6 ms·

I think this is a perfect example of what your parent comment is talking about, being: > unprovable, unfalsifiable, and have the same circular logic as bad rel

I think this is a perfect example of what your parent comment is talking about, being: > unprovable, unfalsifiable, and have the same circular logic as bad religion. You said: > There is no contradiction here. They evolved as the fittest in their own constraint set. If bees weren’t fit enough, they would have gone extinct and replaced by bees with non-barbed stingers. There is no magic that makes them survive. Sure, there's no contradiction, but this is totally circular reasoning that could be used to prove anything. The connection graph between "They survive because they're the fittest" and "They must be the fittest because they survive" is just a circle disconnected from all other knowledge. Your strength as a rationalist is your ability to be more confused by fiction than by reality.[0] But with this circular understanding of natural selection, you could be given a description of absolutely any conceivable configuration of organism and your response would be the same: "they must be the fittest, because they survive, because only the fittest survive" and you haven't gained any understanding at all. There will never be a contradiction, because the argument is disconnected from any larger system of reasoning that could plausibly contradict it. "Hey, there is a random monkey in the Amazon that has 3 hoops on its head and a big hole through its abdomen, isn't that weird? Why are they like that?" "Ah, the hoops and the holes are required for Fitness. Only the Fittest survive, you know. So if they have 3 hoops on their heads and big holes in their abdomens, that is what makes them Fittest. Amen." "Why aren't other monkeys like that then?" "Other monkeys don't need hoops and holes for Fitness. Otherwise they too would have hoops and holes. :)" A better understanding of natural selection would be confused about the hoops and the holes, and that confusion would correlate with either the random monkey species actually not existing, or the model being wrong. As regards the bees: there probably is a reason that dying when stinging confers Fitness. But we should find out what that reason is, rather than state "Fitness because Survival" and feel like we've answered the question. [0] https://www.lesswrong.com/posts/5JDkW4MYXit2CquLs/your-strength-as-a-rationalist https://www.lesswrong.com/posts/5JDkW4MYXit2CquLs/your-stren...

