And yet Newton didn’t develop the theory of gravitation as part of his quest to make cheaper widgets. Nor did Einstein develop relativity because there was anything practical that classical mechanics couldn’t do.
Looking for truth is not a fancy way of saying “looking for the first half-assed solution you can find so you can feel like you know something true and go back to what you were doing before.”
The easiest way to get a bundle of beliefs that completely dissolve previously mysterious questions, don’t contradict one another, don’t contradict experience, don’t contain sacred mysteries, and don’t force you into dark side epistemology to maintain them—as far as I know the easiest way to get such beliefs is to believe things that are true.
If there’s some other weird attractor state of beliefs that also fulfills those requirements, I guess I risk falling into it. But then again, so do you—such beliefs would have to predict experience as successfully as the truth, which means they would have to give you the same widget-making capacity as true beliefs.
Despite starting out at Less Wrong as a believer in the “rationalism helps you win!” school, the more I read the less I think going from the sort of person who’s read all of Overcoming Bias to the sort of person who’s read OB plus all of Less Wrong is going to grant you any hugely significant extra real-world-winning capacity. I should make a post on this sometime.
Never mind widgets. I overemphasized business in the original post. Any kind of reality check will do. Newton had a reality check, what about us? Not many accurate predictions here. Choose any real-world metric that suits you—just don’t degenerate into “what biases have you overcome today” soft-science bullshit.
I once joked that science has four levels, high to low: “this works”, “this is true”, “this sounds true”, “this sounds neat”. We here are still at number three, no?
I think you have too limited a picture of what searching for truth entails, and that we don’t have as great a difference between our views as you think.
Newton and Einstein used rationality to seek truth and bring unity to experience, not for practical results. But they were both smart enough to know they’d better check their results against experience, or they’d get the wrong answer and never be able to move further. If we’re smart, we’ll do the same, whether we’re after truth or whatever.
Someone once said there were two kinds of rich people—those who really like having luxury goods, and those for whom money is just a way to keep score. The same could apply to rationalists; there are those who want some specific practical goal or predictive ability, and there are others for whom the ability to achieve practical goals or make predictions is a way to keep score. Einstein was happy to hear his theory successfully predicted the path of light during an eclipse, I’m sure, but not because he was in it for the eclipse-light-predicting.
You’re right, we are more or less in agreement. The expression “to keep score” captures the topic perfectly. Pickup artists have attained a very accurate/predictive view of female mating psychology because they keep score. :-) I’d love to have something similarly objective for rationalism.
Newton and Einstein used rationality to seek truth and bring unity to experience, not for practical results. But they were both smart enough to know they’d better check their results against experience, or they’d get the wrong answer and never be able to move further.
In 1919, Sir Arthur Eddington led expeditions to Brazil and to the island of Principe, aiming to observe solar eclipses and thereby test an experimental prediction of Einstein’s novel theory of General Relativity. A journalist asked Einstein what he would do if Eddington’s observations failed to match his theory. Einstein famously replied: “Then I would feel sorry for the good Lord. The theory is correct.”
It seems like a rather foolhardy statement, defying the trope of Traditional Reality that experiment above all is sovereign. Einstein seems possessed of an arrogance so great that he would refuse to bend his neck and submit to Nature’s answer, as scientists must do. Who can know that the theory is correct, in advance of experimental test?
Yes. I’ve been a semi-regular reader of OCB for about a year. I think it’s an interesting blog. But have I learned anything useful from it? Has it made any practical difference in the choices I make, either day-to-day or longterm? The answer is no. Admittedly, this may be my own fault. But I recall a post, not too long ago, soliciting people’s feedback on “the most important thing you learned from OCB in the past year,” or something of that sort. And while there were lots of people excitedly posting about how much OCB has taught them, the examples they gave were along the lines of “I learned the power of fundamental attribution error!” or “I learned the importance of continually adjusting my priors!” with curiously few examples of real differences OCB made in anyone’s practical choices. This raises the question: if tweaking our rationality has no appreciable affect on anything, then how can we say we’re really tweaking our rationality at all? Perhaps we’re just swapping new explanations for fundamentally irrational processes that are far too buried and obscure to be accessible to us.
That said, I think things like the recent posts on akrasia are strong moves in the right direction. Intellectually interesting, but with easy to grasp real-world implications.
