Being rich is (as an example) an instrumental goal, not a terminal one. I want it because I will use the money to buy things and experiences that will make me feel good, much more than having the money (and not using it) would.
Treating it as an instrumental goal doesn’t solve the problem, it just moves it back a step. Even if you wouldn’t feel miserable by being poor because you magically eliminated negative incentives you would still feel less of the positive incentives when you are poor than when you were rich, even though richness is just the means to feeling better. All of this:
If your emotions don’t change in response to having or not having a certain something then that something doesn’t count as a goal. You would be wanting something without caring about it, which is silly. You’re saying we should remove the reasons we care about X while still pursuing X, which makes no sense.
still applies.
(Except insofar as it might be altered by relevant differences between positive and negative incentives.)
Better to self-modify to suffer less due to not achieving your goals (yet), while keeping the same goals.
To clarify, what I’m contending is that this would only make sense as a motivational system if you placed positive value on achieving certain goals which you hadn’t yet achieved, I think you agree with this part but am not sure. But I don’t think we can justify treating positive incentives differently than negative ones.
I don’t view the distinction between an absence of a positive incentive and the presence of a negative incentive the same way you do. I’m not even sure that I have any positive incentives which aren’t derived from negative incentives.
Even if you wouldn’t feel miserable by being poor because you magically eliminated negative incentives you would still feel less of the positive incentives when you are poor than when you were rich, even though richness is just the means to feeling better.
Negative and positive feelings are differently wired in the brain. Fewer positive feelings is not the same as more negative ones. Getting rid of negative feelings is very worthwhile even without increasing positive ones.
But the same logic justifies both, even if they are drastically different in other sort of ways.
Forcing yourself to feel maximum happiness would make sense if forcing yourself to feel minimum unhappiness made sense. They both interact with utilitarianism and preference systems which are the only relevant parts of the logic. The degree or direction of the experience doesn’t matter here.
Removing negative incentives justifies maxing out positive incentives = nihilism.
I mean, you can arbitrarily only apply it to certain incentives which is desirable because that precludes the nihilism. But that feels too ad hoc and it still would mean that you can’t remove the reasons you care about something while continuing to think of it as a goal, which is part of what I was trying to get at.
So, given that I don’t like nihilism or preference paralysis but I do support changing values sometimes, I guess that my overall advocacy is that values should only be modified to max out happiness / minimize unhappiness if happiness / no unhappiness is unachievable (or perhaps also if modifying those specific values helps you to achieve more value total through other routes). Maybe that’s the path to an agreement between us.
If you have an insatiable positive preference, satiate it by modifying yourself to be content with what you have. If you can never be rid of a certain negative incentive, try to change your preferences so that you like it. Unfortunately, this does entail losing your initial goals. But it’s not a very big loss to lose unachievable goals while still achieving the reasons the goals matter, so fulfilling your values by modifying them definitely makes sense.
Reducing bad experience was the original subject of discussion. As I said, it’s worthwhile to reduce them even without increasing good experience. I never said I don’t want to increase good experience—I do! As you say, both are justified.
I didn’t mean to imply that I wanted one but not the other; I just said each one is a good thing even without the other. I’m sorry I created the wrong impression with my comments and didn’t clarify this to begin with.
Of course when self-modifying to increase pleasure I’d want to avoid the usual traps—wireheading, certain distortions of my existing balance of values (things I derive pleasure from), etc. But in general I do want to increase pleasure.
I also think reducing negative affect is a much more urgent goal. If I had a choice between reducing pain and increasing pleasure in my life right now, I’d choose reducing pain; and the two cannot (easily) be traded. That’s why I said before that “there’s something wrong about negative [stuff]”.
I agree with a lot of what you’re saying, I made errors too, and IMHO apologizing doesn’t make much sense, especially in the context of errors, but I’ll apologize for my errors too because I desire to compensate for hypothetical status losses that might occur as a result of your apology, and also because I don’t want to miss out any more than necessary on hypothetical status gains that might occur as a result of (unnecessary) apologies. But the desire to reciprocate is also within this apology, I’m not just calculating utilons here.