Retric 2y ago
> totally circular reasoning that could be used to prove anything. No, you can attack the reasoning by looking into actual costs. It seems like it can explain anything because we don’t constantly see examples where it’s false. Looking at the costs to bees you see what percentage of them die from attacking mammal flesh and yep it’s a tiny rounding error. Hypothetically, in a world without constraints mice could have a 100 foot long teeth, but we don’t live in a world without constraints.
- xyzzy_plugh 2y ago
  > Hypothetically, in a world without constraints mice could have a 100 foot long teeth Oh boy, today's the day you learn something new about rodents.
  - stavros 2y ago
    What, the fact that their teeth can grow so long that they'll circle around and stab their brains?
kevinventullo 2y ago
My understanding from the article and the general theory of Superorganisms is that it’s not exactly true that “dying when stinging confers fitness”. Rather, dying when stinging is just not a huge penalty when you’re talking about non-reproducing members of a colony. So, while it may be a good thing for bees to evolve the ability to survive stinging, the selective pressure is not as large as one might intuitively expect. Maybe a better title for the post would be something like, “Isn’t it weird that bees die when they sting? Shouldn’t they have evolved away from that?”
- thaumasiotes 2y ago
  Well, there's a larger problem in the post. The primary reason that a bee dies when it stings you is that you kill the bee. A bee stinging an inanimate hunk of meat is unlikely to die. But they can die, and yeah, a big part of the reason why is that dying isn't as large of a cost for bees as you might expect from a human perspective. And looping back, another part is that given the very high risk of being intentionally killed when stinging an enemy who you want to sting, improving the much smaller rate of accidental death isn't really worth much. But even though it isn't worth much, it's worth something, and work has been done on the project.
  - tsimionescu 2y ago
    Per everything I've ever read on this, a bee that has stung meat is no longer able to survive. It will either try to pull itself out and disembowel itself, or it will remain stack and die of hunger. What makes you think a bee that stings, say, a dog that can't swat it will then go on to survive?
    - thaumasiotes 2y ago
      Watching a youtube video by a beekeeper demonstrating that a bee stuck in his arm will gently wiggle itself around until it can pull free and fly away. Between that one guy with his live demonstration and his plentiful experience being stung by bees, and you with everything you've ever read, I'm obviously going to stick with him. You should read better sources. Here, an example of how bees behave after stinging meat: https://www.youtube.com/shorts/7wH8dYiGbig https://www.youtube.com/shorts/7wH8dYiGbig Or here: https://www.youtube.com/watch?v=G-C77ujnLZo https://www.youtube.com/watch?v=G-C77ujnLZo A bee that stings meat will pull itself out unharmed, if you let it.
iwontberude 2y ago
How is it circular to argue why one species would do better in an environment than another based on phenotype and the physical interactions it enables? It’s all relative to other species. As long as you understand that, there is no logical fallacy. I do very much appreciate the focus on informal logic though.
- jstanley 2y ago
  Because you could encounter absolutely any organism and make the same argument. There is no configuration of organism that would cause you to say "huh, I guess Survival doesn't depend on Fitness after all!" Because it takes the observation of Survival and uses it to infer Fitness, at the same time as saying that Fitness confers Survival.
  - glenstein 2y ago
    >Because you could encounter absolutely any organism and make the same argument. That's a function of the explanation being an extremely good explanation. It rises to the top precisely because it has explanatory power all across nature without evident counter-example. >There is no configuration of organism that would cause you to say "huh, I guess Survival doesn't depend on Fitness after all!" This is where the argument falls apart. For starters, species go extinct all the time for reasons tied to their evolutionary trajectory. And there are species still living that unfortunately seem very imperfectly adapted to their constraints and likely to go extinct without a run of good luck or human intervention (e.g. pandas). We seem perfectly capable of recognizing when such species are "on the ropes". Additionally there are relative advantages we can clearly observe from animals in overlapping niches, and we can marvel at the effectiveness of adaptations in ways that don't involve circular assumptions (e.g. algae's capability for efficient growth is astonishing and without equal on the planet). And, we could surely conceive of preposterous examples that defy expectations (e.g. the other commenter's example of mice with 100ft teeth). It probably feels like it proves too much, because it's confirmed over and over again in nature everywhere, at all times. But in an alternate world where that wasn't the case, counter-examples would abound (such as the mouse with 100ft teeth). So re-iterating the core lesson about the role of natural selection is not just a circular assumption, it's the culmination of hard earned, accumulated evidence, ready at any moment to be falsified. The honeybee is a perfect example, because the stinger does pose a real question about how we understand it's relation to fitness, and it requires delving into all kinds of complicated dynamics about genetically related drones are to the queen, the role of the sacrifice in supporting the hive and so-on. If we didn't have explanations like those, it would indeed pose a problem with explanations that presume fitness. That's a real payoff from being alert to the need to have robust explanations; I don't think anyone is just saying "well it's fitness" and calling it a day so much as they're honoring the explanatory power of a well confirmed theory.
notRobot 2y ago
"Fittest" is what we call those who happen to survive in their context. Systems that successfully replicate themselves in their context tend to stick around. Those who can't, go extinct. We obviously still study why they survived. That's what the article speculates about. So yes, in a sense, any organisms you see is the "fittest" in the sense that it was able to survive (replicate) in its context while countless others were not.
icehawk 2y ago
> The connection graph between "They survive because they're the fittest" and "They must be the fittest because they survive" is just a circle disconnected from all other knowledge. The GP said that bees survived because they're "fit enough" not that they survive because they're "the fittest" and there are definitely species that don't seem to be surviving because they're not fit enough.
shwouchk 2y ago
All of evolution is path dependent. Dying after a sting does not have to confer extra fitness to exist, right now. rather it had to have conferred some fitness relative to the alternative traits circulating at the time it was selected. obviously if you go by gradient descent you are not guaranteed to reach a global minimum or even a local one, given a constantly changing fitness landscape. In most of these discussions, optimizing nature of evolution is taken as granted - we do not need to prove how evolution works yet again - there are plenty of evidence and discussion elsewhere - take it or leave it. This node is well connected to other knowledge, and if you disagree, you need to convince a whole discipline of science, not me. From the optimizing premise of evolution, various inferred hypotheses can be made, explaining a range of phenomena, just like, in physics, from the premise that probabilities of events are given by the amplitudes of solutions of certain pdes with specific initial conditions, we were able to devise tractable mathematical models of various nuclear reactions, here a model of development of certain abstract traits was explained (“altruism”). The author fully acknowledged that this is a simplified model and does not match reality in some cases, and in other cases does not explain well enough. improvements to the model were proposed. Isn’t that how science works, in the best cases?
darkerside 2y ago
This assumes that Fitness is a meaningless aphorism. I would posit that Fitness is a meaningful concept that can be learned. It is defined by what happens to be reproductively helpful to a species, which is tautological, but to understand the definition, it just means you need to understand that particular species ecosystem and lifecycle. If you really analyze any word, it loses all meaning except what we've assigned it.
sedatk 2y ago
It's only circular reasoning if you ignore the mathematical model of evolution. Evolution is simply that your chances of picking a red ball from a jar increases proportionally to the number of red balls in the jar. What you're saying is that "but why did you pick that specific ball, what were the factors at play?". I say "probability", and you call it circular reasoning. You expect me to explain the physical forces at play that pushed forward a specific ball on the top when you filled the jar, and made it collide with your fingers, and your brain's reasoning chain to pick the first ball that you touched, so you pick that one. But, basically, you can't. It's impossible because 1) we can't time travel. 2) we can't stop the universe and examine all of its properties. We work with the mathematical model we have and expect everything to work aligned with it. "Why do certain monkey species have holes in their bellies?". You can reason about that question, you can even come up with some answers, but it would be impossible to prove due to our universe's observability constraints. But the mathematical model works, it even works beyond our physical realm (e.g. it works to solve mathematical problems). And no, it will never explain why certain monkeys have holes in their bellies. That's an entirely different domain of causal analysis.

