8 ms·

The Limits of Rationality

87 points by _gfwu 4y ago

Neil44 4y ago
I'm reading Zen and the art of motorcycle maintainance at the moment which also pushes and pulls at the limits of reason and rationality. Some of the ideas are quite 'dense' and worth making notes over. Worth a read.
- karmakurtisaani 4y ago
  I read it some time ago, really enjoyed it. Lots of layers to it too.
- bilsbie 4y ago
  > pushes and pulls at the limits of reason and rationality Any chance you’d be up for providing a few examples? I read it but I can’t remember that.
  - 0x445442 4y ago
    It’s been a number of years since my last reading but what I remember is it pulling at the thread of the concept of quality. The author stumbles upon this while teaching an introductory writing course at college. What he discovers is that it’s difficult to reduce the act of writing to a prescribed set of rules to follow in order to produce a quality piece. And yet, somehow his students, without training or much effort, are able to judge which pieces exhibit quality. This idea literally leads the author to madness.
    - sifar 4y ago
      And that quality arises at the moment of interaction. It is simple but was a stunning realization for me. This rhymes with what The Tao Te Ching says that you can't be taught - you have to experience it in your own way.
- 0x445442 4y ago
  Fantastic book. It’s exploration of quality is amazing.
- tanseydavid 4y ago
  ZatAoMM was really a great read for me. One of my big takeaways from the book is that the main character actually managed to literally drive himself insane through hyper-rationality and over-thinking.
fictionfuture 4y ago
Nietzsche discusses this at length (the author mentions Jung, but Jung borrowed heavily from Nietzsche in this regard)… Nietzsche believed that the best decisions weren’t the most rationale ones but the ones that were “life furthering.” However he believed rationality is preferable to idealism. Therefore, best to first deal with reality, learn rationality, then make decisions with your feet on the ground
kayodelycaon 4y ago
This is why I like to keep rational thinking as a tool in my box of strategies for life. The biggest change I’ve ever had living my life was the XKCD’s 1 of today’s 10,000. https://xkcd.com/1053/ https://xkcd.com/1053/ To boil it down: “let people enjoy things”. If you like something and it doesn’t have a negative affect on your life or someone else’s, go for. Not everything you do needs to analyzed.
- starkd 4y ago
  To a point. But many joys are very destructive in and of themselves. Without any analysis, one might be tempted to snort meth all day.
  - kayodelycaon 4y ago
    I would argue meth falls under the “negative affect” I mentioned. :)
- chousuke 4y ago
  It's a bit difficult to disagree with "not everything", because that's obviously true; not everything even can be perfectly analyzed. However, I think you can (and should) try to at least maintain awareness of your behaviour and its effects, but only because if you don't, you may actually end up acting in ways that contradict your values; or if your values are ill-defined or vague, you might end up not understanding what it is that really matters to you. I think rationality is a necessary part of having a well-defined value system, even if the axioms are arbitrary. "Letting people enjoy things that aren't harming anyone" is a rational decision based on my value system. What I can use rationality for, really, is figuring out if there's something I'm doing that's actually in conflict with my own values, or if there's something I can do to nudge the world (or at least my own life) to have more of the things I value.
awillen 4y ago
This doesn't feel particularly insightful to me. "We have to realise that rationality, particularly in the form of scientific thought, cannot provide us with a value system." Yeah, that's true, but it's not the point of rationality - rationality is a system of thinking and decision making designed to optimize for the best outcome, but you have to define what's best and thus what you want to optimize for. Rationality won't provide you with a value system, but I don't think anyone who understands it would claim that it does. This is sort of like saying culinary school will teach you how to cook, but it won't pick what's for dinner.
- kiba 4y ago
  Rationality won't provide you with a value system, but I don't think anyone who understands it would claim that it does. This is sort of like saying culinary school will teach you how to cook, but it won't pick what's for dinner. Someone's belief and value system is the difference between being shunned or supported, like a gay teenager being disowned by their parent. We should definitely examine our value system.
  - awillen 4y ago
    Sure, examining value systems is good and we should certainly do that. That's not really relevant to the article or my response, though.
- licebmi__at__ 4y ago
  Plenty of people think Sam Harris understands rationality.
- Dracophoenix 4y ago
  But even the pursuit of optimality is a value, is it not? Even if you define what's best, could one then not choose to do the opposite? No individual is required to live an optimal life, and many purposely and self-righteously shun one.
  - awillen 4y ago
swayvil 4y ago
Rationality is a game. Played within a structure of rules (aka assertions, assumptions, axioms). These rules are not themselves rationally born. They are born from habit, tradition, convention, aesthetics, consensus, utility, authority... Lots of possible sources for rules. Logical consistency is one good rule. Puts the rational in rationality. It's just a game. A dream. Ungrounded. That may be obvious, but it's easy to lose yourself in a dream. (Forever even.) That's important to keep in mind.
- starkd 4y ago