This sounds an awful lot like one of the examples I gave above. Ok, so you’re focused on “risk reduction” and “reducing akrasia.” So what’s that mean? You’ve decided to buckle-up, wear sunscreen, and not be so lazy? Can’t I get that from Reader’s Digest or my mom?
Telling people to buckle up is nothing special. Successfully persuading people to buckle up—helping people understand and fix the internal sources of error that stood in the way of doing so in the past—will save a life if you can do it enough.
The problem is that even though learning to identify and avoid certain biases will affect your behavior, there’s no easy way to articulate those effects. It comes mainly from things not done, not things done.
For instance, upon hearing a fallacious argument, being aware of its fallacies causes the hearer not to believe in it, where he previously would have. Or if he thinks something on his own—previously a bias would have caused him to think a certain thought, which would have led to a certain action. Now, having learned to identify the bias, he doesn’t even generate that thought, but instead another, which leads to him taking a different action. While these things do certainly have an effect, they’re too subtle to identify. You’re not going to know the thoughts you avoided (even if you can try to guess), only the ones you’ve actually thought.
I feel this has largely been the case for me. My behavior has certainly been affected because I now think more clearly. That, I’m pretty certain of. But can I give any concrete examples? I’m afraid not. The effect is on a too subtle of a level for me to properly observe. But that doesn’t mean there aren’t any concrete examples, it only means I can’t verbalize them.
This debate has already played out in attacking and defending Pragmatism.
A lot of the rubrics by which to judge whether or not rationalism wins or whether or not rationalism is an end in itself involve assigning meaning and value on a very abstract level. Eliezer’s posts outline a reductionist, materialist standpoint with some strong beliefs about following the links of causality. Rationalism follows, but rationalism isn’t going to prove itself true.
Deciding that rationalism is the best answer for your axiomatic belief system requires taking a metaphysical stand; I think that if you are looking for a definite metaphysical reason that you should practice rationalism, then you are interested in something that the practice of rationalism is not going to help much.
“Rationalism” as compared to what? Mysticism? Random decision-making? Of course rational behavior is going to be by far the best choice for achieving one’s particular ends. I wasn’t questioning the entire concept of rationalism, which clearly has been the driving force behind human progress for all of history. I was questioning how much making “tweaks”—the kind discussed here and on OCB—can do for us. Or have done for us. Put differently, is perseverating on rationality per se worth my time? Can anyone show that paying special attention to rationality has measurable results, controlling for IQ?
Mysticism and random decision making are both acceptable and highly successful methods of making decisions; most of human history has relied on those two… we still rely on them. If you are a consequentialist, you can ignore the process and just rate the outcome; who cares why nice hair is correlated with success -it just is! Why does democracy work?
What makes rationalism worth the time is probably your regard for the process itself or for its outcomes. If its the outcomes then you might want to consider other options; following your biases and desires almost blindly works out pretty well for most people.
If there’s some other weird attractor state of beliefs that also fulfills those requirements, I guess I risk falling into it. But then again, so do you—such beliefs would have to predict experience as successfully as the truth, which means they would have to give you the same widget-making capacity as true beliefs.
There are plenty of things like this—engineering models, heuristics, etc. You don’t have to have a “true” map to have a “useful” map. An idealized right-angle, not-to-scale map of a city which nonetheless allowed you to logically navigate from point A to point B would be “useful” even if not “true” or “accurate” in certain senses.
Meanwhile, if you wait around for a “true” map, you’re not going anywhere.
But such maps are only useful insofar as they are true. For example, the London Tube Map claims to be a useful representation of which stations are on which lines. It’s useful in doing that because it is correct in its domain—every station it says is on the Piccadilly Line really is on the Piccadilly Line. It doesn’t claim to accurately represent distance, and anyone who tried to use it to determine distances would quickly get some surprises.
There the danger doesn’t seem to be getting something that isn’t the truth, the danger is stopping at something that’s just true enough for a certain purpose, and no more.
And a seeker of truth seems less likely to get stuck there than a seeker of win—witness classical mechanics, which is still close enough to be useful for everything practical, versus relativity, which exists because Einstein wouldn’t accept a theory which worked well enough but had a few little loose ends.
There the danger doesn’t seem to be getting something that isn’t the truth, the danger is stopping at something that’s just true enough for a certain purpose, and no more.
Why is that bad?
And a seeker of truth seems less likely to get stuck there than a seeker of win—witness classical mechanics, which is still close enough to be useful for everything practical, versus relativity, which exists because Einstein wouldn’t accept a theory which worked well enough but had a few little loose ends.
