6 ms·

EA sounds rational and wonderful. And it does make sense. We should follow our logic to its rational conclusions. Our moral intuitions are obviously wrong a lot

EA sounds rational and wonderful. And it does make sense. We should follow our logic to its rational conclusions. Our moral intuitions are obviously wrong a lot - just look at the trolley problem. With reason we can do better. The problem is that it quickly becomes an invitation to ideas like longtermism. Which involve long chains of potentially flawed reasoning, leading to the belief that you're doing tremendous good. And with confirmation bias making it hard for you to doubt your logic, leading to an unbounded potential for error. As the old moral goes, "Nobody is as easy to fool as a person who wants to fool himself." This problem is not original to EA. The history of the 20th century is full of potential utopias. On the basis of the end justifies the means, the prospect of infinite good justifies unlimited harm. Unlimited harm came in the form of wars, famines, and mass repression. But the utopian futures never materialized. That said, there is a lot of good to the idea of EA. It is better to do something effective than to virtue signal. But we should also be biased towards wins we can be more sure are real. Things that are short term and concrete. The more distant and hard to measure the win, the more that we should bias ourselves to the belief that we're missing something.

082349872349872 3y ago
> The more distant and hard to measure the win... known for ages* by the phrase "the end does not justify the means" * exitus acta numquam probat
jltsiren 3y ago
It's the old debate between rationalism and empiricism again. "Rational" is a dangerous word. On the surface, it sounds like "smart". But if you take rationalism to the extreme, it becomes epistemological opposition to evidence. You build mental models and make logical conclusions without considering if the conclusions are also valid in the real world. Scientific worldview is closer to empiricism than rationalism. You start by assuming that your mental models are wrong. They may still be useful, but you have to make observations and experiments and consider the evidence to determine that. Effective altruism is a useful concept. It only becomes problematic once you get too deep into rationalism. The effectiveness of your altruism is fundamentally an empirical question, and it should be answered by empirical means rather than by reasoning.
- analog31 3y ago
  Every rationalist movement eventually ends up at daggers drawn with rational people.
- tim333 3y ago
  Yeah. The trouble with as the dictionary has it "the practice or principle of basing opinions and actions on reason and knowledge" is what are you basing it on? Why are we here? What's the purpose of life. The answer to those is basically don't know, and they you build your reasoning on top of that. However common sense "be nice to people" etc kind of works.
TacticalCoder 3y ago
> The problem is that it quickly becomes an invitation to ideas like longtermism. Which involve long chains of potentially flawed reasoning, leading to the belief that you're doing tremendous good. And with confirmation bias making it hard for you to doubt your logic, leading to an unbounded potential for error. We haven't been reading the same article I think for you've got an extremely gentle take on this. TFA specifically how the "poster boy" of the EA movement bought its way into having a best-seller by using $10m of stolen money gifted by FTX to propel his book ("based on marketing, not merit" according to TFA) and how that same person bought a $16m 15th century mansion in the UK using stolen FTX money. All the while they were presenting SBF as an "altruistic genius driving a Toyota Corolla" while they fully knew he was living an ultra-lavish lifestyle. It's now a fact that SBF is a criminal and the EA movement still hasn't given the ill-acquired donations from SBF back. I hope that John J. Ray III (the person who was in charge of the EA liquidation and now the FTX liquidation) goes after that mansion in the UK and manages to claw money back from the EA movement. There's hope though as 15 years after the Enron scam money was still clawed back.
- RationalDino 3y ago
  While there were bad actors within the EA movement, I don't believe that most involved with EA were truly aware of the level of hypocrisy that you're discussing.
narrator 3y ago
People love to make quasi-religions by introducing flawed premises or arguments early in the reasoning process and then proceeding to build a massive superstructure on top of that. People see this gigantic tower of stuff builded on flawed premises and assume it must be legit. A clever religion founder probably engineered the thing from end to beginning where they found the conclusion they wanted and then worked their way back through their argument to see where they could sneak in a flawed premise or argument to make it work.
throwaway3d3L7 3y ago
"Our moral intuitions are obviously wrong a lot - just look at the trolley problem." I don't understand what I'm supposed to see in it. Could you clarify?
- RationalDino 3y ago
  What should matter is how many lives are lost and how many are saved. But the trolley problem shows that in practice we care more about whether we were the one to throw a switch than the cost in human lives of our decision.
  - Incipient 3y ago
    We had a class of about 30 engineers nearly give an arts ethics professor a nervous breakdown. The entire class went the obvious choice of flipping the switch to kill the one person. She just couldn't get her head around how we all had no issue. Pretty simple, 1<5, and taking no action has no material difference to taking an action. It's a choice either way.
    - RationalDino 3y ago
      Perfect! My favorite line of research involving this is using the trolley problem to get at hidden moral biases that we normally wouldn't admit to ourselves. https://www.discovermagazine.com/mind/given-the-choice-liberals-would-rather-kill-whitey https://www.discovermagazine.com/mind/given-the-choice-liber...

