= 27d567bc2d48d9f40e9ccc20ea370b15)
TheOtherDave comments on Welcome to Less Wrong! (2012) - Less Wrong
You are viewing a comment permalink. View the original post to see all comments and the full post content.
= 783df68a0f980790206b9ea87794c5b6)
Loading…= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
Powered by Reddit
= f037147d6e6c911a85753b9abdedda8d)
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (1430)
If I kill a person, the number of Fun-having-person-moments in the universe is reduced by the remaining lifetime that person would potentially have had. If I kill a baby, the number of Fun-having-person-moments in the universe is reduced by the entire lifetime of the person that baby would potentially have become.
Reasoning sensibly about counterfactuals is hard, but it isn't clear to me why the former involves less total Fun than latter does. If anything, I would expect that removing an entire lifetime's worth of Fun-having reduces total Fun more than removing a fraction of a lifetime's worth.
Probably true, but there's something you seem to be neglecting: Living in fear of being killed will significantly reduce the amount of fun you're having. Making it legal to kill non-person entities doesn't introduce this fear. Making it legal to kill person entities does.
Much less significantly, a culture in which you are obliged to either raise your children or see them put through foster care is also a much less fun culture to live in. And it's not clear that adding a person (ETA: particularly a person who would have been killed by their parents while still a baby given the chance) to the universe (as things stand today) will, on average, increase the amount of fun had down the line; this is why you're not obliged to be trying to have as many children as possible at all times.
If I believed the only reason nobody has killed me yet is because it is illegal to kill people, I wouldn't be very happy.
I don't understand your objection. Are you saying you wouldn't be more afraid, on average, if murder were legal?
I mean that a world where there is someone who would want to kill me, and the only reason why they don't is that they're afraid of ending up in jail, is not so much of a world in which I'd like to live.
It's not that anyone hates you; they might kill you because they're afraid of you killing them first, if there were no legal deterrent against killing.
In particular, if you had any conflict with someone else in a world where killing was legal, it would quite possibly spiral out of control: you're worried they might kill you, so you're tempted to kill them first, but you know they're thinking the same way, so you're even more worried, etc.
At least in my country, killing someone for self-defence is already legal. (Plus, I don't think I'm going to threaten to kill someone in the foreseeable future, anyway.)
Right, but "I accidentally ran over his dog, and I was worried that he might kill me later for it, so I immediately backed up and ran him over" probably won't count as self-defense in your country. But it's the sort of thing that traditional game theory would advise if killing was legal.
This really is a case where imposing an external incentive can stop people from mutually defecting at every turn.
If killing were legal (in a modern state with available firearms, not an ancient tribe with strong reputation effects), threatening to kill someone would be the stupidest possible move. Everyone is a threat to kill you, and they'll probably attempt it the moment they become afraid that you might do the same.
I don't get it... He wouldn't gain anything by killing you (ETA: other than what your father/wife/whoever would gain by killing him after he kills you), so why would you be afraid he would do that? (Also, I'm not sure the assumptions of traditional game theory apply to humans.)
If this was the case, I would expect places with less harsh penalties, or with lower probabilities of being convicted, to have a significantly higher homicide rate (all other things being equal). Does anyone have statistics about that? (Though all other things are seldom equal... Maybe the short/medium term effects of a change in legislation within a given country would be better data.)
I haven't read it yet, but I think this is basically the thesis of Steven Pinker's The Better Angels of our Nature.
Have you seen The Dark Knight? This is exactly the situation with the two boats. (Not going into spoiler-y detail.) Causal decision theory demands that you kill the other person as quickly and safely (to you) as possible, just as it demands that you always defect on the one-shot (or known-iteration) Prisoner's Dilemma.
Anyway, I think you shouldn't end up murdering each other even in that case, and if everyone were timeless decision theorists (and this was mutual knowledge) they wouldn't. But among humans? Plenty of them would.
I'm not sure where you live, but is killing someone who you think will try to kill you some day actually considered self-defense for legal purposes there? I'm pretty sure self-defense doesn't cover that in the US.
No. I guess I misunderstood what orthonormal meant by “afraid of you killing them first”...
As opposed to where? We can ban or allow murder. We can't yet do personality modifications that deep.
As opposed to this world. I don't think that, right now, there's anyone who would want to kill me.