  I would argue most are built on tradition, which offer the ability to build on the results of our forebears without reinventing the wheel. It's also helpful to have a little humility in the face of multiple complexities. That's why challenging tradition, while it can prove useful at times, does not come without a cost.
  - swayvil 4y ago
    Oh good point. That's 2 important powers of rationality. Communicability and buildability. That's how we build things like "chemistry". But yes, tradition. That's a big one.
- anonporridge 4y ago
  It is a game, but we judge games by the fruit they bear, and the harsh reality of life is the ultimate judge. The reason we value the games of rationality and science over the games of religion and mythology is that the former games produce significantly more power. And that's the ultimate game of life. Crudely, might makes right. Those individuals or groups that can most effectively harness various forms of power to first defend, and then extend, their claim of limited resources win the game, by simple virtue of the fact that they remain while their competition fades away. But even games like religion have been, and continue to be valuable frameworks for this ultimate game. There's a strong argument to be made that religion binds large groups of people together towards a common goal and incentivizes individuals to sacrifice the entirety of their life because they believe their rewards will come in the non existent afterlife. That's a powerful game, if you can keep the fantasy alive.
tehchromic 4y ago
It's a well meaning article. My first thought is that it's generated by AI in the service of some moderate yet far reaching religious org. "Finding our intrinsic values is no trivial task, but something that forces us to stare into the core of our humanity." I think it misses a critical topical conclusion which is that "intrinsic values" means little or nothing to many, and many folks are fairly incapable of competently doing the kind of contemplative meditating required to arrive at some set of rules like that, and so they look to others. And here's the crux of the missed point: To really resonate with our intrinsic values, we need to drop the myth that a rational economic system exists, and build one with the value of present and future human life explicitly built in. We already have a massive framework for understanding individual intrinsic values in terms of the sanctity of human life and that's the great mystery religions that have birth to our modern age. What's needed now is a planetary ethos where intrinsic human value is put in the proper context of the ecological reality of the biological and geological systems and processes that sustain us. Building that culture requires myth and magic, but also is an intensely rational project. There is really no way around it, and any philosophical text intended to shift the cultural landscape is incomplete without it.
- openfuture 4y ago
  Please join us on matrix! Starting from datalisp.is.
- jcelerier 4y ago
  > first thought is that it's generated by AI We're in the Yosemite park bear era of AI: there is considerable overlap between the capabilities of the most intelligent AI and the most stupid humans ... And that's very generous to humans
- pas 4y ago
  > sanctity of human life what's so sacrosanct about life? life is common, many argue too common. (see the overpopulation hysteria.) I'd argue there are better things that we should hold dear, like empathy and cooperation, the resilience of getting up after devastating events, our ability to cooperate in even the most abstract frameworks.
cainxinth 4y ago
It's paradoxical, but a certain amount of irrational behavior can be optimal. Consider the balance between depressive realism (seeing the world so accurately that it eats away at your hope for a better future) and positive illusions, the small lies people convince themselves of that they are better, stronger, smarter, and luckier than they are. In small amounts, self-deception and avoidance of discomfort are necessary coping mechanisms. Reality is daunting, mysterious, and random; staring into the abyss can cause despair. It's a better strategy to be slightly optimistic, even if there isn't a strong rational basis for that outlook. Taken too far, you get total self delusion, the people who live in their own reality distortion fields, but there are benefits to a modicum of irrational self confidence.
- srinivgp 4y ago
  Pretty much every direction is a local improvement given _some_ starting point, but if you broaden your overall search, the "more self-deception" direction is gonna be worse.
- kiba 4y ago
  Do people really see the world so 'accurately' that they become depressed? I suspect it's a matter of perspective rather than being a realist.
  - jhbadger 4y ago
    It's basically the argument that poets like Keats had, that science was destroying beauty because it "unweaved the rainbow" -- that is the rainbow is less beautiful if we understand why it occurs rather than being magical. I personally can't understand this argument at all -- but then I always enjoyed figuring out how magic tricks worked as well.
    - nuancebydefault 4y ago
      I also try to find out how and why things work, I suppose that's why I became an engineer. Thinking about the subject further, the beauty is not the rainbow, the beauty is the fantasy around it. For an engineer, fantasising can be a means to enhance creativity and out of the box thinking, which can lead to a solution, the pieces of the puzzle fall into place, in the end the effect can be explained logically. The journey to come to the conclusion is like a game, and is fun and satisfying. Once the solution is known, their interest will sheft to other effects, rendering the former less interesting.
sifar 4y ago
