This seems to be taking down a straw man, and far from “challenging a central tenet of LW: reductionism”, you perfectly describe it and expound on it, if a bit wordily. At least in my mind, it’s very obvious that physical ‘law’ is a map-level concept. Physicists themselves have noticed that for a map-level concept, physical ‘law’ fits the territory so amazingly well, that they have written articles such as “The Unreasonable Effectiveness of Mathematics in the Natural Sciences”
I don’t claim that this post in particular challenges the consensus (at least, I don’t intend to claim that, but I can see how my phrasing in the intro suggests it). It’s mostly just setup. I think the LW consensus is probably closer to what I call “explanatory reductionism” at the end of this post, but attacking that position required that I make it clear how I think about laws of nature. The ultimate position I want to defend is that the only tenable reductionism is the extremely weak mereological kind. Surely this is different from the position generally advocated here.
That said, I don’t think the position I’m attacking in this post is a straw man. As I point out, Paul Davies (hardly a fringe figure in physics) explicitly embraces it. He also says (in the linked excerpt) that “most physicists working on fundamental topics inhabit the prescriptive camp, even if they won’t own up to it explicitly.” In addition, I’ve seen nomic reductionism defended (and upvoted) on LW more than once. As an example, see some of the comments on this thread. Even people who would, if pressed, agree that laws are description often unconsciously infer things that only work if you think of laws as rules.
Do you think the points made in this post are common enough knowledge around here for the post to be of not much use?
I don’t claim that this post in particular challenges the consensus
That would be the bit where you said “This is the first in a planned series of posts challenging a central tenet of the LessWrong consensus”. When you say that you’re challenging the consensus, it appears to the reader as though you’re challenging the consensus.
I agree with David_Gerard: when I say I’m doing something, it appears to the reader as though I’m doing that thing.
I would also agree with various more-strongly-worded equivalents, such as “when I say I’m doing X in a series of acts that includes Y, it’s disingenuous to later claim that Y wasn’t intended to do X.”
Hence, understatement. That is, an expression worded less strongly than, in my opinion, the situation justifies.
Is “challenge the consensus” a performative utterance? By saying “I challenge the consensus regarding foo”, do you thereby challenge the consensus regarding foo?
Consider: If I said, “I challenge the Less Wrong consensus that 2 + 2 = 5. I assert that it’s 4,” by saying this I wouldn’t actually challenge a consensus that 2 + 2 = 5, because there isn’t one to challenge. Rather, all I would be doing is setting up a straw man: falsely asserting the existence of a consensus, and then disagreeing with that imagined consensus.
What is, actually, the difference between laws as rules and laws as descriptions of regularity, except the choice of language? There is in fact a pretty strong LW consensus that beliefs should be distinguishable from each other by different anticipated experiences; I am not sure whether nomic and mereological reductionism predict different observations.
(I agree with the gist of the post, in the sense that it is more elegant to view physical laws as descriptions of regularities in observed universe, rather than rules that push matter around.)
See my comment here. Thinking of laws as rules vs. descriptions may not predict different observations, but they do lead to different cognitive attitudes about scientific inquiry and explanation (e.g. we often think of rules as explaining behavior, but we don’t think of descriptions as explanatory in the same way), and in this regard I think the prescriptive perspective is a recipe for confusion. One could say the same thing about the kind of empiricism advocated on LW as opposed to rationalism (of the traditional philosophical variety, not the LW variety). These two philosophical stances don’t predict different observations, but that doesn’t mean the choice between them is merely a linguistic one. They are associated with different pragmatic attitudes, and the empiricist attitude is a lot more valuable that the rationalist one.
What concrete differences in e.g. scientific achievements would you expect to stem from the difference between the descriptive and prescriptive attitudes?
I know that post and don’t think it invalidates my point. Its main point is that for a proposition to be meaningful it is not necessary that it can be verified directly, but it is often sufficient if the proposition is part of a logically coherent theory that is tested as a whole. This is not in conflict with the demand on difference in anticipated experiences from different beliefs: the hypothesis that a cheesecake materialised in the centre of the Sun may not be falsified by direct observation, but still is incompatible with the picture of the world dictated by perfectly testable and verified physical theories.