082349872349872 3y ago

> The more distant and hard to measure the win... known for ages* by the phrase "the end does not justify the means" * exitus acta numquam probat

jltsiren 3y ago

It's the old debate between rationalism and empiricism again. "Rational" is a dangerous word. On the surface, it sounds like "smart". But if you take rationalism to the extreme, it becomes epistemological opposition to evidence. You build mental models and make logical conclusions without considering if the conclusions are also valid in the real world. Scientific worldview is closer to empiricism than rationalism. You start by assuming that your mental models are wrong. They may still be useful, but you have to make observations and experiments and consider the evidence to determine that. Effective altruism is a useful concept. It only becomes problematic once you get too deep into rationalism. The effectiveness of your altruism is fundamentally an empirical question, and it should be answered by empirical means rather than by reasoning.

analog31 3y ago
Every rationalist movement eventually ends up at daggers drawn with rational people.
tim333 3y ago
Yeah. The trouble with as the dictionary has it "the practice or principle of basing opinions and actions on reason and knowledge" is what are you basing it on? Why are we here? What's the purpose of life. The answer to those is basically don't know, and they you build your reasoning on top of that. However common sense "be nice to people" etc kind of works.

TacticalCoder 3y ago

> The problem is that it quickly becomes an invitation to ideas like longtermism. Which involve long chains of potentially flawed reasoning, leading to the belief that you're doing tremendous good. And with confirmation bias making it hard for you to doubt your logic, leading to an unbounded potential for error. We haven't been reading the same article I think for you've got an extremely gentle take on this. TFA specifically how the "poster boy" of the EA movement bought its way into having a best-seller by using $10m of stolen money gifted by FTX to propel his book ("based on marketing, not merit" according to TFA) and how that same person bought a $16m 15th century mansion in the UK using stolen FTX money. All the while they were presenting SBF as an "altruistic genius driving a Toyota Corolla" while they fully knew he was living an ultra-lavish lifestyle. It's now a fact that SBF is a criminal and the EA movement still hasn't given the ill-acquired donations from SBF back. I hope that John J. Ray III (the person who was in charge of the EA liquidation and now the FTX liquidation) goes after that mansion in the UK and manages to claw money back from the EA movement. There's hope though as 15 years after the Enron scam money was still clawed back.

RationalDino 3y ago
While there were bad actors within the EA movement, I don't believe that most involved with EA were truly aware of the level of hypocrisy that you're discussing.

narrator 3y ago

People love to make quasi-religions by introducing flawed premises or arguments early in the reasoning process and then proceeding to build a massive superstructure on top of that. People see this gigantic tower of stuff builded on flawed premises and assume it must be legit. A clever religion founder probably engineered the thing from end to beginning where they found the conclusion they wanted and then worked their way back through their argument to see where they could sneak in a flawed premise or argument to make it work.

throwaway3d3L7 3y ago

"Our moral intuitions are obviously wrong a lot - just look at the trolley problem." I don't understand what I'm supposed to see in it. Could you clarify?

RationalDino 3y ago
What should matter is how many lives are lost and how many are saved. But the trolley problem shows that in practice we care more about whether we were the one to throw a switch than the cost in human lives of our decision.
- Incipient 3y ago
  We had a class of about 30 engineers nearly give an arts ethics professor a nervous breakdown. The entire class went the obvious choice of flipping the switch to kill the one person. She just couldn't get her head around how we all had no issue. Pretty simple, 1<5, and taking no action has no material difference to taking an action. It's a choice either way.
  - RationalDino 3y ago
    Perfect! My favorite line of research involving this is using the trolley problem to get at hidden moral biases that we normally wouldn't admit to ourselves. https://www.discovermagazine.com/mind/given-the-choice-liberals-would-rather-kill-whitey https://www.discovermagazine.com/mind/given-the-choice-liber...