Like all fundamental things, rationality is what we say it is. Things just are, if we cannot explain them we call them irrational but that doesn't mean that they are inherently bad or to be shunned. Rational thought arises out living, it is not mechanical. I would say the triumph of our rationality has been the ability to deal with this irrationality, something which is on the decline. I would say that we won't be able to build true AI till we understand and reconcile this conflict and find a way to express it.
Joker_vD 4y ago
> Before, my enjoyment of art, music or literature came with a vague angst, urging me to try to understand why it made me feel a particular way, rather than just enjoying the experience itself. I always thought Bazarov from Turgenev's "Fathers and Sons" was a caricature nihilist/rationalist, with his "Nature is just a back wall in a worker's workshop, no need to admire its so called 'beauty'" attitude... but apparently not.
_8j50 4y ago
I would also say that another, perhaps controversial shortcoming of rational thinking is that it presumes the human mind can process and reason with all aspects of reality. But using rational thinking, isn't it rational to also presume that there could be aspects of reality which the human brain simply isn't equipped to process but are still real in that we can sense and be affected by them? This is similar to how a person born blind accepts the reality of the visible world without ever experiencing or being able to rationalize it. Or perhaps a better analogy would be ants and other insects who only think/see in 2d (or so I heard, correct me if this is wrong).
sanroot99 4y ago
I don't think this article was rational /s
e63f67dd-065b 4y ago
It's not helpful to think of rationality as a set of rules. I think rationality is better described, in my view, as an explicit acknowledgement of a search for optima in their own chosen utility functions. Rationality is not thinking like a robot; it's asking the question: if I have a vaguely-defined utility function f, how would a rational optimiser optimise for f? Would you march towards a local optima, or try to look for other better local optima? You see an optima in the distance, what's the shortest path that will lead you there? Other rational agents have function y that conflict with your function; how best to resolve/navigate said conflict? What path traverses along high-value areas such that the expected utility of your path over time is highest? As OP points out, our utility functions are poorly defined. That doesn't mean, however, that we cannot apply mathematical/rational techniques to it. A big part is explicitly acknowledging our own biases and trying to restructure our thinking in a way that best avoids them. A good heuristic, for example, is that having a more accurate view of reality generally allows for more efficient searches of your utility space. Thus, updating your view of the world to better conform to reality is a positive EV investment.
- nuancebydefault 4y ago
  IMHO you missed the point of the article... it says that rationality on itself only can lead to progress in the light of virtues (rules of what is 'right'). Without those as a starting point, there can be no reasoning. Hence reasoning/rationalizing depends on the irrational, ie. subjective virtues, dogmas as it were.
  - pjscott 4y ago
    I think what we have here is a terminological mix-up -- it sounds like you're conflating virtues [1], rules [2], dogma (in the non-Catholic sense), and consequentialist [3] utility functions [4]. The post you're replying to is well aware of the is/ought distinction that the article is talking about. [1] https://en.wikipedia.org/wiki/Virtue_ethics https://en.wikipedia.org/wiki/Virtue_ethics [2] https://en.wikipedia.org/wiki/Deontology https://en.wikipedia.org/wiki/Deontology [3] https://en.wikipedia.org/wiki/Consequentialism https://en.wikipedia.org/wiki/Consequentialism [4] https://en.wikipedia.org/wiki/Utility https://en.wikipedia.org/wiki/Utility
wellbehaved 4y ago
There's nothing more silly than presenting the appearance of reasoning that tries to undercut reasoning as such.
lo_zamoyski 4y ago
A step in the right direction, to be sure. Now for a theoretical course correction... While valuation is not the task of empirical science per se (and indeed, even the choice to pursue science is itself a result of valuation), the view that value and ends aren't rationally knowable, that they are somehow "pre-rational" or "extra-rational" or even "irrational" (Hume famously said that rationality serves the passions) only holds if you reject telos, and this is something the materialist metaphysical doctrine does axiomatically. But if human beings have a nature, then the end or ends of human nature are a defining part of that nature. Human nature is then what determines what is good for human beings and what furthers human flourishing; in short, what results in human happiness. Indeed, it is human nature that is the foundation for any truly defensible objective morality, because without it, we are left with nihilism or relativism or subjectivism or whatever. To reduce rationality to the methods empirical science is not only incoherent, but it is to take a narrow, castrated view of rationality that condemns whole swathes of reality (such as the axiological) to unintelligibility. If the author were to revise his metaphysical views to include telos, I think he would be in a position to make sense of value rather than consigning it to some realm of unintelligible priors. This is where the blog post becomes confusing. On the one hand, value seems to be construed as something outside of rationality. Then we speak of "intrinsic" values which frankly begins to sound like the beginning of an awareness that human nature is a determiner of such things, though still mired in a subjectivist perspective. Then the author speaks of things as if they were objectively valuable which seems to contradict what he said earlier. How can you criticize the hamster wheel of productivity on subjective grounds? Maybe the workaholic loves the hamster wheel. You need human nature to show that workaholism is bad for human beings. It is important here to recognize that telos is not a matter of conscious intent. This is a common misconception (though telos is involves here as well). Telos that toward which something is ordered, especially causally. Indeed, without telos, efficient causality itself becomes unintelligible and scientific explanation itself becomes impossible. You could not explain why striking a match, for instance, predictably results in fire and not something else like confetti falling from the ceiling or the sudden appearance of an elephant.
