Actually, perhaps you should try The Simple Truth, because you seem totally confused.
Yes, a lot of the material on this site is philosophy; I would argue that it is correspondingly more likely to be wrong, precisely because is not subject to the same feedback loops as science. This is why EY keeps asking, “How do I use this to build an AI?”
So...is Gresham;s Law an actual truth about the world?
As far as I can tell, yes (in a limited form), but I’m prepared for an economist to tell me otherwise.
The focus of the question was “about the world”. Gresham’s law, if true, is not a direct empirical fact like the metling point of aluminium, not is it built into the fabric of the universe, since it is indefinable without humans and their economic activity.
perhaps you should try The Simple Truth
Now I’m confused. Is that likely to be wrong or not?
If we consider it as a definition, then it is either useful or not useful.
So this is about the “true” part, not about the “actual world” part? In that case, You are’;t complaining that philosophy ins;t connected to reality, your claiming that it is all false. In that case I will have to ask you when and how you became omniscient.
The focus of the question was “about the world”. Gresham’s law, if true, is not a direct empirical fact like the melting point of aluminium, not is it built into the fabric of the universe, since it is indefinable without humans and their economic activity.
Humans are part of the world.
So this is about the “true” part, not about the “actual world” part? In that case, You aren’t complaining that philosophy isn’t connected to reality, your claiming that it is all false. In that case I will have to ask you when and how you became omniscient.
I’m afraid I don’t understand what you’re saying here. Yes, if you are confused about what truth means, a definition would be useful; I think The Simple Truth is a pretty useful one (if rather long-winded, as is typical for Yudkowsky). It doesn’t tell you much about the actual world (except that it hints at a reasonable justification for induction, which is developed more fully elsewhere).
But I’m not sure why you think I am claiming philosophy is all false.
The focus of the question was “about the world”. Gresham’s law, if true, is not a direct empirical fact like the melting point of aluminium, not is it built into the fabric of the universe, since it is indefinable without humans and their economic activity.
Humans are part of the world.
Then there is no reason why some philosopihical claims about human nature could not count as Actual Truths About The World, refuting your original point.
That depends on what you mean by “human nature,” but yes, some such claims could. However, they aren’t judged based on this (outside of experimental philosophy, of course). So, there is no feedback loop.
OK, it has been established that you attach True to the sentence:
“Philosophers are not judged based on whether their claims accurately describe the world”.
The question is what that means. We have established that philosophical claims
can be about the world, and it seems uncontroversial that some of the make true claims some of the time, since they all disagree with each other and therefore can’t all be wrong.
The problem is presumably the epistemology, the justification. Perhaps you mean that philosophy doesn’t use enough empiricism. Although it does use empiricism sometimes, and it is not that every scientific question can
be settled empirically.
Just a friendly advice. Having looked through your comment history I have noticed that you have trouble interpreting the statements of others charitably. This is fine for debate-style arguments, but is not a great idea on this forum, where winning is defined by collectively constructing a more accurate map, not as an advantage in a zero-sum game. (Admittedly, this is the ideal case, the practice is unfortunately different.) Anyway, consider reading the comments you are replying to in the best possible way first.
If you honestly do not understand the point the comment you are replying to is making, a better choice is asking the commenter to clarify, rather than continuing to argue based on this lack of understanding. TheOtherDave does it almost to a fault, feel free to read some of his threads. Asking me does not help, I did not write the comment you didn’t understand.
It’s been shown that expertise is only valuable in fields where there is a short enough and frequent enough feedback loop for a person to actually develop expertise—and there is something coherent to develop the expertise in. Outside of such fields, experts are just blowhards with status.
I don’t disagree with this, but do you happen to have a cite?
I would also point out that feedback which consists solely of the opinions of other experts probably shouldn’t count as feedback. Too much danger of groupthink.
The finding that expertise is only valuable in fields where there is a sufficiently short and frequent feedback look plausibly explains why professional philosophers are no better than the general population at answering philosophical questions. However, it doesn’t explain the observation that philosophical expertise seems to be negatively correlated with true philosophical beliefs, as opposed to merely uncorrelated. Why are philosophers of religion less likely to believe the truth about religion, moral philosophers less likely to believe the truth about morality, and metaphysicians less likely to believe the truth about reality, than their colleagues with different areas of expertise?