So, if Alice murdered Bob, she had always wanted to kill him since she was born (as opposed to her having changed her mind at some point)? Probably we can't deliberately do personality modifications that deep (or do we? The results of Milgram's experiment lead me to suspect it wouldn't be completely impossible for me to convince someone to want to kill me -- not that I can imagine a reason for me to do that).
(shrug) We're both neglecting lots of things; we couldn't have this conversation otherwise.
I agree with you that the risk of being killed reduces Fun, at least in some contexts. (It increases Fun in other contexts.) Then again, the risk of my baby being killed reduces Fun in some contexts as well. I don't see any principled reason to consider the first factor in my calculations and not the second (or vice-versa), other than the desire to justify a preselected conclusion.
I agree that it's not clear that adding a person to the universe increases the amount of Fun down the line. It's also not clear that subtracting a person from the universe reduces the amount of Fun. Reasoning sensibly about conterfactuals is hard.
You've struck onto something here (taking into account your update about the risk only coming from yourself)
1) Under the current system, parents are somewhat Protected From Themselves. What if a mother, while suffering a state of affect, consciously and subconsciously knew that she was allowed to kill her baby, so she did it, then was hit with regret&remorse?
2) Under the current system, parents feel like society is pressuring them not to commit especially grave failures of parenting, which gives them a feeling of fairness.
Before I respond to this, can you reassure me that you're actually interested in my honest response to it?
Yes, and by asking this you already tipped me off that it's likely to be unpleasant to me, so please fire away.
Does the regret and remorse in case 1 actually matter? If it does, what do you want to say about parents who would feel less regret or remorse given the death of their child than given his or her continued life?
If their life is that terrible, there ought to be social services to take the child away from them and a good mechanism of adoption to place the child into. And I'm willing to pay a huge lot for that in various ways before legalizing infanticide becomes a reasonable alternative to me.
So I repeat my question: does the regret and remorse in case 1 actually matter? For example, what if a parent was regretful and remorseful about having their child forcibly put up for adoption; would that change your position?
I understand the argument that the infant's life is valuable, and am not challenging that here. It was your invoking the parent's regret and remorse as particularly relevant here that I was challenging.
Depends of what kind of parent and what kind of person they would've been if not for that incident. There's certainly evidence that their parenting could've been poor, but I believe that it could've been just fine for a significant minority of cases. I don't sympathize much with completely worthless parents, but what we have here is not a strong enough proof of worthlessness. And I feel really terrible for the "mostly-normal" parent here that I thought of (while somewhat modeling one on myself).
Is there a miscommunication here? parents who would feel more regret or remorse given the death of their child than given his or her continued life - that sounds to be, like, most parents in general, and ALL the parents whom society approves of.
Indeed you're right; I mis-wrote. Fixed.
If the only thing stopping a parent from killing their child is the illegalization of said act, then they shouldn't be parents anyway. If you can't control yourself with an infant, then the probability is pretty high that you are going to be some type of abusive parent. The child is likely going to be a net drain on society because of the low-level of upbringing.
It is probably better for the baby (and society) for it to be killed while it is a blicketless infant, than to grow up under the "care" of such a parent.
I can easily visualize that, in our world, some very quickly passing one-in-a-lifetime temptation to get rid of an infant is experienced by many even slightly unstable or emotionally volatile parents, then forgotten.
Would you really want to give that temptation a chance to realize itself in every case when the (appropriately huge - we're talking about largely normal people here) social stigma extinguishes the temptation today?
Oh, and in no way it's "only the illegalization", it's the meme in general too.
Maybe.
Suppose, for example, that what you're describing here as instability/emotional volatility -- or, more operationally, my likelihood of doing something unrecoverable-from which I generally abhor based on a very quickly passing once-in-a-lifetime temptation -- is hereditable (either genetically or behaviorally, it doesn't matter too much).
In that case, I suspect I would rather that infants born to emotionally volatile/unstable parents ten million years ago had not matured to breeding age, as I'd rather live in a species that's less volatile in that way. So it seems to follow that if the social stigma is a social mechanism for compensating for such poor impulse control in humans, allowing humans with poor impulse control to successfully raise their children, I should also prefer that that stigma not have been implemented ten million years ago.
Of course, I'm not nearly so dispassionate about it when I think about present-day infants and their parents, but it's not clear to me why I should endorse the more passionate view.
Incidentally, I also don't think your hypothetical has much to do with the real reasons for an infanticide social stigma. I support the meme, I just don't think this argument for it holds water.
