Look, scoffing at less-than-optimal philanthropy is ultimately just another form of counterproductive negativity. If you’re really serious about efficacy, you should be adding to the list of causes, not subtracting from it. That is, instead of responding to a post like this by encouraging people to
resist the influence, suppress the… urge [to] take actions
(!)
how about answering with “hey, you know what would be really, really helpful?” and proceeding to list some awesome utility-maximizing charity.
Warm feelings are good. Someone who donates a few spare frequent-flyer miles to help Curt Knox and Edda Mellas visit their daughter imprisoned 6,000 miles away doesn’t need to feel ashamed of themselves for not being “rational”—except in the extremely unlikely event that that action actually prevented them from doing something better. Does anyone honestly, seriously believe that discouraging people from doing things like this is a way of making the world a better place?
Speaking of challenges for LW, I propose a new rule: anybody who comes across an ostensibly good cause, but scoffs at its suboptimality, or thinks “well, it’s not that I’m not willing to sacrifice $10, but surely there are better uses of that money” should be immediately required, right then and there, to donate that $10 to the Singularity Institute—no ifs, ands, or buts.
That is, instead of responding to a post like this by encouraging people to …
how about answering with “hey, you know what would be really, really helpful?” and proceeding to list some awesome utility-maximizing charity.
No, no, NO! I desire to correct a fundamental mistake that is counter to whatever good ‘rationality’ may happen to provide. Raising the sanity waterline is an important goal in itself and particularly applicable in rare communities that have some hope of directing their actions in a way that is actually effective. Not only that, but seeing the very concept of rationality abused to manipulate people into bad decision making is something that makes me feel bad inside. Yes, it is the opposite of a warm fuzzy.
Look, scoffing at less-than-optimal philanthropy is ultimately just another form of counterproductive negativity. If you’re really serious about efficacy, you should be adding to the list of causes, not subtracting from it.
You are fundamentally wrong and the use of labeling things that disagree with you as ‘negative’ is non-rational influence technique that works in most places but should be discouraged here. It is not counterproductive to not do things that are stupid. It is not intrinsically better to add things to a list of normatively demanded behaviors while never removing them. If the list is wrong (for a given value of wrong) then it should be fixed by adding to it or removing from it in whatever way necessary.
Warm feelings are good. Someone who donates a few spare frequent-flyer miles to help Curt Knox and Edda Mellas visit their daughter imprisoned 6,000 miles away doesn’t need to feel ashamed of themselves for not being “rational”—except in the extremely unlikely event that that action actually prevented them from doing something better. Does anyone honestly, seriously believe that discouraging people from doing things like this is a way of making the world a better place?
People being manipulated into actions by the inclusion of irrelevant things in the definition of ‘rational’ is what I am discouraging. Tell people that Knox is a good way to purchase warm fuzzies, that’s fine. But don’t dare try to call it a ‘challenge for rationality’, piggybacking on the human instinct to avoid the shame of not supporting the tribal value (‘rational’).
Speaking of challenges for LW, I propose a new rule: anybody who comes across an ostensibly good cause, but scoffs at its suboptimality, or thinks “well, it’s not that I’m not willing to sacrifice $10, but surely there are better uses of that money” should be immediately required, right then and there, to donate that $10 to the Singularity Institute—no ifs, ands, or buts.
No ifs and buts? Not everyone here needs to consider the SIAI to be the best use of their money. That’s not required by ‘rationality’ either. You’re in the wrong place if you think that approach is at all appropriate. Don’t try to force your obsession with Knox on everyone else. It’s not my priority and for most people it just isn’t the rational way to maximise their preferences either.
While I agree with pretty much all of your points here, you may have better luck persuading those who do not if you take a less confrontational approach (I still fail at this occasionally, despite much effort). It’s easier for me to accept a line of reasoning if that line of reasoning does not include the conclusion that I, personally, am evil. This would not be true if I were a perfect rationalist, but unfortunately it is not yet possible for me to escape my existence as a sack of neurons. And so it is with everyone.
The most persuasive arguments are the ones we want to believe. If you believe you are right (and you should), you should make it as easy as possible for people to agree with you.