Sorry for my previous errors.
You said:
Of course when self-modifying to increase pleasure I’d want to avoid the usual traps—wireheading, certain distortions of my existing balance of values (things I derive pleasure from), etc. But in general I do want to increase pleasure.
I said:
I mean, you can arbitrarily only apply it to certain incentives which is desirable because that precludes the nihilism. But that feels too ad hoc and it still would mean that you can’t remove the reasons you care about something while continuing to think of it as a goal, which is part of what I was trying to get at.
I don’t know how you avoid this problem except by only supporting modifying incentives in cases of unachievable goals. I’d like to avoid it but I would like to see a mechanism for doing so explicitly stated. If you don’t know how to avoid this problem yet, that’s fine, neither do I.
Apologizing is indeed status signaling; I feel better in conversations where it is not necessary or expected.
When I said I was sorry, I meant it in the sense of “I regret”. I didn’t mean it as an apology and wasn’t asking for you to reciprocate. (Also, the level of my idiomatic English tends to vary a lot through the day.)
Now I regret using the expression “sorry”!
I’m glad we agree about apologies :-)
As for the problem of modifying (positive) preferences: I don’t have a general method, and haven’t tried to work one out. This is because I don’t have a way to self-modify like this, and if I acquire one in the future, it will probably have limitations, strengths and weaknesses, which would guide the search for such a general method.
That said, I think that in many particular cases, if I were presented with the option to make a specific change, and enough precautions were available (precommitment, gradual modifications, regret button), making the change might be safe enough—even without solving the general case.
I think this also applies to reducing negative affect (not that we have the ability to that, either) - and the need is more urgent there.
It’s not just about status. It also communicates “This was an accident, not on purpose” and/or “If given the opportunity, I won’t do that again” which are useful information.
It’s not clear to me where the line between status signalling and communicating useful information even is.
My dog, when she does something that hurts me, frequently engages in behaviors that seem designed to mollify me. Now, that might be because she’s afraid I’ll punish her for the pain and decides to mollify me to reduce the chances of that. It might be because she’s afraid I’ll punish her for the status challenge and decides to anti-challenge me. It might be both. It might be that she has made no decisions at all, and that mollifying behavior is just an automatic response to having caused pain, or given challenge. It might be that the behavior isn’t actually mollifying behavior at all, whether intentional or automatic, and it’s a complete coincidence that I respond to it that way. Or it might not be a coincidence, but rather the result of my having been conditioned to respond to it that way. It might be something else altogether, or some combination.
All of that said, I have no problem categorizing her behavior as “apologizing.”
I often find myself apologizing for things in ways that feel automatic, and I sometimes apologize for things in ways that feel deliberate. I have quite a bit more insight into what’s going on in my head than my dog’s head when this happens, but much of it is cognitively impermeable, and a lot of the theories above seem to apply pretty well to me too.
The Dark Arts are as nothing besides the terrible power of signaling!
I’ve read—and I have no idea how much of this is true—that in some Eastern cultures you can get bonus points in a conversation by apologizing for things that weren’t in fact offensive before you started apologizing; or taking the blame for minor things that everyone knows you’re not responsible for; or saying things that amount to “I’m a low status person, and I apologize for it”, when the low-status claim is factually untrue and, again, everyone knows it...
I live in the Northeast US, which isn’t especially “Eastern”, but I’ve nevertheless found that taking the blame for things that everyone knows I’m not responsible for to be a very useful rhetorical trick, at least in business settings.
(Warning: this came out somewhat as a rant. I don’t have the energy to rewrite it better right now.)
Honestly: stories like this terrify me. This is not exaggeration: I feel literal terror when I imagine what you describe.
I like to think that I value honesty in conversations and friendships—not Radical Honesty, the ordinary kind. I take pride in the fact that almost all of my conversations with friends have actual subjects, which are interesting for everyone involved; that we exchange information, or at least opinions and ideas. That at least much of the time, we don’t trade empty, deceptive words whose real purpose is signaling status and social alliance.