Sorry, but I don't like your reasoning.
Look at the Far Eastern and Scandinavian societies. Wouldn't an emotionally unstable person being raised in one of them be trained to control their emotions to a much greater degree than e.g. in Southern Europe?
Further on the "hereditability" part; I'm really emotionally unstable (as you might have witnessed), but my parents are really stable and cool-headed most of the time; however, my aunt from my mother's side is a whole lot like me. I attribute most of my mental weirdness to birth trauma (residual encephalopathy, I don't know if it's pre- or post-natal), but I don't know whether part of it might be due to some recessive gene that manifested in my aunt and me, but not at all in my mother.
I agree that we shouldn't assume that emotionally volatile people fail upon most such temptations.
I agree that my reasoning here is cold (indeed, I said as much myself, though I used the differently-loaded word "dispassionate").
I agree that if impulse control is generally nonhereditable (and, again, I don't just mean genetically), the argument I use above doesn't apply.
I agree that different cultures train their members to "control their emotions" to different degrees. (Or, rather, I don't think that's true in general, but we've specifically been talking about the likelihood of expressing transient rage in the form of violence, and I agree that cultures differ in terms of how acceptable that is.)
I understand that, independent of any of the above, you don't like my reasoning. It doesn't make me especially happy either, come to that.
I still, incidentally, don't believe that the stigma against infanticide is primarily intended to protect infants from transient murderous impulses in their parents.
Neither do I; the reasons for its development do need a lot of looking into. I just listed a function that it can likely accomplish with some success once it's already firmly entrenched.
Yeah. I used "control" in the meaning of "steer", not "rule over".
Eh... we do have to neglect a lot of things, but I don't think I'm being biased in my choice of which to consider. Making it illegal to kill people is a huge net positive to the amount of fun had. And I've never held that other people should be allowed to kill your baby, for precisely that reason: making that legal would significantly reduce the amount of fun you'd have. I'm only arguing that you should be allowed to. So I'm not quite sure what your point is; either you've misread something of mine or, just as likely, I've misread you.
Well, this one is certainly more clear, because you can look at how it hurts the people around them, and so on. I'd give very good odds that most murders do reduce the amount of fun.
(rereads thread) Why, so you haven't. I apologize; the fear of having my baby killed (well, by anyone other than me, anyway) is as you say irrelevant to your point. My error.
This seems to be pointing out that killing could be even worse due to fear but in fact isn't. It's more of a non-argument in favour of the opposing position than an argument in favour of yours, at least is it is framed as "but something you're neglecting".
I had trouble parsing that, could you rephrase?
Keep in mind that my argument isn't just considering the consequence of killing your own child but the consequence of it being legal to kill your own child.
The phrase "but there's something you seem to be neglecting" does not make sense as a reply to the comment you quote.
Ah, we were discussing why killing people should be illegal. The reason I gave was that this reduces the amount of fun in the universe. TheOtherDave said that killing a person didn't seem like it reduced the amount of fun more than killing a newborn; my comment was to explain why I thought it did (namely, he was failing to take into account one significant way in which legalized murder decreased the amount of fun, which is that fear is not fun).
Fear is frequently fun -- ask any carnival promoter, or fans of Silent Hill. (That's small-f fun; from a big-F standpoint, we'd be looking at fear as an aspect of sensual engagement or emotional involvement, but I think the argument still holds.) Without taking into account secondary effects like grief, it's not at all clear to me that an environment containing a suitably calibrated level of lethal interpersonal threats would be less fun or less (instantaneously) Fun than one that didn't, and this holds whether or not the subject is adult.
I do think those secondary effects would end up tipping the balance in favor of adults, though, once we do take them into account. There's also a fairly obvious preference-utilitarian solution to this problem.
But the fear you get from Silent Hill is fear you can walk away from and know you're not going to be attacked by zombies and nor will your loved ones. You choose when to feel it. You choose whether to feel it at all, and how often. Making fear that is known to be unfounded available on demand to those who choose it is not even in the same ballpark as making everyone worry that they're going to be killed.
True enough, and I'm not going to rule out the existence of people calibrated to enjoy low or zero levels of simulated threat (I'm pretty sure they're common, actually). It's also pretty obvious that there are levels of fear which are unFun without qualification, hence the "suitably calibrated" that I edited into the grandparent. But -- and forgive me for the sketchy evopsych tone of what I'm about to say -- the response is there, and I find it unlikely that for some reason we've evolved to respond positively to simulated threats and negatively to real ones.