As a side note, the linked post has been criticised as strawmanning logical positivism.
I know that post and don’t think it invalidates my point. Its main point is that for a proposition to be meaningful it is not necessary that it can be verified directly, but it is often sufficient if the proposition is part of a logically coherent theory that is tested as a whole.
So what testable logically coherent theory is your main point a part of?
Do you want to read about the testable consequences of the belief that beliefs should have testable consequences, or was your question only rhetorical?
Do you want to read about the testable consequences of the belief that beliefs should have testable consequences, or was your question only rhetorical?
One testable consequence of the belief that beliefs should have testable consequences (let’s call it T) is that people who believe T will have higher percentage of true beliefs than people who don’t. Although this is not much a consequence of T as a more precise reformulation thereof.
Well, yes. That was basically pragmatist’s answer to my question which I have accepted.
It is a little bit disturbing since I am forced to give metaphysics more credit that I used to. The possible way out is to limit the testability criterion only to direct logical consequences of beliefs so that it doesn’t apply to “psychological” consequences of form “believing X increases likelihood of believing Y (even if there is no logical connection between X and Y)”. This might be a good idea but I am not sure where precisely to draw the line between direct and psychological consequences of beliefs.
To have an objective criterion for evaluating ideas in case my intuition is ifluenced by bias. To find out what exactly makes most metaphysics appear unsatisfactory and empty to me. Why are people concerned with formalising epistemology, after all?
By the way, you don’t need to link to the Sequences articles for me, I have read them all.
If I aim to apply the criterion “a theory is worthy only if it has direct logical testable consequences”, I better know what do I mean by “direct consequence”.
This seems to be taking down a straw man, and far from “challenging a central tenet of LW: reductionism”, you perfectly describe it and expound on it, if a bit wordily. At least in my mind, it’s very obvious that physical ‘law’ is a map-level concept. Physicists themselves have noticed that for a map-level concept, physical ‘law’ fits the territory so amazingly well, that they have written articles such as “The Unreasonable Effectiveness of Mathematics in the Natural Sciences”
http://www.dartmouth.edu/~matc/MathDrama/reading/Wigner.html
I don’t claim that this post in particular challenges the consensus (at least, I don’t intend to claim that, but I can see how my phrasing in the intro suggests it). It’s mostly just setup. I think the LW consensus is probably closer to what I call “explanatory reductionism” at the end of this post, but attacking that position required that I make it clear how I think about laws of nature. The ultimate position I want to defend is that the only tenable reductionism is the extremely weak mereological kind. Surely this is different from the position generally advocated here.
That said, I don’t think the position I’m attacking in this post is a straw man. As I point out, Paul Davies (hardly a fringe figure in physics) explicitly embraces it. He also says (in the linked excerpt) that “most physicists working on fundamental topics inhabit the prescriptive camp, even if they won’t own up to it explicitly.” In addition, I’ve seen nomic reductionism defended (and upvoted) on LW more than once. As an example, see some of the comments on this thread. Even people who would, if pressed, agree that laws are description often unconsciously infer things that only work if you think of laws as rules.
Do you think the points made in this post are common enough knowledge around here for the post to be of not much use?
That would be the bit where you said “This is the first in a planned series of posts challenging a central tenet of the LessWrong consensus”. When you say that you’re challenging the consensus, it appears to the reader as though you’re challenging the consensus.
Hence my parenthetical concession in the grandparent. But you’re right, I should edit the post itself. Doing that right now.
I hereby nominate this for the 2012 Understatement Award.
How was it an understatement?
I acknowledge that it feels like one when you read it, but defining that way lies madness! Just ask the words “ironic” and “literally”.
I agree with David_Gerard: when I say I’m doing something, it appears to the reader as though I’m doing that thing.