How has relativity made us better off? If you want to pursue truth because you like truth, that’s great—it’s a “win” for you. But if you only need the truth to get to something else, it’s not a win to add useless knowledge.
Are you sure that this isn’t all about signaling being a truth-seeker? (i.e. “Truth-Seeking Isn’t About The Truth”)
After all, credibly signaling that you value the truth could make you a valuable ally, be considered a neutral judge, etc. etc. For these reasons, credibly valuing the truth above all else might be beneficial… for reasons not having anything to do with actually getting to the truth.
So, if you’re saying we should seek truth just because it’s the truth, and not because it brings practical benefit or pleasure or sends good signals, then what is the use of seeking truth?
What if it actually doesn’t and their craft are really only limited by how fast their typical UFO-discs can spin without killing the crew inside (apparently they are spongelike inside) since unlike us they already know how to create anti-gravity to pull their ships forward? In that case the reason we are not dead yet is because they needed to figure out how to construct fast enough motherships capable of a full-scale earth invasion after we apparently killed most of their messengers. In that case our strategy should be to pool resources into defending the earth against an alien invasion and make it so costly to them that they will instead consider a trade agreement with us, which may at some point be more attractive to them than an all-out war. Trade is the way forward. Of course that is only conjecture, I don’t really know if they exist, but assigning literally zero probability to this may be stupid.
There the danger doesn’t seem to be getting something that isn’t the truth, the danger is stopping at something that’s just true enough for a certain purpose, and no more.
Why is that bad?
It can be bad if you mistakenly rest at a local maximum in your results.
You take a theory that is close enough to being true that it gives you results. Let’s say, you make $1000 a month from a certain theory of web advertising on your website. If you worked a little harder to uncover the truth, you might confuse yourself and go down to $500 a month. Yet if you worked even harder, you might make $2000 a month. The $1000 was a local maximum. If so, seeking the truth could help you find it out, if we assume that (on average at least) more truth leads to more results in solving real world problems.
Are you sure that this isn’t all about signaling being a truth-seeker?
Pretty sure. If I wanted to signal, I’d be a lot more high-falutin about it. Actually, my comments do sound a bit high-falutin’ (I was looking for a better word than “truth seeker”, but couldn’t find one) but that wasn’t exactly what I wanted to express. The untangling-wires metaphor works a little better. Nominull’s “I only seek to be right because I hate being wrong.” works too. It’s less of a “I vow to follow the pure light of Truth though it lead me to the very pits of Hell” and more of an “Aaargh, my brain feels so muddled right now, how do I clear this up?”
Also, this would be a terrible community to signal truth-seeking in, considering how entrenched the “rationality as win” metaphor is. As I mentioned in the hair example, I think a lot more people here are signaling a burning interest in real-world application than really have one.
So, if you’re saying we should seek truth just because it’s the truth, and not because it brings practical benefit or pleasure or sends good signals, then what is the use of seeking truth?
Um...this line of argument applies to everything, doesn’t it? What is the use of seeking money, if it doesn’t bring pleasure or send good signals? What is the use of seeking love, if it doesn’t bring pleasure or send good signals? What is the use of seeking ‘practical benefits’, if they don’t bring pleasure or send good signals?
Darned if I know. That’s the way my utility function works. And it certainly is mediated by pleasure and good signals, but I prefer not to say it’s about pleasure and good signals because I’d rather not be turned into orgasmium just yet.
Do you really believe that you engage in Truth-Seeking for utilitarian reasons? I get the impression that you don’t really believe that.
Would you be willing to to enter a computer simulation where you got to investigate higher math puzzles (or metaphysics) with no applications? Spend your days in a fantastic and never-ending Truth-Seeking project (we’ll throw great sex, food and housing into the holodeck for you as well)?
I liked this better at the beginning when you were prodding people who say that they see rationalism as a means to an end! You seem to be going back to consequentialism!
I don’t believe that rationalists WIN because I don’t believe that winning WINS
Would you be willing to to enter a computer simulation where you got to investigate higher math puzzles (or metaphysics) with no applications? Spend your days in a fantastic and never-ending Truth-Seeking project (we’ll throw great sex, food and housing into the holodeck for you as well)?
Maybe a few videogames (or other forms of entertainment in addition to sex) and this sounds like a very sweet deal.