Retric 2y ago

> totally circular reasoning that could be used to prove anything. No, you can attack the reasoning by looking into actual costs. It seems like it can explain anything because we don’t constantly see examples where it’s false. Looking at the costs to bees you see what percentage of them die from attacking mammal flesh and yep it’s a tiny rounding error. Hypothetically, in a world without constraints mice could have a 100 foot long teeth, but we don’t live in a world without constraints.

xyzzy_plugh 2y ago
> Hypothetically, in a world without constraints mice could have a 100 foot long teeth Oh boy, today's the day you learn something new about rodents.
- stavros 2y ago
  What, the fact that their teeth can grow so long that they'll circle around and stab their brains?

kevinventullo 2y ago

My understanding from the article and the general theory of Superorganisms is that it’s not exactly true that “dying when stinging confers fitness”. Rather, dying when stinging is just not a huge penalty when you’re talking about non-reproducing members of a colony. So, while it may be a good thing for bees to evolve the ability to survive stinging, the selective pressure is not as large as one might intuitively expect. Maybe a better title for the post would be something like, “Isn’t it weird that bees die when they sting? Shouldn’t they have evolved away from that?”

thaumasiotes 2y ago
Well, there's a larger problem in the post. The primary reason that a bee dies when it stings you is that you kill the bee. A bee stinging an inanimate hunk of meat is unlikely to die. But they can die, and yeah, a big part of the reason why is that dying isn't as large of a cost for bees as you might expect from a human perspective. And looping back, another part is that given the very high risk of being intentionally killed when stinging an enemy who you want to sting, improving the much smaller rate of accidental death isn't really worth much. But even though it isn't worth much, it's worth something, and work has been done on the project.
- tsimionescu 2y ago
  Per everything I've ever read on this, a bee that has stung meat is no longer able to survive. It will either try to pull itself out and disembowel itself, or it will remain stack and die of hunger. What makes you think a bee that stings, say, a dog that can't swat it will then go on to survive?
  - thaumasiotes 2y ago
    Watching a youtube video by a beekeeper demonstrating that a bee stuck in his arm will gently wiggle itself around until it can pull free and fly away. Between that one guy with his live demonstration and his plentiful experience being stung by bees, and you with everything you've ever read, I'm obviously going to stick with him. You should read better sources. Here, an example of how bees behave after stinging meat: https://www.youtube.com/shorts/7wH8dYiGbig https://www.youtube.com/shorts/7wH8dYiGbig Or here: https://www.youtube.com/watch?v=G-C77ujnLZo https://www.youtube.com/watch?v=G-C77ujnLZo A bee that stings meat will pull itself out unharmed, if you let it.

iwontberude 2y ago

How is it circular to argue why one species would do better in an environment than another based on phenotype and the physical interactions it enables? It’s all relative to other species. As long as you understand that, there is no logical fallacy. I do very much appreciate the focus on informal logic though.