Optimal persuasion was not my priority, emphasizing the nature of disagreement was. If deceptive use ‘rational’ was not a violation of both terminal and instrumental values then accusations of ‘counterproductive negativity’ or generally poor thinking, etc could be taken to have some approximation credibility—“it doesn’t matter so you should shut up” vs “Direct attack on core values! Destroy!”. I did originally explicitly explain things in these terms in the comment, including reference to ethical theory but that was turning into a seriously long winded tangent.
As for optimal persuasion, it would not involve a complete reply, a targeted paragraph responding to a cherry picked quote would be more effective. In fact, better still would be to make no reply at all, but create a new (much needed) post on the subject of ethics and rationality in order to direct all attention away from this one. Arguing with something gives it the credibility of ‘something worth arguing with’.
I perceive the tone of the parent comment as needlessly inflammatory (constituting a violation of niceness) and will therefore take some time out before replying to the substance (no concession on which is to be inferred from my temporary silence).
ETA: The above was written before the sentence calling me “evil” was removed. I continue to take exception to the part about an alleged “obsession with Knox” that I am attempting to “force” on anyone. I defy anyone to justify such a characterization; my charitable interpretation is that wedrifid has misunderstood something I said, and/or forgotten that my comment and the original post were written by two different people.
It should be noted that I observe the tone of the parent of my rebuttal to be aggressive, with vigorous use of shaming to present a position that undermines a core value of this community. A vigorous response should be expected.
At WrongBot’s suggestion I have removed the sentence containing the word ‘evil’. Since almost nobody except myself uses that word in a technical sense it was foolish of me to include it here. I went through planning to edit out anything else that I wrote in haste that I would remove on reflection but I was surprised to find that was the only edit I needed to make. What remains has my reflective endorsement.
It should be noted that I observe the tone of the parent of my rebuttal to be aggressive, with vigorous use of shaming to present a position that undermines a core value of this community
No, I cannot let you get away with that. The position I was presenting was that small good deeds should not be discouraged. If you are going to assert that that undermines a core value of this community (which one?), you are going to have to present a serious (and almost certainly novel) argument before you get to call me “evil”.
Absolutely no “shaming” was used in presenting this position. The charge is an ironic one, because I am in fact attempting to defend myself and any other warm-fuzzy-enthusiasts who may happen to consider themselves members of this community from being “shamed” by those who would regard with contempt any activity not (e.g.) calculated to minimize the expected number of deaths.
Epistemic rationality (which, by the way, is what I presented the Knox case as a lesson in in the first place) is, as you know, not an end in itself. At least, it isn’t the ultimate end. There has to be something to protect. And, at least in my own case, part of what I protect is that part of myself that is capable of caring about specific, individual humans, apart from “humanity” as an aggregate.
For the sake of cutting to the chase, let me now present what I think this disagreement is really about, and you can correct me if necessary. I think what is going on here is that you perceive the kind of “caring” I described above as an obstacle to epistemic rationality, which should therefore be Destroyed. Is that right, or am I being unfair?
At WrongBot’s suggestion I have removed the sentence containing the word ‘evil’...What remains has my reflective endorsement.
The position I was presenting was that small good deeds should not be discouraged.
An inefficient small good deed is a negated greater good deed requiring the same effort. In this framing, the “small good deed” is actually a bad deed, and should be discouraged.
An inefficient small good deed is a negated greater good deed requiring the same effort.
False. Time isn’t fungible, and humans demonstrably don’t make decisions that way.
Among other things, when humans are faced with too many alternatives, we usuallly choose “none of the above”… which means that the moment you complicate the question by even considering what those “greater good deeds might be”, you dramatically reduce the probability that anything whatsoever will be accomplished.
False. Time isn’t fungible, and humans demonstrably don’t make decisions that way.
False (at least I reject the incorrect generalization you use to contradict Vladmir). People who do small goods are less inclined to do subsequent goods. Given that the instincts evaluate ‘good’ more or less independently of any achievement fake ‘good deeds’ can prevent subsequent good deeds that make a difference. (This has been demonstrated.)
(Incidentally, Vladimir did not mention time at all.)