That’s possible and probably partially accurate; if there were more posts taking the form “I believe X because...” on Less Wrong, I might be more open to the idea that people are doing that.
Ciphergoth:
Also, this would be a terrible community to signal truth-seeking in, considering how entrenched the “rationality as win” metaphor is. As I mentioned in the hair example, I think a lot more people here are signaling a burning interest in real-world application than really have one.
I just wanted to get Yvain’s opinion about how much value from posting on Less Wrong was coming from signaling. Yvain suggested that this was not his or her main goal and that LW would be a uniquely poor place to attempt it. I personally doubt both of those points, but I was hoping to get some clarification since the comments about signaling and the nature of truth-seeking don’t seem to be part of a system of beliefs.
Are you worried that signaling truth-seeking is legitimate enough?
I inherently value humanity’s success in understanding as much as we do, but I don’t discount the utility much in time; I don’t much mind if we learn something later rather than earlier.
As a result, it’s not that important to me to try to serve that end directly; I think it’s a bigger gain to serve it indirectly, by trying to reduce the probability of extinction in the next hundred years. This also serves several other goals I value.
There the danger doesn’t seem to be getting something that isn’t the truth, the danger is stopping at something that’s just true enough for a certain purpose, and no more.
Why is that bad?
It’s not, if you know you’re doing it.
This is an interesting debate.
I believe all the truth we’ll ever get will be like the tube map: good for purpose X, and no more. Or at least, bad for purpose Y.
Wanting more is surrendering to metaphysics, realism, platonism, absolutism—whatever you wish to call it.
I believe platonism shaped first the Hellenistic world, then christianity (Paul was of Greek culture, the whole new testament was written in Greek, and books like the one of John are soaked in primary platonic philosophy), and rules until today. It also really sucks. Because it makes people to not want to be less wrong. They want to be completely, absolutely right, in a way you can never claim with the help of mere rationality. Only delusion can help with that.
Wanting more is surrendering to metaphysics, realism, platonism, absolutism—whatever you wish to call it. ….
Because it makes people to not want to be less wrong. They want to be completely, absolutely right, in a way you can never claim with the help of mere rationality. Only delusion can help with that.
The Truth Pilgrim’s progress goes like this:
Slightly Rational → Less Wrong → Delusional
Yep—and that’s probably as close to an “absolute truth” as you can get. Robert Anton Wilson’s “Quantum Psychology” (bad title, awesome book, some parts approach GEB in awesomeness) has some very good information along these lines, along with lots of “class exercises” that might be useful for developing an instrumental rationality group.
Good point! Though inasmuch as one can see the history of ideas as a conflict between Plato and Aristotle (not an entirely fruitless endeavor) it’s worth noting that Aristotle is still alive and kicking.
In a TV tube, an electron gun shoots electrons at a cathode ray tube. An electromagnet bends these rays in a precisely-timed manner to make them scan the screen. Since they’re travelling at relativistic speeds, they are time-dilated from our point of view; and you need to use relativity to bend them the right amount.
Since they’re travelling at relativistic speeds, they are time-dilated from our point of view; and you need to use relativity to bend them the right amount.
And engineers wouldn’t have discovered this fact without relativity theory?
I’m a bit stunned by this question, so maybe it was intended to be rhetorical. But if not, I believe things like GPS relies on relativity. And my life has been so much better ever since I got an iPhone with a GPS receiver. It integrates with google maps and the local public transportation system to actually tell me what time I should leave my house in order to be able to arrive on time at another location at a specific time by crossreferencing the departure and arrival times of all the subwaytrains and busses in the system.
I’m a bit stunned by this question, so maybe it was intended to be rhetorical. But if not, I believe things like GPS relies on relativity. And my life has been so much better ever since I got an iPhone with a GPS receiver.
It was intended to point out that Einstein didn’t seek out relativity in order to produce useful results, and that, with the possible exception of nuclear energy and atomic bombs, it’s quite likely that, had Einstein not come along, most of the “practical” uses of relativity today would’ve prompted engineers to add time dilation fudge factors to their plans, and then inspired some not-Einstein physicists to figure out what the heck was going on.
In other words, there was really no danger of “stopping at something that’s just true enough for a certain purpose, and no more”, in a way that would actually produce a bad result, or deprive us of a good one for more than a limited time.