I would also agree with various more-strongly-worded equivalents, such as “when I say I’m doing X in a series of acts that includes Y, it’s disingenuous to later claim that Y wasn’t intended to do X.”
Hence, understatement. That is, an expression worded less strongly than, in my opinion, the situation justifies.
Is “challenge the consensus” a performative utterance? By saying “I challenge the consensus regarding foo”, do you thereby challenge the consensus regarding foo?
Consider: If I said, “I challenge the Less Wrong consensus that 2 + 2 = 5. I assert that it’s 4,” by saying this I wouldn’t actually challenge a consensus that 2 + 2 = 5, because there isn’t one to challenge. Rather, all I would be doing is setting up a straw man: falsely asserting the existence of a consensus, and then disagreeing with that imagined consensus.
What is, actually, the difference between laws as rules and laws as descriptions of regularity, except the choice of language? There is in fact a pretty strong LW consensus that beliefs should be distinguishable from each other by different anticipated experiences; I am not sure whether nomic and mereological reductionism predict different observations.
(I agree with the gist of the post, in the sense that it is more elegant to view physical laws as descriptions of regularities in observed universe, rather than rules that push matter around.)
See my comment here. Thinking of laws as rules vs. descriptions may not predict different observations, but they do lead to different cognitive attitudes about scientific inquiry and explanation (e.g. we often think of rules as explaining behavior, but we don’t think of descriptions as explanatory in the same way), and in this regard I think the prescriptive perspective is a recipe for confusion. One could say the same thing about the kind of empiricism advocated on LW as opposed to rationalism (of the traditional philosophical variety, not the LW variety). These two philosophical stances don’t predict different observations, but that doesn’t mean the choice between them is merely a linguistic one. They are associated with different pragmatic attitudes, and the empiricist attitude is a lot more valuable that the rationalist one.
What concrete differences in e.g. scientific achievements would you expect to stem from the difference between the descriptive and prescriptive attitudes?
See this comment for an example.
Not really.
I know that post and don’t think it invalidates my point. Its main point is that for a proposition to be meaningful it is not necessary that it can be verified directly, but it is often sufficient if the proposition is part of a logically coherent theory that is tested as a whole. This is not in conflict with the demand on difference in anticipated experiences from different beliefs: the hypothesis that a cheesecake materialised in the centre of the Sun may not be falsified by direct observation, but still is incompatible with the picture of the world dictated by perfectly testable and verified physical theories.
As a side note, the linked post has been criticised as strawmanning logical positivism.
So what testable logically coherent theory is your main point a part of?
Do you want to read about the testable consequences of the belief that beliefs should have testable consequences, or was your question only rhetorical?
I’d actually find that fascinating :)
Are they similar to this?
One testable consequence of the belief that beliefs should have testable consequences (let’s call it T) is that people who believe T will have higher percentage of true beliefs than people who don’t. Although this is not much a consequence of T as a more precise reformulation thereof.
You do realize you can apply that procedure to give any metaphysical belief testable consequences, including the one you were asking about here.
Well, yes. That was basically pragmatist’s answer to my question which I have accepted.
It is a little bit disturbing since I am forced to give metaphysics more credit that I used to. The possible way out is to limit the testability criterion only to direct logical consequences of beliefs so that it doesn’t apply to “psychological” consequences of form “believing X increases likelihood of believing Y (even if there is no logical connection between X and Y)”. This might be a good idea but I am not sure where precisely to draw the line between direct and psychological consequences of beliefs.
Why do you care?
To have an objective criterion for evaluating ideas in case my intuition is ifluenced by bias. To find out what exactly makes most metaphysics appear unsatisfactory and empty to me. Why are people concerned with formalising epistemology, after all?
By the way, you don’t need to link to the Sequences articles for me, I have read them all.
What do any of those have to do with where you “draw the line between direct and psychological consequences of beliefs”?
If I aim to apply the criterion “a theory is worthy only if it has direct logical testable consequences”, I better know what do I mean by “direct consequence”.