He didn’t have to. If time were unlimited, one could do any number of good deeds, and it would literally not matter how many of them you did, you could always do more… and thus there would be no competition between choices of how to use that time.
The assumption that not doing something now lets you do more later is false, however, because the time is already passing—if you choose not to do something now, this doesn’t give you any more time to do it later. Thus, a real thing done now beats an imaginary thing to be done later (which, given human psychology, probably won’t actually be done).
People who do small goods are less inclined to do subsequent goods.
Oh really? What about the FITD effect?
On the other hand, see Doing your good deed for the day (presumably what wedrifid was referring to). Figuring out which effect dominates under which circumstances seems like an important open problem. (My first, simple, guess would be that altruism is depleted in the short term and strengthened in the long term by use, like willpower or muscular strength.)
Figuring out which effect dominates under which circumstances seems like an important open problem. (My first, simple, guess would be that altruism is depleted in the short term and strengthened in the long term by use, like willpower or muscular strength.)
Given evidence for both the FITD effect and the DITF effect, I wonder if both are merely special cases of a broader effect that makes people more likely to accede to a request if they’ve received previous requests from the same source. The low-ball effect would also fit that theory.
Either way, I don’t think those wikipedia pages are very good evidence of anything at all, given that they cite work by only one researcher and do nothing but restate his conclusions with a positive slant. I suspect on those grounds that those pages are the work of a sock-puppet or someone caught in an affective death spiral; even if they’re not, they’re certainly not up to wikipedia’s usual (fairly decent) standard.
He didn’t have to. If time were unlimited, one could do any number of good deeds, and it would literally not matter how many of them you did, you could always do more… and thus there would be no competition between choices of how to use that time.
You’ll note that what he did mention was effort, an entirely different resource, particularly as it applies to humans.
The assumption that not doing something now lets you do more later is false, however, because the time is already passing
This isn’t an assumption of Vladmir’s, it is yours. What we do know is that spending $10 now is $10 that you can not spend later. More importantly given what we know about how humans spend money, $10 you are spending right now on one (completely useless) charity is $10 you are unlikely to spend within this month on an altruistic act that is, in fact, useful.
I guessed wedrifid was referring to this story. There does seem to be some evidence for people feeling that a few virtuous acts give them license to behave badly.
A good example Matt, I hadn’t come across that one specifically but I do know that studies have reliably shown that people who have done one good act feel less obliged to do another one in the short term. This is exactly what we would expect based on signalling needs. I would be rather surprised if pj hadn’t encountered such studies given his chosen occupation.
As for the argument itself: it says nothing more than that the good is bad because it isn’t perfect. That is obviously wrong, because the good is better than nothing. It shouldn’t be discouraged; rather, the better should be (separately) encouraged.
I think what is going on here is that you perceive the kind of “caring” I described above as an obstacle to epistemic rationality, which should therefore be Destroyed. Is that right, or am I being unfair?
That is not right. I disagree specifically with the claims which I quoted in my reply and my disagreement is limited to precisely that which is contained in said reply.
I approve, for example, of seeking warm fuzzies and this is entirely in line with my stated position.
I approve, for example, of seeking warm fuzzies and this is entirely in line with my stated position.
Then what, exactly, do we disagree about?
(Your earlier comment is of no help in clarifying this; in fact you explicitly described the pursuit of warm fuzzies—as would be exemplified by contributing to the causes listed in the post—as “bad decision making”.)
in fact you explicitly described the pursuit of warm fuzzies—as would be exemplified by contributing to the causes listed in the post—as “bad decision making”.
This is not the case. I explicitly describe the equivocation of ‘rational’ with any meaning apart from ‘rational’ (and the application of said equivocation when decision making) as ‘bad decision making’.
Okay, I think I see what happened. Your original point was really this:
This isn’t a religious community and ‘rational’ is not or should not be just the local jargon for ’anything asserted to be morally good
-- with which I agree. However, the following statements distracted from that point and confused me:
The examples listed are not rational. They are examples of ‘altruism’ for the sake of a ‘warm feeling’ and signalling
I make a call for a different response. I encourage people to resist the influence, suppress the irrational urge take actions that are neither optimal signals nor an optimal instrument for satisfying their altruistic values.