In other words, Einstein’s truth-seeking was about his personal desire to “know God’s thoughts”, not to improve the lot of humanity by helping us get iPhone GPS receivers. And as I said earlier in this thread, wanting truth because you’re curious is all well and good, but in the end it’s the search for practical models that drives progress. Science having “true” models saves engineers time and mistakes getting started, but they still have to work out practical models anyway… and sometimes need to be able to deal with things that the scientists haven’t even started figuring out yet.
Case in point: hypnotism. Scientists still don’t have a “true” model for it, AFAIK, but hypnotists have plenty of practical models for it.
Never mind widgets. Maybe I overemphasized business in the original post. Any kind of reality check will do. Newton had a reality check, what about us? I don’t read many accurate predictions here. Choose any real-world metric that suits you—just don’t degenerate into “what biases have you overcome today” soft-science bullshit.
I once joked that science has four levels, high to low: “this works”, “this is true”, “this sounds true” and “this sounds neat”. We here are still at number three, no?
And yet Newton didn’t develop the theory of gravitation as part of his quest to make cheaper widgets. Nor did Einstein develop relativity because there was anything practical that classical mechanics couldn’t do.
Looking for truth is not a fancy way of saying “looking for the first half-assed solution you can find so you can feel like you know something true and go back to what you were doing before.”
The easiest way to get a bundle of beliefs that completely dissolve previously mysterious questions, don’t contradict one another, don’t contradict experience, don’t contain sacred mysteries, and don’t force you into dark side epistemology to maintain them—as far as I know the easiest way to get such beliefs is to believe things that are true.
If there’s some other weird attractor state of beliefs that also fulfills those requirements, I guess I risk falling into it. But then again, so do you—such beliefs would have to predict experience as successfully as the truth, which means they would have to give you the same widget-making capacity as true beliefs.
Despite starting out at Less Wrong as a believer in the “rationalism helps you win!” school, the more I read the less I think going from the sort of person who’s read all of Overcoming Bias to the sort of person who’s read OB plus all of Less Wrong is going to grant you any hugely significant extra real-world-winning capacity. I should make a post on this sometime.
Never mind widgets. I overemphasized business in the original post. Any kind of reality check will do. Newton had a reality check, what about us? Not many accurate predictions here. Choose any real-world metric that suits you—just don’t degenerate into “what biases have you overcome today” soft-science bullshit.
I once joked that science has four levels, high to low: “this works”, “this is true”, “this sounds true”, “this sounds neat”. We here are still at number three, no?
I think you have too limited a picture of what searching for truth entails, and that we don’t have as great a difference between our views as you think.
Newton and Einstein used rationality to seek truth and bring unity to experience, not for practical results. But they were both smart enough to know they’d better check their results against experience, or they’d get the wrong answer and never be able to move further. If we’re smart, we’ll do the same, whether we’re after truth or whatever.
Someone once said there were two kinds of rich people—those who really like having luxury goods, and those for whom money is just a way to keep score. The same could apply to rationalists; there are those who want some specific practical goal or predictive ability, and there are others for whom the ability to achieve practical goals or make predictions is a way to keep score. Einstein was happy to hear his theory successfully predicted the path of light during an eclipse, I’m sure, but not because he was in it for the eclipse-light-predicting.
You’re right, we are more or less in agreement. The expression “to keep score” captures the topic perfectly. Pickup artists have attained a very accurate/predictive view of female mating psychology because they keep score. :-) I’d love to have something similarly objective for rationalism.
According to EY,
A typo in the Yudkowsky’s article: Traditional Reality → Traditional Rationality.
Yes. I’ve been a semi-regular reader of OCB for about a year. I think it’s an interesting blog. But have I learned anything useful from it? Has it made any practical difference in the choices I make, either day-to-day or longterm? The answer is no. Admittedly, this may be my own fault. But I recall a post, not too long ago, soliciting people’s feedback on “the most important thing you learned from OCB in the past year,” or something of that sort. And while there were lots of people excitedly posting about how much OCB has taught them, the examples they gave were along the lines of “I learned the power of fundamental attribution error!” or “I learned the importance of continually adjusting my priors!” with curiously few examples of real differences OCB made in anyone’s practical choices. This raises the question: if tweaking our rationality has no appreciable affect on anything, then how can we say we’re really tweaking our rationality at all? Perhaps we’re just swapping new explanations for fundamentally irrational processes that are far too buried and obscure to be accessible to us.