yamrzou 4y ago
Here is an interesting perspective that I heard recently: The (rational) mind is just another sense. As such, it has limits like the rest of the senses. For example our vision struggles with very close objects (we do not see our own eyelashes) or very distant objects. Similarly, very low or very high sound frequencies are outside the audible range of the human ear. Our rationality must also be limited in that sense.
skybrian 4y ago
This doesn't really get into the differences between rational frameworks and everyday thinking. A better introduction: https://metarationality.com/reasonableness-aspects https://metarationality.com/reasonableness-aspects
machina_ex_deus 4y ago
I couldn't disagree more. The limitations are of the author's rational reasoning, not of rational reasoning itself. You can reason about what brings you pleasure, and make rational decisions to pursue it. The size of the entertainment industry shows the way society is structured can value art. Many big cities still have big green parks. Many strawmen arguments by the author. The author just finally realized that emotions aren't rational relative to the circumstances. That doesn't mean they are beyond rational reasoning. It also doesn't mean society doesn't value them. There's no such thing as society valuing something, there are individual humans making individual decisions. If some of those humans value something, their decisions will reflect that. No such thing as intrinsic value. There's individual humans and their individual decisions in individual situations. Their decisions might not even be consistent, and different humans will make different decisions.
- Cornelius267 4y ago
  "The limitations are of the author's rational reasoning, not of rational reasoning itself. You can reason about what brings you pleasure, and make rational decisions to pursue it." This is just kicking the can down the road. Why would you want to make a decision to pursue this thing that you decided brings you pleasure? Is pleasure in and of itself reason enough? We know that all logical systems contain true statements that cannot be proven. This is a fundamental feature of all logical systems, so why should we expect that this restriction does not apply to our own internal logical systems? I have recently had a similar realization as the author, and so this resonates with me. No matter what, you are engaging in some sort of ontology, some sort of idea that decides what is and is not, and what defines value. You can either engage with this ontology and critically reflect upon it, or you can pretend that you don't have one and therefore miss things that may otherwise have been evident to you.
FrontierPsych 4y ago
This is kind of a silly article to me. It is obvious that nobody can be 100% rational. It is not even rational to think that one can be rational all the time. I've also been in sales a long time, and it is a given that people make decisions irrationally and then use rationality to justify their irrational decisions. If you are a hyper rational person and are tasked to make a decision on what product or service to buy, but this product or service is a key purchase that a lot of people and business units will depend on, you may TRY to act rationally. In reality, many other irrational fears come into play. What if you make the wrong decision, despite all the rationality in the world, you don't KNOW if the product or service will work in your business, despite testimonials from other customers of the vendors. And, if it IS a colossal failure, are you going to have hour job if you make a bad decision? Then, will you be forced to move? Will you have to pull your children out of a school where they seem to be excelling? Will your wife want to move if you get fired? Will you be demoted? Will you be kept there, but transferred to a do-nothing job? All of these issues and more are not rational, and can effect a decision. Lots of people, back in the day, would just buy IBM, because as the saying went, "Nobody gets fired for buying from IBM." Nobody gets fired, because it is the safe choice. Not the best choice, necessarily, but the safest. Because then all you have to do is say, "Hey, it's IBM boss. Everyone buys from them, they have a stellar reputation. This has to be THEIR fault, not mine." And that is true and exactly how it went. Now it is the same for stuff like buying SAP ERP system. Buy another one at the risk of your career. . And in the Scientific Method itself, it acknowledges irrationality. The double-blind experiments are specifically designed to eliminate bias as much as possible. Bias meaning irrationality, at least in my books. Another way to eliminate irrationality is reproducability. By having reproducability, it is impossible to be irrational in an argument if the results cannot be duplicated by someone else doing the exact same experiments using the exact same procedures. This has happened many times when someone has tried to duplicate an experiment, only to find out the original was wrong, and someone didn't irrationally publish their results without others being able to double-check.