These made it sound like you were saying “No! Don’t contribute to those causes! Doing so would be irrational, since they’re not philanthropically optimal!” (I unfortunately have a high prior on that type of argument being made here.) My natural response, which I automatically fired off when I saw that your comment had 17 upvotes, is that there’s nothing irrational about liking to do small good deeds (warm fuzzies) separately from saving the planet.
However, as I understand you now, you don’t necessarily see anything wrong with those causes; it’s just that you disapprove of the label “rationality” being used to describe their goodness—rather than, say, just plain “goodness”.
That’s a decent take. But how do we account for people that are not most effectively modeled as agents with goals? Deontologists for example, can be evil even if their (alleged) preferences entirely match mine.
You’re talking about people who, when acting in a way that they themselves morally endorse, do not pursue the exact same goals you yourself value? In that case, there are few people on Less Wrong who aren’t evilwedrifid, much fewer (by proportion) people in your culture who aren’t evilwedrifid, and hardly anyone at all in the world who isn’t evilwedrifid. I’m not commenting on your own values, wedrifid; practically any two people will disagree about something.
Even if everyone understood your technical usage of “evil”, it wouldn’t convey much information.
That’s okay, dammit!
Look, scoffing at less-than-optimal philanthropy is ultimately just another form of counterproductive negativity. If you’re really serious about efficacy, you should be adding to the list of causes, not subtracting from it. That is, instead of responding to a post like this by encouraging people to
(!)
how about answering with “hey, you know what would be really, really helpful?” and proceeding to list some awesome utility-maximizing charity.
Warm feelings are good. Someone who donates a few spare frequent-flyer miles to help Curt Knox and Edda Mellas visit their daughter imprisoned 6,000 miles away doesn’t need to feel ashamed of themselves for not being “rational”—except in the extremely unlikely event that that action actually prevented them from doing something better. Does anyone honestly, seriously believe that discouraging people from doing things like this is a way of making the world a better place?
Speaking of challenges for LW, I propose a new rule: anybody who comes across an ostensibly good cause, but scoffs at its suboptimality, or thinks “well, it’s not that I’m not willing to sacrifice $10, but surely there are better uses of that money” should be immediately required, right then and there, to donate that $10 to the Singularity Institute—no ifs, ands, or buts.
No, no, NO! I desire to correct a fundamental mistake that is counter to whatever good ‘rationality’ may happen to provide. Raising the sanity waterline is an important goal in itself and particularly applicable in rare communities that have some hope of directing their actions in a way that is actually effective. Not only that, but seeing the very concept of rationality abused to manipulate people into bad decision making is something that makes me feel bad inside. Yes, it is the opposite of a warm fuzzy.
You are fundamentally wrong and the use of labeling things that disagree with you as ‘negative’ is non-rational influence technique that works in most places but should be discouraged here. It is not counterproductive to not do things that are stupid. It is not intrinsically better to add things to a list of normatively demanded behaviors while never removing them. If the list is wrong (for a given value of wrong) then it should be fixed by adding to it or removing from it in whatever way necessary.
People being manipulated into actions by the inclusion of irrelevant things in the definition of ‘rational’ is what I am discouraging. Tell people that Knox is a good way to purchase warm fuzzies, that’s fine. But don’t dare try to call it a ‘challenge for rationality’, piggybacking on the human instinct to avoid the shame of not supporting the tribal value (‘rational’).
No ifs and buts? Not everyone here needs to consider the SIAI to be the best use of their money. That’s not required by ‘rationality’ either. You’re in the wrong place if you think that approach is at all appropriate. Don’t try to force your obsession with Knox on everyone else. It’s not my priority and for most people it just isn’t the rational way to maximise their preferences either.
While I agree with pretty much all of your points here, you may have better luck persuading those who do not if you take a less confrontational approach (I still fail at this occasionally, despite much effort). It’s easier for me to accept a line of reasoning if that line of reasoning does not include the conclusion that I, personally, am evil. This would not be true if I were a perfect rationalist, but unfortunately it is not yet possible for me to escape my existence as a sack of neurons. And so it is with everyone.