That said, I think things like the recent posts on akrasia are strong moves in the right direction. Intellectually interesting, but with easy to grasp real-world implications.
Ob has changed people’s practical lives in some major ways. Not all of these are mine personally:
“I donated more money to anti aging, risk reduction, etc”
“I signed up for cryonics.”
“I wear a seatbelt in a taxi even when no one else does.”
“I stopped going to church but started hanging out socially with aspiring rationalists.”
“I decided rationality works and started writing down my goals and pathways to them.”
“I decided it’s important for me to think carefully about what my ultimate values are.”
Yes! Or even further, “I am now focusing my life on risk reduction and have significantly reduced akrasia in all facets of my life.”
This sounds an awful lot like one of the examples I gave above. Ok, so you’re focused on “risk reduction” and “reducing akrasia.” So what’s that mean? You’ve decided to buckle-up, wear sunscreen, and not be so lazy? Can’t I get that from Reader’s Digest or my mom?
Telling people to buckle up is nothing special. Successfully persuading people to buckle up—helping people understand and fix the internal sources of error that stood in the way of doing so in the past—will save a life if you can do it enough.
The problem is that even though learning to identify and avoid certain biases will affect your behavior, there’s no easy way to articulate those effects. It comes mainly from things not done, not things done.
For instance, upon hearing a fallacious argument, being aware of its fallacies causes the hearer not to believe in it, where he previously would have. Or if he thinks something on his own—previously a bias would have caused him to think a certain thought, which would have led to a certain action. Now, having learned to identify the bias, he doesn’t even generate that thought, but instead another, which leads to him taking a different action. While these things do certainly have an effect, they’re too subtle to identify. You’re not going to know the thoughts you avoided (even if you can try to guess), only the ones you’ve actually thought.
I feel this has largely been the case for me. My behavior has certainly been affected because I now think more clearly. That, I’m pretty certain of. But can I give any concrete examples? I’m afraid not. The effect is on a too subtle of a level for me to properly observe. But that doesn’t mean there aren’t any concrete examples, it only means I can’t verbalize them.
This debate has already played out in attacking and defending Pragmatism.
A lot of the rubrics by which to judge whether or not rationalism wins or whether or not rationalism is an end in itself involve assigning meaning and value on a very abstract level. Eliezer’s posts outline a reductionist, materialist standpoint with some strong beliefs about following the links of causality. Rationalism follows, but rationalism isn’t going to prove itself true.
Deciding that rationalism is the best answer for your axiomatic belief system requires taking a metaphysical stand; I think that if you are looking for a definite metaphysical reason that you should practice rationalism, then you are interested in something that the practice of rationalism is not going to help much.
“Rationalism” as compared to what? Mysticism? Random decision-making? Of course rational behavior is going to be by far the best choice for achieving one’s particular ends. I wasn’t questioning the entire concept of rationalism, which clearly has been the driving force behind human progress for all of history. I was questioning how much making “tweaks”—the kind discussed here and on OCB—can do for us. Or have done for us. Put differently, is perseverating on rationality per se worth my time? Can anyone show that paying special attention to rationality has measurable results, controlling for IQ?
Mysticism and random decision making are both acceptable and highly successful methods of making decisions; most of human history has relied on those two… we still rely on them. If you are a consequentialist, you can ignore the process and just rate the outcome; who cares why nice hair is correlated with success -it just is! Why does democracy work?
What makes rationalism worth the time is probably your regard for the process itself or for its outcomes. If its the outcomes then you might want to consider other options; following your biases and desires almost blindly works out pretty well for most people.
There are plenty of things like this—engineering models, heuristics, etc. You don’t have to have a “true” map to have a “useful” map. An idealized right-angle, not-to-scale map of a city which nonetheless allowed you to logically navigate from point A to point B would be “useful” even if not “true” or “accurate” in certain senses.
Meanwhile, if you wait around for a “true” map, you’re not going anywhere.
But such maps are only useful insofar as they are true. For example, the London Tube Map claims to be a useful representation of which stations are on which lines. It’s useful in doing that because it is correct in its domain—every station it says is on the Piccadilly Line really is on the Piccadilly Line. It doesn’t claim to accurately represent distance, and anyone who tried to use it to determine distances would quickly get some surprises.
There the danger doesn’t seem to be getting something that isn’t the truth, the danger is stopping at something that’s just true enough for a certain purpose, and no more.