87 points by _gfwu 4y ago

I'm reading Zen and the art of motorcycle maintainance at the moment which also pushes and pulls at the limits of reason and rationality. Some of the ideas are quite 'dense' and worth making notes over. Worth a read.

karmakurtisaani 4y ago
I read it some time ago, really enjoyed it. Lots of layers to it too.
bilsbie 4y ago
> pushes and pulls at the limits of reason and rationality Any chance you’d be up for providing a few examples? I read it but I can’t remember that.
- 0x445442 4y ago
  It’s been a number of years since my last reading but what I remember is it pulling at the thread of the concept of quality. The author stumbles upon this while teaching an introductory writing course at college. What he discovers is that it’s difficult to reduce the act of writing to a prescribed set of rules to follow in order to produce a quality piece. And yet, somehow his students, without training or much effort, are able to judge which pieces exhibit quality. This idea literally leads the author to madness.
  - sifar 4y ago
    And that quality arises at the moment of interaction. It is simple but was a stunning realization for me. This rhymes with what The Tao Te Ching says that you can't be taught - you have to experience it in your own way.
0x445442 4y ago
Fantastic book. It’s exploration of quality is amazing.
tanseydavid 4y ago
ZatAoMM was really a great read for me. One of my big takeaways from the book is that the main character actually managed to literally drive himself insane through hyper-rationality and over-thinking.

fictionfuture 4y ago

Nietzsche discusses this at length (the author mentions Jung, but Jung borrowed heavily from Nietzsche in this regard)… Nietzsche believed that the best decisions weren’t the most rationale ones but the ones that were “life furthering.” However he believed rationality is preferable to idealism. Therefore, best to first deal with reality, learn rationality, then make decisions with your feet on the ground

kayodelycaon 4y ago

This is why I like to keep rational thinking as a tool in my box of strategies for life. The biggest change I’ve ever had living my life was the XKCD’s 1 of today’s 10,000. https://xkcd.com/1053/ https://xkcd.com/1053/ To boil it down: “let people enjoy things”. If you like something and it doesn’t have a negative affect on your life or someone else’s, go for. Not everything you do needs to analyzed.

starkd 4y ago
To a point. But many joys are very destructive in and of themselves. Without any analysis, one might be tempted to snort meth all day.
- kayodelycaon 4y ago
  I would argue meth falls under the “negative affect” I mentioned. :)
chousuke 4y ago
It's a bit difficult to disagree with "not everything", because that's obviously true; not everything even can be perfectly analyzed. However, I think you can (and should) try to at least maintain awareness of your behaviour and its effects, but only because if you don't, you may actually end up acting in ways that contradict your values; or if your values are ill-defined or vague, you might end up not understanding what it is that really matters to you. I think rationality is a necessary part of having a well-defined value system, even if the axioms are arbitrary. "Letting people enjoy things that aren't harming anyone" is a rational decision based on my value system. What I can use rationality for, really, is figuring out if there's something I'm doing that's actually in conflict with my own values, or if there's something I can do to nudge the world (or at least my own life) to have more of the things I value.

awillen 4y ago

This doesn't feel particularly insightful to me. "We have to realise that rationality, particularly in the form of scientific thought, cannot provide us with a value system." Yeah, that's true, but it's not the point of rationality - rationality is a system of thinking and decision making designed to optimize for the best outcome, but you have to define what's best and thus what you want to optimize for. Rationality won't provide you with a value system, but I don't think anyone who understands it would claim that it does. This is sort of like saying culinary school will teach you how to cook, but it won't pick what's for dinner.

kiba 4y ago
Rationality won't provide you with a value system, but I don't think anyone who understands it would claim that it does. This is sort of like saying culinary school will teach you how to cook, but it won't pick what's for dinner. Someone's belief and value system is the difference between being shunned or supported, like a gay teenager being disowned by their parent. We should definitely examine our value system.
- awillen 4y ago
  Sure, examining value systems is good and we should certainly do that. That's not really relevant to the article or my response, though.
licebmi__at__ 4y ago
Plenty of people think Sam Harris understands rationality.
Dracophoenix 4y ago
But even the pursuit of optimality is a value, is it not? Even if you define what's best, could one then not choose to do the opposite? No individual is required to live an optimal life, and many purposely and self-righteously shun one.
- awillen 4y ago

swayvil 4y ago

Rationality is a game. Played within a structure of rules (aka assertions, assumptions, axioms). These rules are not themselves rationally born. They are born from habit, tradition, convention, aesthetics, consensus, utility, authority... Lots of possible sources for rules. Logical consistency is one good rule. Puts the rational in rationality. It's just a game. A dream. Ungrounded. That may be obvious, but it's easy to lose yourself in a dream. (Forever even.) That's important to keep in mind.