I would guess that those particular fields look more interesting when you make the wrong assumptions to begin with. I mean, it’s much less interesting to talk about God when you accept there is none. Or to talk about metaphysics, when you accept that the answer will most likely come from physics. (I don’t know about morality.)
We actually see this across a lot of fields besides philosophy, and it’s not LW-specific. For example, simply adding up a few simple scores does better than experts at predicting job performance.
It’s been shown that expertise is only valuable in fields where there is a short enough and frequent enough feedback loop for a person to actually develop expertise—and there is something coherent to develop the expertise in. Outside of such fields, experts are just blowhards with status.
Given the nature of the field, the prior expectation for philosophers having any genuine expertise at anything except impressing people, should be set quite low. (Much like we should expect expert short-term stock pickers to not be expert at anything besides being lucky.)
Of course, one could argue that LW regulars get even less rapid feedback on these issues than the professional philosophers do. The philosophers at least are frequently forced to debate their ideas with people who disagree, while LW posters mostly discuss these things with each other—that is, with a group that is self-selected for thinking in a similar way. We don’t have the kind of diversity of opinion that is exemplified by these survey results.
This seems right to me.
However see my comment above for evidence suggesting that the views of the specialists are those they brought with them to the field (or shifting away from the plurality view), i.e. that the skew of views among specialists is NOT due to such feedback.
What do you think philosophy is lacking? An (analytical) philosopher who makes a logic error is hauled up very quickly by their peers. That’s your feedback loop. So is “something coherent” lacking? Phil. certainly doesn’t have a set of established results like engineering, or the more settled areas of science. It does have a lot of necessary skill in formulating, expressing and criticising ideas and arguments. Musicians aren’t non-experts just because there is barely such a thing as a musical fact. Philosophy isn’t broken science.
OK, so philosophers manage to avoid logical errors. Good for them. However, they make more complicated errors (see A Human’s Guide To Words for some examples), as well as sometimes errors of probability. The thing that philosophers develop expertise in is writing interesting arguments and counterarguments. But these arguments are castles built on air; there is no underlying truth to most of the questions they ask (or, if there is an underlying truth, there is no penalty for being wrong about it). And even some of the “settled” positions are only settled because of path-dependence—that is, once they became popular, anyone with conflicting intuitions would simply never become a philosopher (see Buckwalter and Stich for more on this).
Scientists (at least in theory) have all of the same skills that philosophers should have—formulating theories and arguments, catching logical errors, etc. It’s just that in science, the arguments are (when done correctly) constrained to be about the real world.
How do you know?
How do you know? Are you aware that much philosophy is about science.
To be fair, I have not done an exhaustive survey; “most” was hyperbole.
Sure. But there is no such constraint on philosophy of science.
Why is that a problem? Science deals with empirical reality, philosophy of science deals with meta-level issues. Each to their own.
Because if there is no fact of the matter on the “meta-level issues”, then you’re not actually dealing with “meta-level issues”. You are dealing with words, and your success in dealing with words is what’s being measured. Your argument is that expertise develops by feedback, but the feedback that philosophers get isn’t the right kind of feedback.
I don’t know what you mean by “fact of the matter”. It’s not a problem that meta-level isn’t object level, any more than it’s a problem that cats aren’t dogs. I also don’t think that there is any problem in identifying the meta level. Philosophers “don’t deal with words” in the sense that linguists. They use words to do things, as do many other specialities. You seem to be making the complaint that success isn’t well defined in philosophy, but that would require treating object level science as much more algorithmic than it actually is. What makes a scientific theory a good theory? Most scientists agree on it?
An actual truth about the world.
Have you read A Technical Explanation of Technical Explanation?
I don’t know what you mean by that. Is Gresham’s law such a truth?
My question was rhetorical. Science does not deal entirely in directly observable empirical facts—which might be what you meant by “actual truths about the world”. Those who fly under the Bayesian flag by and large don’t either: most of the material on this site is just as indirect/meta-levle/higher-level as philosophy. I just don’t see anything that justifies the “Boo!” rhetoric.
Actually, perhaps you should try The Simple Truth, because you seem totally confused.