And a seeker of truth seems less likely to get stuck there than a seeker of win—witness classical mechanics, which is still close enough to be useful for everything practical, versus relativity, which exists because Einstein wouldn’t accept a theory which worked well enough but had a few little loose ends.
Why is that bad?
How has relativity made us better off? If you want to pursue truth because you like truth, that’s great—it’s a “win” for you. But if you only need the truth to get to something else, it’s not a win to add useless knowledge.
Are you sure that this isn’t all about signaling being a truth-seeker? (i.e. “Truth-Seeking Isn’t About The Truth”)
After all, credibly signaling that you value the truth could make you a valuable ally, be considered a neutral judge, etc. etc. For these reasons, credibly valuing the truth above all else might be beneficial… for reasons not having anything to do with actually getting to the truth.
So, if you’re saying we should seek truth just because it’s the truth, and not because it brings practical benefit or pleasure or sends good signals, then what is the use of seeking truth?
The lightspeed limit stops aliens from eating us.
What if it actually doesn’t and their craft are really only limited by how fast their typical UFO-discs can spin without killing the crew inside (apparently they are spongelike inside) since unlike us they already know how to create anti-gravity to pull their ships forward? In that case the reason we are not dead yet is because they needed to figure out how to construct fast enough motherships capable of a full-scale earth invasion after we apparently killed most of their messengers. In that case our strategy should be to pool resources into defending the earth against an alien invasion and make it so costly to them that they will instead consider a trade agreement with us, which may at some point be more attractive to them than an all-out war. Trade is the way forward. Of course that is only conjecture, I don’t really know if they exist, but assigning literally zero probability to this may be stupid.
It can be bad if you mistakenly rest at a local maximum in your results.
You take a theory that is close enough to being true that it gives you results. Let’s say, you make $1000 a month from a certain theory of web advertising on your website. If you worked a little harder to uncover the truth, you might confuse yourself and go down to $500 a month. Yet if you worked even harder, you might make $2000 a month. The $1000 was a local maximum. If so, seeking the truth could help you find it out, if we assume that (on average at least) more truth leads to more results in solving real world problems.
It’s not, if you know you’re doing it.
Pretty sure. If I wanted to signal, I’d be a lot more high-falutin about it. Actually, my comments do sound a bit high-falutin’ (I was looking for a better word than “truth seeker”, but couldn’t find one) but that wasn’t exactly what I wanted to express. The untangling-wires metaphor works a little better. Nominull’s “I only seek to be right because I hate being wrong.” works too. It’s less of a “I vow to follow the pure light of Truth though it lead me to the very pits of Hell” and more of an “Aaargh, my brain feels so muddled right now, how do I clear this up?”
Also, this would be a terrible community to signal truth-seeking in, considering how entrenched the “rationality as win” metaphor is. As I mentioned in the hair example, I think a lot more people here are signaling a burning interest in real-world application than really have one.
Um...this line of argument applies to everything, doesn’t it? What is the use of seeking money, if it doesn’t bring pleasure or send good signals? What is the use of seeking love, if it doesn’t bring pleasure or send good signals? What is the use of seeking ‘practical benefits’, if they don’t bring pleasure or send good signals?
Darned if I know. That’s the way my utility function works. And it certainly is mediated by pleasure and good signals, but I prefer not to say it’s about pleasure and good signals because I’d rather not be turned into orgasmium just yet.
Yeah, “rationalists WIN!” is the most widely misued EYism on all of LessWrong.com.
Yvain:
Do you really believe that you engage in Truth-Seeking for utilitarian reasons? I get the impression that you don’t really believe that.
Would you be willing to to enter a computer simulation where you got to investigate higher math puzzles (or metaphysics) with no applications? Spend your days in a fantastic and never-ending Truth-Seeking project (we’ll throw great sex, food and housing into the holodeck for you as well)?
I liked this better at the beginning when you were prodding people who say that they see rationalism as a means to an end! You seem to be going back to consequentialism!
I don’t believe that rationalists WIN because I don’t believe that winning WINS
Maybe a few videogames (or other forms of entertainment in addition to sex) and this sounds like a very sweet deal.
And you must enjoy the signal value you a little bit! You aren’t keeping your Less Wrong postings in your diary under lock and key!
logi:
That’s possible and probably partially accurate; if there were more posts taking the form “I believe X because...” on Less Wrong, I might be more open to the idea that people are doing that.