And then every once in a while, although I try to avoid it, I come up against an example—in real life too—of this sort of interaction. Where the real intent could just as well be transmitted with body language and a few grunts. Where consciousness, intelligence, everything we evolved over the last few million years and everything we learned over the last few thousand, would be discarded in a heartbeat by evolution, if only we didn’t have to compete against each other in backstabbing...
If I let myself become too idealistic, or too attached to Truth, or too ignorant and unskilled at lying, this will have social costs; my goals may diverge too far from many other humans’. I know this, I accept this. But will it mean that the vast majority of humanity, who don’t care about that Truth nonsense, will become literally unintelligible to me? An alien species I can’t understand on a native level?
Will I listen to “ordinary” people talking among themselves one day, and doing ordinary things like taking the blame for things they’re not responsible for so they can gain status by apologizing, and I will simply be unable to understand what they’re saying, or even notice the true level of meaning? Is it even plausible to implement “instinctive” status-oriented behavior on a conscious, deliberate level? (Robin Hanson would say no; deceiving yourself on the conscious level is the first step in lying unconsciously.)
Maybe it’s already happened to an extent. (I’ve also seen descriptions that make mild forms of autism and related conditions sound like what I’m describing.) But should I immerse myself more in interaction with “ordinary” people, even if it’s unpleasant to me, for fear of losing my fluency in Basic Human? (For that matter, can I do it? Others would be good at sensing that I’m not really enjoying a Basic Human conversation, or not being honest in it.)
It’s not actually that hard to accept the blame, especially if people kind of realize that it wasn’t all your fault. Which brings us to the best way of taking the blame: do it for another guy. You’ll feel good for taking the fall, he’ll feel good about not getting blamed, and the guy who lost his whole 36GB porn-collection because of your incompetence will grudgingly admit that you at least didn’t try to weasel out of it.
Then make the developer who really screwed up (if you can find him) know inprivate that he screwed up. Not just so he can avoid it in the future, but so that he knows he owes you one. And, perhaps even more importantly, he’s also likely the person who can fix it. Because, let’s face it, it sure ain’t you.
ETA: I take back my initial reaction. It’s not completely different from what TheOtherDave described. But there are some important differences from at least what I described and had in mind:
If someone else already accepted the blame, it doesn’t advise you to try to take away the blame from him and on yourself, especially if he’s really the one at fault!
It doesn’t paint being blamed as being a net positive in some situations, so no incentive to invent things to be blamed for, or to blow them u pout of all proportion
Telling off the one really at fault, in private, is an important addition—especially if everyone else is tacitly aware you’ll do this, even if they don’t always know who was at fault. That’s taking responsibility more than taking blame.
In addition, there’s a difference between a random person taking blame for the actions of another random person; and a leader taking blame for the mistakes of one of his subordinates. As far as I can tell, the situation described in the article you linked to is a bit closer to the second scenario.
See my above comment, I manage to subvert Basic Human conversation fairly well in real life.
I empathize with all of your complaints. Doing things like explicitly pointing out when you’re manipulating other people (like when I said I empathize with all of your complaints) while still qualifying that within the bounds of the truth (like I will do right now, because despite the manipulativeness of disclosure involved my empathy was still real [although you have no real reason to believe so and acknowledge that {although that acknowledgement was yet another example of manipulation ([{etc}]) }]).
For another less self referential example, see the paragraph I wrote way above this where I explicitly pointed out some problems of the norms involved with apologies, but then proceeded to apologize anyway. I think that one worked very well. My apology for apologizing is yet another example, that one also worked fairly well.
(I hope the fact that I’m explicitly telling you all of this verifies my good intentions, that is what the technique depends upon, also I don’t want you to hate me based on what is a legitimate desire to help [please cross apply the above self referential infinitely recursive disclaimer].)
Although in real life, I’m much less explicit about maniuplation, I just give it a subtle head nob but people usually seem to understand because of things like body language, etc. It probably loses some of its effectiveness without the ability to be subtle (or when you explain the concept itself while simultaneously using the concept, like I attempted to do in this very comment). Explaining the exact parts of the technique is hard without being able to give an example which is hard because I can’t give the example through text because of the nature of real life face-to-face communication.