Yes, a lot of the material on this site is philosophy; I would argue that it is correspondingly more likely to be wrong, precisely because is not subject to the same feedback loops as science. This is why EY keeps asking, “How do I use this to build an AI?”
So...is Gresham;s Law an actual truth about the world?
Now I’m confused. Is that likely to be wrong or not?
As far as I can tell, yes (in a limited form), but I’m prepared for an economist to tell me otherwise.
If we consider it as a definition, then it is either useful or not useful.
The focus of the question was “about the world”. Gresham’s law, if true, is not a direct empirical fact like the metling point of aluminium, not is it built into the fabric of the universe, since it is indefinable without humans and their economic activity.
So this is about the “true” part, not about the “actual world” part? In that case, You are’;t complaining that philosophy ins;t connected to reality, your claiming that it is all false. In that case I will have to ask you when and how you became omniscient.
Humans are part of the world.
I’m afraid I don’t understand what you’re saying here. Yes, if you are confused about what truth means, a definition would be useful; I think The Simple Truth is a pretty useful one (if rather long-winded, as is typical for Yudkowsky). It doesn’t tell you much about the actual world (except that it hints at a reasonable justification for induction, which is developed more fully elsewhere).
But I’m not sure why you think I am claiming philosophy is all false.
Then there is no reason why some philosopihical claims about human nature could not count as Actual Truths About The World, refuting your original point.
That depends on what you mean by “human nature,” but yes, some such claims could. However, they aren’t judged based on this (outside of experimental philosophy, of course). So, there is no feedback loop.
Based on what? Is Gresham’s law based on “this”?
That comment could have been more clear. My apologies.
Philosophers are not judged based on whether their claims accurately describe the world. This was my original point, which I continue to stand by.
OK, it has been established that you attach True to the sentence:
“Philosophers are not judged based on whether their claims accurately describe the world”.
The question is what that means. We have established that philosophical claims can be about the world, and it seems uncontroversial that some of the make true claims some of the time, since they all disagree with each other and therefore can’t all be wrong.
The problem is presumably the epistemology, the justification. Perhaps you mean that philosophy doesn’t use enough empiricism. Although it does use empiricism sometimes, and it is not that every scientific question can be settled empirically.
I’m going to leave this thread here, because I think I’ve made my position clear, and I don’t think we’ll get further if I re-explain it.
Doesn’t follow.
You mean there are ideas no philosopher has contemplated?
Just a friendly advice. Having looked through your comment history I have noticed that you have trouble interpreting the statements of others charitably. This is fine for debate-style arguments, but is not a great idea on this forum, where winning is defined by collectively constructing a more accurate map, not as an advantage in a zero-sum game. (Admittedly, this is the ideal case, the practice is unfortunately different.) Anyway, consider reading the comments you are replying to in the best possible way first.
Speaking of which, I I honestly had no idea what the “this” meant. Do you?
If you honestly do not understand the point the comment you are replying to is making, a better choice is asking the commenter to clarify, rather than continuing to argue based on this lack of understanding. TheOtherDave does it almost to a fault, feel free to read some of his threads. Asking me does not help, I did not write the comment you didn’t understand.
I believe I did:-
′ Based on what? Is Gresham’s law based on “this”?′
The point is that if no one can understand the comment, then I am not uncharitably pretending not to understand the comment:
I don’t disagree with this, but do you happen to have a cite?
I would also point out that feedback which consists solely of the opinions of other experts probably shouldn’t count as feedback. Too much danger of groupthink.
The finding that expertise is only valuable in fields where there is a sufficiently short and frequent feedback look plausibly explains why professional philosophers are no better than the general population at answering philosophical questions. However, it doesn’t explain the observation that philosophical expertise seems to be negatively correlated with true philosophical beliefs, as opposed to merely uncorrelated. Why are philosophers of religion less likely to believe the truth about religion, moral philosophers less likely to believe the truth about morality, and metaphysicians less likely to believe the truth about reality, than their colleagues with different areas of expertise?
Edit: this post is mostly a duplicate of this one
I would guess that those particular fields look more interesting when you make the wrong assumptions to begin with. I mean, it’s much less interesting to talk about God when you accept there is none. Or to talk about metaphysics, when you accept that the answer will most likely come from physics. (I don’t know about morality.)