Ciphergoth:
I just wanted to get Yvain’s opinion about how much value from posting on Less Wrong was coming from signaling. Yvain suggested that this was not his or her main goal and that LW would be a uniquely poor place to attempt it. I personally doubt both of those points, but I was hoping to get some clarification since the comments about signaling and the nature of truth-seeking don’t seem to be part of a system of beliefs.
Are you worried that signaling truth-seeking is legitimate enough?
Sure, but it’s pretty clear that a lot of people are enjoying the WIN! signal too. Let’s try not to get too caught up in who is signalling what.
Even if he did not value the signal, surely you can conjecture a rational strategy of publishing beliefs in order to refine them.
I inherently value humanity’s success in understanding as much as we do, but I don’t discount the utility much in time; I don’t much mind if we learn something later rather than earlier.
As a result, it’s not that important to me to try to serve that end directly; I think it’s a bigger gain to serve it indirectly, by trying to reduce the probability of extinction in the next hundred years. This also serves several other goals I value.
This is an interesting debate. I believe all the truth we’ll ever get will be like the tube map: good for purpose X, and no more. Or at least, bad for purpose Y. Wanting more is surrendering to metaphysics, realism, platonism, absolutism—whatever you wish to call it.
I believe platonism shaped first the Hellenistic world, then christianity (Paul was of Greek culture, the whole new testament was written in Greek, and books like the one of John are soaked in primary platonic philosophy), and rules until today. It also really sucks. Because it makes people to not want to be less wrong. They want to be completely, absolutely right, in a way you can never claim with the help of mere rationality. Only delusion can help with that.
The Truth Pilgrim’s progress goes like this:
Slightly Rational → Less Wrong → Delusional
Yep—and that’s probably as close to an “absolute truth” as you can get. Robert Anton Wilson’s “Quantum Psychology” (bad title, awesome book, some parts approach GEB in awesomeness) has some very good information along these lines, along with lots of “class exercises” that might be useful for developing an instrumental rationality group.
Good point! Though inasmuch as one can see the history of ideas as a conflict between Plato and Aristotle (not an entirely fruitless endeavor) it’s worth noting that Aristotle is still alive and kicking.
In a TV tube, an electron gun shoots electrons at a cathode ray tube. An electromagnet bends these rays in a precisely-timed manner to make them scan the screen. Since they’re travelling at relativistic speeds, they are time-dilated from our point of view; and you need to use relativity to bend them the right amount.
And engineers wouldn’t have discovered this fact without relativity theory?
I’m a bit stunned by this question, so maybe it was intended to be rhetorical. But if not, I believe things like GPS relies on relativity. And my life has been so much better ever since I got an iPhone with a GPS receiver. It integrates with google maps and the local public transportation system to actually tell me what time I should leave my house in order to be able to arrive on time at another location at a specific time by crossreferencing the departure and arrival times of all the subwaytrains and busses in the system.
It was intended to point out that Einstein didn’t seek out relativity in order to produce useful results, and that, with the possible exception of nuclear energy and atomic bombs, it’s quite likely that, had Einstein not come along, most of the “practical” uses of relativity today would’ve prompted engineers to add time dilation fudge factors to their plans, and then inspired some not-Einstein physicists to figure out what the heck was going on.
In other words, there was really no danger of “stopping at something that’s just true enough for a certain purpose, and no more”, in a way that would actually produce a bad result, or deprive us of a good one for more than a limited time.
In other words, Einstein’s truth-seeking was about his personal desire to “know God’s thoughts”, not to improve the lot of humanity by helping us get iPhone GPS receivers. And as I said earlier in this thread, wanting truth because you’re curious is all well and good, but in the end it’s the search for practical models that drives progress. Science having “true” models saves engineers time and mistakes getting started, but they still have to work out practical models anyway… and sometimes need to be able to deal with things that the scientists haven’t even started figuring out yet.
Case in point: hypnotism. Scientists still don’t have a “true” model for it, AFAIK, but hypnotists have plenty of practical models for it.
Never mind widgets. Maybe I overemphasized business in the original post. Any kind of reality check will do. Newton had a reality check, what about us? I don’t read many accurate predictions here. Choose any real-world metric that suits you—just don’t degenerate into “what biases have you overcome today” soft-science bullshit.
I once joked that science has four levels, high to low: “this works”, “this is true”, “this sounds true” and “this sounds neat”. We here are still at number three, no?