starkd 4y ago
I would argue most are built on tradition, which offer the ability to build on the results of our forebears without reinventing the wheel. It's also helpful to have a little humility in the face of multiple complexities. That's why challenging tradition, while it can prove useful at times, does not come without a cost.
- swayvil 4y ago
  Oh good point. That's 2 important powers of rationality. Communicability and buildability. That's how we build things like "chemistry". But yes, tradition. That's a big one.
anonporridge 4y ago
It is a game, but we judge games by the fruit they bear, and the harsh reality of life is the ultimate judge. The reason we value the games of rationality and science over the games of religion and mythology is that the former games produce significantly more power. And that's the ultimate game of life. Crudely, might makes right. Those individuals or groups that can most effectively harness various forms of power to first defend, and then extend, their claim of limited resources win the game, by simple virtue of the fact that they remain while their competition fades away. But even games like religion have been, and continue to be valuable frameworks for this ultimate game. There's a strong argument to be made that religion binds large groups of people together towards a common goal and incentivizes individuals to sacrifice the entirety of their life because they believe their rewards will come in the non existent afterlife. That's a powerful game, if you can keep the fantasy alive.

tehchromic 4y ago

It's a well meaning article. My first thought is that it's generated by AI in the service of some moderate yet far reaching religious org. "Finding our intrinsic values is no trivial task, but something that forces us to stare into the core of our humanity." I think it misses a critical topical conclusion which is that "intrinsic values" means little or nothing to many, and many folks are fairly incapable of competently doing the kind of contemplative meditating required to arrive at some set of rules like that, and so they look to others. And here's the crux of the missed point: To really resonate with our intrinsic values, we need to drop the myth that a rational economic system exists, and build one with the value of present and future human life explicitly built in. We already have a massive framework for understanding individual intrinsic values in terms of the sanctity of human life and that's the great mystery religions that have birth to our modern age. What's needed now is a planetary ethos where intrinsic human value is put in the proper context of the ecological reality of the biological and geological systems and processes that sustain us. Building that culture requires myth and magic, but also is an intensely rational project. There is really no way around it, and any philosophical text intended to shift the cultural landscape is incomplete without it.

openfuture 4y ago
Please join us on matrix! Starting from datalisp.is.
jcelerier 4y ago
> first thought is that it's generated by AI We're in the Yosemite park bear era of AI: there is considerable overlap between the capabilities of the most intelligent AI and the most stupid humans ... And that's very generous to humans
pas 4y ago
> sanctity of human life what's so sacrosanct about life? life is common, many argue too common. (see the overpopulation hysteria.) I'd argue there are better things that we should hold dear, like empathy and cooperation, the resilience of getting up after devastating events, our ability to cooperate in even the most abstract frameworks.

cainxinth 4y ago

It's paradoxical, but a certain amount of irrational behavior can be optimal. Consider the balance between depressive realism (seeing the world so accurately that it eats away at your hope for a better future) and positive illusions, the small lies people convince themselves of that they are better, stronger, smarter, and luckier than they are. In small amounts, self-deception and avoidance of discomfort are necessary coping mechanisms. Reality is daunting, mysterious, and random; staring into the abyss can cause despair. It's a better strategy to be slightly optimistic, even if there isn't a strong rational basis for that outlook. Taken too far, you get total self delusion, the people who live in their own reality distortion fields, but there are benefits to a modicum of irrational self confidence.

srinivgp 4y ago
Pretty much every direction is a local improvement given _some_ starting point, but if you broaden your overall search, the "more self-deception" direction is gonna be worse.
kiba 4y ago
Do people really see the world so 'accurately' that they become depressed? I suspect it's a matter of perspective rather than being a realist.
- jhbadger 4y ago
  It's basically the argument that poets like Keats had, that science was destroying beauty because it "unweaved the rainbow" -- that is the rainbow is less beautiful if we understand why it occurs rather than being magical. I personally can't understand this argument at all -- but then I always enjoyed figuring out how magic tricks worked as well.
  - nuancebydefault 4y ago
    I also try to find out how and why things work, I suppose that's why I became an engineer. Thinking about the subject further, the beauty is not the rainbow, the beauty is the fantasy around it. For an engineer, fantasising can be a means to enhance creativity and out of the box thinking, which can lead to a solution, the pieces of the puzzle fall into place, in the end the effect can be explained logically. The journey to come to the conclusion is like a game, and is fun and satisfying. Once the solution is known, their interest will sheft to other effects, rendering the former less interesting.

sifar 4y ago

Like all fundamental things, rationality is what we say it is. Things just are, if we cannot explain them we call them irrational but that doesn't mean that they are inherently bad or to be shunned. Rational thought arises out living, it is not mechanical. I would say the triumph of our rationality has been the ability to deal with this irrationality, something which is on the decline. I would say that we won't be able to build true AI till we understand and reconcile this conflict and find a way to express it.