jstanley 2y ago
Because you could encounter absolutely any organism and make the same argument. There is no configuration of organism that would cause you to say "huh, I guess Survival doesn't depend on Fitness after all!" Because it takes the observation of Survival and uses it to infer Fitness, at the same time as saying that Fitness confers Survival.
- glenstein 2y ago
  >Because you could encounter absolutely any organism and make the same argument. That's a function of the explanation being an extremely good explanation. It rises to the top precisely because it has explanatory power all across nature without evident counter-example. >There is no configuration of organism that would cause you to say "huh, I guess Survival doesn't depend on Fitness after all!" This is where the argument falls apart. For starters, species go extinct all the time for reasons tied to their evolutionary trajectory. And there are species still living that unfortunately seem very imperfectly adapted to their constraints and likely to go extinct without a run of good luck or human intervention (e.g. pandas). We seem perfectly capable of recognizing when such species are "on the ropes". Additionally there are relative advantages we can clearly observe from animals in overlapping niches, and we can marvel at the effectiveness of adaptations in ways that don't involve circular assumptions (e.g. algae's capability for efficient growth is astonishing and without equal on the planet). And, we could surely conceive of preposterous examples that defy expectations (e.g. the other commenter's example of mice with 100ft teeth). It probably feels like it proves too much, because it's confirmed over and over again in nature everywhere, at all times. But in an alternate world where that wasn't the case, counter-examples would abound (such as the mouse with 100ft teeth). So re-iterating the core lesson about the role of natural selection is not just a circular assumption, it's the culmination of hard earned, accumulated evidence, ready at any moment to be falsified. The honeybee is a perfect example, because the stinger does pose a real question about how we understand it's relation to fitness, and it requires delving into all kinds of complicated dynamics about genetically related drones are to the queen, the role of the sacrifice in supporting the hive and so-on. If we didn't have explanations like those, it would indeed pose a problem with explanations that presume fitness. That's a real payoff from being alert to the need to have robust explanations; I don't think anyone is just saying "well it's fitness" and calling it a day so much as they're honoring the explanatory power of a well confirmed theory.

notRobot 2y ago

"Fittest" is what we call those who happen to survive in their context. Systems that successfully replicate themselves in their context tend to stick around. Those who can't, go extinct. We obviously still study why they survived. That's what the article speculates about. So yes, in a sense, any organisms you see is the "fittest" in the sense that it was able to survive (replicate) in its context while countless others were not.

icehawk 2y ago

> The connection graph between "They survive because they're the fittest" and "They must be the fittest because they survive" is just a circle disconnected from all other knowledge. The GP said that bees survived because they're "fit enough" not that they survive because they're "the fittest" and there are definitely species that don't seem to be surviving because they're not fit enough.

shwouchk 2y ago

All of evolution is path dependent. Dying after a sting does not have to confer extra fitness to exist, right now. rather it had to have conferred some fitness relative to the alternative traits circulating at the time it was selected. obviously if you go by gradient descent you are not guaranteed to reach a global minimum or even a local one, given a constantly changing fitness landscape. In most of these discussions, optimizing nature of evolution is taken as granted - we do not need to prove how evolution works yet again - there are plenty of evidence and discussion elsewhere - take it or leave it. This node is well connected to other knowledge, and if you disagree, you need to convince a whole discipline of science, not me. From the optimizing premise of evolution, various inferred hypotheses can be made, explaining a range of phenomena, just like, in physics, from the premise that probabilities of events are given by the amplitudes of solutions of certain pdes with specific initial conditions, we were able to devise tractable mathematical models of various nuclear reactions, here a model of development of certain abstract traits was explained (“altruism”). The author fully acknowledged that this is a simplified model and does not match reality in some cases, and in other cases does not explain well enough. improvements to the model were proposed. Isn’t that how science works, in the best cases?

darkerside 2y ago

This assumes that Fitness is a meaningless aphorism. I would posit that Fitness is a meaningful concept that can be learned. It is defined by what happens to be reproductively helpful to a species, which is tautological, but to understand the definition, it just means you need to understand that particular species ecosystem and lifecycle. If you really analyze any word, it loses all meaning except what we've assigned it.

sedatk 2y ago

It's only circular reasoning if you ignore the mathematical model of evolution. Evolution is simply that your chances of picking a red ball from a jar increases proportionally to the number of red balls in the jar. What you're saying is that "but why did you pick that specific ball, what were the factors at play?". I say "probability", and you call it circular reasoning. You expect me to explain the physical forces at play that pushed forward a specific ball on the top when you filled the jar, and made it collide with your fingers, and your brain's reasoning chain to pick the first ball that you touched, so you pick that one. But, basically, you can't. It's impossible because 1) we can't time travel. 2) we can't stop the universe and examine all of its properties. We work with the mathematical model we have and expect everything to work aligned with it. "Why do certain monkey species have holes in their bellies?". You can reason about that question, you can even come up with some answers, but it would be impossible to prove due to our universe's observability constraints. But the mathematical model works, it even works beyond our physical realm (e.g. it works to solve mathematical problems). And no, it will never explain why certain monkeys have holes in their bellies. That's an entirely different domain of causal analysis.