I have adopted the meta-meta strategy of being slighly blunt in real life but in such a way that reveals that I am 1. being blunt for the purpose of allowing others to do this do me 2. trying to reveal disdain for these type of practices 3. knowingly taking advantage of these type of practices despite my disdain for them. People love it in real life, when it’s well executed. I’m tearing down the master’s house with the master’s tools in such a way that makes them see me as the master. It’s insidiously evil and I only do it because otherwise everyone would hate me because I’m so naturally outspoken.
That sounds really braggy, please ignore the bragginess, sorry.
Treating it as an instrumental goal doesn’t solve the problem, it just moves it back a step. Even if you wouldn’t feel miserable by being poor because you magically eliminated negative incentives you would still feel less of the positive incentives when you are poor than when you were rich, even though richness is just the means to feeling better. All of this:
still applies.
(Except insofar as it might be altered by relevant differences between positive and negative incentives.)
To clarify, what I’m contending is that this would only make sense as a motivational system if you placed positive value on achieving certain goals which you hadn’t yet achieved, I think you agree with this part but am not sure. But I don’t think we can justify treating positive incentives differently than negative ones.
I don’t view the distinction between an absence of a positive incentive and the presence of a negative incentive the same way you do. I’m not even sure that I have any positive incentives which aren’t derived from negative incentives.
Negative and positive feelings are differently wired in the brain. Fewer positive feelings is not the same as more negative ones. Getting rid of negative feelings is very worthwhile even without increasing positive ones.
But the same logic justifies both, even if they are drastically different in other sort of ways.
Forcing yourself to feel maximum happiness would make sense if forcing yourself to feel minimum unhappiness made sense. They both interact with utilitarianism and preference systems which are the only relevant parts of the logic. The degree or direction of the experience doesn’t matter here.
Removing negative incentives justifies maxing out positive incentives = nihilism.
I mean, you can arbitrarily only apply it to certain incentives which is desirable because that precludes the nihilism. But that feels too ad hoc and it still would mean that you can’t remove the reasons you care about something while continuing to think of it as a goal, which is part of what I was trying to get at.
So, given that I don’t like nihilism or preference paralysis but I do support changing values sometimes, I guess that my overall advocacy is that values should only be modified to max out happiness / minimize unhappiness if happiness / no unhappiness is unachievable (or perhaps also if modifying those specific values helps you to achieve more value total through other routes). Maybe that’s the path to an agreement between us.
If you have an insatiable positive preference, satiate it by modifying yourself to be content with what you have. If you can never be rid of a certain negative incentive, try to change your preferences so that you like it. Unfortunately, this does entail losing your initial goals. But it’s not a very big loss to lose unachievable goals while still achieving the reasons the goals matter, so fulfilling your values by modifying them definitely makes sense.
Reducing bad experience was the original subject of discussion. As I said, it’s worthwhile to reduce them even without increasing good experience. I never said I don’t want to increase good experience—I do! As you say, both are justified.
I didn’t mean to imply that I wanted one but not the other; I just said each one is a good thing even without the other. I’m sorry I created the wrong impression with my comments and didn’t clarify this to begin with.
Of course when self-modifying to increase pleasure I’d want to avoid the usual traps—wireheading, certain distortions of my existing balance of values (things I derive pleasure from), etc. But in general I do want to increase pleasure.
I also think reducing negative affect is a much more urgent goal. If I had a choice between reducing pain and increasing pleasure in my life right now, I’d choose reducing pain; and the two cannot (easily) be traded. That’s why I said before that “there’s something wrong about negative [stuff]”.
I agree with a lot of what you’re saying, I made errors too, and IMHO apologizing doesn’t make much sense, especially in the context of errors, but I’ll apologize for my errors too because I desire to compensate for hypothetical status losses that might occur as a result of your apology, and also because I don’t want to miss out any more than necessary on hypothetical status gains that might occur as a result of (unnecessary) apologies. But the desire to reciprocate is also within this apology, I’m not just calculating utilons here.