Joker_vD 4y ago

> Before, my enjoyment of art, music or literature came with a vague angst, urging me to try to understand why it made me feel a particular way, rather than just enjoying the experience itself. I always thought Bazarov from Turgenev's "Fathers and Sons" was a caricature nihilist/rationalist, with his "Nature is just a back wall in a worker's workshop, no need to admire its so called 'beauty'" attitude... but apparently not.

_8j50 4y ago

I would also say that another, perhaps controversial shortcoming of rational thinking is that it presumes the human mind can process and reason with all aspects of reality. But using rational thinking, isn't it rational to also presume that there could be aspects of reality which the human brain simply isn't equipped to process but are still real in that we can sense and be affected by them? This is similar to how a person born blind accepts the reality of the visible world without ever experiencing or being able to rationalize it. Or perhaps a better analogy would be ants and other insects who only think/see in 2d (or so I heard, correct me if this is wrong).

sanroot99 4y ago

I don't think this article was rational /s

e63f67dd-065b 4y ago

It's not helpful to think of rationality as a set of rules. I think rationality is better described, in my view, as an explicit acknowledgement of a search for optima in their own chosen utility functions. Rationality is not thinking like a robot; it's asking the question: if I have a vaguely-defined utility function f, how would a rational optimiser optimise for f? Would you march towards a local optima, or try to look for other better local optima? You see an optima in the distance, what's the shortest path that will lead you there? Other rational agents have function y that conflict with your function; how best to resolve/navigate said conflict? What path traverses along high-value areas such that the expected utility of your path over time is highest? As OP points out, our utility functions are poorly defined. That doesn't mean, however, that we cannot apply mathematical/rational techniques to it. A big part is explicitly acknowledging our own biases and trying to restructure our thinking in a way that best avoids them. A good heuristic, for example, is that having a more accurate view of reality generally allows for more efficient searches of your utility space. Thus, updating your view of the world to better conform to reality is a positive EV investment.

nuancebydefault 4y ago
IMHO you missed the point of the article... it says that rationality on itself only can lead to progress in the light of virtues (rules of what is 'right'). Without those as a starting point, there can be no reasoning. Hence reasoning/rationalizing depends on the irrational, ie. subjective virtues, dogmas as it were.
- pjscott 4y ago
  I think what we have here is a terminological mix-up -- it sounds like you're conflating virtues [1], rules [2], dogma (in the non-Catholic sense), and consequentialist [3] utility functions [4]. The post you're replying to is well aware of the is/ought distinction that the article is talking about. [1] https://en.wikipedia.org/wiki/Virtue_ethics https://en.wikipedia.org/wiki/Virtue_ethics [2] https://en.wikipedia.org/wiki/Deontology https://en.wikipedia.org/wiki/Deontology [3] https://en.wikipedia.org/wiki/Consequentialism https://en.wikipedia.org/wiki/Consequentialism [4] https://en.wikipedia.org/wiki/Utility https://en.wikipedia.org/wiki/Utility

wellbehaved 4y ago

There's nothing more silly than presenting the appearance of reasoning that tries to undercut reasoning as such.

lo_zamoyski 4y ago

A step in the right direction, to be sure. Now for a theoretical course correction... While valuation is not the task of empirical science per se (and indeed, even the choice to pursue science is itself a result of valuation), the view that value and ends aren't rationally knowable, that they are somehow "pre-rational" or "extra-rational" or even "irrational" (Hume famously said that rationality serves the passions) only holds if you reject telos, and this is something the materialist metaphysical doctrine does axiomatically. But if human beings have a nature, then the end or ends of human nature are a defining part of that nature. Human nature is then what determines what is good for human beings and what furthers human flourishing; in short, what results in human happiness. Indeed, it is human nature that is the foundation for any truly defensible objective morality, because without it, we are left with nihilism or relativism or subjectivism or whatever. To reduce rationality to the methods empirical science is not only incoherent, but it is to take a narrow, castrated view of rationality that condemns whole swathes of reality (such as the axiological) to unintelligibility. If the author were to revise his metaphysical views to include telos, I think he would be in a position to make sense of value rather than consigning it to some realm of unintelligible priors. This is where the blog post becomes confusing. On the one hand, value seems to be construed as something outside of rationality. Then we speak of "intrinsic" values which frankly begins to sound like the beginning of an awareness that human nature is a determiner of such things, though still mired in a subjectivist perspective. Then the author speaks of things as if they were objectively valuable which seems to contradict what he said earlier. How can you criticize the hamster wheel of productivity on subjective grounds? Maybe the workaholic loves the hamster wheel. You need human nature to show that workaholism is bad for human beings. It is important here to recognize that telos is not a matter of conscious intent. This is a common misconception (though telos is involves here as well). Telos that toward which something is ordered, especially causally. Indeed, without telos, efficient causality itself becomes unintelligible and scientific explanation itself becomes impossible. You could not explain why striking a match, for instance, predictably results in fire and not something else like confetti falling from the ceiling or the sudden appearance of an elephant.