Being a participant in one of the safer societies ever to exist, I don't have a huge data set to draw on. But I have been exposed to a few genuinely life-threatening experiences without intending to (mostly while free climbing), and while they were terrifying at the time I think the final fun-theoretic balance came out positive. My best guess, and bear in mind that this is even more speculative, is that levels of risk typical to contemporary life would have been suboptimal in the EEA.
How would you feel about a society otherwise similar to our own which included some designated spaces with, essentially, a sign on the door saying "by entering this room, you waive all criminal and civil liability for violent acts committed against you by other people in this room" and had a subculture of people who hung out in such places, intermittently mutilating and murdering each other?
Death represents pretty significant disutility; if the experience was significantly life-threatening, you're attributing some correspondingly significant utility to the experience of surviving. How confident are you?
Okay, yes. However, I'm almost certain that having killers running around unchecked will not produce the optimal level and type of fear in the greatest possible number of people.
Why? A simulated threat prompts an immediate response, but killers on the loose prompts a lot of worrying over a long period of time. While fighting off a murderer might spike your adrenaline, that's not what killers on the loose will do. Instead people will lock their doors. They'll fear for their safety. They'll be afraid to let strangers into their home. They'll worry about what happens if they have a fight with their friend-- because the friend can commit murder with impunity. They'll look over their shoulders. Parents will spend every second worrying about their children. The children will have little or no freedom, because the parents won't leave them alone and may just keep them inside all the time, which is NOT optimal. People will have a lot of cortisol, depressing immune systems and promoting obesity.
That's NOT THE SAME as a single burst of adrenaline, whether from falling while climbing or from watching a movie or even from fighting for your life. So I guess you're right that it's not about whether it's real or not (though if it's a game, then when it gets too intense, you can just turn it off, and you can't turn off real life), but about the type of threat. However, the simulated threat doesn't actually make you less likely to continue living, whereas a real threat does.
Somewhat regardless of our private feelings on the matter, a tip: Forget OKCupid, do you not see how earnestly stating such beliefs in public gives your handle a reputation you might not mind in general, yet greatly want to avoid at some future point of your LW blogging - such as when wanting to sway someone in an area concerning ethical values and empathy?
Now that's pretty certain.
I'd hope that LessWrong is a community in which having in the past been willing to support controversial opinions would increase your repute, not decrease it. If we always worry about our reputation when having discussions about possibly controversial topics, we're not going to have much discussion at all.
Regardless, the fantastic thing about the internet is that handles are disposable.
We don't mind. You aren't actually going to kill babies and you aren't able to make it legal either (ie. "mostly harmless"). Just don't count too much on your anonymity! Assume that everything you say on the internet will come back to haunt you in the future - when trying to get a job, for example. Or when you are unjustly accused of murder in Italy.
EDIT: Pardon me, when I say "we" don't mind I am speaking for myself and guessing at an overall consensus. I suspect there are one or two who do mind - but that's ok and I consider it their problem.
Fair enough. But my paranoia's only going to go so far; if the choice is a policy of speaking up in relatively accepting communities with no obvious bearing on future life prospects or a policy of remaining silent, I'll take the first one.
That only has a certainty approaching 1 if we all went and forgot about CEV and related prospects.
Really? What's your estimate of the probability that Bakkot's inclusion in a CEV-calculating-algorythm's target mind-space will make it more likely for the resulting CEV to tolerate infanticide?
Pretty negligible, but still orders of magnitude above Bakkot just altering society to tolerate infanticide on his own.
I would tend to agree for what it's worth.
I think I'm not understanding you.
Call P1 the probability that Bakkot's inclusion in a CEV-calculating-algorythm's target mind-space will make it more likely for the resulting CEV to tolerate infanticide. Call P2 the probability that Bakkot isn't capable of making infanticide legal, disregarding P1.
You seem to be saying P1 approximately equals 0 (which is what I understand "negligible" to mean), and P2 approximately equals 1, and that P2-P1 does not approximately equal 1.
I don't see how all three of those can be true at the same time.
Edit: if the downvotes are meant to indicate I'm wrong, I'd love a correction as well. OTOH, if they're just meant to indicate the desire for fewer comments like these, that's fine.
Where do you get "P2 approximately equals 1"?