Sorry for my previous errors.
You said:
I said:
I don’t know how you avoid this problem except by only supporting modifying incentives in cases of unachievable goals. I’d like to avoid it but I would like to see a mechanism for doing so explicitly stated. If you don’t know how to avoid this problem yet, that’s fine, neither do I.
Apologizing is indeed status signaling; I feel better in conversations where it is not necessary or expected.
When I said I was sorry, I meant it in the sense of “I regret”. I didn’t mean it as an apology and wasn’t asking for you to reciprocate. (Also, the level of my idiomatic English tends to vary a lot through the day.)
Now I regret using the expression “sorry”!
I’m glad we agree about apologies :-)
As for the problem of modifying (positive) preferences: I don’t have a general method, and haven’t tried to work one out. This is because I don’t have a way to self-modify like this, and if I acquire one in the future, it will probably have limitations, strengths and weaknesses, which would guide the search for such a general method.
That said, I think that in many particular cases, if I were presented with the option to make a specific change, and enough precautions were available (precommitment, gradual modifications, regret button), making the change might be safe enough—even without solving the general case.
I think this also applies to reducing negative affect (not that we have the ability to that, either) - and the need is more urgent there.
It’s not just about status. It also communicates “This was an accident, not on purpose” and/or “If given the opportunity, I won’t do that again” which are useful information.
It’s not clear to me where the line between status signalling and communicating useful information even is.
My dog, when she does something that hurts me, frequently engages in behaviors that seem designed to mollify me. Now, that might be because she’s afraid I’ll punish her for the pain and decides to mollify me to reduce the chances of that. It might be because she’s afraid I’ll punish her for the status challenge and decides to anti-challenge me. It might be both. It might be that she has made no decisions at all, and that mollifying behavior is just an automatic response to having caused pain, or given challenge. It might be that the behavior isn’t actually mollifying behavior at all, whether intentional or automatic, and it’s a complete coincidence that I respond to it that way. Or it might not be a coincidence, but rather the result of my having been conditioned to respond to it that way. It might be something else altogether, or some combination.
All of that said, I have no problem categorizing her behavior as “apologizing.”
I often find myself apologizing for things in ways that feel automatic, and I sometimes apologize for things in ways that feel deliberate. I have quite a bit more insight into what’s going on in my head than my dog’s head when this happens, but much of it is cognitively impermeable, and a lot of the theories above seem to apply pretty well to me too.
Neat, then we agree on all of that. I also would prefer something ad hoc to the “solution” I thought of.
The Dark Arts are as nothing besides the terrible power of signaling!
I’ve read—and I have no idea how much of this is true—that in some Eastern cultures you can get bonus points in a conversation by apologizing for things that weren’t in fact offensive before you started apologizing; or taking the blame for minor things that everyone knows you’re not responsible for; or saying things that amount to “I’m a low status person, and I apologize for it”, when the low-status claim is factually untrue and, again, everyone knows it...
I live in the Northeast US, which isn’t especially “Eastern”, but I’ve nevertheless found that taking the blame for things that everyone knows I’m not responsible for to be a very useful rhetorical trick, at least in business settings.
(Warning: this came out somewhat as a rant. I don’t have the energy to rewrite it better right now.)
Honestly: stories like this terrify me. This is not exaggeration: I feel literal terror when I imagine what you describe.
I like to think that I value honesty in conversations and friendships—not Radical Honesty, the ordinary kind. I take pride in the fact that almost all of my conversations with friends have actual subjects, which are interesting for everyone involved; that we exchange information, or at least opinions and ideas. That at least much of the time, we don’t trade empty, deceptive words whose real purpose is signaling status and social alliance.
And then every once in a while, although I try to avoid it, I come up against an example—in real life too—of this sort of interaction. Where the real intent could just as well be transmitted with body language and a few grunts. Where consciousness, intelligence, everything we evolved over the last few million years and everything we learned over the last few thousand, would be discarded in a heartbeat by evolution, if only we didn’t have to compete against each other in backstabbing...