The most persuasive arguments are the ones we want to believe. If you believe you are right (and you should), you should make it as easy as possible for people to agree with you.
Optimal persuasion was not my priority, emphasizing the nature of disagreement was. If deceptive use ‘rational’ was not a violation of both terminal and instrumental values then accusations of ‘counterproductive negativity’ or generally poor thinking, etc could be taken to have some approximation credibility—“it doesn’t matter so you should shut up” vs “Direct attack on core values! Destroy!”. I did originally explicitly explain things in these terms in the comment, including reference to ethical theory but that was turning into a seriously long winded tangent.
As for optimal persuasion, it would not involve a complete reply, a targeted paragraph responding to a cherry picked quote would be more effective. In fact, better still would be to make no reply at all, but create a new (much needed) post on the subject of ethics and rationality in order to direct all attention away from this one. Arguing with something gives it the credibility of ‘something worth arguing with’.
I perceive the tone of the parent comment as needlessly inflammatory (constituting a violation of niceness) and will therefore take some time out before replying to the substance (no concession on which is to be inferred from my temporary silence).
ETA: The above was written before the sentence calling me “evil” was removed. I continue to take exception to the part about an alleged “obsession with Knox” that I am attempting to “force” on anyone. I defy anyone to justify such a characterization; my charitable interpretation is that wedrifid has misunderstood something I said, and/or forgotten that my comment and the original post were written by two different people.
It should be noted that I observe the tone of the parent of my rebuttal to be aggressive, with vigorous use of shaming to present a position that undermines a core value of this community. A vigorous response should be expected.
At WrongBot’s suggestion I have removed the sentence containing the word ‘evil’. Since almost nobody except myself uses that word in a technical sense it was foolish of me to include it here. I went through planning to edit out anything else that I wrote in haste that I would remove on reflection but I was surprised to find that was the only edit I needed to make. What remains has my reflective endorsement.
No, I cannot let you get away with that. The position I was presenting was that small good deeds should not be discouraged. If you are going to assert that that undermines a core value of this community (which one?), you are going to have to present a serious (and almost certainly novel) argument before you get to call me “evil”.
Absolutely no “shaming” was used in presenting this position. The charge is an ironic one, because I am in fact attempting to defend myself and any other warm-fuzzy-enthusiasts who may happen to consider themselves members of this community from being “shamed” by those who would regard with contempt any activity not (e.g.) calculated to minimize the expected number of deaths.
Epistemic rationality (which, by the way, is what I presented the Knox case as a lesson in in the first place) is, as you know, not an end in itself. At least, it isn’t the ultimate end. There has to be something to protect. And, at least in my own case, part of what I protect is that part of myself that is capable of caring about specific, individual humans, apart from “humanity” as an aggregate.
For the sake of cutting to the chase, let me now present what I think this disagreement is really about, and you can correct me if necessary. I think what is going on here is that you perceive the kind of “caring” I described above as an obstacle to epistemic rationality, which should therefore be Destroyed. Is that right, or am I being unfair?
See my ETA.
An inefficient small good deed is a negated greater good deed requiring the same effort. In this framing, the “small good deed” is actually a bad deed, and should be discouraged.
False. Time isn’t fungible, and humans demonstrably don’t make decisions that way.
Among other things, when humans are faced with too many alternatives, we usuallly choose “none of the above”… which means that the moment you complicate the question by even considering what those “greater good deeds might be”, you dramatically reduce the probability that anything whatsoever will be accomplished.
False (at least I reject the incorrect generalization you use to contradict Vladmir). People who do small goods are less inclined to do subsequent goods. Given that the instincts evaluate ‘good’ more or less independently of any achievement fake ‘good deeds’ can prevent subsequent good deeds that make a difference. (This has been demonstrated.)
(Incidentally, Vladimir did not mention time at all.)
Oh really? What about the FITD effect?
He didn’t have to. If time were unlimited, one could do any number of good deeds, and it would literally not matter how many of them you did, you could always do more… and thus there would be no competition between choices of how to use that time.