Giving respect to controversy for the sake of controversy is just inviting more trolling and flamewars.
I have respect for true ideas, whether they are outmoded or fashionable or before their time. I don't care whether an idea is original or creative or daring or shocking or boring, I want to know if it's sound.
The fact that you seem to expect increased respect because of controversial opinions makes me think that you when you wrote about your support for infanticide, you were motivated more by the fact that many people disagreed with you, than by the fact that it's actually a good idea that would make the world a better place.
You remind me of Hanson (well, Doherty actually) on Libertarian Purity Duels
This is not what I expect. I expect increased respect for having thought through ideas. I also expect that most people have come to conclusions differing from the LW mainstream on some ideas, and expect - or at least hope - that a willingness to express this is something respected.
The reason I chose to write about infanticide, as opposed to my position of Newcomb's paradox (one-box) - or more illustratively my position on the general shape of planet Earth (roughly a sphere), is that my position on infanticide is controversial and therefore more likely to prompt interesting discussion, whereas my positions on Newcomb's paradox or the shape of planet Earth are not. So yes, the reason I chose to write about infanticide was that many people disagreed with me, and in fact I was explicit about that in my original post. This does not mean I hold this position to be deliberately contrarian, and implying otherwise is insulting.
Let's not go off on that tangent in here, but two-boxing is hardly uncontroversial on LW: lots of one-boxers here, including Yudkowsky. I'm one too. Also, didn't you say you "want to win"?
Good catch. Not what I meant; fixed.
Not always. For any random Lesswrongian with a contrarian position you're nearly sure to find a Lesswrongian with a meta-contrarian one.
Also, notice that your signaling now is so bad from a baseline human standpoint that people's sociopath/Wrong Wiring alarms are going off, or would go off if there's more of such signaling. I think that my alarm's just kinda sensitive* because I had it triggered by and calibrated on myself many times.
*(Alas, this could also be evidence that along the line I subconsciously tweaked this bit of my software to get more excuses for playing inquisitor with strangers)
Perhaps I'll pick another handle for use outside of this thread, then.
FWIW, I disagree with you but you don't set off my "sociopath alarm". I think you and Multiheaded may not be able to have a normal conversation with each other, but each of you seems to get along fine with the rest of LW.
If it helps, I can pretty much envision what's needed for such a conversation, and understand full well that the reasons it's not actually happening are all in myself and not in Bakkot. But I don't have the motivation to modify myself that specific way. On the other hand, it might come along naturally if I just improve in all areas of communication.
Heck, I might be speaking in Runglish. Bed tiem.
I'm curious: did you?
Yes, but on consideration I think I'm just going to switch back to this one.
If it helps, my opinion of you has been raised by this thread, rather than lowered. I think very few LWians actually think less of you for this discussion, but that could just be me projecting typical mind fallacy.
That's lumping a whole lot of things together. I'd gladly hire Bakkot if I was running pretty much any kind of IT business. I'd enjoy some kinds of debate with him. I'd be interested in playing an online game with him. I probably wouldn't share a beer. I definitely would participate in a smear campaign if he was running for public office.
Do you mean that it's pretty certain that I'm not obliged to be trying to have as many children as possible at all times?
Or that it's pretty certain that the fact that it's not clear that adding a person to the universe (as things stand today) will, on average, increase the amount of fun had down the line is why I'm not obliged to be trying to have as many children as possible at all times?
Or both?
Also: how important is it to you to manage your handle's reputation in such a way as to maximize your ability to sway someone on LW in areas concerning ethical values and empathy?
Hmm. Ehhh? ...Feels like both.
Unimportant, because I'm poor at persuading the type of people who care about their status on LW anyway, and am only at all likely to make an impact on the type of person who, like me, cares little/sporadically about their signaling here.
OK, thanks for clarifying.
Quite aside from everything else, this line is needlessly grating to anyone who even slightly adheres to the Western culture's traditional values. You could've phrased that differently... somehow. There's a big difference between denouncing what a largely contrarian audience takes as the standards imposed upon them by society at large and denouncing what they perceive to be their own values. This might be hypocritical, but I guess that many LW readers feel just like that.
Go start breeding now. Or, say, manufacture defective condoms. (Or identify your real reason for not killing babies.)
Please re-read the comment thread. If you still think we're talking about my reasons for doing or not doing anything in particular, let me know, and I'll try to figure out how to prevent such misunderstandings in the future.