If I let myself become too idealistic, or too attached to Truth, or too ignorant and unskilled at lying, this will have social costs; my goals may diverge too far from many other humans’. I know this, I accept this. But will it mean that the vast majority of humanity, who don’t care about that Truth nonsense, will become literally unintelligible to me? An alien species I can’t understand on a native level?
Will I listen to “ordinary” people talking among themselves one day, and doing ordinary things like taking the blame for things they’re not responsible for so they can gain status by apologizing, and I will simply be unable to understand what they’re saying, or even notice the true level of meaning? Is it even plausible to implement “instinctive” status-oriented behavior on a conscious, deliberate level? (Robin Hanson would say no; deceiving yourself on the conscious level is the first step in lying unconsciously.)
Maybe it’s already happened to an extent. (I’ve also seen descriptions that make mild forms of autism and related conditions sound like what I’m describing.) But should I immerse myself more in interaction with “ordinary” people, even if it’s unpleasant to me, for fear of losing my fluency in Basic Human? (For that matter, can I do it? Others would be good at sensing that I’m not really enjoying a Basic Human conversation, or not being honest in it.)
Linux Kernel Management Style says to be greedy when it comes to blame.
Here are some relevant paras:
ETA: I take back my initial reaction. It’s not completely different from what TheOtherDave described. But there are some important differences from at least what I described and had in mind:
If someone else already accepted the blame, it doesn’t advise you to try to take away the blame from him and on yourself, especially if he’s really the one at fault!
It doesn’t paint being blamed as being a net positive in some situations, so no incentive to invent things to be blamed for, or to blow them u pout of all proportion
Telling off the one really at fault, in private, is an important addition—especially if everyone else is tacitly aware you’ll do this, even if they don’t always know who was at fault. That’s taking responsibility more than taking blame.
In addition, there’s a difference between a random person taking blame for the actions of another random person; and a leader taking blame for the mistakes of one of his subordinates. As far as I can tell, the situation described in the article you linked to is a bit closer to the second scenario.
See my above comment, I manage to subvert Basic Human conversation fairly well in real life.
I empathize with all of your complaints. Doing things like explicitly pointing out when you’re manipulating other people (like when I said I empathize with all of your complaints) while still qualifying that within the bounds of the truth (like I will do right now, because despite the manipulativeness of disclosure involved my empathy was still real [although you have no real reason to believe so and acknowledge that {although that acknowledgement was yet another example of manipulation ([{etc}]) }]).
For another less self referential example, see the paragraph I wrote way above this where I explicitly pointed out some problems of the norms involved with apologies, but then proceeded to apologize anyway. I think that one worked very well. My apology for apologizing is yet another example, that one also worked fairly well.
(I hope the fact that I’m explicitly telling you all of this verifies my good intentions, that is what the technique depends upon, also I don’t want you to hate me based on what is a legitimate desire to help [please cross apply the above self referential infinitely recursive disclaimer].)
Although in real life, I’m much less explicit about maniuplation, I just give it a subtle head nob but people usually seem to understand because of things like body language, etc. It probably loses some of its effectiveness without the ability to be subtle (or when you explain the concept itself while simultaneously using the concept, like I attempted to do in this very comment). Explaining the exact parts of the technique is hard without being able to give an example which is hard because I can’t give the example through text because of the nature of real life face-to-face communication.
Blargh,.
I have adopted the meta-meta strategy of being slighly blunt in real life but in such a way that reveals that I am 1. being blunt for the purpose of allowing others to do this do me 2. trying to reveal disdain for these type of practices 3. knowingly taking advantage of these type of practices despite my disdain for them. People love it in real life, when it’s well executed. I’m tearing down the master’s house with the master’s tools in such a way that makes them see me as the master. It’s insidiously evil and I only do it because otherwise everyone would hate me because I’m so naturally outspoken.
That sounds really braggy, please ignore the bragginess, sorry.
I APOLOGIZE FOR MY APOLOGY. :(