The assumption that not doing something now lets you do more later is false, however, because the time is already passing—if you choose not to do something now, this doesn’t give you any more time to do it later. Thus, a real thing done now beats an imaginary thing to be done later (which, given human psychology, probably won’t actually be done).
On the other hand, see Doing your good deed for the day (presumably what wedrifid was referring to). Figuring out which effect dominates under which circumstances seems like an important open problem. (My first, simple, guess would be that altruism is depleted in the short term and strengthened in the long term by use, like willpower or muscular strength.)
That is my guess too.
Given evidence for both the FITD effect and the DITF effect, I wonder if both are merely special cases of a broader effect that makes people more likely to accede to a request if they’ve received previous requests from the same source. The low-ball effect would also fit that theory.
Either way, I don’t think those wikipedia pages are very good evidence of anything at all, given that they cite work by only one researcher and do nothing but restate his conclusions with a positive slant. I suspect on those grounds that those pages are the work of a sock-puppet or someone caught in an affective death spiral; even if they’re not, they’re certainly not up to wikipedia’s usual (fairly decent) standard.
You’ll note that what he did mention was effort, an entirely different resource, particularly as it applies to humans.
This isn’t an assumption of Vladmir’s, it is yours. What we do know is that spending $10 now is $10 that you can not spend later. More importantly given what we know about how humans spend money, $10 you are spending right now on one (completely useless) charity is $10 you are unlikely to spend within this month on an altruistic act that is, in fact, useful.
I guessed wedrifid was referring to this story. There does seem to be some evidence for people feeling that a few virtuous acts give them license to behave badly.
A good example Matt, I hadn’t come across that one specifically but I do know that studies have reliably shown that people who have done one good act feel less obliged to do another one in the short term. This is exactly what we would expect based on signalling needs. I would be rather surprised if pj hadn’t encountered such studies given his chosen occupation.
That was apparently not the argument that wedrifid was making after all.
As for the argument itself: it says nothing more than that the good is bad because it isn’t perfect. That is obviously wrong, because the good is better than nothing. It shouldn’t be discouraged; rather, the better should be (separately) encouraged.
Again, see this post.
That is not right. I disagree specifically with the claims which I quoted in my reply and my disagreement is limited to precisely that which is contained in said reply.
I approve, for example, of seeking warm fuzzies and this is entirely in line with my stated position.
Then what, exactly, do we disagree about?
(Your earlier comment is of no help in clarifying this; in fact you explicitly described the pursuit of warm fuzzies—as would be exemplified by contributing to the causes listed in the post—as “bad decision making”.)
This is not the case. I explicitly describe the equivocation of ‘rational’ with any meaning apart from ‘rational’ (and the application of said equivocation when decision making) as ‘bad decision making’.
Okay, I think I see what happened. Your original point was really this:
-- with which I agree. However, the following statements distracted from that point and confused me:
These made it sound like you were saying “No! Don’t contribute to those causes! Doing so would be irrational, since they’re not philanthropically optimal!” (I unfortunately have a high prior on that type of argument being made here.) My natural response, which I automatically fired off when I saw that your comment had 17 upvotes, is that there’s nothing irrational about liking to do small good deeds (warm fuzzies) separately from saving the planet.
However, as I understand you now, you don’t necessarily see anything wrong with those causes; it’s just that you disapprove of the label “rationality” being used to describe their goodness—rather than, say, just plain “goodness”.
Is this right?
What’s a technical definition of “evil”, then? I would say something about incompatible higher goals, but I’d find your take interesting.
That’s a decent take. But how do we account for people that are not most effectively modeled as agents with goals? Deontologists for example, can be evil even if their (alleged) preferences entirely match mine.
You’re talking about people who, when acting in a way that they themselves morally endorse, do not pursue the exact same goals you yourself value? In that case, there are few people on Less Wrong who aren’t evilwedrifid, much fewer (by proportion) people in your culture who aren’t evilwedrifid, and hardly anyone at all in the world who isn’t evilwedrifid. I’m not commenting on your own values, wedrifid; practically any two people will disagree about something.
Even if everyone understood your technical usage of “evil”, it wouldn’t convey much information.