yamrzou 4y ago

Here is an interesting perspective that I heard recently: The (rational) mind is just another sense. As such, it has limits like the rest of the senses. For example our vision struggles with very close objects (we do not see our own eyelashes) or very distant objects. Similarly, very low or very high sound frequencies are outside the audible range of the human ear. Our rationality must also be limited in that sense.

skybrian 4y ago

This doesn't really get into the differences between rational frameworks and everyday thinking. A better introduction: https://metarationality.com/reasonableness-aspects https://metarationality.com/reasonableness-aspects

machina_ex_deus 4y ago

I couldn't disagree more. The limitations are of the author's rational reasoning, not of rational reasoning itself. You can reason about what brings you pleasure, and make rational decisions to pursue it. The size of the entertainment industry shows the way society is structured can value art. Many big cities still have big green parks. Many strawmen arguments by the author. The author just finally realized that emotions aren't rational relative to the circumstances. That doesn't mean they are beyond rational reasoning. It also doesn't mean society doesn't value them. There's no such thing as society valuing something, there are individual humans making individual decisions. If some of those humans value something, their decisions will reflect that. No such thing as intrinsic value. There's individual humans and their individual decisions in individual situations. Their decisions might not even be consistent, and different humans will make different decisions.

Cornelius267 4y ago
"The limitations are of the author's rational reasoning, not of rational reasoning itself. You can reason about what brings you pleasure, and make rational decisions to pursue it." This is just kicking the can down the road. Why would you want to make a decision to pursue this thing that you decided brings you pleasure? Is pleasure in and of itself reason enough? We know that all logical systems contain true statements that cannot be proven. This is a fundamental feature of all logical systems, so why should we expect that this restriction does not apply to our own internal logical systems? I have recently had a similar realization as the author, and so this resonates with me. No matter what, you are engaging in some sort of ontology, some sort of idea that decides what is and is not, and what defines value. You can either engage with this ontology and critically reflect upon it, or you can pretend that you don't have one and therefore miss things that may otherwise have been evident to you.

FrontierPsych 4y ago

This is kind of a silly article to me. It is obvious that nobody can be 100% rational. It is not even rational to think that one can be rational all the time. I've also been in sales a long time, and it is a given that people make decisions irrationally and then use rationality to justify their irrational decisions. If you are a hyper rational person and are tasked to make a decision on what product or service to buy, but this product or service is a key purchase that a lot of people and business units will depend on, you may TRY to act rationally. In reality, many other irrational fears come into play. What if you make the wrong decision, despite all the rationality in the world, you don't KNOW if the product or service will work in your business, despite testimonials from other customers of the vendors. And, if it IS a colossal failure, are you going to have hour job if you make a bad decision? Then, will you be forced to move? Will you have to pull your children out of a school where they seem to be excelling? Will your wife want to move if you get fired? Will you be demoted? Will you be kept there, but transferred to a do-nothing job? All of these issues and more are not rational, and can effect a decision. Lots of people, back in the day, would just buy IBM, because as the saying went, "Nobody gets fired for buying from IBM." Nobody gets fired, because it is the safe choice. Not the best choice, necessarily, but the safest. Because then all you have to do is say, "Hey, it's IBM boss. Everyone buys from them, they have a stellar reputation. This has to be THEIR fault, not mine." And that is true and exactly how it went. Now it is the same for stuff like buying SAP ERP system. Buy another one at the risk of your career. . And in the Scientific Method itself, it acknowledges irrationality. The double-blind experiments are specifically designed to eliminate bias as much as possible. Bias meaning irrationality, at least in my books. Another way to eliminate irrationality is reproducability. By having reproducability, it is impossible to be irrational in an argument if the results cannot be duplicated by someone else doing the exact same experiments using the exact same procedures. This has happened many times when someone has tried to duplicate an experiment, only to find out the original was wrong, and someone didn't irrationally publish their results without others being able to double-check.